The State of Stylistics (PDFDrive)

The State of Stylistics
5
The State of Stylistics
26
Edited by
Greg Watson
Amsterdam - New York, NY 2008

Cover design: Aart Jan Bergshoeff
The paper on which this book is printed meets the requirements of

“ISO 9706: 1994, Information and documentation - Paper for documents -
Requirements for permanence”.
ISBN: 978-90-420-2428-1
Editions Rodopi B.V., Amsterdam - New York, NY 2008
Printed in The Netherlands
CONTENTS
Notes on Contributors ix
Preface xvii
Part I: Theoretical Outlooks

1. ‘“Where are you going to my pretty maid?”
“For detailed analysis”, sir, she said.’
Mick Short 1
2. A Grammarian's Funeral: On Browning,
Post-Structuralism, and the State of Stylistics
Geoff Hall 31
3. On Genuine Interdisciplinarity: Articulating
Poetics as Theory
Patricia Kolaiti 45
4. Trewe Love at Solentsea? Stylistics Vs. Narratology
in Thomas Hardy
Ken Ireland 61
5. Who Is Afraid of Stylistics? Postgraduate Students’
Responses to Stylistics
Nazan Tutas 75
Part II: Cognitive Stylistics
6. Fusion Style: Towards a Poetics of the Grotesque Body
Shun-liang Chao 91
7. Donald Barthelme’s Art of Storytelling: A Cognitive-
Semiotic Textual Analysis of ‘On the Deck,’
‘At the Tolstoy Museum’ and ‘The Baby’
Ulf Cronquist 119
8. Evoking Interest, Evoking Meaning: The Literary
Theme and the Cognitive Function of Stylistic Devices
Alfonsina Scarinzi 137
9. ‘Why Get Upset Over a Few Cases of Rhinoceritis?’:
Possible Worlds in the Theatre of the Absurd
Katerina Vassilopoulou 155
vi Contents
Part III: Corpus Stylistics

10. A Multivariate Study of Style Differences in Poetry
Vadim Andreev 177
11. e-Lears: a Corpus Approach to Shakespeare and Tate
Maria Cristina Consiglio 191
12. Measuring Text Similarity Between the Two Editions
of John Fowles’s The Magus
Yu-fang Ho 207
13. ‘My Dearest Minnykins’: Style, Gender and Affect
in 19th Century English Letters
Merja Kytö and Suzanne Romaine 229
Part IV: Pragmatics and Discourse Stylistics
14. Functional Stylistics and Peripeteic Texts
Simon Borchmann 265
15. You Can’t Put Your Foot in the Same River Once:
Relevance Stylistics and Rereading
Anne Furlong 283
16. Dialogue and Discourse Structure: A Speech Move
Analysis of Sherman Alexie’s Story
‘What You Pawn I Will Redeem’
Robert A. Troyer 303
17. Stylistic Analysis of Discoursal Identity Formation in
Biographic Interviews with Two Senior Teachers of
English in Hungary
Judit Zerkowitz 333
Part V: Stylistics in the Classroom
18. The Ellipsis of Haiku: The Effects of Poetic Ellipsis
in the Framework of Relevance Theory
Kyoko Arai 347
19. Emotion Tracking Pedagogy: Towards Better Teaching
of the National Curriculum for English
Emma Dawson 363
20. Real People or Verbal Constructs: A Stylistic Analysis
of Character in Fiction
Sarala Krishnamurthy 375
Contents vii
21. Just for Laughs: The Construction of Nonverbal Humour

Nicola Lennon 395
22. On the Phrase ‘Even with a Weight of Pleasure’ in
The Prelude (Bk 2, 178)
Ken Nakagawa 413
23. (Im)Politeness in Dramatic Dialogue: Understanding
Face-Attack in Shakespeare’s Othello
Rachel S Toddington 427
24. Mind-Style, Modality, and Poe’s ‘The Fall of the
House of Usher’
Simon Zupan 451
Bibliography 473
Index 513
Notes on Contributors
Vadim Andreev, PhD, docent and Chair, Smolensk State University

(Russia), teaches stylistics, general linguistics, and different aspects of
English. His area of research and publications include classification of
individual styles in poetry, application of multivariate analysis in the
study of styles and the problem of translation of verse.
Kyoko Arai has been working on the subject of ellipsis for
approximately five years. She is currently associate professor at the
Faculty of Business Administration of Toyo University, Tokyo, where
she teaches business English and Advertising Language.
Simon Borchmann, dr. phil, is a postdoctoral scholar at the
University of Aarhus, Denmark. His main areas of research are: 1)
Information Structure in terms of Logics of Communication, and 2)
Linguistics and Poetics. He has mainly taught in Pragmatics, Textual
Analysis, Text Linguistics, Psycholinguistics, Written Exposition, and
Philosophy of Science. He is a member of The Pragmatic Circle,
Copenhagen (founded 1977) and the initiator of the Danish Poetics
and Linguistics Circle (founded 2006).
Shun-liang Chao is a PhD candidate in Comparative Literature at
University College, London. He had studied and taught in Taiwan and
the US before moving to England. His research interest lies in
comparative arts, with emphasis on the comparison of poetry and
painting. He has published journal articles on the grotesque sublime,
Derrida and the rhetoric of comparative arts, Magritte and Surrealist
poetics, Jean-Luc Godard and (post)modern aesthetics, and so forth.
His most recent article is ‘“To Form a New Compound”: Eliot,
Bergson, and Cubism’ in Études britanniques contemporaines 31
(2006).
Maria Cristina Consiglio holds a PhD in Translation Studies at the
University of Bari, her doctoral dissertation was about Neoclassical
adaptations of Shakespeare’s plays, with a particular focus on Tate’s
King Lear. At present she is participating in a research project on the
reception in England of Alessandro Manzoni’s work. She has
published two articles on the relationship between translation and
adaptation (Adattamento e Traduzione nell’Inghilterra della
Restaurazione and La pratica del remake: una forma di traduzione?)
x Notes on Contributors
one article on the political value of Tate’s Richard II (Political

Manipulation of Shakespeare: the Case of Neoclassical Adaptations)
and a book on the Italian translation of A Clockwork Orange
(Contestazione e slang giovanile). She also collaborates with the
online review Apertamente.
Ulf Cronquist, Ph. D., presently teaches at the Department of
Literature, Gothenburg University, Sweden. He wrote his thesis on
Erotographic Metafiction: Aesthetic Strategies and Ethical Statements
in John Hawkes’s ‘Sex Trilogy’ (Acta Universitatis Gothoburgensis,
Gothenburg Studies in English, 78, 2000). His postdoctoral research
focuses on cognitive poetics and cognitive semiotics and he was a
Visiting Researcher at Center for Semiotics, Aarhus University,
Denmark 2004-2006. He is currently working on a book-length
manuscript: From Cognitive Poetics to Cognitive Semiotics. Theory
and Practice in Literary Studies.
Emma Dawson obtained her doctorate from the University of
Nottingham and is now a Teaching Fellow in Education at the
University of Keele. She has published a number of academic articles
on World Englishes Literature and is the editor of Read Around, a
series of texts in English from various parts of the world designed for
study at secondary education level (forthcoming, CCC Press, 2008).
She is also currently editing anthologies of short stories from
Cameroon and Kenya (forthcoming, New Ventures, 2008).
Anne Furlong teaches linguistics and literature at the University of
Prince Edward Island in eastern Canada where she is Assistant
Professor in the Department of English. She received her PhD from
UCL; her dissertation proposed a relevance-theoretic model of literary
(or non-spontaneous) interpretation. Her work in literary linguistics
and relevance stylistics has resulted in papers on wit, literary and
critical theory, repetition and style, and the curation of cultural
artefacts digitally represented in databases. The article in this volume
constitutes part of an ongoing project investigating the phenomenon of
repetition within a relevance-theoretic framework. Since 2003, she has
developed a significant linguistic concentration of courses in the
English department that focus on metrical theory and stylistic analysis.
Geoff Hall researched a Ph.D in literary theory at Sussex University
before turning more attention to linguistic and stylistic claims in
Notes on Contributors xi
subsequent research and publications. He is currently Senior Lecturer,

Applied Linguistics, Swansea University, U.K. where he teaches
modules in discourse, stylistics and second language learning, and
Assistant Editor of the journal ‘Language and Literature.’
Yu-fang Ho is currently a PhD student at Lancaster University. Her
main interests are in literary stylistics and corpus linguistics. Her work
in stylistics primarily focuses on prose fiction. Her thesis is a
quantitative and qualitative stylistic comparison of the two editions of
John Fowles’s The Magus. Her study explores to what extent stylistic
investigation and corpus techniques can be usefully combined.
Ken Ireland teaches for The Open University, Birkbeck College, the
University of East Anglia, and Cambridge University. He has been a
member of the Poetics and Linguistics Association since 1983, and
has presented some eight papers at PALA conferences. His
publications, in a range of journals, have focussed on comparative,
international and interartistic themes, with particular attention having
been paid to temporality, narrative theory, East/West relations, and
period-style. His most recent books are The Sequential Dynamics of
Narrative (2001), and Cythera Regained? The Rococo Revival in
European Literature and the Arts, 1830-1910 (2006).
Patricia Kolaiti is a UCL-based Greek literary theorist. Her
theoretical interests are mainly in the areas of poetics, pragmatics,
literary theory and the philosophy of the arts. Her doctoral project The
Limits of Expression: Language, Poetry, Thought was awarded a UCL
Graduate Scholarship for research in the Arts and Humanities and an
AHRC doctoral award. As an interdisciplinary thinker, she is
intellectually homeless.
Sarala Krishnamurthy has a PhD has taught at the postgraduate
level for twenty years and supervised the research of many Masters
and PhD Scholars in the areas of Stylistics, English Language
Teaching, Post Colonial Literature and Feminist Theory. She has
presented papers on English Language Teaching and teacher
education at several international conferences all over the world and
was a special invitee to the TESOL conference at Dakar, Senegal in
2004. She has published in international journals and books in
Literature, Applied Linguistics and English Language Teaching. She
is currently the Dean of the School of Communication at the
xii Notes on Contributors
Polytechnic of Namibia and was responsible for starting the Namibian

English Teachers' Association. At present, she is working on a
research project which is studying the impact of the Language Policy
on English at the tertiary level in five SADC countries: Namibia,
Botswana, South Africa, Zimbabwe and Zambia.
Merja Kytö is Professor of English Language at Uppsala University.
She specializes in historical linguistics and corpus studies. She has
participated in the compilation of various historical electronic corpora,
among them the Helsinki Corpus of English Texts, A Corpus of
English Dialogues 1560–1760, and A Corpus of Nineteenth-century
English. In addition to grammaticalization processes, her current
research interests include characteristics of spoken interaction of the
past and manuscript studies of early British and American English
speech-related documents. Her recent project on Early Modern
English depositions aims at an electronic edition comprising
searchable and readable texts transcribed from documents collected
across various localities in England.
Nicola Lennon is a doctoral student and part-time tutor at Queen’s
University, Belfast. She was awarded her MA in English Language
and Linguistics from QUB with Distinction in 2006 and has been the
recipient of a number of scholarships and awards including the Arts
and Humanities Research Council Doctoral Award and Masters
Scheme Award. Her primary research interest lies in the application of
existing theories and methodologies to the analysis of discourse. Her
PhD studies, entitled The Language Game: Networks of Choice in
Strategic Interaction, investigate how techniques from the burgeoning
field of Game Theory and Linguistics may be applied to the study of
media and political discourse.
Ken Nakagawa is Professor of English Philology and Stylistics at
Yasuda Women’s University, Hiroshima, Japan. His major
publications are: The Language of William Wordsworth: A Linguistic
Approach to Poetical Language (in Japanese) (Hiroshima: Research
Institute for Language and Culture, Yasuda Women’s University
[1997]); ‘The Vocabulary that Constitutes The Prelude’ in Studies in
Modern English: The Twentieth Anniversary Publication of the
Modern English Association (Tokyo: Eichosha: 457-71 [2003]);
‘“Through” in The Prelude’ in Voyages of Conception: Essays in
English Romanticism (Tokyo: Kirihara Shoten: 118-33 [2005]), and
Notes on Contributors xiii
‘A Structural Analysis of Wordsworth’s ‘Daffodils’’ in The Writer’s

Craft, the Culture’s Technology (Amsterdam: Rodopi: 85-96 [2005]).
Suzanne Romaine has been Merton Professor of English Language at
the University of Oxford since 1984. Her research interests lie
primarily in historical linguistics and sociolinguistics, especially in
problems of societal multilingualism, linguistic diversity, language
change, language acquisition, and language contact in the broadest
sense. One of her most recent books, Vanishing Voices. The Extinction
of the World's Languages (New York: Oxford University Press,
2000), written jointly with Daniel Nettle, won the British Association
of Applied Linguistics Book of the Year Prize for 2001. Romaine and
Kytö have been collaborating for a number of years on a variety of
corpus-based historical studies, in particular on adjective comparison
and grammaticalization. One of their most recent papers appeared in
the Journal of Historical Pragmatics in 2005 and they are currently
working on a monograph on the grammaticalization of constructions
to be published by Cambridge University Press.
Alfonsina Scarinzi is a doctoral candidate at Georg–August
Universität Göttingen (Germany). Her field of research is literature
and cognitive science, as well as cognitive neuroscience in the
humanities. From 1996 to 2002 she studied German, English and
Communications at Georg–August Universität Göttingen (Germany)
and earned her M.A. in 2002. From 2003 to 2004 she studied French
at ‘Académie de Nice’ (France) and worked as a lector for some
publishers. She has written book reviews and research articles for the
New German Review and Orbis Litterarum.
Mick Short is Professor of English Language and Literature at
Lancaster University, UK. His main interests are in the linguistic
stylistic analysis of literary texts and in stylistics pedagogy. He is the
author of Exploring the Language of Poems, Plays and Prose
(Longman 1996) and, with Elena Semino, Corpus Stylistics: Speech,
Writing and Thought Presentation in a Corpus of English Writing
(Routledge 2004). He is the editor of Reading, Analysing and
Teaching Literature (Longman 1989) and co-editor of Using Corpora
for Language Research (with Jenny Thomas, Longman 1996) and
Exploring the Language of Drama (with Peter Verdonk and Jonathan
Culpeper, Routledge 1998). He has written more than 70 articles on
stylistics and stylistics pedagogy. He is currently working on a book
xiv Notes on Contributors
on the stylistic analysis of drama and film.

Professor Short was the founding editor (1992-6) of Language and
Literature, the international journal of the Poetics and Linguistics
Association (which he also founded). In 2000, he was awarded a
National Teaching Fellowship by the Higher Education Funding
Council for England. He used his prize for a project which produced
an innovative introductory web-based Stylistics course in order to
investigate student responses to web-based vs. more traditional
teaching of Stylistics. Language and Literature 15, 3 (August 2006) is
a special issue of the journal devoted entirely to these investigations.
The web-based course is available free to all at
http://www.lancs.ac.uk/fass/projects/stylistics/start.htm
Rachel Toddington is a student at the University of Huddersfield.
She completed her BA in English with Creative Writing in 2006 and is
now continuing with a Masters degree in Modern English Language.
She is interested in Pragmatics and stylistics. She is particularly
interested in the interface which exists between impoliteness and
humour, and hopes to explore this relationship further at the PhD
level.
Robert Troyer is a lecturer at the Language Institute of
Chulalongkorn University, Bangkok, and a Ph.D. candidate in the
university’s English as an International Language Program. His
research combines World Englishes with literary stylistics for cross-
cultural comparison. In his dissertation he applies conversation and
discourse analysis to literary dialogue composed in different varieties
of English in order to reveal cultural values that are present in
conversations. He has presented papers at conferences of the Poetics
and Linguistics Association (PALA) as well as previous work in
rhetoric and composition pedagogy to the Assembly of Teachers of
English Grammar (ATEG).
Nazan Tutas is Assistant Professor of English Language and
Literature at Ankara University, Turkey. She has published various
articles on the teaching of literature in EFL contexts, reader-response
approach to literature teaching, and stylistic analysis of literary genres.
She is recently interested in stylistics and language teaching.
Katerina Vassilopoulou obtained her PhD from Lancaster
University, UK. Her thesis, entitled “A Cognitive Stylistic Approach
Notes on Contributors xv
to Absurdity in Drama With a Particular Focus on Ionesco’s The Bald

Prima Donna” surveys how absurdity is linguistically conveyed in the
literary movement known as the Theatre of the Absurd through a
comparative approach of dramatic plays by Eugene Ionesco, Harold
Pinter and Samuel Beckett. Her research interests include pragmatics,
stylistics and cognitive linguistics.
Greg Watson is currently Professor of English Language and Culture
at the University of Joensuu, Finland. His primary areas of research
include Linguistic Stylistics and Language Contact studies. His most
recent primary publications include Doin Mudrooroo. Elements of
Style and Involvement in the Early Prose of Mudrooroo; Finno-Ugric
Language Contacts (with P. Hirvonen); and Literature and Stylistics
for Language Learners. Theory and Practice. (with S. Zyngier).
Judit Zerkowitz presently works at Eötvös Loránd University,
Budapest, Hungary, School of English and American Studies,
Department of English Applied Linguistics. Her main research interest
is stylistics, and the discoursal creation of the professional identity of
language teachers. Her teaching includes survey lectures on applied
linguistics, seminars on methodology, stylistics, academic skills, and
practical teacher training.
Simon Zupan is a graduate student of American literature at the
University of Ljubljana, Slovenia. In his MA thesis he studied the
reception of Edgar Allan Poe’s prose in the Slovene literary system,
and the existing Slovene translations of Poe. He is currently
examining Slovene translations of Ernest Hemingway’s prose. His
research interests include American literature, stylistics, and
translation studies. Since 2005, he has worked at the University of
Maribor, where he teaches in the Translation Studies Programme.
Preface
This collection of papers represents the culmination of the 26th annual

conference of The Poetics and Linguistics Society (PALA) which was
held at the University of Joensuu, in Eastern Finland, in July 2006.
The State of Stylistics is the fifth volume of the highly successful and
widely regarded PALA Papers series. The Joensuu conference proved
to be a great success, with over 130 participants in attendance
representing nearly every corner of the globe. The theme of the
conference, The State of Stylistics, was intentionally ambiguous.
Although the title clearly asserts that stylistics can stand alone, that it
is a state unto itself, it also encourages us to consider where we stand
in this field of study as we enter the 21st century, that is, what state is
the field of stylistics currently in. How has the field of stylistics
developed, or has it even, in what directions is it heading, and how do
we stylisticians perceive our own discipline as we move forward in
our respective fields of research and teaching? What, essentially, is
stylistics, how do we or should we use it and how has it evolved to
this point in time?
As evidenced by the number of papers given at the conference, there
were no shortage of takers willing to offer their views on these matters
and this is also reflected in the size of this volume. The State of
Stylistics contains 24 papers by 25 authors arranged into five parts:
theoretical outlooks, cognitive stylistics, corpus stylistics, pragmatics
and discourse analysis, and stylistics in the classroom. Each of these
broad divisions reflect different trends in stylistics and also reflect the
interdependency of our field upon applied linguistics and literary
studies, a factor which, I would argue, makes the study of stylistics so
attractive and useful in the classrooms of today. Apart for Part 1, in
which the papers are arranged according to broad perspectives and
philosophical reflections upon where stylistics should be heading, all
of the papers in each of the subsequent parts are arranged in
alphabetical order according to author.
In Part 1, Theoretical Outlooks, Mick Short’s insightful paper offers a
review of where modern stylistics has come from and where he would
like to see it head towards. In doing so, he warns us not to get caught
up in the agendas and rapidly-changing fashions of nearby fields of
study, that we must be circumspect of these new lines and that we
xviii Preface
must remember to pay close attention to detailed analysis when we do

use these new lines of investigation. Short also wants us to broaden
the range of texts we have traditionally examined but when doing so
we should also re-establish links between detailed analysis on the one
hand and sensitive reading encompassing an aesthetic appreciation on
the other. Geoff Hall’s paper also calls upon us to broaden our
approach towards stylistic enquiry. He wants us to consider cultural
and historical understandings when examining texts, in order to move
beyond isolated and incomplete accounts of individual texts. Hall
would like us to move towards a better rapprochement between an
enquiring stylistics and intelligent literary study. He exemplifies this
with a brief study of Browning’s A Grammarian’s Funeral. Patricia
Kolaiti’s paper is concerned with the idea of a genuine
interdisciplinarity. She argues that poetics/stylistics has been taken
over by too much “applicationization”. To counter this, she offers
some concrete proposals on how cognitive pragmatics could help to
create a genuinely theoretical and interdisciplinary poetics, less
influenced by what she sees as an ever-growing imbalance between
application and theory. Ken Ireland’s paper is also theoretical in
outlook in the sense that he shows us, through an examination of
Thomas Hardy’s An Imaginative Woman, that there are distinct
advantages in questioning the idea of approaching a text from just one
critical approach, but that several approaches, in this case a
narratological and then a stylistic approach, can show the text in
different or more revealing lights. He urges us to question the
advantages and the limits of various singular approaches and to merge
certain approaches when examining a text. Ireland, essentially,
advocates the above-mentioned viewpoints of Short and Hall. The
final paper in Part 1 is by Nazan Tutas, who offers a slightly different
theoretical slant in the sense that Tutas examines the opinions of
postgraduate students towards the use of stylistics. It is interesting
because it reveals to us the opinions of those unfamiliar with the field
of stylistics both before and after they have been exposed to an
introductory course in stylistics. Through this procedure Tutas tries to
debunk certain preconceived assumptions about stylistics, for example
that it considers itself superior to other forms of literary study, that it
excludes any response or appreciation in its interpretations, and that
the claims to objectivity that stylistics advocates is suspicious.
Preface xix
Part 2 of this volume centres around Cognitive Stylistics. Shun-liang

Chao’s paper is concerned with a psychoanalytic reading of the
grotesque body and aims at offering a fusion style that encompasses
the work of Freud, Lacan, Barthes and Kristeva. Ulf Cronquists’s
paper examines three short stories of Donald Barthelme from a
cognitive-semiotic viewpoint. In so doing, he introduces a model for
stratified reading and interpretation and applies three cognitive-
semiotic tools for textual interpretation: the notions of evidentiality
and enunciation, narrative diegesis and blending and conceptual
integration. Alfonsina Scarinzi’s paper examines aspects of reader-
response and is interested in examining how stylistic devices can
affect and guide the reader to a cognitive understanding and thematic
interpretation of a literary text. The final paper of Part 2 is by Katerina
Vassilopoulou, who explores the possible-worlds theory in relation to
absurdist drama, by using selected extracts from Pinter, Ionesco and
Beckett to highlight her claim that an application of possible-world
theory to absurdist drama can prove very useful in understanding the
oddity of this genre.
Corpus Stylistics is the central theme of Part 3. Vadim Andreev’s
paper is a stylometric study of four 19th century American poets,
according to 43 parameters. He found that his model is capable of
clearly discriminating between 95% of all the texts, and is able to
recognise the degree of similarity and difference between the
individual idiostyles of the chosen poets: Bryant, Emerson,
Longfellow, and Poe. Maria Cristina Consiglio’s paper applies corpus
linguistic tools and methodology in order to examine and compare the
work of Shakespeare’s King Lear and Nahum Tate’s 1681
neoclassical adaptation. By using the readily available WordSmith
Tools software, Consiglio examines the three words “fool”, “nature”
and “nothing” in Shakespeare’s King Lear, and that of “love” in
Tate’s version. Her findings confirm the findings of more “traditional”
research and highlight the usefulness of corpus studies. Yu-fang Ho’s
study of two different editions of John Fowles The Magus is an
excellent example of comparative stylistics that enlists the use of
corpus techniques in her analysis, whilst posing two questions: what
difference is there between the two editions, and to what extent can
stylistic investigation and corpus techniques be combined? The final
paper in Part 3 is by Merja Kytö and Suzanne Romaine. This is a
highly thorough study of aspects of affect in 19th century letters based
xx Preface
on the contents of the CONCE (Corpus of Nineteenth Century

English). Kytö and Romaine pay particular attention to how address
terms are useful in understanding interpersonal attitudes and social
relations between participants in speech events.
Part 4 deals with aspects of Pragmatics and Discourse Analysis.
Simon Borchmann’s paper deals with the dilemma faced by those who
wish to use pragmatically based linguistic description to examine
literary language, as it has been shown that literary language often
violates the norms of linguistic description. Borchmann offers a
framework to deal with this problem and a model for the description
of a specific type of literary texts. Anne Furlong’s paper discusses
what she terms relevance stylistics and rereading. Primarily, she is
concerned with understanding what occurs when we reread and
whether this repetition is productive or non-productive. In fact,
Furlong claims it may not produce new effects but rather help the
reader to experience certain effects again, instead of discovering new
ones. Robert A. Troyer’s paper offers a discourse analysis framework
which can be applied to dialogue in fiction. This framework is an
elaboration of Halliday’s functional approach combined with
traditional conversation and discourse analysis and speech act theory.
He then applies this framework to the dialogue of the short story What
You Pawn I Redeem by Sherman Alexie. The final paper in this part is
that of Judit Zerkowitz, who offers a stylistic analysis of discoursal
identity. This is an interesting paper in that it offers reflections on the
rapid shifts that have occurred in socio-political and professional
contexts in Hungary during the past twenty years.
The final part of this volume deals with papers related to Stylistics in
the Classroom, and, perhaps not surprisingly, it contains the largest
number of papers. I find this personally satisfying, as it shows how
active we are in bringing stylistic analysis to the fore and how we are
trying to use stylistic techniques to explain texts to our respective
students. It is a reflection of the vibrancy of stylistic enquiry. I might
add here that although some of the following papers may not have
been specifically written for the classroom, I have placed them in this
part because their stylistic analyses are so thorough and astute that
they could and should be used in the classroom to help highlight the
intricacy and richness of language. Of course, most of the preceding
papers could also be recognised for their pedagogic value. Kyoko
Preface xxi
Arai’s paper shows us how Haiku poetry can be used to examine the
concept of ellipsis from a relevance theory framework. To make a
good Haiku the author has to choose highly relevant words that,
ironically, inspire many weak explicatures and implicatures. Emma
Dawson’s paper discusses emotion tracking. She is worried that
literature from different cultures and traditions are not being properly
taught according to the year 8 English curriculum. In an attempt to
help remedy this, she offers a form of pedagogy which combines
aspects of World Englishes literature, stylistics and emotion study.
Sarala Krishnamurthy uses Things Fall Apart by Chinua Achebe to
help argue her point that characters within literature should only be
viewed as verbal constructs. She claims that critics can be wrong in
their interpretation of a character and that these characters should be
subjected to linguistic analysis just as speech, narration or
focalization. She bases her theoretical premise on the work of
Rimmon Kenan, Halliday and motif analysis from narratologists.
Nicola Lennon’s paper examines a particular form of humour, that of
non-verbal humour. She uses aspects of Nash’s frameworks to
examine the architecture of non-verbal, particularly filmic texts and
shows how this might be used by applying it to the well-known
comedy show Just for Laughs. Ken Nakagawa’s paper is concerned
with a stylistic analysis of the phrase “Even With a Weight Of
Pleasure” from Wordsworth’s The Prelude. This is an exceptionally
astute piece of stylistic investigation that involves historical stylistics,
corpus linguistics and a 21st century reading of the phrases under
investigation. This is an example of the detailed analysis that Mick
Short speaks of in the opening paper of this volume. Rachel S.
Toddington’s paper is no less interesting. She offers a detailed
analysis of face attack in Shakespeare’s Othello, with particular
attention being paid to the devious dialogue of Iago. She arrives at the
term Discourse Disjunctive Politeness to account for the two tiered
effect found within the play and offers a revision of Culpeper’s model
of impoliteness as a consequence, one which takes the context of
drama into account. The final paper within this volume is that of
Simon Zupan’s. Through an excellent reading of Poe’s The Fall of the
House of Usher, Zupan shows us how modality can affect one’s mind-
style, and help to create the gothic atmosphere of this short story,
characterised by elements of fear, suspicion and uncertainty.
xxii Preface
From this brief preview of the papers presented here the vibrancy of
present day stylistic study should be more than apparent. This volume
contains an excellent array of papers from a broad range of scholars,
situated at various stages in their careers, who present a kaleidoscope
of approaches, and represent multiple cultures and language groups. I
would argue that, beyond a shadow of doubt, we truly can claim that
there is a state of stylistics, and like any state today, it is one
experiencing flux and introspection, yet, after having clearly and
firmly established itself in the academic world, it is not stagnant but
rather boldly stepping forward into new territories.
Greg Watson
Joensuu, Finland. 2008
PART I
THEORETICAL OUTLOOKS
‘“Where are you going to my pretty maid?”

“For detailed analysis”, sir, she said.’
Mick Short
University of Lancaster
Abstract
My main aim in this paper is to review what Stylistics has achieved since its modern
Western incarnation in the 1960s, what remains to be done and what I think we need
to concentrate on in the near future. I will argue that the gains we have made have
come about in large part because of a concentration on detailed and systematic textual
analysis related to Stylistics-based theories of textual understanding, and that, if we
are to continue to be successful in what we do, we need to continue to concentrate on
these areas, as well as being clear about what we are trying to achieve. This, in turn,
means that, although we should continue to be a ‘broad church’, (i) we should not be
driven by all the agendas and rapidly-changing fashions of nearby areas, but should be
more circumspect about the new ‘lines’ we move into and (ii) we should subject the
new areas of analysis we embark on to the same level of scrutiny and care that we
have used in our earlier work. I also suggest that we need (i) to broaden in an orderly
way the range of texts that we analyse and (ii) to go back to some ‘old chestnuts’ from
aesthetics and early literary theory, and investigate and debate them in detail and with
care. In particular, I think that we need to examine more carefully than we have so far
what counts as an interpretation of a text, what counts as an alternative interpretation
and what counts as another, slightly different instantiation of an established
interpretation. I also think that we need, for both theoretical and pedagogical reasons,
to re-establish the links between detailed analysis on the one hand and, on the other,
sensitive reading, textual appreciation and the aesthetic properties of texts. These are
all rather broad issues, but I will try to ground them in the discussion of concrete
examples in section 5.
Keywords: aesthetic judgment; affective response; appreciation; effects; evaluation;
interpretations; readings; sensitivity; Stylistics; understanding; value.
2 Mick Short
1. What we have achieved in the last 40 years and how we have

achieved it
Looking back for this paper on the history of Stylistics since the early
seminal work which established the field (e.g. Fowler 1966, 1971;
Freeman 1970; Jakobson 1960; Leech 1969) has been a more pleasant
experience than I had expected. There is plenty still to be done, of
course, and our ability to understand and respond to texts will always
run well in advance of our ability to analyse them, and explain how
meanings and effects are created when we read them. But,
nonetheless, Stylistics has come a long way in what is a short time in
academic terms, even though the number of practitioners is relatively
small. Most of the early work was on the analysis of poetry, but we
have extended our work to cover prose (e.g. Fowler 1977; Leech and
Short 2007 [1981]; Toolan 1990, 2001 [1988] and drama (e.g. Burton
1980, Culpeper, Short and Verdonk 1998, Herman 1995, McIntyre
2006). Chapter 11 of Leech and Short (2007 [1981]) reviews in more
detail the developments of the stylistics of prose fiction between 1996
and 2007. Some of this work has involved us in adding new forms of
analysis to our ‘stylistician’s toolkit’, as I like to call it. For some of
these new approaches we have been dependent, to some degree, on
neighbouring areas like descriptive Linguistics (e.g. discourse and
pragmatic analysis) and Narratology (e.g. story structure analysis), but
we have also developed others for ourselves (e.g. foregrounding
theory, point of view analysis, speech, writing and thought
presentation analysis) – though we should remember that the linguists
and the narratologists, in particular, have done interesting work in
these areas too. We have also begun to explain how major literary
critical concepts like characters (e.g. Culpeper 2001), satire (e.g.
Simpson 2004) and fictional worlds (e.g. Semino 1997, Werth 1999)
are perceived in texts, and this has increasingly involved us in trying
to understand, in general terms, how readers interact with texts to infer
what is ‘behind’ or ‘underneath’ the language of those texts.
One consequence of the felt need of stylisticians to ground their
theories and descriptions in detailed practical analysis is that we now
have a wealth of good analyses of short works and textual extracts.
This is particularly true for poetry and prose fiction, but more work on
drama is coming out all the time, and non-literary texts and popular
fiction have also been examined (e.g. Cook 1992, Nash 1990). This
‘Where are you going to, my pretty maid?’ 3
wealth of good extant analyses is leading to more people being

interested in Stylistics, what it does and how it does it; and our
academic neighbours are beginning to take us a bit more seriously
than in the past. But there is still plenty to be done if we are to achieve
our basic aim of explaining how we understand and respond to
particular texts and texts in general. And if we are to be taken
seriously we need to extend our work to cover not just modern literary
and non-literary texts but also texts from more distant historical
periods.
Another factor which has begun to increase interest in Stylistics is
that, because we have always had to work hard to excite our students
about what we do, we have been consistently interested in producing
books which students will find accessible, and in Stylistics pedagogy
more generally. Many of the books referred to above are read by
students, and a number of others (e.g. Short 1996; Simpson 1997,
2003; Toolan 1997) have been written specifically for students. I am
hoping that the fact that my introductory web-based course
<http://www.lancs.ac.uk/fass/projects/stylistics/start.htm> is free to all
worldwide will help more students to become interested in Stylistics.
Short, Busse and Plummer (2006) contains reports of a series of
pedagogical investigations involving this course. Work on the stylistic
approach to literary texts and how it can be of use for foreign learners
can also be found in, for example, Brumfit and Carter (1986), Carter
and McRae (1996), Short (1989), Watson and Zyngier (2007),
Widdowson (1975 and 1992) and in the journals Language and
Literature and Style.
2. How have we achieved what we have achieved?

We have achieved quite a lot then; but how have we done this? First
and foremost, we have avoided doing the easy thing. Much literary
commentary involves critics writing down, sometimes in reasonable
detail, their responses to the works they have read, and theorizing in
general terms about the nature of literature and how readers respond to
it. We do this too, of course, but Stylistics also brings something extra.
Because we are interested in the ‘nuts and bolts’ of texts and how we
respond to them, we have consistently forced ourselves to be much
more analytical when approaching texts, and to be very detailed and
systematic in that analysis.
4 Mick Short
Making a point of being analytical, detailed and systematic in analysis

and punctilious in argumentation may look to some like mere
inconvenient cussedness. But I think that it brings something very
important to what we do. It is only by being detailed and systematic
that we can properly begin to see how the myriad factors in a text, and
how we respond to it, are integrated to induce specific understandings
and effects in readers.
I also think that forcing ourselves to be analytically detailed and
systematic has helped us to resist, to some degree at least, the
blandishments of (particularly literary) academic fashion. Most
stylisticians work in English departments and understandably feel the
need to relate what they do to the concerns of their literary colleagues,
who are in the majority. But English Literature, as Henry Widdowson
has pointed out, is arguably a subject but not a discipline. This, I
suspect, is a significant factor. All areas of academic life are subject to
changes in fashion (in Biology, genes are in vogue these days, for
example). But literature study, especially literary theory, seems to be
particularly open to such changes, probably because it has less of a
sense of being a discipline than most other subjects (compare History,
for example). As far as I can see, although the sciences and the more
scientific social sciences change what they do, the fashions one finds
in those fields are, most of the time, still clearly connected to the
shared underlying aim of accurately characterizing and accounting for
what they are trying to explain, and many of the changes that have
come about are a consequence of the fact that new discoveries and the
development of new methodologies lead fairly naturally into new
areas needed for that goal to be achieved. So although the changes can
be described as changes in fashion, they are not very like particular
colours being ‘in’ or ‘out’ in the fashion industry or particular
approaches being ‘in’ or ‘out’ in literary theory. If I am right, then, our
analytical discipline, although it sometimes seems a nuisance to have
to adhere to it, is an advantage, not a disadvantage. It is what helps us,
for example, to resist the easy and unhelpful (and in my view
inaccurate and not well justified – see sections 4 and 5) assumption
common in modern literary study that there are as many
understandings of texts as there are readers to understand them.
Indeed, systematic analytical detail, although it certainly does not
guarantee complete agreement by any means, does lead to a healthy
analytical and interpretative overlap when two stylisticians
independently study the same text. Good examples of this are (i)
Fowler’s (1996 [1986]: 168-9) and Leech and Short’s (2007 [1981]:
6.4) analysis of the opening of The Sound and the Fury by William
Faulkner and (ii) Nash’s (1982) and Leech and Short’s (2007 [1981]:
3.4) account of the opening of ‘Odour of Chrysanthemums’ by D. H.
Lawrence.
3. New approaches: new (and old) hopes and dangers

One of the reasons this is an important time in the development of
Stylistics is that, in addition to what we have already achieved (which
has already involved changes in what we do, as the moves from poetry
analysis to prose and then drama have shown), we currently have a
number of exciting new approaches and methodologies which are
becoming available to us. These new possibilities are already leading
to new fashions in our field. We all want to be in the vanguard of new
developments, after all. PhD students wanting university posts and
young lecturers wanting to make their names in the field rarely want
to use ‘old’ methodologies or study texts which they see as ‘old hat’,
even if plenty of useful work still remains to be done in those areas.
And finally, grumpy old stylisticians like me never want to feel we are
falling behind our younger, inventive colleagues either. But it is
important for us to make sure that our changes in fashion are more
like those in science than what happens in much current literary
theory.
The new approaches and analytical methodologies which we now
have available to us, and which are already providing us with new
insights and new results, can perhaps be grouped together under four
headings (though it is worth remembering that there are also overlaps
among them):
1. corpus-based approaches to style and text analysis;
2. narratological approaches to text worlds and story
structure;
3. ‘cognitive’ approaches to these and other areas (e.g.
viewpoint, metaphor, affect); and
4. empirical approaches to text understanding and response.
6 Mick Short
The first thing to notice is that in a sense these approaches are not
completely new, even though they offer us very interesting new things
to do and think about. The use of computers to study style and
compare text-based data sets has been with us ever since computers
became available, as a look at Doležel and Bailey (1969) and the
Literary and Linguistic Computing journal shows. This is because the
accurate analysis of authorial, group and text style has always been
one of our abiding concerns. The study of style became less
fashionable in the 1980s and 1990s, precisely because it had become
difficult, with the tools available at that time, to move forward
significantly in this area, while other areas in Stylistics had become
more ‘do-able’. But the developments in corpus linguistics, with the
advent of corpus tools like Wordsmith, Textant and Wmatrix have
made it much easier for stylisticians to do this kind of work
effectively, as Hardy (2003), Hoover (1999) and Semino and Short
(2004), among others, have shown.
Similarly, narratologists have been working on text worlds and story
structure since at least the 1960s (e.g. Todorov 1969, Brémond 1973),
and some of that work still has significant uses for us (see Leech and
Short 2007 [1981]: 12.3.4). In the period when ‘structuralism’ and
‘formalism’ became demerit terms in most literary circles, work on
story structure virtually disappeared in Stylistics, but the possible
worlds work of Ryan (1991) and others has more recently helped us to
analyse narrative and other texts with more analytical ‘purchase’ (see
Semino 1997: chapters 4 and 5 and Leech and Short 2007 [1981]:
12.3.7) and the recent attempts to integrate narrative theory with
pragmatics and models of cognition (Herman 2002, 2003) hold out yet
more promise.
Semino and Culpeper (2002: ix) have already pointed out in the
foreword to their Cognitive Stylistics collection that cognitive
Stylistics is ‘both old and new’. As they say:
Cognitive stylistics . . . is old in the sense that, in focusing on
the relationship between linguistic choices and effects, stylistics
has always been concerned with both texts and readers’
interpretation of texts. . . . Foregrounding theory (Mukarovsky
1970), which played a major role in the development of modern
Anglo-American stylistics, is concerned with the cognitive
effects of particular linguistic choices and patterns (and this in
spite of the fact that it stemmed from a school known as

Formalism).
Foregrounding theory and analysis has such a ‘cornerstone’ status for
Stylistics that it has never really become unfashionable, but again, the
modern cognitive approach brings significant advantages. For
example, by distinguishing between new linguistic instantiations of
conceptual metaphors and those which are innovative conceptually as
well as linguistically, it helps us to be more accurate about different
kinds of creativity in metaphor analysis. It also helps us to think more
accurately about concepts like affect and viewpoint (other areas which
we have always been concerned with, if to different analytical extents)
and how they work in particular texts, as well as providing significant
added impulse to abiding critical concerns like how fictional worlds
(Semino 1997, Werth 1999, Gavins 2007) and characters (Culpeper
2001) and their ‘mental spaces’ (Emmott 2002, Semino 2002, 2006,
Dancygier 2006) are created. Emmott’s work on narrative
comprehension (e.g. 1997) and Stockwell’s (2002) general
introduction to cognitive approaches have, among others, helped us
towards a better understanding of how readers interact with stories.
It is interesting to note that an important precursor to Stylistics in the
UK, Practical Criticism: A Study of Literary Judgment by I. A.
Richards (2001 [1929]) was empirical in the ‘informant-based’ sense,
in that it analysed a corpus of written reactions to poems by students
(protocols, as Richards called them).1 More modern informant-based
work has become steadily stronger and more accurately undertaken
since the seminal work of van Peer (e.g. 1986), Steen (e.g. 1994) and
others, mainly through the work of IGEL (Internationale Gesellschaft
für Empirische Literaturwissenschaft). In addition, the work of the
Glasgow group, comprised of Emmott and her close colleagues in
Psychology (see, for example, Emmott, Sanford and Dawydiak 2007)
is becoming important in helping us to understand with more accuracy
particular aspects of text-reader interaction.
The four general approaches which I have outlined near the beginning
of this section are not isolated of course. They have considerable
overlaps with one another and with what we have come to think of as
more traditional work in Stylistics. Corpus-based analysis of novels
and plays is beginning to help us to see more clearly lexical patterns
spread through texts which contribute to characterisation and thematic
8 Mick Short
development, as well as to understand better how word association

works in texts. The narratological approaches to story structure
interact in interesting ways with the fictional world construction work
being done in cognitive Stylistics, and narratological work on
viewpoint is clearly related both to the ‘deictic shift’ cognitive
approach and the ‘more traditional’ work on viewpoint done in
Stylistics over the last 25 years. Perhaps most palpable of all are the
connections between the cognitive approaches, empirical reader-
response work and traditional stylistic concerns like foregrounding.
It is clear, then, that there is lots of exciting work going on in our
field, making it vibrant, innovative and fun. We should always
remember, though, that every approach and every method, whether
new or ‘old’ has disadvantages as well as advantages, and brings
dangers as well as excitement. The corpus and empirical informant-
based approaches bring with them problems associated with
technicality. It is easy to become so bound up in the technical issues
involved in using the computer software, or in making sure that all the
variables not under examination are well-controlled in informant- or
reader-based work, that it becomes difficult to see the interpretative
wood for the analytical trees. It is thus very important that we keep
continually in mind the reasons we have embarked on these sorts of
work, as we struggle to control the methodologies and tools we are
employing. With the narratological and cognitive approaches the
dangers seem to be at the opposite end of the scale. It is easy to be
excited by the ‘naturalness’ of approaches like these, and so not notice
that the use of the relatively abstract terms and concepts involved can
cover up hidden difficulties. An obvious issue here within the
cognitive approaches is that what is meant by the term ‘cognitive’ can
vary considerably not just from one work to another, but even within
the same article. Sometimes it is used to talk about general abstract
properties proposed for human minds (as in much of cognitive
Linguistics), arrived at post factum. But when reader interaction with
texts is discussed, ‘cognitive’ often has more to do with online
processing. And writers can move, sometimes imperceptibly, from
one such meaning to another, often without realising they have done
it.2 There is a particular issue here for Stylistics in general which
parallels what I have so far characterised as an issue for the cognitive
and empirical approaches. Critical accounts of texts, including stylistic
accounts, have traditionally been post-processed accounts. The critics
and analysts have read the texts concerned many times and have
considered them very carefully before arriving at a published
commentary. Yet at the same time we ‘cooperatively pretend’ that we
are going through the text for the first time in ‘online processing’
mode (and necessarily so, in order to be able to account for
sequencing effects like surprise, for example). But logic does not
easily allow us to combine online and post-processed accounts. If an
empirical cognitive researcher uncovers what readers do when they
read a text for the first time, how should that research affect an
account of that text which is completely or predominantly post-
processed? In my view, if we are going to find ways through these
sorts of conundrums we need to continue with the traditional stylistic
analytical assumptions I referred to in section 2 above – we need to be
as analytically detailed and systematic as we can, in textual and
empirical analysis and also in our use of terms, if we are to be as exact
as we can about our findings and as sure as we can in our conclusions.
In other words, we need to make enough carefully-made distinctions
to be descriptively and interpretatively adequate.
4. Going back to ‘old’ concerns: interpretation(s), effects,

appreciation and evaluation
As well as developing the new analytical techniques I have referred to
in 3 above, I think that we also need to return to a series of related and
currently unfashionable ‘philosophical’ issues, including those
mentioned in the heading to this section, and use our detailed analytical
approach to explore the consequences that our conclusions about them
have for what we do, and how. I cannot discuss all of the above in any
detail in one article and so will restrict most of what I say below to the
notion of interpretation. But these concepts are intricately connected
with one another and so I want first to make some general remarks
about them. Then I will move on to discuss interpretation (and the
related notion of a reading) in general terms, before looking at some
particular poems in section 5 in order to flesh out some of the more
general points I make here. I use short poems for this purpose as it is
easier to account for them as interpretative wholes than novels, stories,
plays and longer poems.
Traditionally, 20th century criticism was said to be concerned mainly
with the value of literary texts, but in fact very little work on how
textual evaluation or appreciation is arrived at was carried out by the
10 Mick Short
critics. Rather, it seemed to be assumed that, if a particular text was

being discussed by some critic, that in itself was a ‘guarantee’ that it
was to be highly valued, because critics were equipped with special
sensitivity. Then, with the poststructuralist and post-modern turns, any
text, whatever its quality, became considered the equal of any other
text, and as a consequence the notion of text evaluation has almost
disappeared from critical work altogether (though note van Peer’s
forthcoming edited collection). The same is true of Stylistics. We have
concentrated mainly on how textual structure, and how we interact
with it, in our understanding of texts, and as this interpretative work
needs to be in place before the process of evaluation can be explored in
any detail, we have tended to put off a detailed consideration of
evaluation till later (though see Short and van Peer 1989 and Short and
Semino forthcoming). In logical terms, it is difficult to see how one
can value a work before one has understood it but we have done quite a
lot of analytical work on text understanding now, and so it is arguably
time to re-focus to some extent on evaluation and appreciation. We
also have a responsibility, in my view, to reconnect with work in
Aesthetics on aesthetic judgment and to help criticism more widely to
redirect its aim too.
The concept of literary appreciation, which I understand to refer to a
detailed and reasonably well worked out sense of what a text means,
how it affects us and so why it is important for us, has practically
fallen into disuse. Appreciation, which I must say that I was distinctly
puzzled by when I was a student (its meaning was never really
explained, to me at least, and this may help to explain why it is also
currently not much used), seems to be intermediate between
understanding/interpretation and evaluation. For a reasonable
appreciation to be arrived at, the reader needs, as a necessary
condition, a detailed sense of how a text works. It thus assumes, at the
very least, a repeated reading and detailed understanding of the text
concerned,3 and is likely to be improved by analytical examination. In
fact I would want to argue that one of the significant benefits of
stylistic analysis is that it increases students’ appreciation. Certainly
many of my first year English Language undergraduates who
abandoned literary study at school (usually because they felt they could
not connect with what they experienced in class) have told me that
Stylistics has reawakened their interest in reading serious literature
because they now feel that they can connect better (= appreciate?) with
the works we have analysed together in class. My students studying

English literature have not needed to have their interest in literature
reawakened, of course, but many of them have told me that doing
Stylistics has helped them to argue their position better in literature
classes and also to understand better (= appreciate?) how their felt
responses to texts come about. Stylistics thus has an important role to
play in helping our students – and us – to get more out of (=
appreciate?) texts.4
Turning to interpretation, I am conscious of the fact that, in proposing
an account of it, I am creating a hostage to fortune which others are
bound to find fault with. But I am happy to do this in the hope of
helping us all to be clearer about what we mean by it and its associated
concepts. My main aim here is to explore some general parameters,
before looking at some example texts in more detail in the next section,
in the hope that what I say will provoke discussion and debate. A
detailed understanding would seem to be a necessary prerequisite to
appreciation, and well-worked-out interpretations will aid appreciation
further, leading to better and clearer value judgments.
Interpretation, as far as I can see, is a heavily post-processed activity,
involving repeated reading and analytical precision in its best examples
and, similarly, evaluation is also post-processed. Evaluation involves,
at the very least, a careful consideration of a text based on a well-
worked out interpretation. Good understanding and personal response
to literature results from careful reading (online processing plus some
immediate post-processed consideration) and arguably leads to (i)
more careful consideration/study and/or analysis, which in turn leads to
(ii) appreciation (a relatively personal activity, involving at least some
reflection and based partly on a personal response to how a text has
affected us) and finally (iii) a fully-fledged post-processed
interpretation and (iv) a more considered evaluation (which is more
general than personal when compared with appreciation, as well as – or
perhaps as a consequence of – being more post-processed). Note that
appreciation thus also presupposes affective response, which is also
intricately related to understanding. It is difficult to see how a sensitive
response to a text can be achieved without a reasonable understanding
of it.
Now let us come to interpretative variation (including ‘readings’),
which will be the main concern of the rest of this article. I feel that we
12 Mick Short
have to get some general building blocks about the nature of

interpretation in place before we can move on as a community to
understand better the processes of appreciation and evaluation.
The first thing to notice is that in much contemporary literary criticism
the term ‘reading’ has largely replaced the term ‘interpretation’. The
reasons for this are obvious enough, I think. With ‘formalism’ under
attack in critical circles and the preference for many understandings of
the same text already in place, the notion of interpretation seemed too
formal and too ‘exclusive’ for many, particularly as the rhetorical
power of (i) personal readings and (ii) Stanley Fish-like, undefined and
studiously unexamined ‘interpretive communities’ grew. Moreover, the
deconstructive and post-modern moves to undermine the ‘common-
sense’ notion of text-understanding effectively privileged the reader
over the text.5 But although it is possible for individuals to proliferate
such readings it is difficult to believe that the notions of understanding
and communication, which we all need to get by on, can survive the
view that texts have as many meanings as they do readers and that no
reading can be ruled out.
As it happens, I do not think it is too difficult to find interpretations
which are mistaken, particularly (but by no means exclusively) in
student essays. A good candidate is an undergraduate Stylistics essay I
read a couple of years ago on Seamus Heaney’s poem ‘Digging’. The
student thought that, in the fictional world established early in the
poem, Heaney, who depicts himself looking out of the window as he
writes, could literally see his father digging in the garden:
Till his straining rump [ . . .]
[. . .] comes up twenty years away [ . . .]
Where he was digging.
But the TIME IS DISTANCE metaphor in ‘twenty years away’ and the
switch from present simple to past continuous make it difficult to see
Heaney and his father as co-present in the same space-time location.
This leads to an understanding of the lines as relating to Heaney’s
vivid memory of his father digging, and indeed to an understanding of
the whole poem, and the practice of writing poetry as portrayed in this
particular case, at least, as being about memory (as Heaney’s
characterisation of himself at the end of the poem as digging with his
pen further shows). It is difficult to see how ‘Digging’ can be
appreciated sensitively without the ‘memory’ interpretation. Mistakes

like this are, of course, more understandable on first-time reading than
in a considered interpretation (which stylistic analysis essays are
meant to include).
In arguing the unfashionable view that accounts of texts can be wrong
I should make it clear immediately that I do not believe that texts have
one and only one interpretation. Indeed, although stylisticians have
often been criticised by others for holding this belief, I have never met
one in an academic career nearly 40 years long. That said, it is not that
obvious to me that all texts have more than one interpretation or that
even the more interpretatively-wide texts have a wide range of
interpretations. Indeed, I would argue that even long and complex
texts like Shakespeare’s plays each have a relatively small set of
substantially different interpretations (perhaps even countable on the
fingers of one hand?). Part of the issue here is that we need to
understand what counts as a different interpretation and what counts
as a different instantiation of the same interpretation. The fashionable
term ‘reading’ is particularly unhelpful in trying to understand this
distinction, as it enables us to slide, too easily and without
understanding, from individual response to post-processed
interpretation and back again.
In my view, then, interpretations are non-obvious, post-processed and
well-structured hypotheses concerning particular texts which can then
be ‘tested out’ through textual analysis, consideration of prototypical
reading responses and reading contexts, and so on. The descriptor
‘non-obvious’ in the above account distinguishes textual topics (which
are obvious in terms of the textual surface) from themes (which need
more work to determine)6 and the term ‘well-structured’ distinguishes
themes and thematic oppositions from interpretations. In terms of
topic, Joyce’s well-known short story Eveline can be said to be about
an Irish woman and whether she will leave her home for a new life
with her ‘fellow’. In terms of theme, it can be said to be about the
opposition between romantic love and family duty (cf. the opposition
between Ryan’s 1991: 113-23 wish- and obligation-worlds).7 In terms
of interpretation, the story can be seen as an exploration and evocation
of how romantic love and family duty can conflict poignantly with
one another in particular circumstances. Unlike some texts, Joyce’s
story does not, I think, contain a ‘message’ about this issue – he does
14 Mick Short
not tell us whether he thinks Eveline, and others like her, should
liberate themselves from family duty or not. But the way in which the
issue is explored helps us to understand, and so be affected by, the
difficult personal choice which Eveline makes at the end of this
famous story. Note again the intrinsic link between understanding,
affective response and appreciation.
The kind of ‘message’ interpretation absent from Eveline’ is more
likely to be found in ‘didactic’ literature, for example Jane Austen’s
novelistic critique of the 18th century ‘marriage market’ in Pride and
Prejudice and other novels. Other texts make a more studious point of
not ‘giving an answer’, as it were, to the philosophical problem(s)
raised, precisely because one is not available. Good examples of such
texts are R. S. Thomas’s ‘religious struggle’ poems, where he wrestles
with imponderable religious questions.8 But I would argue that critics
and stylisticians should strive to find ‘message interpretations’ for
texts even if, later, they have to retreat to an ‘exploration of’ kind of
account. We need to push ourselves to be as exact and constrained
interpretatively as we can manage if we are not to be merely vague. I
develop this point further below.
One of the problems I have with the ‘many different interpretations’
notion is that often the differences involved between one account and
another do not seem to be enough to warrant the phrase ‘different
interpretations’. Rather, what we often appear to have are slightly
different instantiations of the same interpretation, the rough equivalent
in phonetic terms of two different phonetic realisations of the same
phoneme (e.g. aspirated vs. unaspirated /p/ in English). In this case
there is clearly a phonetic difference, and one which might be
significant in some particular context (e.g. to mark an accent), but the
difference between aspirated and unaspirated /p/ will not distinguish
one morpheme from another, as the voiced/voiceless distinction (i.e.
/p/ vs. /b/) does in the context /-it/. Another problem I have is that
some claimed interpretations do not really seem to be interpretations
at all. For example, I am not convinced that a performance of Macbeth
in modern dress constitutes a different interpretation of the play, as is
sometimes assumed. The modern dress probably does help to make
the play feel a bit more relevant to the modern condition for some
people, but to constitute a different interpretation I would argue that
we would need something more significant to contrast with other

extant interpretations.
A factor which sometimes causes misunderstandings when
interpretation is discussed is that the scope, or domain, of
interpretation can vary considerably. The interpretation of whole texts
is what is most often discussed, but it is also possible to have an
interpretation of a section of a text (e.g. a stanza in a poem), a line, a
sentence, a clause, a phrase or a word. And because human beings
appear to be set up genetically to notice difference rather than
sameness (which helps to explain why interpretative difference is
critical ‘news’) it is not at all unusual to find cases where critics
disagree over some small text part while appearing not to notice the
extent of critical agreement over the rest of the text. However, it is
important to notice that differences of interpretation over a particular
sentence or line may not affect the overall interpretation in any
significant way.
Finally, it is worth noting that those who want to promote the idea of
texts with indefinitely large numbers of readings seem to make a point
of choosing texts to discuss which are particularly difficult to tie down
interpretatively. Modern critics have a predisposition to choose
semantically opaque texts, and many modern poems are, indeed,
difficult. But we should not forget that many poems, particularly,
those earlier in the poetic tradition seem to be fairly straightforward
interpretatively, focusing instead on making the language they use
especially appropriate, or iconic, in relation to the thing described.
Semantically complex, opaque and difficult texts do have to be
catered for in a complete account of interpretation, of course, but so
do texts which are more straightforward interpretatively, of which
there are rather a lot.9
5. Interpretation: some examples and what we can learn from

them
I will now look at a number of short, simple and relatively
straightforward examples to flesh out some of the abstract points I
have been making above. I am not, of course, claiming that all texts
have rather straightforward interpretations. There are plenty of highly
opaque, ambiguous and allusive poems, but I think that the proportion
of interpretatively straightforward poems has been overlooked and
16 Mick Short
that if we are to understand the nature of interpretation we need to

begin by examining relatively straightforward cases. Given the space
available, I will not be able to provide full stylistic analyses of the
texts I examine, but where possible I refer to existing accounts.
5.1 ‘Where are you going to, my pretty maid?’ and ‘yo, Blair’:
resisting readings
One kind of interpretative activity which has become popular in recent
years, particularly in deconstructive and new historical criticism, is the
idea of the ‘resisting reading’, something which is often seen in
modern reactions to earlier texts or CDA-style accounts of (usually)
non-literary texts which the analyst wants to critique for some socio-
political, reason.10 Clearly it is possible for readers to come up with
such critical readings, but to my mind such critiques are often unfair,
in literary-critical terms at least, particularly in relation to earlier
periods. None of us can step outside all the assumptions of our culture
and it is difficult for a writer to ‘critique’ many different, often
deeply-held, cultural assumptions at the same time. There is also an
issue concerning whether such a resisting reading can sensibly count
as an interpretation (as opposed to a critique) of the text concerned,
precisely because it does not usually relate in a receptively sensitive
way (cf. appreciation) to the text under consideration.
I will illustrate this point by looking at the traditional nursery rhyme
which the title of my paper playfully alludes to:
‘Where are you going to, my pretty maid?’
‘I'm going a-milking, sir,’ she said.
‘May I go with you, my pretty maid?’

‘You're kindly welcome, sir,’ she said.
‘Say, will you marry me, my pretty maid?’

‘Yes, if you please, kind sir,’ she said.
‘What is your father, my pretty maid?’

‘My father's a farmer, sir,’ she said.
‘What is your fortune, my pretty maid?’

‘My face is my fortune, sir,’ she said.
‘Then I can't marry you, my pretty maid.’

‘Nobody asked you sir,’ she said.
The above nursery rhyme makes it clear, I think, that the male speaker
has an unreasonable, shallow and sexist attitude towards the young
woman. He apparently proposes to her on their first encounter (the
fact that he addresses her as ‘my pretty maid’ rather than by her name
and she replies with ‘Sir’ suggests they do not know one another)
merely because she is physically attractive. So his proposal is
apparently motivated solely by her beauty but this beauty cannot
prevent him from later withdrawing his ‘offer’11 when he discovers
that she is poor and will bring no dowry to the marriage. The ‘socially
superior’ man is thus portrayed as a traditional male cad. So far, then,
the rhyme is doing reasonably well in 20th century, politically-correct
terms.12 But such an ideological standpoint might lead us in turn to be
critical of the young woman, who, after all, does not herself directly
challenge the sexist and classist assumptions made by the man. She
merely implicates (by flouting the Gricean quantity maxim) that she
did not ask him to propose to her, and, more weakly, that she does not
want to marry him and did not encourage his attentions in the first
place. We have no evidence to help us in determining the motivations
for her implicated lack of interest in the man, but it is clear that she
does not challenge the assumption that female beauty should be a
relevant criterion when long-term relations between men and women
are considered. However, even if one does hold the view that physical
beauty should not be an important criterion for establishing sexual
relationships, it is not at all clear (i) that the associated critique is
reasonable with respect to a text created in a distant era, when
different assumptions may well have been ‘naturalised’ or (ii) that
such a critical stance, if held, counts as a reasonable interpretation of
the nursery rhyme, precisely because the critique is culturally
anachronistic.
Now let us put a more modern, non-literary extract alongside the
above nursery rhyme. It is a (mis)transcription and associated
commentary on what quickly became a celebrated conversational turn
18 Mick Short
addressed by US President George Bush to UK Prime Minister Tony

Blair at the G8 summit on 17 July 2006, as reported in The
Independent the next day:
Bush: Yo, Blair. How are you doing?
Does he regard Mr Blair as an
equal? What about ‘Yo, Tony?’?
(The Independent, 18 July 2006, p. 1)
I am not interested here in whether George Bush said ‘Yo’ or ‘Yeah’,

the circumstances of the recording (was it recorded without the
knowledge of the participants or not?) or the exact turn-taking context
in which the remark was made,13 but merely the use of the last-name-
only formulation and the boxed Independent commentary. In spite of
the hedging interrogative form, it seems that the reporter-commentator
was assuming that George Bush was evincing a ‘socially superior’
attitude to Tony Blair as a consequence of the last-name-only direct
address term. But, although last-name-only can suggest ‘high to low’
speech for many, there are plenty of social contexts, especially in
male-male conversations, where this ‘rule’ does not hold, including
contexts in which the two men may well have habitually taken part,
and in any case there is plenty of other evidence to suggest that they
got on rather well with one another in personal terms. The
Independent commentary thus looks rather like someone with a pre-
established view imposing a reading consistent with that view on the
data, rather than being careful analytically in order to arrive at a
carefully judged account.14 Hence, if my argument holds water, we
would need to reject the Independent’s commentary on what George
Bush said as an unreasonable interpretation of the text, just as we
would have to reject (i) the above hypothetical critique of the nursery
rhyme and (ii) my student’s misunderstanding of the Heaney poem
discussed in section 4.
5.2 ‘Meeting at night’ and ‘He wishes for the cloths of heaven’:
one interpretation or more than one?
Robert Browning’s poem below seems to me to be a good example of
a text of which it is difficult to have many interpretations. Indeed, I
cannot provide more than one, and I know of no significantly different
competing interpretations by other critics, even though it is possible to
have slightly different understandings of particular parts of the text
and possibly different readings of it:
Meeting at Night
The grey sea and the long black land;
And the yellow half-moon large and low;
And the startled little waves that leap
In fiery ringlets from their sleep,
As I gain the cove with pushing prow,
And quench its speed i’ the slushy sand.
Then a mile of warm sea-scented beach;
Three fields to cross till a farm appears;
A tap at the pane, the quick sharp scratch
And blue spurt of a lighted match,
And a voice less loud, thro’ its joys and fears,
Than the two hearts beating each to each!
The account of this poem below is based on my description of it at
<http://www.lancs.ac.uk/fass/projects/stylistics/topic5a/meetingtaskb.
htm>. ‘Meeting at night’ is a description of a meeting between two
lovers. In the first stanza, the man is described in a boat on the sea,
arriving at his apparently secret destination, a small bay (not a
harbour, note, as this would not have the sandy beach referred to in
line 6 and would also not ‘align’ properly with the atmosphere of
secrecy). Given the period when the poem was written, we assume
that the man has probably rowed, or perhaps sailed, secretly to the
bay, where the boat is beached on the sand. Note how our use of
appropriate schematic assumptions, and the need to impose
consistency of understanding, restricts interpretation, (i) ruling out
conceivably possible interpretations of parts of the poem (e.g. that the
‘I’ was in a motor-boat) and (ii) making other differences (e.g.
whether he rowed or sailed the boat) interpretatively irrelevant.
The second stanza describes the man hastening for a mile along the
beach (so he has not landed as near as perhaps he could have to his
final destination, again suggesting the need for extreme secrecy). He
leaves the beach, crosses three fields and arrives at the farm where his
love is waiting secretly for him. He does not knock on the door, but
taps quietly at the window pane, suggesting that there are others in the
house or in a nearby building (maybe the woman is waiting in an
outhouse?), who are asleep and are not to know about the meeting.
Hence the action appears to take place in the middle of the night, not a
20 Mick Short
winter evening, for example. The woman on the other side of the
window lights a match and whispers to her lover. Then she opens the
door and they embrace. We are not told this, but infer it from
schematic assumptions and the fact that at the end of the poem the two
hearts of the lovers are apparently close together, beating as one.
Similarly, we assume that they will spend some time together (they
have both expended too much effort for the reward of a quick kiss to
seem adequate).
The identity of the two lovers is never revealed in the poem. So the
assumption that the person in the boat is male, and the person in the
farmhouse female, is based on schematic assumptions. Even in these
liberated days we would probably assume that the boat is rowed by the
man and that he travels towards the woman for the secret tryst, rather
than the other way round. The poem could conceivably describe a
woman in the boat meeting a man in the building, or two lovers of the
same sex. But given schematic assumptions and the fact that there is
no textual indication that they should be overturned, the male-female
assumption predominates (in any case, this poem was originally
published along with a partner poem ‘Parting at Morning’, under the
joint title ‘Night and Morning’ and this partner poem makes it clear
through pronominal reference that the ‘man in a boat’ assumption is
correct). But even if we insisted on a non-schematic reading, it is not
that obvious that the change would be big enough to warrant the claim
that a different interpretation had been produced. The poem would
still be an evocation of two lovers meeting illicitly in secret.
As it happens, there are two published accounts of this poem, by the
critic F. R. Leavis (1975: 120-2) and the stylisticians Ron Carter and
Walter Nash (1990: 123-9), which I had not read before first I wrote
my web-based description referred to above. Interestingly, all three
accounts are similar with respect to their understanding and positive
judgment of the poem (though there are some minor differences and
Carter and Nash’s description is more precise analytically than
Leavis’s). The fact that the poem is straightforward interpretatively
leads to the issue of why the poem has the status which it does. This, I
think, is because, it captures what it describes (a secret meeting
between two lovers, something we can all easily relate to) in a highly
evocative way, through the use of phonetically and grammatically
mimetic language. In other words, it is an example of what Leavis
called ‘enactment’, where the form of the poem symbolically enacts

what it describes in its fictional world. Evidence for this is the high
density of onomatopoeic words and phrases. Note, for example, the
way in which the narrator-character moves to ‘quench’ the boat’s
speed ‘i’ the slushy sand’ and the onomatopoeic description of his
moment of arrival:
A tap at the pane, the quick sharp scratch
And blue spurt of a lighted match
The grammatical use of elliptical constructions (NPs as clauses)
linked to rapid topic change suggests rapidly changing object
perceptions, and so rapid onward movement. Quite a lot of
interpretatively simple poetry uses enactment in detailed and complex
ways to increase interest and induce emotional responses in readers.
One conceivable way of arriving at a different account of the poem
would be to give it a Freudian reading. It is possible to see the prow
going into the sand as a symbolic prefiguring of the sexual love which
could well have taken place after the lovers meet (they stayed together
till daylight, according to ‘Parting at Morning’). But although this
adds to the felt significance of a poem which does not explicitly say
anything about sexual congress (something which would have been
unacceptable at the time the poem was written), it is not that obvious
that the interpretation changes. Rather, the poem becomes richer, and
so is likely to be valued more highly.
There may be another reasonable interpretation of the poem that I
have not thought of but, as far as I can see, the sorts of differences I
have brought to the fore in my discussion are differences of detail, not
overall interpretation. The following poem by W. B. Yeats, on the
other hand, really does seem to have two possible interpretations:
He Wishes for the Cloths of Heaven
Had I the heavens’ embroidered cloths,
Enwrought with golden and silver light,
The blue and the dim and the dark cloths
Of night and light and the half-light,
I would spread the cloths under your feet:
But I, being poor, have only my dreams;
I have spread my dreams under your feet;
Tread softly because you tread on my dreams.
22 Mick Short
For some years I thought that there was only one way of
understanding this poem, namely that it was a moving statement from
the viewpoint of a lover towards his loved one (biographically Yeats
himself and Maud Gonne, of course), saying that if he had wonderful
things he would give them to his love, but that all he had was his
dreams. The interest of the poem comes mainly the way in which the
simple, repeated and straightforward dreams are contrasted with the
complex, almost ineffable, hypothetical alternative through another,
arguably less straightforward than in Browning’s poem, process of
enactment. The description of the heavens’ embroidered cloths takes
up almost all of the first four lines of the poem and contains a
particularly dense set of interwoven semantically deviant relations,
while ‘the dreams’ receive just the two-word NP. The cloths, while
appearing to be ‘woven’ with threads of golden and silver light are at
the same time described in lines 3 and 4 as being ‘blue’ and ‘dim’ and
‘dark’. Processing is made difficult because these three NP
premodifiers are each relatable semantically to one of the three
postmodifying phrases in the next line, but the two sets of modifiers
are arranged (note also the parallel polysyndetic constructions in each
line) so that none of the postmodifiers in line 4 line up vertically under
the premodifier it most appropriately relates to in line 3. The heavens’
embroidered cloths thus become important and mystical because they
comprise a complex set of ‘impossible’ inter-related features, and the
contrast in value with the lover’s dreams makes those dreams even
more important because dreams are all he has (cf. the Biblical parable
of the widow’s mite).
This interpretation of Yeats’ poem was undermined for me some years
ago by a literature colleague, Richard Dutton, who pointed out
something that had not occurred to me, namely that it could be seen,
like so many poems where a lover addresses a loved one, as being a
poem about writing poetry. In this view, the lover is the poet and is
trying to describe his loved one, but the loved one’s beauty, character,
or whatever, is so complex and ineffable that the poet’s attempt to
capture it is doomed to failure.
This undoubtedly counts as a different possible interpretation. That
said, I have some doubts about its reasonableness. Firstly, although it
is arguable that the ‘writing poetry’ interpretation seems to ‘cover the
text’ as well as the more straightforward one (including the points I
have made above concerning the mystical impressiveness of the

‘heavens’ embroidered cloths), there is no specific part of the text
which helps to trigger that less straightforward interpretation and act
as an argument to support it. In this sense, it looks rather more like a
reading imposed from outside than one generated by the text.
Secondly, I think there is a problem if, in effect, the majority of love
poetry also has to be seen as poetry about writing, as the interpretative
move becomes somewhat reductionist, and so less interesting. Hence,
although I would not want to rule out absolutely a ‘writing poetry’
reading for the poem I think it is less convincing than the more
straightforward one.
The issue here, then, is that even if there is more than one conceivable
interpretation of the same text it may still be the case that one (or more
than one) of those interpretations may be preferable to other
conceivable interpretations. This situation would seem to parallel what
we see in pragmatic understanding more generally. We are usually
involved in inferring the most likely understanding(s) of some
utterance in context, rather than deciding that there is one and one
only possible understanding. Because poems are more
decontextualised than many texts we read, it is, of course, easier for
someone to bring along a different set of contextual conditions for
responding to some poem, and it may well be that better
interpretations can also be generated in this way. We should always be
open to new interpretative possibilities. But it is worth remembering
that, just as the scientists are continually striving to come up with new
ways (hypotheses) of explaining old data, we should be prepared to
test carefully whether those new explanations (i.e. interpretative
hypotheses) (i) hold water in their own terms and when the data (in
this case the text) is measured carefully against them, and (ii) are
definitely new accounts and not merely reformulations of existing
interpretations.
5.3 ‘We dance round in a ring and suppose’: Appropriate levels of
abstraction for interpretations
I have argued above that (i) we should be sure that alternative
‘interpretations’ really are interpretations and not something else (e.g.
‘mere readings’ or interpretative variants) and (ii) that even if there is
more than one conceivable interpretation, it may still be the case that
one (or more than one) is nonetheless preferable. However, this
24 Mick Short
discussion has assumed, unargued, that the interpretations are

involved ‘compete properly with one another’ in the sense that they
operate at the same level of interpretative abstraction. But a
consideration of the final poem which I will examine in this article, a
very short poem by Robert Frost, shows us that we also need to
choose an appropriate level of abstraction for our interpretations and
that two competing interpretations will need to be at the same level of
abstraction if they are to be sensibly compared. I will not analyse this
poem in any detail stylistically as I have already discussed it at some
length in Short (2002) and (2007).
The Secret Sits

We dance round in a ring and suppose,
But the Secret sits in the middle and knows.
It is often the case when people read this poem that they first construe
a rather specific context to relate the poem to. I had just finished
reading a John Le Carré novel when I first read this poem, and so I
was ‘primed’ psychologically to ‘see’ it as about situations like
spying, where one set of people know a secret and others only think
they know it, or are unaware of it. This felt wish to make my
understanding as specific as I could is, I think, symptomatic of a need
we all feel. Certainly I often want to say of modern critical accounts of
texts that they are over-abstract, and so do not tell me enough about
the text they are describing to convince me of their usefulness. I
would argue that we need to strive to make our interpretations as
specific as we reasonably can, and not ‘fly up’ to more abstract levels
unless there is some reasonable warranty for doing so.
In the case of ‘The Secret Sits’, however, I quickly saw that there
were other relevant contexts which could be invoked just as
reasonably to account for the abstract proximal vs. distal deixis of the
poem: for example those who govern a country vs. those who are
governed. And when I first used the poem in class I discovered that
my students could see even more applicable and equivalent contexts,
for example the religious elect vs. the rest and (even more strikingly
for me in class) teacher vs. taught. The reason for the contextual
multi-applicability of this particular poem is obvious enough, I think.
The individuals referred to are not specified or related to any specific
context in the poem, and so as long as the contrasting deictic relations
(centre vs. periphery) can be maintained in some context which

involves people who know something in contrast with others who
don’t, that context can be ‘laid on’ the poem.
But once we have noticed that more than one such specific reading of
a text is equally possible, I think we have a warranty for going to a
higher, more abstract and inclusive, level of interpretation. I quickly
found with ‘The Secret’ that I wanted to describe it in such more
general terms, as encapsulating in an interesting way a commonly felt
human experience whereby ‘others’ are in control or know something
that ‘we’ do not. The spying, government, religious or educational
contexts (and any others that can be thought of) then become specific
instantiations of this more abstract interpretation. In this case, then, a
‘religious’ interpretation, say, does not properly compete with the
more abstract one which subsumes it. Any competing interpretation
would need to be construed at the same level of abstraction. It might
then be asked whether the two possible interpretations of ‘He wishes
for the cloths of heaven’ which I discussed in 5.2 might be resolvable
in this way. But as far as I can see, although one of the interpretations
of that poem is ‘closer to the text’ than the other (and so more
convincing for me), they are at equivalent levels of abstraction.
6. Concluding remarks
In this article I have argued that Stylistics is in a healthy state because
of its attention to analytical and argumentative detail, that we have a
number of new approaches to take our work forward, and that we will
develop these new approaches best by continuing with the detailed,
empirical and self-critical attitude which has become a hallmark of our
work. I have suggested that we also need to concentrate on the
concepts of interpretation and effect, appreciation and the valuing of
texts, and that we need to submit these notions to a similar detailed
and carefully critical examination. I have made some suggestions
about the relationships among these concepts and in particular about
what constitutes an interpretation, and a different interpretation, of the
same text. I give these suggestions, and the arguments and analyses I
have provided to support them, to help make it clear why I have
reached the views I have, in the certain knowledge that others will
find fault with them and in the hope that they will be able to build on
what I say, correct my misconceptions through careful and detailed
analytical discussion, and so help to move us all nearer the goal of
26 Mick Short
understanding better how we understand, interpret, appreciate and

value the texts we prize, and so want to analyse.
Endnotes
1
Richards (1928) was an early entrant to the cognitive field too, of course.
2
I have similar problems with, for example, exactly what ‘mental spaces’ and
‘blends’ are as I move from one description to another.
3
Note, therefore, that appreciation is not the same as the immediate personal response
or personal enjoyment which is so often referred to when literature teaching is
discussed. If I am right about this, accurate and appropriate personal response, which
in itself presupposes understanding, is a precursor to textual appreciation rather than a
substitute for it.
4
My bracketed queries in this paragraph reflect the fact that, in addition to being
closely interconnected with one another, the concepts I refer to in this section have
different meanings for other speakers in different contexts. In using ‘interpretation’
‘appreciation’ and ‘evaluation’ in the considered, post-processed senses I have
outlined, I am aware that others often use them to relate to first-time and immediate
reading responses. I am using ‘understanding’ as the immediate, only just post-
processed and less well worked out equivalent of ‘interpretation’; and ‘personal
response’ as the immediate, only just post-processed and less well worked out
equivalents of ‘appreciation’ and ‘evaluation’. I accept that sometimes the distinctions
I am making are difficult to keep apart, but feel that it is important, for the sake of
clarity, for us to try to make these sorts of distinctions as overtly as we can. Otherwise
we are in danger of sliding unknowingly from one concept to another, in the manner I
often despairingly think of as ‘elegant equivocation’. In my view Stylistics should
mainly be concerned with post-processed and considered analyses, interpretations,
characterisations of effect and evaluation, and as a consequence I also worry that so
much empirical work on literary response (including much of my own to date) has
related to first-time reading responses, as it is not clear how well such work bears on
considered, post-processed discussions of texts, especially in relation to the
characterisation of particular texts rather than to processing in general.
5
In relation to this issue, it is instructive to read David Hoover’s (2005) discussion of
Jerome McGann’s post-modern reading of ‘The Snowman’ by Wallace Stevens in
McGann’s Radiant Textuality (2001).
6
Elena Semino’s MA thesis (1990) made an interesting distinction between ‘topic
titles’ and ‘theme titles’ in poems which is relevant here.
7
Critical characterisations of themes often, but not always, involve an oppositional
contrast between a pair of theme words and I suspect that such oppositions are the
beginning of the process which takes us analytically from theme to interpretation.
Louwerse and van Peer (2002) is an interdisciplinary collection devoted to thematics.
8
Jean Boase-Beier made this point clear to me (and others) at a paper she gave to the
Pragmatics and Stylistics Research Group in Lancaster in March 2007, for which I
thank her.
9
It is an empirical question whether simpler or more difficult poems are in the
ascendant numerically, of course, and how difficulty and simplicity are related to
evaluation.
10
There are interesting parallels between deconstructive criticism and critical
discourse analysis in this, among other, respects.
11
In speech act terms, ‘offer’ seems more appropriate then ‘request’ here in terms of
felicity conditions, given his socially and sexually superior attitude.
12
Whether, as a consequence, the rhyme can be said to be attitudinally ‘before its
time’ is not so clear, however. We often conveniently overlook the ‘modern’ views
that people living in earlier times had.
13
There have, of course, been many debates over this conversational occurrence,
many of which are available via Google. The Wikipedia discusion can be found at
<http://en.wikipedia.org/wiki/Yo,_Blair#The_.22Yo_Blair.22_text>
14
This seems to be equivalent to the kind of thing that Henry Widdowson (1995,
1996) accuses CDA of in his debate in Language and Literature with Norman
Fairclough.
References
Burton, Deirdre. 1980. Dialogue and Discourse. London: Routledge and Kegan Paul.
Brémond, Claude. 1973. Logique du Récit. Paris: du Seuil.
Brumfit, Christopher J., and Ronald A. Carter (eds). 1986. Literature and Language
Teaching. Oxford: Oxford University Press.
Carter, Ronald A. and John McRae. 1996. Language, Literature and the Learner:
Creative Classroom Practice. London: Longman.
__, and Walter Nash. 1990. Seeing through Language. Oxford: Blackwell.
Cook, Guy. 1992. Discourse of Advertising. London: Routledge.
Culpeper, Jonathan. 2001. Language and Characterisation: People in Plays and
Other Texts. London: Longman.
__, Mick Short and Peter Verdonk. 1998. Exploring the Language of Drama: From
Text to Context. London: Routledge.
Dancygier, Barbara. 2007. ‘Narrative Anchors and the Processes of Story
Construction: The Case of Margaret Atwood’s The Blind Assassin’. Style 42(1):
133-52.
Doležel, Lubomír, and Richard W. Bailey. (eds). 1969. Statistics and Style. New
York: American Elsevier.
Emmott, Catherine. 1997. Narrative Comprehension: A Discourse Perspective.
Oxford: Oxford University Press.
__. 2002. ‘“Split Selves” in Fiction and in Medical “Life Stories”: Cognitive
Linguistic Theory and Narrative Practice’ in Semino and Culpeper 2002: 153-82.
__, Anthony J. Sanford and Eugene J. Dawydiak. 2007. ‘Stylistics Meets Cognitive
Science: Studying Style in Fiction and Readers’ Attention from an Inter-
disciplinary Perspective’ in Style 41(2): 204-24.
Fowler, Roger (ed.). 1966. Essays on Style and Language: Linguistic and Critical
Approaches to Literary Style. London: Routledge and Kegan Paul.
__ (ed.). 1971. Style and Structure in Literature: Essays in the New Stylistics.
London: Routledge and Kegan Paul.
28 Mick Short
__. 1977. Linguistics and the Novel. London: Methuen.

__. [1986] 1996. Linguistic Criticism. Oxford: Oxford University Press.
Freeman, Donald C. (ed.). 1970. Linguistics and Literary Style. New York: Holt,
Rinehart and Winston.
Gavins, Joanne. 2007. Text World Theory: An Introduction. Edinburgh: Edinburgh
University Press.
Hardy, Donald E. 2003. Narrating Knowledge in Flannery O’Connor’s Fiction.
Columbia: University of South Carolina Press.
Herman, David. 2002. Story Logic: Problems and Possibilities of Narrative. Lincoln
and London: University of Nebraska Press.
__ (ed.). 2003. Narrative Theory and the Cognitive Sciences. Stanford, California:
Center for the Study of Language and Information.
Herman, Vimala. 1995. Dramatic Discourse: Dialogue as Interaction in Plays.
London: Routledge.
Hoover, David. 1999. Language and Style in The Inheritors. Lanham, Md: University
Press of America.
__. 2005. ‘Hot-air Textuality: Literature after Jerome McGann’ in TEXT Technology:
The Journal of Computer Text Processing 14(2): 71-103.
Jakobson, Roman. 1960. ‘Closing Statement: Linguistics and Poetics’ in Sebeok,
Thomas A. (ed.) Style in Language. Cambridge, Mass.: MIT Press: 350-77.
Leavis, F. R. 1975. The Living Principle: ‘English’ as a Discipline of Thought.
London: Chatto & Windus.
Leech, Geoffrey N. 1969. A Linguistic Guide to English Poetry. London: Longman.
__, and Mick Short. [1981] 2007. Style in Fiction. London: Longman.
Louwerse, Max and Willie van Peer (eds). 2002. Thematics: Interdisciplinary Studies.
Amsterdam: John Benjamins.
McGann, Jerome J. 2001. Radiant Textuality: Literature after the World Wide Web.
New York: Palgrave.
McIntyre, Dan. 2006. Point of View in Plays: A Cognitive Stylistic Approach to
Viewpoint in Drama and other Text Types. Amsterdam: John Benjamins.
Nash, Walter. 1982. ‘On a Passage from Lawrence’s “Odour of Chrysanthemums”’ in
Carter, Ronald A, (ed.) Language and Literature. London: Allen & Unwin: 201-
20.
__. 1990. Language in Popular Fiction. London: Routledge.
van Peer, Willie. 1986. Stylistics and Psychology: Investigations of Foregrounding.
London: Croom Helm.
__ (ed.). Forthcoming. The Quality of Literature: Studies in Literary Evaluation.
Richards, Ivor Armstrong. 1928. Principles of Literary Criticism. London: Routledge.
__. [1929] 2001. Practical Criticism: A Study of Literary Judgment. London:
Routledge.
Ryan, Marie-Laure. 1991 Possible Worlds, Artificial Intelligence, and Narrative
Theory. Bloomington, Indiana: Indiana University Press.
Semino, Elena. 1990. Towards a Theory of Poetic Titles. London: Unpublished MA
dissertation.
__. 1997. Language and World Creation in Poems and Other Texts. London:
Longman.
__. 2002. ‘A Cognitive Stylistic Approach to Mind Style in Narrative Fiction’ in

Semino and Culpeper 2002: 95-122.
__. 2006. ‘Blending and Characters’ Mental Functioning in Virginia Woolf’s “Lappin
and Lapinova”’ in Language and Literature 15(1): 55-72.
__, and Jonathan Culpeper. 2002. Cognitive Stylistics: Language and Cognition in
Text Analysis. Amsterdam and Philadelphia: John Benjamins.
__, and Mick Short. 2004. Corpus Stylistics: Speech, Writing and Thought
Presentation in a Corpus of English Writing. London: Routledge.
Short, Mick (ed.). 1989. Reading, Analysing and Teaching Literature. London:
Longman.
__. 1996. Exploring the Language of Poems, Plays and Prose. London: Longman.
__. 2002. ‘Who Is Stylistics and What Use is She to Students of English Language
and Literature?’ in Poetica 58: 33-54.
__. 2007. ‘Designing and Piloting a World-wide-web-based Stylistics Course’ in
Andrea Gerbig and Anja Müller-Wood (eds.) Rethinking English: Reconciling
Literature, Linguistics and Cultural Studies. Lampeter: Edwin Mellen: 131-66.
__, Beatrix Busse and Patricia Plummer (eds). 2006. Language and Literature special
issue: The Language and Style Pedagogical Investigations, 15.3.
__, and Willie van Peer. 1989. ‘Accident: Stylisticians Evaluate’ in Short, Mick (ed.)
Reading, Analysing and Teaching Literature. London: Longman: 22-71.
__, and Elena Semino. Forthcoming. ‘Stylistics and Literary Evaluation’ in van Peer,
Willie (ed.).
Simpson, Paul. 1997. Language through Literature: An Introduction. London:
Routledge.
__. 2003. Stylistics: A Resource Book for Students. London: Routledge.
__. 2004. On the Discourse of Satire: Towards a Stylistic Model of Satirical Humour.
Amsterdam and Philadelphia: John Benjamins.
Steen, Gerard. 1994. Understanding Metaphor in Literature: An Empirical Approach.
London: Longman.
Stockwell, Peter. 2002. Cognitive Poetics: An Introduction. London: Routledge.
Todorov, Tzvetan. 1969. Grammaire du Décaméron. The Hague: Mouton.
Toolan, Michael. 1990. The Stylistics of Fiction: A Literary-linguistic Approach.
London and New York: Arnold.
__. 1997. Language in Literature: An Introduction to Stylistics. London: Routledge.
__. [1988] 2001. Narrative: A Critical Linguistic Introduction. London: Routledge.
Watson, Greg and Sonia Zyngier (eds). 2007. Literature and Stylistics for Language
Learners. London: Palgrave MacMillan.
Werth, Paul. 1999. Text Worlds: Representing Conceptual Space in Discourse.
London: Longman.
Widdowson, Henry G. 1975. Stylistics and the Teaching of Literature. London:
Longman
__. 1992. Practical Stylistics: An Approach to Poetry. Oxford: Oxford University
Press.
__. 1995. ‘Discourse Analysis: A Critical View’ in Language and Literature 4(3):
157-72.
__. 1996. ‘Reply to Fairclough: Discourse and Interpretation: Conjectures and
Refutations’ in Language and Literature 5(1): 57-69.
A Grammarian's Funeral: On Browning, Post-
Structuralism, and the State of Stylistics
Geoff Hall
Swansea University, UK
Abstract
This paper uses the example of Browning's poem ‘A Grammarian's Funeral’ to both
argue and illustrate the need for a poststructuralist stylistics, sensitive to local and
specific contingencies of meaning and form and the ongoing evolution of language in
use. Such a stylistics examines texts for systematic and salient linguistic and related
features even as it is recognised that a relatively fuller understanding of what and how
those features mean and to who necessarily exceeds a bare and frozen textual account.
The view taken of grammar, form and language use here is ‘emergent’'; the view of
reality is perspectival. The value of literary text lies in its very challenges to simplistic
or idealised views of language use, meaning and communication. The challenge is for
analysis to move beyond isolated and incomplete accounts of individual texts to larger
cultural and historical understandings. Methodologically, this means better
rapprochement between an enquiring stylistics and intelligent literary study, alert to
issues of meaning and interpretation even while recognising their inevitable final
disappointment. The uncertainties and doubts of current stylistics, far from being a
weakness, in fact point the way to a more demanding but more rewarding future
enterprise.
Keywords: poststructuralist stylistics; discourse stylistics; emergent grammar;
Browning’s style; dramatic monologue; stylistics and interpretation; ‘A Grammarian's
Funeral’
‘Ich fürchte, wir werden Gott nicht los, weil wir noch an die
Grammatik glauben ...’ (Götzen-Dämmerung, ‘Die “Vernunft”
in der Philosophie’, 5). ‘I am afraid we are not getting rid of
God because we still believe in grammar ...’ (Twilight of the
Idols, ‘Reason in philosophy’, 1998. p. 19).
1. Introduction. On the state of stylistics.

The 2006 PALA conference in Joensuu was prompted to consider the
State of Stylistics. The stylistician's natural first metalinguistic
response must be to carefully examine that seductively alliterated SoS
in what it implies and connotes for users of English. Does a state, for
example, imply stasis, something fixed and finished? If so, this
intervention will aim to contest the notion that stylistics has achieved
32 Geoff Hall
or should achieve some state of ideal equipoise. The dangerously

ambiguous ‘lying in state’ of Browning’s grammarian, however we
may respect it, signifies ‘not true’ as well as ‘inactive’, indeed ‘dead’.
A more probable interpretation may be prompted by Victorians of
Browning’s generation (Carlyle, Gaskell) who first developed the
‘condition of England’ genre of writings, ‘the state of the state’, so to
speak, of an England perceived as in crisis. State, then, often signifies
a condition which must be righted – a ‘state of affairs’, ‘the state
we’re in’, the state of a teenager’s room, ‘in a sorry state’, ‘a sad state
of affairs’, a ‘state of health’, of war, of emergency, and the list goes
on. Wherever we look for evidence from usage, a ‘state’ seems to
connect to undesirable feelings and emotions. Perhaps stylisticians are
uneasy about their discipline. What is to be done? Hierarchical
authorities like the nation state are now passé in the age of globalism
and postmodernism. (The ‘United States’?) Something of a colonial
inheritance may even attach to the state of stylistics as I have known
it. If, then, as Foucault (1974) suggests, a statement is central to a
discursive system or ‘style’ (1974: 33) by which areas like ‘grammar’
control what can and cannot be said and thought and written in a given
‘discipline’, then, with Foucault’s precursor Nietzsche, we should get
out our Zarathustrian hammers. The ‘style of stylistics’ is also at stake
then. I understand ‘stylistics’ to be a linguistically informed and
systematic close reading of texts (‘grammar’, roughly speaking), but
in a poststructuralist world in which ‘God’ – the guarantor of fixed
meanings, hierarchies, ‘order’ rather than chaos – must be understood
to be dead. But I pass on with these reflections. It would not do, would
it, to get into too much of a state over a few words – unless, as
stylistics suggests, words are exactly what matter. The relevance of
Browning’s poem to all this is that it will help us reflect on the need
for a stylistics adequate (in the case of modern literature in particular)
to its object of attention: a world in which our understandings of
grammar, God and writing have changed in ways recognisable even if
they would not have been fully predictable, to prescient Victorians
like Browning or Nietzsche.
2. On Browning’s style.
Browning the historical-biographical author was clearly committed to
God and to grammar, but as will be shown, his writing constantly
betrays these ideals. Browning, even before Nietzsche, was one of the
A Grammarian’s Funeral 33
first great perspectivist writers. (Nietzsche famously argued that

epistemologically perspectives are all we can ever have.) The Ring
and the Book (1868-9), for example, is an epic of versions of the
‘same’ story from different points of view. The dramatic monologue,
by general consensus Browning’s most assuredly successful form,
exists to open up our understanding to new perspectives beyond those
of the speaker if not also beyond that of the poet too at its best as we
shall see with ‘A Grammarian’s Funeral’. He writes in his ‘Essay on
Shelley’ of ‘the poor limits of our humanity’ (quoted in Bristow 1991:
26), which his poetry will attempt to transcend. ‘Whose words exactly
are these (and therefore what are they worth?)’ the dramatic
monologue constantly prompts an increasingly doubtful reader to ask.
‘The dramatic monologuist is aware of the relativity, the arbitrariness
of any single life or way of looking at the world’, according to Miller
(1963), whose early study is still widely respected as a founding
document for Browning scholars (DG 1963: 108, quoted in Bristow,
67). Browning wrote in and was very much aware of the age of the
elaboration of hermeneutics (Renan, Strauss) throwing doubt on the
narratives of history and the fundamental texts of Christianity. Renan
and Strauss were themselves founding fathers of modern Reception
theory and so another contributory stream to today's poststructuralist
understandings of writing and reading. In the same way, it has been
argued that the dramatic monologue was an appropriate genre for an
age coming to doubt the certainties and value of self and identity of
the canonical Romantic poets (Byron 2003). Browning, Tennyson and
other Victorians were to present their readers with a series of plausible
murderers, lunatics and criminally inclined speakers, in some ways
prefiguring, say, Ian McEwan in our own age. The value of the lives
of these speakers is highly problematic, even as we come to some kind
of understanding of who and how these alien life forms came to be.
The contemporary reader, however, values or at least accepts the
complexity and perceived irony that Browning’s first readers
deprecated or ‘missed’.
I have found non-specialists typically responding to the prompt
‘Browning’ with the title ‘My Last Duchess’. ‘My Last Duchess’ then
can be referred to here as a representative dramatic monologue (with a
homicidal speaker) which sticks in the mind for its defiance of any
certain final interpretation. Tucker (1980), the most subtle and
illuminating of Browning’s critics, argues that the Duke does not fully
34 Geoff Hall
understand himself and his own experience, hence the ongoing

compulsion to draw the curtain and tell the story one more time. Is he
an arrogant fool or a cunning politician? Tucker argues that his
repetitions and uncertainties suggest he is no more sure of this than the
reader can be. He wants ‘fixed relations between form and meaning’
(1980: 179) and reminds Tucker of the inevitably doomed student – a
stylistics student perhaps? – ‘impatient of uncertainties that would fix
the meaning of a text beyond doubt’ (182; quoted in Bristow, 57).
Tucker is himself the nearest to a definitive interpreter of Browning
we are likely to get in our time. He demonstrates convincingly
Browning’s defining inability to ‘say the last word’ to come to a clear
conclusion in any of his writings. ‘Here were the end, had anything an
end’, Browning tries in The Ring and the Book (XII.1.; quoted Tucker,
14) – and then of course the poem continues for another full Book!
‘The Last Word’ of the collection Men and Women rapidly transmutes
into what should be more accurately titled ‘the last few inconclusive
pages’. ‘And yet’ is a recurrent expression in his writing generally.
Browning’s writing constantly recalls those endlessly extended
farewells on the doorstep, in the corridor or on the phone: ‘one more
thing’. Interesting for our purposes here, Tucker concludes that ‘the
way to meaning [in Sordello, specifically] is through its style, not
around or above or in spite of it.’ (87) I argue that this is very much
the case, for example, in ‘A Grammarian’s Funeral’ too. Armstrong
(1993), widely acknowledged as one of the most influential writers on
Victorian poetry and certainly insightful on Browning, characterises
his style as ‘systematically ambiguous’.
Browning’s style continually frustrated and irritated his
contemporaries including admirers like Ruskin, who felt (missing the
point, we now believe) he could do so much better and be more
popular if only he could write more clearly. In a careful response to
Ruskin in a letter explaining why his poetry could never meet those
demands, Browning explains that ‘all poetry is a putting the infinite
within the finite ... You would have to paint it all plain out, which
can’t be’ (quoted in Tucker, 10; full letter in Thomas 1990). Because
of his undoubted and acknowledged influence on modernists like
Pound, who regularly foregrounded form and language as a problem, a
contemporary reader perhaps finds Browning less obscure than his
contemporaries did. Again, to a discourse analyst used to reading
transcripts of natural speech and conversation, regular Victorian

protests that ‘people just don’t speak like that’ remind us uncannily of
the 21st century first year undergraduate. Nevertheless, what Tucker
and the other critics are getting at is certainly a recognisable style of
restless ongoing reformulations, exclamations, breaks in syntax,
colloquialism, ellipsis, overlong sentences and digressions, hesitant
appositions, occasionally obscure vocabulary and neologisms, a
general jerkiness, almost a stuttering because there is just too much to
say at once about this complicated world his speakers inhabit.
(‘Johannes Agricola’ is one of several speakers reduced to the
inarticulate: ‘Gr-gr-gr’, he foams in his rage.) ‘Ungrammatical’
language, then, for a traditionalist at least. This style, as Tucker
suggests, is the full, complex and even confusing world Browning’s
poems offer his readers, and it is easy to see that a Victorian poetry
reading public looking for rest, solace and pious thoughts from its
poets would have been made uneasy by the demands this writing
makes on a reader. The ‘point’ of a Browning poem is careful, patient
reading, rather than any pat conclusion – but also the point is that
reading is not enough. However careful, close, patient and exhaustive,
a reading alone will not give us the comforting answers we seek. This,
it seems to me, is exactly the challenge for a poststructuralist
stylistics, which must remain committed to the value of close and
systematic explorations of a text, but at the same time recognise the
necessary limitations, indeed relativity, of any such endeavour.
Before we turn, however, at last, to our chosen poem, it is important to
acknowledge that for all the twists and turns, incompletions and
incoherencies, Browning himself never seems to have faltered in his
belief that a great Author behind the author had the understanding and
the full story that no human ever seems able to gain. An intellectual
very much up to the minute with the debates of the age, writing in the
age of the ‘disappearance of God’ only made Browning more urgent
in his insistence on the final reality of Christian belief. His is a poetry
of necessary failure but not of final defeat. My own deconstructive
reading of his poetry I justify on the grounds not of clarifying
Browning’s own meanings, but rather on the grounds of showing how
his style and method led him to indeed exceed his own understandings
and knowledge in his poetic practice, if not quite in the ways he would
have wanted. Thus his poem ‘Cleon’, for example, has an unintended
poignancy for readers for whom God has disappeared, where
36 Geoff Hall
Browning clearly intended comfort: his conclusion is inconclusive for

many readers today. We now turn to see how this works in more detail
in ‘A Grammarian's Funeral’, in many ways a typical as well as a
particularly interesting dramatic monologue. The effort in what
follows is to investigate in what ways a new understanding of
grammar – or less cryptically, a stylistics cognizant of poststructuralist
claims – might facilitate understandings of the wider disorder of the
universe which so pervasively informs Browning’s poetry.
3. ‘A Grammarian's Funeral’
A Grammarian's Funeral is found in Browning’s collection Men and
Women (1855), the most widely known and generally the most highly
valued collection for modern readers, though we should note that
initial reception was more of perplexity than of praise. It took some
years for these ‘difficult’ poems to become more widely readable.
What is more, the complexities and instabilities of interpretation
sketched earlier in this piece with respect to ‘My Last Duchess’, the
subtle characterisations we now value Browning for, were not
immediately evident to first readers any more than they are to many
initial readings today. ‘A Grammarian’s Funeral’ invites a reader to
consider the life and works of the dead grammarian as his body is
carried by his followers to the top of a mountain for last rites.
Awareness of Browning’s own veneration for classical Greek
grammar and the achievements of the Renaissance, and a superficial
reading, would suggest that the life of learning for its own sake,
disinterested scholarship, is celebrated as we reflect on the life of this
scholar. What an adequate stylistic account will need to explain,
however, are the increasing doubts readers have had about this ideal
through the twentieth century to the present.
The suspicion begins formally with the respected Victorian scholar
Altick, suggesting in 1963 that the poem should be taken as a mock
encomium on the model of Erasmus’ Praise of Folly: the old stylistic
chestnut of reading not so much the ‘words on the page’ as ‘between
the lines’. Nuttall (2001: 96) concludes ‘the poet does not know what
he means’, that is, whether this life was absurd or glorious. Bohn
(2006) produces a persuasive reading of the grammarian as pederast
rather than pedant, seeking to cover the traces of his culpable life as a
sham scholar. These diverging interpretations of course tell us as
much about their writers and their times as about ‘the’ meaning of the
poem (if we still insist on that hermeneutic quest). But by the same
token it is worth asking ourselves why Browning is at least as highly
valued as he has ever been, and also why this poem has attracted
important essays in prestigious journals in recent years, and what all
this tells us about our ways of reading and understanding texts and
world. (And, stylistically, we note that these critical essays tend to use
questions in their titles: ‘Praise of Folly?’ (Altick, 1963) ‘Apparent
Failure or Real?’ (Svaglic, 1967) ‘Accents Uncertain?’ (Nuttall,
2001)).
In presenting the poem to ‘ordinary’ modern readers, I can report that
the typical response today is certainly one of mockery or disbelief
rather than the respect which seems to have been intended, a post-
Lawrencean disdain, perhaps, for one who chooses to ‘know’ rather
than to ‘live’. The challenge for the teacher – stylistician or otherwise
– in our present achievement-driven, target-obsessed culture is to
show how there is clear evidence that this life of learning for its own
sake was not (necessarily) a pathetic waste of a life, to open up
meanings, not to reduce and close them down. Even a single reader
(me) can change interpretations from one reading to another. How are
such conflicting readings enabled by this text – which is, I have
suggested, a prototypically modern text? A comparable reflection on
the nature of success in a life, likely to be better known to readers, is
‘Andrea del Sarto’ (‘called The Faultless Painter’): Is Andrea a fraud,
a coward and a hopeless flop? Is perfection in a human being a
problem? or is he a real human being who did his best, than which no
praise can be greater? ‘This man’ [the Grammarian] ‘decided not to
Live but Know’ (139): is that praise or condemnation? Or perhaps we
should refrain from either? A stylistic case could be made for the end-
focused affirmation of ‘Know’ as opposed to the negated ‘not to
Live’. On the other hand, a stylistician could argue for an echoic irony
as the grammarian's own motto is repeated for our contemplation –
and there is arguably a negative homonym buried in ‘know’ (no: ‘but
no’). From early publications such as ‘Paracelsus’ (1835), Browning
returns to this theme of the ambitious man who inevitably fails. This is
not the simplistic story of Smiles’s ‘self made’ Victorian heroes, but
neither is it a simple modern counsel of despair. The value of
Browning’s art – its greater readability for us today – is rather the
final undecidability of the large and most important questions.
38 Geoff Hall
To answer such questions – or rather, as we shall see, to come to

understand the final impossibility of answering such questions,
however urgent – we need to turn to a close and systematic
examination of the language of the poem – and the uncertainties of
interpretation it has promoted over time. The poem is spoken by a
group of scholar followers of their dead scholar leader. They tell us of
a man born with good looks (33-4) who gives up ‘dance’ (4) and the
ways of the world, who contemptuously ‘left play for work’ (45) ‘bent
on escaping’ (46) Already doubts creep in: is ‘escaping’ from ‘the
world’ a good or a bad thing to do? We read on, secure in our literary
competence which tells us that questions raised by a text will find an
answer in the text. The scholar’s growing physical decrepitude is
recorded (52-4) along with his single-minded dedication to learning.
Doubts enter again, when it is suggested that like some of the PhDs
we have known, he will start his master work once he has ‘gathered all
books [have] to give’ (67). ‘No end to learning’ (78) – True, but again,
what does that mean for the place of learning in our lives? His
illnesses grow worse as his learning becomes more prodigious – or at
least ‘ground he at grammar’ (126): hard work, surely, unattractive,
surely (the ‘gr’ sound), but how worthwhile? The scholars attempt to
draw a moral in a central passage of the poem (apophthegmatic
present tense):
That low man seeks a little thing to do,
Sees it and does it:
This high man, with a great thing to pursue,
Dies ere he knows it.
That low man goes on adding one to one,
His hundred's soon hit:
This high man, aiming at a million,
Misses an unit. (113-20)
The grammarian is presumably the high man – though here, as
elsewhere, we should note that the binaries Browning often seems to
offer, like those of a pragmatic businessman, do not quite fit the case
at hand. The ‘aspiration’ of the ‘high man’, as with Andrea del Sarto
(‘a man’s reach should exceed his grasp, or what’s a heaven for?’),
seems to be his claim to a worthy life. The self-help books on our
shelves today – an ethic of course already emerging in Browning’s
own time – would clearly argue the greater usefulness and
‘productivity’ of the low man’s life. Glorious failure is oxymoronic, if

not simply absurd for our culture. Yet that is clearly not what the
scholar choir intends, and at the least Browning wants us to consider
an alternative view. What then did the grammarian achieve in his life,
even if we accept that the value of a life may not be reducible to
concrete achievements?
He settled Hoti's business – let it be! –
Properly based Oun –
Gave us the doctrine of the enclitic De,
Dead from the waist down. (129-132)
A useful set of editorial footnotes in the Oxford edition of the poem
(Jacks and Inglesfield 1995: 461) gloss Hoti as ‘what(ever) in indirect
questions’, Oun as ‘therefore’, or ‘certainly, then’, and de as the
inseparable suffix not to be confused with ‘de’ with a different accent,
‘about’. We need these notes because not many today know classical
Greek. Our civilisation has decided it is not a priority for our citizens
– which may be all the more reason to value the grammarian’s work.
Of particular fascination, though, is the fact that Browning is quoted
explaining all this to Tennyson, who shared his love for Greek, if not
his advanced proficiency, and quoting a Greek grammar book which
had evidently been studied minutely. In 1864, Browning wrote a letter
to the Daily News to explain again for a more general readership. The
enthusiasm is evident in both documents, but how finally are we to
judge ‘I wanted the grammarian ... to spend his last breath on the
biggest of the littlenesses: such an one is the “enclitic de”’ (letter to
Tennyson 2 July 1863, quoted Jacks and Inglesfield 1995: 461)? One
common response I have heard to the grammarian’s claims to fame is
the dreaded ‘So what?’ A ‘littleness’ remains after all a littleness.
Another might be ‘did he’ actually ‘fix’ Hoti, oun and de? I am in no
position to judge, but we now know that the idea of a definitive
grammar, except for a specific text or perhaps a speech community at
a certain point in time, is a delusion, or at best ‘idealisation’ (Labov).
Grammar is not like that. Language is not like that. Future scholars
will not ‘let it be’. And yet (more Browningian vacillation) as Sinclair
has taught us in our own time, the little words are indeed the most
fascinating and deserving of the most scholarship by anyone who
considers language is in any way important. But surely ‘Dead from the
waist down’ (132), a complete line on its own with which these claims
40 Geoff Hall
are closed, is the final condemnation? Interestingly, however, I can

find no earlier example of use of what we now think of as a familiar
phrase whose meaning we assume is self-evidently negative, in the
OED or elsewhere, though there is a near parallel in Browning’s hero
Shelley (metaphorically ‘dead to all love’ Queen Mab v. 33). Phrases
like ‘dead boring’, ‘deadly’ date from the 1930s (OED; Deignan
1995). For an age in which one’s sexuality (Life) is one’s identity and
validity, the phrase seems a final indictment (Sir Clifford Chatterley),
but we should remember that it may have been more simply
descriptive as first used, probably here, or even considered a blessing.
Will some of the apparent contradictions and uncertainties be resolved
in the conclusion? The funeral procession (Life as a journey) reaches
the peak of the mountain, the ‘proper place’ (133: ‘proper’ in what
sense?). They are now far from human habitation, among the
‘highfliers’ (135), referring primarily to ‘swallow and curlews’ (136)
but surely also to the Icarian grammarian, whose irrelevance to the
ordinary person was to be argued for a later more disillusioned
generation by Auden (‘About suffering they were never wrong, the
old masters’, ‘Musée des Beaux Arts’). Again, though, note that just
because today it is virtually impossible to speak of a highflier without
a knowing intonation or grin, this has not always and in all places
been the case. To the end:
Lofty designs must close in like effects:
Loftily lying,
Leave him – still loftier than the world suspects,
Living and dying. (145-8; final lines)
The uncertainties seem to proliferate here, far from being resolved.
The reiterated lofty/loftily/loftier resolves nothing concerning the
value of these high aspirations. Also, is the repetition not a little
overblown? overinsistent? overinflated? Is the literally ‘lofty’ end, at
the peak of a bare and obscure mountain, not a somewhat bathetic
version of more metaphorical aspiration? A design, again, is not an
achievement. Effects are not to be mistaken for realities, though it is
important to note that stylistics has normally modestly claimed to
investigate grounds for effects, perhaps especially lofty effects that we
‘like’ (the grand style, the sublime), if that is what literature with its
vexed relations between truth and lying characteristically offers, and
argued for identifiable if complex relations between forms and
meanings. As noted at the outset of this essay, ‘lying’ is a dangerously

ambiguous term for one with high claims to truth, as prone is not
upright. We must ‘leave him’ as must his epigones, but who or what
exactly are we leaving? He is ‘still’, that is dead, as well as ‘even’
loftier than the world expects, ‘even’ today, whatever ‘lofty’ might
represent (a faintly ridiculous nickname for someone excessively
tall?). We can at least agree ‘the world suspects’ his loftiness, whether
they or we are right or wrong to do so. Who, finally, though, is or was
‘Living and dying’ in the last Beckettian circle of apparent mortal
futility? The world or the grammarian? or both? In which case, he was
not so different was he? Questions and more questions multiply as the
poem ends. Irresolvable in Browning’s poem or out of it, they stare us
out of eternity. The unclear reference here is a fitting end to an unclear
poem, an unclear life, for Browning, for the grammarian, for us.
‘Stylistics is the analysis of formal features in the text to show how
they function in its interpretation and in the production of literary
effects’ (Maybin and Pearce 2006: 6). I have found my own attempt to
consider closely and systematically the characteristic formal features
of ‘A Grammarian’s Funeral’ valuable, even though it is important to
note at the same time that they have only succeeded in deepening, not
resolving the interpretative complexities the poem offers. Browning
here, as elsewhere, to my reading, teaches his readers that the answers
they seek are not to be found in the text of his poems, even though the
poems may prompt or encourage worthwhile questions. For me a
stylistic analysis of the kind offered would be a response to more
single minded interpretations, and I would want to believe it was more
accurate or valid than one a Victorian could have seen. But that is of
course from the perspective from which I write.
4. Conclusion. On stylistics and poststructuralism.

Grammar, interpretation and meaning in a godless age.
I fear we are not yet rid of God because we still believe in grammar.
What Nietzsche was asking his Victorian readers to consider in this
passage were the limitations of traditional heuristics, ways of
interpreting the world, in the light of the evolving nature of
contemporary experience. The dangers and limitations of belief in
God and grammar were commitments to fixed and guaranteed
meanings and significance, clear hierarchies, rules and regulations,
42 Geoff Hall
even prescriptive supposedly ‘standard’ grammars as opposed to

human creativity and purposiveness. Nietzsche died in 1900 but has
been seen by many as an influential prophet for the 20th century. In the
first decade of the twenty first century, the ‘state of stylistics’ should
be a recognition of the necessity but impossibility of formalist
stylistics, but this is a promise rather than a threat. It is the best we
have, a consensual, even rhetorical rather than objective knowledge.
Even discourse stylistics is finally inadequate since it relates text to
context, and ‘context, logically, stops nowhere’, as Derrida (1992)
insisted. The strength of stylistics, on the other hand, is the recognition
that a text, particularly a creative text, has its own grammar which
impacts on the system supposedly informing it, as much as being
formed by it. The birth of this ‘grammaring’ (Larsen-Freeman 2003),
for which a grammar or ‘language’ is an abstraction when opposed to
Bakhtinian struggles over language, competing discourses, is the old
grammarian’s funeral. The grammarian’s Greek was a ‘dead
language’. ‘In living speech, messages are, strictly speaking, created
for the first time in the process of transmission, and ultimately there is
no code’ (Bakhtin/Medvedev 1985; quoted in Todorov 1984: 56)
‘Every time language is used, it changes’ in Larsen-Freeman’s more
recent ‘ecological’ formulation, which also draws on the new
metaphors for knowledge coming out of chaos theory (Larsen-
Freeman 1997: 148). The important basic idea is the contingency and
relativity of meaning of any linguistic form, as is particularly evident
in more literary texts in living languages (compare Hardy 2003: Ch. 8;
Hartman 1970).
Despite everything he wanted to believe, and what his own education
told him, Browning had begun to suspect (‘fear’), that traditional
approaches to grammar, or more broadly to language and language
use, were missing the point, just as the Grammarian’s life had
arguably missed it too. The old systems were failing the Victorians, as
our grammars have typically fallen short for aspiring stylisticians,
even if, like those earnest forebears, we are not always sure what to
put in their place. The principle of rigorously respecting the validity of
local and contextual meaning generation is nevertheless increasingly
clearly established. Stylistic analysis must inform interpretation even
as actual texts and interpretations extend stylistic understanding. I
close then with the proposal that stylisticians must build upon such
new ‘emergent’ understandings of grammar, where particular contexts
of use, practices and participants develop a grammar as much as any

grammar determines the language they use. What could be termed
state of the art stylistics, to finish on a more positive lexical note.
References
Altick, R.D. 1963. ‘“A Grammarian’s Funeral”: Browning’s Praise of Folly?’ in
Studies in English Literature 3(4): 449-60.
Armstrong, I. 1993. ‘Browning in the 1850s and after’ in Victorian Poetry. London:
Routledge: 284-317.
Bakhtin, M. and P. M. Medvedev. 1985. The Formal Method in Literary Scholarship
trsl A.J. Wehrle. Cambridge, MA.: Harvard University Press.
Bohn, A. 2006. ‘Increasing suspicion about Browning’s grammarian’ in Victorian
Poetry 44 (2): 165-182, 235.
Bristow, J. (1991) Robert Browning. New York and London: Harvester Wheatsheaf.
Browning, R. [1855] 1995 Men and Women. The Poetical Works of Robert Browning.
Volume 5.(eds) I. Jack and R. Inglesfield. Oxford: Clarendon.
Byron, G. 2003. Dramatic Monologue. The New Critical Idiom. London: Routledge.
Deignan, A. 1995. Collins Cobuild English Guides 7: Metaphor London:
HarperCollins.
Derrida, J. 1992. Acts of Literature. (ed.) D. Attridge. New York and London:
Routledge.
Foucault, M. 1974. The Archaeology of Knowledge trsl. A.M. Sheridan Smith.
London: Tavistock.
Hardy, D.E. 2003. Narrating Knowledge in Flannery O’ Connor’s Fiction. Columbia,
S.C.: University of South Carolina Press.
Hartman, G. 1970. ‘The voice of the shuttle: Language from the point of view of
literature’ in Beyond Formalism. Literary Essays 1958-70. New Haven, CT: Yale
University Press: 337-355.
Jacks, I. and R. Inglesfield (eds). 1995. The Poetical Works of Robert Browning
Volume 5. Men and Women. Oxford: Clarendon Press.
Larsen-Freeman, D. 1997. ‘Chaos/ complexity science and second language
acquisition’ in Applied Linguistics 18(2): 141-65
Larsen-Freeman, D. 2003. Teaching language: from grammar to grammaring.
Boston: Heinle.
Maybin, J. and M. Pearce. 2006. ‘Literature and creativity in English’ in S. Goodman
and K. O’ Halloran (eds) 2006. The Art of English. Literary Creativity. London:
Routledge and Open University Press: 3-24
Miller, J. Hillis. 1963. The Disappearance of God. Cambridge, MA.: Belknap/
Harvard University Press
Nietzsche, F. 1998. Twilight of the Idols. trans. D. Large. Oxford and New York:
Oxford University Press.
Nuttall, A.D. 2001. ‘Browning’s Grammarian: Accents Uncertain?’ in Essays in
Criticism 51(1): 86-100.
Svaglic, M.J. 1967. ‘Browning’s Grammarian: Apparent failure or real?’ in Victorian
Poetry 5: 93-103.
Thomas, D. 1990. The Post Romantics (ed.) D. T. Thomas. London: Routledge.
44 Geoff Hall
Todorov, T. 1984. Mikhail Bakhtin. The Dialogical Principle. trsl. W. Godzich

Manchester: Manchester University Press.
Tucker, H.F. 1980. Browning’s Beginnings. The art of disclosure. Minneapolis:
University of Minnesota Press.
On Genuine Interdisciplinarity: Articulating Poetics as
Theory
Patricia Kolaiti
University College London
Abstract
This paper argues in favour of a genuinely theoretical and interdisciplinary strand
within poetics. It follows the trajectory of a series of ‘inherited confusions’ to show
that the domain of poetics/stylistics has been taken over by massive
‘applicationization’, and proposes that the ever-growing imbalance between
application and theory is one of the most pressing issues for contemporary poetics and
stylistics. In response to this development, I briefly explore the possibility of poetics
as theory and try to blend it with a call for genuinely interdisciplinary practices within
literary study. The paper also considers some concrete proposals on how cognitive
pragmatics could play an important part in the creation of a genuinely theoretical and
interdisciplinary poetics.
Keywords: poetics; interdisciplinarity; theory vs application; interpretation; pre-
theory; cognitive pragmatics; literature as a cognitive/psychological object.
1. Introduction
I would like to start this paper with what seems to me an interesting
paradox, although, as the linguist Neil Smith pointed out, this is not a
paradox in the strict philosophical sense.
The question of the ontology of the work of Art (Wittgenstein 2001,
Danto 1981, Tilghman 1984, Fodor 1993) is perhaps one of the most
widely debated issues in Aesthetics. Simply expressed, the ontological
enquiry in Aesthetics asks what makes an object a work of Art. In the
various sub-domains of Aesthetics one encounters much narrower
varieties of this question: Literary Aesthetics, for instance, asks what
makes a verbal object a work of Art.
The resemblance of this latter question to Jakobson’s influential
statement on Linguistics and Poetics1 is startling. In his seminal

Many thanks to Deirdre Wilson, Billy Clark and Anne Furlong for their very
constructive input, and to Paul Simpson for his interesting comments on the
presentation of this paper at the 26th PALA conference. Also, many thanks to the
‘Lilian Voudouri Public Benefit Foundation’, the AHRC and the UCL Graduate
School for funding my MA and Doctoral studies.
46 Patricia Kolaiti
‘Closing statement’ at the Indiana conference in 1958, Jakobson

suggests:
‘Poetics deals primarily with the question, “What makes a
verbal message a work of art?”’
He continues:
‘(…) [T]he main subject of poetics is the differentia specifica of
verbal art in relation to other arts and in relation to other kinds
of verbal behaviour (…)’ (1958/1996: 10).
I hope you will not disagree with me that the question raised by
Jakobson in this statement is clearly no other than the question of the
ontology of the literary work of Art. And if that is so, then we seem to
be running into a little problem here: simply rephrase Jakobson’s
statement into the legitimate ‘Poetics deals primarily with [a] question
[which is central to Literary Aesthetics]’ and a Pandora’s box
immediately opens. Is Jakobson aware of the mix-up? Does he
purposely put forth the view that Poetics is synonymous with Literary
Aesthetics? Nothing really in his ‘Closing Statement’ strongly
suggests so. The only other reasonable explanation we are left with is
that Jakobson uses the term ‘Poetics’, but what he actually talks about
by use of this term would be more adequately described as Literary
Aesthetics.
Our paradox is unravelling: the most frequently referred to and quoted
statement on Linguistics and Poetics since that seminal conference in
Indiana in 1958 has little to say on the relation between Linguistics
and Poetics and, in fact, illuminates the relation between Linguistics
and Literary Aesthetics.
I am certainly not aiming at any sort of reduction here. Jakobson’s
programme was pioneering for its time and did not in the end reduce
Poetics to Literary Aesthetics alone. His work on parallelism2 or
revolutionary articles such as ‘Poetic grammar in C.P. Cavafy’s
‘‘Body Remember’’’, suggest that he had certainly also envisaged a
systematic theory of literary interpretation. Still, his discipline-shaping
statement does so and it is because of this somewhat unfortunate fact
that it stands as an articulate allegory for the circularities and
confusions that have bedevilled the domain of poetics ever since its
inception.
On Genuine Interdisciplinarity 47
From Aristotle’s essentially canonistic ‘De Arte Poetica’ through to

today’s contending polyphony in Literary Stylistics, one can find a
huge variety of views on what poetics is or what poetics can do and, to
put the cherry on the cake, the present approach will add just one
more.
2. Application, Application, Application!

To permit any constructive proposals as to where poetics/ stylistics
should be heading in the future, one ought – amongst other things – to
have the ability to ‘see’. Allow me an idiosyncratic interpretation of
‘see’ here, to mean: ‘see through inherited confusions’.
Much of the recent history not only of poetics but of literary study in
general has been characterised by such confusions; hence my choice
of introduction. And although the inherited confusion as to whether
Jakobson’s seminal statement concerns the relation between
Linguistics and Poetics or between Linguistics and Aesthetics is
something that I will not return to in this paper – something I only
brought up exempli causa – the confusion I am going to talk about
next will hopefully prove critical to understanding what is bedevilling
poetics/ stylistics at the present.
In Jean Jacques Weber’s (1996: 3) overview of ‘contextualized’
poetics – to choose a random example – stylistic labour is sub-divided
into ‘applied’ and ‘theoretical’. The so-called ‘applied’ strand is taken
to involve a wide array of applications of stylistics to pedagogical
tasks, such as foreign-language teaching or reading and writing skills,
with all remaining analytical activity squeezed under the blanket
‘theoretical’. And this, without doubt, is the standard way of seeing
the division of labour within contemporary stylistics.
But, every time I lift this blanket I get a déjà vu. What contemporary
stylistics treats as a distinction between ‘application’ and ‘theory’ is in
my view more adequately described as a distinction between two
different types of application – let us refer to them as first order and
second order application.
The so-called ‘theoretical stylistics’ could be more adequately
described as a case of first order application, as the standard practice
in theoretical stylistics at this moment is to adopt the theoretical
vocabulary and machinery of other disciplines – e.g. functional
48 Patricia Kolaiti
linguistics, cognitive linguistics, etc – and simply apply them in

interpreting a specific literary text or describing a literary genre.
Hence, this is not the case of a discipline drawing on its interface with
other disciplines and developing an eclecticist dialectics with a view
to enriching its own theoretical repertoire. This is rather the case of a
discipline that has for the most part ignored the possibility of
developing its own theoretical repertoire and has instead adopted the
practice of borrowing theory from other domains, accepting it as a
priori true and then using it to facilitate the process of literary
interpretation. Such practices may well have a subsidiary theoretical
output, but do not really count as theory in any robust sense of the
term.
Along the same lines, ‘applied stylistics’, could be more adequately
described as a case of second order application, given that it takes the
output of ‘theoretical stylistics’ and applies it even further to serve
more practical goals in today’s diverse educational market.
Is it unfair to suggest that, as it stands, the distinction between
‘application’ and ‘theory’ in contemporary stylistics is just another
inherited confusion? ‘Theoretical stylistics’ is theoretical mainly in
name. And the crucial problem for contemporary stylistics is that it
scarcely produces theory in any robust or adequate sense of the term.
3. Beyond interpretation
A healthy affiliation between application and theory within
contemporary poetics/stylistics is desperately needed. And no such
affiliation can obtain until the notion of theory for poetics is put into
place. But first let me make some incidental remarks.
I would be inclined to propose that the turn towards the massive
‘applicationization’ of the discipline is a result of the disproportionate
growth of a single tendency at the expense of all others. The urge for
‘more solid and accurate forms of literary interpretation’, to use
Jackson’s words (2002: 165), i.e. the persistent focus on the
correspondences between the formal features of a text and the
interpretations they trigger (usually referred to as ‘exegesis’, ‘applied
criticism’, ‘analysis’, ‘close reading of the text’, etc.) has nearly taken
over the discipline and obscured the wealth of theoretical ventures
poetics could be setting out on.
In his ‘Introduction to poetics’ Tzvetan Todorov seems soberly aware

of the fact:
‘Poetics (…) is burdened by an age-old tradition that (…)
always arrives at the same result: abandonment of all abstract
reflection, confinement to the description of the specific and the
singular. (…) [F]or each Lessing who describes the laws of
fable, how many exegetes who explain the meaning of one fable
or another! A massive imbalance in favour of interpretation
characterizes the history of literary studies: it is this
disequilibrium that we must oppose, and not the principle of
interpretation’ (1981: 11-12).
Todorov’s implied definition of ‘theory’ can only be comfortably
accommodated within the structuralist agenda; his views, however,
capture an important empirical fact that seems to evade most recent
discussion on poetics: that poetics is not synonymous with
interpretation. That interpretation is merely one amongst an indefinite
number of questions poetics could potentially be contemplating. The
arbitrary equating of the two is yet another inherited confusion, and
one that is now crippling the discipline – you’ll see in a second why I
use the metaphor ‘crippling’– by making theory marginal for poetics3.
Since late antiquity, traditional literary studies have had a binary
arrangement, with traditional literary criticism on the one hand
standardly focusing on literary interpretation, and traditional literary
theory on the other standardly grappling with the vast range of
theoretical and philosophical questions that spring out of the processes
of producing, receiving or being a literary text.
Traditional literary studies
Traditional literary criticism Traditional literary theory

(Interpretation of literary texts) (Theoretical/philosophical enquiry)
About fifty years ago, contemporary poetics/stylistics set out to

modernise traditional literary studies. The emphasis it has placed on
literary interpretation, though, legitimises only a view of it as the
contemporary equivalent of traditional literary criticism; most
stylisticians are fully aware of this and feel very comfortable about it.
50 Patricia Kolaiti
As a result, however, literary theory is still articulated in much the

same terms as it was in the early 20th century and has therefore been
left without a contemporary equivalent. Poetics at the moment is a
discipline with only one leg! And hence my previous choice of
‘crippling’.
Traditional literary studies
Traditional Literary criticism Traditional Literary theory
Contemporary Poetics/ Stylistics (?)
4. Poetics as an inter-discipline
To seek to articulate poetics as theory in the modern day is to plead
for genuine interdisciplinarity within literary study. It has been
pointed out repeatedly that literature is not an autonomous object, and
indeed, any theoretical domain that hopes to take on the totality of
literature as its subject matter should not be anything less than an
inter-discipline.
This statement can be assigned two distinct interpretations. First, the
literary event cannot be addressed by one discipline alone without
being heavily downgraded.
A century ago literary scholars would squarely acknowledge the non-
autonomous nature of the literary object, but at the same time, could
not be accused of theoretical misconduct for not employing
interdisciplinary explanatory tools, simply because there were not any.
Today, there is no excuse either for the literary theorist who chooses
to speak in what Toolan (1996: 118) described as ‘dinosaurian’ Neo-
criticism or for the stylistician who has given up theoretical discourse
altogether. Alan Richardson’s caustic criticism captures something of
the grotesque face of things:
‘Scholars of the future age may well find amusement in the
pretensions of one English professor after another to solve the
riddles of human agency, subject formation, language
acquisition, and consciousness, with little or no awareness of

the spectacular developments in psychology, linguistics,
philosophy of mind, and neuroscience that form the central
story of Anglo-American intellectual life from the 1950s to the
present. . . . The cognitive neurosciences have emerged as (…)
[the] most exciting and rapidly developing interdisciplinary
venture of our era. That this remains news to many working in
literature departments has already become something of an
embarrassment; it will steadily prove more so’ (Richardson
1998: 39).
Second – and more critical – adequate literary theorizing should be
expected to have backward effects on theory produced in the full
range of disciplines – and particularly empirical disciplines – with
which theoretical poetics interacts4.
Interdisciplinarity as used here refers to a reciprocal, bi-directional
relationship between two disciplines, such that theory and practice in
the one discipline can, at least in principle, have a direct bearing on
theory and practice in the other, and vice versa.
Lacking a theory proper, poetics has so far been borrowing from
linguistics and the many scientific domains that have entered the game
of inter-blend since Jakobson’s time without necessarily considering
whether it could or should give something back. Many literary
theorists have viewed this uni-directional game as the only realistic
possibility for poetics and dismissed the idea of genuine
interdisciplinarity as unattainable, particularly when the other side of
the inter-blend involves cognitive domains. T. E. Jackson comments:
‘As far as I can tell, this dialectical relationship (…) cannot be
the case with cognitive literary studies because the originating
theory cannot, even in principle, be recursively affected by the
investigation. An application of that theory to literature may
well change something of our understanding of literature, but it
is difficult to see how the interpretive practice can possibly
change the theory’ (Jackson 2002: 177-178).
Jackson’s pessimism is justifiable and to some extent correct; to the
extent that we commit ourselves once more to the view of ‘poetics’ as
synonymous only with ‘interpretation’. Note how Jackson explicitly
refers to poetics as ‘the interpretive practice’. As soon as we depart
52 Patricia Kolaiti
from this limiting view of the scope of poetic theorizing, it will be

much easier to see that interdisciplinarity is both necessary and
possible for a meaningful and constructive theoretical poetics.
The bi-directionality of the interdisciplinary relation, as I have
presented it in this discussion, gives rise to two parallel questions.
The first is how pertinent scientific enquiry can be of use to literary
studies. The applied strand of poetics has been addressing this
question in various ways and from various methodological standpoints
since 1958, when Jakobson furnished it with all that prestige.
The second question goes hand in hand with the assumption that
theoretical literary study can recursively affect theory in other
disciplines, and this may explain why, although equally important for
any worthwhile notion of interdisciplinary poetics, it has not been
granted the same amount of attention. This question reverses matters
and asks how (theoretical) literary studies can be of use to pertinent
scientific enquiry.
The cognitive neuroscientist Mark Turner steps in and proposes a
possible line of answer:
‘Scholars of literature and art are highly attuned to the intricate
workings of creativity, invention, language, visual
representation, and the construction of meaning. They offer
superb and illuminating examples that often make the
intricacies of mental operation somewhat easier to see. They
have well-trained intuitions about the intricacies of mental and
linguistic phenomena, and they have ideas about meaning and
form. These intricacies and these ideas have, for the most part,
not yet penetrated cognitive neuroscience's field of vision. They
are part of what scholars of literature and art have to offer
cognitive neuroscience’ (2002: 17-18).
Another line of answer assumes that literary scholars having a lot to
offer other disciplines is one thing and literature as an investigative
object having a lot to offer is another5.
In the last twenty years, advances in linguistics, pragmatics and
cognitive science have put the particulars of literary language under
thorough inspection. The alleged uniqueness of literature as a special
linguistic occurrence, rigorously professed by New Criticism and
Formalism, has now collapsed under the pressure of irrefutable

evidence to the contrary. The majority of stylisticians have managed
to adapt sensibly to these changes by pursuing the study of literature
within the spectrum of a new-coined ‘Linguistics of Writing’ (Fabb &
Durant 1987a). Yet a mist of pessimism seems to have taken over
literary study: is there anything about literature – anything beyond its
institutional specifics – worthy of singling out and pursuing as the
subject matter of a dedicated theoretical discipline?
In papers to follow I plan to address this question in greater depth and
propose sober, possible answers. Here, I will only sketch the rationale
with which this question should be tackled.
1) First, I must oppose the view of Literature as a mere institutional
fact. It is very easy, given the evidence we have available today, to
dismiss the idea of the linguistic distinctness of literature as
inadequate or simplistic. For the early 20th century claims about the
uniqueness of literary language were truly revolutionary. And
deservedly so! Pre-eminent literary figures like Eliot, Pound, Breton,
and Blanchot insightfully attacked the then dominant ‘poetics of
convention’ and tried to replace it with a ‘poetics of causation’: i.e. a
poetics motivated by inherent properties of the literary object rather
than by mere institutional and social agreement. It is a contingent fact
that they looked in the wrong area, i.e. that they tried to find these
inherent properties in the language of the literary text.
2) So the accumulating evidence that literature is not a special
linguistic object is indisputable, but does not entail that literature is
not a special object in any other interesting sense.
3) Such distinctness may well be of, say, a psychological/cognitive
and not a linguistic nature6.
4) If literature is indeed distinct as, say, a cognitive/psychological
object, then its investigation should be expected to have retroactive
effects upon other disciplines because it could highlight issues and
questions in a way no other object can.
5. Poetics as theory and Poetics as Theory: the role of pragmatics

Let us now take up again the thread from section 3. In order to discuss
the question of interdisciplinarity, we took a momentary detour from
that section leaving literary theory progressing one-sidedly and poetics
standing on one leg. Remember?
54 Patricia Kolaiti
There is little doubt that traditional literary theory could also be

brought up-to-date quite independently of the poetics/stylistics
paradigm. In which case, one would look into the complexities
involved in updating literary theory without drawing at all on the
contemporary venture of poetics/stylistics. Apart from the fact that
this leaves the noble ambition of contemporary poetics to modernise
literary study partially unfulfilled, there is one at least additional good
reason why this should not be the case.
As Paul Simpson judiciously pointed out during the discussion of this
paper at the 26th PALA conference, traditional literary theory in recent
years seems to be leaping off into greater and greater forms of
abstraction, appearing all the more indifferent towards the actualities
of literary texts. The text seems to have lost its (once-upon-a-time
vital) role as the starting point of all subsequent abstract reflection.
The focus that poetics/stylistics has placed in recent years upon
interpretation and close textual analysis can prove a huge asset in
harnessing literary theory to more text-motivated investigative
practices.
So, where does poetics start with updating literary theory? Well, it
seems that at least some workable varieties of poetic theorizing are
already in place. Early 20th century literary scholarship inaugurated an
ambitious programme of robust theoretical questions and rich
philosophical content. Reviving and re-addressing this agenda –
currently glossed by literary theorists in a semi-primitive theoretical
idiom – in the light of the radical advances of the sciences in the last
30 years is a viable and sensible step. Now more than ever before the
poetician has a vast repertoire of sophisticated descriptive and
explanatory tools to choose from to tackle questions by tradition at the
heart of literary enquiry in a fruitful dialectics with other disciplines.
It is in the very nature of the literary event to excite some curiosity
about human mental processes or old objects out there in the world
that in some way or other relate to the facts of literary production and
reception, or even the fact of just being a literary work of art. Literary
theorists and practising authors have traditionally contemplated the
intricate workings of the subconscious, the nature of creativity, the
interplay between language and thought. They have raised questions
about the limits of linguistic expression, the ontology of intentions, the
machinery of affect.
Highly self-reflexive as a process, literature gives rise to empirical

intuitions and pre-theoretical ideas about production, an aspect of
natural language admittedly neglected by Linguistics. Finally, as an
internally caused creative venture, literature elicits genuine questions
about inspiration, causation, consciousness and free will; here is an
amazing fact that literature does not share with any other non-literary
genre: without deliberation, without being able to explain how or why,
as if a true ‘appearance of the muse’, literature causes itself in a
beautiful, mysterious, uniquely human way.
Up to this point I have been talking about poetics as theory and
intending the term ‘theory’ to contrast with ‘application’. Now, I
would like to momentarily talk about poetics as Theory, as opposed to
‘pre-theory’7.
Twenty or so years since Sperber and Wilson’s Relevance (Sperber &
Wilson 1986/1995, Carston 2002) was first launched one can now
clearly see from the viewpoint temporal distance provides that the
reluctant reception of the theory in literary departments – even
departments that were ground-breaking enough at the time to have
studied ‘Gricean’ Pragmatics – had little to do with the specific
suggestions of the theory per se and more with the deeply entrenched
indisposition of literary scholars towards the cognitive paradigm.
Nigel Fabb and Alan Durant’s (1987a: 10-13) analysis of how the
binary tensions between functionalism and cognitivism have haunted
contemporary linguistics hits the nail on the head. The story is long-
standing and familiar. The aversion towards the cognitive paradigm
itself has very little to do with the specific scientific suggestions of
cognitivism per se and more with its departures from and implications
for well-established ideas about society, culture, religion, the human
being itself. This time it was Chomsky’s (1976: 123-134) turn to
pencil in the picture by showing how ideological prejudice rather than
scientific reason was responsible for the commercial success of the
more romantic Empiricism over the more scientifically sound and far
better evidenced Nativism.
But I hope that even those sceptical towards some of the concrete
suggestions of the theory would not deny that, if nothing else,
Relevance was a forthright epistemic step: the way it broadened the
range of explanatory mechanisms8 available to the humanities, the
56 Patricia Kolaiti
way it replaced pre-theoretical discourse with tractable Theoretical

principles, the way it revived long-forgotten questions in a
surprisingly wide range of other disciplines, makes pragmatics one of
the most influential interdisciplinary ventures of our times. Drawing
on the paradigm of the natural sciences, Relevance attempted a leap
from pre-theory to Theory by producing an investigative language of
tractability, testability and explicitness.
With the notion of what it means to make Theory having changed for
good, the process of updating literary theorising will require a lot
more than simply taking literary theoretic questions and pursuing
them via interdisciplinary investigative practices. In addition to
finding the questions and exploring their overlap with those of other
disciplines, the literary theorist will also have to reflect carefully on
what would be an adequate way of discussing these questions.
I am not one of those who propose the total elimination of pre-
theoretical discourse and its replacement by adequately Theoretical
language. If nothing else, extremisms of this sort are worryingly short
sighted and fail to grasp the contributions pre-theoretical thinking has
made and must continue to make to human intellectual development9.
The weight of this issue is considerable; it has not only theoretical but
also epistemic and ethical implications and could not, of course,
receive the careful consideration it deserves in the scope of this short
paper.
Hence, I am not suggesting that to modernise literary theory by
developing a genuinely theoretical strand within poetics is to produce
a by-product of Relevance Theory or, in fact, a by-product of any
other candidate interdiscipline. Literary theory and poetics are long
standing paradigms; they have their own characteristic discourse and
their own contributions to make to human thought. The challenge is to
let literary theory and poetics be the variety of thinking they are, but
see how this variety of thinking can be partially modified to produce a
more up-to-date and genuinely interdisciplinary theoretical idiom.
What poetics as a theoretical domain could reap from a collaboration
with pragmatics does not stop at purely epistemological rewards. A
common thread that runs through most recent work in stylistics, no
matter how diverse the theoretical camps to which each author
belongs, is a homophonous discontent with the simplistic, fixed,
binary oppositions of the so called ‘Bi-planar’ or ‘Code model of

Communication’ and a parallel desire to locate stylistic enquiry within
pragmatics (Fabb & Durant 1987b: 229-237, Fowler 1996: 199-200,
Kiparsky 1987: 185, Leech 1983, Toolan 1996: 121-124). Together
with the text – which has almost monopolised the attention of poetics
in the last fifty years – , readers and their cognitive environment must
now also be taken into account.
No such thing as literary theory without an underlying theory of
cognition and communication has ever existed. Relevance-theoretic
pragmatics provides a good epistemic example of what investigative
discourse within theoretical poetics could, in part, be like and an
advanced and far reaching theory of communication that theoretical
poetics can certainly do with10. In a healthy interdisciplinary dialectics,
Relevance will also have various rewards to harvest from an inter-
blend with literary-theoretic domains. In the past, some professed
advocates of the theory tried to propose that Relevance holds all the
answers. It does not, and the exchanges go both ways.
6. Epilogue
Contemporary poetics/stylistics has made admirable strides as an
applied discipline, making it possible for stylisticians to play new
roles in today’s competitive industries and explore novel varieties of
social or practical usefulness. But this should not obscure the
importance of striking a golden balance between the ability to use
applied idiom and the ability to grapple with a broad array of
diachronic philosophical and theoretical questions.
Who better than the stylistician to take on the task of reviving and
updating literary theory? The rewards literary study could harvest are
potentially numerous. Let us mention just one par exemplum.
I would like to challenge you to think how the role of the poet as a
theoretician has changed in the last 100 years. A century ago figures
like Eliot, Pound, Breton, Hugo Ball and many others played a central
part in developments within literary theorising. Where is the poet
today? The poet is a theoretical pariah as regards both the discussion
of language – because language is now the subject matter of the quasi-
scientific discipline called Linguistics – and the discussion of poetic
theorising. International events on poetics and literary theory are
58 Patricia Kolaiti
monopolised by academics. With poetics having focused intensely on

literary interpretation, the more abstract and theoretical topics to
which poets could be making interesting contributions are left
untouched11. The bond between poetic theory and poetic practice
seems weaker than ever. Focusing anew on literary theoretical topics
could conceivably enable the poets to reclaim their place within
literary study and put back on the literary theoretic table their valuable
intuitions and insights.
I do not know if the strand I want you to envisage should be called
‘theoretical poetics’ or just ‘poetics’ or ‘contemporary literary
theory’… And it does not matter. One thing is for sure: this genuinely
interdisciplinary and theoretical strand within poetics is a necessary
step towards restoring the breadth and ambition of the field. The
questions are out there. All we need to do is rise to the occasion.
Endnotes
1
For convenience, in this paper ‘stylistics’ and ‘poetics’ will be treated as mere
terminological variants.
2
Parallelism means structural equivalence at the phonological, morphological or
syntactic level.
3
Many theorists have endorsed the convention of talking about ‘Poetics’ in the
consistently narrowed sense ‘the study of interpretation’. However, strictly speaking,
there is no compelling reason to treat the question of interpretation/close reading of
the text as more central to Poetics than other theoretical questions.
4
It is hard to imagine how else it could be. Hard to imagine, for instance, that
theorizing within the Philosophy of Mind (on the issue of mental architecture for
example) could have major implications for Pragmatics by giving rise to questions
such as ‘How does Pragmatics locate within the broader architecture of the mind?’ but
theorizing within Pragmatics would not have retroactive effects on the Philosophy of
Mind. The very minute Pragmatics contemplates the ‘mental location’ of our
pragmatic mechanisms, theory within the Philosophy of Mind is instantaneously
affected. This (if nothing else) is a first-rate example of mutual interdisciplinary
conduct.
5
Turner has an interesting comment to make at this level as well: ‘the theory of
blending, interesting to cognitive neuroscientists because conceptual blending has
been shown to operate throughout everyday thought, language, and action, arose
almost entirely from the study of literary and inventive linguistic expressions’ (2002:
17-18).
6
Can we sense the proportions a project like this could take? An essentialist literary
theory could radically change our view of what literature is. We might find, for
instance, that there is no single common property of literariness running across all
literary genres, that the genres in question inter-relate rather in the fashion of family
resemblances. Let us not forget that the term ‘literature’ itself is a 19th century
invention. So, it might even be the case that some of the genres chosen on the basis of
institutional criteria and treated as ‘literary’ by 19th century theorists are not ‘literary’
in any psychologically interesting sense of the term.
7
To make things simple, I will stick with the convention of using a capital ‘T’ when
referring to theory in this latter sense.
8
Here I use the expression ‘explanatory mechanisms’ to refer to the process of
explaining a phenomenon in psychologically realistic terms rather than merely
describing it. Halliday’s and Hasan (1976) proposals on textual cohesion/textual
grammar, for example, are properly descriptive. Halliday is for the most part
interested in identifying cohesive categories and describing their behaviour in text.
Simultaneously, and precisely because he works mainly within the descriptive
paradigm, he is not interested in whether his notion of textual grammar conflicts with
fundamental assumptions about our mental organisation, the language faculty or even
evolution: is there a textual language module? Is textual grammar distinct from THE
grammar? And what kind of environmental pressures could ever have caused an
organism to evolve a separate textual grammar? If we wanted to furnish his proposals
with explanatory value we would have to test their psychological realism, assess their
compatibility with current views about cognition, language, information processing
and retrieval, evolution and mental structure and make sure they can be
accommodated within a psychologically realistic view of how the mind or human
communication work.
9
I cannot begin to express my distress as a young theoretician about the fact that there
are at this minute Universities and Departments that proudly cultivate a breed of
single-minded intellectuals, for whom a thought without tractability, testability and
explicitness does not count as a worthwhile thought, let alone a thought that deserves
a place within theoretical enquiry. And the reverse, of course.
10
In addition, Relevance-theoretic pragmatics is an exemplary interdiscipline.
Understanding the issues Relevance theory discusses could thus provide the
contemporary literary theorist with immediate insights into the questions and
advances of the many disciplines with which Relevance theory interacts.
11
As Paul Simpson suggested to me, a somewhat different line of thought is also
possible here: poets, he proposed, are preoccupied with issues of form. The
preoccupation of Stylistics with interpretation and formal analysis might in fact be a
possible common ground for collaboration between stylisticians and poets rather than
a deterrent. The more I think about it the more I realise that Paul Simpson is in the
right. I suspect that the very different ways in which we perceive the impact formal
analysis has had on the theoretical role of the poet is because we possibly have very
different poetic traditions in mind. There huge differences between, for instance, the
Anglo-Saxon paradigm and that of Latinogenic literatures (to which Greek poetry also
belongs) and what is true for the one might not be true for the other. The Anglo-Saxon
has by tradition a greater interest in ‘the word’ and in this paradigm formal analysis
might indeed be a luring subject for a poet. The conclusion I make from this is that, if
we decide at some point to consider carefully the overlap between poetic theory and
poetic practice, talk about ‘the poet’ in the abstract will prove a slightly obscure over-
generalisation. There are different paradigms with different traditions and conventions
and, thus, the paradigm from which a poet descends should be seen as a very
important component of his identity and preoccupations as a theorist.
60 Patricia Kolaiti
References
Carston, R. 2002. Thoughts and Utterances: The pragmatics of explicit
communication. Oxford: Blackwell.
Chomsky, N. 1976. Reflections on language. London: Temple Smith.
Danto, A. C. 1981. The transfiguration of the commonplace: a philosophy of art.
Cambridge, Massachusetts: Harvard University Press.
Fabb, N. and A. Durant 1987a. ‘Introduction: The linguistics of writing: retrospect
and prospect after twenty five years’ in Fabb, N. et al. 1987: The linguistics of
writing; Arguments between language and literature. Manchester: Manchester
University Press.
Fabb, N. and A. Durant 1987b. ‘New courses in the linguistics of writing’ in Fabb, N.
et al. 1987: The linguistics of writing; Arguments between language and
literature. Manchester: Manchester University Press.
Fodor, J. A. 1993. ‘Déjà vu all over again: How Danto’s aesthetics recapitulates the
philosophy of mind’ in Rollins, M. (ed.) 1993: Danto and his Critics. Oxford:
Blackwell.
Fowler, R. 1996. ‘Studying literature as language’ in Weber, J. J. 1996: The Stylistics
Reader. London: Arnold.
Halliday, M. A. K. and R. Hasan. 1976. Cohesion in English. London: Longman.
Jackson, T. E. 2002. ‘Issues and problems in the Blending of Cognitive Science,
Evolutionary Psychology and Literary Study’ in Poetics Today 23(1).
Jakobson, R. [1958] 1996. ‘Closing statement: linguistics and poetics’ in Weber, J. J.
1996: The Stylistics Reader. London: Arnold.
Kiparsky, P. 1987. ‘On Theory and Interpretation’ in Fabb, N. et al. 1987: The
linguistics of writing; Arguments between language and literature. Manchester:
Manchester University Press.
Leech, G.1983. Principles of Pragmatics. London: Longman.
Richardson, A. 1998. ‘Brains, Minds, and Texts’ Review 20: 39–48.
Sperber, D. and Wilson, D. 1995. Relevance: Communication and Cognition (2nd ed.).
Oxford: Blackwell.
Tilghman, B. R. 1984. But is it art?; the value of art and the temptation theory.
Oxford: Blackwell.
Todorov, T. 1981. Introduction to Poetics. Brighton: Harvester.
Toolan, M. 1996. ‘Stylistics and its discontents; or, getting off the Fish “hook”’ in
Weber, J. J. 1996: The Stylistics Reader. London: Arnold.
Turner, M. 2002. ‘The Cognitive Study of Art, Language, and Literature’. Poetics
Today 23(1): 9-20.
Weber, J. J. 1996. ‘Towards contextualized stylistics: An overview’ in Weber, J. J.
1996: The Stylistics Reader. London: Arnold.
Wellek, R. and Austin W. 1966. Theory of literature. (3rd rev. ed.) London: Cape.
Wittgenstein, L. 2001. Philosophical investigations: the German text, with a revised
English translation (3rd ed.). Oxford: Blackwell.
Trewe Love at Solentsea? Stylistics Vs. Narratology in
Thomas Hardy
Ken Ireland,
The Open University, UK
Abstract
In assessing the relative merits of stylistic and narratological approaches to a given
text, in this case, Hardy’s short story ‘An Imaginative Woman,’ we might apply a pair
of concepts by the film critic, Laura Mulvey. For the spatial relationships of film
narrativity, she introduces the metaphor of map, for temporal movement that of motor
(Mulvey, 1996). The terms could be usefully transferred to the detailed linguistic
patterning which stylistics involves, and to the elements of plot, action and forward
momentum embraced by narratology. How the two fields interact or mutually
demarcate themselves, to what extent they possess inherent limits and represent
necessarily partial readings of any text, will constitute the larger rubric of this article.
Keywords: Hardy; short story; temporality; Novelle; stylistic foregrounding;
narratology.
1. Introduction
A brief synopsis, first, of the content of Hardy’s story. Ella Marchmill
is the emotionally repressed and poetically inclined wife of a
Midlands gunmaker. On holiday at Solentsea, the couple and their
family take rooms in a house normally occupied by a poet, Robert
Trewe, whose work Ella has long admired. Despite attempts to meet
him (he has retreated to a cottage on the nearby Isle of Wight), she has
to content herself with dressing up in his mackintosh and hat, and
gazing at his photograph and versified scribbles on the bedroom wall,
and on both occasions she is interrupted by her husband. Once home,
she corresponds with Trewe, using the pseudonym of ‘John Ivy’, and
invites him to visit, but he again fails to appear. When his suicide is
announced in the papers, and a letter at the inquest refers to his
inspiration by an imaginary woman, she requests his photograph and a
lock of hair from his landlady, and sets off to visit his grave, where
Marchmill discovers her. Several months later, she dies in childbirth,
and in the dénouement, a couple of years afterwards, as he prepares
for remarriage, her husband finds the photograph together with the
lock of hair. Comparing both with Ella’s youngest child, he concludes
62 Ken Ireland
that the boy must have been conceived by his wife and the poet during
the summer stay at Solentsea, and angrily rejects him.
2. Genre theory
Our first sphere of concern in assessing the differing roles and values
of stylistic and narratological investigation, is genre theory. Dating
from 1893, during the final stage of Hardy’s prose-writing career,
when he was working on what was effectively his last novel, Jude the
Obscure, ‘An Imaginative Woman’ was claimed by Hardy himself to
be his favourite story. After appearing in the Pall Mall Magazine
(1894), it was gathered in the Wessex Tales (1896), but shifted to the
collection, Life’s Little Ironies, for Macmillan’s Wessex Edition of
Hardy (1912), occupying a prominent first place in each edition.
Unlike his contemporary, Henry James, however, Hardy does not
draw any theoretical distinctions between types of prose narrative. If
he first conceives of Jude itself as a short story (Millgate, 1984: 216),
he typically refers in his Prefatory Note of 1913 to the tales in A
Changed Man, as ‘a dozen minor novels’ (Hardy, 1977: 196). His
criterion for good fiction is that the story be ‘worth the telling’
(Hardy, 1930: 158), and form for him, as he records in ‘The Profitable
Reading of Fiction,’ is more a matter of ‘beauty of shape’ than length
(Orel, 1966: 120). In a note written in the same year as ‘An
Imaginative Woman’, Hardy stresses the need for exceptional subject-
matter and presentation: ‘We tale-tellers,’ he suggests, ‘are all Ancient
Mariners, and none of us is warranted in stopping Wedding Guests (in
other words, the hurrying public) unless he has something more
unusual to relate than the ordinary experience of every average man
and woman. The whole secret of fiction and the drama – in the
constructional part – lies in the adjustment of things unusual to things
eternal and universal. The writer who knows exactly how exceptional,
and how non-exceptional, his events should be made, possesses the
key to the art’ (Millgate, 1984: 268).
Hardy’s concern about the degree to which events are exceptional and
unusual in relation to ordinary experience, is one shared by the
German Novelle, the main 19th century prose form in Germany,
Austria and Switzerland. Goethe’s characterization of the Novelle as
‘eine sich ereignete, unerhörte Begebenheit’ (Eckermann, 1948: 225)
directs attention to the nature of the event: one that has actually
occurred, but is still unprecedented. This central event may range from
Trewe Love at Solentsea? 63
an earthquake in Heinrich von Kleist’s Das Erdbeben in Chili, a

murder in Annette von Droste-Hülshoff’s Die Judenbuche, to the
vision of a water-lily in Theodor Storm’s Immensee, and a plucked
orange in Eduard Mörike’s Mozart auf der Reise nach Prag. Writers
of Novellen in the nineteenth century are, however, strongly aware of
an evolving fictional tradition as well as an accompanying critical
theory, in contrast to the relative scarcity of short fiction and
associated criticism in the anglophone world, dominated, as it is, by
the novel.
Other key features of the Novelle include: the presence of a turning-
point (or turning-points), whereby events can take one of several paths
forward; the role of chance or fate, implied by the exceptional nature
of the central event; the concentration upon a single plot or slice of
life, producing a mood of urgency; the testing rather than development
of character; the importance of the narrative mode, often employing a
framework technique; the seeming objectivity of the Novelle, allowing
the indirect expression of subjective feelings; and a strongly
delineated outline, often mnemonic or symbolic, crystallizing into a
concrete object (Dingssymbol): Boccaccio’s falcon (Decameron),
Adalbert von Chamisso’s lost shadow (Peter Schlemihl), Gottfried
Keller’s love-letters (Die missbrauchten Liebesbriefe). As Martin
Swales cautions, though, a genre concept like the Novelle is valid
insofar as it functions as a reservoir of potentiality or as a structuring
force, rather than as an aggregate of observable features (Swales,
1977: 15).
Anglophone theory and practice in short fiction, apart from American
contributors, has little to offer until the closing decades of the century.
Edgar Allan Poe’s stress, in his essay of 1842 on Hawthorne’s Twice-
Told Tales, is on the dynamics of authorial intent and reader response:
the author working out a single preconceived effect by means of
invented incidents and appropriate tone, the reader sharing in a sense
of unity by perusing the work at a single sitting, a procedure
unattainable with the novel. Applying German Novelle theory to
Hardy’s story can be instructive. Thus, its central event, in terms of
importance and number of pages, is the scene where Ella, in dressing-
gown and candle-lit atmosphere, imagines a sexual union with Trewe.
The ‘unprecedented’ element posited by Goethe would be the chance
occupation by the family of the rooms occupied by an admired poet. It
64 Ken Ireland
is that initial coincidence which triggers all the later developments.

Ella’s obsession with the absent poet represents the ostensible plot of
the story, while events test her innately ‘impressionable, palpitating’
nature (Hardy, 1977: 12), rather than develop new aspects of
character. Although, elsewhere, Hardy exploits framework techniques
and other narrative modes, this late story strikes a contemporary note
in its direct narrative expression: one reason perhaps why Hardy later
shifted the story from the more historically-based Wessex Tales to the
more contemporary settings of Life’s Little Ironies.
The Novelle’s concrete object is represented in Hardy’s story by a
photograph of the poet Robert Trewe. In the central scene, Ella is
emotionally overcome by his image, and the photograph prompts her
fantasy of sexual union. She guiltily slips his image under the pillow
when her husband unexpectedly returns, and in a cancelled passage of
the MS pushes the picture down as far as she could reach with her
toes: ‘Trewe cannot get any further into her bed than that, and Hardy
must have felt that this was inappropriate for his conception of the
story’, one critic comments (Ray, 1997: 175). Marchmill sees the
photograph next morning, and Ella defends Trewe against her
husband’s offhanded ridicule. After the poet’s suicide, the photograph
serves as her memorial to him, while it functions, after her own death,
together with Ella’s dated inscription on the back, to convince
Marchmill of his wife’s infidelity.
The photograph also projects the inner symbolism explored by the
Novelle form. Thus, the poet, before giving up his rooms to the
Marchmills, asked his landlady to conceal the picture under that of the
Royal Duke and Duchess in the portrait frame. Palimpsest as image
may also be transferred to the level of plot: royal couple overlaying
poet echoes the narrative palimpsest of the Marchmill/Trewe plot
overlaying another private, covert plot, Hardy’s own, as if to realize
the Novelle’s notion of subjective feelings projected on to an objective
surface. At this point, a stylistic investigation proves efficacious, by
tracing out the textual changes and variants between Hardy’s MS,
with its own deletions and alterations, and the three later versions of
the text. What then emerges, is Hardy’s personal subtext: his attraction
to a married woman, the writer and poet, the Hon. Florence Ellen
Hungerford Henniker, concealed beneath the fictional text of the
Ella/Trewe relationship. Martin Ray, in his textual study of Hardy’s

short stories, sums up the evidence (Ray, 1997: 172-73).
The name Ella Marchmill seems inspired by Mrs Henniker’s second
name Ellen, and maiden name Milnes; in the MS, the poet is called
Crewe, like Florence Henniker’s uncle Lord Crewe, with his seat at
Crewe Hall, Cheshire. Her father owned a lock of Keats’s hair, while
she herself, like Ella, is attracted to Shelley’s poetry (quoted in the
story’s central scene), and her soldier-husband, eventually to become
Major-General Henniker, shares military interests with the gunmaker
Marchmill. The Hennikers lived on South Parade, Southsea, Robert
Trewe on New Parade, Solentsea, both houses facing the sea, and
Florence’s house likely inspiring Ella’s. Hardy’s close friendship with
her in 1893, fictionalized in ‘An Imaginative Woman’, was clearly
intense enough to warrant his resort to the disguise of names and
places, such that it is only twenty years later, in the Wessex Edition of
1912, that he feels sufficient distance from the events to allow
possible identifications to be made.
The Henniker narrative below the main Ella/Trewe narrative also
resonates in poems of the period such as ‘The Ivy-Wife’ and ‘The
Division’ (Ebbatson, 1993: 86-7), as well as furnishing aspects of Sue
Bridehead, whose second name also happens to be Florence, the
heroine of Jude the Obscure, on which Hardy was working during
1893. By that autumn, Hardy concedes, in a letter to Mrs Henniker,
the ‘one-sidedness’ of their relationship, his feelings being
unreciprocated by hers (Ray, 1997: 174). In the story, it is the woman
who desires the male poet, so that the device of role-reversal
supplements Hardy’s distortion and erasure of real-life names. It is
also intriguing to speculate that Ella and Trewe never meet in the flesh
throughout the story, and her courtship of him remains a virtual one, a
metaphor perhaps for Mrs Henniker’s blank response to Hardy’s
overtures. While the interplay of presence and absence here would
delight Deconstructionist critics, another commentator finds ‘An
Imaginative Woman’ very relevant to our own times: ‘Trewe is indeed
the ideal post-structuralist writer, silent and invisible’ (Ebbatson,
1993: 90).
In this account of generic concerns, then, a stylistic investigation
proves valuable in terms of the changes and revisions of Hardy’s short
story, which a textual history is able to reveal. Such transformations
66 Ken Ireland
testify to shifts of attitude on Hardy’s part, and uncover a thematic

subtext, to create a double narrative. The majority of features relating
to the German Novelle, and applied to Hardy’s text, however, tend
towards narratology, aspects of motor rather than map.
3. Perceptual distance
With our second sphere of concern, the control of perceptual distance,
stylistics comes more fully into its own. The term ‘perceptual
distance’ itself, refers to the techniques used to focus closely or
remotely on events or characters in a narrative, alternating between
different perspectives, so as to elicit sympathy and understanding, or
to create tension and detachment in relation to the objects represented.
A reader’s perceptions are influenced by particular modes of
representation: verbal deixis, for instance, can produce effects
comparable to the zoom-in movement of a film camera, as do
unmediated speech and thought renditions. In a reverse operation,
narratorial summary and paraphrase echo the pull-back movement of a
camera, whereby panorama replaces close-up, and specifics give way
to generalities.
Early in the first section of the story, Hardy contrasts the differing
temper and tastes of the Marchmills, suggesting that William’s
character is stylistically projected in ‘squarely shaped sentences’ (p.
12), whereas Ella is introduced as ‘a votary of the muse’ (p. 12).
When the landlady draws attention to Trewe’s minute scribblings on
the wallpaper behind the bed, spatial deictics join with temporal
deictics, as she informs a flushing Ella that one particular couplet,
‘must have been done only a few days ago’ (p. 17). The diminutive
scale of the script, Ella’s need to inspect it at extremely close range,
and its very recent origins, reinforce the physical and emotional
impact of Trewe on a receptive admirer, which the sight of the same
couplet in published form shortly afterwards, can only intensify.
In a second case of the control of perceptual distance prior to the
central section, Ella, in a fine example of clothes fetishism, dons
Trewe’s mackintosh and waterproof cap, declaring it to be the mantle
of Elijah, in an allusion to her desire to be, like the Biblical Elisha, a
worthy successor to a poetic, if not prophetic genius: ‘Her eyes always
grew wet when she thought like that, and she turned to look at herself
in the glass. His heart had beat inside that coat, and his brain had
worked under that hat at levels of thought she would never reach’ (p.
18). Spatial deictics are compounded here by italics and Free Indirect
Thought, to produce an emotional climax, sadly deflated by her
husband’s entry a couple of lines later.
In the third and most striking instance of the control of perceptual
distance, drawn from the central section of the story, and reproduced
in the Appendix, Ella has exchanged mackintosh and cap for her own
dressing-gown, and garish afternoon sunlight for a romantic
atmosphere, nicely rendered in trochaics, of ‘silence, candles, solemn
sea and stars’ (pp. 20-21). Marchmill has gone off yachting with his
friends and is not expected back until next day, the children are in bed,
and Ella, ‘with a serene sense of something ecstatic to come’ (p. 20),
reads some of Trewe’s ‘tenderest utterances’ (p. 21), before finally
opening up the picture-frame. In what follows, arguably the pivot of
the story, motor yields to map: narratological elements of plot, action
and forward momentum are supplanted by stylistic concerns of
detailed linguistic patterning, as Ella slips into a state of rhapsodic
contemplation.
The widest range and most intensive concentration of foregrounded
devices in the story, then serves to project the semantic importance of
the section onto the stylistic surface of the text. At the same time, the
reader comes as close to her thoughts, feelings and obsessions as is
fictionally possible in the 1890s, within a sequence that embraces
tactile immediacy (ll. 1-3, 22, 24, 26, 34-6), lexical repetition
(1/4/6/7/30, 4/60, 5/6, 11/28, 13/17, 18/25, 21, 23/24, 30/32) and
verbal punning (14/16), Free Indirect Thought (6, 24-26, 30-34),
ellipsis (33) and enumeration (19-20), spatial and temporal deixis (5,
16, 19, 22, 24, 27, 34, 38) often implying an accompanying gesture
(24, 26), and intertextuality both direct and indirect.1 Her husband’s
unexpected return is signalled by the first instance of direct speech in
the scene (41), and a subsequent shift away from Ella’s perspective,
together with an end to foregrounded devices. Marchmill’s
unexpected show of affection to Ella may surprise the reader, but the
transition to the events of the following day and week is stylistically
unmarked (62-63), and it is only at the story’s close, that the
retrospective force of a narratological approach makes itself felt.
68 Ken Ireland
4. Temporal relationships
A convenient point, perhaps, to broach our third and final sphere of
concern: temporal relationships, and their relevance for stylistics and
narratology. While our focus on the central section above was
predominantly stylistic, its chief temporal feature was a drastic
slowing-down of tempo, almost to the point of narrative stasis, until
Marchmill’s rude interruption of Ella’s reverie. If we treat speech as
the closest to an ideal pace constant, where story-time=discourse-time,
this near-isochrony soon gives way to an increase in narrative tempo,
as Marchmill, finding the stay at Solentsea ‘getting rather slow’ (p.
23), decides to leave in three days’ time, thus scuppering Ella’s
chances of meeting Trewe. The tempo then increases, as she makes a
hasty and unsuccessful crossing to the Isle of Wight to track him down
(conveyed in a single paragraph), before Marchmill agrees to let the
family stay on longer. A single, terse, typographically foregrounded,
but balanced sentence-paragraph renders this: ‘But the week passed,
and Trewe did not call’ (p. 24).
With Ella back in her Midlands home, the narrative pace accelerates,
and Günther Müller’s correlation of erzählte Zeit/Erzählzeit, story-
time (the period covered by the narrated events) and discourse-time
(the time taken by representation expressed in pages of text), could be
usefully applied (Müller, 1968). Ella and Trewe begin a
correspondence lasting two months, until her invitation to visit is
accepted, and the conversation that afternoon with Trewe’s painter-
friend (Trewe himself has, typically, cried off) brings a singulative
rhythm and deceleration. A couple of days later, she reads a
newspaper account of his suicide, and his letter presented at the
inquest. Real duration is involved here: we as readers presumably take
as long to peruse the verbatim embedded documents as Ella herself,
and her stunned reaction matches the lento pace of her earlier bedroom
reverie.
In the third and shortest part of the story, events move swiftly,
resulting in a heightened narrative tempo. Ella’s response to Trewe’s
suicide is to make a nocturnal visit to his grave, where her husband
discovers her. Months then pass, before her gloomy prediction of
dying in childbirth. Six months later (the whole period rendered in
little more than a page), her prediction is borne out. The final
singulative event occurs a couple of years later, and comprises barely
half a page, as Marchmill compares the features of Ella’s fourth child

with the hair and photo of the poet, and discerns a ‘trewe’ likeness.
The tripartite division of the story is far less obtrusive than in other
stories, being unnumbered, but marked by typographical gaps. ‘The
Romantic Adventures of a Milkmaid’, by contrast, contains no fewer
than seventeen marked divisions; ‘The Withered Arm’ and ‘The
Distracted Preacher,’ moreover, include descriptive headings as well
as segmentation. Given its small scale and single plot-line, too, the
kind of sequential dynamics, involving complex continuity relations
and categories of transition in the more expansive form of the novel
(see Ireland, 2001), are scarcely relevant. One particular temporal
device, familiar to a narratological investigation, is, however, of
significance .
In the central section of ‘An Imaginative Woman’, Marchmill’s
surprise return and show of affection to Ella, concluding the scene
with direct speech (‘“I wanted to be with you to-night’” [ll. 61-62]), is
followed by a new sentence which registers a nocturnal gap: ‘Next
morning Marchmill was called at six o’clock …’ (l. 63). Nothing
untoward here, apart from his noticing Trewe’s photograph in the bed,
before he goes off. Only at the very end of the story, when he
compares photo with latest offspring, does Marchmill calculate that
the date of the child’s conception must have been in the second week
of August at Solentsea, and that Trewe must be the responsible party.
Though the text in the central section maintains a discreet Victorian
silence about Marchmill’s further shows of affection, it is apparent
that the nocturnal gap masks an informational gap or paralipsis
relating to conjugal procreation. Thus, although the story is not
susceptible to the order transforms or linear/nonlinear interplay
characteristic of the novel, this instance of paralipsis is nonetheless
structurally or sequentially pivotal in triggering off the dénouement.
Marchmill’s calendar computations reflect Hardy’s attention in all his
work to temporal measurement and accurate record, exemplified in his
first published novel, Desperate Remedies (1872), where every
chapter and subdivision bears a note of its date and temporal duration.
In the novels of Jane Austen, a time-span of eight months is the norm,
but comparisons with her shorter fiction are naturally ruled out. With
Hardy, however, producing both novels and short stories, the situation
is more intriguing. Conventional assumptions that novels cover longer
70 Ken Ireland
stretches of time than short stories, are likely to be tested by Hardy,

whose unconventionality is again to the fore. Thus, the main action in
more than half-a-dozen of his major novels ranges between one and
five years. In his short stories, by contrast, time-spans range
enormously, from the single evening covered in ‘The Three Strangers’
to the three decades and more in ‘The Waiting Supper’. It is
suggestive, nonetheless, that the one-year-long main action of The
Return of the Native is easily surpassed by the three or more years
accounted for in ‘An Imaginative Woman’. Ella’s pregnancy, whether
in novel or short story, could obviously not be despatched in much
less than nine months, in any case, while Marchmill’s remarriage has
to allow for a decent interval following Ella’s death, such that nature,
in both cases, determines the time-span of art. The ending of the story,
in its practical application of temporal relationships, and its ironical
turn, whereby the imaginative element has ultimately been transferred
from Ella to Marchmill, in his fancy that a ‘transmitted idea’ has been
at work (see Brady, 1982: 103), may be regarded as the ‘twist in the
tail’ often associated with the short story as a genre. 1893 is not only
the year of ‘An Imaginative Woman’, but it also marks the year of the
death of a contemporary master, often associated with the ironic
dénouement: Guy de Maupassant.
5. Conclusion
What emerges, then, from this review of our third sphere of concern,
temporal relationships, is Hardy’s elasticity of approach, his
avoidance of pigeonholing and categorizing material to match specific
containers. Such features relate closely to genre theory, our first area
of concern. In that discussion, it will be recalled, the application of
German Novelle theory tended towards narratological investigation,
and only in the contribution of textual history did stylistics play a
significant role. With temporal relationships, likewise: motor
overshadows map. In terms of perceptual distance, however, the
longest, central and most striking section of the story makes its impact
on the reader as a linguistic ‘tour de force’, richly exploiting the
resources of verbal style. The very concentration and diversity of
foregrounded devices serves here to suspend any interest in plot
events outside, and to focus exclusively on Ella’s inward vision.
If we view the story as a triptych, the flanking panels lead up to and

away from this central scene, but narratological and stylistic elements
assume differing degrees of relevance at differing junctures of the
narrative, only to underscore the overall principle of interdependence.
This might appear to be a truism (or ‘trewism’), though what is self-
evident still needs to be demonstrated in practice. Hardy’s short story,
like all texts, is open to a whole range of critical approaches, of which
narratology and stylistics represent only two possibilities. It is to be
hoped that this analysis has shown, at least, the advantages of
questioning limits and exporting insights across them. Removing
border controls has proved problematic in political terms; could
stylistico-narratologists, even narratological stylisticians, point the
way?
Endnote
1
Between the two lines of direct quotation from Shelley’s Prometheus Unbound, I (ll.
748-49), and Hardy’s paraphrase of l. 737 at the end of the paragraph (‘she was
sleeping on a poet’s lips’), is an allusion in l. 745 to the poet’s watching from dawn to
gloom, the sun illuminating ‘the yellow bees in the ivy-bloom.’ Given Ella’s adoption
of ‘John Ivy’ as her nom de plume, the veiled reference intriguingly underlines her
own position.
References
Brady, Kristin. 1982. The Short Stories of Thomas Hardy: Tales of Past and Present.
London: Macmillan.
Ebbatson, Roger. 1993. Hardy: The Margin of the Unexpressed. Sheffield: Sheffield
Academic Press.
Eckermann, Johann Peter. [1827] 1948. Gespräche mit Goethe, in Ernst Beutler (ed.)
Gedenkausgabe der Werke, Briefe und Gespräche [Goethes] 24. Zürich: Artemis-
Verlag.
Hardy, Florence Emily. 1930. The Later Years of Thomas Hardy, 1892-1928. London:
Macmillan.
Hardy, Thomas. 1977. Life’s Little Ironies and A Changed Man. The New Wessex
Edition. London: Macmillan.
Ireland, Ken. 2001. The Sequential Dynamics of Narrative: Energies at the Margins
of Fiction. Cranbury, NJ and London: Associated University Presses.
Millgate, Michael (ed.). 1984. The Life and Work of Thomas Hardy by Thomas
Hardy. London: Macmillan.
Müller, Günther. 1968. Morphologische Poetik: Gesammelte Aufsätze. Darmstadt:
Wissenschaftliche Buchgesellschaft.
Mulvey, Laura. 1996. ‘Film Narrativity.’ Lecture given at the University of
Cambridge, 12 November 1996.
Orel, Harold (ed.). 1966. Thomas Hardy’s Personal Writings. London: Macmillan.
72 Ken Ireland
Ray, Martin. 1997. Thomas Hardy: A Textual Study of the Short Stories. Aldershot:
Ashgate.
Swales, Martin. 1977. The German ‘Novelle’. Princeton NJ: Princeton University
Press.
Appendix
As she gazed long at the portrait she fell into thought, till her eyes filled
with tears, and she touched the cardboard with her lips. Then she
laughed with a nervous lightness, and wiped her eyes.
She thought how wicked she was, a woman having a husband and
three children, to let her mind stray to a stranger in this unconscionable
manner. No, he was not a stranger! She knew his thoughts and feelings
as well as she knew her own; they were, in fact, the self-same thoughts
and feelings as hers, which her husband distinctly lacked; perhaps
luckily for himself, considering that he had to provide for family ex-
10 penses.
‘He’s nearer my real self, he’s more intimate with the real me than
Will is, after all, even though I’ve never seen him,’ she said.
She laid his book and picture on the table at the bedside, and when
she was reclining on the pillow she re-read those of Robert Trewe’s
verses which she had marked from time to time as most touching and
true. Putting these aside she set up the photograph on its edge upon the
coverlet, and contemplated it as she lay. Then she scanned again by the
light of the candle the half-obliterated pencillings on the wall-paper
beside her head. There they were – phrases, couplets, bouts-rimés,
20 beginnings and middles of lines, ideas in the rough, like Shelley’s scraps,
and the least of them so intense, so sweet, so palpitating, that it seemed
as if his very breath, warm and loving, fanned her cheeks from those
walls, walls that had surrounded his head times and times as they
surrounded her own now. He must often have put up his hand so –
with the pencil in it. Yes, the writing was sideways, as it would be if
executed by one who extended his arm thus.
These inscribed shapes of the poet’s world,
‘Forms more real than living man,
Nurslings of immortality’,
30 were, no doubt, the thoughts and spirit-strivings which had come to
him in the dead of night, when he could let himself go and have no fear
of the frost of criticism. No doubt they had often been written up hastily
by the light of the moon, the rays of the lamp, in the blue-grey dawn,
in full daylight perhaps never. And now her hair was dragging where
his arm had lain when he secured the fugitive fancies; she was sleeping
on a poet’s lips, immersed in the very essence of him, permeated by his
spirit as by an ether.
While she was dreaming the minutes away thus, a footstep came upon
the stairs, and in a moment she heard her husband’s heavy step on the
40 landing immediately without.
‘Ell, where are you?’

What possessed her she could not have described, but, with an
instinctive objection to let her husband know what she had been doing,
she slipped the photograph under the pillow just as he flung open the
door with the air of a man who had dined not badly.
‘O, I beg pardon,’ said William Marchmill. ‘Have you a headache?
I am afraid I have disturbed you.’
‘No, I’ve not got a headache,’ said she. ‘How is it you’ve come?’
‘Well, we found we could get back in very good time after all, and I
50 didn’t want to make another day of it, because of going somewhere else
to-morrow.’
‘Shall I come down again?’
‘O no. I’m as tired as a dog. I’ve had a good feed, and I shall turn in
straight off. I want to get out at six o’clock to-morrow if I can…. I
shan’t disturb you by my getting up; it will be long before you are
awake.’ And he came forward into the room.
While her eyes followed his movements, Ella softly pushed the photo-
graph further out of sight.
‘Sure you’re not ill?’ he asked, bending over her.
60 ‘No, only wicked!’
‘Never mind that.’ And he stooped and kissed her. ‘I wanted to be
with you to-night.’
Next morning Marchmill was called at six o’clock …
(Hardy, 1977: 21-23)
[my underlinings]
Who Is Afraid of Stylistics? Postgraduate Students’
Responses to Stylistics
Nazan Tutas
Ankara University, Turkey
Abstract
The aim of this paper is to give a brief description of the postgraduate ‘Stylistics in
Literature’ course I teach at Ankara University, in the Department of English
Language and Literature, and to discuss how we defined the state of stylistics and
describe the methodology we used in this course. It presents the feelings and the
struggles these postgraduates experience in defining the aims of this field of
investigation and its benefits for them. The promising papers they submitted and the
positive responses I received after the course encouraged me to reflect upon this topic.
Keywords: stylistics; student responses; stylistics course; teaching stylistics; state of

stylistics.
1. Introduction
Many stylisticians agree that stylistics has evolved in many ways and
is continuing to evolve. What ‘stylistics’ means in 2006 is not what it
meant in 1970s. It has been misunderstood or misinterpreted by many
critics. Macleod (2005: 61) thinks that part of the reason for the
misunderstanding and the distrust of stylistics comes from some false
assumptions. First of all, there is an assumption that stylistics claims
to be superior to other forms of literary study. Short (2006) responds
to this and says that stylistics does not claim to replace literary study
or to be able to explain everything in textual understanding and
response. He says ‘the detail of stylistic analysis means that it can
only be applied sensibly to short texts or extracts of longer texts,
leaving plenty of other aspects of texts in need of exploration, as well
as the relations between literary texts and the personal, historical and
social contexts of their production and reception’ (Short 2006: 5).
The second assumption Macleod (2005) identifies sees stylistics as
being so committed to the objective and verifiable description of
relevant linguistic features of a text that it excludes any involvement
of response or appreciation. Stylistics does not ‘deny the relevance of
imagination and response and sympathetic understanding’ (Macleod,
76 Nazan Tutas
2005: 61). It simply develops verifiable criteria for the evaluation of

textual language.
Thirdly, there is a suspicion about the claims stylistics makes to
scientific objectivity. Barry (1995: 207) defends stylistics and says,
‘stylistic analysis attempts to provide a commentary which is
objective and scientific, based on concrete quantifiable data, and
applied in a systematic way’. He asserts that its aim is partly the
‘demystification’ of both literature and criticism, stressing that its
method and procedures can be learned and applied by all (Ibid, 208).
Macleod (2005: 61) agrees with Barry, saying that stylistics ‘offers an
approach, relying on objective description and systematic observation,
which facilitates the exploration of the roots of literary response and
interpretation in accurate description and annotation of the ordinary
and systematic facts of language’.
As Hamilton (2004) states, no matter what people say about
‘stylistics,’ it simply will not go away. One reason for this fact,
according to Hamilton, is that the linguistic theories that provide a
foundation for stylistics in particular and literary criticism in general
have evolved over the years. The generative and systemic functional
theories that Fish (1973) was familiar with in the 1970s have
gradually given way to theories from cognitive science or
sociolinguistics.
Recent stylistics includes work in narratology and psychology
concerning textual understanding, as well as the more context-based
accounts of language understanding deriving from the study of
pragmatics, discourse analysis and sociolinguistics within linguistics.
Consequently, the approaches and analytical techniques used within
stylistic analysis have widened considerably in recent years, helping to
make stylistics more effective in describing texts, how these texts
work and how we respond to them. Short (2006: 6) presents the
following recent trends, which started some years ago and are
becoming more evident in the latest works in stylistics:
Informant-based work to see whether the predictions arising
from the stylistic analysis of texts actually occur when real
readers read texts (e.g. van Peer 1986, Miall & Kuiken 2001),
Who Is Afraid of Stylistics? 77
Corpus Stylistics, the use of corpus-based work to test stylistic

theories and accounts of texts etc. (e.g. Hoover 1999, Semino
and Short 2004),
The use of stylistic analysis to elucidate how we infer fictional
worlds and characterisation from texts (e.g. Semino 1997 and
Culpeper 2001),
Cognitive Stylistics or Cognitive Poetics, the development of
cognitive models of processing to put alongside more text-
analytical work (e.g. Stockwell 2002, Semino and Culpeper
2002, Gavins and Steen 2003).
2. Description of the course

I teach in an English Language and Literature department in Ankara,
Turkey. English Language and Literature departments in Turkey
mostly concentrate on the study of the history of English Literature
and theories of literary criticism. The students study selected literary
pieces and analyse them according to literary theories. Stylistic
analysis is not very common in these departments. Linguistics
departments usually offer stylistics courses in which study is restricted
largely to the formal and highly ‘theoretical’ aspects of Linguistics.
My engagement with stylistics actually started a couple of years ago
when I was asked to teach stylistics in our PhD programme. As the
Department of English Language and Literature at Ankara University,
we had concerns about the lack of knowledge of linguistics and
stylistics among the students, especially when they need to further
their studies at MA or PhD levels. We thought that it would be useful
for these graduate students to study stylistics as well. We believe that
stylistic analysis is potentially beneficial in the integration of language
and literature study since it is an area on the borders of the two. This
means that stylistics can help students understand how to integrate
their study of language and literature. Because it tries to reveal what is
involved in the process of textual understanding and interpretation, it
has a number of specific advantages for students. As Short (2006) also
states, stylistics pushes the students to be more precise and analytical
in thinking about textual understanding and interpretation, helping
them to think harder about the linguistic structure of texts and the
cognitive processes involved in understanding them. This is
something, which is important academically and can also be seen as
the development of a valuable skill that they can transfer into use in
78 Nazan Tutas
later life. When we are trying to understand a text, knowledge of

stylistics can often give us a starting point or, as Short says,
‘something to do’ when you get stuck interpretatively or are not sure
whether your feelings about a text are accurate. In such cases, Stylistic
analysis is particularly useful for non-native speakers of English as it
helps them to infer meaning in context without having to use a
dictionary. Short (2006: 2) argues that a detailed stylistic analysis can
also help students to learn not just how an individual literary text
works but also some valuable general lessons concerning how they
interact with texts, how they infer fictional worlds from texts, how
textual structure constrains interpretation and how an interpretation of
a text involves decisions concerning the appropriate level of
abstraction of that interpretation.
The main aim of our course was to enable students to read texts
sensitively, and perform stylistic analysis on texts they are
encountering for the first time. All the three main literary genres and
relevant comparisons with non-literary texts were explored, and a
wide range of texts and textual extracts were used. Discourse
presentation in news reports and advertisements was also studied.
The participants in this course were five Turkish teachers of English
as a Foreign Language working at different universities and were
graduates of English Language and Literature Departments at different
universities in Turkey. They had no previous knowledge of stylistics
but they knew quite a lot about theories of literary criticism. The
course was mainly devoted to the stylistic analysis of poetry, prose
fiction and drama and lasted one semester (15 weeks-3 hours per
week). The first three weeks were spent on looking at the theoretical
basis of stylistics and introductory discussions.
I found Mick Short’s introductory Web-based course ‘Language and
Style’1 quite useful, and I used a combination of both Short’s Web-
based course format and the traditional lecture-seminar based teaching
approach to ensure the involvement of the students and to encourage
the students to share their views and ideas. The students were asked to
study the topics from Short’s web-based stylistics course, his book
Exploring the Language of Poems, Plays and Prose (1996), Leech and
Short’s (1981) Style in Fiction and other books and articles on our
reading list. One student each week was asked to present and discuss
the topics presented on the webpage in class. They had to read and
listen to the presentations and comment on each other’s papers. I

devoted the second half of the semester to the stylistic analysis of
poetry, prose fiction and drama. The students had to present and hand
in three assignments: a stylistic analysis of a poem, a piece of fictional
prose and a dramatic text of their choice.
In the following sections, I will briefly explain how we defined the
state of stylistics and what we think about the place of stylistics in
literary analysis and the benefits of it for the further studies the
students will undertake.
3. Stylistics and its place

As Short (2006: 4) states, because stylistics predominantly takes the
techniques of linguistics and applies them to literary texts, the stylistic
analysis of English language has always existed on the edges of two
academic worlds: English Literary Studies and Linguistics. As a
consequence, there has always been an issue as to whether stylistics
constitutes a valuable link between language and literary study.
Although a minority see stylistics as providing a positive contribution
to literary study, many literary critics do not see the need to have their
intuitive understanding of texts, and their responses to them, explained
or supported by detailed stylistic analysis. They are most interested in
texts themselves, the social contexts in which they were created and
intuitive personal response to them. Similarly, linguists, too, tend to
ignore stylistics. That is to say, it has been neglected by both fields of
study.
After reading all these debates between the critics, we were actually
not sure how to define stylistics and where to put it. Especially after
reading Bradford (1997: i) who starts his book with the daunting
statement that ‘stylistics is an elusive and slippery topic’, my students
expressed their concerns about studying ‘an issue which was not
established as a field of study with set methods and principles in itself,
which had rather evolved from the ancient art of rhetoric through
centuries’ (Student-1: Taner). I, therefore, warned my students against
the definitional conundrum that the term presents. We thought it better
to accept it as an eclectic approach, which is drawn from disciplines
such as linguistics, literary studies and social sciences. Another
student, Ozlem, for instance, was happy that it combines the
80 Nazan Tutas
techniques of many aspects of other theories, which makes it more

practical:
Stylistic analysis appears to be a combination of various
approaches to language and literature. Instead of only analyzing
the formal aspects of a work like formalist criticism, or instead
of analyzing a work from a certain theoretical approach like
Marxist, Freudian or feminist, or referring to reader’s process of
reading to make meaning, this approach combines them all and
gives us a chance to analyze texts by making use of linguistics,
semantics, phonology, graphology, reader’s response technique,
psychoanalysis etc. Therefore, it is much healthier than sticking
to just one approach to make meaning out of a text. (Student-2:
Ozlem)
We also agreed with Wales’s (2001: v) recognition of stylistics as a
‘tool kit’ – an application of linguistic know-how, of linguistic
observation, of descriptively led linguistic insights brought to bear on
the practical analysis of language in literary and non-literary texts.
One student, Taner, for example agrees that
It is an approach to show how the technical linguistic features
of a literary work contribute to overall meaning and effect. It
can be used as a ‘tool’ in helping us reach our aim in
application of other literary theories such as Marksist,
Psychoanalytic, Feminist, etc. (Student-1: Taner)
He also thinks that almost all literary theories, either textual or
contextual, at some point make use of stylistics anyway. Maybe they
do not name it ‘stylistics’ but what they do is to apply techniques from
stylistics:
For example, while making a feminist criticism of a text, it is
also possible to be able to apply stylistic analysis when
necessary. This will help us base our analysis of literary text on
firmer grounds. The most important value stylistics adds into
the field of literary analysis is that it teaches students analytical
thinking and enables to support his thinking with hard data.
(Student-1: Taner)
Another student, Ayca, also agrees with this idea saying that
I think stylistic makes one to be much more confident about

his/her analysis. For, analysing a text by a particular approach
or school of criticism can lead to a distancing from the
dynamics of the text to other issues, to external aspects.
However, if such an analysis is supported with a stylistic
analysis it will be even more fruitful. (Student-3: Ayca)
Therefore, we agreed that stylistics is a critical approach rather than a
critical theory. It applies ‘practical work distinctly different in tone
and method’ from what they are accustomed to (Barry, 1995: 95). It
constitutes a valuable link between language and literature study as it
takes the techniques of language analysis and applies them to literary
texts. Here are some of the definitions the students came up with:
It is a fundamental link between literature and linguistics, or
rather between art and science. Although it is possible to apply
stylistic analysis to any text, it is most popular with literary
texts. I now believe that through stylistics, it is possible to make
‘real’ scientific comments on literary texts, rather than merely
speculative ones. It changed my habit of analysing texts.
(Student-4: Berkan)
Stylistic analysis seems to be a natural way of approaching to a
text, as you are engaged with the inner dynamics, the very
building stones of the text, and it justifies that each element,
each word, or construction has a function, shaped by the
writer’s own style. Moreover, a change in the expressions, or
the choice of a different narrative technique would lead to a
different work. (Student-3: Ayca)
As I have understood so far, stylistics has some similarities with
New Criticism, Russian Formalism and Structuralism, which I
had studied previously. The basic similarity lies in the fact that
all focus on text and linguistic patterns in the text. On the other
hand, different from the three literary theories, stylistics
analyses texts of all types both synchronically and
diachronically, and with its concerns of intertextual relations
and authorial style together with text style, it goes beyond the
text as a closed whole. In addition, along with its focus on
linguistic structures, it requires and makes use of intuition and
background knowledge to some extent. (Student-5: Isık)
82 Nazan Tutas
4. Benefits of stylistics in the students’ future career

During the evaluation of the course, the general feelings of the
students were positive. They stated that they had learnt to look at
literature with different eyes. They see stylistics as providing a
positive contribution to literary study. They feel secure when their
intuitive understandings of texts and their responses to these texts are
supported by detailed stylistic analysis. They said they always felt that
their interpretations and argumentation needed detailed linguistic
support. Studying stylistics inspired two of the students to present
papers at two different conferences. One student, Ozlem, is preparing
a paper for the METU- British Novelists Seminar on the stylistic
analysis of Jeanette Winterson’s ‘Sexing the Cherry’. It especially
encouraged another student, Berkan, to formulise - what he calls a
‘holistic approach’:
It helped me to formulise the ‘Holistic’ approach, a three-
stepped stylistic analysis that also includes the author and the
reader. It is, basically, a combination of stylistics, Reader
Response criticism, New Historicism, and the biographical data
about the author. It seeks to find the possible connections
between the author, text, reader, and the inner and outer
contexts of texts. It also offers new terms: coretext (the
stylistics features of the author in a given text that can be traced
and analysed) and supertext (the contextual dynamics of a text).
The Holistic approach was first presented during the class.
Later on, an introductory paper (titled ‘It has a Style: A Short
Holistic Analysis of George Herbert’s “The Collar”’) was
presented in the First International IDEA Conference: Studies
in English (24-26 April 2006, Istanbul-Bogazici University).
(Student-4: Berkan)
Student-3, Ayca, said that at the beginning she thought stylistics was
just another literary theory that can only be applied to certain texts. As
the course proceeded, it became clear for her that stylistics has a much
broader field of application than any other literary theory and
therefore it can be employed for various purposes by students of
literature. Here is a brief account of the benefits of stylistics according
to the students.
Before taking this course, I had been familiar with structuralism

and I think there are many aspects common to a stylistic and a
structural analysis. Before, I was analysing texts on the
contextual level. Yet, stylistic analysis helped me justify my
comments on the level of language, or surface level. I’ve always
conceived of form and content as inseparable, and the benefit of
a stylistic analysis, I think, is that it provides a whole
interpretation of a text, as it relies on both formal and
contextual features, and being an almost statistical analysis.
(Student-3: Ayca)
As the results drawn from a textual analysis are supported with
linguistic properties of the text, students of literature, just like
those of science, express their findings and opinions with
concrete evidences. As a result, they become more self-
confident and are motivated to analyse more challenging texts.
(Student-1: Taner)
The outcome of starting to learn stylistics is that, my
acquaintance with poetry improved due to this course. It will
also be useful for me while preparing my academic papers.
(Student-2: Ozlem)
I must also refer to the advantage of this course in terms of its
contribution to my understanding, recognition, and
interpretation of style in non-literary field, particularly in
advertisements. (Student-3: Ayca)
Almost all of the students stated that they want to make use of their
stylistics knowledge in their doctoral thesis. It was particularly
influential on the following students:
I want to make use of Stylistics in my Ph.D. dissertation as
well. Supporting one’s intuitive conclusions with hard data is a
common problem in literary studies. Based on the scientific
objectivity of Linguistics, Stylistics provides hard data
necessary to conduct literary studies. Moreover, employing
Stylistics along with some contextual literary criticisms, such as
Feminism, New Historicism and Postcolonial theory will not
only increase the objectivity of the study but also bestow it an
interdisciplinary scope, which is regarded as an important
84 Nazan Tutas
criteria particularly for comprehensive studies like Ph.D.

dissertations. (Student-1: Taner)
The course was so influential that I even thought about writing
my PhD thesis on/within stylistics for a time though my
attention turned to ‘ekphrastic poetry.’ However, I think that I
will make use of stylistics to a great extent in my PhD studies.
Stylistics analysis, in my opinion, is best applicable to poetry
and I plan to make stylistics analyses of the poems I will study
when I look for the artistic, textual, and contextual connections
between poetry and painting (of course I will touch upon it in
my Introduction). Last of all, I would like to state that it
changed all our conceptions in reading and analysing texts.
Now, a different world of literature stands on our way and I
want to go for it with that stylistics tool-kit in my hand.
(Student-4: Berkan)
The advantages of stylistic knowledge, they believed, will be
especially influential in their teaching:
As a lecturer of English at Ankara University, I am planning to
use basic stylistic analysis in reading classes. I believe that
stylistic analysis will enhance students’ reading skills and make
classes more enjoyable as it requires students’ active
participation. Students will be assigned to make stylistic
analysis of a given text either in groups or as individuals.
Analysing the writer’s word-choice, use of active/passive
structures, formal/informal patterns and so forth, students will
become aware of how language functions and how it can be
employed in different manners to create different effects on the
reader. Surely, such analyses will eventually lead students to
employ similar strategies in their own writing papers and
presentations. I will make use of Short’s web-based course
devoted to Stylistics to an optimum level. The exercises
presented with animations and videos on the Internet will help
students understand the elements of stylistic analysis. (Student-
1: Taner)
Learning stylistic analysis gives me much confidence as an
instructor of literature and literary translation-interpretation. I
am planning to employ it during my courses on both literature
and translation, because both areas require a close analysis of

texts. (Student-2: Ozlem)
Stylistics course has been both a challenge for me and a great
influence and aid in my studies in literature. It has especially
been useful in my teaching experience. In the undergraduate
course I’ve been teaching for four years called ‘Textual
Analysis’, I am concentrating on the analysis of literary texts,
mainly of drama, short story, and the essay. We are rather
concerned with the contextual level in the texts rather than the
formal aspects. Yet, as it is an introductory course aiming to
make the students familiar with the elements of fiction, style is
one of the elements we deal with. I must confess the difficulty
of teaching style. I encourage my students to see the unity of a
text, of its form and content, by dealing with the language, and
narration, and other aspects of style. In addition, they were
more comfortable when they find that their comments about the
text can be justified through the style. That style is a key to the
meaning and essence of a text is what I emphasize to my
students. (Student-3: Ayca)
5. Difficulties and concerns

As can be seen from the above responses, stylistics appears to be
potentially beneficial for our students. However, there were some
concerns as well. One of them is related to the terminology. Since
stylistics includes linguistic study as well, it uses specialised technical
terms and concepts derived from the science of linguistics. Terms like
‘transitivity’, ‘collocation’, ‘cohesion’ ‘under-lexicalization’, and so
on. As literature students, they have not been taught linguistics and
therefore they were unhappy about the way stylisticians use this
terminology. For example, they did not understand why stylisticians
instead of using set terms in literary analysis like ‘ambiguity’, ‘irony’,
‘paradox’ ‘metaphor’ prefer ‘hybrid meaning’, ‘deviation’, ‘cohesion’
and so on just for the sake of being different. One student says:
Through the course, I learned that the stages one follows in a
textual analysis are in fact parts of Stylistic analysis identified
with specific terminology. In other words, when dealing with
texts, students actually make stylistic analysis without knowing
it. When they learn how to make a complete Stylistics analysis
86 Nazan Tutas
and familiarise with all the necessary terminology, the

arguments they present in their papers or oral presentations gain
a more serious and scientific tone. (Student-1: Taner)
However, he criticises this thinking that this desire for inventing new
terminology or using the terminology of linguistics for already
existing literary terms makes stylistics more obscure.
Another issue the students were confused about was the name of the
approach. The indecisiveness of stylisticians on what to call it
surprised them a lot. The name ‘Stylistics’ came about because the
nature of style and its relation to meaning, and the characterisation of
the style of particular writers was an important consideration for those
who practised it in its early years (Leech and Short 1981: 10-73). But
this name became problematic as the central interest for stylisticians
gradually became the characterisation of textual meaning and effect.
During the 1980s it came to be called the ‘new stylistics’ which had a
limited degree of eclecticism in that it drew on the findings of other
new kinds of criticism such as feminist, structuralist, post-structuralist
and so on. Macleod (2005: 61) thinks that stylistics today might be
called ‘post-stylistics’ and thinks that the best definition of it today is
that it is the linguistically informed study of texts of all kinds. Other
names have been proposed from time to time, the most influential
being ‘Literary Linguistics’, ‘Critical Linguistics’ and ‘Poetics’. But,
as Short (2006) says, each of these names has its own problems in
covering the range of activities involved in the ‘stylistic’ approach,
and none of them have gained widespread acceptance. Short (2006: 4)
refers to stylistics today as ‘modern stylistics’ whose central aim is to
understand how we get from the words on the page of (literary) texts
to our understanding of those texts (how we see its style, the fictional
world and its characters in our mind) and how they affect us. As a
consequence, modern stylistics involves not just linguistic textual
analysis but also an attempt to account for how readers interact with
textual structure, via psychological and pragmatic processing, to infer
meaning etc. from texts. So we decided to continue to use the original
name, ‘Stylistics’, despite its shortcomings, since we agreed with
Short who thinks that no other name covers the area of study any
better.
Student-1, Taner, stated a disadvantage of stylistics. He said he is
worried that stylistics might lead to a tendency on the part of students
to analyse literary works solely on the textual level, ignoring

contextual elements present in the text. However, this can easily be
avoided by employing stylistics, along with other literary theories, as a
means to draw conclusions from literary texts, rather than an ultimate
aim in itself.
Another student, Ayca, expressed her concerns about the difficulty of
stylistic analysis as an EFL student:
I always regarded style as an intimidating and extremely
difficult concept. I had not been much familiar with the analysis
of style. I still think stylistic analysis is a very demanding
attempt, since you have to be an expert on language to penetrate
the style of the text, and it is a handicap for most of us who
learned English as a foreign language. The same difficulty is
valid for me, especially when I was faced with the phonological
analysis. It is more manageable to analyse the structural and
contextual aspects in a text than its lexical (especially phonetic)
and grammatical aspects. (Student-3: Ayca)
Some of them also said that they were sometimes hesitant to express
strong personal views on some literary works. They said that since
they were foreign speakers of the language they usually have the
tendency to refer to the opinions of experts, or critics since they do not
always trust their intuitions and their knowledge of language.
6. Conclusion
The feedback I collected from the students helped me to redesign the
course outline and course materials. For example, when they said that
they were afraid to start stylistic analysis, I prepared some guidelines
which were adapted from Leech and Short (1981), Short (1996) and
Lazar (1993) for poetry, prose and plays to eliminate this anxiety. The
guidelines help to have a starting point in their analysis.
I also had to add some topics in the syllabus. At first, I thought that
giving some linguistic and grammatical information would be
inappropriate for them, as they were English language teachers. I
assumed that as foreign language teachers, they would probably have
a considerable awareness of English grammatical structure and they
would be consciously aware of linguistic structure and equipped to
analyse it. Nevertheless, I saw that they still lacked some information,
88 Nazan Tutas
especially on linguistics, and they felt insecure about it. Therefore, I

decided to include that information in the syllabus.
In conclusion, I can say that my students are not anxious about
stylistics any more. They submitted successful first attempt
assignments. They were glad to learn how to make stylistic analysis
and found it very beneficial and advantageous in their analysis of
literature. In answer to Short’s deviant and foregrounded question,
‘Who is stylistics?’ we can now say, ‘She became a friend of ours’
(1996: 1). Therefore, ‘Who is afraid of Stylistics?’
Endonote
1
http://www.lancs.ac.uk/fass/projects/stylistics/index.htm
References
Barry, P. 1995. Beginning Theory. Manchester: Manchester University Press.
Bradford, R. 1997. Stylistics. London: Routledge.
Culpeper, J. 2001. Language and Characterisation: People in Plays and Other Texts.
London: Longman.
Fish, S. E. 1973. ‘What is stylistics and why are they saying such terrible things about
it?’ in S. Chatman (ed.) Approaches to Poetics: Selected Papers from the English
Institute. New York: Columbia University Press: 109-52.
Gavins, J. and G. Steen (eds). 2003. Cognitive Poetics in Practice. London:
Routledge.
Hamilton, C.A. 2004. Preface. Style 38(4).
Hoover, D. L. 1999. Language and Style in ‘The Inheritors’. Lanham, Maryland:
University Press of America.
Lazar, G. 1993. Literature and Language Teaching. Cambridge: Cambridge
University Press.
Leech, G. N. and M. Short. 1981. Style in Fiction. London: Longman.
Macleod, N. 2005. ‘Stylistics and point of view in fiction: a credo and some
examples’ in The European English Messenger 14(2): 61-73.
Miall, D. S., and D. Kuiken. 2001. ‘Shifting perspectives: Readers’feelings and
literary response’ in Willie van Peer and Seymour Chatman (eds.) New
Perspectives on Narrative Perspective. Albany, NY: State University of New
York Press: 289-301.
van Peer, W. 1986. Stylistics and Psychology: Investigations of Foregrounding.
London: Croom Helm.
Semino, E. 1997. Language and World Creation in Poems and Other Texts. London:
Longman.
Semino, E. and J. Culpeper (eds). 2002. Cognitive Stylistics: Language and Cognition
in Text Analysis. Amsterdam: John Benjamins.
Semino, E. and M. Short. 2004. Corpus Stylistics: A Corpus-based Study of Speech,
Thought and Writing Presentation in Narratives. London: Routledge.
Short, M. 1996. Exploring the Language of Poems, Plays and Prose. London:
Longman.
Short, M. 2006. ‘Designing and piloting a world-wide-web-based stylistics course’ in
Andrea Gerbig and Anja Müller-Wood (eds) Rethinking English: Reconciling
Literature, Linguistics and Cultural Studies (Lampeter: Edwin Mellen).
(Avaliable at http://www.lanc.ac.uk/fass/projects/stylistics/tutors/Paperversions/
Trier paper.doc).
Stockwell, P. 2002. Cognitive Poetics: An Introduction. London: Routledge.
Wales, K. 2001. A Dictionary of Stylistics. 2nd edition, Harlow: Longman.
PART II
COGNITIVE STYLISTICS
Fusion Style: Towards a Poetics of the Grotesque Body1
Shun-liang Chao
University College, London
Abstract
This paper aims to construct a psychoanalytic reading of the grotesque body as a kind of
(avant-garde) poetic language, one that deforms and destabilises the normal process of
signification by its fragmented syntax and bold metaphors. To do so, I combine the
Freudian dream-work, the Lacanian real, the Barthesian writerly, and the Kristevan
semiotic to contend that the grotesque body, as a fusion of different or discordant
objects, embodies an articulation of unconscious drives through which (avant-garde)
poetic language carries out its major function: to foster the polysemic function of the
signifying practice. By reading the grotesque body as poetic language, I also seek to
develop as a possible index to verbal grotesques the concept of the flesh-made
metaphor, one that contains the metamorphic images of human, animal, and/or vegetal
bodies. The grotesque occurs when the components of a flesh-made metaphor are only
weakly functionally or characteristically similar so much so as to interrupt the cognitive
process of figuratively apprehending the metaphor, and thus the mind’s eye cannot help
but dwell on its literal level where visually incongruous images are born.
Keywords: grotesque; metaphor; metonymy; metamorphosis; Barthes; the Freudian
dream-work; the Lacanian real; the Kristevan semiotic.
1. Introduction
There Scylla came; she waded into the water,
Waist-deep, and suddenly saw her loins disfigured
With barking monsters, and at first she could not
Believe that these were parts of her own body.
She tried to drive them off, the barking creatures,
And flees in panic, but what she runs away from
92 Shun-liang Chao
She still takes with her; feeling for her thighs,

Her legs, her feet, she finds, in all these parts,
The heads of dogs, jaws gaping wide, and hellish.
She stands on dogs gone mad, and loins and belly
Are circled by those monstrous forms.
– Ovid (1955: 340), Metamorphoses, XIV
The term ‘grotesque’ was coined in late fifteenth-century Italy to refer
to a style of painting and ornamentation that offered images composed
of human and animal forms, fantastically interwoven with fruits,
flowers, and foliage. Fusing contraries into a hybrid or collage, the
grotesque gives birth to, as Bakhtin (1984: 24) puts it, ‘a phenomenon
in transformation, an as yet unfinished metamorphosis, of death and
birth, growth and becoming’. It is proper to say that the very nature of
the grotesque lies in its perpetual, never-ending metamorphosis of one
substance into another. In other words, the complete transformation of,
say, Daphne into a laurel tree is not grotesque per se, whereas Scylla in
Ovid’s passage above is no doubt grotesque – insofar as she exists as a
cross between a human being and ‘barking monsters’. In Daphne, the
conflict between a human and a plant no longer exists, since her
metamorphosis is finished. The tension between bodily forms of two
different kinds, however, is taking place in Scylla, who is both a human
being and dogs. Indeed, Scylla ‘is experiencing the grotesque, suffering
the logically impossible though undeniable recognition that “her loins”
are also “dogs gone mad” (italics mine)’ (Harpham 1982: 16-17). Then,
as long as her metamorphosis remains incomplete, her body –
composed of both human and animal forms – is grotesque or
grotesquely monstrous. Scylla’s body is the very prototype of the
grotesque image in which the pattern of ‘either-or’ is replaced with that
of ‘both-and’.
This model of ‘both-and’, or logical impossibility, features the Freudian
dream-work (Traumarbeit) of ‘condensation’ (Verdichtung), whose
name, as Jacques Lacan (1977: 60) has noted, ‘condensing in itself the
word Dichtung [poetry; composition; versification], shows how the
mechanism is connatural with poetry to the point that it envelops the
traditional function [i.e. metaphor] proper to poetry’.2 With this in mind,
I propose to argue that the grotesque body, as a product of
dream-condensation, is the pictorial representation of the poetry or
poetic language that, critics such as Kristeva have observed, brings to
Fusion Style 93
the fore the polysemic function of the signifying practice, i.e. that
which standard language tends to repress. The grotesque body, we shall
see, deforms and destabilises the normal functioning of standard
language by fragmenting its body, its ‘syntax’, and foregrounding its
metaphorical dimensions – or rather, the primary process of the
Freudian dream-work. In this respect, the grotesque body functions like
avant-garde poetic language whose syntax is disconnected and whose
metaphors are audacious and ambiguous. By linking grotesque bodies
to poetic language and metaphor, I also aim to bring forth a notion of
the flesh-made metaphor as an index to verbal grotesque imagery.
My theoretical approach is drawn from psychoanalysis: I seek to tease
out a psychoanalytic reading of the grotesque body to illustrate a kind
of avant-garde poetic language that diversifies or pluralizes signifying
practices. To do so, I shall first draw on Lacan’s idea of the fragmented
body to make the point that the grotesque body or bodily experience
marks the return of the real. I shall then go on to weave together the
Lacanian real, the Barthesian writerly, and the Kristevan semiotic to
argue that the grotesque body produces a text or poetic language that
unsettles the logical functioning of the linguistic system. This will bring
us to the ways in which the grotesque body semioticizes itself as an
avant-garde poetic language and as a bold (flesh-made) metaphor. Last
of all, I shall attempt to construct a definition of verbal grotesque
metaphor, verbal metaphor that presents the imagery of the grotesque
body.
2. The Return of the Real

Freud (1991c: 105-06) describes the id, ‘the dark and inaccessible part’
of the mind, as ‘a chaos, a cauldron full of seething excitations’: it is
filled with concurrent contradictory drives or impulses, and as such, the
governing rules of logic do not apply there. The id, so to speak, is the
very kernel of the unconscious, ‘the Realm of the Illogical’ (Freud 1993:
401). Since the id or the unconscious is hard to access, it only becomes
visible under conditions of dreaming, of neurosis, of parapraxes, of
jokes, and so forth. Among these conditions, dreams are considered to
be the royal road to the unconscious, primarily because the dream state,
like the unconscious system, is characterised by the exemption from
mutual contradiction, that is, by the logic of ‘both-and’. As Freud
(1991a: 429) explains in The Interpretation of Dreams of 1900,
94 Shun-liang Chao
The way in which dreams treat the category of contraries and

contradictions is highly remarkable. It is simply disregarded.
‘No’ seems not to exist so far as dreams are concerned. They
show a particular preference for combining contraries into a
unity or for representing them as one and the same thing. Dreams
feel themselves at liberty, moreover, to represent any element by
its wishful contrary (italics mine).
Here Freud refers to two types of images created by the so-called
dream-condensation in contrast to dream-displacement (Verschiebung),
which transfers the reference of one object onto another alluding to it,
thereby eliding or disguising the original object (1991c: 208).
Composite images of condensation fuse seemingly dissonant figures
that yet ‘have something [latent] in common’ into a unity, a
photomontage, with parts of the original figures being discernible;
collective images present only one figure as the representative of the
others in absentia whilst they all share some attributes (Freud 1991b:
205-06; 1991a: 431-34).3 Simply put, having something in common is
the prerequisite for the dream-work of condensation.
Clearly, composite images are more pertinent to the study of the
grotesque. Indeed, Freud goes on to liken them to fantastic figures such
as centaurs or fabulous beasts in Arnold Böcklin’s paintings: ‘The
psychical process of constructing composite images in dreams is
evidently the same as when we imagine or portray a centaur or a dragon
in waking life’ (1991a: 436; 1991b: 206). Lacan (1977: 11; 2-5) links
the psychically charged composite images to the ‘imagos of the
fragmented body’, which oftentimes appear in dreams ‘in the form of
disjointed limbs, or of those organs represented in exoscopy, growing
wings and taking up arms for intestinal persecutions—the very same
that the visionary Hieronymus Bosch has fixed, for all time, in painting,
in their ascent from the fifteenth century to the imaginary zenith of
modern man’. In Lacan (4; see also Evans 1996: 67), ‘the fragmented
body’ (le corps morcelé) is the matrix of any sense of fragmentation and
disintegration; it refers to the fact that the infant, due to his/her sensory
and ‘motor un-coordination’, experiences his/her body as piecemeal or
shapeless before s/he (mis)identifies with the unified (and yet alienated)
image of his/her own body in the mirror. Hence, the birth of the
narcissistic ‘ideal ego’ (moi idéal), the function of which is ‘one of
mis-recognition [méconnaissance]; of refusing to accept the truth of
Fusion Style 95
fragmentation and alienation’ (Homer 2005: 25; see also Laplanche &
Pontalis 1973: 250-52); and of establishing a unified consciousness.
Paradoxically, the méconnaissance, or the illusion of totality, ‘in which
a human being is always looking forward to self-mastery, entails a
constant danger of sliding back again into the chaos from which he
started; it hangs over the abyss of dizzy Ascent in which one can
perhaps see the very essence of Anxiety’,4 says Lacan (1953: 15). Put
another way, from the mirror phase onwards, the infant, having
internalized the ideal ego, will continue to identify with his/her images
of wholeness as a promise of ‘self-mastery’ throughout his/her life –
even when s/he enters the symbolic order of language and then learns to
face up to the inevitable anxiety of fragmentation or incompleteness. At
the same time, s/he will continue to be haunted and tormented by the
surviving memory of the fragmented body, which usually crops up in
fantasies and dreams of body parts being dislocated, devoured, or
distorted (Lacan 1977: 11-12). In other words, the fear or anxiety of
‘sliding back again into the chaos’ will remain present as long as the
ego carries with it the desire for self-mastery or completeness resulting
from the idealized illusion of unity in the ‘mirror phase’ (le stade du
miroir) or the imaginary order.
Seen in this light, Ovid’s Scylla, one can say, is being struck with fear or
anxiety arising from the return of and to the fragmented body and motor
incapacity. She anticipates seeing in the river the integrated image of
her body, only to find that her body parts are being ‘disfigured’ or
trans-formed into barking monsters. She then ‘flees in panic’ and yet
cannot master her own bodily movements: ‘what she runs away from /
She still takes with her’. Her loss of corporeal or formal integration and
integrity is, then, the lapse into the chaotic state preceding the mirror
phase, the state that Lacan has compared to Bosch’s paintings of
deformed creatures or grotesque bodies. They present a radically untidy,
incoherent, and undifferentiated world, in contrast to the imaginary
world of perfectly defined objects implied in the unified ego or
consciousness. The grotesque body or monster thus serves as a
‘dehiscence’ (in Lacan’s terms; 1977: 4)5 of the unified consciousness;
it opens onto ‘vertiginous new perspectives characterized by the
destruction of logic and regression to the unconscious – madness,
hysteria, or nightmare’ (Harpham 1976: 462); it shows that which, as R.
Grant Williams (2001: 605-06) observes,
96 Shun-liang Chao
subjectivity in its attempts to master identity perpetually strives

to repress. Constituting the return of the real, the monster is the
ineffable residue of the symbolic and imaginary orders, that is,
the subjective realms of language and projection. It is only the
imaginary gestalt of the body that psychically protects the
subject from slipping into an awareness of his or her fundamental
monstrosity.
It is fair to say, then, that the grotesque body lays bare the chaotic,
turbulent nature of the real experience, a ‘fundamental monstrosity’, by
peeling off the illusory gestalt veneer of rational unity created by the
imaginary projection and sustained by the symbolic order of language.
The grotesque thus marks ‘the return of the real’ rather than of the
imaginary, as indicated by Rosemary Jackson (1981: 90) in her
psychoanalytic reading of fantastic arts (including the grotesque):
Fantasies try to reverse or rupture the process of ego formation
which took place during the mirror stage, i.e. they attempt to
re-enter the imaginary. Dualism and dismemberment are
symptoms of this desire for the imaginary. . . . A fantasy of
physical fragmentation corresponds, then, to a breakdown of
rational unity. That linguistic order which creates and constitutes
a whole self, a total body, is un-done.
Jackson is certainly right in explicating the disruptive or revolutionary
power of fantastic arts. Nevertheless, she seems to confuse the
imaginary with the real (Jackson 1981: 91): she goes on to underscore
that fantastic arts ‘express a desire for the imaginary, for that which has
not yet been caught and confined by a symbolic order’. In Lacan, the
very basis of the imaginary order is the mirror phase, during which, as
we have seen, the (ideal) ego is constructed by the (mis)identification
with the visual gestalt of the body in the mirror. ‘It is this identification
and unification of a self as a self’s image’, Allon White (1993: 76; 64)
observes, ‘which is important for the generation of a unified
consciousness capable of producing speech’ as well as ‘reason through
the propositional structures embedded in syntactic order’. Accordingly,
what fantastic arts seek to ‘reverse or rupture’ – to wit, unified
consciousness and its accompaniments – is in fact the very product of
the imaginary.
That is to say, fantastic arts seek to re-enter or desire the real, which is
Fusion Style 97
characteristic of physical dismemberment and resistance to

symbolization – the phenomena Jackson mistakenly ascribes to the
imaginary. The real, as Lacan (qtd. in Evans 1996: 159) has put it, is
‘that which resists symbolization absolutely’; is ‘the domain of
whatever subsists outside symbolisation’. Though in principle
un-symbolizable or un-presentable, the real, Bruce Fink (1995: 24)
suggests, can be conceived of as ‘an infant’s body “before” it comes
under the sway of the symbolic order’ of socialization, as ‘a time before
words, to some sort of presymbolic or prelinguistic moment in the
development of homo sapiens or in our own individual development’.
Or if the subject in the symbolic is structured by syntactico-semantic
coherence, by the proper use of language, to produce meanings, then
‘parcelled-out, broken-up, separated pieces of body, language, thought
comprise the subject in the real’ (Ragland 1993: 82). It is tempting to
say, then, that E. H. Gombrich (1979: 256) is in fact describing the
experience of the real when elaborating on the viewer’s response to
grotesque hybrids:
[T]he reaction of exasperated helplessness [is] provoked by
hybrid creatures, part plant, part human; part woman, part fish;
part horse, part goat. [For] [t]here are no names in our language,
no categories in our thought, to come to grips with this elusive
dream-imagery in which ‘all things are mixed’. It outrages both
our ‘sense of order’ and our search for meaning.
In (the form of) the grotesque hybrid, then, occurs the real: Lacan (1988:
64) regards Medusa’s grotesque head as an example of ‘the apparition
of the terrifying anxiety-provoking image’ that reveals the real, namely,
what ‘properly speaking is unnameable’ and what is ‘the essential
object which isn’t an object any longer, but this something faced with
which all words cease and all categories fail’. In a word, the grotesque
hybrid presents or, one may prefer, embodies the un-nameable and
un-classifiable nature of the real. ‘Lire’, says Roland Barthes (1994b:
562) in S/Z, ‘c’est trouver des sens, et trouver des sens, c’est les
nommer’ (‘To read is to find meanings; to find meanings is to name
them’). But the grotesque hybrid, with its body composed of
heterogeneous bits and pieces, fractures the orderly use of language as a
means of conveying meaning, thereby returning the viewer/reader to
the pre-verbal or pre-symbolic stage of the inability to name things.
98 Shun-liang Chao
3. ‘Le scriptible’
The transgression of denomination or categorization is a defining trait
of Barthes’s notion of ‘le Texte’ (‘the Text’), as opposed to ‘l’œuvre’
(‘the work’). Their opposition, Barthes (1994c: 1212) exemplifies, can
be compared to ‘la distinction proposée par Lacan: la « réalité » se
montre, le « réel » se démontre’ (‘Lacan’s distinction: ‘reality’ is
displayed; ‘the real’ is revealed’). By this, Barthes (1994c: 1212)
suggests that the work is a palpable and classifiable existence as
displayed in bookstores or libraries; the Text, however, resists being
properly designated or classified or symbolized (in Lacan’s sense) and
‘ne s’éprouve [ou se démontre] que dans un travail, une production’ (‘is
experienced [or revealed] only in a labour of production’). For the Text
is ‘un espace à dimensions multiples, où se marient et se contestent des
écritures variées, dont aucune n’est originelle’ (1994b: 493-94) (‘a
multi-dimensional space, wherein various writings, none of which is
original, mingle and collide’). The Text, so to speak, is irreducible to a
closed meaning, a proper name, a transcendental signified. The work is
‘readerly’ (‘lisible’) because it closes on a signified and therefore
plunges the reader into ‘une sorte d’oisiveté’ (‘a kind of idleness’). By
contrast, the Text is ‘writerly’ (‘scriptible’), insofar as it demands the
reader to ‘apprécier de quel pluriel il est fait’ (1994b: 558-59)
(‘appreciate what plural it is made of’) and to co-produce the plurality
of signification:
Dans ce texte idéal, les réseaux sont multiples et jouent entre eux,
sans qu’aucun puisse coiffer les autres; ce texte est une galaxie
de signifiants, non une structure de signifiés; il n’a pas de
commencement; il est réversible; on y accède par plusieurs
entrées dont aucune ne peut être à coup sûr déclarée principale;
les codes qu’il mobilise se profilent à perte de vue, ils sont
indécidables (le sens n’y est jamais soumis à un principe de
décision, sinon par coup de dés).6
It is proper to say, then, that grotesque hybrids are the writerly text par
excellence: for ‘we cannot even tell’, as Gombrich (1979: 256) writes of
the grotesque, ‘where they begin or end – they are not individuals,
because their bodies merge or join with those plants and tendrils…
Thus, there is nothing to hold on to, nothing fixed, the deformitas is
hard to “code”’.
It is because the grotesque hybrid is difficult to ‘code’ or ‘name’,
Fusion Style 99
Barthes would argue, that the reader/viewer is required to practically

collaborate in re-writing or re-producing it. Therein lies the jouissance
of the text: the grotesque hybrid belongs to the kind of text that, Barthes
(1994d: 1502) reckons, has ‘un corps de jouissance fait uniquement de
relations érotiques’ (‘a body of jouissance made solely of erotic
relations’). To put this in Freud’s terms (1991b: 246), the grotesque
hybrid creates a ‘polymorphously perverse’ body, in which various
body parts function as erogenous zones – none of which predominates –
and create an interplay. In short, the grotesque body, per se, is a floating
signifier, a text of polysemic nature that breaks through or tears open
the determinate aspects of the signifying process and as such, to quote
Julia Kristeva (1984: 62) on the semiotic, ‘brings about all the various
transformations of the signifying practice that are called “creation”’.
Kristeva divides language, or rather, the signifying process that forms
language, into two modes: the semiotic (le sémiotique) and the symbolic
(le symbolique). The former is governed by drive energies and their
indeterminate articulations that ‘[do] not yet refer (for young children)
or no longer refers (in psychotic discourse) to a signified object for a
thetic consciousness’ (Kristeva 1980: 133), i.e. to sense. The semiotic
chronologically precedes the mirror phase, with which comes the
‘thetic’ phase, the one that posits signification, or the signified, and acts
as ‘the “deepest structure” of the possibility of enunciation’ (Kristeva
1984: 62; 43-44). The thetic, so to speak, is that which renders any
signifying practices possible and thus serves as the entrance into the
symbolic, which is the ‘inevitable attribute of meaning, sign, and the
signified object for the [thetic] consciousness’ necessary for
communication proper (Kristeva 1980: 134). In a nutshell, ‘[t]he
semiotic could be seen as the modes of expression that originate in the
unconscious whereas the symbolic could be seen as the conscious way
a person tries to express using a stable sign system (whether written,
spoken or gestured with sign language)’ (McAfee 2004: 17).
Notably, in Kristeva (1980: 134), the semiotic and the symbolic are not
mutually exclusive but rather dialectically and dynamically ‘combined
in different ways to constitute types of discourse, types of signifying
practices’. Scientific or rational discourse, for instance, is dominated by
the symbolic component in order to convey meaning as unequivocally
and determinately as possible. On the other hand, in the discourse of
art – for example, poetry – the semiotic component transgresses the
100 Shun-liang Chao
symbolic function or ‘unsettle[s] the position of the signified’, by

violating certain grammatical rules or syntactic coherence; by
introducing or reactivating the stream of semiotic drives, the ‘deluge of
signifiers’ (Kristeva 1980: 134-35; 1984: 79).7 Poetry, especially
avant-garde poetry, thereby develops to the full the unstable function of
the signifying system. In this way, ‘poetry – more precisely, poetic
language – reminds us of its eternal function: to introduce through the
symbolic that which works on, moves through, and threatens it. The
theory of the unconscious seeks the very thing that poetic language
practices within and against the [symbolic] order’ (Kristeva 1984: 81).
The semiotic component of (avant-garde) poetic language, so to say,
‘can be seen as an articulation of unconscious processes’ or drives that
disturb or destroy the ossified structure of the symbolic signification
and thereby ‘operate a return or, in Kristeva’s language, a revolution
from [the symbolic] to [the semiotic]’ (Smith 1998: 16; 21). As a result,
the function of (avant-garde) poetic language, says Kristeva (1980:
138), is to retrieve that which the symbolic tends to reduce, or really,
repress into the unconscious: ‘“signifier,” “primary processes,”
displacement and condensation, metaphor and metonymy, rhetorical
figures – [that] which always remains subordinate – subjacent to the
principal function of naming-predicating’. Briefly put, (avant-garde)
poetic language manifests the return of the repressed in the unconscious
in the form of the semiotic or the Freudian dream-work or the Lacanian
real.8
Then, the grotesque image, as a ‘creation’ of the (illogical) dream-work,
serves to be a pictorial presentation of poetic language in its greatly
disruptive form by mixing diverse sorts of body parts into a locus of the
displacement and condensation of semiotic fragments, into a
‘semioticizing body, heterogeneous to signification’ (in Kristeva’s
words) (1980: 139). Here is where we turn towards ways in which the
grotesque body deforms the normal process of signification, or rather,
semioticizes itself as poetic language.
4. The Semiotization of the Symbolic

To better apprehend and appreciate the semiotic utilization of language
by and in the grotesque body, it is necessary to obtain a sense of the
normal or conventional operation of language. Of special significance
to our concern is Saussure’s concept of syntagm and association (or
Fusion Style 101
paradigm), ‘two forms of our mental activity, both indispensable to the

life of language’ (1959: 123). Whilst, according to Saussure (1959:
123), different signs in the former ‘are arranged in sequence on the
chain of speaking [or writing]’, a sign in the latter ‘will unconsciously
call to mind a host of other [signs]’ which ‘have something in common’
with it. To use a non-verbal example: syntagmatic is the co-ordination
of different body parts (the head, the trunk, and the limbs) in praesentia
to form a human body as a whole or corpus; on the other hand,
paradigmatic is the ‘[m]ental association’ (Saussure 1959: 125) of, say,
a human head with other signs in absentia – e.g., a gorilla’s or an
ostrich’s head or, more remotely, a sunflower. The former comprises an
inter-locking grammatical chain or context; the latter suggests
associations or substitutions of signs and ‘discards [those] that becloud
the intelligibility of discourse’ (Saussure 1959: 127, n. 10). Together,
they provide a stable, coherent structure in which words or discourse
makes sense.
Nevertheless poetic language (of the grotesque body) tampers with the
proper or Saussurean structure of language. This structure may break
down in various ways, among which Jakobson’s two types of aphasia
have become a vital guide to the literary or artistic disturbance of
language: the similarity disorder and the contiguity disorder. Aphasics
of the first type are confined to the syntagmatic connection or
contiguity of the whole to the part; therefore, they are only able to speak
about one object with reference to its parts or the context it belongs to
rather than by means of synonyms, antonyms, or paraphrases existing
in the paradigmatic axis (Jakobson 1987b: 100-06).9 They will, for
instance, refer to a human body as walking on two legs (as the body and
its legs are physically or spatially contiguous). In Jakobson (Jakobson
1987: 109), the pattern of the similarity disorder matches with that of
metonymy, which depends on the transfer of reference between, say, a
human body as a whole and its various parts.
On the other hand, aphasics of the second type are restricted to the
paradigmatic substitution of words that are semantically (or
characteristically) similar, i.e., the process of naming, defining,
paraphrasing, etc. In other words, they are unable to organize words to
form a grammatical context or syntactic coherence; as such, sentences
fall apart into agrammatical heaps of words: ‘Word order becomes
chaotic; the ties of grammatical coordination and subordination . . . are
102 Shun-liang Chao
dissolved’ (Jakobson 1987: 106). Here we encounter a verbal version of

Lacan’s images of the fragmented body (stemming from the child’s
motor un-coordination) or grotesque images in general. A human trunk
may be combined with an ostrich’s head (substituting a human head) on
its top and with a gorilla’s limbs (substituting human limbs); this body
is thus with no unified form, a deformitas. Markedly, the pattern of the
contiguity disorder, Jakobson (Jakobson 1987: 109) points out,
coincides with that of metaphor, which entails a transfer of sense
between, say, a human head and an ostrich’s head. Metaphor, he
(Jakobson 1987: 111) continues, ‘dominates’ in (Romantic/Symbolist)
poetry and Surrealist painting; yet metonymy in (Realist) prose and
Cubist painting.10 It is tempting to say, then, that the language of art and
literature serves to be the aesthetic performance of aphasia or, to quote
Jan MukaĜovský (1964: 18), ‘the esthetically intentional distortion of
the linguistic components’.
For MukaĜovský (1964: 19), literature is possible only when the norm
of the standard language is violated – viz., the act of ‘foregrounding
[aktualisace]’ – and poetry, or rather, poetic language, is ‘the maximum
of foregrounding of the utterance’. Foregrounding, Kristeva (1984: 79)
would agree, is an example of the ‘semiotization of the symbolic’. The
standard language (or the symbolic mode of signification) renders our
consciousness or stimuli automatized, so the function of foregrounding
is to use linguistic devices, says Bohuslav Havránek (1964: 10), ‘in
such a way that this use itself attracts attention and is perceived as
uncommon, as deprived of automatization, as deautomatized, such as a
live poetic metaphor’. In a similar vein, Paul de Man (1986: 14) bases
the distinction between literary and non-literary language on the fact
that literary language ‘foregrounds’ what he calls ‘the rhetorical
dimension of discourse’ – i.e. figures of speech such as metaphor – to
undermine or destabilize its grammatical and logical function.11
Metaphor, especially bold metaphor, is indeed a royal road to poetic
foregrounding or semiotization, inasmuch as metaphor has been the
signature of poetry or poetic novelty at least since Aristotle (1984b:
1495a, 255) in the Poetics lauds it as ‘a sign of genius’; that poetic
language, P. B. Shelley (2002: 512) declared, ‘is vitally metaphorical’,
for ‘it marks the before unapprehended relations of things’; that
‘[m]etaphor’, according to Donald Davidson (1979: 29), ‘is the
dreamwork of language’; and so forth.
Fusion Style 103
Davidson’s concept of metaphor recalls Lacan’s correlation of

metaphor with dream-condensation. Taking a cue from Saussure and
Jakobson, Lacan (1977: 161) understands the primary process of
displacement and condensation as that of metonymy and metaphor and
thus re-writes the Freudian dream-work: ‘the dream-work follows the
laws of the signifier’. For displacement and condensation are in fact
‘two “sides” of the effect of the signifier on the signified’ (Lacan 1977:
160). In Lacan (Lacan 1977: 156), the displacement of, say, a human
body by its head – which does not involve signification – ‘is nowhere
but in the signifier, and . . . it is in the word-to-word
[signifier-to-signifier] connexion that metonymy is based’. By contrast,
condensation is based on the substitution of one signifier for another in
which the poetic spark of metaphor ‘flashes between two signifiers[,]
one of which has taken the place of the other in the signifying chain, the
occulted signifier remaining present through its (metonymic)
connexion with the rest of the chain’ (Lacan 1977: 157). It is from
metaphor that emerges signification (Lacan 1977: 164) or, as I. A.
Richards (1936: 94) writes of metaphor, ‘a borrowing and intercourse
of thoughts, a transaction between contexts’. What is implied in
Richards’s concept of metaphor is that two signifiers or terms in a
metaphor are both metonymies for the contexts they belong to. This
manifests itself in most grotesque bodies. For instance, in the case of an
ostrich’s head being placed on top of a human body, the substituted
human head, ‘the occulted signifier’, remains visible in the mind’s eye
through its invisible metonymic link (or reference) to the human body;
at the same time, the ostrich’s head also metonymically alludes to its
context, an ostrich as a corpus. In this grotesque body, then, the shift or
transaction of sense or signified occurs not only between a human head
and an ostrich’s head, but, we shall see, between their contexts, i.e.
human beings or bodies and ostriches or birds.
Attention should be paid to the fact that the grotesque body as a type of
discourse proceeds like modern poetic language, which is fragmented
and thus full of lexical lacunas. The grotesque body consists of diverse
body parts which are so mysteriously juxtaposed as to render the body
full of imaginative gaps. To put it in Kristeva’s terms, semiotic
fragments are arranged one alongside the other without obvious logical
sequence, which leads to the breakdown in the signifying chain. The
absence of causal connections is characteristic of modern poetic
language, insofar as it ‘does not proceed discursively, in unison with the
104 Shun-liang Chao
laws of language’ but jettisons ‘syntactical sequence’, writes Joseph

Frank (1991: 10, 14), ‘for a structure depending on the [simultaneous]
perception of relationships between disconnected word-groups’.
Simply put, modern poetry presents its material in the manner of
collage, montage, or jump cut (as in the case of Apollinaire, cummings,
Eliot, etc.). Take Apollinaire’s lines as an example (1956: 363):
Ta langue
Le poisson rouge dans le bocal
De ta voix (ll. 7-9)
(‘Your tongue / The goldfish in the bowl / Of your voice’)
Obvious discursive relations do not exist between these lines; instead,
they proceed like montage, in which several images are perplexingly
juxtaposed: what is the relationship between ‘your tongue’ and ‘the
goldfish’, and between ‘the bowl’ and ‘of your voice’? The latter seems
to be possessive; if so, what does ‘the bowl of your voice’ refer to?
These lines are arranged as a heap of broken images. With the loss of ‘la
nature relationnelle’ (‘the relational nature’) of language, modern
poetry, as Barthes (1993a: 163, 165) asserts, divests itself of ‘la nature
spontanément fonctionnelle du langage’ (‘the spontaneously functional
nature of language’) and reveals ‘la splendeur et la fraîcheur d’un
langage rêvé’ (‘the splendour and freshness of a dream language’). The
‘language’ of the grotesque body, then, is akin to that of modern poetry:
the inter-locking syntagmatic bonds – which constitute an utterance –
between body parts are shattered; miscellaneous body parts are so
esoterically combined that it requires a great deal of imaginative work
to fill in the gaps between them or, Barthes (1994c: 1216) would say, to
‘play with the text’: ‘le texte lui-meme joue. . . ; et le lecteur joue, lui,
deux fois’ (‘the text itself plays. . . ; and the reader plays twice’).
To fill in imaginative gaps in the grotesque body is to restore or disclose
the connections that have been removed or repressed in the signifying
chain. This is where one’s imagination exerts itself most to set off
poetic sparks. To do so, one may return to the operation of the Freudian
condensation: likeness in unlike objects is the precondition for the
dream-work of condensing them into hybrid or composite figures. This
brings us to the idea of metaphor: Aristotle (1984b: 1459a, 255) defines
metaphor as the expression ‘of the similarity in dissimilars’. There are
certainly several aspects of similarity or analogy that allow two
Fusion Style 105
disparate objects to be combined. In addition to semantic similarity, on

which Jakobson grounds his ideas of metaphor, of great help to
discover the similar aspects of two combined objects are Willard
Bohn’s categorizations of Surrealist images: ‘those that depend on
physical/formal similarity, functional similarity, or similarity involving
other characteristics’ (2002: 155).
Let us, then, apply the above-mentioned aspects of similarity to the
composite figure of a human body with an ostrich’s head. A human
being and an ostrich can be combined on the grounds of semantic
similarity: ‘ostrich’ is used as a metaphor for a person who refuses to
face the music. This significance in fact derives from the popular view–
which has been seen as a characteristic of the ostrich – that the ostrich
hides from danger by burying its head in the sand. Consequently, the
substitution of an ostrich’s head for a human head draws attention to the
fact that a person’s intention of ignoring reality arises in his/her head, or
rather, brain. Moreover, the transformation of a human head (as a
metonymy for a human body) into an ostrich’s head (as a metonymy for
an ostrich) brings out/back an uncanny function of human flesh: ostrich
meat is edible and so is human flesh. This not merely renders this
grotesque hybrid even more grotesque, but serves to illustrate
Harpham’s view that ‘the grotesque is embodied in an act of transition,
of metonymy becoming metaphor’ (1982: 47): the grotesque happens
when that which should remain contiguous/metonymic turns out to be
identical/metaphorical. It is natural that human beings subsist on
ostrich meat, but it is unnatural, or really, grotesque, that human beings
are ostrich meat. To sum up, a human body and an ostrich should be
essentially kept apart and yet fused together: the former is naturally
inedible and at the same time naturally edible when its head grows into
that of an ostrich.
5. The Flesh Made Metaphor

Unquestionably, the grotesque body is metaphorical. More precisely,
the grotesque body is a violent metaphor, one which endows itself with,
to quote Breton (1988: 327) on Surrealist images, ‘un très haut degré
d’absurdité immediate’ (‘a high degree of immediate absurdity’) and
which, as Clark Hulse (1981: 7) says of metamorphic images, intimates
‘ecstasy or terror of the flesh made free to move across the categories of
substance’. To better understand the grotesque body as a (violent)
metaphor, one should have a close look at the notion of metaphor and
106 Shun-liang Chao
its conventional operation. In general, metaphor urges the reader to

look beneath, or rather, disregard, its surface to bring to light the latent
or figurative similarities between its two components. As Hegel (1975:
403) explains in his Aesthetics: ‘When, e.g., we hear “the springtime of
these cheeks” or a “sea of tears” we are compelled to take this
expression not literally but only as an image, the meaning of which the
context expressly indicates to us’ (italics mine). Nevertheless, as a
metaphor, the grotesque body, whilst demanding figurative
interpretations, brings into relief the superficial or literal level of
dissimilarity in order to stimulate strong monstrous or absurd effects.
The grotesque body, as Peter Stockwell (2003: 17) writes of surreal
images, represents ‘a literalisation of metaphor’. This is especially the
case when it comes to the construction of the verbal type of grotesque
metaphor, since the visual type, due to its immediate visibility, never
fails to strike the viewer with grotesqueness at the literal level – no
matter how close the imaginative distance between its components.
That is to say, in order to reach the striking effects of visual grotesques,
the verbal type needs to foreground its literal incompatibility or
absurdity or, to quote T. E. Hulme on fresh poetic metaphors (1971:
134), ‘to make you continuously see a physical thing, to prevent you
gliding through an abstract process’. I shall return to the abstraction – as
opposed to literalisation – of metaphor.
To throw light on the borderline between metaphor in general and
grotesque metaphor in particular, one may resort to the two processes of
the Freudian dream-work: the secondary process (normal thinking) is
foregrounded in the former and yet the primary process (logical
impossibility) in the latter. The literal level of metaphor operates like
the primary process, in which the rules of logic carry no weight at all
because, incited by the id, it discharges psychic energy to construct a
‘perceptual identity’ of wishes. On the other hand, the function of the
figurative level parallels that of the secondary process, which follows
the lead of the ego to establish a ‘thought identity’, that is, to search for
objects in reality that match the psychic images created in the primary
process (Freud 1991a: 761-62). Simply put, the primary process is of
the signifier and the secondary process of the signified. It is fair to say,
then, that Jackson (1981: 84-85) is in fact referring to the primary/literal
process of metaphor when she argues that ‘the fantastic does not
proceed by analogy – it is not based on simile and comparison (like, as,
as if) but upon equation (this did happen). With the problem of
Fusion Style 107
“character”, the fantastic does not introduce scenes as if they were

real . . . : it insists upon the actuality of the transformation (as in . . .
Kafka’s Metamorphosis)’. Or in the fantastic, as Barthes (1995: 856)
speaks of Giuseppe Arcimboldo’s paintings of composite heads,
‘l’analogie devient folle, parce qu’elle est exploitée radicalement,
poussée jusqu’à se détruire elle-même comme analogie: la comparaison
devient métaphore’ (‘the analogy goes mad, because it is radically
exploited, pushed to the point of demolishing itself as analogy: the
simile turns into metaphor’). That is, the fantastic burgeons from actual
rather than virtual metamorphoses. Whence comes the prima facie
absurdity of the fantastic or the primary process, i.e. that which
metaphor in general wants to withdraw and yet grotesque metaphor –
especially, the verbal type – desires to spotlight.
Mention should be made briefly of the distinction between metaphor
and simile before we continue to elaborate on the idea of verbal
grotesque metaphor. Whilst metaphor is for Aristotle (1984a: 1406b,
173) only slightly different from simile, Quintilian (1921: Book viii, vi.
8-9, 305) in the Institutio Oratoria draws a rather clear line between
them: ‘in the latter we compare some object to the thing which we wish
to describe, whereas in the former this object is actually substituted for
the thing’. That is, simile depends on comparison but metaphor on
actual substitution. It is the actual transformation of one item into
another that gives birth to literal absurdity or actual falsehood that
Davidson (1979: 39) ascribes to metaphor: ‘all similes are true and
most metaphors are false. The earth is like a floor, the Assyrian did
come down like a wolf on the fold, because everything is like
everything. But turn these sentences into metaphors, and you turn them
false’. It is, so to say, simply beyond common sense or ‘the sense of
order’ to equate or identify the Assyrian with wolves (i.e. the Assyrian
are wolves) because they are by nature dissimilar or discordant.
Hence, in order for it to make sense, metaphor asks for figurative
understanding at a deeper level to locate the underlying common
ground which justifies the complete identification of one component
with the other. Rudolf Arnheim (1966: 279; see also Bohn 2002: 144-45)
sees the hidden common ground in a metaphor as its ‘structural unity’,
which ‘can be obtained on the basis of certain physiognomic qualities
the components have in common. Therefore the discordant qualities of
the components will retreat, the common ones will come to the fore’. In
108 Shun-liang Chao
this manner, the incongruous components lose their individual

concreteness and ‘become more abstract’, that is, they turn on the
‘intercourse of thoughts’ (in Richards’s words). ‘Otherwise’, Arnheim
(1966: 279) underscores, ‘the [metaphorical] construction would either
split up into incompatible elements or give birth, on the reality level, to
a Surrealistic monster’. In Arnheim, the proper operation of metaphor
relies on abstraction or abstracted meaningful characteristics that
outshine the concrete or visual incompatibility of the components. In
Christine Brooke-Rose (1958: 156, 155), literalisation of metaphor
occurs and ambiguity arises when ‘the functional element [of a
metaphor] is lacking or weak, or not easily apprehensible or
far-fetched’. By ‘functional element’ she refers to functional similarity
(‘A is called B by virtue of what it does’) – as opposed to ‘sensuous’
similarity (‘A is called by virtue of what it looks like, or more rarely,
sounds like’, etc. – which triggers figurative understanding or
metaphorical abstraction. To sum up, then, to weaken the characteristic
and/or functional element of a metaphor is to hinder its abstract process
and foreground its visual incongruity or literal absurdity.
Let us look at several metaphors to illustrate this assumption. In
Thomas Campion’s metaphor as follows (1969: 174):
There is a Garden in her face
Where Roses and White Lillies grow; (ll. 1-2)
The visual conflict between a face and a garden of roses and lilies arises
immediately, but it does not take long for the conflict to give way to
their abstracted common features: the lady’s face has the pink and pure
qualities of roses and lilies. For the characteristic similarities are not
far-fetched at all and are strong enough to reach the structural unity of
the metaphor. By the same token, it is not hard to distil from the
following metaphor by Shakespeare a functional element which is
strong enough for the components to lose their individual concreteness
(1997: 52):
When fortie Winters shall beseige thy brow,
And digge deep trenches in thy beauties field, (ll. 1-2)
As an army will relentlessly ravage the city of the enemy, so ‘fortie
Winters’ (as a metonymy for time) will mercilessly destroy the
beautiful face of a youth. Once this prominent functional similarity is
unveiled, the literal or visual collision retreats between a youth’s face
Fusion Style 109
and a besieged city, his eyebrows and city walls, his forehead and the
field before the walls, and the like. Campion’s and Shakespeare’s
metaphor are predominantly characteristic or functional, whereas
Richard Crashaw’s metaphor of Magdalene’s tears is overtly sensuous
(1957: 83):
What hath our world that can entice
You to be borne? what is’t can borrow
You from her eyes swolne wombes of sorrow. (st. 21)
Although wombs and eyes or eye balls may share a similar shape, they
are unlike characteristically, semantically, and so forth. Arguably,
wombs and eyes may be also functionally similar: they both shed
liquids or produce something; this functional similarity, however, is too
feeble or far-fetched to stop the reader continuously seeing the horrible
image of (the wombs of) Magdalene’s eyes literally giving birth to tears
(“You”). This situation is even more so in T. S. Eliot’s completely
sensuous metaphor (1963: 45),
Daffodil bulbs instead of balls
Stared from the sockets of the eyes! (ll. 5-6)
Daffodil bulbs may look like eyeballs, but they are dissimilar otherwise.
In Eliot’s metaphor as in Crashaw’s, then, the physical similarity that
may initially allow the components to be mingled into a metaphor
nevertheless brings into relief their visual incompatibility because they
are different in so many concrete details, or rather, because
sensuous/physical similarity alone is not able to spark metaphorical
abstraction. Hence, a surreal image steps to the fore.
It is proper to say, then, that without the support of strong functional or
characteristic or semantic similarity, a metaphor would spawn visually
jarring or surreal images which, as Reuven Tsur (2002: 294) claims,
serve to ‘delay the smooth cognitive process’ or ‘prolong a state of
[cognitive and emotional] disorientation’ so much so that the reader
‘lingers at the visual images without appraising their significance’. This
situation is for Breton (1988: 338) the greatest virtue of the Surrealist
image: the virtue rests upon ‘celle qui présente le degré d’arbitraire le
plus élevé . . . ; celle qu’on met le plus longtemps à traduire en langage
pratique, soit qu’elle recèle une dose énorme de contradiction apparente,
soit que l’un de ses termes en soit curieusement dérobé. . . , soit qu’elle
déchaîne le rire’ (‘the one that presents the highest degree of
110 Shun-liang Chao
arbitrariness . . . ; the one that takes the longest time to translate into
practical language, either because it possesses an enormous amount of
obvious contradiction, or because one of its terms is strangely
hidden. . . , or because it stirs up laughter’). When contradictory or
arbitrary combinations of this sort happen, the primary process resists
readily succumbing to the secondary process, that is, the signifier
refuses to be anchored to the signified but instead continues to float.
This situation would act as a requisite for the birth of verbal grotesque
metaphor.
One must be aware of the fact that not all surreal metaphors or images
would breed grotesque monsters. In a broad sense, the seed of the
grotesque, as Harpham (1982: 124) points out, lies in ‘all metaphors
with a spark of life’, wherein
the referential (usually called the literal) always confronts us. It
is a prior phase, whose self-annihilating absurdity motivates us
to the act of interpretation that completes the understanding of
the metaphor. It is the phase of the grotesque, which . . . occurs
primarily as a naive experience, a function of the literal, in the
context of referential art. Considered referentially, metaphors are
grotesques.
It is nevertheless necessary to further explain Harpham’s observations.
Apposite to this case are Quintilian’s classifications of metaphor (1921:
Book viii, vi. 7-10, 305; examples are mine): the transformation of 1)
the animate into the animate (wounds pant); 2) the animate into the
inanimate (iron-hearted); 3) the inanimate into the animate (barking
bells); 4) the inanimate into the inanimate (poetry is eternal treasure).
Here one finds that the first three kinds are potentially grotesque
because they obviously contain ‘a spark of life’ or are flesh-made.
Flesh-made metaphors are able to arouse grotesqueness in that, as
Philip Thomson (1972: 56) (following Bakhtin) puts it, ‘the grotesque
is essentially physical, referring always to the body and bodily
excesses’. This situation, Lacan would concur, has to do with the
surviving anxiety-provoking memory of the fragmented body: the
transference between body parts or between bodies and (in)animate
objects would expose to sight for the reader his/her fundamental
monstrosity. This would constitute a rationale for Robert Rogers’s
comment that ‘Poets instinctively turn to images of the body when they
mean to disturb the reader most’, for body imagery tends to encourage
Fusion Style 111
‘functional regression in the reader’s mentation’ (1978: 90, 78) or, as

Edmund Burke would put it, to sharpen the reader’s consciousness of
‘self-preservation’, which turns on pain or terror, the strongest of all the
emotions (1990: 35-36).
Despite the fact that all flesh-made metaphors harbour grotesqueness to
some extent, a high degree of grotesqueness arises, I would suggest,
when the components of a flesh-made metaphor are not strongly similar
in terms of function, characteristic, and/or semantics so much so that its
referential absurdity refuses to easily annihilate itself and become
abstract; or a war of domination occurs between the primary and
secondary processes; or the five senses are confused in and by images
not only absurd but marvellous and phantasmagorical. We have seen
conspicuously flesh-made metaphors by Crashaw and Eliot which give
birth to highly grotesque images. Let us look at another example by
recalling Apollinaire’s complex metaphor: a tongue transforms itself
into a goldfish, the oral cavity into a bowl, and a woman’s voice into
water. On one hand, we have a mouth inside which a tongue is moving
(to speak) and, on the other, a bowl inside which a goldfish is
swimming. One set of objects are in-corporated into the other. Rather,
that which should remain contiguous suddenly turns out to be identical:
as one may drink from a bowl of water (with her mouth), her mouth is
suddenly turning into the bowl because of functional similarity (both
are a kind of container); the tongue inside is growing into a goldfish
because of characteristic similarity (both are red) and/or because of
physical similarity (they may look alike), an association sparked by the
tongue’s contiguity with water. Again, the grotesque, as we have seen,
takes place in the transition of metonymy into metaphor, which no
doubt instantly ignites a strong sense of absurdity, monstrosity, and
perplexity.
Other aspects of similarity can be found: the soft tactile sensations of a
goldfish and a tongue (sensuous/physical similarity), together with the
tender qualities of female voice and water (characteristic similarity),
could suggest the idea of comfort or agreeableness. Nevertheless, even
after the above common elements are found, the individual
concreteness of these various objects does not disappear but sticks
around in the conscious mind. For none of them is strong enough to
serve as a central thought or signified which defines this complex
metaphor so that the logical impossibility or grotesqueness can retreat
112 Shun-liang Chao
to the background; for the (reproductive) imagination seems inadequate

to present in the mind’s eye a totality or complete metamorphosis in
which one set of objects takes full place of the other. Instead, we have
composite figures such as those born of the Freudian
dream-condensation. It is tempting to say, then, that in metaphors that
kindle a high degree of grotesqueness, the primary process (nonsense)
is always reluctant to fully give up its seat to the secondary process
(sense). Such a foregrounding of the primary process speaks to
Kristeva’s argument that avant-garde arts ‘wipe out sense through
nonsense and laughter’ (1980: 142): ‘Laughter always indicates an act
of aggression against the Creator, or rather, a rejection of the Creator’;
every new poetic/art device is ‘a practice of laughter’ (1984: 223-24).
For Baudelaire (1961: 986, 985), ‘[i]l n’y a qu’une vérification du
grotesque, c’est le rire’ (‘there is only one proof of the grotesque, which
is laughter’). The grotesque, he highlighted, is mostly ‘une création
mêlée d’une certaine faculté imitatrice d’éléments préexistants dans la
nature’ (‘a creation, mixed with a certain faculty of imitating elements
pre-existing in nature’), and the laughter it provokes expresses ‘l’idée
de supériorité, non plus de l’homme sur l’homme, mais de l’homme sur
la nature’ (‘the idea of superiority, no longer of man over man, but of
man over nature’). For the grotesque is not the product of imitation but
of imagination; it is, so to speak, the triumph of human (creative)
imagination over the order of nature to unexpectedly or
unconventionally have, say, a tongue transformed into a goldfish.
Barthes (1995: 861, 860) calls metamorphoses of this sort ‘métaphores
improbables’ (‘improbable metaphors’): ‘le travail métaphorique [dans
les têtes composées d’Arcimboldo] est si audacieux (comme celui d’un
poète très précieux ou très moderne) qu’il n’y a aucun rapport « naturel
» entre la chose représentée et sa représentation’ (‘the metaphorical
operation [in Arcimboldo’s composite heads] is so audacious (as the
one of a very precious or very modern poet) that there is no “natural”
connection between the thing represented and its representation’).
Monsters are thereby born (868):
Or la « merveille » – ou le « monstre » – c’est essentiellement ce
qui transgresse la séparation des règnes, mêle l’animal et le
végétal, l’animal et l’humain; c’est l’excès, en tant qu’il change
la qualité des choses auxquelles Dieu a assigné un nom: c’est la
métamorphose, qui fait basculer d’un ordre dans un autre.12
Fusion Style 113
Bodies or body parts are made free to grow from one order of substance
into another, as the free flow of drives in the primary process of dreams;
borders fall apart and so do complete body forms. That is to say, as one
order is turning itself into another without limit, that which is
determinate, clean, and continuous turns out to be ambiguous, untidy,
and fragmented. Grotesque monsters embody, to quote Kristeva (1984:
58) in avant-garde art, ‘the flow of the semiotic into the symbolic’.
6. Conclusion
‘Only in dream logic’, Kristeva (1984: 29) notes, do the semiotic
practices of poetic language dominate the signifying process. The
grotesque body exactly proceeds by dream logic: it is composed of the
displacement and condensation of semiotic fragments and thus full of
(both syntactic and semantic) fissures or hiatuses. More precisely, it
performs the semiotic by giving birth to a kind of flesh-made metaphor
whose primary process (nonsense; referential absurdity; concreteness;
the signifier) is at least as spotlighted as its secondary process (sense;
figurative similarities; abstraction; the signified), thereby, as Thomson
(1972: 65) puts it, ‘producing in the reader a strange sensation – making
one suddenly doubt one’s comfortable relationship with the language’.
This is even more patent in verbal grotesque metaphor. Moreover, since
the grotesque body is one without a unified form, a structural unity, a
central signified, i.e. a decentred or ‘writerly’ body, it demands the
second type of the two interpretations Derrida (1978: 192) brings forth:
‘The one seeks to decipher, dreams of deciphering a truth or an origin
which escapes play and the order of the sign. . . . The other, which is no
longer turned toward the origin, affirms play and tries to pass beyond
man and humanism . . . .’ In short, the grotesque body performs a
poetics of contradiction and ambiguity, through the (con)fusion of
forms or objects that, logically, should be kept separate.
Endnotes
1
This paper is dedicated to Tze-ming Hu (1972-2003) as a token of everlasting love and
memory.
2
For Lacan (1981: 247), poetry is, first and foremost, metaphor: ‘[C]e pourrait être une
définition du style poetique que de dire qu’il commence à la métaphore, et que là où la
métaphore cesse, la poesie aussi’ (‘[I]t could be a definition of the poetic style to say
that it begins with metaphor and that where metaphors stops, poetry stops as well’).
3
To illustrate the second type of images, Freud (1991a: 399-400) cites his own (famous)
dream of Irma’s injection, wherein Irma ‘turn[s] into a collective image with . . . a
114 Shun-liang Chao
number of contradictory characteristics’ which ‘allu[de] to a whole series of other

figures’, none of whom, however, ‘appear[s] in the dream in bodily shape’; as such, she
‘became the representative of all these other figures which had been sacrificed to the
work of condensation, since I passed over to her, point by point, everything that
reminded me of them’.
4
Jane Gallop (1985: 84-85) offers a brilliant exposition of these remarks by Lacan:
‘That which is not organized or totalized or unified cannot be violated. The anxiety that
Lacan represents as the risk of “sliding back again into. . . chaos” can be experienced
only by the ego with its “illusion of unity.” But the mirror stage is only the first step in
the “dizzy Ascent.” At this point the subject can “look forward” without the fear of
“sliding back,” since she is just beginning her climb. The ego is only just being formed
and as yet has no ongoing organization to be endangered. The mirror stage is a fleeting
moment of jubilation before an inevitable anxiety sets in. The mirror stage is thus high
tragedy: a brief moment of doomed glory, a paradise lost’.
5
Lacan, as Michael Payne (1993: 31) illuminates, ‘employs the image of dehiscence for
its sense of explosive development, as in time-lapse photographs of the bursting
opening of seed pods. Just as the fruit, anther, or sporangium splits open its sides as it
ripens, so the once unified fiction of the child’s ego is destined to become fragmented
and alienated. Rather than being cast free by this process, the child is destined to retain
the remnants of its early undeveloped state as a kind of “fetalizatoin,” or persistence of
its immaturity, into its later life. The lure or mirage of future integrity will persist even
in the face of a continuous present in which wholeness has not been achieved’.
6
‘In this ideal text, the networks are multiple and interplay; none of them is able to top
all the others; this text is a galaxy of signifiers, not a structure of signifieds; it has no
beginning; it is reversible; one reaches it via several entrances, none of which can be
definitely declared to be the principal one; the codes it mobilizes spread out as far as
the eye can see, they are unfixed (meaning is never subject to a rule of determination,
unless by tossing dice)’.
7
The symbolic function, Kristeva (1980: 134) reminds us, still maintains its presence in
the discourse of art; otherwise the discourse is no longer a language but falls into
delirium.
8
Noticeably, I am not implying here that these three terms are interchangeable. In our
case, it is perhaps proper to describe their relationships as follows: The semiotic and the
dream-work, though each operates at a different level in the psychic apparatus, tend to
trouble the signifying process by veering themselves towards the real which they will
never represent.
9
For a critique of Jakobson’s article, see the first section in Paul Ricœur’s ‘The Work of
Resemblance’ (1977: 173-215) in The Rule of Metaphor.
10
Since, as Barthes (1993b: 1499) explains, the operation of syntagm and of paradigm
are necessary to all discourse, the typical use of metonymy or of metaphor in certain
types of discourse only implies ‘la dominance de l’un ou l’autre’ (‘the dominance of
one or the other’). Jakobson (1987a: 85) himself also once made observations on the
fact that metonymy and metaphor are not mutually exclusive: ‘in poetry, where
similarity is superinduced upon contiguity, any metonymy is slightly metaphoric and
any metaphor has a metonymic tint’.
11
De Man (15-16) goes so far as to say that ‘no grammatical decoding, however refined,
Fusion Style 115
could claim to reach the determining figural dimensions of a text. There are elements in
all texts that are by no means ungrammatical, but whose semantic function is not
grammatically definable, neither in themselves nor in context’. Accordingly, reading a
literary text can be likened to looking through a stained glass window; the meaning of a
text thus constantly flickers and slides, becomes full of blanks that ‘[have] to be, but
cannot be’ (15), filled by grammatical means.
12
‘Now the “marvel” – or the “monster” – is essentially that which transgresses the
separation of realms, mixes the animal and the vegetable, the animal and the human; it
is excess, since it changes the quality of the things to which God has assigned a name: it
is metamorphosis, which turns one order into another’.
References
Apollinaire, Guillaume. 1956. ‘Fusée-Signal’ in Œuvres poétiques (ed.) Marcel Adéma
and Michel Décaudin. Paris: Gallimard: 363.
Arnheim, Rudolf. 1966. ‘Abstract Language and the Metaphor’ in Towards a
Psychology of Art. Berkeley and Los Angeles: University of California Press:
266-82.
Aristotle. 1984a. Rhetoric (tr. W. Rhys Roberts) in The Rhetoric and the Poetics of
Aristotle. New York: Modern Library: 19-218
__. 1984b. Poetics (tr. Ingram Bywater) in The Rhetoric and the Poetics of Aristotle.
New York: Modern Library: 223-66.
Bakhtin, Mikhail. [1965] 1984. Rabelais and His World (tr. Helene Iswolsky).
Bloomington: Indiana University Press.
Barthes, Roland. 1993a. Le degreé zero de l’écriture in Œuvres complètes, Vol. I (ed.)
Éric Marty. Paris: Seuil: 139-86.
__. 1993b. Éléments de sémiologie in Œuvres complètes, Vol. I (ed.) Éric Marty. Paris:
Seuil: 1469-1522.
__. 1994a. ‘La mort de l’auteur’ in Œuvres complètes, Vol. II (ed.) Éric Marty. Paris:
Seuil: 491-95.
__. 1994b. S/Z in Œuvres complètes, Vol. II (ed.) Éric Marty. Paris: Seuil: 557-728.
__. 1994c. ‘De l’œuvre au texte’ in Œuvres complètes, Vol. II (ed.) Éric Marty. Paris:
Seuil: 1211-17.
__. 1994d. Le Plaisir du texte in Œuvres complètes, Vol. II (ed.) Éric Marty. Paris :
Seuil: 1495-1529.
__. 1995. ‘Arcimboldo ou Rhétoriqueur et magician’ in Œuvres complètes, Vol. III (ed.)
Baudelaire, Charles. 1961. ‘De l’essence du rire’ in Œuvres completes (ed.) Claude
Pichois. Paris: Gallimard: 975-93.
Bohn, Willard. 2002. The Rise of Surrealism: Cubism, Dada, and the Pursuit of the
Marvelous. Albany: State University of New York Press.
Breton, André. 1988. Manifestete du Surréalisme in Œuvres complètes, Vol. I (ed.)
Marguerite Bonnet. Paris: Gallimard, 1988: 311-46.
Brooke-Rose, Christine. 1958. A Grammar of Metaphor. London: Secker & Warburg.
Burke, Edmund. 1990. A Philosophical Enquiry into the Origin of our Ideas of the
Sublime and Beautiful (ed.) Adam Philips. Oxford and New York: Oxford
University Press.
Campion, Thomas. 1969. ‘There Is a Garden in Her Face’ in The Works of Thomas
116 Shun-liang Chao
Campion: Complete Songs, Masques, and Treaties on a Selection of the Latin Verse
(ed.) Walter R. Davis. London: Faber & Faber: 174.
Crashaw, Richard. 1957. ‘The Weeper [1646]’ in The Poems, English, Latin and Greek
of Richard Crashaw (ed.) L. C. Martin. Oxford: Clarendon Press: 83.
Davidson, Donald. 1979. ‘What Metaphors Mean’ in On Metaphor (ed.) Sheldon Sacks.
Chicago: University of Chicago Press: 29-45.
de Man, Paul. 1986. The Resistance to Theory. Minneapolis: University of Minnesota
Press.
Derrida, Jacques. [1967] 1978. Writing and Difference (tr. Alan Bass). Chicago:
University of Chicago Press.
Eliot, T. S. 1963. ‘Whispers of Immortality’ in Collected Poems: 1909-1962 (ed.) T. S.
Eliot. New York and London: Harcourt: 45.
Evans, Dylan. 1996. An Introductory Dictionary of Lacanian Psychoanalysis. Hove
and New York: Brunner-Routledge.
Fink, Bruce. 1995. The Lacanian Subject: Between Language and Jouissance.
Princeton: Princeton University Press.
Frank, Joseph. 1991. The Idea of Spatial Form. New Brunswick: Rutgers University
Press.
Freud, Sigmund. [1900] 1991a. The Interpretation of Dreams (tr. James Strachey).
Penguin Freud Library, 4. London: Penguin Books.
__. [1916-17] 1991b. Introductory Lectures on Psychoanalysis (tr. James Strachey).
__. [1933] 1911c. New Introductory Lectures on Psychoanalysis (tr. James Strachey).
__. [1940] 1993. An Outline of Psychoanalysis (tr. James Strachey). Penguin Freud
Library, 15. London: Penguin Books: 375-443.
Gallop, Jane. 1985. Reading Lacan. Ithaca: Cornell University Press.
Gombrich, E. H. 1979. The Sense of Order: A Study in the Psychology of Decorative Art.
London: Phaidon.
Harpham, Geoffrey G. 1976. ‘The Grotesque: First Principles’ in Journal of Aesthetic
and Art Criticism 34(4): 461-68.
__. 1982. On the Grotesque: Strategies of Contradiction in Art and Literature.
Havránek, Bohuslav. 1964. ‘The Functional Differentiation of the Standard Language’
in A Prague School Reader on Esthetic, Literary Structure, and Style (ed. and tr.
Paul L. Garvin). Washington D.C.: Georgetown University Press: 3-16.
Hegel, G. W. F. 1975. Aesthetics: Lectures on Fine Art, Vol. I (tr. T. M. Knox). Oxford:
Clarendon Press.
Homer, Sean. 2005. Jacques Lacan. London and New York: Routledge.
Hulme, T. E. 1971. Speculations: Essays on Humanism and the Philosophy of Art (Rpt.
ed.) (ed.) Herbert Read. London: Routledge & Kegan Paul.
Hulse, Clark. 1981. Metamorphic Verse: The Elizabethan Minor Epic. Princeton:
Princeton University Press.
Jackson, Rosemary. 1981. Fantasy: The Literature of Subversion. London and New
York: Methuen.
Jakobson, Roman. 1987a. ‘Linguistics and Poetics’ in Language in Literature (ed.)
Krystyna Pomorska and Stephen Rudy. Cambridge, Mass.: Belknap Press of
Harvard University Press: 62-94.
Fusion Style 117
__. 1987b. ‘Two Aspects of Language and Two Types of Aphasic Disturbances’ in
Language in Literature (ed.) Krystyna Pomorska and Stephen Rudy. Cambridge,
Mass.: Belknap Press of Harvard University Press: 95-114.
Kristeva, Julia. [1977] 1980. Desire in Language: A Semiotic Approach to Literature
and Art (Rpt. ed.) (ed.) Leon S. Roudiez; (tr. Thomas Gora et al.). Oxford: Basil
Blackwell.
__. [1974] 1984. Revolution in Poetic Language (tr. Margaret Walter). New York:
Columbia University Press.
Lacan, Jacques. 1953. ‘Some Reflections on the Ego’ in International Journal of
Psycho-analysis 34: 11-17.
__. [1966] 1977. Écrits: A Selection (tr. Alan Sheridan). New York and London: W. W.
Norton.
__. [1978] 1988. The Seminar of Jacques Lacan: Book II (tr. Sylvana Tomaselli). New
York: Cambridge University Press.
__. 1981. Les Psychoses (ed.) Jacques-Alain Miller. Paris: Seuil.
Laplanche, J. and J.-B. Pontalis. [1967] 1973. The Language of Psycho-Analysis (tr.
Donald Nicholson-Smith). New York: W. W. Norton, 1973.
McAfee, Noëlle. 2004. Julia Kristeva. London and New York: Routledge.
MukaĜovský, Jan. 1964. ‘Standard Language and Poetic Language’ in A Prague School
Reader on Esthetic, Literary Structure, and Style (ed. and tr. Paul L. Garvin).
Washington D.C.: Georgetown University Press: 31-69.
Ovid. 1955. Metamorphoses (tr. Rolfe Humphries). Bloomington: Indiana University
Press.
Payne, Michael. 1993. Reading Theory: An Introduction to Lacan, Derrida, and
Kristeva. Oxford: Blackwell.
Quintilian. 1921. The Institutio Oratoria of Quintilian, Vol. IV (trans. H. E. Butler).
Cambridge, Mass.: Harvard University Press..
Ragland, Ellie. 1993. ‘Lacan, the Death Drive, and the Dream of the Burning Child’ in
Death and Representation (ed.) Sarah Goodwin and Elisabeth Bronfen. Baltimore
and London: Johns Hopkins University Press: 80-102.
Richards, I. A. 1936. The Philosophy of Rhetoric. Oxford: Oxford University Press.
Ricœur, Paul. [1975] 1977. ‘The Work of Resemblance’ in The Rule of Metaphor:
Multi-disciplinary Studies of the Creation of Meaning in Language (tr. Robert
Czerny et al.). Toronto: University of Toronto Press: 173-215.
Rogers, Robert. 1978. Metaphor: A Psychoanalytic View. Berkeley and Los Angeles:
University of California Press.
de Saussure, Ferdinand. [1955] 1959. Course in General Linguistics (ed.) Charles Bally
and Albert Sechehaye; (tr. Wade Baskin). New York: Philosophical Library.
Shakespeare, William. 1997. ‘Sonnet 2’ in The Art of Shakespeare’s Sonnets (Helen
Vendler). Cambridge, Mass.: Belknap Press of Harvard University Press: 52.
Shelley, Percy Bysshe. 2002. A Defence of Poetry in Shelley’s Poetry and Prose (2nd ed.)
(ed.) Donald H. Reiman and Neil Fraistat. New York: W. W. Norton: 510-35.
Smith, Anne-Marie. 1998. Julia Kristeva: Speaking the Unspeakable. London: Pluto
Press.
Stockwell, Peter. 2003. ‘Surreal Figures’ in Cognitive Poetics in Practice (ed.) Joanna
Gavins and Gerard Steen. London and New York: Routledge: 13-25.
Thomson, Philip. 1972. The Grotesque. London: Methuen.
Tsur, Reuven. 2002. ‘Aspects of Cognitive Poetics’ in Cognitive Stylistics: Language
118 Shun-liang Chao
and Cognition in Text Analysis (ed.) Elena Semino and Jonathan Culpeper.
Amsterdam/Philadelphia: John Benjamins Publishing Co.: 279-318.
White, Allon. 1993. ‘L’Éclatement du sujet: The Work of Julia Kristeva’ in Carnival,
Hysteria, and Writing: Collected Essays and Autobiography. Oxford: Clarendon
Press: 62-87.
Williams, R. Grant. 2001. ‘Disfiguring the Body of Knowledge: Anatomical Discourse
and Robert Burton's The Anatomy of Melancholy’ in ELH 68(3): 591-613.
Donald Barthelme’s Art of Storytelling: A Cognitive-
Semiotic Textual Analysis of ‘On the Deck,’ ‘At the
Tolstoy Museum’ and ‘The Baby’
Ulf Cronquist
Gothenburg University, Sweden
Abstract
The overriding purpose of this article is to analyse three randomly chosen short stories
by Donald Barthelme from a cognitive-semiotic perspective. In order to supply an
analysis that considers literary texts as whole gestalts, I introduce a model for
stratified reading and interpretation developed by Line Brandt and Per Aage Brandt
(Brandt & Brandt 2005a, Brandt (forthcoming)). Also, I propose a diagrammatic
poetics that is grounded in the methodology of cognitive science per se, rather than
the sometimes too reductive applications of cognitive linguistics for readings of
literary texts in the recently emerging paradigm of cognitive poetics. In my analyses
of Barthelme’s stories I introduce and apply three cognitive-semiotic tools for textual
interpretation: Per Aage Brandt’s notion of evidentiality and enunciation (Brandt
2004b), his narrative diegesis (Brandt 1983, 1989) and Line and Per Aage Brandt’s
model for blending/conceptual integration (Brandt & Brandt 2005b). My intention is
to show that traditional literary reading combined with methodology from cognitive
science and cognitive semiotics will map out the pertinent structures and details of
Barthelme’s textuality, in an analytic mode that generally demystifies Barthelme’s
reputation as a writer of ‘unreal’ texts about ‘unreal’ worlds.
Keywords: cognitive poetics/cognitive semiotics; enunciation & evidentiality;
narrative diegesis; blending theory; Donald Barthelme; Per Aage Brandt; Line Brandt.
1. Introduction
Donald Barthelme has been called ‘the pioneer of American
postmodernism’ (Sloboda 1997: 109). And certainly together with
writers like e.g. John Hawkes, William Gass and Robert Coover,
Barthelme enters new prose territory beginning in a historical parallel
with the French Roman Nouveau. Barthelme and his ‘iconoclast’ peers
are per definition a challenge to readers of Realist fiction since their
texts seem to be out of joint with respect mimetic and diegetic
expectations. A somewhat paradoxical summary of Barthelme’s
fiction reads like this: ‘In his short stories, Barthelme describes a
world so unreal that traditional modes of fiction can no longer
encompass it’ (The Columbia Encyclopedia: 4381). We are thus
instead invited to textual worlds where there is a strong uncertainty
120 Ulf Cronquist
relation between the signifier and the signified. However, in

considering the effort it takes to come to grips with Barthelme’s text
worlds, instead of speculating about ‘the floating signifier,’ I focus in
the following analysis on the construction of signifieds, as meaning
production is a question of cognitive-semiotic intersubjectivity, i.e. the
exchange of signs between intentional minds.1
A complicating factor in Barthelme’s production is that on occasion
he inserts graphics and images in his texts, sometimes with (mock)
ekphrastic effects. Postmodern American writers, as Nicholas
Zurbrugg remarks, often create a literary montage that ‘interweaves
and accepts the copresence of differing discourses and conflicting
categories’ (Zurbrugg 1993: 56; qtd. in Sloboda 1997: 109). In
Barthelme’s textuality we frequently see this montage- or collage-like
quality strongly accentuated, and in one of the stories analysed here,
‘At the Tolstoy Museum,’ I will consider the relation of text and image
from different angles. ‘On the Deck’ will be analysed as a textual
montage or collage, while ‘The Baby’ presents itself as a more
traditional piece of fiction with a narrative structure that comes to
closure.
The purpose of this article is grounded in a straight-forward question:
how do we process ‘difficult’ postmodern writing like the stories of
Barthelme? And while the basic hypothesis is that we process texts
cognitively as information patterns, the method used is a cognitive-
semiotic analysis where careful literary reading is combined with
findings from cognitive science.2 Let us therefore first look at what a
literary reading implies – and then what we should expect from an
analysis grounded in procedures from cognitive science.
2. Stratified analysis
Brandt & Brandt (2005a) suggest a philological model for textual
analysis that both precedes the cognitive analysis and which can be
(continually) related to cognitive tools of analysis in relation to
language and literature. In what Brandt & Brandt call structural
stratification, the investigation concerns strata in the texual analysis
that are related to each other on four levels:
1. Language (grammatical structure) and enunciation;
2. Semantic context of the text including imagery and narrative;
3. Compositional form including phonetics, graphics etc.;
Donald Barthelme’s Art of Storytelling 121
4. Interpretive aesthetic status in the framework of a genre.

(Brandt & Brandt 2005a: 1203)
Thus, the literary text is philologically investigated, before and in
combination with analyses from a cognitive-semiotic perspective. It is
important to stress that a stratified analysis considers the text as a
global configuration, its gestalt as a whole.. A stratified analysis is
always concerned with real texts and is therefore an empirical
investigation of what is given – far from the (cognitive) linguist’s
construction of ‘unreal’ examples (such as ‘John came into the room’
etc.). A stratified procedure begins with the question who is speaking
and what is this text, proceeds to the how-question (poetics proper)
and then to the interpretive why-question, and while these four entities
may not always be distinguishable in practice they can and will be
separated for analytic purposes in the following.
3. A diagrammatic poetics
Patrick Colm Hogan remarks that cognitive science contains a number
of different research areas that often seem to have little or nothing at
all to do with each other (Hogan 2003: 29). But there is a fundamental
methodological principle that most researchers in the discipline agree
upon: that the working process within cognitive science is always
performed in three steps. First, a given problem is formulated in terms
of how information is processed; secondly, since our consciousness
does not receive ’pure’ information – it is our cognitive apparatus that
organises the information – we have to specify what cognitive
architecture we will use to study the problem; thirdly the analysis is
performed in terms of algorithmic sequences. These three steps are
taken so that the cognitively scientific analysis as explicitly as
possible accounts for how the processing of information happens from
inputs to output.4
But the most basic point about cognitive architecture and algorithmic
sequencing is not about how graphics or descriptive prose should be
used and combined. Rather it is so, as Per Aage Brandt underlines,
that diagrams in general give us a better and more lucid access to our
semiotic and cognitive conceptualisations, they satisfy the descriptive
needs of our consciousness in a better way than rhetoric or formal
logic (Brandt 2004a).5 The advantage with diagrams probably depends
on the fact that the scale is reduced in a fruitful way and gives us a
122 Ulf Cronquist
well functioning synoptic overview. In other words what takes place is

a mapping between different scales in the cognitive architecture – as
Brandt puts it: a natural geometry in consciousness. On cognitive
scientific grounds we have all the reason to assume that the brain
processes representations and conceptualisations in a geometric-
prototypical way, and that therefore the usage of well-wrought
diagrams not only ’mirrors’ this processing but also makes clear the
processes in the cognitive architecture.
Brandt points to the fact that our mental equipment consists of
relatively few types of diagrams, the basic ones being arrow
diagrams, flow-chart diagrams, mereological diagrams and tree
diagrams: arrow diagrams typically describe horisontal movements
with a beginning and an end, flow-chart diagrams specify movements
in the form of boxes with different contents, mereological diagrams
typically relate parts and wholes in ’cake-forms,’ tree diagrams
structure different forms of hierarchies, often vertically. These types
are probably relatively direct graphic expressions of our schematically
memory-oriented processes in consciousness and their structures shed
light on how we think whenever we think in terms of relations. From
an evolutionary point of view the four respective diagram types can be
compared to the basal activities of hunting, food preparation, storage
and forms of domestication. As Brandt underlines there are no definite
diagrammatical borders between everyday occurrences and abstract
thinking, we cogitate through diagrams when we sketch them or just
think of them – man is a veritable homo diagrammaticus.
Thus, there are good reasons for people working in language and
literature to analyze diagrammatical representations of textual
architecture in consciousness. For metricians and musicologists
models and diagrams are natural forms for representing the many
expressions of rhythm that are all cognitively-bodily grounded. It
should be as natural to investigate the bodily form of the literary text
as a whole, its cognitive rhythm and diagrammatic architecture in both
narrative structures and in lyrical expression and content.
4. Initial explication / ‘first reading’

From the point of view of textual analysis it is important to approach
the literary text as a global configuration. But it is equally important
that we bring pertinent descriptive categories to the object of study.
That is, whereas much time and energy can be spent in the framework
of ‘cultural studies’ debating which discourse-analytical framework is
the most appropriate (gender, Marxist, postcolonial etc.), I claim that a
first explication of the text is necessary to position the text in time and
space, looking at basic grammar and semantics. By initial explication
or a ‘first reading’ I simply mean that we should not get into the
analysis of meaning production before we have mapped the basic
structure and content of our object of analysis. This first step will then
necessarily define the directions for further consideration of the text.
I would describe the basics of the three Barthelme stories thus:
‘On the Deck’ consists of nine paragraphs of what an Olympic
narrator sees on the deck of a boat. The Olympic narrator switches to
first-person narrator in the last paragraph, possibly the voice in lines
5-7 of paragraph six belongs to him, too – but there is nothing overly
problematic about the voice of narration. As regards continuation in
time and space, the story is mainly temporally static in its spatial
description of figures on the deck’s ground but also involves the
change of several seasons in the last three paragraphs;
‘At the Tolstoy Museum’ consists of fourteen paragraphs and ten
illustrations, some of the latter accompanied by text giving a (mock)
ekphrastic effect. The narrator position circulates in the text between
third-person ‘we,’ Olympic, and first-person, which disrupts the
temporal, linear, processing of the text. And since the illustrations are
not directly connected to the running text, they too, of course, have an
effect on the temporality. The spatial aspect of the story is instead
emphasized, the fragmented positionality of the graphics read
somewhat like literary tableaux vivants;
‘The Baby’ consists of five paragraphs where the narration alternates
between third-person and first-person while the story proceeds linearly
towards a problem resolution. On another level, however, there is an
absurd temporality as the baby accumulates punitative hours to be
spent in his/her room – finally in the number of years. Spatially, the
story takes place in the living room and the baby’s room in a family
home, but the sense of being locked in is broken in the last phrase of
the text.
124 Ulf Cronquist
Thus, I propose that a first reading should be as ‘neutral’ as possible

as regards culture-specificity. We shall return to the relations of
culture, ideology and aesthetics below, in the section on interpretation.
5. Enunciation and evidentiality

The reader always meets a particular voice in a text, in some form,
some kind of narratorship. Per Aage Brandt suggests a model for
analyzing enunciation in conjunction with evidentiality, i.e. the mode
of the narrative voice is analyzed in relation to specifying the source
or nature of evidence – the semantic validity – in a given utterance
(Brandt 2004b). In Brandt’s model, beginning from a degree zero, the
here and now of a speech act, there are four axis that denote the
aspects and conjunctions of evidentiality and enunciation – Refer to
Figure 1.
I. Experiental (subjectivized)
IV. Aphonic
III. Polyphonic Ø Speech Act (imaginarized)
(relativized)
II. Epistemic (objectivized)
Figure 1: Brandt: Evidentiality and Enunciation (Brandt 2004b)

The experiental position (I.) is subjective (‘I see/hear/feel that…’), the
mental domain of the speaker, while the epistemic position (II.) is
objective (‘I know/ conclude/ find that…’), the speaker’s outer world.
Both these are strongly realistic in contrast with the other two types.
The polyphonic position (III.) denotes a speaker’s voice that
reproduces another person’s voice (‘As Plato said…’), and can also
involve the speaker’s own attitude e.g. irony (‘Indeed a platonic
thought…’), while the aphonic position (IV.) denotes a mode where
the speaker emphatically withdraws from investing in the utterance
(‘Let us imagine that…’ or ‘Once upon a time there was…’). For our
purposes, it should be stressed that literary fiction necessarily must

begin with the delegation from Ø > IV – otherwise it would not be
fiction – and can move from there to one of the three other positions.6
For example:
Ø > IV > I – ‘In Munich I felt as if I were dead’
Ø > IV > II – ‘In Munich I met many people who felt as if they
were dead’
Ø > IV > III – ‘In Munich Willie felt as if he were dead’
A basic point here is that when we follow the inscription of the
enunciator in the semantics – through the deictic processing –
considering relations of evidentiality and enunciation, we see how the
whole text, the global configuration, is constituted and processed
rather than remaining on sentence level. Furthermore, this model
could be used to investigate the difference in style between authors to
determine salient patterns of evidentiality and enunciation as pertinent
to how we determine author styles on an empirical basis – rather than
talking about e.g. Hemingway’s iceberg technique, Salinger’s zen-like
emptiness the postmodern erotographics of Duras etc., this model is a
tool for investigating questions of style in a comparative mode. Let us
then look at Barthelme’s stories.
The enunciative pattern for ‘On the Deck’ is:
Paragraphs 1-8: Ø > IV > II
Paragraph 9: Ø > IV > II (S(entence)1-2)
Paragraph 9: Ø > IV > III (S3)
Paragraph 9: Ø > IV > II (S4)
We can note the strong Ø > IV > II frequency in the story, which may
seem a bit surprising, since this would be the objective mode of, for
example, Hemingway’s hardboiled prose. Apparently, in terms of
evidentiality and enunciation Barthelme’s text is not as relativized as
his critics would have it. In the last paragraph, possibly, in S3 the
phrase ‘I thought how good it was’ (pragmatically: ‘it felt good’),
there is a delegation to subjectivized experience.
The enunciative pattern for ‘At the Tolstoy Museum’ is:

126 Ulf Cronquist
Again, the voice is objectivized, ‘showing’ rather than ‘telling.’ The

meta-comment in paragraph 7, S3, possibly denotes a polyphonic
moment, Ø > IV > III (a narrator narrating within the narration), but it
is in no way disrupting the epistemic mode.
The enunciative pattern for ‘The Baby’ is:
Similarly here, the pattern of enuniciation is never relativized, for
example the baby’s behavior is displayed but the narrator does not
speculate about her feelings – or his own for that matter! Possibly he
expresses something felt, experiental, in the last paragraph, S2,
parents’ opportunities being ‘each one good as gold.’
To summarize: the enunciation and evidentiality of Barthelme’s
narrators in these three stories should pose no difficulty for the reader.
All three proceed from a once-upon-a-time aphonic platform and there
is the usual mode of suspension of disbelief – all related in a lucid,
epistemic discourse. It is clear that the narrating voices are not
iconoclastic or ‘unreal’ in response to an unreal reality. Further study
in this area could involve more empirical evidence from Barthelme’s
collected works, and, as mentioned above, comparative studies
between different author styles would most probably be an
enlightening project.
6. Narratives
From a cognitive-semiotic perspective there is always a temporal
aspect – i.e. some form of narrative – to be analyzed in a literary text.
A narrative is relatively stable once its structure has been established,
while the semantics of a text is more or less processually spatial
depending on the malleability of imagination. The narratological
model developed by Per Aage Brandt is called diegesis (Brandt 1983,
1989) and it can be summarized in four steps:
1. In the beginning the agent, S1, is in a contractual relationship
with another subject or object, S0.
2. With the appearance of a new subject or object, S2, who
demands attention from S1, there is a crisis between S1-S0-S2.
3. Then with the appearance of a new (further) development, S3,
which demands even more attention from S1, there is an
escalation into a catastrophe.
4. With S4, there is a transformation, a development of S1 > S4

and where the discourse perspective involves an evaluation of
the whole narrative structure. Refer to Figure 2.
Discourse Contract Crisis Catastrophe
S4 S0 S1 S2 S3
Narrative exchange (subjects/objects)

Trans-
formation
Figure 2: Per Aage Brandt: Narrative Diegesis (Brandt 1983, 1989)

As an example, we can consider Hemingway’s short story ‘A Very
Short Story.’ There is an initial contract between ‘He,’ a soldier, and
‘Luz,’ a nurse, in Italy during World War I. The crisis appears as they
quarrel over a future together in America after the armistice. Luz stays
in Italy and makes love to an Italian major; the catastrophe appears in
the form of the break-up. Soon after, ‘He’ then contracts gonnorhoea
from a department sales girl in a taxi cab: S1 transforms into S4 and
possibly the discourse perspective is that the good soldier was not
such a good guy after all.
The diegesis for ‘On the Deck’ concerns the narrator, S1, and his
deck, S0, where a contractual spatial stasis is upheld. As the seasons
suddenly progress fast in the last three paragraphs there is narratorial
crisis. When the narrator then meets ‘the other’ the catastrophe is a
fact: his Olympic view of the world is disrupted, the narration comes
to an end. S1 transforms into S4 through bodily touch with the
possible discourse perspective that every human subject is a subject-
in-process in relation to another human being: ‘no man is an island,’
as Donne put it. Refer to Figure 3.
128 Ulf Cronquist
‘No man is
an island’ Contract Crisis Catastrophe
Subject-in Deck: Seasons: ‘I’ meets

-process stasis Narrator process ‘other’

Bodily
touch
Figure 3: ‘On the Deck’: diegesis.

‘At the Tolstoy Museum’ begins with a contract between the agency of
the narrator, S1, and a possible recipient, S0, in the aphonic mode of
relating what it is like at this museum. The crisis appears as the
texture of words and images turns surreal. And the escalation into the
catastrophe takes place as there is an obvious blur of fact and fiction,
transforming the aphonic agency into an absurd act of delegating
words and images, with the possible discourse perspective being an
ironic comment on Horace’s ut pictura poesis. Refer to Figure 4.
‘Ut pictura
poesis’ Contract Crisis Catastrophe
Word/ Texture: Fact /

Image Recipien Agency surreal Fiction
t

The absurd:
escalating
Figure 4: ‘At the Tolstoy Museum’ diegesis.

In ‘The Baby’ the contractual event is the father’s, S1, parental
relation with his baby, S0, a matter of authority and house rules. The
crisis concerns the baby’s habit of tearing pages out of books, which
is not allowed. As the rules defined by the father are broken by
himself, the catastrophe appears, which includes a volta in the text
when the father discovers/decides that he can do whatever he likes.

Possibly, a discourse perspective illuminates the fact that the father is
as childish himself, and further that the baby’s ‘genuine’ destructive
behavior is in contrast with the adult’s repressive rules, hence the
perspectival Wordsworth quote. Refer to Figure 5.
‘The child is the

father of man’
Contract Crisis Catastrophe
Rule- Tearing Rules

bending Baby Father of pages broken

Escalation:
windshields
Figure 5: ‘The Baby’ diegesis.

These are the overriding narrative structures in Barthelme’s stories,
but it is always possible that a narrative may have several diegeses.
With ‘On the Deck’ we might consider different diegeses for the
different people on the boat, with ‘At the Tolstoy Museum’ we could
consider diegeses for almost each separate paragraph and with ‘The
Baby’ perhaps a narrative exchange with the baby as S1. The purpose
here, however, has been to outline the structure of the texts as global
configurations, and the space here is too limited to detail e.g. the many
possible diegeses in ‘On the Deck.’
7. Blends
Brandt & Brandt (2005b) present a cognitive-semiotic model for
analyzing blends, which is an elaboration of Fauconnier & Turner
(2002) and involves a critique of their model of blending/ conceptual
integration networks especially as regards a ‘generic space.’ An
emergent blend cannot be understood without context and thus Brandt
& Brandt add a relevance space in the diagram. That is, where
Fauconnier & Turner claim that the ‘blend develops emergent
structure that is not in the inputs’ (Fauconnier & Turner 2002: 42),
130 Ulf Cronquist
Brandt & Brandt emphasize that context is analyzable and should

figure in the network diagrams.
It is clear that Brandt & Brandt’s model accounts more lucidly as to
how the semantic processing takes place – their relevance space
specifies why a certain meaning arises in a factually existing context.
Fauconnier & Turner (cf. above), and many of their followers within
cognitive poetics, instead use a kind of opaque metaphoric prose to
describe emergent meaning as (magically) coming together
somewhere in consciousness.7 This does not quite meet the standards
of the discipline of cognitive science, as defined by Hogan above, in
terms of as explicitly as possible mapping out how the information
processing takes place. Brandt & Brandt’s model is also more fully
elaborated as it makes a distinction between a first literal blend and
the emergent blend that is contextually defined. Refer to Figure 6.
Semiotic Space Presentation Reference Space

(Speech Act) Space
Relevance Literal
Space Blend
Emergent Blend
= Meaning
Figure 6: Brandt & Brandt Blending Model (Brandt & Brandt 2005b)
The global blend in ‘On the Deck’ emerges through a network defined
by spatiality in the presentation space with temporality in the
reference space, i.e. the steady rhythm of persons and objects
appearing on the surface of the deck is presented as if there would be a
linear, temporal progression. What we would expect from a story is
thus subverted, and, in fact, the spatial connections themselves are
mainly undetermined – we do not know where on the deck the
different persons and objects are situated, due to the repeated use of
locatives like ‘in front of’ and ‘next to.’ Thus, the literal blend is that
space is time, which may seem mind-boggling. But as the text
introduces a defined temporality in the last three paragraphs, several
season changes, a dynamics of stasis and process comes into focus.
And this dynamics is the meta-textual moment of relevance in ‘On the
Deck,’ the context through which the literal blend produces an
emergent blend where the meaning production concerns the embodied
nature of language: temporal structures demand that the reader move
about in space and spatial structures demand that the reader process
objects in some kind of linear fashion. Refer to Figure 7.
‘There is a lion Space: objects Time: seasons

on the deck...’ on the deck on the deck
Dynamics of Space is
stasis / process Time
Embodiment:
Space / Time
Time / Space
Figure 7: ‘On the Deck’ time/space blend

The global blend in ‘At the Tolstoy Museum’ concerns the narrative:
what is presented as a factual description in time and space blends
with fictional elements of text and images, which makes for a literal
blend where fact is fiction. It is interesting to consider here what
context will structure the network into a meaningful emergent blend.
Generally we could say that it is a question of fragmentation: some
paragraphs of the narration are more or less connected, some are not –
and then there are the images that do not quite fit in, sometimes with
added (mock) ekphrastic qualities. The fact is that the text offers eight
small narratives: paragraphs 1-6 one narrative each, paragraphs 7-9
one narrative, paragraph 10 one narrative. Similarly the images offer
eight small narratives: pictures 1-3 is one narrative, pictures 4-10 one
narrative each. As regards narrative uncertainty and relevance, we are
132 Ulf Cronquist
in a meta-textual context here, i.e. what structures the emergent blend

is the notion of the difference between grand and small stories with
the effect that the emergent meaning is an affirmation of narrative
dissemination. Refer to Figure 8.
‘At the Tolstoy Factual Fictional

Museum we sat..’ narrative narrative
Grand / small Factual =

stories Fictional
Narrative
dissemination
Figure 8: ‘At the Tolstoy Museum’ narrative blend

The global blend in ‘The Baby’ begins with rules for the infant in the
presentation space and rules made by adults in the reference space. A
first, literal blend is produced where the infant becomes an adult
through the authoritarian voice of the parent-narrator, in a suspension
of disbelief the reader will accept – if only initially – that the baby
should not tear books apart and that s/he should be severely punished
by being locked into her room for four hours for each page torn out.
This harsh punishment is a rule made by adults, but not really for
infants – hence the literal infant-adult blend. As regards moral
behaviour and rules, laws, there are, however, always dynamic
relations as to what is accepted and what the punishment should be.
Western law is defined by being subject to change through the
dynamics of interpretation and this is the context that structures the
literal blend into an emergent blend where meaning-production is
focused on the fact that, through procedures in law, values in
themselves in a societal context are subject to change. Refer to Figure
9.
‘The first thing Infant Adult

the baby did…’ rules rules
Moral behavior: Infant =

law dynamics Adult
Values subject
to change
Figure 9: ‘The Baby’ rules’ blend
8. Interpretation
I referred above to Line and Per Aage Brandt’s model for stratified
textual analysis (Brandt & Brandt 2005a). Line Brandt has recently
developed a more elaborate model for stratified reading and
interpretation, with reference to cognitive poetics and cognitive
semiotics (Brandt (forthcoming))8. Here she distinguishes between the
interpretive reading process and the process of literary interpretation.
The initial reading, or the textual/semantic interpretation of the text,
has three levels of semiotic integration: enunciation (who), textual
semantics (what) and literary rhetoric (how). As Brandt remarks this
‘deciphering’ of the text produces a reading where the text is grasped
as a whole. Similarly, I have ventured to map out the global
configurations in Barthelme’s stories above – textual interpretation
focused on enunciation, narratives and blends to produce a reading9.
What remains to consider is literary interpretation, which also, as
Brandt puts it, ‘has the whole text in its scope, based on a reading of
the text, and at this level, the text is interpreted as a meaningful
aesthetic artifact . . . [we are ] asking what the text means as an
expressive whole – in a moral or existential perspective.’
Literary interpretation is thus a question of why and from the
perspective of cognitive semiotics ‘an orientation away from norms in
post-structuralist and culturally oriented literary studies, including
134 Ulf Cronquist
political and historical criticism, e.g. feminist and ‘queer’ readings of

texts, that is, away from particular minds and communities and
towards minds as such: human minds.’ Of utmost importance here is
how the literary mind works. What distinguishes man from other
animals is the ability to understand intentionality and what defines the
reading of a literary text is that the reading mind interprets the text as
an intentional sign, a bounded whole authored by an intentional agent.
By contrast, as Brandt remarks: ‘A computer-generated text would not
solicit the same interpretive attention – unless fictively imagined as
originating in some sort of intentionality.’
Thus, in producing a literary interpretation from a cognitive-semiotic
perspective we do two things: (i) we proceed from the material
gathered in the semantic reading of the text as a whole; (ii) we
consider the text as an intentional sign, as textual intention or aesthetic
intention. In a certain way we can talk of affordances and constraints
here. For example Hemingway’s A Farewell to Arms affords an
intertextual reading, or blend, in relation to Shakespeare’s Romeo and
Juliet, but is entirely constrained to be intertextually analyzed as
blended with e.g. Twelfth Night; Wallace Stevens’ poem ‘Theory’
affords considerations on writing and art in general, but there are
severe constraints on considering the text e.g. in a quantum physics’
context.
Based on the semantic/textual analyses above, what can we then
surmise about Barthelme’s aesthetic intentions and what do these
three stories tell us about human existence? We have seen that notions
of time and space are disrupted in Barthelme’s aesthetics, but also that
the enunciative voice is far less ‘unreal’ than some critics would have
it, and we can also find relatively natural patterns of narrative in the
texts. That is, on the whole Barthelme’s narrators are interested in
telling ‘good stories’ around the campfire. Global blends in the texts
concern classical themes: diegetic stasis/process problematics,
factual/fictional dissemination and moral behaviour in relation to
changing human values, respectively. Thus the two first stories are
decidedly meta-textual, while the third discusses rules and morals. All
three are about change and epistemic and ethic uncertainty.
These stories invite the reader to an aesthetic intentionality that is
poignantly humanistic, rather than inducing a search for postmodern
‘floating signifiers.’ Heraclit’s panta rei is applicable in most of what
we value as good and carefully crafted literature, but this does not
mean that we cannot present relatively stable interpretations – literary
interpretations that are grounded in and motivated by our first
semantic reading. If there is but one discourse perspective in relation
to Barthelme’s writing it has to do with human existential awareness
spiced with gentle satire and warm humour. Cognitive semiotics,
based in cognitive science – the analysis of information patterns,
cognitive architecture and algorithmic sequencing – is also a
humanistic venture, grounded in the craft of reading, and also includes
ludic perspectives, which I hope to have contributed to in the above.
Endnotes
1
As Line Brandt puts it: ‘Meaning, within cognitive semiotics, is taken to refer to the
signified (signifié) side of signs occurring in communication and other expressive
practices, and ‘construction’ is taken to be a mental endeavor engaging multiple
minds, as the exchange of signs (semiosis) is essentially an intersubjective enterprise’
(Brandt 2006).
2
More specifically, my method of analysis proceeds from the research program in
cognitive semiotics initiated and developed by Per Aage Brandt over the last 10-15
years at the Center for Semiotics, Aarhus University, Denmark.
3
For a more detailed exposition on stratified textual analysis see Brandt
(forthcoming), ‘Cognitive Poetics from a Semiotic Perspective: A Stratified Model for
Reading and Interpretation.’ I return to this article below, in the section on
interpretation.
4
Hogan (2003: 29-34), describes the three steps of the working process within
cognitive science with reference to Michael R. W. Dawson’s Understanding
Cognitive Science (Oxford: Basil Blackwell, 1998).
5
I refer in this and the following paragraph to Brandt: 2004: ‘The Semantics of
Diagrams,’ pp. 87-102.
6
Brandt remarks that most enunciative values are probably based on only two or three
delegations.
7
For this opaqueness see e.g. Stockwell 2002: 98: ‘‘running the blend’ through its
emergent logic’; Crisp 2002: 111: ‘the composed input spaces are drawn on to create
a single fantastic imaginary space’; Semino 2002: 115: ‘It is in the blend that,
according to blending theory, meanings are generated.’
8
The three quotes below are from Ms Brandt’s manuscript, p. 3, 2 and 4 respectively,
and can be obtained from her at cogsemlb@yahoo.com.
9
If, to the impatient reader, my reading seems reductive, I must remind her/him that it
is based in textual/semantic procedures that cannot, per definition, be relativized.
However, my generalisations can of course be disputed, bettered, falsified.
References
Barthelme, Donald. 1994. 40 Stories. New York: Penguin.
136 Ulf Cronquist
Brandt, Line. 2006. ‘Dramatization in the Semiotic Base Space: A Semiotic Approach
to Fictive Interaction as a Representational Strategy in Communicative Meaning
Construction’ in Oakley, Todd, and Anders Hougaard (eds) Mental Spaces
Approaches to Discourse and Interactio. Amsterdam: John Benjamins. See also:
http://www.hum.au.dk/semiotics/docs2/pdf/brandt_line/dramatization.pdf
__. (forthcoming). ‘Cognitive Poetics from a Semiotic Perspective: A Stratified
Model for Reading and Interpretation’ in Brône and J. Vandaele (eds)
Applications of Cognitive Linguistics. Berlin & New York: Mouton de Gruyter.
Brandt, Line, and Per Aage Brandt. 2005a. ‘Cognitive Poetics and Imagery’ in
European Journal of English Studies Vol. 9, No. 2 August 2005: 117-130. See
also:
http://www.hum.au.dk/semiotics/docs2/pdf/brandt&brandt/cognitive_poetics.pdf
Brandt, Line, and Per Aage Brandt. 2005b. Making Sense of a Blend – A Cognitive-
Semiotic Approach to Metaphor. Annual Review of Cognitive Linguistics 3, 2005:
216-249. See also:
http://www.hum.au.dk/semiotics/docs2/pdf/brandt&brandt/making_sense.pdf
Brandt, Per Aage. 1983. Sandheden, saetningen og döden – Semiotiske aspekter af
kulturanalysen. Copenhagen: Basilisk.
__. 1989. ‘Genese og diegese – Et problem i den almene narratologi.’
Religionsvidenskabeligt tidsskrift, vol 14: 75-85.
__. 2004a. Spaces, Domains and Meaning. Essays in Cognitive Semiotics. Bern: Peter
Lang.
__. 2004b. ‘Evidentiality and Enunciation. A Cognitive and Semiotic Approach’ in
Arrese, J.M. (ed.) Perspectives on Evidentiality and Modality. Madrid: Editorial
Complutense. Columbia Encyclopedia, The, Sixth Edition, 2004. New York:
Columbia University Press.
Crisp, Peter. 2002. ‘Conceptual Metaphor and Its Expressions’ in Gavins, Joanna, and
Gerard Steen (eds) Cognitive Poetics in Practice. London: Routledge: 99-113.
Dawson, Michael R. W. 1998. Understanding Cognitive Science. Oxford: Basil
Blackwell.
Fauconnier, Gilles, and Mark Turner. 2002. The Way We Think. Conceptual Blending
and the Mind’s Hidden Complexities. New York: Basic Books.
Gavins, Joanna, and Gerard Steen (eds). 2003. Cognitive Poetics in Practice. London:
Routledge.
Hogan, Patrick Colm. 2003. Cognitive Science, Literature and the Arts. A Guide for
Humanists. New York and London: Routledge.
Semino, Elena. 2002. ‘A Cognitive-Stylistic Approach to Mind Style in Narrative
Fiction’ in Semino, Elena, and Jonathan Culpeper (eds) Cognitive Stylistics.
Language and Cognition in Text Analysis. Amsterdam: John Benjamins: 95-122.
Semino, Elena, and Jonathan Culpeper (eds). 2002. Cognitive Stylistics. Language
and Cognition in Text Analysis. Amsterdam: John Benjamins.
Sloboda, Nicholas. 1997. ‘Heteroglossia and Collage: Donald Barthelme's “Snow
White”’ in Mosaic 30(4).
Stockwell, Peter. 2002. Cognitive Poetics. An Introduction. London: Routledge.
Zurbrugg, Nicholas. 1993. The Parameters of Postmodernism. Carbondale: Southern
Illinois University Press.
Evoking Interest, Evoking Meaning: The Literary
Theme and the Cognitive Function of Stylistic Devices
Alfonsina Scarinzi
Georg - August Universität Göttingen, Germany
Abstract
In the field of thematics the literary theme of a literary text is usually considered to be
an abstract situation formulated as a declarative statement. The process of
understanding and interpretation of the literary theme while reading a literary text
depends on the cognitive operations of the human mind. Discussing the state and the
cognitive role of stylistics in the field of thematics and considering the role of
‘literariness’ and of cognitive science within literary studies, this article focuses on the
question how stylistic devices can affect and guide the cognitive pathway that leads
the reader to the cognitive process of thematic understanding and thematic
interpretation. It discusses how the interaction of form and content of a literary text
with the reader’s background knowledge and the reader’s emotional cognitive activity
can affect the mental pathway leading to the attribution of thematic meanings. Several
studies are cited. Following cognitive studies in the field of reader response, it is
suggested that the thematic understanding process and the thematic interpretation
process are based on a control element that guides the whole thematic pathway that
leads the human mind to the abstraction of the literary theme. It is argued that stylistic
devices help to evoke this control element in both the understanding and thematic
interpretation process.
Keywords: thematics; literariness; reader response; stylistics; cognitive mental
operations; interpretation; understanding; emotions.
1. Introduction
In her article ‘Responding to Style: Cohesion, Foregrounding, and
Thematic Interpretation.’ Catherine Emmott (2002) argues that
stylistic devices can convey the literary theme of a literary text and
stimulate an ‘affective response’ on the part of the reader, appealing to
his emotion. In this article, I will discuss this cognitive function of
stylistic devices in the thematic understanding and interpretation
process of a literary text based on the interaction of the reader, of his
mental operations and of his background knowledge of the literary
text. I will not concentrate on a linguistic analysis of stylistic devices.
I will accommodate the discussion on the cognitive function of
stylistic devices within a theoretical discussion on the development of
a theoretical framework for the abstraction and interpretation of the
138 Alfonsina Scarinzi
theme of a literary text. I will concentrate on the cognitive mechanism

of the reader’s mind which stylistic devices can trigger while the
reader interacts with a literary text.
The first question that arises here is how the state of stylistics is in the
field of thematics. Certainly a promising one for future studies on
literary themes, but not a well-founded one, for thematics is a field
that is still looking for its own methodological framework:
[...]thematics is a rather undisciplined discipline, beset with
subjective strategies and terminological disputes;[...] what is
needed is a methodological framework, theory or a set of
theories[...](Bremond & Landy & Pavel, 1995: 1)
In contrast to Bremond & Landy & Pavel (1995), Max Louwerse and
Willie van Peer (2002) are more optimistic. Instead of considering
thematics an undisciplined discipline, they prefer to regard this field
as an interdisciplinary discipline and show how such an approach can
be productive for this literary field. Even if thematics still lacks a
metalanguage and there is still little consensus about what exactly a
literary theme is, their anthology shows that especially ‘going
cognitive’ (Hamilton & Schneider, 2002: 640) in the approach to the
literary theme can be a first step toward the development of a theory
for thematics. By cognitive approach to literature I mean the stream
within literary studies that focuses on the mental processes of the
individual reader who creates literary meaning while interacting with
the literary text (Tsur, 2002).
Against such a background, stylistics can contribute to the
development of a theoretical framework for thematics and to find an
answer to the questions that can convey a literary theme and how the
literary aspect of a theme can be conveyed, being the literary aspect of
a theme linked to the concept of ‘literariness’ that is also determined
by stylistic variations (see Miall & Kuiken 1999).
Before explaining what I will discuss in the next two sections, I will
state more precisely in the following passage what exactly is meant in
this article by a cognitive approach to the thematic understanding and
interpretation process.
According to the cognitive approach to literature and literary
interpretation, thematic meanings are the result of the interaction
Evoking Interest, Evoking Meaning 139
between the reader’s background knowledge and the literary text. This
is based on the cognitive assumption that meanings are not contained
within the text but are constructed in the interaction between the text
and the reader’s background knowledge. The text and the reader’s set
or system of beliefs enable the process of interpretation (Hobbs, 1990;
Semino, 1997; Lásló, 1999). Thematic meanings are the product of a
knowledge–based inference process on the part of the reader (Lásló,
1999). The result of this process, called ‘thematic abstraction’
(Graesser & Pomeroy & Craig, 2002; Zwaan & Radvansky &
Whitten, 2002), is conditioned by the extent to which the reader shares
the cultural background knowledge implied in the text and leads to a
work’s individual thematic interpretation by the reader (Semino, 1997;
Meister, 2002; van Peer, 2002). In terms of the well-known schema
theory, this means, according to Elena Semino (1997), that:
[…] text worlds are cognitive constructs that arise in the
interaction between readers and the language of the texts. […]
The sum of the reader’s existing schemata makes up a skeleton
of that person’s model of reality […] which serves as a frame of
reference in the construction and evaluation of text worlds. The
way in which a particular reader will perceive a particular text
world will depend on how his or her various instantiated
schemata interact with one another in comprehension, and on
whether the reader’s current model of the world is reinforced or
challenged in the process. (Semino, 1997: 161)
As Hoppe-Graff & Schell (1989) point out, comprehension is a
constructive process that is not restricted to the extraction of meaning
from the linguistic message alone. Comprehension is a process of
looking for meaning, the tendency to make sense of events in the
world. In trying to make sense of a literary text, the reader always
relies on his knowledge of the world. The elements of world
knowledge are concepts that are functionally interwoven and therefore
interact in cognition (see also Hoppe-Graff & Schell, 1989: 100).
This interaction in cognition reminds one of Roger Schank’s (1982)
concept of understanding. By understanding, Schank means in his
Dynamic Memory (1982) being reminded of the closest previously
experienced phenomena. While processing a new input we cannot
help but pass through old memories. Adapted to the readers of a
literary text and to the interaction between the reader’s background
knowledge and the literary text, this means that there must be in the
literary text elements which remind the reader of previous experiences
in order to be able through analogical reasoning to understand the
theme of the text. In view of such an interaction, we can assume that
inferences leading to the literary theme also result from the fact that
particular elements in the text trigger the activation of certain
schemata in the reader’s mind and that the activated schemata
generate expectations that fill in what is not explicitly mentioned in
the text (Cook, 1994; Semino, 1997).
From a cognitive point of view, the mental activities of the reader
leading to the thematic interpretation of a text cannot be extended ad
infinitum. A control element is needed. While Schank (1979) focuses
on a control element installed in the knowledge structure of the reader
that reminds the reader of previous knowledge and that enables the
reader to recognize the deviation from such knowledge during
reading, such as the concept of unusualness, van Peer (1992) stresses
the fact that the text and its stylistic foregrounding devices
pragmatically control its interpretation. Van Peer (1992: 139) points
out that foregrounding refers to the fact that literary texts, by making
use of some special devices, direct the attention of the reader to their
own formal and semantic structure. Some parts of the text are thereby
promoted into the foreground. These textual locations are given more
attention, and in the reader’s perception they play a relatively more
important role in the act of interpretation (see also Short, 1973;
Verdonk, 2002). In this article, I will start from the assumption that
Schank’s approach and van Peer’s approach complement one another.
The former focuses on the process of understanding and the latter on
the process of interpretation. I will argue that the combination of these
two processes leads to the formulation of the literary theme of a
literary text. I will discuss the mental mechanism that is triggered off
by stylistic devices in the process of thematic understanding and
thematic interpretation.
In section 2, I will focus on the cognitive mechanism of evoking
interest in readers and on its cognitive function in the thematic
meaning construction. In this section I will just touch up on the
cognitive role of stylistic devices. In section 3, I will concentrate on it
and on its role for the concept of ‘literariness’. I will explore the
cognitive mechanism of evoking thematic meaning in the activation of

the pathway that leads to the thematic interpretation of a literary work.
2. Interestingness, foregrounded striking situations and the

concept of ‘meaning maximizing’
How can we define the concept of literary theme? In the field of
literary studies there is little consensus about what exactly a literary
theme is. A discussion on how a literary theme can be defined and on
the many definitions of ‘theme’ existing in literary studies is beyond
the scope of this article. In this article I will follow the definition
proposed by WordNet®, the database maintained by the Cognitive
Science Laboratory at Princeton University, and the observations on
the characteristics of literary themes proposed by van Peer (2002).
According to WordNet®, a theme can be defined as motif. A theme–
motif is the moral of a story, an abstract situation formulated as a
declarative statement. Hence, for example, ‘romantic ideas’ in
Madame Bovary cannot be a theme-motif, but ‘romantic ideas are
inadequate for everyday life’ can (see also Graesser & Pomeroy &
Craig, 2002; Zwaan & Radvansky & Whitten, 2002).
According to van Peer (2002: 255-256), literary themes are
characterized by these issues: 1. they must be relatable to human
activity of a non-routine character and are involved in human interests
and concerns; 2. they are emotionally charged; 3. they are
foregrounded.
These characteristics for literary themes remind us of the interaction
between the literary text and the cognitive activity of the reader in the
thematic interpretation process of a literary text. They are the starting-
point for the exploration of cognitive and stylistic control elements in
this article.
A concept used by van Peer (2002) points to the interaction of the
cognitive activity of the reader with the text and its stylistic devices:
meaning maximizing. This is defined as:
The tendency of literary themes to be linked to concerns that
most readers will recognize as of significant importance to the
persons involved. (van Peer, 2002: 255)
The example that van Peer gives is when in Kafka’s The
Metamorphosis Gregor Samsa finds himself changed into a large
insect. This unusual concern is of significant importance to Samsa and

the reader tends to maximize the significance of this event in the life
of the character. Van Peer’s (2002) proposal to regard the literary
theme as linked to foregrounded, striking situations implies an
extension of the foregrounding theory: instead of concentrating on
deviations on the levels of phonology, grammar and semantics, the
analysis of deviations of general cultural knowledge seems to be
required (see also van Peer, 2002: 255, note 81).
To return to Kafka’s The Metamorphosis, it can be argued that
unusualness in the narrated concerns triggers off the meaning
maximizing. But the question now is what Gregor Samsa’s
transformation into a large insect tells us about the theme of Kafka’s
The Metamorphosis being the moral of the story. It tells us nothing
about the moral itself. But the unusual concern can be considered to
be a trigger of a cognitive pattern and the first step towards the
pathway leading to the thematic interpretation of the literary text.
As Schank (1979) argues, unusual things are for humans more
interesting than usual ones. As, according to Schank’s Dynamic
Memory (1982), understanding means being reminded of the closest
previously experienced phenomena and is based on analogical
reasoning, the control element in the understanding process leading to
the thematic interpretation of the text can be only outside the text, in
the knowledge structure of the reader. The mental activity of the
reader who recognizes the similarities of experienced phenomena is
the ability to create levels of identification of similarities between
concepts. This is called in the cognitive approach to literature
‘attribute mapping’ (Freeman, 2002: 468). This kind of cognitive
mapping is the ability of human beings to articulate their thoughts
from the present reality spaces into hypothetical situations. Attribute
mapping is the perception or creation of similarities between objects.
In my view, in the interaction of the reader with the literary text,
meaning maximizing and, for example, the identification of the reader
with the character of the story are based on analogical reasoning. They
can occur only if the reader recognizes a concern in the story of the
character as familiar, as usual according to one’s own knowledge
structure and previous experiences, they can occur only if there is
knowledge that something normally happens, and if the reader
recognizes the unusualness present in the text as the trigger of an
expectation failure (Schank, 1982; Cook, 1994; Lásló, 1999), that is as

a trigger of a deviation from something familiar, from existing cultural
schemata. This experienced deviation is in my view emotionally
charged. It is, hence, likely that the reader’s individual level of
knowledge and experience in a literary context guides the thematic
interpretation, which is likely subjective.
In the thematic abstraction and interpretation in a literary context,
experience within the literary system (see Meutsch & Schmidt, 1985)
plays an important role in order to identify the literariness of a literary
theme, for the literary theme is considered to have an extra-literary
origin. Menachem Brinker (1995) points out that the literary aspect of
a theme depends upon our familiarity with literary texts. He argues:
‘Themes’ are loci where artistic literary texts encounter other
texts: texts of philosophy, or the social and human science, texts
of religion and social ideologies, journalistic texts, including
gossip columns, and personal texts such as diaries and letters.
The various degrees to which a specific theme obviously
dominates and generate other elements of an individual work
indicate the various degrees to which the work allows, seeks, or
resists consideration as an artistic equivalent of nonartistic texts.
(Brinker, 1995: 36).
Experience in a literary context is linked to the concept of
‘literariness’. Cook (1994) defines ‘literariness’ as deviation at the
level of language and text. It poses a challenge to the reader’s
schemata and may result in schema change. A schema or schema
knowledge is defined as background knowledge related to a particular
type of object, person, situation or event. Each schema contains
generic information about the content elements of a particular domain
and about the relationships between such elements. A ‘Washing
Clothes’ schema, for example, will include information about things
that are used to deal with dirty clothes, about the way in which such
elements relate to one another, and about the sequencing of the actions
in which one has to engage. The content of schemata will vary from
individual to individual, and from culture to culture. The activation of
schemata allows us to make predictions and to draw inferences in the
process of comprehension (Semino, 1997). According to Cook’s
concept of literariness, a schema change may involve the destruction
of old schemata, the creation of new ones, or the establishment of new
connections between existing schemata. This is based on the claim

that the act of understanding does not simply consist in the
reactivation of existing conceptual schemata. It always implies
change. A text world may be regarded as deviant or alternative
because it presents a view of reality that goes against the reader’s
existing assumption. Literary texts correspond to deviation at the level
of the background knowledge and go against the reader’s existing
assumption.
As Semino (1997) observes, Cook’s central claim is that literary
discourse performs the cultural function of creating the conditions for
schema change. This is due to the fact that literature is a type of
discourse with no immediate practical or social consequences. As a
result, it provides readers with an opportunity to reorganize their
schemata without the risk of incurring socially inconvenient practical
consequences (Semino, 1997). This experience of deviation while
reading a literary text can likely characterize a literary experience. The
deviation at the level of language and text from the background
knowledge of the reader creates unusualness. Schank (1979) points
out that the mental pathway readers follow to understand a story is
controlled by interestingness. When readers find something in a text
which is unusual they recognize it as being deemed to be of interest.
They pay attention to it. It seems to be convincing that the dominant
element controlling the story processing and the pathway that leads to
the thematic interpretation of a literary text remains interest. Along the
interesting pattern of events, readers give free rein to their thematic
inference process. Interest is considered by Tan (1994) to be a special
instance of anticipatory emotion in general. If interest consists of an
inclination to further elaborate the stimulus in anticipation of the final
interpretation of the text, it is expected that interesting elements have a
higher level of probability of being included in the end of the path
leading to the thematic interpretation and hence, in my view, in the
formulation of the theme of a literary text.
Against this background, van Peer’s (2002) concept of meaning
maximizing as reader response to the presentation of a deviation from
expectations can be considered to be a reaction to anticipation in the
story that can be included in the final interpretation of the text. To
return to the previous example given by van Peer (2002), the textual
description of a person in a room does not present the theme of a
literary text, but the description of a person in a room who finds

himself transformed into an insect does (van Peer, 2002: 255). The
question is what such a deviation from routine activities tells us about
the theme of the text, for, as was said, the transformation into an insect
can not be considered to be the moral of Kafka’s The Metamorphosis.
Such a deviation is a kind of foregrounding device aimed at evoking
interest in the reader. Meaning maximizing seems to be a response to
foregrounding devices in the text or, according to van Peer (2002:
255), to a foregrounded striking situation corresponding to an
expectation failure and hence, in my opinion, not a response to the
literary theme itself.
What does the control function of foregrounding devices in
understanding and interpreting the theme of a literary text consist of?
That foregrounding devices in a literary text contribute to the control
of the pattern of understanding and interpretation of a literary text
refers to the fact that they help create the meaning of a sentence (Miall
& Kuiken, 1994). In his investigation, van Peer (1986) points out that
regardless of their prior level of knowledge, readers respond to
foregrounding devices – meaning in this case by foregrounding a
deviation from standard stylistic practices – and that foregrounding
strikes readers’ interest. As it is the activation of interestingness that
leads to control the thematic pattern in text understanding, and not the
text itself, foregrounding devices in the text can be considered to be
elements that activate interestingness and hence elements that trigger
off the control element of the cognitive thematic interpretation pattern.
As foregrounding triggers off interest, which is a special instance of
anticipatory emotion, it can be hence considered to be emotionally
charged.
In the next section I will concentrate on the cognitive role of stylistic
devices and on thematic meanings.
3. Evoking thematic meaning and responding to style: readers’

emotions and the literariness of the thematic pathway
Not only as a trigger of an anticipatory emotion, but also as a
stimulation of an affective response on the part of the reader,
foregrounding can contribute to the activation of the thematic pathway
in the reader’s mind. Emmott (2002: 93-94) points out that this aspect
of stylistic foregrounding can have thematic significance only if the
reader succeeds through this in projecting himself in the described

situation. ‘Madness’, for example, can be conveyed in a text by an
explicit statement, by plot actions or by portraying the state of mind
(Semino & Culpeper (eds), 2002) of an insane person by using, for
example, excessive linguistic repetition. Emmott (2002) points out
that the repeated use of a particular referring expression, that is of
linguistic expressions used when referring to an entity, can produce
stylistic foregrounding effects which may contribute to convey the
theme of a story, that is ‘madness’, for example. The stylistic
foregrounding appeals to the reader’s emotion. But to enable this, the
reader’s schema knowledge of ‘madness’ is needed (Semino, 1997;
Emmott, 2002). Schema knowledge enables the readers to make a link
between the psychological state of a person and the style used in the
text, for example.
The repetitive style in the description of ‘madness’, according to
Emmott (2002: 95), can be considered a stimulus for the unstable
psychological state of mind of the described person and will trigger
the reader’s knowledge about such a state. But she does not consider it
to be an unusual element. The weakness of Emmott’s (2002) approach
is, in my opinion, the fact that she does not consider the thematic
interpretation as a process where the moral of a story is abstracted by
the reader and formulated as a declarative statement, such as
‘romantic ideas are inadequate for everyday life’ in Madame Bovary,
for example. Considering ‘madness’ as a theme reminds one of
Schank’s (1982) concept of theme and of his TOPs – thematic
organization points – in memory. According to Schank (1982),
thematic information is independent of most content features of the
text. A theme provides background information about the origin of
people’s goals. Semino (1997: 140) gives the following example: if
we know that someone is a vegetarian Animal Rights campaigner we
have no difficulties dealing with the news that he or she is
demonstrating outside an abattoir. Thematic knowledge structures
represent patterns of goals and plans in an episode in memory. Their
form and content depend on their use and they are based on personal
experiences. This is the reason why TOPs structures formed by any
individual are idiosyncratic. The commonalities of experiences in a
variety of settings determine the organization of the TOPs in memory.
They capture similarities of experiences in memory and allow
analogical reasoning. (Seifert & Abelson & McKoon, 1986). Hence,
‘madness’ can be considered to be the concise point of a narrated

episode conveying elements which readers recognize as characteristic
of their TOP ‘madness’. ‘Madness’ as TOP is characterized by a
sequence of actions based on goal and plan: it has the pattern of the
actor being psychologically insane, giving rise to his goal of
expressing his disease. This says nothing about the cognitive process
leading the reader to the interpretation of the moral of a whole literary
text.
It seems that Catherine Emmott’s (2002) approach to the concept of
theme fails to account for the shift from a sequence of actions, that is a
sequence of TOPs, to an underlying theme, to an underlying abstract
situation conveying a moral (see also MacKenzie, 1987). The
interaction of stylistic deviations and of the characteristics of literary
themes – meaning by theme the moral of a literary text – as elements
being linked to unusual concerns and hence emotionally charged is not
taken into account by Emmott (2002). She takes an unusual concern,
such as ‘madness’, into account, but not how such an unusual issue
can be linked to an underlying literary theme and interact with it. In
my opinion, the analysis of such an interaction in studying stylistic
devices and literary themes might be promising.
Being a literary theme, an abstract situation formulated as a
declarative statement, it is likely that in order to identify it, the reader
thinks abstractly while searching for the aboutness of what he reads.
The pathway leading to the moral of a story is likely based on thinking
abstractically. Like Schank (1979), Emmott (2002) as well points to
unusualness as the trigger of the activation of a schema. She
concentrates on stylistic unusualness. On the other hand, even if
Emmott (2002) recognizes foregrounding as having the function of
evoking a schema for an extreme emotional state she does not
recognize foregrounding achieved through repetitive style as having a
defamiliarization effect, that is the effect of making things strange in
order to make us see them anew (Cook, 1994). Defamiliarization
being departure from expectation, a deviation from a norm (Cook,
1994: 138), the repetitive style can be considered in my view certainly
as having a defamiliarization effect, for the repeated use of some
expressions can be considered to be unusual. Hence, it seems likely
that also from a stylistic point of view evoking a schema for an
extreme emotional state passes through unusualness and hence
through evoking interest as a control element of the cognitive pathway

of interpretation.
As Miall (1995) points out, neuropsychological evidence also supports
the view that readers need a control system to cope with unusualness,
such as surprise or contradiction, in a literary text. The prefrontal
cortex has the neurobiological function to control the response to
varying stimuli and organize the information. It has the function of
forming a programme for monitoring and searching memory. The
activation of schemata for emotions triggered off by stylistic
foregrounding is likely neurobiologically controlled by the prefrontal
cortex as well as the role of feelings in the thought process (Miall,
1995: 282). The response to foregrounding devices passes through
affects and feelings, not only through interest as anticipatory emotion.
The fact that the empirical studies conducted by van Peer (1986) and
Miall & Kuiken (1994) on readers’ response to foregrounding show
that response to foregrounding is independent of literary training or
experience provides evidence that all readers competent in the
language appear to possess a basic sensitivity to it. This is supported
by the fact that the prefrontal cortex in all human beings independent
of their knowledge about stylistics is responsible for anticipation in
the process of meaning construction. The anticipatory function is
created and supported by feelings. Miall & Kuiken (1994) concentrate
on the defamiliarization effects and on the emergence of affect by
studying readers with different levels of literary competence and
interest. They support the view that literary response follows a
distinctive course in which foregrounding prompts defamiliarization,
defamiliarization evokes affect, and affect guides ‘refamiliarizing’
interpretive effort. In my view, it is likely that stylistic foregrounding
through stylistic deviation contributes to triggering off the reader’s
socio–cultural experiences that form the meaning of a literary text
according to the socio-cultural knowledge invoked in the text (Lásló,
1999).
Foregrounding devices affect the emotional response to literary
content and contribute towards guiding literary interpretation. In this
respect, foregrounding and its devices are triggers of emotional
experiences. Their contribution to the thematic interpretation process
supports the assumption that in the process of meaning attribution the
human brain and its ability to cope with unusualness play an important
role (see Turner, 1994). I suppose that thematic interpretation occurs

along the lines of the triggered emotional experiences. It is likely that
emotions have a control role in the process of thematic interpretation
and that, in such a process, ‘literariness’ also has an emotional side
that plays a role in the activation of a theme in the reader’s mind.
Regarding the emotional experience of a person in a certain situation
as the result of the way a person assigns meaning to that situation,
Kneepkens & Zwaan (1994) point out that the emotional experience
can influence cognitive processing and trigger cognitive structures that
are characteristic of given emotional experiences. Emotions help
people to determine what knowledge is relevant to the situation and
what knowledge has to be activated. Kneepkens & Zwaan (1994) link
a form of deviation in the events and situations of the story to readers’
expectations. When these are disturbed, surprise and interest are
evoked. This supports from a cognitive point of view what was said
about van Peer’s (2002) concept of meaning maximizing. Being this,
an experience triggered off by a disturbed expectation, it can be
considered to be the result of a cognitive emotional experience.
Emotional involvement and meaning maximizing are controlled not
only by interestingness but also by emotional patterns. These are
called, by Tsur (1992), emotional archetypal patterns. The emotional
archetypical pattern lends significance to elements that occur in a
story or poem and is in the back of one’s mind. Tsur (1992) points out
that an archetypal pattern is brought into existence when it comes into
contact with corresponding elements of the verbal context. In terms of
literary structures, archetypal patterns are regulative concepts which
emphasize certain elements of the pattern and relegate the poorly
articulated ones. In this way, they create a hierarchic system. The
archetypal pattern has just an organizing regulative effect in the sense
that, through having an associative cluster, it puts in the foreground
certain aspects of a structurally weak verbal texture, while some
others are backgrounded (Tsur, 1992). A kind of emotional archetypal
pattern is, according to Tsur (1992), the David–and–Goliath
archetype. The ‘operation of foregrounding’ (Tsur, 1992) is regulated
by knowledge about emotional patterns. Hence, the isolation of the
pattern readers decide to follow seems likely to depend on the type of
knowledge about patterns being in the back of one’s mind. The type of
knowledge readers have determines what triggers off interest in
readers, so that the possible pathways leading to the thematic

interpretation correspond to the level of readers’ knowledge and
experience in a literary domain.
The assumption that emotions play a role in literary thematic
interpretation seems not to be compatible with Cook’s concept of
literariness. In his concept, emotions are not taken into account. Miall
& Kuiken (1999) accommodate emotions in their definition of
literariness. According to them, feeling appears to implicate the
reader's self concept and to provide a route to specific issues relating
to the self, as well as to experiences or memories that may provide a
new interpretive context following the moment of defamiliarization.
Their main claim is that while all readers appear to be sensitive to
foregrounding in a literary text their construals of its meaning often
differ widely. It is in this respect that literature seems to invoke an
individual response and that we can assert that a literary text may
have, hence, more than one thematic interpretation. According to
Miall & Kuiken (1999), stylistic features or striking features due to
narrative, and the reader's defamiliarizing response to them, are
necessary but insufficient to identify literariness. They are just part of
literariness. The reader’s defamiliarizing response to them can be
considered to be an emotional response. In this sense, readers can
respond emotionally to a category of literariness without identifying
the theme. The third component of literariness, according to Miall &
Kuiken (1999), is constituted by the reader's attempts to articulate the
phenomena within the text that are found striking and evocative of
feeling. The modification or transformation of readers' concepts or
feelings is specific to the individual reader: it is in this respect, indeed,
that literature seems to invoke what is individual in readers. In my
opinion, the identification of the pathway leading to the formulation of
the moral of the text is likely determined by this third element. It is
likely that the formulation of the theme of Madame Bovary ‘romantic
ideas are inadequate for everyday life’ occurs thanks to the interaction
at a cognitive level of stylistic features striking interest, reader’s
defamiliarization response to them and the evocation of feelings in
readers. This interaction contributes to the determination of the
pathway readers follow to formulate the moral of a literary text.
4. Conclusion
In the field of thematics, the cognitive role of stylistic devices seems
to be a promising one in the attempt to develop a theoretical
framework for the analysis of cognitive processes that lead the reader
to the literary theme. Even if the analysis of the form of a literary text
helps to explain what kind of cognitive operations readers follow in
the process of thematic meaning attribution, stylistics is still relegated
to a role of secondary importance within thematics. Stylistic devices
evoke interest in the reader and hence activate the cognitive pattern
for text processing. In the pattern readers follow to abstract and
interpret the literary theme of a literary text, interest is a cognitive
control element and the starting-point for activating an emotionally
charged knowledge-based cognitive pathway that determines the
interpretation of the ‘aboutness’ of a literary text being the emotional
experiences of a person in a certain situation, the result of the way a
person assigns meaning to that situation.
In order to find out how readers control their thematic interpretation
process in a literary context, concentration on literary experience and
on literariness is necessary. Without the element ‘literariness’ and
without focusing on the reader’s literary competence I doubt that any
contribution to the study of the cognitive pathway leading to literary
thematic interpretation can be promising for the study of literary
themes. It is, in my view, concentration on the emotional experience
as part of literariness that enables the exploration of the expectation
failure, of the deviation from existing background knowledge about
language use and emotionally charged schemata, of meaning
maximizing as the cognitive starting-point for the thematic
interpretation of a literary work.
Stylistic devices provide the pathway to literariness and hence to the
literary aspects of themes. The development of an approach to
stylistics aimed at explaining the formulation of literary themes as
abstract situations and as the moral of a story can be but a valuable
addition to thematics.
References
Bremond, C., J. Landy and T. Pavel (eds). 1995. Thematics. New Approaches.
Albany: State University of New York Press.
Brinker, M. 1995. ‘Theme and Interpretation’ in Bremond, C., J. Landy and T. Pavel
(eds) Thematics. New Approaches. New York: State University of New York
Press: 33-44.
Cook, G. 1994. Discourse and Literature: The Interplay of Form and Mind. Oxford:
Emmott, C. 2002. ‘Responding to Style: Cohesion, Foregrounding, and Thematic
Interpretation’ in Louwerse, M., and W. van Peer (eds) Thematics.
Interdisciplinary Studies. Amsterdam/Philadelphia: John Benjamins Publishing
Company: 91- 118.
Freeman, M. H. 2002. ‘Cognitive Mapping in Literary Analysis’ in Style 36: 466-
483.
Graesser, A., V. Pomeroy and S. Craig. 2002. ‘Psychological and Computational
Research on Theme Comprehension’ in Louwerse, M., and W. van Peer (eds)
Thematics. Interdisciplinary Studies. Amsterdam/Philadelphia: John Benjamins
Publishing Company: 19-34.
Hamilton C. A. and R. Schneider. 2002. ‘From Iser to Turner and Beyond: Reception
Theory Meets Cognitive Criticism’ in Style 36: 640- 658.
Hobbs, J. 1990. Literature and Cognition. Stanford, CA: Center for the Study of
Language and Information.
Hogan, P. C. 2003. Cognitive Science, Literature, and the Arts. A Guide for
Humanists. New York and London: Routledge.
Hoppe-Graff, S. and M. Schell. 1989. ‘The Comprehension of Literary Texts:
Developmental Considerations’ in Meutsch, D., and R. Viehoff (eds)
Comprehension of Literary Discourse: Results and Problems of Interdisciplinary
Approaches. Berlin: de Gruyter: 89-110.
Kneepkens, E. W. E. M., and R. A. Zwaan. 1994. ‘Emotions and Literary
Comprehension’ in Poetics 23: 125-138.
Lásló, J. 1999. Cognition and Representation in Literature. The Psychology of
Literature. Budapest: Akadémiai Kiadó.
Louwerse, M. and W. van Peer (eds). 2002. Thematics. Interdisciplinary Studies.
Amsterdam/Philadelphia: John Benjamins Publishing Company.
MacKenzie, I. 1987. ‘Narratology and Thematics’ Modern Fiction Studies 33(1): 535-
544.
Meister, J. C. 2002. ‘Parsing for the Theme. A Computer Based Approach’ in
Louwerse, M., and W. van Peer (eds) Thematics. Interdisciplinary Studies.
Amsterdam/Philadelphia: John Benjamins Publishing Company: 407- 432.
Meutsch, D. and S. J. Schmidt. 1985. ‘On the Role of Conventions in Understanding
Literary Texts’ Poetics 14: 551-574.
Miall, D. S. and D. Kuiken. 1994. ‘Foregrounding, Defamiliarization, and Affect:
Response to Literary Stories’ in Poetics 22: 389-407.
Miall, D. S. and D. Kuiken. 1999. ‘What is Literariness? Three Components of
Literary Reading’ in Discourse Processes 28: 121-138.
Miall, D. S. 1995. ‘Anticipation and Feeling in Literary Response: A Neurological
Perspective’ in Poetics 23: 275-298.
Schank, R. 1979. ‘Interestingness: Controlling Inferences’ in Artificial Intelligence
12(3): 273-297.
Schank, R. 1982. Dynamic Memory. A Theory of Riminding in Computers and People.
Cambridge: Cambridge University Press.
Seifert, C. M., R. P. Abelson and G. McKoon. 1986. ‘The Role of Thematic

Knowledge Structures in Reminding’ in Galambos, J. A., R. P. Abelson and J. B.
Black (eds) Knowledge Structures. London: LEA: 185-210.
Semino, E. 1997. Language and World Creation in Poems and Other Texts. London
and New York: Longman.
Semino, E., and J. Culpeper (eds). 2002. Cognitive Stylistics: Language and
Cognition in Text Analyses. Amsterdam: Benjamins.
Short, M. H. 1973. ‘Some Thoughts on Foregrounding and Interpretation’ in
Language and Style 6(2): 97-108.
Tan, E. S. 1994. ‘Story Processing as an Emotion Episode’ in van Oostendorp, H., and
R. A. Zwaan (eds) Naturalistic Text Comprehension. Norwood, New Jersey:
Ablex Publishing Corporation: 165-185.
Tsur, R. 2002. ‘Some Cognitive Foundations of “Cultural Programs”’ in Poetics
Today 23(1): 63- 89.
Tsur, R. 1992. Toward a Theory of Cognitive Poetics. Amsterdam: Elsevier Science
Publisher.
Turner, M. 1994. ‘Design for a Theory of Meaning’ in Overton, Willis F., and David
S. Palermo (eds) The Nature and Ontogenesis of Meaning. Hillsdale, New York:
Lawrence Erlbaum Associates: 91-107.
van Peer, W. 2002. ‘Where Do Literary Themes Come from?’ in Louwerse, M., and
W. van Peer (eds) Thematics. Interdisciplinary Studies. Amsterdam/Philadelphia:
John Benjamins Publishing Company: 253-263.
__. 1992. ‘Literary Theory and Reader Response’ in Nardocchio, E. F. (ed.) Reader
Response to Literature. The Empirical Dimension. Berlin/New York: de Gruyter:
137-152.
__. 1986. Stylistics and Psychology. Investigation of Foregrounding. London, Sydney,
Wolfeboro, New Hampshire: Croom Helm.
Verdonk, P. 2002. Stylistics. Oxford: Oxford University Press.
Zwaan, R. A., G. A. Radvansky and S. N. Whitten 2002. ‘Situation Models and
Themes’ in Louwerse, M., and W. van Peer (eds) Thematics. Interdisciplinary
Studies Amsterdam/Philadelphia: John Benjamins Publishing Company: 35-53.
‘Why Get Upset Over a Few Cases of Rhinoceritis?’:
Possible Worlds in the Theatre of the Absurd
Katerina Vassilopoulou
University of Lancaster, England
Abstract
This present paper applies possible-worlds theory – as developed by literary theorists–
to an analysis of absurdist drama, a genre that has, to date, been unexplored in these
terms. I argue that this framework can prove very useful in the approach to absurdity.
I discuss some selected extracts from Pinter’s Old Times, Ionesco’s The Bald Prima
Donna, Jacques or Obedience and Rhinoceros, and Beckett’s Waiting for Godot. The
analysis is based on Ryan’s (1991) typology of accessibility relations, as well as on
her catalogue of types of alternative possible worlds that can be included within a
fictional universe (1985). A discussion of the plays in terms of the first typology
shows that some partial impossibilities can often be captured by accessibility relations
other than logical compatibility, which is typically associated with absurdist drama. I
further examine whether it is the relaxation of these relations alone that is responsible
for the created oddity. Additionally, in discussing the conflicts within the fictional
universe I argue that a further factor for the creation of absurdity lies in the fact that
the mismatches fail to move the plot forward, contrary to what happens in other
genres.
Keywords: Theatre of the Absurd; possible-worlds theory; partially impossible
worlds; accessibility relations; alternative worlds; authentication.
1. Introduction
The aim of this paper is to examine whether possible-worlds theory
can be relevant to the study of absurdist drama. In other words, I seek
to extend the applicability of possible-worlds models to this genre, in
order to examine whether these models can contribute to a thorough
interpretation of absurdity. First of all, I check whether there are any
other accessibility relations, apart from that of logical compatibility
(Ryan, 1991: 32) that are relaxed in absurdist plays as well as the
reasons for the creation of the subsequent oddity. Moreover, I examine
the role that the inter-world clashes play in the development of the
plot of these plays. My study is a comparative one and seeks to show
whether there are any similarities and differences in the three
playwrights’ preferences for the ways of building up absurdity as far
as the projection of possible worlds is concerned. This discussion is
part of a larger study, the text corpus of which consists of twenty
156 Katerina Vassilopoulou
plays, nine from Ionesco, nine from Pinter and two from Beckett, all
belonging to the playwrights’ early periods.1 In this paper, I will focus
on selected extracts from Ionesco’s The Bald Prima Donna, Jacques
or Obedience and Rhinoceros, Pinter’s Old Times and Beckett’s
Waiting for Godot. Since the study is based on the written text of the
plays, from this point onwards I will be referring to the readers of the
plays rather than the theatre audiences. Before turning to the analysis
of specific examples, the possible-worlds framework as applied in
logic and in fictionality will be briefly described.
2. Possible-worlds theory: from logic to fictionality

The notion of possible worlds can be traced back to the 17th century
and Leibniz, who expressed the belief that our actual world was
chosen as the best among an infinity of possible worlds that exist as
thoughts in God’s mind (1969: 333-4 and throughout). This notion has
been broadly exploited in the field of philosophical logic in order to
deal with some important logical issues to which the one-world model
could not provide solutions. Taking the ‘actual world’ as the only
frame of reference creates problems, for example, in the attribution of
truth-values to propositions of the type (1) ‘The Eiffel Tower is not in
Paris’ or (2) ‘The Eiffel Tower is in Paris and the Eiffel Tower is not
in Paris’, which should then be described as false. Therefore, logicians
adopted a frame of reference where, apart from the ‘actual world’,
there is also an infinite number of possible worlds surrounding it that
are defined as abstract and complete sets of states of affairs.
Within this system the classification of a proposition as either true or
false is extended, as the system also includes the modal operators of
necessity and possibility. Both operators lie beyond the limits of the
actual world. Possible truth applies to propositions that are true in at
least one possible world, while possible falsity to propositions that are
false in at least one possible world. On the other hand, necessary truth
applies to propositions that are true in all possible worlds, whereas
necessary falsity to propositions that are false in all possible worlds
(Semino, 1997: 59). Seen in this light, proposition (1) is possibly
false, because it is false in our actual world, although it may be true in
an alternative world, while proposition (2) is necessarily false, since it
contains a logical contradiction and thus cannot be true in any
logically possible world.
‘Why Get Upset Over a Few Cases of Rhinoceritis?’ 157
Any attempt to interpret fictional worlds within the framework of

traditional logical semantics has led to their treatment either as false or
as neither-true-nor-false, since they were situated outside the ‘actual
world’ of the readers. A sentence of the type ‘Emma Bovary
committed suicide’, even though it accords with Flaubert’s book,
would have to be interpreted either as false, because it assumes the
existence of a fictional character, i.e. a non-existent individual, or as
neither-true-nor-false, because it refers to an imaginary entity with no
referent in the actual world. Therefore, since the late 1970s literary
theorists (Eco, 1979; Doležel, 1988, 1989; Pavel, 1986) adapted and
further extended the notion of possible worlds and have developed a
semantics of fictionality based on the idea that the semantic domain
projected by the literary text is an alternative possible world (APW)
that acts as actual the moment we are immersed in a fiction. Through
this act of ‘recentering’ (Ryan, 1991), which is an essential part of
fiction-making, the actual world of the readers becomes only one of
the many alternative possible worlds that revolve around the world
that the narrator presents as actual. In this sense, the proposition
expressed by the sentence ‘Emma Bovary committed suicide’ is true
in relation to the world of Flaubert’s novel, whereas a proposition
expressed by the sentence ‘Emma Bovary did not commit suicide’
would be considered false, because it does not accord with the
fictional possible world.
Within the limits of logic, the term ‘possible’ describes those sets of
states of affairs that do not break the logical laws of non-contradiction
(given a proposition x, it is not possible that both x and not-x are true
in a given world) and of the excluded middle (given two contradictory
propositions, x and not-x, only one must apply in a given world, while
the ‘middle’ option where neither x nor not-x is true is ruled out). In
crossing over from logic to the field of literary studies, possible-
worlds theory has undergone a drastic change so as to deal precisely
with impossibility in fiction. Fictional worlds can thus be perceived as
possible even when they are ‘inconsistent’, namely when they violate
the laws of non-contradiction and of the excluded middle, whereas in
logic such worlds would be considered impossible. It is this
broadening of the theory that establishes its applicability to absurdist
drama. However, by and large, absurdist drama has not been exploited
as a source of data within possible-worlds approaches to the study of
fiction. Even though the particular genre has been referred to as an
example of a textual universe contradicting the actual world of the

readers (Ryan, 1991) and therefore the need for a closer look from this
perspective has been recognized, there has been no extended
discussion of absurdist drama in such terms. Post-modernist literature,
on the other hand, which also often contains logical incompatibilities,
seems to have monopolized the theorists’ interest. It is characteristic,
for example, that in his typology of impossible fictions, Ashline
(1995) draws examples primarily from postmodernist literature. It is
the aim of the present paper to compensate for this gap in research.
3. Ryan’s typology of accessibility relations and the notion of

authentication
Ryan (1991) suggests a typology of fictional worlds that are projected
by texts that belong to different genres. She forms her typology with
the aim to complement the deficits of previous approaches2 and to
provide a theory of fictional genres, since an interpretation of fiction
within the limits of logically possible worlds cannot cover the wide
range of fictional worlds because it excludes worlds that contain
logical impossibilities. In her view, there is no such thing as an
impossible fictional world and a world’s actuality, possibility or
impossibility is rather a matter of degree. In order to avoid talking
about an ‘impossible possible world’ in fiction, a wider range of
accessibility relations is required. These accessibility relations exhibit
the various ways in which the textual actual world (TAW) can be
associated with the actual world (AW) of the readers:
In decreasing order of stringency, the relevant types of
accessibility relations from AW involved in the construction of
TAW include the following:
(A) Identity of properties (abbreviated A/properties): TAW
is accessible from AW if the objects common to TAW and AW
have the same properties.
(B) Identity of inventory (B/same inventory): TAW is
accessible from AW if TAW and AW are furnished by the same
objects.
(C) Compatibility of inventory (C/expanded inventory):
TAW is accessible from AW if TAW’s inventory includes all
the members of AW, as well as some native members.
(D) Chronological compatibility (D/chronology): TAW is
accessible from AW if it takes no temporal relocation for a
member of AW to contemplate the entire history of TAW. (This

condition means that TAW is not older than AW, i.e. that its
present is not posterior in absolute time to AW’s present. We
can contemplate facts of the past from the viewpoint of the
present, but since the future holds no facts, only projections, it
takes a relocation beyond the time of their occurrence to regard
as facts events located in the future.)
(E) Physical compatibility (E/natural laws): TAW is
accessible from AW if they share natural laws.
(F) Taxonomic compatibility (F/taxonomy): TAW is
accessible from AW if both worlds contain the same species,
and the species are characterized by the same properties. Within
F, it may be useful to distinguish a narrower version Fǯ
stipulating that TAW must contain not only the same inventory
of natural species, but also the same types of manufactured
objects as found in AW up to the present.
(G) Logical compatibility (G/logic): TAW is accessible from
AW if both worlds respect the principles of noncontradiction
and of excluded middle.
(H) Analytical compatibility (H/analytical): TAW is
accessible from AW if they share analytical truths, i.e. if objects
designated by the same words have the same essential
properties.
(I) Linguistic compatibility (I/linguistic): TAW is accessible
from AW if the language in which TAW is described can be
understood in AW. (Ryan, 1991: 32-3; author’s italics)
As Ryan suggests, the world of absurdist texts results from the
relaxation of logical compatibility, which leads to types of worlds that
are described not as wholly impossible but only as logically
impossible, because, for example, something has both happened and
not happened. These logical contradictions in the way the characters
describe their world make it impossible for the readers to have any
reliable access to that world. Consequently, authentication becomes
extremely problematic, because virtually nothing can be ascertained
about the TAW. The term ‘authentication’ (Doležel, 1989) is
associated with establishing the existence of entities and of
occurrences of events in fiction via a reliable voice. As Doležel points
out, ‘the construction of impossible worlds is part and parcel of a
more general anomaly of fiction making, the misuse of authentication’
(1998: 160). In relation to absurdist plays, each conflicting version is

fully authentic when taken on its own, but when these versions are
juxtaposed, they form logically impossible worlds that prevent the
authentication of fictional existence.
In Ryan’s view, the relaxation of accessibility relations that are on the
top (A/properties, B/same inventory, C/expanded inventory) or in the
middle (D/chronology, E/natural laws, F/taxonomy) of her list results
in worlds that do not depart a great deal from the actual world of the
readers. Lifting F/taxonomy, for example, results in science-fiction
worlds, whereas lifting both F/taxonomy and E/natural laws results in
the world of fairy tales. However, the discussion of some extracts
from Ionesco’s Jacques or Obedience and Rhinoceros will show that
absurdist drama can also be associated with the relaxation of these
accessibility relations. A further issue that requires exploration is to
which extent the relaxation of these relations is responsible for the
subsequent absurdity and whether it is equally exploited by the three
playwrights as a technique for the creation of an odd textual world.
4. The relaxation of G/logic in absurdist drama

The fictional worlds of absurdist plays are often inconsistent due to
the logical impossibilities that they contain. These plays depict a
seemingly realistic world inhabited by human beings; readers are thus
highly likely to expect that the characters will also share their own
assumptions about the logical laws pervading their world. These
expectations, however, can be disrupted when the law of non-
contradiction is violated and G/logic is lifted, leading to worlds that
contain contradictory states of affairs. Ionesco’s The Bald Prima
Donna (1948) constitutes an interesting example of such an
inconsistent world. Logical contradictions make their appearance quite
early in the text, when the protagonist couple, Mr. and Mrs. Smith
begin to talk about someone called Bobby Watson, who is first
discussed as dead. The discussion abounds in contradictory claims
about the exact time of Bobby Watson’s death or the time of his
funeral:
MR. SMITH: [still with his paper] Well, well, well! According
to this, Bobby Watson’s dead.
MRS. SMITH: Good Heavens! Poor fellow! When did it
happen?
MR. SMITH: What are you looking so surprised about? You

knew perfectly well he was dead. He died about two years ago.
You remember, we went to the funeral about eighteen months
ago, it must be.
MRS. SMITH: Of course, I remember perfectly. It came back to
me at once; but I fail to understand why you had to look so
surprised to see it in the paper.
MR. SMITH: It wasn’t in the paper! It must be three years ago
now since there was talk of his passing away. I was reminded of
it by an association of ideas. (Ionesco, 1958: 88; turns 14-8)
(my emphasis)
[…]
MR. SMITH: He made the best-looking corpse in Great Britain!
And he never looked his age. Poor old Bobby! He’d been dead
for four years and he was still warm. A living corpse if ever
there was one. And how cheerful he always was! (Ionesco,
1958: 89; turn 20) (my emphasis)
At first, Mr. Smith claims that Bobby Watson’s death is announced in
the newspaper and this causes his wife’s astonishment. Her reaction
suggests that she perceived her husband’s words the same way that
readers are highly likely to do, that is, as a reproduction of what he has
just read. Yet, all the subsequent discussion demolishes this
impression and creates absurdity. Mr. Smith then asserts that Watson
died two years ago and accuses his wife of pretence since she must
also remember the funeral that took place a year and a half ago. This
claim contradicts our schematic assumptions in the sense that a funeral
normally takes place a few days after one’s death. Quite unexpectedly,
Mrs. Smith admits that she remembers Watson’s death and funeral
and her previous expression of surprise remains unjustified.
Mr. Smith denies having read about Bobby Watson’s death in the
newspaper, as this would not be possible since he died three years ago,
while later he says that Bobby Watson had been dead for four years
and he still hadn’t been buried. The relaxation of G/logic leads to the
projection of a logically impossible world that Mrs. Smith also shares,
within which it is possible for a person to have died many different
times in the past, and results in the subsequent blocking of
authentication. Apart from the logical contradictions concerning
Watson’s death, Mr. Smith also disrupts once again our schematic
assumptions regarding funerals. The corpse’s description as ‘warm’

and ‘veritable living’ (turn 20) could be interpreted as hiding a hint
that Bobby Watson was still alive or preserving some biological
functions of the living at that time. This claim may sound arbitrary,
but it could also be seen as totally compatible with the world-view that
the characters appear to share.
Despite these logical contradictions, both characters agree that Bobby
Watson is dead until Mrs. Smith asks an odd question:
MRS. SMITH: And when are they thinking of getting married,
the two of them?
MR. SMITH: Next spring, at the latest. (Ionesco, 1958: 89;
turns 27-8)
Much to the readers’ surprise, the two characters refer to Bobby
Watson as if he were alive and about to get married, although both
have just claimed to have attended his funeral and have talked about
his widow. The wh-question ‘And when are they thinking of getting
married, the two of them?’ (turn 27) presupposes that the Watsons’
forthcoming wedding is taken for granted, although a little earlier the
death of one of the two had been asserted. Interestingly, Mr. Smith
replies to his wife’s question without any objection to the fact that her
claim comes in sheer contrast with what the two characters were so far
presenting as the truth. The fact that a person is discussed as being
both dead and alive constitutes a further logical impossibility that
contributes to the projection of a world where the distinction between
authenticated and non-authenticated states of affairs becomes
extremely difficult.
When the conversation revolves around the Watsons’ children, logical
impossibilities continue and further prevent readers from fully
constructing and exploring the fictional world in their minds:
MRS. SMITH: […] It’s sad for her to have been widowed so
young.
MR. SMITH: Lucky they didn’t have any children.
MRS. SMITH: Oh! That would have been too much! Children!
What on earth would she have done with them?
MR. SMITH: She’s still a young woman. She may quite well
marry again. Anyway, mourning suits her extremely well.
MRS. SMITH: But who will take care of the children? They’ve
a girl and a boy, you know. How do they call them? (Ionesco,
1958: 89-90; turns 31-5)
As it appears, the Smiths return to the scenario according to which
Bobby Watson is dead. Not only that, but also Mrs. Smith first agrees
with her husband’s statement that the Watsons are lucky not to have
any children but then asserts their existence, as the phrase ‘you know’
(turn 35) suggests. Again, the two contradictory versions are discussed
as equally true and further establish the logically impossible world-
view that the couple shares.
Through the application of possible-worlds theory to The Bald Prima
Donna it has been demonstrated that absurdity does not result only
from the logical contradictions that render the world partially
impossible but mainly from the characters’ reaction to these
impossibilities. The logical distortions that take place in the text,
which are a means of demonstrating the disintegration and emptiness
of language, appear to sound completely normal to the characters.
They thus constitute a further indication that their world is determined
by logic and conventions that are totally different from those of the
readers and serve Ionesco’s purpose to enact the futility of existence
and communication.
5. Other accessibility relations as sources of absurdity

5.1 The relaxation of F/taxonomy
Ionesco’s Jacques or Obedience (1950) constitutes a parody of
bourgeois society. A great part of the play’s plot revolves around the
selection of a proper bride for the protagonist, Jacques, after he has
succumbed to the family creed (‘I love potatoes in their jackets’, p.
128).3 Readers are then gradually faced with a textual world that
relaxes taxonomic compatibility, because it includes human beings
with different properties from those in the actual world of the readers.
When Roberta, the bride-to-be, first appears on stage, her face is
covered by a white veil and her body is also hidden because of the
bridal dress that she wears. Roberta’s face is revealed after a long
interaction in which the two families, the Jacques and the Roberts,
enumerate her virtues, which turn out to be totally non-human, like,
for example, the ‘green pimples on a beige skin’, the ‘red breasts on a
mauve ground’ or a ‘shoulder of lamb’ (p. 132, turn 129). The stage
directions inform readers that Roberta has two noses:
JACQUELINE: Come on, then, the face of the bride!
[ROBERT FATHER pulls aside the white veil that hides
ROBERTA’S face. She is all smiles and has two noses; a
murmur of admiration from all except JACQUES.]
JACQUELINE: Oh! Lovely!
ROBERT MOTHER: What do you think of her?
JACQUES FATHER: Ah, if I were twenty years younger!
JACQUES GRANDFATHER: And me … ah … er … and me!
(Ionesco, 1958: 134; turns 173-6)
One main reason for the creation of absurdity is the fact that the
relaxation of taxonomic compatibility takes place in a textual world
that otherwise looks entirely realistic, as the setting is reminiscent of a
bourgeois interior and the characters are connected with recognizable
family bonds. Additionally, and contrary to what readers are highly
likely to expect, in seeing Roberta’s appearance, the other characters
express their admiration for her. Desirability is associated with a
monstrous appearance, as Jacques Father and Jacques Grandfather
actually admit that if they were younger they would fall in love with
her. Their attitude is thus a reversal of that expected in the real world,
were one to face such a creature. In this sense, the two men’s
comments further build up the absurdity and potential funniness of the
scene.
Jacques’s silence may at first be regarded as a reaction to the overall
abnormality and thus fool readers to assume that he shares their
assumptions about what is considered normal regarding one’s
appearance. As it turns out, however, Jacques is not pleased with
Roberta because he wants a woman with at least three noses:
JACQUES: No! No! She hasn’t got enough! I must have
one with three noses; three noses, I say, at least! (Ionesco,
1958: 135; turn 195)
The relaxation of F/taxonomy is maintained, as Robert Father appears
prepared for this demand and presents a daughter with three noses,
although when he first introduced Roberta to Jacques he had claimed
that she is their only daughter (p. 133):
ROBERT FATHER: [jovially] Never mind, ladies and gents,

there’s no harm done. [He slaps JACQUES, who is still tense,
on the back.] We’d already foreseen this difficulty. We have a
second only child at your disposal. And she has her three noses
all complete. (Ionesco, 1958: 136; turn 201)
Of course, logical compatibility is also relaxed at this point, since the
Roberts have two daughters, but, according to them, each is their only
child. The two claims are contradictory and mutually refuted, since the
notions ‘only’ and ‘second’ are incompatible. Yet the fact that the
other characters accept this claim as rational does not allow Robert
Father’s verbal behaviour to be interpreted as a breaking of the maxim
of quality (cf. Grice’s Cooperative Principle).4 Rather, the notion of
‘only’ expands its actual meaning to serve the characters’ purposes
and to adjust to the reality of their world.
Even after Jacques is presented with the second potential wife, he
keeps complaining that she is not ugly enough and asks for one that is
much uglier: No, she won’t do. Not ugly enough! Why, she’s quite
passable. There are uglier ones. I want one much uglier (p. 137, turn
220). The seemingly realistic world of the play, then, turns out to be
totally absurd, not only because it is inhabited by human beings with
non-human characteristics, and thus relaxes taxonomic compatibility,
but also because the characters share the odd belief that the more
monstrous one looks the more beautiful they are considered.
5.2 Relaxing E/natural laws

Ionesco’s Rhinoceros (1958) is another play where the conventional,
bourgeois settings, as presented to the readers in the initial stage
directions and throughout, raise expectations about a realistic plot but
then readers are confronted with the absurd situation where all human
beings are gradually transformed into rhinoceroses except from the
protagonist, Berenger. The play is thus interesting to analyze from a
possible-worlds perspective as it violates physical compatibility.
For the purposes of this paper I will focus on some extracts that are
taken from the second half of the play. By the end of the first scene of
Act Two, many of the citizens have already transformed into
pachyderms but Berenger deals with the situation with surprising
calm. He intends to visit his friend Jean and mend their friendship,
which was unsettled after a quarrel they had in Act One. In the second
scene of Act Two Berenger witnesses the metamorphosis of Jean,

which takes place in front of his eyes. At first, Jean does not accept
that there is something wrong with him but then considers the changes
on him as normal and gets angry with Berenger for pointing them out
to him:
BERENGER: Your voice is hoarse, too.
JEAN: Hoarse?
BERENGER: A bit hoarse, yes. That’s why I didn’t recognize
it.
JEAN: Why should I be hoarse? My voice hasn’t changed; it’s
yours that’s changed! (Ionesco, 1960: 60; turns 1064-7)
[…]
BERENGER: It’s your skin…
JEAN: What’s my skin got to do with you? I don’t go on about
your skin, do I?
BERENGER: It’s just that … it seems to be changing colour all
the time. It’s going green. [He tries to take JEAN’S hand.] It’s
hardening as well. (Ionesco, 1960: 63; turns 1116-8)
Jean is outspokenly expressing his preference for his new skin over
that of a normal human being and even finds some ‘advantages’ to
support this preference. For Berenger, however, this change is
horrifying:
BERENGER: But whatever’s the matter with your skin?
JEAN: Can’t you leave my skin alone? I certainly wouldn’t
want to change it for yours.
BERENGER: It’s gone like leather.
JEAN: That makes it more solid. It’s weatherproof.
BERENGER: You’re getting greener and greener.
JEAN: You’ve got colour mania today. You’re seeing things,
you’ve been drinking again. (Ionesco, 1960: 64; turns 1144-9)
As it becomes apparent, the whole process of transforming into a
rhinoceros does not shake Jean’s complacency. In fact, he uses a
series of arguments to rationalize his choice and, as his words reveal,
he thinks that being a rhinoceros is a much more preferable situation
than being a human being. A few turns later Jean confesses his wish to
become one of the pachyderms and accuses Berenger of prejudice for
being against these transformations:
BERENGER: I’m amazed to hear you say that, Jean, really!

You must be out of your mind. You wouldn’t like to be a
rhinoceros yourself, now would you?
JEAN: Why not? I’m not a victim of prejudice like you.
(Ionesco, 1960: 68; turns 1220-1)
[…]
JEAN: Keep your ears open. I said what’s wrong with being a
rhinoceros? I’m all for change. (Ionesco, 1960: 68; turn 1225)
Once again, it is not the relaxation of physical compatibility itself that
creates absurdity but the way the characters deal with the situation of
human beings turning into rhinoceroses. Jean has not only accepted
this change but is actually looking forward to it, while Berenger, who
has so far been calm and indifferent, is shocked for the first time,
primarily because he now witnesses the transformation and it is his
friend who has chosen to become a pachyderm.5 His words as he
narrates the transformation to his colleague Dudard in act three reveal
that he is more surprised by his friend’s choice to transform than by
the fact itself:
BERENGER: […] he was such a warm-hearted person, always
so human! Who’d have thought it of him! We’d known each
other for…for donkey’s years. He was the last person I’d have
expected to change like that. I felt more sure of him than of
myself! And then to do that to me! (Ionesco, 1960: 74; turn
1292)
In Dudard’s case, the transformation takes place on a moral plane.
Although he admits that he cannot find a satisfactory explanation for
this phenomenon (p. 74), he begins to rationalize the choice of
becoming a rhinoceros by mentioning some privileges that these
pachyderms have and that human beings lack, much like Jean had
earlier done:
DUDARD: Perhaps he felt an urge for some fresh air, the
country, the wide-open spaces…perhaps he felt a need to relax.
I’m not saying that’s any excuse…
BERENGER: I understand what you mean, at least I’m trying
to. But you know –if someone accused me of being a bad sport,
or hopelessly middle class, or completely out of touch with life,
I’d still want to stay as I am.
DUDARD: We’ll all stay as we are, don’t worry. So why get

upset over a few cases of rhinoceritis. Perhaps it’s just another
disease.
BERENGER: Exactly! And I’m frightened of catching it.
(Ionesco, 1960: 75; turns 1301-4)
Referring to both Dudard and Jean, Hoy (1964: 253) suggests that this
is the most insidious kind of rationalization, as it serves to cover
humanity’s retreat into animality. Berenger’s attitude suggests that he
does not perceive the situation in his town as absurd. In fact, he even
considers the possibility of him catching the disease too, which is a
further source of anxiety for him. Dudard accuses Berenger of not
having any sense of humour (p. 78) and suggests keeping an open
mind when judging those who have decided to turn into rhinoceroses
(p. 83). The moral transformation of Dudard has already begun. His
way of facing the people’s transformation into rhinoceroses as normal,
as a simple decision to change their skin is indicative of his own
gradual infection from what they describe as a spreading disease:
BERENGER: And you consider all this natural?
DUDARD: What could be more natural than a rhinoceros?
BERENGER: Yes, but for a man to turn into a rhinoceros is
abnormal beyond question
DUDARD: Well, of course, that’s a matter of opinion…
(Ionesco, 1960: 84; turns 1420-23)
To Berenger’s prediction that Dudard is going to follow the
rhinoceroses soon, the latter pleads objectivity and a tendency to
always look at the positive side of things. After Daisy, the woman
with whom Berenger is in love, appears on stage, Dudard continues to
support those who choose to become pachyderms. When he sees them
streaming out and crowding the streets he too runs out and joins them.
At the end of the play Berenger is the only citizen left to insist that it
is normal to be human and abnormal to be a rhinoceros. Yet he never
rejects the whole situation as something impossible to happen. He
therefore shares the odd belief that all characters in the play hold,
namely that such transformations can in fact happen. Much like in
Jacques, it is not the relaxation of an accessibility relation as such, in
this case of physical compatibility, that is responsible for the created
absurdity but primarily the fact that it takes place in a seemingly
realistic world and the way it is dealt with by the characters that
results in the subsequent absurdity.
6. The internal structure of fictional worlds

Most fictional worlds can be described as universes, namely systems
of worlds, where one world functions as actual and is surrounded by a
variety of possible worlds that function as non-actualized alternatives
of this actual world (Pavel, 1986: 64; Ryan, 1985: 719, 1991: 109).
These alternative possible worlds (APWs) correspond to the private
worlds of the characters. The commonest types of private worlds are
Knowledge Worlds (K-Worlds), Obligation Worlds (O-Worlds), Wish
Worlds (W-Worlds) and Fantasy Worlds (F-Worlds), represented
respectively by the characters’ beliefs, obligations, wishes and
fantasies, dreams or hallucinations. A perfect correspondence between
the actual world and the private worlds of the characters creates a
situation of equilibrium in the narrative universe. In order for a plot to
begin, a situation of conflict must be created within the narrative
universe. The nature of the conflict may vary, but in any case it
contributes to the creation of a ‘successful’ plot that guarantees the
tellability of the narrative universe. With regard to absurdist plays,
whose main characteristic is the lack of plot, there arises the question
whether these conflicts either between the characters’ private worlds
and the TAW or between different private worlds do in fact lead to the
undertaking of action on the part of the characters and whether this
action or inaction is associated with the absurdist nature of the TAW.
7. The role of internal conflicts in the creation of absurdity

7.1 Internal conflicts between the actual domain and the
characters’ private worlds
Beckett’s Waiting for Godot (1952) describes a perpetual act of
waiting. Beckett dramatizes this waiting during which nothing
happens by repeating the same pattern of actions twice. The two acts
of the play have a repetitive structure, as they describe two different
days, during which, however, similar activities and events take place.
Vladimir and Estragon are waiting for Godot, a man whom they do
not know but to whom they have imbued the attributes of a hero or a
Christ-Saviour. Throughout the play, the characters repeat the
following exchange:
ESTRAGON: […] [He turns to VLADIMIR.] Let’s go.

VLADIMIR: We can’t.
ESTRAGON: Why not?
VLADIMIR: We’re waiting for Godot.
ESTRAGON: [Despairingly.] Ah! [Pause.] […] (Beckett, 1986:
15; turns 91-5)
As it appears, the two tramps have to fulfil an Obligation World,
namely to remain in the same place and wait for Godot who will come
and save them from their purposeless and meaningless life. This
world, however, comes in conflict with their desire to leave, which
reflects their joint Wish World. Each time the two tramps reach a dead
end out of desperation, Estragon suggests to Vladimir that they leave,
but the thought of Godot prevents them from departing. These shared
private worlds, however, do not accord with the actual world of the
play, in which Godot never arrives nor is he likely to do so at some
point after the end of the play. At the end of both acts, Godot’s
emissary, a little boy, enters the stage to inform the tramps that his
master will not come that evening but he will definitely make it the
following day:
VLADIMIR: […] [Pause.] Speak.
BOY: [In a rush.] Mr. Godot told me to tell you he won’t come
this evening but surely tomorrow.
[Silence.]
VLADIMIR: Is that all?
BOY: Yes, sir.
[Silence.] (Beckett, 1986: 49; turns 772-5)
[…]
VLADIMIR: You have a message from Mr. Godot.
BOY: Yes, sir.
VLADIMIR: He won’t come this evening.
BOY: No, sir.
VLADIMIR: But he’ll come tomorrow.
BOY: Yes, sir.
VLADIMIR: Without fail.
BOY: Yes, sir.
[Silence.] (Beckett, 1986: 85; turns 1655-62)
Both times the tramps’ Wish World is frustrated, but the second time
Vladimir’s disappointment becomes more apparent due to his
anticipation of the Boy’s lines. Although, quite surprisingly, the Boy

fails to remember Vladimir, he recognizes the Boy as soon as the
latter enters the stage, and can therefore predict the reason for his visit.
This further builds up the repetitive character of the play and suggests
that the tramps’ private World is recurrently frustrated. After the
second interaction with the Boy, Vladimir admits to himself that they
are waiting in vain. This realization, nevertheless, cannot endure for
long, but is rather another glimmer of the truth that they are refusing
to face, namely that their life is sterile, purposeless and thus absurd.
The play closes with another rejection on Vladimir’s part of
Estragon’s suggestion to leave, using again as an excuse that they are
waiting for Godot. As we come to realize, nothing changes throughout
the play nor do we expect anything to change in the future. The
tramps’ Wish World will remain in conflict with the actual domain
even after the end of the play but it seems beyond the characters’
power to do anything but wait. In other words, the conflict within the
fictional universe fails to make the characters undertake an action that
will move the plot forward. This odd status of the expectation for
Godot to come, the Wish World that is never realized, accounts to a
great extent for the created absurdity.
7.2 Internal conflicts between the private worlds of different

characters
A different kind of heterogeneity is the one that results from a conflict
between the private worlds that different characters possess. I have
chosen to discuss Pinter’s Old Times (1970), because the whole of this
play revolves around the different versions of the same events that two
characters, Deeley and Anna, offer in order to gain power over Kate,
who is Deeley’s wife and Anna’s old friend and perhaps lover. This
play is very interesting in possible-worlds terms in the sense that the
boundaries between the real and the unreal seem to have collapsed
completely and as a consequence the contradictory versions of the past
are all accepted as true.
All the contradictory claims and versions of events in the play should
be interpreted in the light of the words that Anna says in the middle of
Act One, in which the core of the play could be said to lay:
ANNA: […] There are some things one remembers even though
they may never have happened. There are things I remember
which may never have happened but as I recall them so they

take place. (Pinter, 1971: 31-2; turn 193)
Anna’s statement is absurd for two reasons. First, she associates
conflicting notions by arguing that she remembers things that have
never happened. Second, although for the readers memory is the result
of a certain experience, for Anna –and apparently for the other
characters too– it is the driving force for the creation of experiences,
so that everything that is remembered is considered true. This turn
reflects the Knowledge World that Anna possesses, which becomes an
actuality in the play so that the characters use their memories as a
weapon and reappropriate them in order to serve their present conflicts
in an attempt to gain sexual power over Kate (Hughes, 1974: 468). As
Batty appositely remarks, with Anna’s words ‘Pinter powerfully
demonstrates how the past is a terrain to be plundered and exploited
for confirmation of identity’ (2001: 64).
The extracts I have chosen to quote here show the contradictory
Knowledge Worlds that Deeley and Anna hold with regard to whom
Kate saw a film with about twenty years ago, when they were all
young and lived in London:
DEELEY: What happened to me was this. I popped into a
fleapit to see Odd Man Out. […] And there was only one other
person in the cinema, one other person in the whole of the
whole cinema, and there she is. And there she was, very dim,
very still, placed more or less I would say at the dead centre of
the auditorium. I was off centre and have remained so. […] So
it was Robert Newton who brought us together and it is only
Robert Newton who can tear us apart.
Pause
ANNA: F.J. McCormick was good too. (Pinter, 1971: 29-30;
turns 185-6)
[…]
ANNA: […] For example, I remember one Sunday she said to
me, looking up from the paper, come quick, quick, come with
me quickly, and we seized our handbags and went, on a bus, to
some totally obscure, some totally unfamiliar district and,
almost alone, saw a wonderful film called Odd Man Out.
Silence
DEELEY: Yes, I do quite a bit of travelling in my job. (Pinter,

1971: 38; turns 230-1)
As it becomes apparent, a seemingly innocent topic, namely a memory
from over twenty years ago, turns out to be a battleground for rivalry
between the two competitors. It is interesting that neither openly
questions the veracity of the other’s story, although they present their
own as the true one. It appears, however, that Anna’s story has a
stronger impact on Deeley than the other way round. In my view, this
is revealed by the fact that after Deeley’s turn there is a pause before
Anna speaks, whereas after Anna’s turn there follows a silence.
Tannen claims that a silence ‘represents climaxes of emotion in
interaction, the point at which the most damaging information has just
been introduced into the dialogue, directly or indirectly’ (1990: 263).
This interpretation accords with the sense of hidden conflict and the
struggle for domination over Kate between Deeley and Anna. This is
further supported by the way conversation continues in both cases.
After the pause, Anna takes the turn and, based on a rather
unimportant detail of the story regarding the protagonist of the film,
disagrees with Deeley and expresses her preference for another actor.
Thus, with this indifferent remark she diminishes Deeley’s assertion.
After the silence following Anna’s story, however, Deeley leaps to a
new topic about travelling around the world on business and his
reaction could be seen as an attempt to ‘polish’ his hurt self-image
(Homan, 1993: 169) after Anna’s brief reminiscence, which
annihilates his own earlier claim.6
In this paper I have shown that possible-worlds theory is a powerful
instrument in the study of absurdity. The discussion has been two-
fold, focusing both on the cases where accessibility relations are
relaxed as well as on those cases of conflicts within the narrative
universe. As far as the first issue is concerned, it has been shown that
apart from G/logic, absurdist plays also relax accessibility relations
that are prototypically associated with fictional worlds that do not
depart a great deal from the actual world. In a corpus of twenty plays,
the extracts discussed are only a small sample of those cases. As has
been claimed throughout, it is not the relaxation as such that is
responsible for the creation of absurdist effects but rather the
contribution of certain factors. First of all, the lifting of relations takes
place in seemingly realistic settings. This is where the main difference

between absurdist drama and fairy-tales lay. Although the world of
fairy-tales results primarily from the lifting of physical laws and
taxonomic compatibility, as it includes witches, talking animals or
magical transformations, readers are aware that this world is different
from their own and they can thus construct it in their mind without
considering it absurd. In the case of absurdist plays, on the other hand,
readers are faced with a world that is similar to theirs but all their
expectations for a realistic plot are then disrupted due to the
impossibilities that take place. This disruption is further reinforced by
the characters’ unexpected reaction to the impossibilities. Their
attitude in no way accords with the way one would react in the real
world as characters do not appear to hold the same assumptions with
the readers about the laws that govern their world. As a consequence,
any attempts on the readers’ part to construct a coherent text world are
frustrated.
The discussion about the conflicts within the fictional world has not
been exhaustive either. However, it has demonstrated that another
crucial factor for the creation of absurdity is the fact that these inter-
world conflicts do not lead to the undertaking of action that will move
the plot forward. Both plays discussed in these terms are characteristic
examples of static drama with no external action. Thus, the failure of
this mismatch to lead to action confirms one of the main
characteristics of absurdist drama, namely a lack of plot (cf. Esslin,
1978; 1980).
A comparison of the three playwrights has led to some interesting
conclusions regarding the application of possible-worlds theory. First
of all, Ionesco exploits the technique of creating absurdity through the
projection of possible worlds that deviate from the actual world much
more frequently than Pinter or Beckett. In fact, cases of absurdity
associated with the relaxation of accessibility relations other than
G/logic are found primarily in Ionesco’s plays. This can be associated
with the fact that Ionesco exploits the ‘language of images’ (Lane,
1994: 12), primarily in his early plays, in order to reveal the
strangeness of the world, since he considers theatre, as it was
developing in his time, an inadequate medium of expression. As far as
the analysis of Pinter and Beckett’s plays is concerned, possible-
worlds theory can prove particularly useful in relation to the
discussion of internal conflicts of the projected worlds. The

relationship between the various worlds in the textual universe never
changes and, consequently, these conflicts fail to move the plot
forward, which inevitably results in the creation of absurdity.
In conducting this analysis I neither wish to claim that these are the
only extracts that can be discussed in possible-worlds terms nor that
possible-worlds theory is the only appropriate tool for shedding light
on the absurdity created in these cases. Schema theory, for example,
can also be useful, and the discussion of the extracts has often
revolved around the disruption of the expectations that the text world
created for the readers. Besides, the logical contradictions included in
the plays can raise questions regarding the applicability of Grice’s
maxim of quality. Nor is it necessary to choose only one theoretical
tool, as a combinatory approach is highly likely to lead to a more
thorough interpretation of absurdity. The aim of the present paper has
been to compensate for the deficits in the work that has been done so
far in relation to the possible worlds projected by absurdist plays and
their role in the creation of oddity. At the same time, there have been
some interesting findings as far as the dramatic technique of Ionesco,
Pinter and Beckett is concerned. This is only a first step in the
application of possible-worlds theory to this genre but is indicative of
the important role that this theoretical tool can play in the stylistics of
absurdist drama.
Endnotes
1
The main criterion for the selection of these plays, which mainly justifies the
unequal number selected from each playwright, has been the presence of dialogue,
since I decided to focus on prototypical plays that consist of interactions between
characters. Moreover, the plays are all intended for stage performance. Sketches and
very short texts (less than ten pages long) are excluded.
2
As Semino (1997: 80-1) notices, the main deficit of typologies such as Doležel’s or
Maitre’s is that they lack an accurate account of the way in which the readers perceive
the distance between the fictional world and the actual world during text processing.
3
The triviality of the issue at hand comes in sheer contrast with the verbal violence
that Jacques suffers from his family as long as he refuses to submit.
4
Grice’s maxim of quality suggests the following: ‘Do not say what you believe to be
false. Do not say that for which you lack adequate evidence.’ (1975: 46)
5
When Berenger talks about transforming into a rhinoceros with Dudard in Act
Three, he refers to it as ‘a nervous disease’ (p. 76).
6
Some of the oddities I have discussed in previous sections could also potentially be
interpreted as conflicts between the private worlds of different characters, cf. the
absurd situation in Ionesco’s Rhinoceros, where Berenger is the only character that
holds the Belief World that transforming into a rhinoceros is something negative.
References
Ashline, W.L. 1995. ‘The problem of impossible fictions’ in Style 29(2): 215-34.
Batty, M. 2001. Harold Pinter. Tavistock: Northcote House.
Beckett, S. 1986. Samuel Beckett: The Complete Dramatic Works. London: Faber and
Faber.
Doležel, L. 1988. ‘Mimesis and possible worlds’ in Poetics Today 1(3): 7-25.
Doležel, L. 1989. ‘Possible worlds and literary fictions’ in Allén, S. (ed.) Possible
Worlds in Humanities, Arts and Sciences: Proceedings of Nobel Symposium 65.
New York: de Gruyter: 223-42.
Doležel, L. 1998. Heterocosmica: Fiction and Possible Worlds. Baltimore: The Johns
Hopkins University Press.
Eco, U. 1979. The Role of the Reader. London: Hutchinson.
Esslin, M. 1978. An Anatomy of Drama. London: Abacus. 2nd edition.
Esslin, M. 1980. The Theatre of the Absurd. London: Penguin. 3rd edition.
Grice, H.P. 1975. ‘Logic and Conversation’ in Cole, P., and J.L. Morgan (eds) Syntax
and Semantics, Volume 3: Speech Acts. New York: Academic Press: 41-58.
Homan, S. 1993. Pinter’s Odd Man Out: Staging and Filming Old Times. Lewisburg:
Bucknell University Press.
Hoy, C. 1964. The Hyacinth Room: An Investigation Into the Nature of Comedy,
Tragedy, and Tragicomedy. London: Chatto and Windus.
Hughes, A. 1974. ‘“They Can’t Take That Away From Me”: Myth and Memory in
Pinter’s Old Times’ in Modern Drama 17(4): 467-76.
Ionesco, E. 1958. Plays. Volume I: The Lesson, The Chairs, The Bald Prima Donna,
Jacques or Obedience. London: John Calder. (trans. D. Watson)
Ionesco, E. 1960. Plays, Volume IV: Rhinoceros, The Leader, The Future is in Eggs
or It Takes All Sorts to Make a World. London: John Calder. (trans. D. Watson)
Lane, N. 1994. Understanding Eugène Ionesco. Columbia: University of South
Carolina Press.
Leibniz, G.W. 1969. Philosophical Papers and Letters, 2nd edition, translated and
edited by L.E. Loemker. Dordrecht: D. Reidel Publishing Company.
Pavel, T.G. 1986. Fictional Worlds. Cambridge: Harvard University Press.
Pinter, H. 1971. Old Times. London: Methuen.
Ryan, M.L. 1985. ‘The modal structure of narrative universes’ in Poetics Today 6(4):
717-755.
Ryan, M.L. 1991. Possible Worlds, Artificial Intelligence and Narrative Theory.
Bloomington: Indiana University Press.
Semino, E. 1997. Language and world creation in poems and other plays. London:
Longman.
Tannen, D. 1990. ‘Language as conflict management in fiction and drama: Pinter’s
Betrayal and a short story, “Great Wits”’ in Grimshaw, A.D. (ed.) Conflict Talk:
Sociolinguistic investigations of arguments in conversations. Cambridge:
Cambridge University Press: 260-79.
PART III
CORPUS STYLISTICS
A Multivariate Study of Style Differences in Poetry
Vadim Andreev
Smolensk State University, Russia
Abstract
The paper deals with the comparative analysis of individual styles (idiostyles) in
verse. The material includes lyrics of four leading American poets of the 19th century
– W.C. Bryant, H.W. Longfellow, R.W. Emerson and A.E. Poe. Verse texts
simultaneously demonstrate language relations and specific versification tendencies
and thus have a much more complicated structure than prose (Gasparov, 1974). The
article focuses on linguistic peculiarities of idiostyles of the four authors. Each of the
texts has been analyzed with 43 parameters (phonetic, morphological, syntactic,
rhythmic, etc.)
The characteristics discriminating between the texts of the poets were established with
the help of multivariate discriminant analysis. These characteristics were found to be
highly effective: they discriminate over 95 % of the texts of the four authors. The
results obtained formed the basis for a comparative study of the individual styles of
the authors. The degree of similarity/difference between the idiostyles was measured.
Keywords: style; classification; multivariate discriminant analysis.
1. Introduction
Quantitative study of style (stylometry) attracts more and more
attention. Methods based on measuring the variability of text
elements, belonging to different language subsystems, are used in
research on genre distinctions, to establish authorship, for gender
categorization, and so on.
Most of the studies in the field of stylometry in English are devoted to
the analysis of prose. Verse texts have been attracting much less
178 Vadim Andreev
attention. Such lack of interest in this sphere seems unjust, because

verse presents a very interesting material for such studies due to its
specific features.
Being divided into equal parts (lines) which possess a certain kind of
connection between their elements, verse demonstrates not one (like
prose) but two ‘dimensions’: horizontal (interrelation of language
elements in one and the same line) and vertical (interrelation of
elements of different lines). The latter is not usually found in prose.
2. Data sources
The material includes lyrics from four leading American romance
poets of the 19th century – W.C. Bryant, H.W. Longfellow, R.W.
Emerson and A.E. Poe – 141 texts with a total amount of more than
4000 lines.
Lyrics were chosen because they express in the most vivid way the
individuality of an author’s style, the essential characteristics of his
poetry. In order to achieve a common basis for comparison of the four
poets we investigated only iambic lyrics, generally not exceeding 60
lines. It should be noted that this meter was used by American
romance poets in most of their lyrics.
Sonnets were not taken for analysis because they possess specific
structural organization.
Four classes of texts were formed:
Class 1: Bryant – 31 texts (1160 lines);
Class 2: Longfellow – 53 texts (1670 lines);
Class 3: Emerson – 34 poems (701 lines);
Class 4: Poe – 23 works (628 lines).
3. Method
One of the possible methods, which gives an opportunity to compare
different texts by simultaneously using a large number of
characteristics, is multivariate discriminant analysis.
Discriminant analysis has been used in literature studies for various
purposes. Stamatatos, Fakatakis and Kokkinakis (2001) use
discriminant analysis to solve the authorship detection problem on the
basis of what they call low-level measures (sentence length,
A Multivariate Study of Style Differences in Poetry 179
punctuation mark count, etc.), syntax-based measures (noun phrase

count, verb phrase count, etc.) and a set of style markers obtained by a
natural language processing tool (percentage of rare or foreign words,
a measure that indicates morphological ambiguity, etc.) Additionally,
they also use the frequencies of most frequent words. Baayen, Van
Halteren, and Tweedie (1996) apply discriminant analysis to
determine authorship attribution using syntax-based methods.
The study reported by Holmes and Forsyth (1995) uses this method to
determine which vocabulary richness measures better discriminate
between texts written by Alexander Hamilton and those by James
Madison. The study made by Martindale and Tuffin (1996) uses this
method to find differences between Homer’s Illiad and Odyssey.
Wayne Larsen, Alvin Rencher and Tim Layton (1980) used
discriminant analysis for classification of various parts of the Book of
Mormon, establishing and measuring differences that exist between
them.
In Karlgen and Cutting (1994) multivariate discriminant analysis was
used to establish the difference between texts of various genres on the
basis of such quantitative parameters as word-length, sentence length,
adverb counts, and so on. Similarly, in a paper by Minori Murata
(2000), genres were differentiated on the basis of the rate of
appearance (per sentence) of the 62 selected conjunctive words and
particle-phrases.
In research on verse texts, statistical methods are not used so widely
as in prose, but have proved to be very effective in Andreev (2006),
Baevskij (1993), Kelih, Antiü, Grzybek, and Stadlober (2004),
Stadlober and Djuzelic (2004).
Discriminant analysis is a procedure whose purpose is to find
characteristics that discriminate between naturally occurring (or a
priori formed) classes, and to classify into these classes separate
(unique) cases which are often doubtful and ‘borderline’ (Klecka,
1989).
More exactly, discriminant analysis can give answers to the following
questions:
x Is there any relevant difference between several (naturally
occurring) classes of objects?
180 Vadim Andreev
x What characteristics differentiate these classes?

x What is the discriminating force of each characteristic?
In our research, verse texts are objects for analysis and groups of texts
written by the poets are the classes, which we compare.
4. Characteristics
Each text was described according to 43 characteristics, reflecting its
phonetic, morphological, syntactic, rhythmic properties, and also its
rhyme and stanza peculiarities.
Rhythmic Characteristics
By rhythm we understand a concrete realization of metric scheme in a
verse. Meter is the ordered alteration of strong (predominantly
stressed) and weak (predominantly unstressed) syllabic positions,
abstracted from the accentual structure of a concrete verse text. The
strong position is called the ictus, the weak position is called the non-
ictus. For our study, the most relevant are the first, the second and the
last (final) strong positions in a line.
In the actual verse text the metrical scheme is sometimes violated:
unstressed syllables may occupy ictuses (omission of an ictus stress).
This serves as a basis for the following characteristics (Baevskij,
1993; Tarlinskaja, 1976).
Unstressed first strong position
O'er the fair woods the sun looks down
Upon the many-twinkling leaves (...)
(Longfellow)
If the red slayer thinks he slays
(Emerson)
Unstressed second strong position

Where darkly the green turf upheaves (...)
(Longfellow)
Whose heart-strings are a lute
(Poe)
Unstressed final strong position

And sinking silently (...)
(Longfellow)
The prosperous and beautiful
(Emerson)
Two more characteristics are based on whether there are syllables

(unstressed or stressed) preceding the first strong position.
Number of syllables preceding the first strong position
The new moon's modest bow grow bright (1 syllable)
(Bryant)
And the cloud that took the form (2 syllables)
(Poe)
Number of stressed syllables preceding the first strong position
Songs flush with purple bloom the rye (...) (1 syllable)
(Longfellow)
Space grants beyond his fated road (…) (1 syllable)
(Emerson)
The number of feet in a line and the number of measures (the length of
the line in feet) form two more rhythmic characteristics.
Number of feet in a line
Am I a king, that I should call my own (5 feet)
This splendid ebon throne? (3 feet)
Or by what reason, or what right divine, (5 feet)
Can I proclaim it mine? (3 feet)
(Longfellow)
Number of measures
They brought me rubies from the mine, (4 feet)
And held them to the sun; (3 feet)
I said, they are drops of frozen wine (4 feet)
From Eden's vats that run. (3 feet)
There are 2 measures used (3-feet and 4-feet iamb)
(Emerson)
In the first case, the average number of feet is counted, in the second –
only the number of measures. Two measures are present in this poem.
182 Vadim Andreev
Morphological Characteristics
Morphological characteristics are represented in terms of traditional
morphological classes (noun, verb, adjective, adverb and pronoun).
We counted how many times each of them occurs in the first and the
final strong positions.
Phonetic Characteristics
The number of syllables in the words, occurring in the first and the
final strong positions was established.
Number of syllables in the word in the first strong position
Becalmed upon the sea of Thought, (2 syllables)
Still unattained the land is sought (...) (3 syllables)
(Longfellow)
Number of syllables in the word in the final strong position

And trophies buried (...) (2 syllables)
(Emerson)
Syntactic Characteristics
Most of syntactic characteristics are based on the use of traditional
notions of the members of the sentence (subject, predicate, object,
adverbial modifier). We also take into account cases when the word in
this position does not belong to any member of the sentence (address,
parenthesis). As with the morphological level we count how many
times in a poem each of the members of the sentence is used in the
first and the final strong positions.
In addition to this, we take into account if there is inversion (complete
– with the inversion of subject-predicate structure, or partial – with the
inversion of the secondary parts of the sentence).
Complete inversion
Arrives the snow, and, driving o'er the fields (...)
(Emerson)
Partial inversion
Of Merlin wise I learned a song (...)
(Emerson)
And on the gravelled pathway

The light and shadow played (...)
(Longfellow)
Two more characteristics are based on the number of clauses in (a)
complex and (b) compound sentences.
There are also several characteristics which represent what can be
called poetical syntax. They are enjambements and syntactical pauses.
Enjambement takes place when a clause does not end at the end of the
line and continues on the other line:
For so I must interpret still
Thy sweet dominion o'er my will.
(Emerson)
Pause is a break in a line, caused by a subordinate clause or another
sentence:
The very tones in which we spake
Had something strange I could but mark (...)
(Longfellow)
One more parameter is the number of lines, which end in exclamation
marks or question marks.
Number of lines ending in exclamation marks, question marks
How shall the burial rite be read?
(Poe)
Characteristics Of Rhyme
Exact rhyme
The rhyme is exact when all the sounds, starting with the stressed
vowel are the same (task – ask).
Thy parent sun, who bade thee view
Pale skies, and chilling moisture sip,
Has bathed thee in his own bright hue,
And streaked with jet thy glowing lip.
(Bryant)
184 Vadim Andreev
Partial rhyme
Partial rhymes include cases when the stressed syllable is rhymed with
unstressed (eyes – Paradise) and the so-called ‘eye-rhymes’ as in
vague – Prague; wreath – breath.
In my young boyhood – should it thus be given,
'Twere folly still to hope for higher Heaven!
(Poe)
Masculine rhyme
Rhodora! If the sages ask thee why
This charm is wasted on the earth and sky
(Emerson)
Stanza Characteristics
The number of stanzas and the number of types of stanzas in a poem
are taken into account on this level. Types of stanza in our research
are defined depending on the number of lines in them. Thus, stanzas
with the same amount of lines are considered to be of the same type.
5. Results
At the first stage of our analysis we established which of these
characteristics (if any) possess discriminating force. 35 of the 43
parameters were found relevant for the discrimination of text classes.
In other words, about 80% of the characteristics were found to
differentiate the texts written by different authors. We shall say that
these 35 characteristics ‘formed the model’, which will be used in our
further analysis.
It is important to check how reliable the results are. For this purpose
we only used characteristics that possess discriminant power and
classified the texts automatically into four groups, ignoring their a
priori class indications. This gave us a possibility to compare the
automatic classification (predicted) with the a priori classification
(observed). The results of this text distribution are given in Table 1.
Table 1 The predicted classifications versus the observed

classifications of texts
Percentage Classes of texts predicted on the basis of
of the model
Classes correctly
of texts predicted Long- Emer-
Bryant Poe
classifica- fellow son
tion
Bryant 96,77 30 1 0 0
Long-
90,57 3 48 1 1
fellow
Emerson 97,06 0 1 33 0
Poe 100,00 0 0 0 23
Total 95,04 33 50 34 24
In this table rows are observed classifications and columns are pre-
dicted classifications. The percentage of the correctly predicted classi-
fication of texts into classes is the indicator of the reliability of the
model.
In the first line, we can see that approximately 96% of Bryant’s works
were attributed correctly. 30 poems of Bryant’s fall into his own class
and one poem was included into the class of Longfellow. In the case
of Longfellow, 90% of the poems were correctly attributed. Three
texts were placed into the class of Bryant, one poem that of Emerson
and one was included into the class of Poe. The predicted attribution
of Emerson’s texts is very similar to that of Bryant: 97% were
attributed correctly, only one work was included in a different class
(the class of Longfellow). For Poe, all works were correctly placed in
his class. The total percentage of correct attribution is 95%, which is a
very high rate of prediction and can be considered a good result. The
expected rate of correct random prediction for four classes is only
25%.
Using the characteristics of the model, we can establish the degree of
similarity (or difference) between the texts of the four classes – and,
consequently, between the individual styles of the authors. The
186 Vadim Andreev
difference is established with the help of a method, called Squared

Mahalanobis Distances.
The results are given in Table 2. It shows how far the centroids of the
classes are positioned (centroids are the ‘mean points’ that represent
the means for all characteristics in the multidimensional space in
which each poem (observation) was plotted). The bigger the value –
the farther are the centroids, and, consequently, the bigger is the
difference between classes of texts.
Table 2 The data about the distances between text groups
Classes of Longfel-
Bryant Emerson Poe
texts low
Bryant 0,00 9,11 23,39 29,06
Longfel-
9,11 0,00 16,15 19,61
low
Emerson 23,39 16,15 0,00 22,90
Poe 29,06 19,61 22,90 0,00
As we see, works by Bryant and Longfellow are the closest, texts

written by Bryant and Poe are the farthest from one another. The
styles of these two authors, thus, are the most different. The group of
texts by Longfellow occupies the central position, his style is
relatively similar to the style of all other poets, while the group of
texts by Poe is rather distant from the rest of the classes. His
individual style is the most different from the others. The strongest
opposition exists between the styles of Bryant and Poe.
The results obtained show that there are two opposed clusters. The
basic division of text groups into clusters is illustrated by Picture 1.
Emerson
Bryant Longfellow
Poe
Cluster 1 Cluster 2
Picture 1 Basic division of text groups into clusters
One of the clusters is formed by the texts by Bryant and Longfellow,

the other includes texts by Emerson and Poe. Classes of texts by
Bryant and Longfellow are closer to each other than the text classes of
Poe and Emerson.
It is interesting that Bryant and Emerson in their essays, lectures and
letters stood for achieving the literary independence of America. They
expressed the necessity to abandon topics and images common for
Europe and to write instead about objects specific to America.
Poe disagreed with the idea of such limitation of the choice of poetic
images. Of all the four poets Poe mentioned Bryant as the most
talented. It would be natural to expect that the class of texts by Poe
would be either opposed to the classes of both Bryant and Emerson or
would be closer to the text class of Bryant. Nevertheless, Poe’s style is
strongly opposed to the style of Bryant and the poems written by Poe
are united with the works of Emerson into one cluster.
Another unexpected result is the central position of Longfellow’s texts
in the system. As we see, the structure of relations between the styles
of the four poets is rather complex. The relationship of a poet’s vision
of poetry and his subconscious stylistic preferences are far from
straightforward.
188 Vadim Andreev
Another important aspect of a poet’s individual style is the degree of

its stability, the range of its variation. It is possible to establish the
degree of homogeneity of the text classes, for this purpose the mean
distance (x) and standard deviation (ı) of each text from the centroid
of its class. The obtained results are presented in Table 3.
Table 3 Mean distances and standard deviations of the texts to their
centroids
Classes of texts Standard deviation Mean values
Bryant 9,869860 24,14358
Longfellow 13,50279 30,36947
Emerson 14,62560 42,88500
Poe 16,24002 42,55988
Using the data, we calculated the variation coefficient (V), which is

the ratio of standard deviation to the mean value. The value of the
coefficient lies within the range of 0 to 1. The less the coefficient, the
more homogeneous the class is. The following results were obtained
for the four classes:
Bryant V = 0,41
Longfellow V = 0,44
Emerson V = 0,34
Poe V = 0,38
The data shows that the investigated characteristics of individual
styles of different authors vary by different degrees. The class of texts
by Emerson turned out to be the most homogeneous and the group of
texts by Longfellow has the least degree of homogeneity.
This result together with Longfellow’s central position in respect to
other poets can be interpreted as proof of the attention paid by
Longfellow to works of other American poets. It is possible that
willingly or unwillingly he was learning from their works and
introduced certain features of their style into his own lyrics. Of course,
this is only a hypothesis, which has to be tested.
6. Conclusion
Two major clusters opposed to each other were established among
American romance poets. One of them is formed from the texts by W.
Bryant and H. Longfellow, the other includes texts by R. Emerson and

E. Poe, the strongest opposition existing between the style of W.
Bryant and E. Poe. The individual style of H. Longfellow occupies the
central position and is relatively close to the style of the other poets.
This result shows that a poet’s opinion on poetry and the ‘proper’ way
for its development and the actual stylistic characteristics of his texts
are not directly interrelated.
Morphological and syntactic characteristics, as well as some
parameters traditionally used in analysis of poetry, enable us to
effectively discriminate between verse texts of different authors in
more than 95% of cases. 35 of 43 parameters (about 80%) were found
to be relevant for discrimination. The class of texts by Emerson is the
most homogeneous and the group of texts by Longfellow has the least
degree of homogeneity. Longfellow’s style is relatively close to the
styles of the other authors (5 of his texts were attributed on the basis
of the parameters of the model to other authors) and Poe’s style is the
most different from the rest.
References
Andreev, S. 2006. ‘A Diachronic Study of the Style of Longfellow’ in Journal of
Quantitative Linguistics 10: 2-12.
Baayen, R.H., H. Van Halteren and F. Tweedie. 1996. ‘Outside the cave of shadows:
Using syntactic annotation to enhance authorship attribution’ in Literary and
Linguistic Computing 11: 121-131.
Baevskij, V.S. 1993. Pasternak - Lirik. [Pasternak the lyric poet]. Smolensk: Trust-
Imakom.
Gasparov, M.L. 1974. Sovremennij russkij stikh. Metrika i ritmika. [Contemporary
Russian verse. Metrics and rhythmics.]. Moscow: Nauka.
Holmes, D.I., and D. Forsyth. 1995. ‘The Federalist Revisited: New Directions in
Authorship Attribution’ in Literary and Linguistic Computing vol. 10: 111.
Karlgen, J., and D. Cutting. 1994. ‘Recognizing text genres with simple metrics using
discriminant analysis’ in Proceedings of COLING 94, Kyoto: 1071-1075.
Kelih, E., G. Antiü, P. Grzybek and E. Stadlober. 2004. Classification of Author
and/or Genre? The Impact of Word Length. Available at http://www-gewi.uni-
graz.at/quanta/publ/2004_ke_ga_pg-est_do_author-genre.pdf.
Klecka, W.R. 1989. Faktornyj, diskriminantnyj i klasternyj analiz. [Factor, discrimi-
nant and cluster analysis]. Moscow: Finansi i statistica.
Larsen, Wayne A., Alvin C. Rancher and Tim Layton. 1980. ‘Who Wrote the Book of
Mormon? An Analysis of Wordprints’ in BYU Studies 20: 225-251
Martindale, C., and P. Tuffin. 1996. ‘If Homer is the Poet of the Iliad, then he may not
be the Poet of the Odyssey’ in Literary and Linguistic Computing 11(3): 109-120.
190 Vadim Andreev
Minori, Murata. 2000. ‘Identify a Text's Genre by Multivariate Analysis – Using

Selected Conjunctive Words and Particle-phrases’ in Proceedings of the Institute
of Statistical Mathematics 48(2): 311-326.
Stadlober, E., and M. Djuzelic. 2004. Multivariate Statistical Methods in Quantitative
Text Analyses. Available at: http://www.stat.tugraz.at/stadl/papers/stdj05.pdf.
Stamatatos, E., N. Fakatakis and G. Kokkinakis. 2001. ‘Computer-Based Authorship
Attribution Without Lexical Measures’ in Computers and the Humanities 35: 193-
214.
e-Lears: a Corpus Approach to Shakespeare and
Tate
Maria Cristina Consiglio

University of Bari, Italy
Abstract
This article is an attempt to apply Corpus Linguistics tools and methodology to a
comparative analysis of William Shakespeare’s King Lear (1605) and its neoclassical
adaptation by Nahum Tate (1681). A ‘traditional’ analysis has highlighted in Tate the
presence of three major shifts from Shakespeare – in genre, in characterization and
plot, in language – which are the starting points of the present study aiming to find out
whether and to what extent the results obtained ‘manually’ are confirmed by Corpus
Linguistics. This article presents some results obtained by processing the two texts
with the software WordSmith Tools. In particular, it discusses the quantitative data
obtained and a qualitative analysis of three thematic words in Shakespeare – ‘fool’,
‘nature’, ‘nothing’ – and one thematic word in Tate – ‘love’.
Keywords: Corpus Linguistics; Shakespeare; Tate; King Lear; language shifts;
semantic prosody.
1. Corpus linguistics and literary studies

The importance of the role of Corpus Linguistics in Translation
Studies is ackowledged; the possibility of processing large quantities
of data in a very short time has opened new frontiers in linguistics,
particularly in lexicography. A lot of scholars are focusing their
attention on corpora for translation purposes: Sinclair, Baker, Laviosa,
Tognini-Bonelli, Barlow, Scott, to quote only a few. What is still
under consideration is the possibility to apply corpus linguistics
methodology and tools to literary studies. As Martin Wynne has
suggested, ‘there are interesting similarities in the approaches of
stylistics and corpus liguistics’ (2005: 1) and there are several possible
approaches to literary texts, which allow both intralingual and
interlingual analyses.
One of the first scholars to realize the importance of corpora in literary
studies was Bill Louw who, starting from a personal intuition, studied
what he has called semantic prosody in Philip Larkin’s poems (Louw,
1993). Other uses of corpora involve the study of linguistic creativity
(Carter, 2004; Stubbs, 2005); the classification of speech presentation
192 Maria Cristina Consiglio
in novels (Semino and Short, 2004); the investigation of a writer’s

style (Mahlberg, 2007); the identification of keywords in
Shakespeare’s plays (Scott, 2006); computer-assisted readings of
literary texts (Tribble, 2006; Brusasco, 2004).
With respect to interlingual studies, Jeremy Munday has analysed the
lexical shifts in an English translation of Gabriel Garcìa Màrquez’s
Diecisiete ingleses envenenados (Munday, 1998); Charlotte Bosseaux
has compared two French translations of Virginia Woolf’s The Waves
in order to check the translator’s (in)visibility (Bosseaux, 2001);
Federico Zanettin has used a bilingual parallel corpus to study stylistic
variations across translations of Salman Rushdie’s works (Zanettin,
2001).
A corpus approach allows scholars to make both quantitative and
qualitative analyses. Statistics and word lists are useful to carry out a
comparison between source text and target text, for example the
type/token ratio is a measure of lexical density, frequency lists are
indicators of lexical creativity, and scaling plots of the most frequent
words are a measure of linguistic characterization. Qualitative
analysis, instead, involves the investigation of semantic prosody,
translation choices and translation shifts.
This article is about Shakespeare’s King Lear (1605) and its
neoclassical adaptation by Nahum Tate (1681). The study is part of
wider research work on neoclassical adaptations of Shakespeare’s
plays and is an attempt to use corpus linguistics in order to investigate
the linguistic shifts which characterize Tate’s play. ‘Traditional’
research has pointed out various shifts in characterization and plot
which are strictly linked to language changes and transformations, for
the nature of drama itself as a genre. This study makes use of
WordSmith Tools in order to find out whether and to what extent the
results obtained manually are confirmed by Corpus Linguistics.
2. Starting points
In 1681 Nahum Tate adapted Shakespeare’s King Lear in order to
make it fit the Restoration stage. This means that the text was
refurbished in compliance with the new aesthetics of the time,
strongly infuenced by French classicism. The adeherence to the new
model led to a manipulation of the source consisting in ‘substantial
cuts of scenes, speeches, and speech assignments; much alteration of
e-Lears 193
language; and at least one and usually several important (or scene-
length) additions’ (Spencer, 1965: 5)
In fact, a comparative analysis of the two texts has highlighted some
important shifts: a) in genre, b) in characterization and plot, and c) in
language. With respect to genre, Shakespeare’s play has always been
considered a tragedy – despite the presence of some elements typical
of chronicles – whereas Tate’s play presents a happy ending with a
regenerating marriage between Edgar and Cordelia after Lear’s
restoration to the throne. The moral order is restored, the good are
awarded and the bad are punished in accordance with the neoclassical
rule of poetic justice as it was formulated by Thomas Rymer.
In relation to characterization, all Restoration adaptations present a
standardization of characters, who tend to become types in compliance
with the rule of decorum which wants the characters to be static and to
behave and speak according to their social status. In Tate’s remake,
for example, Lear is the portrait of a feudal king, royal and majestic
even in his madness, Edgar becomes a romantic hero and Cordelia a
sentimental heroine, while Edmund is the villain par excellance (even
Goneril and Regan become more wicked). This clear-cut
characterization which prevents the characters from changing and
‘growing’ through the play has its effects on the plot. The story
develops along one line only, focusing attention on the love affairs –
the relationship between Edgar and Cordelia and by contrast the
relationship between Edmund and the two wicked sisters. The shift in
genre determines a shift in emphasis, Edgar and Cordelia become the
protagonists and love becomes the dominating theme. On the level of
language, Tate used a more refined language – conversation was witty
and sophisticated at his time – consequently, he eliminated all that he
could not understand in Shakespeare making his language plain,
unambiguous, with the consequence that the metaphorical quality of
Shakespeare’s language was lost in his ‘translation’. (Consiglio, 2006)
3. Methodology
In a corpus based study the first problem to solve is the creation of
corpora. Texts belonging to past ages are available on the Web and
they are usually not subject to strict copyright norms; it is possible to
search for them with a commmercial search engine (like google) and
download them in the format needed in order to be processed by the
software. I have eliminated all stage directions since I believe that

they could have altered the results (first of all it is not possible to have
concordance lines with stage directions, secondly the insertion of
stage directions in frequency lists would have given false results, for
instance the word ‘fool’ appears a lot of times as a stage direction in
Shakespeare and is completely absent as such in Tate, and words like
‘enter’ or ‘exeunt’ would have been in the first positions of the list).
After creating the corpora, the first step was a quantitative and
statistical analysis of the two texts, in particular the present study
analyses the frequency lists of the two plays in order to find out
whether there is a shift in theme. The second step was a qualitative
analysis consisting of a study of concordance lines in order to identify
the semantic prosody – defined by Bill Louw (1993: 157) as ‘a
consistent aura of meaning with which a form is imbued by its
collocates’ – of some thematic words. In particular, this article focuses
on four words: ‘love’ – central to Tate’s play where a love story
between Edgar and Cordelia is the expedient the author chose in order
to ‘rectifie’ what he perceived as ‘the irregularity and improbability of
the tale’ (Tate, 1965: 201) – and ‘nature’, ‘fool’ and ‘nothing’ –
central to Shakespeare’s play, as has been highlighted by Northrop
Frye in his Nine Lessons (Frye, 1986: 113).
4. Quantitative and statistical analysis

The first step of this phase was a stastistical study of the composition
of the two corpora. The following table shows the results obtained:
Table 1: statistical data
Shakespeare Tate
Bytes 140.254 100.992
Tokens 25.580 18.220
Types 4.122 3.436
Type/token ratio 16,11 18,86
Even if the two corpora are rather small and unlike in size, the
different type/token ratio is worth noting; it represents the ratio
between the total number of running words of the corpus (tokens) and
the number of the different words employed in it (types). This ratio is
useful when studying lexical variety. As can be seen from the table,
Tate’s remake presents a higher percentage, this means that it contains
a lower number of repetitions; Shakespeare’s play is characterized by
e-Lears 195
the presence of some thematic words that echo throughout the tragedy
which Tate eliminated in his process of ‘polishing’ Shakespeare’s
language, as he himslef stated in the ‘Dedication’ to his King Lear.
The second step of this phase consisted of the creation of two
frequency lists in order to identify the most used words. After
excluding grammatical words (prepositions, conjunctions, articles,
pronouns and auxiliary verbs) and the appellatives ‘sir’, ‘lord’, and
‘king’ which are obviously much used throughout the play, I have
seen that the most used content word in Shakespeare is ‘know’ – with
the variants ‘known’, ‘know’st’, ‘knows’, ‘know’t’, ‘knew’ – for 158
occurrences, plus 4 occurrences for ‘knowledge’ and 6 for
‘understand’. This result is not surprising since one of the main
themes in King Lear is knowledge. This semantic preeminence is
further evident taking into account the occurrences of the words
referring to sight, a semantic field thematically linked to that of
knowledge; words like ‘eyes’, ‘sight’, ‘behold’, ‘looks’, and ‘see’
occur in the play 224 times. In order to test the centrality of the couple
‘know-see’, the occurrences of the words belonging to the antonym
semantic field, that referring to blindness, have been taken into
account. Words like ‘night’, ‘dark’, ‘blind’, and ‘darkness’ occur in
the text 116 times.
A similar analysis was made for Tate’s text, in order to investigate a
shift in theme. Words referring to the semantic field of knowledge –
‘know’, with its variants, and ‘understand’ (‘knowledge’ is absent in
Tate’s text) – occur in the corpus 64 times; those referring to sight –
‘eyes’, ‘sight’, and ‘see’ – occur 112 times; those referring to
blindness – ‘night’, ‘dark’, ‘blind’, and ‘darkness’ – show 58
occurrences. This study of the frequency list of Tate’s text shows that
the most frequent content word is once again the verb ‘know’, which
is not surprising since Tate’s remake follows its source text quite
faithfully up to the fourth act. The shift in genre with the introduction
of a romantic happy ending only arises in the fifth act.
Since romance is central in Tate’s play I have also considered the
frequency of the word ‘love’. In Tate, it occurs 53 times and in
Shakespeare, surprisingly, 109 times. The numbers seem not to
confirm the importance the love story between Edgar and Cordelia has
in the remake. Only a qualitative analysis of the concordances of the
word ‘love’ may help interpret the value of these numbers.
A further analysis was concerned with the frequency of the words

‘fool’, ‘nature’, and ‘nothing’. This choice was suggested by an essay
by Northrop Frye, in which the critic maintained that in order to
orientate himself in the complex structure of Shakespeare’s King Lear,
the reader should look for hints among those words the author repeats
in the text so insistently that he seems to influence the public by
means of suggestion. Among these words Frye indicated ‘fool’,
‘nature’, and ‘nothing’. From a quantitative point of view, the three
words are used much more in Shakespeare than in Tate, as the
following table clearly shows:
Table 2: occurrences of ‘fool’, ‘nature’ and ‘nothing’
Shakespeare Tate
Fool 54 5
Nature 40 25
Nothing 34 12
With respect to the word ‘fool’, there may be two main reasons for the
few occurrences of it in the remake. First, the elimination of the
character of the Fool, which makes all references to him unusable;
secondly, the shift in characterization of Lear, which determines the
elimination of several references to madness and folly.
In relation to the word ‘nature’, the reduction of its occurrences may
have been determined by the process of stylistic ‘polishing’ wanted by
Tate, in accordance to what he himself wrote in the preface, a process
which tended to eliminate all those repetitions he believed useless, as
well as by his wish to eliminate all references to a chaotic and
insinctive nature, which was contrary to the neoclassical view of an
ordered and hierarchical nature.
Finally, the drastic reduction of the occurrences of ‘nothing’, that in
Shakespeare echoes throughout the play communicating a pessimistic,
almost nihilistic, view of the world, may be a consequence of Tate’s
wish to give his public a more positive image of the world, in line
with the perception of the Restoration as a rebirth.
e-Lears 197
5. Qualitative analysis
The second step of this study was the qualitative analysis of the data
obtained through the reading of the frequency lists. First of all, I
considered the concordances of the word ‘love’ in both plays. The
table shows some of the results with reference to Shakespeare’s
tragedy:
Table 3: concordance of ‘love’ in Shakespeare
ave begot me, bred me, loved me: I Return thos
As much as child e'er loved, or father found;
hat shall Cordelia do? Love, and be silent. Of
In your dear highness' love. Then poor Cordeli
nk it. I know you do not love me; for your sisters
gest daughter does not love thee least; Nor are
ved, or father found; A love that makes breath p
Without our grace, our love, our benison. Com
ot been little: he always loved our sister most; an
d a dearer father in my love. Here is better tha
y heart into my mouth: I love your majesty Acco
r honour'd as my king, Loved as my father, as
your land: Our father's love is to the bastard E
arry like my sisters, To love my father all. But
l, Goneril, To the great love I bear you,-- Pray
thy master, whom thou lovest, Shall find thee f
ragon and his wrath. I loved her most, and thou
k the truth, Do you not love my sister? In hono
ur youngest daughter's love, Long in our court
The analysis of concordance lines allows one to verify the different

uses and meanings a given word assumes in a given text. As it can be
seen from the table, in Shakespeare the word ‘love’ is quite always
used with reference to filial or paternal love, sometimes with reference
to the love of the subjects for their king, only in I,i it is used with
reference to romantic love.
A similar analysis was made of Tate’s text. The table below shows the
most representative concordances:
Table 4: concordance of ‘love’ in Tate

And not a mutual Love? just Nature then
You both deserv'd my Love and both possest it
omen dar'd for vicious Love, And we'll be shin
by Denying: But if his Love be fixt, such Const
n the Race of Men, His Love was Int'rest, so ma
witness How much thy Love to Empire I prefer!
as due To my aspiring Love, for 'twas presumpt
nge As in my Gloster's Love, my Jealousie Ins
weigh'd the merit of my Love, Or is't the raving
promis'd Harvest of our Love. A word more
some expression of our Love. On, to the Sports
d live dependent on my Lover's Fortune. I cann
ord's reeking point; But Love detains me from D
en what Reception must Love's Language find F
isters have I sworn my Love, Each jealous of t
'offend Thee? Talk't of Love. Then I've offende
must have the Wings of Love; where's Albany.
In Tate’s play the word ‘love’ is used with reference to both the
romantic love between Edgar and Cordelia and the lustful relationship
between Edmund and Goneril and Regan. These results are not always
evident from the reading of the concordance lines, in many cases it
has been necessary to expand the span, which means to increase the
co-text in order to see other words that can clarify the meaning of the
words under study.
A similar study was made with the three words which, according to
Frye, are the key to Shakespeare’s play. The following table shows
some of the results concerning the word ‘fool’ in Shakespeare:
Table 5: concordance of ‘fool’ in Shakespeare 1
y boy, between a bitter fool and a sweet fool?
s; Whiles thou, a moral fool, sit'st still, and cries
u wouldst make a good fool. To take 't again p
t believe a fool. A bitter fool! Dost thou know th
ld night will turn us all to fools and madmen. Tak
ir, this your all-licensed fool, But other of your i
an thou art now; I am a fool, thou art nothing.
hat's a wise man and a fool. Alas, sir, are you
e-Lears 199
here. Dost thou call me fool, boy? All thy other

l? Go you, and call my fool hither. You, you, si
s services are due: My fool usurps my body. M
y! Now, by my life, Old fools are babes again; a
s neither wise man nor fool. Rumble thy bellyful
, fool? Not i' the stocks, fool. Deny to speak wit
speeches, as I were a fool? Goose, if I had y
atitude! If thou wert my fool, nuncle, I'ld have th
As it can be noticed, in Shakespeare’s play the word ‘fool’ has two
different meanings: one refers to the ‘motley fool’, the one who earns
his living from folly; the other refers metaphorically to those who do
not understand the surrounding world, and in this case it is directed to
more than one character in the play. The table shows how the term
assumes the former meaning when it is preceded by a possessive or by
another adjective specifying it; instead, when it is used in the plural
form, or in a general sense, or when it is preceded by indefinite article,
the word should be intended metaphorically. Besides, as Northrop
Frye has suggested, the word ‘fool’ is used with reference to any
character in the play endowed with human qualities, like Lear, Kent
and Albany. The concordance lines analysis seems to highlight
another meaning of the word ‘fool’ as a victim of misfortune; in this
sense all the characters in the play who find themselves on the wrong
side of the wheel of fortune are fools. The table shows the occurrences
of ‘fool’ when it is used with this latter meaning, as can be seen from
the analysis of the co-text:
Table 6: concordance of ‘fool’ in Shakespeare 2
e To this great stage of fools: this a good block;
I am even The natural fool of fortune. Use me
re villains by necessity; fools by heavenly comp
y bo-peep, And go the fools among. Prithee, n
y will not let me have all fool to myself; they'll b
ed that first finds it so. Fools had ne'er less wit
an fly: The knave turns fool that runs away; The
see, see! And my poor fool is hang'd! No, no,
id as in woman. O vain fool! Thou changed and
any kind o' thing than a fool: and yet I would not
In Tate’s play, as was pointed out by the quantitative analysis, the

word ‘fool’ is much less frequent; the only meaningful examples are
those also present in the source text, in two other cases the word refers
to Edgar disguised as poor Tom and, therefore, makes reference to a
staged madness:
Table 7: concordance of ‘fool’ in Tate
beauties Are due: my Fool usurps my Bed -
Tom's a cold; - I cannot fool it longer, And yet I
To this great Stage of Fools. - O here he is, l
r? I am even the natural Fool of Fortune: Use me
, to sooth us back To a Fool's Paradise of Hope,
The last example in the table is very interesting because it is not
present in the source text; in the co-text there are expressions like ‘the
wheel of fortune’, ‘misfortune’, and the word ‘hope’ repeated twice,
but in negative sentences, which suggests that Tate uses the word
‘fool’ in the meaning of victim of misfortune.
With respect to the word ‘nature’, Shakespeare’s characters use it with
different meanings. The Elizabethan Age was a period of transition
and crisis, which was reflected on the concept of nature. On the one
hand, nature was intended as a positive and benign force since it was
perceived as natural order; on the other hand, nature was seen as
negative when it referred to animal nature, the lowest level of the
chain of being. As Frye has suggested, the Elizabethans thought of
nature as a hierarchy, an ordered structure where good was at the top
and bad at the bottom of the ladder; for this reason the various
characters use the word nature with different intentions. Lear and the
characters associated with him perceive nature as order, for them love,
authority, compliance, and loyalty are natural because they are human.
Edmund, instead, makes his vow to the lowest level of nature, its
instinctive and predatory part. The following table shows some of the
concordance lines for the word ‘nature’ in Shakespeare:
Table 8: concordance of ‘nature’ in Shakespeare
er! O, sir, you are old. Nature in you stands on
e king falls from bias of nature; there's father ag
k conduct. Oppressed nature sleeps: This rest
. You cowardly rascal, nature disclaims in thee:
r your disposition: That nature, which contemns
e-Lears 201
a case. I will forget my nature. So kind a father

, wrench'd my frame of nature From the fix'd pl
the garb Quite from his nature: he cannot flatter,
ing could have subdued nature To such a lown
, and i' the heat. Thou, nature, art my goddess;
at breach in his abused nature! The untuned an
y be so, my lord. Hear, nature, hear; dear godd
o, Despite of mine own nature. Quickly send, B
brother noble, Whose nature is so far from do
m: Our foster-nurse of nature is repose, The w
r know'st The offices of nature, bond of childhoo
re not ourselves When nature, being oppress'd,
nuff and loathed part of nature should Burn itsel
tality. O ruin'd piece of nature! This great world
o, in the lusty stealth of nature, take More comp
s: though the wisdom of nature can reason it thu
As can be easily seen, the co-text helps interpret the word. The
presence of expressions like ‘lusty stealth’or ‘loathed part’ convey the
idea of nature as instinct, whereas expressions like ‘wisdom of nature’
or ‘offices of nature’ make reference to nature as a harmonious set of
rules. Moreover, the word ‘nature’ is also used with the meaning of
temperament, disposition, in both a positive and negative sense; in this
case the word is preceded by possessive or demonstrative adjectives.
Tate, as a Neoclassical writer, thought of nature as a linear structure, a
just nature ordered according to fixed rules, whose laws should also
regulate human behaviour. The following table shows some
concordances of ‘nature’ in Tate:
Table 9: concordance of ‘nature’ in Tate
lory of my Artifice, His Nature is so far from do
s discord, The bond of Nature crack't 'twixt Son
repair This Breach of Nature. We have empl
mpest. I will forget my Nature, what? so kind a
ing cou'd have subdu'd Nature To such a Lown
Kent see here Inverted Nature, Gloster's Shame
Repent, for know Our nature cannot brook A
not a mutual Love? just Nature then Had err'd:
know'st The Offices of Nature, bond of Child-h
art feels yet A Pang of Nature for their wretche

nuff and feebler Part of Nature shou'd Burn it s
urch-men Plot. Thou Nature art my Goddess,
as got it. The Pride of Nature Dies. Away, the
ho in the lusty stealth of Nature Take fiercer Qu
r. Mark ye that. Hear Nature! Dear Goddess
or; But least thy tender Nature shou'd relent At
Edmund, the villain, is the only character who uses the word ‘nature’
in reference to animal instinct and he defines himself as ‘the pride of
nature’, but he is doomed to die – as all the other wicked characters in
the play – and no repentence is possible for him, so that a hierarchical
vision of the natural world which, as the neoclassical writers believed,
was reflected in the social order, may be preserved.
Finally, the following table shows some concordances of ‘nothing’ in
Shakespeare:
Table 10: concordance of ‘nothing’ in Shakespeare
in their fury, and make nothing of; Strives in hi
ids me, though you say nothing. Mum, mum, He
n unto the worst Owes nothing to thy blasts. But
f all patience; I will say nothing. Who's there?
an you make no use of nothing, nuncle? Why,
ass? Couldst thou save nothing? Didst thou give
w; I am a fool, thou art nothing. Yes, forsooth,
tens to a score. This is nothing, fool. Then 'tis l
itfully offered. There is nothing done, if he retur
n your sisters? Speak. Nothing, my lord. Nothi
natural! Go to; say you nothing. There's a divisi
ing can be made out of nothing. Prithee, tell hi
Duchess of Burgundy. Nothing: I have sworn; I
rd. Nothing! Nothing. Nothing will come of not
nuncle? Why, no, boy; nothing can be made ou
eak. Nothing, my lord. Nothing! Nothing. Nothi
mund; it shall lose thee nothing; do it carefully.
ing, my lord. Nothing! Nothing. Nothing will co
Let's see: come, if it be nothing, I shall not need
is sword Away! I have nothing to do with thee.
gan with him: have you nothing said Upon his p
e-Lears 203
something yet: Edgar I nothing am. 'Tis strange

u're much deceived: in nothing am I changed B
ters, sir. Death, traitor! nothing could have subd
g. Nothing will come of nothing: speak again. U
I taste bread, thou art in nothing less Than I hav
it o' both sides, and left nothing i' the middle: her
As can be seen from the table, the word ‘nothing’ is often associated
with the act of speaking. The analysis of the co-text seems to reinforce
the negativity of the word and communicates a sense of void and
disintegration; as Frye has highlighted, the characters question their
identity but they find no answer since the social upheaval due to the
division of the kingdom – but above all to Lear’s pretension to
maintain his identity after becoming a ‘nothing’, in the Fool’s words
‘an 0 without a figure’ (I,iv 185-6) – is reflected in nature as the storm
is reflected in the protagonists’ minds. Such an apocalyptic world
finds no place in Tate’s text, as can be seen from the drastic reduction
of the occurrences of ‘nothing’ and the consequent elimination of the
echo effect that the word has in Shakespeare. As can be seen from the
following table showing the concordances of ‘nothing’, its
occurrences are all taken directly from Shakespeare’s Lear:
Table 11: concordance of ‘nothing’ in Tate
Duke Albany's Party? Nothing, why ask you?
Nothing can come of Nothing, speak agen.
s, Sir. Death, Traytor, nothing cou'd have subd
u are much deceiv'd, in nothing am I Alter'd But
me first to House 'em. Nothing but this dear De
rethee; I tell thee I have nothing to do with thee.
as it exceeds in Truth - Nothing my Lord. Nothi
lly And it shall lose thee nothing. So, now my p
uth - Nothing my Lord. Nothing can come of No
ass? Cou'dst thou save Nothing? didst thou give
s bold to say If you like nothing you have seen t
6. Conclusion
These quantitative and qualitative analyses of the data seem to
confirm the conclusions of ‘traditional’ research, that is the choices of
Tate’s rewrite are in line with neoclassical aesthetics, and this is
confirmed by both the numbers and the concordance lines. However,
this brief analysis is not indented to be final, rather it should be

considered as an attempt to apply Corpus Linguistic notions and tools
to literary studies. Indeed, as Louw has pointed out, the opportunity
for corpora to play a role in literary criticism has increased greatly
over the last decade. From computer-assisted critical readings that aim
at identifying textual features that may support or contradict intuitive
criticism to collocational studies which may provide insights into
aspects of ‘literariness’ and peculiar styles, that is from data-assisted
to data-driven analyses, corpora have demonstrated their usefulness in
literary studies (Louw 1997: 240). As Wynne has stated, even if
theoretical objections to the use of corpora in the study of literature
persist, corpus linguistics and stylistics will converge and overlap
(Wynne 2005: 5-6).
References
Primary sources
Tate, N. 1681. King Lear, ed. by C. Spencer 1965 Five Restoration Adaptations of
Shakespeare. Urbana: University of Illinois Press: 201-274.
Tate, N. 1681. King Lear, ed. by J. Lynch.
http://andromeda.rutgers.edu/~jlynch/Texts/tatelear.html
Shakespeare, W. 1623. King Lear, ed. by M. Best.
http://ise.uvic.ca/texts/Lr_F1.html
Secondary sources
Bosseaux, C. 2001. ‘A Study of the Translator’s Voice and Style in the French
Translations of Virginia Woolf’s The Waves’ in CTIS Occasional Papers 1: 55-
72.
Brusasco, P. 2004. ‘A Computer-assisted Reading of Nathaniel Hawthorne’s The
Scarlet Letter’ in English Studies 2003. Torino: Trauben: 83-95.
Carter, R. 2004. Language and Creativity. The Art of Common Talk. London:
Routledge.
Consiglio, M.C. 2006. Traduzione remake manipolazione: King Lear neoclassico,
unpublished doctoral dissertation.
Frye, N. 1986. Shakespeare. Nove Lezioni. Torino: Einaudi.
Louw, B. 1993. ‘Irony in the Text or Insincerity in the Writer? The Diagnostic
Potential of Semantic Prosodies’ in Baker, M., G. Francis and E. Tognini-Bonelli
(eds) Text and Technology. Amsterdam: John Benjamins: 157-176.
Louw, B. 1997. ‘The Role of Corpora in Critical Literary Appreciation’ in Wichman,
A., S. Fligelstone, T. McEnery and G. Knowles (eds). Teaching and Language
Corpora. Harlow: Addison Wesley Longman: 240-251.
e-Lears 205
Mahlberg, M. 2007. ‘Corpora and Translation Studies: Textual Functions of Lexis in

Bleak House and in a Translation of the Novel into German’ in Gatto, M., and G.
Todisco (eds) Translation. The State of the Art/La traduzione. Lo stato dell’arte.
Ravenna: Longo.
Munday, J. 1998. ‘Computer-assisted Approach to the Analysis of Translation Shifts’
in Meta 43.
Scott, M. 2006. ‘Key Words of Individual Texts’ in Scott, M., and C. Tribble Textual
Patterns. Key Words and Corpus Analysis in Language Education. Amsterdam:
Benjamins: 55-72.
Semino, E., and M. Short. 2004. Corpus Stylistics. Speech, Writing and Thought
Presentation in a Corpus of English Writing. London: Routledge.
Spencer, C. 1965. Five Restoration Adaptations of Shakespeare. Urbana: University
of Illinois Press.
Stubbs, M. 2005. ‘Conrad in the Computer: Examples of Quantitative Stylistics
Analysis’ in Language and Literature 14(1): 5-24.
Tribble, C. 2006. ‘Counting Things in Texts You Can’t Count on’ in Scott, M., and C.
Tribble Textual Patterns. Key Words and Corpus Analysis in Language
Education. Amsterdam: Benjamins: 179-193.
Wynne, M. (ed.). 2005. Developing Linguistic Corpora: a Guide to Good Practice.
Oxford: Oxbow Books.
Zanettin, F. 2001. Ipergrimus, intralinea, ipermedia (online)
http://www.intralinea.it/hypermedia/eng_open.php
Measuring Text Similarity Between the Two Editions of
John Fowles’s The Magus
Yu-fang Ho
Lancaster University
Abstract
John Fowles’s The Magus was first published in 1966. He then revised and republished
it in 1977. My doctoral research is a comparative stylistic analysis of the two editions of
the novel. In this paper, I will explore (i) what differences there are between the two
versions by using particular corpus techniques on them and (ii) to what extent stylistic
investigation and corpus techniques can be usefully combined.
I will briefly introduce how I have used TESAS/Crouch and WCopyfind software to
detect and measure text similarity (in terms of ‘matched’ words, i.e. n-gram overlaps,
stemmers and synonyms) between the two versions of the novel. With the aid of these
two corpus tools, I will present the statistical results of a chapter-by-chapter comparison
to show in quantitative terms the general pattern of revision between the first and
second editions of the novel. I will then discuss the limitations in applying the corpus
tools in relation to my stylistic comparison. I will use textual examples to illustrate the
limitations of applying surface linguistic criteria to computational measurement of text
similarity or text content, especially the difficulty in measuring the content of the texts
involving extensive revision.
Keywords: corpus stylistics; revision; text reuse measurement; n-gram overlap; text
similarity; John Fowles; The Magus; TESAS; Crouch; Wcopyfind.
1. Introduction
Given that my research is a comparative analysis of the two editions of
The Magus, the first question I have to answer is: what exactly are the
differences between them? Hence, I conduct a corpus-based
quantitative comparison, in order to find out the similarities and
differences between the two editions in statistical terms.
I will introduce the two corpus tools that I use for initial text
comparison: (i) the TESAS/Crouch tool for measuring the extent of text
reuse in the domain of journalism, which was developed at the
University of Sheffield and tested at Lancaster University; and (ii)
WCopyfind, a software package developed in 2002 by Professor Lou
Bloomfield of the Physics Department at the University of Virginia,
which is mainly used to detect the possibility, and quantify the degree,
of one text being copied from another.
208 Yu-fang Ho
John Fowles’s revision of The Magus is similar in some ways to the

plagiarizing reuse of the existing written source to create a new version,
given that the revised edition of The Magus contains the same number
of chapters and there is no major structural or narrative overhaul of the
original edition. Hence, I will consider the first edition of The Magus as
the source text and the second edition as the derived/rewritten target
text and use these two corpus tools to assess the degree of similarity
between the two versions of the book. I will present the statistical
results of each chapter-pair comparison and then examine the general
pattern of the changes between the two editions of The Magus.
2. The Magus Corpus

John Fowles’s The Magus is divided into three parts, narrated in
retrospect by the main character, Nicholas Urfe, a middle-class young
English man. Part I (chs. 1 to 9) is set principally in London, where he
meets a young Australian woman called Alison, finishing with
Nicholas on Phraxos, an enchanting Greek island where he has
accepted a teaching post. Part II (chs. 10 to 67) is almost entirely
composed of Nicholas’s bizarre experiences on Phraxos, where he
meets an old man, Maurice Conchis, and is continually involved in a
quest to find out the truth behind Conchis’s manipulations, especially
the true identity of a beautiful woman, ‘Lily’. In Part III (chs. 68 to 78)
Nicholas returns to London, and after a prolonged period of waiting, he
is at last contacted by Alison, the woman he finally realizes he loves.
In order to combine the corpus techniques with my stylistic
comparative research work, I first built a corpus containing the full
texts of the two editions of the book. The content of the corpus is shown
in Figure 1.
No. of word tokens % of tokens
The Magus st nd increased or
1 Edition 2 Edition
Corpus decreased in
(1966) (1977)
M2
Part I (chs.1-9) 18,370 18,237 -0.72%
Part II (chs.10-67) 172,221 191,665 +11.29%
Part III (chs. 68-78) 31,848 31,843 -0.02%
Total 222,439 241,745 +8.68%
Figure 1: The content of The Magus corpus.
Measuring Text Similarity 209
The total number of tokens is 222,439 in the first edition of The Magus
and 241,745 in the revised edition, which indicates an increase of
8.68% in tokens in the revision. In fact, Part I and III show small
decreases, the increase taking place in Part II.
I then identify the changes firstly with TESAS/Crouch, to examine,
chapter by chapter, the overall pattern of text similarity between the
original and the revised version. In section 3, I will introduce the
matching approach of this corpus tool in more detail, especially the two
major functions that are particularly helpful to my initial comparative
research work: (a) text alignment at the sentence level, and (b)
assessing the degree of overall text similarity at the whole document
level. In section 4, I will use another corpus tool, WCopyfind, for text
comparison, to confirm the statistical results obtained from the
TESAS/Crouch tool.
3. Corpus tool for text comparison 1: TESAS/Crouch

3.1 The concept of text reuse and the TESAS/Crouch tool
To explain the development of the TESAS/Crouch tool, we shall start
from the idea of text reuse in the domain of journalism. Although in an
academic environment the reuse of other people’s texts without
acknowledgement is considered to be plagiarism, text copying and
reuse in journalism is standard business practice. In the newspaper
industry most newspapers rely heavily upon press agencies, such as the
UK Press Association (PA), as their primary source of written news
material. Upon payment of a subscription fee, the newspaper is free to
reuse this material verbatim, or modify it in whatever way, without any
need to acknowledge the source (Gaizauskas et al. 2001; Clough et al.
2002a; McEnery and Piao 2003).
In the past few years, some corpora have been built for the study and
analysis of journalistic text reuse, notably (a) the METER (MEasuring
TExt Reuse) project which was carried out at the University of
Sheffield to identify British newspaper articles reuse of texts released
by PA and (b) a similar project carried out at Lancaster University to
explore text reuse in mid-17th century English newspapers. The main
goal of these two research projects was to explore approaches and
algorithms for automatic detection, comparison and measurement of
journalistic text reuse (i.e. text similarity). They wanted to find out not
only whether a source text has been reused or not, but also ‘to what
210 Yu-fang Ho
extent and subject to what transformations’ (Clough et al. 2002b, 152)

it has been reused.
In these two projects, a number of algorithms were tested. The
prototype tool developed from the METER project is called TESAS;
the one developed in Lancaster University is called Crouch. In Crouch,
the corpus tool I am using for this study, the researchers have modified
the existing TESAS functions and added new functions to it. Since
there is not much literature on Crouch, as it is based heavily on TESAS
with only slight modifications, I will introduce in more detail the
METER project and the matching approaches of TESAS algorithm. I
will particularly focus on the major functions which are relevant to my
research and which remain unchanged in Crouch.
3.1.1 The METER Corpus and its annotation
The METER Corpus consists of a set of news stories written by the
major UK news agency (PA), and a set of stories about the same news
events, as published in nine British daily newspapers. Some newspaper
stories are rewritten from the PA source; some are independently
written by the newspapers’ own journalists (for detail of the METER
Corpus, see Clough et al. 2002a, 2002b). The corpus is manually
classified and annotated at two levels, the document level and the
lexical level, by expert journalists with years of practical experience of
the newspaper industry.
3.1.2 General classification at the document level: WD, PD, ND
In order to capture general information about the reliance of a
newspaper story upon PA source text, each newspaper text as a whole is
classified into one of the following three categories (see Gaizauskas et
al. 2001, 219):
Wholly-derived (WD): all content of the target text is derived
only from the PA;
Partially-derived (PD): some content of the target text is
derived from the PA, but other sources have also been used;
Non-derived (ND): no content of the target text is derived from
the PA.
The classification of each text as belonging to one of the above
categories was based upon the journalists’ professional judgements, not
upon linguistic criteria, such as presence of a certain number or length
of shared tokens. By ‘content’ they mean the provision of ‘facts’. In a
wholly-derived newspaper text, all of the facts in it can be mapped to its

PA counterpart, with varying degrees of directness, from verbatim copy
to change of word order, substitution of synonyms or paraphrase. In a
partially-derived newspaper text, some facts can be mapped on to its
corresponding PA text, but it also contains new facts that cannot be
found in the PA. This category represents an intermediate degree of
dependency of newspaper text on the PA. The last category covers
those newspaper articles that are written independently from PA;
namely, none of the facts in the newspaper text can be found in the PA.
3.1.3 Detailed annotation at the lexical/phrasal level:
verbatim/rewrite/new
After the document-level classification, about 400 wholly- or partially-
derived newspaper articles were further annotated in detail. The
detailed annotation attempts to capture reuse within the newspaper text
itself down to the sentence, phrase and word level. Similarly, based on
the judgement of the professional journalists, three categories are used
(Gaizauskas et al. 2001, 219-20; see also Clough et al. 2002a, 2002b):
VERBATIM: text that is reused from PA word-for-word in the
same context;
REWRITE: text that is reused from PA, but paraphrased to
create a different surface appearance. The context is still the
same;
NEW: text not appearing in PA, or apparently verbatim or
rewritten, but used in a different context.
In their annotation, the tagging of the REWRITE materials is generally
less straightforward than the tagging of VERBATIM and NEW, as
rewrite has many forms. According to Gaizauskas et al. (2001) and
Clough (2003), a PA text can be modified or rewritten in the following
ways: a) Reordering: rearrangement of word/phrase/sentence order or
position; b) Substitution: change of original terms with synonyms or
other context dependent substitutable terms; c) Deletion: deletion of
original materials; d) Insertion: addition of minor new materials (e.g.
words like by in passivisation). The newspaper materials falling into
any one of the above four transformations are tagged as REWRITE.
With these classifications in mind, we can now proceed to see in detail
the matching approaches of the TESAS algorithm tested in the METER
project. I will focus in particular on how it measures similarity, i.e. the
degree of overlap, between a text pair.
212 Yu-fang Ho
3.2 Approaches to the measurement of text similarity

For the sake of manageability, the researchers of the METER project
used sentence as the unit to be mapped. Piao (2001) proposes an
algorithm for identifying text reuse based on sentence alignment. It is
assumed that the relationship between a pair of texts can be determined
by examining the relationship between sub-units of the texts, i.e.
sentences. If they can detect relationships between sentences, they can
then assess the overall relationship between whole texts based on these
sentence level relationships. The algorithm has two main stages. In the
first stage, the algorithm splits the derived text and source text into
sentences and searches for alignments at the sentence level. Each
sentence in the derived text is scanned and compared against all of the
sentences in source text to find the best match. This ‘cognate’ matching
approach to sentence alignment will be discussed in 3.2.1. In the second
stage, the algorithm measures the similarity between each aligned
sentence pair. Four statistical scores are extracted to suggest the
likelihood of derivation. The computational measurement will be
discussed in 3.2.2.
3.2.1 Cognate matching approach: n-gram, stemming, synonym
In order to cope with various changes during text reuse, a
cognate-based approach to sentence alignment is adopted for the
METER task. In the task, cognates are defined as a pair of terms ‘that
are identical, share the same stems, or are substitutable in the given
context’ (Clough et al. 2002b, 155). In practice, three types of cognates
are considered (for detail, see Piao 2001; McEnery and Piao 2003):
Verbatim N-grams matching (n 2). N-gram means the
consecutive word sequences. Measuring the overlap of
consecutive word sequences between sentence pair is an initial
and the most straightforward approach to assessing text
similarity.
An extended version of Porter’s English stemmer (Porter,
1980). Porter’s algorithm reduces inflected variants of an
English word into a morphological base form (e.g. to convert the
word ‘degree’ and ‘degrees’ into the base form ‘degre’). It is an
efficient tool for identifying inflectional variants of a single
English word.1 However, the original Porter stemmer cannot deal
with irregular inflectional forms such as ‘thought’ and ‘drank’.
Hence, it was extended to cope with such words in the METER

project.
Synonym list extracted from WordNet. A list of English
synonyms containing about 46,000 entries was extracted from
WordNet.2 Each entry contains two or more basic synonyms.
3.2.2 The TESAS algorithm
Following the cognate approach, three kinds of shared terms between
the aligned sentence pair are identified: a) n-grams, b) identical single
words, and c) word pairs that are synonyms or have the same stem.
Based on the matches, the relationship between the sentence pair is
quantified in terms of three scores: PSD, PS, and PSNG. Each of these
scores reflects different aspects of the relationship between the
sentence pair. The PSD score indicates the extent to which a derived
sentence is dependent on its source sentence(s). The PS score indicates
the proportion of shared terms in the pair of sentences, which reflects
the similarity between the sentences. The PSNG score denotes the
degree of significance of the matched items, i.e. the proportion of
n-grams among the matched terms (see Clough et al. 2002b, 156;
McEnery and Piao 2003, 639). The value of the three scores all range
between 0 and 1. To put it simply, the maximum score, 1, implies a
complete match between the sentence pair; the minimum score, 0,
means that the sentence pair are completely unrelated.
In order to obtain a single overall metric for measuring the similarity
between the whole mapped sentence pair, the three scores are combined
together to create a weighted score, WS. A default threshold of WS,
0.65, is used to determine whether or not a pair of sentences are truly
related. Those sentence pairs which produce a WS score higher than the
threshold are taken to be truly related (for detailed computation,
indication and evaluation of the scores, see Piao 2001; Clough et al.
2002b; McEnery and Piao 2003).
The tool then displays the comparison of a given pair of texts in an
alignment-map, mapping sentences in the derived text to their source
sentences, including null alignment if no match can be found, as shown
in Figure 2.
214 Yu-fang Ho
Figure 2: TESAS/Crouch sentence-alignment mapping example.

The three statistical scores (PSD, PS, PSNG) and the alignment
weighted score (WS) are extracted and displayed in the first column of
the table. The user-friendly graphical interface helps us to identify the
local changes more easily, as the mapped sentences are displayed side
by side in a table, and different types of matched words are highlighted
with different colours. As Figure 2 makes clear, the n-grams shared by
each pair of aligned sentences, and the shared single words (including
those which have undergone inflectional and letter-case changes) are
highlighted in different colors (green and red on screen).
Then, based on the measurements of text similarity between pairs of
sentences discussed above, the TESAS/Crouch tool can assess the
degree of text similarity at a higher level, namely, the document level,
and produce a pie chart reporting the total number of words in the
derived text, the numbers of matched and unmatched words, and the
similarity score of the text pair. Figure 3 shows the comparison result of
ch. 59 shown in the pie-chart.
Figure 3: TESAS/Crouch similarity score sample report.

Ranging between 0% and 100%, a greater percentage indicates a
greater similarity between a text pair, while a smaller percentage
indicates less similarity or greater difference between a text pair. The
extent of the text similarity in chapter 59 of The Magus as measured by
TESAS/Crouch is 29.39%, which indicates a high percentage of
unmatched words, and so much greater revision, in this chapter.
Given that The Magus is a novel of over 600 pages, it is impossible for
the human analyst to keep in mind all the changes scattered over the
two editions. The benefit of the TESAS/Crouch tool for my study is that
it helps me to ascertain the degree of similarity between each chapter
pair and hence the general pattern of revision for the whole novel.
According to McEnery and Piao (2003, 645), the evaluation of the
TESAS/Crouch corpus tool shows that it is ‘capable of detecting and
measuring text re-use with a reasonably high rate of precision.’ Hence,
I will use the similarity scores of each chapter-pair comparison as
statistical indicators to describe the pattern of Fowles’s revision of The
Magus.
216 Yu-fang Ho
3.3 Statistical Result

The similarity scores of the 78 chapter pairs are shown in Figure 4
below (the scores below the average of 87.53% are highlighted with a
darker colour).
Figure 4: TESAS/Crouch Statistics: 78 chapter-pair similarity scores.

Fowles himself has commented on his revision of The Magus in an
interview: ‘I haven’t changed the general line of the story, but I’ve
rewritten a large, mostly the central, part’ (Singh 1980/1981, 186). As
shown in Figure 4, the average similarity score of the novel as a whole
is 87.53%, which is quite high and indicates great similarity (in terms of
matched words) between the two versions. The average similarity score
of Part I of the novel is 97.58% and Part III 90.16%, which reflect
Fowles’s statement that he hardly changed these two parts, Part I in
particular. Most of the changes occur in Part II, which explores the
main character Nicholas’s journey to self-realisation. The average
similarity score for Part II, 74.85%, is considerably lower than the other
two parts.
The general pattern of Fowles’s revision is clearly shown in the

graphical presentation in Figure 5.
TESAS/Crouch:
78 Chapter-Pair Similarity Scores
100%
Similarity Score (%)
80%
60%
40%
20%
0%
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77
Chapter-Pair No.
Figure 5: TESAS/Crouch comparison.

The tripartite divisions of the book are indicated by the dark lines in
ch.9 and ch.67 (the final chapters of Part I and Part II). The chapters
with low similarity scores (e.g. chs. 33-5, 43, 45-7, 55-9) are also the
chapters where, as Binns (1977) and Boccia (1980/1981) point out,
Fowles has made drastic revision on the dialogue among the characters
to make the theme more explicit to his readers.
Next I will explore another corpus tool for text comparison, to see if the
statistical result retrieved from TESAS/Crouch can be upheld.
4. Corpus Tool for Text Comparison 2: WCopyfind

Similar to TESAS/Crouch, the focus of plagiarism detection is also on
how to detect similarity between texts and what to measure. Given that
plagiarism is a form of unacceptable text reuse, it has received
considerable attention and relatively sophisticated computer
programmes have been written for automatic plagiarism detection (see
Clough 2002 for a review). WCopyfind is one such programme.
WCopyfind was developed in 2002 by Professor Lou Bloomfield at the
University of Virginia. This program compares a collection of papers
by ‘look[ing] through them for matching words in phrases of a
specified minimum length. When it finds two files that share enough
words in those phrases, WCopyfind generates html report files. These
reports contain the document text with the matching phrases
underlined’ (Bloomfield 2002). To put it simply, WCopyfind also
218 Yu-fang Ho
adopts an n-grams overlap approach, namely, finding the overlap of

matching consecutive words of length n (where n is derived
empirically). The software allows the user to specify what the
minimum and maximum sizes are for phrases to be checked.
Bloomfield recommends leaving this parameter at 6 words; that is, to
measure the overlap of 6-grams (n 6).
This system also allows for the adjustment of other comparison rule
parameters. Given limitations of space, I do not intend to discuss the
settings of the scanning parameters in detail (for more information, see
Bloomfield 2002). In my study, I used the default parameter settings
recommended by Bloomfield, to find ‘absolute matching’.
Figure 6: A sample html report of WCopyfind text comparison
Figure 6 illustrates chapter 59 comparison report. To examine the texts

in detail, we can click on the files individually or click on the
‘side-by-side’ option to display the pair of files in adjacent panels of a
new browser window. When we view the two files side-by-side, perfect
matches are indicated by red-underlined words. All the matching
phrases are links. If we click on a matching phrase, we will be taken to
the equivalent phrase in the other document in the pair.
In terms of statistical results, WCopyfind generates two numbers of
matches: Total Match and Basic Match. According to Bloomfield
(2002), if the parameter of ‘Most Imperfections to Allow’ is set to zero

(i.e. no imperfections are allowed in the matching), Total Match and
Basic Match will be the same, and Basic Match will be essentially the
value that would have been obtained. Both match entries have 3
subparts (cf. Figure 6): (a) the number of matching words (e.g. 153); (b)
what percentage of File 1 is accounted for by these matching words (e.g.
25%); (c) what percentage of File 2 is accounted for by these matching
words (e.g. 22%). To put it simply, the percentage figures tell us how
much of the files are common to each other.
Following the same procedures as with the TESAS/Crouch corpus tool,
I ran the 78 chapter pairs through the WCopyfind software. Notice, the
TESAS/Crouch statistics is based on M1 as the source text and M2 as
the derived text. In order to make the WCopyfind statistics comparable
with the TESAS/Crouch statistics, I adopt the first percentage of the
Basic Match. The matching percentages of the 78 chapter pairs are
shown in Figure 7 (the chapters with the scores below the average 86%
are highlighted with a darker colour).
Figure 7: WCopyfind Statistics: the 78 chapter-pair matching

percentages.
220 Yu-fang Ho
I then compared the two sets of statistics, the similarity scores

generated from TESAS/Crouch (Figure 4) and the matching
percentages generated from WCopyfind (Figure 7), to ascertain the
difference in the overall pattern of revision across the 78 chapters
between the two editions of The Magus. The general pattern of the
changes retrieved from the two corpus tools is almost identical, as
shown in Figure 8. The result of text comparison via WCopyfind seems
to uphold the result retrieved from the TESAS/Crouch tool.
TESAS/Crouch vs. WCopyfind

78 Chapter-Pair Comparison Result
100%
Similarity Degree
80%
60%
Crouch
40% WCopyfind
20%
0%
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77
Chapter-Pair No.
Figure 8: TESAS/Crouch vs. WCopyfind comparison.

As pointed out earlier, the areas with the lower degrees of similarity in
the bar chart are mainly the chapters where Nicholas has active
interactions or conversations with the other characters (especially chs.
33-5, 43, 45-7, 55-9). The low similarity scores or matching
percentages of these chapters result mainly from the fact that higher
percentage of words in the M2 version are detected as ‘unmatched’. We
then have to ask: are the words ‘unmatched’ because they do not appear
in the M1 version (i.e. are they all new content being added to M2)? The
question can only be answered by conducting a manual textual
comparison. I examined some of these chapters manually and found
that the answer seems to be no. Some of the M1 extracts which have
been paraphrased or rewritten extensively but with the content invariant
in the M2 versions are mapped as ‘unmatched’. The computational
mapping of the ‘matched’ and ‘unmatched’ leads us to consider further
the complexity and difficulty of measuring the ‘content’ of two
expressions. Below I explore in more detail what is implicated in the
statistics of the ‘matched’ and the ‘unmatched’, and show in what

aspects the pure quantitative measurement is advantageous as well as
limited with regard to my stylistic comparison.
5. Advantages and disadvantages of the corpus tools: measuring

the unmeasurable?
It is clear that the two corpus tools explored in sections 3 and 4 can be
of some assistance in textual comparison. By and large, there are two
major advantages in applying the two corpus techniques to my
comparative study. Firstly, with the aid of the corpus tools, I am able to
detect and measure text similarity in terms of ‘matched’ words (i.e.
n-grams overlaps and substitutable terms) between the two versions of
the novel. We can thus know roughly how much of the M1 edition
remains unchanged. The statistical chapter-by-chapter comparison also
helps to show the general pattern of the revision in quantitative terms
(cf. Figure 8).
Secondly, as The Magus is a novel of over 600 pages, there is great
difficulty in identifying manually the various kinds of linguistic
changes over the two versions. With TESAS/ Crouch, its sentence
alignment mapping approach and the user-friendly interface help us to
identify the subtle changes in the texts to be compared (cf. Figure 2).
With WCopyfind, all the matching phrases (n-gram overlaps) are
highlighted and inter-linked (cf. Figure 6). If we click on a matching
phrase, we are taken to the equivalent phrase in the other document of
the pair. It thus becomes easy for us to identify the texts that remain
unchanged in both versions, and accordingly to trace the texts that have
been revised.
However, there are still some limitations in applying the corpus tools in
relation to my stylistic comparison. Given that a) WCopyfind is merely
supporting software I used to confirm the TESAS/Crouch result, and b)
it shares more or less the same problems with TESAS/Crouch, I will
focus the discussion below on the limitations of the TESAS/Crouch
tool only. The major limitation in the corpus technique is related to the
‘content’ issue, i.e. the difficulty of measuring text content in statistical
terms.
222 Yu-fang Ho
(1) Limitations in applying surface linguistic criteria to measuring

text content.
It is difficult for the software to define precisely the similarity or
comparability of ‘content’ between two expressions. In terms of
deciding whether two different expressions mean the same thing or not,
many scholars have suggested different criteria. Enkvist (1988)
proposes using ‘syntactic roles’ and ‘quantifier scope’ as criteria to
define the content of different linguistic expressions. As he states,
‘those word-order permutations that do not change the basic syntactic
roles of constituents, and which do not change quantifier scope, are
cognitively equivalent’ (1988, 147-8). For example, John kicked Mary
and Mary kicked John are non-equivalent because syntactic roles have
been switched. Everybody in this room speaks three languages and
Three languages are spoken by everybody in this room are
non-equivalent because they differ in quantifier scope. Enkvist’s two
examples illustrate the limitations of the corpus tool in textual
comparison.
If we compare the sentence pair in the first example, TESAS/Crouch
reports: the number of matched single words is 3.0 (i.e. John, Mary,
kicked), and 0.0 words are unmatched. The similarity score is 100%.
The two sentences contain exactly the same words, with the syntactic
roles of John and Mary switched. The switch of syntactic roles makes
the sentences non-equivalent in content, yet they are mapped as 100%
‘matched’.
Similarly, if we compare Enkvist’s second example, their similarity
score is 98.303%, which resulted from the matched n-grams (in this
room) and the matched substitutable terms (Everybody-everybody,
speaks-spoken, Three-three). However, although they are mapped as
highly similar, the content of the two sentences are actually
non-equivalent in that they differ in quantifier scope, as Enkvist
suggests.
In both cases, the high similarity scores between the obviously
non-equivalent sentence pairs reveal the limitation of measuring the
text similarity by simply applying surface linguistic criteria, i.e. by
counting matched and unmatched words.
(2) The issue of identifying synonyms and stemmers as ‘matched’.
As the TESAS/Crouch statistics shows, 87.53% of words are ‘shared’,
i.e. ‘matched’ between the two versions. Those matched words include
verbatim n-grams, synonyms and stemmers (see the discussion on the
Cognate approach in 3.2). We have to bear in mind that the main
purpose of developing TESAS/Crouch was to detect and measure
whether a source text has been reused or not in the domain of
journalism. What has been reused was more important in their research
than how a text has been reused. Hence stylistic variation between the
source text from the UK news agency and the reused texts in the other
newspapers is not an issue for the journalistic business practice. This is
the reason why synonyms and inflectional variants of a single word are
identified as ‘matched’. However, the main purpose of my study is to
explore whether there is any change in text style between the two
versions, and if so, how their text styles are different. If I take the
similarity scores at face value, I might miss some important stylistic
variation in the scattered changes in synonyms and stemmed words
which have been counted by the software as ‘matched’ words between
M1 and M2.
For example, let us compare the PA news agency source with the
subsequent rewritten text published in The Sun (the example is taken
from Clough et al. 2002a, p. 1678):
Example 1:
Original (PA) A drink-driver who ran into the Queen Mother’s
official Daimler was fined £700 and banned from driving for two
years.
Rewrite (The Sun) A DRUNK driver who ploughed into the
Queen Mother’s limo was fined £700 and banned for two years
yesterday.
This simple example illustrates the types of rewrite that occur in a
single short sentence. TESAS/Crouch reports as below:
224 Yu-fang Ho
The size of the derived text from The Sun is 20 words; 4 words are
unmatched (ran/ploughed, official Daimler/limo, yesterday). The
similarity score of the sentence pair is 80%, including the shared
n-grams 15 words and 1 matched substitutable term (drink/ DRUNK).
Despite the high degree of text re-use, however, the style of rewrite
from The Sun is markedly different from the PA report. The style
variation between the two news texts is conveyed both in the ‘matched’
and ‘unmatched’ words.
Let us first look at the ‘unmatched’ words. The addition of yesterday is
typical of all newspaper stories in that the re-used news text is usually
published the day after the PA copy is produced. It is a non-stylistic
deictic change due to a time-frame change. Nonetheless, the use of
ploughed and limo in The Sun is relatively informal or colloquial
compared with ran and official Daimler in the PA text. The lexical
changes result in different stylistic effects.
With regard to the ‘matched’ words, apart from the verbatim n-grams,
the word drunk is counted as a matched substitutable stemmer of drink.
If our comparison focuses only on the unmatched words, the stylistic
effect (i.e. using capital letters to capture readers’ attention) is likely to
be unnoticed simply because they are counted as ‘matched’. Example 1
illustrates the reason why we can not take the TESAS/Crouch similarity
score at face value in stylistic terms, since style variation can be found
both in the ‘matched’ and ‘unmatched’ linguistic items.
(3) Limitations in measuring content of texts involving extensive
revision.
In Example 1, the expressions official Daimler and limo refer to the
same object in the text world context, yet they are counted as
‘unmatched’ linguistic items. The example shows that it is difficult for
the software to define precisely similarity or comparability of ‘content’
between two lexical items. The difficulty is even greater in texts
involving extensive revision.
In regard to the definition of content, Doležel (1971) suggests that ‘Text
content can be defined as the aggregate of meaning associated with a
text paraphrase which is referentially equivalent to the original text; in
other words, the original text expression and its content paraphrase
denote the same content’ (Doležel 1971, 103, emphasis mine). Here is
an adapted example taken from Jean Boase-Beier (2004, 26-7), which
exemplifies what Doležel suggests:
Example 2:
(a) Tim is a good teacher.
(b) Tim always prepares well for classes and gets on really well
with the students and he gets the best feedback reports in the
Department.
Although the linguistic forms of the two sentences are so different, we
can still say quite confidently that they convey the same content. There
are three factors that help us to make such a judgment. Firstly, and most
importantly, sentence (b) is referentially equivalent to sentence (a), in
that they have the same referent, Tim. Secondly, although his role as a
teacher is not explicitly mentioned in sentence (b), it can be easily
inferred from the terms associated schematically with teacher, i.e.
classes, students, feedback reports, Department. Thirdly, the evaluative
judgments on Tim’s role as a teacher are positive in both expressions:
from good in sentence (a) to well, really well, the best in sentence (b).
However, TESAS/Crouch reports that none of the words in (b) is
detected as ‘matched’ with (a). The similarity score between the
sentence pair is 0.000%, even though (a) is a summary of (b). The
following sentence pair from The Magus is the sample texts involving
extensive revision.
Example 3:
(a) A different kind of tension had arisen, mainly because there
were things in him that I could not relate (and which he knew
and intended I could not).... (M1, ch. 30, 184)
(b) He clearly meant me not to be able to relate the conflicting
sides of his personality. (M2, ch. 30, 192)
There is a set of referential invariants between the two texts (e.g. He
refers to Conchis; I refers to Nicholas.). Secondly, there is certain
semantic association between the M1 and M2 expressions. For example,
things in him that I could not relate can be associated with the
conflicting sides of his personality; He knew and intended I could not
can be associated with He clearly meant me not to be able to, etc.
However, the TESAS/Crouch report of this example is similar to that of
Example 2. None of the words in (b) is matched with (a). The similarity
score between the sentence pair is 0.000%.
With the broader definition of ‘content’, we perceive the above
examples of text revision (Examples 2 and 3) as two different
226 Yu-fang Ho
expressions saying roughly the same thing, even though their linguistic
structures are so different. These examples illustrate the limitation of
applying surface linguistic criteria to the computational measurement
of text similarity, especially that of the texts involving extensive
revision.
Figure 9 shows what is implicated in the ‘matched’ and ‘unmatched’
statistics in stylistic terms.
Figure 9: Implications of the text similarity statistics.

The similarity score between M1 and M2 as a whole indicates that
87.53% of words are shared or matched between the two versions, and
roughly 12.47% of words are unmatched. The matched words include
verbatim n-grams, synonyms and stemmers. The unmatched words
include: (i) texts with roughly equivalent content but which have been
detected as ‘unmatched’ in that their surface linguistic structures are so
different (as illustrated in example 3); and (ii) texts that appear in the
M2 edition only, i.e. new scenes, new dialogues or new narrative
description, etc. As I have exemplified, stylistic variation can actually
be found in texts with different degrees of revision, from the matched to
the unmatched, from a one-word change to a change at sentence level,
or possibly even at discourse level.
What I intend to explore in this research is whether the changes Fowles
has made across the chapters result in any difference in text style
between the two editions. The statistic is not revealing in this regard.
Moreover, the new content which is included in the 12.47% of
unmatched words raises an issue I have to consider further in my
comparative stylistic study. Is invariant content or the same content
really a necessary condition of my stylistic comparison of the two
editions? If ‘content’ should be held constant for every stylistic
comparison, how do we account for these additional words (i.e. new
scenes, new content)? How can we compare the new content if it does
not exist in the original version? Because there is no alternative to be
compared with, does it mean that the additional new content is not
important? Certainly not. Some new content (i.e. new scenes, dialogues,
or narrative descriptions) might also be important in relation to readers’
interpretation.
According to Short (1994), stylistic comparison does not have to be
conducted in a restricted way. It is possible for someone to talk about
different authors’ styles, in spite of the fact that the content expressed
are unlikely to be equivalent. As he says, ‘[f]or styles to be established
irrespective of content, regularity of choice with respect to particular
style features becomes paramount’ (Short 1994, 4376, emphasis added).
To put it simply, stylistic comparison in my study can be made at
different levels. At the micro-level, we can compare the stylistic effect
of two linguistic choices that express roughly the same content. At the
macro-level, we can explore if there are any recurrent linguistic
features that Fowles has frequently used in the revised texts,
particularly in the additional passages in M2, and which result in
different text styles between the two versions of The Magus. The
micro-level qualitative stylistic comparison of some equivalent extracts
and the macro-level quantitative linguistic comparison will be
conducted in my research afterwards.
Endnotes
1
For details of Porter’s stemmer, see
http://www.tartarus.org/martin/PorterStemmer/index.html or
http://www.comp.lancs.ac.uk/computing/research/stemming/general/porter.htm.
2
WordNet is a lexical database for the English language. It groups English words into
sets of synonyms called synsets, provides short definitions, and records the various
semantic relations between these synonym sets. WordNet was created in 1985 and is
being maintained at the Cognitive Science Laboratory of Princeton University under
the direction of Professor George A. Miller (See http://wordnet.princeton.edu/).
References
Binns, Ronald. 1977. ‘A New Version of The Magus’ in Critical Quarterly 19(4):
79-84.
Bloomfield, Lou. 2002. The Plagiarism Resource Center.
http://plagiarism.phys.virgina.edu.
Boase-Beier, Jean. 2004. ‘Knowing and not knowing: Style, intention and the
translation of a Holocaust poem’ in Language and Literature 13(1): 25-35.
Boccia, Michael. 1980/1981. ‘“Visions and Revisions”: John Fowles’s New Version of
The Magus’ in Journal of Modern Literature 8(2): 235-46.
Clough, Paul. 2002. ‘Plagiarism in natural and programming languages: an overview
228 Yu-fang Ho
of current tools and technologies.’ (No. CS-00-05): Department of Computer

Science, University of Sheffield, UK.
Clough, Paul. 2003. Measuring text reuse. PhD Thesis, University of Sheffield.
Clough, Paul, and Robert Gaizauskas et al. 2002a. ‘Building and annotating a corpus
for the study of journalistic text reuse.’ Conference Proceedings of the 3rd
International Conference on Language Resources and Evaluation (LREC-2002)
(pp.1678-91), Los Palmas de Gran Canaria, Spain.
Clough, Paul, and Robert Gaizauskas et al. 2002b. ‘METER: MEasuring TExt Reuse.’
Conference Proceedings of the ACL-2002, University of Pennsylvania,
Philadelphia, USA.
Doležel, Lubomir. 1971. ‘Toward a structural theory of content in prose fiction’ in
Chatman, S. (ed.) Literary style: a symposium. London: Oxford University Press:
95-110.
Enkvist, Nils Erik. 1988. ‘Style as parameters in text strategy ’ in Willie van Peer (ed.)
The taming of the text: explorations in language, literature and culture. London:
Routledge: 125-51.
Fowles, J. 1966. The Magus. London: World Book.
Fowles, J. 1977. The Magus. London: Vintage.
Gaizauskas, Robert, and Jonathan Foster et al. 2001. ‘The METER corpus: a corpus for
analysing journalistic text reuse.’ Conference Proceedings of Corpus Linguistics
2001, Lancaster, UK.
McEnery, Tony, and Scott S. L. Piao. 2003. ‘A tool for text comparison.’ Conference
Proceedings of the Corpus Linguistics 2003 Conference, Lancaster, UK.
Piao, Scott S.L. 2001. ‘Detecting and measuring text reuse via aligning texts’ Technical
paper: CS-01-15. University of Sheffield, UK.
Porter, M.F. 1980. ‘An algorithm for suffix stripping’ in Program 14(3): 130í137
Short, Mick. 1994. ‘Style: Definitions’ in Asher, R. (ed.) The Encyclopaedia of
Language and Linguistics. Amsterdam: Pergamon Press: 4375-4378.
Singh, Raman K. 1980/1981. ‘An Encounter with John Fowles’ in Journal of Modern
Literature 8(2): 181-202.
‘My Dearest Minnykins’: Style, Gender and Affect in
19th Century English Letters
Merja Kytö, Uppsala University, Sweden

and Suzanne Romaine, Oxford University, UK
Abstract
Personal letters may offer a potential glimpse into more informal and colloquial language
because they are concerned primarily with the creation of intimacy and positive face
relations. Address terms constitute a key site for the encoding of interpersonal
attitudes, and are thus central to understanding social relations among participants in
speech events. We show how the use of intensifying superlatives such as dearest in 19th
century epistolary formulae emerged as markers of stylistic affect against the backdrop of
increasing conventionalization of dear. We trace a path of semantic evolution involving
increasing subjectification of dear as it progressed from adjectival status to become an
address term.
Keywords: adjectives; adjective comparison; comparative adjectives; superlative
adjectives; diminutives; affect; politeness; face relations; gender differences; address
forms; subjectification; epistolary formulae; corpora; CONCE (Corpus of Nineteenth-
Century English); BNC (British National Corpus); letters; 19th century English; genre.
1. Introduction
In this paper we examine style, gender and affect as factors
constraining patterns of variation in address terms in 19th century
English letters contained in the CONCE (Corpus of Nineteenth-
Century English). The letters present an excellent opportunity to study
individual linguistic features, and in particular, to compare male and
female usage. As historians of language have often pointed out, letters
may offer a potential glimpse into more informal and colloquial
language. Görlach (1999: 149), for example, argues that letters mirror
social relations between sender and addressee to a very high degree,
equaled perhaps only by spoken texts. Letters function in a sense like
conversations between correspondents.
This impression is indeed empirically confirmed by Biber’s (1988)
multidimensional analysis of a variety of spoken and written genres in
modern English. Although personal letters are written and thus share
some features with other written genres, they are closer on the whole
230 Merja Kytö and Suzanne Romaine
to spoken genres than other forms of writing. They typically have an

involved (rather than informational) focus, and share many of the
interactive and affective characteristics of conversation (e.g. high
frequency of personal pronouns I and you, contractions, emphatics
(including more and most), and private verbs such as feel, love, etc.).
Personal letters scored higher in terms of the use of these features than
did all other non-conversational spoken genres such as broadcasts.1
Like Trials, they are high in wh-questions, as one would expect,
because these are used primarily in interactive discourse. Biber and
Finegan’s (1989a) analysis of 500 texts and 24 genre categories
revealed that personal letters were actually considerably more
affective than conversation. Only fifteen texts (3%) made extensive
use of affect markers; 7 of the 15 texts were personal letters,
comprising 70% of the cluster they designated as ‘emphatic marking
of affect’.
Scholars have examined some of the linguistic and pragmatic
conventions of letter writing in different historical periods. From a
diachronic perspective, Nevalainen (2004: 181) observes that both the
material circumstances of letter writing and its discursive practices
have changed with the times. Biber and Finegan (1989b) argue that
over the last four centuries letters have drifted toward a more oral
style as manifested in an increase in linguistic features indicative of
increased involvement, but less elaboration and less abstraction. More
specifically, they found that 17th and 18th century texts tended to be
moderately or extremely literate, but the 19th century was marked by a
transition towards more oral styles (Biber and Finegan 1989: 507).
Fitzmaurice (2002) has explored the use of particular epistolary
strategies in the construction of subjectivity between sender and
receiver in 18th century literary letters. She explains that letters are
concerned primarily with the creation of intimacy and positive face
relations.
Because address terms constitute a key site for the encoding of
interpersonal attitudes, they are central to understanding social
relations among participants in speech events. The topic has been
prominent within sociolinguistics dating back to Brown and Gilman’s
(1960) pioneering work. More recent diachronic investigations of the
evolution of address terms have also proved illuminating (see the
‘My Dearest Minnykins’ 231
papers in Taavitsainen and Jucker 2003). Nevalainen (2004: 5) has

emphasized that an analysis of letter writing as a social and discursive
practice cannot ignore the social hierarchy reflected in forms of
address or in other words, how sender and receiver position
themselves in relation to one another. In this task attention must be
paid to concepts such as politeness, face, power and solidarity.
Certainly, the availability of historical corpora such as the Helsinki
Corpus and the Corpus of Early English Correspondence (CEEC) has
opened new vistas that have been exploited by researchers such as
Nevalainen and Raumolin-Brunberg (1996, 2003). A recent issue of
the Journal of Historical Pragmatics (2004) devoted to letter writing
covers the period from 1400 to 1800. Much of our knowledge about
earlier forms of address in personal letters comes from studies based
on CEEC, and its extensions, databases of personal letters written
between 1410 and 1800. According to Nevala (2004: 275), the letters
in CEEC from the 17th and 18th centuries are marked by an increase in
the epistolary expression of intimacy. Bijkerk (2004) has examined
how new formulas expressing positive politeness such as yours
sincerely arose during the 18th century to replace older ones such as
your most humble servant. These studies still leave the 19th century
relatively neglected, but they suggest promising avenues to pursue. As
Kytö, Rudanko and Smitterberg (2000: 85) have pointed out, although
the 19th century was an age of great exploration and new discoveries,
the English language in that period remains largely unexplored
territory. After providing a brief overview of CONCE, especially the
letters, we will show how the use of intensifying inflectional
superlatives such as dearest in epistolary formulae emerged as
markers of stylistic affect against the backdrop of increasing
conventionalization of dear. We trace a path of semantic evolution
involving increasing subjectification of dear as it progressed from
adjectival status to become an address term and then part of an
epistolary formula.
2. Brief overview of CONCE

CONCE is a joint project between the University of Tampere and
Uppsala University that fills a much needed gap in multigenre corpora
covering specific periods in the history of English. The corpus
contains ca. one million words of material from seven genres (debates,
trials, drama, fiction, letters, history and science) divided into three
sub-periods corresponding roughly speaking to the beginning, middle,
and end of the 19th century: 1800 to 1830, 1850 to 1870, and 1870 to
1900. Although the number and length of texts vary with genre, the
corpus is well suited for in-depth studies of individual linguistic
features. The inclusion of a large number of personal letters written by
both women and men provides an opportunity to look at both genre-
and gender-related variation. The letters amount to 343,631 words or
slightly more than a third of CONCE. Table 1 provides word counts
for male and female writers in the three periods.
Table 1 Word counts for the letters by women and men writers in
CONCE, excluding the words within reference codes and text-level
codes (from Kytö, Rudanko and Smitterberg 2000: 90, Table 5)
Period Men Women
1 52,353 69,271
2 68,776 62,340
3 40,737 50,154
The twenty-seven authors (14 men and 13 women) whose letters are
contained in CONCE include some of the most well known literary
and intellectual figures of the 19th century. Professional, friendship
and family circles intersect and overlap in various ways in these
letters. To begin with rather a dramatic example of a complicated
intertwining, Percy Shelley and Lord Byron were friends, and both
had an affair with the half-sister of Shelley’s wife, Mary. The literary
community known as the Wordsworth circle centered on William
Wordsworth and Samuel Taylor Coleridge. Robert Browning and
other literary figures regularly visited Coleridge at home. Coleridge
was married to Sarah Hutchinson, the younger sister of William
Wordsworth’s wife, Mary, who was a childhood friend of William’s
beloved sister, Dorothy. Robert Southey was a close associate of
Coleridge and Wordsworth, with the three sometimes referred to as
‘the Lake poets’. Geraldine Jewsbury maintained a correspondence
with her friend, Jane Welsh Carlyle, wife of Scottish historian,
Thomas Carlyle. Although his letters are not in CONCE, Carlyle was
a formidable influence on his contemporaries such as Matthew
Arnold, the Brownings, and Charles Dickens. He enjoyed worldwide
fame after the publication of his book on the French revolution in
1837, which Dickens relied on for historical details in his novel A Tale
of Two Cities (1859). Dickens was a friend and rival of novelist
William Thackeray, and also published some of Elizabeth Gaskell’s
work. Anne Thackeray Ritchie, oldest daughter of William Thackeray,
was her father’s biographer and a novelist in her own right. During the
1860s everyone was discussing Charles Darwin’s theory of the
evolution of species, but especially so Samuel Butler, who had his
own ideas on the subject. Thomas Huxley also corresponded with
Darwin.
The material also includes a considerable amount of correspondence
between husbands and wives (e.g. Robert and Elizabeth Barrett
Browning, Mary and William Wordsworth). Indeed, the Brownings’
courtship can be said to have begun in a fan letter that Robert wrote to
Elizabeth in 1845 praising her poetry. The courtship continued over
the course of a 20 month exchange of nearly 600 letters. These
examples hint at just some of the intricate, interpersonal and
professional connections among the authors of the letters in CONCE.
Epistolary networks such as these may have been significant vectors
for the rise and spread of linguistic innovations. Tieken-Boon (1999),
for example, has argued that John Gay may have been responsible for
introducing the epistolary formula yours sincerely in letters to his
closest friends, thus breaking with previous formulae such as your
most humble servant.
3. Superlatives and patterns of adjective comparison in CONCE

letters
The point of departure for understanding the use of superlatives in
epistolary forms of address lies in our earlier corpus-based
investigations of adjective comparison from Middle English to
contemporary English (see Kytö 1996; Kytö and Romaine 1997, 2000,
2006). We have primarily focused on the competition between the
so-called inflectional comparative/superlative (e.g. commoner/
commonest), historically the older form, and the newer periphrastic
construction (e.g. more/most common). In a mnemonic nutshell, our
studies have addressed the question of whether it is more
common/most common to say commoner/commonest or to say more
common/most common. Our analysis of patterns of variation in
adjective comparison in CONCE established a steady increase in the
use of inflectional forms throughout the 19th century, with a

significant difference in the proportion of inflectional to periphrastic
forms at the beginning and end of the century (Kytö and Romaine
2006). This is a continuation of a long term trend we have traced
through various corpora in an on-going series of studies referred to
above.
Even by the end of the 19th century, however, the system does not
match the present-day distribution exactly. If we take the British
National Corpus (BNC) as representative of contemporary written
English, inflectional forms account for 74.7% of the instances of
comparative forms (and periphrastic forms for 25.3%; Leech and
Culpeper 1997: 373).2 In CONCE, inflectional forms also predominate
over the periphrastic forms for both comparative and superlative
adjectives, but the figures are somewhat lower than in the BNC,
namely, 63% of comparatives and 60.2% of superlatives are
inflectional (compared with 37% and 39.8% periphrastic,
respectively). Our analysis of the use of comparative forms revealed
that the last decades of the 19th century had already come close to
contemporary usage while the early decades of the century still
belonged to the past. The picture for superlatives, however, is more
interesting as well as more complex because 19th century English
displays more variation. We realized that further quantitative and
qualitative analysis would be required in order to illuminate some
potentially interesting gender variation, particularly in Period 2.
Figure 1 shows the incidence per 100,000 words of inflectional and
periphrastic comparative and superlative forms of adjectives in
CONCE as a whole compared with the letters.3 Although there is some
variation across the three sub-periods and seven genres, this chart
relies on averages for the 19th century as a whole.4
200
180 CONCE
Incidence per 100,000 words
160 Letters
140
120
100
80
60
40
20
0
inflectional periphrastic inflectional periphrastic
comparatives comparatives superlatives superlatives
Figure 1 Incidence of inflectional and periphrastic comparative and

superlative forms of adjectives in CONCE as a whole compared with
letters
Generally speaking, although letters contain more superlatives than
other genres, Figure 1 reveals a roughly similar pattern in the letters
and CONCE as a whole, with the exception of the category of
inflectional superlatives. This is due to a rise in inflectional
superlatives in Period 2 that is quite a bit steeper in the letters than in
other genres. This can be seen more clearly in Figure 2, showing the
incidence of inflectional and periphrastic comparative and superlative
adjective forms per 100,000 words in the letters for each of the three
time periods. Nevertheless, in both the corpus as a whole and the
letters in particular, this steep rise is due largely to the use of
epistolary formulae of the type contained in our title example ‘dearest
Minnykins’.
300 inflectional
comparatives
250
periphrastic
200 comparatives
150 inflectional
superlatives
100
periphrastic
50 superlatives
0
Period 1 1800-1830 Period 2 1850-1870 Period 3 1870-1900
Figure 2 Incidence of inflectional and periphrastic comparative and

superlative adjective forms per 100,000 words in CONCE letters (see
Kytö and Romaine 2006: 202-203, Tables 7.2a and 7.2b for raw
figures).
Period 2 contains 200 (60.6%) out of a total of 330 forms of dearest,
most of which occur in opening and closing formulae. The superlative
dearest can be considered a highly marked affective form contrasting
with the unmarked address term dear, which is nearly four times as
frequent (N=950). Using the WordSmith Tools keyword analysis to
compare the letters with the rest of the CONCE as a reference corpus
revealed that both forms are among the top 15 keywords in the letters
(dear is 7th and dearest, 13th). This is one indication of the
conventionalization of dear as an address term during the period, and
of the intimate, involved nature of the letters.
Superlatives are, however, complex from a semantic and pragmatic
point of view because they admit three possible interpretations that
grammarians have referred to as relative, absolute and intensifying.5
As Huddleston and Pullum (2002: 1161) note, although it might seem
that plain, comparative and superlative forms express progressively
higher degrees of the property denoted by the adjective in question
(e.g. kind, kinder, kindest), it would be a mistake to interpret them in
this way. In other words, if Ellen is kinder than Bill, this does not
entail that Ellen is kind. Ellen is not kind in any absolute sense, but
merely kind relative to Bill. She might well be cruel; all that is
asserted is that she is better with respect to the attribute of kindness

than Bill. Likewise, if Ellen is the kindest of the three, this means the
same thing as saying that Ellen is kinder than the other two. There is
no difference of degree, as Jespersen (1949: 392) insisted some time
ago when he said that ‘the superlative does not indicate a higher
degree than the comparative’ in ordinary usage. Huddleston and
Pullum (2002: 1161) sum up the situation as follows:
The system of grade, therefore, is not a matter of different
degrees ordered on a scale. The plain form differs from the
others in that it does not express comparison. The main
difference between the comparative and the superlative is that
for the most part they express different kinds of comparison: the
comparative is used predominantly for term comparison while
the superlative is used exclusively for set comparison.
Thus, when Mary Wordsworth writes to her husband William in (1)
about ‘that dearest & sweetest of all thy letters’, she is singling out a
particular letter from the set of all his letters and anchoring it in an
explicit comparative set. Namely, where X is the entity being
compared, A is the adjective and Z is the set of relevant items for
comparison, she is in effect saying that X is the A-est among Xsmod as
delimited by Z.6 The quality denoted by the adjective is not predicated
in a strict or absolute sense because some explicit modification is
contained in the sentence or surrounding context that restricts the
application of the superlative. Claridge (2006: 76) found that the
majority of instances of relative superlatives in CONCE belonged to
this type. Nevertheless, Wordworth’s combination of dearest and
sweetest heightens the superlative effect.
(1) William that dearest & sweetest of all thy letters, which
affected me so much on Saturday when I received it that I could
not trust myself to speak in reply to it – that dear letter as I was
reading it over in bed before I slept – last night (as I had done
the preceeding one) caught fire, & the corner was burn off
where thou hadst so feelingly traced the progress of that
affection which in different situation thou hast felt towards
me – (Mary Wordsworth, 1800-1830, p. 246).
In (2) Mary Wordsworth writes to her sister Sarah Hutchinson,
describing D as ‘the kindest creature in the world’ and ‘the best of all
persons’. Although the frame of reference encompasses the whole
world, the explicit modifying prepositional phrase, in the world, limits
the proper name syntactically. Likewise, in the case of the noun
phrase, the best of all persons, D is at the extreme end, i.e. the best of
all people imaginable to Mary Wordsworth. This suggests that
superlative expressions involve quantification over degrees (see
Sharvit and Stateva 2002: 457). Thus, kindest and best indicate a very
high degree along some scale.
(2) D., who is the kindest creature in the world, being able to
speak German – and the best of all persons in making her way
amongst all sorts of strange people (Mary Wordsworth, 1800-
1830, p. 60).
While relative superlatives are fairly easy to identify due to the
criterion of contextual restriction, absolute uses leave matters more
open to subjective interpretation, and have a wider scope. The limits
to absolute superlatives are often extreme and/or objectively
unverifiable (see Farkas and Kiss 2000: 437 on ‘absolute absolute
superlatives’). As Bolinger (1977: 28) put it, ‘the superlative can jump
any adjective to the outer limits of the scale.’ The meaning of the
adjective itself and the range of constructions it occurs in can also play
a role in affecting the uses made of it. In (3) the superlative noun
phrases the best most attached tender-hearted creature and the
greatest comfort occur without any scope qualifying expressions and
are therefore absolute in an unqualified way.7 Absolute superlatives
are less well anchored referentially and always have connotations of
extremity. Hence, they have strong emotive potential.
(3) The ship is all in a bustle and honest dear faithful Eyre is
blundering away at the baggage; he is the best most attached
tender-hearted creature: and its the greatest comfort to me to
have him. (William Thackeray, 1850-1870, p. 110).
Intensifying uses of the superlative are, like the absolute ones, usually
unmodified as well as close to them semantically and pragmatically.
Grammarians are not always in agreement concerning their
classification. Claridge (2007: 136-138) observes that absolute uses of
the superlative may shade into intensifying uses, with no clear
boundaries between them. Nevertheless, there are relatively clear

cases. In (4) George Eliot writes to her intimate friend, Cara Bray,
addressing her as dearest Cara, which is clearly intensifying. We will
be discussing these uses later in some detail in section 4, so we focus
for the moment on the other superlative, namely, dearest in the phrase
three or four dearest dears. To have three or four dearest dears should
be a logical impossibility in the absolute reading because there can be
only one unique instance, the most extreme one (the most dear) on a
scale (of dears). This example is also intensifying, and clearly
formulaic, like the salutation dearest Cara, and the formulaic
expression nearest and dearest.
(4) I am busy today – have three long letters to write, so good
bye, dearest dear. I have three or four dearest dears – they
take it in turn week and week about. (George Eliot 1850-1870,
p. 16)
These examples illustrate that presence or absence of modification
clearly restricting the comparison set is important, but not always
sufficient for delimiting the types. Many cases are tricky because
superlative adjectives are themselves largely evaluative in nature.
Consider, for instance, the closing in (5) to a letter Charles Darwin
wrote to his wife (and cousin), Emma.
(5) I have got some more letters to write, though I wrote six
longish ones yesterday so farewell my best & dearest of wives.
(Charles Darwin, 1850-1870, Vol. VII, p.80).
Again, the prepositional modifier of wives might suggest a relative
reading. However, Darwin had only one wife, so he is not comparing
her to a set of his wives, but to all possible wives imaginable. This
example is therefore also an instance of the intensifying superlative in
a formulaic usage. This example illustrates a rather common variant of
the intensifying superlative, namely (the) + superlative adjective + of
+ NP. Leech and Culpeper (1997: 369) claim that this construction
carries an emphatic (i.e. intensifying) meaning more often in
inflectional than in periphrastic superlatives.
Although the relative and absolute senses do not appear to be sensitive
to a morphological type of comparison, some have claimed that the
intensifying superlative occurs only with periphrastic comparison
(Rusiecki 1985: 137, Huddleston and Pullum 2002: 1165). However,

where evaluative adjectives occur in predicative position without a
definite determiner as in (6), the meaning is potentially ambiguous
between a relative superlative (i.e. ‘You are the most kind and good of
all people’) or an intensifying superlative (i.e. ‘You are extremely
kind and extremely good’, see Quirk et al. 1985: 466).8
(6) My dear Mrs. Kinney, You are always most kind and good.
(Robert Browning, 1850-1870, p. 98).
The intensifying meaning can also occur in both inflectional and
periphrastic superlative noun phrases with an indefinite determiner, as
in (7a) and (7b).
(7a) – altogether make me think Dr. Lane innocent & that it is
a most cruel case.
(Charles Darwin, 1850-1870, Vol. VII, p. 116)
(7b) I leave it to T[om] M. to make my apologies for troubling
you with a commission – which is to procure me 5 shades of
drab rug worsted from the darkest drab nearly black to white
included – 2 pounds of each shade; and 6 pounds of a lightest
bright yellow. (Sara Hutchinson, 1800-1830, p. 54)
Further corroboration of the high degree of interpersonal and involved
in the letters can be found in Claridge’s (2006: 86) figures for the rate
of occurrence of different functions of superlatives across the different
genres. Letters show the highest use of so-called intensifying
superlatives, with nearly half (49%) of the superlatives belonging to
this type. Letters also show the lowest incidence of relative
superlatives (22%).9 In addition, Geisler’s (2002) multidimensional
analysis of the CONCE genres confirms that Letters (along with
Drama and Trials) score high on Dimension 1 (involved v.
informational production), as one would expect, but rather low on
Dimension 5 (abstract vs. non-abstract information). We turn now to
some of the gender differences in CONCE, particularly in Period 2.
4. Gender-differentiated patterns of variation in the use of

dear/dearest
Women letter writers in CONCE use more instances of adjective
comparison, both comparatives and superlatives (4.6 instances per
1,000 words, N=834) than men (4.2 instances per 1,000 words,
N=676, Kytö and Romaine 2006: 206). Although there are gender
differences in the use of both inflectional and periphrastic
comparatives, the differences over time are not statistically significant,
so we will not discuss them further here. On the whole, most of the
adjectives in CONCE that are used in both inflectional and
periphrastic superlatives are evaluative in tone and are used to express
the speaker’s attitude. Some of the differences observed in men’s and
women’s inflectional superlatives are related to the word structure
obtaining for particular adjectives (see Kytö and Romaine 2006: 209-
210).
However, it may not be coincidental that the two double superlatives
in (8) and (9) were produced by a woman.10 There were only three
clear instances of double forms in CONCE, two of which were found
in the letters of Anne Thackeray Ritchie.11 Both these examples are
intensifying, but the emphasis of most delightfulest is further
reinforced by the intensifier very.
(8) I needn’t tell you that Hester and Billy instantly sprang from
their beds when I appeared the messenger of sweetness and
delight. The bonbon tongs had an immense success, the
bonbons are an immenser, and Hester says those red ones are
“the most deliciousest things she ever ate.” (Anne Thackeray
Ritchie, 1870-1900, p. 192).
(9) Mrs. Ritchie presents her compliments to Monsieur Denis.
She also hugs him and gives him a kiss.
His father has piles of maps of Normandy and Brittany and
is making out the very most delightfulest tour. (Anne Thackeray
Ritchie, 1870-1900, p. 247).
The possibility of this hybrid variant containing more/most with
inflectional adjectives arose when the new periphrastic construction
appeared in the thirteenth century. This meant that from the Middle
English period onwards three alternative forms of comparison for an
adjective such as dear existed: inflectional (dearer/dearest),
periphrastic (more dear/most dear) and double (more dearer/most
dearest). The continuing influence of standardization led prescriptive
grammarians to condemn such forms as non-standard, and they have
gradually disappeared from standard written English. Kytö (1996), for

example, did not find any double forms for the post-1640s sub-period
of the Early Modern English period represented in the Helsinki
Corpus. Today they are found primarily in the most colloquial
registers of spoken English.
Figure 3 shows the results for the incidence of inflectional and
periphrastic superlatives in men’s and women’s letters for the three
time periods. The use of superlatives noticeably peaks in Period 2 for
both men and women, but women lead in the use of the inflectional
superlatives over men in all three periods.
Men inflectional
3.5 superlatives
3
2.5 Men periphrastic
2 superlatives
1.5
Women inflectional
1
superlatives
0.5
0 Women periphrastic
Period 1 Period 2 Period 3 superlatives
1800-1830 1850-1870 1870-1900
Figure 3 Inflectional and periphrastic superlatives in men’s and

women’s letters; incidence per 1000 words (see Kytö and Romaine
2006: 207, Table 7.3b for raw figures).
Despite the fact that both men and women employ dearest, women
use the form to a much greater extent. The ratio of dear to dearest is
5.97 for men and 1.83 for women.12 For women, dearest ranks 6th in
keyness (compared to dear in 14th place), while for men dear ranks
5th, and dearest, 25th. The differences can be seen more clearly in
Figure 4 containing the figures for inflectional superlatives for men
and women with and without tokens of dearest. If all tokens of
dearest are removed, this eliminates 319 out of 638 adjectives. This
reduces the amount of data by half! The reduction in data is less
dramatic for men than women. As far as men are concerned, 175
superlatives remain after removing 90 tokens of dearest (34% of the

data) from a total of 265. For women, however, eliminating 229
examples (61% of the data) leaves only 144 superlative adjectives out
of the original total of 373. Although men’s and women’s usage still
differs slightly in Periods 1 and 2 after tokens of dearest have been
excised, it is virtually the same for Period 3. Now, however, men lead
slightly in the use of inflectional superlatives in the first two periods,
but the differences are minimal. By examining the letters more
carefully, we will show how certain patterns of adjective comparison,
particularly the use of superlatives in epistolary formulae, function as
stylistic markers of affect.
3.5
3
2.5
Men [-dearest]
2
Women [-dearest]
1.5
Men [+dearest]
1
0.5 Women [+dearest]
0
Period 1 1800-1830 Period 2 1850-1870 Period 3 1870-1900
Figure 4 Inflectional superlatives in men’s and women’s letters (with

and without dearest); incidence per 1000 words (see Kytö and
Romaine 2006: 208, Figure 7.2)
In order to understand the differences between men’s and women’s
use of dearest and dear, we must first establish what is conventional
for the 19th century in order to see whether and how individuals depart
from the norm, and what rhetorical functions, if any, such departures
may serve. Since the 17th century, dear has become the ordinary polite
form for addressing an equal, but the kind of formula in which it was
embedded varied according to the nature of the relationship between
the writer and addressee in terms of parameters such as social
position/distance, age and sex. Nevala (2004: 378) suggests that
address forms in letters may be positively or negatively polite or a
mixture of both. Forms of address that orient to the addressee’s
positive face are usually informal and intimate, e.g. first names,
nicknames. By contrast, negative politeness manifests itself in terms
of titles and honorifics. Where power and status differences as well as
social distance are minimal (e.g. among family members), reciprocal
positive politeness may prevail. Yet, even where status differences are
minimal, but distance goes beyond the nuclear family, positive
politeness cannot be entirely taken for granted. As distance grows, the
options become more limited. Positively polite address is directed
toward inferior correspondents by superior ones and the latter receive
forms indicating negative politeness. Those equal in power may
address one another with positive or negative politeness, depending on
relative distance.
Even in family letters husbands usually addressed their wives with
titles such as madam and your ladyship, and occasionally as dear wife.
Wives used deferential address to their socially superior husbands. In
letters between mutually distant correspondents titles and honorifics
such as your lordship, sir, etc. prevailed. Nevala (2004: 283) found
that among gentry and professionals, whose letters make up the largest
portion of the material in CEEC, a wider variety of formulae is
employed, including terms of endearment such as use of first names,
nicknames, etc. such as my dearest love, my beloved nephew. In the
18th century, however, forms of address in literary correspondence
became responsive to the writer’s attitude or mood toward the
recipient. This meant that writers could depart from epistolary
conventions in order to signal affective overtones.
Although Nevala’s remarks pertain to the 17th and 18th centuries, they
are relevant to the 19th. Charles Dickens, for instance, addressed
Elizabeth Gaskell as (My) dear Mrs. Gaskell, but his wife as my
dearest Kate. In his fan letter to Elizabeth Barrett, Robert Browning
wrote: ‘I love your verses with all my heart, dear Miss Barrett.’
Although Elizabeth Barrett Browning addressed her friends Isabella
Blagden and Anna Brownell Jameson as dear/dearest Isa/Mona Nina,
she never progressed to the use of reciprocal first names with her
lifelong friend Mrs. James Martin, whom she always addressed as My
dear/dearest Mrs. Martin. During this time the use of first names was
largely confined to family. The use of nicknames and hypocoristics
was even more intimate. For men, the use of bare surnames was
halfway between formal and intimate (Görlach 1999: 41).

Despite the nuances available in the 19th century system of epistolary
address, it represents a considerable simplification of the situation that
obtained in the Middle English and early modern periods. Raumolin-
Brunberg (1996) shows that forms of address during this time were
quite complex, containing a salutation consisting of a head noun with
one or more modifiers that could be coordinated and/or intensified.
Even if Elizabeth Barrett Browning’s My dearest Mrs. Martin sounds
somewhat distant and formal to modern English ears, it becomes
extremely familiar and almost intimate compared with Margaret
Paston’s greeting to her husband John as Ryth reuerent and worsepful
husbon, I recomawnde me to Ǻow wyth alle myn sympyl herte
(Margaret Paston, Letter 124, 1441?).
Davis (1965: 236) observed that 15th century English letters of a
formal, respectful type often opened ‘with a long sequence of
conventional phrases and sentences constructed with minor variations
on a regular pattern’. Although the fullest exemplars can be seen in
letters from children to parent, Davis (1965: 237-238) writes that the
same conventions were widely used throughout the fifteenth century
and later in many official and private letters. The formula contained
seven divisions, some with subdivisions:
i. address term commonly beginning with right followed by an
adjective of respect such as worshipful, beloved, worthy, and a
noun such as sir, husband, father, etc.
ii. a formula commending the writer to the recipient, often
accompanied by an expression of humility. If the letter is
addressed to a parent, there may be a request for a blessing.
iii. expression of desire to hear of the addressee’s wellbeing.
iv. a prayer for the continuation and increase of welfare.
v. news of the sender’s welfare.
vi. a report of the writer’s good health.
vii. offering of thanks to God for wellbeing.
The regular observance of the seven part formula in the same order
across 15th century English letters led Davis (1965: 240-241) to
conclude that although no letter writing manuals survive from the
period, there are many manuscripts written in England during the 14th
and 15th centuries containing guides to the composition of letters in

French. They offer model letters suitable for various occasions, e.g.
from son to mother, etc. Davis (1965: 243) speculates that various
expressions of politeness and respect generally observed in English
15th century letters were regularized in polite letter writing
conventions in France.
The seven part pattern is a continuation from the previous century.
Davis attributes the earliest use of these epistolary conventions to
Chaucer’s Troilus and Criseyde, conventionally dated ca. 1383. The
earliest genuine examples found by Davis come from two letters
written by Sir John Hawkwood in 1392 and 1393 to Thomas Cogsale.
The first contains the opening ‘Dere trusty & welbiloved frend
hertliche I grete you…’, and the second, ‘Dere S’ I grete you wel…’
Davis (1965: 238) singles out as significant Hawkwood’s use of
simple dear instead of the more humble and elaborate salutation
contained in the first letter. Despite its modern predominance, the use
of dear on its own was rare until the 17th century. Davis’s citations for
this sense of dear antedate those found in the OED (see section 5).
Nevala (2004) credits Hill’s letter writing manual of 1721 for
introducing a trend towards the expression of increasing intimacy,
especially between friends and other correspondents equal in power.
Terms of endearment and nicknames from wives to husbands started
to increase gradually in the course of the 17th century. Raumolin-
Brunberg (1996: 175-176) found that the typical form of address
adopted at that time by couples from all ranks was the reciprocal use
of my dear(est). As Nevala (2004: 275) points out, the social setting of
the late modern English period was still too strongly hierarchical for
expressions of intimacy to permit the spread of terms of endearment to
all levels of usage. It is against the backdrop of the trend towards
expression of intimacy and the introduction and subsequent
conventionalization of dear that we must examine men’s and women’s
use of dearest.
Table 2 reveals a great range of individual variability in the incidence
of dearest in men’s and women’s letters per 1,000 words. The men’s
scores range from 0 (Samuel Butler) to 1.54 (William Thackeray),
with the average at .52. The women’s scores range from .25 (Mary
Shelley) to 3.32 (Eliza Wilson), with the average at 1.27. Table 2
pinpoints the individuals who are high/low users of superlative

dearest relative to the norms for men and women. Among the men
Thackeray and Macaulay stand out as high users, whereas among the
women, Elizabeth Barrett Browning, Eliza Wilson and May Butler
greatly exceed the average for their group. Infrequent users among
women include Anne Thackeray Ritchie, Mary Shelley, and Sarah
Hutchinson; among men the infrequent users are William Blake,
Samuel Butler, Robert Southey, Samuel Coleridge, Lord Byron and
Charles Darwin. We can now see more clearly why the rise in Period
2 for inflectional superlatives is so steep in Figure 2. All of the women
in Period 1 and Period 3 (with the exception of May Butler) are below
average users of dearest, and all those in Period 2 (except Elizabeth
Gaskell) are above average, in particular, Elizabeth Barrett Browning
and Eliza Wilson. Similarly for the men, the high users tend to be
concentrated in Period 2, with Darwin being the exception. Periods 1
and 3 contain low users, with the exception of John Keats in Period 1
(who frequently addressed his lover Fanny Brawn as dearest) and
Matthew Arnold in Period 3 (who addressed his mother and son as
dearest).
Some of the individual variability is due to the nature of the
correspondence included in the sample, in particular, the extent to
which the authors are writing to those within or outside the family
circle. All of Eliza Wilson’s letters containing dearest are addressed to
her husband Walter, whom she usually addressed as dearest Walter.
All of May Butler’s letters containing dearest are addressed to her
older brother Samuel, whom she usually addressed as dearest Sam. In
Butler’s letters to her in CONCE, he addresses her only as (My) dear
May.
Table 2 Incidence of dearest in men’s and women’s letters per 1,000

words, with number of instances in square brackets
Period Men dearest/1,000 Women dearest/1,000
words words
Period 1 William 0 Jane Austen .48 [N = 5]
1800- Blake
1830
Lord Byron .31 [N = 4] Sarah .34 [N = 7]
Hutchinson
Samuel .21 [N =2] Mary Shelley .25 [N = 3]
Coleridge
John Keats .75 [N = 7] Mary .99 [N = 28]
Wordsworth
Robert .08 [N = 1]
Southey
Period 2 Charles .29 [N = 6] Elizabeth B. 2.72 [N = 39]
1850- Darwin Browning
1870 Robert .86 [N = 7] George Eliot .7 [N = 8]
Browning
Thomas 1.25 [N = 18] Elizabeth 1.62 [N = 18]
Macaulay Gaskell
W.M. 1.54 [N = 26] Geraldine 1.25 [N = 14]
Thackeray Jewsbury
Charles .65 [N = 8] Eliza Wilson 3.32 [N = 56]
Dickens
Period 3 Thomas .37 [N = 2] May Butler 2.99 [N = 26]
1870- Huxley
1900 Matthew .56 [N = 5] Mary Sibylla .76 [N = 10]
Arnold Holland
Samuel .08 [N = 1] Christina .76 [N = 8]
Butler Rossetti
Thomas .43 [N = 2] Anne T. Ritchie .35 [N = 7]
Hardy
TOTAL .52 [N = 89] 1.27 [N = 229]
Some of the differences are due to variables that would be difficult or

impossible to control. Nineteenth century women, for instance, did not
have access to the same professional circles that their male
contemporaries moved in. Hence, there are no letters between
professional women to match Darwin’s letters to his scientific
colleagues. Naturally one would expect the tone of professional
correspondence between men to be rather different from that of
friendly personal correspondence between women. Indeed, Biber

(1988: 132–33) found that professional letters are quite different in
terms of their degree of involved v. informational focus. They
contained fewer features referring directly to personal emotions, or to
interaction between readers and writers. Moreover, professional letters
can be highly informational and writers often take considerable
trouble to revise them. Such letters therefore often show a high degree
of lexical variety and informational density compared with personal
letters. Bergs (2004) proposes a typology of letters based on two
dimensions, author-addressee roles and relationships on the one hand
and communicative function on the other. As noted earlier, however,
the social relationships between correspondents represented in
CONCE were often dense and multiplex, so that it is not always
possible to neatly distinguish these two parameters. Darwin and Sir
Joseph Dalton Hooker, for instance, were professional colleagues as
well as close friends. Interestingly, Claridge (2006: 84-85) has
independently identified Darwin as a frequent user of superlatives
more generally. He exceeds the corpus average in both his letters and
especially in his scientific writing, where he produced 21.5% of the
examples for this genre. This suggests that some individuals may
show a particular proclivity towards the use of the superlative.
Our findings concerning the high use of inflectional superlatives in
Period 2 receive additional support from Geisler’s (2003)
multidimensional analysis of the CONCE letters, which singles out
Period 2 as extremely high in Dimension 1 (involved vs. informational
production). Geisler (2003: 92-95) found a large jump from Period 1
to 2 for women, indicating a substantial increase in involvement. His
analysis pinpoints Eliza Wilson and Geraldine Jewsbury as women
with very high scores. Interestingly, our results for dearest also put
Wilson at the top for Period 2. In fact, with the exception of Geraldine
Jewsbury, who is slightly above the norm in her use of dearest, our
ranking of the Period 2 women based on dearest follows the one
Geisler obtained for Dimension 1. Geisler’s list from lowest to highest
involvement is: Eliot < Gaskell < Browning < Jewsbury < Wilson.
Ours for the use of dearest from lowest to highest is: Eliot < Jewsbury
< Gaskell < Browning < Wilson. Male letter writers have lower scores
than women for Dimension 1 in Periods 1 and 2, but higher ones in
Period 3. Geisler (2003: 104) also found significant differences
between men and women on Dimension 5 (impersonal vs. personal

style) that were due to men’s higher use of agentless passives. This
may also be an indication of the more mixed nature of the men’s
letters in terms of their role relations to their addressees.
Closer examination is required in order to reveal the differing
conventions for male and female letter writers in the use of dearest as
a term of address. Most of the men use dearest in addressing women,
usually wives, female lovers, sisters or mothers, but rarely for male
addressees. In fact, there are only four instances, where men address
males with dearest. Two of these involve fathers writing to sons.
Matthew Arnold addresses his son Richard as my dearest Dick. The
other three cases involve men writing to close male colleagues or
friends, as in (10), a letter from Charles Darwin to Joseph Dalton
Hooker, one of his closest friends, and (11) in a letter from Robert
Browning to his friend, journalist and editor John Forster, and (12)
from William Makepeace Thackeray to Edward Fitzgerald, his best
friend from university days at Cambridge.
(10) My dearest Hooker. You will, & so will Mrs Hooker, be
most sorry for us when you hear that poor Baby died yesterday
evening. (Charles Darwin, 1850-1860, Vol. VII, p.121).
(11) And let my son be friends with yours (“as Shafalus to
Procrus, I to you”) as you, dearest old friend, with your ever
affectionate R.B. (Robert Browning, 1850-1870, p. 62).
(12) My dearest old friend I mustn’t go away without shaking
your hand and saying Farewell and God bless you – (William
Makepeace Thackeray, 1850-1870, p. 98).
The use of dearest followed by surname or noun phrase old friend is
less intimate than the use of dearest followed by first name or
nickname. Huxley also corresponded with Hooker, but addressed him
as Dear Hooker. It is significant that male equals do not address one
another with dearest followed by first name. The only example where
this occurs is in (13), where Samuel Coleridge is writing to the
Wordsworth household, which included William, his wife Mary, his
sister Dorothy, and Sarah Hutchinson, Mary’s younger sister, whom
Coleridge later married. His opening salutation greets them
collectively as my dearest friends, and so we understand the closing
dearest to refer to them collectively also rather than to single out

William.
(13) In the mean time, God bless you, dearest William,
Dorothy, Mary, S., and my godchild, (Samuel Coleridge, 1800-
1830, p. 459).
By contrast, women address each other as well as men (usually
husbands, but also brothers) with dearest followed by first name or
nickname. Example (14) contains three superlative forms of dear, two
inflectional and one periphrastic. In addition to addressing her
husband William as dearest, Mary Wordsworth also uses the
periphrastic most dear in referring to one of his previous letters, and
the inflectional dearest to refer to a letter from William’s sister
Dorothy. Sometimes the women even use forms such as dearest,
dearest, or dearest dear, or the superlative followed by title rather
than first name, e.g. My dearest Mrs. Martin. Repeating the
superlative dearest is also a means of intensifying it even further as in
(15).
(14) My dearest William I have to thank you for a most dear
letter which I received on friday & for the envelope which
brought me dearest Dorothy’s letter to day – for which with
you I am very thankful, for your uneasiness had made me
uneasy also – (Mary Wordsworth, 1800-1830, p. 168).
(15) It will cost me nothing to bid it farewell, fear not, dearest,
dearest Walter, for the charms it had for me were always
superficial in the extreme. (Eliza Wilson, 1850-1870, p. 514).
What is perhaps the most salient example of the affective use of
dearest in CONCE can be seen in (16), where William Makepeace
Thackeray addresses his youngest daughter, Harriet.
(16) My dearest Minnykins. I am in beautiful rooms that are so
awfully noisy that they drive me out of the town and yesterday I
ran away to a place called Balloch at the end of Loch Lomond
and had a row on the lake for 3 hours with Mr. James in
attendance and could not see Ben Lomond for the mist: –
(William Thackeray, 1850-1870, p. 33).
The affectionate intimate nature of the greeting is heightened by his
use of the pet name Minnykins. The source for Harriett’s nickname
Minny is not entirely clear, but it may be etymologically linked to her
middle name, Marian, since Minny/Minnie is often a diminutive of
Mary or Wilhelmina. Minny/Minnie also became popular in the 19th
century as an independent female name. Minnykins is morphologically
actually a triple diminutive because the suffix -kins is itself composed
of two suffixes with diminutive overtones. The -kin suffix from Dutch
-kijn/-ken ‘small’ is combined with the hypocoristic -s suffix. Adding
these two suffixes to the pet name Minny (which itself contains the
diminutive -y often, but not exclusively, found in female nicknames)
heightens the affective tone to an even greater extent. Jespersen (1948:
9-10) called such suffixes ‘fondling-endings’, and thought that
English used them sparingly by comparison with other European
languages such as Italian and Dutch. However, they turn up frequently
in baby talk (e.g. beddie-byes, dindins) and diminutive affectionate
names (e.g. cuddles, sweetiepies). Mühlhäusler (1983: 78) collected a
wide variety of such pet names in Valentine messages published in
newspapers such as The Guardian and The Times (among them the
following, Tiddles, Nibbles, Snoops, Toots).
There may be even more significance to the name Minnykins, as
further investigation reveals that it may be related to manikin from the
Dutch compound minnekijn/minneken [minne ‘love’ + kijn/ken
’small’] meaning ‘sweetheart, beloved, darling, friend’. According to
the OED, there are several senses in which the term manikin as noun
and adjective is used. As a noun it may be used in reference to a
young girl or woman as a term of endearment. It is now rare or
obsolete, but in this meaning it appears to be related to manikin pin, a
kind of small pin used by women for their clothing (also now
obsolete). As an adjective it refers to a person or thing diminutive in
size or form, and was employed in expressions of affection.
Interestingly, a minikin name was a pet name. It is not clear whether
ordinary English speakers would have identified the -kin of minikin as
a diminutive suffix in its own right (albeit originally from Dutch).
Nevertheless, it is relevant to raise the possibility that there may be
some influence from minikin in the choice of Harriet’s nickname
because Thackeray clearly was familiar with the word, and used it, as
the citation in (17) for adjectival minikin from the OED attests. Thus,
Minnykins may be the ultimate affectionate minikin name!
(17) They [sc. pastorals] are to poetry what charming little

Dresden china figures are to sculpture: graceful, minikin,
fantastic. (Thackeray, 1853, English Humourists iv. 176).
Another greeting of this type in (18) including a pet name with a
hypocoristic suffix is T.H. Huxley’s salutation, dearest Babs, to Ethel,
the youngest of his five daughters. It is not entirely clear where this
pet name comes from because Babs is usually regarded as a nickname
for Barbara. At the time this letter was written in 1891, ‘Babs’ was, at
age 25, already an adult. In another letter written in 1885 Huxley
addresses her as dearest Ethel, but this letter is not included in
CONCE. He mainly addressed her as dearest Babs, but in other
correspondence not included in CONCE he addressed her as dearest
Pabelunza. Is this macaronic? He addressed his eldest daughter as
(my) dearest Jess.
(18) DEAREST BABS – 1. “Ornary” or not “ornary” B is merely
A turned upside down and viewed with the imperfect
appreciation of the mere artistic eye! (T.H. Huxley, 1870-1900,
p. 310).
5. Discussion
By comparison with correspondence from prior centuries, 19th century
personal letters have a distinctly different emotional tone. Although
there are a variety of reasons for this in the changing historical
circumstances, a telling indicator of a shift towards increasing
intimacy and familiarity can be seen in the choice of address forms.
Our examination of some of the parameters affecting the choice of
address terms has paid particular attention to the use of superlative
evaluative adjectives. By combining quantitative and qualitative
methods of analysis, we have tried to show how dearest functioned as
a sensitive marker of involved style, particularly for women.
Some of the typical features of personal letters as opposed to written
genres have been identified as characteristic of women’s
conversational style. As noted in 4, some of these have been linked to
Dimension 1 in Geisler’s (2003: 104) multidimensional analysis. On
the whole, men’s letters tended to be more information-oriented, and
abstract, whereas women tended to use more involved and situated
language. Gender scholars, such as Coates (1996), have argued that
the greater use of involvement, display of emotions and personal

affect reflects the functions conversation serves for women, namely, to
show concern for others, and to maintain solidary and cooperative
relationships. Indeed, the top two words identified for their keyness by
a Wordsmith Tools keyword analysis are the personal pronouns I and
we. For men, the top two words are I and my, and me is 6th; we is 43rd.
The pronouns my and me are 6th and 13th respectively in the women’s
keywords list. The reflexive myself is 37th for men, but 99th for
women. These differences are in line with suggestions from Coates
and other gender scholars that men put themselves first as discourse
topics, while women relate more to others. However, they are also tied
to the more personal nature of women’s letters in CONCE. Geisler
(2003: 95) also found that women letter writers in CONCE used more
emphatics and private verbs such as feel, love, think, etc. Our keyword
analysis ranked feel as 21st in the women’s letters, but 81st in the
men’s, while it ranked love as 28th in the women’s keyword list, but
255th in the men’s.
Further linguistic indicators of the involved tone of the letters support
Fitzmaurice’s (2002: 19-20) suggestion that letters are constructed as
an ongoing conversational exchange similar to a telephone call. Eliza
Wilson refers to one of the letters her husband, economist and
journalist Walter Bagehot, wrote from his club as ‘chatty’. One might
almost imagine the text of (19) as a transcript of a telephone
conversation rather than a letter from Elizabeth Barrett Browning to
her friend, Anna Brownell Jameson, because Browning uses terms one
would employ in reference to the spoken channel such as voice, hear,
talk(ing), listen (rather than handwriting, write, read). Indeed, it is
almost as if they were heeding the advice offered by Busbridge (1909:
10), author of a letter writing manual published at the beginning of the
following century, when he recommended that ‘every effort should be
made in order that a letter may resemble a pleasant conversation as
much as possible, so that the description of it as ‘a talk on paper’ may
really be true.’
(19) How pleasant, dearest Mona Nina, to hear you, though the
voice sounds far! Try and come back to us soon, and let us talk,
or listen, rather, to your talking. (Elizabeth Barrett Browning,
1850-1870, p. 220).
Dimension 1 of Geisler’s multidimensional analysis also included

attributive adjectives, but he found that men scored higher than
women. Although Geisler does not provide details, this difference is
no doubt due to the choice of particular adjectives and their functions,
along the lines we have argued above. Some time ago, Lakoff (1975)
highlighted the use of both emphatics and certain so-called ‘empty’
adjectives (e.g. cute, divine, adorable, etc.) as a hallmark of ‘women’s
language’. Biber and Finegan (1989b) also identified adjectives as a
linguistic feature contributing to affect in English texts, but they did
not explore the dimension of gender. In quite another line of research
Hall (1995: 199), who interviewed workers engaged in creating sexual
fantasies over the telephone, found that women consciously employed
strategies to make their speech sound stereotypically feminine, among
them, some of the hallmarks Lakoff identified as belonging to
‘women’s language’. One woman explained how she worked to create
a marketable feminine persona ‘by using lots of adjectives’ and
‘words which are very feminine’. Our analysis of dearest suggests that
from being ‘empty’, it is charged with affect, while dear has become
empty or semantically bleached and pragmatically weakened over
time.
These two facts need to be seen in tandem. Once a formula becomes
routinized, it depreciates in value or loses its rhetorical force. Another
is then recruited to take its place. In terms reminiscent of Gabelentz’s
(1891: 241) comparison of linguistic forms to state employees who are
hired, promoted and pensioned off, as new workers line up to take
their place, Dahl (2001) suggests that intensifying modifiers lose their
force over time. Dahl uses the term ‘rhetorical devaluation’ to refer to
this process, but sees it not as an end stage, but as part of an on-going
cycle of inflation, depreciation and renewal of linguistic forms. Titles
and forms of address are particularly susceptible to this kind of
inflationary cycle. As Raumolin-Brunberg (1996: 170) has explained,
the inventory of adjectives and participles used as honorifics and
terms of endearment has undergone continual change in response to ‘a
need to create more expressive and less stereotyped forms of
expression’. In fact, she found in her examination of letters in CEEC
from 1420 to 1680 that within each subperiod of 50 to 80 years
structurally simpler forms of address were followed by structurally
more complex ones.
Indeed, Thomas Carlyle could well have been writing about what has
happened to address forms when he wrote that ‘so many highest
superlatives achieved by man are followed by new higher; and
dwindle into comparatives and positives!’ (1837 The French
Revolution I. v. ix, cited in OED). As Jespersen (1949: 395) pointed
out, ‘the almost universal tendency to exaggerate’ often leads people
‘to use the superlative where they mean only a very high degree’. If an
element that was originally emphatic or intensifying in function gets
used too often when no emphasis is required, it may then become
useless in its original function. Does this, however, point to the demise
of dear as a formal courtesy and general address term, as Safire (2006)
has suggested?
Once the use of dear had become so routinized that it lost its affective
meaning in formulae such as Dear Sir/Madam/Mr./Ms, etc., it required
intensification in order to convey expressive meaning. Writers could
then use most dear or dearest to intensify the conventional formula, or
repeat the terms, or add other intensifiers such as very, etc. (e.g. my
dearest Mona Nina, my dear friend). Such usages hark back to 16th
century practices, as can be seen in the use of My most dere lorde and
fader in 1503 by Queen Margaret of Scotland (eldest daughter of
Henry VII of England) to address her father, and her use of Derest
broder in 1516 to her brother, Henry VIII. The range of intensifiers
and other modifiers have also varied over time, as can be seen in 16th
century formulas such as right dere and welbeloved, right trusty trusty
and wellbeloved, etc. During the 19th century, however, the fact that
George Eliot addresses her friend Cara as dearest dear and then refers
to her three or four dearest dears in (4) is indicative of both the
pragmatic weakening of dear and the intensification of dearest.
Figure 5 traces the pragmatic progression of dear from an adjective of
common Germanic origin expressing personal feelings of regard to its
use as an address term indicating the same in an epistolary formula. In
other words, it follows a path of increasing subjectification so that it
becomes more embedded in the speaker’s egocentric world view. The
earliest OED citations for some of its senses date back to Old English
(e.g. 1000, se deora sunu ‘the dear Son’ referring to Christ). The
meaning most closely related to the use of dear as a term of address is
that of someone beloved or regarded with personal feelings of high
esteem and affection. The OED suggests, however, that this sense is
derived from an earlier one denoting ‘esteemed, valued’ rather than
‘loved’, but that ‘the passage of the one notion into the other is too
gradual to admit separation.’ As an address term, the OED explains
that dear indicates affection or regard, with citations such as fader
dere beginning in 1250, with other early citations including Mi dere
frende (1314) and Dere syr (1489). This is particularly the case in
introductory salutations in letters, where the OED adds that
contemporary ‘Dear Father, Brother, Friend, Dear John, and the like,
are still affectionate and intimate, and made more so by prefixing My;
but Dear Sir (or Dear Mr. A.).’ The OED also notes that the adjective
is often used absolutely in the sense of ‘dear one’, especially in
dear/dearest or my dear/my dearest addressed to a person. Safire
(2006) observes that the casualness of e-mail has led to the
replacement of dear with first name, Hello/Hi or nothing.
Such an event would have been unthinkable to 19th century
correspondents. We can gauge the dramatic reversal in tone by
replacing Becky Sharp’s greeting in (20) to Amelia Sedley, her best
friend from school, with ‘Hi or Hi Amelia’.
(20) MY DEAREST, SWEETEST AMELIA
With what mingled joy and sorrow do I take up the pen to write
to my dearest friend! Oh, what a change between to-day and
yesterday! Now I am friendless and alone; yesterday I was at
home, in the sweet company of a sister, whom I shall ever, ever
cherish! (Thackeray, 1847-1848, Vanity Fair, Chapter VIII,
lines 1-5).
Linguistic ADJECTIVE ADDRESS TERM EPISTOLARY

form FORMULA
Meaning esteemed, valued affection, beloved greeting/salutation
Citations • se deora sunu • fader dere • Dere trusty &
‘the dear Son’ (1250, Genesis and welbiloved frend
(1000, Cynewulf, Exodus) (1392, Sir John
Juliana) Hawkwood)
• Mi dere frende • Dere S’
(1314, Romance of (1393, Sir John
Guy of Warwick) Hawkwood)
• Right dere and
• ‘Dere syre’, sayd welbeloved
the duke Naymes, (1450, Queen
‘ye sende vs for Margaret
noughte.’ • My most dere lord
(1489, Caxton, and fader
Four Sonnes of (1503, Queen
Aymon) Margaret to Henry
VII)
• Dear Cara
• Dear Mrs. Gaskell
• Dear Sir/Madam
• Dearest Mrs. Martin
• Dearest Babs
• My dearest
Time span 1000 Minnykins, etc.
1250 (CONCE)
1392
>>>> increasing subjectification >>>>>>>>>
Figure 5 Semantic evolution of dear

In this letter from Thackeray’s novel, Vanity Fair, Becky has just left
the Sedley household. It seems to us no accident that Thackeray, who
scored the highest among the men in use of dearest, has captured well
the feminine emotional overtones in its use in epistolary exchanges
between close women friends. We have tried to show how a detailed
quantitative and qualitative examination of one adjective can

illuminate both changing conventions in the evolution of address
systems in English and the long term change towards an increasing
use of inflectional adjective comparison.
Endnotes
1
These generalizations apply only to personal letters.
2
Leech and Culpeper (1997) examined only ten of the most frequent disyllabic
adjectives in the written portion of the BNC. Kytö and Romaine (1997: 335) looked at all
the adjectives in the spoken portion, where the figures are even higher: 84% of
comparatives and 73% of superlatives are inflectional.
3
Our analysis relies on the classification of adjectives adopted in our previous work in
order to facilitate direct comparison. As a whole, CONCE contains a total of 1,779 tokens
of comparative adjectives and 1,901 superlative adjectives, but we have included only
non-defective adjectives, i.e. cases where forms of comparison are based on the same root
as the positive form. Instances of comparative forms containing ‘umlaut’ (e.g. elder/older,
eldest/oldest) are also included in the group of non-defective adjectives. The excluded
defective group involves instances where the comparative and superlative do not derive
from the same root as the positive (e.g. good/better/best). Kytö and Romaine (1997: 331-
333) provide further details on the use of the non-defective and defective forms of
adjective comparison in the history of English.
4
Both comparatives and superlatives show a statistically significant trend towards
increasing use of the inflectional forms (see Kytö and Romaine 2006: 199, Tables 7.1a
and 7.1b for raw figures).
5
Unfortunately, not all grammarians use the same terminology.
6
The notation is adapted from Rusiecki (1985).
7
Superlative noun phrases such as the simplest problem/the most difficult problem,
for instance, have a quantificational reading that Fauconnier (1980: 60) has argued is
not referential, but pragmatic. A sentence such as John can solve the most difficult
problem implies that John can solve any problem. Likewise, John cannot solve the
simplest problem has a corresponding quantificational interpretation of John cannot
solve any problem. Their meaning is located in the speaker’s epistemic world rather
than in the real world. The simplest/the most difficult problem does not refer to or
entail the existence of a problem with the property of being simple/difficult, but to a
pragmatic scale whose extremes are denoted by superlatives (see Veloudis 1998 for
an alternative account).
8
Quirk et al. (1972: 287) argue that there is a difference between American and
British English in such examples so that a sentence such as She is most beautiful is
‘not the superlative in Br[itish]E[nglish], though it can be in Am[erican]E[nglish].’ In
Br[itish]E[nglish] the sentence can only mean she is extremely beautiful and not that
she is more beautiful than all others. This absolute sense is common in
Am[erican]E[nglish] too. In British English most is a superlative only when preceded
by the definite article: She is the most beautiful. The syntactic position of the
adjective is important, but we will consider here only adjectives occurring in

attributive and not predicative position. The former favor inflectional and the latter,
periphrastic comparison for both comparatives and superlatives in 19th century as well
as in contemporary English (Kytö and Romaine 2006: 199-200).
9
Claridge (2006: 85), who also included defective adjectives in her analysis of
superlatives in CONCE, found that the number of occurrences of superlatives in letters
(36.4) clearly exceeded the corpus average (24.5). Claridge (2006: 75) also found over
300 occurrences of best.
10
The form most deliciousest could be simply a childish variant. Hester was Anne
Thackeray Ritchie’s only daughter and was 8 years old in 1886 at the time this letter was
written. The other form, however, is clearly Ritchie’s own. It occurs in a letter to her son,
William Thackeray Denis, born in 1880, who was 18 when this letter was written in
1898. She uses formal address forms (Mrs. Ritchie to refer to herself, and Monsieur
Denis to refer to her son), but these are embedded in the context of the formulaic
expression ‘presents her compliments’. The fact that this apparent formal tone is
obviously and deliberately undermined in her next sentence, where she offers her son
hugs and kisses, suggests a mock formality.
11
The third is a double comparative, more truerer, in Thomas Morton’s play, The School
of Reform (1805).
12
The ratio is calculated by dividing the number of occurrences of dear (male = 531,
female = 419) by the number of occurrences of dearest (male = 89, female = 229).
Ratios greater than 1 indicate that dear is more frequent than dearest.
References
Bergs, Alexander T. 2004. ‘Letters. A new approach to text typology’ in Journal of
Historical Pragmatics 5(2): 207-227.
Biber, Douglas. 1988. Variation across speech and writing. Cambridge: Cambridge
University Press.
Biber, Douglas, and Edward Finegan. 1989a. ‘Styles of stance in English: Lexical and
grammatical marking of evidentiality and affect’ in Text 9(1): 93–124.
Biber, Douglas, and Edward Finegan. 1989b. ‘Drift and the evolution of English
style. A history of three genres’ in Language 65(3): 487-517.
Bijkerk, Annemieke. 2004. ‘Yours sincerely and yours affectionately. On the origin
and development of two positive politeness markers’ in Journal of Historical
Pragmatics 5(2): 297-311.
Bolinger, Dwight. 1977. ‘Neutrality, norm and bias.’ Bloomington: Indiana
University Linguistics Club.
Brown, Roger W. and Albert Gilman. 1960. ‘The pronouns of power and solidarity’
in Sebeok, Thomas A. (ed.) Style in language. New York: Wiley: 253-176.
Busbridge, E.M. 1909. Letter-writing and etiquette. London: Collins.
Claridge, Claudia. 2006. ‘“With the most superlative felicitee”: Functions of the
superlative in 19th-century English’ in Dalton-Puffer, Christiane, Dieter
Kastovsky, Nikolaus Ritt and Herbert Schendl (eds) Syntax, style and
grammatical norms. Bern: Peter Lang AG: 73-89.
Claridge, Claudia. 2007. ‘The superlative in spoken English’ in Facchinetti, Roberta

ed. Corpus linguistics 25 years on. Amsterdam: Rodopi: 123-150.
Coates, Jennifer. 1996. Women talk. Conversation between women friends. Oxford:
Blackwell.
CONCE = A Corpus of Nineteenth-Century English, being compiled by Merja Kytö
(Uppsala University) and Juhani Rudanko (University of Tampere).
Dahl, Östen. 2001. ‘Grammaticalization and the life cycles of constructions’ in RASK
- Internationalt tidsskrift for sprog og kommunikation 14: 91-134.
Davis, Norman. 1965. The Litera Troili and English letters. R.E.S. New Series. Vol.
XVI, No. 63: 233-244.
Farkas, Donna F., and K.E. Kiss. 2000. ‘On the comparative and absolute readings of
superlatives’ Natural language and linguistic theory 18: 417-455.
Fauconnier, Gilles. 1980. ‘Pragmatic entailment and questions’ in Searle, John R.,
Ferenc Kiefer and Manfred Bierwisch (eds) Speech act theory and pragmatics.
Dordrecht: Reidel: 57-69.
Fitzmaurice, Susan. 2002. The familiar letter in early modern English: A pragmatic
approach. Amsterdam: John Benjamins.
Gabelentz, Georg van der. 1891. Die Sprachwissenschaft. Ihre Aufgaben, Methoden,
und bisherige Ergebnisse. Leipzig: Weigel.
Geisler, Christer. 2002. ‘Investigating register variation in nineteenth-century
English: A multi-dimensional comparison’ in Reppen, Randi, Susan M.
Fitzmaurice and Douglas Biber (eds) Using corpora to explore linguistic
variation. Amsterdam/Philadelphia: John Benjamins: 249–271.
Geisler, Christer. 2003. ‘Gender-based variation in nineteenth-century English letter
writing’ in Leistyna, Pepi and Charles F. Meyer (eds) Corpus analysis: Language
structure and language use. Amsterdam: Rodopi: 87-106.
Görlach, Manfred. 1999. English in nineteenth-century England: An introduction.
Hall, Kira. 1995. ‘Lip service on the fantasy lines’ in Hall, Kira and Mary Bucholtz
(eds) Gender articulated: Language and the socially constructed self. New York:
Routledge: 183-216.
Helsinki Corpus = The Helsinki Corpus of English Texts. 1991. Helsinki: Department
of English, University of Helsinki.
Huddleston, Rodney and Geoffrey K. Pullum 2002. The Cambridge grammar of the
English language. Cambridge: Cambridge University Press.
Jespersen, Otto. 1948. Growth and structure of the English language. New York:
Doubleday.
Jespersen, Otto. 1949. A modern English grammar on historical principles. Part VII:
Syntax. Copenhagen: Ejnar Munksgaard and London: George Allen & Unwin.
Kytö, Merja. 1996. ‘“The best and most excellentest way”: The rivalling forms of
adjective comparison in Late Middle and Early Modern English’ in Svartvik, Jan
(ed.) Words. Proceedings of an international symposium, Lund, 25–26 August
1995. (Konferenser 36.). Stockholm: Kungl. Vitterhets Historie och Antikvitets
Akademien: 123–144.
Kytö, Merja, and Suzanne Romaine. 1997. ‘Competing forms of adjective comparison
in modern English: “What could be more quicker and easier and more effective?”’
in Nevalainen, Terttu, and Leena Kahlas-Tarkka (eds) To explain the present:
Studies in the changing English language in honour of Matti Rissanen. Helsinki:
Société Néophilologique: 329–352.
Kytö, Merja, and Suzanne Romaine. 2000. ‘Adjective comparison and standardisation
processes in American and British English from 1620 to the present’ in Wright,
Laura (ed.) The development of standard English, 1300–1800: Theories,
descriptions, conflicts. Cambridge: Cambridge University Press: 171–194.
Kytö, Merja, and Suzanne Romaine. 2006. ‘Adjective comparison in 19th-century
English’ in Kytö, Merja, Mats Rydén and Erik Smitterberg (eds) Nineteenth-
century English: Stability and change. Cambridge: Cambridge University Press:
194-214.
Kytö, Merja, Juhani Rudanko and Erik Smitterberg. 2000. ’Building a bridge between
the present and the past: A corpus of 19th-century English’ in ICAME Journal 24:
85–97.
Lakoff, Robin. 1975. Language and woman’s place. New York: Harper & Row [out
of print].
Leech, Geoffrey, and Jonathan Culpeper. 1997. ‘The comparison of adjectives in
recent British English’ in Nevalainen, Terttu and Leena Kahlas-Tarkka (eds) To
explain the present: Studies in the changing English language in honour of Matti
Rissanen. Helsinki: Société Néophilologique: 353–373.
Mühlhäusler, Peter. 1983. ‘Stinkiepoos, cuddles and related matters’ in Australian
Journal of Linguistics 3(1): 75-91.
Nevala, Minna. 2004. ‘Inside and out. Forms of address in seventeenth and eighteenth
century letters’ in Journal of Historical Pragmatics 5(2): 271-296.
Nevalainen, Terttu. 2004. ‘Letter writing’ in Journal of Historical Pragmatics 5(2):
181-191.
Nevalainen, Terttu, and Helena Raumolin-Brunberg (eds). 1996. Sociolinguistics and
language history. Studies based on the Corpus of Early English Correspondence.
Amsterdam: Rodopi.
Nevalainen, Terttu, and Helena Raumolin-Brunberg. 2003. Historical
sociolinguistics. Language change in Tudor and Stuart England. London:
Longman.
OED = Oxford English Dictionary. On-line edition. URL:http://dictionary.oed.com/,
accessed November 2006.
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik. 1972. A
grammar of contemporary English. London and New York: Longman.
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik. 1985. A
comprehensive grammar of the English language. London and New York:
Longman.
Raumolin-Brunberg, Helena. 1996. ‘Forms of address in early English
correspondence’ in Nevalainen, Terttu and Helena Raumolin-Brunberg (eds)
Sociolinguistics and language history. Studies based on the Corpus of Early
English Correspondence. Amsterdam: Rodopi: 167-183.
Rusiecki, Jan. 1985. Adjectives and comparison in English: A semantic study.
London: Longman.
Safire, William. 2006. ‘A change of address: the demise of “dear”’ in International
Herald Tribune, October 23, p. 7.
Sharvit, Yael, and Penka Stateva. 2002. ‘Superlative expressions, context and focus.’
in Linguistics and Philosophy 25: 433-504.
Taavitsainen, Irma, and Andreas H. Jucker (eds). 2003. Diachronic perspectives in
address term systems. Amsterdam: John Benjamins.
Tieken-Boon van Ostade, Ingrid. 1999. ‘Of formulas and friends: Expressions of
politeness in John Gay’s letters’ in Tops, Guy A.J., Betty Devriendt and Steven
Geukens (eds) Thinking English grammar. To honour Xavier Dekeyser, Professor
Emeritus. Leuven: Peeters: 99-112.
Veloudis, Ioannis. 1998. ‘“Quantifying superlatives” and homo sapiens’ in Journal of
Semantics 15(3): 215-237.
PART IV
PRAGMATICS AND DISCOURSE STYLISTICS
Functional Stylistics and Peripeteic Texts
Simon Borchmann
Scandinavian Institute, University of Aarhus
Abstract
Using a pragmatically based linguistic description apparatus to study literary use of
language is not unproblematic. Observations show that literary use of language
violates the norms contained by this apparatus. With this paper I suggest how we can
deal with this problem by setting up a frame for the use of a functional linguistic
description apparatus on literary texts. As an extension of this suggestion I present a
model for the description of a specific type of literary texts.
Keywords: literary pragmatics; deviation; text understanding.
1. Functional stylistics
1.1. The model
In the following, I will outline a stylistics that forms a part of a text
model. The text model is set up to serve didactic purposes in the field
of language and literature. The demand on the model is that it can
describe, explain and deal with a specific type of texts. By description
I take to mean a systematic account of the linguistic and textual
prerequisites for text comprehension; by explanation an account of
why a text has the properties it has; by dealing with a methodical
utilization of a text’s potential for understanding.
The descriptive part of the model corresponds to the stylistics, and it
consists of applying a functional linguistic and textual description
apparatus to a literary text. This application serves to refine the user’s
awareness of the relation between text comprehension processes, on
266 Simon Borchmann
the one hand, and the linguistic choices and textual organization
forms, on the other.
1.2. The functional approach

The functional approach employed here belongs to normative
pragmatics, and it is characterized by the following three assumptions.
1) The primary purpose of language is communication. A natural
language is formed in order to serve this purpose. When we say
function, we mean a structure that is determined by the purpose
it is serving (Harder 1996). We distinguish between the
following four subordinate communicative functions:
representational (e.g. semantic roles, time, space), informational
(e.g. presupposition, message structure, relevance structure),
sender relational (e.g. subjective relation, perspective), and
interactional (e.g. illocutionary force indication) (Hansen &
Heltoft 1999). These functions are assumed to be coded in a
natural language – some more than others.
2) Knowing a language is only part of the knowledge required to
take part in everyday language use. Using a language for
communicative purposes requires what we call ‘communicative
competence’ (Habermas 2001 (1971), de Beaugrande 1980, Dik
1989). The communicative competence can be reconstructed as
a set of norms on different levels of communication, and this
reconstruction must be based on a generalization from everyday
language use.
3) A sequence of sentences is assumed to be an intentional act
that serves a specific purpose (Harder 1979). Thus, it is
assumed that functions can be attributed to each and every
linguistic choice and textual organization form in terms of an
overall function (Togeby 1993).
In accordance with this, the description apparatus comprises: 1) four
aspects corresponding to the four subordinate functions, namley the
aspect of representation, the aspect of informativity, the aspect of
sender relations, and the aspect of genre (Borchmann 2005), 2) a
system of norms, and 3) an assumption as to the overall function of
the text (a text type and a genre).
Functional Stylistics and Peripeteic Texts 267
1.3. The problem

The application of this description apparatus to a literary text is not
unproblematic. The stumbling block is this: we can observe
discordances between literary language use and the linguistic and
textual norms contained in the descriptive apparatus, and furthermore:
we can observe them consistently. Consequently, the following
objection is raised the minute we apply – or even suggest to apply –
the description apparatus on a literary text: how can you apply a
description apparatus to language use that deviates from norms
contained in this apparatus?
Now, this is a fundamental problem as to the co-ordination of
linguistics and science of literature. I have dealt with this problem
with the project Funktionel tekstteori og fiktivt fortællende tekster med
refleksiv funktion. What I will present in the following, are some of
the results of this project. The questions addressed here are:
1) How can we utilize a functional description apparatus within a
frame of understanding that embraces violations of norms?
2) What are the limits of such utilization?
2. Peripeteic texts
2.1. An example
The project took a starting point in an observation of a phenomenon of
text understanding. I will begin by illustrating this phenomenon with a
piece of text.
Jeg så engang i et selskap en ung kvinde forelsket. Hendes øine
var da dobbelt blå og dobbelt strålende, og hun kunde slet ikke
skjule sine følelser. Hvem elsket hun? Den unge herre borte ved
vinduet, husets søn, en mand med uniform og løverøst. Og Gud,
hvor hendes øine elsket den unge mand, og hvor hun sat urolig
på stolen!
Da vi gik hjem om natten, sa jeg, fordi jeg kjendte hende så
godt:
’Hvor veiret er lyst og herligt! Har du moret dig i nat?’
Og for å imøtekomme hendes ønske, trak jeg min
forlovelsesring av fingeren og sa videre: ’Se, din ring den er blit
mig for trang ... (Knut Hamsun. Ringen).
268 Simon Borchmann
I once at a party saw a young woman in love. Her eyes were

then doubly blue and doubly radiant, and she could not hide her
feelings at all. Who did she love? The young gentleman by the
window, the son of the house, a man with uniform and a lion’s
voice. And by God, how her eyes loved the young man, and
how restlessly she sat on the chair!
When we went home at night I said, because I knew her so
well:
‘How bright and splendid the weather is! Have you had fun
tonight?’
And to comply with her wish, I took off my engagement ring
and continued: “Look, your ring, it has become too tight to me
...
When readers read this piece of text from left to right in standard
reading pace, it occasions an understanding in which the narrator
contemplates a woman who is not related to him in any particular
way. This understanding, however, must be abandoned as the reading
progresses. Exactly when this abandoning occurs depends, to some
extent, on the individual reader. But in any case, it must be abandoned
when the subordinate clause, ‘fordi jeg kendte hende så godt’
(‘because I knew her so well’), is read. Here it is presupposed that the
narrator knew her, and with the following sentence he even
presupposes that the woman is his fiancée. At this point there is a
basis for a reconfiguration of the text meaning, that is, in such a way
that the narrator in the party scenario is the witness of the deceit of his
fiancée.
One could note that there are contra indications to the first
understanding in the second sentence, namely the adverb ‘dobbelt’
(‘doubly’), which presupposes a standard of reference, and the adverb
‘da’ (‘then’) which implicates a present state of the eyes. However,
psycholinguistic experiments have shown that readers tend to ignore
problems when they read at standard pace (Sanford 1990).
Anyway, exactly when the reader acknowledges that she has been put
on the wrong track, is not decisive for the illustration of the
phenomenon of text understanding. What is decisive is that the text
occasions a process of understanding that embraces two discordant
configurations of the text meaning, with a textual discordance as the
point of reversal. I call this phenomenon of understanding the

peripeteic process of understanding.
2.2. P-texts and their basic functions

The phenomenon has been illustrated with a short piece of text.
However, for a number of texts it is the case that they occasion a
peripeteic process of understanding. That is, the overall frame of
understanding corresponds to a peripeteic process of understanding
such that the point of reversal will occur when the reader reads the
very last part – sometimes even the last word – of the text. This goes
for a lot of jokes, but also for a great many detective novels and for a
set of short stories (James Joyce’s Araby is a particularly good
example). I call such texts peripeteic texts or P-texts for short. There
are five criteria for P-texts (Borchmann 2005), but as we are dealing
with a general problem they should not be mentioned here. The
relevant observations in this connection are the following:
1) The process of understanding includes a reversal.
2) This reversal is occasioned by discordance.
3) The reconfiguration of the text meaning is more exhaustive
and therefore necessary for the realization of the meaning
potential of the text.
4) Thus, the discordance can be attributed to a function in an
overall frame of understanding, that is, as a catalyst for a
reconfiguration of the text meaning.
On this basis we can assume two basic P-text functions: inhibitors and
catalysts. Inhibitors are linguistic choices that are particularly decisive
for the first interpretation that results in inhibition. Catalysts have a
double function; they are discordances in the first interpretation, and
cues to the second interpretation.
The project concentrated on peripeteic short stories by Danish authors
(among others Svend Aage Madsen, Jan Sonnergaard and Knud
Sørensen). The reason for using peripeteic short stories in the attempt
to deal with the problems described above is, of course, that peripeteic
texts deviate from norms of language use by definition.
270 Simon Borchmann
3. Linguistic norms and literary language use

3.1. Traditional accounts
The fact that literary use of language violates norms has received a lot
of attention in scientific approaches to the description of literature.
Indeed, literature has been defined as a deviation from the norm.
There are two widespread deviation assumptions: the structural,
linguistically based deviation assumption (Mukarovsky 1964 (1939),
1971 (1940)) and the pragmatic deviation assumption (Iser 1980).
According to the structural, linguistically based deviation assumption
literary use of language deviates from the rules of language and thus
constitutes a separate language. According to the pragmatic deviation
assumption literary language use deviates from norms of language use
and thus constitutes a separate language use.
Regrettably, the explanatory value of these assumptions is severely
restricted. As to the structural, linguistically based assumption,
observations show that a great part of what we call literary language
use perfectly complies with the constitutive rules of language. Thus,
‘violence against language’ (Mukarovsky 1971 (1940)) is not a
common feature of the literary use of language. Nor, in fact, are the
norm violations mentioned in such account violations of the
constitutive rules of a language. Consequently, it appears that this
assumption is based on an unclear distinction between different types
of norms and different types of norm violations.
As to the pragmatic deviation assumption, the flaws are very clear in
Wolgang Iser’s account. Iser’s deviation assumption is based on a
generalisation of a very specific norm: Austin’s criteria for strongly
institutionalized, spoken types of language use like the act of baptism
and other exercitives that presuppose precisely defined situations.
Therefore, it is unclear how literary use of language distinguishes
itself from all other language uses except the little corner of language
use that satisfies these criteria. Furthermore, Iser rejects the imitation
assumption (Searle 1996 (1979)) of the speech act theory by asserting
that an imitation of a text should have the same consequences as the
text it is imitating. But the literary texts have exactly the consequences
they are expected to have, that is, they are sold, reviewed, read, cause
artistic experiences etc. Clearly, Iser does not distinguish between an
imitation act and the result of the imitation act. In other words, the
pragmatic deviation assumption is based on an insufficiently analyzed
concept of text as an act. The project has dealt with the flaws of these
accounts.
3.2. The functional account of norms

As to the norms of language use, the functional approach distinguishes
between four types of norms. 1) The normative validity claims are
constitutive for language use that aims at mutual understanding
(Habermas 2001 (1971) & 1981). 2) The rules of language are
constitutive for a particular natural language. 3) The general
principles are regulatory principles for the exchange of information
within the framework of an accepted purpose (Grice 2001 (1975),
Levinson 2000). 4) The genre specific principles are regulatory
principles for choosing linguistic constructions and for organizing
them sequentially such that coherent configurations can be configured
under each of the four aspects of the text (Borchmann 2005). This
account of the system of norms is a theoretical reconstruction of
communicative competence.
Now, if we employ this system of norms to the Hamsun excerpt, it is
clear that there are no violations of the rules of language. The
language realized by the Hamsun excerpt complies with Norwegian
bokmål. Thus, there is no ‘violence’ against language, as Mukarovsky
assumes, and thus there is no renewal; it is simply reproduction and
use of Norwegian, bokmål.
Nevertheless, the linguistic choices and the textual organization forms
of the text do give rise to inhibition in the process of understanding,
and therefore, presumably, norms are violated. The norms violated,
however, concern our understanding of the language use as an
intentional act, and the violation must be described as such (ff. 4.1.).
3.3. The functional text definition

As to the concept of text, the functional approach employs the
following definition (Borchmann 2005): a text is a sequence of
sentences and must form the basis for a communicative understanding
process that comprises an intentional whole. If we apply this
definition to a literary text the flaw of the pragmatic deviation
assumption becomes clear. This can be illustrated with the following
example:
272 Simon Borchmann
Hun slyngede sin cognac i ansigtet på ham og rejste sig rasende.

Hun løb ud af baren, og de andre kunder så forbavset på os, og
den lille mand begyndte at græde. Vi vakte opmærksomhed, og
det var meget ubehageligt, og jeg skyndte mig ud til Ulla, for
det var naturligvis hende jeg holdt med, (Jan Sonnergaard.
William).
She threw her cognac in his face and got up furiously. She ran
out of the bar, and the other customers looked at us, and the
little man <whom she threw the cognac at> began to cry.
We were attracting attention, and it was very unpleasant, and I
hurried out to Ulla, for of course it was her that I was siding
with.
The narrator is communicating that he is siding with Ulla. However, if
we consider more thoroughly what the narrator communicates, there
seems to be discordance. On the one hand, the sentence adverbial
‘naturligvis’(of course) indicates that he has a solid knowledge as to
whom he is siding with (Andersen 1986). On the other hand, he
presupposes, with the contrastive focus on ‘hende’(her), that there is a
relevant alternative to the focus value, the alternative being the little
man (Grønnum 2004, von Heusinger 1999). This alternative
interpretation, as to whom the narrator is siding with, can only be
occasioned by what the narrator himself has communicated. Thus, the
linguistic choices of the narrator give rise to inconsistency. This
inconsistency could occasion a suspicious interpretation of the
narrator, and in fact it should, if we want to understand what is going
on at all. A preliminary result of this interpretation would be that the
narrator, at least, is uncertain as to whether he is siding with Ulla or
the little man.
Now, if we consider this suspicious interpretation as necessary for the
understanding of the short story, and thus for the realization of the
overall function of the text, we can consider it as intended. This
intention, however, is different from the narrator’s intention. Thus,
there are two intentions at work in the understanding: the narrator’s
intention to communicate about his relation to his girlfriend, and an
intention to cause a suspicious interpretation of the narrator.
According to the functional text definition, if there are two intentions,
there are two texts. In line with this, the functional approach
distinguishes between two texts in the literary use of language: the

literary text and the narrator’s text. The following relation is assumed:
the literary text forms the basis for an imitation of a text, that is, the
narrator’s text.
On the basis of the application of this concept of text the following
framework is set up:
1) The literary text forms the basis for an imitation.
2) If the literary text is understood as an imitation act, then it
can only deviate from the normativity it establishes by being
understood as an imitation act. And literary texts conform to
the normativity of the imitation act.
3) What deviates normatively is the text imitation for which the
literary text, as an imitation act, forms the basis.
4) The norm violations of the text imitation can be attributed to
a function in the frame of understanding of the literary text.
4. The potential and limitations in the utilization of the functional

linguistic descriptive apparatus
4.1. The potential
On the basis of the framework sketched out above, we can pinpoint 1)
how a functional linguistic and textual description apparatus can be
used in the description of literary use of language, and 2) what are the
limits of this usage.
Firstly, we can explain why the text imitation gives rise to inhibition,
and we can locate the linguistic choices and textual organization forms
that are decisive for the interpretation which leads to inhibition. This
can be illustrated by the Hamsun excerpt.
According to the validity claims the narrator claims, and the reader
assumes, that behind each and every linguistic choice and textual
organization form there is a communicative intention.
According to the rules of Norwegian bokmål the indefinite article ‘en’
marks that the element can not be identified by the receiver, but must
be introduced.
According to the general principles the sender must – at every stage of
the exchange – make his contribution as informative as required for
the purpose, but not more informative than required.
274 Simon Borchmann
According to the specific principles for narrative texts the sender must
make linguistic choices and organize the sentences in such a way that
the receiver can configure a coherent succession of events and actions.
One of the fundamental activities in the configuration of a coherent
succession of actions and events is to assign roles to the characters by
mapping them into scenarios (Garrod 1995). This role assignment is
crucial for deciding which inferences are relevant in order to achieve
coherence.
Now, the application of this system of norms can explain why the text
occasions inhibition in the process of understanding: if the woman
could be assigned the role as the narrator’s fiancée, then the narrator
should have chosen a more informative designation. Therefore, when
the narrator chooses the indefinite article and the abstract notion
‘kvinde’ (woman), the reader assumes that the woman is not related in
any relevant way to the narrator in the actual scenario. Otherwise, the
narrator would have chosen another designation. In other words, the
indefinite form is an inhibitor (2.2., ff. 5.1.).
Secondly, we can describe the violations of the text imitation with
reference to norms of language use, and we can localise the violations
to linguistic choices and textual organization form. This usage can be
illustrated by the Sonnergaard excerpt. Here we can describe the
violation with reference to the principle of accommodation for the
marking of sender relations (Togeby 1993, Borchmann 2005), and we
can localise the violation to discordances between a contrastive focus
and a sentence adverbial used to communicate the sender’s relation to
his feelings for his girlfriend Ulla. In other words, the combination of
the sentence adverbial and the contrastive focus is a catalyst (2.2., ff.
5.2.).
It is very important to notice that this description and localization of
violation is not a meaningless linguistic drill; it contributes to the
interpretation. Thus, on the basis of these descriptions and
localizations we can set up a heuristics for the second, more
exhaustive interpretation. For example: look for sentence adverbials or
other linguistic choices that indicate subjective relation when the
narrator communicates about the behaviour of Ulla.
If we apply this heuristics to the short story, we will indeed get results
relevant to the interpretation, e.g.:
Da jeg lod som om hun havde såret mig – men såret var jeg
naturligvis ikke, men der var alligevel noget ubehageligt ved
tanken om det der med, at det kunne være ’hvem som helst’, og
på en måde var det ikke så morsomt igen – greb hun fat i min
arm og kyssede mig ... (Jan Sonnergaard. William)
When I pretended that she had hurt me – but, of course, I was
not hurt, but there was something uncomfortable about the
thought of this whole ‘it could be anybody’, and in a way it was
not so funny after all – she caught my arm and kissed me…
og jeg kunne se hvor glad hun var over, at jeg håbede at hun
langt om længe var blevet gravid – selvom hun naturligvis
spillede skuespil og lod som om det irriterede hende ... (ibid.)
and I could see how happy she was that I was hoping that she at
long last had become pregnant – even though, of course, she
was play-acting and pretended that it annoyed her…
4.2. The limits

The limits of the utilization of the description apparatus is also very
clear by virtue of the functional definitions. We can use the
description apparatus to locate the discordances that forms the basis of
the reconfiguration of the text meaning, but functions cannot be
attributed to the violations, neither can we interpret them with
reference to the functional description apparatus.
The reason for this is obvious: the limits of the description apparatus
are the communicative competence, and the interpretation of
violations of norms exceeds the communicative competence.
Therefore, other frameworks of understanding and another type of
understanding strategy are necessary. These frameworks and strategies
have been described by literary theorists, among others.
As to the co-ordination of linguistics and literary theory, the point is
this: if we want to co-ordinate linguistics/pragmatics and literary
theory, it is very important to acknowledge the limits of the
linguistically based description apparatus.
5. The model
The P-text model is to be seen as a stylistic concretization of
Ricoeur’s three fold mimesis: prefiguration, configuration, and
276 Simon Borchmann
refiguration (Ricoeur 1984 (1983)). Thus, reflexive function is

attributed to periepteic short stories.
In the application of the P-text model the analyst assumes, by
abduction, that the fictitious narrative text that is the object of analysis
is a P-text. By deduction the analyst must thus partly be able to
achieve an exhaustive understanding of the succession of actions and
events (and the persons), partly be able to account for a textual basis
for the peripeteic process of understanding. And further, the analyst
must thus, by induction, partly describe and localize a system of
inhibitors and catalysts and a peripeteic point, partly account for being
able to achieve an exhaustive understanding of the succession of
actions and events (and the persons) with a basis in the system of
catalysts.
5.1. The inhibition

The analysis of inhibition consists of 1) accounting for the inhibiting
text meaning and 2) tracing it back to the text as a sequence of
linguistic constructions. The procedure for the analysis of the
inhibition is as follows:
1. A Paraphrase Of The First Configuration.
Under this point the analyst is instructed to paraphrase the
configuration that had to be abandoned.
2. An Account Of The Textual Basis For The First
Configuration.
2.1. An analysis of the textual basis for establishing the
situational model. Under this point the analyst is instructed to
undertake an analysis of the informative choices and the
informative organization corresponding to the analysis that is
presented under the aspect of informativity <an information
structure analysis>.
2.2. A specification of the inhibitors. Under this point the
analyst is instructed to specify the markers and indicators that
were particularly decisive for the configuring that led to
inhibition.
2.3. A specification of the peripeteic point. Under this point the
analyst is instructed to localize the point at which the first
configuring had to be abandoned.
2.4. An analysis of the sequential organization of the textual

basis for the first configuring. Under this point the analyst is
instructed to undertake an analysis of the composition of the
inhibitory text sequence corresponding to the analysis that is
presented under the aspect of genre/interaction <a composition
analysis>.
5.2. The catalysis

When the functional reader cannot configure a coherent succession of
action and events with confidence in the narrator as a communicator,
he alters – and must alter – his strategy of understanding. This is
reflected in the P-text model by the catalysis’ dissolution of the
inhibition with a view to arriving at a coherent succession of actions
and events.
The catalysis comprises two phases that correspond to dissolution and
reconstruction in the hermeneutics of suspicion, namely 1) the
dissolution in which the reader retrospectively attempts to find
everything that is discordant in the first configuring and 2) the
reconfiguring in which the reader attempts to reconfigure the
succession of actions and events with a basis in these violations.
The central poetic textual function in catalysis – the one the reader
looks for in the dissolution, and the one he takes as his point of
departure in the reconfiguration – is the catalyst. The criteria for a
catalyst are 1) that it can be localized to the text as a sequence of
linguistic constructions and can be described under one of the four
aspects as a violation of the specific text principles and 2) that it can
form the basis for a reconfiguring of the succession of actions and
events and of the persons.
I distinguish between catalysts according to their argumentative status.
Catalysts that are related by means of similarity relations, localized to
the same subject matter, for example, always mutually strengthen each
other according to the hermeneutic principle. Some catalysts,
however, provide strong arguments in themselves for a reconfiguring,
whereas others do so only when they are related to other catalysts. The
former can be called primary catalysts, the latter secondary catalysts.
When I distinguish between primary and secondary catalysts, it is to
indicate the catalysts that justify the reconfiguration in practice and
thus anchor the catalysis argumentatively. This is not a distinction
278 Simon Borchmann
having to do with their contribution to the reconfiguring of the

succession of actions and events and of the persons. Furthermore, I
distinguish between catalysts and reflectors which is a distinction with
regard to intersubjectivity: Catalysts can be recognized by an
interpretive community as violations of intersubjective norms of
language use; reflectors cannot be recognized as violations with
intersubjective norms of language use, but are subjective projections.
The two phases in catalysis, dissolution and reconfiguring, each
comprise four levels:
1. The Dissolution
1.1. An indication of the first primary catalyst. Under this point
the analyst is instructed to identify the peripeteic catalyst that
justifies the catalysis.
1.2. The intuitive catalysis. Under this point the analyst is
instructed to take a retrospective look at his understanding in
order to recall problems in understanding that were kept in
mind during the progressive reading process with a view to
subsequent resolution, but which were not resolved. These
problems are candidates for the category of primary catalysts.
1.3. The primary catalysts. Under this point the analyst is
instructed, with the help of the functional text model, to
describe the problems of understanding that were found in the
intuitive catalysis as violations of the specific text principles at
a localized point in the text.
1.4. The analytical catalysis – secondary catalysts. Under this
point the analyst is instructed to examine the text analytically,
selectively on the basis of the catalysts that have been found
with a view to finding catalysts that are associated with these by
means of similarity relations.
2. The Reconfiguring
2.1. An interpretation of groupings of catalysts. Under this
point the analyst is instructed to interpret groupings of catalysts
as partial meanings, i.e. as thematically determined meaningful
connections.
2.2. An investigation of the text on the basis of relevance
criteria that have been established through catalysis. Under this
point the analyst is instructed to examine the text again on the
basis of the newly established relevance criteria.
2.3. Abductive reconfiguring. Under this point the analyst is

instructed to form a hypothesis which can cover all of the
partial meanings and which can thus explain all of the
violations of the specific text principles.
2.4. Verifying the hypothesis. Under this point the analyst is
instructed to verify the hypothesis from both a text internal and
text external point of view. The text internal verification
requires that the hypothesis be historically, societally, socially
and/or psychologically plausible. The text external verification
can require a systematic look at historical, societal, social and
psychological knowledge. This type of systematics is not the
concern of text theory.
6. Conclusion
In 1973, Fish wrote: ‘Linguists resolutely maintain that literature is,
after all, language, and that therefore linguistic description of a text is
necessarily relevant to the critical act; critics just as resolutely maintain
that linguistic analyses leave out something, and that what they leave
out is precisely what constitutes literature’ (Fish 1996 (1973), p. 97).
Today, more than 30 years later, we still have to accept this as a
reasonable description of the state of affairs in the field of language and
literature. With this paper I have suggested how we can bring stylistics
beyond this intolerable situation: when employing stylistics we have to
define the limits of the use of the linguistic description apparatus within
a specific literary frame of understanding.
In so far as the literary use of language always imitates non-literary use
of language to some extent, and readers accordingly rely on their
communicative competence in the proces of understanding, a
pragmatically based linguistic description apparatus must have a
potential with regard to the process of interpretation. This potential,
however, is limited to the imitative part, that is, the part where the
norms of language use contained in the description apparatus are
observed. Thus, in so far as the literary use of language by definition
violates the norms contained in the description apparatus, the
linguistic description is not sufficient for an exhaustive interpretation.
Now, literary critics might claim that the relevant literary
interpretation process begins exactly where the linguistic description
ends, that is, where the norms are violated. This, however, does not
280 Simon Borchmann
mean that the linguistic description is irrelevant. As I have shown, the

description and localization of violations can contribute to a second,
more exhaustive interpretation within the frame of a peripeteic process
of understanding. Though not sufficient, the linguistic description is
necessary for an exhaustive interpretation within a frame of
understanding that embraces violations of norms contained by the
linguistic description apparatus.
References
Andersen, John Edelsgaard. 1986. ‘Adverbier – sprogvidenskabens stedbørn’ in
Beiträge zur nordischen Philologie 15.
De Beaugrande, Robert. 1980. Text, Discourse and Process. Newark, NJ: ABLEX
Publishing Corporation.
Booth, Wayne C. [1961] 1983. The Rhetoric of Fiction. London: Penguin Books.
Borchmann, Simon Uffe. 2005. Funktionel tekstteori og fiktivt fortællende tekster med
refleksiv funktion. Frederiksberg: Roskilde Universitetsforlag.
Dik, Simon. 1989. The Theory of Functional Grammar. Part 1: The Structure of the
Clause. Dordrecht: Foris Publications.
Fish, Stanley. [1973] 1995. ‘How Ordinary Is Ordinary Language’ in Is There a Text
in This Class? Cambridge: Harvard University Press.
Garrod, Simon. 1995. ‘Distinguishing between Explicit and Implicit Focus during Text
Comprehension’ in Richeit, G., and C. Habel (ed.) Focus and Coherence in
Discourse Processing. Berlin: Walter de Gruyter.
Grice, Paul. [1975] 2001. ‘Logic and Conversation’ in Henriksen, Carol (ed.) Can you
reach the salt? Frederiksberg: Roskilde Universitetsforlag.
Grønnum, Nina. [2001] (2004). Revideret kapitel 9 fra Fonetik og Fonologi – Almen og
Dansk. 2. edition. København: Akademisk Forlag.
Habermas, Jürgen. [1971] 2001. ’Forberedende bemærkninger til en teori om den
kommunikative kompetens’ in Henriksen, Carol (ed.) Can you reach the salt?
Frederiksberg: Roskilde Universitetsforlag.
Habermas, Jürgen. 1981. ’Hvad er universalpragmatik?’ in Teorier om samfund og
sprog. København: Gyldendal.
Hansen, Erik, and Lars Heltoft. 1999. Udkast til Grammatik over det Danske Sprog.
Roskilde: Afdelingen for Dansk og Public Relations Roskilde Universitetscenter.
Harder, Peter. 1979. ’Tekstpragmatik. En kritisk vurdering af nogle principielle og
praktiske tilgange til tekstbeskrivelsen, med ansatser til et alternativ’ in Andersen,
John E., and Lars Heltoft (ed.) Nydanske Studier & Almen kommunikationsteori. 10-
11. Sprogteori og Tekstanalyse. København: Akademisk Forlag.
Harder, Peter. 1996. Functional semantics. Berlin: Mouton de Gruyter.
Von Heusinger, Klaus. 1999. Intonation and Information Structure. The Representation
of Focus in Phonology and Semantics. Habilitationsschrift: Universität Konstanz.
Iser, Wolfgang. 1980. The Act of Reading. Baltimore: The Johns Hopkins University
Press
Levinson, Stephen C. 2000. Presumptive meanings. Cambridge: The MIT press.
Mukarovsky, Jan. [1939] 1964. ‘Standard Language and Poetic Languages’ in Garvin,
Paul L. (ed.) A Prague school reader on esthetics, literary structure and style.
Washington: Georgetown University Press.
Mukarovsky, Jan. [1940] 1971. ‘Om det digteriske sprog’ in Kock, Christian (ed.)
Tjekkisk strukturalisme. København: Munksgaard.
Ricoeur, Paul. [1983] 1984. Time and narrative. Vol 1. Chicago: The University of
Chicago Press.
Sanford, Anthony J. 1990. ‘On the nature of text-driven inference’ in Balota, D.A., G. B.
Flores d´Arcais and K. Rayner (ed.) Comprehension processes in reading. Hilsdale,
New Jersey: Lawrence Earlbaum.
Searle, John. [1979] 1996. Expression and meaning. Cambridge: Cambridge University
Press.
Togeby, Ole. 1993. PRAXT – pragmatisk tekstteori. Århus: Aarhus Universitetsforlag.
You Can’t Put Your Foot in the Same River Once:
Relevance Stylistics and Rereading
Anne Furlong
University of Prince Edward Island
The vibe … from people who reread a lot … [it] doesn’t seem
like they’re reading because they’ve forgotten the book. So
what’s the appeal? (kyuuketsukirui.livejournal.com)
Abstract
Rereading – the repetition of whole texts – represents the general phenomenon of
repetition at the largest possible scale. For this reason alone, it invites close attention.
The phenomenon of rereading poses interesting issues for stylistics: whether we can
distinguish in principle between productive and non-productive repetition, and
whether it is possible to explain what exactly is being experienced in productive
repetition, and why. This level of repetition is under-explored in pragmatic stylistics;
such treatments as do exist focus on cultural or literary aspects of rereading. The
complexity of this phenomenon can be fruitfully addressed through a stylistic
treatment: not just of individual cases, but of types of readers, texts, situations, and
goals. Relevance stylistics sheds light on the distinctions between productive and non-
productive repetition, and thus on the phenomenon of rereading. I argue that rereading
may not produce new effects: the reader's goal may be to experience certain effects
again, not to find new ones.
Keywords: stylistics; relevance theory; pragmatics; repetition; rereading; literary
interpretation.
1. Introduction
Verbatim repetition – i.e., the repetition of utterances or texts – strikes
most of us as an utter waste of time and effort. When a speaker repeats
herself, she risks putting her audience to gratuitous effort for no gain.1
When a hearer discovers that he is being asked to process an utterance
identical to one which he has heard and understood already, then his
attention may wander elsewhere. For example, most passengers on
airplanes rarely attend to the explanations of safety procedures given
by flight attendants; some airline companies introduce novel effects or
performances (incorporating jokes or encouraging interaction between
crew and passengers) to overcome the effects of verbatim repetition.2
284 Anne Furlong
Consequently, the question arises: when readers re-read a text,

voluntarily, independent of an assigned task, what are they doing? The
effort involved in processing verbal utterances is already significant;
but reading places demands on a range of resources that are greater
still. If rereading produces additional effects, then we should know
what they are. If there is no gain, then we should find a drop in
processing effort. And we need explanations as to why the efforts are
reduced, or the effects increased. If neither of these two conditions
obtains – and I will argue that this is the case, at least some times –
then we need an account of how repetition that does not produce new
effects, or come at less processing cost, may be not only tolerated but
sought out.
Relevance stylistics can shed light on what happens when people
reread, because rereading concerns cognition, about which relevance
theory has a great deal to say. In this paper I will outline some of the
issues raised by rereading, and propose a framework in which to make
sense of it. For the purposes of this paper, I have drawn on the
comments of ‘naïve readers’ – that is, readers who spontaneously
reread, and who discuss their habits, rationales, and states of mind
without prompting.3 To do so, I have sampled comments from
rereaders posted on a variety of weblogs and from one published
source (Lesser 2002).4 Their remarks suggest possibilities for future
research within the framework outlined in the paper.
Theories on rereading have tended to emerge from three main areas:
the psychology of reading (including developmental psychology: see
Zabrucky & Moore (1999); Stine-Morrow et al. (2004)); the writing
process (especially the teaching of writing: see Cornis-Pope &
Woodlief (2000); Sample (2005)); and aesthetics (philosophical
examinations of emotional responses to fiction: see Walton (1978);
Skulsky (1980); Neill (1991); Morreall (1993); McCormick (1985);
Wilkinson (2000).). Some address the rhetorical or stylistic uses of
repetition in texts (see Neal (1987); Bursey & Furlong (2006):
Johnstone (1994)). Many researchers recognise the oddity of
rereading, but few have pursued the fundamental issues raised by
voluntary rereaders. Brown (2002) suggests rereading is best studied
by ‘aesthetics’, arguing that the phenomenon falls within ‘literary
studies’, where it is commonplace.
You Can’t Put Your Foot in the Same River Once 285
I don’t think that’s enough. Rereading – the repetition of texts –

makes serious demands on cognitive resources, and cognition is not
really what literary studies is concerned with. It is not clear why
readers experience the same (or similar) emotions of suspense, fear,
excitement when rereading a story whose outcome is familiar,
unchanged, and unchangeable, a fact which is the topic of Gerrig’s
article on the relationship between fiction and non-fiction (1989: 277).
He asks, specifically, why ‘some works do not tolerate re-experiences’
(279). His argument is that rereading is dependent in part on the
degree to which a text satisfies the reader’s expectations; as a result,
some texts will tolerate rereading while others will not. I would argue,
on the contrary, that virtually any text is rereadable, and that
satisfaction is not a quality of the text but of the experience. The key
to the problem may lie, then, not in the ‘world of the text’ but in the
mind of the reader.
2. Kinds of rereading
There are many situations in which people reread texts; most of them
are not relevant here. For the purpose of this paper, I will focus on a
particular group of readers engaged in a particular kind of rereading.
First, I want to exclude specific rereaders. Professionals such as
lawyers and teachers reread texts in preparing their work. Students,
and the subjects of experiments, may be directed to reread texts (see
Millis & King (2001), Caws (1989)). Editors reread in order to
correct, amend, and reshape the text they are working on. In all these
cases, the rereading, though it may be voluntary, is in pursuit of a goal
separate from the reading experience. Such readers typically aim at an
interpretation (a set of assumptions) that includes implications and
implicatures that are relevant to them in pursuit of some other goal:
presenting an argument, teaching a class, passing a test.
Some of these situations have been investigated in the fields I have
mentioned. I want to consider the case of reading that is self-directed,
voluntary, and non-professional. This is the kind of rereading which
many people engage in and which is clearly related to aesthetics and
literary studies. Indeed, a text’s capacity to inspire or withstand
rereading is often given as necessary (but not sufficient) evidence of
its ‘literariness’ or value. It is hardly enough on its own; a great many
people have read The Da Vinci Code more than once. On the other
286 Anne Furlong
hand, literary value does not lead to rereading, nor does its absence
preclude textual repetition. As Susan Ives (mysanantonio.com) writes,
‘I read trash. I reread trash. Then I reread it again’.
We can look at the phenomenon of rereading in various ways. I
propose to discuss rereading by examining three elements:
Productive and non-productive repetition
Spontaneous and non-spontaneous interpretation
Cumulative and non-cumulative rereading(s)
These are not mutually exclusive categories. Rather, they are
properties of a reading; hence a reader may engage in reading which is
more productive than non-productive, more spontaneous than non-
spontaneous, more cumulative than non-cumulative. Moreover, these
properties interact with one another depending on the reader’s goals,
circumstances, and experience. To explore their connections, I will
take each in turn, and then show how they interact.
3. Productive and non-productive repetition

Verbatim repetition within a text is unproblematically productive.
Persson’s 1974 survey identified a wide range of effects of repetition
in texts. Kawin (1972) and Suleiman & Crosman (1980) have
extensively analysed the occurrence and function of repetition in
literary works. Verbatim repetition may be productive, according to
relevance theory, if it leads the reader to expand the context in ways
suggested by the text as a whole, and in which new effects are
experienced.5 However, the repetition of entire texts raises questions
which none of these studies addresses. Rereading puts the reader to
the effort of reprocessing a whole work, without the guarantee of
gaining any new insights or experiences, and with little reduction in
processing effort. I want to show that relevance theory can
accommodate this behaviour, account for it, and provide insight into
reader behaviour.
Relevance theory makes two fundamental claims about cognition and
communication. It claims that ‘human cognition tends to be organised
so as to maximise relevance’, and that ‘every act of ostensive
communication communicates a presumption of its own optimal
relevance’ (Sperber & Wilson 1995, 260). These claims mean that
there is a single criterion guiding the interpretation process, both
generating and evaluating interpretations. Furthermore, relevance

theory accounts for individual variations, and for general consensus
among readers. It is at once extensive and exceptionless.6
‘Relevance’ is a property (261), and is quantifiable: something is
relevant to the extent that the cognitive effects it produces are large,
and to the extent that the cognitive effort needed to achieve these
effects is small. The greater the cognitive effects that an utterance
produces, the more relevant it is; and the more effort it takes to
achieve these effects, the less relevant it is (122-132). Relevance thus
involves a dynamic balance between effort and effect; when there are
no longer adequate effects for the effort expended, we stop processing
the text – we stop paying attention.
Repetition presents something of a paradox for relevance theory, as
Jucker (1994) has noted. The processing of a repeated text would
seem to yield roughly the same set of assumptions for the effort
expended during the (second) reading. That expenditure of effort may
sound reasonable, but it means that the reader has put double the effort
(allocating the same resources twice) for exactly the same result (ie,
without any gain). This is clearly wasteful of cognitive resources;
according to the extent conditions of the theory, such an interpretation
must be literally irrelevant.
Against a reduction of positive cognitive effects, though, we might
count savings in cognitive resources. Rereading makes certain
assumptions highly salient, or ‘manifest’. A previously read text has
already been through the process of interpretation, so that cognitive
resources expended in the first place in hypothesis construction and
confirmation can be redirected. Rather than exactly doubling the
cognitive effort, the reader may be economising his effort, achieving
adequate relevance at a considerable reduction of processing. The
remarks of some rereaders indicate that they experience such a
reduction:
(1) On a second (or subsequent) reading, you’re not reading
for the story … [the storyline will] absorb too much of my
attention in the first reading. (zeitgeist)
(2) And then, of course, I had to remember the first reading
well enough to get something new out of the rereading.
(Lesser, 5)
288 Anne Furlong
These readers point to different aspects of this reduction in cognitive

effort: zeitgeist’s remark suggests that rereaders are aware of the
quantitative difference between first (and subsequent) readings;
Lesser’s comment shows that rereaders depend on assumptions from
the first interpretation as they process the text a second (or later) time.
These readers are not just repeating their initial experience; their
reduced effort achieves an interpretation that includes a large subset of
the assumptions constructed or entertained in the first reading process.
Lesser’s observation also demonstrates that readers know that, in
order to make this reinvestment worth their while, they will have to do
more than just process the text in the same way a second time.
Readers may cast about for more contextual premises that will yield
adequate contextual effects, paying more attention to details of the
text that they passed over the first time. Thus the ‘savings’ in one area
of processing may seem to be matched, or even outweighed, by
expenditure in another, resulting in effects not achieved in the first
reading.
Repetition will therefore be productive if the process yields new
assumptions, or strengthens or overturns existing assumptions. Such
rereading aims at contextual effects produced by an expanded context,
which includes not only the repeated text, but the original context, the
original interpretation, and the fact of its repetition. Even the prior
interpretation of the repeated text contributes actively to the
construction of the core set of assumptions in which the entire text is
interpreted, adding to the reader’s understanding of complex aspects
of the text, such as structure, imagery, allusion, and so forth. The
reader is aiming at a ‘literary’, or ‘non-spontaneous’ interpretation, a
concept I will cover briefly.
4. Spontaneous and non-spontaneous interpretation

Relevance theory was developed to account for spontaneous
interpretation – the kind of thing everyone does, all the time, whether
reading the paper or watching Lost or catching up with a friend over
the telephone. The contrasting notion of ‘non-spontaneous’
interpretation has been proposed in relevance theory.
Spontaneous interpretation aims at and is satisfied by adequate
relevance. The reader aiming at a non-spontaneous interpretation will
aim for an optimally relevant interpretation of the whole text. This

kind of interpretation can be understood in terms of exhaustiveness,
plausibility, and unity.7
Exhaustiveness: The reader makes a conscious effort to
account for all the evidence that the text presents.
Plausibility: The reader aims for an interpretation warranted by
the text and any other evidence that he can discover. The
stronger the evidence, the more plausible the interpretation,
constrained by the second principle of relevance (Sperber &
Wilson 1995, p. 266).
Unity: The reader spends considerable cognitive effort setting
up large, complex contexts. The ‘cashing-out’ of these efforts
depends on the resulting effects interacting with one another.
The reader aims to strengthen a core set of assumptions (or
global inferences).8 This core set strengthens the plausibility of
the interpretation, since all the evidence points toward this set;
and its exhaustiveness, since one test of global inferences is that
all parts of the work can be processed in a context in which
these inferences are highly manifest.
Non-spontaneous interpretation requires both a significant investment
of cognitive effort and far more time than spontaneous interpretation:
it constitutes productive repetition. It is practically impossible without
rereading, and is productive for it produces ‘extra’ effects. Rereaders’
reports and comments suggest not only that they are conscious that
second readings produce interpretations which differ from their initial
experience, but that they reread precisely in order to achieve such
interpretations:
(3) Rereading a book isn’t about going back over what
happened. It’s about returning to the work with context
gleaned from additional experiences and gaining new
insights. (Mrs Peel)
(4) On a second (or subsequent) reading, you’re not reading
for story … there will be subtleties that I’ll have raced past
while I looked out for what was going to happen … In the
best cases there’s a beauty in the arrangement of words,
sentences, and paragraphs that doesn’t wither with time.
(zeitgeist)
290 Anne Furlong
Readers’ terms for their goals – ‘gaining new insights’, ‘beauty’,

‘something new’ – suggest that typically, subsequent readings are
‘richer’ than the first ones. Their anecdotal evidence points to two
different kinds of readings, which accord well with the
spontaneous/non-spontaneous distinction I have drawn.
But what if the reader is not aiming at a non-spontaneous
interpretation, but at a second spontaneous interpretation? In that case,
the same effects are produced with (as we have seen) some reduction
in effort: repetition makes highly salient the same cast of characters,
the same plot, the same images. This sort of rereading could lead to
boredom, and relevance theory explains why this happens (when it
does). Since the text cannot vary, a second spontaneous interpretation
cannot produce the very wide range of very weak implicatures which
constitute adequate contextual effects for the effort expended. And in
general, the more closely the interpretations resemble one another, the
less relevant they are likely to be.
This account exposes the peculiarity of rereading aimed at producing
a spontaneous interpretation: the interpretation resembles itself in
virtually all particulars, and so ought to be less relevant, perhaps even
irrelevant. Nevertheless, there are many readers who, far from wanting
something ‘new’, reread because they want to experience exactly the
same thing a second time. Why bother to reread at all in this way?
5. Cumulative and non-cumulative readings

With respect to the phenomenon of rereading, productive repetition
may have two results. It may yield new implicatures (ie, amount to a
new experience). Or it may repeat desired effects at slightly less
processing cost (ie, result in a similar experience). I call the outcome
of these two types of repetition cumulative or non-cumulative
readings. These follow the division between spontaneous and non-
spontaneous interpretations, not because they are necessarily linked,
but because of the nature of these kinds of interpretive processes.
Non-spontaneous interpretation has as its goal optimal interpretation;
spontaneous interpretation aims for an interpretation that is adequate
for the effort expended.9
a. Cumulative readings and Non-spontaneous interpretation
Readers who anticipate that rereading will alter their interpretation (by
adding new assumptions, strengthening existing ones, weakening or
overturning and eliminating others) are aiming at producing non-

spontaneous interpretations. The reader’s expectations of relevance
are raised significantly, and so he is prepared to expend a great deal of
cognitive effort in deriving an interpretation. Instead of aiming for an
interpretation that is relevant enough, he is explicitly and consciously
aiming at an optimal interpretation. Rereaders indicate that they have
such expectations, and that they evaluate some texts at least on the
degree to which these can sustain such repetition. Their accounts also
support the suggestion (Krug et al., quoted in Brown 2002: 3) ‘that
“full processing of the text will occur” only when “readers’
representations of the text are absent from working memory”’.
(5) I definitely make sure to wait until the details have blurred
out in my mind a little bit so that I can regain a little bit of
surprise and suspense around the plot. (cimorene)
(6) The fact that I already knew the plot the second time
around did not deter me: at the age of twenty-six, I still
zoomed, suspense-driven, toward the final pages [of The
Portrait of a Lady], as if only the ending counted. (Lesser,
1)
Compare cimorene’s and Lesser’s (6) remarks with Mrs Peel’s and
zeitgeist’s (4). All are reading a second or subsequent time in order to
experience something new (or additional); but while Mrs Peel and
zeitgeist (4) assume that knowing the plot frees up the reader’s
cognitive resources to focus on other aspects of the text, both
cimorene and Lesser (6) claim that they are hoping to recapture
something of the original excitement – that is, to duplicate the
experience to some extent. Furthermore, they claim that they can do
precisely this, so long as they are actually rereading the text, and not
just retrieving an existing interpretation from memory. Their reports
suggest that if all the details of the text are highly manifest and readily
accessible in ‘working memory’, then the rereading will be
unproductive, and no accumulation of effects will occur. Their
comments are in line with the relevance theoretic account of
repetition: even a significant savings in effort will not be enough to
make a rereading relevant if no new effects are forthcoming; indeed,
the greater the savings in effort, the less likely it is that the rereading
will be worthwhile.
292 Anne Furlong
Some rereaders are not looking to replicate the original experience;

they want instead to ensure that their original interpretations will not
be duplicated, but rather altered and probably expanded. However, as
we have seen, not everyone who rereads is aiming at this sort of
experience – something new and different. Many are looking to
replicate an experience. It is to this kind of rereading that I now turn.
b. Non-cumulative readings and spontaneous interpretation
In relevance theory terms, a reader who is constructing a spontaneous
interpretation of a text may achieve one that is relevant enough to
warrant the effort he is prepared to expend on it, but this effort may
not be enough to achieve an optimally relevant interpretation of the
text as a whole. The reader whose goal is spontaneous interpretation
of a text will not search in any systematic way for connections, say,
between the passage he is currently reading and previous ones except
the most superficial way (keeping track of characters, plot, setting and
so forth). This is likely typical for the ‘first reading’ of most texts.
Nevertheless, there are readers who come back to a text for the
express purpose of duplicating their first experience. Rereading
without the expectation of an enhanced or expanded interpretation is
thus a puzzling phenomenon, not least from a relevance-theoretic
perspective. In this case, it would seem that the second reading
(ideally) resembles the first, and as we have seen, this is problematic
for relevance theory.
Nevertheless, some important elements are different. The reader
invests less effort constructing and confirming hypotheses about the
grammatical and narrative structures, since his previous interpretation
is accessible through memory. Furthermore, since the characters are
now familiar, no introductions are necessary, and the characters
behave in anticipated ways. Both conditions obtain because the reader
is already familiar with the work as a whole. Certainly, some aspects
of the text will have been forgotten; but even those who forget rarely
forget everything; their memory will be jogged at various points.
(7) I might not be surprised by anything anymore, but I can
still be sad, or happy, or creeped out. (wingstar)
(8) Every time I read [the Narnia books] again, I seem to …
suspend my recollection of them? I can read mysteries and
still be surprised by whodoneit [sic] every time.

(bibliotech)
In relevance theoretic terms, assumptions comprising the previously
constructed interpretation will become more manifest and readily
accessible as the reader processes the familiar material. The rereaders
who post their comments indicate that, when they are not aiming at a
non-spontaneous interpretation, they deliberately space out their
readings to minimise the effects of memory, which means that they
increase the cognitive effort required. This suggests that they may be
trying to replicate the original reading situation. Again, they are
seeking a balance between familiarity and novelty; according to their
own accounts, they return to a text because they want to experience
what they have undergone before.
I propose, however, that some readers – contrary to their own
accounts – do not in fact expect to repeat the original experience,
complete with ‘suspense’ and ‘surprise’; instead, they appear to be
interested in recapturing the effect of that experience. In fact, this kind
of rereader is aiming for an experience that follows the completion of
the first reading: that is, he is interested in what happens upon
completing the second reading. If that rereading yields adequate
effects, then he is likely to read the text a third time (and a fourth, and
so on). It may well be, then, that these readers spend the time with a
text in order to decide whether they want to reread it: for them, the
first reading tests whether the book will be worth coming back to.
(9) I reread for exactly the same reason I'm in fandom in the
first place: because I have a - possibly pathological! -
fascination/obsession with certain kinds of emotional
states, which can be induced by consuming certain texts
(movies, tv shows, books, fanfic etc.). A large part of
fandom is about preserving/ prolonging/ intensifying
certain emotional states, IMO, and for most people that
does seem to involve a repeated consumption of the
emotions-triggering material. When I find a text that
delivers exactly the kind of 'kick' I'm looking for, I try to
get the most out of it. One of the ways to do that is by
reading or watching it several times. (hmpf)
294 Anne Furlong
This reader, like many others, is avoiding an accumulation of effects;

he does not want each reading to bring something new, but rather to
return him to a desired state.
Clearly, this process is subject to the law of diminishing returns. If the
reader returns too soon or too often to a text (though some readers
report starting a book again immediately after completing the first
reading), then he will not produce the intended effect, since the
interpretation is only satisfactory if it is replicated rather than
retrieved from memory. Or the text may be too slight to withstand
rereading (such texts generally do not sustain non-spontaneous
interpretations, either), because the prior interpretation has made
manifest or easily accessible most of the assumptions in the previous
reading: they are retrieved in spite of the reader’s efforts. If this
occurs, then one of two things will happen.
In the first case, some readers give up: there are inadequate effects for
the effort expended (and each rereading brings the reader closer to
absolute minimum both of effort and effect), and they set the text
aside. In the second case, others (weblogs suggest that there may be a
great many) will go back to the same slender text time and again. This
second group of readers, despite the fact that the interpretation yields
inadequate effects, experience cognitive effects which adequately
repay the investment of their resources but which are not part of the
intended interpretation; they return to the text over and over.
For the first group, the interpretation of the text has failed: the set of
assumptions constructed in the search for relevance is not longer
adequate for the effort expended. These readers must move towards
non-spontaneous interpretation for the work to remain relevant and
readable. The interpretative process may also have ‘failed’ for the
second group. However, for them the context in which the text is
processed now includes a wide range of weak assumptions which do
not count as implicatures at all, but which are part of an eisegetical
interpretation – that is, the result of treating the text, or the reading
process, as a phenomenon. Now relevance results from treating the
text, not as a communicative act, but as the occasion for the accessing
and entertaining of a set of cognitive effects, not forming part of the
intended interpretation, that are produced almost entirely on the
reader’s responsibility. The term that some rereaders use for this
experience is ‘comfort reading’.
(10) Comfort reading is part of the appeal of genre fiction – you

know more or less what sort of story you’re getting, and
that it’s the kind of story you like – re-reading just
multiplies that assurance! (makesmewannadie)
(11) I want to re-experience the feel of the story. (ilyena_sylph)
The goal of such rereading resembles the effects of pornography
(which no one rereads spontaneously in order to undergo unexpected
and novel experiences) in that the reader’s purpose is to ‘induce’
‘certain kinds of emotional states’ independent of the intended
interpretation: indeed, it’s the repetition of standard tropes which is
crucial to genre fiction such as pornography (or romance novels, or
sci-fi).
This account may help explain why genre fiction attracts so many
rereaders. Much popular genre fiction is escapist in nature. Those who
reread the texts prefer spending time in the universe of the fiction; for
them, each rereading represents not just the opportunity to escape
from the reality of their lives, but to move into a world whose
inhabitants and conditions are preferable, because familiar and
reassuring. As flambeau remarks – ‘Rereading is a lot like revisiting
places I really like’, adding that his ‘aesthetic experiences are similar,
and gain by repetition.’ (flambeau, livejournal.com) Either flambeau
selects texts which can support non-spontaneous readings, or he has
constructed a context which permits the construction of a state of
mind which is strengthened with each reading. For many readers of
genre fiction, the first possibility is more likely, though the context in
which the (increasingly) non-spontaneous interpretations are taking
place includes non-propositional assumptions (the emotional and
psychological effects which the interpretation induces).
6. Conclusion
The non-spontaneous interpretation of a text very quickly reveals
whether the text can support such a reading (let alone rereading). The
effort expended will always be significant, even after multiple
encounters. The reader may well experience diminishing effects over
time, but unless the interpretation of the text as a whole fails to cash
out the effort, he is likely to regard the rereading as profitable. He
engages in rereading precisely because he is actively seeking new
implicatures, and is engaged in a process which must positively alter
296 Anne Furlong
his cognitive environment. The readings allow the accumulation of

effects over time; they are thus ‘cumulative’ rereadings.
Non-cumulative rereading – multiple spontaneous interpretations of a
text by a single reader over a period of time which do not result in
anything like a non-spontaneous interpretation – will usually demand
significantly less processing effort. Multiple readings will eventually
fail to yield adequate effects from the interpretation itself (regardless
of whether the text could support non-spontaneous interpretation), in
part because the reader is not actively seeking new implicatures, but
rather a repetition of the effect that followed the first reading. Readers
who produce non-cumulative rereadings of a work over a long period
of time may be able to create an adequately similar experience for low
processing effort; or they may incorporate their emotional and
psychological states into the context in which the text is processed,
thus paying less attention to the implicatures; or they may begin to
notice new aspects of the text, and thus drift toward non-spontaneous
interpretation.
The rereader who returns to a book and finds – whether or not he has
anticipated this outcome – that his interpretation has developed and is
richer and more meaningful, is engaged in cumulative rereading, in
which productive repetition is leading to the construction of a non-
spontaneous interpretation. The rereader who returns to a book and
finds that his interpretation has not materially altered – but that his
experience remains as satisfactory – is engaged in non-cumulative
rereading, in which productive repetition leads to the construction of
a spontaneous interpretation which satisfies the expectations of
relevance. In his case, the cashing out of effort produces largely non-
propositional assumptions: the emotional and psychological state
induced by the experience of reading the text a second (or subsequent)
time.
You can’t put your foot in the same river twice: it’s not the same river.
The spontaneous rereader can’t put his foot in the same river once.
(12) I reread books. I don’t remember which ones right now
though … which might explain why I reread them.
(ENeRgy)
Endnotes
1
Following the convention of relevance theory discourse, I refer to the speaker
(communicator, writer) as she and the hearer (addressee, reader) as he.
2
Such demonstrations in fact must be repeated verbatim; they fall into a set of
utterances or texts which must not deviate from one iteration to the next. Like
definitions, technical instructions, and protocols in general, their validity and their
efficacy depend on their accurate reproduction. Not all repetition is a waste of time.
3
For other approaches to analysing rereading, see Foster 1996.
4
The writers of the weblogs are identified by their aliases (eg, flambeau), which have
been italicised.
5
For some relevance-theoretic approaches to repetition, see Curl et al. (2006), Jucker
(1994), Padilla Cruz (1991). None of these papers addresses the issues central to this
paper.
6
There are several excellent introductions to relevance theory. Beside the foundation
text (Sperber & Wilson 1995), which includes a level of detail which is quite
formidable, there is the summary of relevance theory in the Handbook of Pragmatics
(Wilson & Sperber 2004).
7
This concept is explored and explained at length in Furlong (1996).
8
Global inferences are explored at length in Clark (1996).
9
Optimal relevance demands that the interpretation be the best that the reader could
construct under the circumstances, given his preferences and abilities. It is highly
demanding, and is in fact the kind of relevance which a non-spontaneous
interpretation aims at. See Sperber & Wilson (1995: 157-60, 261-3, 266-72) for a
discussion of the relationship between maximal relevance, optimal relevance, and
adequate relevance.
References
Brown, S. 2002. ‘What Happens When People Reread?’ Paper presented at the
Annual World Congress of the Association Internationale de Linguistique
Appliquée. Singapore. December 16-21.
Bursey, J., and A. Furlong. 2006. ‘Cognitive Gothic: Relevance theory, iteration, and
style’ in Tabbi, J., and R. Shavers (eds) Paper Empire: William Gaddis and the
World System. Tuscaloosa: U of Alabama P.
Caws, M. A. 1989. ‘Narrative Voice and Second Reading: Relation and Response’ in
Poetics Today 10(2): 243-253.
Chris. ‘Rereads.’ Message posted to Fool.Co.UK. Retrieved 15 July, 2006, from
http://boards.fool.co.uk/Message.asp?mid=10037804&sort=postdate.
Clark, B. 1996. ‘Stylistic analysis and relevance theory’ in Language and Literature
5(3): 163-178.
Claudia. ‘Comfort re-reading’. Weblog. Retrieved 7 July, 2006, from
http://clauclauclaudia.livejournal.com/tag/books.
Cornis-Pope, M., and A. Woodlief. 2000. ‘The Rereading/Rewriting Process: Theory
and Collaborative, On-line Pedagogy’ in M. Helmers (ed.) Intertexts: Reading
Pedagogy in College Writing Classrooms. Mahwah. Princeton, NJ: Lawrence
Erlbaum.
298 Anne Furlong
Curl, T.S., J. Local and G. Walker. 2006. ‘Repetition and the prosody-pragmatics
interface’ in Journal of Pragmatics 38(10): 1721-1751.
Foster, J. B. 1996. ‘Faces of Rereading’ in Poetics Today 17(2): 253-261.
Furlong, A. 1996. ‘Relevance theory and literary interpretation.’ Unpublished doctoral
dissertation, University College, London, UK.
Gerrig, R. J. 1989. ‘Reexperiencing Fiction and Non-Fiction’ in The Journal of
Aesthetics and Art Criticism 47(3): 277-280.
Ives, Susan. ‘Lanny Budd is gone, but his war-ravaged world still remains.’
Homepage of San Antonio, Texas. Retrieved 7 July, 2006, from
http://www.mysanantonio.com/opinion/columnists/sives/stories/MYSA111305.3
H.ives.c47f929.html.
Johnstone, B. (ed.). 1994. Repetition in Discourse: Interdisciplinary perspectives.
Princeton, NJ: Ablex Publishing Corporation.
Jucker, A.H. 1994. ‘Irrelevant repetitions: A challenge to relevance theory’ in
Repetition. SPELL (Swiss Papers in English Language and Literature) 7. A.
Fischer (ed.). Tübingen: Narr: 47-60.
Kawin, B. 1972. Telling it again and again: Repetition in literature and film. New
York: Cornell University Press.
Lesser, W. 2002. Nothing Remains the Same: Rereading and Remembering. New
York: Houghton.
Marie. ‘Booking through Thursday: Rereading.’ A Knit’s Tale. Weblog. Retrieved 18
July, 2006, from
http://knittale.blogspot.com/2006/06/booking-through-thursday-with-repeats.html.
Mary-Lue. ‘Rereads.’ Life, the universe, and everything. Weblog. Retrieved 9 July,
2006, from http://www.ltuande.blogspot.com/2006/06/booking-through-thursday-
with-repeats.html.
McCormick, P. 1985. ‘Feelings and Fictions’ in The Journal of Aesthetics and Art
Criticism 43: 375-383.
McQuade, M. 2002, March 1. ‘Wendy Lesser: A Portrait of the Writer as a Reader
(and Rereader).’ Booklist: 1091.
Millis, K. K., and A. King. 2001. ‘Rereading Strategically: The Influences of
Comprehension Ability and A Prior Reading on the Memory for Expository Text’
in Reading Psychology 22: 41-65.
Morreall, J. 1993. ‘Fear Without Belief’ in The Journal of Philosophy 90(7): 359-366.
Mrs Peel. ‘Review: A Book of Five Rings.’ Skylark of Valeron. Weblog. Retrieved 7
July, 2006, from
http://skylarkofvaleron.blogspot.com/2006/06/review-book-of-five-rings.html.
Neal, N. 1987. ‘Functions of repetition in conversation’ in Text 7: 245-64.
Neill, A. 1991. ‘Fear, Fiction and Make-Believe’ in The Journal of Aesthetics and Art
Criticism 49(1): 47-56.
Padilla Cruz, M. 2001. ‘Learning about the role of repetitions of phatic sequences: A
relevance theoretic approach.’ Paper given at V International Conference on
Languages for Specific Purposes. E.U. d’Enginyeria Tècnica en Teixits de Punt
(Canet de Mar). University of Barcelona. 6 September, 2001.
Persson, G. 1974. Repetition in English: Part 1. Uppsala: Uppsala University Press.
‘Rereading.’ Weblog. Retrieved 7 July, 2006, from
http://kyuuketsukirui.livejournal.com/427066.html.
Sample, K. J. 2005. ‘Promoting Fluency in Adolescents with Reading Difficulties’ in

Intervention in School and Clinic 40(4): 243-246.
Schroeder, Al. ‘Everything you wanted to know about Mindmistress (*but were afraid
to ask).’ Mindmistress. Online graphic novel. Retrieved 15 July, 2006 from
http://mindmistress.comicgenesis.com/mmknow.htm.
Skulsky, H. 1980. ‘On Being Moved by Fiction’ in The Journal of Aesthetics and Art
Criticism 39: 5-14.
Sperber, D., and D. Wilson. 1995. Relevance: Communication and Cognition. 2nd
Edition. Oxford: Blackwell.
Stine-Morrow, E. A. L., D. D. Gagne, D. G. Morrow and B. H. Dewall. 2004. ‘Age
Differences in Rereading’ in Memory & Cognition 32(5): 696-710.
Suleiman, S., and I. Crosman (eds). 1980. The Reader in the Text: Essays on Audience
and Interpretation. Princeton, New Jersey: Princeton University Press.
Walton, K. 1978. ‘Fearing Fictions’ in The Journal of Philosophy 75(1): 5-27.
Wellthen13. ‘Currently reading.’ Weblog. Retrieved 15 July, 2006, from
http://www.xanga.com/wellthen13.
‘Who rereads books?’ Weblog (archive). Retrieved 7 July, 2006, from
http://forums.ogresnet.com/archive/index.php/t-2054.html.
Wilkinson, J. 2000. ‘The Paradox(es) of Pitying and Fearing Fictions’ in South
African Journal of Philosophy 19(1): 62-74.
Wilson, D., and D. Sperber. 2004. ‘Relevance theory’ in Handbook of Pragmatics. G.
Ward and L. Horn. (eds). Oxford: Blackwell: 607-632
Zabrucky, K. M., and D. Moore. 1999. ‘Influence of Text Genre on Adults’
Monitoring of Understanding and Recall’ in Educational Gerontology 25: 691-
710.
300 Anne Furlong
Appendix A: Reported reasons for rereading

Seeking ‘new’ experiences
Rereading a book isn’t about going back over what happened. It’s about returning
to the work with context gleaned from additional experiences and gaining new
insights. (Mrs Peel, skylarkofvaleron.blogspot.com)
On a second (or subsequent) reading, you’re not reading for story … [the storyline
will] absorb too much of my attention in the first reading, so that there will be
subtleties that I’ll have raced past while I looked out for what was going to happen
… in the best cases there’s a beauty in the arrangement of words, sentences, and
paragraphs that doesn’t wither with time. (zeitgeist, livejournal.com)
When I reread a book, I don’t expect the words to change. It is I who have
changed; the world has sadly remained the same. (Susan Ives, mysanantonio.com)
Seeking ‘same’ experiences

Comfort reading is part of the appeal of genre fiction – you know more or less
what sort of story you’re getting, and that it’s the kind of story you like – re-
reading just multiplies that assurance! (makesmewannadie, livejournal.com)
I have a (possibly pathological) fascination/obsession with certain kinds of
emotional states, which can be induced by consuming certain texts … a repeated
consumption of the emotions-triggering material … Ironically, though, I get bored
with other forms of repetition quickly. (hmpf, livejournal.com)
Every time I read [the Narnia books] again, I seem to … suspend my recollection
of them? I can read mysteries and still be surprised by whodoneit every time.
(bibliotech, livejournal.com)
I definitely make sure to wait until the details have blurred out in my mind a little
bit so that I can regain a little bit of surprise and suspense around the plot.
(cimorene, livejournal.com)
Even though the element of suspense was absent, I still derived pleasure from
rereading. (XsMarbles, bloggerforum.com)
I might not be surprised by anything anymore, but I can still be sad, or happy, or
creeped out. (wingstar, livejournal.com)
I want to ‘see’ those people again. I want to re-experience the feel of the story … I
want those particular people to be part of my life again … (ilyena_sylph,
livejournal.com)
Rereading is a lot like revisiting places I really like … those aesthetic experiences
are similar, and gain by repetition. I’m not sure I can explain why, though.
(flambeau, livejournal.com)
Other comments
My whole library is things I might theoretically re-read. Otherwise, why own ‘em?
(clauclauclaudia, livejournal.com)
Appendix B: Do you ever reread your books?

Christine F, Mary So-So Cal,
Marie, blogspot.com
livejournal.com blogspot.com
Given that I asked
YES. these questions
Do you I need some ‘comfort partially in hopes of
ever reading’. confirming that I’m not
reread Haven’t read the such an odd duck in
Oh, yeah!
your book in a long time obsessively reading
books? and think it is time to books over and over
do so again. again, well, you have
your answer.
Rereading a beloved
I always forget
book is like coming
If so, enough detail to
Just about any book I home.
which enjoy them over
read and enjoy first Books containing
ones? If and over again.
time … is a potential worlds that I loved
not, why The payoff comes
candidate inhabiting the first time
not? at the end [of an
around get picked up
epic series].
again and again.
I almost always read
It depends on both my
from beginning to
mood and the book. …
end…
I almost always Sometimes … a
Do you Only exceptions are if
re-read a book the particular line or scene
read I get really fed up and
whole way will suddenly crop up
books the don’t finish the
through. in my thoughts and
whole book…
… a few nothing will do until
way [or] I get started on
occasions … I I’ve revisited it in the
through another book and the
just wanted to book.
or pick first gets neglected or
read a specific …there are those books
through forgotten…
paragraph or that contain scenes I
favourite [or] I’m looking for a
section because it find unpleasant to
scenes? specific quote or
is that good. revisit … and I will
section and not really
often skip these scenes
aiming to reread the
on a rereading.
whole book.
…very enjoyable
and quick reads
Has to be a book I
I know that I will
What really like, with … special ones that
like them, maybe
qualifies characters I like or a render me insensible to
don’t remember
a book really interesting the world while reading
all the details and
for the plot… and send my
so it is fresh and I
reread Other than that, no imagination racing
know what I’m
pile? universal after I am finished.
getting.
requirements.
I’m bored and it is
there.
302 Anne Furlong
She has total recall; she can literally remember, in perfect detail,
everything that has happened to her. She has memorized the Atlanta
phone book, and can tell you whose name a phone number is under –
just from hearing any random number in the phone book. She can read
an entire page at a glance – and often reads twelve books a night, and
will hold it in perfect memory, forever. As a consequence, she never
rereads a book. (‘Everything you wanted to know about Mindmistress,
Com- but were afraid to ask’, http://mindmistress.comicgenesis.com)
ments I’m not one of those people that rereads books all the time. … But
from now I’m desperate. … One [sic] you read the books once, that’s it.
non- (wellthen13, http://www.xanga.com/wellthen13)
rereaders … I naturally read stuff carefully and slowly, word by word. Most
people just scan or read quickly. Means I remember stuff I’ve read,
but get through less. (r3negade,
http://www.ogrenet.com/forums/showthread.php?t=2054)
I can’t reread books. For some reason I don’t remember waht [sic]
happens until I start reading the book, and since I remember what will
happen it is less enjoyable. The same goes for watching movies.
(lorax, http://www.ogrenet.com/forums/showthread.php?t=2054)
Dialogue and Discourse Structure:
A Speech Move Analysis of Sherman Alexie’s Story
‘What You Pawn I Will Redeem’
Robert A. Troyer
Chulalongkorn University, Bangkok
Abstract
This paper presents a discourse analysis framework that can be applied to dialogue in
fiction. Based on an elaboration of Halliday’s functional approach to conversational
interaction combined with traditional conversation and discourse analysis and speech
act theory, the framework posits a hierarchical categorization of opening and
responding speech moves. When applied to fictional dialogue, this analytical method
offers a descriptive apparatus that can be simple or complex depending on one’s needs
(i.e., pedagogical or research oriented) while also providing insight for interpretation
of character interaction. The major strength of the approach is its ability to capture not
isolated speech acts, but the interactive nature of conversation – the verbal dance of
dialogue between characters in a narrative. Initiating and continuing speech moves
(both verbal and non-verbal) with various discourse functions are followed by
responding moves that can be grouped into the two broad classes of supporting or
confronting. Quantificational analysis of such description provides empirical support
for readers’ intuitions about conversational exchanges.
As a sample analysis, the framework is applied to all of the dialogue in the short story
‘What You Pawn I Will Redeem’ by Sherman Alexie (2004). This particular story,
with its fourteen distinct conversational interactions between the main character and a
variety of other characters of differing degrees of status and solidarity, provides an
ideal demonstration of the proposed method of analysis. The main character, a
homeless Native American Indian in Seattle, Washington, exhibits distinctly different
patterns of discourse or conversational styles in his interactions with friends,
strangers, and acquaintances of higher status. Such discoursal indications of power
and solidarity are not only inherent in the dialogue, but central to the story’s broader
themes of the individual’s role in society as well as distinctly Native American
concerns for heritage and preservation of cultural identity. In keeping with the
descriptive perspective of conversation analysis though, generalizations about
interaction in different situations should only serve as guidelines – likewise, the
power of stylistic analysis lies in its ability to help interpret the linguistic subtleties of
a given text. This study demonstrates that analysis of the discourse functions of
speech moves in the dialogue of fictional narratives serves the purposes of
explication, which are central to stylistics and literary study.
Keywords: dialogue; Sherman Alexie; discourse move analysis; conversation
analysis; sociolinguistic variation.
304 Robert A. Troyer
1. Introduction
Nearly all works of prose fiction contain represented speech.
Commonly referred to as direct and indirect speech, much of it is not
single utterances by one speaker but multi-turn conversations between
characters: the dialogue. While dialogue has been addressed in non-
academic writing – handbooks for writers about how to create
successful dialogue in fiction (Chiarella 1998, Kempton 2004,
Stanbrough 2004, Turco 1991, 2004) – academic research has
provided much more specific analysis. Detailed studies of methods of
speech presentation and their effect on narrative and perspective have
been performed (Leech and Short 1981, Short 1996, Semino, Short
and Culpeper 1997) with special emphasis on free indirect discourse
(Fludernik 1993), and attention has been given to authors’ methods of
imitating real speech in its many varieties both idiosyncratic and
dialectical (Page 1988, Short 1996). Another approach is to analyze
dialogue in terms of interaction between characters.
The emergence of methodologies in the fields of pragmatics,
discourse, and conversation analysis has equipped stylisticians with
means to examine the verbal interaction that is represented in plays,
novels, and stories. Leech and Short (1981) and Short (1996) offer
overviews both theoretical and practical of how to apply speech act
theory, Gricean maxims, conversation analysis (CA), and politeness
theories to characters’ interactions. Such methods have been shown by
many stylisticians to profitably illuminate the subtlety and
significance of fictional dialogue in a variety of dramatic and prose
literary works.
What most of these studies in pragmatics share is an emphasis on
close analysis of selected passages with detailed explanations of the
characters’ verbal interaction. Such studies effectively highlight the
relationships between characters and the functions of dialogue while
explicating the work as a whole; however, they are less able to capture
the interactive nature of conversational exchanges in a way that is
quantifiable, thus making them less amenable to broader stylistic
comparison and contrast.
The analytical framework proposed in this study is a compromise
between two earlier stylisticians’ approaches to discourse analysis of
fictional dialogues. The first of these methods was proposed by
Burton (1980, 1982). Her methodology is a development of the
Dialogue and Discourse Structure 305
‘Birmingham School’ of discourse analysis, which is based on Sinclair

and Coulthard’s (1975) observations of teacher-student interaction.
Burton posits (as others have since) a broader, more inclusive
categorization of Exchanges, Moves, and Acts which can be applied to
any conversational situation.
In this system, any transaction between two speakers must be made up
of at least one exchange. A conversational exchange must contain at
least two moves: the first speaker uttering a type of initiating move
(Burton identifies 3 types of ‘Opening’ moves) to which the second
speaker can Respond with a Supporting or Challenging move. Each
move is composed of at least one act, the smallest building blocks of
conversation, which are classified into twenty-one types that can
account for all basic functions of conversational interaction. A sample
of Burton’s method of analysis is shown below in Example 1.
Exchange Move Act
1 Gus I want to ask you pre-topic opening metastatement
something
2 Ben What are you doing topic 1 opening elicitation
out there
3 Gus Well, I was just support reply
4 Ben What about the tea? topic 2 opening elicitation
5 Gus I’m just going to make it support reply
6 Ben Well, go on, make it bound-opening directive
7 Gus Yes, I will. support react
Example 1 Sample of Burton’s (1982) analysis of the opening lines of
The Dumbwaiter
Due to the relatively large number of act functions (twenty-one)
classified in this framework, and since some acts can realize Opening
or responding moves depending on the context, using acts for
quantificational stylistic analysis is problematic. An analysis of the
move structure, however, is a more profitable application of this
approach because the discourse function of moves encodes the
speakers’ interaction. Unfortunately, the three categories of initial
moves (Bound-, Re-, and Opening) followed by either a Supporting or
Confronting Move is a limited characterization of what happens in a
conversation. What is needed is a more specific taxonomy of initial
moves that is more useful for descriptive analysis. Furthermore, any
quantificational data must be amenable to comparative analysis that

foregrounds stylistic differences between characters, works of fiction,
authors, or time periods, depending on the scope of the stylistic
analysis.
Another stylistician who has applied this kind of structural discourse
analysis to fictional dialogues is Michael Toolan (1985, 1990, 1998,
2000). Toolan proposed a simplified schematic based on a functional-
systemic approach to verbal interaction and applied this to passages
from plays. His framework (shown below) proposes four basic
initiating moves and one class of response. Initial moves are classified
according to whether the reference is mental or physical (action or
information), and whether the orientation of the exchange is to or from
the addressee (giving or seeking).
INITIAL MOVES Actions Information
(proposals: intention to (propositions: intention to
do) know)
Seeking Request Question
(speaker from Will you give me a hand? Have you got a good hold?
addressee)
Giving Undertaking Inform
(speaker to Can I give you a hand? I mustn’t do any heavy
addressee) lifting.
PROSPECTED Request—(Acknowledgement +) non-verbal performance

RESPONSES Undertaking—Acknowledgement (accept/decline)
to the four Question—Inform
Initial Moves Inform—Acknowledgement
Figure 1 Toolan’s (2000) taxonomy of basic discourse functions

Thus, within Toolan’s framework, five types of discourse moves can
describe the unfolding production of conversational interaction in
dialogue. Below in Example 2 is a sample of Toolan’s analysis of the
‘interrogating Stanley’ scene in Pinter’s The Birthday Party.
MCCANN. Nat. Request
GOLDBERG. What? Question
MCCANN. He won’t sit down. Inform
GOLDBERG. Well, ask him. Request
MCCANN. I’ve asked him. Request
GOLDBERG. Ask him again. Request
MCCANN (to STANLEY). Sit down. Request

STANLEY. Why? Question
MCCANN. You’d be more comfortable. Inform
STANLEY. So would you. Inform
Pause.
MCCANN. All right. If you will I will. Acknowledge;
Inform/Undertake
STANLEY. You first. Request
MCCANN slowly sits at the table, left
Example 2 Sample of Toolan’s (2000) analysis of a scene from The
Birthday Party
While this method of analysis offers a way of approaching how the
dialogue works stylistically at the discourse level, most of Toolan’s
more insightful comments come from attention to the content and
grammar of the lines and de facto classification of responses rather
than from his discourse move framework. His use of a move-level
analysis and his taxonomy of initial moves is the strength of the
approach. However, by not distinguishing in his coding between
initiating and responding moves and by not classifying responses, he
greatly reduced the descriptive power of the move framework.
This paper presents an analytical method that systematically
categorizes functional initiating and responding move types. The basic
framework, explained and shown in Figure 2 in the Methods section
below, is based mainly on Eggins and Slade’s (1997) elaboration of
Halliday’s functional-systemic approach as applied to casual
conversation, along with insights from Tannen’s (1990) work on
silence. Such a framework can be a reliable and accurate tool for
describing the interactions between characters in a way that can be
applied to a wide body of conversations for quantifiable analysis of
literary style which can then be interpreted for what it reveals about
relationships between characters.
While Burton and Toolan focus primarily on the dialogue of plays, the
analysis of fictional dialogue not only applies to prose fiction but also
highlights the differences between drama and prose. In plays, the
dialogue normally does all the work of telling the story and must by
necessity bear the weight of plot development and contextual
information such as characters’ names and events that occurred prior
to the action of the play. In prose fiction, however, such practical

information is generally conveyed by the narrator, thus, dialogue is
typically reserved for highlighting the verbal interaction of characters.
While such interaction does convey plot-developing events and
introduces new information, the fact that the author chose to place
such information in the form of dialogue foregrounds the relationships
between characters by placing it in spoken form – that which is the
greatest contrast to written prose and most similar to the interactions
that structure people’s daily social roles.
2. Purpose
One goal of literary stylistics is to raise conscious awareness of the
means by which authors convey their fictional worlds. This study
seeks to achieve this goal by closely analyzing the dialogue in a short
story by using methods derived from conversation analysis, speech act
theory, and functional-systemic grammar. A discourse move analysis
can explain how an author uses dialogue to develop unique individuals
and to create relationships between characters who are members of
larger socio-cultural groups. This paper offers a discourse move
framework combined with a sociolinguistic approach to language
variation. An analysis of dialogue in Sherman Alexie’s short story
‘What You Pawn I Will Redeem’ (2004) will reveal stylistic
tendencies (at the discourse level) for certain characters and variation
in characters’ conversational styles along social dimensions of
language use which index relationships of hierarchy and social
distance between the characters in the story.
3. Methodology
For this study, the larger issue of the relationship between author and
reader will be set aside in order to focus on the discourse world of the
characters. One reason dialogue ‘works’ is that authors portray subtle
linguistic variation along the dimensions of age, gender, social status,
degree of intimacy, and formality of situation. This variation is present
not only at the level of word choice and pronunciation (as indicated by
spelling), but more subtly at the level of discourse moves – their
interactional tendencies. To reveal conversational styles, a
comparative analysis of spoken discourse must account for the social
aspects of characters’ identities and context. One way of doing this is
to classify dialogues according to the relationship between the
speakers. In Analyzing Casual Conversation, Eggins and Slade

(1997), base their classification on degree of intimacy/social contact.
For application to the wider variety of interactions contained in
stories, I have modified their classes, which will subsequently be
referred to as relationship domains, as shown in Figure 2 below.
Non-service Service
brief extended
voluntary non-voluntary
romantic friendly family acquaintances strangers

Figure 2 Relationship domains of conversational interaction
To the degree that social relationships are at all classifiable, these
domains are a useful starting point for grouping interactions
encountered in stories. Further distinctions based on age and status
relationships or other variables can be made within the above
domains. For example, adult-child friends will interact differently and
should be distinguished from peer friends; likewise, acquaintances
may be of equal or unequal status which will significantly effect their
conversations. Typically, the characters contained in any one story
represent a limited social network of interactions that can easily be
analyzed into a few relationship types according to the above
variables.
Though traditional CA divides conversations into turns, sequences,
and pairs as the fundamental units of conversation, this paper follows
the Birmingham school’s approach because it allows every utterance
in a conversation to be classified systematically according to its
function in the discourse. Rather than analyzing selected sequences of
pairs of turns from a conversation, the analysis applied in this study
classifies each functional move in the dialogue in order to describe the
types of interaction that are present.
The move level (similar to traditional CA’s Turn Constructional Unit
or TCU) describes an utterance after which speaker change could
occur. While a turn must consist of at least one move, multiple moves
per turn are common. Moves can be realized as one or more clauses,
an elliptical clause, a conventional utterance (interpersonal moves
such as ‘Thanks’), or a non-verbal action. Context and authors’ use of

punctuation and placement of speech reporting clauses (i.e., he said)
usually provide unambiguous indications of move boundaries.
Though the basic move categories used in this analysis (see Figure 2)
could be further classified and given labels that correspond to more
specific speech acts (as in Burton’s work), a move analysis with
sufficient discrimination between discourse functions can capture the
outcomes of interaction in a way that is amenable to quantification
and comparison of the interactive nature of dialogue. Figure 2
provides a schematic of basic discourse move functions, based on
Eggins and Slade (1997), which were used to describe the dialogues in
this study.
This framework for move analysis allows the researcher to describe
the discourse function of each move in a conversation according to a
hierarchical classification system. For this study, I have modified
Eggins and Slade’s (1997) framework in several respects. The basic
schematic consists of a few functional choices of Initiating moves that
can be followed by either a Continuing move by the same or next
speaker, or a next speaker’s Response to an initial move. All
Responses can be placed into one of two groups: supporting or
confronting. For further descriptive power, supporting and confronting
moves can be divided into more specific categories.
-------------------------------------------------
info
give
action
Initiate
info
seek
action
Interpersonal*
-------------------------------------------------
support
Respond
confront
-------------------------------------------------
-------------------------------------------------
prolong
Continue
append
-------------------------------------------------
attend
Textual
monitor
-------------------------------------------------
Delay
-------------------------------------------------
Figure 3 Simplified discourse function move schematic
* Interpersonal moves convey mainly interpersonal rather than
referential meaning such as acts of greeting, thanking, or apologizing.
The benefit of such a framework for this study is that it describes
various types of both initiations and responses. The following
descriptions of move types should help clarify their respective uses
from a functional perspective.
x Initiating Moves (I) are functionally geared toward giving
or seeking information or action or expressing interpersonal
relationships.
x All moves that are Responses (R) to a previous speaker can

be classified as either supporting (sup) or confronting (con)
– even interpersonal moves that are reactions and non-
verbal moves such as laughter or silence.
x Continuations (Cont) are moves that can either directly
follow a speaker’s move, thus prolonging the turn, or can
follow another speaker’s response but still extend, enhance,
or elaborate a speaker’s own previous move.
x Textual Moves (TM), or Turn Management moves, serve
mainly as structuring or organizing devices that signal the
start of a speaker’s turn or the maintenance of a turn. Such
moves have no actual informational value.
x The Delay (D) as displayed in stories is generally a silence
(indicated by the narrator) prior to a move which, despite its
non-verbal realization, makes a functional contribution to
the ongoing discourse. Tannen (1990) discussed the
functions of silence and its role in British and American
theater. Since pauses and silences are often indicated in
stories as part of the conversational interaction, they are
also given a functional label in this study.
When applied to conversational dialogue, this basic framework
provides a way of describing the amount and kinds of verbal
interaction between characters. As this methodology is a development
of speech act theory, it should be understood that the use of ‘act’ or
‘move’ here is different from Searle’s notion in that acts are viewed as
functionally motivated within the context of a move and exchange.
Thus, moves and acts do not refer to the ‘speech acts’ of Searle’s
logical-philosophical theory which focuses on the force of individual
utterances, but to ‘discourse moves and acts’ which denote the
function of utterances in the context of a conversation – not speech
acts, but speech interacting.
4. ‘What You Pawn I Will Redeem’

With the above purpose in mind, the short story ‘What You Pawn I
Will Redeem’ by American author Sherman Alexie was chosen for
analysis. The story, published in The New Yorker in 2004 and selected
for inclusion in America’s Best Short Stories 2004, contains 6,756
words of which 2,949 (43.65%) represent conversational interaction,

the remainder constituting the narration of the story. The story
contains a total of fourteen conversations ranging in length from 37 to
996 words. Placed into this fictional world of discourse are a range of
characters of varying social distance, group memberships, and relative
status levels. As these characters and their relationships are central to
the story, a stylistic analysis of their conversations serves the above
purpose and proves the validity of this method of analysis.
While stylistic analysis can be used to make generalizations about a
novel, author, genre, or time period, such approaches can also be used
to explicate a single story as is the case with this study. Some
background on the story and author are, thus, crucial to placing the
data below in context. At forty years old, Sherman Alexie, born in
1966 and a registered member of the Native American Spokane tribe,
has been prolific enough to fill several pages of his personal website
<http://www.fallsapart.com/index.html> with more honors and awards
and grants for his creative writing than can be listed here. He is
currently an Artist in Residence at the University of Washington in
Seattle where he teaches courses in American Ethnic Studies.
Alexie’s writing is most often praised for its creative vocabulary,
vivid imagery, and irreverent humor (Tabur-Jogi 2004). Most of his
stories take place on the Spokane reservation and offer an unflinching
portrayal of modern Native Americans, who despite stereotypical
behavior, become distinct individuals. Common themes of Native
American authors like community and audience, place and history,
spiritual experiences, and the nature of language and storytelling are
present in his works (Fast 1999). By Alexie’s own explanation in
interviews, his world view is summarized in the statement, ‘I’m a
Spokane/Coeur d’Alene Indian’ (Banks 1995).
‘What You Pawn I Will Redeem’ is a first-person account of twenty-
four hours in the life of a homeless, alcoholic Native American named
Jackson Jackson who lives on the streets of Seattle. The plot of the
story follows an episodic quest structure after Jackson sees what he
believes to be his deceased grandmother’s powwow regalia displayed
in a mysterious pawnshop window. The regalia which had been stolen
years earlier (one of the great tragedies of his family) appears to be
authentic owing to the presence of his family’s signature yellow bead
hidden in the armpit. The mildly sympathetic pawnbroker allows
Jackson twenty-four hours to gather 999 dollars to buy the regalia, and
gives him twenty dollars for encouragement.
Thus begins Jackson’s quixotic odyssey, in which a former boss gives
him a gift of free newspapers to sell, he encounters three lonely Aleut
Indians who watch the sea and sing of their faraway home in Alaska,
he recollects stories of his grandmother, vomits his first solid food of
the day, wins the lottery, loses, gains, and loses friends, is beaten, and
passes out penniless again on the railroad tracks where he is rescued in
the morning by a friendly police officer who donates money to
Jackson’s hopeless cause: twenty dollars which Jackson uses to buy
breakfast for the Aleuts and himself. Returning to the pawnshop with
only five dollars, he resigns himself to his fate, but the shopkeeper
insists that Jackson take the regalia for free. Wearing the regalia,
Jackson, ‘stepped off the sidewalk and into the intersection.
Pedestrians stopped. Cars stopped. The city stopped. They all watched
me dance with my grandmother. I was my grandmother, dancing.’
Jackson Jackson is one of Alexie’s asocial/antisocial characters who
typically drop out of mainstream society by choosing to give up
family, employment, and home in exchange for a wandering self-
destructive alcoholism. However, and this is a big ‘however’, Jackson
Jackson is so charming both in his interactions with other characters
and in his story-telling ability that his negative qualities are
downplayed. He is the kind of happy-go-lucky drunk who buys
alcohol for his friends and himself with the money from the
pawnbroker, gives twenty dollars of the 100-dollar winning lottery
ticket to the girl at the Korean market where he bought the ticket, and
spends the rest on drinks for his ‘cousins’ (strangers he befriends at a
bar frequented by Native Americans). Even Officer Williams extols
Jackson’s wit and sense of humor in the face of being picked-up off
the railroad tracks. As a narrator, Jackson lives up to all the praises
given to Alexie. He is humorous, often in a wry, self-aware way,
honest yet sympathetic while describing others, inventive and precise
with words, and willing to share his feelings and stories about his
grandmother and family with the reader.
The above summary and interpretation may seem excessive in a paper
proposing a stylistic analysis, yet such detail is important if we are to
understand the context of the conversational interaction represented in
the story. Whether one is reading for enjoyment or study, or in the role
of a teacher, a story like ‘What You Pawn I Will Redeem’ requires

acknowledgement of Alexie’s Native American world view. Related
to common themes in Native American literature mentioned above, is
the shared sense of brotherhood among many Native Americans, who
despite tribal differences that hundreds of years ago may have divided
them, are now united under their common plight of loss (of land,
heritage, culture, autonomy) following years of oppression and
discrimination. Thus, in the following analysis, conversations between
strangers and acquaintances who are Native Americans and/or of
similar social status are seen to be more like conversations with
friends (highly interactive and with much non-verbal communication)
than conversations with acquaintances from other cultural groups and
social classes.
5. Procedures
Having scanned and converted the text of the story to a word
processing file, the conversations were cut and pasted into a coding
sheet for manual marking and numbering of turns and moves per
speaker and subsequent labeling of the discourse function of each
move. Below is a sample of a coded dialogue, formatted vertically to
fit this page. Shown are turns 14 to 18 from the third conversation in
the story with each move on its own line, allowing the reader to read
downward, move by move, through the conversation.
Example of a coded dialogue

T/M Speaker Text SR Discourse function
14a Pawn- There it is, ds I:give
broker info:fact:state
14b … You were right. ds R:sup:acquiesce
14c … This is your grandmother's ds Cont:R:sup:acquies
regalia. ce
15 Junior It's been missing for fifty ds R:sup:dev:(I:give
years. info:claim)
16a Jackson Hey, Junior, ds TM:attend
16b … It's my family's story. ds R:con:counter
16c … Let me tell it. ds I:seek action:direct
17a Junior All right. ds R:sup:acquiesce
17b … I apologize. ds R:sup:interpers:apo
logize
17c … You go ahead. ds R:sup:comply
18 Jackson It's been missing for fifty ds Cont:I:give
years. info:claim
T/M = Turn and Move; SR = method of Speech Representation

I = Initiate; R = Respond; Cont = Continue; TM = Turn Management
Non-verbal moves in fictional dialogue
An additional goal of this study was to account for the non-verbal
communication described in dialogue. Three types of non-verbal
action were identified: only one offers functional contributions to the
discourse of a dialogue while the others provide accompanying or
contextual detail.
1) Non-verbals that contribute a functional move to the
discourse: these were coded and counted as any other move
but labeled NV rather than ds (direct speech), is (indirect
speech), or nrsa (narrative report of speech act).
2) Non-verbal descriptions that accompany a functional move:
these were not labeled or counted as moves but were placed
on a separate line either directly before or after the verbal
accompaniment which carried the coding. These were
included in the analysis for additional study of the non-

verbal interactions of characters.
3) Non-verbal descriptions that provide context but do not
significantly contribute to the character’s interaction.
Because these can be helpful during coding, they were
included in [brackets] in the text column of the coding
sheets, but not labeled as moves or methods of representing
interaction.
Non-verbal Type 1 (move #4, bold-faced type)
3a Jackson That’s my ds I:give info:claim
grandmother’s powwow
regalia in your window
3b … Somebody stole it from ds Cont:I:give info:claim
her fifty years ago, and
my family has been
searching for it ever
since
4 Pawn- looked at him like he NV R:con:counter
broker was a liar
5a Jackson I'm not lying, ds R:con:refute
5b … Ask my friends here. ds Cont:R:con:refute
5c … They'll tell you. ds Cont:R:con:refute
Non-verbal Type 2 (line 1, bold-faced type)

Williams kicked Jackson in the ribs N
V
1a … Jackson ds TM:attend
1b … Is that you ds I:interpersonal:greeting
2a Jackson Officer Williams ds R:sup:engage
Non-verbal Type 2 (lines 4+5) and Type 3 (line 6, bold-faced type)

35a Pawn- We have a deal. ds R:sup:dev:(R:sup:ac
broker cept)
35b … And I'll get you started. ds I:give action:offer
35c … Here's twenty bucks. ds Cont:I:give
action:offer
… opened up his wallet, NV
pulled out a twenty-dollar
bill, and gave it to Jackson
[Rose, Junior, Jackson
walked out]
Overlapping moves (interruptions) and shared moves in fictional

dialogue
One advantage of analyzing fictional dialogue as opposed to real
speech, is the clear distinctions between speaker turns. Interruptions,
however, may be indicated with ellipsis, dashes, line breaks, or
syntactic contrast which are clear indicators of turn change when
speakers make functional moves. There are cases, however, when
narrators indicate moves shared by more than one speaker by
reporting non-verbal action or speech acts without direct or indirect
speech. In ‘What You Pawn, I Will Redeem’ this is seen frequently
when conversationalists share laughter.
‘Honey Boy,’ I said, ‘you can try to seduce me, but my heart
belongs to a woman named Mary.’
‘Is your Mary a virgin?’ Honey Boy asked.
We laughed.
This type of laughter is classified in this paper as a supporting
response. Because each character contributes to the laughter and this
analysis attempts to account for each character’s moves, the coding is
as shown below. In this way, the coding of moves can account for
each character’s contribution to a shared interaction which is not
captured in traditional analysis of represented speech.
Shared discourse moves (moves 32 and 33, bold-faced type)

30a Jackson Honey Boy ds TM:attend
30b … you can try to seduce ds R:con:re-chall
me,
30c … but my heart belongs to ds Cont:R:con:re-chall
a woman named Mary.
31 Honey Is your Mary a virgin? ds R:sup:confirm
Boy
32 Jackson laughed NV R:sup:register
33 Honey laughed NV R:sup:register
Boy
6. Data analysis and results

The majority of dialogue in the story was rendered in direct or free
indirect speech (97.41%) with a small amount presented as narrative
report of speech act (1.65%) and indirect speech (0.94%). These
figures are based on a word count of the reported speech rather than
counting instances of each type of representation which could yield
slightly different results. Table 1 presents a list in order of occurrence,
of the fourteen conversations in the story, their participants, and their
relationship domains along with the number of turns and moves in
each dialogue. It should be noted here that all of the strangers who
Jackson spoke to were Native Americans of similar social status to
Jackson. As will be shown below, the most significant influence on
Jackson’s style of interaction with others is not degree of intimacy
(friend-acquaintance-stranger), but social status and cultural identity.
Table 1 Conversations in order of appearance with total turns and

moves
participants (relationship domain) Turns Moves
1. Jackson + Plains Indian (friends) 4 6
2. Jackson + Junior (friends) 3 3
3. Jackson + Rose of Sharon + Junior + Pawnbroker 35 73
(extended service)
4. Jackson + 3 Aleuts (strangers) 4 5
5. Jackson + Big Boss (acquaintances, unequal status) 24 34
6. Nurse (J’s Grandmother) + Patient (Maori Indian) 20 32
(acquaint, equal status)
7. Jackson + Mary (friends) 24 42
8. Jackson + Bartender (brief service) 5 6
9. Jackson + Irene + Honey Boy, other Indians 45 60
(strangers)
10. Jackson + Bartender (extended service) 17 21
11. Jackson + Williams (acquaintances, unequal 79 186
status)
12. Jackson + 3 Aleuts (acquaintances, equal status) 18 26
13. Jackson + 3 Aleuts + Waitress (extended service) 17 28
14. Jackson + Pawnbroker (extended service) 23 31
Total 318 553
Given the above dialogues and the move analysis which was
performed, two main methods of comparison were used to reveal
stylistic tendencies. The first method examines the types of moves
present in dialogues grouped according to the relationship domains.
This analysis provides an overview of the interaction from a socio-
linguistic perspective with comments like: in conversations between
friends, about 20% of the moves were non-verbal whereas in
conversations between acquaintances of unequal status, only 6% of
the moves were non-verbal.
Chart 1 indicates the amount of ‘interaction’ between speakers as
expressed by the total moves per turn. If each turn is composed of
only one move, speaker change is more rapid – a constant back and
forth interaction with each contribution serving only one main

purpose; thus, the conversation is characterized as highly interactive.
The presence of multiple moves per turn, however, signals less
interaction in this sense. As is shown in Chart 1, conversations
between strangers were more interactive in this respect (1.30 M/T)
than those with acquaintances of unequal status (2.14 M/T) with the
other domains falling in between. The figures in Chart 1 were
calculated by adding the total moves and total turns within a domain
(rather than averaging the M/T of each dialogue within a domain
which would have yielded slightly different results but not effected the
overall trend).
Chart 1 Interaction by relationship domain (measured in moves/turn)
2.40
2.14
2.20
2.00
1.80 1.65 1.64
1.53
1.60
1.40 1.30
1.20
1.00
0.80
0.60
0.40
0.20
0.00
friends strangers acquaintances, service acquaintances,
equal status unequal status
Chart 2 Non-verbal moves as percentages of total moves in each

domain
100.00%
90.00%
80.00%
70.00%
60.00%
50.00%
40.00% 32.31%
30.00%
19.61%
20.00% 12.07% 12.58%
10.00% 5.91%
0.00%
friends strangers acquaintances, service acquaintances,
equal status unequal status
Chart 2 highlights the presence of non-verbal moves in the dialogues

(i.e., functional discourse moves signaled non-verbally). The degree of
non-verbal interaction varied inversely to the amount of interaction, as
shown in Chart 1. Jackson’s interactions with friends and strangers
contain much more non-verbal communication (19.61% and 32.31%
respectively) than his conversations with people of higher status
(5.91%). These differences in communicative style are influenced not
only by social distance and power hierarchies but also by cultural
norms. The relationship domains of friends, strangers, and
acquaintances for Jackson are composed of other Native Americans
(and a Korean American), whereas the Boss and Officer Williams are
part of mainstream Anglo-American culture.
Table 2 Discourse function of moves by relationship domain (using
domain totals)
Init. Resp. Textual Delay total Cont
&Cont &Cont # # # #
# % # % % % % %
friends 9 38 4 0 51 12
17.65% 74.51% 7.84% 0% 100% 23.53%
strangers 13 45 7 0 65 7
20.00% 69.23% 10.77% 0% 100% 10.77%
acquaint., 20 34 1 3 58 14
equal status 34.48 % 58.62% 1.72% 5.17% 100% 24.14%
service 42 102 8 7 159 33
26.42 % 64.15% 5.03% 4.40% 100% 20.75%
54 154 9 3 220 92
acquaint.,
unequal 24.55 % 70.00% 4.09% 1.36% 100% 41.82%
status
&Cont = Continuations of Initiating and Responding moves

Table 2 presents a summary of move functions in the five domains.
Counting the amount of initial and responding moves and their
continuations provides an indication of how frequently speakers
exerted themselves to introduce new topics as opposed to playing
responding roles. Interactions between friends and strangers (who
were Native Americans) had fewer initiations in relation to responses
than the other domains. Acquaintances of unequal status had many
responding moves due partly to the extremely high percentage of
continuing moves which indicate less speaker change than in the other
domains. Conversations between strangers and friends were also
similar in their amount of textual (turn managing) moves and lack of
delay moves.
Due to the importance of social status and culture in the dialogues of
the story, the subsequent analysis combines the data for friends,
strangers, and acquaintances of equal status (dialogues 1, 2, 4, 7, 9, 12
– hereafter referred to as the ‘friends’ domain) for comparison to the
acquaintances of unequal status domain (dialogues 5 and 11). Though
data from the service encounters (dialogues 3, 8, 10, 13, 14) is also
interesting, they will not be addressed in this paper.
Likewise, the anomalous dialogue 6 will only be briefly addressed
here. This conversation (the only one not including Jackson) is
actually Jackson’s report of a conversation between his Grandmother
who was a WWII nurse and a Maori (New Zealander) wounded
soldier who she cared for. Jackson’s Grandmother had reported this
conversation to Jackson many times in his youth. The Native
American emphasis on oral cultural transmission is displayed here as
Jackson recreates the conversation for the reader. Given the situation,
a nurse and a severely wounded soldier who is taking morphine, it is
no surprise that there are several short threads of conversation, thus,
many initiations and continuations as the nurse and patient discuss
various topics.
The second main approach to the data is to look at which characters
produced the moves within the different domains. In this case there
are, likewise, two different ways of approaching the data, here termed:
1) inter-character analysis and 2) intra-character analysis. Inter-
character study examines how the moves in a dialogue or domain are
distributed in order to describe how characters interact with each
other. Looking at discourse moves in this way allows us to make
statements such as: in conversations between friends, 77.78% of all
the Initiating moves (and their Continuations) were made by Jackson
whereas in conversations with acquaintances of higher status Jackson
made only 11.11% of these moves.
On the other hand, an intra-character perspective looks at the
distribution of a specific character’s moves in order to describe his or
her interactional choices. Thus, we can make statements such as: in
conversations with friends, 33% of Jackson’s moves are Initiations,

but in conversations with acquaintances of higher status, only 5% of
Jackson’s moves are Initiations. In this case both views demonstrate
that Jackson takes a leading role when conversing with friends, but
with acquaintances of higher status, he assumes the opposite role, that
of responder.
Table 3 is the first of eight tables that compare Jackson’s
conversational behavior to that of the other characters in the different
domains. Tables 3, 4, and 5 provide an inter-character view which
indicates who contributed the moves analyzed in a domain. In other
words, of all of the Responding moves (and their Continuations)
between friends, Jackson made 34.65% with others producing the
remaining 65.35%. As with Initiations, this figure is reversed in the
domain of acquaintances of unequal status. This view of the data
highlights the role that speakers play as either initiators or responders
in conversations.
Table 3 Inter-character comparison: Initiating and Responding moves
Initiate &Cont Respond &Cont
Jackson others total Jackson others total
friends 77.78% 22.22% 100% 34.65% 65.35% 100%
acqaint. 11.11% 88.89% 100% 65.58% 34.42% 100%
unequal
status
Table 4 Inter-character comparison: Continuing moves

Continue
Jackson others total
friends 43.48% 56.52% 100%
acquaintances, unequal status 58.70% 41.30% 100%
It is notable though that if Continuing moves are counted as a separate

type of move (Table 4), Jackson does less than half of the Continuing
with friends but more than half with his former employer and Officer
Williams. Thus, in both domains, the person who plays the role of
initiator of interactions allows the responder to continue moves more
than he does. In all of the non-service dialogues that Alexie presents
in this story, most conversations are controlled by one person, but that
person listens sympathetically to the other speaker as reflected in the
discourse moves shown in Tables 3 and 4.
Table 5 examines the moves that serve functions other than Initiating
and Responding. Textual (or turn manipulating/controlling) moves,
Delay moves, and the amount of moves expressed non-verbally also
describe the interaction within different domains. Out of all the
Textual moves between friends, Jackson made 66.67% while others
made the remaining 33.33%, but with acquaintances of unequal status,
others made the majority of these moves. This is in accord with the
variation shown above between the roles of initiator and responder.
Table 5 Inter-character comparison: Other discourse functions
Textual Delay Non-verbal
Jackson others Jackson other Jackson others
friends 66.67% 33.33% 0% 100% 26.47% 73.53%
acqaint. 44.44% 55.56% 0% 100% 23.08% 76.92%
unequal
status
More noteworthy is the fact that Jackson never used a Delay move
whereas other characters did use them, and in both domains Jackson
made only about one-fourth of the non-verbal moves. This is one
method by which Alexie portrays Jackson as an assured, quick-
thinking, verbally-skilled conversationalist – qualities that are also
explicitly stated by Officer Williams in their conversation.
Tables 6 to 9, offer a complementary intra-character analysis by
giving the percentage of each discourse function used by the
characters relative to their own repertoire of moves. Of all of
Jackson’s moves with friends, 32.81% are Initiations and their
Continuations whereas with acquaintances of unequal status only
5.41% of his moves are of this type.
Table 6 Intra-character comparison: Jackson’s total moves in two

domains
Jackson
I&C R&C Textual Delay total
friends 32.81% 54.69% 12.50% 0% 100%
acquaintances, 5.41% 90.99% 3.60% 0% 100%
unequal status
Table 7 Intra-character comparison: Jackson’s continuations and non-

verbal moves
Jackson
Cont Non-verbal
friends 15.63% 14.06%
acquaintances, unequal status 48.65% 2.70%
Table 8 Intra-character comparison: Other’s total moves in two

domains
Others
I &Cont R &Cont Textual Delay total
friends 7.59% 83.54% 5.06% 3.80% 100%
acquaintances, 44.04% 48.62% 4.59% 2.57% 100%
unequal status
Table 9 Intra-character comparison: Other’s continuations and non-

verbal moves
Others
Cont Non-verbal
friends 16.46% 31.65%
acquaintances, unequal status 34.86% 9.17%
The figures in these tables are meant to provide an alternative example

of how the data from a move analysis can be used to describe the
stylistic tendencies present in dialogue. This view makes it easier to
see similarities and differences in the conversational behavior of the
people Jackson speaks with. The percentage of Textual and Delay
moves are similar among ‘friends’ and ‘acquaintances’, but the

characters Jackson speaks with play very different initiator and
responder roles and display very different use of non-verbal
interaction (as also shown above), depending on their culture and
status.
Finally, Table 10 focuses on Responding moves and their
continuations by dividing them between Supporting and Confronting
types (and their continuations respectively). The results reveal that in
all domains, and for all speakers (except for dialogue 6) there is a
striking consistency of characters’ ratios of Supporting to Confronting
moves: approximately 70% supporting to 30% confronting.
Table 10 Intra-character comparison: supporting and confronting
moves
Jackson Others
Support Confront Support Confront
friends 71.43% 28.57% 72.73% 27.27%
service 70.45% 29.55% 72.41% 27.59%
acquaintances, unequal 68.32% 31.68% 71.70% 28.30%
status
Nurse Patient
(J’s Grandmother) (Maori)
dialogue #6 90.91% 9.09% 100% 0%
7. Discussion
The overview given in Table 1 shows that the participants and
situations of the dialogues serve as a narrative structuring device. The
story begins with talk between friends, moves to the conflict-raising
interaction with the pawnbroker, then to the three Aleut Indians, and a
medium length dialogue with the sympathetic Big Boss. The central
conversation differs as it is reported by Jackson (an embedded
discourse level). This dialogue in which two marginalized people
commiserate about their individual plights and their peoples’
relationship with white people serves as the thematic center of the
story. From there, conversations between friends are interspersed with
service encounters until the long dialogue between Jackson and
Officer Williams which mirrors and extends the earlier conversation
with the Boss. The reappearance of the three Aleuts signals the end of
the story which comes full-circle as Jackson meets with the
pawnbroker.
Though further analysis of the move data into more specific categories
can reveal more detailed interactional tendencies, the results of the
discourse move analysis from a sociolinguistic perspective reveals
distinct stylistic tendencies. To begin with, the amount and type of
interaction between speakers is more influenced by cultural and status
hierarchies than degree of intimacy. Though Jackson is familiar with
his former boss and with Officer Williams, they are members of
white, Anglo-American middle class culture. They dominate the
conversations with Jackson by making far more initial moves and
performing more turn controlling actions. It may also be said that
Jackson defers to these men as he assumes the role of responder in the
conversations – this can also be seen as one of his strategies for
securing the aid of those with more social and economic power.
However, the Boss and Williams are sympathetic to Jackson, as
shown by their many supportive responses and the frequency with
which they allow Jackson to continue his responses to their initiations.
Furthermore, the relatively high number of moves per turn and the
conversely minimal degree of non-verbal interaction are typical of
what Hall (1990) calls Low Context communication. Anglo-American
culture relies heavily on verbal clarification rather than non-verbal
cues and cultural assumptions. High Context communication,
however, which is more common in Eastern cultures and mono-
cultural orientations, features greater use of non-verbal messages and
presupposed meanings. This orientation can clearly be seen in the
greater amount of non-verbal moves and lower moves per turn when
Jackson interacts with other Native Americans whether they are
friends, acquaintances, or strangers.
Examination of Jackson as the main character of the story reveals that
his conversational style with other Native Americans (and the Korean
shop clerk) is much different from the other dialogues. In these
situations, he plays the role of initiator, makes the most turn
controlling moves, and allows others to continue their responses to
him. However it is also important that across all domains Jackson
displays a much lower percentage of non-verbal moves in comparison
to other characters. This portrays Jackson as a more verbal rather than
physical character which is in keeping with his role as the narrator and
his interest in stories and metaphysical concerns. Likewise, the fact
that Jackson never pauses before responding (Delay moves, which
almost all other characters display) demonstrates his aforementioned
verbal skill in conversations.
In this respect, Jackson can be seen as using a communicative style
closer to that of typical Anglo-American Low Context culture. Most
of the other Native Americans in the story never speak with white
characters; Jackson, who does, serves as a bridge between cultures
due, in part, to his ability to communicate effectively with white,
Anglo-Americans of higher social status. From a larger perspective,
and considering that most of the characterizations ascribed to Jackson
have also been ascribed to Sherman Alexie, this story can be seen as
part of a cross-cultural discourse in which Native American themes
presented by a Native American voice are communicated in a way that
makes them receptive (via a short story in mainstream publications) to
the Anglo-American and multi-ethnic reading cultures in the U.S.
Because the results above deal with represented conversations, the
data can be viewed within the discourse world of the characters as
description of stylistic tendencies that reflect characters’ personalities.
On a different level, the data can be used to explain how the author
used dialogue for characterization, or to speculate about how readers
interpret characters based on their conversational interactions. Such is
the realm of stylistics – to make empirical sense of what characters,
authors, and readers do both consciously and unconsciously. The lack
of studies of discoursal interaction in dialogue implies that the
importance of this level of language is generally unacknowledged, yet
the results of this study demonstrate that how characters ‘move’
together contributes to what readers and writers of fiction do with
dialogue.
In ‘What You Pawn, I Will Redeem’ the characters, especially the
main character, move differently depending on who they are talking
with, yet one commonality across dialogues in the story is the high
percentage of positive as opposed to confronting responses. There are
no heated arguments or debates in this story. One cause of this, it can
be reasoned, is the influence of the central dialogue, number 6. The
memory of Jackson’s Grandmother’s story of her interaction with the
wounded Maori solider, with its almost completely supportive
responses, is the thematic center of the story and a model of

conversational interaction that shapes Jackson’s discourse style.
8. Conclusion
Not all short story authors and novelists use dialogue to the extent that
Alexie does; nonetheless, most prose fiction writers do rely on
represented conversations, and one purpose is to foreground character
relationships. This paper demonstrates that this function of dialogue
can be explored with insightful results by categorizing the participants
of dialogues and performing a discourse move analysis. The counting
of turns and types of moves and comparison of dialogues and
conversational interactants is consistent with other methods of stylistic
description which seek to explicate texts, explain readers’ intuitions,
and support aesthetic judgments using empirical data. The
fundamental sociolinguistic principles and simplified discourse move
schematic applied here are flexible enough to be applied by teachers
and students at almost any level to a story or novel. Alternatively a
more rigorous application by researchers studying a wider body of
works and dialogues by one or more authors from different cultures or
time periods is also possible. The dance of dialogue exists not in a
separate fictional discourse world, but one that is embedded in our
own world, shaped by and a reflection of our own social interactions –
a discourse move analysis captures the structure of this dance so that
we can better understand its role in both worlds.
References
Alexie, Sherman. 2004. ‘What You Pawn, I Will Redeem’ in Moore, Lorrie, and
Katrina Kenison (eds) The Best American Short Stories 2004. Boston: Houghton
Mifflin.
Banks, Wishelle. 1995. ‘Interviews: Alexie finds inspiration in his home, humor in his
family and their life’ The Native Voice. Available at
<http://www.fallsapart.com/art-nv.html> accessed 28 Feb 2005.
Burton, D. 1982. ‘Conversation Pieces’ in Carter, Ronald and Deirdre Burton (eds)
Literary Text and Language Study. London: Edward Arnold.
__. 1980. Dialogue and Discourse: A Sociolinguistic Approach to Modern Drama
Dialogue and Naturally Occurring Conversation. London: Routledge and Kegan
Paul.
Chiarella, Tom. 1998. Writing Dialogue: How to create memorable voices and
fictional conversations that crackle with wit, tension and nuance. Cincinnati:
Story Press.
Eggins, Suzanne, and Diana Slade. 1997. Analyzing Casual Conversation. London:
Cassell.
Fast, Robin Riley. 1999. The Heart as a Drum: Continuance and Resistance in
American Indian Poetry. Chicago: University of Michigan Press.
Fludernik, Monika. 1993. The Fictions of Language and the Languages of Fiction.
London: Routledge.
Hall, E. T., and M. R. Hall. 1990. Understanding Cultural Differences. Yarmouth:
Intercultural Press Inc.
Kempton, Gloria. 2004. Dialogue: Techniques and exercises for crafting effective
dialogue (Write Great Fiction series). Cincinnati: F & W Publications.
Leech, Geoffrey N., and Michael H. Short. 1981. Style in Fiction: A Linguistic
Introduction to English Fictional Prose. London: Longman.
Page, Norman. 1988. Speech in the English Novel. 2nd ed. London: Macmillan. (1st ed
1973).
Semino, Elena, Mick Short and Jonathan Culpeper. 1997. ‘Using a corpus to test a
model of speech and thought presentation’ in Poetics 25: 17-43.
Short, Michael. 1996. Exploring the Language of Poems, Plays and Prose. London:
Longman.
Sinclair, J. McH and R.M. Coulthard. 1975. Towards an Analysis of Discourse: The
English Used by Teachers and Pupils. London: Oxford University Press.
Stanbrough, Harvey. 2004. Writing Realistic Dialogue and Flash Fiction.
Albuquerque: Central Avenue Press.
Tabur-Jogi, Helena. 2004. Stereotypical Self-Images of Native Americans in the Novel
Reservation Blues and the Short Story Collections The Lone Ranger and Tonto
Fistfight in Heaven and The Toughest Indian in the World by Sherman Alexie.
Unpublished Masters Thesis, University of Tartu.
Tannen, Deborah. 1990. ‘Silence as conflict management in fiction and drama:
Pinter’s Betrayal and a short story, “Great Wits”’ in Grimshaw, Allen (ed.)
Conflict talk: Sociolinguistic investigations of arguments in conversations.
Toolan, Michael. 2000. ‘“What makes you think you exist?”: A speech move
schematic and its application to Pinter’s The Birthday Party’ in Journal of
Pragmatics 32: 177-201.
__. 1998. ‘The give and take of talk, and Caryl Churchill’s Cloud Nine’ in Culpeper,
J., M. Short and P. Verdonk (eds) Exploring the Language of Drama: From Text
to Context. London: Routledge: 142-160.
__. 1990. The Stylistics of Fiction. London: Routledge.
__. 1985. ‘Analyzing fictional dialogue’ in Language and Communication 5(3): 193-
206.
Turco, Lewis. 2004. The Book of Dialogue. How to write effective conversation in
fiction, screenplays, drama and poetry. Lebanon: University Press of New
England.
__. 1991. Dialogue (Elements of Fiction Writing): A Socratic dialogue on the art of
writing dialogue in fiction. London: Robinson Publishing. (originally pub. 1989.
Cincinnati: Writer’s digest books).
Stylistic Analysis of Discoursal Identity Formation in
Biographic Interviews with Two Senior Teachers of
English in Hungary
Judit Zerkowitz
Eötvös Loránd University
Budapest, Hungary
Abstract
The issue of professional identity formation in biographic discourse is of special
interest when both the socio-political and professional contexts undergo rapid change
– as has been the case in Hungary during the past two decades. Being a seasoned
language teacher I came to realise that not only the methodology of teaching but the
nature of my subject, English, has radically changed since I started out in the sixties,
and I thus became interested in whether colleagues of the same vintage have also
experienced speedy adjustments between their assigned and claimed identities. I made
a number of biographic interviews, observed some classes, read publications of the
interviewed colleagues and from the multiple sources, multiple texts, observing the
requirements of qualitative research, attempted to draft a picture of professional
identity development in the Hungarian historical climate of change. I analysed the
transcripts stylistically as well and this paper concentrates on a stylistic analysis of
two extracts, each reflecting a different personal style of professional identity
formation.
Keywords: discoursal identity formation; professional identity; agency; point of view;
stylistic analysis.
1. Historical background
Today, as I am beginning to write up the talk I gave in Joensuu, is the
23rd October, 2006, a holiday in Hungary, this being the 50th
anniversary of the 1956 revolution. Teaching English fifty years ago
was almost an act of defiance, as English was considered an
imperialistic language. As late as in the seventies a methodology book
(Banó, I. 1973) divided languages into two groups: Russian was to be
taught with integrative motivation, and the western languages with
instrumental motivation. Although Russian was the compulsory
language in schools, many people studied English in private, and
gradually, from being hardly tolerated, English became the market
leader of foreign languages (Enyedi, Medgyes 1998).
334 Judit Zerkowitz
In the sixties, when I chose to become a teacher of English, the

message of English to me was literature, culture, Churchill, fair-play,
sense of humour and the glamour of the West. By now, forty years
later, English has lost much of its romantic resonance. It has become a
practical necessity and one cannot help teaching globalism (Jenkins
2006) and a business mentality through it. As far as methodology is
concerned, we are moving from the printed word to living speech and
the flickering screen, from grammar translation to post-
communicative eclecticism. I became curious whether any of my
contemporaries had similar impressions of a historical change from
idealism to utility and what impact the marked changes had on their
identities.
1.1. Definition of terms

1.1.1 Language teacher identity
Ever since classroom research revealed that classrooms are complex
places where there is no direct causal relation between methods and
results (Allwright 2005) and it came to light that teachers are also
people, interest arose in teacher beliefs, (Borg 2003, Elbaz 1990,
Pajares 1992, Wood 1996), biography (Kelchtermans, 1993) and
identity (Johnson, 2006 Holmes 2006).
Identity is a category into which a person or thing can be read as
belonging. It is inseparable from language, for it is language that
enables people to form a concept of self and others. (Joseph, 2004,
chapter 2) Neither is style separable from discoursal identity creation.
The formation of professional identity in biographic interview
transcripts reflects how the primarily knowledge-related construction
of inclusionary and exclusionary situatedness is built-up from the
interviewee’s present perspective. Style in speech, not unlike other
signals that come from people’s appearance, tone of voice, unguarded
glances, etc,. keeps changing. Identities however, being part of
representation, are situated transitory coherences that can be
interpreted in a particular context. Discoursal identity exists by
negotiating its transitory coherence, and the textual pattern of these
brief units of willed coherence is what I shall attempt to analyse
stylistically. Identity is socially constructed and people are not free to
take on any identity they choose, (claiming), nor to interpret the
signals coming from the other in any way (ascription). In everyday-
Stylistic Analysis of Discoursal Identity Formation 335
life building on scant information we can go quite far, a practice that

Joseph calls over-reading (2004: 38), since the data on which it is
based cannot adequately support the inferences made.
The stylistic reading of one singled out text cannot escape the
predicament of over-reading, but it can add its contribution to the
multiplicity of approaches of identity capturing. Varghese et al. (2005:
24) recommend an openness to multiple theoretical approaches, a
dialogue across paradigms. In their article they discuss three
approaches based on three new understandings of identity. ‘First,
identity is not a fixed, stable, unitary, and internally coherent
phenomenon but is multiple, shifting, and in conflict’, which, in other
words, says that individuals are intentional beings, not confined once
and for all by circumstances. ‘Second, identity is not context-free but
is crucially related to social, cultural and political context –
interlocutors, institutional settings, etc.’ an aspect of which contextual
feature is the distinction between claimed and assigned identity.
‘Third, identity is constructed, maintained, and negotiated to a
significant extent through language and discourse.’ (Varghese, 2005:
23.) The performance of identity, therefore, greatly hinges on the
impression we want to create through words, the identity we claim.
However, Varghese et al.. make a point about teachers being mostly
invisible to themselves, as it is rather the students who read our
‘image text’, creating a composite portrait of us in interaction
(Varghese: 32). How others see us, however, will influence our own
view of ourselves and certainly we want to manage the impression we
create. We constantly ‘re-play’ the situations we have been in from
our own personal point of view trying to make peace with the figure
we cut during the experience and who we are at present. Conflating
the three approaches one can look for an interplay between contextual
determination and personal freedom in discoursal identity re-plays. By
providing a time, place, topic, purpose, interviewer, one fixed
positioning is assured, as individuals position themselves differently at
different moments and places, in different situations and roles. The
transcript of a biographic interview, therefore, is worth studying for it
contains one moment of discoursal identity creation.
1.1.2 Stylistic analysis

Transcripts have been elicited and studied by discourse analysts,
conversational analysts and interactional sociolinguistics (see in
336 Judit Zerkowitz
Discourse and Identity, 2006) to detect identity creation. What can a

stylistic analysis add to the various research already done on
discoursal identity formation? The former approaches look for the
general in the particular, sieving and comparing a lot of data. Stylistics
is more interested in the particular in its particularity, the one
positioning of identity formation, the analysis of one text, which does
not preclude that stylistics cannot be practiced across texts, but that
the form and content of one textual moment provides plenty to
describe and interpret. Style raises expectation, helps interpretation,
offers a shortcut to claiming and ascribing identity; it is part and
parcel of (see in Leech and Short 1981, chapter 1, the monist, dualist
and pluralist views of style) the fleeting and scant messages we
project and over-read in others.
Since Buffon’s famous verdict that ‘le stile c’est l’homme même,’ no
one can ignore the relation between style and ‘man’, that the style
adopted has a role to play in identifying the speaker in the ongoing
exchange. The speaker’s discoursal identity is created by what and
how is said, when, where and to whom. If we transform the narration
model Rimmon-Kenan (1983) describes and put identity in, we can
visually present that what we are interested in is the constructed
temporary identity of the speaker as it appears in one text.
Real Implied Narrator Narratee Implied Real
speaker Speaker Listener listener
Construct Construct
/identity/ /voice/ /voice/ /identity/
In this paper I shall deal with two aspects of textual analysis: point of
view, and agency. Point of view indicates the perspective of the
speaker who in the biographic interviews is both the focalizer and the
focalized. As there is also a time shift, we deal with the special case of
the double perspective, for in biographic interviews the interviewee
will present their younger self from the professional maturity of the
present. The speakers are certainly both ideologically and
psychologically involved in what they relate about their own lives.
Even in literature, of the four aspects of point of view, space, time,
ideology, psychology, Simpson considers the last, the ‘slant’, the way
in which ‘events are mediated through the consciousness of the teller’
(Simpson, 1993: 11) to be the most important. Indicators of point of

view may reveal the attitude of the interviewee to the present and past
relation between context and free will, to the present configuration of
allegiances. Context determined vs free will leads us to agency,
names, pronouns, the subjects of the finite and non-finite verbs used,
how the speaker presents action.
For ease of handling, I put the transcripts in stanzas and lines,
following the model of Gee (1996: 94) ‘Lines are usually clauses
(simple sentences); stanzas are set of lines about a single minimal
topic, organized rhythmically and syntactically so as to hang together
in a particularly tight way.’ I noted carefully the rhythm of the talk, a
longer silence before a stanza and when the pause is just perceptible
only a new line can begin. Great emphasis is indicated with capital
letters.
2. Analysis
Extract one: only a girl
1 Actually
2 my brothers have all done very well,
3 er, my, they,
4 but THAT’s an interesting thing too,

5 as it’s,
6 in terms of education,
7 that’s typical of the class system at, at home,
8 er, in England,
9 that my BROthers
10 were all put down for a fairly,
11 you know,
12 elite boarding school,
13 which my father had been to.
14 Almost at birth,
15 as of birthright,
16 they wanted them to get into it,
17 and they were prepared to pay for them,
18 whereas I,
19 I would have gone to the grammar school,
338 Judit Zerkowitz
20 if I passed the eleven plus,

21 but then I got a second rate girls’ school,
22 because they could not afford it
23 But they could afford my brothers to go to school,

24 I mean,
25 because of the prioritizing was like that,
26 because a girl,
27 at that point,
28 which is logical,
29 I cannot blame them for it,
30 because I was supposed to marry and be a housewife
31and what’s the point investing huge amounts of education?
32(laugh)
32 Er,
33 but my brothers,
34 in fact did,
35 I mean,
36 all the three of them went to university.”
In this extract the contrast between the ‘birthright’ of the brothers and
the only sister is explained in class terms: boys of their class were
expected to don the old school tie and girls to marry. For the girls
good education was not a social requirement. For the boys attending
the socially desirable public school was necessary, but the brothers did
very well, for they went on to tertiary education at a time when
university degrees only just started to become expected of the men of
their class. Elsewhere CB said that of her schoolmates only one other
girl has had a career, while everybody else stayed at home, brought up
children, at times drifted into voluntary work for cancer wards or
churches. In contrast, five years earlier all the girls remained
housewives and five years later the majority worked outside the home.
The interviewee’s generation was a transitory generation in relation to
education and work in general, and in gender role terms.
The text is framed by extolling the merits of the brothers, while in the
bracket there appears a thought, an aside, indicated by the emphatic
but THAT. ‘THAT’ can be unpacked as the difference in prioritizing
the investment into the education of boys to that of girls, on the
grounds detailed above.’ The textual pattern hinges on repetitions of

but, THAT, it:
but THAT (4)
, as it’s (5), that’s typical, (7)
that my BROthers, (9)whereas I, I, (18)
but then I,(21) but they could, (23)
it,(16) it (22) it (29)
but my brothers (33)
Expressing contrast, CB starts the digression from the praise of the
brothers with the first ‘but’, (4) moving on to the comment on their
differential educational birthrights ’that’(7) was typical. Now she
states the contrast, saying ’that my BROthers (9) - whereas I’, (18)
and further qualifies the contrast with the money issue: when it came
to paying for a private school the parents felt they could not afford to
pay a lot for the first born girl, while paying the expensive fees after
the boys was a matter of course, see ‘but then I,(21) but they
could’(23). The ‘its’ of the passage, as it’s, (5) get them into it (16)
afford it, (22) blame them for it, (29) refer back to the starting
THAT,(4) emphasizing CB’s angle to the contrast. I cannot blame
them for it exonerates the parents for ‘it’, the THAT that was so
typical, because, and suddenly the only question appears: What’s the
point in investing huge amounts of education? The question deviates
from the series of statements leading up to it, and to some extent from
the context as well. It can be read both as what’s the point in investing
huge amounts of education in this particular case, or in general. The
question is followed by laughter, an indication of emotional
involvement, which foregrounds the slight ironic distance between a
possibly earnest vehicle and an obviously rhetorical tenor in the past
context. The problem is more revealed than resolved, and this done,
the last ‘but my brothers’ (33) leads us back to the main line of
narration, the embedding is over.
CB’s point of view in this extract is present to past, here to there,
conscious of both the ideological and psychological planes. Her style
is colloquial, as she talks to me, an old colleague, and her
contextualisation signals point from the deictic centre of here and now
to the past by the colloquial ‘that’ and ‘it, well aware of what ‘that’
and ‘it’ meant then and why she will not blame her parents from the
340 Judit Zerkowitz
perspective of today, although the fact that she was hurt by their
calculations is very clear from her present perspective.
Concerning agency, the parents and the brothers dominate until (18)
where ‘I’ appears for the first time, following the only ‘whereas’ in
the text, and starting a new stanza as the break in the rhythm of speech
was long. Contrasting ‘I’ with the family is foregrounded here, but the
strongest foregrounding occurs in (29), in ‘I cannot blame them for it’,
as it does not relate to the plans of the parents with her, on the
contrary, it belongs to what she thinks now of their action. This
sentence stands out as the only direct commentary and as a clear
statement of her suppressed resentment. Connecting ‘THAT’ with ‘it’
against the structure of contrasts reveals here the underlying topic:
access to preparing for her future profession was made difficult, and
yet the blame is more on the context than on the parents’ action. Her
own agency is backgrounded, for the situation would not have been
much better had she passed the eleven plus with flying colours. It is
interesting to note that she, modestly, narrates the whole digression
not for her own personal sake, but because her predicament is ‘typical
of the English class system’ (7) as if the system were the primary
agent, hence the general interest, and not her parents, which was her
particular plight. This modesty, though, is not surprising, given the
British preference of understatement and the female self-deprecating
way of talking.
2.2
Extract two: one of the boys
1 My professional life
2 all starts with me being 4 or 5 year old kid,
3 whose parents were clever enough to send him to an English
speaking nursery school,
4 which in fact wasn’t an English speaking nursery school
5 because the teacher there was a lady who was originally German
6 and she also spoke good Hungarian,
7 but I don’t know whether her English was as good as her German
and Hungarian.
8 We used to learn a lot of nursery songs
9 but normally the language of communication was Hungarian.
10 But perhaps it all started with this
11 because it must have made an incredible imprint,

12 that at the age of 5 I began to learn in a very spontaneous
manner, English
13 Incidentally,
14 one of the other boys was Nádasdy Ádám,
15 so our friendship dates back
16 to at least 50 years,
17 more than 50 years in fact.
18 Then after nursery school,

19 my parents insisted that I carry on with English
20 and they hired various language teachers,
21 English language teachers,
22 but I never learned English at school,
23 so it was all through private tutoring.
24 And I mean the second language that I came in contact with at

the age of 10,
25 when like any other schoolboy in Hungary
26 I had to learn Russian.
27 Funnily enough
28 my Russian teacher left Hungary in 1956
29 and went to live in Australia.
30 But Russian remained.

31 And when I went to what by Hungarian standards is secondary
school at the age of 15,
32 my father insisted that I learn not only Latin,
33 the dead language,
34 but also Russian in a special Russian class,
35 so I had something like 6 or 7 Russian lessons per week
36 which helped me master to some extent the Russian language,
37 so it was pretty obvious

38 when I reached school leaving age at the age of 18,
39 I decided to major in English and Russian.
342 Judit Zerkowitz
40 Despite the fact that my last private English tutor also taught to
me German,
41 he was also a German native speaker
42 and he would have liked me to study German to major in
German rather than Russian,
43 but then I still decided to take Russian and English.
44 During my university years, not for a moment did I think

45 that I was going to become an English teacher:
46 my plans were far more highfalutin,
47 to become a translator and a creative writer myself,
48 which never materialised.
This non-native male speaker of English has a more formal style than
the first native female speaker. The stanzas narrate the topic, the
beginning of the professional career, step by step. The sequence, with
two digressions, resembles the density and order one expects in
written paragraphs. The interviewee distances himself from his past
identity by talking about himself in the third person, being a child
whose professional life started by the parents sending him to an
English speaking nursery (1-3). The clever parents are followed by a
number of teachers, (7) (20), (28), (40) all contributing to MP’s
development according to their languages and lights. The father
insisted (19) (32) that he should get private tutoring, the German
teacher would have liked him to do German (41), he had to study
Russian at school (26) show the educational context of his early years.
This context influenced his decision to take Russian and English (39)
but in his heart of hearts it was not teaching he was interested in but
translating or creative writing (47). Other people’s designs upon him
are contrasted with his free will and so in the last stanza he distances
himself again from his former self by disclosing his ‘highfalutin’ plans
to become a creative writer or translator. People and the context are
acting and he is only reacting. In Greimas’s subordination of
characters to the functions they serve (quoted in Rimmon Kenan,
1983: 35) MP has many helpers (parents, friend, teachers) and his
opponents are only some political aspects of the context. Key ideas are
foregrounded by repetition, such as it all started (2), (10), the parents
insisting (18), (31), contrasts expressed by the word but (7), (9), (22)
(30)(34)(43) and the foregrounded ‘not for a moment did I think’ (44)
that leads up to the final denial of the obvious choice (46), with an
ironic afterthought (48) that the highfalutin plans never materialised.

The second speaker has a more powerful position in society than the
first one as his parents invested in his education and private tutoring,
which assured access to his future profession. The present point of
view is in every respect far above the past: the time-lag is stated to be
over 50 years, (16) (17) self-irony in the last stanza shows that no face
threat is involved, digressions are included, see stanzas 2 and 5, and
last but not least, he deems his parents have been ‘clever enough’ (3),
a jocular praise from a self-assured position. The quoted passage is
framed by how M thinks his professional life ‘must have started’ and
‘not for a moment’ did he think he would become an English teacher,
having highfalutin plans, which would by definition exclude the
teaching profession. Apparently he came down to teaching, while the
first speaker, C, worked her way up to teaching. I have already
pointed out C’s modesty in backgrounding her importance in her own
life story, as if her story were mainly interesting as an example of
class customs, MP’s narration is ego-centred. The grown-up man,
from the height of the present, looks back on his past and benignly
smiles even at his younger self, and of course the times and the people
who contributed to his development. Both speakers play on exclusion
and inclusion, but one does it with feminine charm, relegating herself
into the background and the other with masculine charm, placing
himself centre-stage.
As MP also wrote about this early period of his life, to compare the
style of the spoken identity creation with the written one, I quote a
short passage:
At the time I reached nursery school age, Communists had
seized power, nationalised ‘bourgeois’ property and banned all
foreign languages other than Russian. Nevertheless, I was
fortunate enough to attend a private nursery school in Budapest
run by Aunt Ida. It boggles the mind how she managed to stay
private, and, on top of that, to teach English, the chief language
of the imperialists. I never learnt English at school. Instead, my
parents paid for private lessons. The first were with Aunt
Franciska, a soft-spoken former nun, whose contribution to my
education ended when, on entering the children’s room, my
mother saw me turning Hungarian somersaults instead of
practising English grammar. My second teacher was Aunt Ila,
344 Judit Zerkowitz
who had lived the better part of her life in India married to an
Englishman. Once she presented our family with a mah-jong set
from China, which I still treasure though I can’t play.
However, the teacher I remember most vividly is Privatdozent
Dr Koncz, who had lost his university job by being stigmatised
as a relic of the old regime. He was a severe and unhappy man
in his fifties. When I rang the doorbell his wife would come and
open the door. As I sat down, I heard Dr Koncz cleaning his
teeth in the adjacent bathroom. A minute later he would join
me, …
This text highlights what MP remembers of his first English teachers.
The description is again in a gently ironic tone, praising the parents
and smiling at the most salient episodic memories of past teachers, the
private nursery, Hungarian somersaults, mah-jong set, and pre-class
teeth cleaning. MP was fortunate, so many people were trying to
outweigh the communist influence in his life. All the teachers are past
their prime in both the spoken and the written version, which may
explain that first MP did not wish to become a teacher. The mah-jong
set he still cherishes, though he can’t play – the memory is still
pleasant, yet the game itself is no longer timely. The contrast between
context and intentional behaviour is indicated with ‘but’ in speech,
‘nevertheless’ in writing, a difference only in the choice of the word.
Reliance on shared knowledge is negligible as a reader – writer type
of distance is maintained both in the written and the oral texts.
The two extracts examined can only give an impression, the identity
of a moment as read by the interlocutor. Wagner and Wodak (2006) in
their article entitled Performing success: Identifying strategies of self-
presentation in women’s biographic narratives establish through
deconstructing strategies of self presentation general features of what
successful women attribute their success to. The stylistic comments of
the above two extracts, although quite revealing of the social, the
gender, the historic situation, do not attempt to draw general
conclusions about types of teacher identity, but hope to describe
textual patterns, such as ways of negotiating the contrast between
context and intention, of two particular identity-projecting biographic
discourses.
3. Conclusion
The various views of colleagues about their career development in our
historically-changing context may yield characteristic clusters of
answers. The actual words of the speakers, however, in our case two
senior teachers of English, a British woman and a Hungarian man, are
also a rich source of data. In the present happy state of stylistics,
analysing the individual texts themselves of such instances of identity
projection has a contribution to make to identity research.
References
Allwright, D. 2005. ‘From teaching points to learning opportunities and beyond’ in
TESOL Quarterly 39(1) 9-31.
Banó, I. 1973. Módszertan. Budapest: Tankönyvkadó.
Borg, S. 2003. ‘Teacher cognition in language teaching: a review of researh on what
teachers think, know, believe and do’ in Language Teaching 36: 81-109.
Elbaz, F. 1990. ‘Knowledge and Discourse. The evolution of research on teacher
thinking’ in Day, C., M. Pope and P. Denicolo (eds) Insight into Teachers’
Thinking and Practice. London: Falmer: 1-19.
Enyedi, Á., and P. Medgyes. 1995. ‘Angol nyelvoktatás Közép- és Kelet-Európában a
rendszerváltozás óta’ in Modern Nyelvoktatás 4/23: 1232.
Gee, J. P. 1996. Social Linguistics and Literacies. London: Taylor & Francis, first
edition in 1990.
Holmes, J. 2006. ‘Workplace narratives, professional identity and relational practice’
in De Fina, A., D. Schiffrin and M. Bamberg (eds) Discourse and Identity.
Jenkins, J. 2006. ‘The spread of EIL: a testing time for testers’ ELT journal 60(1) Jan
2006. Oxford: Oxford University Press.
Johnson, G. C. 2006. ‘The discursive construction of teacher identities in a research
interview’ in De Fina, A., D. Schiffrin and M. Bamberg (eds) Discourse and
Identity. Cambridge: Cambridge University Press.
Joseph, J.E. 2004. Language and Identity, national, ethnic and religious. Basingstoke:
Palmer Macmillan.
Kelchtermans, G. 1993. ‘Teacher and their Career Story: A Biographical Perspective
on Professional Development’ in Day, Ch, J. Calderhead and P. Denicolo (eds)
Research on teacher thinking, understanding professional development. London,
Washington D.C.: The Falmer Press: 198-220.
Leech, G., and M. Short. 1981. Style in Fiction. London: Longman.
Pajares, M. F. 1992. ‘Teachers’Beliefs and Educational Research: cleaning Up a
Messy Construct’ in Review of Educational Research Fall 1992 62(3): 307-332.
Rimmon-Kenan, S. 1983. Narrative Fiction, Contemporary Poetics. New Accents,
(gen. ed.) Hawkes, T. London and New York: Routledge.
Simpson, P. 1993. Language, ideology and point of view. London and New York:
Routledge.
Varghese, M et al. 2005. ‘Theorizing Language Teacher Identity: Three Perspectives
and Beyond’ in Journal of Language, Identity and Education 4(1): 21-44.
346 Judit Zerkowitz
Wagner, I., and R. Wodak. 2006. ‘Performing success: identifying strategies of self-
presentation in women’s biographical narratives’ in Discourse and Society.
London: Thousand Oaks, Sage.
Woods, D. 1996. Teacher Cognition in Language Teaching. Beliefs, decision –
making and classroom practice. Cambridge: Cambridge University Press .
PART V
STYLISTICS IN THE CLASSROOM
The Ellipsis of Haiku: The Effects of Poetic Ellipsis1

in the Framework of Relevance Theory
Kyoko Arai
Toyo University, Tokyo
Abstract
In the framework of relevance theory, certain mental effects we experience during
conversations are defined as cognitive effects. These kinds of effects include
obtaining contextual implications and revising one’s present assumptions. Another
kind of effect that is well known is the poetic effect, which assumes the special effects
we experience when a speaker intends her utterance to have multiple weak
implicatures.
The effects of literary texts are often analyzed using this notion, poetic effect.
However, for the very short type of literary texts, such as Japanese short poems,
Haiku and Waka, the value of the shortness cannot be explained by the notion. This
paper introduces a third type, which contrasts with poetic effects, that is, the effects of
‘poetic ellipsis’. By posing the effects, we will be able to distinguish between the
essential value of verse and prose. In other words, we will explain why the shorter
forms of literature sometimes tell us as much as, or more than, the longer expressions.
The main purpose of this paper is to define the special type of ellipsis of Haiku by
using the notion of poetic ellipsis. Haiku is called the literature of ellipsis and it is said
that the essential value of this literature is its shortness.
The first section is an overview of some basic concepts about the meanings of an
utterance (sentence) and what ellipsis is in the framework of relevance theory, along
with a review of the overall attitude of relevance theory towards analyzing literary
text and the well-known concept of poetic effects. In the second section, the new
concept, poetic ellipsis, is introduced as a way of analyzing literary texts from a
different angle, and illustrated by examples of Haiku and Haiku’s ellipsis as they are
explained by the notion. Section three is an examination of English translations of
Japanese Haiku in order to gain insight into Haiku’s special ellipsis, and includes a
348 Kyoko Arai
discussion on how to translate the shortness of Haiku. Finally, the importance of the
quality of poetic ellipsis in Haiku’s ellipsis is explained by using the notion of highly
relevant words.
Keywords: relevance; poetic effects; ellipsis; and weak explicatures.
1. A review of utterance meanings, ellipsis and poetic effects in

relevance theory
1.1. Two kinds of meaning
First of all, in relevance theory the meanings that the hearer recovers
from the speaker’s utterance are categorized below.
(1) Sentence meaning (Logical Form) = the meaning that was
decoded from the literal form of the utterance.
Speaker’s meaning
a. explicature: the meaning that was pragmatically enriched
from the sentence meaning. (Pragmatic enrichment: saturation,
ad-hoc concept construction, and free enrichment)
b. implicature: the meaning that was inferred only pragmatically
from the sentence meaning and the context as a premise.
The sentence meaning is the meaning that the hearer encodes from the
utterance, and the other meaning is the one she draws only by
pragmatic inference. The speaker’s meaning is the main subject of the
study of pragmatics, and it consists of two kinds of meanings,
explicatures and implicatures, as shown above. An explicature is
usually a proposition expressed by an utterance.
1.2. Two types of ellipsis

Based on the categorization of the utterance meanings, the
phenomenon of ellipsis can be defined as follows.
Ellipsis Type A:
The ellipsis of the constituents of an explicature
To omit some constituents of the speaker’s utterance that she assumes
that the hearer can infer in the course of his pragmatic enrichment
(saturation and free enrichment), based on the principle of relevance.
Ellipsis Type B:
The ellipsis of one of the propositions in a series of utterances
To omit some propositions in a series of the speaker’s utterances that
she assumes that the hearer can infer as implicatures in the course of
The Ellipsis of Haiku 349
his inference, based on the principle of relevance.

Type A ellipsis is that which is called syntactic ellipsis in syntax and
other faculties of linguistics. In relevance theory, when we only focus
on the explicature of an utterance, the ellipsis is defined as Type A.
However, when we consider the implicature of an utterance, another
type of ellipsis, the ellipsis of whole sentences in the series of
utterances, is identified as type B. Consider the following examples:
(2) a. Do you like dogs?
b. Yes, I do (like dogs).
(3) Tell me when (I can stop pouring the liquid).
(4) Get up, John! It’s a fine day. (We can go fishing.) Get ready.
The sentence, (2b) is generally called a syntactic ellipsis because the
speaker expects the hearer to recover omitted parts of the utterance by
saturation, that is, the knowledge of syntax. If the hearer tries to
recover the omitted parts of the sentence in example (3) by only his
syntactic knowledge, there will be a limitation. Therefore, the hearer
will use another kind of inference, called free enrichment, to recover
the explicature. On the other hand, in example (4), the whole sentence
‘we can go fishing’ was omitted. The hearer recovers this sentence as
an implicature. This is type B ellipsis.
Poetic effects are produced when the speaker (writer) uses a special
case of type B ellipsis, on the other hand, the effects of poetic ellipsis
are generated in a special case of type A ellipsis.
1.3 Poetic effects

Before we discuss poetic effects, let us review the basic attitude of
relevance theory toward analyzing literary texts. The early reference
to rhetorical expression in relevance theory is in Wilson and Sperber
(1990) as follows:
(5) From a cognitive view, the teaching of rhetoric turns out to
have been less a source of self-understanding than a source of
self-misunderstanding. Because rhetoric effects are achieved in
the normal course of the ever-present pursuit of relevance, the
institution of rhetoric as a separate subject for teaching and
study defeats its avowed purpose. (Wilson and Sperber 1990:
155)
350 Kyoko Arai
This was a rather shocking remark for those studying rhetorical

expressions, and stimulated many scholars to reconsider the literary
values and effects from the cognitive viewpoint. Many figurative
expressions, such as metaphor, simile and irony, had been examined in
the framework of relevance theory by using the concepts of ad-hoc
concept constructions, echoic usage of lexical items, and others.
Furthermore, this movement influenced the field of literary criticism.
The analysis of literary works became very active, especially after the
new concept, ‘poetic effects’, was introduced by Wilson and Sperber
(1997).
(6) Let us give the name of poetic effects to the particular
effects of an utterance which achieves most of its relevance
through a wide range of weak implicature. (Wilson and Sperber
1997: 222)
(7) Poetic effects, we claim, result from the accessing of a large
array of very weak implicatures in the otherwise ordinary
pursuit of relevance. Stylistic differences are just difference in
the way relevance is achieved. (Wilson and Sperber 1997: 224)
Pilkington (2000) and many other scholars apply poetic effects to
study literary texts, and again, everybody assumes that these effects
are not only for rhetorical expressions, but also for ordinary everyday
ones. Let us view a concrete example of these effects:
(8) The scenery of Walden is on a humble scale, and, though
very beautiful, does not approach to grandeur, nor can it much
concern one who has not long frequented it or lived by its
shore; yet this pond is so remarkable for its depth and purity as
to merit a particular description. It is a clear and deep green
well, half a mile long and a mile and three quarters in
circumference, and contains about sixty-one and a half acres; a
perennial spring in the midst of pine and oak woods, without
any visible inlet or outlet except by the clouds and
evaporation… (‘The Writings of Henry D. Thoreau, Walden’
1971: 175)
In this essay, Thoreau wrote about the many detailed descriptions of
the pond and its surroundings, so that all readers of this can imagine
the scenery very clearly. Although Thoreau did not express any of his
feelings in these passages, the readers, somehow, can understand his
attitude toward the natural beauty of the scenery including the pond.
But why? In the theory of poetic effects it is explained that the
explicature of each sentence in (8) has many weak implicatures and
the reader usually tries to look for them by making an extra effort,
hopefully compensated for by an enhanced appreciation of literature.
Let us consider the following famous Haiku poem by Basho Matsuo:
(9) Furuike ya
Kawazu tobikomu, Mizuno oto (Basho Matsuo)
The old pond:
A frog jumps in, – The sound of the water (Translated by R.H.
Blyth)
This is also the description of the scenery of a pond. It is said that this
Haiku was made in a garden, called Seitou Teien. From only a few
sentences of this Haiku it is said that many implicatures are recovered.
For instance, they are ‘How quiet and calm this area is’, ‘How
beautiful the scenery of the pond is’, ‘I really appreciate the old pond’,
and so on. That there are many weak implicatures, in fact, made Haiku
poems one of the world’s most highly valued literary forms.
Compare this Haiku and the passage of ‘Walden’ in (8). Thoreau’s rich
descriptions make us imagine the exact view of the pond. We can
draw a picture of the scenery, including the pond, easily, according to
the details he wrote. On the other hand, it is more challenging for us to
draw a picture of the scenery from (9), only seventeen syllables of a
poem. Somehow, we can draw that the pond is not so big, perhaps
surrounded by some rocks with moss. We can imagine that the water
is clear and clean, with a small green frog jumping in the water, and so
on. Some people can even draw the surroundings of the pond. The
scenery that the readers of (9) imagine from the Haiku is not much
simpler than the one drawn by the reader of (8). This explains that the
quantity (or length) of the description does not so much matter
regarding the scenery that the reader imagines from the description.
From the viewpoint of relevance theory, these facts can be explained
by the concept of ‘weak explicatures’. This is the key notion in
answering the question, ‘Why do poems sometimes tell us more than
works written in prose?’ Or more simply, ‘Why do we want to write
poems?’
352 Kyoko Arai
2. Poetic Ellipsis
2.1 Ellipsis and weak explicature
In Wilson & Sperber (2000: 247) the strength of explicature was
explained as follows:
(10) Explicature can be weaker or stronger, depending on the
degree of indeterminacy introduced by the inferential aspect of
comprehension. (Wilson and Sperber 2000: 247)
Consider the following cases:
(11) a. [A man sees that smoke is coming out of a window and
says] Fire!
b. [A man stops walking on the street, and says] Fire!
Both utterances have the same logical form. The explicature of (11a)
must be ‘The house is on fire!’ but how about the explicature of (11b)?
The man might have wanted to say ‘I forgot to turn off the gas oven’s
fire!’ Or, the man is a plainclothesman pursuing a criminal with other
policemen and has just found the criminal and ordered, ‘Fire a volley!’
Since the degree of determinacy of the proposition expressed by the
utterance (11b) is quite low, we won’t be able to recover a very strong
single explicature, but many weak candidates of an explicature; in
other words, an utterance like (11b) has many weak explicatures. Let
us compare a proverb and a Haiku:
(12) a. No sweet without sweat
b. Ara Touto,
Aoba Wakaba no,
Hi no Hikari (Basho Matsuo)
Oh, holy,
Green leaves, young leaves,
Light of the sun (Translated by R.H. Blyth)
From the syntax point of view, both examples consist of noun phrases.
The explicature of (12a) can be recovered very easily, as in, ‘We
cannot get sweets without sweat.’ On the other hand, it is not easy to
find a single clear explicature for (12b). We cannot tell what is so
holy, and what is the matter with the leaves and light. There are many
candidates for the explicature of (12b).
(13) a. There are holy green leaves and young leaves under the
light of the sun.
b. I feel very holy to see green and young leaves and the light
of the sun.
Only people who know that this Haiku is about Nikko Toshogu (a
shrine)2 and that Basho wrote it during his famous journey called
‘Okuno-hosomichi’ (Basho’s Narrow Road to a Far Province) may be
able to recover the following explicatures.
(14) a. I am impressed by the holy temple. There are beautiful
green and young leaves under the light of the sun.
b. Nikko (=sunlight) Toshogu is very holy and green and
young leaves are also very beautiful and holy.
Aside from those who know the story behind the Haiku, people cannot
recover a single definite explicature from it. However, even the reader
who does not know anything about the story can still appreciate the
Haiku by imagining beautiful new leaves shining on the trees in the
woods or a churchyard.
For the proverb there should be a single strong explicature that states a
fact or instruction, while Haiku is designed to have many weak
explicatures by omitting many constituents of a sentence.
In the framework of relevance theory, certain mental effects we have
during conversations are defined as cognitive effects. These kinds of
effects include obtaining contextual implications and revising one’s
present assumptions. Another kind of effect that is well known is the
poetic effect, as reviewed in the previous section, which describes the
special effects we experience when a speaker intends her utterance to
have multiple weak implicatures. The effects of poetic ellipsis are the
third type, contrastive effects with poetic effects.
(15) 3 Kinds of Effects
1. Cognitive effects
2. Poetic effects
3. The effects of poetic ellipsis
While poetic effects can be found in many rhetorical forms of ordinary
utterances, poetic ellipsis is also used in not only literary works, but
also in everyday conversation. Poetic ellipsis is where speakers
354 Kyoko Arai
assume multiple weak explicatures to their utterances by omitting

some parts of the sentence.
2.2 Ellipsis in Haiku

In this section, the relationship between the ellipsis of good Haiku
poems in general and poetic ellipsis will be discussed. In Ishihara
(1989) the special type of ellipsis in Haiku is explained:
(16) In the shortest form of poem, Haiku, it is impossible to
describe everything about the object, therefore, a drastic
omission is needed … According to the rhythm prepositional
particles, auxiliary verbs and verbs, which do not influence the
understanding the meaning of the Haiku, are often omitted. This
is an effective method made by this kind of special ellipsis.
(Ishihara 1989: 232)
In pragmatic terms, ‘which do not influence the understanding the
meaning of the Haiku’ means that the reader can recover the
explicature of the Haiku without those words. From the viewpoint of
syntax, it is not allowed to omit these kinds of parts of the sentence.
This is the reason why all Haiku, such as (9), consist of only some
noun phrases.
Moreover, in the first stanza of Haiku (9) the word ‘ya’ is called
‘Kireji’, which indicates the period of a sentence. Kireji is a special
technique that allows Haiku poems to include multiple sentences in
only seventeen syllables. Therefore, this stanza is assumed to be one
sentence, so the Haiku (9) consists of two sentences and we can notice
that there are many words that were omitted. One of the explicatures
of (9) could be as follows:
(17) There is an old pond. I hear the sound that a frog jumps into
the pond.
Consider another example of Haiku.
(18) Kikunoka ya
Naraniha Furuki Hotoketachi (Basho Matsuo)
The scent of chrysanthemum
In Nara, Many old images of Buddha
(Translated by the author of this paper)
This Haiku also consists of two sentences. The explicature of the first
stanza may be ‘The scent of chrysanthemum pervades the air,’ or ‘I
can smell the scent of chrysanthemum.’ The noun phrase can be the
subject or the direct object. From the noun phrase in the second stanza
the reader can also recover ‘There are many images of Buddha,’ or
‘Many images of Buddha are found.’ The parts of the explicatures
aside from the noun phrases that exist in Haiku are recovered by free
enrichment. (See Type A ellipsis) There is no evidence to judge the
syntactic structure of the sentence, so the reader can freely construct
and enrich it until his expectation of optimal relevance is satisfied.
Above all, Haiku’s ellipsis, in relevance theoretical terms, omits many
constituents of the sentence, which the reader can infer in the course
of his pragmatic enrichment (especially free enrichment) based on the
principle of relevance. And, the ellipsis aims at increasing the literary
effects by including many weak explicatures, that is, poetic ellipsis.
The reader consumes his efforts to infer many weak explicatures,
however, the efforts will be compensated by a greater appreciation of
the literature.
3. The translation of the shortness

3.1. Haiku in an English translation
To consider more carefully the poetic ellipsis in Haiku let us look at
some English translations of Haiku. Here, there are two different
translations of the Haiku of (9):
(19) Furuike ya
Kawazu tobikomu, Mizuno oto (Basho Matsuo)
(20) The old pond:
A frog jumps in, –
The sound of the water (R.H. Blyth)
(21) Breaking the silence of an ancient pond
A frog jumps into water –
A deep resonance. (N.Yuasa)
Since there are not definite and indefinite articles in the Japanese
language and in Haiku demonstratives are often omitted, in the
translation of (20) the translator added these words to the Haiku to
make its English sentence syntactically correct and this process can be
called saturation, especially reference assignment. The word ‘furuike’
356 Kyoko Arai
in Japanese can be any old pond, but in the English translation it is

‘the old pond’, the definite pond. Also ‘kawazu’ can mean a frog or
multiple frogs, but again the translator defined it as a single frog. ‘The
sound of water’ in English means that the sound was heard when the
frog jumped in the pond. In the original Haiku, the cause of the sound
is rather ambiguous. In this translation the translator recovered the
explicature only by saturation.
On the other hand, for translation (21), the part ‘Breaking the silence
of’ does not have the semantic representation in the original Haiku.
The translator might have thought the surroundings of the pond must
be very quiet. Also the last part, ‘deep resonance’, is completely due
to the imagination of the translator. The translator of (20) recovered
only its explicature, but the translator of (21) recovered an omitted
proposition, that is, an implicature of the Haiku. The difference
between the translations depends on how familiar with Haiku and
Japanese culture the reader of each translation is. For the beginners of
Haiku, the translator has to give much of the additional information to
the translation. Gutt (2000) defined such additional information as
‘communicative clues’.
(22) One might well argue that the point of preserving stylistic
properties lies not in their intrinsic value, but rather in the fact
that they provide clues that guide the audience to the
interpretation intended by the communicator. We shall refer to
such clues as communicative clues. (Gutt 2000: 134)
Gutt quoted that when Yuasa, the translator of (20), translated the
Haiku he thought ‘the old pond’ was ‘too weak, far too abstract and
general’, and explained as follows:
(23) From the point of view of relevance theory, it is fairly clear
that the problem Yuasa is grappling with here is the fact that the
meaning communicated by a text is not attributed to the
stimulus alone, but results from the interaction between
stimulus and cognitive environment. (Gutt 2000: 139)
The explicature is too weak and too abstract and general, so the
translator adds the communicative clues to the translation in the
course of his pragmatic enrichment. In general, the main issue of
translation is that simply decoding the text and encoding it into the
new language is not enough to convey what the author wants to say to
the foreign reader. To make up for the deficit they use communicative
clues.
Then, how many translators have to add communicative clues for the
text each time? How far or deep do they have to convey the intended
meaning of the author to the reader? Let us discuss this issue in the
case of Haiku in the next section.
3.3. The translation of shortness (the weakness of the explicature)

As mentioned before, when applying the concept of poetic ellipsis, it
can be said that the weakness of the explicature is very important.
Therefore, the more the translator adds the parts of the explicature of
the Haiku, the stronger the explicature becomes. If the translator
writes a single strong explicature, the foreign reader does not need to
do any pragmatic enrichment and all he/she has to do is to recover
implicatures. Compare another set of translations (examples (24) and
(25)) with Haiku (12b), repeated here for ease of comparison:
(12b) Ara touto
Aoba, wakaba no
Hi no hikari (Basho Matsuto)
(24) Oh, holy,
Green leaves, young leaves,
Light of sun. (R.H. Blyth)
(25) Oh holy, hallowed shrine!
How green all the fresh young leaves
In thy bright sunshine! (D. Britton)
In the Haiku of (12b) the place the Haiku refers to was not mentioned.
The translator of (24) changed each word into an English word;
therefore, it is a direct translation. On the other hand, the translator of
(25) knew this Haiku was about Nikko Toshogu (a shrine) so she
added ‘hallowed shrine’ to her translation.
Which translation is better depends on the purpose of the translation,
just as in the comparison between (21) and (22). If the reader wants to
study Basho’s Haiku, the more communicative clues there are, the
more they help his study. In contrast, if the reader wants only to
appreciate the Haiku, the minimum number of communicative clues
are recommended, because the shortness is one of the literary values
358 Kyoko Arai
of Haiku and recovering many weak explicatures must be part of the

appreciation of the literature. Look at the next example:
(26) Suzushisa ya
Kane wo hanaruru
Kane no koe (Buson Yosa)
(27) Coolness
Leaving the bell
The sound of the bell
When the author of this paper read this Haiku for the first time, she
thought it was about a church bell in England. This Haiku reminded
her of the cool air of an English summer morning and the sound of a
church bell from a mile off. In fact, this Haiku was composed during
the Edo era in Japan, and the author would never have intended a
reference to the sound of a church bell in England. However, since
each word of this Haiku is not definitive, the reader can enjoy finding
many weak explicatures by themselves for this Haiku. Whether the
pond in Haiku (9) can be Walden or one of the lakes in the Lake
District, foreign readers can also appreciate the Haiku in different
ways. This should be one of the pleasures of reading Haiku and its
value of shortness. Therefore, if the translator wants to preserve the
value of poetic ellipsis in Haiku, they should not add too many
communicative clues to their translations.
4. The ellipsis of Haiku

4.1 The literary quality of poetic ellipsis
In the previous section, it was demonstrated that the literary value of
Haiku depends on its poetic ellipsis. To let poetic ellipsis have many
explicatures what kind of words do we omit, or what kind of words do
we chose to leave? Consider the following Haiku:
(28) Matsushima ya
Aa Matsushima ya, Matsushima ya
Even now, many people believe Basho wrote this Haiku. The story
behind it is that Basho was too impressed by the scenery of the
Matsushima seashore to compose anything more than these
exclamations. However, this Haiku and the story behind it were both
made by somebody else. We could say this Haiku doesn’t have
multiple explicatures so that it is not a good Haiku, although the

repetition of the same exclamation (or the noun phrase) might inspire
some implicatures.
Pilkington (2000) distinguished the epizeuxis with poetic effects from
the one without any literary effects.
(29) My childhood days are gone, gone.
(30) Oh, Fred, my colleague, my colleague.
(31) The pubs have closed, closed.
Pilkington said that without any special context we look for many
meanings for the word ‘childhood’ while we cannot find any rich
meanings for the word ‘colleague’. The word that has poetic effects
makes the reader access the wide range of implicatures, he explained.
On the other hand, such a word used in (30) is not poetic but
sentimental, and the sentence of (31) is simply ridiculous.
As for Haiku’s poetic ellipsis, the words which were chosen in Haiku
should be the same kind of word Pilkington mentioned. We can say
that a good Haiku consists of these effective words that inspire readers
to discover many weak implicatures and explicatures. Since the
refrained noun phrases in (28) inspire readers to discover neither many
explicatures nor many implicatures, it can be said that it does not have
much literary value as a good poem.
Recall the example of (10b). Although the explicature of the utterance
is very weak, the utterance is not poetic because the word ‘fire’ in this
context does not inspire many weak explicatures or implicatures. In
this case, the utterance does not have the effects of poetic ellipsis and
is just an ambiguous statement or an exclamation. Exclamations are
very similar to Haiku’s expression. However, they are, in Pilkington’s
words, just sentimental.
Above all, the quality of Haiku’s poetic ellipsis is determined by the
words comprising the sentence. As Pilkington defined it, the words
inspire the reader, many implicatures are poetic and have literary
value. ‘Literary value’ means, in this case, highly relevant in that the
text is highly effective but is energy consuming to process.
4.2. Metaphors and season words

One effective way of expressing multiple concepts in a few noun
360 Kyoko Arai
phrases is to use metaphors. In relevance theory, metaphors are

analyzed by using a theory called ad-hoc concept construction. (See
details in Carston, 2002) The hearers can construct the concept of a
word when they process each utterance. The ad-hoc concept is
temporal and it changes every time the hearer processes the word.
This is one example of the metaphors in Haiku, and the flowing
passage is an essay by the author, Basho:
(32) Yuku haru ya
Tori naki, Uo no me ha namida. (Basho)
Loath to let spring go
Birds cry, and even fishes’
Eyes are wet with tears. (Translated by D. Britton)
(33) My closest friends, who had been with us since the night
before, came on the riverboat to see us off. We disembarked at a
place called Senju, and my heart was heavy at the thought of the
miles that lay ahead. And though this ephemeral world is but an
illusion, I could not bear to part from it and wept. (Translated
by D. Britton, Britton 2002: 30)
The journey at that time was not very easy, so many of Basho’s friends
saw him off at the boat harbor and cried. Leaving spring implies
Basho and the birds and fish are metaphors for those friends. These
kinds of metaphors are used only at the moment Basho wrote this
Haiku; in other words, they are not conventional.
Metaphors inspire many meanings (mainly implicatures); therefore,
they are highly relevant words that should be used in the literature.
In addition to this, Haiku literature has metaphors with fixed images
and meaning, called ‘Kigo’ (season words). They are collected in
official glossaries called Saijiki and Haiku poets always refer to these
glossaries when they create Haiku poems.
(34) Samidare o
Atsumete Hayashi, Mogami-gawa (Basho)
Gathering as it goes
All the rains of June, how swiftly
The Mogami flows! (Translated by D. Britton)
‘Samidare,’ according to the glossary, is a season word for the rainy
season of mid-summer, and also the sound of the word is very pleasant
in Japanese.
(35) Shizukasa ya
Iwa ni shimiiru, Semi no koe (Basho)
In this hush profound,
Into the very rocks it seeps –
The cicada sound (Translated by D. Britton)
‘Semi (cicada)’ is one of the season words for summer. For many
Japanese the sound of cicadas inspires many images and feelings, such
as good old memories of catching cicadas in our childhood, the
beautiful nature of the countryside in summer, etc. In Haiku, these
season words are used effectively and they can help to expand the
concept of words. By using them, poets can express many weak
explicatures and implicatures.
By using highly relevant words, including metaphors and season
words, Haiku’s poetic ellipsis achieves literary value. It is said that
readers should appreciate Haiku beyond the author’s intention by
using the readers’ free imagination. The more weak explicatures are
recovered, the more weak implicatures are found for each explicature.
The idea that it is possible to recover an infinite number of meanings
from a Haiku gives us a hint as to why Haiku is sometimes referred to
as a ‘microcosm’.
5. Conclusion
By reviewing some basic concepts about the meanings of an utterance
and the relevance theoretic account of ellipsis, and by describing the
concept of poetic effects, the literary value of the shortness of Haiku
was explained, thereby setting the stage for the introduction of a new
tool to analyze literary texts: poetic ellipsis. Next, Haiku’s ellipsis was
explained through this notion by using examples of some English
translations of Japanese Haiku. Haiku’s poetic ellipsis, above all, is a
kind of poetic ellipsis but the quality of ellipsis, that is, which words
are chosen, is a very important issue. To make a good Haiku the
author has to choose highly relevant words that inspire many weak
explicatures and implicatures.
In the near future, I would like to analyze many other ‘the shorter, the
better’ examples, such as catch phrases and headlines, using the
362 Kyoko Arai
concept of poetic ellipsis.
Endnotes
1
The author first introduced the word ‘poetic ellipsis’ during the ‘IfR, Interpreting for
Relevance Conference’ in Poland in 2003.
2
Nikko Toshogu is a shrine in Tochigi prefecture, which was built in 1617 to enshrine
the first Shogun of the Edo Era, Ieyasu Tokugawa, by his son, Hidetada.
References
Arai, K. 2002. Syntactic Ellipsis and Non-Syntactic Ellipsis – Haiku, the Literature of
Ellipsis, Eigo Gohou Bunpou Kenkyu 9. Tokyo: Kaitakusha.
Blyth. R.H. 1981. Haiku, Volume one: Eastern Culture.Tokyo: The Hokuseido Press.
Britton, D. 2002. A Haiku Journey – Basho’s Narrow Road to a Far Province. Tokyo:
Kodansha International.
Carston, R. 2002. Thoughts and Utterances, The Pragmatics of Explicit
Communication. Blackwell Publishers, Ltd.
Gutt. E.A. 2000. Translation and Relevance – Cognition and Context. Cambridge,
MA: St. Jerome Publishing.
Ishihara, Y. 1989. Haiku-Bunpo Nyumon (An Introduction to Haiku Grammar).
Tokyo: Iizuka Shoten.
Pilkington. A. 2000. Poetic Effects – A Relevance Theory Perspective.
Sperber, D., and D. Wilson. 1997. Relevance, communication and cognition, second
edition. Blackwell Publishers, Ltd.
Thoreau, H. D. ‘Walden’ The Writings of Henry David Thoreau, (Shanley, J.L. edited
1971). Princeton: Princeton University Press.
Wilson, D., and D. Sperber. 1990. ‘Rhetoric and Relevance’ in The Ends of Rhetoric
History, Theory, Practice. Stanford: Stanford University Press: 140-155.
Wilson, D., and D. Sperber. 2000. ‘Truthfulness and Relevance’ in UCL Working
Papers in Linguistics 12: 215-254.
Wilson, D. and D. Sperber. 2002. ‘Relevance Theory’ in UCL Working Papers in
Linguistics 14: 249-287.
Emotion Tracking Pedagogy:
Towards Better Teaching of the National Curriculum
for English
Emma Dawson
The University of Nottingham
Abstract
This paper presents research which reveals that ‘literature from different cultures and
traditions’ is not being properly fulfilled as part of the National Curriculum
requirement at Year 8 in schools in England. Reasons why this part of the Curriculum
is currently neglected are presented here and a solution is offered in the form of a
pedagogy. World Englishes literature is defined and offered as a literature used to
represent ‘fiction from different cultures and traditions’. This paper brings together
aspects of World Englishes literature, stylistics and emotion study.
Keywords: affective response; basic emotions; educational research; emotional
response to literature; emotion study; Emotion Tracking Pedagogy; Key Stage 3: Year
8; National Curriculum; pedagogy; post colonial literature; stylistics; universal
emotions; World Englishes literature.
1. Introduction
This paper argues that in the 21st century stylistics is a viable
pedagogical tool, helping to teach new literatures in an ever changing
world. The literature discussed here is World Englishes literature, not
to be mistaken as world literature in translation nor as Diaspora
literature; a definition of World Englishes literature will be offered in
this paper. In general, the teaching of this type of literature is often
neglected as it is perceived to be too ‘Other’ (Said 1978), wrapped up
in issues of complex cultural phenomena, including foreign lexemes
and culturally dependent concepts. I argue that the teaching of World
Englishes literature should not be viewed as problematic; it is simply a
question of approach.
Turning to research by Short (1999), stylistics ‘upside down’ is
presented for the teaching of language and literature to both native and
non-native students and this ‘upside-down’ approach to language and
literature has provided a useful model for this current research on the
teaching of World Englishes literature. Short speaks of native speaker,
undergraduate students and how many of them have little basic
364 Emma Dawson
grammar knowledge due to little or no grammar instruction in their

schooling (1999: 43). The teaching context Short describes is similar
to the one of this research area, as the pupils are at Year 8 in the
English school system (12 years old) and have little basic grammar
knowledge; a purely stylistic approach, as with Short’s undergraduate
students, would not be viable.
It should be noted that the approach developed as part of this research
does not explicitly demonstrate a stylistic approach, however, in line
with work by Carter (1987, 1991, 1999), Brumfit (1987), Long (1991)
and McRae (1991, 1992, 1999, 2003) the approach has a strong
element of creative reading techniques using stylistics embedded
within. This study has its limitations and I do not present it as a piece
of research which has tested a purely stylistic approach for the reading
of World Englishes texts, rather, as a stylistic approach moulded into a
pedagogical tool. This tool aims to engage pupils with a World
Englishes text and to better understand the interface between language
and literature.
2. World Englishes and Emotion Tracking Pedagogy

As part of the National Curriculum for English at Year 8 (Key Stage
3), drama, fiction and poetry by major writers from different cultures
and traditions, (1999: 36) is required to be taught. The research has
documented that the fiction element of this National Curriculum
stipulation is not being fulfilled at present. The research involved
eight schools, in two cities in Derbyshire and Staffordshire and the
aim was to assess the reality of the situation with regards to the
teaching of literature from different cultures and traditions. A
questionnaire was distributed to a teacher of Year 8 English in each of
the eight schools and the questionnaire was designed to:
1. assess whether this aspect of the National Curriculum was
being fulfilled at all.
2. if the aspect was being fulfilled, assess which authors/works
were being taught to comply with the National Curriculum.
3. assess the range and availability of resources for teachers
that may be used to fulfil this aspect of the National
Curriculum.
The questionnaire included presenting the ‘Example Author’ list from
the National Curriculum; this list offers example authors in order to
Emotion Tracking Pedagogy 365
teach this aspect of the Curriculum (see 1999: 36). The questionnaire
used this list as a basis to assess which authors/works were being
taught, although space was given on the questionnaire to indicate
other authors/works taught which did not appear on the
aforementioned list.
Overall, the questionnaire indicated that drama, fiction and poetry by
major writers from different cultures and traditions is not being taught
at Year 8, at least not in these eight schools. What is interesting about
the results from the questionnaire is that reasons why this element is
not being taught in these eight schools include lack of resources,
budgeting and training. These issues could all be considered as
nationwide concerns thus it is highly possible that the results from this
limited research in eight schools represents a nation-wide situation.
According to the questionnaires, all eight schools stated that the range
of resources for teaching the ‘Example Author’ list from the National
Curriculum as well as any extra available resources they have within
the Department remain ‘under-developed’. The teachers were asked in
the questionnaire to speculate why the resources are perceived to be
under-developed and four main reasons for this were given:
1. lack of resources (dated resources/translation works)
2. financial constraints (budgeting within the department)
3. training (how to teach this type of literature, question of
methodology)
4. time in class and also in preparation (prioritisation of
literature to be covered at Yr 8)
Having completed this stage of the research, it was clear that in order
for the study of literature from different cultures and traditions to be
properly fulfilled, what is needed is not only a resource but also some
kind of methodology for its implementation.
Emotion Tracking Pedagogy was created primarily to realise the latter
of these issues, although the practical application of the pedagogy
produced an apparatus and therefore a resource. At the heart of the
apparatus is a World Englishes text and this is in response to the
National Curriculum requirement to teach ‘fiction by major writers
from different cultures and traditions’ (1999: 36). Definitions of
366 Emma Dawson
World Englishes are offered by prominent authors in the field such as

Jenkins (2006), Melchers & Shaw (2003), and McArthur (2003)
however, these publications define World Englishes linguistically and
sociologically; little is said of the literature of World Englishes. World
Englishes continues to remain a purely (socio)linguistic phenomenon
pertaining to the notion that many people speak English around the
world and that these people speak in varieties or ‘Englishes’. World
Englishes in this sense has been formally recognised since the seminal
work of Kachru (1986, 1992). Kachru’s model of the spread of
English around the world remains one of several base models from
which we understand the phenomenon of World Englishes.
For Kachru (1992) the Englishes of the world can be divided into ‘The
Inner, Outer and Expanding circles’ and these three roughly
correspond to the concepts of English as a native language (ENL),
English as a second language (ESL) and English as a foreign language
(EFL) respectively. The ‘Inner circle’ includes USA, UK, Canada,
Australia and New Zealand. The ‘Outer circle’ nations such as: India,
Kenya, Malaysia, Singapore and for the ‘Expanding circle’ countries
such as: China, Egypt, Israel, Japan. In a sense it would be logical to
transpose Kachru’s model onto the phenomenon of World Englishes
literature but it does not work so well. Kachru’s model is in essence a
sociological framework, showing the reality of English use around the
world, but the defining factor of World Englishes literature is that it
does not include literature written from the ‘Inner circle’ (Dawson
PhD thesis 2007). Writing from the ‘Inner circle’ I argue, is defined
by the country from which it is produced, for example England
produces English literature and also ‘Diaspora’ literature may feature
as a ‘literature’ of that country.
I therefore argue that World Englishes literature is defined exactly
through the fact that it is written outside of the ‘Inner circle’ (Kachru
1992). I offer a definition of World Englishes literature here:
‘Most (but not all), World Englishes writing explores the
country and people from which it is written, often the literature
employs an English of that place (to a lesser or greater degree)
and moreover, the writer chooses to write in that English over
other languages he/she writes in.’
(Dawson PhD 2007)
Often misunderstood, World Englishes literature is not a synonym for

post-colonial literature, although many countries which have a history
of English colonialism produce World Englishes literature. The voice
of World Englishes literature is not one which necessarily laments
post-coloniality nor one which wishes for the ‘subaltern to speak’
(Spivak: 1988), rather World Englishes I suggest, is ‘post’ post-
colonialism and although it represents those writers who may
remember the moment of independence of their country (India 1947,
Nigeria 1960, Kenya 1963, Philippines 1946), it also represents a
generation of writers who do not.
The definition above is a definition I offer in order to define the
literature as something which is not ‘Diaspora literature’ nor
‘literature in translation’. An example of World Englishes literature
would be Arundhati Roy’s The God of Small Things whereas 26a by
Diana Evans is not an example of ‘World Englishes’ literature rather,
an example of ‘Diaspora literature’ according to the research.
The World Englishes literary text is part of what I refer to as the
‘apparatus’ and this itself is a physical manifestation of the pedagogy,
created through the research. A brief overview of the theory behind
Emotion Tracking Pedagogy and extracts of an apparatus based on
Emotion Tracking Pedagogy are to follow. Emotion Tracking
Pedagogy will be referred to as E.T.P.
With the belief that there are four basic emotions which are universal
to human beings, E.T.P focuses on these, highlighting the importance
of ‘affect’ as a vehicle of response to literature. The four universal
emotions employed in E.T.P are: happiness, sadness, fear and anger
and the choice of these four emotions in particular is due to my
adaptation of Turner’s (2000) work on basic emotion types. These
four universal emotions are explored in what constitutes the first phase
of the E.T.P apparatus (The Emotion Tracking phase) and there are
four phases in total.
Happiness and sadness are specifically explored through the first half
of the first phase; here a World Englishes short story is split into six
‘chunks’ with prompts eliciting (affective) responses for each chunk
of text (the rationale of how and why the text is ‘chunked’ is not
within the scope of this paper). Fear and anger are explored in the
second half of the first phase through a thought shower exercise,
368 Emma Dawson
revisiting the narrative and exploring empathy with the character/s

portrayed.
The second phase of the apparatus is known as the ‘Language and
Emotion’ phase. The first part of this phase is a set of questions
which help the pupils to consider why and how they felt as they did
whilst following the text (World Englishes short story) through. The
first question in this phase always asks the pupils to look back on the
emotions noted in the first phase (‘Emotion Tracking’ phase). This is
then expanded by asking the pupils to plot the feelings that they noted
down in response to the six chunks of text. It is at this point that the
four universal emotions are central; in plotting the emotion
experience, the four basic emotions provide a framework for focusing
on the emotions experienced through the text - happiness, sadness,
fear and anger.
Within this second phase we then move to consider the linguistic
aspect of ‘Language and Emotion’ on further, and it is here where
‘stylistics’ is used as a pedagogical tool to engage the pupils with the
text, eliciting response and heightening understanding between
language and literature. The pupils are not equipped explicitly with a
stylistic toolkit, rather they are led to discover how language works for
literary effect and only then are literary devices as ‘labels’ or
‘definitions’ offered. At this point in the apparatus, the pupils have
already formulated a meaning of the story and possibly an opinion
about the main character’s feelings and experience. Short reminds us:
stylistics is usually thought of as an analytical technique to help
support or test already-formed interpretative hypotheses. (1999: 42)
and this notion is echoed in the formulation of the tasks in this part of
the apparatus. The tasks in this second phase attempt to help the
pupils realise how their understanding of the text and experience of
the narrative has direct links with the words on the page. Below is an
example taken from the apparatus to highlight this point:
So, we can see that when we read, we experience different feelings
and emotions. But why and how does this happen ?
We’re going to think about what it is that makes you feel these
different emotions as you read through a story.
Let’s concentrate on this part of the story - as you read through
circle any words which you think made you feel different
emotions.
An unexpected shove jolts me out of my dream and back to the moment. Then
I’m wobbling, fighting for control. I fall.
Unbelievable!
I swallow the girt on my tongue and shake my head to clear the ringing in my
ears. I feel confused. Not quite on this earth. My hands are grazed with white
track chalk mixed with brown soil and smudges of blood. I shape them into
fists and press hard to force the pain away. A blue shirt whizzes by, kicking
dust in my face.
When I was in my dream, Kip must have pushed me with his elbow. Mami
would be proud of a son like Kip, who knows winning is what matters.
Legs zoom past me in a whir of hot air and dust. I glance toward the side of
the track. The crowd probably thinks Kip and I touched accidentally.
Which words have you circled?
----------------------------------------------------------------------------------
Now get into a group of 4 or 5.
We’re going to make this part of the text come alive through
SOUND only.
Read through and circle any words that add SOUND to the piece.
He’s in a cluster, but I know that Kip always goes for the flashy sprint finish.
I have to catch up with him now if I’m to have a chance. Concentration,
concentration, concentration now begins.
Amid all the crowd noises, I think I hear Baba yell, “Run, son!”
A new energy tingles from my feet, up along my legs, loosens my hips and
expands my chest. I tear past Chris, who is panting like a horse. Uganga
magic is with me!
The cluster is breaking up. Kip is racing ahead. My heart hammers in my
ribs. I open my mouth wider to take in more air. I’m catching up. I’m in the
dispersing cluster. I overtake one, two, three boys.
I’m flying, my feet almost slapping my bottom, half a step behind Kip.
Now decide who will make each sound – share them out.
Somebody in the group will read the piece of text and each person
must come in with their sound in the right place….
Have a go….
When a word sounds like its meaning we call this ‘onomatopoeia’.
370 Emma Dawson
Look back over the text and write down a couple of words that
sound like the words’ meanings.
When the initial sounds of words are repeated we call this
‘alliteration’.
Look back over the text and find where two words have the same
initial sounds repeated.
----------------------------------------------------------------------
Can you find an example of ‘alliteration’ in this part of the story?
Suddenly I see a tall figure approaching from a distance and shoot up again.
But Baba is half bald, and this man had tight clumps that look like sleeping
safari ants scattered about his head.
“Down, Kamau!” barks Mr Juma.
My race will start in a few minutes. I close my eyes and slowly mouth the
secret word. Ndigidigimazlpixkarumbeta!
Please let Baba be here by the end of this blink.
Here an extract from the short story is presented in order to perform a
more detailed reading. The aim with this task is to subtly demonstrate
the link between the emotion experienced when reading the text and
the language present on the page. In brief, this task attempts to
highlight to the pupils the important link between the experience as a
reader and the words written by the author. A more stylistic approach
would involve the use of metalanguage, however, because the pupils
concerned hold little basic grammar knowledge a more stylistic
approach would prove ineffective. Yet, the result, I argue, from
employing either a traditional stylistic approach or the E.T.P approach
will be very similar as both approaches help the pupils to gain entry
into a text, making the text mean. For World Englishes literary texts,
‘making the text mean’ is particularly valuable as World Englishes
literature displays ‘unfamiliar’ or ‘other’ (cultural) concepts/lexemes
and so engagement with the text is paramount.
As we see above, the second and third parts of this second phase lead
to discovery of the literary devices: onomatopoeia and alliteration In
the example from the apparatus, the tasks demand that the pupils
perform ‘onomatopoeia’ and ‘alliteration’ in their groups. Performing
‘onomatopoeia’ and ‘alliteration’ (rather than learning them through a
dictionary definition) is more effective for those pupils who remember

both the literary device as well as its effect.
What the second phase aims to do overall, is to highlight the
relationship behind the choice of words used in the text and their
effect on the reader and the meaning. This approach Short refers to as
the ‘upside-down’ approach and for Short this includes the element of
‘softening-up’. The aim of softening-up according to Short is
described as ‘getting students interested in doing analytical work in its
various aspects’ (Short 1999: 43). Short advocates the use of a
‘checklist’ once the students have been ‘softened up’ in order to
encourage students to be more ‘systematic analytically’. For the Year
8 pupils in this research context it is questionable whether the
Emotion Tracking apparatus allows them to become more ‘systematic
analytically’ but Short’s reassuring remarks below do offer hope:
You can not teach stylistics to everyone. Some will not be able
to cope with the specialised vocabulary or analytical approach.
But where the ‘lower’ limit is for stylistics is not clear, and the
educational and age levels where stylistics can be appropriate is
probably lower than most assume.
(Short 1999: 42) (Author’s own emphasis.)
The Year 8 pupils’ experience of the Emotion Tracking apparatus
could be considered as one of the ‘lower’ limits of stylistics Short
mentions and this would mean that stylistics has proved an appropriate
pedagogical tool for pupils as young as 12 years old.
The third phase of the apparatus is entitled ‘Cultural Exploration’. The
World Englishes short story provides a cultural backdrop and this is
explored through the third phase of the apparatus. Continuing on from
the previous phase (Language and Emotion), the ‘Cultural
Exploration’ phase begins linguistically in that it probes the
culture/geography/history of the text by turning to linguistic clues
embedded in the text itself. These clues may be place names or
people’s names. Below is an example taken from the apparatus.
In which country do you think this short story is set ?
Can you get any clues from the first paragraph ?
‘Wooyay, please with sugar cane juice,’ I silently pray. ‘Let me be
one of the lucky ones today.’ Although Kenyatta Primary Academy in
372 Emma Dawson
Nairobi has almost four hundred students, not many parents have showed up
for Sports Day. I don’t care about other parents as long as Baba is there for
me.
While the headmistress screeches something or other on the squeaky
microphone, I scan the group standing on the other side of the track. Baba is
not among them. He’s tall and big like Meja Rhino the champion wrestler, so
you can’t miss him.
What tells you where the story may be set ?
…………………………………………………………………
Once one of these linguistic clues has been presented (as above), this
third phase goes on to explore the cultural aspect further. Exploration
of the country, people, culture or geography may be carried out
through I.T resources, film or cultural markers such as dance, food or
maps; images and music may be used to encourage learning about the
culture/country presented, as well as to appeal to different learning
styles.
The final phase in the apparatus, the ‘Discussion & Task’ phase is
primarily a bank of resources and stimuli for creativity. Built around
the World Englishes literature and also the culture/country presented
through this literature, the tasks offered in this phase are creative, in
the sense that they do not dictate wholly what should be achieved.
Rather, as the pupil has experienced the apparatus individually, so the
pupil will produce what he/she is inspired to produce. The tasks which
constitute this final phase may be individual or group tasks and they
range from class projects and discussion of ideas to individual creative
writing tasks such as writing personal diary entries as a given
character in the story or writing acrostic poems.
3. Conclusion
In summary, this paper documents research which has used stylistics
as a pedagogical tool for the teaching of World Englishes literature.
Although this paper does not present the opportunity to review the
results of the research in detail, I can reveal that the results do
generally indicate that Emotion Tracking Pedagogy has been
successful in getting the pupils to engage with World Englishes texts.
It has also been successful in heightening the pupils’ awareness of the
interface between language and literature.
To the critics who believe that stylistics ‘has had its day’, I offer this
research on Emotion Tracking Pedagogy to demonstrate that stylistics
may be employed to successfully raise awareness of the interface
between language and literature in World Englishes short fiction, as
well as being a successful pedagogical tool for pupils as young as 12
in the English secondary school system of the 21st century.
N.B Since the writing of this paper, the National Curriculum has been
updated (2007). However, the changes that have occurred from this
update have not greatly impacted the teaching of literature from
different cultures and traditions with particular regard to my
research.
References
Roy, A. 1995. The God of Small Things. London: Flamingo.
Brumfit, C., and R. Carter. 1987. Literature and Language Teaching. Oxford: Oxford
University Press.
Carter, R., and M.N. Long. 1991. Teaching Literature. London: Longman.
Carter, R., and J. McRae (eds). 1999. Language, Literature and the Learner. London:
Pearson Education.
Dawson, E. 2007. PhD thesis: Emotion Tracking Pedagogy. (The University of
Nottingham, UK).
Department for Education and Employment, Qualifications and Curriculum
Authority. 1999. The National Curriculum for England Key Stages 1-4 English
London: Qualifications and Curriculum Authority.
Evans, D. 2005. 26a. London: Chatto & Windus.
Jenkins, J. 2006. World Englishes. Oxford/New York: Routledge.
Kachru, B. 1986. The Alchemy of English: The Spread, Functions and Models of Non-
native Englishes. Oxford: Pergamon, reprinted 1990 Urbana: University of Illinois
Press.
Kachru, B. 1992. ‘Teaching world Englishes’ in Kachru B. (ed.) 1992.
Kachru, B. 1992. The Other Tongue. English Across Cultures. (2nd ed) Urbana:
University of Illinois Press.
McArthur, T. 2003. Oxford Guide to World English. Oxford: Oxford University Press.
McRae, J. 1991. Literature With a Small ‘l’. Basingstoke: MEP/Macmillan.
McRae, J. 1992. Wordsplay. Basingstoke: Macmillan.
McRae, J., and R. Carter (eds). 1999. Language, Literature and the Learner. Essex:
Pearson Education.
McRae, J., and M.E. Vethamani. 2003. Now Read On. London: Routledge.
Melchers, G., and P. Shaw. 2003. World Englishes. London: Arnold.
Said, E. 1978. Orientalism. New York: Pantheon.
Short, M. 1999. ‘Stylistics “upside down”: using stylistic analysis in the teaching of
language and literature’ in Carter, R., and J. McRae (eds) Language, Literature
and the Learner. Essex: Pearson Education: 41-64.
374 Emma Dawson
Spivak, G. 1988. ‘Can the Subaltern Speak?’ in Marxism and the Interpretation of
Culture (eds) Nelson, C., and L. Grossberg. London: Macmillan.
Turner 2000. On The Origins Of Human Emotion. Stanford, USA; Stanford
University Press.
Further Reading
Ashcroft, B., G. Griffiths and H. Tiffin (eds). 2003. The Empire Writes Back.
London: Routledge.
Bhabha, H.K. 1994. The Location of Culture. London: Routledge.
Boehmer, E. 1995. Colonial and Postcolonial Literature. Oxford: Oxford University
Press.
Brathwaite, E.K. 1984. History of the Voice. London: New Beacon.
Damascio, A. 1994. Descartes’ Error. New York: Penguin.
Fanon, F. 1997. Black Skin. White Mask. New York: Grove Press.
Hogan, P. 2003. The Mind and Its Stories. Cambridge: Cambridge University Press.
Iser, W. 1978. The Act of Reading: A Theory of Aesthetic Response. USA: John
Kovecses, Z. 2003. Metaphor and Emotion. Cambridge: Cambridge University Press.
Loomba, A. 2002. Colonialism/Postcolonialism. London: Routledge.
McLeod, J. 2000. Beginning Postcolonialism. Manchester: Manchester University
Press.
Plutchik, R. 1994. The Psychology and Biology of Emotion. New York: Harper
Collins.
Rosenblatt, L. 1938. Literature As Exploration. New York: Appleton-Century.
Richards, I.A. 1978. Practical Criticism. London: Routledge & Kegan Paul Ltd.
Short, M. (ed.). 1989. Reading, Analysing and Teaching Literature. London:
Longman.
Turner, M. 1996. The Literary Mind. Oxford: Oxford University Press.
Wierzbicka, A. 1999. Emotions Across Languages and Cultures. Cambridge:
Cambridge University Press.
Real People or Verbal Constructs: A Stylistic Analysis
of Character in Fiction
Sarala Krishnamurthy
Polytechnic of Namibia, Windhoek, Namibia
Abstract
The novel is an art form that belongs to the modern age and as such has been
subjected to scrutiny for various reasons and varied purposes. For any lecturer
teaching at the postgraduate level, the present scenario is fraught with danger. She has
to contend with many ‘theories’ and negotiate her way through a minefield which puts
her in a precarious position vis a vis her subject matter and her students. New
developments that have taken place in the field of Literary Criticism and, Literary and
Critical Theory have brought in a plethora of terms and expressions, and new ways of
thinking and dealing with literary texts. So much so that the text itself has disappeared
from view. These days, Literature is seen more as a tool to be used to expound a thesis
and it is not studied in its own right. However, in order to restore literature to its
primacy of place in Literature Studies a paradigm shift has to take place. One of the
ways of doing this is by using Stylistics, which is a subject and a discipline of
thinking divested of any ideology because it deals with the text in and by itself. My
paper takes one aspect of the novel which is highly contested by critics and which has
not received due attention from Stylisticians. This is Character and Characterisation.
The paper reviews the hypotheses that deal with characters in novels as independent
people and argues that characters are only verbal constructs. Indeed, characters should
be subjected to linguistic analysis just as speech, narration or focalization. My
theoretical premise derives from Rimmon Kenan (1983) and I extend it further by
developing a framework for the analysis of character in fiction by using Halliday’s
Functional Grammar (1985) and motif analysis of Narratologists. Through my
analysis, I show how critics can be wrong in their interpretation of a character and for
the purposes of illustration, I take Chinua Achebe’s novel, Things Fall Apart. (1958)
for analysis.
Keywords: motifs; direct definition; indirect presentation; physical traits;
psychological traits; utterances; elaboration; enhancement; repetition; reinforcement.
1. Introduction
‘True wit is nature to advantage dress'd
What oft was thought but ne'er so well express'd’
- Alexander Pope
Style and Stylistics have been defined and described variously during
the past few decades. While in the past the focus of Stylistic analysis
was primarily restricted to one genre of literature, that is, poetry, now
376 Sarala Krishnamurthy
Stylistic analysis includes under its umbrella other genres such as

fiction and drama and other forms of discourse such as news writing,
academic writing, business communication etc. Other methods of
analyses, such as the structuralist approach have evolved in-keeping
with the development of narratology, semiotics, pragmatics, and
discourse stylistics which have expanded the field considerably and
made it more exciting and challenging for all of us. Early studies on
fiction used stylistic analysis largely in the same way as it was done
for poetry. The novel itself was not subjected to close scrutiny but
sections of the text were analysed and the theory extrapolated or
derived from the analysis to include the complete novel.
The structuralist approaches to narrative can be categorised in two
ways mainly. While some, like Bal (1980), use the term ‘Narratology’
to refer only to ‘the study of literary narrative from the point of view
of “narrative discourse”’ (Pavel 1985: 86) and exclude from the object
of the discipline the ‘text independent plot’ (p.86), others like Todorov
(1977) include both ‘research about plot structure (histoire) and about
text structure (discours) within its scope.’ Among the structuralists,
some like Propp (1968), Levi Strauss (1968), and Griemas (Hawkes:
1977) consider the story, whereas others like Barthes (1977), Todorov
(1977), Prince (1982), van Dijk (1985), Fowler (1977), Chatman
(1978) and Leech and Short (1982) could be said to concern
themselves more with text structure rather than plot. All of them
reduce narratives to patterns of abstract entities such as ‘functions’,
‘semes’, ‘mythemes’, etc in discovering the deep structural patterns
underlying concrete narratives. Hence, analysis carried-out according
to these approaches often results in disturbingly uniform patterns of
very dissimilar texts. What is needed here is an approach to the novel
which deals with both plot structure to include the study of the story
and character in the novel and narrative structure which examines
narration, focalization, characterisation and the study of time in
concrete terms.
In order to delimit the topic, this paper is illustrative of an attempt to
analyse character and characterisation in a novel through a stylistic
analysis that draws upon insights gained from Narratology and
Functional Grammar. Toolan (1992) points out that extra-textual
knowledge is brought to bear on our reading of any text and adds ‘a
semantic feature analysis of the characters of a text involves
Real People or Verbal Constructs 377
specifying a limited list of what the analyst takes to be as crucial

features or attributes which distinguish particular characters’(p. 99).
But his own analysis does not include a comprehensive study of a
character in any novel and he chooses for his analysis James Joyce’s
short story, Eveline.
With the emergence of the new form of fiction in the twentieth
century, the notion of characters as real people has given way to
character as verbal constructs. Since literary critics can no longer
describe characters as they do real people, they have declared the
character ‘dead’. The Structuralists with their new found notions on
language have added ‘nails to its coffin’ (Rimmon Kenan 1983: 30).
There have been two main trends in criticism with reference to
characters in fiction. The first is the realistic argument which insists
that in the course of the novel characters acquire a certain
independence from the events in which they live and they can be
discussed at some distance from their context. Realists like Watt
(1957), Ghent (1953), Hardy (1964), Allen (1954) and Forster (1927)
discuss characters as they would people. The ‘purists’ like Propp
(1968), Griemas (1977) etc, on the other hand, insist that characters do
not exist at all except ‘in so far as they are a part of the images which
bear and move them’ (Hawkes 1977: 104).
2. A realistic approach to characters

Fiction’s distinctive literary form vis-a-vis the romance was because
of its emphasis on ‘realism’. The romance tended towards triumphant
adventure and a heroic protagonist. Shroder (1967: 22) states
‘romance depends on the art of inflation: the romanesque world is one
in which every youth is a hero, every antagonist an ogre, every maiden
a masterpiece of nature’. The romance idealised heroes and heroines.
Their names were chosen precisely because they echoed previous
myths and legends. Sometimes their names were symbolic of a single
quality which they represented, much like ‘humours’ for e.g.
Everyman, (Every Man in His Humour) ‘Christian’, ‘Badman’
(Pilgrim’s Progress) etc. However with the introduction of realism in
the art of the novel, proper names were given to characters, almost as
if to give them their own identity in the world of the novel. Watt
(1957: 19) quotes Hobbes who says, ‘Proper names bring to mind one
thing only, universals recall one of many’. Therefore, writers like
Defoe, Sterne and Richardson were interested in creating a fictional

character whose has an identity of his/her own. The realistic novel, as
written by these authors, portrays varieties of human experience.
Bradbury (1973: 13) points out that ‘realism has represented a
humanistic balance in which the claims of individuals as persons grow
coherently in a world solid and substantial enough for them to
encounter it in its force and value, so that the reality of persons meets
the realistic of society, or history’. Therefore, a character has a
historical and social existence as well as his/her own individuality.
Perhaps the character represents the social milieu that he inhabits or
he is, maybe a radical, fighting against the mores and conventions of
his society. Realism of the novel is thus a product of sociological,
cultural, philosophical and ideological causes. This realism, says
Bradbury (1973:14), tends ‘to regard characters as a matter of
propriety, individualism and its progressive view of history produces
lively dealings between the two’.
Hardy (1964: 14) however, insists that novels are not about characters
as much as about human relationships. According to her ‘the novel is
usually concerned with giving a substantive picture of human
relationships and if this fails no amount of serious purpose or poetic
unity can make it a good novel’. She points out that the unpopularity
of character analysis in Shakespeare’s plays has resulted in an over-
emphasis on the study of theme and symbol in novels. However, if we
were to reduce the novel to its ‘didactic capacity’ (ibid) we lose ‘the
necessary respect for the intent of local appearances of truth’.
Muir (1974: 23) has distinguished between the novel of ‘action’ and
novel of ‘character’. The novel of action concerns mainly the plot. But
in the novel of character, ‘the characters are not conceived as parts of
the plot, on the contrary they exist independently, and the action is
subservient to them’.
Forster (1927: 73-81) makes a distinction between ‘round’ and ‘flat’
character. Flat characters are ‘humours’ and are constructed around a
single quality or idea. ‘Round’ characters are ‘those which are capable
of surprising us in a convincing way’. Thus, critics advocating
‘realism’ in the novel discuss the characters either as individuals in
their own right, or as individuals poised in relationships, or as
individuals who are a part of history.
Though there have been detailed discussions of characters in novels,

not enough attention seems to have been given to the techniques of
characterisation that novelists use. Occasionally, a few evaluative
remarks are made about the skill or lack thereof of a certain novelist in
portraying characters. No attempt is made to substantiate these
remarks on the basis of a close study of the novels under
consideration.
3. A purist approach to character in fiction

Rimmon Kenan (1983: 33) points out that ‘in the text characters are
nodes in the verbal design. Structuralists do not believe in the
centrality of character in the novel, as literary critics did. For them
character is important only in so far they are necessary for action.
Propp subordinates characters to ‘spheres of action’ (Kenan 1983: 84)
and designates seven general roles to them: the villain, the donor, the
helper, the sought-for-person, the father, the dispatcher, the hero and
the false hero. Griemas (1977) reduces the seven categories outlined
by Propp into three pairs of opposed ‘actants’ which emphasise not
the individual item but the structural relationships between them. For
example, his categories are (Hawkes 1977: 91):
A. Subject versus Object
B. Sender versus Receiver
C. Helper versus Opponent
Todorov (1977) isolates three dimensions of the narrative: ‘semantic’
aspect, its ‘syntactical’ aspect and its ‘verbal’ aspect. He analyses the
stories of the Decameron in terms of ‘propositions’. Propositions are
formed by the combination of a noun (character) with either an
adjective (attribute) or a verb (action). All attributes are reducible to
these adjectival categories: ‘states’, ‘interior properties’ and
‘conditions’ (Hawkes 1977: 97).
4. Framework of Analysis
A stylistic analysis of characters in fiction provides a cogent, coherent
and comprehensive way of dealing with characters’ constructs. For
purposes of illustration, Chinua Achebe’s Things Fall Apart (1958),
(henceforth TFA), is taken for analysis. In order to study the
techniques of characterisation as well as the characters in the novels,
all motifs from the texts which represent a particular character trait
have been considered. The notion of ‘motif’ is taken from
Tomashevskij (quoted in Dolezel 1971: 96) who defines it as ‘minimal
dissection of thematic material’. Since it is the use of particular
techniques of characterisation which makes it possible for the reader
to ‘reconstruct’ (Rimmon Kenan 1983: 36) a character in a particular
way, these techniques have been studied in detail here. According to
Rimmon Kenan (1983) there are basically two types of textual
indicators of characters. These are direct definition and indirect
presentation:
I Direct Definition
Directly defines a character trait by:
A. an adjective
(1) Okonkwo was tall and huge. (TFA: 3)
B. an abstract noun
(2) But his (Okonkwo’s) whole life was dominated by fear.
(TFA: 9)
II Indirect Presentation
Whenever a character trait instead of being described, is
exemplified or ‘displayed’ (p. 61) then it is indirect presentation
e.g.,
(3) Okonkwo did not taste food for two days after the death of
Ikemefuna. (TFA: 44) (this motif indirectly reveals Okonkwo’s
love for the boy, Ikemefuna)
Text motifs which directly define or indirectly present character
traits have been classified into three groups:
A. Those which present physical traits
B. Those which present psychological states
C. Utterances of characters about themselves or others which
reveal certain traits.
Description of physical traits is a crucial indication of character

according to Rimmon Kenan (1983: 40), because a set of physical
attributes implies a psychological trait as an attributive proposition.
Further, in any attempt to investigate the techniques of
characterisation it is important to study how much physical

description a novelist uses with reference to his major character.
Hence, motifs which describe physical attributes of characters have
been considered in this analysis of characterisation and characters. As
for the description of psychological states this is the typical means
novelists use to present the internal perspective of a character.
Another technique of characterisation that novelists use, is to describe
a character through the words of other characters. Hence, these are the
three categories considered here. Motifs under A and B are
attributable to the narrator, whereas those under (C) are attributable to
some character in the novel, e.g.,
A. Physical traits
DD
(4) He was tall and huge, and his bushy eyebrows and wide
nose gave him a severe look. (TFA: 3)
IP
(5) Okonkwo’s wives … might have noticed that the second
egwugwu had the springy walk of Okonkwo. (TFA: 63)
B. Psychological traits
DD
(6) Inwardly he was repentant. But he was not the man to go
about telling his neighbours that he was in error. (TFA: 22)
IP
(7) Okonkwo never showed any fear of showing emotion
openly, unless it be the emotion of anger. (TFA: 20)
C. Utterances of characters
DD
(8) ‘I am not afraid of blood.’said Obierika. (TFA: 60)
IP
(9) ‘Go home and sleep’, said Okonkwo (to his wife), ‘I shall
wait here’. (TFA: 76)
These are the examples of utterances of characters who by means of

those utterances reveal some of their own character traits. In (8),
Obierika condemns Okonkwo’s killing of Ikemefuna and proclaims
that he is not a coward. But in the next motif (9), Okonkwo
admonishes his wife and tells her to go home and rest. His favourite
child, Ezinwa has been taken away by the Priestess, Chielo. His
speech reveals his love for both his wife and his daughter indirectly.
Further, it also reveals his anxiety. There are other instances in the
novel of Achebe where characters describe directly or indirectly other
people’s traits e.g.,
DD
(10) ‘She should have been a boy’, he (Okonkwo) thought as he
looked at his ten year old daughter. (TFA: 44)
IP
(11) ‘Looking at a king’s mouth’, said an old man, ‘one would
think he (Okonkwo) never sucked at his mother’s breast’. (TFA:
19)
In addition to considering whether a character trait is directly defined
or indirectly presented, we can also study how a particular, trait is
‘expanded’ by the novelist. The notion of ‘expansion’ is taken from
Halliday (1985: 132-250). In his treatment of the clause complex he
discusses two types of relationships namely ‘expansion’ and
‘projection’. According to him a clause can be ‘expanded’ by another
clause in three ways: (a) elaboration, (b) extension and (c)
enhancement.
(a) Elaboration
In elaboration ‘one clause elaborates the meaning of the author by
further specifying it or describing it’ (Halliday 1985: 203). He gives
the example: Alice could only look puzzled: she was thinking of the
pudding.
(b) Extension
According to Halliday (1985: 207), ‘one clause extends the meaning
of another by adding something new to it’ e.g.,’I said you looked like
an egg, Sir and some eggs are very pretty you know.’
(c) Enhancement
In enhancement ‘one clause enhances the meaning of another by
qualifying it in one of a number of possible ways: by reference to
time, place, manner, cause or condition’ e.g. Alice looked up, and
there stood the Queen in front of them. (Halliday 1985: 211)
Halliday’s analysis of the clause complex deals with how a clause is
‘expanded’. But here the same notion is applied to how a text motif
representing a character trait is ‘expanded’. Since we are concerned
with the treatment of individual traits only ‘elaboration’ and
‘enhancement’ are considered and not ‘extension’, which relates to
progression from one entity to another. For e.g.,
a. Elaboration
In TFA the motif:
(12) But his whole life was dominated by fear is elaborated as:
But his whole life was dominated by fear, the fear of failure and
weakness. It was deeper and more intimate than the fear of evil
and capricious gods and of magic, the fear of the forest, and the
forces of nature, malevolent, red in tooth and claw. Okonkwo’s
fear was greater than these. It was not external but lay deep
within himself. It was the fear of himself, lest he should be
found to resemble his father (TFA: 9).
b. Enhancement
(13) And so when Okonkwo of Umuofia arrived at Mbaino as
the proud emissary of war, he was treated with great honour and
respect. (TFA: 9)
Okonkwo’s leadership qualities are enhanced by the respect with
which he is treated both by his own tribe and also by the neighbouring
villages.
Another way of looking at the treatment of a character trait in a novel
is to consider whether it is ‘repeated’ or ‘reinforced’. Repetition is a
trait that is repeated on more than one occasion. ‘Reinforcement’ of a
trait is defined here as representation of different aspects of the same
character trait, on different occasions in different situations, e.g.,
Repetition
(14) Okonkwo ruled his household with a heavy hand (TFA: 9)
(trait: strictness).
(15) He treated Ikemefuna as he treated everybody else, with a
heavy hand (TFA: 20) (trait: strictness).
(16) ‘Sit like a woman’, said Okonkwo (to Ezinwa). (TFA: 32)
(trait: strictness)
Reinforcement
(17) Inwardly he was repentant. But he was not the man to go
about telling his neighbours that he was in error. (TFA: 22)
(desire to appear superior).
(18) Inwardly Okonkwo knew that the boys were still too
young, but he thought one could not begin too early. (TFA: 23)
(desire to appear strict).
(19) Dazed with fear, Okonkwo drew his matchet and cut him
(Ikemefuna) down he was afraid of being though weak (TFA:
43) (desire to appear strong).
In the first three examples Okonkwo’s strict nature is revealed in all
the motifs. This device is called repetition. In the second set of
examples, the text motifs reveal a trait of Okonkwo, namely his desire
to appear strong and warrior like according to his and his society’s
concept of manliness. Not only does he desire to appear superior, he
also wants to be known as strict and takes great pride in being known
as a fearless and ruthless man. It is this trait of his that gets reinforced
in the novel by the addition of a new dimension to the same quality.
When different characters in a novel are considered together, an
important technique of characterisation that novelists use relates to
similarity or contrast between characters. The purpose for which such
similarities or contrasts between characters are used is also significant.
For instance in TFA
(20) During the planting season Okonkwo worked daily (p. 10)
(21) Okonkwo’s prosperity was visible in his household (p. 10)
Unoka (Okonkwo’s father)
(22) In his day he was lazy and improvident (p. 3)
(23) Unoka, the grown up was a failure (p. 4)
The traits derived from the motifs that describe Okonkwo are clearly
contrasted with the traits desirable from the motifs that describe his
father, Unoka.
5. Comments by Critics on the characters in Achebe’s novel

Most literary critics have addressed the issue of the protagonists in
Achebe’s novel. They discuss Okonkwo in detail. They consider
Okonkwo either as a representative of his society or a tragic figure
fighting against the conventions established by his society. Irele
(1967: 178) claims ‘in many ways Okonkwo represents his society in
so far as the society has made the man by proposing to him certain
values and lines of conduct’, and that Okonkwo’s way of conforming
is a ‘sort of perversity’ (Irele 1967: 179) because of the unusual
importance he attaches to manliness. Nwoye, on the other hand, says
Irele (ibid) ‘is presented all along as a sensitive young man whose
psychology turns against certain customs of the village’. Thus, Nwoye
stands as a ‘symbolic negation of his father, the living denial of all
that Okonkwo accepts and stands for’ (ibid).
Palmer (1972: 53-55) states that Okonkwo is what his society has
made him, for ‘his most conspicuous qualities are a response to the
demands of his society’. ‘Okonkwo is’, Palmer continues, ‘the
personification of his society’s values and he is determined to succeed
in this rat-race’. He adds that even though Unoka, Okonkwo’s father,
is presented to us as lazy and cowardly, we should be cognizant of the
fact that he has ‘artistic imagination: he “loves” good food, good
fellowship; lives for the day and is in harmony with the forces of
nature’.
Laurence (1968: 103) describes Okonkwo as ‘a true representative of
the Ibo of that period, living in villages which regarded one another
with a mutual suspicion born of insecurity, a highly individualistic
society which did not acknowledge inherited rules but in which the
wealthy became virtual rulers’.
Others critics see Okonkwo as a rebel who is constantly fighting
against the norms of his society. Cook (1977: 66-67) states that by
flouting the norms of his society ‘Okonkwo brings many of his ills
upon his head’. He, however, describes Okonkwo as ‘heroic’ because
he ‘shows exceptional bravery, firmness, even greatness of soul’.
Further, Cook adds, ‘he makes a final, grandiloquent assertion of the
values of his society before the established patterns of that society is

changed beyond all recognition’.
Carroll (1980: 40) is of the opinion that Okonkwo’s desire for success
stems from his father’s failure. He says that Okonkwo’s ‘inflexible
will’ brings ‘him success in a society remarkable for its flexibility’.
He says that we sense a growing alienation between Okonkwo and the
members of his society as success and wealth grow. According to
Carroll it is Obierika who ‘recapitulates the painful tensions within the
tribe in a way which proves in the long run to be more ominous than
the temporary eclipse of the hero’ (ibid).
Most of the critics discuss the characters in Achebe’s novels as they
would discuss real people, attributing human characteristics and
motives to them. Only one critic makes a passing remark on
characterisation in Arrow of God (1964), Achebe’s third novel, and
none at all on TFA.
6. Findings
In this study an attempt is made to examine the characters in TFA,
along the lines of both the traits attributed to them and the methods
which are used to present these traits. The features of characterisation
are as follows:
A. The focus is on the protagonist, Okonkwo, therefore, of the 115
motifs identified as those devoted to characterization, 94 deal with
Okonkwo.
B. In TFA, the character traits which directly define Okonkwo are 82
motifs and indirectly present 12.
C. Achebe describes psychological states more than physical traits and
he does not often use ‘speech’ for characterisation. Whenever he does
this, the protagonist’s speech provides an indirect presentation of a
certain character trait, rather than direct definition. The following list
gives the number of motifs which deal with physical, psychological
traits and speech.
TFA – Okonkwo Phy 28
Psy 34
Spe 22
D. Another notable feature about Achebe’s technique of

characterisation is that the character traits that are directly defined are
further elaborated upon in order to give details of certain traits, for
instance, in TFA, an important motif about Okonkwo:
Okonkwo is well known throughout the nine villages and even beyond
(TFA: 3). Achebe elaborates in order to explain why Okonkwo was
well known.
His fame rested on solid personal achievements. As a
young man of eighteen he had brought fame to his village
by throwing Amalinze the cat. Amalinze was the great
wrestler who for several years was unbeaten, from
Umuofia to Mbaino. He was called the cat because his
back would never touch the earth. It was this man that
Okonkwo threw in a fight which the old man agreed was
one of the fiercest since the founder of their two engaged a
spirit of the wild for seven days and seven nights. The
drums beat and the flutes sang and the spectators held
their breath. Amalinze was a wily craftsman, but Okonkwo
was a slippery as a fish in water. Every nerve and every
muscle stood out on their arms, on their backs and their
thighs and one almost heard them stretching to breaking
point. In the end Okonkwo threw the cat.
E. In all his novels, Achebe builds up a certain picture of his
protagonist by repeating certain characters traits, for instance, in TFA
Okonkwo’s traits; ‘hardworking’; ‘successful’, ‘proud’; ‘strict’;
‘respected’; ‘desire to conform to the mores of his society’ are
repeated several times in the course of the novel.
F. The minor characters in Achebe’s novels highlight certain traits of
the main characters by providing contrasts to them. In TFA, Okonkwo
is contrasted with his father, Unoka, on the one hand and his son,
Nwoye on the other. Okonkwo is hardworking and successful. He has
won two titles in his tribe. His father is lazy and a failure by his
society’s norms. Okonkwo is violent by nature, but his son Nwoye
dislikes any form of violence and therefore rejects the notion of
manliness as espoused by his father.
TFA, Achebe’s first novel, has Okonkwo as its protagonist. Achebe,

by making use of direct definition gives a complete picture of
Okonkwo. The narrative voice which relates the story of Okonkwo
can be assumed to be an old man in the village. Therefore, all
statements that proceed from the narrative voice are to be regarded as
authoritative, objective and unbiased. Thus, when this voice says that
Okonkwo was ‘a man of action and not of words’ (p.76), it is
substantiated by the motifs in the novel which are directly connected
with the characterisation of Okonkwo. Therefore, Achebe, as a true
story teller, describes the life of the Ibo village and through his
narration uses direct definition to portray Okonkwo’s traits. His
physical traits are described in great detail as are his psychological
states. Of the 17 speech motifs that constitute the speech motifs in the
novel, 15 are Okonkwo’s own speech. Through his words we
indirectly come to know some character traits of his. All the speech
utterances of Okonkwo are addressed to his children, (specifically to
his daughter, Ezinwa) and to his friend, Obierika. Only in two
instances does Achebe use free indirect style to voice Okonkwo’s
thoughts: both times he has lost his sons, the first time soon after the
death of Ikemefuna, the surrogate son whom he loves dearly, and the
second time when Nwoye, his son abandons the traditions of his
ancestors and joins the Christians. The motifs are:
(24) When did you become a shivering old woman? (TFA: 45)
(25) Why, he cried in his heart should he, Okonkwo, of all the
people, be cursed with such a son? (TFA: 180)
Though Achebe clearly defines Okonkwo’s physical and mental
states, just once in the novel, he steps backs and refuses to either
directly define or indirectly present any trait. This is when the elders
of the Ibo clan inform him about the decision to sacrifice Ikemefuna to
the Gods.
(26) But when they went, Okonkwo sat still for a very long time
supporting his chin in his palm.(TFA: 40)
Okonkwo is a man of few words indeed because every time he speaks,
it is the attendant circumstances that give his words their significance.
Thus, what he says is not elaborated but enhanced. Achebe uses
repetition and reinforcement to emphasise certain character traits of
Okonkwo. Okonkwo is a representative of the Ibo clan therefore all
the motifs which represent his conventionality, his desire to conform

to the rules of the tribe, his stress on masculinity, his violence are
repeated again and again. His desire to appear strong is reinforced by
his desire to appear superior and his desire to appear strict. His fear of
being thought weak is reinforced by his fear of showing emotion,
which is reinforced by his desire to appear strong. His fondness for his
children is reinforced by his love for his daughter, Ezinwa, and his
concern for his first born, Nwoye whose desertion brings him despair
and immense pain, and his extreme attachment to Ikemefuna whose
death causes him untold misery. His loyalty to his tribe is reinforced
by his stress on masculinity that is compounded by his short temper
and violent nature.
Comments by other characters in the novel on the protagonist often
help us to understand Okonkwo better. Three people comment on
Okonkwo. The first is his father, Unoka, who says, ‘you have a manly
and proud heart. A proud heart can survive a general failure because
such a failure does not prick its pride. It is more difficult and bitter
when a man fails alone!’ (TFA: 18). He is described as arrogant by an
old man in the village: ‘Looking at a king’s mouth … one would think
he never sucked at his mother’s breast’ (p. 9). The last comment is by
his friend Obierika who says, ‘That man was one of the greatest men
in Umuofia. You (the white man) drove him to kill himself.’ (p. 147).
This analysis of characterisation presented here provides us with a
comprehensive picture of characters in a novel. Okonkwo, in TFA is
delineated as a man with both faults and good characteristics, as can
be seen from my analysis. While literary critics interpret a character
from a theoretical premise that is suited to their argument, stylistic
analysis of a character presents an objective point of view. The
framework of analysis developed can be utilised to settle many
disputes that arise because of ideological differences. Thus, it can be
said that the discipline of Stylistics is thriving and further, teaching
and research in this field will prove to be both useful and rewarding.
References
Achebe, Chinua. 1958. Things Fall Apart. London: Heinemann.
Achebe, Chinua. 1964. Arrow of God. London: Heinemann.
Allen, Walter. 1954. The English Novel. London: Phoenix House.
Bal, M. 1980. Narratology: Introduction to the Theory of Narrative. Toronto:
University of Toronto Press.
Barthes, R. 1977. ‘Introduction to the structural analysis of narratives’ in Image-

Music Text. London: Fontana.
Bradbury, Malcolm. 1973. Possibilities: Essays on the State of the Novel. London:
Carroll, David. 1980. Chinua Achebe. London: Macmillan & Co. Ltd.
Chatman, S. (ed.). 1971. Literary Style: A Symposium. London: Oxford University
Press.
Chatman, S. 1978. Story and Discourse. Ithaca, Cornell University Press.
Cook, David. 1977. African Literature: A Critical View. London: Longman.
Dijk, Teun A. van (ed.). 1985. Discourse and Literature. Amsterdam: John Benjamin
Publishing Co.
Dolezel, Lubomir. 1971. ‘Towards a Structural Theory of Content in Prose Fiction’ in
Literary Style: A Symposium (ed.) Chatman, Seymour. London: Oxford University
Press.
Forster, E.M. 1927. Aspects of the Novel. Hammondsworth, Middlesex: Penguin
Books Ltd.
Fowler, R. 1977. Linguistics and the Novel. London: Methuen.
Ghent, Dorothy van. 1953. The English Novel: Form and Function. New York:
Cornell University Press.
Griemas A. 1977. in Structuralism and Semiotics (ed.) Hawkes, Terence. London:
Methuen.
Halliday, M.A.K. 1985. An Introduction to Functional Grammar. London: Edward
Arnold.
Hardy, Barbara. 1964. The Appropriate form: An Essay on the Novel. London:
University of London: Althone Press.
Hawkes, Terence. 1977. Structuralism and Semiotics. London: Methuen.
Irele, Abiola. 1967. ‘Tragic conflict in Achebe’s novels’ in Introduction to African
Literature (ed.) Beier, Ulli. London: Longman.
Laurence, Margaret. 1968. Long Drums and Cannon. London: Macmillan & Co Ltd.
Levi-Strauss, Claude. 1968. Structural Anthropology. New York: Doubleday Anchor.
Leech, G.N., and M.H. Short. 1982. Style in Fiction. London: Longman.
Muir, Edwin. 1974. The Structure of the Novel. London: Hogarth Press.
Palmer, Eustace. 1972. Introduction to the African Novel London: Heinemann.
Pavel, Thomas G. 1985. ‘Literary narratives’ in Discourse and Literature (ed.) Teun
A. van Dijk. Amsterdam: John Benjamins Publishing Co.
Prince, G. 1982. Narratology: The Form and Function of Narrative. The Hague:
Mouton.
Propp, V. 1968. The Morphology of the Folktale. Austin: University of Texas Press.
Rimmon-Kenan, Shlomith. 1983. Narrative Fiction. London: Methuen.
Shroder, Maurice Z. 1967. ‘The Novel as a Genre’ in The Theory of the Novel (ed.)
Stevick, P. New York: The Free Press.
Stevick, Philip. 1967. The Theory of the Novel. New York: The Free Press (First
published by Collier Macmillan Ltd, London).
Todorov, T. 1977. The Poetics of Prose. Oxford: Blackwell.
Toolan, Michael J. 1992. Narrative: A Critical Linguistic Introduction. London:
Routledge. (First published in 1988).
Watt, Ian. 1957. The Rise of the Novel, Harmondsworth: Penguin (First published by
Chatto and Windus).
APPENDIX –LIST OF MOTIFS (A Sample)

_____________________________________________________________________
Page No Motif DD/IP Phy/Psy Ela/Enh Trait
Rep/Rein
Comp/Cont
_____________________________________________________________________
P3 Okonkwo was well known DD Psy Ela Good wrestler
throughout the nine villages
P3 He was tall and huge, and his DD Phy Ela

bushy eyebrows and wide
nose give him a severe look
P6 Okonkwo was clearly cut out DD Psy Ela Successful and

for great things showing promise
P 9 And so when Okonkwo of IP Psy Enh Respected

Umuofia arrived at Mbaino as
the proud emissary of war, he
was treated with great honour
and respect
P9 Okonkwo ruled his household DD Psy Ela Strict

with a heavy hand
P9 But his whole life was DD Psy Ela Fear

dominated by fear Contrast with Unoka
P10 During the planting a season DD Phy Ela Hardworking
P11 Okonkwo’s prosperity IP Psy Enh Dominating

was visible Rein
P12 Okonkwo did not have DD Phy Ela Hardworking

the start in life which many Rep
young men had usually
P13 With a father like Unoka, DD Phy Ela Hardworking

Okonkwo did not have a Rep
start in life which many Cont with Unoka
young men had
P15 ‘You have a manly and DD (Speech) Proud

a proud heart ….’ Rein
P18 I tried Okonkwo’s DD Psy Enh Impatient

patience beyond words
P19 ‘Looking at a king’s mouth’, DD (Speech) Ela Proud

said an old man, ‘one would Rep
think he never sucked at his
mother’s breast’
P19 He (the old man) respected DD Phy Enh Hardworking

him for his industry and Respected
success Successful
P19 But he was struck by DD Psy Ela Brusque

Okonkwo’s brusqueness Intolerant of
in dealing with less less successful men
successful men
P19 Anyone who knew him DD Phy/psy Ela Hardworking

grim struggle against poverty Rep
and misfortune could not
say he had been lucky
P19 Anyone who knew his grim DD Phy/psy Ela Hardworking

struggle against poverty and Rep
misfortune could not say he
had been lucky
P20 Okonkwo never showed any DD Psy Ela Fear of showing

emotion openly, unless it be emotion
the emotion of anger
P20 He treated Ikemefuna as DD Psy Enh Strict/fair

he treated everybody else Rep
with a heavy hand
P21 And that was the year DD Phy Ela Anger

Okonkwo broke the peace Impulsive
and was punished
P22 Okonkwo tried to explain DD Psy Ela Desire to

to him (Ezeani) what his conform
wife had done Rein
P22 Inwardly he was repentant. DD Psy Ela Desire to

But he was not the man appear superior
to go about telling his Rein
neighbours that he was in error
P23 Okonkwo spent the next DD Phy Ela Industrious

few days preparing his Careful
seed yams Rep
P23 Inwardly Okonkwo knew DD Psy Ela Desire to

that the boys were still appear strict
too young … Rep
P23 But he though one could DD Psy Ela Strict

not begin too early
P26 Okonkwo always asked his DD Phy Enh Desire to

wives’ Relations conform
Rep
P27 But somehow Okonkwo DD Psy Ela Loner

could never become as
enthusiastic over feasts
as most people
P27 Without further argument DD Psy/phy Ela Authoritative,

Okonkwo gave her a violent
second beating
P32 ‘Sit like a woman’ DD (Speech) Enh Strict

Okonkwo shouted Rep
Just for Laughs: The Construction of Nonverbal
Humour
Nicola Lennon
Queen’s University, Belfast, UK
Abstract
This article focuses on the discursive construction of the nonverbal candid camera
television programme Just for Laughs, Volumes 1 & 2. This Saturday evening prime-
time show consists of a series of extended practical jokes designed to outwit
unsuspecting members of the public. Previous research into the structural design of
the joke in discourse studies has focused predominantly upon issues of verbal and
textual detail. For example, Nash (1985) discusses the narrative shape and locative
formulae of joke design in relation to spoken and written discourse. However, the
potential for exploiting such existing methodologies as a means of dismantling the
architecture of nonverbal, particularly filmic texts, has yet to be fully explored. This
paper applies a number of Nash’s frameworks for analysing phase structure, narrative
shape and locative formulae, to the fabric of popular television production Just for
Laughs. The article, which also draws upon interview evidence from the show’s
executive producer and editors, argues that humour creation in the show may be
adequately researched using techniques such as those presented in this paper. Given
the 21st Century trend for texts to branch away from spoken and written media, the
methodological treatment of nonverbal filmic texts to the fields of discourse studies
and stylistics warrants further investigation.
Keywords: humour; candid camera; gag, filmic; nonverbal; narrative; shape; structure.
1. Introduction
Just for Laughs is a popular candid camera television production, with
origins in the annual Quebec comedy festival Juste pour Rire, which
began in 1983. While the Canadian version of the show, Just for
Laughs Gags (filmed for Canadian Broadcasting Corporation and the
Comedy Network) is filmed mainly in Montreal, the UK version of
Just for Laughs is produced by the British Broadcasting Corporation,
and shot predominantly in the Northern Irish city of Belfast and the
Scottish city of Glasgow. The rationale behind an in-depth exploration
of the television programme emanates from its broad cross-linguistic
and cross-cultural appeal. In other words, the majority of the gags (a
term synonymous with the practical joke) used in the programme may
be understood by anyone who is visually exposed to them, regardless
of his/ her linguistic or cultural background. It is for this reason that
396 Nicola Lennon
the show has been widely adopted as a form of in-flight entertainment,

to the amusement of passengers flying to locations across the world.
The reason behind the show carrying such a universal interpretation is
grounded in the way in which it is consciously fabricated by the
production and post-production (editing) team, who continually strive
to deliver a series of well-timed broadcast performances. Moreover,
the steps involved in the composition of the programme directly
correlate to the various interrelated linguistic stages of a typical joke,
proposed by a number of scholars. The viewing public are able to
perceive and interpret such visual gags as humourous, mapping their
knowledge of recognisable forms of verbal and textual humour onto
the nonverbal and visual space.
This article examines how humour is constructed in Just for Laughs,
an analysis built upon a detailed discoursal investigation of 15
episodes of the programme, totalling 244 individual gags. It was
through this analysis that the triad of linguistic formats exploited by
Just for Laughs were revealed. These are summarized in Figure 1. In
the first format type, (non-linguistic and nonverbal), language is
completely exempt from the production and editing processes. Rather,
the practical joke is explained using only gestures and facial
expressions, and is accompanied by music which highlights key
events in the narrative. The second format type, the verbal-visual gag,
is accompanied by some speech, together with a layer of music. The
language may be English, but may also be a foreign language where
the producers know that the majority of participants in the joke will
not be able to comprehend. The third type of gag, termed the verbal-
visual joke, involves the participant reading, or becoming implicated
in, some type of text, typically a map, poster or questionnaire. This
paper will focus solely on the first format, although scope does exist
to explore the other two formats in more detail at a later stage.
Just for Laughs 397
Figure 1: Linguistic and Non-Linguistic Formats in Just for Laughs

(I) Non-Linguistic & This format of gag uses no
nonverbal language at all in its production
or post-production. The gag is
explained using gestures and
facial expression only, and is
accompanied by music which
highlights key events in the
narrative.
(II) Verbal-Visual There are two types of verbal-
visual gag:
(a) English is used to some

degree in order to communicate
the message of the joke, and is
accompanied by music.
(b) A Foreign language is used by

the actors, to the exclusion and
linguistic mis-comprehension by
the majority of the punters /
audience.
(III) Visual-Textual This format involves the punter

reading, or becoming implicated
in, a text, typically a map, poster
or questionnaire.
Additional primary research involved an illuminating interview with

one of the show’s executive producers. I also had the opportunity to
visit the post-production company where much of the show is edited
in preparation for broadcasting, and was able to observe an editor as
she cut a gag from start to finish. Throughout the process of
researching the humour mechanisms of Just for Laughs, which is
chiefly nonverbal, a number of similarities emerged between the show
and some existing linguistic models which have sought to explain how
humour is achieved in verbal and textual media. In other words, while
398 Nicola Lennon
the content of the comedic material in Just for Laughs is chiefly

communicated through nonverbal, visual and non-linguistic means,
the underlying design of the gag in the programme may be uncovered
using linguistic models provided by Nash (1985). It is possible to
map the configuration of verbal and textual humour onto its
nonverbal, visual equivalent, while incorporating a range of multi-
disciplinary perspectives, such as the Psychology and Anthropology
of humour and laughter. Moreover, there is something tantalizing
about our human communicative and linguistic abilities, that we are
unconsciously able to read filmic texts in the same way as verbal and
textual ones. This reveals much about what Carrell (1997) calls
‘humour competence’ and ‘joke competence,’ and what such
competence can reveal about our innate human quest for cross-cultural
communication.
2. The narrative shape of the gag: formats and functions

As a starting point for analysing the underlying mechanisms of
humour construction in Just for Laughs, it was decided that an
interview with a producer of the show could shed some interesting
light on the attitudes of those directly responsible for creating humour,
towards its underlying features.1 The interview took place between the
producer (identified from this point forward with the acronym PR) and
myself, on 15th June 2006. According to PR, the production team are
aware that they are essentially composing a narrative that mimics a
children’s story, and structure the introduction to the practical joke to
imitate this form. Mundane, pre-practical joke information must be
filmed, such as an ordinary member of the public strolling down the
street, in order to build a sense of anticipation and surprise when the
farce is actually enacted upon the participants.
In order for the production team to complete the nonverbal narrative,
cameramen, actors and the director of the programme must adhere to a
set of cinematographic rules. A member of the editing team revealed
that most gags require two cameras, disguised from the public by a
tent-like structure. The two cameras must then obey the 180º Rule, a
cinematographic term used in television production which stipulates
that a ‘camera can move through an arc of 180 degrees relative to the
center point between the subjects,’ such as two subjects A and B, as
shown in the diagram in Figure 2 below.2
Just for Laughs 399
Figure 2: The 180 Degree Rule 3
The first camera is situated at an angle of 0 – 45° in relation to the

second camera. One camera must directly face the person who is
being tricked, producing a series of establishing shots to show the gag
in its entirety. The second camera needs to film a series of close-up
shots in order to capture the reactions, expressed through facial
expression and physical gestures, by the punter. Throughout the
editing process, the general rule of thumb is to cut from a wide-angle
shot to a close-up shot, varying the filmic pattern occasionally by
introducing a mid shot. This creates an extremely fluid narrative
pattern. Both the production and post-production (editing) teams have
an explicit awareness that, in order to create humour nonverbally, they
must adhere to a strict narrative formula.
According to PR, the choice as to whether the practical joke is
produced as a purely nonverbal text, or with some speech, is
dependent upon what the public actually say, and the decision is
reached in the final editing process. Each gag, PR continues, should be
filmed so that it may be understood without any speech, and that
verbal content supplements the events in the narrative. Additionally, a
distinction can be made between verbal material from the actors,
which serves an explanatory function, and verbal content from the
participating public, who may produce a comical statement or noise.
400 Nicola Lennon
On the other hand, PR attests, attempts may be made to censor the

verbal material, in a bid to protect the viewing audience from, for
example, expletive exclamations. In this sense, as the producer of the
show would argue, television becomes a game of smoke and mirrors.
As this is primarily a study of nonverbal humour construction, the
paper’s analytical focus uses a range of narrative examples from
format type (I), as defined in Figure 1. How then, can the non-
linguistic and nonverbal gag be related to the palette of linguistic
models for analysing humour? There are a range of linguistic models
and theories which could have been used, but Nash’s 1985
examination of the linguistic and stylistic structure of humour in
literary and non-literary texts was particularly attractive. Nash
provides a critical method for systematizing joke design, and in doing
so refers to the two aspects of narrative shape and the locative
witticism. (Nash, 1985: 27-30). Beginning with narrative shape,
Nash’s model of narrative phase structure (Nash, 1985: 28) can be
mapped onto a specific example of a classic gag recounted by the
producer.
Figure 3: Typical Phase Structure in a Just for Laughs Gag: Elderly
Crossing
Phase Action A Action B
I Old Lady begins to Motorist drives
walk across slowly along the road
the road
Motorist notices the

Old lady notices old lady
the motorist
Just for Laughs 401
Phase Action A Action B

II Old lady hesitates as Motorist waves her
she walks across
slowly across the road
Old lady looks Motorist reassures

insulted by these her, through more
exaggerated hand hand gestures, that
gestures she will be quite safe
III Young female Old lady accepts help
appears and offers to and the pair continue
help old lady cross to walk across the
the road road
Another old lady

starts walking across
the road, beginning
the process again
The above gag, termed Elderly Crossing, may be broken down in
terms of its narrative phase structure. Following Nash’s critical
method, Phases I and II of the above Just for Laughs gag can be
classified as containing a symmetrical structure, with each move being
matched by a counter-move. Therefore, in phase 1, action (a) involves
the old lady crossing the road, which is matched by the counter move
(action (b)) of the motorist driving slowly across it. The motorist then
notices the old lady, which is counterbalanced by the old lady noticing
the motorist. In phase II, the old lady initially hesitates as she crosses
the road, countered by action (b), where the motorist waves her across
the road. The motorist then reassures her, through more exaggerated
hand gestures, that there is no danger in her continued passage. In a
counter move, the old lady looks insulted by these impatient and
frustrated signs.
It is only in Phase III, however, that the asymmetricality of the
narrative is posited: there is no counter or consequences, only a single
402 Nicola Lennon
concluding act, terminal statement or outcome, (Nash, 1985: 29),

which takes the form of another old lady, willing to try the patience of
motorists once again. In obeying the comic narrative shape, this final
act cannot be resolved. Viewers are simply left laughing at the
asymmetry of the final phase. The models above establish a binding
relationship between the narrative phases of the gag, and the
symmetrical flow from Camera 1 (the wide-angled shot) to Camera 2
(the close-up shot), and then back again to Camera 1, as consolidated
in the post-production process. In the absence of speech or text, it is
this binding relationship which provides the viewer with the crucial
mechanisms for interpreting the gag, in its entirety, as inherently
comical. In fact, as PR argues, each gag constitutes a self-contained,
mini narrative with an individual set-up phase, exposition and
punchline, something which is explored in more depth below.
In a contrasting joke type, Nash attests that action symmetry is
replaced by a pattern of symmetrically matching attributes. (Nash,
1985: 29). This alternative joke format has been replicated in Figure 4.
Figure 4: The Narrative Shape of Joke Design- Nash (1985)
Action Stagecoach bandits lurk
counterpoise approaches vs behind boulders
Attribute counterpoise: Sheriff in vs bandits in
(a) generic clean dirty clothes,
clothes on a with
named horse nameless horses
Attributes counterpoise: Sheriff is deaf vs bandit leader is
(b) caricature Jewish
As Nash demonstrates, the generic attributes above are observed in

Western- style films, while the caricature attributes in part (b) are
devised and engineered by the humourist (Nash, 1985: 29). Nash’s
intriguing model, while opening up a possible discussion of filmic
humour, warrants a fresh and extended application. Figure 5 applies
Nash’s model to another typical gag from Just for Laughs:
Just for Laughs 403
Figure 5: The Narrative Shape of Joke Design- Drag Net

Action counterpoise: Police car vs Motorist
pulls motorist stops car
into layby and waits
for police
Attribute counterpoise: Policemen vs Motorist
(a) generic are dressed in waits while
uniform and policeman
appear normal appears
behind wheel
Attributes counterpoise: Policeman (1) is vs Motorist
(b) caricature wearing only his is shocked
socks and shoes; at the anti-
policeman (2) social
is wearing behaviour of
stockings the policemen.
(pantyhose)
and suspenders
In the above joke, the punter experiences a reversal of expectations.

The unsuspecting citizen, as PR outlines, is reluctant to question
figures of authority, especially the police. The comedy actors dress in
a manner to suggest that they are law abiding members of the
constabulary, and then perform in such a manner which subverts their
traditional social role, as Figure 5 illustrates. During this time, the
punter is genuinely perplexed and this, the show’s producer attests, is
essentially at the core of stimulating humour from a joke of this kind.
In terms of Nash’s second aspect of joke design, the locative
witticism, the viewer responds to the joke through their knowledge of
the underlying narrative. In relation to this point, Nash argues that oral
humour typically involves a simple relationship between the formulaic
superstructure of the joke and a substructure of generic detail. (Nash,
1985: 30-31). To illustrate the structural interdependency between
superstructure and substructure, Nash selects a textual joke which
refers to the universal unpopularity of the then Conservative
government in the United Kingdom, led by British Prime Minister
Margaret Thatcher:
404 Nicola Lennon
SUPERSTRUCTURE (FORMULA)
‘Guy Fawkes, where are you now that we need you?’
SUBSTRUCTURE (GENERIC DETAIL)

In 1605, Guy Fawkes and his associates plotted to
blow up Parliament while the House was in session –
the so-called Gunpowder Plot.
In 1981 the Conservative government was not
universally popular.
(Nash, 1985: 31)
Distinguishing between the Superstructure and the Substructure, or
generic detail inherent in Just for Laughs is a relatively
straightforward task. Using the example of the Nudist Supermarket
joke, from the collection of 244 gags analysed, one may provide a
similar nonverbal analysis to the above, using textual description. The
visual information of the joke is unconsciously decoded by the
viewing audience using cognitive and psychological mechanisms,
combined with a perceptual awareness of social norms, a common
axiom in the process of creating humour:
SUPERSTRUCTURE (FORMULA)
A shopper walks into a supermarket. The supermarket is filled
with naked shoppers, who carry on with their chores in a
perfectly normal fashion.
(narrative account of a gag in Series 2, episode 2)
SUBSTRUCTURE (GENERIC DETAIL)

Supermarkets are public areas.
Supermarkets (in the western world) are filled with
fully-clothed shoppers.
Western society is generally disapproving of people
who wish to practice nudism in public.
It is also possible to systematically apply Nash’s analysis of the
locative formulae in narrative humour to the humour created in
nonverbal texts, as Figure 6 below demonstrates. (see Nash, 1985: 34).
Just for Laughs 405
Figure 6: Locative Formulae in Just for Laughs’ Gag Production

LOCATION PRE-LOCATION
Verbal Did you The Irish Who missed a but Scored on

or hear centre- penalty the action
Textual about forward replay?
Humour
Locus
Signal (of Orientation Context (word or
the intention (to the type (in which phrase
to joke) of joke) joke operates) which
clinches or
discharges joke)
Non- Gag Trickster Location Comic Reversal

verbal, intro form/ / setting Timing of
Visual Identity Expectati
Humour ons
This adapted model of Nash’s locative formulae in linguistic humour

attempts to account for nonverbal humour in its structural, essentially
grammatical form. Nash provides the top layer of analysis, accounting
for verbal and textual humour, while I have added an additional
interpretative tier below the original analysis, in order to attempt an
equally exhaustive account of the nonverbal and visual humour
mechanisms which prevail in Just for Laughs.
Therefore, one might say that the two elderly women being introduced
at the start of the gag examined in Figure 3 consist of the gag intro,
which then orientates us to the type of gag, chiefly one where two old
ladies assume the role of trickster, operating within the context of an
urban road crossing. Visually a prank of this type reminds one of the
popular ‘why did the chicken cross the road?’ joke, (to get, of course,
to the other side). However, and this is where the coordinating
conjunction but becomes relevant, the expectations of the audience are
reversed, as the first elderly lady backtracks to her initial position at
the very end of the joke, signalling the futility of her mission.
406 Nicola Lennon
Comic timing, which has been associated with but in the textual joke,
performs a paratactic function, acting as a marker of coordination as
the audience anticipate the ‘locus’ of the prank. According to PR,
comic timing is absolutely paramount, and a key mechanism which
operates in humour creation on the visual space of the television
screen. The actors, director and to a larger extent, the editing team, cut
each gag with precision, to coherently and consistently demonstrate a
pattern of reversed expectations. By adjusting the timing in terms of
where it is least expected, the resultant effect can be additional surprise
for the punter, and consequently, the scale of laughter is increased.
Conceptualizing a visual gag in terms of its syntactic structure is
useful in understanding its additional phases. For example, each
practical joke finishes with a freeze frame, denoting the gag’s
denouement. According to PR, the freeze frame functions as a tag, a
visual tag which captures the revelation that the punter has been
tricked, and the humour and laughter rendered by the joke. Just as
comic timing serves a paratactic and coordinating function, the freeze
frame tag reiterates the grammatical components of the joke’s
essentially sentential construction. Further discussion on the correlates
between humour and syntax is explored in Godkewitsch (1974),
Hetzron (1991) and Oaks (1994).
3. Establishing trickster relationships

A joke...must be told to someone else. The psychical process of
constructing a joke seems not to be completed when the joke
occurs to one: something remains over which seeks, by
communicating the idea, to bring the unknown process of
constructing the joke to a conclusion.
Freud ([1905] 1966: 195)
In this section, the communicative function of the joke is surveyed,
which, as Freud attests, is absolutely essential to joke construction
and, as this article argues, to nonverbal as much as verbal and textual
comedy. Freud continues his thesis by acknowledging the importance
of interpersonal relationships in the comic process, stipulating that it
‘is content with these two persons: the self and the person who is the
object; a third person may come into it, but is not essential.’ (Freud,
1966: 195).
The specific form of trickster dynamics in Just for Laughs was alluded
Just for Laughs 407
to in consultation with PR, who contends that gags are edited in two
very different ways. In the first scenario, the audience at home are
explicitly shown how the action of the prank is going to unfold,
through an introductory or set-up phase. This consists of a couple of
shots which condense the narrative action, making the viewer
immediately aware of the form and nature of the prank. In a converse
form, this set-up phase is deliberately omitted, thereby consciously
surprising the audience at home with a sequence of events with an
ambiguous resolution, thus intensifying the humour experience. In a
typical episode, with a range of 14-20 gags (and within each gag, a
number of surprise reactions from a variety of participants), the
dynamics of the trickster-viewer relationship are continually modified
in order to retain the entertainment factor of the show, and reverse the
general expectation of the viewing audience. Therefore, the trickster
(television actor/ actresses/ producers) may play a trick on a
participant, but the ‘trickster’ may also outwit the viewer at home,
using the second strategy discussed above. And, as we have seen, the
trickster and audience at home may also act as co-conspirators on the
participant, in that the audience is aware of the joke through the
explanatory set-up phase. In an even more complex dynamic, the
trickster may be the friend or partner of the punter, who agree to play
along with the main trickster as the practical joke is wrought.
Additionally, in many cases, it is the trickster who, through costume,
simultaneously assumes the role of the clown or fool.
In Figure 7, Nash identifies the main players of the comic poetic
narrative, as executant, executant-within-the-text, respondent-within-
the-text, and respondent. Like Nash’s previous models, this analytical
model has also been adapted, using bold typeface, to include each of
the characters implicated in the Just for Laughs trickster framework. It
is possible to apply the model systematically to nonverbal, filmic
comedy in a manner that likens, for example, the show’s executive
producers to the executant of a poetic text, as the original authors of
the trickster narrative. The actors also operate within the filmic text,
and, like the persona who speaks for the author in a poetic narrative, it
is the actors who are hired to carry out the nasty work of the
executant. Moreover, the role of the respondent within the text is filled
by the participant who is outwitted in the comic narrative, also
referred to by the producers and editors as the punter. Finally, the role
of the viewer of the television audience may be likened to the
408 Nicola Lennon
respondent of a text, as reader and observer.

Figure 7: The Respondent in the Filmic Text (Adapted from Nash,
1985: 19)
(E ) Et Rt ( R)
E = executant: author, poet, wit, original ‘I.’ PRODUCER

Et = executant-within-the-text: the persona who speaks for the author,
perhaps, without necessarily being the author. ACTOR
Rt = respondent-within-the-text: the persona controlled by the
executant-within-the-text, and making responses shared or disclaimed
by the respondent-outside-the-text. PUNTER
R = respondent: the reader, as observer and censor. VIEWER
Finally, the nature of the audience requires further explanation. Most
obviously, the audience constitutes BBC, CBC and other television
viewers, but the programme is also broadcast on aeroplanes as a form
of in-flight entertainment, given its broad cross-linguistic appeal.
When questioned about the nature of Just for Laughs as an
international product, PR admitted that the show does not overtly
focus on its cross-cultural and cross-linguistic communicative
message. However, the show is generally concerned with creating an
adequate level of universality, so that the communicative intent of the
joke is not ‘lost’ on the audience.
4. Conclusion
This paper has exposed some of the ways in which the analytical
strategies pioneered by established Stylistician Walter Nash may be
practically applied to nonverbal comedy such as Just for Laughs, as
much as verbal and textual humour. Four of Nash’s models for
uncovering the phase structure, narrative shape, locative witticism and
locative formula of joke design, were each related to specific
examples of individual Just for Laughs gags. Some models proved to
be more relevant to certain types of gags than others, and a future
research project would centre on tracking a more extensive corpus of
gags in relation to the relevance of each of these analytical templates.
In addition, Nash’s concept of the respondent in the filmic text was
Just for Laughs 409
discussed in relation to each of the key players involved in the

television programme.
In spite of the fact that Nash’s thesis was written in 1985, this paper
has attempted to demonstrate its continued relevance to the linguistic
investigation of new and emergent types of discourse. In writing the
paper, I sought to demonstrate how Nash’s work may be moulded and
transferred onto texts lying outside of the spoken and written medium.
While conducting primary research with Just for Laughs’ executive
producer and editors, it became apparent that the programme makers
are, like stylisticians and discourse analysts, acutely aware of the
subtleties of narrative shape and phase structure. Moreover, each
component of the gag is carefully co-ordinated in order to
communicate a universally comical message to the televised audience.
The limitations of restricting the linguistic analysis of Just for Laughs
to the type of models proposed by Nash outlined in this paper, become
apparent when the television programme is approached from other
disciplinary perspectives, both inside and outside the field of
Stylistics.
Room certainly exists for extending the analysis above to include a
discussion of cognitive poetics (see, for instance, Stockwell, 2002).
Emmott (2003), for example, analyses plot reversals in narrative texts
using cognitive poetics. Anthropological studies would also contribute
important insights into the fabric of the television comedy. For
example, Apte (1985: 231) has written extensively on the nature of the
trickster in folklore, whom he defines as being dependent upon the
manipulation of verbal and visual modalities, social collectivity and
interaction. Moreover, the identity of the trickster is conditioned by
their ability to change form, shape or sex. They may change into
different objects, Apte attests, and this process of morphing can be
traced back to, for example, the Native American trickster, the African
trickster Ture, and the mythical Hindu trickster Indra. The trickster,
Apte continues, may also bear a physically grotesque appearance
(Apte, 1985: 226). Therefore, one needs to exert caution that, in
working solely within the parameters of Nash’s framework, one does
not omit significant theoretical contributions offered by other
disciplines.
While the models provided by Nash could not hope to account directly
410 Nicola Lennon
for the anthropological or psychological undercurrents of this or

indeed any other television series, they do provide linguists with the
necessary tools and techniques for approaching nonverbal and filmic
texts in the 21st Century.
Endnotes
1
With thanks to Wild Rover Production Company, Belfast, for granting access to
archives of Just for Laughs and for taking the time to host an interview with its
Producer, Phillip Morrow. I would also like to thank Offline Central Editing
Company., Belfast, for their insights into the editing process.
2
http://www.tv-handbook.com/Editing%20and%20Program%20Continuity.html
16th October 2006
3
http://www.tv-handbook.com/Editing%20and%20Program%20Continuity.html
16th October 2006
References
Apte, Mahadev L. 1985. Humour and Laughter: An Anthropological Approach. New
York: Cornell University Press.
Carrell, Amy. 1997. ‘Joke competence and humour competence’ in Humor 10(2):
173-185.
Emmott, C. 2003. ‘Reading for pleasure: A cognitive poetic analysis of “twists in the
tale” and other plot reversals in narrative texts’ in Cognitive Poetics in Practice
(ed.) Gavins, J., and G. Steen. London: Routledge: 145-159.
Freud, Sigmund. [1905] 1966. Jokes and their Relation to the Unconscious. London:
Routledge &. Kegan Paul.
Godkewitsch, M. 1974. ‘Correlates of humor: Verbal and nonverbal aesthetic
reactions as functions of semantic distance within adjective-noun pairs’ in Studies
in the new experimental aesthetics (ed.) Berlyne, D. E. Washington, DC:
Hemisphere: 279-304.
Hetzron, Robert. 1991. ‘On the structure of punchlines’ in Humor 4(1): 61-108.
Nash, Walter. 1985. The Language of Humour. New York: Longman.
Oaks, Dallin D. 1994. ‘Creating structural ambiguities in humor: getting English
grammar to cooperate’ in Humor 7(4): 377-401.
Stockwell, P. 2002. Cognitive Poetics: An Introduction. London: Routledge.
CBC Television: Comedy

<http://www.cbc.ca/justforlaughs/index.htm/> 5th May 2006.
Television Production Handbook: Editing and Program Continuity
<http://www.tv-handbook.com/Editing%20and%20Program%20Continuity.html> 16
October 2006
Official Just for Laughs/ Juste pour Rire Website
<http://www.hahaha.com/> 5th May 2006.
Official Wild Rover Productions Ltd Website
<http://www.wild-rover.com/framez.htm/> 5th May 2006.
Just for Laughs 411
Official Offline Central Website

<http://www.offlinecentral.com/> 5th May 2006.
Just for Laughs Gags: Volume 1. DVD, Image Entertainment, 2004.
Just for Laughs Series 2, Episodes 1- 6. VHS, Wild Rover Productions Ltd, 2006.
Just for Laughs Series 4, Episodes 2 & 3. DVD, Wild Rover Productions Ltd, 2006.
Sources
P. Morrow, personal interview, 15th May 2006.
D. Lennon & G. Brady, personal interviews, 12th October 2006.
On the Phrase ‘Even with a Weight of Pleasure’ in The
Prelude (Bk 2, 178)
Ken Nakagawa
Yasuda Women’s University, Hiroshima, Japan
Abstract
It is my opinion that Wordsworth is the one and only poet in English literature that
boldly connected ‘weight’ and ‘pleasure’ with the use of the particle ‘of.’ He captures
the strong feeling of joy not from the direction of <intensity> but from the direction of
<weight>. In The Prelude (1805 edition), Wordsworth uses ‘weight’ fourteen times,
which is far fewer than I expected. I divide the 14 instances of ‘weight’ into three
groups in terms of <value>, i.e. according to whether it is valuable to the poet or not:
(A) negative value, (B) positive value, and (C) neutral value. In order to explore his
idiosyncrasy in the combination of ‘weight’ and ‘pleasure,’ I investigated Chaucer
(1340?-1400) and Shakespeare (1564-1616) with regard to the combination. I did not
find any positive use of ‘weight of ~’ in both writers. For other writers, I also
consulted the Collins COBUILD Wordbanks Online and the Nineteenth-Century
Fiction: Full-Text Database, but the combination of ‘weight’ with ‘pleasure’ was
nowhere to be found. The Gutenberg Files gives us three examples of ‘weight of
displeasure’ and one example of ‘weight of a secret joy.’ What these four examples
reveal is that ‘weight’ is hard to combine directly with ‘pleasure’ or ‘joy.’ I conclude
that the oxymoronic word combination in the phrase ‘Even with a weight of pleasure,’
which seems to break the selectional restriction between lexical items, is an
expression peculiar to William Wordsworth, poet of nature and joy.
Keywords: deviation; ‘pleasure’; ‘weight’; word combination; Wordsworth’s Prelude.
1. Introduction
In this paper I will demonstrate how one of Wordsworth’s uses of
‘weight’ is a deviation from the norm of the English language.
Before we look at the deviational use of ‘weight’, let us turn our
attention to the non-deviational, conventional use of ‘weight.’ The
following passage appears in ‘Tintern Abbey’
that blessed mood,
In which the burthen of the mystery, A of B
In which the heavy and the weary weight Aƍ
Of all this unintelligible world, Of Bƍ
Is lightened:
(T. A. 37-41)
414 Ken Nakagawa
The A of B structure, ‘the burthen of the mystery’ in line 2, is repeated

appositionally and amplified to the Aƍ of Bƍ structure, ‘the heavy and
the weary weight / Of all this unintelligible world’ in lines 3-4.
Wordsworth regards the shackles of this world as a ‘weight.’ This is
certainly one of the most famous passages in English literature, and
The Oxford Dictionary of Quotations gives it as the first of eight
instances for the item ‘weight.’ Collins Dictionary of Quotations, too,
quotes this passage for one of its six citations.
What essentially expresses the same purport as the above is the next
passage from The Prelude.
(0)
Though doing wrong, and suffering, and full oft
Bending beneath our life’s mysterious weight A
Of pain and fear; yet still in happiness Of B
Not yielding to the happiest upon earth.
(The Prelude (1805) bk 5, 441-44)
Here Wordsworth states one aspect of his image of an ideal child.
Here, too, ‘weight’ is used. The noun form ‘mystery’ in the ‘Tintern
Abbey’ lines transforms into the adjective form ‘mysterious.’
Incidentally, Shakespeare (1564-1616) has Albany speak on the sad
occasion of King Lear’s death in similar terms:
The weight of this sad time we must obey,
Speak what we feel, not what we ought to say.
(Lear 5. 3. 324-25)
Japan’s own Tokugawa Ieyasu (1543-1616), originator of the
Tokugawa shogunate, is quoted to have said:
A man’s life is like a long journey with a heavy load on his
back.
(Watanabe et al: 2003, s.v. omoni)
Recently, Lakoff and Turner (1989: 25-26, 52) proposed a
metaphorical proposition:
LIFE IS A BURDEN.
From what I have discussed, the conception that ‘the shackles of this
world are regarded as weight’ is true for all ages and in all places.
On the Phrase ‘Even with a Weight of Pleasure’ 415
There is general agreement that carrying on with our work and

surviving in our life is a burden that necessarily falls upon human
beings. Put simply, LIFE IS A BURDEN is a universally acknowledged
truth.
2. Purpose
Now, the purpose of this paper, as previously stated, is to argue that
Wordsworth sometimes uses ‘weight’ in a deviant way from other
writers. Let me explore where and how he uses ‘weight’ in The
Prelude (1805).
There are a total of 14 examples of ‘weight’ in The Prelude, which is
far fewer than I expected. I scrutinized in what sense each ‘weight’ is
exploited in its own context.
2.1 Categorization of the examples of ‘weight’ in The Prelude

When categorizing the 14 examples, I had difficulty in finding an
appropriate criterion. OED (2nd edition) groups the meaning of
‘weight’ into five larger groupings and twenty-four smaller groupings,
which seems too complicated. Longman Lexicon of Contemporary
English classifies it into seven, and Roget’s II The New Thesaurus
classifies it into six, both of which classifications did not satisfy my
purpose, either. Finally, I divided the 14 instances of ‘weight’ into
three categories in terms of ‘value,’ that is, according to whether it is
valuable to the poet or not:
(A) negative value
(B) positive value
(C) neutral value
Here let me find a synonym suitable for each ‘value.’ Synonyms
corresponding to ‘weight’ with (A) negative value are ‘burden’ and
‘affliction’; those expressing the meaning of (C) neutral value are
‘counterpoise,’ that is to say, a ‘counterbalancing weight’ (OED2) and
‘restraint.’ When it comes to the choice of synonyms for (B) positive
value, I am quite at a loss, because I cannot find any appropriate
synonyms, however hard I try. Is there any other way of expressing it
than ‘weight’? It is rather easy to find synonyms for (A) negative
value and (C) neutral value, but difficult to allocate a proper synonym
for (B) positive value, which is quite strange to me. Perhaps in
416 Ken Nakagawa
everyday English ‘weight’ does not carry a <positive value>, whereas

Wordsworth does use this term in such a way. The following Chart
displays the semantic and syntactic features of the 14 examples.
Chart 1: ‘Weight’ and its surroundings
t)
in
tra
es
,r
s
es
ise
n)
in
po
io
pp
ct
er
ha
fli
nt
ne ( ? , af
ou
e,
ur
c
C) lue den
n
as
tio
ue
)
le
va bur
al
na
,p
in al v
bi
(
ht
n
(B lue
m
io
lig
iv
r
co
sit
ut
va
s it
de
er
o
)+
rb
t
)-
po
ep
w
ar
y,
ve
(A
po
he
ap
pr
m
jo
(1) + + + + burthen is shaken off

(2) + + + by
(3) + + + with
(4) + hang on days
(5) + + woes
(6) + + + + beneath
(7) + some personal concerns
(8) + + + against
(9) + + + descend / Upon my heart
(10) + +
(11) + + with
(12) + + like
(13) + + + of
(14) + + under
(1) The heavy weight of many a weary day (1, 24 (A))

(2) by its own weight (1, 625 (B)) <ACCOMPANIMENT>
(3) Even with a weight of pleasure (2, 178 (B))<COOPERATIVENESS>
(4) A weight (3, 419 (C))
(5) that weight (5, 6 (A))
(6) Beneath our life’s mysterious weight / Of pain and fear (5, 442-43
(A))
(7) no heavy weight (6, 36 (A))
(8) Against the weight of meanness, selfish caress, / Coarse manners,
vulgar passions (8, 454-55 (A))
(9) A weight of Ages (8, 703 (B))
(10) no ~ but weight of power (8, 705 (B))
(11) Power growing with the weight (8, 706 (B))
(12) there lay it like a weight (10, 252 (A))
(13) aught of heavier or more deadly weight, / In trivial occupations,
and the round / Of ordinary intercourse (11, 262-64 (A))
(14) under all the weight / Of that injustice (12, 102-03 (A))
2.2 How to read the chart

Let me explain how to look at the chart. Bracketed figures in the
leftmost side of the chart correspond to those of the phrases quoted
just below the chart itself. The sequence follows the order of
occurrence of ‘weight’ in The Prelude (1805 version), which is
comprised of 13 Books. Roughly speaking, one example of ‘weight’
appears in each Book.
The next group of three columns in the chart indicates, in terms of
semantics, or rather synonyms, whether each example is valuable or
not; in other words, favorable to the poet or not.
The following group of four shaded columns represents, in terms of
syntax, how each ‘weight’ co-occurs with ‘mind’; ‘heart’; ‘power’;
and ‘joy, delight, pleasure, and happiness’ in its own context. These
lexical items are all key words for understanding Wordsworth’s poetry.
Particularly ‘joy,’ ‘delight,’ ‘pleasure’ and ‘happiness’ are what
Josephine Miles (1942, rpt. 1965) calls the ‘vocabulary of emotion.’
‘Mind’ and ‘heart’ are the recipients that receive those emotions
strongly, that is to say, powerfully, i.e. ‘in a way that has a lot of
energy, power, or force’ (Spears: s.v. powerfully).
In the following three columns of the chart, in terms of syntax as well,
are represented: first, with what appositional lexical item each
‘weight’ occurs; second, with what preposition (also shaded) each
‘weight’ co-occurs; and last, with what verb each ‘weight’ combines.
Let us pay special attention to the prepositions listed in the right-hand
shaded column. They are prepositions which stand before ‘weight.’ I
have two points to make: one is that human beings have to endure
‘beneath’ and ‘under’ the heavy burden; the other is that relations
between human beings and these ‘weights’ are hostile, which is clearly
realized by ‘against.’
What I want to discuss earnestly here is the preposition, ‘by’ and
‘with.’ They are realized respectively in (2) ‘by its [joy’s] own weight’
and (3) ‘Even with a weight of pleasure.’ They both describe the
inner feelings of Wordsworth, that is to say, the poet’s joy and
pleasure. The prepositions ‘by’ and ‘with,’ entail the meaning of
<ACCOMPANIMENT> and <COOPERATIVENESS>. Both meanings are
favorable to human beings.
418 Ken Nakagawa
3. Fourteen examples
In order to ascertain whether the semantic classification above is
appropriate, let me show you concrete examples. I will examine the
three groups in order of occurrence: eight times for (A) negative
value, five times for (B) positive value, and one time for (C) neutral
value.
3.1 Examples of (A) negative value (8 instances)

Examples of negative value, those relevant to the idea that ‘life is a
burden’ are (1), (5), (6), (7), (8), (12), (13), and (14). All of the eight
examples show negative vectors in a downward direction. As I
mentioned at the beginning, the conception that LIFE IS A BURDEN is
universal to all mankind in all ages, so I will not deliberate further
about this group.
(1)
it is shaken off,
As by miraculous gift ’tis shaken off,
That burthen of my own unnatural self,
The heavy weight of many a weary day
Not mine, and such as were not made for me.
(1, 21-25)
(5)
…, it grieves me for thy state, O Man,
Thou paramount Creature! And thy race, while ye
Shall sojourn on this planet; not for woes
Which thou endur’st; that weight, albeit huge,
I charm away; but for those palms atchiev’d
Through length of time, by study and hard thought,
The honours of thy high endowments, there
My sadness finds its fuel.
(5, 3-10)
(6)
Though doing wrong, and suffering, and full oft
Beneath our life’s mysterious weight
Of pain and fear; yet still in happiness
Not yielding to the happiest upon earth.
(5, 441-44, the same as (0))
(7)
and [I] should have been
Even such, but for some personal concerns
That hung about me in my own despite
Perpetually, no heavy weight, but still
A baffling and a hindrance, a controul
(6, 33-37)
(8)
And thus
Was founded a sure safeguard and defence
Against the weight of meanness, selfish cares,
Coarse manners, vulgar passions, that beat in
On all sides from the ordinary world
In which we traffic.
(8, 452-57)
(12)
Now had I other business for I felt
The ravage of this most unnatural strife
In my own heart; there lay it like a weight
At enmity with all the tenderest springs
Of my enjoyments.
(10, 250-54)
(13)
There are in our existence spots of time,
Which with distinct pre-eminence retain
A vivifying Virtue, whence, depress’d
By false opinion and contentious thought,
Or aught of heavier or more deadly weight,
In trivial occupations, and the round
420 Ken Nakagawa
Of ordinary intercourse, our minds

Are nourished and invisibly repair’d,
A virtue by which pleasure is enhanced
That penetrates, enables us to mount
When high, more high, and lifts us up when fallen.
(11, 258-68)
(14)
… how much of real worth
And genuine knowledge, and true power of mind
Did at this day exist in those who liv’d
By bodily labour, labour far exceeding
Their due proportion, under all the weight
Of that injustice which upon ourselves
By composition of society
Ourselves entail.
(12, 98-105)
3.2 Examples of (B) positive value (5 instances)

Contrary to the above, I can give examples (2), (3), (9), (10), and (11)
as those with positive value which are favorable to human existence:
(2)
And if the vulgar joy by its own weight
Wearied itself out of the memory,
The scenes which were a witness of that joy
Remained, in their substantial lineaments
Depicted on the brain, and to the eye
Were visible, a daily sight;
(1, 625-30)
(3)
Oh! then the calm
And dead still water lay upon my mind
Even with a weight of pleasure, and the sky
Never before so beautiful, sank down
Into my heart, and held me like a dream.
(2, 176-80)
(9), (10), (11)

The very moment that I seem’d to know
The threshold now is overpass’d, Great God!
That aught external to the living mind
Should have such mighty sway! Yet so it was
A weight of Ages did at once descend
Upon my heart; no thought embodied, no
Distinct remembrances; but weight and power,
Power growing with the weight: alas! I feel
That I am trifling: [original italics]
(8, 699-707)
Now, let me look at the most Wordsworthian examples in citations (2)
and (3). Unlike the representative example (0) with ‘pain’ and ‘fear’
illustrated at the beginning of this essay, ‘weight’ here is used together
with ‘joy’ and ‘pleasure,’ which bears a positive meaning, not a
negative one.
3.2.1 How to intensify the degree of ‘pleasure’

I will make a special mention of citation (3). When one wants to
intensify the degree of ‘pleasure,’ what kind of expression will he or
she use? It would seem normal to modify ‘pleasure’ with an adjective
such as ‘great,’ ‘genuine,’ ‘real’ and ‘much.’ Instead, here, ‘weight’ is
used in the form of ‘a weight of pleasure.’ Wordsworth combines
‘weight’ with ‘pleasure.’ His purpose is to heighten the intensity of
‘pleasure,’ although there seems to be little semantic affinity between
‘weight’ and ‘pleasure.’
3.2.2 Finite verbs used in citation (3)

At the same time I have to point out the meanings of the finite verbs
used in this citation: ‘Oh! Then the calm / And dead still water lay
upon my mind / … and the sky / Never before so beautiful, sank
down / Into my heart, and held me like a dream.’ All these verbs,
especially the first two, combine suitably with ‘weight.’ The verbs
‘lie,’ ‘sink’ and ‘hold’ perform their functions most properly when
they collaborate with ‘weight,’ in other words, when they are given
added load, or ‘weight.’
422 Ken Nakagawa
3.2.3 ‘Even’ prefixed to ‘with-phrase’

Moreover, one cannot ignore the fact that ‘Even’ is prefixed to the
with-phrase. Oxford Advanced Learner’s Dictionary refers to ‘even’ as
‘used to emphasize something unexpected or surprising in what one is
saying or writing.’ When he wrote this phrase, Wordsworth must have
wanted to emphasize the expression (i.e. ‘with a weight of pleasure’)
as something unexpected or surprising. In addition, Collins Dictionary
of the English Language defines ‘even’ as ‘intensifier; used to suggest
that the content of a statement is unexpected or paradoxical (my
underline)’. There is something paradoxical in the combination of
‘weight’ and ‘pleasure.’ There is a contradiction in the combination of
these two words. This is an extremely marked linguistic phenomenon.
This is why Wordsworth must have inserted a word of emphasis:
‘even.’
3.2.4 Brief mention of (9), (10), and (11)

Next, in these examples I can see a conspicuous convergence of three
occurrences of ‘weight.’ Here, too, ‘weight’ is charged with a positive
meaning which carries favorable implications. Wordsworth is
overwhelmed by ‘A weight of Ages,’ when the poet sets foot in
metropolis London for the first time in his life. His heart is
overpowered by the human lives which have lived assiduously and
strenuously there for the past thousands of years.
3.3 Example of (C) neutral value (1 instance)

Finally, I give an example of (C) neutral value, of which there is only
one example:
(4)
A weight must surely hang on days begun
And ended with worst mockery: be wise,
Ye Presidents and Deans, and to your Bells
Give seasonable rest;
(3, 419-22)
Wordsworth feels outrage over the archaic system of management that
has not changed over the years in Cambridge University. He does not
want to be forced to attend formal morning and evening services
which have remained unchanged for a long time. In this case he insists
that such an old custom should be abolished by putting on it a
‘weight.’
4. Observation
Thus far I have cited fourteen examples of ‘weight’ that appear in The
Prelude and divided their meanings into three major categories. The
first category (A) negative value is the stress that unavoidably
accompanies human life. The heavy burden and strong pressure
inevitably befalls human beings who come into this world and lead a
social life. The perception that LIFE IS A BURDEN is common to people
of all ages and countries. This usage of ‘weight’ with an unfavorable
and uncooperative sense accounts for as many as eight out of fourteen
examples. Therefore, the first group with negative meaning can be
said to be the normal and ordinary usage of ‘weight.’ This is why little
explanation has been made on the (A) usage.
Incidentally, I have put a question mark in the diagonal column of (B)
positive meaning on the chart, because I was unable to find a
corresponding synonym when I was making the chart. Why? It occurs
to me that these word combinations, that is to say, ‘weight’ with ‘joy’
and ‘pleasure’ are particularly peculiar to Wordsworth, and that is the
very reason why I can not think of a synonym. If the combination of
‘weight’ with ‘pleasure’ is a common practice in English, perhaps I
could have hit upon a synonym commonly used with this positive
meaning.
Now in order to examine the question whether my speculation is right
or wrong, I investigated whether the same collocation might be found
in other writers. Take Shakespeare (1564-1616), for example, who
flourished prior to Wordsworth (1770-1850). According to a
concordance of Shakespeare’s work there are nine examples under the
item ‘weight of ~’ in his works. None of these, however, could be
considered positive. Tracing further back to Chaucer (1340?-1400), I
could not find any such positive combinations in a concordance of his
work, either.
Extending the scope of my search, I consulted the Gutenberg Files.
This file, as of August 1997, contains 550 copies of books, both verse
and prose. It includes twentieth century novels, to say nothing of the
Bible and Shakespeare. In total, 1281 examples which include ‘weight
of ~’ are found, and out of these there are but four instances with
‘pleasure’ and ‘joy.’ These concrete examples are the following:
424 Ken Nakagawa
the whole weight of her displeasure, was obvious.

(Hardy, The Return of the Native, 1878)
the want of an army saved France from the full weight of his
displeasure.
(Schiller, The History of the Thirty Years’ War,
1789? Translated by Morrison)
would let Plato feel the weight of her displeasure.
(Chesnutt, The House Behind the
Cedars, 1900)
bear the weight of a secret joy or of a secret sorrow,
(Defoe, Moll Flanders, 1722)
As one can see in the above quotations, ‘weight’ is combined,
however, not with ‘pleasure’ but with ‘displeasure.’ When it comes to
‘joy,’ it unites with ‘secret joy,’ not with ‘genuine joy.’ In either case,
‘weight’ does not tie well with ‘pleasure’ or ‘joy.’ ‘Weight’ collocates
with the words of a negative meaning (i.e. prefixed ‘pleasure’ and
secret ‘joy’). What these four examples reveal is that ‘weight’ is hard
to combine directly with either ‘pleasure’ or ‘joy.’ In everyday
English, ‘weight’ is semantically in harmony with words carrying
negative, passive, and regressive meanings. In short, ‘weight’ does not
go well with ‘joy’ and ‘pleasure.’
Now let me consider the matter from another view point. To use
Lakoff and Johnson’s (1980: 14-21) ‘orientational metaphor’ <UP>
and <DOWN>, ‘weight’ usually goes with ‘down.’ In Wordsworth’s
case, however, there is one exception: ‘a weight of pleasure.’
‘Pleasure’ is <UP> and ‘weight’ is <DOWN>. The two words clash in
meaning. This idiosyncratic or oxymoronic combination of words
would appear to be a uniquely Wordsworthian turn of phrase.
Furthermore, in order to investigate words that may fit in the slot of
‘weight of ~,’ I referred to Collins COBUILD Wordbanks Online. I
found only one ‘weight of joy’ combination in all 525 examples of
‘weight of ~.’ This, from the context, seems to be part of a translation
from a Greek version of Oedipus Rex. However, the combination of
‘weight’ with ‘pleasure’ is nowhere to be found in the 525 examples.
Other evidence to reinforce my argument comes from the Nineteenth-
Century Fiction: Full-Text Database. It reveals 646 examples of the
‘weight of ~’ combination, but ‘weight of pleasure’ cannot be found.

On the other hand, Wordsworth uses a combination of ‘Even with a
weight of pleasure’ (2, 178) contrary to the above-mentioned general
tendency of popular usage. He, as it were, forcibly unites ‘weight’
with ‘pleasure’ and that directly, that is to say, without the intervention
of ‘dis-’ and ‘secret.’ He combines in this way: ‘weight of Ø pleasure.’
For the so-called poet of Joy, ‘pleasure’ must not be a ‘secret’ one but
must be a ‘true’ and ‘genuine’ one.
This world is filled with heavy pressure and distress. Indeed the poet
captures that aspect of the world in a word ‘weight.’ But he
compensates a downward vector component Ļ of ‘weight’ with
upward vector component Ĺ of ‘joy.’ In addition, he overwhelms such
a ‘weight’ lying heavily on human beings with his feeling of ‘pleasure’
and ‘joy.’
5. Conclusion
It is apparent from the discussion above that Wordsworth is the poet
who boldly connected ‘weight’ and ‘pleasure’ with the use of the
particle ‘of.’ He captures the strong feeling of joy not from the
direction of <intensity> but from the direction of <weight>. He is the
first poet in English literature that realized ‘the weight of pleasure’
deep down in his heart. Therefore, I can conclude that this oxymoronic
word combination ‘Even with a weight of pleasure,’ which seems to
break the selectional restriction between lexical items, is actually an
expression peculiar to Wordsworth, poet of nature and joy.
References
Lakoff, G., and M. Johnson 1980. Metaphors We Live By. Chicago: The University of
Chicago Press.
Lakoff, G., and M. Turner 1989. More than Cool Reason: A Field Guide to Poetic
Metaphor. Chicago: The University of Chicago Press.
Miles, J. [1942] rpt. 1965. Wordsworth and the Vocabulary of Emotion. New York:
Octagon Books.
Texts
Wordsworth, W. 1952. The Poetical Works of William Wordsworth (2nd edn). Vol. 2.
E. de Selincourt (ed.) Oxford: Clarendon Press.
__. 1959. William Wordsworth: The Prelude or Growth of a Poet’s Mind (2nd edn). E.
de Selincourt (ed.) and H. Darbishire (rev). Oxford: Clarendon Press.
426 Ken Nakagawa
Dictionaries & Concordances

Blake, N. F., D. Burnley, M. Matsuo and Y. Nakao (eds). 1994. A New Concordance
to The Canterbury Tales Based on Blake’s Text Edited from the Hengwrt
Manuscript. Okayama: University Education Press.
Collins COBUILD Wordbanks Online.
Crowther, J. et al. (eds). 2000. Oxford Advanced Learner’s Dictionary of Current
English (6th edn). Oxford: Oxford University Press.
Gutenberg Files.
Hanks, P. et al. (eds). 1979. Collins Dictionary of the English Language. London:
Collins.
Jeffares, A. N., and M. Gray (eds). 1995. Collins Dictionary of Quotations. Glasgow:
Harper Collins Publishers.
McArthur, T. 1981. Longman Lexicon of Contemporary English. London: Longman.
The Nineteenth-Century Fiction: Full-Text Database.
The Oxford Dictionary of Quotations (3rd edn). 1979. Oxford: Oxford University
Press.
The Oxford English Dictionary (2nd edn) on CD-ROM Version 3.0. 2002. New York:
Roget’s II The New Thesaurus. 1980. Boston: Houghton Mifflin Company.
Shakespeare, W. 1970. A Complete and Systematic Concordance to the Works of
Shakespeare. Vol. 6. M. Spevack (ed.). Hildesheim: Georg Olms.
__. 1997. The Riverside Shakespeare (2nd edn). G. B. Evans (ed.). Boston: Houghton
Mifflin Company.
Spears, R. A. 1998. NTC’S American English Learner’s Dictionary. Lincolnwood
(Chicago): NTC Publishing Group.
Watanabe, T., E. R Skrzypczak. and P. Snowden. 2003. Kenkyusha’s New Japanese-
English Dictionary (5th edn). Tokyo: Kenkyusha.
(Im)Politeness in Dramatic Dialogue: Understanding
Face-Attack in Shakespeare’s Othello
Rachel S Toddington
University of Huddersfield, UK
Abstract
Within this paper I consider how Brown and Levinson’s (1987) model of politeness –
i.e. the mitigation of Face Threatening Acts (FTAs) and Culpeper (1996, 2005) and
Culpeper et al.’s (2003) model of impoliteness i.e. the communicative strategies used
to deliberately attack face, may be manifested within the context of drama – in this
case Shakespeare’s Othello. As a tragedy, Othello is rich in scenes of confrontation
between characters, and hence provides a good model in which to analyse (im)
politeness phenomena.
Central to Culpeper’s notion of impoliteness within his 2005 paper-an analysis of the
‘exploitative’ quiz show The Weakest Link, in which face-damage is part of the
show’s format-is the assertion that ‘impoliteness is not unintentional’. However, I
argue that when his model is applied within the context of a play – in this case
Shakespeare’s Othello – an anomaly within his definition becomes apparent. This is
due to the two-tier discourse structure of the play which allows for differing
interpretations of face attack for the audience and the characters, which ultimately
shows how offence can be perceived as unintentional.
I introduce the term Discourse Disjunctive Politeness to account for this type of
‘incidental’ offence and propose a revised definition of his 2005 model which takes
the context of drama into account.
Keywords: politeness; Brown and Levinson; face attack; Shakespeare; confrontation.
1. Introduction
Although much research has concentrated on how and why people are
linguistically ‘polite’ i.e. ‘how communicative strategies are employed
to promote or maintain social harmony in interaction’ (Culpeper 1996:
349), relatively little by comparison appears to have been undertaken
to account for ‘impoliteness’, or, ‘communicative strategies designed
to attack face, and thereby cause social conflict and disharmony
(Culpeper 2005: 38).
Studies of linguistic politeness within drama have been useful in
highlighting critical issues such as plot and characterisation (Bennison
1993, Culpeper 2001, Leech 1992, Simpson 1989). Conflict in drama
is a device which is used habitually in order to generate entertainment
428 Rachel S Toddington
and anticipation for the audience; after all, the idea of a ‘happy
ending’ (and in some cases not-so-happy ending) would not be as
climactic without the equilibrium - disequilibrium-equilibrium outline
which characterises a dramatic plot (Bremond 1966, 1973, cited in
Culpeper et al. 1998).
As Culpeper (2005) notes, the fact that audiences ‘enjoy’ conflict in a
performance can be explained via a consideration of the pleasure that
the members gain as they view conflict from a distance and hence feel
safe:
it is pleasant, when on the great sea the winds are agitating the
waters, to look from the land on another’s great struggle, not
because it is delectable joy that anyone be distressed, but
because it is pleasant to see what ills you yourself are free from1
(Lucretius (De Rerum Natura, Book II, 1-4 cited in Culpeper
2005: 45)
It is this feeling of ‘safety’ which allows us to ‘enjoy’ conflict. As
such, plays are a particularly rich genre in which to explore the
phenomena of (im) politeness, and as Culpeper (1998) has pointed
out:
Impoliteness is a type of aggression, and aggression has been a
source of entertainment for thousands of years (Culpeper 1998:
86)
Models of (im) politeness cannot just be tested on a speaker’s words,
it is sometimes necessary to know their unspoken thoughts – more
specifically their intentions – and Shakespeare allows us access to
these through the use of soliloquies and asides (Brown and Gilman
1989: 171). Culpeper et al. maintain that knowing a person’s intention
is ‘a key difference between politeness – [intention to support face],
and impoliteness’ – [intention to attack it] (Culpeper et al. 2003: 1550)
It is this notion of intentionality which I would like to expand on
within this paper. In particular, it is the notion of intentionality within
Culpeper’s 2005 definition of impoliteness which I believe becomes
problematic when considered within the context of a two tier
discourse play such as Othello. This is because of the
audience/reader’s privileged position as ‘overhearers’ which allows us
to ‘see’ a character’s real intention. Ultimately, this enables us to
judge whether an utterance within the context surrounding it is polite,
(Im)Politeness in Dramatic Dialogue 429
impolite or as Watts (2003) maintains politic i.e. simply appropriate to

the situation of context at that time.
The two-tier discourse structure of plays in general enables the
dramatic irony to be set up, but as I aim to show in an analysis of
Othello, may also result in interpretations of face-attack (impoliteness)
being different for the audience and the characters.
Because Culpeper’s definition of impoliteness is built around Brown
and Levinson’s (1978, 1987) model of politeness, a brief overview of
their influential strategies is necessary in order to lay down the
groundwork of what is to follow in this paper.
2. Brown and Levinson’s model of politeness

Following on from Goffman’s (1967) concept of ‘face’, and
incorporating Grice’s (1975) Cooperative Principle (CP), Brown and
Levinson (1987: 6) maintain that:
It is the mutual awareness of ‘face’ sensitivity, and the kinds of
means-ends reasoning that this induces, that together with the
CP allows the inference of implicatures of politeness.
Their model – arguably the most influential study of politeness to have
emerged in the last twenty years – argues politeness to be
Like formal diplomatic protocol (for which it must surely be the
model), [and] presupposes that potential for aggression as it
seeks to disarm it, and makes possible communication between
potentially aggressive parties (1987, 1)2
They take Goffman’s (1967) concept of ‘face’ which he defines as
The positive social value a person effectively claims for himself
by the line others assume he has taken during a particular
contact. Face is an image of self delineated in terms of approved
social attributes’ (1967, 1)
and construct it around ‘two basic socio-psychological wants’
(Culpeper 1998: 84). Hence, positive face is described as:
The want of every member that his wants be desirable to at least
some others…in particular, it includes the desire to be ratified,
understood, approved of, liked or admired (1987: 62)
Negative face as is described as:

The want of every competent adult member that his actions be
unimpeded by others (1987: 62)
Damage to either face is labelled a Face Threatening Act, and within
their model various strategies may be employed by a speaker in order
to reduce or mitigate such acts.
2.1 Examples of Brown and Levinson’s Politeness Strategies in

Othello
2.1.1 Bald on record politeness
This super-strategy is associated with the lowest amount of face threat
in which the FTA is performed ‘in the most direct, clear, unambiguous
and concise way possible’ (1987, 69).There is no need to consider the
hearer’s face wants because as Brown and Levinson maintain ‘the
speaker does not fear retribution from the addressee [because it is
used] in the interests of urgency or efficiency’.
Othello
What is the matter here?
Montano
Zounds, I bleed still
I am hurt to th’ death: he dies! [Lunges at Cassio]
Othello
Hold for your lives!
(Othello act 2 scene 3 lines 157-161)
2.1.2 Positive Politeness

This strategy pays attention to the hearer’s positive face wants and
indicates ‘in some respects, S wants H’s wants’ such as:
Use of in-group identity markers:
Cassio (to Iago):
Do, good my friend
(Othello act 3 scene 1 lines 29)
Notice admirable qualities:
Montano (to Othello) Worthy Othello, I am hurt to danger
(Othello act 2 scene 3 line 93)
Exaggerate sympathy, approval etc:

Emilia (to Cassio):
Good morrow, good lieutenant. I am sorry
For your displeasure
Thus, positive politeness is:
a ‘general appreciation of the hearer’s wants [and] will serve to
counter-balance the specific imposition…[and] in using them
one indicates a wish to be closer to the addressee’ (Culpeper
2001: 244)
2.1.3 Negative Politeness

This strategy indicates that the speaker is paying attention to the
hearer’s negative face wants and is reluctant to interfere with the
hearer’s freedom of action. Examples of this would be:
Giving deference:
Othello (to the Duke and Senators)
Most potent, grave and reverend signiors, my
Very noble and approved good masters
(Othello act 1 scene3 lines 77-78)
Go on record as incurring a debt
Cassio (to Emilia):
I am much bound to you
(Othello act3 scene 1 line 58)
Apologize. Admit the impingement, express reluctance, ask
forgiveness:
Desdemona (to Othello):
O, banish me, my lord, but kill me not!
(Othello act 5 scene 2 line 77)
2.1.4 Off-record
This strategy involves the use of indirectness and ambiguity in the
performance of the FTA. In typically Gricean terms (1975), it involves
the flouting of a maxim; the reason being ‘so that the actor cannot be
held to have committed himself to one particular intent’ (Brown and

Levinson 1987: 69), and involves the generation of a trigger which
will make the addressee look for an additional level of meaning (See
examples 4 and 5)
During interaction, Brown and Levinson argue that people
instinctively assess the potential for face damage by considering three
sociological variables:
1) The ‘social distance’ (D) of S and H (a symmetric relation)
2) The relative ‘power’ (P) of S and H (an asymmetric relation)
3) The absolute ranking (R) of impositions in the particular
culture (Brown and Levinson 1987: 74)
Hence their equation is as follows (where x is the FTA):
Wx = D(S, H) + P (H, S) + Rx
This means that the weightiness3 (Wx) of the face threatening act
depends on the distance between the speaker and the hearer, plus the
relative power of the hearer over the speaker, plus the degree to which
the FTA x is rated an imposition in that culture.
There are numerous issues concerning the validity and accessibility of
Brown and Levinson’s ideas, not least the fact that their politeness
strategies are based on single grammatical utterances and do not take
context or extended discourse into account. Recent work by Watts
(2003) highlights the ambiguousness of their strategies when applied
to different contexts. He maintains that peoples’ notions of politeness
are negotiated through a ‘discursive struggle’ depending on their own
ideas of the social value of politeness. Ultimately, he suggests that the
‘work’ we do when we decide what strategy to adopt is actually a
‘facework’ strategy rather than a politeness one because
[the] facework strategies are by no means always associated
with linguistic politeness (Watts 2003: 93)
Culpeper et al. point out that because Brown and Levinson concern
themselves only with linguistic form, ‘impolite implicatures can slip
through their framework’ (Culpeper et al. 2003: 1555 my emphasis)
A detailed analysis of Brown and Levinson’s politeness model
including its deficiencies is not within the remit of this paper;
however, one final drawback needs pointing out, namely the model’s
inadequacy in accounting for deliberately offensive behaviour. Their

bald, on-record strategy appears at first sight to accommodate
impolite utterances since they are couched ‘in the most direct, clear,
unambiguous and concise way possible’; however, this strategy is first
and foremost a politeness one i.e. with the intention of ‘disarm[ing]
that potential for aggression’ (1987: 1 my emphasis).
3. Culpeper’s model of impoliteness

Whereas Brown and Levinson’s model is primarily about conflict
avoidance, Culpeper’s impoliteness strategies show how interlocutors
apply various linguistic devices in order to deliberately attack an
addressee’s face.
In a reformulation of his (1996) definition of impoliteness, Culpeper
(2005) maintains that:
Impoliteness comes about when: (1) the speaker communicates
face-attack intentionally, or (2) the hearer perceives and/or
constructs behaviour as intentionally face-attacking, or a
combination of (1) and (2). (Culpeper 2005: 38)
Taking Brown and Levinson’s politeness strategies, Culpeper’s model
may be considered a parallel structure; although his superstrategies are
opposite to Brown and Levinson in terms of orientation (they are
designed to attack and not maintain face) they are not necessarily
opposite pragmatically. For example, from a Gricean point of view,
bald on record politeness strategies have off-record strategies as the
opposite (where the FTA is committed via an implicature).
3.1 Impoliteness Strategies4

Bald, on-record Impoliteness – this occurs where it is the intention
of the speaker to attack the face of the hearer, and where there is much
face at stake.
Positive Impoliteness – the use of strategies designed to damage the
addressee’s positive face wants (‘ignore’, ‘snub the other’, ‘exclude
the other from the activity’, ‘disassociate from the other’, ‘be
disinterested, unconcerned, unsympathetic’, ‘use inappropriate
identity markers, ‘use obscure or secretive language’, ‘seek
disagreement’, ‘make the other feel uncomfortable’ ( e.g. do not avoid
silence, joke or use small talk), ‘use taboo words’, ‘call the other
names’).
Negative Impoliteness – The use of strategies designed to damage the
addressee’s negative face wants (‘frighten’, ‘condescend’, ‘scorn’ or
‘ridicule’,’ invade the other’s space’, ‘explicitly associate the other
with a negative aspect’, ‘put the other’s indebtedness on record’,
‘hinder’ or ‘block the other-physically or linguistically’)
Sarcasm or mock politeness – The use of politeness strategies that
are obviously insincere and thus remain surface realisations. Sarcasm
(mock politeness for social disharmony) is clearly the opposite of
banter (mock for social harmony). This strategy is not based not on
Brown and Levinson, but on Leech’s (1983) Irony Principle, and
highlights how non-observance of Grice’s maxims can generate
impolite implicatures as in the following example taken from Leech
(1983):
A: Geoff has just borrowed your car.
B: Well, I like that!
Leech explains the ‘irony’ here is as an ‘exploit[ation of] the
Politeness Principle (PP) in order to uphold, at a remoter level, the
CP. A flouting of the Maxim of Quality shows that ‘what B says is
polite to Geoff and is clearly not true. Therefore, what B really means
is impolite to Geoff and true’ (Leech 1983: 83).
As a point, it needs noting here that Leech’s predominantly content
based approach to politeness can be used alongside Brown and
Levinson’s form based approach to complement the two models –
something which Culpeper points out in his (1996) study.
Withhold politeness – Keep silent or fail to act where politeness work
is expected. Tanaka (1993) has pointed out that saying nothing within
a situation ‘where there is a strong expectation that something will be
said is in itself a massive FTA’ (Tanaka 1993 cited in Thomas 1995:
175).
4. (Im) politeness in dramatic dialogue

In this section I look at how (im)politeness is realized in extended
dialogue beginning with a single character and a single strategy,
before moving on to look at the idea of how combinations of strategies
(see Culpeper et al. 2003: 1561) work to exacerbate face attack across
exchanges between characters.
For this study I used the Arden 3rd edition of Shakespeare’s Othello.
Tragedies typically have scenes which are rich in conflict between
various characters i.e. scenes in which there are ‘general
disagreements in interaction which are displayed by the occurrence of
some sort of opposition to an antecedent event’ (Corsaro and Rizzo
1990, 26 cited in Culpeper et al.: 1547).
There is also a variety of social classes between the characters and the
complex themes of manipulation, jealousy and hate. These are used as
an intriguing exploration with respect to the power and social role of
the characters.
The notion of ‘power’ needs defining at this point: the characters of
Iago, Brabantio and Othello all possess ‘power’; taking Wartenberg
(1990: 5, 28) and Barnes’ (1988: 6) concept of ‘power-over’ and
‘power-to’ I believe that the characters of Brabantio and Othello
exhibit ‘power-over’ due to their hierarchical position in society,
whilst the character of Iago exhibits ‘power-to’ due to his linguistic
dexterity and manipulation of the other characters. Locher (2003)
refers to this type of ‘power-to’ as ‘the ability an individual may
(temporarily) possess and use [and is what] many linguists are
interested in [when] examining the potential power an interactant has
when entering into a speech event’ (Locher 2004: 11 emphasis in
original)
4.1 The story of Othello

The story concerns the Moorish5 General (Othello), who secretly
marries the daughter (Desdemona) of a Venetian Senator (Brabantio).
Othello’s supposedly trustworthy ensign6 (Iago) is jealous and
resentful at being passed over for promotion by Othello to a lesser
experienced soldier (and Othello’s personal friend) Cassio. Roderigo,
Iago’s dupe and unsuccessful suitor of Desdemona, also bears a
grudge against Othello and assists Iago in his machinations. In order
to vent his hatred and resentment against Othello and Cassio, Iago
schemes and plots to bring about the downfall of the two men, using
his ‘honest’ reputation as a shield, eventually destroying everything
the two men hold dear.
Example [1]
Context: Angry at being passed over for promotion by Othello, Iago
exacts revenge by hysterically informing Brabantio of the secret
marriage between his daughter and the Moor – news which Iago
knows will incense and frighten Brabantio.
Iago
Zounds, sir, you’re robbed, for shame put on your
gown!
Your heart is burst, you have lost half your soul,
Even now, now, very now, an old black ram
Is tupping your white ewe! Arise, arise,
Awake the snorting citizens with the bell
Or else the devil will make a grandsire of you,
Arise I say!
(Othello Act 1 Scene 1, lines 85-90)
Bald, on-record impoliteness occurs throughout this scene with Iago’s
imperative commands, which are issued ‘in a direct, clear,
unambiguous and concise way in circumstances where face is not
irrelevant or minimized’ (Culpeper 1996: 356).
This should not be mistaken for Brown and Levinson’s bald, on-
record, which is specifically a politeness strategy and maintains that
this type of FTA ‘will be done…where the danger to H’s face is very
small’ (Brown and Levinson 1987: 69). As Iago is all too aware, the
power differential between himself and Brabantio makes this FTA
enormous and explains why Iago is so keen to preserve his anonymity
whilst standing in the shadows under Brabantio’s window. As
Culpeper (1996) notes:
The greater the imposition of the act, the more powerful and
distant the other is, the more face-damaging the act is likely to
be (Culpeper 1996: 357).
Iago’s use of the personal pronouns ‘you’ towards Brabantio, along
with the contemptuous manner in which his speech is conveyed, are
all negatively impolite, as is his frightening prediction at the end – the
devil will make a grandsire of you. In terms of positive impoliteness,
Iago’s description of Othello as an old black ram is indicative of
‘inappropriate identity markers’ (Culpeper et al. 2003: 1555), but the
fact that he makes Brabantio feel uncomfortable at all is also a huge

positive impoliteness strategy.
Iago’s repeated imperatives for Brabantio to arise can be seen as
parallelism, a strategy which occurs when ‘words, grammatical
structures, intonational contours [are used as] a pragmatic strategy’
(Culpeper et al. 2003: 1561). Such a device serves not only to
exacerbate the imposition on Brabantio’s negative face (by not giving
him chance to speak), it also, as Holmes (1984) points out:
Serves as a rhetorical device to increase the force of the
repeated speech act (Holmes 1984 cited in Culpeper et al. 2003:
1561)
As a result, Iago’s impolite strategies are significantly increased with
both positive and negative strategies being used.
A point to make here for both politeness and impoliteness is that
context is crucial; a consideration of single grammatical utterances as
‘intrinsically threatening face’ (Brown and Levinson (1987, 65)
cannot be accepted. As Fraser and Nolan (1981) correctly point out:
No sentence is inherently polite or impolite…it is not the
expressions themselves but the conditions under which they are
used that determine the judgment of politeness. (Fraser and
Nolan 1981: 96)
As such, impolite acts can take place over a series of turns as is
demonstrated in the next example.
Example [2]
Context Following Cassio’s disgraced demotion from his position as
lieutenant, Iago has been insinuating to Othello that Cassio and
Desdemona are committing adultery behind his back. Knowing the
agitated state of Othello’s mind at this news, Iago then tells Othello
that he has seen Cassio wiping his beard with Desdemona’s
handkerchief. Othello, determined that the procurement (or not) of
this handkerchief will prove his wife’s loyalty or deceit, demands that
she show it to him. Desdemona, unaware of Othello’s agitation, is
more concerned with the reinstatement of Cassio as lieutenant.
Othello
Fetch me the handkerchief, my mind misgives.
Desdemona
Come, come.
You’ll never meet a more sufficient man.
Othello
The handkerchief!
Desdemona
I pray, talk me of Cassio.
Othello
The handkerchief!
Desdemona
A man that all his time
Hath founded his good fortunes on your love,
Shared dangers with you-
Othello
The handkerchief!
Desdemona
I’faith, you are to blame
Othello
Zounds!
Exit.
In this scene, Othello’s ‘disinterested’, ‘unconcerned’ and
‘unsympathetic’ frame of mind with regards to Cassio’s plight is
damaging to Desdemona’s positive face wants (Othello does not want
what Desdemona wants), plus his interruptions and repeated demand
for the handkerchief can be seen as a ‘linguistic block’ – a negative
strategy – in which he attempts to prevent Desdemona from talking
about Cassio by ‘hogging’ the conversational floor (Culpeper et al.
2003: 1561). However, a secondary (and potentially more damaging)
implication from this type of negative impoliteness strategy is also
apparent:
An interruption may, in specific contexts, attack negative face

by impeding someone, but it may also imply that the
interuptee’s opinion wasn’t valued-a positive face issue.
(Culpeper 2005: 42)
Positive impoliteness (the secondary implication) would be more
damaging to Desdemona because it interferes with her self worth and
feelings of security with Othello.
By shouting at Desdemona7, Othello’s impoliteness is intensified
whilst at the same time making Desdemona aware of his anger. As
Culpeper et al. point out, this prosodic strategy is not just about self-
expression; it is used to make the other person feel responsible for the
extreme emotional state of the addresser (Culpeper et al. 2003, 1573).
As a combination of these strategies, Othello’s behaviour is
significantly foregrounded, and as such ‘it increases the imposition
upon the target and/or emphasizes the negative attitude of the speaker
towards the target (Culpeper et al. 2003, 1561).
The fact that Desdemona has no idea how significant Othello’s request
is (and the audience does) adds to the increasing tension in this scene,
and results in a climactic build up for the plot as it effectively seals
Desdemona’s fate, and it highlights the metamorphosis that Othello is
going through from loving husband into jealous monster. Ultimately,
these strategies demonstrate how impoliteness can be seen as ‘crucial
to the construction of [Othello’s] character (Culpeper 2001: 260).
Desdemona’s words can also be seen as (an albeit unintentional)
attack on Othello’s positive face wants because she repeatedly ignores
Othello’s demands for the handkerchief, thereby failing to attend to
H’s ‘wants…goods, beliefs or values’ (Brown and Levinson 1987:66)
As a reader or member of the audience we know that Desdemona is
innocent of any wrong doing, and is therefore ignorant of Othello’s
tortured state of mind. Her words therefore can be interpreted
differently from the way Othello interprets them. Her remark at the
end of this exchange ‘I’faith, you are to blame’ is interpreted by
Othello with the shocked exclamation: ‘zounds’! because by
defending herself she inadvertently produces the secondary effect of
actually attacking Othello’s positive face. As well as the shock of
interpreting Desdemona’s words as critical towards him (and therefore
in breach of the deferment he expects from her as his wife), this

unintentional face attack contributes to Othello also inferring (false)
information from her linguistic behaviour i.e. that she is in fact guilty
of being deceitful with Cassio.
The process of inferring causal information from someone’s behaviour
is known in social psychology terms as attribution (Culpeper 2001:
115). At this point in the play, Othello (who is nearly half-mad with
tormented thoughts about his wife and Cassio), is only concerned with
one thing: the procurement of the now symbolic handkerchief, the
presenting of which will provide Othello with the proof of
Desdemona’s loyalty and innocence. Therefore, as Desdemona
desperately side-steps Othello’s insistence for the handkerchief,
believing him to be side-stepping the issue of Cassio’s reinstatement,
Othello’s fears are growing exponentially. All this of course adds to
the increasing tension of the scene, because as the reader or audience
we are aware via the dramatic irony (which is created by the discourse
structure of the play) of the true state of affairs and of Othello’s
murderous thoughts.
As Short (1996: 169) points out, the discourse structure of drama
typically comprises at least two levels: the author-audience/reader
level, and the character-character level. The discourse of characters is
‘embedded’ in the higher level which effectively allows an audience
or reader to ‘listen in’ to conversations between the characters.
What this means is that Desdemona cannot (for the audience) be
interpreted as being impolite, because we know her offence was
unintentional. It was produced as a by-product of defending herself
against Othello’s own FTAs:
Defensive strategies may have, to some degree, the secondary
[implication] of offending the speaker of the original
impoliteness act (Culpeper et al. 2003, 1563)
Within everyday interaction, secondary offences such as this may or
may not be intentional. I maintain that within the context of a two-tier
discourse structured play such as Othello, where the audience knows
more of the fictional world than the characters, such an offence as
produced by Desdemona cannot be seen as intentional and therefore
cannot be considered impolite – even though Othello has perceived it
as such.
She appears to have committed what Goffman termed a ‘faux pas’ i.e.
[her] offence seems to be unintended and unwitting (Goffman
1967, 14)
As Culpeper et al. point out (2003: 1551), the concept of accidental
offence is captured within a politeness framework and as such, I
would like to introduce the term Discourse Disjunctive Politeness to
account for this type of unintentional offence.
However, the fact that unintentional offence can be perceived throws
up an inconsistency with Culpeper’s (2005) definition, a reminder of
which is perhaps called for at this stage:
Impoliteness comes about when: (1) the speaker communicates
face-attack intentionally, or (2) the hearer perceives and/or
constructs behaviour as intentionally face-attacking, or a
combination of (1) and (2). (Culpeper 2005: 38)
The problem appears to hinge on the word intention. The above
example shows how impoliteness can be interpreted when none was
intended. It is also the case that impoliteness can be intentionally
conveyed and yet not be interpreted as such, as in the following
example:
Example [3]
Context: After the demotion of Cassio in act 2, Iago insists on
referring to Cassio as lieutenant:
Iago
What, are you hurt, lieutenant?
Cassio
Ay, past all surgery.
Iago
Good-night, lieutenant, I
Must to the watch.
Cassio
Good-night, honest Iago.
Iago
How do you now, lieutenant?
Cassio
The worser, that you give me the addition
Whose want even kills me.
Iago
O me, lieutenant! What villains have done this?
Cassio
I think that one of them is hereabout
And cannot make away.
The form of Iago’s words within these examples appears positively
polite – he is ‘claiming common ground’ by showing ‘interest’,
‘approval’ and ‘sympathy’ towards Cassio who also interprets them as
such. We can tell this by Cassio’s responses, the importance of which
Culpeper et al. note when they maintain that
The response to an utterance can reveal much about how that
utterance is to be taken (Culpeper et al. 2003: 1562)
The function of Iago’s words however is very different, and show how
FTAs can be conveyed indirectly and hence
Some impolite utterances are far from the directness associated
with bald, on-record. (Culpeper et al. 2003: 1549).
Cassio has no reason to interpret Iago’s words as insincere; Iago’s
‘honesty’ serves as a mask which enables him to perform a variety of
face attacks on the rest of the characters without them realising this.
As audience/readers though, we know of Iago’s contempt and hatred
for Cassio, and we can interpret Iago’s words accordingly.
Rudanko (2005) identifies two types of speaker intention when
interpreting a character’s actions. Iago’s utterances in example [3] are
what Rudanko calls an ‘overt first order intention’; it is an intention
about the world which Iago intends Cassio to recognize, i.e. Iago
intends Cassio to recognize that Iago is his ‘honest’ friend and staunch
supporter. Iago’s ‘second order intention’ is to make Cassio recognize
his first order intention (i.e. to believe Iago is genuine) – and Cassio’s
words indicate that he does.
However, in our role of audience or reader, we know that Iago’s
agenda is completely different. Iago’s true first order intention is
actually to discredit Cassio in Othello’s eyes, and his true second
order intention is to prevent any of the other characters (save
Roderigo) from recognizing this. In Rudanko’s terminology then,
Iago’s ‘interactional move’ means that Cassio believes that Iago is his
genuine friend and supporter. (Rudanko 2005: 6)
However, as well as appearing positively polite, Iago’s repeated
address to Cassio as lieutenant can also be viewed as the impolite
superstrategy ‘sarcasm or mock politeness’ (Culpeper 2005, 42). Iago
uses ‘inappropriate identity markers’ – a positive impoliteness strategy
(Culpeper 1996: 357), which are an attack on Cassio’s positive face
wants, but also Iago’s repeated use of this address indicates his
contempt and scorn for Cassio, and as such, is indicative of a negative
impoliteness strategy. The fact that negative impoliteness is occurring
here at the same time as positive impoliteness indicates that a multiple
strategy is in effect. As Culpeper et al. point out
as with all politeness phenomena, impoliteness does not simply
arise from any one particular strategy, but is highly dependent
on context (Culpeper et al. 2003: 1555).
Also, by repeatedly addressing Cassio as lieutenant, Iago is constantly
reminding Cassio of what he has lost (his position of lieutenant, but
more importantly Othello’s high regard for him).What Iago is really
doing is a slow, communicative manipulation of Cassio’s positive and
negative face and his repetitions are maximising the cost to him, an
inversion of Leech’s ‘Tact Maxim’ which states that we should
‘minimise the expression of beliefs which express or imply cost to the
other’ (Leech 1983: 132).
For the audience, Iago’s words can be seen as being a huge violation
of the Maxim of Quality ‘do not say what you believe to be false’. In
Gricean terms it is the ‘unostentatious non-observance’ of a maxim
with the intention to ‘mislead’ (Grice 1975: 49).
The fact that Iago’s violation is so quiet means that Cassio is
completely deceived. Iago’s words in example 3 are indicative of what
Rudanko has termed ‘metadiscursive deception’ (2005: 8). He cites

Galasinski (2000) who maintains that such a strategy as Iago’s is an
Attempt of the speaker/deceiver to make the addressee believe
that the utterance the speaker is issuing is cooperative, whereas
it is not (Galasinski 2000: 82 cited in Rudanko 2005: 8)
Whilst mapping out his criteria for what impoliteness is and is not,
Culpeper (2005, 37) maintains that ‘impoliteness is not unintentional’,
however by page 39 he appears to concede that his definition can have
‘other permutations’ (as example 3 nicely illustrates), these being that:
Face-attack may be intentionally communicated but fail to find
its mark in any way, or, conversely, the hearer may perceive or
construct intentional face-attack on the part of the speaker,
when none was intended. (Culpeper 2005, 39 my emphasis)
This definition suggests that impoliteness can be unintentional. Also,
the fact that he stresses that the hearer may ‘perceive and/or construct
behaviour as intentionally face-attacking’ suggests that any behaviour
can be reconstructed as impolite if the hearer so desires.
Similarly, the fact that a single hearer is specified holds implications
for this model, and not just in the case of dramatic dialogue, because
in any discourse where overhearers are present, impoliteness may still
be interpreted even though the intended recipient may not perceive it.
I believe a slight amendment to this model is needed in order to take
into account an audience’s overarching knowledge within the
discourse structure. Bousfield’s definition (2007a, 2007b) plugs the
gaps within Culpeper’s (2005) definition. Hence, successful
impoliteness comes about when:
[a speaker issues] intentionally gratuitous and conflictive face-
threatening acts that are purposefully performed. [They are]
unmitigated (i.e. not polite), in contexts where mitigation (i.e.
politeness) is required and/or [they are issued ] with deliberate
aggression, that is, with the face-threat exacerbated, ‘boosted’,
or maximised in some way to heighten the face damage
inflicted…the intention of the speaker (or ‘author’) to ‘offend’
(threaten/damage face) must be understood by those in a
receiver role[s]. (Bousfield 2007a, 2007b)
The point to make here is that impoliteness is not wired into any one
form of words spoken. Depending on the context, the form of words
can also have additional and multiple functions. This, then, implies
that we cannot have just one single definition of impoliteness because
depending on the context, different interpretations are able to be
generated.
If we consider the bigger picture, each individual has a particular
social context with a set of expectations regarding appropriate
behaviour. Therefore, there can never be just one definition of
impoliteness because everyone interprets face-attack differently which
means that impoliteness is a different phenomenon from person to
person.
4.2 When politeness can be impolite

Example [4]
Context: Iago begins to hint at Desdemona’s (supposed) deceit with
Cassio
Iago
Ha, I like not that.
Othello
What dost thou say?
Iago
Nothing, my lord; or if-I know not what.
Othello
Was not that Cassio parted from my wife?
Iago
Cassio, my lord? No, sure, I cannot think it
That he would steal away so guilty-like
Seeing you coming.
(Othello 3.3.34-40)
Iago’s first sentence is deliberately elliptical and thus exploits the
Maxim of Quantity. Othello immediately notices the flout and asks
Iago to repeat himself. Iago, knowing that he has Othello’s full
attention now can be even vaguer, hesitant and contradictory – all of
which exploit the Maxims of Quality, Quantity and Manner. His ploy
appears to work; Othello ‘reads’ a deeper meaning into Iago’s words

and mentions Cassio by name even though Iago has not. The fact that
Othello has mentioned him in the same sentence as Desdemona gives
Iago a perfect opportunity to use the negative words ‘steal’ and
‘guilty’ in the context of Cassio’s name, thus implicating the impolite
belief that Cassio has ‘stolen’ something of Othello’s ( i.e. his wife).
Example [5]
Iago
Good name in man and woman, dear my lord,
Is the immediate jewel of their souls:
Who steals my purse steals trash – ‘tis something –
Nothing,
‘twas mine, ’tis his, and has been slave to thousands –
but he that filches from me my good name
robs me of that which not enriches him
and makes me poor indeed.
Othello
By heaven, I’ll know thy thoughts!
Many of Iago’s intentional FTAs towards Othello in act 3 are cleverly
instigated via Brown and Levinson’s off-record politeness strategy.
This is the only way that Iago can sow his seeds of doubt in Othello’s
mind but it is also highly dangerous. In Brown and Levinson’s terms,
the weightiness factor is extreme and the power of the hearer (Othello)
over the speaker (Iago) is very great. No amount of redressive
politeness strategies could possibly mitigate what Iago intends to
convey to Othello. As such, this strategy allows a speaker to produce
‘all kinds of hints as to what [they] want or mean to communicate,
without doing so directly’ (Brown and Levinson 1987:69). In Gricean
terms, Iago is able to hint at Desdemona’s deceit with Cassio via an
implicature. In the above examples, Iago breaches the ‘Maxim of
Manner’ which, as part of the overarching Principle of Cooperation
within a conversation, maintains that a speaker should:
avoid obscurity of expression
avoid ambiguity
be brief
be orderly
The fact that both speakers are adhering to Grice’s (1975)

Cooperative Principle means that Iago’s deliberately ambiguous and
obscure words set in motion a chain of reasoning for Othello who, so
far, has had no reason to mistrust Iago (remember Iago’s ‘honest’
reputation8), and it is clear that Othello does pick up on Iago’s
implicatures: by heaven, I’ll know thy thoughts. By line 200 Iago is
able to be more explicit whilst still upholding his mask of ‘honesty’:
Example [6]
Iago
I am glad of this, for now I shall have reason
To show the love and duty that I bear you
With franker spirit: therefore, as I am bound,
Receive it from me. I speak not yet of proof:
Look to your wife, observe her well with Cassio.
Wear your eyes thus, not jealous nor secure,
I would not have your free and noble nature
Out of self bounty be abused.
(Othello Act 3 Scene 3 lines 196-203)
Iago’s ostensibly polite utterances can however be seen as positively
impolite because although Iago is not saying anything derogatory
about Othello himself, he is in fact making suggestions about the two
people closest to Othello. As Culpeper points out ‘the notion of face is
not confined to the immediate properties of the self, but can be
invested in a wide range of phenomena such as one’s family, job [or]
nationality’ (1996: 361). Liu (1986) has suggested that the idea of face
can be viewed as a series of concentric circles with those closest to a
person’s ego as being the most face-laden, which is why Iago’s
‘impolite beliefs’ are so damaging to Othello (Culpeper 1996: 361).
To criticise Desdemona therefore, is ultimately to criticise Othello.
The fact that Iago can use ostensibly polite strategies to convey
impoliteness to Othello shows how Brown and Levinson’s model is
deficient in accounting for ‘impolite desires’; they admit that their
‘system ‘overgenerates’ and needs to be complemented with a set of
‘filters’ that check that a chosen utterance form has no impolite
implicatures’ (Brown and Levinson 1987, 11). Iago also demonstrates
how impoliteness can be conveyed in an indirect way.
5. Conclusion
Within this study I have attempted to highlight a discrepancy with the
notion of intentionality within Culpeper’s (2005) definition of
impoliteness. Within the context of drama, the two-tier discourse
structure of a play means that interpretations of face-attack can be
different for the audience than for the characters. This is because of
the audience’s privileged position as ‘overhearers’ which allows them
to ‘see’ a character’s ‘real’ intentions, and hence promotes dramatic
tension within the scenes. Ultimately, this enables the audience to
judge whether an utterance is polite, impolite or as Watts (2003) has
termed politic i.e. ‘linguistic behaviour which is perceived to be
appropriate to the social constraints of the on-going interaction’
(Watts 2003: 19 my emphasis)
I maintain that face-attack can be unintentional, but rather than call it
impoliteness I would like to use the term Discourse Disjunctive
Politeness because, as we have seen, Desdemona’s intention is not to
attack Othello’s face, but in her attempt at defence she still succeeds
in doing so.
There is still much research to do regarding (im)politeness in drama,
not least considering how discourse structures may affect its
interpretation. Othello typically has a two-tier structure; however,
there are plays with more than two tiers- as in the case of plays with
narrators. It is the play’s structure which generates the dramatic irony
and allows an audience to understand more of what is going on in the
fictional world. Analysing impoliteness interpretation in a three tier
play structure is perhaps an area for future study, as is the role of
impoliteness within a different setting. As a tragedy, Othello was
obviously rich in scenes of conflict and tension. Culpeper has already
demonstrated how entertaining impoliteness can be within the genre
of film (1998) and television quiz shows (2005). Further research into
this area of politeness study could be undertaken within the genre of
comedy, to see how impoliteness may generate humour.
Endnotes
1
Culpeper’s translation of: Sauve, mari mango turbantibus aequora ventis,
e terra magnum alterius spectare laborem;
non quia vexari quemquamst iucunda voluptas,
sed quibus ipse malis careas quia cernere sauve est.
2
However, what Brown and Levinson don’t account for are the implications which
arise after aggression has surfaced within communication i.e. when impoliteness has
been generated.
3
Brown and Levinson define weightiness as ‘seriousness’ (p. 76).
4
I have omitted to show examples of impoliteness here, as they are explicated within
section 4.
5
Honigmann notes that during the middle ages, the word ‘Moor’ was synonymous
with ‘Negro’ (Honigman 2001:14).
6
Second Lieutenant.
7
I am aware that the prosodic effects of a character’s utterances are very difficult to
ascribe purely from a text analysis. The reading of Othello within this paper is based
on my own interpretation of what is happening within the fictional world. Knowing
the context in which the words are spoken, and the suggestion by another character
immediately after Othello’s exit that he might be jealous, lead me to ascertain that this
scene is fraught with tension and that the only plausible way to read it is to see
Othello’s words as having a certain amount of aggression and, hence, increasing
loudness to them.
8
It is Iago’s ‘honesty’ which allows him to mask many of his FTA’s.
References
Primary Source:
Honigmann, E.A.J. 2001. Othello (3rd Ed). London: Arden Shakespeare.
Secondary Sources:
Barnes, B. 1988. The Nature of Power. Cambridge: Polity Press.
Bennison, N. 1993. ‘Discourse Analysis, pragmatics and the dramatic character: Tom
Stoppard’s Professional Foul’ in Language and Literature 2(2): 79-99.
Bousfield, D. In press 2007a. ‘Impoliteness, Preference Organization and
Conducivity’in Multilingua.
Bousfield, D. In press 2007b. ‘Beginnings, middles and ends: towards a biopsy of the
dynamics of impolite exchanges’in Journal of Pragmatics.
Bremond, C. 1966. ‘La Logique des Possibles Narratifs’ in Communications 4 : 4-32.
Brown, R., and A. Gilman. 1989. ‘Politeness Theory and Shakespeare’s Four Major
Tragedies’ in Language in Society 18: 159-212.
Brown, P., and S.C. Levinson. 1987. Politeness: Some Universals in Language
Usage. Cambridge: Cambridge University Press.
Corsaro,W., and T.A. Rizzo. 1990. ‘Disputes in the peer culture of American and
Italian nursery-school children’ in Grimshaw, A.D (ed.) Conflict Talk:
Socioloinguistic Investigations of Arguments and Conversations. Cambridge:
Culpeper, J. 1996. ‘Towards an Anatomy of Impoliteness’ in Journal of Pragmatics
25: 349-367.
Culpeper, J. 1998. ‘Impoliteness in Drama’ in Culpeper, J., M. Short and P. Verdonk
(eds) Exploring the Language of Drama. London: Routledge.
Culpeper, J. 2001. Language and Characterisation. London: Pearson Education Ltd.
Culpeper, J., D. Bousfield and A. Wichmann. 2003. ‘Impoliteness Revisited: with

special reference to dynamic and prosodic effects’ in Journal of Pragmatics 35:
1545-1579.
Culpeper, J. 2005. ‘Impoliteness and Entertainment in the television quiz show The
Weakest Link’ in Journal of Politeness Research 1: 35-72.
Fraser, B. and W. Nolan. 1981. ‘The Association of Deference with Linguistic Form’
in International Journal of the Sociology of Language 27: 93-109.
Galasinski, D. 2000. The Language of Deception. A Discourse Analytical Study.
Thousand Oaks, CA: Sage.
Goffman, E. 1967. Interactional Ritual: Essays on Face-to-Face Behaviour. New
York: Anchor Books.
Grice, P. 1975. ‘Logic and Conversation’ in Cole, P., and J. Morgan (eds) Syntax and
Semantics Vol 3: Speech Acts. New York: Academic Press.
Holmes, J. 1984. ‘Modifying Illocutionary Force’ in Journal of Pragmatics 8: 345-
365.
Leech, G. 1992. ‘Pragmatic principles in Shaw’s “You Never Can Tell” in Toolan, M.
(ed.) Language, Text and Context. London: Routledge.
Leech, G. 1983. Principles of Pragmatics. London: Longman.
Liu, R 1986. The politeness principle in A Dream of Red Mansion’s. Unpublished
M.Phil.dissertation, Lancaster University.
Locher, M.A. 2004. Power and Politeness in Action: Disagreements in Oral
Communication. Berlin and New York: Mouton de Gruyter.
Lucretius De Rerum Natura, Book II: 1-4.
Rudanko, J. 2005. ‘Aggravated impoliteness and two types of speaker intention in an
episode in Shakespeare’s Timon of Athens’ in Journal of Pragmatics. Article in
press.
Simpson, P. 1989. ‘Politeness Phenomenon in Ionesco’s The Lesson’ in Carter, R.,
and P. Simpson (eds) Language, Discourse and Literature: An Introductory
Reader in Discourse Stylistics. London: Unwin Hyman: 171-193.
Short, M. 1996. Exploring the Language of poems, Plays and Prose. London:
Longman.
Tanaka, N. 1993. The pragmatics of uncertainty: its realisation and interpretation in
English and Japanese. Unpublished PhD Thesis, Lancaster University.
Thomas, J. 1995. Meaning In Interaction. London: Longman.
Wartenberg, T.E. 1989. The Forms of Power. From Domination to Transformation.
Philadelphia: Temple University Press.
Watts, R.J. 2003. Politeness. Cambridge: Cambridge University Press.
Mind-Style, Modality, and Poe’s ‘The Fall of the House
of Usher’
Simon Zupan
University of Ljubljana
Abstract
This paper examines the interrelatedness of modality and the narrator’s mind-style in
Edgar Allan Poe’s story ‘The Fall of the House of Usher.’ Uncertainty modality
directly influences the way the reader perceives the fictional world. By adding
uncertainty modality to his narrative, the first-person narrator admits that he has
difficulties dealing with his visual and aural perceptions and, ultimately, his own
mental and cognitive processes. This projects a view of the world characterized by
pockets of uncertainty. These, in turn, serve as a vehicle for many of the Gothic
effects the story has often been praised for: fear, terror, anxiety and discomfort.
Keywords: mind-style; modality; E.A.Poe, ‘The Fall of the House of Usher’; Gothic
effects.
1. Introduction
‘The Fall of the House of Usher’ is one of Edgar Allan Poe’s best
known short stories. It is particularly famous for its Gothic properties.
Critics have ascribed its capacity for creating Gothic effects to
different features of the narrative: its ‘utterly strange atmosphere’
(Voloshin 1986), its ‘unique mood and tone’ (Evans 1977: 142), the
‘morbidity’ (Haggerty 1989: 93, Peeples 2002: 182), ‘the instructive
terror of the narrator’ (Rountree 1972: 128), the reader’s confrontation
with ‘the terror in one’s losing his mind’ (Obuchowski 1975: 407), the
combination of the ‘preternatural, the natural, and the psychological’
in the tale (Voloshin 1986: 420), its proto-Gothic structural principles
that allow the reader to identify themselves with the narrator (Hustis
1999: 17, Haggerty 1989: 92), and so on.
So far, however, critics (with the exception of Haggerty (1989: 92))
have paid almost no attention to another element of the narrative that
contributes significantly to the story’s Gothicness: the narrator’s
uncertainty. The relative neglect of uncertainty is the more surprising
because it manifests itself clearly in at least two different ways: first,
through the narrator’s repeated and overt complaints about the
difficulty of relating to what he had experienced at the House of Usher
452 Simon Zupan
(e.g. ‘I know not how it was’; ‘I knew not why’; ‘What was it?’); and
second, through the modality of the narrative. It is particularly the
latter that is important because it is directly indicative of the way the
narrator perceives his experience. It reveals how difficult it was for
him to provide rationally for the events and phenomena he had
witnessed. It shows that he was unable to do so in most cases. This, in
turn, is important because these pockets of uncertainty at the same
time make his experience appear even more mysterious and
incomprehensible, a situation which, consequently, also becomes the
source of fear and anxiety for the reader. The modality of his narrative
thus directly influences the way the reader perceives the narrator’s
view of his own experience.
As a detailed examination of the narrative reveals, uncertainty is not
restricted only to certain areas of the narrator’s experience. Instead, it
turns out to determine his view of the experience as a whole. Thus, it
seems reasonable that his experience be examined with the help of
Roger Fowler’s concept of mind-style. This, as we will explain later,
will allow us to observe one type of discourse structure – modals –
and the cumulative effect of individual occurrences of modality
against the narrator’s view of the experience as a whole. We will first,
however, briefly define each of the two categories.
2. Modality
At its most general, modality has traditionally been understood as ‘the
manner in which the meaning of a clause is qualified so as to reflect
the speaker’s judgement of the likelihood of the proposition it
expresses being true’ (Quirk et al. 1992: 219). This definition rests on
an assumption that the speaker, when making a statement, can
basically adopt two opposite positions: they maintain either that what
they are saying is true, or, in contrast, that the proposition they are
making is not true. Michael Halliday refers to this as polarity
(Halliday and Matthiessen 2004: 143). However, speakers can also
adopt many positions in between these two poles. They can determine
the degree to which they believe the proposition is true or not true. It
is exactly these intermediate degrees that are strictly referred to as
modality. At the same time, it has to be pointed out that, strictly
speaking, there exist two types of modality. Halliday distinguishes
between modalization, which refers to the propositions that cover the
intermediate degrees between asserting and denying, and modulation,
Mind-Style, Modality, and Poe’s ‘The Fall of the House of Usher’ 453
which refers to proposals that cover the area between prescribing and
proscribing (Halliday and Matthiessen 2004: 147). Both types, which
otherwise more or less correspond to extrinsic and intrinsic modality,
respectively (Quirk et al. 1992: 219), have different intermediate
possibilities. From the point of view of the text we are going to
analyse, we will be particularly interested in propositions equipped
with probability modality. These are equivalent to ‘either yes or no’,
that is, maybe yes, maybe no, with different degrees of likelihood
attached (Halliday and Matthiessen 2004: 147).
3. Mind-Style and Modality

In his first major exposition of the notion, Roger Fowler defined
mind-style in the following way (1977: 76):
Cumulatively, consistent structural options, agreeing in cutting
the presented world to one pattern or another, give rise to an
impression of a world-view, what I shall call a ‘mind style’.
As Elena Semino and Kate Swindlehurst have pointed out, the idea
rests on two central assumptions: the first is that each speaker has his
or her own ‘reality’ that is the result of their perceptual and cognitive
processes; the second that language is a central part of the process by
which each person makes sense of the world around them. The way in
which their experience is put into words is thus reflective of the
conceptualization of reality pertaining to a particular speaker (Semino
and Swindlehurst 1996).
Since its introduction, mind-style has proved a useful analytical tool
for examining different world-views. It has been used to analyse a
literary interpretation of the world-view of a Neanderthal (Fowler
1977: 104-106, Halliday 1996), and the world-view of a protagonist of
a typical Gothic narrative (Fowler 1996: 223-226), while Leech and
Short, for example, have analysed a whole range of different mind-
styles from the “normal’ to very deviant ones, including that of the
mentally handicapped Benjy in William Faulkner’s novel The Sound
and the Fury (1995: 202-207). Another original application of the
notion was Swindlehurst and Semino’s study of the narrator’s mind-
style in Ken Kesey’s One Flew Over the Cuckoo’s Nest, whose
idiosyncrasy originates from the consistent employment of an
underlying conceptual machine-room metaphor (Semino and
Swindlehurst 1996).
454 Simon Zupan
Most of these studies examine mind-style with what has by now

become the ‘standard’ analytical method. It includes analysis of both
the micro- and the macrostructural levels of the text. For the former,
the ‘consistent structural options’ are studied, in other words, the
particular speaker’s discourse. In most cases, the following three
features of the discourse are examined in detail: syntactic properties,
lexical properties, and transitivity. (With the exception of the
previously mentioned study by Semino and Swindlehurst, which
examines the occurrence of metaphors in the discourse.) Implications
of the established ‘consistent structural options’ are then studied on
the macrostructural level of the text: complex syntax thus suggests the
capacity for complex thinking and logical sophistication, simple
syntax a lack thereof; similarly, elaborate, abstract vocabulary
suggests intellectual sophistication, basic, concrete one usually
simplicity and naivety.
Somewhat surprisingly, almost no study to date has examined the role
of modality in the creation of mind-style. This is the more surprising
because modality of some type is an integral part of practically every
discourse, even when not explicitly marked. Hence, it should logically
play a role in the creation of mind-style. One reason for the relative
neglect may coincide with what Semino and Swindlehurst identified
as the main reason for the relative neglect of the role of modality in
mind-style studies: that the use of metaphors in itself is not necessarily
a sign of the deviant or idiosyncratic in discourse, which has been the
focus of most mind-style studies (Semino and Swindlehurst 1996).
Analogously, modality in itself is not necessarily indicative of oddities
in the speaker’s view of the world.
The only study that at least hints at a possible interrelatedness between
modality and mind-style is Fowler’s Linguistic Criticism. The link is
hinted at through the notion of point of view. As we know, Fowler
developed mind-style out of Boris Uspensky’s notion of the ‘point of
view of the ideological plane’ (Uspenski 1979). He even explicitly
said that the two notions were equivalent (Fowler 1996: 214) (but see
Short (1992: 32)), which means that a particular ideological stance is
at the same time a mind-style (for problems in equating mind-style
with ideological point of view, see Semino and Swindlehurst (1996)).
At the same time Fowler maintains that modality, too, is closely
related to point of view. For him modality is one of two fairly distinct
ways in which point of view on the ideological plane manifests itself.

He even calls it the ‘grammar of explicit comment’ (Fowler 1996:
166). However, he did not go on to examine the interrelationship in
any concrete examples in his mind-style analyses.
In our view, the need to examine modality with regard to mind-style
shows itself in at least two ways. Firstly, as we have mentioned,
modality per se is an integral part of every discourse. As such, it is
necessarily reflective of the way the speaker perceives of their reality
and the way they put their experience of reality into words. A good
way to illustrate this is to compare propositions with modality
attached to them with their demodalized counterparts. If, for example,
two people observe the same situation and one says ‘perhaps I saw a
ghost’, and the other ‘I saw a ghost’, then the two would not have
experienced the situation in the same way. Whereas the latter
apparently has no doubt about seeing the ghost, the former does. This
also means that the reader or listener sees the situation differently. If
they relied only on the information they got from the two speakers,
they would think that each speaker’s view of the situation was
different. Secondly, the notion of mind-style can capture the
cumulative effect of modal structures. This is important because
modal structures normally do not occur only sporadically; instead
modality marks longer stretches of text. Mind-style therefore seems
particularly appropriate for examining the overall effect of individual
occurrences of a particular type of modality.
The latter point is at the same time the main reason for Poe’s
employing it in ‘The Fall of the House of Usher’. As we will see, the
narrator’s discourse is marked by uncertainty modality throughout the
narrative. Analysis of individual examples will help us establish the
influence these exert on the reader’s perception of the narrator’s view
of events.
‘The Fall of the House of Usher’
‘The Fall of the House of Usher’ is a story about a series of unusual
events that the first-person narrator witnessed sometime in the past,
while visiting his childhood friend, Roderick Usher. Immediately after
his arrival at the eerie Usher mansion, Roderick’s twin-sister Lady
Madeline, whom he saw alive the day before, suddenly dies
mysteriously. He then helps Roderick bury her temporarily in one of
456 Simon Zupan
the vaults below the mansion. A few days later, however, it turns out
that Roderick’s dark premonition that drew him to the verge of
insanity after the burial was justified: after a few days’ struggle,
Madeline somehow manages to get out of the tomb only to die in his
arms; it appears that they had buried her while she was still alive.
Overcome by awe, the narrator ‘flees aghast’ from the mansion, only
to see it disappear in ruins in the tarn in front of it.
4. Modality and the narrator’s visual perception

Even though brief and schematic, the summary nevertheless should
make it clear that the story encapsulates many proto-Gothic properties,
including mystery, fear, anxiety, terror, eeriness, and the like. Most of
these effects are the result of the narrator’s having witnessed events
whose background he could not rationally explain. For this reason, it
is unsurprising that the narrator’s observations are marked with
uncertainty.
The first examples of the kind can be found at the very opening of the
story. There, the narrator describes the moments just at nightfall,
when, after riding through a ‘singularly dreary tract of country’ for the
whole day, he finally reaches the House of Usher. He reports that he
then began studying the externals of the house, upon which
immediately ‘a sense of insufferable gloom pervaded [his] spirit’.
Nevertheless, he goes on to describe the externals of the house. He
does so in the following way (sentences have been numbered and
parts of the text have been underlined for easy reference) (Poe 1986:
140-41):
(1) Its principal feature seemed to be that of an excessive
antiquity. (2) The discoloration of ages had been great. (3)
Minute fungi overspread the whole exterior, hanging in a fine
tangled web-work from the eaves. (4) Yet all this was apart
from any extraordinary dilapidation. (5) No portion of the
masonry had fallen; and there appeared to be a wild
inconsistency between its still perfect adaptation of parts, and
the crumbling condition of the individual stones. (6) In this
there was much that reminded me of the specious totality of old
wood-work which has rotted for long years in some neglected
vault, with no disturbance from the breath of external air. (7)
Beyond this indication of extensive decay, however, the fabric
gave little token of instability. (8) Perhaps the eye of a

scrutinizing observer might have discovered a barely
perceptible fissure, which, extending from the roof of the
building in front, made its way down the wall in a zigzag
direction, until it became lost in the sullen waters of the tarn.
As we can see, the narrator focuses on different features of the
externals of the house. Among them are its old age, colour, the fungi,
the general condition of both the individual stones and the walls as a
whole, the stability of the house, the fissure in the wall, and others.
We can notice that attention is paid both to detail and to the overall
picture. It is evident, however, from the first part of the extract that the
way the narrator describes the image shows that there are at least two
points where he is not certain that what he saw was indeed as he saw
it. The key factor in this is the modality at the propositions. In
sentence (1), for example, we can see that the narrator does not know
for certain whether ‘excessive antiquity’ was the principal feature of
the house or not, because he adds a proviso that it only seemed to him
that this was so, i.e. perhaps it was as he describes it, perhaps it was
not. Something similar can be noticed in sentence (5). Here, the
narrator is focusing on the ‘wild inconsistency’ between the condition
of the wall as a whole and the individual stones composing it. Again,
as his wording reveals, the narrator admits that his observation is not
an objective but clearly a subjective one, and that it again only
appeared to be as he describes it, i.e. that he does know that for
certain.
Both examples thus point at the discrepancy between what the narrator
saw and what the actual state of affairs was. The narrator is uncertain
whether these always matched. A good way to illustrate this is to
compare these sentences with their demodalized counterparts, for
example with polar sentences of the type ‘Its principal feature was that
of an excessive antiquity’ and ‘No portion of the masonry had fallen;
and there was a wild inconsistency between its still perfect adaptation
of parts, and the crumbling condition of the individual stones’,
respectively. As we can see, the narrator’s view of the phenomena
described would be different in this case: instead of their evidently
being subject to speculation and guessing, their description through
polar sentences would make them appear factual. The epistemic status
of the described entity with the narrator would thus change. It is
458 Simon Zupan
important to stress that this would also change the effect of the
uncertainty on the reader. In the original text, the uncertainty that
characterizes the narrator’s view of the house adds to the mystery of
the image; if the uncertainty is taken away, the narrator appears much
more confident about the object of his report, and, consequently, the
reader too will very likely feel less uncertain about what is being
reported.
The passage contains other signs of uncertainty besides those just
described. Thus we can see that in sentence (8) the narrator adds
modality to his observation about the fissure in the wall. Here two
discourse markers are employed, the adverb ‘perhaps’ and the ‘modal’
verb ‘might’. However, modality is used slightly differently in this
case. It does not refer directly to the character of the narrator’s
observations, i.e. to whether or not the fissure existed. The narrator
makes it clear in the proposition that the fissure was there. Instead, he
speculates about its prominence, i.e. whether other observers would
have noticed it. Uncertainty thus does not refer directly to the
existence of the fissure, but rather to its prominence.
Other examples in the text indicate that the narrator had difficulties
coming to terms with his visual perceptions. It is noticeable that
uncertainty is present no matter what is being described, be it the
inanimate world of the House of Usher or the people he met in it. The
following example is also taken from the introductory part of the
story. Immediately after entering the house, as the narrator recalls, he
was taken by one of the servants to his master, Roderick Usher. On
the way there ‘through many dark and intricate passages’, he
remembers meeting the family doctor on one of the staircases. He
describes the encounter thus (Poe 1986: 141):
(9) On one of the staircases, I met the physician of the family.
(10) His countenance, I thought, wore a mingled expression of
low cunning and perplexity. (11) He accosted me with
trepidation and passed on.
In his short description of the situation, the narrator chooses to focus
on the physician’s countenance. He singles out two characteristics he
apparently read on it, ‘low cunning, and perplexity’. As in the
previous examples, however, here too the narrator is not entirely sure
that the two personal traits were indeed on the physician’s
countenance. That is only his speculation. The narrator signals the

subjectivity of his observation by inserting a proviso: he does not
describe the situation with a polar sentence (e.g. ‘His countenance
wore a mingled expression of low cunning and perplexity.’), but
admits that he thought he had seen them, i.e. perhaps the countenance
wore the expression described, perhaps it did not.
There is, at the same time, another feature to be pointed out with this
particular example: the narrator manages to add modality to his
proposition without using any of the common modality markers, like
the one, for example, we used in our paraphrase (the modal adverb
‘perhaps’). Instead, another means of expressing modality is
employed, the verb ‘to think’. Michael Toolan, for example, regards
this as a more advanced way of appending modality to a proposition,
since the verbs of thinking in such cases are not used with their literal
meaning, but metaphorically (1998: 53). The effect, however, as can
be seen through the paraphrase, is the same as when more
conventional modality markers are employed. The narrator’s
uncertainty shows that he does not know for certain what he saw, and
this, in turn, creates another locally salient pocket of the unknown.
This consequently determines the narrator’s view of the situation, i.e.
his mind-style.
The fact that the narrator often employs this means of marking
modality is further corroborated by another example that can be found
in the immediate textual vicinity of the previous sentence. According
to the narrator, having led him through the long corridors of the House
of Usher, the servant finally ‘ushered’ him into his master’s studio.
The narrator describes his first encounter with Roderick in the
following way (Poe 1986: 142):
(12) Upon my entrance, Usher arose from a sofa on which he
had been lying at full length, and greeted me with a vivacious
warmth which had much in it, I at first thought, of an overdone
cordiality – of the constrained effort of the ennuyé man of the
world.
The narrator in this scene does not focus principally on what he
noticed physically upon encountering Roderick after a long time, but
is instead concerned with how he interpreted the manner in which his
host greeted him. He thought Roderick’s ‘vivacious warmth’ had
460 Simon Zupan
much ‘overdone cordiality’ in it. Again, however, it turns out that the
narrator is not entirely certain that the initial interpretation was
correct. There exist two indications of his being aware of this: the first
is that he corrects himself by additionally explaining that it was not
only ‘overdone cordiality’, but also the ‘constrained effort of the
ennuyé man of the world’, which is how he tries to pin down what he
saw more precisely. At the same time, we can see once more that he
tags his own proposition with uncertainty modality. This, as the
underlined part of the text reveals, is done in the same way as in the
previous example: the narrator does not use any of the explicit
modality markers; instead, the verb of thinking is employed.
The effect, however, is the same as with other more conventional
modality markers: the narrator’s sentence is evidently not a polar
statement in the sense of ‘Roderick’s vivacious warmth had much in it
of an overdone cordiality’; instead the narrator adds a proviso to his
observation. For this reason, this micro-segment of the text is another
spot marked by uncertainty. Here, the reader once more is left feeling
that despite the narrator’s description, in fact because of it, he is once
more being denied access to reality in its entirety.
5. Modality and the narrator’s aural perception

As it turns out, the narrator does not only have difficulties dealing
with what he saw or the interpretation of what he saw during his visit
to the House of Usher. It is also difficult for him to relate to what he
heard there. Sometimes this shows through his uncertainty whether he
had heard a sound at all, other times in his wondering what exactly he
had heard. The difficulties are to a large extent analogous to those
with visual perception.
One part of the text where this shows is the scene we considered
above, the moment when the narrator meets Roderick for the first
time. The narrator reports that the two had a conversation. Roderick
apparently talked about the reasons why he wanted the narrator to
come over, and what he expected from the visit. According to the
narrator’s report, Roderick, among other things, also addressed the
cause of his mysterious condition. This is how the narrator remembers
Roderick’s explanation (Poe 1986: 143):
(13) It was, he said, a constitutional, and a family evil, and one
for which he despaired to find a remedy – a mere nervous
affection, he immediately added, which would undoubtedly

soon pass off. (14) It displayed itself in a host of unnatural
sensations. (15) Some of these, as he detailed them, interested
and bewildered me; although, perhaps, the terms, and the
general manner of the narration had their weight.
The passage includes two explicit signs of modality. The first is the
adverb ‘undoubtedly’ in sentence (13), and the second the adverb
‘perhaps’ in sentence (15). Even though the first is of interest, we will
focus only on the second sentence, the reason being that the first part
of the passage is, strictly speaking, originally not the narrator’s but
Roderick’s discourse.
In sentence (15), the narrator is talking about his interest in what he
had heard from Roderick and at the same time trying to explain to
himself why that ‘interested and bewildered’ him. The narrator
believes two reasons were responsible for this: the first was that he
was told the ‘details’ about the ‘unnatural sensations’, and the second
‘the terms, and the general manner of their narration’, i.e. the way
Roderick described the ‘sensations’ in words. However, the narrator is
again not entirely certain that he knows whether the two causes were
indeed responsible for the effect Roderick’s explanation had on him.
The proviso ‘perhaps’ shows he is particularly uncertain about the
second cause, ‘the terms, and the general manner of the narration’,
whether it was responsible for the aforementioned effect. Moreover,
his wording shows that this is another of his speculations, and that the
true cause why Roderick’s explanation ‘interested and bewildered’
him, remains unknown to him. In turn, the narrator’s view of this
particular incident is at least partially veiled in mystery.
The previous example shows that the narrator appears not to have had
difficulties only with what he heard from Roderick, but also with the
manner in which he spoke. Other examples in the narrative indicate
that the narrator had problems physically relating to what he heard. An
illustrative example comes from that part of the story in which the
narrator describes Roderick’s recitals. The narrator reports that in the
days after Lady Madeline’s entombment he often listened to Roderick
recite poems and accompany himself upon the guitar. He reports that
these recitals made a deep impression on him. Moreover, like a
number of other phenomena, they had the capacity to evoke peculiar
feelings in him. As with other phenomena and events, the narrator
462 Simon Zupan
tries to find a rational explanation for these strange effects on him.

This is how he tries to explain it (Poe 1986: 146):
(16) It was, perhaps, the narrow limits to which he thus
confined himself upon the guitar, which gave birth, in great
measure, to the fantastic character of his performances. (17) But
the fervid facility of his impromptus could not be so accounted
for. (18) They must have been, and were, in the notes, as well as
in the words of his wild fantasias (for he not unfrequently
accompanied himself with rhymed verbal improvisations), the
result of that intense mental collectedness and concentration to
which I have previously alluded as observable only in particular
moments of the highest artificial excitement.
The passage reveals that the recital made a strong impression on the
narrator, who is now trying to pin down the exact cause of that effect.
His thinking runs in two different directions: on the one hand, he
ascribes the effect of the recital partly to the ‘narrow limits upon the
guitar’; on the other, he believes that it was also the ‘notes and the
words’ of Roderick’s ‘wild fantasias’ that were responsible for the
effect.
As it turns out, however, the narrator is once more not certain that his
explanation is necessarily correct. Moreover, the two modal structures
‘perhaps’ and ‘must have been’ indicate that his explanation is yet
another instance of speculation. Again, the best way to illustrate the
role modality plays with regard to the propositions is to compare these
sentences with their demodalized counterparts. In this case we would
be comparing them with something like ‘It was the narrow limits to
which he thus confined himself on the guitar…’, and ‘They were in
the notes, as well as in the words…’, respectively. As we can see, the
narrator’s perception of the reasons why the recital had such an effect
on him would be completely different. While in the original text the
adverb ‘perhaps’ and the modal verb that expresses the so-called
‘logical’ or ‘epistemic’ necessity, would suggest that there is only a
high likelihood that what the narrator is saying is true, but that that is
not necessarily so (Quirk et al. 1992: 226), the demodalized version
effectively allows only one interpretation: that the narrator has no
doubt whatsoever that his proposition holds true. This, in turn, would
also alter the narrator’s mind-style. While in the original version his
wording reveals uncertainty as to why Roderick’s performances had a
‘fantastic character’ and why his impromptus were characterized by

‘fervid facility’, in the demodalized version the narrator’s view would
suggest that he has an objective and rational explanation for the
strange effect. Unlike in the original version, this would also mean
that the traits of mysteriousness have been removed, and that,
consequently, the sentence’s potential to create discomfort would be
weakened.
Immediately following the extract we have just analysed, the narrator
provides a concrete example of his reaction to one of Roderick’s
recitals. He describes it in the following way (Poe 1986: 146-147):
(19) The words of one of these rhapsodies I have easily
remembered. (20) I was, perhaps, the more forcibly impressed
with it, as he gave it, because, in the under or mystic current of
its meaning, I fancied that I perceived, and for the first time, a
full consciousness on the part of Usher, of the tottering of his
lofty reason upon her throne.
As can be seen, the narrator is trying to work out the reasons behind
the ‘impression’ that Roderick’s rhapsody has made on him. He
believes that it was because of his becoming aware for the first time
that Roderick knew that he was gradually losing his mind. Again,
however, it is obvious that the narrator does not know that for certain.
Instead, the proviso that he adds to the proposition with the adverb
‘perhaps’, as well as with the verb ‘fancy’ (‘I fancied’) shows that the
reason for the rhapsody’s effect on him is at best his speculation. The
narrator denies access to the ‘truth’ about a phenomenon. As a
consequence, this part of the narrator’s experience too remains partly
characterized by mystery. The narrator’s view of the phenomenon, in
turn, will also be marked by uncertainty.
There are other examples to show the narrator’s confusion about what
he heard at the House of Usher. One of the most illustrative is the
narrator’s confusion about the sounds they heard in the moments prior
to Lady Madeline’s resurrection. At the time, as the narrator reports,
Roderick’s premonitions about Madeline’s chilling fate had already
brought him to the verge of insanity. The narrator tried to calm him by
reading him a passage from a book. Yet as he did so, something
unusual happened: the sounds of ripping and tearing that the story
described began matching the sounds that Roderick and the narrator
464 Simon Zupan
could actually hear in the House of Usher. The narrator describes

these moments in the following way (Poe 1986: 154):
(21) At the termination of this sentence I started, and for a
moment, paused; for it appeared to me (although I at once
concluded that my excited fancy had deceived me) – it appeared
to me that, from some very remote portion of the mansion, there
came, indistinctly, to my ears, what might have been, in its
exact similarity of character, the echo (but a stifled and dull one
certainly) of the very cracking, and ripping sound which Sir
Launcelot had so particularly described. (22) It was, beyond
doubt, the coincidence alone which had arrested my attention;
for, amid the rattling of the sashes of the casements, and the
ordinary commingled noises of the still increasing storm, the
sound, in itself, had nothing, surely, which should have
interested or disturbed me.
The narrator describes the sounds that he and Roderick heard while
reading the book by Sir Launcelot. In contrast with other sentences
where modality is not always marked so explicitly, the narrator makes
it clear in this one that he is not certain as to what exactly he heard. He
uses a series of modals to signal this. In sentence (21), for example,
the narrator expresses his reservation very clearly by saying that it
only ‘appeared’ to him that he had heard the ‘echo’ of the sound Sir
Launcelot’s story described. His reservation receives additional
emphasis: first, by being repeated verbatim twice, and, second, by
appearing immediately after the brief pause that is indicated by the
dash. Moreover, his uncertainty regarding the nature of the sounds is
further corroborated by yet another marker of the so-called epistemic
possibility modal, the verbal phrase ‘might have been’. This one, too,
indicates the ‘possibility of a given proposition’s being or becoming
true’ (Quirk et al. 1992: 223), which in this particular example means
that the narrator is unclear whether the sounds he heard were indeed
the same as those Sir Launcelot’s story ‘had so particularly described’.
Another important modality feature of the passage involves the
adverbial structures ‘certainly’, ‘beyond doubt’, and ‘surely’ in
sentence (22). These differ in their function from those we have just
considered. In contrast to the verb ‘appear’ and the verbal phrase
‘might have been,’ they do not express uncertainty on behalf of the
narrator. Instead, they are to indicate the opposite, certainty. The
narrator uses them to emphasize the fact that, in contrast to the

uncertainty about whether he and Roderick had heard any sounds
similar to those in Sir Launcelot’s story, he has no doubt that the
‘echo’ was ‘a stifled and dull one’, that it was only because of the
‘coincidence’ that he came to hear the sounds, and that ‘the sound, in
itself, had nothing which should have interested or disturbed’ him.
That, at least, is what his explicit assurance, expressed via the three
adverbial structures, should suggest. However, it is precisely because
of this that the truth-value of his proposition is questionable. This
might seem surprising at first because the narrator uses high value
modals that apparently give the proposition almost the full weight of
the polar form. We know, however, that even high value modals (like
those in the passage) are less determinate than the polar forms, which
means that, as Halliday (2004: 147) has astutely observed, speakers
only say that they are certain when they are not. This has clear
implications for our example. Were the narrator indeed certain about
the true character of the ‘echo’, the role of ‘coincidence’, and the
effect of the ‘noises’ upon him, he would not have to explicitly assure
the reader that he did. In so doing he reveals that the narrator at this
particular point of the narrative is again in the dark about the true
nature of the phenomenon he is describing. It is certain that he is
himself not certain about the sounds, be it whether they existed, and, if
they did, what their character was. Consequently, the passage
indicates that at this textual point, too, the narrator’s perception, and
hence his mind-style, is marked by uncertainty.
6. Modality and the narrator’s thinking

The third major area in the narrative where uncertainty modality is
prominent includes the narrator’s thoughts in general. Uncertainty is
not only evident in the narrator’s perceptions. Instead, he is often
uncertain about his own mental or cognitive processes, be it with
regard to the information he gained through perception during the visit
itself, or the information about the Usher family he had prior to the
visit to their mansion. The two different areas may even be causally
linked to some extent: it appears that the narrator’s uncertainty may be
responsible for his occasionally being distrustful of his own mind and
thinking.
Our first example is taken from the introductory part of the narrative.
Here, as has already been pointed out, the narrator is observing the
466 Simon Zupan
exterior of the house, at the same time thinking about what he had
known about the Ushers. One line of thought comprised the unusual
genealogical characteristic of the family that it developed without ‘any
enduring branches’. This peculiarity was of interest for the narrator
because he saw a possible connection between this unusual fact and
the fact that for some, the name of the family denoted both the family
and the house (Poe 1986: 139-140):
(23) It was this deficiency, I considered, while running over in
thought the perfect keeping of the character of the premises
with the accredited character of the people, and while
speculating upon the possible influence which the one, in the
long lapse of centuries, might have exercised upon the other – it
was this deficiency, perhaps, of collateral issue, and the
consequent undeviating transmission, from sire to son, of the
patrimony with the name, which had, at length, so identified the
two as to merge the original title of the estate in the quaint and
equivocal appellation of the ‘House of Usher’ – an appellation
which seemed to include, in the minds of the peasantry who
used it, both the family and the family mansion.
A close examination of the sentence reveals that the narrator’s
thinking ran simultaneously along two different lines: on the one
hand, he was thinking about the ‘possible influence’ between ‘the
character of the premises’ and ‘the character of the people’ over time,
and on the other, about the possibility that this unusual fact about the
family caused the name ‘House of Usher’ to denote both the family
and the mansion. The sentence can be divided into two parts: one part
of the sentence, that between the first comma and the first dash,
indirectly presents the first line of thinking, and, with the verbs of
thinking (‘consider’, ‘run over in thought’, ‘speculate’) at the same
time serves as a reporting clause that introduces the other line of
thinking that is presented in the rest of the sentence. The latter, in
contrast to the first line of thinking, is presented almost verbatim
(grammatically adapted to the narrative technique used).
This distinction is important to keep in mind for two reasons. First, the
fact that the narrator thinks about so many things at the same time,
shows how hard he was trying to find a (rational) explanation for what
he was thinking about, and second, more importantly, it allows us to
see the extent to which uncertainty and speculation marked his
thinking. Here, modality once more plays the crucial role. As we can
see, modal structures occur in both streams of the narrator’s thoughts.
In the first one there are two modal structures: the adjective ‘possible’
and the modal verb in the verbal phrase ‘might have exercised’. The
adjective indicates that he does not know for certain whether the
‘influence’ between the ‘character of the premises’ and the ‘character
of the people’ existed, and the verbal structure, whether the first
‘exercised’ it over the other. It is evident that this chain of thought is,
at best, the narrator’s speculation, his interpretation (Hustis 1999: 12),
which the narrator at the same time indicates openly by denoting the
character of his thoughts as such (‘while speculating’).
The sentence, however, includes even more signs of speculation. We
find these in the second part of the sentence. Here modality is also
marked twice, with the adverb ‘perhaps’ as well as with the verb
‘seem’. In this part, the narrator is thinking about the connection
between the peculiar ‘deficiency’ of the Usher family and the fact that
the name came to refer both to the family and the family mansion. As
the modal structures show, however, this too is no objective
observation, since it is not described as a polar statement. Instead, the
uncertainty modality that both modals add to the proposition shows
that the narrator does not know for certain whether the connection is
his thoughts was indeed responsible for the fact that with some people
the syntagm ‘House of Usher’ came to refer to both the family and
their mansion. He also admits that it only ‘seemed’ to him that the
‘appellation’ the peasantry used indeed meant both the family and the
mansion to them, i.e. he does not know that for sure.
Consequently, the effect of this segment of the text is similar to that of
other parts that are marked by uncertainty modality. The narrator’s
wording reveals that he did know something about the Ushers and
their mansion, yet that his information was only fragmentary, which is
why his picture of the family was not complete. It is not surprising,
therefore, that the narrator tried to fill the gaps somehow. He did so by
guessing and speculation. The result is a view of the world that is to a
large extent marked by uncertainty. This uncertainty, in turn, opens up
more space for mystery, speculation, guesswork, and, consequently,
discomfort, anxiety, and fear.
Finally, we can notice a similar functional paradigm in our last
example. This sentence comes from the introductory part of the
468 Simon Zupan
narrative and describes the moments when the narrator is observing

the exterior of the mansion. As we have already pointed out, the
observation had a most unusual effect on the narrator, filling him with
discomfort, anxiety, and fear. That did not stop him from trying to
find out why those perceptions had such an effect on him. He began
seeking even more desperately, both when he was in front of the
house and later, in retrospect, a rational explanation for what he had
witnessed. For this reason, he not only re-examined the perceptions
themselves, but also engaged in meta-processing, analysing his own
way of thinking.
The following is an example of how he tried to explain the cause of
the unusual effect his looking at the house and its reflection in the tarn
had on him (Poe 1986: 140):
(24) I have said that the sole effect of my somewhat childish
experiment – that of looking down within the tarn – had been to
deepen the first singular impression. (25) There can be no doubt
that the consciousness of the rapid increase of my superstition –
for why should I not so term it? – served mainly to accelerate
the increase itself. (26) Such, I have long known, is the
paradoxical law of all sentiments having terror as a basis. (27)
And it might have been for this reason only, that, when I again
uplifted my eyes to the house itself, from its image in the pool,
there grew in my mind a strange fancy – a fancy so ridiculous,
indeed, that I but mention it to show the vivid force of the
sensations which oppressed me.
In this passage, the narrator ascribes the cause of the ‘first singular
impression’ and ‘strange fancy’ to his becoming aware of his own
superstitiousness. It is interesting that here, too, the narrator seems
confident about the truth-value of his explanation. This shows itself in
his employing a high-degree modal to assert that ‘there can be no
doubt’ that he is right. However, as we have seen with other examples,
it is exactly such high-degree modals that require caution, since
speakers only say that they are certain when they are not (Halliday and
Matthiessen 2004: 147). In this sentence this means that the narrator in
reality is not entirely convinced that ‘consciousness of the rapid
increase of his superstition’ indeed served to ‘accelerate the increase
itself’. If he were, he could have expressed the proposition as a polar
statement.
There are other signs of the narrator’s uncertainty in the passage.

Another can be found in sentence (27) when he explains why ‘there
grew in [his] mind a strange fancy’ when he ‘uplifted his eyes to the
house itself, from its image in the pool’. He believes that this, too, was
caused by his becoming aware of his own superstition. However, he is
once more reluctant to use a polar statement. Instead, he again adds a
proviso to the proposition, saying that it ‘might have been’ this reason
that caused a ‘strange fancy’ to grow in his mind, i.e. maybe it was
this reason, maybe it was not.
Finally, we should also not overlook the third indication of his
uncertainty, the generic sentence. As we can see in sentence (26), the
narrator claims that it is the ‘law of all sentiments having terror as a
basis’ that becoming conscious of one’s superstition only accelerates
the increase of that superstition. The narrator formulates the sentence
as if he were making a claim that is necessarily always and invariably
true. The two signs of its ‘genericness’ are the present tense and the
fact that the sentence could easily be introduced by the formulation ‘It
is always the case that…’ (Toolan 1998: 60). At the same time,
however, the narrator does not have any evidence to support his claim.
All that he has to prove the validity of his claim is his saying that he
‘has long known’ the ‘law’ to be as he describes it. It is clear,
therefore, that the narrator is trying to find yet another way of saying
that ‘there is no doubt’ about the truth of the proposition. What that
means, we have already seen in other examples where high-degree
modals were used.
7. Conclusion
This last passage appropriately rounds off our examination of the role
of modality, particular that of uncertainty, in the narrative. Like a
number of other examples from the text, it shows how important a
component it is of the narrator’s discourse. It is the means which the
narrator uses to add uncertainty to his narrative. It is indicative of the
narrator’s difficulties in dealing with the events and phenomena he
had witnessed at the House of Usher. It shows that in a number of
cases throughout the narrative, the narrator is not able to provide a
rational explanation for what he had experienced, whether in terms of
his perceptions, of his mental processing of those perceptions, or of
what he had known before the visit. For all these reasons, the narrator
lapses into the domain of speculation and guesswork.
470 Simon Zupan
The study has examined the implications of uncertainty for the

narrator’s mind-style. It has shown that uncertainty is directly
connected with the way the narrator, and consequently the reader,
perceive the developments in the story world. The uncertainty that
marks the narrator’s perception of his visit to the House of Usher is
the vehicle for at least some of the Gothic effects of the narrative. It
turns out to create pockets of the unknown, the mysterious, and the
incomprehensible throughout the text. These pockets then have the
potential to become sources of discomfort, anxiety, and fear. This
study has thus shown that the categories of mind-style and modality
can be successfully combined and made complementary, when
exploring the effects that modal structures at the micro-level can exert
on the macro-level of the text.
References
Evans, W. 1977. ‘“The Fall of the House of Usher” and Poe’s theory of the tale’ in
Studies in Short Fiction 14: 137-144.
Fowler, R. 1977. Linguistics and the Novel, New accents. London: Methuen.
__. 1996. Linguistic Criticism, Opus. Oxford: Oxford University Press.
Haggerty, G. E. 1989. Gothic fiction/Gothic form. University Park: Pennsylvania State
University Press.
Halliday, M. 1996. ‘Linguistic function and literary style: An inquiry into the
language of William Golding’s The Inheritors’ in The Stylistics Reader: From
Roman Jakobson to the present (ed.) J. J. Weber. London: Arnold: 56-86.
Halliday, M. A. K., and C. M. I. M. Matthiessen. 2004. An Introduction to Functional
Grammar. London: Arnold.
Hustis, H. 1999. ‘“Reading Encrypted But Persistent”: The Gothic of Reading and
Poe's “The Fall of the House of Usher”’ in Studies in American Fiction 27 (1): 3-
20.
Leech, G. N., and M. H. Short. 1995. Style in Fiction: A Linguistic Introduction to
English Fictional Prose (ed.) R. Quirk, English Language Series. London and
New York: Longman. Original edition, 1981.
Obuchowski, P. 1975. ‘Unity of effect in Poe’s “The Fall of the House of Usher”’ in
Studies in Short Fiction 12: 407-412.
Peeples, S. 2002. ‘Poe’s “constructiveness” and “The Fall of the House of Usher”’ in
The Cambridge Companion to Edgar Allan Poe (ed.) K. J. Hayes. Cambridge:
Cambridge University Press: 178-190.
Poe, E. A. 1986. The Fall of the House of Usher and Other Writings. London:
Penguin.
Quirk, R., S. Greenbaum, G. N. Leech and J. Svartvik. 1992. A Comprehensive
Grammar of the English Language. London in New York: Longman.
Rountree, T. J. 1972. ‘Poe’s Universe: The House of Usher and the Narrator’ in
Tulane Studies in English 20 (4): 123-134.
Semino, E., and K. Swindlehurst. 1996. ‘Metaphor and mind style in Ken Kesey’s
One Flew Over the Cuckoo’s Nest’ in Style 30 (1): 143-167.
http://search.epnet.com/login.aspx?direct=true&db=aph&an=9608291448.
2/2/2006.
Short, M. H. 1992. ‘Mind-Style’ in The European English Messenger 1 (3): 31-33.
Toolan, M. J. 1998. Language in Literature: An Introduction to Stylistics. London:
Arnold.
Uspenski, B. 1979. Poetika kompozicije. [Trans. N. Petkoviü]. Beograd: Nolit.
Voloshin, B. R. 1986. ‘Explanation in “The Fall of the House of Usher”’ in Studies in
Short Fiction 23 (4): 419-428.
Bibliography
Achebe, Chinua. 1958. Things Fall Apart. London: Heinemann.

__. 1964. Arrow of God. London: Heinemann.
Alexie, Sherman. 2004. ‘What You Pawn, I Will Redeem’ in Moore,
Lorrie, and Katrina Kenison (eds) The Best American Short Stories
2004. Boston: Houghton Mifflin.
Allen, Walter. 1954. The English Novel. London: Phoenix House.
Allwright, D. 2005. ‘From teaching points to learning opportunities
and beyond’ in TESOL Quarterly 39(1) 9-31.
Altick, R.D. 1963. ‘“A Grammarian’s Funeral”: Browning’s Praise of
Folly?’ in Studies in English Literature 3(4): 449-60.
Andersen, John Edelsgaard. 1986. ‘Adverbier – sprogvidenskabens
stedbørn’ in Beiträge zur nordischen Philologie 15.
Andreev, S. 2006. ‘A Diachronic Study of the Style of Longfellow’ in
Journal of Quantitative Linguistics 10: 2-12.
Apollinaire, Guillaume. 1956. ‘Fusée-Signal’ in Œuvres poétiques
(ed.) Marcel Adéma and Michel Décaudin. Paris: Gallimard: 363.
Apte, Mahadev L. 1985. Humour and Laughter: An Anthropological
Approach. New York: Cornell University Press.
Arai, K. 2002. Syntactic Ellipsis and Non-Syntactic Ellipsis – Haiku,
the Literature of Ellipsis, Eigo Gohou Bunpou Kenkyu 9. Tokyo:
Kaitakusha.
Aristotle. 1984a. Rhetoric (tr. W. Rhys Roberts) in The Rhetoric and
the Poetics of Aristotle. New York: Modern Library: 19-218
__. Poetics of Aristotle. New York: Modern Library: 223-66.
Armstrong, I. 1993. ‘Browning in the 1850s and after’ in Victorian
Poetry. London: Routledge: 284-317.
Arnheim, Rudolf. 1966. ‘Abstract Language and the Metaphor’ in
Towards a Psychology of Art. Berkeley and Los Angeles:
University of California Press: 266-82.
Ashline, W.L. 1995. ‘The problem of impossible fictions’ in Style
29(2): 215-34.
Baayen, R.H., H. Van Halteren and F. Tweedie. 1996. ‘Outside the
cave of shadows: Using syntactic annotation to enhance authorship
attribution’ in Literary and Linguistic Computing 11: 121-131.
Baevskij, V.S. 1993. Pasternak - Lirik. [Pasternak the lyric poet].
Smolensk: Trust-Imakom.
474 Bibliography
Bakhtin, Mikhail. [1965] 1984. Rabelais and His World (tr. Helene
Iswolsky). Bloomington: Indiana University Press.
__, and P. M. Medvedev. 1985. The Formal Method in Literary
Scholarship trsl A.J. Wehrle. Cambridge, MA.: Harvard University
Press.
Bal, M. 1980. Narratology: Introduction to the Theory of Narrative.
Toronto: University of Toronto Press.
Banks, Wishelle. 1995. ‘Interviews: Alexie finds inspiration in his
home, humor in his family and their life’ The Native Voice.
Available at <http://www.fallsapart.com/art-nv.html> accessed 28
Feb 2005.
Banó, I. 1973. Módszertan. Budapest: Tankönyvkadó.
Barnes, B. 1988. The Nature of Power. Cambridge: Polity Press.
Barry, P. 1995. Beginning Theory. Manchester: Manchester
University Press.
Barthelme, Donald. 1994. 40 Stories. New York: Penguin.
Barthes, Roland. 1977. ‘Introduction to the structural analysis of
narratives’ in Image-Music Text. London: Fontana.
__. 1993a. Le degreé zero de l’écriture in Œuvres complètes, Vol. I
(ed.) Éric Marty. Paris: Seuil: 139-86.
__. 1993b. Éléments de sémiologie in Œuvres complètes, Vol. I (ed.)
__. 1994a. ‘La mort de l’auteur’ in Œuvres complètes, Vol. II (ed.)
__. 1994b. S/Z in Œuvres complètes, Vol. II (ed.) Éric Marty. Paris:
Seuil: 557-728.
__. 1994c. ‘De l’œuvre au texte’ in Œuvres complètes, Vol. II (ed.)
__. 1994d. Le Plaisir du texte in Œuvres complètes, Vol. II (ed.) Éric
Marty. Paris : Seuil: 1495-1529.
__. 1995. ‘Arcimboldo ou Rhétoriqueur et magician’ in Œuvres
complètes, Vol. III (ed.) Éric Marty. Paris: Seuil: 854-69.
Batty, M. 2001. Harold Pinter. Tavistock: Northcote House.
Baudelaire, Charles. 1961. ‘De l’essence du rire’ in Œuvres completes
(ed.) Claude Pichois. Paris: Gallimard: 975-93.
de Beaugrande, Robert. 1980. Text, Discourse and Process. Newark,
NJ: ABLEX Publishing Corporation.
Beckett, S. 1986. Samuel Beckett: The Complete Dramatic Works.
London: Faber and Faber.
Bibliography 475
Bennison, N. 1993. ‘Discourse Analysis, pragmatics and the dramatic

character: Tom Stoppard’s Professional Foul’ in Language and
Bergs, Alexander T. 2004. ‘Letters. A new approach to text typology’
in Journal of Historical Pragmatics 5(2): 207-227.
Biber, Douglas. 1988. Variation across speech and writing.
__, and Edward Finegan. 1989a. ‘Styles of stance in English: Lexical
and grammatical marking of evidentiality and affect’ in Text 9(1):
93–124.
__, and Edward Finegan. 1989b. ‘Drift and the evolution of English
style. A history of three genres’ in Language 65(3): 487-517.
Bijkerk, Annemieke. 2004. ‘Yours sincerely and yours affectionately.
On the origin and development of two positive politeness markers’
in Journal of Historical Pragmatics 5(2): 297-311.
Binns, Ronald. 1977. ‘A New Version of The Magus’ in Critical
Quarterly 19(4): 79-84.
Blake, N. F., D. Burnley, M. Matsuo and Y. Nakao (eds). 1994. A
New Concordance to The Canterbury Tales Based on Blake’s Text
Edited from the Hengwrt Manuscript. Okayama: University
Education Press.
Bloomfield, Lou. 2002. The Plagiarism Resource Center.
http://plagiarism.phys.virgina.edu.
Blyth. R.H. 1981. Haiku, Volume one: Eastern Culture.Tokyo: The
Hokuseido Press.
Boase-Beier, Jean. 2004. ‘Knowing and not knowing: Style, intention
and the translation of a Holocaust poem’ in Language and
Boccia, Michael. 1980/1981. ‘“Visions and Revisions”: John Fowles’s
New Version of The Magus’ in Journal of Modern Literature 8(2):
235-46
Bohn, A. 2006. ‘Increasing suspicion about Browning’s grammarian’
in Victorian Poetry 44 (2): 165-182, 235.
Bohn, Willard. 2002. The Rise of Surrealism: Cubism, Dada, and the
Pursuit of the Marvelous. Albany: State University of New York
Press.
Bolinger, Dwight. 1977. ‘Neutrality, norm and bias.’ Bloomington:
Indiana University Linguistics Club.
476 Bibliography
Booth, Wayne C. [1961] 1983. The Rhetoric of Fiction. London:

Penguin Books.
Borchmann, Simon Uffe. 2005. Funktionel tekstteori og fiktivt
fortællende tekster med refleksiv funktion. Frederiksberg: Roskilde
Universitetsforlag.
Borg, S. 2003. ‘Teacher cognition in language teaching: a review of
researh on what teachers think, know, believe and do’ in Language
Teaching 36: 81-109.
Bosseaux, C. 2001. ‘A Study of the Translator’s Voice and Style in
the French Translations of Virginia Woolf’s The Waves’ in CTIS
Occasional Papers 1: 55-72.
Bousfield, D. In press 2007a. ‘Impoliteness, Preference Organization
and Conducivity’in Multilingua.
__. In press 2007b. ‘Beginnings, middles and ends: towards a biopsy
of the dynamics of impolite exchanges’in Journal of Pragmatics.
Bradbury, Malcolm. 1973. Possibilities: Essays on the State of the
Novel. London: Oxford University Press.
Bradford, R. 1997. Stylistics. London: Routledge.
Brady, Kristin. 1982. The Short Stories of Thomas Hardy: Tales of
Past and Present. London: Macmillan.
Brandt, Line. 2006. ‘Dramatization in the Semiotic Base Space: A
Semiotic Approach to Fictive Interaction as a Representational
Strategy in Communicative Meaning Construction’ in Oakley,
Todd, and Anders Hougaard (eds) Mental Spaces Approaches to
Discourse and Interactio. Amsterdam: John Benjamins. See also:
http://www.hum.au.dk/semiotics/docs2/pdf/brandt_line/dramatizati
on.pdf
__. (forthcoming). ‘Cognitive Poetics from a Semiotic Perspective: A
Stratified Model for Reading and Interpretation’ in Brône and J.
Vandaele (eds) Applications of Cognitive Linguistics. Berlin &
New York: Mouton de Gruyter.
__, and Per Aage Brandt. 2005a. ‘Cognitive Poetics and Imagery’ in
European Journal of English Studies Vol. 9, No. 2 August 2005:
117-130. See also: http://www.hum.au.dk/semiotics/docs2/pdf/
brandt&brandt/cognitive_poetics.pdf
__, and Per Aage Brandt. 2005b. Making Sense of a Blend – A
Cognitive-Semiotic Approach to Metaphor. Annual Review of
Cognitive Linguistics 3, 2005: 216-249. See also: http://www.hum.
au.dk/semiotics/docs2/pdf/brandt&brandt/making_sense.pdf
Bibliography 477
Brandt, Per Aage. 1983. Sandheden, saetningen og döden –

Semiotiske aspekter af kulturanalysen. Copenhagen: Basilisk.
__. 1989. ‘Genese og diegese – Et problem i den almene narratologi.’
Religionsvidenskabeligt tidsskrift, vol 14: 75-85.
__. 2004a. Spaces, Domains and Meaning. Essays in Cognitive
Semiotics. Bern: Peter Lang.
__. 2004b. ‘Evidentiality and Enunciation. A Cognitive and Semiotic
Approach’ in Arrese, J.M. (ed.) Perspectives on Evidentiality and
Modality. Madrid: Editorial Complutense. Columbia Encyclopedia,
The, Sixth Edition, 2004. New York: Columbia University Press.
Brémond, Claude. 1973. Logique du Récit. Paris: du Seuil.
Bremond, C. 1966. ‘La Logique des Possibles Narratifs’ in
Communications 4 : 4-32.
Bremond, C., J. Landy and T. Pavel (eds). 1995. Thematics. New
Approaches. Albany: State University of New York Press.
Breton, André. 1988. Manifestete du Surréalisme in Œuvres
complètes, Vol. I (ed.) Marguerite Bonnet. Paris: Gallimard, 1988:
311-46.
Brinker, M. 1995. ‘Theme and Interpretation’ in Bremond, C., J.
Landy and T. Pavel (eds) Thematics. New Approaches. New York:
State University of New York Press: 33-44.
Bristow, J. (1991) Robert Browning. New York and London:
Harvester Wheatsheaf.
Britton, D. 2002. A Haiku Journey – Basho’s Narrow Road to a Far
Province. Tokyo: Kodansha International.
Brooke-Rose, Christine. 1958. A Grammar of Metaphor. London:
Secker & Warburg.
Brown, P., and S.C. Levinson. 1987. Politeness: Some Universals in
Language Usage. Cambridge: Cambridge University Press.
Brown, Roger W. and Albert Gilman. 1960. ‘The pronouns of power
and solidarity’ in Sebeok, Thomas A. (ed.) Style in language. New
York: Wiley: 253-176.
__, and A. Gilman. 1989. ‘Politeness Theory and Shakespeare’s Four
Major Tragedies’ in Language in Society 18: 159-212.
Brown, S. 2002. ‘What Happens When People Reread?’ Paper
presented at the Annual World Congress of the Association
Internationale de Linguistique Appliquée. Singapore. December
16-21.
478 Bibliography
Browning, R. [1855] 1995 Men and Women. The Poetical Works of

Robert Browning. Volume 5.(eds) I. Jack and R. Inglesfield.
Oxford: Clarendon.
Brumfit, Christopher J., and Ronald A. Carter (eds). 1986. Literature
and Language Teaching. Oxford: Oxford University Press.
Brusasco, P. 2004. ‘A Computer-assisted Reading of Nathaniel
Hawthorne’s The Scarlet Letter’ in English Studies 2003. Torino:
Trauben: 83-95.
Burke, Edmund. 1990. A Philosophical Enquiry into the Origin of our
Ideas of the Sublime and Beautiful (ed.) Adam Philips. Oxford and
New York: Oxford University Press.
Bursey, J., and A. Furlong. 2006. ‘Cognitive Gothic: Relevance
theory, iteration, and style’ in Tabbi, J., and R. Shavers (eds) Paper
Empire: William Gaddis and the World System. Tuscaloosa: U of
Alabama P.
Burton, Deirdre. 1980. Dialogue and Discourse: A Sociolinguistic
Approach to Modern Drama Dialogue and Naturally Occurring
Conversation. London: Routledge and Kegan Paul.
__. 1982. ‘Conversation Pieces’ in Carter, Ronald and Deirdre Burton
(eds) Literary Text and Language Study. London: Edward Arnold.
Busbridge, E.M. 1909. Letter-writing and etiquette. London: Collins.
Byron, G. 2003. Dramatic Monologue. The New Critical Idiom.
London: Routledge.
Campion, Thomas. 1969. ‘There Is a Garden in Her Face’ in The
Works of Thomas Campion: Complete Songs, Masques, and
Treaties on a Selection of the Latin Verse (ed.) Walter R. Davis.
London: Faber & Faber: 174.
Carrell, Amy. 1997. ‘Joke competence and humour competence’ in
Humor 10(2): 173-185.
Carroll, David. 1980. Chinua Achebe. London: Macmillan & Co. Ltd.
Carston, R. 2002. Thoughts and Utterances: The pragmatics of
explicit communication. Oxford: Blackwell.
Carter, Ronald A. 2004. Language and Creativity. The Art of Common
Talk. London: Routledge.
__, and M.N. Long. 1991. Teaching Literature. London: Longman.
__, and John McRae. 1996. Language, Literature and the Learner:
Creative Classroom Practice. London: Longman.
__, and J. McRae (eds). 1999. Language, Literature and the Learner.
London: Pearson Education.
Bibliography 479
__, and Walter Nash. 1990. Seeing through Language. Oxford:

Blackwell.
Caws, M. A. 1989. ‘Narrative Voice and Second Reading: Relation
and Response’ in Poetics Today 10(2): 243-253.
CBC Television: Comedy <http://www.cbc.ca/justforlaughs/index.
htm/> 5th May 2006.
Chatman, S. (ed.). 1971. Literary Style: A Symposium. London:
__. 1978. Story and Discourse. Ithaca, Cornell University Press.
Chiarella, Tom. 1998. Writing Dialogue: How to create memorable
voices and fictional conversations that crackle with wit, tension
and nuance. Cincinnati: Story Press.
Chomsky, N. 1976. Reflections on language. London: Temple Smith.
Chris. ‘Rereads.’ Message posted to Fool.Co.UK. Retrieved 15 July,
2006, from http://boards.fool.co.uk/Message.asp?mid=10037804&
sort=postdate.
Claridge, Claudia. 2006. ‘“With the most superlative felicitee”:
Functions of the superlative in 19th-century English’ in Dalton-
Puffer, Christiane, Dieter Kastovsky, Nikolaus Ritt and Herbert
Schendl (eds) Syntax, style and grammatical norms. Bern: Peter
Lang AG: 73-89.
__. 2007. ‘The superlative in spoken English’ in Facchinetti, Roberta
ed. Corpus linguistics 25 years on. Amsterdam: Rodopi: 123-150.
Clark, B. 1996. ‘Stylistic analysis and relevance theory’ in Language
and Literature 5(3): 163-178.
Claudia. ‘Comfort re-reading’. Weblog. Retrieved 7 July, 2006, from
http://clauclauclaudia.livejournal.com/tag/books.
Clough, Paul. 2002. ‘Plagiarism in natural and programming
languages: an overview of current tools and technologies.’ (No.
CS-00-05): Department of Computer Science, University of
Sheffield, UK.
__. 2003. Measuring text reuse. PhD Thesis, University of Sheffield.
__, and Robert Gaizauskas et al. 2002a. ‘Building and annotating a
corpus for the study of journalistic text reuse.’ Conference
Proceedings of the 3rd International Conference on Language
Resources and Evaluation (LREC-2002) (pp.1678-91), Los Palmas
de Gran Canaria, Spain.
480 Bibliography
__, and Robert Gaizauskas et al. 2002b. ‘METER: MEasuring TExt

Reuse.’ Conference Proceedings of the ACL-2002, University of
Pennsylvania, Philadelphia, USA.
Coates, Jennifer. 1996. Women talk. Conversation between women
friends. Oxford: Blackwell.
Collins COBUILD Wordbanks Online.
CONCE = A Corpus of Nineteenth-Century English, being compiled
by Merja Kytö (Uppsala University) and Juhani Rudanko
(University of Tampere).
Consiglio, M.C. 2006. Traduzione remake manipolazione: King Lear
neoclassico, unpublished doctoral dissertation.
Cook, David. 1977. African Literature: A Critical View. London:
Longman.
Cook, Guy. 1992. Discourse of Advertising. London: Routledge.
__. 1994. Discourse and Literature: The Interplay of Form and Mind.
Oxford: Oxford University Press.
Cornis-Pope, M., and A. Woodlief. 2000. ‘The Rereading/Rewriting
Process: Theory and Collaborative, On-line Pedagogy’ in M.
Helmers (ed.) Intertexts: Reading Pedagogy in College Writing
Classrooms. Mahwah. Princeton, NJ: Lawrence Erlbaum.
Corsaro,W., and T.A. Rizzo. 1990. ‘Disputes in the peer culture of
American and Italian nursery-school children’ in Grimshaw, A.D
(ed.) Conflict Talk: Socioloinguistic Investigations of Arguments
and Conversations. Cambridge: Cambridge University Press.
Crashaw, Richard. 1957. ‘The Weeper [1646]’ in The Poems, English,
Latin and Greek of Richard Crashaw (ed.) L. C. Martin. Oxford:
Clarendon Press: 83.
Crisp, Peter. 2002. ‘Conceptual Metaphor and Its Expressions’ in
Gavins, Joanna, and Gerard Steen (eds) Cognitive Poetics in
Practice. London: Routledge: 99-113.
Crowther, J. et al. (eds). 2000. Oxford Advanced Learner’s Dictionary
of Current English (6th edn). Oxford: Oxford University Press.
Culpeper, J. 1996. ‘Towards an Anatomy of Impoliteness’ in Journal
of Pragmatics 25: 349-367.
__. 1998. ‘Impoliteness in Drama’ in Culpeper, J., M. Short and P.
Verdonk (eds) Exploring the Language of Drama. London:
Routledge.
__. 2001. Language and Characterisation: People in Plays and Other
Texts. London: Longman.
Bibliography 481
__. 2005. ‘Impoliteness and Entertainment in the television quiz show

The Weakest Link’ in Journal of Politeness Research 1: 35-72.
__, D. Bousfield and A. Wichmann. 2003. ‘Impoliteness Revisited:
with special reference to dynamic and prosodic effects’ in Journal
of Pragmatics 35: 1545-1579.
__, Mick Short and Peter Verdonk. 1998. Exploring the Language of
Drama: From Text to Context. London: Routledge.
Curl, T.S., J. Local and G. Walker. 2006. ‘Repetition and the prosody-
pragmatics interface’ in Journal of Pragmatics 38(10): 1721-1751.
Dahl, Östen. 2001. ‘Grammaticalization and the life cycles of
constructions’ in RASK - Internationalt tidsskrift for sprog og
kommunikation 14: 91-134.
Dancygier, Barbara. 2007. ‘Narrative Anchors and the Processes of
Story Construction: The Case of Margaret Atwood’s The Blind
Assassin’. Style 42(1): 133-52.
Danto, A. C. 1981. The transfiguration of the commonplace: a
philosophy of art. Cambridge, Massachusetts: Harvard University
Press.
Davidson, Donald. 1979. ‘What Metaphors Mean’ in On Metaphor
(ed.) Sheldon Sacks. Chicago: University of Chicago Press: 29-45.
Davis, Norman. 1965. The Litera Troili and English letters. R.E.S.
New Series. Vol. XVI, No. 63: 233-244.
Dawson, E. 2007. PhD thesis: Emotion Tracking Pedagogy. (The
University of Nottingham, UK).
Dawson, Michael R. W. 1998. Understanding Cognitive Science.
Oxford: Basil Blackwell.
Deignan, A. 1995. Collins Cobuild English Guides 7: Metaphor
London: HarperCollins.
Department for Education and Employment, Qualifications and
Curriculum Authority. 1999. The National Curriculum for England
Key Stages 1-4 English London: Qualifications and Curriculum
Authority.
Derrida, Jacques. [1967] 1978. Writing and Difference (tr. Alan Bass).
Chicago: University of Chicago Press.
__. 1992. Acts of Literature. (ed.) D. Attridge. New York and London:
Routledge.
Dik, Simon. 1989. The Theory of Functional Grammar. Part 1: The
Structure of the Clause. Dordrecht: Foris Publications.
482 Bibliography
Dijk, Teun A. van (ed.). 1985. Discourse and Literature. Amsterdam:

John Benjamin Publishing Co.
Doležel, Lubomir. 1971. ‘Toward a structural theory of content in
prose fiction’ in Chatman, S. (ed.) Literary style: a symposium.
London: Oxford University Press: 95-110.
__. 1988. ‘Mimesis and possible worlds’ in Poetics Today 1(3): 7-25.
__. 1989. ‘Possible worlds and literary fictions’ in Allén, S. (ed.)
Possible Worlds in Humanities, Arts and Sciences: Proceedings of
Nobel Symposium 65. New York: de Gruyter: 223-42.
__. 1998. Heterocosmica: Fiction and Possible Worlds. Baltimore:
The Johns Hopkins University Press.
__, and Richard W. Bailey. (eds). 1969. Statistics and Style. New
York: American Elsevier.
Ebbatson, Roger. 1993. Hardy: The Margin of the Unexpressed.
Sheffield: Sheffield Academic Press.
Eckermann, Johann Peter. [1827] 1948. Gespräche mit Goethe, in
Ernst Beutler (ed.) Gedenkausgabe der Werke, Briefe und
Gespräche [Goethes] 24. Zürich: Artemis-Verlag.
Eco, U. 1979. The Role of the Reader. London: Hutchinson.
Eggins, Suzanne, and Diana Slade. 1997. Analyzing Casual
Conversation. London: Cassell.
Elbaz, F. 1990. ‘Knowledge and Discourse. The evolution of research
on teacher thinking’ in Day, C., M. Pope and P. Denicolo (eds)
Insight into Teachers’ Thinking and Practice. London: Falmer: 1-
19.
Eliot, T. S. 1963. ‘Whispers of Immortality’ in Collected Poems:
1909-1962 (ed.) T. S. Eliot. New York and London: Harcourt: 45.
Emmott, Catherine. 1997. Narrative Comprehension: A Discourse
Perspective. Oxford: Oxford University Press.
__. 2002. ‘“Split Selves” in Fiction and in Medical “Life Stories”:
Cognitive Linguistic Theory and Narrative Practice’ in Semino and
Culpeper 2002: 153-82.
__. 2002. ‘Responding to Style: Cohesion, Foregrounding, and
Thematic Interpretation’ in Louwerse, M., and W. van Peer (eds)
Thematics. Interdisciplinary Studies. Amsterdam/Philadelphia:
John Benjamins Publishing Company: 91- 118.
__. 2003. ‘Reading for pleasure: A cognitive poetic analysis of “twists
in the tale” and other plot reversals in narrative texts’ in Cognitive
Bibliography 483
Poetics in Practice (ed.) Gavins, J., and G. Steen. London:

Routledge: 145-159.
__, Anthony J. Sanford and Eugene J. Dawydiak. 2007. ‘Stylistics
Meets Cognitive Science: Studying Style in Fiction and Readers’
Attention from an Inter-disciplinary Perspective’ in Style 41(2):
204-24.
Enkvist, Nils Erik. 1988. ‘Style as parameters in text strategy ’ in
Willie van Peer (ed.) The taming of the text: explorations in
language, literature and culture. London: Routledge: 125-51.
Enyedi, Á., and P. Medgyes. 1995. ‘Angol nyelvoktatás Közép- és
Kelet-Európában a rendszerváltozás óta’ in Modern Nyelvoktatás
4/23: 1232.
Esslin, M. 1978. An Anatomy of Drama. London: Abacus. 2nd edition.
__. 1980. The Theatre of the Absurd. London: Penguin. 3rd edition.
Evans, Dylan. 1996. An Introductory Dictionary of Lacanian
Psychoanalysis. Hove and New York: Brunner-Routledge.
Evans, D. 2005. 26a. London: Chatto & Windus.
Evans, W. 1977. ‘“The Fall of the House of Usher” and Poe’s theory
of the tale’ in Studies in Short Fiction 14: 137-144.
Fabb, N. and A. Durant 1987a. ‘Introduction: The linguistics of
writing: retrospect and prospect after twenty five years’ in Fabb, N.
et al. 1987: The linguistics of writing; Arguments between
language and literature. Manchester: Manchester University Press.
__, and A. Durant 1987b. ‘New courses in the linguistics of writing’
in Fabb, N. et al. 1987: The linguistics of writing; Arguments
between language and literature. Manchester: Manchester
University Press.
Farkas, Donna F., and K.E. Kiss. 2000. ‘On the comparative and
absolute readings of superlatives’ Natural language and linguistic
theory 18: 417-455.
Fast, Robin Riley. 1999. The Heart as a Drum: Continuance and
Resistance in American Indian Poetry. Chicago: University of
Michigan Press.
Fauconnier, Gilles. 1980. ‘Pragmatic entailment and questions’ in
Searle, John R., Ferenc Kiefer and Manfred Bierwisch (eds)
Speech act theory and pragmatics. Dordrecht: Reidel: 57-69.
__, and Mark Turner. 2002. The Way We Think. Conceptual Blending
and the Mind’s Hidden Complexities. New York: Basic Books.
484 Bibliography
Fink, Bruce. 1995. The Lacanian Subject: Between Language and

Jouissance. Princeton: Princeton University Press.
Fish, S. E. 1973. ‘What is stylistics and why are they saying such
terrible things about it?’ in S. Chatman (ed.) Approaches to
Poetics: Selected Papers from the English Institute. New York:
Columbia University Press: 109-52.
Fish, Stanley. [1973] 1995. ‘How Ordinary Is Ordinary Language’ in
Is There a Text in This Class? Cambridge: Harvard University
Press.
Fitzmaurice, Susan. 2002. The familiar letter in early modern English:
A pragmatic approach. Amsterdam: John Benjamins.
Fludernik, Monika. 1993. The Fictions of Language and the
Languages of Fiction. London: Routledge.
Fodor, J. A. 1993. ‘Déjà vu all over again: How Danto’s aesthetics
recapitulates the philosophy of mind’ in Rollins, M. (ed.) 1993:
Danto and his Critics. Oxford: Blackwell.
Forster, E.M. 1927. Aspects of the Novel. Hammondsworth,
Middlesex: Penguin Books Ltd.v
Foster, J. B. 1996. ‘Faces of Rereading’ in Poetics Today 17(2): 253-
261.
Foucault, M. 1974. The Archaeology of Knowledge trsl. A.M.
Sheridan Smith. London: Tavistock.
Fowler, Roger (ed.). 1966. Essays on Style and Language: Linguistic
and Critical Approaches to Literary Style. London: Routledge and
Kegan Paul.
__ (ed.). 1971. Style and Structure in Literature: Essays in the New
Stylistics. London: Routledge and Kegan Paul.
__. 1977. Linguistics and the Novel. London: Methuen.
__. [1986] 1996. Linguistic Criticism. Oxford: Oxford University
Press.
__. 1996. ‘Studying literature as language’ in Weber, J. J. 1996: The
Stylistics Reader. London: Arnold.
Fowles, J. 1966. The Magus. London: World Book.
__. 1977. The Magus. London: Vintage.
Frank, Joseph. 1991. The Idea of Spatial Form. New Brunswick:
Rutgers University Press.
Fraser, B. and W. Nolan. 1981. ‘The Association of Deference with
Linguistic Form’ in International Journal of the Sociology of
Language 27: 93-109.
Bibliography 485
Freeman, Donald C. (ed.). 1970. Linguistics and Literary Style. New

York: Holt, Rinehart and Winston.
Freeman, M. H. 2002. ‘Cognitive Mapping in Literary Analysis’ in
Style 36: 466- 483.
Freud, Sigmund. [1933] 1911c. New Introductory Lectures on
Psychoanalysis (tr. James Strachey). Penguin Freud Library, 2.
London: Penguin Books.
__. [1905] 1966. Jokes and their Relation to the Unconscious.
London: Routledge &. Kegan Paul.
__. [1900] 1991a. The Interpretation of Dreams (tr. James Strachey).
__. [1916-17] 1991b. Introductory Lectures on Psychoanalysis (tr.
James Strachey). Penguin Freud Library, 1. London: Penguin
Books.
__. [1940] 1993. An Outline of Psychoanalysis (tr. James Strachey).
Penguin Freud Library, 15. London: Penguin Books: 375-443.
Frye, N. 1986. Shakespeare. Nove Lezioni. Torino: Einaudi.
Furlong, A. 1996. ‘Relevance theory and literary interpretation.’
Unpublished doctoral dissertation, University College, London,
UK.
Gabelentz, Georg van der. 1891. Die Sprachwissenschaft. Ihre
Aufgaben, Methoden, und bisherige Ergebnisse. Leipzig: Weigel.
Gaizauskas, Robert, and Jonathan Foster et al. 2001. ‘The METER
corpus: a corpus for analysing journalistic text reuse.’ Conference
Proceedings of Corpus Linguistics 2001, Lancaster, UK.
Galasinski, D. 2000. The Language of Deception. A Discourse
Analytical Study. Thousand Oaks, CA: Sage.
Gallop, Jane. 1985. Reading Lacan. Ithaca: Cornell University Press.
Garrod, Simon. 1995. ‘Distinguishing between Explicit and Implicit
Focus during Text Comprehension’ in Richeit, G., and C. Habel
(ed.) Focus and Coherence in Discourse Processing. Berlin: Walter
de Gruyter.
Gasparov, M.L. 1974. Sovremennij russkij stikh. Metrika i ritmika.
[Contemporary Russian verse. Metrics and rhythmics.]. Moscow:
Nauka.
Gavins, Joanne. 2007. Text World Theory: An Introduction.
Edinburgh: Edinburgh University Press.
Gavins, Joanna, and Gerard Steen (eds). 2003. Cognitive Poetics in
Practice. London: Routledge.
486 Bibliography
Gee, J. P. 1996. Social Linguistics and Literacies. London: Taylor &

Francis, first edition in 1990.
Geisler, Christer. 2002. ‘Investigating register variation in nineteenth-
century English: A multi-dimensional comparison’ in Reppen,
Randi, Susan M. Fitzmaurice and Douglas Biber (eds) Using
corpora to explore linguistic variation. Amsterdam/Philadelphia:
John Benjamins: 249–271.
__. 2003. ‘Gender-based variation in nineteenth-century English letter
writing’ in Leistyna, Pepi and Charles F. Meyer (eds) Corpus
analysis: Language structure and language use. Amsterdam:
Rodopi: 87-106.
Gerrig, R. J. 1989. ‘Reexperiencing Fiction and Non-Fiction’ in The
Journal of Aesthetics and Art Criticism 47(3): 277-280.
Ghent, Dorothy van. 1953. The English Novel: Form and Function.
New York: Cornell University Press.
Godkewitsch, M. 1974. ‘Correlates of humor: Verbal and nonverbal
aesthetic reactions as functions of semantic distance within
adjective-noun pairs’ in Studies in the new experimental aesthetics
(ed.) Berlyne, D. E. Washington, DC: Hemisphere: 279-304.
Goffman, E. 1967. Interactional Ritual: Essays on Face-to-Face
Behaviour. New York: Anchor Books.
Gombrich, E. H. 1979. The Sense of Order: A Study in the Psychology
of Decorative Art. London: Phaidon.
Graesser, A., V. Pomeroy and S. Craig. 2002. ‘Psychological and
Computational Research on Theme Comprehension’ in Louwerse,
M., and W. van Peer (eds) Thematics. Interdisciplinary Studies.
Amsterdam/Philadelphia: John Benjamins Publishing Company:
19-34.
Griemas A. 1977. in Structuralism and Semiotics (ed.) Hawkes,
Terence. London: Methuen.
Grice, P. 1975. ‘Logic and Conversation’ in Cole, P., and J. Morgan
(eds) Syntax and Semantics Vol 3: Speech Acts. New York:
Academic Press.
__. [1975] 2001. ‘Logic and Conversation’ in Henriksen, Carol (ed.)
Can you reach the salt? Frederiksberg: Roskilde Universitetsforlag.
Grønnum, Nina. [2001] (2004). Revideret kapitel 9 fra Fonetik og
Fonologi – Almen og Dansk. 2. edition. København: Akademisk
Forlag.
Gutenberg Files.
Bibliography 487
Gutt. E.A. 2000. Translation and Relevance – Cognition and Context.

Cambridge, MA: St. Jerome Publishing.
Görlach, Manfred. 1999. English in nineteenth-century England: An
introduction. Cambridge: Cambridge University Press.
Habermas, Jürgen. [1971] 2001. ’Forberedende bemærkninger til en
teori om den kommunikative kompetens’ in Henriksen, Carol (ed.)
Can you reach the salt? Frederiksberg: Roskilde Universitetsforlag.
__. 1981. ’Hvad er universalpragmatik?’ in Teorier om samfund og
sprog. København: Gyldendal.
Haggerty, G. E. 1989. Gothic fiction/Gothic form. University Park:
Pennsylvania State University Press.
Hall, E. T., and M. R. Hall. 1990. Understanding Cultural
Differences. Yarmouth: Intercultural Press Inc.
Hall, Kira. 1995. ‘Lip service on the fantasy lines’ in Hall, Kira and
Mary Bucholtz (eds) Gender articulated: Language and the
socially constructed self. New York: Routledge: 183-216.
Halliday, M.A.K. 1985. An Introduction to Functional Grammar.
London: Edward Arnold.
Halliday, M. 1996. ‘Linguistic function and literary style: An inquiry
into the language of William Golding’s The Inheritors’ in The
Stylistics Reader: From Roman Jakobson to the present (ed.) J. J.
Weber. London: Arnold: 56-86.
Halliday, M. A. K. and R. Hasan. 1976. Cohesion in English. London:
Longman.
Hamilton, C.A. 2004. Preface. Style 38(4).
__. and R. Schneider. 2002. ‘From Iser to Turner and Beyond:
Reception Theory Meets Cognitive Criticism’ in Style 36: 640-
658.
Hanks, P. et al. (eds). 1979. Collins Dictionary of the English
Language. London: Collins.
Hansen, Erik, and Lars Heltoft. 1999. Udkast til Grammatik over det
Danske Sprog. Roskilde: Afdelingen for Dansk og Public Relations
Roskilde Universitetscenter.
Harder, Peter. 1979. ’Tekstpragmatik. En kritisk vurdering af nogle
principielle og praktiske tilgange til tekstbeskrivelsen, med ansatser
til et alternativ’ in Andersen, John E., and Lars Heltoft (ed.)
Nydanske Studier & Almen kommunikationsteori. 10-11. Sprogteori
og Tekstanalyse. København: Akademisk Forlag.
488 Bibliography
Hardy, Barbara. 1964. The Appropriate form: An Essay on the Novel.

London: University of London: Althone Press.
Hardy, Donald E. 2003. Narrating Knowledge in Flannery
O’Connor’s Fiction. Columbia: University of South Carolina
Press.
__. 1996. Functional semantics. Berlin: Mouton de Gruyter.
Hardy, Florence Emily. 1930. The Later Years of Thomas Hardy,
1892-1928. London: Macmillan.
Hardy, Thomas. 1977. Life’s Little Ironies and A Changed Man. The
New Wessex Edition. London: Macmillan.
Harpham, Geoffrey G. 1976. ‘The Grotesque: First Principles’ in
Journal of Aesthetic and Art Criticism 34(4): 461-68.
__. 1982. On the Grotesque: Strategies of Contradiction in Art and
Literature. Princeton: Princeton University Press.
Hartman, G. 1970. ‘The voice of the shuttle: Language from the point
of view of literature’ in Beyond Formalism. Literary Essays 1958-
70. New Haven, CT: Yale University Press: 337-355.
Havránek, Bohuslav. 1964. ‘The Functional Differentiation of the
Standard Language’ in A Prague School Reader on Esthetic,
Literary Structure, and Style (ed. and tr. Paul L. Garvin).
Washington D.C.: Georgetown University Press: 3-16.
Hawkes, Terence. 1977. Structuralism and Semiotics. London:
Methuen.
Hegel, G. W. F. 1975. Aesthetics: Lectures on Fine Art, Vol. I (tr. T.
M. Knox). Oxford: Clarendon Press.
Helsinki Corpus = The Helsinki Corpus of English Texts. 1991.
Helsinki: Department of English, University of Helsinki.
Herman, David. 2002. Story Logic: Problems and Possibilities of
Narrative. Lincoln and London: University of Nebraska Press.
__ (ed.). 2003. Narrative Theory and the Cognitive Sciences. Stanford,
California: Center for the Study of Language and Information.
Herman, Vimala. 1995. Dramatic Discourse: Dialogue as Interaction
in Plays. London: Routledge.
Hetzron, Robert. 1991. ‘On the structure of punchlines’ in Humor
4(1): 61-108.
von Heusinger, Klaus. 1999. Intonation and Information Structure. The
Representation of Focus in Phonology and Semantics.
Habilitationsschrift: Universität Konstanz.
Bibliography 489
Hobbs, J. 1990. Literature and Cognition. Stanford, CA: Center for

the Study of Language and Information.
Hogan, Patrick Colm. 2003. Cognitive Science, Literature and the
Arts. A Guide for Humanists. New York and London: Routledge.
Holmes, D.I., and D. Forsyth. 1995. ‘The Federalist Revisited: New
Directions in Authorship Attribution’ in Literary and Linguistic
Computing vol. 10: 111.
Holmes, J. 1984. ‘Modifying Illocutionary Force’ in Journal of
Pragmatics 8: 345-365.
__. 2006. ‘Workplace narratives, professional identity and relational
practice’ in De Fina, A., D. Schiffrin and M. Bamberg (eds)
Discourse and Identity. Cambridge: Cambridge University Press.
Homan, S. 1993. Pinter’s Odd Man Out: Staging and Filming Old
Times. Lewisburg: Bucknell University Press.
Homer, Sean. 2005. Jacques Lacan. London and New York:
Routledge.
Honigmann, E.A.J. 2001. Othello (3rd Ed). London: Arden
Shakespeare.
Hoover, David. 1999. Language and Style in The Inheritors. Lanham,
Md: University Press of America.
__. 2005. ‘Hot-air Textuality: Literature after Jerome McGann’ in
TEXT Technology: The Journal of Computer Text Processing
14(2): 71-103.
Hoppe-Graff, S. and M. Schell. 1989. ‘The Comprehension of Literary
Texts: Developmental Considerations’ in Meutsch, D., and R.
Viehoff (eds) Comprehension of Literary Discourse: Results and
Problems of Interdisciplinary Approaches. Berlin: de Gruyter: 89-
110.
Hoy, C. 1964. The Hyacinth Room: An Investigation Into the Nature
of Comedy, Tragedy, and Tragicomedy. London: Chatto and
Windus.
Huddleston, Rodney and Geoffrey K. Pullum 2002. The Cambridge
grammar of the English language. Cambridge: Cambridge
University Press.
Hughes, A. 1974. ‘“They Can’t Take That Away From Me”: Myth
and Memory in Pinter’s Old Times’ in Modern Drama 17(4): 467-
76.
490 Bibliography
Hulme, T. E. 1971. Speculations: Essays on Humanism and the

Philosophy of Art (Rpt. ed.) (ed.) Herbert Read. London:
Routledge & Kegan Paul.
Hulse, Clark. 1981. Metamorphic Verse: The Elizabethan Minor Epic.
Hustis, H. 1999. ‘“Reading Encrypted But Persistent”: The Gothic of
Reading and Poe's “The Fall of the House of Usher”’ in Studies in
American Fiction 27 (1): 3-20.
Ionesco, E. 1958. Plays. Volume I: The Lesson, The Chairs, The Bald
Prima Donna, Jacques or Obedience. London: John Calder. (trans.
D. Watson)
__. 1960. Plays, Volume IV: Rhinoceros, The Leader, The Future is in
Eggs or It Takes All Sorts to Make a World. London: John Calder.
(trans. D. Watson)
Ireland, Ken. 2001. The Sequential Dynamics of Narrative: Energies
at the Margins of Fiction. Cranbury, NJ and London: Associated
University Presses.
Irele, Abiola. 1967. ‘Tragic conflict in Achebe’s novels’ in
Introduction to African Literature (ed.) Beier, Ulli. London:
Longman.
Iser, Wolfgang. 1980. The Act of Reading. Baltimore: The Johns
Ishihara, Y. 1989. Haiku-Bunpo Nyumon (An Introduction to Haiku
Grammar). Tokyo: Iizuka Shoten.
Ives, Susan. ‘Lanny Budd is gone, but his war-ravaged world still
remains.’ Homepage of San Antonio, Texas. Retrieved 7 July,
2006, from http://www.mysanantonio.com/opinion/columnists/
sives/stories/MYSA111305.3H.ives.c47f929.html.
Jacks, I. and R. Inglesfield (eds). 1995. The Poetical Works of Robert
Browning Volume 5. Men and Women. Oxford: Clarendon Press.
Jackson, Rosemary. 1981. Fantasy: The Literature of Subversion.
London and New York: Methuen.
Jackson, T. E. 2002. ‘Issues and problems in the Blending of
Cognitive Science, Evolutionary Psychology and Literary Study’ in
Poetics Today 23(1).
Jakobson, Roman. [1958]1996. ‘Closing statement: linguistics and
poetics’ in Weber, J. J. 1996: The Stylistics Reader. London:
Arnold.
Bibliography 491
__. 1960. ‘Closing Statement: Linguistics and Poetics’ in Sebeok,

Thomas A. (ed.) Style in Language. Cambridge, Mass.: MIT Press:
350-77.
__. 1987a. ‘Linguistics and Poetics’ in Language in Literature (ed.)
Krystyna Pomorska and Stephen Rudy. Cambridge, Mass.:
Belknap Press of Harvard University Press: 62-94.
__. 1987b. ‘Two Aspects of Language and Two Types of Aphasic
Disturbances’ in Language in Literature (ed.) Krystyna Pomorska
and Stephen Rudy. Cambridge, Mass.: Belknap Press of Harvard
Jeffares, A. N., and M. Gray (eds). 1995. Collins Dictionary of
Quotations. Glasgow: Harper Collins Publishers.
Jenkins, J. 2006. ‘The spread of EIL: a testing time for testers’ ELT
journal 60(1) Jan 2006. Oxford: Oxford University Press.
__. 2006. World Englishes. Oxford/New York: Routledge.
Jespersen, Otto. 1948. Growth and structure of the English language.
New York: Doubleday.
__. 1949. A modern English grammar on historical principles. Part
VII: Syntax. Copenhagen: Ejnar Munksgaard and London: George
Allen & Unwin.
Johnson, G. C. 2006. ‘The discursive construction of teacher identities
in a research interview’ in De Fina, A., D. Schiffrin and M.
Bamberg (eds) Discourse and Identity. Cambridge: Cambridge
University Press.
Johnstone, B. (ed.). 1994. Repetition in Discourse: Interdisciplinary
perspectives. Princeton, NJ: Ablex Publishing Corporation.
Joseph, J.E. 2004. Language and Identity, national, ethnic and
religious. Basingstoke: Palmer Macmillan.
Jucker, A.H. 1994. ‘Irrelevant repetitions: A challenge to relevance
theory’ in Repetition. SPELL (Swiss Papers in English Language
and Literature) 7. A. Fischer (ed.). Tübingen: Narr: 47-60.
Just for Laughs Series 2, Episodes 1- 6. VHS, Wild Rover Productions
Ltd, 2006.
Just for Laughs Series 4, Episodes 2 & 3. DVD, Wild Rover
Productions Ltd, 2006.
492 Bibliography
Kachru, B. 1986. The Alchemy of English: The Spread, Functions and

Models of Non-native Englishes. Oxford: Pergamon, reprinted
1990 Urbana: University of Illinois Press.
__. 1992. ‘Teaching world Englishes’ in Kachru B. (ed.) 1992.
__. 1992. The Other Tongue. English Across Cultures. (2nd ed)
Urbana: University of Illinois Press.
Karlgen, J., and D. Cutting. 1994. ‘Recognizing text genres with
simple metrics using discriminant analysis’ in Proceedings of
COLING 94, Kyoto: 1071-1075.
Kawin, B. 1972. Telling it again and again: Repetition in literature
and film. New York: Cornell University Press.
Kelchtermans, G. 1993. ‘Teacher and their Career Story: A
Biographical Perspective on Professional Development’ in Day,
Ch, J. Calderhead and P. Denicolo (eds) Research on teacher
thinking, understanding professional development. London,
Washington D.C.: The Falmer Press: 198-220.
Kelih, E., G. Antiü, P. Grzybek and E. Stadlober. 2004. Classification
of Author and/or Genre? The Impact of Word Length. Available at
http://www-gewi.uni-graz.at/quanta/publ/2004_ke_ga_pg-
est_do_author-genre.pdf.
Kempton, Gloria. 2004. Dialogue: Techniques and exercises for
crafting effective dialogue (Write Great Fiction series). Cincinnati:
F & W Publications.
Kiparsky, P. 1987. ‘On Theory and Interpretation’ in Fabb, N. et al.
1987: The linguistics of writing; Arguments between language and
literature. Manchester: Manchester University Press.
Klecka, W.R. 1989. Faktornyj, diskriminantnyj i klasternyj analiz.
[Factor, discriminant and cluster analysis]. Moscow: Finansi i
statistica.
Kneepkens, E. W. E. M., and R. A. Zwaan. 1994. ‘Emotions and
Literary Comprehension’ in Poetics 23: 125-138.
Kristeva, Julia. [1977] 1980. Desire in Language: A Semiotic
Approach to Literature and Art (Rpt. ed.) (ed.) Leon S. Roudiez;
(tr. Thomas Gora et al.). Oxford: Basil Blackwell.
__. [1974] 1984. Revolution in Poetic Language (tr. Margaret Walter).
New York: Columbia University Press.
Kytö, Merja. 1996. ‘“The best and most excellentest way”: The
rivalling forms of adjective comparison in Late Middle and Early
Modern English’ in Svartvik, Jan (ed.) Words. Proceedings of an
Bibliography 493
international symposium, Lund, 25–26 August 1995. (Konferenser

36.). Stockholm: Kungl. Vitterhets Historie och Antikvitets
Akademien: 123–144.
__, and Suzanne Romaine. 1997. ‘Competing forms of adjective
comparison in modern English: “What could be more quicker and
easier and more effective?”’ in Nevalainen, Terttu, and Leena
Kahlas-Tarkka (eds) To explain the present: Studies in the
changing English language in honour of Matti Rissanen. Helsinki:
Société Néophilologique: 329–352.
__, and Suzanne Romaine. 2000. ‘Adjective comparison and
standardisation processes in American and British English from
1620 to the present’ in Wright, Laura (ed.) The development of
standard English, 1300–1800: Theories, descriptions, conflicts.
Cambridge: Cambridge University Press: 171–194.
__, and Suzanne Romaine. 2006. ‘Adjective comparison in 19th-
century English’ in Kytö, Merja, Mats Rydén and Erik Smitterberg
(eds) Nineteenth-century English: Stability and change.
Cambridge: Cambridge University Press: 194-214.
__, Juhani Rudanko and Erik Smitterberg. 2000. ’Building a bridge
between the present and the past: A corpus of 19th-century
English’ in ICAME Journal 24: 85–97.
Lacan, Jacques. 1953. ‘Some Reflections on the Ego’ in International
Journal of Psycho-analysis 34: 11-17.
__. [1966] 1977. Écrits: A Selection (tr. Alan Sheridan). New York
and London: W. W. Norton.
__. [1978] 1988. The Seminar of Jacques Lacan: Book II (tr. Sylvana
Tomaselli). New York: Cambridge University Press.
__. 1981. Les Psychoses (ed.) Jacques-Alain Miller. Paris: Seuil.
Lakoff, G., and M. Johnson 1980. Metaphors We Live By. Chicago:
The University of Chicago Press.
__, and M. Turner 1989. More than Cool Reason: A Field Guide to
Poetic Metaphor. Chicago: The University of Chicago Press.
Lakoff, Robin. 1975. Language and woman’s place. New York:
Harper & Row [out of print].
Lane, N. 1994. Understanding Eugène Ionesco. Columbia: University
of South Carolina Press.
Laplanche, J. and J.-B. Pontalis. [1967] 1973. The Language of
Psycho-Analysis (tr. Donald Nicholson-Smith). New York: W. W.
Norton, 1973.
494 Bibliography
Larsen, Wayne A., Alvin C. Rancher and Tim Layton. 1980. ‘Who
Wrote the Book of Mormon? An Analysis of Wordprints’ in BYU
Studies 20: 225-251.
Larsen-Freeman, D. 1997. ‘Chaos/ complexity science and second
language acquisition’ in Applied Linguistics 18(2): 141-65.
__. 2003. Teaching language: from grammar to grammaring. Boston:
Heinle.
Lásló, J. 1999. Cognition and Representation in Literature. The
Psychology of Literature. Budapest: Akadémiai Kiadó.
Laurence, Margaret. 1968. Long Drums and Cannon. London:
Macmillan & Co Ltd.
Lazar, G. 1993. Literature and Language Teaching. Cambridge:
Leavis, F. R. 1975. The Living Principle: ‘English’ as a Discipline of
Thought. London: Chatto & Windus.
Leech, Geoffrey N. 1969. A Linguistic Guide to English Poetry.
London: Longman.
__. 1983. Principles of Pragmatics. London: Longman.
__. 1992. ‘Pragmatic principles in Shaw’s “You Never Can Tell” in
Toolan, M. (ed.) Language, Text and Context. London: Routledge.
__, and Jonathan Culpeper. 1997. ‘The comparison of adjectives in
recent British English’ in Nevalainen, Terttu and Leena Kahlas-
Tarkka (eds) To explain the present: Studies in the changing
English language in honour of Matti Rissanen. Helsinki: Société
Néophilologique: 353–373.
__, and M. Short. 1981. Style in Fiction. London: Longman.
__, and M. H. Short. 1995. Style in Fiction: A Linguistic Introduction
to English Fictional Prose (ed.) R. Quirk, English Language
Series. London and New York: Longman. Original edition, 1981.
__, and Mick Short. [1981] 2007. Style in Fiction. London: Longman.
Leibniz, G.W. 1969. Philosophical Papers and Letters, 2nd edition,
translated and edited by L.E. Loemker. Dordrecht: D. Reidel
Publishing Company.
Lesser, W. 2002. Nothing Remains the Same: Rereading and
Remembering. New York: Houghton.
Levinson, Stephen C. 2000. Presumptive meanings. Cambridge: The
MIT press.
Levi-Strauss, Claude. 1968. Structural Anthropology. New York:
Doubleday Anchor.
Bibliography 495
Liu, R 1986. The politeness principle in A Dream of Red Mansion’s.

Unpublished M.Phil.dissertation, Lancaster University.
Locher, M.A. 2004. Power and Politeness in Action: Disagreements in
Oral Communication. Berlin and New York: Mouton de Gruyter.
Louw, B. 1993. ‘Irony in the Text or Insincerity in the Writer? The
Diagnostic Potential of Semantic Prosodies’ in Baker, M., G.
Francis and E. Tognini-Bonelli (eds) Text and Technology.
Amsterdam: John Benjamins: 157-176.
__. 1997. ‘The Role of Corpora in Critical Literary Appreciation’ in
Wichman, A., S. Fligelstone, T. McEnery and G. Knowles (eds).
Teaching and Language Corpora. Harlow: Addison Wesley
Longman: 240-251.
Louwerse, M. and W. van Peer (eds). 2002. Thematics.
Interdisciplinary Studies. Amsterdam/Philadelphia: John
Benjamins Publishing Company.
Lucretius De Rerum Natura, Book II: 1-4.
MacKenzie, I. 1987. ‘Narratology and Thematics’ in Modern Fiction
Studies 33(1): 535-544.
Macleod, N. 2005. ‘Stylistics and point of view in fiction: a credo and
some examples’ in The European English Messenger 14(2): 61-73.
Mahlberg, M. 2007. ‘Corpora and Translation Studies: Textual
Functions of Lexis in Bleak House and in a Translation of the
Novel into German’ in Gatto, M., and G. Todisco (eds)
Translation. The State of the Art/La traduzione. Lo stato dell’arte.
Ravenna: Longo.
de Man, Paul. 1986. The Resistance to Theory. Minneapolis:
University of Minnesota Press.
Marie. ‘Booking through Thursday: Rereading.’ A Knit’s Tale.
Weblog. Retrieved 18 July, 2006, from
http://knittale.blogspot.com/2006/06/booking-through-thursday-
with-repeats.html.
Martindale, C., and P. Tuffin. 1996. ‘If Homer is the Poet of the Iliad,
then he may not be the Poet of the Odyssey’ in Literary and
Linguistic Computing 11(3): 109-120.
Mary-Lue. ‘Rereads.’ Life, the universe, and everything. Weblog.
Retrieved 9 July, 2006, from http://www.ltuande.blogspot.com/
2006/06/booking-through-thursday-with-repeats.html.
Maybin, J. and M. Pearce. 2006. ‘Literature and creativity in English’
in S. Goodman and K. O’ Halloran (eds) 2006. The Art of English.
496 Bibliography
Literary Creativity. London: Routledge and Open University Press:

3-24
McAfee, Noëlle. 2004. Julia Kristeva. London and New York:
Routledge.
McArthur, T. 1981. Longman Lexicon of Contemporary English.
London: Longman.
__. 2003. Oxford Guide to World English. Oxford: Oxford University
Press.
McCormick, P. 1985. ‘Feelings and Fictions’ in The Journal of
Aesthetics and Art Criticism 43: 375-383.
McGann, Jerome J. 2001. Radiant Textuality: Literature after the
World Wide Web. New York: Palgrave.
McEnery, Tony, and Scott S. L. Piao. 2003. ‘A tool for text
comparison.’ Conference Proceedings of the Corpus Linguistics
2003 Conference, Lancaster, UK.
McIntyre, Dan. 2006. Point of View in Plays: A Cognitive Stylistic
Approach to Viewpoint in Drama and other Text Types.
McQuade, M. 2002, March 1. ‘Wendy Lesser: A Portrait of the Writer
as a Reader (and Rereader).’ Booklist: 1091.
McRae, J. 1991. Literature With a Small ‘l’. Basingstoke:
MEP/Macmillan.
__. 1992. Wordsplay. Basingstoke: Macmillan.
__, and R. Carter (eds). 1999. Language, Literature and the Learner.
Essex: Pearson Education.
__, and M.E. Vethamani. 2003. Now Read On. London: Routledge.
Meister, J. C. 2002. ‘Parsing for the Theme. A Computer Based
Approach’ in Louwerse, M., and W. van Peer (eds) Thematics.
Interdisciplinary Studies. Amsterdam/Philadelphia: John
Benjamins Publishing Company: 407- 432.
Melchers, G., and P. Shaw. 2003. World Englishes. London: Arnold.
Meutsch, D. and S. J. Schmidt. 1985. ‘On the Role of Conventions in
Understanding Literary Texts’ Poetics 14: 551-574.
Miall, D. S. 1995. ‘Anticipation and Feeling in Literary Response: A
Neurological Perspective’ in Poetics 23: 275-298.
__, and D. Kuiken. 1994. ‘Foregrounding, Defamiliarization, and
Affect: Response to Literary Stories’ in Poetics 22: 389-407.
__, and D. Kuiken. 1999. ‘What is Literariness? Three Components of
Literary Reading’ in Discourse Processes 28: 121-138.
Bibliography 497
__, and D. Kuiken. 2001. ‘Shifting perspectives: Readers’feelings and

literary response’ in Willie van Peer and Seymour Chatman (eds.)
New Perspectives on Narrative Perspective. Albany, NY: State
University of New York Press: 289-301.
Miles, J. [1942] rpt. 1965. Wordsworth and the Vocabulary of
Emotion. New York: Octagon Books.
Millgate, Michael (ed.). 1984. The Life and Work of Thomas Hardy by
Thomas Hardy. London: Macmillan.
Miller, J. Hillis. 1963. The Disappearance of God. Cambridge, MA.:
Belknap/ Harvard University Press
Millis, K. K., and A. King. 2001. ‘Rereading Strategically: The
Influences of Comprehension Ability and A Prior Reading on the
Memory for Expository Text’ in Reading Psychology 22: 41-65.
Minori, Murata. 2000. ‘Identify a Text's Genre by Multivariate
Analysis – Using Selected Conjunctive Words and Particle-
phrases’ in Proceedings of the Institute of Statistical Mathematics
48(2): 311-326.
Morreall, J. 1993. ‘Fear Without Belief’ in The Journal of Philosophy
90(7): 359-366.
Mrs Peel. ‘Review: A Book of Five Rings.’ Skylark of Valeron.
Weblog. Retrieved 7 July, 2006, from http://skylarkofvaleron.
blogspot.com/2006/06/review-book-of-five-rings.html.
Muir, Edwin. 1974. The Structure of the Novel. London: Hogarth
Press.
Mukarovsky, Jan. [1939] 1964. ‘Standard Language and Poetic
Languages’ in Garvin, Paul L. (ed.) A Prague school reader on
esthetics, literary structure and style. Washington: Georgetown
__. [1940] 1971. ‘Om det digteriske sprog’ in Kock, Christian (ed.)
Tjekkisk strukturalisme. København: Munksgaard.
Mulvey, Laura. 1996. ‘Film Narrativity.’ Lecture given at the
University of Cambridge, 12 November 1996.
Munday, J. 1998. ‘Computer-assisted Approach to the Analysis of
Translation Shifts’ in Meta 43.
Mühlhäusler, Peter. 1983. ‘Stinkiepoos, cuddles and related matters’
in Australian Journal of Linguistics 3(1): 75-91.
Müller, Günther. 1968. Morphologische Poetik: Gesammelte Aufsätze.
Darmstadt: Wissenschaftliche Buchgesellschaft.
498 Bibliography
Nash, Walter. 1982. ‘On a Passage from Lawrence’s “Odour of

Chrysanthemums”’ in Carter, Ronald A, (ed.) Language and
Literature. London: Allen & Unwin: 201-20.
__. 1985. The Language of Humour. New York: Longman.
__. 1990. Language in Popular Fiction. London: Routledge.
Neal, N. 1987. ‘Functions of repetition in conversation’ in Text 7:
245-64.
Neill, A. 1991. ‘Fear, Fiction and Make-Believe’ in The Journal of
Aesthetics and Art Criticism 49(1): 47-56.
Nevala, Minna. 2004. ‘Inside and out. Forms of address in seventeenth
and eighteenth century letters’ in Journal of Historical Pragmatics
5(2): 271-296.
Nevalainen, Terttu. 2004. ‘Letter writing’ in Journal of Historical
Pragmatics 5(2): 181-191.
__, and Helena Raumolin-Brunberg (eds). 1996. Sociolinguistics and
language history. Studies based on the Corpus of Early English
Correspondence. Amsterdam: Rodopi.
__, and Helena Raumolin-Brunberg. 2003. Historical sociolinguistics.
Language change in Tudor and Stuart England. London:
Longman.
Nietzsche, F. 1998. Twilight of the Idols. trans. D. Large. Oxford and
New York: Oxford University Press.
The Nineteenth-Century Fiction: Full-Text Database.
Nuttall, A.D. 2001. ‘Browning’s Grammarian: Accents Uncertain?’ in
Essays in Criticism 51(1): 86-100.
Oaks, Dallin D. 1994. ‘Creating structural ambiguities in humor:
getting English grammar to cooperate’ in Humor 7(4): 377-401.
Obuchowski, P. 1975. ‘Unity of effect in Poe’s “The Fall of the House
of Usher”’ in Studies in Short Fiction 12: 407-412.
OED = Oxford English Dictionary. On-line edition.
URL:http://dictionary.oed.com/, accessed November 2006.
Official Just for Laughs/ Juste pour Rire Website
<http://www.hahaha.com/> 5th May 2006.
Official Wild Rover Productions Ltd Website
<http://www.wild-rover.com/framez.htm/> 5th May 2006.
Official Offline Central Website
<http://www.offlinecentral.com/> 5th May 2006.
Orel, Harold (ed.). 1966. Thomas Hardy’s Personal Writings. London:
Macmillan.
Bibliography 499
Ovid. 1955. Metamorphoses (tr. Rolfe Humphries). Bloomington:

Indiana University Press.
The Oxford Dictionary of Quotations (3rd edn). 1979. Oxford: Oxford
University Press.
The Oxford English Dictionary (2nd edn) on CD-ROM Version 3.0.
2002. New York: Oxford University Press.
Padilla Cruz, M. 2001. ‘Learning about the role of repetitions of
phatic sequences: A relevance theoretic approach.’ Paper given at
V International Conference on Languages for Specific Purposes.
E.U. d’Enginyeria Tècnica en Teixits de Punt (Canet de Mar).
University of Barcelona. 6 September, 2001.
Page, Norman. 1988. Speech in the English Novel. 2nd ed. London:
Macmillan. (1st ed 1973).
Pajares, M. F. 1992. ‘Teachers’Beliefs and Educational Research:
cleaning Up a Messy Construct’ in Review of Educational
Research Fall 1992 62(3): 307-332.
Palmer, Eustace. 1972. Introduction to the African Novel London:
Heinemann.
Pavel, Thomas G. 1985. ‘Literary narratives’ in Discourse and
Literature (ed.) Teun van A. Dijk. Amsterdam: John Benjamins
Publishing Co.
__. 1986. Fictional Worlds. Cambridge: Harvard University Press.
Payne, Michael. 1993. Reading Theory: An Introduction to Lacan,
Derrida, and Kristeva. Oxford: Blackwell.
Peeples, S. 2002. ‘Poe’s “constructiveness” and “The Fall of the
House of Usher”’ in The Cambridge Companion to Edgar Allan
Poe (ed.) K. J. Hayes. Cambridge: Cambridge University Press:
178-190.
van Peer, W. 1986. Stylistics and Psychology: Investigations of
Foregrounding. London: Croom Helm.
__. 1992. ‘Literary Theory and Reader Response’ in Nardocchio, E. F.
(ed.) Reader Response to Literature. The Empirical Dimension.
Berlin/New York: de Gruyter: 137-152.
__. 2002. ‘Where Do Literary Themes Come from?’ in Louwerse, M.,
and W. van Peer (eds) Thematics. Interdisciplinary Studies.
Amsterdam/Philadelphia: John Benjamins Publishing Company:
253-263.
__ (ed.). Forthcoming. The Quality of Literature: Studies in Literary
Evaluation. Amsterdam: John Benjamins.
500 Bibliography
Persson, G. 1974. Repetition in English: Part 1. Uppsala: Uppsala

University Press.
Piao, Scott S.L. 2001. ‘Detecting and measuring text reuse via
aligning texts’ Technical paper: CS-01-15. University of Sheffield,
UK.
Pilkington. A. 2000. Poetic Effects – A Relevance Theory Perspective.
Pinter, H. 1971. Old Times. London: Methuen.
Poe, E. A. 1986. The Fall of the House of Usher and Other Writings.
London: Penguin.
Porter, M.F. 1980. ‘An algorithm for suffix stripping’ in Program
14(3): 130í137.
Prince, G. 1982. Narratology: The Form and Function of Narrative.
The Hague: Mouton.
Propp, V. 1968. The Morphology of the Folktale. Austin: University
of Texas Press.
Quintilian. 1921. The Institutio Oratoria of Quintilian, Vol. IV (trans.
H. E. Butler). Cambridge, Mass.: Harvard University Press.
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan
Svartvik. 1972. A grammar of contemporary English. London and
New York: Longman.
__, S. Greenbaum, G. Leech and J. Svartvik. 1985. A comprehensive
grammar of the English language. London and New York:
Longman.
__, S. Greenbaum, G. Leech and J. Svartvik. 1992. A Comprehensive
Grammar of the English Language. London in New York:
Longman.
Ragland, Ellie. 1993. ‘Lacan, the Death Drive, and the Dream of the
Burning Child’ in Death and Representation (ed.) Sarah Goodwin
and Elisabeth Bronfen. Baltimore and London: Johns Hopkins
Raumolin-Brunberg, Helena. 1996. ‘Forms of address in early English
correspondence’ in Nevalainen, Terttu and Helena Raumolin-
Brunberg (eds) Sociolinguistics and language history. Studies
based on the Corpus of Early English Correspondence.
Amsterdam: Rodopi: 167-183.
Ray, Martin. 1997. Thomas Hardy: A Textual Study of the Short
Stories. Aldershot: Ashgate.
‘Rereading.’ Weblog. Retrieved 7 July, 2006, from
Bibliography 501
http://kyuuketsukirui.livejournal.com/427066.html.
Richards, Ivor Armstrong. 1928. Principles of Literary Criticism.
London: Routledge.
__. 1936. The Philosophy of Rhetoric. Oxford: Oxford University
Press.
__. [1929] 2001. Practical Criticism: A Study of Literary Judgment.
London: Routledge.
Richardson, A. 1998. ‘Brains, Minds, and Texts’ Review 20: 39–48.
Ricœur, Paul. [1975] 1977. ‘The Work of Resemblance’ in The Rule
of Metaphor: Multi-disciplinary Studies of the Creation of
Meaning in Language (tr. Robert Czerny et al.). Toronto:
University of Toronto Press: 173-215.
__. [1983] 1984. Time and narrative. Vol 1. Chicago: The University of
Chicago Press.
Rimmon-Kenan, Shlomith. 1983. Narrative Fiction. London:
Methuen.
__. 1983. Narrative Fiction, Contemporary Poetics. New Accents,
(gen. ed.) Hawkes, T. London and New York: Routledge.
Rogers, Robert. 1978. Metaphor: A Psychoanalytic View. Berkeley
and Los Angeles: University of California Press.
Roget’s II The New Thesaurus. 1980. Boston: Houghton Mifflin
Company.
Rountree, T. J. 1972. ‘Poe’s Universe: The House of Usher and the
Narrator’ in Tulane Studies in English 20 (4): 123-134.
Roy, A. 1995. The God of Small Things. London: Flamingo.
Rudanko, J. 2005. ‘Aggravated impoliteness and two types of speaker
intention in an episode in Shakespeare’s Timon of Athens’ in
Journal of Pragmatics. Article in press.
Rusiecki, Jan. 1985. Adjectives and comparison in English: A
semantic study. London: Longman.
Ryan, M.L. 1985. ‘The modal structure of narrative universes’ in
Poetics Today 6(4): 717-755.
__. 1991. Possible Worlds, Artificial Intelligence and Narrative
Theory. Bloomington: Indiana University Press.
Safire, William. 2006. ‘A change of address: the demise of “dear”’ in
International Herald Tribune, October 23, p. 7.
Said, E. 1978. Orientalism. New York: Pantheon.
Sample, K. J. 2005. ‘Promoting Fluency in Adolescents with Reading
Difficulties’ in Intervention in School and Clinic 40(4): 243-246.
502 Bibliography
Sanford, Anthony J. 1990. ‘On the nature of text-driven inference’ in

Balota, D.A., G.B. Flores d´Arcais and K. Rayner (ed.)
Comprehension processes in reading. Hilsdale, New Jersey:
Lawrence Earlbaum.
de Saussure, Ferdinand. [1955] 1959. Course in General Linguistics
(ed.) Charles Bally and Albert Sechehaye; (tr. Wade Baskin). New
York: Philosophical Library.
Schank, R. 1979. ‘Interestingness: Controlling Inferences’ in Artificial
Intelligence 12(3): 273-297.
__. 1982. Dynamic Memory. A Theory of Riminding in Computers
and People. Cambridge: Cambridge University Press.
Schroeder, Al. ‘Everything you wanted to know about Mindmistress
(*but were afraid to ask).’ Mindmistress. Online graphic novel.
Retrieved 15 July, 2006 from
http://mindmistress.comicgenesis.com/mmknow.htm.
Scott, M. 2006. ‘Key Words of Individual Texts’ in Scott, M., and C.
Tribble Textual Patterns. Key Words and Corpus Analysis in
Language Education. Amsterdam: Benjamins: 55-72.
Searle, John. [1979] 1996. Expression and meaning. Cambridge:
Seifert, C. M., R. P. Abelson and G. McKoon. 1986. ‘The Role of
Thematic Knowledge Structures in Reminding’ in Galambos, J. A.,
R. P. Abelson and J. B. Black (eds) Knowledge Structures.
London: LEA: 185-210.
Semino, Elena. 1990. Towards a Theory of Poetic Titles. London:
Unpublished MA dissertation.
__. 1997. Language and World Creation in Poems and Other Texts.
London: Longman.
__. 2002. ‘A Cognitive-Stylistic Approach to Mind Style in Narrative
Fiction’ in Semino, Elena, and Jonathan Culpeper (eds) Cognitive
Stylistics. Language and Cognition in Text Analysis. Amsterdam:
John Benjamins: 95-122.
__. 2006. ‘Blending and Characters’ Mental Functioning in Virginia
Woolf’s “Lappin and Lapinova”’ in Language and Literature
15(1): 55-72.
__, and Jonathan Culpeper (eds). 2002. Cognitive Stylistics. Language
and Cognition in Text Analysis. Amsterdam: John Benjamins.
Bibliography 503
__, and M. Short. 2004. Corpus Stylistics. Speech, Writing and

Thought Presentation in a Corpus of English Writing. London:
Routledge.
__, Mick Short and Jonathan Culpeper. 1997. ‘Using a corpus to test a
model of speech and thought presentation’ in Poetics 25: 17-43.
__, and K. Swindlehurst. 1996. ‘Metaphor and mind style in Ken
Kesey’s One Flew Over the Cuckoo’s Nest’ in Style 30 (1): 143-
167.http://search.epnet.com/login.aspx?direct=true&db=aph&an=9
608291448. 2/2/2006.
Shakespeare, W. 1623. King Lear, ed. by M. Best.
Shakespeare, W. 1970. A Complete and Systematic Concordance to
the Works of Shakespeare. Vol. 6. M. Spevack (ed.). Hildesheim:
Georg Olms.
__. 1997. The Riverside Shakespeare (2nd edn). G. B. Evans (ed.).
Boston: Houghton Mifflin Company.
__. 1997. ‘Sonnet 2’ in The Art of Shakespeare’s Sonnets (Helen
Vendler). Cambridge, Mass.: Belknap Press of Harvard University
Press: 52.
Sharvit, Yael, and Penka Stateva. 2002. ‘Superlative expressions,
context and focus.’ in Linguistics and Philosophy 25: 433-504.
Shelley, Percy Bysshe. 2002. A Defence of Poetry in Shelley’s Poetry
and Prose (2nd ed.) (ed.) Donald H. Reiman and Neil Fraistat. New
York: W. W. Norton: 510-35.
Short, Mick. 1973. ‘Some Thoughts on Foregrounding and
Interpretation’ in Language and Style 6(2): 97-108.
__. (ed.). 1989. Reading, Analysing and Teaching Literature. London:
Longman.
__. 1992. ‘Mind-Style’ in The European English Messenger 1 (3): 31-
33.
__. 1994. ‘Style: Definitions’ in Asher, R. (ed.) The Encyclopaedia of
Language and Linguistics. Amsterdam: Pergamon Press: 4375-
4378.
__. 1996. Exploring the Language of Poems, Plays and Prose.
London: Longman.
__. 1999. ‘Stylistics “upside down”: using stylistic analysis in the
teaching of language and literature’ in Carter, R., and J. McRae
504 Bibliography
(eds) Language, Literature and the Learner. Essex: Pearson

Education: 41-64.
__. 2002. ‘Who Is Stylistics and What Use is She to Students of
English Language and Literature?’ in Poetica 58: 33-54.
__. 2006. ‘Designing and piloting a world-wide-web-based stylistics
course’ in Andrea Gerbig and Anja Müller-Wood (eds) Rethinking
English: Reconciling Literature, Linguistics and Cultural Studies
(Lampeter: Edwin Mellen). (Avaliable at http://www.lanc. ac.uk/
fass/projects/stylistics/tutors/Paperversions/ Trier paper.doc).
__. 2007. ‘Designing and Piloting a World-wide-web-based Stylistics
Course’ in Andrea Gerbig and Anja Müller-Wood (eds.)
Rethinking English: Reconciling Literature, Linguistics and
Cultural Studies. Lampeter: Edwin Mellen: 131-66.
__, Beatrix Busse and Patricia Plummer (eds). 2006. Language and
Literature special issue: The Language and Style Pedagogical
Investigations, 15.3.
__, and Willie van Peer. 1989. ‘Accident: Stylisticians Evaluate’ in
Short, Mick (ed.) Reading, Analysing and Teaching Literature.
London: Longman: 22-71.
__, and Elena Semino. Forthcoming. ‘Stylistics and Literary
Evaluation’ in van Peer, Willie (ed.).
Shroder, Maurice Z. 1967. ‘The Novel as a Genre’ in The Theory of
the Novel (ed.) Stevick, P. New York: The Free Press.
Simpson, Paul. 1989. ‘Politeness Phenomenon in Ionesco’s The
Lesson’ in Carter, R., and P. Simpson (eds) Language, Discourse
and Literature: An Introductory Reader in Discourse Stylistics.
London: Unwin Hyman: 171-193.
__. 1993. Language, ideology and point of view. London and New
York: Routledge.
__. 1997. Language through Literature: An Introduction. London:
Routledge.
__. 2003. Stylistics: A Resource Book for Students. London:
Routledge.
__. 2004. On the Discourse of Satire: Towards a Stylistic Model of
Satirical Humour. Amsterdam and Philadelphia: John Benjamins.
Sinclair, J. McH and R.M. Coulthard. 1975. Towards an Analysis of
Discourse: The English Used by Teachers and Pupils. London:
Bibliography 505
Singh, Raman K. 1980/1981. ‘An Encounter with John Fowles’ in

Journal of Modern Literature 8(2): 181-202.
Skulsky, H. 1980. ‘On Being Moved by Fiction’ in The Journal of
Aesthetics and Art Criticism 39: 5-14.
Sloboda, Nicholas. 1997. ‘Heteroglossia and Collage: Donald
Barthelme's “Snow White”’ in Mosaic 30(4).
Smith, Anne-Marie. 1998. Julia Kristeva: Speaking the Unspeakable.
London: Pluto Press.
Spears, R. A. 1998. NTC’S American English Learner’s Dictionary.
Lincolnwood (Chicago): NTC Publishing Group.
Spencer, C. 1965. Five Restoration Adaptations of Shakespeare.
Urbana: University of Illinois Press.
Sperber, D., and D. Wilson. 1995. Relevance: Communication and
Cognition. 2nd Edition. Oxford: Blackwell.
__, and D. Wilson. 1997. Relevance, communication and cognition,
second edition. Oxford: Blackwell Publishers, Ltd.
Spivak, G. 1988. ‘Can the Subaltern Speak?’ in Marxism and the
Interpretation of Culture (eds) Nelson, C., and L. Grossberg.
London: Macmillan.
Stadlober, E., and M. Djuzelic. 2004. Multivariate Statistical Methods
in Quantitative Text Analyses. Available at:
http://www.stat.tugraz.at/stadl/papers/stdj05.pdf.
Stamatatos, E., N. Fakatakis and G. Kokkinakis. 2001. ‘Computer-
Based Authorship Attribution Without Lexical Measures’ in
Computers and the Humanities 35: 193-214.
Stanbrough, Harvey. 2004. Writing Realistic Dialogue and Flash
Fiction. Albuquerque: Central Avenue Press.
Steen, Gerard. 1994. Understanding Metaphor in Literature: An
Empirical Approach. London: Longman.
Stevick, Philip. 1967. The Theory of the Novel. New York: The Free
Press (First published by Collier Macmillan Ltd, London).
Stine-Morrow, E. A. L., D. D. Gagne, D. G. Morrow and B. H.
Dewall. 2004. ‘Age Differences in Rereading’ in Memory &
Cognition 32(5): 696-710.
Stockwell, Peter. 2002. Cognitive Poetics. An Introduction. London:
Routledge.
__. 2003. ‘Surreal Figures’ in Cognitive Poetics in Practice (ed.)
Joanna Gavins and Gerard Steen. London and New York:
Routledge: 13-25.
506 Bibliography
Stubbs, M. 2005. ‘Conrad in the Computer: Examples of Quantitative

Stylistics Analysis’ in Language and Literature 14(1): 5-24.
Suleiman, S., and I. Crosman (eds). 1980. The Reader in the Text:
Essays on Audience and Interpretation. Princeton, New Jersey:
Princeton University Press.
Svaglic, M.J. 1967. ‘Browning’s Grammarian: Apparent failure or
real?’ in Victorian Poetry 5: 93-103.
Swales, Martin. 1977. The German ‘Novelle’. Princeton NJ: Princeton
University Press.
Taavitsainen, Irma, and Andreas H. Jucker (eds). 2003. Diachronic
perspectives in address term systems. Amsterdam: John
Benjamins.
Tabur-Jogi, Helena. 2004. Stereotypical Self-Images of Native
Americans in the Novel Reservation Blues and the Short Story
Collections The Lone Ranger and Tonto Fistfight in Heaven and
The Toughest Indian in the World by Sherman Alexie. Unpublished
Masters Thesis, University of Tartu.
Tan, E. S. 1994. ‘Story Processing as an Emotion Episode’ in van
Oostendorp, H., and R. A. Zwaan (eds) Naturalistic Text
Comprehension. Norwood, New Jersey: Ablex Publishing
Corporation: 165-185.
Tanaka, N. 1993. The pragmatics of uncertainty: its realisation and
interpretation in English and Japanese. Unpublished PhD Thesis,
Lancaster University.
Tannen, Deborah. 1990. ‘Silence as conflict management in fiction
and drama: Pinter’s Betrayal and a short story, “Great Wits”’ in
Grimshaw, Allen (ed.) Conflict talk: Sociolinguistic investigations
of arguments in conversations. Cambridge: Cambridge University
Press.
Tate, N. 1681. King Lear, ed. by C. Spencer 1965 Five Restoration
Adaptations of Shakespeare. Urbana: University of Illinois Press:
201-274.
Tate, N. 1681. King Lear, ed. by J. Lynch.
http://andromeda.rutgers.edu/~jlynch/Texts/tatelear.html
Television Production Handbook: Editing and Program Continuity
<http://www.tvhandbook.com/Editing%20and%20Program%20Co
ntinuity.html> 16 October 2006
Thomas, D. 1990. The Post Romantics (ed.) D. T. Thomas. London:
Routledge.
Bibliography 507
Thomas, J. 1995. Meaning In Interaction. London: Longman.

Thomson, Philip. 1972. The Grotesque. London: Methuen.
Thoreau, H. D. ‘Walden’ The Writings of Henry David Thoreau,
(Shanley, J.L. edited 1971). Princeton, NJ: Princeton University
Press.
Tieken-Boon van Ostade, Ingrid. 1999. ‘Of formulas and friends:
Expressions of politeness in John Gay’s letters’ in Tops, Guy A.J.,
Betty Devriendt and Steven Geukens (eds) Thinking English
grammar. To honour Xavier Dekeyser, Professor Emeritus.
Leuven: Peeters: 99-112.
Tilghman, B. R. 1984. But is it art?; the value of art and the
temptation theory. Oxford: Blackwell.
Todorov, Tzvetan. 1969. Grammaire du Décaméron. The Hague:
Mouton.
__. 1977. The Poetics of Prose. Oxford: Blackwell.
__. 1981. Introduction to Poetics. Brighton: Harvester.
__. 1984. Mikhail Bakhtin. The Dialogical Principle. trsl. W. Godzich
Manchester: Manchester University Press.
Togeby, Ole. 1993. PRAXT – pragmatisk tekstteori. Århus: Aarhus
Universitetsforlag.
Toolan, Michael. 1985. ‘Analyzing fictional dialogue’ in Language
and Communication 5(3): 193-206.
__. 1990. The Stylistics of Fiction: A Literary-linguistic Approach.
London and New York: Arnold.
__. 1992. Narrative: A Critical Linguistic Introduction. London:
Routledge. (First published in 1988).
__. 1996. ‘Stylistics and its discontents; or, getting off the Fish
“hook”’ in Weber, J. J. 1996: The Stylistics Reader. London:
Arnold.
__. 1997. Language in Literature: An Introduction to Stylistics.
London: Routledge.
__. 1998. ‘The give and take of talk, and Caryl Churchill’s Cloud
Nine’ in Culpeper, J., M. Short and P. Verdonk (eds) Exploring the
Language of Drama: From Text to Context. London: Routledge:
142-160.
__. 2000. ‘“What makes you think you exist?”: A speech move
schematic and its application to Pinter’s The Birthday Party’ in
Journal of Pragmatics 32: 177-201.
508 Bibliography
__. [1988] 2001. Narrative: A Critical Linguistic Introduction.

London: Routledge.
Tribble, C. 2006. ‘Counting Things in Texts You Can’t Count on’ in
Scott, M., and C. Tribble Textual Patterns. Key Words and Corpus
Analysis in Language Education. Amsterdam: Benjamins: 179-
193.
Tsur, Reuven. 1992. Toward a Theory of Cognitive Poetics.
Amsterdam: Elsevier Science Publisher.
__. 2002. ‘Some Cognitive Foundations of “Cultural Programs”’ in
Poetics Today 23(1): 63- 89.
__. 2002. ‘Aspects of Cognitive Poetics’ in Cognitive Stylistics:
Language and Cognition in Text Analysis (ed.) Elena Semino and
Jonathan Culpeper. Amsterdam/Philadelphia: John Benjamins
Publishing Co.: 279-318.
Tucker, H.F. 1980. Browning’s Beginnings. The art of disclosure.
Minneapolis: University of Minnesota Press.
Turco, Lewis. 1991. Dialogue (Elements of Fiction Writing): A
Socratic dialogue on the art of writing dialogue in fiction. London:
Robinson Publishing. (originally pub. 1989. Cincinnati: Writer’s
digest books).
__. 2004. The Book of Dialogue. How to write effective conversation
in fiction, screenplays, drama and poetry. Lebanon: University
Press of New England.
Turner, M. 1994. ‘Design for a Theory of Meaning’ in Overton, Willis
F., and David S. Palermo (eds) The Nature and Ontogenesis of
Meaning. Hillsdale, New York: Lawrence Erlbaum Associates: 91-
107.
Turner 2000. On The Origins Of Human Emotion. Stanford, USA;
Stanford University Press.
Turner, M. 2002. ‘The Cognitive Study of Art, Language, and
Literature’. Poetics Today 23(1): 9-20.
Uspenski, B. 1979. Poetika kompozicije. [Trans. N. Petkoviü].
Beograd: Nolit.
Varghese, M et al. 2005. ‘Theorizing Language Teacher Identity:
Three Perspectives and Beyond’ in Journal of Language, Identity
and Education 4(1): 21-44.
Veloudis, Ioannis. 1998. ‘“Quantifying superlatives” and homo
sapiens’ in Journal of Semantics 15(3): 215-237.
Verdonk, P. 2002. Stylistics. Oxford: Oxford University Press.
Bibliography 509
Voloshin, B. R. 1986. ‘Explanation in “The Fall of the House of

Usher”’ in Studies in Short Fiction 23 (4): 419-428.
Wagner, I., and R. Wodak. 2006. ‘Performing success: identifying
strategies of self-presentation in women’s biographical narratives’
in Discourse and Society. London: Thousand Oaks, Sage.
Wales, K. 2001. A Dictionary of Stylistics. 2nd edition, Harlow:
Longman.
Walton, K. 1978. ‘Fearing Fictions’ in The Journal of Philosophy
75(1): 5-27.
Wartenberg, T.E. 1989. The Forms of Power. From Domination to
Transformation. Philadelphia: Temple University Press.
Watanabe, T., E. R Skrzypczak. and P. Snowden. 2003. Kenkyusha’s
New Japanese-English Dictionary (5th edn). Tokyo: Kenkyusha.
Watson, Greg and Sonia Zyngier (eds). 2007. Literature and Stylistics
for Language Learners. London: Palgrave MacMillan.
Watt, Ian. 1957. The Rise of the Novel, Harmondsworth: Penguin
(First published by Chatto and Windus).
Watts, R.J. 2003. Politeness. Cambridge: Cambridge University Press.
Weber, J. J. 1996. ‘Towards contextualized stylistics: An overview’ in
Weber, J. J. 1996: The Stylistics Reader. London: Arnold.
Wellek, R. and Austin W. 1966. Theory of literature. (3rd rev.
ed.) London: Cape.
Wellthen13. ‘Currently reading.’ Weblog. Retrieved 15 July, 2006,
from http://www.xanga.com/wellthen13.
Werth, Paul. 1999. Text Worlds: Representing Conceptual Space in
Discourse. London: Longman.
White, Allon. 1993. ‘L’Éclatement du sujet: The Work of Julia
Kristeva’ in Carnival, Hysteria, and Writing: Collected Essays and
Autobiography. Oxford: Clarendon Press: 62-87.
‘Who rereads books?’ Weblog (archive). Retrieved 7 July, 2006, from
http://forums.ogresnet.com/archive/index.php/t-2054.html.
Widdowson, Henry G. 1975. Stylistics and the Teaching of Literature.
London: Longman.
__. 1992. Practical Stylistics: An Approach to Poetry. Oxford: Oxford
University Press.
__. 1995. ‘Discourse Analysis: A Critical View’ in Language and
510 Bibliography
__. 1996. ‘Reply to Fairclough: Discourse and Interpretation:

Conjectures and Refutations’ in Language and Literature 5(1): 57-
69.
Wilkinson, J. 2000. ‘The Paradox(es) of Pitying and Fearing Fictions’
in South African Journal of Philosophy 19(1): 62-74.
Williams, R. Grant. 2001. ‘Disfiguring the Body of Knowledge:
Anatomical Discourse and Robert Burton's The Anatomy of
Melancholy’ in ELH 68(3): 591-613.
Wilson, D., and D. Sperber. 1990. ‘Rhetoric and Relevance’ in The
Ends of Rhetoric History, Theory, Practice. Stanford: Stanford
Wilson, D., and D. Sperber. 2000. ‘Truthfulness and Relevance’ in
UCL Working Papers in Linguistics 12: 215-254.
__, and D. Sperber. 2002. ‘Relevance Theory’ in UCL Working
Papers in Linguistics 14: 249-287.
__, and D. Sperber. 2004. ‘Relevance theory’ in Handbook of
Pragmatics. G. Ward and L. Horn. (eds). Oxford: Blackwell: 607-
632.
Wittgenstein, L. 2001. Philosophical investigations: the German text,
with a revised English translation (3rd ed.). Oxford: Blackwell.
Woods, D. 1996. Teacher Cognition in Language Teaching. Beliefs,
decision – making and classroom practice. Cambridge: Cambridge
University Press .
Wordsworth, W. 1952. The Poetical Works of William Wordsworth
(2nd edn). Vol. 2. E. de Selincourt (ed.) Oxford: Clarendon Press.
__. 1959. William Wordsworth: The Prelude or Growth of a Poet’s
Mind (2nd edn). E. de Selincourt (ed.) and H. Darbishire (rev).
Oxford: Clarendon Press.
Wynne, M. (ed.). 2005. Developing Linguistic Corpora: a Guide to
Good Practice. Oxford: Oxbow Books.
Zabrucky, K. M., and D. Moore. 1999. ‘Influence of Text Genre on
Adults’ Monitoring of Understanding and Recall’ in Educational
Gerontology 25: 691-710.
Zanettin, F. 2001. Ipergrimus, intralinea, ipermedia (online)
http://www.intralinea.it/hypermedia/eng_open.php
Zwaan, R. A., G. A. Radvansky and S. N. Whitten 2002. ‘Situation
Models and Themes’ in Louwerse, M., and W. van Peer (eds)
Thematics. Interdisciplinary Studies Amsterdam/Philadelphia:
John Benjamins Publishing Company: 35-53.
Bibliography 511
Zurbrugg, Nicholas. 1993. The Parameters of Postmodernism.

Carbondale: Southern Illinois University Press.
Index
A C
accessibility relations 155, candid camera 395
158,160,163, 173, 174 classification ix, 110, 156, 177,
address forms 231, 243, 253, 179, 184-5, 191, 210-11, 238,
256, 260 259, 307, 309-10, 415, 418
adjective comparison xiii, 229, cognitive mental operations
233, 240, 243, 259 137
adjectives 201, 229, 234-5, cognitive poetics/cognitive
239-43, 253, 255, 259-60 semiotics 119
aesthetic judgment 1, 10, 330 cognitive pragmatics xviii, 45
affective response 1, 11, 14, comparative adjectives 229,
137, 145, 363, 367 259
agency 50, 128, 210, 223, 333, CONCE (Corpus of
336, 337, 340 Nineteenth-Century English)
Alexie, Sherman xx, 303, 308, xx, xxii, 229, 231-7, 240-1,
312-5, 324-5, 329-30 247, 249, 251, 253-4, 258-60
alternative worlds 155 confrontation 427, 451
appreciation xviii, 1, 9-12, 14, conversation analysis 303-4,
16, 25-6, 75, 351, 355, 358 308
authentication 155, 158-61 corpora xii, xiii, 191, 193-4,
204, 209, 229, 231, 234
B Corpus Linguistics xi, xxi, 6,
Barthelme, Donald xix, 119- 191-2, 204
20, 123, 125-6, 129, 133-5 corpus stylistics xiii, xvii, xix,
Barthes, Roland xix, 91, 97-9, 77, 207
104, 107, 112, 114, 376 Crouch 207, 209-10, 214-17,
basic emotions 363, 367-8 219-25
blending theory 119, 135
BNC (British National Corpus) D
229, 234, 259 deviation 85, 140, 142-5, 147-
Brandt, Line 119, 133, 135 9, 151, 188, 265, 270-1, 413
Brandt, Per Aage 119, 121, dialogue xiv, xx, xxi, 173, 175,
124, 126-7, 133, 135 217, 226-7, 303-4, 306-10,
Brown and Levinson 427, 429- 312, 315-6, 318-24, 326-30,
30, 432-4, 436-7, 439, 446-7, 335, 428, 434, 444
449 diminutives 229
Browning’s style 31-2, 34
514 Index
direct definition 376, 380, 386, filmic xxi, 395, 398-9, 402,
388 407-8, 410
discoursal identity formation Fowles, John xi, xix, 207-8,
333, 336 215-7, 226-7
discourse move analysis 303, the Freudian dream-work 91-3,
308, 328, 330 100, 103, 106
discourse stylistics 31, 42, 376
dramatic monologue 31, 33, 36 G
gag 395-402, 404-9
E gender differences 229, 240-1
educational research 363 genre xiv, xix, 32-3, 48, 55,
elaboration xx, 33, 129, 230, 58-9, 62-3, 70, 78, 121, 155,
303, 307, 375, 382-3 157-8, 175, 178-9, 191-3,
ellipsis ix, xxi, 35, 67, 318, 195, 229-32, 234-5, 240, 249,
347-9, 352-5, 357-9, 361-2 253, 266, 271, 277, 295, 300,
emergent grammar 31 313, 375-6, 428, 448
emotional response to Gothic effects 451, 470
literature 364 ‘A Grammarian's Funeral’ 31,
emotions 32, 111, 137, 145, 33-4, 36, 41
148-50, 249, 254, 285, 293, grotesque ix, xix, 50, 91-107,
300, 363, 367-8, 417 110-3, 409
emotion study xxi, 363
Emotion Tracking Pedagogy H
363-5, 367, 373 Hardy, Thomas xviii, 61-6, 69-
enhancement 375, 382, 383 71, 73, 248
enunciation & evidentiality humour xiv, xxi, 135, 168,
119 334, 377-8, 395-400, 402-8,
epistolary formulae 229, 231, 448
235, 243
evaluation 1, 9-12, 26-7, 76, I
82, 127, 139, 213, 215, 255 indirect presentation 375, 380,
386
F interdisciplinarity xviii, 45, 50-
face attack xxi, 427, 429, 433, 3
435, 440-2, 444-5, 448
face relations 229-30 K
‘The Fall of the House of Key Stage 3: Year 8 363
Usher’ 452, 455 King Lear ix, xix, 191-2, 195-
6, 414
Index 515
the Kristevan semiotic 91, 93 narratology 2, 61, 66, 68, 71,

76, 376
L National Curriculum 363-5,
the Lacanian real 91, 93, 100 373
language shifts 191 n-gram overlap 2, 61, 66, 68,
lettersxix, 143, 187, 229-37, 71, 76, 376
239-51, 253-5, 257, 259, 260 Nineteenth (19th) century
literariness 58, 137-8, 140, English 229, 234
143, 145, 149-51, 204, 285 nonverbal 395-400, 404-8, 410
literature as a Novelle 61-4, 66, 70
cognitive/psychological
object 45 P
literary interpretation 46, 48-9, partially impossible worlds
58, 133-5, 138, 148, 279, 155
283, 453 pedagogy xxi, 363, 365, 367
literary pragmatics 265 physical traits 375, 380-1, 386,
388
M ‘pleasure’ 413, 417, 421, 423-
The Magus xi, xix, 207-9, 215- 5
6, 220-1, 225, 227 pre-theory 45, 55-6
metaphor 5, 7, 12, 42, 49, 61, Poe, Edgar Allan 451, 455-6,
65, 85, 91-3, 100, 102-14, 458-9, 462-4, 466, 468
350, 359-61, 424, 453-4 poetic effects 347-51, 353,
metamorphosis 91-2, 107, 112, 359, 361
115, 141-2, 145, 166, 439 point of view 333, 335-7, 339,
metonymy 91, 100-3, 105, 343
108, 111, 114 politeness 229, 231, 244, 246,
mind-style xxi, 451-5, 459, 427-34, 436-7, 441, 443-6,
462, 465, 470 448
modality xxi, 451-62, 464-5, possible-worlds theory xix,
467, 469-70 155-7, 163, 173-5
motifs 375, 380-1, 384-6, 388- post colonial literature xi, 363,
9, 391 367
multivariate discriminant poststructuralist stylistics 31,
analysis 177-9 35
pragmatics ix, xi, xiv, xv, xvii,
N xx, 6, 26, 45, 52-3, 55, 56-9,
narrative diegesis xix, 119, 127 76, 265-6, 275, 283, 304,
348, 376
516 Index
professional identity xv, 333-4 stylistic analysis x, xiii, xiv,

psychological traits 375, 381, xx, 10, 13, 41, 75-85, 87-8,
386 207, 303, 305-6, 313-4, 333,
335-6, 375-6, 379, 389
R stylistic foregrounding 61,
reader response xiv, xix, 8, 63, 140, 145-6, 148
82, 63, 82, 137, 144 stylistics and interpretation 31
readings 1, 11-2, 15-6, 18, 23, stylistics course xiv, 77-8, 85
36-7, 61, 119, 134, 192, 204, subjectification 229, 231, 256,
288-90, 292-3, 295-6 258
reinforcement 375, 383-4, 388 superlative adjectives 229,
relevance theory xxi, 56, 59, 234, 239, 243, 259
283-4, 286-8, 290, 292, 297,
347-51, 353, 356, 360 T
repetition x, xx, 34, 40, 67, Tate, Nahum ix-x, xix, 191-8,
146, 194, 196, 283-91, 295-7, 200-1, 203
300, 339, 342, 359, 375, 383- teaching stylistics 75
4, 388, 443 temporality xi, 61, 123, 130-1
rereading xx, 283-97, 300-1 TESAS 207-11, 213-7, 219-25
revision xxi, 65, 207-9, 215-7, text reuse measurement 207
220-1, 224-6 text similarity 207, 209, 212,
214-5, 221-2, 226
S text understanding 5, 10, 12,
semantic prosody 191-2, 194 145, 265, 267-8
sensitivity 1, 10, 148, 429 Theatre of the Absurd xv, 155
Shakespeare, William ix, xix, thematics 26, 137-8, 151
xxi, 13, 108-9, 134, 191-200, theory vs application 45
202-3, 378, 413-4, 423, 427,
428, 435 U
shape 395, 398, 400, 402-3, universal emotions 363, 367-8
408-9 utterances 283-4, 297, 304,
short story xx, xxi, 13, 61-2, 312, 348-9, 352-4, 375, 380-
65, 70-1, 83, 127, 272, 274, 2, 388, 432-3, 437, 442, 447,
303, 308, 312, 329-30, 367-8, 449
370-1, 377
sociolinguistic variation 303 W
state of stylistics xvii, xxii, 31- Wcopyfind 207, 209, 217-21
2, 42, 75, 79, 138, 345 weak explicatures xxi, 348,
student responses xiv, 75 351-5, 358-9, 361
Index 517
‘weight’ 313-7, 421-5

word combination 413, 423,
425
Wordsworth’s Prelude 413
World Englishes literature x,
xxi, 363, 366-7, 372

The State of Stylistics (PDFDrive)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The State of Stylistics (PDFDrive)

Uploaded by

Copyright:

Available Formats

The State of Stylistics

Amsterdam - New York, NY 2008

The paper on which this book is printed meets the requirements of

Part I: Theoretical Outlooks

Part III: Corpus Stylistics

21. Just for Laughs: The Construction of Nonverbal Humour

Vadim Andreev, PhD, docent and Chair, Smolensk State University

one article on the political value of Tate’s Richard II (Political

subsequent research and publications. He is currently Senior Lecturer,

Polytechnic of Namibia and was responsible for starting the Namibian

‘A Structural Analysis of Wordsworth’s ‘Daffodils’’ in The Writer’s

on the stylistic analysis of drama and film.

to Absurdity in Drama With a Particular Focus on Ionesco’s The Bald

This collection of papers represents the culmination of the 26th annual

must remember to pay close attention to detailed analysis when we do

Part 2 of this volume centres around Cognitive Stylistics. Shun-liang

on the contents of the CONCE (Corpus of Nineteenth Century

‘“Where are you going to my pretty maid?”

1. What we have achieved in the last 40 years and how we have

wealth of good extant analyses is leading to more people being

2. How have we achieved what we have achieved?

Making a point of being analytical, detailed and systematic in analysis

3. New approaches: new (and old) hopes and dangers

spite of the fact that it stemmed from a school known as

development, as well as to understand better how word association

4. Going back to ‘old’ concerns: interpretation(s), effects,

critics. Rather, it seemed to be assumed that, if a particular text was

the works we have analysed together in class. My students studying

have to get some general building blocks about the nature of

appreciated sensitively without the ‘memory’ interpretation. Mistakes

we would need something more significant to contrast with other

5. Interpretation: some examples and what we can learn from

that if we are to understand the nature of interpretation we need to

‘May I go with you, my pretty maid?’

‘Say, will you marry me, my pretty maid?’

‘What is your father, my pretty maid?’

‘What is your fortune, my pretty maid?’

‘Then I can't marry you, my pretty maid.’

addressed by US President George Bush to UK Prime Minister Tony

I am not interested here in whether George Bush said ‘Yo’ or ‘Yeah’,

called ‘enactment’, where the form of the poem symbolically enacts

have made above concerning the mystical impressiveness of the

discussion has assumed, unargued, that the interpretations are

The Secret Sits

(centre vs. periphery) can be maintained in some context which

understanding better how we understand, interpret, appreciate and

__. 1977. Linguistics and the Novel. London: Methuen.

__. 2002. ‘A Cognitive Stylistic Approach to Mind Style in Narrative Fiction’ in

1. Introduction. On the state of stylistics.

or should achieve some state of ideal equipoise. The dangerously

first great perspectivist writers. (Nietzsche famously argued that

understand himself and his own experience, hence the ongoing

transcripts of natural speech and conversation, regular Victorian

Browning clearly intended comfort: his conclusion is inconclusive for

To answer such questions – or rather, as we shall see, to come to

‘productivity’ of the low man’s life. Glorious failure is oxymoronic, if

are closed, is the final condemnation? Interestingly, however, I can

meanings. As noted at the outset of this essay, ‘lying’ is a dangerously

4. Conclusion. On stylistics and poststructuralism.

even prescriptive supposedly ‘standard’ grammars as opposed to

of use, practices and participants develop a grammar as much as any

Todorov, T. 1984. Mikhail Bakhtin. The Dialogical Principle. trsl. W. Godzich

‘Closing statement’ at the Indiana conference in 1958, Jakobson