Rethinking Language, Text and Context : Interdisciplinary Research in Stylistics in Honour of Michael Toolan 9780815395768, 0815395760

"This collection of original research highlights the legacy of Michael Toolan's pioneering contributions to th

1,873 42 5MB

English Pages 332 [349] Year 2019

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Rethinking Language, Text and Context : Interdisciplinary Research in Stylistics in Honour of Michael Toolan
 9780815395768, 0815395760

Table of contents :
Introduction: Language, text and context revisited / Ruth Page, Beatrix Busse and Nina Nørgaard --
Section 1: Plots and progression. Covert progression, language and context / Dan Shen
La La Land: counterfactuality, disnarration and the forked (motorway) path / Marina Lambrou
Scribbling suspense and surprise / Matthew Collins and Mel Evans --
Section 2: Patterns and predictions. Investigating syntactic simplicity in popular fiction: a corpus stylistics approach / Rocío Montoro
'We work hard here': exploring person and place deixis in a corpus of historical migrant letters / Emma Moreton
The Devil has all the best tunes: an investigation of the lexical phenomenon of Brexit / Lesley Jeffries and Dan McIntyre
Corpus stylistics, norms and comparisons: studying speech in Great Expectations / Michaela Mahlberg and Viola Wiegand --
Section 3: Pragmatics and perception. 'Intending to mean, pretending to be': reflections on the limits on genre / Rukmini Bhaya Nair
Indeterminacy and interpretation: what is shown and what is hidden in Michael Haneke's Caché / Billy Clark
Reliability, unreliability, reader manipulation and plot reversals: strategies for constructing and challenging the credibility of characters in Agatha Christie's detective fiction / Catherine Emmott and Marc Alexander
Metaphoric interpretations of a short story by J.D. Salinger: a reader-response study / Laura Hidalgo Downing --
Section 4: Projection and positioning. Winnie-the-Pooh goes paralinguistic / Manuel Jobert
Quotation and overhearing in Austen / Joe Bray
Suppression, silencing and failure to project: ways of losing voice while using it / Chris Heffer --
'Hey YouTube': positioning the viewer in vlogs / Stephen Pihlaja --
Section 5: Politics. The value of intertextual associations: how GM technologies are given value through association / Alexanne Don
Public women: power, gender and semiotic representations / Carmen Rosa Caldas-Coulthard
The citizen caught between dialogue and bureaucracy / Wolfgang Teubert

Citation preview

Rethinking Language, Text and Context

This collection of original research highlights the legacy of Michael Toolan’s pioneering contributions to the field of stylistics and in so doing provides a critical overview of the ways in which language, text and context are analysed in the field and its related disciplines. Featuring work from an international range of contributors, the book illustrates how the field of stylistics has evolved in the 25 years since the publication of Toolan’s seminal Language, Text and Context, which laid the foundation for the analysis of the language and style in literary texts. The volume demonstrates how technological innovations and the development of new interdisciplinary methodologies, including those from corpus, cognitive and multimodal stylistics, point to the greater degree of interplay between language, text and context exemplified in current research and how this dynamic relationship can be understood by featuring examples from a variety of texts and media. Underscoring the significance of Michael Toolan’s extensive work in the field in the evolution of literary linguistic research, this volume is key reading for students and researchers in stylistics, discourse studies, corpus linguistics and interdisciplinary literary studies. Ruth Page is a Senior Lecturer at the University of Birmingham. Her publications include Stories and Social Media (Routledge, 2012), Narratives Online: Shared Stories and Social Media (Cambridge University Press, 2018) and Researching the Language of Social Media (Routledge, 2014). She is the editor of Discourse, Context & Media. Beatrix Busse is Professor of English Linguistics at Heidelberg University. Her research interests include the history of English, English historical linguistic and Shakespeare studies, stylistics and corpus linguistics. She is the co-editor of the series Discourse Patterns and reviews editor of the International Journal of Corpus Linguistics. Nina Nørgaard is Associate Professor of Applied Linguistics and Director of the Centre for Multimodal Communication at the University of Southern Denmark. Her research interests lie within the fields of stylistics, multimodality, multimodal stylistics, critical discourse analysis and the semiotics of architecture. She has a keen interest in combining work and insights from the (traditionally separated) fields of stylistics and multimodality. Her major publications include Key Terms in Stylistics (2010; co-edited with Beatrix Busse and Rocío Montoro) and Multimodal Stylistics of the Novel: More than Words (forth., 2019).

Routledge Studies in Rhetoric and Stylistics Edited by Michael Burke, Christiana Gregoriou, and Joe Bray

Style and Rhetoric of Short Narrative Fiction Covert Progressions Behind Overt Plots Dan Shen Kafka’s Cognitive Realism Emily T. Troscianko Stories, Meaning, and Experience Narrativity and Enaction Yanna B. Popova From Conversation to Oral Tradition A Simplest Systematics for Oral Traditions Raymond F. Person, Jr. Making Sense of Narrative Text Situation, Repetition, and Picturing in the Reading of Short Stories Michael Toolan A Prosody of Free Verse Explorations in Rhythm Richard Andrews Similes, Puns and Counterfactuals in Literary Narrative Jennifer Riddle Harding Rethinking Language, Text and Context Interdisciplinary Research in Stylistics in Honour of Michael Toolan Edited by Ruth Page, Beatrix Busse and Nina Nørgaard For more information about this series, please visit: www.routledge.com/ Routledge-Studies-in-Rhetoric-and-Stylistics/book-series/RSTYL

Rethinking Language, Text and Context Interdisciplinary Research in Stylistics in Honour of Michael Toolan Edited by Ruth Page, Beatrix Busse and Nina Nørgaard

First published 2019 by Routledge 711 Third Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2019 Taylor & Francis The right of the editors to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-0-815-39576-8 (hbk) ISBN: 978-1-351-18322-2 (ebk) Typeset in Sabon by Apex CoVantage, LLC

For Michael, our friend and fellow scholar, with thanks.

Contents

List of Tables List of Figures Acknowledgements Preface

x xii xiii xiv

M I C H A E L B U RKE

1 Introduction: Language, Text and Context Revisited

1

RU TH PAG E , B E ATRIX B USSE A N D N IN A N Ø RGAAR D

SECTION 1

Plots and Progression 2 Covert Progression, Language and Context

15 17

DA N S H E N

3 La La Land: Counterfactuality, Disnarration and the Forked (Motorway) Path

29

M A R I N A L A MB RO U

4 Scribbling Suspense and Surprise

43

M ATTH E W C O L L IN S AN D ME L E VAN S

SECTION 2

Patterns and Predictions 5 Investigating Syntactic Simplicity in Popular Fiction: A Corpus Stylistics Approach RO C Í O M O N TO RO

61

63

viii

Contents

6 ‘We work hard here’: Exploring Person and Place Deixis in a Corpus of Historical Migrant Letters

79

E M M A M O R E TO N

7 The Devil Has All the Best Tunes: An Investigation of the Lexical Phenomenon of Brexit

103

L E S L E Y J E F FRIE S AN D DA N McIN TYRE

8 Corpus Stylistics, Norms and Comparisons: Studying Speech in Great Expectations

123

M I C H A E L A MA H L B E RG AN D VIO L A WIE GA ND

SECTION 3

Pragmatics and Perception 9 ‘Intending to Mean, Pretending to Be’: Reflections on the Limits on Genre

145

147

RU K M I N I B H AYA N AIR

10 Indeterminacy and Interpretation: What Is Shown and What Is Hidden in Michael Haneke’s Caché

164

B I L LY C L A R K

11 Reliability, Unreliability, Reader Manipulation and Plot Reversals: Strategies for Constructing and Challenging the Credibility of Characters in Agatha Christie’s Detective Fiction

177

CATH E R I N E E MMOTT AN D MA RC AL E XAN D ER

12 Metaphoric Interpretations of a Short Story by J. D. Salinger: A Reader-Response Study

191

L AU R A H I DAL GO DOWN IN G

SECTION 4

Projection and Positioning

207

13 Winnie-the-Pooh Goes Paralinguistic . . .

209

M A N U E L J O B E RT

14 Quotation and Overhearing in Austen J O E B R AY

225

Contents 15 Suppression, Silencing and Failure to Project: Ways of Losing Voice While Using It

ix 237

CHRIS HEFFER

16 ‘Hey YouTube’: Positioning the Viewer in Vlogs

254

S TE P H E N P I HL AJA

SECTION 5

Politics

267

17 The Value of Intertextual Associations: How GM Technologies Are Given Value Through Association

269

A L E X A N N E DO N

18 Public Women: Power, Gender and Semiotic Representations

286

CA R M E N ROSA CAL DAS- CO ULTH A RD

19 The Citizen Caught Between Dialogue and Bureaucracy

303

WO L F G A N G TE UB E RT

List of Contributors Index

321 329

Tables

4.1 4.2 4.3 5.1 5.2 5.3 5.4 6.1 6.2 6.3 6.4

6.5

6.6 6.7 7.1 7.2 7.3 7.4

7.5 7.6

Proportion of Letters in the Corpus Containing Narratives Proportion of Letters Containing Narrative Authorship of Letters Non-Nominal Post-Modification Non-Nominal Post-Modification Wildcards and Tags Non-Nominal Post-Modification in the Chick Lit Corpus Non-Nominal Post-Modification in the MVC Overview of the ALC Overview of the JLC Topics in the ALC and JLC, Listed Alphabetically Statistically Significant Words in the ALC and JLC (Using a Biased Odds Ratio of 3 With a Minimum of 10 Occurrences) Statistically Significant Semantic Domains in the ALC and JLC (Using a Biased Odds Ratio of 3 With a Minimum of 10 Occurrences) Frequencies for ‘We’, ‘Here’ and ‘There’ (in the Topic ‘Ireland and America’) in the ALC and JLC References to ‘We’, ‘Here’ and ‘There’ in the ALC and JLC Human Agent Nouns With Suffixes From the NOW Corpus Top 22 Collocates of Human Agent Nouns With Suffixes From COHA, by Collocational Strength (MI 3+) Earliest Occurrences of Brexit and Bremain in British National Press in Order of Occurrence Number of Articles in ProQuest International Newsstand Containing Tokens of Brexit and Bremain in Each Month of 2016 and 2017 Number of Tokens of Brexit and Bremain per Million Words in the NOW Corpus Raw Numbers of Occurrences of Brexit means Brexit in NOW

47 48 49 72 73 73 73 83 85 90

92

92 94 95 107 108 112

114 114 115

Tables 7.7 8.1 8.2 8.3 12.1 12.2 12.3 12.4 12.5 12.6 12.7 12.8 17.1 17.2 17.3 18.1

Frequency of Collocation of soft and hard With Brexit in NOW Corpora Used in Comparison with Great Expectations Clusters in GE that are Highly Frequent in DNov and the Spoken BNC Top Ten Four-word Clusters for Key Comparison Quotes vs. Non-quotes in DNov, for all Key Clusters p < 0.0001 Number and Distribution of Words per Scene in the Short Story Number of Responses per Theme in Question 1 Answers to the Question: Why Does the Ending Make Sense in Terms of Story Development? Cues for Metaphoric Explanations in Students’ Responses Sources and Targets for the war Metaphorical Scenario Sources and Targets for the mental illness Metaphorical Scenario Sources and Targets for the Seymour’s traumatic war experience and suicide Metaphorical Scenario Answers to the Question: Why Do You Think the Story Is Called ‘A Perfect Day for Bananafish’? First 17 Keywords of Corpus and Seven Individual Articles Frequency of Use of ‘Exhorts’ in Specific Wordbanks Corpus Domains Wordsketch of Subjects and Objects of ‘Exhort’ in Rank Order of Frequency Categorisation Framework Following Van Leeuwen (1996)

xi 117 132 133 138 192 197 197 199 201 202 203 203 274 278 279 290

Figures

6.1 8.1 8.2

8.3 8.4 8.5 13.1 17.1 17.2 18.1

Comparative Charts Showing Length of Letters in the ALC and JLC Screenshot of the CLiC Chapter View for Example (2) from Great Expectations Sample of the 106 Occurrences of you know in Quotes— the Tick at the End of the Concordance Lines Indicates Occurrences of the Discourse Marker Concordance of all Eight Instances of I don’t want to in Great Expectations Concordance Showing 16 of 43 instances of old chap in GE CLiC view of chapter 40 in Bleak House showing of such a thing being used in non-quotes Descriptive Framework for Paralinguistic Vocal Features Newspapers as a Field of Recontextualisation (after Maton, 2014) Clusters of Associations of Four Main Signifiers in the Corpus CEO of Petrobras

87 128

130 135 136 139 212 272 281 296

Acknowledgements

If anyone deserves a book in his honour, it is Michael Toolan! Michael is a key figure in stylistics whose publications and research leadership have played a major role in turning stylistics into what it is today. When first envisioning the present volume, we did so under the headline of E. M. Foster’s famous phrase ‘only connect!’ because it so aptly captures a central aspect of Michael’s professional life and personal qualities: his ability to connect. With a theoretical basis in integrationist linguistics, his academic oeuvre is characterised by its fruitful interdisciplinary connection and integration of classical stylistics, pragmatics, corpus linguistics, critical discourse studies and narrative theory, just as his 1992 collection of essays in stylistics—with which the present collection critically engages—emphasises the significance of connecting language, text and context in stylistics. And, on a personal level, Michael has a unique talent of genuinely connecting with other people, be they colleagues, students or people he just happens to meet on his way. The contributors to this volume include Michael’s former and present colleagues, his former and present PhD students and fellow stylisticians with whom he has connected over the years. We represent but a small part of the community with whom Michael has engaged, but also reflect the international scope of his influence across countries and continents. That he decidedly deserves a book in his honour was testified by everybody’s prompt and enthusiastic agreeing to contribute in spite of tight schedules and other obligations. We are grateful for the generosity and intellectual contributions of each author and for their good spirit which has infused the process of developing this project. We are also very grateful to Jennifer Smith and Nina Dumrukcic for their help in the final stages of preparing the manuscript. For an editorial team, this book has very much been a joint effort with emails, documents and skype calls flying to and fro between Denmark, Germany and the United Kingdom in a truly collaborative process which fittingly reflects Michael’s capacity to ‘only connect’. In closing, we hope this book will inspire further connections and conversations between ideas, people and ways of thinking about texts that continue Michael’s legacy to our field for many years to come. Ruth Page, Beatrix Busse and Nina Nørgaard

Preface

I wanted to call this Preface ‘I don’t know what he’s saying half the time, but I’m hooked on his publications: Comprehensible dialogue and integrated multimodal characterisation in the works of Michael Toolan’, as I know Michael loves a bit of rhetoric and punning and it would also be a respectful nod to what I think is one of the finest stylistics articles ever written. But prefaces should be called ‘Preface’, so I have curbed my enthusiasm and gone for the plain style rather than the florid. This is a truly wonderful book that honours not only Michael Toolan the stylistician, the writer, the teacher and the researcher, but also Michael Toolan, the colleague, the mentor, the leader and just generally Michael Toolan the splendid human being. There are very few people in our community who are wiser than Michael, who are more intelligent and who have as much empathy and compassion as he does. On top of all this, he is a damned good madrigal singer too, as many a Palan who has attended a Poetics and Linguistics Association (PALA) conference dinner can attest, and perhaps a slightly less good actor, though in Pala-thespian terms Mick Short is a very hard act to follow. Michael has been inextricably interwoven with PALA as far back as I can remember. I can recall my old PhD supervisor, and one of the founding fathers of the Poetics and Linguistics Association, Peter Verdonk (Professor of Stylistics at the University of Amsterdam), telling me that at the first ever meeting of PALA that took place at the University of East Anglia, organised by Roger Fowler back in the late 1970s, there were two very young scholars hanging around: a certain wide-eyed Paul Simpson and an equally fresh-faced Michael Toolan. Peter’s memory may of course be playing tricks on him; it is, after all, a long time ago, but it is very nice to imagine a youthful Michael as the stylistician’s apprentice: learning his trade, honing his skills, finding his linguistic line. But what of Michael in that youthful time? Well, before he found his way to PALA he had completed a degree in English Language and Literature at Edinburgh University and from there it was down to St John’s College in Oxford where he completed a DPhil on William Faulkner’s style. Here you can see how the seeds of the stylistician were being sown

Preface

xv

early. He went on to teach at the National University of Singapore and thereafter at the University of Washington in Seattle. In the 1990s, he joined the English department at the University of Birmingham, where he has worked since and where to this day he holds the post of Professor of English Language. He is a very active teacher in both undergraduate and graduate modules and has a great ability to teach courses across the curriculum, from modules on grammar, syntax, phonology and morphology to ones about ‘language and law’, ‘the language poets use’ and, of course, his core specialisations of narrative and stylistics. We know Michael best in his role as the Chair of the Poetics and Linguistics Association and of course in his guise as Editor-in-Chief of the Journal of Literary Semantics, a post he has held for more than 15 years. In both these positions he is a natural, kind and inspirational leader. This year, 2018, he is the host of the PALA conference and how fitting it is that stylisticians from around the world, from China and the USA, Japan and Brazil, South Africa and Canada, the Middle East and Europe have converged on leafy Birmingham to celebrate stylistics in Michael’s very own garden of narrative eloquence. Michael may be an inspirational teacher but he is also a prolific writer. He has an extensive number of single-authored books, all of which I can find on my bookshelves at home. These include The Stylistics of Fiction, Total Speech, Language in Literature and Narrative. He has also edited many collections of essays and articles including Language, Text and Context: Essays in Contextualised Stylistics and Language Teaching: Integrational Linguistic Approaches. I am guessing he must have published close to 100 articles, essays and book chapters in leading journals and multi-authored books on various linguistic topics. This book, Rethinking Language, Text and Context, acclaims Michael’s work. In it, there are sections that map out his many areas of expertise, namely, plot progression, patterns, pragmatics, perception and politics. And then there are chapters in these sections on wide-ranging topics, including narrative, genre, reader manipulation, corpora and collocations. If you ever take a moment to look at Michael’ s university webpage, as I did recently when I was pointing one of my Dutch undergraduate students in his direction to do graduate work with him in Birmingham (something I have done several times in the past), then you might very well be struck by the opening word on that page. It is ‘Welcome!’ (exclamation mark included). It is highly unusual in that position and in that context, and readers may be immediately drawn to it, not just because of its highly foregrounded location but also because of its deviant contextual usage. If we were to conduct a corpus analysis of how academics address the outside world and present themselves digitally, then I am guessing that the word ‘welcome’ won’t be breaking any pots in the frequency lists (with or without accompanying exclamation mark). However, run a search for that same word in the written and spoken discourse

xvi

Preface

of Michael Toolan and I am sure those frequency rates will spike (though Michael might have to run a log-likelihood analysis on that to check its significance and generalisability). Back in 1999 when I was assembling authors for Peter Verdonk’s Festschrift (something I should have started with earlier), I approached Michael as a leading stylistician even then and a good friend of Peter’s to ask if he had time to write something. I remember him saying something along the lines of ‘No, I don’t, but I am going to write something for Peter anyhow’. Such is the generosity of the man. And he ended up writing a wonderful piece, at very short notice, on speech patterns in Anglo-Irish literature. In sum, and in hindsight, I will rename this Preface, for this is no occasion for the plain style but for the grand. This encomium will be called ‘We do know what he’s saying all the time, and that’s why we are all hooked on his publications: The wise words and excellent analyses of Michael Toolan; scholar, teacher, writer, friend’. Michael Burke Middelburg, The Netherlands April 13, 2018

1

Introduction Language, Text and Context Revisited Ruth Page, Beatrix Busse and Nina Nørgaard

1. ‘Only Connect’: Language, Text and Context The connections between language, its use, text and context have always been central to stylistics and stylistic analysis. Stylistics is an empirical discipline, where usually, although not exclusively, the analyst focuses on the textual patterns created by (literary) language. The process of interpreting these linguistic patterns does not take place in a vacuum, but, by necessity, always in a combination of at least cognitive, cultural and historical contexts. In this respect, since its emergence, stylistics has tended to avoid the abstraction of theoretical linguistics and the universalising assumptions found in structuralist models of the text. The stylistic conceptualisation of context has also moved from its product-oriented view outside text and discourse to a procedural and dynamic perspective which sees context as being construed in and constructive of language and other semiotic modes as well as by the reader. The stylistic emphasis on contextualisation was made explicit in Michael Toolan’s (1992) seminal collection of essays: Language, Text and Context. The title of our present collection is a deliberate echo of that earlier project—to honour Michael Toolan’s work as well as to highlight the trendsetting potential of his view of context then. The essays brought together in the 1992 collection illustrate clearly the breadth of each of the title’s key terms as these were conceived at the time. First, the collection illustrated the wide range of texts which are of interest within stylistics; texts which are by no means limited to a canonical set of text types, but instead consider a range of (literary) genres such as poetry, prose and plays alongside advertisements in a matrimonial column and the headlines of reports published in a tabloid newspaper. Second, the interpretive frameworks used to explore the language in these texts were similarly broad and showed the branches to which stylistics had already spread out, such as pragmatic stylistics, functional stylistics and critical stylistics. Finally, the kinds of contexts that were important in these essays varied and included the textual context of particular passages which were subject to stylistic scrutiny, as well as the historical and ideological contexts in which those texts and the language

2

Ruth Page, Beatrix Busse and Nina Nørgaard

therein were situated. As the editor’s Preface pointed out, Toolan’s 1992 collection was a landmark that demonstrated the value of making connections between text and context, and between different disciplinary contexts, especially those found at the interface between linguistic and literary scholarship. Some 25 years on, it is time to revisit the connections between language, text and context, for whilst each of these three key terms has remained central to the stylistic enterprise, much has also changed. There are several reasons for this. First, as stylistics has matured over the decades, it has continued to broaden and deepen its interdisciplinary connections. The contextual range of stylistics now incorporates a number of sub-branches, such as cognitive stylistics, corpus stylistics, multimodal stylistics, historical stylistics, postcolonial stylistics and pragmatic stylistics, to name but a few. This widening interdisciplinarity opens up different methods for exploring language, draws attention to different aspects of context and expands the linguistic meaning-making that might be of interest to the stylistician. Second, significant theoretical shifts have taken place within linguistics over the last 25 years and have had important implications for stylistics. It is now less easy to pin down what we might mean by ‘language’ or ‘context’, or even what counts as the ‘text’. For example, a ‘text’ may be considered a single, written artefact, such as a poem, but can also extend beyond this to include semiotic resources and paratextual elements, such as the ways in which a poem performed, recorded and then uploaded to a social media platform like YouTube will also incorporate the sound, gestures and camera angle presented in the video and the resources of the YouTube template itself (including comments, ranking and metadata). Third, the last decade of the twentieth century was marked—among others—by significant social, cultural and technological changes. Technological changes made it possible to investigate language on a large-scale basis and to store natural language data in large quantities in a computer-readable format. These facilitate the quantitative investigation of language in language corpora by means of specific corpus linguistic tools and methodology and allow for innovative ways of interpreting the relationship between language use, text and context. Technological changes have also facilitated the development of new kinds of texts and uses of language. In this Introduction, taking Toolan’s 1992 collection of essays as a starting point, we examine what the theoretical, technological and methodological implications of these changes are for exploring the connections between language, text and context.

2. Looking Back to Look Forward In order to show why we need to revisit the connections between language, text and context, we begin by setting Toolan’s 1992 collection in

Introduction

3

its wider scholarly context. Stylistics was not the only discipline where ‘context’ was gaining particular attention at the end of the twentieth century. The 1990s saw the publication of several key scholarly texts in the social sciences that reoriented the way researchers thought about the relationship between language, text and context. These publications can be understood as a part of a longer, scholarly trajectory, where from the 1970s and 1980s scholars had reacted against the earlier, structuralist attempts to identify universal accounts of language. In these preceding decades, into which modern stylistics was ‘born’, we observe a general shift towards building theoretical models about language and language use based on the empirical observation of texts (which include spoken interactions as well as written data). For example, in their seminal work on narrative structure, Labov and Waletzky (1967) claimed that we could not understand the universal patterns of storytelling set out by narratologists without first attending to the minimal stories that people actually told about their experiences. Whilst the ‘text’ began to gain greater precedence as the means by which scholars developed linguistic theory, so too were the last decades of the twentieth century characterised by an increasing attention to the socially situated nature of language use. In this regard, scholars in the social sciences and the humanities were drawing on models of context whereby they could describe the contrast in different forms of language use according to social categories, such as age, class or gender, which in the classic works of variationist sociolinguistics are said to be reflected in a speaker’s use of language variables. During the 1970s and 1980s, the interest in the social and cultural contexts within applied linguistics, for some scholars, became decidedly political. Emerging from the theoretical starting point of Marxism, the field of critical linguistics led by Fairclough (1989) and Fowler (1986) began to think about how the differences in social categories might be understood as shaped by and contributing to various forms of power. The impact of these ways of thinking about context is evident in the 1992 collection Language, Text and Context. There is particular interest in the social category of gender, with a section devoted to the framing of women in various literary and non-literary texts, just as the lexico-grammatical analysis found in several papers shows the scholarly interest in interpreting the relative agency of particular persons and/ or ways of thinking. However, it is perhaps telling that whilst celebrating this contextualist turn, there is little in the essays from our 1992 predecessor that really explores what context might be. Instead, context seems to be thought of as a relatively stable entity, somehow outside the text (and thus, extra-textual), but accessible by means of interpreting the language and the patterns found therein. It is this view of context which is challenged by other publications contemporary with Toolan’s collection in the 1990s and which led to a more complex, nuanced and multi-layered view of the connections between language, text and context.

4

Ruth Page, Beatrix Busse and Nina Nørgaard

3. Language, Text and Context: From Product to Process Within the social sciences, one of the pivotal texts which reoriented scholarly concepts of context was Alessandro Duranti’s (1992) collection Rethinking Context. There is no mention of stylistics in Duranti’s collection, nor indeed do any of the essays interrogate literary texts but rather focus on spoken interactions of various kinds. Nonetheless, the influence of Rethinking Context on discourse studies, pragmatics, linguistic anthropology and sociolinguistics—all fields which intersect with developments in stylistics—means that the central claims proposed by Duranti should be considered here. Perhaps the most far-reaching claim was the repositioning of context as a process that emerged from language use rather than a predefined form that constrains language performance. In many ways, this view of context is implicit in the way in which stylistics proceeds: contextual factors are inferred from the language in the text. However, typically the interpretation of context within stylistics tends to treat it in a less fluid and co-constructed manner than proposed by Duranti’s work. This opens up further challenges for the stylistician, who might then question how far these claims about the co-constructed nature of context operate also for textual interactions where the sender and recipients of the text are not co-present, and how we as analysts might begin to observe this co-construction of context on an empirical basis. The kinds of texts interrogated by the authors of the essays in Duranti’s collection illustrate neatly the turn towards examining the interactional context as the site where meaning is negotiated by the interlocutors. The essays focused on spoken interaction, observed and analysed in relation to the performances situated in particular, physical places. This led to greater attention to and conceptualisation of the behavioural context and to the physical spaces and places in which texts are produced and consumed, seen, for example, in the sociolinguistic work of Scollon and Scollon (2003). This began to expand the range of communicative resources of interest to the linguist, to include gesture, facial expression and other kinds of embodied interactions. Again, at first sight, this might seem to bear little influence on the stylistic analysis of an individual, print-based text. However, the empirical turn towards context can be seen as part of the wider developments more recently in stylistics which have analysed the experiences of actual readers and their interpretation of particular texts. The tendency for stylistics to approach context via the language found in a pre-existing, often written, text means that the complexities of the textual production and reproduction are not always as obvious as they appear when other kinds of linguistic performances are considered. However, at the same time that Toolan’s 1992 collection was published, within linguistic anthropology, Bauman and Briggs (1990) drew attention to the

Introduction

5

complex ways in which texts are made and remade across different contexts. In their influential essay ‘Poetics and Performance’, Bauman and Briggs set out three elegant concepts: entextualisation, contextualisation and recontextualisation. These ‘telling resources’ have become part of a crucial toolkit within discourse studies (De Fina and Georgakopoulou, 2012, p. 133) which have shaped the understanding of text and context as a dynamic, renegotiated process that unfolds over time. These concepts point to the strategic ways in which language is turned into a stretch of text, how the meanings of those texts are constructed and reconstructed as the interaction unfolds and how those texts can be reinterpreted as they are inserted into new contexts. These concepts are important, for they also emphasise the fluid, interactional process of contextualisation, but, in addition, point to the ways in which texts can be renegotiated iteratively across successive contexts. This demonstrates how the view of context has expanded in terms of the dimensions of time, space and setting, all of which might be of importance in understanding what the language in a particular text might mean. The relationship between one text and another, prior text is an aspect of context that has long been of interest to stylisticians, literary critics and linguists. The ways in which interdiscursive contexts have been explored is varied, including work that draws on Bakhtin’s (1986) notions of heteroglossia and dialogism, the field of fidelity studies and within stylistics. But Briggs and Bauman’s (1990, 1992) work has been particularly pivotal within later scholarship emerging from mediated discourse analysis and geosemiotics, where these chains of recontextualisation are interpreted from a socio-political perspective. The work of Scollon (2008) in particular brought this together with systemic functional linguistics to explore how the processes of recontextualisation were combined in texts with different semiotic qualities, across textual genres in ways that might obscure or promote the agency of one textual producer rather than another. This work was also influenced by a separate but inter-related theoretical development that took place in the 1990s, namely, the emergence of social semiotics as a distinct field of research, led by the work of Kress and van Leeuwen (1996, 2001). The impact of this work within stylistics has been to expand the range of semiotic resources that are perceived as meaningful within any given text and its context, beyond the verbal focus typical of much earlier stylistic research. Thus, whilst the essays in Toolan’s 1992 collection focus solely on verbal patterns, in our present collection we find essays that also include attention to visual aspects of the text such as image and (within the image) gaze (see chapters by Caldas-Coulthard and Pihlaja) and to aural forms of communication (see Jobert’s chapter). The co-constructed nature of context does not result alone from the externally observable, multimodal interactions in which interlocutors engage. The interpretation of language, texts and their context is also a

6

Ruth Page, Beatrix Busse and Nina Nørgaard

cognitive matter. One of the most influential developments in stylistics since the 1990s has been what has been described as ‘the cognitive turn’. Although the start of this ‘cognitive turn’ pre-dates the 1990s, the last decade of the twentieth century saw the publication of a number of texts which signal the gaining traction and influence of cognitive linguistics on stylistics and its understanding of ‘context’. In relation to the interpretation of literary texts in particular, Tsur’s (1992) Towards a Theory of Cognitive Poetics, Turner’s (1991) Reading Minds and (1998) The Literary Mind were landmark publications that marked the increasing interest in modelling how readers made sense of the complexities of literary texts. In the same decade, Ryan’s (1990) work Possible Worlds and the posthumous publication of Paul Werth’s (1999) work Text Worlds set out influential frameworks for describing the ways in which readers make sense of fictional worlds and their various imagined configurations. In many ways, the cognitive turn is a contrasting direction within stylistics that runs counter to the empirical turn in Rethinking Context. As Gavins and Steen (2003) point out, the work of Tsur (1992) and Turner (1991) is in many ways aligned with the structuralist poetics of the 1970s, whilst at the same time sharing with the contextualist turn a commitment to exploring the relationship between the text and the reader. It is perhaps no surprise then that the models proposed in the key texts that were published in the 1990s within cognitive poetics and the interpretations proposed are not founded on empirical observations of neuroscience—that is, the reactions and responses of actual readers— but rather are heuristic models that allowed the analyst to map carefully the readings that were possible from particular literary texts. It is only with later developments that cognitive science has taken into account the behaviour of actual readers, for example, as modelled through eyetracking technologies, or inferred from the responses gathered in focus groups or interviews. Although the co-constructed, fluid and multi-layered nature of ‘context’ is underpinned to at least some extent by empirical observations of textual interactions, explaining the relationship between the patterns of text and context often seems a matter of interpretation. In particular, certain aspects, such as the historical, social and cultural contexts, are not tractable in the same way that the situational contexts of face-toface interaction might be. One influential model, from social psychology, more or less contemporary with the publication of Toolan’s 1992 collection, was put forward by Davies and Harré in their 1990 paper. In this paper, Davies and Harré (1990) suggested that interlocutors could ‘position’ themselves relative to the master narratives of their cultural context. As with stylistics, and indeed other frameworks from conversation analysis (such as membership category analysis), broader social meanings are interpreted from the language in a text, but, in particular, positioning theory has drawn attention to the normative assumptions that are made

Introduction

7

visible through the analysis of the linguistic choices that a speaker might make. The ongoing power of the stereotypes which are associated with apparently taken-for-granted, normative cultural assumptions reminds us that ways of thinking about the world which are reflected and reinforced through the language used in particular texts continue to interact with extra-textual aspects of context. It is to these extra-textual factors in the socio-historical context at the turn of the twentieth century that we now turn.

4. Social and Historical Context: 1990–2018 The conceptual shifts in understanding context we have examined thus far emerge from examining interactions beyond the written word. A second impetus for rethinking the possibilities of what ‘context’ might entail has been driven by another range of texts that were not available at the time that Toolan’s 1992 collection was published: texts produced and consumed by digital technology. It is hard to underestimate the sociocultural influence of the technological changes which dominated the 1990s. This was the decade in which the internet became mainstream, and in which devices used to access the internet became increasingly mobile. As a result of these changes, the range of online genres that have become of interest to stylisticians continues to diversify. Of course, as Herring (2013) reminds us, innovation in communicative formats is relatively rare and, more often, the forms of communication available online have antecedents in earlier genres. Nonetheless, digital fiction (Bell et al., 2014), literary gaming (Ensslin, 2014), fanfiction (Thomas, 2011) and online reading groups (Fuller and Rehberg Sedo, 2013) have become objects of interest for stylistics, both for their distinctive characteristics and for their relationship with earlier, offline forms of literature. Other rather less literary genres have also been of interest within media stylistics, for example, blogs and YouTube videos (as discussed by Pihlaja in his chapter). The development of these online genres is particularly important for probing the relationship between text and context, for they make certain paratextual aspects of context open to scrutiny in ways that the production and consumption of offline forms of literature and other mediated texts do not. The time and place of publication, the number of views and the immediacy and visibility of ranking of these online texts are observable through the metadata appended to digital texts in the way that turning the pages of a book and the time at which it is read are not. Indeed, the metadata available to researchers, such as the connections between the readers who interact with a text, can be visualised in ways that show us how these interactions give rise to various online communities and contexts which transcend physical time and space. More recently, Jones (2015) has pointed to the ways in which online interactions challenge us to think about what lies hidden beneath the surface of the text in the

8

Ruth Page, Beatrix Busse and Nina Nørgaard

ways that metadata can be observed and repurposed by third parties who might be thought of as ‘overhearers’ who carry out forms of surveillance. The technological changes from the 1990s to the present thus deepen, complicate and politicise the forms of context in which texts are produced, consumed and interpreted. Whilst the facets of context that may be used in the service of interpreting textual meaning have expanded in the light of these technological developments, the question of the socio-political contexts and their relationship to particular texts has remained an ongoing concern for stylistics. However, the ways in which the socio-political contexts are understood have shifted somewhat. In part, this has been stimulated by the theoretical developments in particular fields. For example, the changes in thinking about identity politics related to gender are characterised by multiple, competing trajectories. In 1990, Butler’s publication of Gender Trouble marked a key turning point where her discussion of gender as a discursive performance radically opened up far more complex ways of ‘queering’ identity than are evident in the discussion of gender in Toolan’s 1992 collection. In gender studies, the rejection of binary thinking went hand-in-hand with a more contextualised way of approaching gendered identity that drew attention to the intersection between gender and other aspects of identity (Milani, 2018). The consequences of a more performative and decentred approach to gendered politics have been much debated within the field during the last two decades. By the end of the twentieth century, so-called post-feminism had begun to be proposed as a counterpoint to the earlier second-wave feminism. Texts often associated with post-feminism, such as ‘Chick Lit’, began to be explored within stylistics as a result (Montoro, 2012). However, whilst gender politics may have become more complex and contradictory, they remain an ongoing priority for many scholars, who rightly point out that the inequalities in social power that are differentiated along gendered points of division are far from resolved (see Caldas-Coulthard’s discussion of the representation of politicians in Brazil). The more fluid and complex ways of understanding social identity have also impacted the ways in which other kinds of social categorisations have been reconceptualised. This has emerged particularly within the use of Vertovec’s (2007) concept of superdiversity. In the ethnographic and sociolinguistic studies that have drawn on Vertovec’s work, the kinds of identities that are constructed are no longer thought of as fixed, but rather multi-layered, and often have explored the language practices of diasporic communities where the participants’ linguistic repertoires can shift from moment to moment rather than being clearly demarcated within a single language variety. Once again, the interpretation of these interactions is highly contextualised, and often relies on the participants working with the researchers in order to make sense of the  meanings that emerge. One outcome of this increased attention to the polyphonic

Introduction

9

nature of language is that for some researchers, ‘language’ is not thought of as a decontextualised, abstract system which is governed by a single set of syntactic rules. Instead, researchers have begun to speak of ‘languaging’ as a more process-oriented account of how speakers draw on a range of linguistic resources in their interactions. Second, the shift to a more polyphonic perspective takes multilingualism as its norm and thereby decentres a single, nationalistic emphasis on any one language variety as the primary standard against which other varieties are measured. The impact of superdiversity within stylistics has yet to be fully felt (see Busse, forthcoming), but nonetheless serves as an important challenge to us as researchers as we seek to explore the voices that are ‘heard’ within texts of many different kinds not to simplify or impose analytical interpretations that miss the rich and hybrid way in which these voices might interact. The politics of dialogicity remain open to debate, because, whilst accounting for diversity may appear to make certain voices and identities more articulable than previously, this may obscure other kinds of political inequality that become less visible as a result (see Heffer’s discussion of voices in the courtroom).

5. Methodological Implications: How Might We Analyse Language, Text and Context? Thus far, we have seen how the developments in discourse studies, linguistic anthropology and cognitive linguistics have enriched the understanding of context as a fluid, co-constructed and multi-dimensional resource. The influence of these changes can be seen in the kinds of qualitative and quantitative methods that are used to explore the relationship between text and context. In particular, the empirical emphasis within stylistics has been increasingly used to explore different aspects of the context. For example, in order to explore how readers actually do interpret texts, Experimental Studies of Literature (ESL) and Naturalistic Studies of Reading (NSR) have begun to emerge. These show how rather different methods can be used to elicit and observe the interpretations that readers generate, which may expand or challenge the analytical assumptions made by a researcher alone in her reading of a text. Experimental methods have also been used to complement cognitive stylistic methods, for example, by using eye-tracking technologies to document reaction time studies (Jacobs, 2015), and thereby explore differences in the way that readerly responses such as ‘immersion’ might be experienced. In Toolan’s 1992 collection, there is no evidence of this empirical attention to the readerly experience of context beyond that of the analyst’s own experience alone. In our present volume, we see the interpretation of ‘real readers’ in the chapter by Hidalgo Downing. The development of digital tools for exploring the patterns of language use has also increased exponentially since the time of Toolan’s 1992

10

Ruth Page, Beatrix Busse and Nina Nørgaard

collection. This is felt most prominently in the developments of corpus stylistics. Although the use of concordancing tools and the development of digital archives pre-dates the 1990s, again, in Toolan’s 1992 collection, these do not feature as tools for analysis or as objects of scrutiny. This is no doubt due to the fact that the 1990s saw key developments which facilitated much of the work in corpus stylistics. Large-scale corpora such as the British National Corpus began to be compiled in the 1990s, along with historical corpora, such as the Corpus of Early English Correspondence (Nevalainen et al., 1998). Likewise, the corpus linguistic tools that are commonly used by stylisticians, for example, WordSmith tools (Scott, 1996) and Antconc (Anthony, 2014), were developed and released in the same decade. As a result, stylisticians began to be able to probe the linguistic patterns in both literary and non-literary texts drawing on corpus linguistic methodology and attempting to discuss critically the extent to which style and foregrounding can actually be measured. The use of corpus stylistics to explore contextual effects, such as suspense and surprise, is evident in Toolan’s later work (2009, 2016) in his exploration of short stories and more recently in his discussions of media texts. It is perhaps no surprise that corpus stylistics feature in many of the chapters in the present volume, including the exploration of historical texts (see those discussed by Moreton and by Evans and Collins), contemporary literature (see Montoro’s essay) and news media (see the discussion by Jeffries and McIntyre). As technological innovations and big data continue to play a crucial role in the analysis of language and other semiotic modes and digital archives continue to be more widely available, this direction for the exploration of texts and the co-texts looks set to flourish for decades to come.

6. About This Book The chapters in this book demonstrate the ongoing importance of stylistics as a contextual endeavour, rooted in the empirical observation of particular texts and relating them to a range of ideological, historical and textual concerns. The texts that the authors explore in the chapters are wide-ranging, including canonical literary texts (such as novels by Jane Austen discussed by Bray), but also detective fiction (as discussed by Emmott and Alexander), Chick Lit (Montoro) and children’s literature (Jobert). Alongside these literary texts, other genres such as historical letters (Moreton; Evans and Collins) and media texts such as news (Jeffries and McIntyre; Don; Caldas-Coulthard) and courtroom proceedings (Heffer) feature in different chapters. The increasing interest in multimodal texts is also evident in this collection, where films and other kinds of audiovisual content (vlogs) are the objects of interest in four of the 19 chapters (Lambrou, Clark and Teubert all discuss films, whilst Pihlaja discusses vlogs).

Introduction

11

The frameworks and methods that the authors use to analyse the language, texts and contexts in each chapter are also highly varied and reflect the changes in the ways in which context has been conceptualised in different fields, as well as points of continuity with earlier models. For example, Bauman and Briggs’ work on entextualisation is taken up by Heffer in his analysis of courtroom discourse, whilst both Don and Pihlaja make use of Davis and Harré’s concepts of positioning. The influence of possible world theory is felt in Lambrou’s discussion of the film La La Land (2016), and attention to the interpretations made by actual readers is explored by Laura Hidalgo Downing in her discussion of J. D. Salinger’s short story ‘A Perfect Day for Bananafish’ (1953). The influence of corpus methods as a means of exploring digitised bodies of texts is felt strongly in this collection and is taken up to a greater or lesser extent in six of the chapters as a means of exploring how particular contextual effects are achieved. The kinds of specialised corpora that feature in these chapters include bespoke collections of news articles (as examined by Jeffries and McIntyre and by Don), archives of personal letters (Moreton) and other kinds of historical correspondence (Evans and Collins), as well as literary texts (Montoro; Mahlberg and Wiegand). These specialised corpora are compared with larger reference corpora that are now increasingly available, for example, the BNC, COCA and GLOWBE. Twenty-five years ago, it would not have been possible to carry out this kind of analysis. The multi-layered nature of context is teased out by the authors of the chapters in different ways. The complexity of the readers’ interpretation is a common theme in many chapters, whether that is making sense of complex literary or filmic texts, such as a popular film (like La La Land), psychological thrillers (like Caché), detective fiction or a short story by Chekhov. Other chapters focus on the ideological natures of context, whether they relate to contemporary politics, gender or ecological concerns. The linguistic focus of the chapters is also varied, ranging from the morphological analysis of creative blends used in the media coverage of Brexit, to paralinguistic discussions of voice, to the deictic positioning of place in historical letters and in contemporary video-blogs. The majority of the chapters remain (perhaps frustratingly) monolingual, but the cultural and geographical contexts which are included in the book range from Brazil to India to North American and European literature. Given this diversity, the eclectic nature of stylistics and the multi-faceted nature of context, there are many ways in which we could have grouped the chapters that are brought together in this book. We have chosen to group the chapters into five sections, which move (broadly speaking) from a focus on more textually oriented forms of context to the wider matters of political and ideological context. The first section, ‘Plots and Progression’, explores the different ways in which the structural contexts created within particular texts give rise to particular effects, such as suspense, and can generate multi-layered interpretations. The second section,

12

Ruth Page, Beatrix Busse and Nina Nørgaard

‘Patterns and Predictions’, includes the corpus-assisted and corpus-driven analysis of contextual patterns in particular sets of texts. These chapters use word lists, keywords and collocational searches to explore specific linguistic phenomena, such as grammatical complexity, deictics and reported speech, and use these to test and challenge claims that have been made about particular types of text (such as lowbrow literature, nineteenth-century literature or migrant letters). The third section, ‘Pragmatics and Perception’, begins to focus more on the interpretation of texts. For example, Rukmini Bhaya Nair shows how the perlocutionary effects of pretend behaviour and different language games can vary across different cultures but also retain points of commonality. Pragmatic concepts are also used by Billy Clark in his reading of the film Caché. The extent to which readers can be manipulated by effects of reliability and unreliability is the main concern of Catherine Emmott and Marc Alexander in their reading of Agatha Christie’s detective fiction, whilst Laura Hidalgo Downing uses reader-response to illuminate the metaphoric meanings in a short story by J. D. Salinger. The fourth section, ‘Projection and Positioning’, brings together four chapters, each of which investigates how voice is interpreted within particular contexts. In each case, the authors show how the question of ‘who speaks’ is contextually interpreted within particular norms, but is also open to multiple interpretations. Joe Bray’s essay focuses on voices as they are represented in reported speech in literary texts, revisiting the dialogue in Jane Austen’s novel Emma (1816) as a form of ‘overhearing’. Manuel Jobert analyses the paralinguistic framing of the characters’ voices in the classic tales of Winnie-the-Pooh, showing how these vocal notations are an important element in the stylistic analysis of the text’s narrative effects. Chris Heffer’s analysis of voices as hearable within courtroom discourse points to the crucial real-world consequences of how voices are positioned within particular interactions. Stephen Pihlaja also questions the power relations of positioning the video-logger Casey Neistat and his audience, showing how the apparent address to the audience by the celebrity in his videos seems to be a strategy for enhancing the video-logger’s visibility and fan base, rather than actually enabling direct interaction to take place. The final section, ‘Politics’, contains three chapters, each of which interrogates the political consequences of particular textual representations. Alexanne Don uses Appraisal Theory to show how positioning is achieved through an intertextual analysis of news reporting about genetically modified foods. Carmen Rosa Caldas-Coulthard analyses the multimodal representation of Brazilian politicians to show the ongoing gender inequalities that constrain the representation of women in the workplace. Finally, Wolfgang Teubert challenges us to question the discourses that surround us, where inequalities should not be hidden, but where the relationship between language, text and context can be used to question and challenge those inequalities.

Introduction

13

References Anthony, L. (2014) AntConc (3.4.4) [Computer Software]. Tokyo, Japan: Waseda University. Available at: www.laurenceanthony.net/ (Accessed 27 March 2017). Austen, J. (2005) [1815] Emma. Edited by R. Cronin and D. MacMillan. Cambridge: Cambridge University Press. Bakhtin, M. M. (1986) Speech genres and other late essays. Trans. V. W. McGee and Edited by C. Emerson and M. Holquist. Austin: University of Texas Press. Bauman, R. and Briggs, C. (1990) ‘Poetics and performances as critical perspectives on language and social life’, Annual Review of Anthropology, 191 (1), pp. 59–88. Bell, A., Ensslin, A. and Rustand, H. (2014) ‘From theorizing to analyzing digital fiction’, in Bell, A., Ensslin, A. and Rustad, H. (eds.) Analyzing digital fiction. New York: Routledge, pp. 3–17. Briggs, C. L. and Bauman, R. (1992) ‘Genre, intertextuality and social power’, Journal of Linguistic Anthropology, 2, pp. 131–172. Busse, B. (forthcoming) ‘Patterns of discursive urban place-making in Brooklyn, New York’, in Wiegand, V. and Mahlberg, M. (eds.) Corpus linguistics, context and culture. Berlin: Mouton de Gruyter. Butler, J. (1990) Gender trouble. London and New York: Routledge. Corpus of Early English Correspondence. (1998) Compiled by T. Nevalainen, H. Raumolin-Brunberg, J. Keränen, M. Nevala, A. Nurmi and Palander-Collin, M. at the Department of Modern Languages, University of Helsinki. Available at: http://purl.ox.ac.uk/ota/2461 (Accessed 16 August 2017). Davies, B. and Harré, R. (1990) ‘Positioning: The discursive construction of selves’, Journal for the Theory of Social Behaviour, 20, pp. 43–63. De Fina, A. and Georgakopoulou, A. (2012) Analyzing narrative: Discourse and sociolinguistic perspectives. Cambridge: Cambridge University Press. Duranti, A. (1992) Rethinking context: Language as an interactive phenomenon. Cambridge: Cambridge University Press. Ensslin, A. (2014) Literary gaming. Cambridge, MA: MIT Press. Fairclough, N. (1989) Language and power. Harlow, Essex and New York: Longman. Fowler, R. (1986) Linguistic criticism. Oxford: Oxford University Press. Fuller, D. and Rehberg Sedo, D. (2013) Reading beyond the book: The social practices of contemporary literary culture. London and New York: Routledge. Gavins, J. and Steen, G. (2003) Cognitive poetics in practice. London: Routledge. Herring, S. (2013) ‘Discourse in web 2.0: Familiar, reconfigured, and emergent’, in Tannen, D. (ed.) Georgetown University round table on language and linguistics 2011: Discourse 2.0: Language and new media. Washington, DC: Georgetown University Press, pp. 1–25. Jacobs, A. (2015) ‘Towards a neurocognitive poetics model of literary reading’, in Willems, R. M. (ed.) Cognitive neuroscience of natural language use. Cambridge: Cambridge University Press, pp. 135–159. Jones, R. H. (2015) ‘Surveillance’, in Georgakopoulou, A. and Spilloti, T. (eds.) The Routledge handbook of language and digital communication. London: Routledge, pp. 408–411. Kress, G. and van Leeuwen, T. (1996) Reading images. London and New York: Routledge. Kress, G. and van Leeuwen, T. (2001) Multimodal discourse. London: Edward Arnold.

14

Ruth Page, Beatrix Busse and Nina Nørgaard

Labov, W. and Waletzky, J. (1967) ‘Narrative analysis’, in Helm, J. (ed.) Essays on the verbal and visual arts. Seattle: University of Washington Press, pp. 12–44. Milani, T. (2018) Queering language, gender and sexuality. Sheffield: Equinox. Montoro, R. (2012) Chick lit: The stylistics of cappuccino fiction. London: Continuum. Ryan, M. (1990) Possible worlds, artificial intelligence and narrative theory. Bloomington: Indiana University Press. Salinger, J. D. (1953) ‘A perfect day for bananafish’, in Nine stories. Boston, MA: Little, Brown and Company. Scollon, R. (2008) Discourse itineraries: Nine processes of resemioticization. London: Routledge. Scollon, R. and Scollon, S. W. (2003) Discourses in place: Language in the material world. New York: Routledge. Scott, M. (1996) WordSmith tools. Oxford: Oxford University Press. Thomas, B. (2011) ‘What is fanfiction and why are people saying such nice things about it?’ Storyworlds, 3, pp. 1–24. Toolan, M. (1992) Language, text and context. London: Routledge. Toolan, M. (2009) Narrative progression in the short story. Amsterdam: John Benjamins. Toolan, M. (2016) Making sense of narrative text: Situation, repetition and picturing in the reading of short stories. London and New York: Routledge. Tsur, R. (1992) Towards a theory of cognitive poetics. Eastbourne: Sussex Academic Press. Turner, M. (1991) Reading minds: The study of English in the age of cognitive science. Princeton: Princeton University Press. Turner, M. (1998) The literary mind: The origins of thought and language. Oxford: Oxford University Press. Vertovec, S. (2007) ‘Super-diversity and its implications’, Ethnic and Racial Studies, 30 (6), pp. 1024–1054. Werth, P. (1999) Text worlds: Representing conceptual space in discourse. New York: Pearson Education.

Section 1

Plots and Progression

2

Covert Progression, Language and Context Dan Shen

1. Introduction Ever since Aristotle, investigations of narrative fiction have focused on the plot development as the only narrative movement (which may have different branches or layers). But in many fictional narratives, there in effect exists a ‘covert progression’—a thematic-aesthetic undercurrent paralleling the plot development throughout the text (Shen, 2013; Shen [2014] 2016). Significantly, the language choices in the covert progression, in comparison with those in the plot development, are often more closely connected with the sociocultural context. There are various kinds of relationships between the covert progression and the plot development, ranging from harmonious complementation to radical subversion, which invariably complicates (or has the potential to complicate) the readers’ response (Shen, [2014] 2016, 2015, 2017, 2018). This chapter will focus on one kind of complementary relation: The plot development is concerned with family or personal conflict, but the linguistic choices in the covert progression centre on the conflict between individual and society. Significantly, the thematic function of the subtle linguistic patterning in the covert progression often cannot be fully understood unless we take into account biographical and historical contexts. The stylistic uncovering of the covert progression, which tends to elude intuitive interpretation, helps to show ‘the principle of contextualization in stylistic investigation’ (Toolan, 1992, p. xiv) or the interrelation between language, text and context.

2. Individual-Societal Conflict Behind Family Conflict: Kafka’s ‘The Judgment’ In some literary texts with double narrative movements, the plot development focuses on the conflict between family members, but the covert progression concentrates on the social pressure individuals are subject to. Franz Kafka’s ‘The Judgment’ ([1913]1977) is a typical case in point.

18

Dan Shen

Its plot development may be summarised as: Georg Bendemann, a young businessman, writes a letter to a childhood friend (also a businessman) in St. Petersburg, announcing his engagement to a girl from a rich family. He then goes to tell his father about the composition of the letter. The father first questions the existence of the friend, but then says that he has kept correspondence with him. He announces his alliance with the friend (whom he says Georg betrays) and with Georg’s deceased mother. The father accuses his son of being a devilish human being and condemns him to death by drowning, and the son rushes out and drowns himself in the river. Although this narrative has aroused heated controversy, critics have reached the consensus that it centres on ‘the conflict between father and son’ (Binder, 1977, p. 14; see also Rolleston, 1977, p. 134; Brod, 1995, pp. 129–130; Berman, 2002), a consensus backed up by Kafka’s own description of the story as ‘a journey around father and son, and the friend’s changing shape may be a change in perspective in the relationship between father and son’ (Kafka, 1973, p. 267; see also Kafka, 1954, p. 278).1 Behind the plot development, there exists a parallel narrative movement focusing on the conflict between individual and society. In the opening paragraph, the omniscient narrator tells the reader that Georg has written a letter to his friend, and the following seven paragraphs, which take up about one-third of the textual space, all focus on this friend, representing mainly Georg’s thoughts on the friend in free indirect style. From Georg’s reflection, we get to know that the friend is unsuccessful in doing business in Russia and is suffering from some serious disease. Moreover, he is unmarried and has to remain so all his life. Further, he is very lonely, isolated from society—both from his fellow-countrymen and from Russian families. Georg considers the possibility of advising him to return home, to transfer his business to his native country, so that he can benefit from the concern and help of his friends. However, Georg is worried about the severe consequences of such a move: But that would amount to telling him [. . .] that all his efforts so far had failed, that he should finally abandon them, come back home, and be gaped at on all sides as a prodigal who has returned for good, that only his friends understood things and that he himself was a great baby who must simply do as he was told by these friends of his who had stayed put and been successful. And besides, was it even certain that all the pain that one would have to inflict on him would serve any purpose? [. . .] But if he really did follow their advice, only to find himself driven under on his return—not as the result of any malice, of course, but through force of circumstance—if he failed to get on either with his friends or without them, felt humiliated, and so became homeless (keine Heimat) and friendless in all earnest, wouldn’t it be far better for him, in that case, to stay abroad as he was?2

Covert Progression, Language and Context

19

In terms of plot development, Georg’s letter to his friend forms the immediate cause of the conflict between father and son, and the death sentence the father passes on the son has to do with this friend. But it is puzzling that Georg’s reflection on his friend should take up one-third of the textual space. Previous criticisms have focused on the contrasts between the friend and Georg: the former’s being a bachelor, unsuccessful, ascetic and spiritual, dissatisfied with his conditions, versus the latter’s being engaged, materialistic, successful and self-satisfied (see White, 1977 for a summary of previous criticisms and his own comment). The italicised linguistic choices in the passage quoted earlier appear to be trivial or digressive to the plot development since they are not directly associated with the conflict between the father and the son. But these linguistic choices are very significant to the covert progression centering on the conflict between individual and society. In Georg’s view, what prevents his friend from returning home is mainly the opinion of other people or the damage of social pressure on the individual. The italicised adversative ‘But’ ushers in the social judgement on the friend’s work abroad: ‘all his efforts so far had failed’, where the premodifier ‘all’ and the temporal adjunct ‘so far’ function to present a totally negative picture. The choice of passive voice and perception process ‘be gaped at’ subjects the friend to the ‘gaping’ of other people (notice the echoing between this process and the equally passive ‘find himself driven under’). Moreover, the adjunct ‘on all sides’, coupled with the prepositional group ‘as a prodigal’, indicates the overwhelming humiliation the society may impose on the individual. The exclusive mental process ‘only his friends understood things’ and the relational process ‘he himself was a great baby’ show how the comparison with other people will deprive Georg’s friend of his adulthood and reduce him to the deplorable position of being a ‘baby’. Most significant is the cause and effect relationship between the mental process and the relational process: ‘felt humiliated, and so became homeless (keine Heimat)’. If Georg’s friend feels humiliated, he will become homeless or hometownless—cannot survive in his own hometown, which points to how unbearable social pressure is. And the pressure does not come from people’s lack of kindness: The negating and adversative adjuncts ‘not as the result of any malice’ ‘but through force of circumstance’ and the modal of obligation ‘would have to inflict on him’ all point to the unavoidability of social pressure despite people’s good intention. Significantly, what we have here is not an objective description, but Georg’s own subjective conjecture of what situation his friend may face if he comes home. The friendship between the two starts from childhood and they are now both young businessmen, and so Georg can put himself in his friend’s position. From Georg’s reflection on his friend we can infer a prominent characteristic of Georg’s mentality: being very much concerned with social opinion on the individual. If people around, who

20

Dan Shen

represent the society, have a negative evaluation of the individual, the individual will suffer ‘all the pain’, feel humiliated and find the social pressure unbearable. In this light, we can infer why the friend in Russia, who has failed in his business, has chosen not to have contact with his fellow-countrymen, nor with Russian families, and to remain single—he is forced to cut himself off from other people in order to avoid social pressure. Following this quotation comes Georg’s conclusion that because of these reasons ‘it was impossible [.  .  .] to send him any real news such as could be given unhesitatingly to even the most distant acquaintance’ (p. 2). Georg himself not only has been very successful in business for the past few years but also recently got engaged to a girl from a rich family. Considering that his friend in Russia has become a failure both in career and in life, Georg is afraid that telling him ‘real news’ will bring him great social pressure (‘only his friends understood things’ and ‘felt humiliated’). But Georg’s fiancée cannot accept the fact that the uninformed friend will not attend their wedding, and she asserts, ‘If you’ve got friends [Freunde] like that, Georg, you should never have got engaged’ (p. 4). Judging from the use of the plural form ‘Freunde’, the friend in Russia is taken to represent Georg’s friends in general. Since the wedding is a most important social occasion, if friends do not come to one’s wedding, one may lose face. Giving social pressure primary consideration, the fiancée even regards friends’ attending the wedding as the prerequisite, without which one ‘should never have got engaged’. And Georg (p. 4, italics added) responds, ‘Well, we’re both of us to blame there; but I wouldn’t have it any other way now.’ And when, breathing faster under his kisses, she still objected: ‘All the same, it does upset me,’ he thought it really couldn’t do any harm to tell his friend the whole story. ‘That’s how I’m made and he must just take me as I am,’ he said to himself, ‘I can’t fashion myself into a different kind of person who might perhaps make him a more suitable friend.’ The inclusive relational process ‘we’re both of us to blame’ betrays the fact that Georg shares his fiancée’s view about the importance of his friend attending their wedding. Although he tries to silence his fiancée by passionate kisses, she still insists that the friend’s not attending the wedding ‘does upset’ her. Under such pressure, Georg gives in and decides to tell his friend about his engagement. His reasoning at this moment forms a striking contrast to his earlier words, explaining to his fiancée why he cannot inform the friend about the wedding: ‘he would feel awkward and at a disadvantage [. . .] he’d have to go back again alone. Alone—do you realize what that means?’ (p. 4). The linguistic choices ‘awkward’, ‘at a disadvantage’ and ‘alone’ all point to the psychological pain the

Covert Progression, Language and Context

21

friend will suffer if he comes to Georg’s wedding. Indeed, the modality of obligation and the action process ‘would have to go back’, the marked theme and repeated ‘Alone’ and the interrogative ‘do you realize what that means?’ unequivocally point to the severe consequences of the friend’s attending the wedding, and that is why Georg has refrained from informing the friend. But under social pressure, Georg now decides to sacrifice the long-term friendship and he gives himself excuses for doing so, including ‘I can’t fashion myself into a different kind of person who might perhaps make him a more suitable friend’. Significantly, the author uses a comma instead of a full stop in between the fiancée’s words and Georg’s thought: she still objected: ‘All the same, it does upset me’, he thought it really couldn’t do any harm to tell his friend the whole story. The deviant use of the comma subtly indicates that it is the fiancée’s objection that directly and immediately leads to Georg’s changing his mind.3 In analysing ‘The Judgment’, numerous critics focusing on the fatherson conflict have referred to Kafka’s letter to his father for evidence. What they have overlooked is the concern with social pressure Kafka also expresses in this letter. I remember, for example, how we often undressed together in the same cubicle. I skinny, frail, fragile, you strong, tall, thickset. Even in the cubicle I felt a puny wretch, and not only in front of you but in front of the whole world, because for me you were the measure of all things. [. . .] I felt happiest when you sometimes undressed first and I could stay in the cubicle alone and delay the shame of my public appearance. (Kafka, 1973, pp. 24–25, italics added) I could not suddenly transform myself when I came into contact with other people, rather I developed a deeper sense of guilt towards them [.  .  .] Moreover, you objected secretly or openly to everyone with whom I associated [. . .] that mistrust developed within me into mistrust of myself, and a perpetual fear of all other people. (ibid., pp. 55–56, italics added) From such words, we can see that Kafka’s letter to his father is also concerned with the conflict between individual and society. This kind of conflict is emphatically yet implicitly conveyed through the linguistic choices in the covert progression of ‘The Judgment’. Significantly, in this undercurrent, social pressure is embodied in people around or closely associated. The social pressure the friend in Russia may be subject to

22

Dan Shen

when returning home comes from the comparison with Georg and other friends, and the social pressure Georg’s fiancée feels arises from people attending the wedding. As for Georg’s father, the pressure stems from the comparison with his son. When Georg comes to tell his father about the letter he has written to the friend in Russia, the father launches various false accusations against the son, turning himself into a ‘comedian’ (p. 10). The father unequivocally admits that he is ‘playing a comedy’ since that is the only comfort he could get as a widowed old man (p. 10). He asks his son whether it is possible for him to have any other comfort: Tell me—and for the space of your answer you shall still be my living son—what else was left to me, in my back room, hounded by disloyal staff, decrepit to the marrow of my bones? And my son went about the world exulting, concluding deals that I had prepared, falling over himself with glee, and stalking away from his father in the stiff mask of an honorable man! Do you suppose that I didn’t love you, from whom you sprang? (pp. 10–11) The epithet ‘decrepit’ and the adjunct ‘to the marrow of my bones’ point to the pathetic condition the old father is in, which makes him extremely vulnerable to social pressure. In striking contrast, the son goes ‘about the world exulting’, ‘falling over himself with glee’. The action process ‘stalking’ and the adjuncts ‘away from his father’ and ‘in the stiff mask of an honorable man’ convey the father’s personal envy and resentment in the plot development, but in the covert progression, they also indicate the unbearable social pressure the son’s success has subject the father to. If the friend in Russia feels great social pressure because Georg and other friends have become successful, we can imagine what social pressure the widowed, betrayed (‘hounded by disloyal staff’) and extremely ‘decrepit’ father would feel in the face of the robust son’s success in business and in life (engaged to a rich girl). When Georg was young, his parents loved him very much (‘Do you suppose that I didn’t love you, from whom you sprang?’) and let him live in a sunny front room, while they themselves chose to live in a back room. But the situation has changed, and the son’s success now gives rise to unbearable social pressure on the widowed and senile father. Earlier in the text, the father admits that he knows Georg’s friend and claims that the friend ‘would have been a son after my [his] own heart’ (p. 9). The lonely, sick and unsuccessful friend in Russia would not bring any social pressure on Georg’s father, and therefore he wins the father’s heart. Usually a father would wish his son to be successful both in career and in life, but in the covert progression of ‘The Judgment’, social pressure is most undesirable and unbearable, a kind of pressure that renders the father morbid in mentality and makes him prefer to have a son who is a total

Covert Progression, Language and Context

23

failure. As regards his successful son, the father even wishes him to be dead to get rid of the pressure—the clause ‘for the space of your answer you shall still be my living son’ foreshadows the father’s sentencing his son to death later on. At this moment, to reduce the pressure, the father plays ‘a comedy’ and falsely accuses his son in various ways to make the son feel inferior, so as to shift the pressure onto his son. And the son feels overwhelmed. He retreats to a corner of the room, trying to keep away as far as possible from his father (p. 10). From the detailed description on page 10 of the text, we get to know that for a long time in the past, Georg had always maintained the determination to keep high vigilance against other people, protecting himself from possible attacks in all directions from the outside world. At this moment, he recalls ‘this long-forgotten resolve’, but it immediately ‘slipped’ out of his mind. This description of Georg’s fear of the outside world reminds us of his words in the letter to his father ‘a perpetual fear of all other people’. For the past few years, Georg has been very successful, enabling him to be free from social pressure. At this moment, however, under his father’s attack, his vigilance comes back, but he immediately forgets about it, which echoes ‘for all the time he kept forgetting everything’ (p. 11). This forgetfulness underlies his taking his father’s false accusations and claims to be true. In order to kill Georg so as to get rid of the social pressure, the father falsely accuses Georg of betraying or maltreating his friend, his mother and himself, threatens that he will ‘sweep’ the fiancée away from Georg and claims that he himself has formed an alliance with all other people around—the friend, (the dead) mother and Georg’s clients—to fight against Georg (pp. 9–12). The father says loudly, ‘So now you know what else there’s been in the world besides you, until now you’ve known of nothing but yourself. You were an innocent child, it’s true, but it’s even more true that you’ve been a devilish human being! And so hearken to me: I sentence you now to death by drowning!’ (p. 12). The adjunct ‘in the world besides you’ and the phenomenon of the mental process ‘know’ (‘nothing but yourself’) point to the conflict between individual and society. As a child, Georg did not bring any social pressure on his robust and successful father and his parents were very proud of him at that time (p. 12), but as an adult, his being strong and successful involves much social pressure on his widowed and weak father, and so he has turned ‘devilish’ in his father’s eyes. The father, who has become morbid under social pressure, would rather kill his own son to free himself from the unbearable pressure. At the father’s death sentence, Georg finds himself ‘driven from the room’ to rush to the river to drown himself (p. 12), which reminds us of Georg’s imagining his friend ‘to find himself driven under on his return  .  .  . through force of circumstance’. Social pressure forces the friend to stay abroad in a deplorable condition and social pressure leads to the morbid father’s killing his own son. Before drowning himself, Georg calls out in

24

Dan Shen

a soft voice: ‘Dear parents, I did always love you’ (p. 12). The affectionate relationship between father and son is destroyed by social pressure, which deprives the father of the only relative to look after him. Both the son and the father, as well as the friend in Russia and Georg’s fiancée, are victims of the modern society. In modern times, people are faced with ‘overwhelming social force’ and ‘the metropolitan type of man—which, of course, exists in a thousand individual variants—develops an organ protecting him against the threatening currents and discrepancies of his external environment which would uproot him’ (Simmel, 1969, pp. 47–48). There is a critical consensus that ‘The Judgment’ is ‘Kafka’s seminal story. It contains in miniature the essence of his themes and techniques in his later work’ (Flores, 1977, Foreword). If we look at only the plot development of ‘The Judgment’, we can see only family conflict, but this is not a major thematic concern of Kafka’s later work. If we can uncover the covert progression through a careful examination of the linguistic choices in ‘The Judgment’, we will be able to discover that the ‘themes’ shared by this text and Kafka’s later texts, such as Metamorphosis ([1915]2015), The Trial ([1925]2015) and The Castle ([1926]1998), centre on the conflict between individual and society or the predicament of people in the modern society.

3. Individual-Societal Conflict Behind Personal Conflict: Mansfield’s ‘The Singing Lesson’ Katherine Mansfield’s ‘The Singing Lesson’ ([1920]1945) is set in a girls’ school. The plot development focuses on the conflict between the female protagonist Miss Meadows and her fiancé, as well as between Miss Meadows and another female teacher or her pupils. At the beginning of the narrative, Miss Meadows is in a very bad mood and is hostile towards the Science Mistress on her way to teach a singing lesson, and when in the classroom, she treats her pupils harshly and asks them to sing a mournful lament, all because her fiancé Basil has broken off their engagement. Then she is summoned to the school office where she receives a telegram from Basil in which he promises to marry her. This transforms Miss Meadows into a warm, happy and kind woman. And when back in the classroom she asks her pupils to sing a joyful paean instead. This plot development has received drastically different interpretations. Some critics take the narrative optimistically as an illustration of Mansfield’s effort to show ‘“that marvelous triumph” when beauty holds the balance over the ugliness in life’ (O’Sullivan, 1996, p. 137; Kobler, 1990, p. 141). By contrast, some critics take the story as a satirical depiction of a shallow, stupid and unkind spinster who blindly depends on a man, and the end of the narrative is not seen as being triumphant, but ‘mournful’, since Miss Meadows ‘would be better off to suffer the initial hurt than to be bound to Basil forever’ (Morrow, 1993, p. 85).

Covert Progression, Language and Context

25

If we analyse the linguistic choices carefully, we may find that behind the plot development, there in effect exists a covert progression centering on the conflict between woman and the patriarchal society (for a detailed discussion, see Shen, [2014] 2016, pp. 111–124). Let us first examine the opening sentence of the narrative: ‘With despair—cold, sharp despair— buried deep in her heart like a wicked knife, Miss Meadows, in cap and gown and carrying a little baton, trod the cold corridors that led to the music hall’ (Mansfield, 1945, p. 343). Although despair is Miss Meadows’s own internal feeling, Mansfield metaphorises it into an external knife—a ‘cold, sharp’ and ‘wicked’ knife—piercing Miss Meadows’s heart, starting to invite us to see her as a victim of external forces. From Mansfield’s pen, the metaphorical adjunct ‘With . . . like a wicked knife’ occupies the prominent position of the very beginning of the narrative, and its syntactic status as the marked theme of the sentence contributes to the foregrounding of the metaphor in the reading process. The Science Mistress stopped Miss Meadows. ‘Good mor-ning,’ she cried, in her sweet, affected drawl. ‘Isn’t it cold? It might be win-ter.’ Miss Meadows, hugging the knife, stared in hatred at the Science Mistress . . . ‘It is rather sharp,’ said Miss Meadows, grimly. The other smiled her sugary smile. ‘You look fro-zen,’ said she. Her blue eyes opened wide; there came a mocking light in them. (Had she noticed anything?) ‘Oh, not quite as bad as that,’ said Miss Meadows, and she gave the Science Mistress, in exchange for her smile, a quick grimace and passed on. (ibid., p. 344, italics added) Here the omniscient narrator’s perspective interacts with Miss Meadows’s limited perspective. As is revealed later in the narrative, Miss Meadows is in despair because she has received Basil’s letter breaking off their engagement. Miss Meadows’s hostility (‘in hatred’, ‘grimace’) to the Science Mistress is ill grounded since, without access to the personal letter, the Science Mistress is undoubtedly in the dark about the happening. The material process ‘there came a mocking light in them’ indicates Miss Meadows’s ungrounded suspicion instead of being an objective narratorial description. This point can gain support from the following free indirect thought: ‘Had she noticed anything?’ Through the free indirect thought, Mansfield signals to us that what Miss Meadows is afraid of is other people getting to know the breaking off of the engagement. The use of brackets surrounding Miss Meadows’s speculative thought functions paradoxically to draw attention to it, subtly highlighting her worry about other people’s opinion. This invites us to experience the

26

Dan Shen

social pressure Miss Meadows is in fear of, which is further expressed by the following words: She would have to leave the school, too. She could never face the Science Mistress or the girls after it got known. She would have to disappear somewhere. Passes away. The voices began to die, to fade, to whisper . . . to vanish . . . (ibid., p. 348, original italics marking words from a song) Here we have Miss Meadows’s inner thoughts presented in free indirect discourse. The repeated modal of obligation ‘would have to’ and the action processes ‘leave the school’ and ‘disappear somewhere’ show that Miss Meadows would lose her job because she ‘could never face’ people in the school. Significantly, from Mansfield’s ingenious pen, the words in the song ‘Passes away’ echo both the preceding ‘disappear’ and the following ‘to die’ and ‘to vanish’, to form a series of death images, subtly indicating that Miss Meadows may not be able to survive if the breaking off of the engagement ‘got known’. Why, then, does other people’s opinion concerning the engagement or marriage matter so much to the female protagonist? To answer this question, we need to look into the Victorian social context. Victorian England was marked by ‘the universal obsession with marriage’, which ‘the entire weight of nineteenth-century ideology put forward as being the culminating point of a woman’s life’ (Basch, 1974, p. 16). The Victorian society believed that a woman who failed to catch a husband was unattractive, unintelligent and useless, often making her an outcast (Auerbach, 1982, p. 111). In ‘The Singing Lesson’, although there is no reference to any historical period, the most severe consequences the discarded Miss Meadows would have to suffer suggest that Mansfield has in mind a phallocentric society like that of the Victorian age where marriage is of paramount importance to a woman and where a man’s desertion of a woman can deprive the woman of all her value. Mansfield once wrote in her diary: ‘We are firmly held with the self-fashioned chains of slavery . . . It is the hopelessly insipid doctrine that love is the only thing in the world, taught, hammered into women, from generation to generation, which hampers us so cruelly’ (Murry, 1954, pp. 36–37). In light of the diary, we may gain a better understanding of the metaphorical action process ‘hugging the knife’ at the beginning of ‘The Singing Lesson’, with the process ‘hugging’ subtly hinting at Miss Meadows’s embracing the ‘self-fashioned chain of slavery’. But, of course, even in the Victorian society, a woman unwanted by men would not have to ‘disappear’ from society. In ‘The Singing Lesson’, Mansfield is using the fictive covert progression to indicate and satirise, in a highly dramatised yet implicit manner, the phallocentric discrimination against a woman discarded by a man, threatening her job, even her very existence.

Covert Progression, Language and Context

27

As regards Miss Meadows’s relation with other characters, in the overt plot, her hostility towards the Science Mistress and harsh treatment of her pupils is only a matter of her own bad temper, and we see her unreasonably venting her frustration on other people. In the covert progression, however, Mansfield leads us with subtle linguistic choices to perceive that in the phallocentric society whether Miss Meadows can get married is a life-and-death matter for her and that the Science Mistress or the pupils will unwittingly form part of the social forces functioning to drive Miss Meadows away, even driving her into death. Seen in this light, Miss Meadows’s hostility and bad temper become quite understandable.

4. Conclusion In ‘The Judgment’ and ‘The Singing Lesson’, a careful stylistic analysis has enabled us to uncover that, behind the plot development centering on family or personal conflict, there is a covert progression where the linguistic choices are culturally conditioned and are aimed at conveying the conflict between individual and society. It should have become clear that, if we miss the covert progression in narratives like ‘The Judgment’ and ‘The Singing Lesson’, we can gain only a one-sided understanding of the thematic import of the text, a partial image of the characters, and we can appreciate the narrative’s aesthetic value in only a limited way. Significantly, the stylistic uncovering of the covert progression enables us to gain not only a fuller picture of the text, but also a much better understanding of the relation between the text and the context.

Notes 1. Some critics regard the father as a tyrant, a representative of oppressive patriarchal authority (see, for instance, Neumann, 1981), but some take the narrative as an affirmation of patriarchal authority (see, for instance, Pan, 2000), or take the son as being too egocentric, deserving punishment (see, for instance, Berman, 2002). 2. Franz Kafka, ‘The Judgment’, translated by Malcolm Pasley, in Angel Flores, ed. The Problem of ‘The Judgment,’ New York: Gordian Press, 1977, pp. 1–2, italics added. Further references are to this edition and are given parenthetically in the text. 3. Although the linkage between the fiancée’s spoken words and Georg’s inner thought appears to be loose and ungrammatical, another translator has also preserved it: ‘she still brought out: “All the same, I do feel upset”, he thought it could not really involve him in trouble were he to send the news to his friend’ (Kafka, [1913]1993, p. 7).

References Auerbach, N. (1982) Woman and the demon. London: Harvard University Press. Basch, F. (1974) Relative creatures. New York: Schocken. Berman, R. A. (2002) ‘Tradition and betrayal in Das Urteil’, in Rolleston, J. (ed.) A companion to the works of Franz Kafka. New York: Camden House, pp. 85–99.

28

Dan Shen

Binder, H. (1977) ‘The background’, in Flores, A. (ed.) The problem of “The Judgment”. New York: Gordian Press, pp. 13–38. Brod, M. (1995) Franz Kafka: A biography. Boston, MA: Da Capo Press. Flores, A. (ed.) (1977) The problem of ‘The Judgment’. New York: Gordian Press. Kafka, F. (1954) The diaries of Franz Kafka, 1910–1913. Trans. J. Kresh and Edited by M. Brod. New York: Schocken. Kafka, F. (1973) Letters to Felice. Trans. J. Stern and E. Duckworth. New York: Schocken. Kafka, F. ([1913]1977) ‘The judgment’. Trans. M. Pasley, in Flores, A. (ed.) The problem of ‘The Judgment’. New York: Gordian Press, pp. 1–12. Kafka, F. ([1913]1993) ‘The judgment’. Trans. W. Muir and E. Muir, in Selected short stories of Franz Kafka. New York: The Modern Library, pp. 3–19. Kafka, F. ([1915]2015) Metamorphosis. Trans. D. Wyllie. Sweden: Wisehouse Classics. Kafka, F. ([1925]2015) The trial. Trans. D. Wyllie. Sweden: Wisehouse Classics. Kafka, F. ([1926]1998) The castle. Trans. M. Harman. New York: Schocken Books. Kobler, J. F. (1990) Katherine Mansfield. Boston, MA: Hall. Mansfield, K. ([1920]1945) ‘The singing lesson’, in Collected stories of Katherine Mansfield. London: Constable, pp. 343–350. Morrow, P. D. (1993) Katherine Mansfield’s fiction. Bowling Green: Bowling Green State University Popular Press. Murry, J. (1954) Journal of Katherine Mansfield. London: Constable, 1954. Neumann, G. (1981) Franz Kafka, ‘Das Urteil’. Munich: Hanser. O’Sullivan, V. (1996) ‘The magnetic chain: Notes and approaches to K.K.’, in Pilditch, J. (ed.) The critical responses to Katherine Mansfield. Westport: Greenwood Press, pp. 129–155. Pan, D. (2000) ‘The persistence of patriarchy in Franz Kafka’s “The Judgment” ’, Orbis Litterarum, 55, pp. 135–160. Rolleston, J. (1977) ‘Strategy and language: Georg Bendemann’s theater of the self’, in Flores, A. (ed.). The problem of ‘The Judgment’. New York: Gordian Press, pp. 133–145. Shen, D. (2013) ‘Covert progression behind plot development: Katherine Mansfield’s “The fly”‘, Poetics Today, 34 (1/2), pp. 147–175. Shen, D. (2016) [2014] Style and rhetoric of short narrative fiction: Covert progressions behind overt plots. London and New York: Routledge. Shen, D. (2015) ‘Dual textual dynamics and dual readerly dynamics: Double narrative movements in Mansfield’s “Psychology”‘, Style, 49 (4), pp. 428–435. Shen, D. (2017) ‘Dual narrative progression as dual authorial communication: Extending the rhetorical model’, Style, 51 (4), pp. 61–66. Shen, D. (2018). ‘Dual narrative movement and dual ethics’, Symplokē, 25 (1-2), pp. 511-515. Simmel, G. (1969) ‘The metropolis and mental life’, in Sennett, R. (ed.) Classic essays on the culture of cities. NJ: Prentice-Hall, pp. 47–60. Toolan, M. (1992) ‘Preface and acknowledgements’, in Toolan, M. (ed.) Language, text and context. London: Routledge, pp. xiv–xv. White, J. J. (1977) ‘Georg Bendemann’s friend in Russia: Symbolic correspondences’, in Flores, A. (ed.) The problem of ‘The Judgment’. New York: Gordian Press, pp. 97–113.

3

La La Land Counterfactuality, Disnarration and the Forked (Motorway) Path Marina Lambrou

1. Introduction In the preface of Language, Text and Context: Essays in Stylistics published in 1992, Michael Toolan explains that ‘Stylisticians have a particular expertise with the form, function and structure of language in discourse’ and with the usefulness of using ‘literary linguistic procedures’ to interpret and understand the text (p. xiv). For the purposes of this discussion, I define text as ‘a stretch of language that, in terms of communicative meaning, is complete in itself’ (Lambrou, 2015, p. 93.) Toolan’s edited volume offers essays on a range of discourse types including literary and media texts to illustrate how context is as important as the stylistic analysis for their interpretation. Moreover, it is a testament to the development and success of stylistics providing an interdisciplinary approach for analysing discourses for meaning and insights. In later work that supports Toolan’s earlier claims that stylistic approaches can be applied more broadly, he also discusses issues surrounding whether film can be subjected to a stylistic analysis. In his chapter ‘Stylistics and Film’ Toolan (2014, p. 455) points out that stylistics is usually concerned with language but ‘film is film’: a metonym for a particular kind of cultural artefact: the output of a recording, on film or in digital files, of a sequence of images and sounds . . . which tell a fictional narrative; and further (except in TV series), usually between one and two hours in duration. Where Nørgaard et al. (2010, p. 21) state that like textual forms, ‘film stylistics aims for a more retrievable way of analysing cinematic forms based on frameworks which have already proven successful’ Toolan (2014, p. 455) contends that ‘[l]anguage is not a necessary ingredient’, especially in silent films, and this is problematic when discussing the application of existing textual stylistic tools to texts other than those that are spoken and written. Toolan (2014) acknowledges that the visual and auditory channels of communication found within film can be subjected to

30

Marina Lambrou

multimodal levels of analysis, and within the visual and auditory modes, language is present. He then goes on to discuss the procedural challenge for the stylistics of narrative fiction which provides the important systematic and retrievable criteria that are central to a stylistic analysis of any text type. Films specifically have a grammar of film design-features, such as perspective and camera-angles (see Bordwell and Thompson, 2012), which coexist with the narrative—text grammar (i.e. genre, intertextual sense-making, mental scripts) and other characteristics, which like literary texts have their own characteristics that are open to a stylistic analysis. For example, Hansen (2014, pp. 223–224) in his discussion of film and ‘the narrative of a musical piece’ looks to models and concepts from ‘formalism, structuralism and semiotics rather on the models developed within literary studies’ for their understanding. The points raised by Toolan (2014) in his discussion of film stylistics are useful and this chapter builds on this work in order to contribute to the broadening interest in the stylistic analysis of film. This chapter focuses on the musical romantic comedy La La Land and draws on narratology, film theory and stylistics for the analysis and interpretation of plot development and the unusual and unexpected storytelling strategies. The key terms that will be discussed in the analysis of the narrative dimensions in La La Land are counterfactual divergence (Dannenberg, 2008), the forked path (Borges, [1944] 2000) and disnarration (Prince, 1988), which will be discussed in turn after a synopsis of La La Land.

2. Synopsis of La La Land The film La La Land was written and directed by Damien Chazelle (2016) and can be described as a musical romantic comedy. Central to the plot is the relationship between a jazz pianist, Sebastian (played by the actor Ryan Gosling), and aspiring actor Mia (played by the actor Emma Stone). The plotline (or story, as I use both terms interchangeably) at first appears to adhere to the genre of a romantic comedy following the conventional and idealised Hollywood film genre, that is, the ‘boy meets girl and they fall in love’ trope, though it is not with several ‘hit and miss’ scenarios (described in Excerpts 1 and 2 later in this chapter). With the inclusion of musical song and dance numbers from the opening big production number of ‘Another Day of Sun’ that involves a huge cast bursting into song, or what Hansen (2014, p. 223) calls ‘the frame-breaking aspect of the genre’, to the iconic song and dance scene between Mia and Sebastian to the song ‘A Lovely Night’, overlooking Los Angeles, the film also references musicals from Hollywood’s ‘Golden Age’ (which refers to the period of film-making from the early 1930s to the mid-1950s). Both Sebastian and Mia are passionate about their careers and desire for success—Sebastian as a pianist aspiring to own his own jazz club and Mia as an actor—and audiences see them both struggle for most of the

La La Land

31

film before going on to achieve their dreams. It is not until the final part of the film, however, that audiences understand more fully the sacrifices each has made to achieve their success and the cost this has taken on their relationship. The final plot twist is delivered in an anachronistic flashforward where audiences are transported five years into the future when they learn that Mia is not with Sebastian but is instead married to another man, David, and they are parents of a child. In a last-minute change to their evening plans, and by chance, Mia and David walk into a jazz club called Seb’s and sit down at a table to watch the entertainment. Mia is startled to see Sebastian at the piano: Seb’s is Sebastian’s jazz club. At this moment, Sebastian and Mia’s eyes meet and time appears to stop, whilst both imagine the same what-might-have-been scenario had they stayed together, presented in a dreamlike sequence. Audiences are transported along with Mia and Sebastian to their alternative dimension, almost as voyeurs, and caught in a wishful state of reparation for the idealised Hollywood ending before being brought back to the present time and reality in which Mia and Sebastian did not have a traditional happy-ever-after closure to their romance narrative.

3. Musical Romantic Comedy Film Genre Film genres are identified by assessing three key components ‘descending from the surface deeper into the film: the visual characteristics, narrative patterns and wider ideology’ (McDonald, 2007, p. 11). Specifically, visual characteristics deal with iconography that can include locations, props, costumes and stock characters, while narrative patterns describe the tropes that commonly appear within specific film genres. In romantic comedies, the ‘“master narrative” is the boy meets, loses, regains girl’ (McDonald, 2007, p. 12) beginning with the ‘meet cute’, when the couple first cross paths (with cute suggesting a scenario where the couple meet in an often humorous encounter to suggest to audiences their eventual coupling). The final component, ideology, is best described as reflecting the underlying values of the genre, e.g. a romantic comedy will inevitably end with the happy-ever-after closure and possibly marriage. A master definition of the romantic comedy genre, therefore, can be said to describe ‘a film which has as its central narrative motor a quest for love, which portrays this quest in a light-hearted way and almost always to a successful conclusion’ (McDonald, 2007, p. 9). In many ways, La La Land adheres to the romantic comedy genre despite the ‘almost always’ conclusion of happiness together. However, a true definition of a romantic comedy has love as the central motivation for the characters and drives the plot, whereas I argue that Mia and Sebastian are driven by their determination to succeed in their careers and that their romance is something that happens along their way as a counterpoint to their achievement. In other words, romance is a distraction

32

Marina Lambrou

rather than motivation for La La Land although both characters are spurred on by each other to succeed and perhaps they needed to meet to help each other to reach this outcome. A question that may be asked is why make La La Land a musical as this subgenre of a romantic comedy further suggests the film will lead to a happy ending based on previous musical romantic comedies of the Hollywood Golden Age. For example, audiences are immediately presented with the choreographed song and dance opening sequence of the film with its elaborate camera shot tracking and overhead and panoramic views of the location. Later in the film, the iconic dance routine between Mia and Sebastian (which is also the image that is used for the film’s publicity posters) allows the couple to play out their romance against the romantic starry background of night time Los Angeles (with the urban location being an example of the location iconography of the romantic comedy genre, as described earlier). As Altman (1987, p. 85) explains, ‘If lovers’ hearts beat in time to each other, then the dance provides the opportunity to rehearse that rhythm, to slide imperceptibly from indifference to passion’. Moreover, a familiar trope will see ‘their dances beginning as bickering then moving towards union’ (p. 161). Mostly, the song and dance routines are integrated into the narrative plot in the storyworld, for example, Sebastian singing and dancing on his way home after arranging a date with Mia, which is a reflection of his emotional state or what Hansen (2014, p. 235) describes as ‘the musical (emotional) flow’ that does not undermine the narrative. This particular scene in La La Land recalls the nostalgic tropes from the Golden Age of Hollywood musicals both visually and technically by making direct references to the films of Fred Astaire and Gene Kelly, for example. Keen audiences will see Sebastian wearing ‘spats’ and dancing with a trilby hat that alludes to Fred Astaire’s top hat and Gene Kelly’s fedora in the most memorable dance routines. There is also reference to the silent movie genre later in the film which will be discussed late in this chapter.

4. The What-Might-Have-Been ‘What If?’ Scenario As the title of the chapter suggests, the main emphasis of this chapter, ‘La La Land: Counterfactuality, Disnarration and the Forked (Motorway) Path’, is to explore the various dimensions of narrative and the multiple possibilities that emerge from counterfactual divergence and disnarration as one set of possibilities is dismissed. The final, anachronistic time sequence presented as a storyworld within the film’s storyworld purposely plays with the audience’s expectation of the plot outcome, particularly if the rules of the musical romantic comedy film genre are applied and audiences rely on their schematic knowledge based on their previous exposure to the romantic comedy genre and ‘happy-ever-after’ trope. In the alternative what-might-have-been scenario that arises when

La La Land

33

Mia and Sebastian accidentally see each other five years in the future of the film’s original storyworld timeline, audiences are shown what is best described as a counterfactual alternative reality for the couple and their life together. This story is told as a film montage showing a temporally ordered sequence of short scenes, depicting generic fragments of significant moments of their life together, such as Mia pregnant, Mia with their new-born baby, the baby as a toddler and birthday celebrations with the family, etc. Their hypothetical life together shows the couple living a blissful and domesticated life in what can only be described as an idealised romantic Hollywood ending. Curiously, the montage of their what if? life together is triggered when Sebastian and Mia sit down in front of a screen as if to watch a home movie creating a break from the storyworld. Moreover, the montage is presented as a silent home movie with the requisite flickering, out of focus and sunlit ‘home movie’ style of the cinematography (shot on 16mm anamorphic Kodak film as opposed to digital film to create this particular effect). The wordless nature of the montage with just the underlying musical score of the song ‘City of Stars’, previously sung by both characters, which I suggest is their signature tune, then changes to ‘Epilogue’ when the home movie part of the montage ends, both of which function to complement the visual quality where the focus is on the narrative. The what if? moment passes and Sebastian and Mia are back in real time in Seb’s jazz café with Sebastian on piano and Mia sitting next to her husband, David. The film ends with Mia and her husband getting up to leave, but on her way out, Mia hesitates and turns to look back at Sebastian. Sebastian looks up and as their eyes meet a final, knowing smile passes between them where both acknowledge their past and present, and accept the sacrifices they have made to achieve their success. At the time of writing, Sebastian and Mia’s alternative ending is widely available on the internet to watch and readers unfamiliar with La La Land may wish to view this section of the film.

5. Counterfactual Divergence The what if? scenario in La La Land presents an alternative fictional storyworld or a counterfactual plot that is best described by Dannenberg (2008, p. 119) as ‘a hypothetical alteration in a past sequence of events that changes the events in a factual sequence in order to create a different, counterfactual outcome’. The point when the alteration is made is the antecedent and the point of the counterfactual divergence is the consequence. This shift in time and dimension is the moving from Mia and David’s storyworld (the antecedent, which for these characters is their discourse world) to the counterfactual storyworld of Mia and Sebastian (the consequence) and then back again to the real world of Mia and David. As fictional characters, these ‘people’ are likely to be accepted by audiences as merely constructs from the mind of the author and

34

Marina Lambrou

director, Chazelle, who do not exist outside of the film: They are nothing more than a manifestation of the author’s imagination (see Margolin, 1983). However, to understand Mia and Sebastian and their motivations that lead to the what if? scene, it is helpful to consider characterisation according to Forster’s ([1927] 1985) flat and round characters. Here, ‘Flat characters [. . .] are constructed round a single idea or quality’ and present stereotypical representations of ‘people’, while round characters are ‘more highly organized’ (p. 75) and therefore ‘capable of surprising in a convincing way’ (p. 78). This presents a fairly reductive way of describing fictional characters, who as Rimmon-Kenan (1983, p. 40) points out do not always fit within these categories giving the example of the flat characters in Dickens ‘who are not only felt as very much “alive” but also create the impression of depth’. Ewen (1971, p. 7) offers another useful model of character to consider alongside Forster’s that presents character points existing along a continuum. These character points, which are described in terms of ‘complexity’, ‘development’ and penetration into the ‘inner life’, exist along three separate axes. For example, on character complexity, at one end of the continuum, there exists one trait by which the character is known, and this presents characters who are caricatures. At the other end are more complex or rounded characters so what exists in-between and along the axis are degrees of this aspect of characterisation. The what if? scenario allows audiences a brief glimpse into Mia and Sebastian’s inner life (and possible regrets) as fully formed characters through counterfactual divergence. It could be argued that Mia and Sebastian are round characters because they develop in the course of the plot action and are closer to real people with complex rather than straightforward life choices. In fact, the counterfactual what if? scene is only possible because of the complexity and development of their character, despite the anti-happy-together ending that subverts the conventional romantic trope of Hollywood films. Counterfactual divergence is also possible as narratives are not fixed at the start but, as Phelan (1989, p. 15) argues, are ‘developing wholes’ and arise as characters and plot develop. Because Mia and Sebastian’s motivations have been established early in the film, La La Land’s ending with both having pursued and chosen their dreams over their relationship provides the more satisfying conclusion if audiences accept that characters consider situations and have choices; they are not so much constructs but conscious beings in their storyworld with their own hopes, wishes and desires. In a discussion of ‘counterfactual emotions’ in ‘self-focused autobiographical counterfactuals’, Dannenberg (2008, p. 120) explains that satisfaction is often created through imagining downward counterfactuals (that is, imagining what their life might have been), whereas regret is generated by upward counterfactuals (for an improved version of reality). Mia and Sebastian’s alternative thinking or what social psychologists

La La Land

35

Roese and Olson (1996, p. 200) describe as ‘counterfactual thoughts’ gives consideration to past outcomes that ‘often focus prescriptively on what “should have been”’ . . . ‘that is, on what decisions or actions might have led directly to a more desired outcome and thus “ought” to have been implemented’. Roese and Olson (1996, p. 200) point out that ‘recent research has focused on the beneficial effect of generating such counterfactuals, in that they may often elucidate plans and behaviour that lead to future betterment’. It is difficult to say which counterfactual emotion took place for Mia and Sebastian and whether the what if? scenario led to a future betterment. What is evident is that somewhere in the five-year gap, these characters made choices to pursue their goals even if it meant they could not be together, and the final smile exchanged in the club suggests that both are accepting of their sacrifices.

6. Counterfactual Divergence and the Forked Path Counterfactual stories and counterfactual thinking can also be understood in terms of options that are taken, which allows for some kind of change or diversion to continue along another trajectory or path. Individuals in literature and real life are constantly faced with having to make choices and when one choice is made, all other possibilities are cancelled. Baroni (2016) describes the virtualities of a narrative as being presented through the coexistence of contradictory pathways, or forked paths (Borges ([1944] 2000). Counterfactual divergence as a plot device is an interesting and often challenging way to tell a story. One well-known film where counterfactual divergence is central to the plot is Sliding Doors (1998, directed by Peter Howitt) which tells of the parallel lives of the protagonist, Helen, whose life splits into two parallel universes as a result of a 2-second delay when catching the train. Since the film was released, it has given rise to the expression ‘sliding doors moment’ to describe the experience of a life splitting into two or being faced with two choices (see Lambrou, 2018a for a fuller discussion). Another example of counterfactuality in film is Frank Capra’s 1946 classic It’s a Wonderful Life where the protagonist George Bailey (played by James Stewart) finds himself in a ‘downward counterfactual world where he was never born’. He goes on to experience the nightmarish world that would have existed as a result of his absence until everything is returned to normal. The counterfactual world is ‘disnarrated’ (Prince, 1988; see later discussion) and the what-mighthave-been is dismissed. Dannenberg (2012), in a discussion of counterfactuality, discusses the plots of the television series Doctor Who with specific reference to the 2008 episode ‘Turn Left’, in which the title itself foregrounds the alternative pathway taken. The plot centres on the Doctor’s companion Donna who is tricked and turns right at the traffic lights instead of left. As a result, she finds herself in a counterfactual world

36

Marina Lambrou

where she has never met the Doctor and ‘is not there to step in at a crucial moment; as a result, the Doctor dies and all the catastrophes that threatened earth after he met Donna are no longer averted’ (p. 139). Dannenberg makes the point that the alternative world that is triggered fits within the established fictional history of the series and not ‘real-world history’ and allows for historical employment of events to be fictionalised. The same can be said of La La Land though the characters and plot mirror those of the real world. The idea that separate paths exist or that characters may create an alternative pathway for themselves whenever there is a decision to make (or triggered in some cases by a wish or desire for something to happen) is a concept that Bremond (1966) and Bremond and Cancalon (1980) theorise. Bremond offers a model of story that explains the bifurcations or forking in a narrative at various points along its trajectory to illustrate that stories are more logically rather than temporally ordered. Bremond’s theorisation of narrative structure is a direct reaction to the linear model proposed by structuralists such as Propp ([1928] 1968) as Rimmon-Kenan (1983, p. 22) explains: As with Propp, the function is the basic unit for Bremond. Every three functions combine to form a sequence in which they punctuate three logical stages: possibility (or potentiality), process and outcome. Rather than leading to the next function, as in Propp, each function opens two alternatives, two directions the story can subsequently take. By theorising narrative plots as having numerous alternatives along points in the story, it can be reasoned that characters can also be responsible and influence change and the direction of their lives. In other words, the actualisation of the act (the choice made or the option taken) can either take place or not and the event ‘can follow or not follow its course up to the end which was foreseen’ (Bremond and Cancalon, 1980, p. 387). This process of storytelling opens a network of possibilities to create the virtualities of plot and the consequent counterfactual divergence. Another way to think of the points of departure along two (or even more) paths is best described by the idea of ‘forked paths’, a term that first originated in the short story ‘The Garden of Forking Paths’ by Jorge Luis Borges ([1944] 2000). The short story refers to the chaotic and senseless manuscripts written by one Ts’ui Pên, thought to describe a labyrinth, yet whose ‘novel was incoherent and no one found the labyrinth’ (p. 48). Readers together with the narrator, who is Ts’ui Pên’s ancestor, are led to believe in the existence of a physical labyrinth until the owner of the manuscript, Stephen Albert, makes it clear that the mythical labyrinth is not a conventional labyrinth in physical space such as a garden but instead a labyrinth in time. In other words, the confusing manuscript

La La Land

37

itself is the labyrinth and Borges goes on to explain through the character of Stephen Albert that all texts offer a labyrinth of plot options and therefore multiple possibilities: In all fictions, each time a man meets diverse alternatives, he chooses one and eliminates the others; in the work of the virtually impossible to disentangle Ts’ui Pên, the character chooses—simultaneously—all of them. He creates, thereby, ‘several futures,’ several times, which themselves proliferate and fork. (p. 51) As with Bremond and Cancalon’s (1980, p. 388) network of plot possibilities, the suggestion is that all plots can be non-linear to offer a labyrinth of plot options and multiple possibilities. These alternative choices of what characters may or may not experience are subject to the paths they follow, whether it is determined by the author, narrator or character. With Mia and Sebastian, each were motivated by different goals and desires that directed them separately along different paths that at some point forked in the ellipted five-year gap to create an alternative future. It is worth pointing out that forking paths are present throughout La La Land with several ‘hit-and-miss’ moments in Mia and Sebastian’s relationship. Audiences who watch the film closely will be able to identify that this happens several times between Mia and Sebastian and these occurrences are later ‘repaired’ in the counterfactual sequence. For example, towards the beginning of the film, audiences see Mia walk into a jazz club drawn by the piano music being played by Sebastian. She watches as Sebastian argues with the club’s manager, is fired and leaves. As he storms out of the club, Sebastian bumps into Mia, curses, then walks out. In the what if? montage, instead of ignoring Mia, Sebastian walks directly to Mia, takes her in his arms and kisses her. Their relationship begins from that moment (whereas in the real plot it is much later).

7. The Forked Motorway Path: A Literal, Forked Path The concept of a forked path is embedded in the understanding of a literal and physical branching of a path that offers two routes, as well as the metaphorical or conceptual understanding of being faced with choices and making one of two decisions. In a stroke of genius, Chazelle inserts a literal forked path in La La Land that leads Mia to Sebastian’s jazz club and triggers the what if? counterfactual montage. Audiences see Mia and her husband, David, driving on a motorway to a theatre (or to see a film) but they are stuck in a traffic jam. When they think they would arrive too late they change plans and turn off a forked motorway path to go to dinner instead. It is this decision that inadvertently takes them to Seb’s

38

Marina Lambrou

jazz club. Excerpt 1 from La La Land’s film script describes this moment (with the turn off or point of forking shown in bold): INT. CAR—LATER Mia and David are seated. Still not moving. Mia looks at the clock again: 8:27. DAVID Do you want to just skip it . . .? Turn off here and get dinner? MIA Where are we . . .? Excerpt 1: The forked motorway path (Chazelle, 2013, p. 90) Mia and David’s sudden decision to change plans because of external circumstances (the traffic jam) is a further example of characters making choices as a result of having their own volition (see Lambrou, 2018a). Their choice is to either continue on their original path or turn off at a fork. Interestingly, in the what if? counterfactual scenario of their life together, Sebastian and Mia’s re-imagined life replicates Mia and David’s, as in both plots, each couple is caught in a traffic jam, change plans and turn off the same motorway fork. Other examples of forking paths in La La Land’s plot that hinder Mia and Sebastian’s relationship occur towards the beginning of the film. It is worth noting that the hit-and-miss scenarios faced by Mia and Sebastian are common to the romantic comedy genre, which sees couples undergo a series of hindrances before uniting. A further example is given in Excerpt 2 (showing two examples, a and b, with the point of the forking paths shown in bold): (a) The cars stagger forward. Sebastian’s car almost BUMPS into Mia’s as he changes lanes. He and she see each other for a second—before their cars head their separate ways. With that—we follow Mia . . . (p. 3) (b) EXT. 101 FREEWAY—EARLIER THAT MORNING The same 101 traffic jam we began the movie with. This time, we’re on Sebastian.

La La Land

39

As we saw before, he nearly bumps into Mia’s car. Turns, then cuts out of traffic. Merges onto a more free-flowing freeway . . . (p. 11) Excerpt 2: two further examples of Mia and Sebastian’s forking paths (Chazelle, 2013) The complex plotlines and non-linear story structure all contribute to the film’s tellability, the quality in the story that makes it worth telling in the first place. As Ryan (2005, p. 590) points out: some events make better stories than others because they project a wider variety of forking paths on the narrative map. Even though the story can only follow one path, the understanding of these events involves a consideration of the ‘virtual narratives’ of the unrealised sequences that branch out of the event. In La La Land, the unrealised sequences are shown but then disnarrated— a further dimension of storytelling interlinked with counterfactual divergence and forked paths, which is discussed next.

8. Disnarration The analysis of counterfactual divergence and forking paths in La La Land highlights the proposition that when one path is taken, the other is eliminated—forever. This idea of presenting and then dismissing the virtualities of plot can also be described in terms of the dimension of disnarration. In his article in Style entitled ‘The Disnarrated’, Prince (1988, p. 3) describes disnarration (from the French dénarré) as one of three modes in which a text remains unmentioned, described as categories of the ‘unnarratable’ and ‘non-narratable’. Disnarration is the third of these categories, the other two being: events that cannot be narrated or are not worth narrating; and intentional gaps or ellipses to invoke surprise or suspense. In Lambrou (2018a), I present a fuller discussion of Prince’s disnarration and Warhol’s (2010) taxonomy of the unnarratable in relation to what is narratable, which extends and broadens Prince’s categories. Disnarration, put simply, describes ‘the events that do not happen but, nonetheless, are referred to (in a negative or hypothetical mode) by the narrative text’ (Prince, 1988, p. 2) and in narrative fiction, it can be expressed by the narration, narrator or character. Disnarration according to Prince (1988, p. 2) is constructed through: alethic expressions of impossibility or unrealized possibility, deontic expressions of observed prohibition, epistemic expressions of ignorance, ontologic expressions of nonexistence, purely imagined

40

Marina Lambrou worlds, desired worlds, or intended worlds, unfulfilled expectations, unwarranted beliefs, failed attempts, crushed hopes, suppositions and false calculations, errors and lies, and so forth.

As disnarration developed from Prince’s work in literature, hypothetical scenarios are expressed linguistically through devices such as negation, comparators and modality. In La La Land, the counterfactual what if? scenario clearly shows ‘the events that do not happen but, nonetheless, are referred to (in a negative or hypothetical mode)’ (Prince, 1988 p. 2) which are communicated visually. Audiences are in no doubt of the ‘choices not made, the roads not taken, possibilities not actualized, goals not reached’ (Prince, 1992, p. 36) as Mia and Sebastian are returned to their actual storyworld.

9. Conclusion The unusual narrative structure of La La Land that manipulates logical and spatial dimensions in the plot resulting in counterfactual divergence and virtual narratives creates an original and perhaps unconventional film in the musical romantic comedy genre. Drawing on theories and models in stylistics, narratology and film theory, La La Land appears to allude to the Golden Age of Hollywood musicals with its familiar romantic comedy tropes. However, La La Land goes on to subvert the audience’s expectation of idealised Hollywood romantic comedy by ending with a not-so-happy-ever-after for the film’s protagonists possibly because ambition is the central motivation rather than love. The use of counterfactual storytelling, forked paths and disnarration presents audiences with a plot for what might have been had the what if? outcome happened and is likely to have contributed to the film’s success and six Oscar wins in 2016. In an attempt to explain the ending of his film, the film writer and director Damien Chazelle at the 2016 New York Film Critics Circle Awards referred to the plot of the 1927 silent film 7th Heaven by Frank Borzage: It’s a love story about a man and a woman, Charles Farrell and Janet Gaynor. Charles Farrell goes off to war, dies, Janet Gaynor is informed of his death and holds on to this irrational hope that he’s not dead. Her friends, her family tell her she’s crazy, stop dreaming, be realistic, get on with your life. She insists he’s coming back. There’s an abrupt cut back to the battlefield. Charles Farrell is suddenly alive, inexplicably, and makes his way back home. The last scene, he comes home, they kiss, swell of music, fade to black. [How did this come about? Possibly,] the studio fucked over the director [by insisting on a happy ending]. They reshot something, cobbled it

La La Land

41

together, and that’s what you got. [Possibly, we could be seeing her] descent into total insanity. (Reddit, 2017) Chazelle explains further that Charles Farrell did die and came back from the dead because both possibilities can coexist alongside each other. The reason both can coexist ‘is because of how deeply this woman loves him. The emotion was so deep and profound that the laws of time and reality and physics stop existing’. With this reasoning, it is possible to consider that in a parallel fictional dimension, Mia and Sebastian never separated and are living happily ever after but that in this scenario, both may have compromised their dreams of becoming an actor and jazz club owning pianist. Ultimately, audiences can decide on what ending they prefer and why.

References Altman, R. (1987) The American film musical. London: BFI Publishing. Baroni, R. (2016) ‘Virtualities of plot and dynamics of re-reading’, in Baroni, R. and F. Revaz (eds.) Narrative sequence in contemporary narratologies. Columbus: Ohio State University Press, pp. 87–106. Bordwell, D. and Thompson, K. (2012) Film art: An introduction. New York: McGraw Hill. Borges, J. [1944] (2000) ‘The garden of forking paths’, in Yates, D. A. and J. E. Irby. (eds.) Labyrinths. London: Penguin Books, pp. 44–54. Bremond, C. (1966) ‘La logique des possibilities narratifs’, Communications, 8, pp. 60–76. Bremond, C. and Cancalon, E. D. (1980) ‘The logic of narrative possibilities’, New Literary History, 11 (3), pp. 387–411. Chazelle, D. (2013) ‘La La land’, Available at: www.la-screenwriter.com/wp-con tent/uploads/2017/02/LA-LA-Land.pdf (Accessed 7 September 2017). Dannenberg, H. P. (2008) Coincidence and counterfactuality: Plotting time and space in narrative fiction. Lincoln and London: University of Nebraska Press. Dannenberg, H. P. (2012) ‘Fleshing out the blend: The representation of counterfactuals in alternate history in print, film, and television narratives’, in Schneider, R. and M. Hartner (eds.) Blending and the study of narratives, approaches and applications. Berlin and Boston, MA: Walter de Gruyter, pp. 121–145. Ewen, J. (1971) ‘The theory of character in narrative fiction’, Hasifrut, 3, pp. 1–30. Forster, Edward M. [1927] (1985) Aspects of the novel. San Diego: Harcourt. Hansen, P. K. (2014) ‘Flow-stoppers and frame-breakers: The cognitive complexities of the film musical exemplified by Lars von Trier’s Dancer in the dark (2000)’, in Alber, J. and P. R. Hansen (eds.) Beyond classical narration: Transmedial and unnatural challenges. Berlin and Boston, MA: Walter de Gruyter, pp. 221–237. It’s a Wonderful Life (1946) Frank Capra (director) [DVD]. Available at: Universal Pictures. Lambrou, M. (2015) ‘Discourse stylistics’, in Sotirova, V. (ed.) The Bloomsbury companion to stylistics. London: Bloomsbury, pp. 92–108.

42

Marina Lambrou

Lambrou, M. (forthcominga) Disnarration and the unmentioned in fact and fiction. London: Palgrave Macmillan. Lambrou, M. (forthcomingb) ‘Metalepsis, counterfactuality and being led up the “forked” garden path’, in Sorline, E. (ed.) Stylistic manipulation of the reader in contemporary literature. London: Bloomsbury. Margolin, U. (1983) ‘Characterisation in narrative: Some theoretical prolegomena’, Neophilologus, 67, 1–14. McDonald, Y. J. (2007) Romantic comedy. London: Wallflower. Nørgaard, N., Busse, B. and Montoro, M. (2010) Key terms in stylistics. London: Continuum. Phelan, J. (1989) Reading people, reading plots: Character, progression, and the interpretation of narrative. Chicago: University of Chicago Press. Prince, G. (1988) ‘The disnarrated’, Style, 22 (1), pp. 1–8. Prince, G. (1992) Narratives as theme: Studies in French fiction. London: University of Nebraska Press. Propp, V. [1928] (1968) Morphology of the folktale. Austin: University of Texas Press. Reddit (2017) La La Land: ’Time travel and alternate universes’, Available at: www.reddit.com/r/FanTheories/comments/5wavk7/la_la_land_time_travel_ and_alternate_universes/ (Accessed 7 September 2017). Rimmon-Kenan, S. (1983) Narrative fiction: Contemporary poetics. London: Routledge. Roese, J. N. and Olson, J. M. (1996) ‘Counterfactuals, causal attributions, and the hindsight bias: A conceptual integration’, Journal of Experimental Social Psychology, 32, pp. 197–227. Ryan, M-L. (2005) ‘Tellability’, in Herman, D., Manfred, J. and Ryan, M. (eds.) Routledge encyclopedia of narrative theory. London: Routledge, pp. 589–591. Sliding Doors (1998) Peter Howitt (director) [DVD]. Available at: Paramount Pictures. Toolan, M. (1992) Language, text and context: Essays in stylistics. London: Routledge. Toolan, M. (2014) ‘Stylistics and film’, in Burke, M (ed.) The Routledge handbook of stylistics. London: Routledge, pp. 455–470. Warhol, R. (2010) ‘“What might have been is not what is”: Dickens’s narrative refusals’, Dickens Studies Annual, 41, pp. 45–60.

4

Scribbling Suspense and Surprise Matthew Collins and Mel Evans

1. Introduction In October 1628, Lord Carleton wrote to Queen Henrietta. His topic was not a happy one: ‘I am to trouble your Grace’, his letter begins, ‘with a most Lamentable Relation’. His news concerned the events surrounding the assassination of the Duke of Buckingham, which Carleton relays across a 1,000-word narrative. His conclusion justifies the letter’s composition and length: ‘I thought it my bounden duty howsoever to let your Ma[jes]tie. Have the first intelligence of it, by the hand of Maddam Yo[u] r sorrowfull servant’ (Origin3_025, CEECS). His letter, analysed further later in the chapter, illustrates the communicative affordances of narratives in early modern correspondence. As the primary means of written communication in the period (Markus, 2006, p. 107), letters enabled the rapid transmission of information: A letter could travel from London to Carlisle in four days (Beale, 2005, p. 174). Letters also provided means of authorial control over content, rather than risk distortion through third-party oral transmission, and the recipient could be assured of the authenticity of those contents through the sender’s handwriting and/or signature (see Daybell, 2012). The centrality of letters for communication, alongside their potential for intimacy and association with oral as well as visual modes of language and meaning (Daybell, 2012, p. 6), makes them a rich text type for the exploration of the forms and functions of epistolary narrative, and, in particular, the ongoing narratological debate regarding what constitutes ‘narrativity’. Fludernik posits that not all sequences of events should be classified as ‘narrative’ in epistolary discourse, but rather as ‘reports’. Reclassification is required, in Fludernik’s view, because many of the relayed sequences of events lack the key narrative criterion: experientiality. A legitimate narrative is ‘always both evaluative and steeped in emotional reactions to experience’ (Fludernik, 2004, p. 132). Our study adopts a corpus-based methodology to explore empirically Fludernik’s proposition regarding the ‘narrativity’ of epistolary narratives, focusing on sixteenth- and seventeenth-century correspondence. Using the Corpus of Early English Correspondence Sampler (CEECS),

44

Matthew Collins and Mel Evans

a 0.45-million-word subset of the CEEC that is freely available (copyright restricts distribution of some CEEC letter collections), we seek to understand who, in addition to Lord Carleton, told stories within their correspondence in the early modern period, and how, from a stylistic perspective, this was achieved. To provide a measure of these stories’ possible ‘narrativity’, we focus in particular on the use of two ‘experientially engaging’ (Toolan, 2001, p. 99) narrative features: suspense and surprise.

2. Background The study of early modern letters is a vibrant area of historical and linguistic research that explores the relationship between form and function in a highly context-dependent text type (e.g. Auer et al., 2015; Culpeper, 2011; Daybell and Gordon, 2016; Evans, 2013; Marcus, 2018). For present-day English it is said that letters generally have more ‘oral’ characteristics than other kinds of writing (Biber and Finegan, 1989), typically (although not always) using an intimate, personal register. Historical socio-pragmatic work (e.g. Nurmi et al., 2009; Palander-Collin, 2011) investigates how correspondents’ relationships can shape a letter’s discursive properties; for example, how linguistic forms construct and reflect epistolary relationships (e.g. Barton and Hall, 2000, p. 7), and articulate the correspondents’ subjectivity and intersubjectivity (Adamson, 1994, p. 199; Wright, 1995, p. 152). The study of narrative in correspondence falls within the same frame of inquiry, in that senders’ motivations for, and manner of, relaying a sequence of events encompass interpersonal intentions and effects. However, defining a narrative in epistolary discourse is not straightforward (Altman, 1982, pp. 206–207). Previous studies have found correspondence narratives to be stylistically and pragmatically diverse, requiring a nuanced theorisation of their narrativity. Fludernik, who focuses on the level of experientiality, proposes a report-to-narrative scale which runs from narrative epistles with ‘zero-degree narrativity’ to ‘more pronounced forms’ (2007, p. 249). Applying this definition through a survey of historical correspondence, she concludes that ‘[l]etters between the fifteenth and mid-seventeenth century are not predominantly narrative’ (2007, p. 258). For many letter-writers, a story’s emphasis appears to have been on the reporting (informational) dimension rather than the engaging (interpersonal) dimension. However, epistolary narratives, like Carleton’s, suggest that some letter-writers did produce a highly narrativised discourse. This raises three related lines of inquiry: which letterwriters included ‘more pronounced’ kinds of narratives in their letters; what motivations did they have; and, at the heart of a stylistic investigation, how was ‘higher narrativity’ achieved? Multiple factors likely contribute to the presence of high or low narrativity in early modern correspondence. One aspect concerns the

Scribbling Suspense and Surprise

45

relevance of literacy and writing experience for the production of epistolary discourse with high narrativity; these factors are known to shape epistolary practice across the linguistic spectrum (Rutten and van der Wal, 2014), although the connection with discourse types, such as narrative, is less clear. Local factors, such as the author-recipient relationship, topic and context of composition, may also be relevant. Studies of present-day narrative demonstrate the interpersonal effects of storytelling (e.g. Lampropoulou, 2012), and the same considerations might be considered to apply to the dyadic communicative structure of an early modern letter. Producing a discourse with high narrativity potentially enhances the interpersonal, rather than informational, dimension of the communication. In order to narrow the remit of what constitutes a high- or low-narrativity epistle, our study focuses on two specific features that are thought to enhance experientiality and thereby narrativity: suspense and surprise (Toolan, 2001, p. 99). Hitherto, surprise and suspense have, like many other narratological concepts, attracted the attention of analysts dealing with contemporary narratives, and their relevance to historical writing has been largely overlooked. A further aim of our investigation, therefore, is to interrogate how applicable these concepts are to historical texts. After outlining the methodology, our investigation begins with an examination of the social and stylistic dimensions of narratives in early modern epistolary discourse. We then explore the forms and effects of suspense and surprise in those narratives, combining qualitative analysis with quantitative information of key trends across the dataset, exploring the topics that motivate storytelling and the social profile (e.g. social rank, gender) of the letter-writer, to provide a rounded analysis of a text type supposedly disinclined towards highly narrativised tellings.

3. Methodology Our investigation uses the CEECS (1998). CEECS comprises 450,000 words, and spans the time period 1418–1680. A ‘fairly accurate smallscale copy’ of the full (2.7-million-word) Corpus of Early English Correspondence that also provides similar linguistic results (Nurmi, 1998), the sampler corpus is of a suitable size and scale for our case study exploration of the distribution and types of narrative in early modern correspondence. The lack of a reliable automated system for narrative identification entails that narratives require manual mark-up. This initial analytic stage used a minimal definition of narrative as events containing a complicating action or change of state. Helpfully, letter-writers typically mark points of narrative transition linguistically, which provides evidence for

46

Matthew Collins and Mel Evans

our identification of narrative epistles; for instance, the discourse marker ‘and so’ (our emphasis): but the parishioners would not suffer a Traytor’s corpes to be layed in the earthe where theire parents, wyeffs, chyldren, kynred, maisters, and old neighbors did rest: and so his carcase was retourned to the buryall grounde neere Tyborne, and there I leave yt. (ORIGIN2_032, CEECS) A manual reading of all of the 1,116 letters that make up CEECS found that 81 (7.3%) had narrative content that satisfied these criteria. These form the main dataset, which was analysed to record social and stylistic features, including: author/recipient biographical information (social rank, gender, relationship), topic, letter and narrative length; and the linguistic features associated with suspense and/or surprise. These features, based on present-day narrative analyses, include readerly investment in an outcome, telling delay, descriptive passages, ana/prolepsis and hypothetical narration, as well as lexical items known to entextualise suspense in present-day narratives, e.g. if-clauses, temporal phrases, repetition (Toolan, 2009, pp. 166–177). The extract from William Fleetwood’s letter given earlier illustrates suspense and surprise entextualised through hypotheticals (‘would not’), nominal repetition and telling delay. To ascertain the relevance of these features, we consider their use in three epistolary narratives in detail, before expanding our analysis to consider their quantitative distribution across the corpus. Of course, all narrative analysis entails a somewhat subjective classification; for example, identifying surprise requires an interpretation of the likely extra-linguistic context, which is difficult to confidently establish at such historical distance. Furthermore, as the features we focus on are derived from studies of more recent (literary) texts, our approach likely overlooks alternative expressions of surprise and suspense used in early modern English. As the following discussion illustrates, some recalibration of modern criteria is required for the analysis of historical letters.

4. Narrative Epistles: CEECS 4.1 Distribution of Narrative Letters The initial analysis identified a dataset of 81 narrative letters, 7.3% of the CEECS total. This is a relatively low proportion of correspondence, but as the main means of written communication in the period, letters had to serve multiple functions (e.g. instruction, description). However, the proportion of narrative letters does increase over time (see Table 4.1). This modest rise could be due to changes in the type of extant correspondence. Personal letters, as classified by subject matter (e.g. domestic affairs) and author-recipient relationship, increase in frequency

Scribbling Suspense and Surprise

47

Table 4.1 Proportion of Letters in the Corpus Containing Narratives Century

Corpus

Narratives

Percentage

15th 16th 17th Total

242 304 570 1,116

14 19 48 81

5.8% 6.3% 8.4% 7.3%

over the period, with the linguistic features showing a parallel movement towards positive politeness (e.g. kinship address terms) and a reduction in formulaic language (Nevala, 2004). Narrative may therefore be part of the suite of discursive features associated with the transition from professional to personal letter-writing. Overall, the findings suggest that narrative letters are a small but emergent sub-group of correspondence in the period, largely corroborating Fludernik’s (2007) survey. Topic offers further insight into narrative practices in correspondence, as it indicates the kinds of information that letter-writers were compelled to exchange. We classified topics according to the first major complicating action, alongside any evidence in the opening (abstract) and/or closing (coda) sections. For example, the following is classified as ‘family life’ as following the salutation Thomas Betson writes: ‘seth I come home to London I mette with my lady, your modyr, and God wote she made me right sulleyn chere’ (STONOR_072, CEECS). The topics comprise ten areas: military events, political events, family life, crime, financial affairs, religious activity, social events, death/illness, marriage/love and historical events. Of these, military and political events are most common, comprising almost half of the narratives. Crime is also a favoured topic, found in 14 of 81 letters. It is striking that these are topics considered ‘newsworthy’ in present-day reporting, and are commonly found in newspapers from the seventeenth century onwards (Stephens, 2007, p. 100). Conversely, local and personal events are infrequent: for instance, only three letters provide accounts of marriage. This supports previously identified connections between news reporting and correspondence genres, with the former developing through the epistolary mode (Bazerman, 1999), and differs from present-day conversational narratives, in which the mundane and familiar (qualities more likely associated with local events) are often shared due to their interpersonal significance and potential for high involvement (Norrick, 2000, p. 106). Another property to consider is narrative length, and its relationship to the length of the entire letter. The majority (61%) of narratives are between 100 and 499 words. The shortest is 46 words, in which Nathaniel Bacon reports a (not necessarily unimpressive) seventeenth-century street fight. The longest, at 3,001 words, is by John Shillingford, describing a 1447 council meeting. Viewed as a component of a letter text, the figures suggest that narratives take up the majority of the letters in which they occur (see Table 4.2).

48

Matthew Collins and Mel Evans

Table 4.2 Proportion of Letters Containing Narrative Narrative-to-Letter Length

No. Letters

Percentage Letters

0–25% 25–50% 50–75% 75–100% Total

11 19 28 23 81

14% 23% 35% 28%

Narrative length is plausibly connected to topic. The most ‘newsworthy’ topics, such as crime narratives, typically comprise at least half of the letter in which they occur. However, narrative duration may also be influenced by the context of composition. Writing time was dictated by letterwriter’s health, quality of writing materials and the scheduled departure of the letter’s bearer (see Daybell, 2012). In our dataset, Nehemiah Wharton orients his reader to the place of composition on the seventeenthcentury battlefield, noting, ‘[a]s time permits, I desire to acquaint you with the passages of my pilgrimage’ (WHARTON_003, CEECS). There are thus similarities between narratives in correspondence and conversational narratives, which are also affected by contextual constraints (Norrick, 2000, p. 12). From a ‘narrativity’ perspective, a truncated epistolary narrative (such as Bacon’s 46-word fight account) is likely to contain less interpersonal linguistic features, i.e. lack of evaluation. This would result in its classification as report rather than narrative. As is discussed in section 5.1, the relationship between context and narrative length thus has particular implications for the use of suspense and (to a lesser degree) surprise by early modern letter-writers. 4.2 Social Distribution of Narrative Letters Before considering how the narratives are told, we use demographic information available in CEECS to provide an overview of the users of epistolary narratives. Our approach recognises that discourse practices, including devices like suspense and surprise, are bound-up with social factors such as writing experience, literacy and engagement with various social domains. Whilst information on these aspects is not directly available for each writer in the corpus, insights can be gleaned indirectly by considering traditional sociolinguistic variables such as gender and social rank. The relationship between narrative and gender has been explored in present-day sociolinguistic work (e.g. Lampropoulou, 2012), but is less established for historical material. This is partly due to the paucity of evidence. Women are under-represented in the archives, reflecting different literacy rates. In the early modern period, particularly in the sixteenth century, only upper-ranking women usually received an education that included writing, alongside reading and more domestic skills, such as

Scribbling Suspense and Surprise

49

needlework (Eales, 1998, pp. 34–35; Frye, 2010, p. 6). Consequently, CEECS does not contain an equal gender representation (74% male to 26% female by letter; 79% to 21% by author), which needs to be accounted for in analysis. The CEECS data suggests gender has limited relevance for the use of narrative by early modern letter-writers. Proportionally, male letterwriters are responsible for the majority of narrative epistles (65 letters, 80%), which mirrors the distribution of male/female letter-writers in the CEEC as a whole. The distribution of recipients by gender is similar: 70% male, 22% female (8% unknown/mixed). The figures suggest male and female letter-writers are as equally likely to include narratives. This is somewhat surprising given the longevity of contemporary stereotypes that characterised women as lovers of gossip and tall tales (Fox, 2000, pp. 174–178), as well as differences in gender literacies. That said, one property that may correlate with gender is the distribution of topics: military, politics and crime are over-represented in male-authored letters. One explanation is that male topic preferences are associated with the public sphere, to which they had greater access than their female counterparts (Eales, 1998, p. 22), but further research is needed. Social rank is another factor that informed literacy rates in the period, and here we do find more convincing evidence of a correlation with narrative. The highest-ranked authors (nobility and gentry) are responsible for more than half of the narratives (see Table 4.3). Although the numbers are small, higher-ranked authors include narratives on a greater variety of topics, including those with a private or domestic focus. This could indicate the greater opportunity for them to send correspondence, and therefore address a range of subjects relevant to daily life, compared with lower-ranked authors more likely to compose letters in exceptional circumstances, such as war. Calculating the distribution by individual letters, rather than by author, supports this proposed correlation of topic and rank. Professionals, a CEEC socialrank category that includes army officers, government officials, lawyers, doctors, clergymen and teachers (Nevalainen and Raumolin-Brunberg,

Table 4.3 Authorship of Letters Rank

By Author

By Individual Letter

Nobility Gentry Clergy Merchants Professional Unknown TOTAL

12 (26.7%) 17 (37.8%) 8 2 5 1 45

13 32 9 7 19 (23.4%) 1 81

50

Matthew Collins and Mel Evans

2003, p. 36), author a quarter of texts (23.4%), the greatest contribution of all social ranks. Their frequency arises from the voluminous output of particular professionally ranked male writers, such as Nehemiah Wharton, a military officer. The preferred topics, military, political and criminal events, are event types for which professionals, due to their social roles, would often have first-hand experience. Their prominence in the corpus suggests the importance of epistolary narrative for conveying that experience. The CEECS data suggests that letter-writers from all (literate) walks of life use epistolary narratives, although social experience and literacy may influence the kinds of narrative reported. The interpersonal significance of narratives entails that more local facets of epistolary communication, such as register, may also be significant. Here, register is understood as, in part, the author-recipient relationship (Hallidayan ‘tenor’; see Halliday and Hasan, 1976, p. 22). Register in CEECS can thus be analysed using author-recipient relationship (Nevalainen and Raumolin-Brunberg, 2003, pp. 188–191). For simplicity, 36 letters were classified as ‘personal’ (kin or close friendship) and 38 ‘professional’ (all other relationships).1 Personal letters cover a range of topics, whereas professional relationships focus on the aforementioned ‘newsworthy’ matters: crime, military and political events account for 31 of 36 narratives. This distribution mirrors the findings of conversational narratives in present-day English, in which intimates often share familiar and/or mundane narratives due to their interpersonal significance (Norrick, 2000, pp. 106–107). Professional relationships maintain the preference for the three topics throughout the fifteenth to seventeenth centuries, whereas personal letters broaden their range of topics, suggesting a connection with the diversification of correspondence functions in the period. In sum, narratives are a small but growing feature of epistolary discourse in the early modern period, used with a broadening set of topics by writers from all (literate) social backgrounds. We now consider the extent to which these events were narrativised through the use of suspense and surprise.

5. Suspense 5.1 Qualitative Analysis Our exploration combines a qualitative analysis of suspense and surprise in three letters, to define and illustrate their stylistic operation in context, with a quantitative overview of the associated features across our dataset. Suspense, as defined by Toolan (2009), manifests through multiple linguistic and discursive features, including delayed clauses, temporal adjuncts, prolepsis, readerly investment, telling delay, description, hypothetical narration, repetition and evaluative (stance) markers. Yet the

Scribbling Suspense and Surprise

51

presentation and conceptualisation of suspense in historical narratives are not necessarily equivalent. Aristotle’s Poetics notes the importance of dramatic suspense for audience investment, but the rhetorical figure for suspense, sustentatio, does not occur in ars dictaminis or in relation to narrative until 1527 (Cave, 2009, p. 159), making it difficult to confidently assert the historical provenance of particular stylistic effects. For example, Auerbach argues that in classical literature ‘[descriptive] digressions are not meant to keep the reader in suspense, but rather to relax the tension’ (1974 [1953], p. 36). It is therefore essential to consider examples of potentially ‘suspenseful’ linguistic features in detail and in context. Returning to Carleton’s letter narrating Buckingham’s assassination, multiple features contribute to a suspenseful telling. The opening section of his letter, the salutation and exordium (see Perelman, 1991, pp. 103–104), explicitly frames the ensuing narrative as ‘a most Lamentable Relation’ and proleptically advertises the tale’s narrative point. The formal elements of the letter are thus exploited for their narrative potential (cf. Altman’s ‘epistolarity’, 1982, p. 4): The exordium overlaps in functional terms with Labov and Waletzky’s (1967) narrative abstract. The salutation also grounds suspense in evaluation and engagement, by drawing explicitly on the letter’s dyadic dimension. Elsewhere, Carleton also exploits the dyadic communicative structure for suspenseful effect. For example, the direct address: ‘Maddam, you may easily guesse what outcryes were then made’, and self-reference: ‘ah poore Ladies, such was their screechings [. . .] that I never in my Life heard the like before’ delay the telling of the narrative, whilst promoting engagement. Evaluation, stance and suspense are linked by virtue of their interpersonal effect. These features are found in other narrative letters in CEECS, and appear to be characteristic of suspenseful early modern epistolary narratives. Carleton’s letter also exploits the syntactical arrangement of the narrative action for the purpose of suspenseful delay: This day betwixt nine and ten of the clock in the morning, the Duke of Buckingham then comming out of a Parlor, into a Hall, to goe to his coach and soe to the King, (who was four miles of[f]) having about him diverse Lords, Colonells, and Captains, & many of his owne Servants, was by one Felton (once a Lieutenant of this our Army) slaine at one blow, with a dagger-knife. Here the main action contained in the final, main clause is delayed by a series of preceding subordinate clauses and orientational adjuncts of time, place and participants, some containing further subordination in parentheses. Suspense and subordination combine to create a narratorial antipathy towards the ‘villaine’, Felton. The reader must first engage with the activities of the Duke and his men, before encountering the assassin who is relegated to a passive prepositional phrase at the periphery of the

52

Matthew Collins and Mel Evans

complicating action. Passivisation ensures that the reader’s attention is on Buckingham. With reference to modern narratives, Toolan cautions against identifying syntactical delay with narrative suspense; the latter ‘is considerably more experiential, and emotionally as well as intellectually engaging’ (2009, p. 166). But in early modern epistolary narratives, we suggest that the compressed form of correspondence means that syntactical and narrative suspension often coincide. Carleton’s (combined) use of local/syntactical and narrative/discourse suspense indicates a calculatedness in the manner of telling, and parallels might be drawn with styles of suspense found in more literary contexts. This contrasts with some other narratives in CEECS, particularly those depicting military events, which are often summarised; a distinction that may arise from differing communicative objectives (e.g. sharing information), and context of composition (e.g. the battlefield). However, 41% of the letters contain syntactic postponement (i.e. a subclause + main clause construction), and two-thirds of these contain narrative suspense. This syntactic-narrative co-occurrence becomes more pronounced over time (rising from 7% to 31% between the fifteenth and seventeenth centuries), although further research is required to see to what extent this is due to the adoption of Latinate and hypotactic syntax in English prose styles more generally (see Adamson, 1999). Suspense, manifested in delay through description, evaluation and syntactical postponement, constructs an experiential narrative. The point of this telling is reliving the horror of the event: Queen Henrietta can still experience suspense, even if she already knows the outcome (Hoeken and van Vilet, 2000; van Peer, 2007). That this assassination provided content for other correspondence (see Fludernik, 2004, pp. 134–135) and Dumas’s The Three Musketeers is further indicative of its tellable qualities. Carleton ends, ‘Maddam, this is the truth, the whole truth, and nothing but the truth’, a coda that casts his narrative as testimony; perhaps an attempt to mitigate the moments of high narrativity in what contextually might be expected to be a more neutral report. 5.2 Quantitative Analysis The exploration of Carleton’s letter reveals syntactic, lexical and discursive features that contribute to a suspenseful telling. When these features are considered across the dataset, it is possible to extrapolate some quantitative trends relating to the distribution of suspense. Using the features identified earlier, we categorised all narratives in the dataset as to whether they do, or do not, include one or more suspenseful elements. On this calculation, 48% (39/81 letters) contain suspense, suggesting it was an established practice for epistolary narratives. Viewed textually, the results suggest that longer letters are more likely to include suspenseful features: Only a quarter of letters of fewer than 100 words (2/8 letters) included

Scribbling Suspense and Surprise

53

suspenseful features, compared with the two letters in the dataset totalling more than 1,500 words. However, it is not clear whether this correlation arises due to suspense’s requirement of delay (and therefore textual space), or from the greater probability of suspense occurring because of the greater quantity of text (its verbosity). It is plausible that, in longer narratives, the writer is devoting enough effort to their narrative telling to elaborate beyond the complicating action baseline. There is no strong evidence that topic influences the inclusion of suspense. It occurs in half of the best-represented topics (crime, military and political events), mirroring the overall distribution. Whilst male and female letter-writers include suspenseful elements fairly equally, the sex of recipient may have a small effect, with letter-writers more likely to include suspense in letters sent to a recipient of the same sex. This could suggest that narrative practice was sensitive to social differences. It may be that the degree of involvement required on behalf of the writer and recipient, and the (self-)belief in the story’s narrativity, shows sensitivity to contemporary gender hierarchies. Further research is needed to confirm this interpretation. When categorised by social rank, professional-authored letters include suspenseful elements more frequently than other groups, 74% (14/19 letters), possibly due to their ‘newsworthy’ topic choices. Among gentry and nobility narratives, the proportion is lower at around 40%. This suggests that the use of narrative suspense is not strongly associated with (epistolary) literacy, with other factors such as topic and relationship-type plausibly as, or more, relevant. Thus, the greater range of topics covered by gentry and nobility writers may dilute the potential opportunities for a suspenseful telling. As the analysis of Carleton’s letter demonstrates, high-ranking letter-writers could make use of a variety of features for a suspenseful narrative telling, when appropriate.

6. Surprise 6.1 Qualitative Analysis To illustrate the formal properties of surprise, we turn to a series of letters in CEECS written by a professionally ranked writer, Nehemiah Wharton. Although the CEECS data is one-sided (i.e. only letters from Nehemiah to his recipients), it reveals how narrative was used to develop relationships, and, correspondingly, how the manner of telling may change over time. For Wharton’s letters, the later narratives in the sequence include more detail, personal experiences and evaluation than in his earlier letters, perhaps in response to his recipient’s (encouraging) responses. Notably, surprise (as well as suspense) shows a parallel rise, suggesting a link between high narrativity features and reduced social distance. Wharton’s letters use features associated with a (military) report, including summary, lists, numbers and a diarised structure. However,

54

Matthew Collins and Mel Evans

these features do not completely inhibit the use of linguistic and discursive features for the purpose of surprise. One such feature found in his succinct narrative is lexical deviation. When giving an account of an unexpected attack by the enemy in his first letter, dated 26 August 1642, Wharton describes his own men as ‘cannybals’ (cf. Carelton’s ‘villaine’). This word was conventionally associated with voyager narratives and typically used to describe outsiders. It would presumably strike the reader as a strange description in a military report of one’s fellow soldiers. Of course, lexical deviation presupposes norms, highlighting the importance of the extra-linguistic context for the creation of surprise. The in situ basis of Wharton’s correspondence results in inaccuracies, which we also consider to be a form of surprise. In one letter, Wharton corrects a previous narrative, adopting a ‘statement + “but not”’ structure. The lexico-syntactic parallelism creates mini-suspense repetitions, whereby narrative unfolds and is then withdrawn: e.g. ‘First, I wrote that the Earle of Northampton and the Lord Craven were with the Prince, but they were not’ and ‘The third was, that Colonell Sands was dead; but both he and Captain Sands, though both wounded, yet are still livinge’ (WHARTON_008, CEECS). These corrections, which include the mis-reported deaths of individuals known to writer and recipient, would presumably have had a surprising impact with significant emotional resonance. Finally, as the previous example indicates, Wharton’s letters also include suspense. However, unlike Carleton’s controlled telling of his story, with its already completed real-world action, Wharton’s narrative suspense often arises as a consequence of the unresolved context of the letters’ composition. For example, in his letter of 26 August 1642, he makes the tantilising observation that ‘Fryday morninge our drummes beat for our regiment to march away, but wheither I know not’ (WHARTON_008, CEECS). The resolution of the cliffhanger depends, for the recipient, on the arrival of Wharton’s next letter. 6.2 Quantitative Analysis Surprise is an infrequent narrative feature, identified in only 20 of the letters (24%). This finding is itself somewhat surprising, as newsworthiness, and by implication surprise, is a plausible motive for letter-writing. Its scarcity possibly arises from the nature of epistolary communication. Surprise is a largely extra-textual phenomenon, requiring deviation from an expected event sequence. Whilst general schematic sequences, such as going to a restaurant (Schank and Abelson, 1977), would be widely recognisable for narratees, Wright emphasises that letters are ‘built on a large body of mutual background knowledge, of shared presuppositions [.  .  .] elements of narrative [.  .  .] may be allusive, possibly vague and often opaque to the epistolary eavesdropper’ (1989, pp. 551–552). This

Scribbling Suspense and Surprise

55

inevitably complicates the picture for historical narratives, with recipients’ schemas only partially retrievable due to temporal distance. This entails that our classification of surprising events may over- or (more likely) underestimate the feature. Whilst suspense is more likely to occur in longer narratives, surprise occurs most frequently in letters of mid-range length (500–1,500 words), and is relatively uncommon (16%) in shorter narratives (100–499 words). This plausibly reflects the intra- and extra-textual context of the two narrative devices. For a suspenseful telling, the letter-writer is required to extend the duration of their narrative. For a surprising telling, the letterwriter needs only the same reference points as their recipient, permitting a much briefer narrative. Moreover, despite its infrequency, surprise occurs in all topic groups, excepting financial affairs and historical events. Differences between suspense and surprise are evident in the social group analysis. Unlike suspense, gender distribution of surprise is imbalanced. Proportionally, male writers use surprise more frequently (18/65 letters) than women (2/16 letters). It seems unlikely that this difference is connected to literacy levels, given the hypothesised extra-textual basis of surprise, and instead may reflect the kinds of narrative male and female letter-writers were able to tell. Surprising military and political events, as narrated by Wharton, are unlikely subjects for early modern women’s letters. This interpretation gains support from the social rank distribution. Whilst all ranks, excepting merchants, use surprise, professional writers are the most likely to include the feature (9/19 letters). As illustrated in Wharton’s letter, professional writers largely narrate ongoing military, political and criminal events: Their very nature is unscripted and unpredictable.

7. Suspense and Surprise The CEECS data suggests that surprise tends to co-occur with suspense. In this final (brief) analysis we show how surprise and suspense combine to enhance tellability in a letter from Sir Robert Cecil to Robert Devereux, Earl of Essex, resulting in highly narrativised epistolary discourse. The letter describes an infamous exchange in which the Polish ambassador insulted Queen Elizabeth I, who, in ex tempore Latin, put him firmly in his place, much to the delight of on-looking councillors. The narrative commences by appropriating the epistolary salutation and exordium as narrative abstract. Cecil promises his reader an interesting tale will follow, with the supension metalingusitically drawing attention to the delayed telling: My Lord, after this longe preface, lett me take a little tyme for you with that, which I assure you would have pleased you, if you had ben a beholder. (ORIGIN3_003, CEECS)

56

Matthew Collins and Mel Evans

Descriptive pause is a tool for suspense (Toolan, 2009, p. 167), but can also enhance subsequent narrative surprise. Cecil sets up the narrative twist by providing a lengthy and positive description of the Polish ambassador, including his attire, his proximity to the Queen ‘tenne yardes of[f]’ and his character: ‘such a gallant countenance, as in my lyfe I never beheld’. The extended positive description aligns narratorial and recipient points of view. The latter therefore shares in the narrator’s original surprise when the ambassador’s unprecedented criticism of Elizabeth is finally revealed, and the queen gives her Latin response. Narratorial stance, and its association with suspense and surprise effects, foregrounds the experiential dimension of the event. The relationship between writer and recipient likely informs the use of suspense and surprise. The recipient, Essex, regularly disobeyed court protocols (Hammer, 2004), and, to emphasise the didactic potential of his narrative, Cecil may have chosen these stylistic strategies to enhance the letter’s rhetorical persuasiveness. At the narrative’s conclusion, Cecil requests Essex to ‘take notice that you were pleased to heare of her wise and eloquent aunswer’. The evalution foregrounds the narrative event Cecil believes warrants the recipient’s attention, and perhaps, his subsequent reflection; that is, the majesty of their sovereign. This example reminds us that effects such as surprise and suspense rely on social and textual norms, and thus necessitate the alignment of writer/reader expectations if they are to have the intended interpersonal impact.

8. Conclusion Our analysis contributes to the ongoing debate surrounding the definition of narrative. Narrative features such as surprise and suspense offer a means of identifying narrativity by attending to the interpersonal and tellable qualities of a story. Our examination has found that narrative is a distinct discourse function in early modern correspondence, and that its use and form are informed and shaped by social context and letterwriting practices. For those writers who include narrative, it plays an important role in their correspondence, evident in features of high narrativity such as suspense and, to a lesser extent, surprise. The writers use a range of linguistic and discursive strategies, and their letters reveal an acute awareness of the need to shape a narrative’s telling for their context and audience. In some cases, the narratives are self-contained and show sophisticated forethought in their construction, whereas other letter-writers’ narratives are swept along with their context of composition. Our study shows, too, that the epistolary context is interwoven with the narrative form and function; for instance, the use of salutations in service of a narrative abstract, or the placement of direct address to convey narratorial stance and point of view.

Scribbling Suspense and Surprise

57

The analysis of CEECS confirms that narratives are infrequent in early modern correspondence. However, around half of the narratives contain suspenseful elements, and at least a quarter contain facets of surprise. The presence of these discourse features, with their experiential effects, suggests a high level of narrativity. Thus, the infrequency of narratives in early modern correspondence perhaps underplays the linguistically elaborate and multi-faceted functions of those narratives that are present. Whilst some epistolary narratives are succinct reports, with few features associated with high narrativity, others show a richness and complexity comparable with that of narratives in other text types.

Note 1. Seven relationships could not be classified due to lack of evidence.

References Adamson, S. (1994) ‘From empathetic deixis to empathetic narrative: Stylisation and (de-) subjectivisation as processes of language change’, Transactions of the Philological Society, 92 (1), pp. 55–88. Adamson, S. (1999) ‘The Literary Language’, in Lass, R. (ed.) The Cambridge history of the English language, 1476–1776. Cambridge: Cambridge University Press, pp. 539–653. Altman, J. (1982) Epistolarity. Columbus: Ohio State University Press. Auer, A., Schreier, D. and Watts, R. (eds.) (2015) Letter writing and language change. Cambridge: Cambridge University Press. Auerbach, E. ([1953] (1974) Mimesis: The representation of reality in Western literature. Trans. Willard R. Trask. Princeton: Princeton University Press. Barton, D. and Hall, N. (2000) Letter writing as a social practice. Amsterdam and Philadelphia: John Benjamins. Bazerman, C. (1999) ‘Letters and the social grounding of differentiated genres’, in Barton, D. and Hall, N. (eds.) Letter writing as social practice. Philadelphia: John Benjamins, pp. 15–29. Beale, P. (2005) England’s mail: Two millenia of letter writing. Stroud: Tempus Publishing. Biber, D. and Finegan, E. (1989) ‘Styles of stance in English: Lexical and grammatical marking of evidentiality and affect’, Text, 9 (1), pp. 93–124. Cave, T. (2009) Retrospectives: Essays in literature, poetics and cultural history. London: Legenda. Corpus of early English correspondence. 1998. Compiled by T. Nevalainen, H. Raumolin-Brunberg, J. Keränen, M. Nevala, A. Nurmi, and M. Palander-Collin at the Department of Modern Languages, University of Helsinki. Available at: http://purl.ox.ac.uk/ota/2461 (Accessed 16 August 2017). Culpeper, J. (ed.) (2011) Historical sociopragmatics. Amsterdam: John Benjamins. Daybell, J. (2012) The material letter in early modern England. Houndmills, Basingstoke and Hampshire: Palgrave Macmillan. Daybell, J. and Gordon, A. (2016) Cultures of correspondence in early modern Britain. Philadelphia: University of Pennsylvania Press.

58

Matthew Collins and Mel Evans

Eales, J. (1998) Women in early modern England, 1500–1700. London: Taylor & Francis. Evans, M. (2013) ‘The language of Queen Elizabeth I: A sociolinguistic perspective on royal style and identity’, in Transactions of the philological society monograph series, 45, Chichester: Wiley-Blackwell. Fludernik, M. (2004) ‘Letters and chronicles: How narrative are they?’ in Rossholm, G. (ed.) Essays on fiction and perspective. Frankfurt am Main: Lang, pp. 129–154. Fludernik M. (2007) ‘Letters as narrative: Narrative pattern and episode structure in early letters, 1400 to 1650’, in Fitzmaurice S.M. and Taavitsainen, I. (eds.) Methods in historical pragmatics. Berlin: Mouton de Gruyter, pp. 241–266. Fox, A. (2000) Oral and literate culture in England, 1500–1700. Oxford: Clarendon Press. Frye, S. (2010) Pens and needles: Women’s textualities in early modern England. Philadelphia and Oxford: University of Pennsylvania Press. Halliday, M. and Hasan, R. (1976) Cohesion in English. London: Longman. Hammer, P. (2004) ‘Devereux, Robert, second earl of Essex (1565–1601), soldier and politician’, Oxford Dictionary of National Biography. Available at: www. oxforddnb.com/view/10.1093/ref:odnb/9780198614128.001.0001/odnb9780198614128-e-7565. (Access 30 November 2017). Hoeken, H. and van Vilet, M. (2000) ‘Suspense, curiosity, and surprise: How discourse structure influences the affective and cognitive processing of a story’, Poetics, 27 (4), pp. 277–286. Labov, W. and Waletzky, J. (1967) ‘Narrative analysis: Oral versions of personal experience’, in Helm, J. (ed.) Essays on the verbal and visual arts. Seattle: University of Washington Press. Lampropoulou, S. (2012) Direct speech, self-presentation and communities of practice. London: Continuum. Marcus, I. (2018) The linguistics of spoken communication in early modern English writing: Exploring Bess of Hardwick’s manuscript letters. Basingstoke: Palgrave Macmillan. Markus, M. (2006) ‘Abbreviations in early modern English correspondence’, in Dossena, M. and Fitzmaurice, S. (eds.) Business and official correspondence: Historical investigations. Bern: Peter Lang, pp. 107–128. Nevala, M. (2004) Address in early English correspondence: Its forms and sociopragmatic functions. Helsinki: Société Néophilologique. Nevalainen, T. and Raumolin-Brunberg, H. (2003) Historical sociolinguistics: Language change in Tudor and Stuart England. London: Longman. Norrick, N. (2000) Conversational narrative: Storytelling in everyday talk. Amsterdam and Philadelphia: John Benjamins. Nurmi, A. (ed.) (1998) Manual for the corpus of early English correspondence sampler CEECS. Department of Modern Languages, University of Helsinki. Available at: http://khnt.hit.uib.no/icame/manuals/ceecs/ (Accessed 28 March 2018). Nurmi, A., Nevala, M. and Palander-Collin, M. (2009) The language of daily life in England (1400–1800). Amsterdam and Philadelphia: John Benjamins. Palander-Collin, M. (2011) ‘Variation and change in patterns of self-reference in early English correspondence’, in Culpeper, J. (ed.) Historical sociopragmatics. Amsterdam: John Benjamins, pp. 83–108.

Scribbling Suspense and Surprise

59

Perelman, L. (1991) ‘The medieval art of letter writing: Rhetoric as institutional expression’, in Bazerman, C. and Paradis, J. (eds.) Textual dynamics of the professions: Historical and contemporary studies of writing in academic and other professional communities. Madison: University of Wisconsin Press, pp. 97–119. Rutten, G. and van der Wal, M. (2014) Letters as loot: A sociolinguistic approach to seventeenth- and eighteenth-century Dutch. Amsterdam: John Benjamins. Schank, R. C. and Abelson, R. P. (1977) Scripts, plans, goals and understanding: An inquiry into human knowledge structures. Hillsdale, NJ: Lawrence Erlbaum. Stephens, M. (2007) A history of news. Oxford: Oxford University Press. Toolan, M. (2001) Narrative: A critical linguistic introduction. London: Routledge. Toolan, M. (2009) Narrative progression in the short story. Amsterdam: John Benjamins. van Peer, W. (2007) ‘Towards a new narratology: An extended review of Psychonarratology’, Language and Literature, 16 (2), pp. 214–224. Wright, S. (1989) ‘Private language made public: The language of letters as literature’, Poetics, 18 (1989), pp. 549–578. Wright, S. (1995) ‘Subjectivity and experiential syntax’, in Stein, D. and Wright, S. (eds.) Subjectivity and subjectivisation: Linguistic perspectives. Cambridge: Cambridge University Press.

Section 2

Patterns and Predictions

5

Investigating Syntactic Simplicity in Popular Fiction A Corpus Stylistics Approach Rocío Montoro

1. Introduction Michael Toolan’s contribution to the study of language has been as plentiful as it has been influential. His diverse research interests have inspired scholars concerned with the study of narrative and stylistics among many other issues. Corpus stylistics has, of late, informed much of Toolan’s research, especially the way in which patterns can be exploited in the description of narrative progression in the short story (Toolan, 2009) and to make sense of the narrative text (Toolan, 2016). Toolan’s discussion of the explanatory and illustrative power of linguistic patterns and the methodological advantages of corpus stylistics has informed my own research as I explain later. In this chapter I look into a claim made about popular fiction concerning its apparent syntactic simplicity. Literary critics and readers (Montoro, 2012) often characterise popular literature as ‘easy to read’ and this easiness tends to be equated with a kind of syntactic simplicity which separates genre fiction from Literature with a capital ‘L’. Describing popular fiction in a comprehensive way falls outside the remit of this chapter, as does accounting fully for the notion of ‘complexity’ which, of course, encompasses many aspects besides syntactic components.1 In this chapter, I confine my analysis to a few specific linguistic features which seem to explain and support the claims from critics and readers with regard to the simplicity of popular writing when compared to Literature. As I explain below, a corpus stylistics methodology enables me to account for the notion of syntactic simplicity in a rigorous and thorough fashion. Literary criticism has approached the distinction of popular vs. serious fiction in a variety of ways. Some literary critics (McCracken, 1998), for instance, view popular novels in terms of their large readership, whereas others (Gelder, 2004) focus on the dichotomy between popular writing and Literature by situating the canon from the twentieth and twentyfirst centuries in the latter category, leaving the rest outside such a privileged club. Both approaches, though sound in their own terms, fall short of providing testable and replicable criteria as to how to classify works

64

Rocío Montoro

within either category. Other authors (Ramet, 1999; Moretti, 2010) have addressed the popular vs. serious fiction distinction on the basis of linguistic differences but their attempts do not always result in a rigorous account of how each text type differs. Moretti (2010, p. 1), for instance, describes (literary) prose as prototypically hypotactic which, in turn, entails a hierarchical clausal structure and hence complex meaning: Prose is not a gift; it’s work: [. . .] hypotaxis is not only laborious [. . .] but truly productive: the outcome is usually more than the sum of its parts because subordination establishes a hierarchy among clauses, meaning becomes articulated, aspects emerge that didn’t exist before. That’s how complexity comes into being. Although I do not necessarily disagree with the possible effects of subordination described here, I find such description methodologically lax and hermeneutically poor. In what follows, I further previous research on the matter (McIntyre and Montoro, 2012; Montoro and McIntyre, 2019, fc.) by focusing on the structure of just one grammatical category which can account for the simplicity often associated with popular fiction in an insightful manner: the Noun Phrase (NP).2 Needless to say, this does not imply that my results exclude additional linguistic components from also characterising popular and serious writing differently. My focus on the NP emanates from the results obtained by Montoro and McIntyre (2019, fc). Based on Moretti (2010), among others, in this earlier research, we (McIntyre and Montoro, 2012; Montoro and McIntyre, 2019,fc) investigated the claims that (a) subordination projects complexity and (b) both phenomena are prototypically found in Literature. Montoro and McIntyre (2019, fc) argue that some of the claims made by literary critics tend to be too imprecise to be testable (for example, that ‘hypotaxis is [. . .] truly productive’) or made on the basis of little evidence. This, of course, does not imply that these critics draw erroneous conclusions but simply that they do not draw upon replicable and/or falsifiable tenets. As I explain in the next section, Montoro and McIntyre (2019, fc) look at subordinating conjunctions in popular and serious fiction and conclude that, on the basis of that feature alone, the latter is not more syntactically complex than the former (in contrast to what Moretti (2010) seemed to suggest earlier). However, our research also highlights that genre fiction does indeed display a kind of syntactic simplicity, albeit one that works at phrasal rather than at clausal level, more specifically (though not exclusively) in relation to one component of the NP, the post-modifying of-phrase. I therefore develop our (earlier) research by looking in more detail at the NP. Specifically, I consider whether one of the distinguishing features of genre fiction which prompts critics and readers to view it as somehow less structurally developed is the underoccurrence of nominal post-modification by means of the of-phrase. As

Syntactic Simplicity in Popular Fiction

65

stated earlier, the claims I make here exclusively apply to the grammatical features I focus on and do not preclude others not dealt with in this chapter from equally having some bearing on the apparent syntactic simplicity of popular fiction. For ease of reference and due to space constraints, I use the following terms interchangeably: serious, highbrow fiction and Literature, on the one hand, and popular, lowbrow and genre fiction, on the other, though I am aware that these terms are not fully synonymous.

2. Previous Research Montoro and McIntyre (2019, fc) use a corpus stylistics methodology to test some of the claims emerging from the literary critical field with regard to the dichotomy of popular vs. serious fiction. They circumvent the circularity problem by using two corpora categorised as literary and popular by experts who have applied their own independent set of criteria. The first corpus is the Lancaster Speech, Writing and Thought presentation corpus (SW&TP) (Semino and Short, 2004), compiled at Lancaster University (available at the Oxford Text Archive website).3 The second is The Chick Lit Corpus (Montoro, 2012),4 inclusive of six Chick Lit novels5 amounting to a total of about 500,000 words. Montoro and McIntyre (2019, fc) first analysed the prose fiction section of the SW&TP by tagging it for part-of-speech (POS) information using the Constituent Likelihood Automatic Word-Tagging System (CLAWS) tagger in the software Wmatrix (Rayson, 2009). The target corpus was then compared to a reference corpus, Wmatrix’s one-million-word sample of the written section of the British National Corpus (BNC Sampler Written) and smaller subsections thereof (mainly the BNC Sampler Written Imaginative sub-corpus).6 Wmatrix uses, among others, the log-likelihood test (Dunning, 1993) which measures the likelihood of the results obtained being due to chance alone. The higher the log-likelihood value, the higher the statistical significance of the results, that is, the more likely it is that the results are not accidental. Wmatrix (Rayson, 2009) uses the following cut-off values: 95th percentile; 5% level; p < 0.05; critical value = 3.84 99th percentile; 1% level; p < 0.01; critical value = 6.63 99.9th percentile; 0.1% level; p < 0.001; critical value = 10.83 99.99th percentile; 0.01% level; p < 0.0001; critical value = 15.13 (Rayson, 2009) Thus, anything over the critical value of 3.84 is statistically significant, so any conclusion drawn on that basis is backed up by statistical evidence. Montoro and McIntyre (2019, fc) found that both the serious and the popular prose fiction sections of the SW&TP corpus overuse subordinating conjunctions (POS tag CS) when compared to the BNC Sampler Written (inclusive of fiction):

66

Rocío Montoro Serious fiction compared to BNC Sampler Written: CS217 CS CSA8 CS31

= = = =

+26.14 +14.76 +09.38 +05.08

Popular fiction compared to BNC SamplerWritten: CS = CS32 = CS31 =

+13.86 +06.56 +06.27

So, on the basis of this POS tag alone, that is, subordinating conjunctions, there is no evidence that popular fiction is less syntactically complex than Literature. Besides subsequently looking in more detail into these subordinators’ functionality, Montoro and McIntyre (2019, fc) also highlight that the same overuse of subordinating conjunctions is found in the Chick Lit Corpus (compared against the BNC Sampler Written Imaginative): Chick Lit Corpus compared to BNC Sampler Written Imaginative: CS CS21 CSW9 CST10 CS31

= = = = =

+37.56 +35.92 +21.00 +17.41 +06.65

These values confirm that should syntactical simplicity be prototypical of popular forms, then it cannot be explained by the under-representation of subordinating conjunctions.11 Having said that, Montoro and McIntyre (2019, fc) consider other keyness results which not only illustrate that simplicity does indeed feature in genre fiction but also clarify the level at which it appears to work. They consider the under- rather than the overuse of certain POS and focus on three under-represented categories, namely prepositions, adjectives and the wh-determiner (whose), all prototypical components of the NP; for the purposes of this chapter, I only refer to the first of those three: Prepositions: Chick Lit Corpus compared to BNC Sampler Written Imaginative: IO (of, as preposition) II (general preposition)

= =

−264.68 −047.65

The IO (of, as preposition) is in the top ten most significantly underrepresented categories in the Chick Lit novels. On closer inspection,

Syntactic Simplicity in Popular Fiction

67

concordances show that the IO tags are mainly (though not exclusively as I develop later) used to introduce post-modifying prepositional phrases in the NP. This underuse therefore suggests that a lack of nominal postmodification could explain the stylistic simplicity prototypically seen as characteristic of popular fiction, at least with regard to Chick Lit novels. In the remainder of this chapter, I expand on these results by analysing the under-representation of certain POS categories in more detail. I look at two corpora of popular literature amounting to a total of more than one and a half million words, the Chick Lit Corpus, as described earlier, and the Modern Vampire Corpus12 (MVC) (Montoro, 2015), a corpus of one million words inclusive of eight novels by two different authors, Anne Rice and Charlaine Harris.13 The MVC is also tagged for partsof-speech using Wmatrix (Rayson, 2009) and compared to the BNC Written Imaginative sub-corpus to ascertain, as done before, whether subordinating conjunctions are over-represented or not. Other key POS are also analysed, namely the Noun category, following which I focus on the occurrences and functions of the post-modifying of-phrase as suggested by the results in Montoro and McIntyre (2019, fc). I then run the tagged versions of both corpora through AntConc (Anthony, 2014), an additional piece of software which allows me to discriminate the various functions of the of-phrase in more sophisticated ways than is possible with Wmatrix (Rayson, 2009). This two-stage approach enabled me to reassess the original results discussed in Montoro and McIntyre (2019, fc) independently but equally methodically.

3. Results and Discussion Montoro and McIntyre (2019, fc) demonstrate that claims regarding the syntactic simplicity of genre fiction due to the underuse of subordinating conjunctions simply do not stand; yet, they also conclude that further investigation can provide more evidence on how that simplicity is actually attained. Thus, I first replicate Montoro and McIntyre’s (2019, fc) analysis to test whether the overuse of subordinating conjunctions they document is also prevalent in a different corpus, namely the MVC: Subordinating conjunctions: MVC compared to BNC Sampler Written Imaginative: CS21 CSA CST CS

= = = =

+140.89 +075.41 +047.56 +015.98

As this analysis demonstrates, were syntactic simplicity to be singled out as a defining feature of the novels contained in the MVC, then it cannot

68

Rocío Montoro

be said to be effected by the under-representation of subordinating conjunctions. These results endorse those discussed so far and also underscore the usefulness of a corpus stylistics methodology to test literary critical claims. The next step, as also implemented before, is to consider whether the statistically significant under-representation of any other POS tag could shed light on the characteristics of genre fiction. In this respect, the noun category (NN) stands out suggesting that the NP structure requires further attention: Nouns: Chick Lit Corpus compared to BNC Sampler Written Imaginative: NN114 = NN215 = NN16 =

−400.91 −057.53 −032.30

Nouns: MVC compared to BNC Sampler Written Imaginative: NN1 NN

= =

−247.20 −151.70

The functional role of the NP, its diachronic evolution in English as well as its structural composition are amply analysed topics (Biber and Clark, 2002; Biber, 2003; Leech et al., 2009; Biber and Gray, 2011). In what follows, I focus on three phenomena which affect the NP (besides other linguistic units) and which appear to characterise twentieth-century English: ‘colloquialisation’ and ‘densification’ as defined by Leech et al. (2009), and ‘compressed expression’ as described in Biber and Clark (2002). I argue that popular writing, by failing to follow the densification effect generally observed in twentieth-century English, appears more aligned with colloquial, less informational text types. Syntactic simplicity, therefore, could be accounted for by popular forms only partially following twentieth-century English syntactic trends. Leech et al. (2009) examine the grammatical changes in written standard English in the twentieth century and conclude that twentieth-century English has undergone a series of processes they label grammaticalisation, colloquialisation, densification, Americanisation and other processes (e.g. prescription and language planning) (Leech et al., 2009, p. 237). Colloquialisation is defined as a ‘shift to a more speech-like style’ (Leech et al., 2009, p. 239) and it explains certain changes in the frequency of grammatical forms such as the ‘continuing rise of certain semi-modals [. . .] and the progressive’ and ‘in the decline of wh-relatives’ (Leech et al., 2009, p. 239).17 Thus, although, as Leech et al. (2009, p. 245) argue, this colloquialisation hypothesis ‘cannot be adopted wholesale’, ‘the colloquialisation of written English is a real linguistic trend’ (2009, p. 249). Conversely, Leech et al. (2009) also contend that, in twentieth-century

Syntactic Simplicity in Popular Fiction

69

English, densification or ‘compacting meaning into a smaller number of words’ (2009, p. 249) works concurrently with colloquialisation so much so that they (2009, p. 252) claim: We often find cases where colloquialization and densification compete with one another in the same text. Putting it flippantly, one could argue that the current trend in written English is to pack ever more information into a given length of text and then to ‘sell’ this fairly heavy intellectual diet in a somewhat more informal/colloquial style than used to be the case. Crucially for my own interests, densification is primarily accounted for by changes in the structure of the NP although it is also effected by lexical density, morphological changes and processes of word formation. In connection with the NP, Leech et al. (2009) argue that: We were surprised to note an increase in the occurrence of nouns as a very highly significant trend. [. . .] The high increase of frequency of certain noun classes [. . .] pointed to an overall pattern of condensation of information in the noun phrase, to which we gave the label ‘densification’. (Leech et al., 2009, p. 206) Densification, therefore, is due to the increasing frequency of nouns which causes information to become highly condensed and compact. Of course, the frequency of nouns as well as the complexity of NPs are highly dependent on register differences so on a written-informational vs. spokencolloquial gradient we would respectively expect processes of densification or colloquialisation (see also Biber et al., 1999). This upward trend of NP use is clearly at odds with the under-representation of nouns I find in my own corpora. If the densification process described by Leech et al. (2009) is a feature characterising the written language of twentieth-century English, then that tendency is subverted in the popular novels I investigate; this subversion, thus, would indicate an opposite stylistic effect to that of densification. It could be argued that the underuse of nouns found in genre fiction veers away from twentieth-century prototypical written language which endows it with a more ‘colloquial feel’ than that found in other written varieties, including Literature. If, as Leech et al. (2009) contend, colloquialisation and densification compete in twentieth-century English, the former certainly wins the battle with regard to popular literature. Biber and Clark (2002) also look at NP structure but their analysis is register based. Based on the results discussed in Biber et al. (1999),18 Biber and Clark’s (2002) conclusions differ from those of Leech et al. (2009) in so far as NPs appear to be as frequent in conversation, news, academic writing or fiction; crucially, though, they contend that it is the internal

70

Rocío Montoro

structure of the NP that differs in each register. Biber and Clark (2002, p. 46) echo Biber et al. (1999) and confirm that ‘prepositional phrases (rather than relative clauses) are overwhelmingly the most common type of post-modification’ in present-day English (PDE) giving rise to what they term ‘compressed expression’ (Biber and Clark, 2002, p. 63). Thus, whereas the clausal post-modification (non-finite and relative clauses) prototypical of the past (seventeenth to twentieth centuries) afforded a more ‘expanded expression’ (Biber and Clark, 2002, p. 63), PDE is located at the opposite end of the cline favouring a more compressed way of delivering information. Additionally, and crucially for my own interests, Biber and Clark (2002, p. 44) also clarify that though non-clausal modifiers are much more common than clausal modifiers, their occurrence differs among registers so that non-clausal modification is particularly prevalent in academic writing (a type of written-informational register) and it seems virtually absent from normal conversation (a spoken and colloquial register). Thus, the underuse of phrasal post-modification would align popular fiction with prototypically colloquial registers and, hence, would endow it with a more colloquial feel than that associated with Literature, for instance. These two processes, i.e. densification and compressed expression, therefore highlight a particular trend regarding the NP structure which characterises certain registers in the twentieth century and to which popular writing seems to run counter. I argue that these discrepancies endow popular fiction with the kind of ‘colloquial feel’ that makes readers and critics view it as (a) less syntactically complex and (b) easier to read. Following the results in Montoro and McIntyre (2019, fc) I focus on phrasal post-modification, in particular on the of-phrase, to assess whether the behaviour of this syntactic form contributes to the general informal feel of this text type. Leech et al. (2009) confirm a significant decline in the use of prepositions in twentieth-century English: Although the loss of prepositions cannot necessarily be connected with the noun phrase (since the prepositional phrases they introduce can function adverbially as well as as modifiers in the NP), a larger number of them are in noun phrases than are adverbial in function, so the decline in prepositions may indeed reflect the tendency to make noun phrases more compact. Certainly, this is an obvious interpretation of the declining use of the most frequent preposition, of, which characteristically functions in the post-modification of noun phrases. (Leech et al., 2009, p. 209) Significantly, this decline is corroborated by Biber and Gray (2011, p.  232) who underscore the loss of prepositions introducing NP postmodifiers in fiction: ‘Fiction writing actually shows a decrease in the use of PP noun modifiers’. These results allow me to hypothesise that

Syntactic Simplicity in Popular Fiction

71

a statistically significant underuse of prepositions specifically functioning as nominal post-modifiers (in particular the preposition ‘of’) would strengthen the case of phrasal simplicity in popular fiction observed by Montoro and McIntyre (2019, fc) and would align lowbrow novels with registers prototypically seen as colloquial: Preposition ‘of’: Chick Lit Corpus compared to BNC Sampler Written Imaginative (p < 0.05) IO =

−264.68

Preposition ‘of’: MVC compared to BNC Sampler Written Imaginative (p < 0.05) IO =

+45.28

These figures are telling as the over-representation of the of-preposition (IO) in the MVC is clearly at odds with the underuse of the same POS in the Chick Lit Corpus. On the one hand, Chick Lit novels underuse nouns as well as prepositions so it could be argued that NPs in Chick Lit are less frequent and less syntactically complex than those in the BNC Written Imaginative, characteristically associated with highbrow writing. However, although the MVC uses fewer nouns, the IO tag is over-represented which, as Biber and Gray (2011) have observed, prevails in academic writing not in fiction. A manual analysis of the IO tag concordances in both corpora, though, demonstrates that the under-representation in the Chick Lit Corpus as well as the overuse in the MVC require further analysis. Toolan (2008) addresses the issue of interpretation in corpus-based studies as follows: The software counts words the way a garden fork digs an allotment [. . .] Computers have enabled us to see texts, and language, differently. By comparison with the old so-called ‘manual’ computational methods for grappling with large amounts of linguistic material, it is perhaps analogous to displacement of the handfork by the dieselpowered tractor in the cultivation of 100-acre fields. But a tractor is not a farmer and corpus linguistic programmes do not do analysis. (Toolan, 2008, pp. 86, 87) Toolan’s (2008) horticultural metaphor usefully illustrates some of the drawbacks associated with corpus analysis. Corpus linguistic programmes are those diesel-powered tractors capable of unearthing linguistic patterns but it is up to the farmer to dredge up the actual function of particular patterns. Looking in more detail at the various concordances of the IO tag in both corpora reveals that the of-preposition is multifunctional, i.e. it is employed in contexts other than that of nominal

72

Rocío Montoro Table 5.1 Non-Nominal Post-Modification Verb ‘he was thinking

IO (of) of

other POSi leaving’

Adjective ‘fond

IO (of) of

other POS him’

Prepositions ‘out

IO (of) of

other POS mortal flesh’

Superlative after-determiner ‘most

IO (of) of

other POS whom’

Adverb ‘sort

IO (of) of

other POS steered’

Non-head pronounii IO (of) ‘[accused] me of

other POS being’

i

Various categories occur after the IO (of) form, including verbs, pronouns, nouns, determiners, adjectives, adverbs, etc.; these are, however, irrelevant to my analysis here. ii All these categories correspond to the same tagset used in the software Wmatrix (Rayson, 2009).

post-modification. The manual analysis of the IO concordances shows that, besides nominal post-modification, the IO (of) tag features in the following syntactic contexts. Table 5.1 illustrates that the IO tag complements verbs, adjectives and adverbs; it is also part of complex prepositions and follows superlative after-determiners (most, least, fewest); moreover, this manual approach confirms that the IO (of) tag has the same six functions besides that of nominal post-modification in the Chick Lit Corpus and the MVC respectively. The statistical significance of Wmatrix (Rayson, 2009) results, therefore, requires further contextualisation so I complement the initial analysis with a second one using AntConc (Anthony, 2014), a freeware corpus analysis toolkit for concordancing and text analysis. Although Wmatrix (Rayson, 2009) also offers search and concordancing options, AntConc (Anthony, 2014) allows the analyst to search for terms as words or as character strings. Thus, I loaded the POS tagged versions of my target corpora into AntConc (Anthony, 2014) in order to scrutinise fully the six functionalities of the IO (of) tag brought to light thanks to a manual analysis (Table 5.1). By using a combination of wildcards and the tagset employed in the original software (Wmatrix; Rayson, 2009), I was able to isolate instances of nominal post-modification and separate them from non-nominal functions (e.g. verb, adverb or adjective complement) which eventually helped me quantify each occurrence separately and accurately. Using a combination of wildcards and tagsets as illustrated in Table 5.2, AntConc (Anthony, 2014) quantified the occurrences of each kind of IO (of) tag separately as shown in Tables 5.3 and 5.4.

Syntactic Simplicity in Popular Fiction

73

Table 5.2 Non-Nominal Post-Modification Wildcards and Tags Verb

*V* of_IO *_V*

. . . what comes of being a lawyer . . .

Adjective

*JJ* of_IO *_V*

. . . incapable of guarding you . . .

Prepositions

*II* of_I* *_V*

. . . instead of watching . . .

Superlative after-determiner

*DAT* of_IO *_P*

. . . least of anyone

Adverb

*RR* of_RR* *_V*

. . . kind of liked . . .

Non-head pronoun *PP*O* of_IO *_V* . . . deceiving her of creating . . .

Table 5.3 Non-Nominal Post-Modification in the Chick Lit Corpus Verb (V) Adjective (JJ) Prepositions (II) Superlative after-determiner (DAT) Adverb (RR) Non-head pronoun (PP) Total

283 266 622 45 87 18 1,321

Table 5.4 Non-Nominal Post-Modification in the MVC Verb (V) Adjective (JJ) Prepositions (II) Superlative after-determiner (DAT) Adverb (RR) Non-head pronoun (PP) Total

1,099 874 1,364 62 149 96 3,644

These figures clearly prove that not all the occurrences of the IO (of) tag function in post-nominal position and indicate that the statistical significance measured by Wmatrix (Rayson, 2009) should be reconsidered. That is, since the focus of my investigation is ascertaining whether the underuse of nominal post-modification prevails in genre fiction, it is crucial that those figures are revisited in the light of these results. In order to single out the actual instances of nominal post-modification, I subtracted the number of non-nominal post-modification occurrences as shown in Tables 5.3 and 5.4 from the total number of IO (of) tags which both Wmatrix (Rayson, 2009) and AntConc (Anthony, 2014) computed: Original IO (of) count in the Chick Lit Corpus: 5,810 Non-nominal post-mod in the Chick Lit Corpus: 1,321 Total nominal post-modification: 4,489

74

Rocío Montoro Original IO (of) count in the MVC: Non-nominal post-mod in the MVC: Total nominal post-modification:

19,969 3,644 16,325

This new reading exclusively includes occurrences of nominal postmodification which entails that the statistical significance measure, i.e. log-likelihood ratio,19 of this tag as a noun post-modifier must be recalculated: Nominal post-modification in the Chick Lit Corpus Original LL: −264.68  New LL: −898.87 Nominal post-modification in the MVC Original LL: +45.28  New LL: −43.98 Judging by the newly obtained LL values, it seems that the two-stage methodology I apply in this chapter qualifies the original results considerably. On the one hand, using a second piece of software allowed me to focus specifically on the particular function of the IO (of) tag which can help me assess whether phrasal simplicity is indeed a feature of popular fiction. On the other hand, this not only confirms that the IO (of) tag is statistically underused in the Chick Lit Corpus but, more importantly, it also demonstrates that the post-nominal of-phrase is also statistically under-represented in the MVC, something which disproves the initial reading in which all the functions of the IO (of) tag were combined as one. The underuse of nouns and the loss of nominal post-modification, therefore, jointly account for the kind of phrasal simplicity which endows genre writing with its colloquial, informal feel.

4. Conclusion In this chapter I addressed the distinction between serious vs. popular fiction from a corpus stylistics perspective and assessed some literary critical claims which, inaccurately, distinguish the former from the latter on the under-specified basis of syntactic simplicity. In the previous pages, I have argued, first, that the issue of simplicity is not dichotomous and should always be viewed in relation to many other factors, such as, for example, register. Syntactic simplicity is first viewed with regard to the use of subordinating conjunctions but this measure proves inconclusive as subordinating conjunctions are overused in both serious and popular fiction. I have implemented a corpus stylistics methodology, though, following Toolan (2008), I have argued that computer-based methods cannot replace the creative input of the analyst. In my case, for instance, using one piece of software proves insufficient in whittling out the functional nuances of the investigated POS and, as shown, using additional software proves far more effective.

Syntactic Simplicity in Popular Fiction

75

Leech et al. (2009), on the one hand, and Biber and Clark (2002), on the other, describe twentieth-century English as having undergone processes of densification, colloquialisation and compressed expression. In this chapter, I hypothesised that a statistically significant under-representation of prepositions introducing nominal post-modifiers would strengthen the case hinted at in Montoro and McIntyre (2019, fc) with regard to phrasal rather than clausal simplicity in popular fiction. As the results discussed in section 3 demonstrate, the nominal post-modifying IO (of) category is statistically underused when compared to the BNC Written Imaginative so phrasal simplicity with regard to the NP is indeed a feature that characterises my popular fiction corpora. Moreover, the under-representation of nouns, on the one hand, as well as the underuse of the nominal postmodifying of-phrase, on the other, situate popular fiction at the opposite end of the densification trend prototypically associated with twentiethcentury English. Instead, these features situate genre fiction at the ‘colloquialisation’ end which could explain the recurrent categorisation of popular writing as ‘easy to read’. In sum, my research underscores that the simplicity associated with popular fiction can be accounted for by means of the lower frequency of nouns as well as the lower frequency of post-modifying prepositional of-phrases. These results also highlight that further research on other types of popular writing should be considered in order to ascertain whether these results are, indeed, prevalent across lowbrow fiction or simply associated with the kind of novels included in my corpora.

Notes 1. ‘Simplicity’ and ‘complexity’ as I address them in this paper should more accurately be situated along a grammatical continuum and not as discrete states; however, I maintain the simple vs. complex distinction because of how recurrently lowbrow fiction is described in terms of the former. 2. The NP has received plenty of attention from a variety of perspectives. For a summary of the various ways in which this category has been discussed see Martínez-Insua and Pérez Guerra (2011); for the relationship between NP type and sentence complexity see Gordon et al. (2004). 3. The criteria Semino and Short use for the distinction of serious vs. popular, especially with regard to the fiction section, are, for the most part, not linguistic but ‘the cultural distinction between popular fiction on the one hand and highbrow fiction on the other. Apart from anything else, this distinction has concrete manifestations, such as the design of front covers and the physical positioning of books in bookshops and libraries’ (Semino and Short, 2004, p. 22). 4. Ferriss and Young (2006) and Montoro (2012) provide a set of criteria which define Chick Lit as a popular genre. 5. The six novels are Bridget Jones’s Diary (Fielding, 1996), Getting Rid of Matthew (Fallon, 2007), Pants on Fire (Alderson, 2000), The Last Year of Being Single (Tucker, 2003), The Other Woman (Green, 2004) and The Secret Dreamworld of a Shopaholic (Kinsella, 2000).

76

Rocío Montoro

6. The BNC Written Imaginative is part of the BNC Sampler Written and contains only fiction, primarily though not exclusively, literary prose, which fits Montoro and McIntyre’s (2019, fc) purpose of comparison appropriately. For more information see the BNC User Reference Guide available at: www. natcorp.ox.ac.uk/docs/URG/BNCdes.html. 7. The first of the two digits immediately following the grammatical tag ‘CS’ indicates the number of words/tags in the sequence; the second digit marks the position within that sequence. Examples of the CS2 category are as if, as though, so that; examples of the CS3 category are as long as, as far as, as soon as. 8. CSA: as (as conjunction). 9. CSW: whether (as conjunction). 10. CST: that (as conjunction). 11. Biber and Clark (2002, p. 43) equally underscore that subordination is not more prevalent in written language than in conversation despite common assumptions; it is, instead, the kind of subordination or embedded clauses that distinguish one register from another. 12. For an account of vampire literature as popular fiction see Glover (1996). 13. From Anne Rice’s The Vampire Chronicles: Interview with the Vampire (1976), The Vampire Lestat (1985), The Queen of the Damned (1988), The Tale of the Body Thief (1992). From Charlaine Harris’s Sookie Stackhouse series: Dead Until Dark (2001), Living Dead in Dallas (2002), Club Dead (2003), Dead to the World (2004). 14. NN1: singular common noun (e.g. book, girl). 15. NN2: plural common noun (e.g. books, girls). 16. NN: common noun, neutral for number (e.g. sheep, cod, headquarters). 17. This speech-like tendency has also been attested by Biber (2003). 18. Unlike Leech et al. (2009), Biber et al.’s (1999) focus is synchronic and based on register variation, namely conversation, fiction, news and academic prose. 19. A log-likelihood calculator is available at: http://ucrel.lancs.ac.uk/llwizard. html.

References Alderson, M. (2000) Pants on fire. London: Penguin Books. Anthony, L. (2014) AntConc (Version 3.4.3) [Computer Software]. Tokyo, Japan: Waseda University. Available at: www.laurenceanthony.net/. Biber, D. (2003) ‘Compressed noun-phrase structures in newspaper discourse: The competing demands of popularization vs. economy’, in Aitchison, J. and Lewis, D. (eds.) New media language. London and New York: Routledge, pp. 169–181. Biber, D. and Clark, V. (2002) ‘Historical shifts in modification patterns with complex noun phrase structures: How long can you go without a verb?’ in Fanego, T., López-Couso, M. J. and Pérez-Guerra, J. (eds.) English historical syntax and morphology. Selected Papers from 11 ICEHL, Santiago de Compostela, 7–11 September 2000. Amsterdam: John Benjamins, pp. 43–66. Biber, D. and Gray, B. (2011) ‘Grammatical change in the noun phrase: The influence of written language use’, English Language and Linguistics, 15 (2), pp. 223–250. Biber, D., Johansson, S., Leech, G., Conrad, S. and Finnegan, E. (1999) Longman grammar of spoken and written English. Harlow, Essex: Pearson Education.

Syntactic Simplicity in Popular Fiction

77

BNC User Reference Guide. Available at: www.natcorp.ox.ac.uk/docs/URG/ BNCdes.html (Accessed 15 September 2017). Dunning, T. (1993) ‘Accurate methods for the statistics of surprise and coincidence’, Computational Linguistics, 19 (1), pp. 61–74. Fallon, J. (2007) Getting rid of Matthew. London: Penguin Books. Ferriss, S. and Young, M. (eds) (2006) Chick lit: The new woman’s fiction. New York and London: Routledge. Fielding, H. (1996) Bridget Jones’s diary. London: Picador. Gelder, K. (2004) Popular fiction: The logics and practices of a literary field. London: Routledge. Glover, D. (1996) Vampires, mummies, and liberals: Bram Stoker and the politics of popular fiction. Durham, NC and London: Duke University Press. Gordon, P. C., Randall, H. and Johnson, M. (2004) ‘Effects of noun phrase type on sentence complexity’, Journal of Memory and Language, 51 (1), pp. 97–114. Green, J. (2004) The other woman. London: Penguin Books. Harris, Ch. (2001) Dead until dark. London: Gollancz. Harris, Ch. (2002) Living dead in Dallas. London: Gollancz. Harris, Ch. (2003) Club dead. London: Gollancz. Harris, Ch. (2004) Dead to the world. London: Gollancz. Kinsella, S. (2000) The secret dreamworld of a shopaholic. London: Black Swan. Leech, G., Hundt, M., Mair, Ch. and Smith, N. (2009) Change in contemporary English: A grammatical study. Cambridge: Cambridge University Press. Martínez-Insua, A. E. and Pérez-Guerra, J. (2011) ‘An open-sesame approach to English noun phrases: Defining the NP (with an introduction to the special issue)’, English Language and Linguistics, 15 (2), pp. 201–221. McCracken, S. (1998) Pulp: Reading popular fiction. Manchester: Manchester University Press. McIntyre, D. and Montoro, R. (2012) ‘Complex sentences in serious and popular fiction’, IV International Conference of Corpus Linguistics (CILC_12). University of Jaén, Spain, 22–24 March. Montoro, R. (2012) Chick lit: The stylistics of cappuccino fiction. London: Bloomsbury. Montoro, R. (2015) ‘Style in popular literature’, in Sotirova, V. (ed.) The Bloomsbury companion to stylistics. London: Bloomsbury, pp. 673–689. Montoro, R. and McIntyre, D. (2019, fc) ‘Subordination as a potential marker of complexity in serious and popular fiction: A corpus stylistic approach to the testing of literary critical claims’, Corpora, 14 (3). Moretti, F. (2010) ‘History of the novel, theory of the novel’, Novel: A Forum on Fiction, 43 (1), pp. 1–10. Oxford Text Archive. Available at: https://ota.ox.ac.uk/ (Accessed 15 September 2017). Ramet, C. (1999) Ken Follett: The transformation of a writer. Bowling Green, OH: Popular Press. Rayson, P. (2009) Wmatrix: A web-based corpus processing environment. Computing Department, Lancaster University. Available at: http://ucrel.lancs.ac.uk/ wmatrix3.html (Accessed 15 September 2017). Rice, A. (1976) Interview with the vampire. London: Sphere. Rice, A. (1985) The vampire lestat. London: Sphere. Rice, A. (1988) The queen of the damned. London: Sphere.

78

Rocío Montoro

Rice, A. (1992) The tale of the body thief. London: Sphere. Semino, E. and Short, M. (2004) Corpus stylistics. Speech, writing and thought presentation in a corpus of English writing. London and New York: Routledge. Toolan, M. J. (2008) ‘“Around 1990, Corpus Linguistics set in”: Combining systemic linguistics, corpus methods, and stylistics in the study of narrative’, Systemic functional linguistics in use. Odense working papers in language and communication (Edited by N. Nørgaard), 29, pp. 85–110. Toolan, M. J. (2009) Narrative progression in the short story: A corpus stylistic approach. Amsterdam and Philadelphia: John Benjamins. Toolan, M. J. (2016) Making sense of narrative text: Situation, repetition, and picturing in the reading of short stories. London and New York: Routledge. Tucker, S. (2003) The last year of being single. London: Red Dress Ink.

6

‘We work hard here’ Exploring Person and Place Deixis in a Corpus of Historical Migrant Letters Emma Moreton

1. Introduction This essay uses a combination of qualitative and corpus-based approaches to explore the letters of two sisters, Annie and Julia Lough, who migrated from Ireland to America in the late nineteenth century. First, a close qualitative reading of the letters is carried out to identify topics in the discourse. Visualisation and corpus tools (the Interesting Items Visualisation Tool (IIVT) (Culy, 2016) and Sketch Engine (Kilgarriff and Kosem, 2012)) are then used to examine the language of one topic in particular—‘Ireland and America’—to see if any linguistic patterns can be identified. First, the IIVT is used to identify statistically significant semantic domains and words in the topic ‘Ireland and America’ when compared with all other topics combined. Next, Sketch Engine is used to examine some of the linguistic findings in context, focusing in particular on person deixis (specifically the firstperson pronoun ‘we’) and place deixis (specifically the adverbs ‘here’ and ‘there’). As summarised by Zubin and Hewitt (1995, p. 6), the term ‘deixis’ refers to the fact that certain linguistic forms have direct pragmatic interpretation dependent on parameters of the speech situation, rather than a stable semantic value. In particular, their interpretation is contextually anchored to the identity of the speaker and addressee, their locations, and the time of the utterance. An analysis of deictic terms, in the context of writing about life in Ireland and/or America, will, therefore, reveal something about how the Lough sisters positioned themselves in relation to the recipient of the letter in terms of physical and social distance and proximity and how they self-identified—as ‘Daughters of Erin’ (Diner, 1983) and/or as American citizens.

80

Emma Moreton

2. The (Female) Irish Migrant The post-famine period (circa. 1850s–1920s) was a time that saw a significant increase in female migration from Ireland to America. Economic changes in Ireland, including declining wage-earning capabilities due to the de-industrialisation of the Irish countryside, as well as changes in inheritance practices from partible to impartible inheritance systems (in turn, leading to changes in marriage trends), contributed to ‘a massive post-famine emigration by young, unmarried women’ (Miller, 1985, p. 3). As Diner explains, ‘in the society created in the wake of the Famine only one son per family could inherit land’; the same son typically entered into matrimony (1983, p. 10). Additionally, the widespread application of the dowry system meant that only one daughter within a family could realistically expect to marry. Her unmarried sisters had, as Diner puts it, ‘limited options’: ‘They could remain single in the countryside’ with very little chance of work; they could seek employment in cities such as Dublin, or further afield; or they could join the ‘millions of other young Irish women crossing the Atlantic to seek fortune or family in the United States’ (1983, p. 12). By the second half of the nineteenth century Ireland had become ‘a nation characterized by late and reluctant marriage as well as by a massive voluntary exodus’ (ibid., p. 8). According to Diner, Irish women differed from other migrant women in several ways. For one, ‘they were the only significant group of foreignborn women who outnumbered men’ (1983, p. xiv). Indeed, after 1880, young women constituted the majority of the departing Irish (Miller, 1985, p. 392). However, while female Irish migrants appeared to enjoy a certain level of autonomy and independence—they migrated alone, played an important economic role within the notional family hierarchy and sought to advance themselves socially—this, Diner argues, ‘did not mean that they thought only of themselves’ (1983, p. xiv). Rather, their actions were largely motivated by ‘family loyalties’ and ‘a commitment to Irish Catholic culture and to its way of life’ (ibid.). Nevertheless, migration would surely have represented significant change for these women. For Meaney et al. (2013, p. 101), although the female Irish migrant ‘may not have contrasted her new autonomous life with the patriarchal, restricted life predominant in rural Ireland’, neither did she ‘reject what she encountered in America’. Letters home relayed information about experiences of, and activities in, modern American society (ibid.) and would almost certainly have influenced relatives in Ireland: Family and friends in Ireland, Meaney et al. (2013, p. 100) suggest, ‘learned from letters that conditions in America could be transforming’. Previous research which uses Irish migrant letters as a primary data source has typically focused on one of two areas: (1) the language of the

‘We work hard here’

81

letters and (2) the subject matter of the letters. While linguists have been especially interested in exploring the linguistic features of Irish English as well as aspects of language change and variation (see, for instance, McCafferty and Amador, 2012, 2014; Amador and McCafferty, 2015; Van Hattum, 2014; Montgomery, 1995), historians have been more concerned with what the migrants wrote about. Two of the most notable studies by social historians are Miller’s (1985) book Emigrants and Exiles and Fitzpatrick’s (1994) book Oceans of Consolation. Miller carries out a close reading of 5,000 letters by Irish migrants to North America from 1607 to 1921. He argues that although most Irish who crossed the Atlantic were ‘voluntary emigrants who went abroad in search of better economic and social opportunities—that is, for the same reasons motivating emigrants from other parts of Europe’ (Miller, 1985, p. 6), they often viewed themselves as involuntary exiles, ‘compelled to leave home by forces beyond individual control, particularly by British and landlord oppression’ (ibid., p. 556). Fitzpatrick, using a much smaller dataset and focusing on a much shorter time span, explores nineteenth-century Irish migration to Australia. Fitzpatrick observes themes such as ‘home’, ‘loneliness’ and ‘nostalgia’—features of migrant correspondence that are also noted by Miller; however, he ‘reports no comparable use of the [‘exile’ trope] among the Irish migrants in Australia’ (Elliott et al., 2006, p. 11). The current study builds on research in both of these areas: social history and linguistics. Similar to Miller and Fitzpatrick, it begins by examining what the Lough sisters wrote about to see whether similar topics and themes emerge. It then uses linguistic tools to compare the language of the different topics to see whether there is a propensity for using certain words and phrases (i.e. a local grammar) when talking about specific subjects. Finally, it examines the language of one topic—‘Ireland and America’—in more detail, using the concept of deixis as a framework within which to discuss the findings.

3. The Lough Letters The Lough (pronounced ‘lok’) family letters are taken from a much larger body of Irish migrant correspondence collected by Kerby Miller, Emeritus Professor, University of Missouri. Miller himself has explored this wider corpus in several pioneering works on Irish migration (see, for instance, Miller, 1985, 2008; Miller et al., 1995, 2003), and his archive of more than 5,000 letters has been referred to by many scholars, including Emmons (1990), Koos (2001), Bruce (2006), Corrigan (1992), Nolan (1989) and Noonan (2011). The six Lough sisters—Elizabeth, Alice, Annie, Julia, Mary and Maggie—came from a Roman Catholic family in Meelick, in what was then called Queen’s County (now County Laois), Ireland.1 The sisters

82

Emma Moreton

were daughters of Elizabeth McDonald Lough and James Lough who lived on a small holding consisting of two fields, one of which, according to family legend, was sold to pay for the sisters’ passages. The Lough family was, according to Miller, not of the lowest class as both parents and daughters were able to write. Apart from Mary and Maggie, all the Lough sisters migrated to America between 1870 and 1884. There are 99 letters in the Lough collection, the majority of which were written by Annie (39 letters / 20,405 tokens), the third sister to migrate in 1878, and Julia (35 letters / 12,220 tokens), the fourth sister to migrate in 1884. This essay will focus on these two larger collections, which, from hereon in, will be referred to as the ALC (Annie Lough Corpus) and the JLC (Julia Lough Corpus). Tables 6.1 and 6.2 show how frequently Annie and Julia wrote home and whom they wrote to. Focusing on Annie’s correspondence first of all, Table 6.1 shows that her earlier letters were addressed to her mother, the first of which was sent in around 1878 (although the letter itself is not dated) from Queenstown, County Cork, Ireland, just before Annie was about to set sail for America. After 1895 (around the time of her mother’s death), Annie began writing to her sister Mary, and the correspondence continued into the late 1920s. Annie wrote to Mary regularly during this 30- to 35-year period, often sending letters at Easter and Christmas, or on the anniversary of a family member’s death. Several of Annie’s letters are not dated but their content would suggest they were written from 1920 onwards. All but the first letter were sent from Winsted, Litchfield County, Connecticut. Focusing on Julia’s correspondence, Table 6.2 shows that 23 of the 35 letters were addressed to Julia’s mother, while 12 were addressed to her sister Mary. Most of Julia’s letters (33 out of 35) were sent between 1884 to 1895. Some of the letters are not dated, but their content has allowed them to be placed within an approximate time frame. Most of the letters dated between 1884, when Julia first migrated, and 1894 were sent from Winsted, Connecticut. Around 1895, Julia relocated to Torrington, Connecticut. Julia’s pattern of letter-writing differs quite noticeably from her sister Annie’s. Annie’s letters are fairly evenly distributed and there are no major gaps in her correspondence. Most of Annie’s letters (27 out of 39) were addressed to their sister Mary and were sent after their mother’s death. In contrast, most of Julia’s letters were addressed to their mother. After their mother’s death, Julia wrote somewhat sporadically to Mary and there was a 24-year gap in Julia’s writing between letter 33 (sent in 1895) and letter 34 (sent in 1919–1920). Overall, Annie tended to write longer letters than Julia with an average word count of 523 (versus 349 for Julia). Additionally, Annie’s letters were more varied in length than Julia’s (see Figure 6.1).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

18 03 29 15 23 30 – – 17 18 16 21 10 03 20 30 12 08 11 31 06 14 14 21 21 07 –

Day

June March October December March March December – March May February September December April June November December December December April May July August March March December –

Month

Table 6.1 Overview of the ALC

– 1890 1891 1891 1892 1893 – – 1895 1899 1901 1901 1902 1906 1906 1906 1912 1913 1914 1918 1918 1918 1919 1920 1920 1919/1920 –

Year Queenstown, Ireland Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America

From (Location) Mother Mother & Sister Mother Mother Mother Mother Mother Mother Sister Sister Sister Sister Sister Sister Sister Sister Sister Sister Niece Sister Sister Sister Sister Niece Sister Sister Sister

Recipient Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland

To (Location)

356 480 1,055 487 1,017 971 208 645 612 541 394 441 365 431 332 302 632 514 398 664 884 863 857 469 649 396 435

Tokens

28 29 30 31 32 33 34 35 36 37 38 39

31 29 28 18 04 – – – – – – 01

March September March October November – – – – – – December

1924 1925 1928 1928 – – – – – – – 1919

Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America

Sister Sister Sister Sister Nephew Sister Sister Sister Sister Sister Sister Sister

Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland

538 237 206 870 513 261 400 489 476 207 307 503

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

27 – 20 – – 03 02 – – 09 10 – 18 25 30 18 14 11 01 – 21 – – 03

Day

September – December – December November December – – March August December January January March October December May September – March May July September

Month

Table 6.2 Overview of the JLC

1884 1884 1884 1884–1894 1888 1889 1889 1889–1890 1889–1894 1890 1890 1890 1891 1891 1891 1891 1891 pre–1892 1892 1892–1893 1893 1893 1893 1893

Year Queenstown, Ireland Ireland or England Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America Winsted, America

From (Location) Mother Mother Sister Mother Mother Mother Mother Mother Mother Mother Mother Mother Sister Mother Mother Mother Mother Mother Mother Mother Sister Mother Mother Mother

Recipient Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland

To (Location)

40 98 519 190 342 444 436 487 259 463 366 350 348 351 225 317 300 400 396 321 423 305 340 356

Tokens

25 26 27 28 29 30 31 32 33 34 35

10 – 25 24 – 04 – 08 – 17 09

October December March May – June November July August March November

1893 1893 1894 1893–1894 1889–1894 1894 1895 1895 1895 1919–1920 1927

Winsted, America Winsted, America Torrington, America Torrington, America Torrington, America Winsted, America Winsted, America Queenstown, Ireland Torrington, America Torrington, America Torrington, America

Mother Mother Sister Sister Sister Sister Sister Sister Sister Sister Sister

Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland Meelick, Ireland

334 451 183 477 354 736 469 44 416 331 349

‘We work hard here’

87

ALC

JLC

Figure 6.1 Comparative Charts Showing Length of Letters in the ALC and JLC

4. Topic Identification To prepare the ALC and JLC for analysis, the letters were transcribed and saved in Plain Text format. While automated semantic taggers provide a useful overview of the content of a corpus, they can sometimes miss more subtle topics and themes in the discourse. A trial study using Wmatrix (Rayson, 2009) as a starting point, for example, showed that it was sometimes difficult to ascertain those sections of the discourse that were describing, for instance, homesickness, which is often expressed indirectly or through the use of metaphor. Therefore, a close reading of the letters was carried out to identify sequences in the discourse that

88

Emma Moreton

appeared to be lexically related to see what topics emerged. The topics were then annotated following XML conventions. The following section, for example, is about education and is marked with the opening tag to show where this section begins and the closing tag (with forward slash) to show where it ends, as follows: I hope you keep them to school all you can when they grow bigger you can not send them very well I suppose there are not many of the nuns alive now that was there when I went to school we call them Sisters here I think we have 15 or more of them here they have a very nice convent they go to Mass with the school children and go see the sick we have two other schools besides publick schools that is for any one wishes to go there and there is a High school also. (ALC, 10 December 1902) In cases where the discourse could be interpreted in more than one way, two or more tags were assigned. This meant that a section could be said to be ‘about’ just one topic, or it could be said to be ‘about’ a number of topics. In the previous example, where the text is annotated with the tags , an alternative interpretation might be the topic ‘Ireland and America’ (any reference to life in Ireland and/or America). In this situation, the annotation would be as follows: I hope you keep them to school all you can when they grow bigger you can not send them very well I suppose there are not many of the nuns alive now that was there when I went to school we call them Sisters here I think we have 15 or more of them here they have a very nice convent they go to Mass with the school children and go see the sick we have two other schools besides publick schools that is for any one wishes to go there and there is a High school also. (ALC, 10 December 1902) Additionally, it is possible for one or more topics to be embedded within another main topic. In the following example, for instance, the main theme is work; however, within this section Annie makes a comment regarding attitudes towards work in Ireland versus America, so the tag has been embedded within , as follows: is Maggie home yet or is she going to stay home all winter idle spending her money and weering out her nice clothes I think she is very foolish for her self she had ought to stay in a good place when she had one she wont get one like it for a while again I wonder she comes home to stay there now she ought to be working for herself and considering the wey things are at home

‘We work hard here’

89

now what she would do if she was here no one ever thinks of staying a week away from work unless they were sick or out of work. (ALC, 15 December 1891) Tagging for topics allowed me to get a sense of what the sisters wrote about—their preoccupations, perceptions and experiences. It also allowed me to extract all instances of a particular topic to explore how language is used within that specific context. Using corpus tools to compare language use across topics made it possible to identify local grammars— words, phrases and structures that are statistically more likely to occur in one topic over another. As this study involved comparing one section of the corpus against all other sections combined, a separate reference corpus was not necessary. Table 6.3 lists the topics that were identified in the ALC and JLC, organised alphabetically. Many of the topics identified can be mapped onto Fitzpatrick’s (1994) thematic index; however, there are a few alterations and additions, namely: there are now separate categories for ‘Homesickness’, ‘Future letters’ and ‘Previous letters’, and a distinction is made between remittances and other types of enclosure. Additionally, the topics ‘Postal system’, ‘World War I’, ‘Reassurance’ and ‘Identity’ have been added. The ‘Freq.’ column shows how often the topics occurred across all letters and the ‘Av. p/letter’ column gives the average frequency of a topic per letter. Of course, it is possible that a topic may be mentioned several times in the same letter, or it may not be mentioned at all. Counting the number of times a topic occurs thus offers only one way into the letters. Counting the number of words attributed to each occurrence of a particular topic, on the other hand, arguably provides a more accurate reflection of the content of a letter, or a letter collection (i.e. it tells us what percentage of the letter or letter collection was spent discussing health, education or work, for instance). The ‘Tokens’ column, therefore, provides the raw word count for each topic while the ‘%’ column shows what percentage of the corpus was attributed to that particular topic. What is perhaps striking about Table 6.3 is how rarely topics such as ‘Migration’, ‘Work’ and ‘Education’ seem to crop up, even though these topics are precisely the ones many researchers on migrant letters have focused on (e.g. Koos, 2001; Miller, 1985; Miller et al., 1995). Instead, the far greater focus of both Annie’s and Julia’s letters is on ‘Family and friends’ (82 occurrences (39.94% of the overall content) in the ALC and 58 occurrences (34.81%) in the JLC), as well as the subtle national comparisons identified under ‘Ireland and America’ (59 occurrences (22.14%) in the ALC and 66 occurrences (18.57%) in the JLC). It is perhaps unsurprising that these are the two most frequent topics in the Lough sisters’ letters given the context in which they wrote; however, it

Freq.

1 18 6 49 82 86 25 7 1 59 6 7 12 4 71 5 39 17 15 6 1 15 27 4

ALC Topics

Childbirth Death Education Enclosure Family and friends Future letters Health and illness Homesickness & separation Identity Ireland and America Marriage Migration News event Postal system Previous letters Reassurance Recollection Religion Remittance Reunion Transportation Weather and seasons Work World War I

0.02 0.46 0.15 1.26 2.10 2.21 0.64 0.18 0.03 1.51 0.15 0.18 0.31 0.10 1.82 0.13 1 0.44 0.38 0.15 0.03 0.38 0.69 0.10

Av. p/letter 92 1712 255 1354 8149 1580 1197 230 9 4517 274 453 800 129 1681 231 1274 961 438 152 69 616 823 396

Tokens

Table 6.3 Topics in the ALC and JLC, Listed Alphabetically

0.45 8.39 1.25 0.24 39.94 7.74 5.87 1.13 0.04 22.14 1.34 2.22 3.92 0.63 8.24 1.13 6.24 4.71 2.15 0.75 0.34 3.02 4.03 1.94

% Daily life Death Education Enclosure Family and friends Future letters Health and illness Homesickness & separation Identity Ireland and America Migration News event Previous letters Recollections Religion Remittance Reunion Transportation Weather and seasons Work

JLC Topics 8 9 2 17 58 41 24 28 6 66 1 10 49 31 48 16 10 1 31 23

Freq. 0.23 0.26 0.06 0.49 1.66 1.17 0.69 0.8 0.17 1.89 0.03 0.29 1.4 0.89 1.37 0.46 0.29 0.03 0.89 0.66

Av. p/letter

320 691 53 699 4255 642 563 737 234 2269 136 476 1186 978 1854 539 213 60 774 885

Tokens

2.62 5.65 0.43 5.72 34.81 5.25 4.61 6.03 1.91 18.57 1.11 3.90 9.71 8.00 15.17 4.41 1.74 0.49 6.33 7.24

%

‘We work hard here’

91

would be interesting to compare different letter collections, taking into consideration socio-biographic variables, to see whether these topics are typical of migrant correspondence in general. Other topics, meanwhile, feature heavily because they provide a recurring structure to the letters, reflecting the rituals and demands of letterwriting itself. For instance, the topics ‘Future letters’ (any reference to letters that are about to be written or sent) and ‘Previous letters’ (any reference to letters already sent or received) are a significant part of Annie’s and Julia’s correspondence (and, arguably, migrant correspondence more generally), with 7.74% and 8.24% in the ALC and 5.25% and 9.71% in the JLC. Once all 74 letters were annotated for topics, it was possible to write a programme to automatically extract all instances of a particular topic to look more closely at the language using the IIVT and Sketch Engine. This chapter focuses on one of the most frequent topics: ‘Ireland and America’ to see what the language might reveal about how the sisters negotiated their new life in America with, as Diner puts it, ‘family loyalties’ in Ireland (1983, p. xiv).

5. Using the IIVT to Explore Words and Semantic Domains The IIVT was used to identify statistically significant items, specifically words and semantic domains, that are more likely to occur in the topic ‘Ireland and America’ when compared with all other topics combined. The IIVT offers a useful starting point as it allows the user to quickly and easily compare the linguistic features found in a subsection of the corpus (in this case the topic ‘Ireland and America’) against the rest of the corpus, minus the subsection. Additionally, it is possible to visualise all of the results on a single screen (i.e. it is possible to look at ‘interesting items’ for each of the topics in the ALC and JLC simultaneously). For the analysis using semantic information, the texts were examined using the online version of the UCREL Semantic Analysis System (USAS).2 This involved a small amount of pre-processing work to standardise some of the spelling variations found in the Lough letters. The complex semantic tags of USAS were split into their components of base, intensity and multi-word-expression. Only the base component is used in the analysis.3 To calculate significance, a biased odds ratio threshold of 3 was used together with a required minimum of 10 occurrences across the topics. Tables 6.4 and 6.5 show the results for statistically significant words and semantic domains in the topic ‘Ireland and America’ when compared with all other topics combined. The ‘(Freq.)/BOR’ column shows the number of times the word or semantic tag occurs in the topic ‘Ireland and America’, together with the biased odds ratio. For example, looking at Table 6.4, we can see that the word ‘country’ occurs 39 times in the ALC and has a biased odds ratio of 6.24—that is to say, the

Table 6.4 Statistically Significant Words in the ALC and JLC (Using a Biased Odds Ratio of 3 With a Minimum of 10 Occurrences) ALC

JLC

Word

(Freq.)/BOR

Word

(Freq.)/BOR

country Ireland terrible flowers doing several planting white thing snow every man people big here over winter

(39) 6.24× (34) 5.93× (15) 5.22× (10) 4.10× (10) 4.10× (10) 4.10× (10) 4.10× (10) 4.10× (22) 3.47× (18) 3.86× (67) 3.34× (20) 3.29× (26) 3.21× (33) 3.03× (159) 2.99× (39) 2.69× (67) 2.56×

ye mission country far weather down years here there went we

(11) 6.31× (14) 5.71× (10) 5.16× (11) 4.47× (19) 3.59× (16) 3.57× (20) 3.34× (117) 3.37× (24) 3.20× (29) 2.95× (56) 2.32×

Table 6.5 Statistically Significant Semantic Domains in the ALC and JLC (Using a Biased Odds Ratio of 3 With a Minimum of 10 Occurrences) ALC

JLC

Code

Semantic Domain

G2.1

law and order

(Freq.)/ BOR

Code

Semantic Domain (Freq.)/ BOR

(21) 9.42× M1/N6 moving, coming, going G1.1 government (50) 5.94× X6 deciding S7.1 power / organising (18) 5.92× W4 weather Z8/N5.1 pronouns / (12) 5.92× O4.6 temperature entirety; maximum G3 warfare, defence, (13) 5.13× Z2 geographical army, weapons names O1.1 substances and (11) 5.01× materials: solid I1.3 money: cost and (21) 4.50× price N4/T1.3 linear order / time: (15) 4.05× period W4 weather (53) 3.35× S2 people (26) 3.21× A1.7 constraint / no (26) 3.21× constraint F4 farming and (21) 3.06× horticulture Z8/Z6 pronouns / (32) 2.76× negative I1 money generally (32) 2.76× O4.6 temperature (29) 2.74×

(10) 5.16× (18) 3.88× (40) 3.28× (24) 3.20× (63) 3.04×

‘We work hard here’

93

word ‘country’ is 6.24 times more likely to occur in the topic ‘Ireland and America’ than in all other topics combined in the ALC.4 Focusing on Table 6.4, first of all, perhaps unsurprisingly for the topic ‘Ireland and America’, the noun ‘country’ is significant in both the ALC and JLC (with a BOR of 6.24× and 5.16× respectively). The adverb ‘here’ is also significant (2.99× in the ALC and 3.37× in the JLC), while ‘there’ is significant in the JLC (3.20×). Additionally, words to do with ‘weather’, ‘seasons’ and ‘farming’ are significant in both the ALC and JLC: ‘flowers’ (4.10×), ‘planting’ (4.10×), ‘snow’ (3.86×), ‘winter’ (2.56×) and ‘weather’ (3.59×). Some of these findings are further supported in Table 6.5 where the semantic domain ‘weather’ (W4) is significant in both the ALC and JLC, as well as the domains ‘farming and horticulture’ (F4) in the ALC and ‘temperature’ (O4.6) in the JLC. The results in Table 6.5 might also point to other sorts of national comparisons (in addition to ‘weather’ and ‘farming and horticulture’). In the ALC, for instance, the domains ‘law and order’ (G2.1), ‘government’ (G1.1), ‘power/organising’ (S7.1) and ‘warfare, defence, army, weapons’ (G3) are all highly significant, perhaps suggesting that Annie’s letters contain more political content. The domain ‘money: cost and price’ (I1.3) is also significant in the ALC and may gesture to content relating to the economic situation in Ireland and/or America. Julia’s letters, on the other hand, appear to focus more on physical spaces and geographic locations with domains such as ‘geographical names’ (Z2) and ‘moving, coming, going’ (M1/N6). While historical context will always influence what authors write about, the results in Tables 6.4 and 6.5 at least gesture to possible lines of inquiry which might provide useful insights into how the sisters conceptualised and contrasted their new lives in America with their old lives in Ireland. It is worth noting here what the role is of the IIVT and statistics more generally. When working with small corpora such as the ALC and JLC, which are even smaller given that we are really dealing with the topic subcorpora, the IIVT can only serve to make suggestions for possible lines of inquiry. In other words, the statistics do not tell us what is interesting, but rather they suggest places to look for things that might be interesting. The next stage, therefore, involves looking at some of the IIVT’s findings in more detail using the corpus tool Sketch Engine.

6. Using Sketch Engine to Explore Deictic Terms in the Topic ‘Ireland and America’ Section 5 showed that the words ‘we’, ‘here’ and ‘there’ were statistically significant in the topic ‘Ireland and America’ in the ALC and/ or the JLC, when compared with all other topics. While person deixis (such as the pronoun ‘we’) encodes the ‘speakers and hearers within a situational context and their relative distance from the speaker or writer’ (Jeffries and McIntyre, 2010, p. 158), place deixis (such as the

94

Emma Moreton

adverbs ‘here’ and ‘there’) encodes ‘the relative position in space of specific objects and entities’ (ibid., p. 157). A closer examination of these deictic terms—in the context of writing about Ireland and America— may, therefore, reveal something about how the sisters perceived their new lives in America (their situational context) and how they balanced this with their previous lives and their ongoing relationships with family back home in Ireland. The frequencies for ‘we’/‘here’/‘there’ are shown in Table 6.6 (the ‘Norm.’ column provides normalised frequencies per hundred words). The findings show that the pronoun ‘we’ is used slightly more by Annie (1.48) than Julia (1.23), and Julia tends to use the adverbs ‘here’ and ‘there’ slightly more than her sister (1.98 versus 1.17 for ‘here’ and 1.36 versus 1.28 for ‘there’). However, a closer examination of the language in context is needed to identify what these words are referring to. Specifically, in instances such as ‘we never had one here’ and ‘we used to like to go there’, who is the ‘we’ and where is the ‘here’ and ‘there’? Table 6.7 shows that the sisters typically use the pronoun ‘we’ to align themselves with one of five groups: i. Their family in America, e.g. ‘Dear Mother you need not ever be afraid when we all meet and talk about home that we will ever say you were the cause of Sending us away’ (ALC, 29 October 1891) and ‘I suppose no one cares to see us but we are not without friends and good neighbors’ (ALC, n.d.). ii. The local Irish Catholic community in America, e.g. ‘two week ago we celebrated our Pastors silver Jublee’ (ALC, 14 July 1918) and ‘we had a grand mission for a week I went twice a day and three times some days’ (JLC, 2 December 1889). iii. American citizens, e.g. ‘this country is doing a whole lot for ireland and we hope that before long congress will recognise the republic of Ireland’ (ALC, 21 March 1920); ‘I like you to read some of what the country is doing to help win the war we are all trying to do our bit’ (ALC, 6 May 1918); and ‘I hope James will have good success in his profession as Solisiter lawyers we call them here’ (ALC, 14 July 1918).

Table 6.6 Frequencies for ‘We’, ‘Here’ and ‘There’ (in the Topic ‘Ireland and America’) in the ALC and JLC ALC

we here there

JLC

Raw

Norm.

Raw

Norm.

67 53 58

1.48 1.17 1.28

28 45 31

1.23 1.98 1.36

‘We work hard here’

95

Table 6.7 References to ‘We’, ‘Here’ and ‘There’ in the ALC and JLC

we

here there

ALC

Raw

Norm. JLC p/100

Family in America Irish Catholic community in America American citizens Family in Ireland Local community in America

30 9

0.66 0.20

national (America) local (Winsted) world Ireland (local) Ireland (national) locations in America France Mexico existential ‘there’

29 23 1 14 6 5

0.64 0.51 0.02 0.31 0.13 0.11

3 1 29

0.07 0.02 0.64

10 6 12

0.22 0.13 0.27

Raw

Norm. p/100 0.31

7 Family in America 11 Irish Catholic community in America 1 American citizens Family in Ireland 2 Local community 7 in America national (America) 18 local (Winsted) 27

0.79 1.19

Ireland (local) 10 Ireland (national) 3 existential ‘there’ 18

0.44 0.13 0.79

0.48 0.04 0.09 0.30

iv. Family in Ireland, e.g. ‘you will find no change in me Mother when you see me again I hope we will spent a happy time together yet perhaps sooner than you think’ (JLC, n.d. December 1888); ‘I remember long ago when we were counting the days before Christmas’ (ALC, 3 April 1906); and ‘I have thought of Mother very much all through May I remember the prayers we used to say during May let me know do you pray as much now as when I was at home’ (JLC, 4 June 1894). v. The local community in Winsted, e.g. ‘our winter did not begin yet we have not any snow yet or any cold weather’ (ALC, 15 December 1891) and ‘but in summer we do often have terrible thunder showers’ (ALC, 3 March 1890). In general, Annie tends to align herself with her family in America (her sisters, husband, nieces, nephews, aunts and uncles)—there are 30 occurrences of this in the ALC (a normalised frequency of 0.66), but only 7 (0.31) in the JLC. Julia, in contrast, is much more likely to use the pronoun ‘we’ to align herself with the Irish Catholic community in America— a normalised frequency of 0.48 in the JLC versus just 0.20 in the ALC. The most noticeable difference between the two sisters, however, is the extent to which the pronoun ‘we’ is used to self-identify as an American citizen. While Annie uses ‘we’ in this way 10 times (a normalised frequency of 0.22), Julia does this just once (a normalised frequency of

96

Emma Moreton

0.04). Both sisters use the pronoun ‘we’ to align themselves with their family in Ireland (usually their mother and/or sisters Mary and Maggie). Typically, ‘we’ is used in this way when the sisters recount their childhood memories. Through this imagined world, based on shared experiences, Annie and Julia are able to bridge the psychological and physical distance between themselves and their family in Ireland. In terms of place deixis, in the JLC, the adverb ‘here’ most frequently refers to Winsted (the sisters’ main place of residence in America), with a normalised frequency of 1.19 in the JLC and 0.51 in the ALC. These occurrences are typically formulaic greetings: ‘We are all well here thank God’ (JLC, 9 March 1890), or comments about the weather: ‘There has been no rain here this fall and there is a great scarsity of water’ (JLC, 18 October 1891). Both sisters use the adverb ‘here’ to refer to America; however, Julia does this slightly more often than her sister (a normalised frequency of 0.79 versus 0.64). In both corpora, the adverb ‘there’ is primarily used to refer to Ireland, although in the ALC it can also refer to France (3 occurrences), Mexico (1 occurrence) and towns and cities within America (5 occurrences). When referring to Ireland as a country, the adverb ‘there’ has a normalised frequency of 0.13 in both corpora. When referring to local towns and villages in Ireland, particularly the sisters’ hometown of Meelick, the adverb ‘there’ is used more by Julia than Annie (a normalised frequency of 0.44 in the JLC versus just 0.31 in the ALC). By extending the context in which these deictic terms occur (the parameters were set at 150 characters), three key themes emerged: (a) Life in Ireland Is Easy; Life in America Is Hard Both Annie and Julia write about how difficult life is in America compared to what they perceive to be a relatively easy life in Ireland. Often this comparison centres around the theme of work (as in the following examples 1–3). In example (1) Julia’s mother and younger sister Mary are directly addressed (‘I think you two ought to be very comfortable’). Julia makes a direct comparison between her mother in Ireland and her aunt in America, suggesting that Julia’s mother is lucky to have one of her daughters at home to help with the running of the farm. In (2) and (4), Julia and Annie compare the work ethics of American citizens (themselves included) with their Irish counterparts, and in (3)—a letter to younger sister Mary—Annie comments on the work ethics of her sister Maggie (who also remained in Ireland). As with Julia’s letters to her mother, Annie passes judgement on Maggie’s behaviour, suggesting that such a relaxed attitude to work would be unthinkable in America. 1. I think she only fancies she may be sick. I was going to write to Mary but I will wait now till I get her next letter I think you two ought to be very comfortable there and us here working hard My Aunt can

‘We work hard here’

97

not keep one of her daughters to wait on her like you She has no body but Uncle George and I am sure Mary is better off and has more clothes than any one of us (JLC, 09 March 1890). 2. every thing now. I am sure you work hard but Lizze will soon be able to help you work that seen good We all work hard here All friends are well Lizzie was here last week. I was making a suit for her I do all their dressmaking and for a few friends so I assure you I work hard too. Thomas does not want to have but (JLC, n.d. 1889–1894). 3. our winter did not begin yet we have not any snow yet or any cold weather she ought to be working for herself and considering the wey things are at home now what she would do if she was here no one ever thinks of staying a week away from work unless they were sick or out of work the times has changed very much in a few years I suppose he likes to live in Meelick I think we all (ALC, 15 December 1891). 4. there is lot of work here for any one who has a family it is all together different from Ireland people here are always up early and has early breakfast and dinner sharp at twelve (ALC, 29 October 1891).

(b) Ireland Represents Tradition; America Represents Progress Another theme which is found in both Annie’s and Julia’s letters is the idea that Ireland represents tradition and ‘sameness’, while America represents progress and change. Notions of tradition versus progress are discussed with reference to various topics including industry (1), work (2), education (3), religion (4) and marriage (5), as well as topics such as parenting, childbirth and health. Interestingly, in the following instances, we see how the position of women outside Ireland, in the Diaspora, in effect offers a critique of womanhood in Ireland. 1. here the 2 of June it wont seem long now untill Alice will be also it must seem strange to have the mill all closed up perhap something else might be done there if that was only here it would be made good use of I hope John will be kept there for work well Dear Sister I hope you get good health and I hope you do not work yourself too hard is times any better there at the (ALC, 20 June 1906). 2. last week begins to look like spring but the frost is on the ground yet and the snow drifts is not all gone yet I suppose you have all the planting done at home before this time the people here never thinks of planting untill about the time Charley Conroy used to, I have not seen her all winter they can not very come out in winter the roads does be so bad we reed by the papers (ALC, 30 March 1893).

98

Emma Moreton

3. her grow big for I know there Lizzie grows big also in most all cases here where the parents need their childrens help they go to work at 14 years they are not allowed to work any younger here in the shops and some more if they are better off do nt send them to work for a little while longer. I must tell you some time about the work here and what the manufacture myself and I am (ALC, n.d.). 4. remember happy to hear Maggie and family are well where does she live at present please give her my loving regards I suppose you people over there do not fast in Lent any more. We do not have to here But we try to pray all we can. Well Dear Sister I will wish you all a very happy Easter with all the Joys and blessings attached to that holy season Thomas Joins with me in sending (JLC, n.d. 1889–1894). 5. I am but if it was Gods will for me have a family I might be still happier God know best, it is very hard to bring up a famely and bring them up good there are a great number of married people here dont have children but my husband thinks just as much of me as if I had half a dozen and is as good and kind as could be. Dear Mother you asked me if I thought Julie would ever go home I dont (ALC, n.d.).

(c) Ireland Is Home While themes (a) and (b) seem rather critical of life in Ireland, Ireland is, nevertheless, home for the two sisters. In the following example (1), Annie writes about her ‘great regard for all in the old homestead’ stating that it was ‘there’ (Meelick) that the sisters always wrote to, and in (2) and (3) Julia reassures her mother that she (and the family home) are not forgotten despite the passing of time and physical distance between them. In these occurrences, Annie and Julia textually perform their commitment to family in Ireland; while the physical distance may be great, through the act of remembering (or not forgetting), Ireland remains very much in their imaginations. 1. we could not live I read of the raise of wages all over Ireland and about the asylum also. you know I have always a great regard for all in the old homestead could to be sure it was always there we wrote I was very much interested in Fr Edmunds case of any of the papers I like the Lenster leader the best there is more of our county notes there in any of the others I was reading in a (ALC, 7 December 1919). 2. see your darling Julia once again, but this is worth writing about I am sending you two pounds Annie is sending you one pound and Liz is sending you ten shillings so you see you are not forgotten here

‘We work hard here’

99

although Liz is a great many years here I dont think I would forget you either if I was away so long Indeed I never could forget my darling Mother Winsted Conn December 20–84 (JLC, n.d. December 1888). 3. happy in having you to write to again this Christmas and I hope and trust in In God it will not be the last and I hope you will try and be very happy and enjoy yourself [ ] do not worry about us here I am always lonely and thinking of home at xmas you have Maggies home and you will not mind so much I hope she will soon get an nice place. I am sure she will not stop home all winter I (JLC, 14 December 1891).

7. Conclusion This chapter began by identifying topics in the letters of two sisters— Annie and Julia Lough—who migrated from Ireland to America in the late nineteenth century. It then used the IIVT to identify statistically significant words and semantic domains in one of the topics, ‘Ireland and America’, to see what is distinctive, peculiar or marked about the language of this topic when compared with all other topics combined. Finally, Sketch Engine was used to examine some of the language in more detail, focusing in particular on the use of person deixis (the pronoun ‘we’) and place deixis (the adverbs ‘here’ and ‘there’). Although this chapter examined very small datasets, some patterns and themes began to emerge. In both the ALC and JLC, the pronoun ‘we’ most frequently referred to the sisters’ family in America. However, the findings showed that Annie was much more likely than her sister to explicitly self-identify as an American citizen. In terms of place deixis, Julia used the adverb ‘there’ to refer to local places in Ireland (specifically the sisters’ hometown of Meelick) much more than Annie did. Indeed, Annie’s worldview appears to have been more wide-reaching than her sister’s, with the adverb ‘there’ being used to refer to France, Mexico and various locations within America, as well as Ireland. Additionally, while Julia mostly used the adverb ‘here’ to refer to her local environment (Winsted), Annie mostly used it to refer to America. In short, Julia’s deictic centre (see Stockwell, 2002; McIntyre, 2006; Jeffries and McIntyre, 2010) appears to be grounded in the Irish Catholic community in Winsted; Annie’s, on the other hand, appears to be grounded more broadly in America and American society as a whole. It is worth pointing out, however, that while most of Annie’s letters were addressed to her younger sister Mary, Julia wrote mainly to her mother. This would certainly have influenced the content of the letters and the way in which the two sisters talked about their experiences in America. In both the ALC and the JLC, America represented hard work, progress and change, often discursively constructed in direct contrast to the perceived easy life and traditional values of Ireland. Nevertheless,

100

Emma Moreton

Ireland, for both sisters, ultimately represented home and their concern for, and commitment to, family in Ireland was expressed in various ways. Through recounting memories and past experiences, Annie and Julia projected a deictic centre in which both author and recipient were reunited in their shared past. In so doing, the sisters demonstrated a closeness to their family in Ireland, whilst simultaneously constructing new lives in America. The focus of this chapter was narrow, examining just one collection of correspondence from the much larger archive of 5,000 migrant letters held by Kerby Miller. But through repeating the process I have described here (tagging for topic and using corpus-based approaches to analyse the language of those topics), a more comprehensive lexicogrammar of my key topics may begin to emerge, providing a fuller picture of the language and functions of migrant correspondence. Equally too, of course, this further research may show that the linguistic features and themes I have identified here need to be expanded or refined as other, more typical ones emerge. Additionally, comparing migrant letters by authors from a range of socio-historical, economic and cultural backgrounds may provide new insights into how different individuals and groups conceptualised ‘place’. And, indeed, it is worth considering that the discourses and topics that do not emerge may be as telling as the ones that do.

Notes 1. I am indebted to Kerby Miller for the information he generously provided relating to the Lough family. 2. Rayson, P. USAS Online English Tagger. Available from: http://ucrel.lancs. ac.uk/usas/tagger.html. 3. I would like to thank Chris Culy (www.chrisculy.net) for preparing the Lough data for analysis using the IIVT. 4. Items highlighted in italics are ‘borderline significant’ (in terms of the parameters that have been set)—i.e. they have a biased odds ratio of 2.5 or more.

References Amador Moreno, C. P. and McCafferty, K. (2015) ‘“Sure this is a great country for drink and rowing at elections”: Discourse markers in the Corpus of Irish English Correspondence, 1750–1940’, in Amador Moreno, C. P, McCafferty, K. and Vaughan, E. (eds.) Pragmatic markers in Irish English. Amsterdam: John Benjamins, pp. 270–291. Bruce, S. U. (2006) The harp and the Eagle: Irish-American volunteers and the Union Army, 1861–1865. New York: New York University Press. Corrigan, K. P. (1992) ‘“I gcuntas De muin Bearla do na leanbhain”: Eisimirce agus an Ghaeilge sa naou aois deag’, in O’Sullivan, P. (ed.) The Irish world wide. Leicester: Leicester University Press, pp. 143–161. Culy, C. (2016) ‘Interesting items visualisation tool’, Available at: http://vistola. pythonanywhere.com.

‘We work hard here’

101

Diner, H. R. (1983) Erin’s daughters in America. Baltimore, MD: The Johns Hopkins University Press. Elliott, B. S., Gerber, D. A. and Sinke, S. (eds.) (2006) Letters across borders : Epistolary practices of international migrants. London: Palgrave Macmillan. Emmons, D. M. (1990) The Butte Irish: Class and ethnicity in an American mining town, 1875–1925. Urbana: University of Illinois Press. Fitzpatrick, D. (1994) Oceans of consolation: Personal accounts of Irish migration to Australia. Cork: Cork University Press. Jeffries, L. and McIntyre, D. (2010) Stylistics. Cambridge: Cambridge University Press. Kilgarriff, A. and Kosem, I. (2012), ‘Corpus tools for lexicographers’, in Granger, S. and Paquot, M. (eds.) Electronic lexicography. New York: Oxford University Press, pp. 31–56. Koos, G. (2001) ‘The Irish hedge schoolmaster in the American backcountry’, New Hibernia Review, 5, pp. 9–26. McCafferty, K. and Amador Moreno, C. P. (2012) ‘“I will be expecting a letter from you before this reaches you”: A corpus-based study of shall/will variation in Irish English correspondence’, in Dossena, M. (ed.) Letter writing in late modern Europe. Amsterdam: John Benjamins, pp. 179–204. McCafferty, K. and Amador Moreno, C. P. (2014) ‘“If you write soon I shall get it & will reply at once” The spread of first-person future will in Irish English’, English Language and Linguistics, 18, pp. 407–429. McIntyre, D. (2006) Point of wiew in plays. Amsterdam and Philidelphia: John Benjamins. Meaney, G., O’Dowd, M. and Whelan B. (2013) Reading the Irish woman: Studies in cultural encounters and exchange, 1714–1960. Oxford: Oxford University Press. Miller, K. A. (1985) Emigrants and exiles: Ireland and the Irish exodus to North America. New York: Oxford University Press. Miller, K. A. (2008) Ireland and Irish America: Culture, class, and transatlantic migration. Dublin: Field Day. Miller, K. A., Doyle, D. N. and Kelleher, P. (1995) ‘For love and liberty: Irish women, migration and domesticity in Ireland and America, 1815–1920’, in O’Sullivan, P. (ed.) The Irish world wide. Leicester: Leicester University Press, pp. 54–61. Miller, K. A., Schrier, A., Boling, B. D. and Doyle, D. N. (2003) Irish immigrants in the land of Canaan: Letters and memoirs from colonial and revolutionary America, 1675–1815. New York: Oxford University Press. Montgomery, M. (1995) ‘The linguistic value of Ulster emigrant letters’, Ulster Folklife, 41, pp. 26–41. Nolan, J. A. (1989) Ourselves alone: Women’s emigration from Ireland, 18851920. Lexington: University Press of Kentucky. Noonan, A. J. M. (2011) ‘Oh those long months without a word from home’, Migrant letters from mining frontiers. The Boolean: University College Cork. Available at: http://publish.ucc.ie/boolean/pdf/2011/00/31-Noonan-2011-00en.pdf. Rayson, P. (2009) Wmatrix. Lancaster: Lancaster University. Available at: http:// ucrel.lancs.ac.uk/wmatrix/. Stockwell, P. (2002) Cognitive poetics: An introduction. London: Routledge.

102

Emma Moreton

Van Hattum, M. (2014) ‘“Queensland for rver & Augus un ballybug go braugh”: The expression of identity in nineteenth-century Irish emigrant letters’, in Evans, D. (ed.) Language and identity: Discourse in the world. London: Bloomsbury, pp. 105–122. Zubin, A. D. and Hewitt, L. E. (1995) ‘The deictic centre: A theory of deixis in narrative’, in Duchan, J. F., Bruder, G. A. and Hewitt, L. E. (eds.) Deixis in narrative: A cognitive science perspective. Hillsdale, NJ: Lawrence Erlbaum, pp. 1–27.

7

The Devil Has All the Best Tunes An Investigation of the Lexical Phenomenon of Brexit Lesley Jeffries and Dan McIntyre

1. Introduction On 23 June 2016, voters in the United Kingdom and Gibraltar cast their votes in a referendum on whether Britain should remain a member of or leave the European Union. The referendum followed months of fierce campaigning by both ‘Leave’ and ‘Remain’ supporters. The result was that 51.9% of voters were in favour of Britain leaving the EU. The referendum was purely advisory in status, yet despite the enormous complexity and potential long-term social, economic and cultural damage to the UK of doing so, on 29 March 2017, British Prime Minister Theresa May initiated the EU withdrawal process, or Brexit as it has come to be described. Brexit continues to be a hugely divisive issue in British society. Our focus in this chapter is not on the process itself, but on the linguistic issues inherent in how the Leave and Remain campaigns were promoted. Professional linguists are inevitably interested in phenomena like Brexit, where a new lexical item appears and becomes the label of choice for an unclear and shifting referent. Research into socio-political keywords (Jeffries and Walker, 2017) has demonstrated that everyday words often take on particular political significance, albeit rather empty of meaning, and become shorthand for a complex but poorly defined set of semantic properties. Jeffries and Walker’s (2017) previous work, then, has shown how words like ‘choice’ and ‘reform’ take on specific socio-political significance but with a vague referential meaning during relatively narrow time periods. Here, we see a similar phenomenon play out over an even shorter time period and with a new lexical item, rather than an existing one, playing the main part. Our starting point in this investigation was the intuition that those who wanted the UK to leave the EU had an advantage in the label ‘Brexit’, since its composition and usage displays a degree of linguistic creativity that was lacking in the Remain campaign, and that consequently resulted in comparatively greater foregrounding of the Leave position. This chapter aims to marshal linguistically based argument and corpus-based evidence to try to persuade the reader that this is a reasonable view to take.

104

Lesley Jeffries and Dan McIntyre

We begin by discussing the morphological structure of the term ‘Brexit’, its competitor term ‘Bremain’ and the related terms coined to describe adherents of both the Leave and Remain positions. We then move on to discuss usage of the term ‘Brexit’ particularly, taking a corpus-based approach to examining its function and ideological effects in context. In this respect, our focus is very much inspired by Michael Toolan’s approach to the analysis of style and meaning, an approach that is often eclectic in its analytical methods and which is demonstrated in his edited collection Language, Text and Context (Toolan, 1992). Like Toolan, we are as interested in the style of non-literary texts as literary ones and we have consistently argued that non-literary style should attract equal attention in stylistics (Jeffries and McIntyre, 2010); we focus here primarily on newspaper discourse. We are also convinced, as all stylisticians are, of the interdependence of theoretical and applied research (see McIntyre and Price, 2018); our initial analysis of the morphological structure of Brexit and its related terms serves to contextualise and underpin our later corpus-based analysis of usage. Finally, as our analysis shows, context is key to understanding the relationship between linguistic form and meaning.

2. The Morphology of Brexit From a stylistic perspective, the referendum campaign was a rich source of lexical innovation and creativity. Moreover, these linguistic innovations gave rise to significant ideological effects which, we will argue, may have had a substantial impact on the respective outcomes for the Leave and Remain campaigns. Although it would be difficult to test such a causative relation directly, we will make a case for the effectiveness of Brexit as a label under which people supporting the Leave campaign could gather. In this section we discuss some of the neologisms that were coined during the referendum campaign and the word formation processes that underpinned them. We then identify the associated ideological effects of these neologisms, arguing that the Leave campaign’s stylistic choices were ultimately more impactful than those of the Remain campaign. 2.1 Brexit and Bremain Both the Leave and Remain campaigns employed neologisms as shorthand terms for the referendum result they were seeking. Leave took ownership of Brexit1 (henceforth, [1]) while Remain used Bremain (henceforth, [2]). Both [1] and [2] are linguistic blends demonstrating graphemic and phonemic overlap between their constituent source words. Term [1] has a clear graphemic structure (albeit with two variant pronunciations), while [2] has an ambiguous graphemic structure. They are both what Kemmer (2003) terms sequential blends; that is, blends in which one element from one source word is appended to one element from another.

The Devil Has All the Best Tunes

105

In [1], the first source word (Britain’s or British) is shortened to its initial two letters while the second (exit) is not: Graphemic structure

Pronunciation

[1]
+

[1a] [ˈbɹɛkˌsɪʔ] [1b] [ˈbɹɛgˌzɪʔ]

In the case of [2], however, the two possible graphemic structures are: Graphemic structure

Pronunciation

[2] [2a]
+ [2b] +

[ˌbɹɪˈmeɪn] [ˌbɹɪˈmeɪn]

It is also the case that the two-word noun phrase that is blended into one lexeme in [1] is syntactically complete: Britain’s exit or British exit. In [2], however, this is not the case: *Britain/British remain. The only way to make sense of [2] is to see it as a blend of two lexemes taken from a sentence with an elided element—something like Britain should remain. The importance of designated terms to describe the processes favoured by the two campaign groups is that the act of naming a concept produces an existential presupposition about the existence of that concept (see Levinson, 1983; Simpson, 1993; Jeffries, 2010a). By naming the concept of Britain leaving the EU, we bring into existence the possibility of this course of action happening, however difficult or ill-advised this may be. The same is true of naming the process of Britain remaining in the EU. However, we would argue that [2] (Bremain) was less impactful than [1] (Brexit) for a number of reasons: 1. The second source word of [1] (exit) can function as both a noun and a verb. Consequently, it is possible to see Brexit as a nominalisation of a verbal process. In the case of [2], however, the second source word (remain) functions only as a verb. The potential for nominalisation to create ideological effects by reifying a process and simultaneously backgrounding its actor(s) and goal(s) (see Fairclough, 1989 and Fowler, 1991) is consequently not possible with [2], as a result of the structural ambiguity referred to earlier. Note that Fontaine (2017) has examined a large number of tokens of Brexit to determine the extent to which Brexit was actively used as a grammatical metaphor (Halliday, 2003, p. 69) by retaining some of the process-related meaning from its verbal origins, and to what extent it had already become, in May 2016, a full nominal, despite its origins. Her findings were that in the first three years of its usage, there was still a strong verbal/process element in its usage, though there was also evidence of some full nominal use too.

106

Lesley Jeffries and Dan McIntyre

2. The stress in [1] is on the first syllable, which foregrounds the blended element of the word, thereby drawing attention to the creativity of the neologism. And as experimental research into foregrounding has shown (see Van Peer et al., 2007), foregrounded elements of language are more memorable than conventional elements and, in cognitive terms, function as attractors (Stockwell, 2009). By contrast, in [2] the stress is on the second syllable, which is simply the second syllable of the second source word (remain). The first syllable, in which the blend occurs, is unstressed and the creativity of the neologism is consequently backgrounded. Moreover, in English, most polysyllabic words that begin with
have their stress on the first syllable (e.g. breathlessness, breakaway, brilliant, broadcast) rather than the second, meaning that Bremain, with its stress on the second syllable, is unnatural for speakers of British English. Although this is also, thereby, foregrounded, the lack of ease of pronunciation as well as the larger departure from regular neologisms in English make it less likely to be taken up by popular campaigns than the more comfortable, but nevertheless still foregrounded, Brexit. Additionally, the fact that the first syllable of Bremain is unstressed may even mean that in conversation hearers do not recognise it as a blend, instead processing it simply as remain. 3. The blend in [2] is likely to be more difficult to comprehend than the blend in [1]. Making sense of [2] requires reconstructing a sentence with an elided element, whereas interpreting [1] requires only the reconstruction of a two-word noun phrase. 4. The fact that exit is a dynamic verb while remain is a stative one means that, in effect, [1] and [2] connote action and inaction respectively. It is easy to see how the campaigns might as a result have been associated (at least for some voters) with dynamism and passivity. The significance of this is that both campaigns were trying to attract floating voters in particular. And those voters who were sceptical of the value of EU membership tended to take that view precisely because they saw the UK as being too passive in its dealings with the EU (such a view was no doubt fuelled by years of unsubstantiated scare stories in the UK press about the EU dictating that bananas should be straight rather than bendy, and other such misinformation2). What is suggested from the previous analysis is that, on all measures, Brexit is a more impactful blend than Bremain in terms of its potential to attract the attention and support of floating voters. We would argue that as a consequence of both the congruence of foregrounding in Brexit and the linguistic factors working against Bremain (e.g. its unusual stress pattern and the difficulty of parsing the blend), Bremain failed to gain traction in public discourse.

The Devil Has All the Best Tunes

107

2.2 Brexiters, Brexiteers, Remainers, Remoaners and Remainiacs Once the campaigns for and against leaving the EU were established, it was not long before the advocates of each position were being referred to in the media by derivational forms. A Brexiter is someone in favour of Britain leaving the EU. A Remainer is someone who favours the opposite. Both words are formed via the addition of an suffix on the root morpheme. However, while Brexiter is a neologism derived from Brexit, Remainer (albeit without word-initial capitalisation and with the meaning of a person who remains in general terms) has a much longer history, being first attested in 1565 (OED). The consequence of this is that Remainer is unlikely to be considered as foregrounded while Brexiter almost certainly is. Both Brexiter and Remainer are semantically neutral insofar as they convey only the course of action that the referent is committed to, with no particular value judgement attached. The same cannot be said of the many other terms that have been coined to refer to the adherents of both the Leave and Remain campaigns. Brexiteer is another derivational neologism, formed via affixation of the suffix to Brexit. The suffix comes from French , and is used to form nouns that denote people, where ‘the usual sense is “one who is concerned with”, or “one who deals in”’ (OED). However, unlike Brexiter, Brexiteer is arguably ideologically loaded as a result of its use in other contexts. Table 7.1 shows a list of human agent nouns with suffixes from the Corpus of News on the Web (Davies, 2013), henceforth the NOW Corpus3 (note that Brexiteer is among the results). These were extracted from the first 100 hits for the search term *eer.[nn*].4 Table 7.2 shows the top 22 adjective collocates that have MI (mutual information) scores above 3 (i.e. denoting statistical significance and collocational strength) of each of the nouns in Table 7.1 with the exception of Brexiteer. The collocational span used is a nine-word window of up to four words to the left and right of the node word. It should be noted that these collocates are derived from the Corpus of Historical American English (COHA) since the British National Corpus of Contemporary British English is not large enough to generate results for each of the target nouns. Despite COHA sampling historic American English, in our view these results are valid since (i) each of the words in question is also found in British English, and (ii) the historical perspective that COHA

Table 7.1 Human Agent Nouns With Suffixes From the NOW Corpus Engineer, Volunteer, Pioneer, Overseer, Auctioneer, Mountaineer, Puppeteer, Buccaneer, Balladeer, Privateer, Brexiteer, Racketeer, Gazetteer, Marketeer, Charioteer, Musketeer, Profiteer, Mutineer, Scrutineer, Rocketeer, Bioengineer, Wagoneer, Pamphleteer

Table 7.2 Top 22 Collocates of Human Agent Nouns With Suffixes From COHA, by Collocational Strength (MI 3+) Noun

Collocates

Engineer

transmission-line, acoustical, power-transformer, waxhouse, enable, defense-contracting, conversation-starved, Bessarabian-born, Berkley-trained, block-anchoring, aircraft company, American, brake-development, boywanted, bushy-mustachioed, bydrographical, even-tempered, ex-steamboat, federal-highway, good-fellow-well-met hand-irked, ex-aerospace, at-through, career-sample, Bisayan, Indo-Gallic, material-even, pink-clad, provisional army, citizen-training, defense, French-gray, over-the-transom, united, white-starched, golden-flowered, ex-peace, ex-campaign, dead-of-night, cartridge paper commercial-aviation, expatriated, energy-drink, dancerchoreographer, custom-beauty, citizens-to-be, brown-amber, Brobdignagian, business-dramatic, all-adventuring, freedistribution, ground-to-ground, hypernoisy, in-the-moment, italan-born, juicy-diarist, light-cooking, meat-and-sausage, nonengineered, phosphate-keyed pen-nit, flu-ashed, brutal, lucre-loving, uzzi, spoiled-child, half-earnest, black-skinned, grazier, cybernetic, bulletheaded, self-elected, leaved, two-pound, thread-like, inhumanly, handpicked, ill-used, asinine, deadlier dinner-master, lraordinary (sic), up-coming, raucous-voiced, others, loud-tongued, silver-fox, boiler-room, gloomylooking, squared-off, slave-trader, pastel-colored, twangy, sloven, knock-down, lithographic, vituperative, u-shaped, semi-annual, twenty flat-terrain, fern-bedded, foxy-eyed, jet-skinned, Jacksonarea, apraxic, mean-dog-mean, jug-band, Punch-and-Judy, tolidest, stoutest-hearted, aist, down-on-his-luck, bushybearded, outdoor-loving, home-state, singular-looking, asturian, tyrolese, freedom-loving script, high-tension, palsied, fishy, yawning, drunken, skilled, romantic, mad, upward, cruel, apparent, brilliant, professional, fun, worse, chief, able, live, national well-sailed, disease-splotched, hoysted, hand-and-glove, tripledecked, buccaneering, black-bearded, knavish, whiskered, blood-thirsty, unbuckled, mutinous, ruthless, heretical, licentious, massy, ever-increasing, pampered, vengeful, satirical gold-mining, esthetic, cerebral, moody, middle-aged, lone, shy, revolutionary, musical, supreme, highest, popular, lower, blue, French, beautiful, late, political, general thirty-gun, malartic, blue-chinned, uncommissioned, 24-pound, four-pound, Indiaman, well-aimed, selfcomplacent, Fenian, hour-long, unrecognised, posthumous, Colombian, cursory, belligerent, disarmed, sagacious, sensual, Caribbean

Volunteer

Pioneer

Overseer

Auctioneer

Mountaineer

Puppeteer Buccaneer

Balladeer Privateer

(Continued)

The Devil Has All the Best Tunes

109

Table 7.2 (Continued) Noun

Collocates

Racketeer

policy-slip, run-of-themine, flower-show, two-billion-dollar, hot-oil, biggish, pasty-faced, nonsectarian, pearl-handled, black-bearded, man-eating, out-, ex-convict, hollow-eyed, feeble-minded, small-time, mannered, substandard, corrupt, reputed Biblical, national leftist-turned-free, state-hating, self-begotten, hutt-like, bookswop, glinty, honey-tongued, future-gazing, blue-skinned, pony-tailed, glad-handing, doctrinaire, well-intended, savvier, anti-tax, business-as-usual, self-interested, Tokyobased, Seattle-based, indefatigable rearer, farm-boy, dextrous, liveried, scintillating, knotty, humbled, winged, sweating, gaunt, worthless, melancholy, reluctant, glowing, aged, finest, shouting, glorious, naked, experienced seneschal, pesky, plumed, seasoned, tempting, clumsy, mature, graceful, fatal, armed, favorite, handsome, broken, military, French, young vulturous, swinish, hard-bitten, unwitting, canny, blackest, blatant, cocky, teen-age, illicit, khaki, aspiring, outraged, systematic, convicted, protected, so-called, helpless, prominent, fat hanging-work, shanghaied, congenital, tipsy, roan, brawny, audacious, eventful, surviving, relieving, impracticable, accusing, dire, tremulous, stony, slippery, resolute, convicted, aggressive, prompt No collocates foremost No collocates rust-rotted, sloping, fake, brown, red, black, sure ex-professor, anti-militarist, owi, street-corner, readier, pygmy, cinematic, prolific, slashing, tory, fanatic, pungent, manifold, anonymous, irresponsible, questionable, controversial, revolutionary, eloquent, furious

Gazetteer Marketeer

Charioteer

Musketeer Profiteer

Mutineer

Scrutineer Rocketeer Bioengineer Wagoneer Pamphleteer

provides gives an insight into the long-standing semantic associations of each word, which we might reasonably assume to contribute to readers’ individual lexical schemas. Table 7.2 begins to shed light on why it is that we might intuitively feel Brexiteer to have negative associations. While we cannot attribute a particular semantic prosody (Louw, 1993) to Brexiteer (on the grounds that semantic prosodies belong to units of meaning rather than individual words [Sinclair, 1998]), what we can observe is that all of the nouns from Table 7.1 have among their collocates some that are semantically negative (e.g. ‘Brobdignagian’, ‘spoiled-child’, ‘self-elected’, ‘cruel’, ‘mutinous’, etc.) and that connote carelessness and a general cavalier attitude.

110

Lesley Jeffries and Dan McIntyre

This is not to say that there are not positive or neutral collocates too; there are. However, the presence of the semantically negative collocates explains any interpretation of Brexiteer which sees it as inherently negative. It may well be the case, of course, that being described as a Brexiteer has come to be seen as a badge of honour to Leave supporters, with a certain amount of covert prestige attached. Brexiteer is a derivational noun whose root is a blend. Remoaner, on the other hand, is a blend whose root is a derivational noun. Unlike Remainer, Remoaner displays a high degree of creativity in its formation. Remain and moan are blended to create remoan, a verb meaning something like ‘to remain in a mindset of moaning [about the outcome of the referendum]’. An suffix is appended to create a noun from the verb. Remoaner is thus what Kemmer (2003) describes as an intercalative blend; that is, a blend in which ‘the sounds of one source word are interspersed between the other’ (Kemmer, 2003, p. 72). In this case, moan is interspersed with Remainer. In this respect, there is a greater degree of linguistic creativity and consequent foregrounding in Remoaner than in Brexiteer. The same might also be said of Remainiac, which is a sequential (i.e. non-intercalative) blend in which the two source words (remain and maniac) are present in their entirety, at least in terms of their constituent phonemes. Of the five terms (Brexiter, Brexiteer, Remainer, Remoaner and Remainiac), Brexiteer is by far the most prevalent in the NOW Corpus, with 831 instances. The majority of these are adjectival references (e.g. ‘staunch Brexiteer Jacob Rees-Mogg MP’) with few instances of firstperson self-reference. The same appears to be the case for Remainer (601 instances, primarily third-person references) and Brexiter (195 instances, primarily third-person references). Four of the ten instances of Remainiac in the NOW Corpus are self-references (e.g. ‘Alastair Campbell told the crowd he was himself a “Remainer, Remoaner, Remainiac” who would not stop talking about the issue’). The proportion of self-references is much lower for Remoaner, at just four out of a total of 94. These results come from a corpus of news texts. There are no instances of any of the five words in the Corpus of Global Web-Based English (GloWbE) or the Corpus of Online Registers of English (CORE). It would seem reasonable to suggest, then, that despite the fact that Brexiter, Brexiteer, Remainer, Remoaner and Remainiac are prevalent in public discourse, they primarily function as media shorthand for stakeholders in the Leave and Remain campaigns. In the next section, we turn our attention to corpus evidence of how the terms ‘Brexit’ and ‘Bremain’ in particular have been used.

3. Brexit and Bremain: Corpus Evidence of Usage In addition to bringing insights from an analysis of morphology and grammar to bear on the apparent power of the term ‘Brexit’, when

The Devil Has All the Best Tunes

111

compared with the term ‘Bremain’, we used corpus linguistic techniques to see how the words have regularly been used since their first appearance. Examining the lexical phenomenon of Brexit using corpus techniques may appear relatively easy at first sight, since it is a word that has existed for only a short while, is clearly a blend (with consequent foregrounded status) and has no history beyond the political events of the UK leaving the European Union. However, the question of what data to investigate in order to examine the behaviour of Brexit as a lexical item is not an easy one. The options include conventional news reporting, online news, social media debate, political discussion, commentary on broadcast media and so on. With such studies as this one, there is a practical imperative, which is that the study needs to be manageable, particularly if the corpus is being used mainly to retrieve representative examples of the socio-political keyword in context for qualitative analysis. There is a Brexit corpus in existence, available via Sketch Engine,5 consisting primarily of data from Twitter that contains the term. However, when we started searching this corpus for patterns of usage, it became clear that normal processes of corpus analysis were not adequate for this kind of data, since there are very many repeated occurrences as a result of the practice of retweeting. Whilst this may be of interest in itself, the potential for skew caused by repeated data makes it difficult to be clear about the prevalence of particular patterns of usage. For this reason we have made use of a range of other sources of information to try to determine the behaviour of Brexit and Bremain in order to test our intuition about their functions. This has included searching the ProQuest International Newsstream database6 for information on usage in mainstream UK newspapers, as well as examining the NOW Corpus referred to in section 2. The option to construct and examine a bespoke corpus was rejected in favour of searching existing data sources for a wider range of evidence to answer our research questions. Since ProQuest and NOW are made up of different kinds of data and allow different kinds of searches, we have used them where relevant to answer the questions we had which we saw as fundamental to testing our intuition that Brexit was advantageous to the Leave campaign. These questions are as follows: 1. When did Brexit (and Bremain) first begin to be used in the press? 2. How comparable are the frequencies of use of these two terms? 3. Has the frequency of occurrence of Brexit peaked at the time of writing (December 2017)? 4. What is (are?) the referent(s) of Brexit? 5. When did soft and hard Brexit start to be talked about? 6. What are the potential ideological effects of the words ‘Brexit’ and ‘soft/hard Brexit’?

112

Lesley Jeffries and Dan McIntyre

3.1 Earliest and Most Frequent Uses of Brexit and Bremain To answer question 1 (When did Brexit (and Bremain) first begin to be used in the press?), we consulted ProQuest and searched for the item in the UK-based press. There is some evidence7 that Brexit was based on the word Grexit (see note 1) which was a response to the financial crisis in Greece in 2009 which led to a fear that it might leave the EU’s monetary union, if not the EU itself. Even so, the British version started occurring earlier in the press than we might expect and a few years before the referendum on Britain leaving the EU was announced.8 Although it became ubiquitous in the year of the referendum itself (2016), the earliest occurrences in the British mainstream press (according to ProQuest), compared to the appearance of Bremain, were as shown in Table 7.3. According to ProQuest, then, in the print media it was the broadsheets which were first to adopt the term, though the tabloid newspapers also adopted it well before UK Prime Minister David Cameron set the date for the referendum, on 20 February 2016. As debate about the possibility of a referendum heated up, it was to be expected that the emerging campaigns on either side would try to persuade voters by the use of labels which might give them a linguistic and political advantage. However, the competing term ‘Bremain’ was not used alongside Brexit until well into the campaign and in most cases only after the referendum was announced. This information alone points to the notion that the naming of the process/concept of leaving the EU was established well in advance of the referendum campaign. We would argue that in addition to the linguistic advantages discussed in the previous section, the early establishment of Brexit as a recognisable nominal form, despite its referential vagueness (which we will discuss later), lent it credibility compared to Bremain. We examine the contextual power of Brexit later, but here we answer question 2 (How comparable are the frequencies of use of these two terms?) and question 3 (Has the frequency of occurrence of Brexit peaked at the time of writing (December 2017)?) by examining the numbers of Table 7.3 Earliest Occurrences of Brexit and Bremain in British National Press in Order of Occurrence Newspaper

Earliest Brexit

Earliest Bremain

Financial Times Telegraph Guardian Independent Times Sun Mail Express Mirror

18/10/12 20/10/12 15/11/12 24/11/12 26/11/12 10/05/13 18/06/13 09/04/14 02/06/15

03/03/16 15/04/16 26/01/16 06/05/16 03/02/16 29/02/16 05/12/15 03/05/16 09/10/15

The Devil Has All the Best Tunes

113

hits in ProQuest and in the NOW Corpus. In the case of ProQuest, the hits represent numbers of articles, not numbers of Brexit tokens but we can see in Table 7.4 that the months around the referendum itself (June and July 2016) were the apparent high water mark for Brexit in news reporting. Although the negotiations for Brexit are continuing at the time of writing, and the word ‘Brexit’ is in common parlance, it nevertheless looks from this data as though there is a falling off of press reporting using this word. However, we may yet see a second peak before the final date of the UK’s departure from the EU and broadcast news might present a rather different story. Nevertheless, by comparison, Bremain hardly features on this measure at all. The numbers peak at the same point (in June 2016) but are otherwise in single or double figures only. This comparison of the frequencies of the two main campaigning labels in the same data and over the same time period seems to us to support the notion that Brexit was much more evident in the run up to the referendum and beyond than Bremain which simply did not take hold. In the NOW Corpus, there is a similar picture, with Brexit far outdoing Bremain in frequency. The figures for tokens of Brexit include 291,863 tokens from the beginning of 2015 to the second half of 2017. The results in Table 7.5 are based on tokens per million words to produce comparable figures, since there is no consistency of corpus size per year in this data. The figures for Bremain, again, are so low that they barely register. At their highest, in 2016, they reached a maximum of 176 tokens (0.04 per million words), so again, the alternative to Brexit is almost invisible. 3.2 The Referent(s) of Brexit and the Role of Brexit Means Brexit Question 4 (What is (are?) the referent(s) of Brexit?) is prompted by the prior work of Jeffries and Walker (2017) on socio-political keywords, in which it was observed that the pattern of such keywords is of great significance in political debates, despite their having relatively undefined meaning. There is a paradoxical pattern of behaviour of these words which are frequently unmodified and are thus intended to be understood without further elaboration. They are therefore symbolic of something important and political without having a clear denotation. Our question, then, is to ask whether Brexit belongs in this category of socio-political keyword. There are many potential ways to try to answer this question, including observing its very frequent use as an unmodified head noun, but in this case we decided to focus particularly on the well-known mantra used by British Prime Minister Theresa May as she embarked on the process of disentangling the UK from the EU: ‘Brexit means Brexit’. In response to questions from other parliamentarians and from journalists asking what Brexit would entail, she consistently repeated this sentence. Others then took it up, sometimes with ironic intent and sometimes in all seriousness. The first recorded reporting of the use of this sentence in

1669 19,950

Jan

3 3

2016 2017

Bremain

2016 2017

14 6

Feb

5157 19,198

Feb

8 12

Mar

5798 25,687

Mar

31 3

Apr

6712 19,542

Apr

47 5

May

9957 19,757

May

324 3

Jun

39,571 24,793

Jun

55 1

Jul

36,336 16,902

Jul

4 2

Aug

16,651 13,682

Aug

Brexit Bremain

0.16 0

Jan–Jun 2015

0.16 0

Jul–Dec 2015 24.79 0.04

Jan–Jun 2016 24.78 0.01

Jul–Dec 2016

Table 7.5 Number of Tokens of Brexit and Bremain per Million Words in the NOW Corpus

Jan

Brexit

13 0

Sep 3 2

Oct

21,140 17,897

Oct

26.18 0

Jan–Jun 2017

18,558 17,556

Sep

8 0

8 2

Dec

18,015 16,847

Dec

25.75 0

Jul–Dec 2017

Nov

24,122 17,623

Nov

Table 7.4 Number of Articles in ProQuest International Newsstand9 Containing Tokens of Brexit and Bremain in Each Month of 2016 and 2017

The Devil Has All the Best Tunes

115

ProQuest is from 7 June 2016 as May put herself forward to be Prime Minister and June 2016 is the high point when ProQuest reports it in 1,042 articles in total. Table 7.6 shows a similar high point in the number of tokens of this clause in the NOW Corpus: Table 7.6 Raw Numbers of Occurrences of Brexit means Brexit in NOW

Brexit means Brexit

Jun 2016

Jul–Dec 2016

Jan–Jun 2017

Jul–Nov 24 201710

21

585

179

75

In the earliest uses of this phrase by Theresa May as she put herself forward to be the party leader (and by default Prime Minister), it was reported straightforwardly, all 21 tokens in the NOW Corpus in June 2016 being speech reporting of the following kind: •





‘Like other candidates, May rules out a second referendum and said that “Brexit means Brexit.” She also ruled out a general election until 2020’ (30/6/16 CNBC) ‘Clear . . . Patriotic . . . Strong . . . Reassure. She asserted that “Brexit means Brexit” and there must be “no attempts” to stay in the EU’ (30/6/16 Daily Mail) ‘She underlined that “Brexit means Brexit” as she announced that she will be running for party leadership contest’ (30/6/16 Daily News and Analysis)

What the phrase appeared to mean at this point was that May would take the referendum as an instruction to her if she became Prime Minister, to enact the UK’s withdrawal from the EU, even though the referendum was, in fact, only advisory. This phrase, then, was probably intended to be taken in the spirit of the kind of tautological phrase used by those in authority over others who are trying to instil obedience. Thus, parents might say to children ‘Bed means bed’ or teachers to their students might say ‘homework means homework’. In such cases the referent of the repeated noun is not in question, but there is some debate about whether the actions it implies are obligatory. Although this meaning of the phrase continued to occur, in the six months that followed her uncontested ‘election’ as party leader—and thence to Prime Ministerial office—the debate in the media became ever more critical of this phrase, now that it was clear that there was in fact no agreed referent for the repeated noun ‘Brexit’. The following are typical of many similar examples in the NOW Corpus from the second half of 2016: •

‘the cabinet is struggling to agree a common position on what “Brexit means Brexit” really means’ (15/11/16 BBC News)

116 • • •

Lesley Jeffries and Dan McIntyre ‘We don’t know what it means when it says, “Brexit means Brexit”’ (14/11/16 BBC News) ‘the content of Brexit remains a largely open question. May’s mantra “Brexit means Brexit” has become a laughing stock’ (7/11/16 Red Flag) ‘frankly, there aren’t any plans, beyond the hollow rhetoric of “Brexit means Brexit”’ (5/11/16 The Guardian)

As the preparations for Brexit continued in 2017, this phrase started to be used mainly as a criticism by opponents of Theresa May’s government and the perceived lack of clarity about what kind of agreement they were trying to achieve with the EU: •

• •

‘she never thought ahead to actually having to deliver on the details of what “Brexit means Brexit” is about. This is no way to run negotiations’ (26/6/17 The Independent) ‘We’ve heard a lot about “Brexit means Brexit” but it still hasn’t been clarified what that will actually entail’ (23/6/17 The Argus) ‘Phrases such as “Brexit means Brexit” are not very informative and serve only to increase the uncertainty’ (23/6/17 Irish Times)

However, among the later examples in 2017, there also appeared to be an increasing tendency to interpret this phrase as equivalent to what has also often been labelled ‘hard Brexit’: • •





‘Making these concessions would be very difficult for any UK prime minister. “Brexit means Brexit”, after all’ (9/6/17, Irish Times) ‘Our country is now on a course to leave the EU. Brexit means Brexit, a hard Brexit, a red white and blue Brexit’ (26/5/17 Ardrossan and Saltcoats Herald) ‘She literally could not say. All she could articulate were two slogans: “Brexit means Brexit” and “No deal is better than a bad deal”’ (10/6/17 The New York Review of Books) ‘there would be no Border problem in Ireland. Instead they have decided that Brexit means Brexit, there has to be a total divorce and Europe has to become something quite alien’ (3/9/17 Irish Times)

As the difficulties of negotiating the ‘divorce agreement’ multiplied in the later part of 2017, there was also some evidence that the European side was beginning to use May’s phrase against her, also implying that Brexit could not mean anything other than a complete separation between the UK and the EU: •

‘to make the point that the UK can not have it all. “Brexit means Brexit”, leaving the single market, means leaving the single market’ (31/8/17 Malay Mail Online)

The Devil Has All the Best Tunes •



117

‘The phrase “Brexit means Brexit” has now become a bit of a tongue-in-cheek catchphrase in European circles’ (10/7/17 BBC News) ‘In a frank speech to an audience in Brussels, Barnier said “Brexit means Brexit” everywhere, and confirmed that Britain will lose its financial passporting rights’ (20/11/17 Business Insider)

3.3 Hard and Soft Brexit As the negotiations on the terms of the UK’s departure from the EU progressed, there began to be a recognition on all sides that there was no single, clear and generally agreed referent of Brexit. However, though there are in fact many possible outcomes to the process, a tendency to binary divisions began to be evident. Work on the creation of unconventional oppositions (see, for example, Jeffries, 2010b; and Davies, 2012) within texts has shown in recent years that politically contentious events are often reported and debated using a newly constructed set of oppositional terms. In the case of Brexit, this opposition has been based on the pair of adjectives hard and soft, which are conventional opposites in English. The fact that these are gradable opposites, of course, technically allows for a range between the two extremes, but as with all gradable opposites, they are often treated in textual terms as though they were complementary, or mutually exclusive, opposites. In order to investigate these phrases more closely, we began by searching for the co-occurrence of soft and hard with Brexit within four words of each other in the NOW Corpus (see Table 7.7). The low numbers in some of these periods mean that trends cannot be described on their basis, but it is interesting to note that out of the five examples of soft + Brexit in early 2016, only three are actually phrasal as in: •



‘Even if there is no “soft Brexit” of this variety, there is still the economy to potentially derail Scottish independence’ (29/6/16 The Conversation UK) ‘Geert Bourgeois, who has pleaded for a soft Brexit in order to limit damage to Flemish exporters’ (28/6/16 Telegraph.co.uk)

Table 7.7 Frequency of Collocation of soft and hard With Brexit in NOW

soft hard

Jan–Jun 2016

Jul–Dec 2016

Jan–Jun 2017

Jul–Nov 24 2017

5 37

594 2,328

499 3,461

474 1,285

118

Lesley Jeffries and Dan McIntyre

The other two are incidental co-occurrences which do not necessarily link to the oppositional meaning: •



‘Leave campaigners began to think that Mr Johnson had gone soft on Brexit, though sources close to Mr Johnson insist the article was co-edited by Mr Gove’ (30/6/16 Telegraph.co.uk) ‘Combined with soft growth figures and Brexit fears, today’s data will likely dampen any expectations of a UK rate rise’ (12/1/16 Business Recorder)

Whilst we may choose to ignore these incidental co-occurrences, the first of these may link to the idea of soft Brexit being negatively evaluated, by bringing into play the connotations of being soft in the face of challenges (i.e. weak) rather than any more positive concept of softness (i.e. gentle). Of the 37 hard + Brexit occurrences in early 2016, not a single one is phrasal and they appear unrelated to the phrasal meaning: • •

‘Financials and tech stocks were hard hit after the Brexit vote but recovered well’ (28/6/16 Stuff.co.nz) ‘Moreover, in Scotland, where it seems it was rather hard to find a Brexit supporter, the leader of the Scottish National Party and First Minister of the Scottish National Party . . .’ (26/6/16 Daily Maverick)

The data from the first half of 2016, then, shows hard Brexit as apparently not being regularly named as such, even though soft Brexit was already in use. The second half of 2016, however, saw a change. Not only were there many times more hard (2,328) than soft (594) collocates, they almost all appear to be of the phrasal kind, and not incidental collocates.11 On many occasions, both hard and soft appear in the same sentence, and there is also the beginnings of a new opposition appearing between hard Brexit and no Brexit, which implies that soft Brexit is equivalent to not leaving the EU at all. Examples of each of these cases follow: •







‘debate in Britain over whether to seek a “hard” or “soft” Brexit, depending on how close a relationship was preserved at the end of the talks’ (6/12/16 malaysiandigest.com) ‘At the other end of the scale, a “soft” Brexit might involve some form of membership of the European Union single market, in return’ (3/12/16 BBC News) ‘. . . training and change the way we build. Be it hard or soft Brexit, we need to take back control of the construction industry’ (13/12/16 The Irish World Newspaper) ‘the need for a transitional deal to stop Britain falling into a hard, destructive Brexit that would take our economy over a cliff edge’ (12/12/16 The Guardian)

The Devil Has All the Best Tunes •

119

‘has been firmly told that there would either be a “hard” Brexit or no Brexit’ (11/12/16 DAWN.com)

This emerging opposition between hard Brexit and no Brexit implies that contrary to the binary but gradable opposition allowing for different kinds of relationship with the EU, Brexit actually is a singular (and implicitly most extreme) kind of separation which takes us back to May’s original Brexit means Brexit mantra. The first half of 2017 sees soft Brexit staying fairly stable (499) but hard Brexit continues to rise (3,461) and raises the question of whether, like Brexit itself, the ubiquity of (phrasal) hard Brexit makes it difficult to imagine any other kind of withdrawal, leaving the Brexiteers with their favoured interpretation and no clear alternative. There is, in this period, some debate about what exactly these (phrasal) terms mean, with a range of definitions such as: •





‘If the government moves toward a “soft” Brexit which permits high immigration, Ukip can accuse the government of “selling out”’ (15/2/17 eNCA) ‘as strong an association with Europe as possible – a so-called “soft” Brexit – including membership of the EU’s free-trade zone, the single market’ (18/4/17 CNN) ‘There is the so-called “soft” Brexit, whereby we would leave the EU but negotiate to stay in the free trade’ (9/3/17 Gazette)

However, some uses appear to question the very division into hard and soft: •

‘So far, May has clearly dismissed the idea of a soft or hard Brexit and she has emphasised that the UK needs to carve up new deals’ (17/1/17 Telegraph.co.uk)

The first half of 2017 sees the pinnacle of hard Brexit occurrences in this corpus (3,461) and most of them are phrasal: • •

‘Talk of hard Brexit and soft Brexit dominates as Parliament rages over what leaving the EU should mean’ (28/3/17 Stoke Sentinel) ‘holding Theresa May to account over her headlong rush into a hard, destructive Brexit’ (21/3/17 Pakistan Today) In this period, too, the first signs of soft border also occur:



‘good position given the recognition on all sides to maintain a soft border after Brexit’ (30/5/17 The Scotsman)

120

Lesley Jeffries and Dan McIntyre

There are also increasing numbers (ten from the first 100 examples) of hard border: •

‘tariffs would be important in whether or not there would be a hard border after Brexit and said there may need to be a change in EU law’ (29/4/17 BBC News)

It is difficult to be certain, but there is the potential for the rise of hard border (which is universally recognised to be a harmful development for Ireland) as having an incidental effect on the connotation of hard Brexit, which though hated by those who support remaining in the EU may nevertheless have started to seem as though it were the only possible outcome for Brexit at all. Such lateral effects of regular collocations would be very difficult to test even were it possible to turn the clock back and carry out psychological experiments on people as they live through the Brexit years, so we have to remain speculative about this point. In the second half of 2017, it looks as though the soft/hard peak has been reached and is now in decline. There are a couple of examples only of soft border and a few more of hard border (though the negotiations at this stage could see that number rise) and a small number of non-phrasal hard + Brexit examples remain, though the majority are of the phrasal kind.

4. Conclusion Our study of Brexit and related terms has taken a range of types and sources of evidence to try to demonstrate what we take to be the success of the Leave campaigners in branding their campaign to the point where there is no satisfactory way to refer to the process of the UK leaving the EU except with Brexit. This term, resisted in the first place by those who did not support it, became the common noun for a referent which is recognised by all, but defined by nobody. Once serious broadcasters and newspapers like the BBC and The Guardian started using it without any scare quotes and without even commenting on the word itself, it had won the day. Our claims here are that the reason for this success can be at least partly put down to its superior invention in phonological, morphological, syntactic and semantic terms. Once it had been established, there was relatively little to impede its growth in usage and this may have influenced the perceptions of the voting public both before and after the referendum itself. We certainly find that Brexit outnumbers its competitors (e.g. Bremain) and though it is never clear what, beyond a process, its referent is, it nevertheless starts to take part in fixed phrases (Brexit means Brexit; hard Brexit; soft Brexit) as though it were a common noun like any other. What we do not know at this point is whether Brexit will turn out to be a socio-political keyword that is very short-lived or whether it will continue to designate the decision of the UK to leave the EU and the process of doing so.

The Devil Has All the Best Tunes

121

Notes 1. The Oxford English Dictionary (OED) notes that Brexit was inspired by Grexit (i.e. Greek exit), the linguistic blend that was coined in 2012 to describe the potential process of Greece leaving the EU’s monetary union. The OED also claims Brixit to be an alternative form, though Fontaine (2017) notes that this was an early blend soon superseded by Brexit. 2. The story about the EU dictating the shape of bananas is a (possibly deliberate) misreading of the purposes of Commission Regulation (EC) No. 2257/94 (see http://bit.ly/1hilGhs). 3. Quantitative information in this chapter deriving from the NOW Corpus was sourced on 11 December 2017. Since NOW is a monitor corpus, figures may be different subsequent to this date. The approximate size of the NOW Corpus at the time of writing is 5.6 billion words. 4. The initial * is a wildcard and *eer. therefore indicates ‘any character/s preceding the letter combination ’. Following this [nn*] is the partof-speech (POS) tag for all nouns. Taken together, the search term means ‘extract from the corpus any noun that ends with ’. 5. Kilgarriff, Adam, Baisa, Vít, Bušta, Jan, Jakubíček, Miloš, Kovář, Vojtěch, Michelfeit, Jan, Rychlý, Pavel, Suchomel, Vít (2014) ‘The Sketch Engine: Ten years on’, Lexicography, 1: 7–36. https://www.sketchengine.eu/. 6. www.proquest.com/products-services/pq_newsstand_intl.html. 7. Friederichsen, P. (27 June 2016) ‘Coining catchy “Brexit” term helped Brits determine EU vote’, NYDailyNews.com. Retrieved 15 December 2017. 8. Both Grexit and Brexit are recorded in the OED as first occurring in 2012, in February and May respectively, so the press apparently picked up on them early on. 9. ProQuest figures were correct as of 7 January 2018, when the searches were carried out. This database is added to over time, so the figures may differ if accessed on a different date. 10. These figures are up to the time of writing. The comparable figures from 2016 (i.e. up to 24 November) are 538, so the apparent high point in second half of 2016 is consistent when the figures are made more comparable. 11. Note that there are still occasional examples of the other (non-phrasal) kind: ‘Remainers are still finding it hard to accept Brexit’ (23/12/16 Express.co.uk).

References Davies, M. (2012) Oppositions and ideology in news discourse. London: Bloomsbury. Davies, M. (2013) Corpus of News on the Web (NOW): 3+ billion words from 20 countries, updated every day. Available online at https://corpus.byu.edu/now/. Fairclough, N. (1989) Discourse and social change. Cambridge: Polity Press. Fontaine, L. (2017) ‘The early semantics of the neologism Brexit: A lexicogrammatical approach’, Functional Linguistics, 4 (6). Fowler, R. (1991) Language in the news: Discourse and ideology in the press. London: Routledge. Halliday, M. (2003) [2003] ‘Grammar, society, and the noun’, Reprinted in Halliday, M. and Webster, J. (eds.) On language and linguistics: Collected works of MAK Halliday, Vol. 3, pp. 50–73. London: Continuum. Jeffries, L. (2010a) Critical stylistics. Basingstoke: Palgrave Macmillan. Jeffries, L. (2010b) Opposition in discourse. London: Bloomsbury. Jeffries, L. and McIntyre, D. (2010) Stylistics. Cambridge: Cambridge University Press.

122

Lesley Jeffries and Dan McIntyre

Jeffries, L. and Walker, B. (2017) Keywords in the press: The New Labour years. London: Bloomsbury. Kemmer, S. (2003) ‘Schemas and lexical blends’, in Cuykens, H., Berg, T., Dirven, R. and Panther, K-U. (eds.) Motivation in language: From case grammar to cognitive linguistics: A Festschrift for Günter Radden. Amsterdam: John Benjamins, pp. 69–97. Levinson, S. (1983) Pragmatics. Cambridge: Cambridge University Press. Louw, B. (1993) ‘Irony in the text or insincerity in the writer? The diagnostic potential of semantic prosodies’, in Baker, M., Francis, G. and Tognini-Bonelli, E. (eds.) Text and technology: In honour of John Sinclair. Amsterdam: John Benjamins, pp. 157–176. McIntyre, D. and Price, H. (eds.) (2018) Applying linguistics: Language and the impact agenda. Abingdon: Routledge. Simpson, P. (1993) Language, ideology and point of view. London: Routledge. Sinclair, J. (1998) ‘The lexical item’, in Weigand, E. (ed.) Contrastive lexical semantics. Amsterdam: John Benjamins, pp. 1–24. Stockwell, P. (2009) ‘The cognitive poetics of literary resonance’, Language and Cognition, 1 (1), pp. 25–44. Toolan, M. (1992) Language, text and context: Essays in stylistics. London: Routledge. Van Peer, W., Zyngier, S. and Hakemulder, J. (2007) ‘Foregrounding: Past, present, future’, in Hoover, D. and Lattig, S. (eds.) Stylistics: Prospect and retrospect. Amsterdam: Rodopi, pp. 1–22.

8

Corpus Stylistics, Norms and Comparisons Studying Speech in Great Expectations Michaela Mahlberg and Viola Wiegand

1. Introduction Literary stylistics is often described as a discipline that studies literary texts by drawing on a linguistic toolkit. Or, as Simpson (2014, p.  3) describes it, stylistics is ‘a method of textual interpretation in which primacy of place is assigned to language’. Because of the centrality of language, stylisticians can be seen to inherit theories ‘from the particular kind of linguistics (systemic functional, corpus, cognitive, etc.) they chiefly employ’ (Toolan, 2014, p. 13). While literary stylistics thus brings the study of language and literature together, this does not automatically mean language and literature are seen as fully integrated. Leech and Short (2007, p. 12) refer to Spitzer’s ‘philological circle’ to explain fundamental tenets of literary stylistics. According to Spitzer (1948), a literary text can be seen both as a work of art and as a sample of language. Hence an analysis can start from the literary effects of the text and study the language to explain these effects or equally it can begin with the study of the language and then seek to explain the literary effects the language might create. Ideally, a stylistic analysis would then proceed in a cyclic fashion moving between the linguistic and literary view. Any approach, however, that aims to emphasise the integration of the study of language and literature necessarily also stresses that language and literature are fundamentally different and so non-literary and literary texts need different treatment. This difference is sometimes expressed through value judgements. Stockwell and Whiteley (2014, p. 1) go as far as to say that stylistics is ‘the proper study of literature’, while Toolan (2014) points out that literary texts operate like any other type of language, but are more intellectually interesting. Carter (2016) approaches the distinction through the notion of ‘literariness’. He argues that literariness is best seen as a cline, so that it is possible to see literary features to varying degrees in all types of texts. The way in which early dictionaries used examples from literary texts is another more practical illustration that the boundaries are fuzzier than the terminology around literary and non-literary language suggests.

124

Michaela Mahlberg and Viola Wiegand

In this chapter, we look at fictional speech in Dickens’s Great Expectations (GE) to lend further support to the argument that the notion of a clear-cut distinction between literary and non-literary language is ‘an unhelpful one’, as Carter (2016, p. 69) puts it. In particular, we aim to demonstrate that literary stylistics cannot just rely on linguistic models and methods but also needs to push the boundaries of the field by stressing how the analysis of the language of literature impacts on how we describe language more generally. Toolan (2014, p. 13) lists corpus linguistics among the examples of linguistic approaches from which stylistics has inherited theories. Leech and Short (2007, p. 286) even talk about a ‘corpus turn’ in stylistics and the term ‘corpus stylistics’ is used to position work in this field. In this chapter, we want to argue that, while existing corpus linguistic methods and emerging tools for textual analysis within the digital humanities offer useful potential for stylistic analysis, there is still a need for more approaches that are tailored towards the textual analysis of literature. We will introduce the CLiC web app (see CLiC web app, n.d.) that has been specifically designed for the study of literary texts and demonstrate how adjustments to conventional corpus tools make it possible to approach literary questions in a more focused way. We will propose an innovative method to identify linguistic patterns that are associated with ‘spokenness’ in fiction and suggest that corpus methods can also help to identify likely candidates for free indirect speech.

2. Corpus Stylistic Methods and Theory Many corpus stylistic studies have applied standard corpus methods to literary texts. A popular corpus linguistic approach to literary texts is the analysis of keywords. Keywords are words that are statistically significantly more frequent in a specific text compared to a reference corpus. For example, Scott (2006) examines keywords of Romeo and Juliet retrieved in comparison with all the Shakespeare plays added as a reference corpus. His study illustrates the keyword procedure for examining the ‘aboutness’ of an individual text by identifying keywords (love, lips, light, death, poison, etc.). Unsurprisingly, Scott (2006) finds proper nouns (Romeo, Juliet, Mercutio, Verona, etc.) among the keywords, but also exclamations (O, Ah). He points out that unexpected items like exclamations should be investigated further in a concordance analysis; which is generally a popular procedure for corpus stylistics (see e.g. Fischer-Starcke, 2010). Scott’s interest is not primarily a literary stylistic one. He mainly uses the Shakespeare examples to illustrate the concept of keyness. Still, the corpus methodological approach he takes is similar to studies in corpus stylistics. Stylistic concerns especially come into play when decisions are made on the type of comparisons. Culpeper (2009) illustrates keyword (as well as key part-of-speech and semantic category) comparisons between subcorpora that consist of the speech by individual characters in Romeo and

Corpus Stylistics, Norms and Comparisons

125

Juliet. This approach allows him to observe textual features of characterisation. Because of the text format of plays, subcorpora for character speech are straightforward to select. Murphy (2015) also runs keywords for Shakespeare, but with the focus on the language of soliloquies in Shakespeare’s plays. He creates specific subcorpora based on existing typologies of soliloquy and dialogue, drawing on criteria of direction of address. These corpora make it possible for Murphy (2015) to compare the language of soliloquies with dialogue. The approach of ‘text-internal’ keyword comparisons has also been used in studies of narrative texts. Toolan (2006) draws on the keywords procedure in his study of narrative progression in short stories. As Toolan’s general framework is sentence-based (see Toolan, 2009, 2016), instead of carrying out a conventional concordance analysis of the keywords, he pulls out all full sentences of a particular keyword. Aiming at tracing progression and coherence in a short story, Toolan (2006) argues that this set of keyword sentences acts as an ‘abridgement’ of the story. In addition to the keywords procedure, other comparative methods from corpus linguistics include ‘multi-dimensional (MD) analysis’. MD analysis was developed by Biber (1988) for the comparison of spoken and written registers. It works with corpora that have been tagged for a selection of mainly grammatical features. These features are then quantified and assigned to five dimensions. Shepherd and Berber Sardinha (2013) illustrate the use of MD analysis to compare works by the writer Julian Barnes to a number of registers such as telephone conversations, face-to-face conversations, professional letters and adventure fiction. Generally, for corpus linguistics, any type of analysis will involve comparison. The notion of comparison is also important to capture theoretical links between corpus linguistics and literary stylistics by relating comparison to the notion of ‘foregrounding’. Foregrounding means that linguistic features are made prominent and stand out against the norms of general language or against the background of their textual context. It is the psychological effect brought about in the reader (hearer) when deviant features of a text are made perceptually prominent (cf. Leech, 1985, p. 47; Short, 1996, p. 11). The two main textual patterns that can account for linguistic means to achieve foregrounding are ‘deviation’ and ‘repetition’. Deviations from linguistic norms are, for instance, ungrammatical forms or uncommon semantic combinations. Repetition is also a form of deviation in that it goes against normal usage patterns by being overfrequent (cf. Wales, 2001, p. 157). Deviant textual patterns can theoretically be described as the result of comparisons against various norms. From a corpus linguistic point of view, deviation can be practically identified through various forms of corpus comparison. Leech (1985) distinguishes  three types of deviation: ‘primary deviation’ is deviation from norms of the language as a whole (Leech, 1985, p. 45); ‘secondary deviation’ is deviation from norms of literary composition, including norms of author or genre

126

Michaela Mahlberg and Viola Wiegand

(Leech, 1985, p. 48); and ‘tertiary deviation’ (also called ‘internal deviation’) is deviation from norms internal to a text (Leech, 1985, p. 49). Louw (1993) is now a classic example of how corpus methods can aid the identification of primary deviation. He refers to his method as ‘matching texts against corpora’ (Louw, 1993, p. 161). This means individual words or phrases that are identified in a specific text passage are compared to a general reference corpus. With the help of a concordance analysis in the general corpus, Louw (1993) shows how the meaning in the text passage can be described as unusual and thus creating literary effects. An example of secondary deviation is Scott’s (2006) comparison of Romeo and Juliet to the Shakespeare plays, whereas Culpeper’s (2009) comparison of individual characters’ speech can be seen as describing a form of internal deviation. While the framework by Leech (1985) is useful to systematically relate the notion of deviation to corpus comparison, corpus linguistics emphasises that norms can only ever represent a simplified view. Focusing on parameters for comparison and the description of variation might be more productive (cf. Mahlberg, 2013 for a more detailed discussion). Shifting the emphasis from deviation to comparison, concordances, as a standard corpus tool, also provide crucial methods for corpus stylistic research. Running concordance searches across an individual text will retrieve all the instances of a word or phrase in that text and thus makes it possible to identify a cumulative picture of how specific textual meanings are created. Ruano San Segundo’s (2017) study of reporting verbs in Nicholas Nickleby is one example of the potential of concordance searches. Common corpus tools such as WordSmith Tools (Scott, 2017), AntConc (Anthony, 2018) and Wmatrix (Rayson, 2008) are popular choices for corpus stylisticians. Reasons for this popularity include their availability and accessibility. A keyword comparison with any of these three tools will be easier to run than, for instance, an MD analysis not only because of the statistical complexity but also because the Biber tagger that is used for the initial tagging is less readily available.1 In addition to standard corpus tools, which typically include concordance, keyword, cluster and collocation functions, a handful of new tools have been developed that are particularly suited to support stylistic analyses. Although in principle built for the analysis of any type of text, the WordWanderer web app (Dörk and Knight, 2015) currently features exclusively literary texts as preloaded examples (but allows users to upload their own texts). This tool was developed to provide both novices and experts with the opportunity to explore a text via a ‘playful’ navigation through the lexical links in the text (Dörk and Knight, 2015, p. 84). By contrast, the WorldBuilder tool (Wang et al., 2016) works with conceptual rather than lexical patterns. Having been developed to facilitate the annotation of a text with the elements of the cognitive

Corpus Stylistics, Norms and Comparisons

127

stylistic framework of Text World Theory (Gavins, 2007; Werth, 1999), the tool makes it possible to quantify these categories and produce ‘cognitive diagrams’ based on the text worlds. Work in corpus stylistics also needs to be seen in the context of wider developments in computational linguistics and digital humanities. Secondary deviation as referred to by Leech (1985) is especially relevant to studies in stylometry. Hoover (2007), for instance, studies the distinctive features of Henry James’s style and the division into early and late James. Burrows and Craig (2012, p. 293) discuss wider issues in authorship attribution studies to ‘show the literary fundamentals of the relations between character styles and authorial styles’. Drawing on examples from Shakespeare and his contemporaries, Burrows and Craig (2012) show that while character idiolects are identifiable, authorial differentiation transcends variation in characters’ speech styles. Eder (2017, p. 51) argues that methods of stylometry are popular with literary scholars, ‘because they offer convincing visualizations’. Eder (2017, p. 51) also refers to the ‘immense popularity of beautiful yet relatively simple plots’ of research concerned, for instance, with literary history or ‘distant reading’ and ‘macroanalysis’ (Moretti, 2005; Jockers, 2013). Other terms that have come to be used for work of a similar nature are ‘cultural analytics’ and ‘culturomics’. What tends to distinguish such studies from research in corpus stylistics is the amount of data under analysis and the emphasis on larger trends. But the results of such studies provide an important context to assess the norms, in the sense of Leech (1985), which impact on the literary stylistic analysis of individual texts or even text extracts. Underwood et al. (2018), for instance, investigate how the language used to describe fictional men and women has changed since the eighteenth century, which can provide a useful reference point for the study of gendered language in an individual novel.

3. A Tool for Corpus Stylistics: The CLiC Web App Corpus linguistics has not yet used the range of data visualisation tools that other areas of digital humanities might have. However, concordances are fundamentally a form of linguistic visualisation. Concordances enable the researcher to see patterns and identify meanings associated with these patterns. This approach to meaning is central to the innovative contribution that corpus linguistics has made to modern linguistics. Sinclair (1991, p. 100) already makes the point: ‘[t]he language looks rather different when you look at a lot of it at once’. To identify meanings in literary texts, and be able to focus on individual texts and even text extracts in the way that literary stylistics does, concordances and related tools also play an important role. In the present chapter, we propose an approach that adjusts standard corpus methods and tools so that they best serve the exploration of literary texts.

128

Michaela Mahlberg and Viola Wiegand

The Corpus Linguistics in Context (CLiC) web app (Mahlberg et al., 2016) has been specifically designed for this purpose. It offers standard corpus functionalities (concordancing, generating clusters and keywords) but also additional options to aid the analysis of literary features. These provide the user with access to different textual subsets (e.g. character speech and narration) along with the possibility to ‘KWICGroup’ concordance lines based on shared lexical patterns. CLiC also contains a tag menu that allows researchers to ‘tag’ concordance lines as part of their analysis and there is an option for different users to merge their results to facilitate the measurement of inter-rater agreement. The basic principle behind the CLiC interface is designed around ‘subsets’ or intratextual subcorpora. Accordingly, apart from accessing ‘all text’ of a particular book, it is possible to navigate to ‘quotes’ (mostly corresponding to character speech) and ‘non-quotes’ (narration). A specific type of non-quote is the ‘suspension’. Based on Lambert’s (1981) concept of the ‘suspended quotation’, suspensions are narratorial interruptions of character speech, which tend to contain reporting clauses and/or additional character information such as body language. In CLiC, suspensions are categorised according to length into ‘short suspensions’ containing up to four words and ‘long suspensions’ for any stretches of five or more words. As Examples (1) and (2)—both from chapter 2 of Great Expectations— illustrate, the longer suspensions provide the narrator with more opportunity to give additional character information such as body language.2 (1) ‘Yes, Pip,’ said Joe; ‘and what’s worse, she’s got Tickler with her.’ (GE, chapter 2) (2) ‘You know, Pip,’ said Joe, solemnly, with his last bite in his cheek and speaking in a confidential voice, as if we two were quite alone, ‘you and me is always friends, and I’d be the last to tell upon you, any time. But such a . . .’ he moved his chair and looked about the floor between us, and then again at me – ‘such a most oncommon Bolt as that!’ (GE, chapter 2) While CLiC makes it possible to read the novels with this annotation in the Chapter view, as illustrated for Example (2) in Figure 8.1, the main advantage of the corpus stylistic features of the CLiC web app is that the subsets are searchable.

Figure 8.1 Screenshot of the CLiC Chapter View for Example (2) from Great Expectations (quotes are tagged in red in the interface and printed here in a darker grey; long suspensions are tagged in orange and printed in a lighter grey here)

Corpus Stylistics, Norms and Comparisons

129

The searchable subsets make it possible to focus on the language used in quotes, non-quotes and suspensions. A simple option is, for instance, to check the distribution of a word or phrase that is found when reading a small text extract. Example (2) contains the phrase you know. Checking quotes and non-quotes shows that you know occurs 106 times in quotes and once in non-quotes of Great Expectations. In Example (2), it is used as a discourse marker. Going through the concordance lines, the tagging function allows the researcher to highlight all occurrences where you know functions as a discourse marker by adding a tick to the line (see Figure 8.2). This makes it easy to count the examples, but also to sort the concordance so that examples with a tick are displayed together. If a word or phrase occurs predominantly in quotes, the occurrences in non-quotes can point to specific textual functions. The only example of you know in non-quotes is shown in context in Example (3). Here the narrator Pip uses the discourse marker to pick up on Herbert’s wording in the preceding sentence. (3) “But the thing is,” said Herbert Pocket, “that you look about you. That’s the grand thing. You are in a counting-house, you know, and you look about you.” It struck me as a singular implication that you couldn’t be out of a counting-house, you know, and look about you; but I silently deferred to his experience. (GE, chapter 22)

4. Common and Idiosyncratic Speech Patterns in Great Expectations The corpus interpretation of the three types of deviation outlined by Leech (1985) has already pointed towards how corpus comparison can contribute to literary stylistic concerns. In this section, we are particularly interested in how corpus methods can support the analysis of fictional speech in a single novel. Page (1988) has argued that for the description of fictional speech the main question is not how similar it is to real spoken language. Because of the challenges that lie in representing spoken language in written form a detailed comparison would be an unproductive exercise. He also raises the point that the conceptualisation of fictional speech will depend on the current model of real spoken language that he still sees as inadequate. However, especially because of corpus advances, today’s current model of real spoken language is significantly different from the one that Page (1988) was referring to and our corpus approach takes account of these developments. In literary stylistics, the analysis of fictional speech has drawn on various linguistic models and approaches, including pragmatic principles, conversation analysis and discourse analysis. The speech and thought

Figure 8.2 Sample of the 106 Occurrences of you know in Quotes—the Tick at the End of the Concordance Lines Indicates Occurrences of the Discourse Marker

Corpus Stylistics, Norms and Comparisons

131

presentation model by Leech and Short (1981) has been an extremely influential approach in literary stylistics not least because it directly accounts for the specific forms in which speech is presented in literature. An updated model also includes writing presentation (Leech and Short, 2007), and the large-scale corpus study by Semino and Short (2004) has provided an empirical basis that has shown the relevance of the model beyond literary texts. Busse (2010) has specifically developed the speech, writing and thought presentation categories based on a corpus of nineteenth-century fiction and identified repetitive linguistic features that can support automatic annotation. From a theoretical point of view, direct speech generally seems to be the most straightforward category as it is formally indicated by quotation marks. In CLiC, the identification of quotes is also entirely based on punctuation. Direct thought and direct writing appear to be less frequent in the CLiC corpora so, as a shorthand, we refer to all quotes as fictional speech to start with. As part of the speech, writing and thought presentation model, direct speech tends to be seen in relation to the other categories so that the actual patterns of the content of what is said receive less attention than the structures that define what category the speech presentation falls into. In corpus stylistics, patterns of what is said are mainly studied for drama (see the earlier example of Shakespeare) or television dialogue (e.g. Bednarek, 2011), as in both cases direct speech comes in a more straightforward format than in narrative fiction. CLiC, however, makes it easy to focus on patterns in speech. Beginning from the example of Great Expectations (GE), in this section we will illustrate how CLiC can help identify patterns in direct speech that appear particularly ‘speech-like’. We use ‘clusters’, i.e. repeated sequences of words such as I don’t want to to look at three types of examples: 1. General speech clusters in GE quotes; 2. Idiosyncratic character clusters in GE; and 3. Clusters for which the tendency to occur in quotes affects their function when occurring in non-quotes. GE forms a relatively small corpus of around 185,000 words and we mostly focus on the even smaller ‘quotes’ subset of the corpus of approximately 53,000 words. We compare GE to the CLiC corpus of Dickens’s novels (DNov3) and to the spoken part of the British National Corpus (BNC). There is no general reference corpus of transcribed spoken nineteenthcentury English that we could have used. More importantly, however, we use the spoken component of the BNC (Spoken BNC) as a proxy for linguistic background knowledge of today’s reader. This knowledge is

132

Michaela Mahlberg and Viola Wiegand

Table 8.1 Corpora Used in Comparison with Great Expectations Corpus

Corpus Size

Speech Subset Used

BNC Dickens’s novels (DNov) Great Expectations (GE)

97,639,023 3,835,807 185,199

9,899,403 1,369,029 53,221

relevant to the extent to which examples from GE can create an effect of spokenness in the reader. Table 8.1 provides an overview of the corpora used in this chapter. 4.1 General Speech Clusters in GE Quotes This section examines which ‘general speech clusters’—i.e. repeated sequences of words that are frequent in authentic twentieth-century spoken language and Dickens’s overall fictional speech—can also be found in GE. Our argument is that frequent clusters in corpora of both authentic and fictional speech can act as a list of candidates that we use to look through an individual text. Even if these speech clusters occur only once in the individual text, there is evidence for their speech-like quality. Table 8.2 shows the most frequent clusters in GE quotes that also reach high frequency thresholds in the Spoken BNC and the entire quotes subset of all of DNov. The three main columns following the individual clusters provide the frequency information for each of the three corpora. The clusters are displayed in the order of their frequency in GE, which is shown in the first column following the clusters (‘GE—Freq.’). Accordingly, I don’t want to is with eight occurrences the novel’s most frequent quote cluster that is also highly frequent in the two larger corpora—DNov and the Spoken BNC. Note that purely based on cluster frequency in GE quotes, the top cluster would be I am going to (occurring ten times), but this cluster does not reach the thresholds in the other corpora and hence is not included in Table 8.2. As all three corpora are of different sizes (cf. Table 8.1), Table 8.2 displays both the raw and relative frequencies (normalised per million words). For very small corpora, relative frequencies generally lead to inflated numbers. So even a single occurrence such as I don’t know what reaches a relative frequency of almost 19 per million words in GE. However, this is not a problem in the present chapter as we are mainly interested in the frequency of the clusters in the larger corpora to make our point. Importantly, as our method enables us to find clusters that are defined by comparison, we can pick up the examples at the bottom of Table 8.2 that occur only once in GE. If clusters are run for an individual text, they need to be found at least twice to become noticeable. And often, minimum thresholds, e.g. at least five occurrences in the text under investigation, as in one of our earlier studies on GE

Corpus Stylistics, Norms and Comparisons

133

Table 8.2 Clusters in GE that are Highly Frequent in DNov and the Spoken BNC Cluster

GE

DNov

Spoken BNC

Freq. Relative Freq. # of Relative Freq. # of Relative Freq. Texts Freq. Texts Freq. 1 i don’t want to 2 what do you mean 3 what do you think 4 what do you want 5 i don’t know how 6 you know what i 7 are you going to 8 i tell you what 9 i don’t think i 10 i was going to 11 nothing to do with 12 i don’t know what 13 at the same time 14 i don’t know why 15 but i don’t know 16 is one of the 17 for a long time

8 7

150.32 131.53

51 196

11 15

37.25 143.17

529 337

261 165

53.44 34.04

5

93.95

117

15

85.46

508

242

51.32

5

93.95

74

14

54.05

359

123

36.26

4

75.16

66

14

48.21

413

221

41.72

3

56.37

30

12

21.91

541

176

54.65

3

56.37

41

13

29.95

474

190

47.88

3 3 2 2

56.37 56.37 37.58 37.58

84 39 70 34

13 14 13 11

61.36 28.49 51.13 24.84

293 284 416 246

119 162 223 149

29.60 28.69 42.02 24.85

1

18.79

162

15

118.33

854

283

86.27

1

18.79

74

14

54.05

459

260

46.37

1

18.79

28

12

20.45

276

142

27.88

1

18.79

37

12

27.03

236

140

23.84

1 1

18.79 18.79

30 36

10 13

21.91 26.30

224 215

157 151

22.63 21.72

(Mahlberg, 2007), introduce a particular selection that prevents us from finding the kind of examples we identify in this chapter. In this study, we still operate with thresholds, but they help take a broader view first before narrowing down the selection in GE. Table 8.2 contains only clusters that reach a frequency of at least 20 words per million both in DNov and in the Spoken BNC and also appear in a minimum of five texts in both of these corpora. By meeting these thresholds, the clusters qualify as what has been termed ‘lexical bundles’ in the corpus linguistic literature (Biber et al., 1999, p. 989): ‘the sequences of words that most commonly co-occur in a register’. Clusters that meet these criteria in DNov and the Spoken BNC will bring a speech-like quality to GE, even if they appear there only once. The clusters are common in the speech that readers are habitually exposed to so in the fictional text they create an effect of spokenness.

134

Michaela Mahlberg and Viola Wiegand

Cluster lists for GE, DNov and all other CLiC corpora can be retrieved with the CLiC web app. Currently the app interface displays only the most frequent clusters (occurring at least five times), but the complete cluster information can be retrieved via the CLiC API (see CLiC API, 2018 for documentation). Accordingly, the GE and DNov frequencies in Table 8.2 have been collected via the CLiC API. The BNC clusters have been generated from the spoken component of the XML edition of the BNC from 1994 (British National Corpus XML Edition, n.d.). The concordance in Figure 8.3 shows all instances of the most frequent quote cluster in GE—I don’t want to—that also qualifies as a lexical bundle in the Spoken BNC and DNov quotes. I don’t want to is followed by know in half of the cases, as well as be betrayed, get into trouble, go and once it is part of an answer (‘Why don’t you cry?’ ‘Because I don’t want to.’).4 This phrase is not tied to a particular character. Table 8.2 includes three examples of the trigram I don’t know followed by how, what, and why respectively which reflects that I don’t know is among the top frequent clusters in spoken English. Importantly, however, not all clusters in Table 8.2 should automatically be regarded as ‘speech’ bundles. The cluster at the same time makes the lexical bundles threshold of 20 that we have set here, but it is also a lexical bundle in non-quotes— which is unsurprising as time is a general noun, i.e. it is among the top most frequent nouns in the language based on general reference corpora (Mahlberg, 2005). In fact, in our non-quotes set the relative frequency of at the same time is even higher (73.26 per million) than in quotes. A comparison as in Table 8.2 can provide a reference point for clusters that occur infrequently or even just once if a single text is the only dataset. By interpreting their functions based on their common patterns valuable information is provided on what makes fictional speech speechlike. The identification of such speech bundles in fiction is particularly important, because idiosyncratic speech clusters, such as those discussed in the next section, tend to be more noticeable to the reader (and the critic) and so receive more attention. 4.2 Idiosyncratic Character Clusters in GE In Mahlberg’s (2007, 2013) terms, a cluster that is specific to a particular character is referred to as a ‘label’. Table 8.2 does not include any labels but these can be found by examining a list of all GE clusters. We can distinguish between idiosyncratic clusters that occur only locally at one particular point of the text and those that appear across various chapters. One well-known example of the second type is an extended cluster of Joe’s affective address for Pip, based on the core old chap, a sample of which is shown in Figure 8.4. We have used the KWICGrouper to identify the displayed concordance lines based on matches of dear, old and Pip within three words to the left of old chap. The top six lines, printed in black and white here, but highlighted in purple in the interface, are the six

Figure 8.3 Concordance of all Eight Instances of I don’t want to in Great Expectations

Figure 8.4 Concordance Showing 16 of 43 instances of old chap in GE

Corpus Stylistics, Norms and Comparisons

137

examples of the five-word cluster dear old Pip old chap, which contains two four-word clusters that are relevant to the data in this chapter, the two fragments dear old Pip old and old Pip old chap. The context shows that line 1 is different from the others (see Example (4)). Although the sequence summarises Joe’s speech habit of addressing Pip, Pip describes the phrase as one of Joe’s ‘old names’ for him, commenting generally on the use of this phrase rather than narrating a particular speech instance. (4) As I became stronger and better, Joe became a little less easy with me. In my weakness and entire dependence on him, the dear fellow had fallen into the old tone, and called me by the old names, the dear “old Pip, old chap,” that now were music in my ears. I too had fallen into the old ways, only happy and thankful that he let me. But, imperceptibly, though I held by them fast, Joe’s hold upon them began to slacken; and whereas I wondered at this, at first, I soon began to understand that the cause of it was in me, and that the fault of it was all mine. (GE, chapter 57) As illustrated in Figure 8.4, the CLiC tags menu makes it possible to annotate an example like this with a customised tag; in this concordance, we have opted to tag this particular instance with ‘pip’ and all others with ‘joe’ in order to mark the speaker. Another example of an idiosyncratic cluster that is not restricted to one place in the text is all right John all (and its variation right John all right). The Aged P.’s notorious response to his son John Wemmick occurs across four different chapters. By contrast, the two four-word clusters put the case that and Tom, Jack, or Richard are each used in just one chapter, by Jaggers and Wemmick, respectively. Both cases are noteworthy in that they occur frequently within the space of only a few paragraphs. 4.3 Quote Clusters Occurring in Non-Quotes The final type of clusters we look at are clusters that have a tendency to occur in quotes. When such clusters are found in non-quotes they take on particular funtions in the text. The 11th cluster in Table 8.2, nothing to do with, is the first cluster that occurs in GE and is a lexical bundle in both spoken subcorpora of the BNC and DNov, but does not contain a personal pronoun. If clusters contain a first- or second-person personal pronoun their association with speech is more direct—the pronouns reflect interpersonal relationships between speaker and hearer. In Mahlberg (2013: 66), which was not based on the CLiC subsets, the occurrence of the pronouns is used as a defining feature for ‘speech clusters’. The speech bundle nothing to do with does not make the lexical bundle threshold in DNov non-quotes where it occurs 23 times, i.e. 9.36 per million. In GE, the cluster occurs three times altogether, twice in quotes and once in nonquotes. Without the comparison with the reference corpora, these would

138

Michaela Mahlberg and Viola Wiegand

not be reliable figures to claim a tendency to occur in speech. Based on the comparison in Table 8.2, however, Example (5), where nothing to do with appears in non-quotes, can be seen to illustrate the speech-like quality of the first-person narrator’s interpretation of Jaggers’s speech. (5) ‘When that person discloses,’ said Mr. Jaggers, straightening himself, ‘you and that person will settle your own affairs. When that person discloses, my part in this business will cease and determine. When that person discloses, it will not be necessary for me to know anything about it. And that’s all I have got to say.’ We looked at one another until I withdrew my eyes, and looked thoughtfully at the floor. From this last speech I derived the notion that Miss Havisham, for some reason or no reason, had not taken him into her confidence as to her designing me for Estella; that he resented this, and felt a jealousy about it; or that he really did object to that scheme, and would have nothing to do with it. When I raised my eyes again, I found that he had been shrewdly looking at me all the time, and was doing so still. ‘If that is all you have to say, sir,’ I remarked, ‘there can be nothing left for me to say.’ (GE, chapter 36) If we broaden the focus from one novel to the entire DNov, another way to account for differences between quotes and non-quotes is a key comparison. As we argued earlier, key comparison is a popular method in corpus stylistics. A comparison across subsets illustrates another dimension of this approach for the study of fiction. Table 8.3 shows the top ten four-word ‘key clusters’, i.e. those clusters that are significantly more frequent in DNov quotes compared to non-quotes, which also include I don’t know what from Table 8.2.

Table 8.3 Top Ten Four-word Clusters for Key Comparison Quotes vs. Non-quotes in DNov, for all Key Clusters p < 0.0001

what do you mean i beg your pardon i should like to how do you do what do you think i am going to i don’t know what what is the matter i am sure you i tell you what

Freq. in Quotes

Freq. in Non-Quotes

Keyness

196 197 137 120 117 128 162 92 83 84

2 3 7 3 3 9 37 2 0 2

413.03 407.25 252.84 239.78 233.29 224.11 197.1 185.86 183.69 168.51

Figure 8.5 CLiC view of chapter 40 in Bleak House showing of such a thing being used in non-quotes

140

Michaela Mahlberg and Viola Wiegand

If we now move from the broader perspective back to individual textual examples, the key comparison can help find cases where speech clusters take on specific functions when they appear in non-quotes. One of the key clusters is of such a thing—occurring 28 times in quotes and ten times in non-quotes. It is also part of the cluster I never heard of such a thing. Figure 8.5 shows one of the non-quote examples (in Bleak House). In Figure 8.5, of such a thing or even the longer heard of such a thing uses spoken features in narration, which is part of the way in which Sir Leicester’s reaction and the reaction of the debilitated cousin are presented. Such examples of spokenness can make an important contribution to the discussion of criteria for free indirect speech. As both Toolan (2009) and Busse (2010) have argued, the definition of free indirect speech is difficult to formalise so that examples could be automatically retrieved. The type of comparisons we suggest in this chapter will not be able to resolve the issue of automatisation, but taking the spoken-like qualities of speech clusters into account that we have described does add a new method for finding candidate examples for free indirect speech.

5. Conclusion In this chapter, we moved away from a clear-cut distinction into literary and non-literary language by linking the notions of deviations and norms that are drawn on in literary stylistics to corpus linguistic comparisons of different corpora. In this way, we have also situated corpus stylistics within wider trends in digital humanities. Corpus stylistics strikes a balance between the interest of the stylistician in a detailed textual analysis and the concerns of data science and digitial humanties approaches that look for trends across large amounts of data. With functionalities of the CLiC web app we have highlighted how standard corpus linguistic tools can be adjusted to literary stylistic concerns. Our discussion of spokenness in particular stresses that the theoretical concept of the norm hides that categories found in an actual text are much fuzzier. The thresholds we used in Table 8.2 as well as the statistics that underlie key comparisons only seemingly suggest a clear cut-off and they also depend on the corpora we use. By using the Spoken BNC for a comparison with GE and DNov, we have shown how fictional speech shares lexico-grammatical elements with real spoken language. At the same time, our approach underlines that comparison is crucial and tendencies are more important than exact thresholds. Especially the examples of clusters that tend to occur in quotes but are also found, although less frequently so, in nonquotes show how frequency tendencies affect textual functions in local contexts. These insights have implications for fundamental stylistic models and categories. In particular, the concept of free indirect speech—a thorny category in literary stylistics—will need to be reconsidered with regard to quantitative measures to account for features of spokenness.

Corpus Stylistics, Norms and Comparisons

141

Notes 1. Andrea Nini developed the Multidimensional Analysis Tagger (MAT; Nini, 2014) that replicates Biber’s (1988) tagger. However, it is also important that the underlying selection of features can be modified to fit the set of texts under analysis (cf. Biber, 2006, pp. 181f.). 2. All examples are taken from the CLiC text, so no page numbers are included. 3. DNov contains GE, but this does not affect the kind of comparison we present here. 4. I don’t want to is an example for which the automatic speech annotation shows a mistake. There are also two instances found in non-quotes, because quotation marks were missed. This annotation mistake, however, does not affect our argument.

References Anthony, L. (2018) AntConc (Version 3.5.6). Tokyo, Japan: Waseda University. Available at: www.laurenceanthony.net/software (Accessed 16 April 2018). Bednarek, M. (2011) ‘The language of fictional television: A case study of the ‘dramedy’ Gilmore Girls’, English Text Construction, 4 (1), pp. 54–84. Biber, B. (2006) University language: A corpus-based study of spoken and written registers. Amsterdam: John Benjamins. Biber, D. (1988) Variation across speech and writing. Cambridge: Cambridge University Press. Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. (1999) Longman grammar of spoken and written English. Harlow, Essex: Longman. British National Corpus XML Edition (n.d.) Available at: http://ota.ox.ac.uk/ desc/2554. (Accessed 16 April 2018). Burrows, J. and Craig, H. (2012) ‘Authors and characters’, English Studies, 93 (3), pp. 292–309. Busse, B. (2010) Speech, writing and thought presentation in a corpus of 19thcentury narrative fiction. Bern: University of Bern. Carter, R. (2016) Language and creativity: The art of common talk. 2nd ed. London: Routledge. CLiC API (2018) Available at: https://github.com/birmingham-ccr/clic/blob/1.6/ doc/api_usage.rst (Accessed 16 April 2018). CLiC Web App. (n.d.) Available at: http://clic.bham.ac.uk/ (Accessed 16 April 2018). Culpeper, J. (2009) ‘Keyness: Words, parts-of-speech and semantic categories in the character-talk of Shakespeare’s Romeo and Juliet’, International Journal of Corpus Linguistics, 14 (1), pp. 29–59. Dörk, M. and Knight, D. (2015) ‘WordWanderer: A navigational approach to text visualisation’, Corpora, 10 (1), pp. 83–94. Eder, M. (2017) ‘Visualization in stylometry: Cluster analysis using networks’, Digital Scholarship in the Humanities, 32 (1), pp. 50–64. Fischer-Starcke, B. (2010) Corpus linguistics in literary analysis: Jane Austen and her contemporaries. London and New York: Continuum. Gavins, J. (2007) Text world theory: An introduction. 1st ed. Edinburgh: Edinburgh University Press. Hoover, D. L. (2007) ‘Corpus stylistics, stylometry, and the styles of Henry James’, Style, 41 (2), pp. 174–203.

142

Michaela Mahlberg and Viola Wiegand

Jockers, M. L. (2013) Macroanalysis: Digital methods and literary history. Champaign, IL: University of Illinois Press. Lambert, M. (1981) Dickens and the suspended quotation. New Haven: Yale University Press. Leech, G. (1985) ‘Stylistics’, in van Dijk, T. A. (ed.) Discourse and literature. Amsterdam: John Benjamins, pp. 39–57. Leech, G. and Short, M. (2007) [1981] Style in fiction: A linguistic introduction to English Fictional prose. 2nd ed. Harlow, Essex: Pearson Education. Louw, W. E. (1993) ‘Irony in the text or insincerity in the writer? The diagnostic potential of semantic prosodies’, in Baker, M., Francis, G. and Tognini-Bonelli, E. (eds.) Text and technology: In honour of John Sinclair. Amsterdam: John Benjamins, pp. 157–174. Mahlberg, M. (2005) English general nouns: A corpus theoretical approach. Amsterdam: John Benjamins. Mahlberg, M. (2007) ‘A corpus stylistic perspective on Dickens’ Great Expectations’, in Lambrou, M. and Stockwell, P. (eds.) Contemporary stylistics. London: Continuum, pp. 19–31. Mahlberg, M. (2013) Corpus stylistics and Dickens’s fiction. London: Routledge. Mahlberg, M., Stockwell, P., Joode, J. de, Smith, J. C. and O’Donnell, M. B. (2016) ‘CLiC Dickens: Novel uses of concordances for the integration of corpus stylistics and cognitive poetics’, Corpora, 11 (3), pp. 433–463. Moretti, F. (2005) Graphs, maps, trees: Abstract models for a literary history. London and New York: Verso. Murphy, S. (2015) ‘I will proclaim myself what I am: Corpus stylistics and the language of Shakespeare’s soliloquies’, Language and Literature, 24 (4), pp. 338–354. Nini, A. (2014) ‘Multidimensional analysis tagger 1.2—manual’, Available at: http://sites.google.com/site/multidimensionaltagger (Accessed 16 April 2018). Page, N. (1988) [1973] Speech in the English novel. 2nd ed. Houndmills: Palgrave Macmillan. Rayson, P. (2008) ‘From key words to key semantic domains’, International Journal of Corpus Linguistics, 13 (4), pp. 519–549. Ruano San Segundo, P. (2017) ‘Reporting verbs as a stylistic device in the creation of fictional personalities in literary texts’, Atlantis, 39 (2), pp. 105–124. Scott, M. (2006) ‘Keywords of individual texts: Aboutness and style’, in Scott, M. and Tribble, C. (eds.) Textual patterns: Key words and corpus analysis in language education. Amsterdam: John Benjamins, pp. 55–72. Scott, M. (2017) WordSmith Tools version 7. Stroud: Lexical Analysis Software. Available at: http://lexically.net (Accessed 16 April 2018). Semino, E. and Short, M. (2004) Corpus stylistics: Speech, writing and thought presentation in a corpus of English writing. London: Routledge. Shepherd, T. M. G. and Berber Sardinha, T. (2013) ‘A rough guide to doing corpus stylistics’, Matraga, Rio de Janeiro, 20 (32), pp. 66–89. Short, M. (1996) Exploring the language of poems, plays and prose. London: Longman. Simpson, P. (2014) [2004] Stylistics: A resource book for students. 2nd ed. London and New York: Routledge. Sinclair, J. (1991) Corpus, concordance, collocation. Oxford: Oxford University Press.

Corpus Stylistics, Norms and Comparisons

143

Spitzer, L. (1948) Linguistics and literary history: Essays in stylistics. Princeton: Princeton University Press. Stockwell, P. and Whiteley, S. (2014) ‘Introduction’, in Stockwell, P. and Whiteley, S. (eds.) The Cambridge handbook of stylistics. Cambridge: Cambridge University Press, pp. 1–9. Toolan, M. (2006) ‘Top keyword abridgements of short stories: A corpus linguistic resource?’ Journal of Literary Semantics, 35 (2), pp. 181–194. Toolan, M. (2009) Narrative progression in the short story: A corpus stylistic approach. Amsterdam: John Benjamins. Toolan, M. (2014) ‘The theory and philosophy of stylistics’, in Stockwell, P. and Whiteley, S. (eds.) The Cambridge handbook of stylistics. Cambridge: Cambridge University Press, pp. 13–31. Toolan, M. (2016) Making sense of narrative text: Situation, repetition, and picturing in the reading of short stories. New York: Routledge. Underwood, T., Bamman, D. and Lee, S. (2018) ‘The transformation of gender in English-language fiction’, Journal of Cultural Analytics. Available at: http:// culturalanalytics.org/2018/02/the-transformation-of-gender-in-english-languagefiction/ (Accessed 16 April 2018). Wales, K. (2001) A dictionary of stylistics. 2nd ed. Harlow, Essex: Longman. Wang, J., Ho, Y., Xu, Z., McIntyre, D. and Lugea, J. (2016) ‘The visualisation of cognitive structures in forensic statements’, 20th International Conference Information Visualisation, Universidade NOVA de Lisboa, Lisbon, Portugal, 20 July. Available at: http://eprints.hud.ac.uk/id/eprint/29086/ (Accessed 16 April 2018). Werth, P. (1999) Text worlds: Representing conceptual space in discourse. Harlow, Essex: Longman.

Section 3

Pragmatics and Perception

9

‘Intending to Mean, Pretending to Be’ Reflections on the Limits on Genre Rukmini Bhaya Nair

1. Imponderable Evidence: ‘This Very Special Weave’ The question is: What does imponderable evidence accomplish? [.  .  .] Imponderable evidence includes subtleties of glance, of gesture, of tone. I may recognize a genuine loving tone, distinguish it from a pretended one [. . .] (but) why should it be that pretending is always taking place—this very special weave in the pattern of our lives? (Wittgenstein, 1953, p. 228) There are many cultures around the world where, one is heartened to realize, the story opening, ‘I have a pet budgie called Billy. He swore at the cat today’, would not translate directly and remain interpretable. The broader cultural context would need to be explained too. The distinction I have drawn here [. . .] is implicit in the broader systemic linguistic distinction between context of situation and context of culture. (Toolan, 1988, p. 227; bold in original)

Do our lives indeed possess pattern, as Wittgenstein so sunnily presupposes, or are we creatures designated by nature to read meaning into all sorts of ‘contexts of situation’, into any set of random objects that just happen to meet our gaze? Cognitive scientists maintain that, as a species, we still tend to discern faces in clouds, figures in trees and serpents in ropes even though, objectively speaking, ‘there’s no there there’. This ability not only to apprehend design(s) but also instantaneously to re-design the world around us is, they argue, an evolutionarily pre-programmed anticipatory response (see Atran and Norenzayan, 2004). It signals the possibility of harm so that we can get out of its way. Literature, I suggest, carries forward into modernity that ancient impulse towards reading harm and panic, not to mention various other more complex emotions from shame to schadenfreude, as we navigate each day those tossing seas of circumstance. That we do not necessarily recognise first-off such patterns of shame or rituals of enjoyment in other cultures, all those myriad subtleties of tone and glance, is to acknowledge Toolan’s further emphasis that we also need a fair degree of immersion in a culture before we can adequately interpret its stories. Mere contextual

148

Rukmini Bhaya Nair

surmise is insufficient and may indeed place us in harm’s way, earning us social ridicule, even opprobrium, if we are culturally unschooled or inattentive. Consider Billy the budgie, for example. He has, on the face of it, a narrative analogue in the Indian parrot (budgies not being native to the subcontinent but to another former English colony, Australia). Yet, caution is advised. A parrot in Indian stories is in fact disallowed by cultural convention to parrot or swear; no birdbrain this! Had J. L. Austin considered the case, he might have allowed that a parrot could perform a locutionary act, that is, a verbal act with an ostensible meaning and reference, but certainly not a full-fledged illocutionary act, complete with intention. Indian legend, on the other hand, accords parrots superior language abilities and intentionality, not to mention the capacity to create impressive perlocutionary effects. Indeed, you find fortune-tellers on the streets of Delhi who to this day will, for a small consideration, direct their parrots to divulge your thrilling future. Parrots are to be found, as well, perched on the delicate hands of princesses in Mughal miniatures, avian companions of beauty, purveyors of truth, meaning savants par excellence. In short, the cultural and literary complexes around ‘talking birds’ might well differ radically, depending on social context, as Toolan wisely suggests—but the more general premise remains. Across cultures, literary forms, oral and written, train us in social survival; through their textual iterations, they bring about a powerful combination of representation and emotion. That is why listening to a good story or reading a fine satirical poem often moves one irresistibly to tears or laughter, even if one is well aware that what’s being so seductively presented is indubitably false, no more than a farrago of words. It is in this inherent capacity of language not just to rivet our attention but also to mislead us in our quest for a comforting stability of meaning that the import of what Austin called ‘pretend behaviour’ (or PB) resides. True, Austin himself felt that philosophers had exaggerated the scope and importance of PB (Austin and Anscombe, 1958, p. 278). However, in the light of more recent work on language acquisition, PB seems to be a corollary of our mastery of language. Cognitive studies show that the vast majority of children have told their first lie by the age of 4 (Talwar and Lee, 2002) at about the same time that they have acquired full competence over language, while clinical ‘theory of mind’ studies show that an identifying characteristic of autistic children, for example, is their inability to process certain forms of PB, putting them at a considerable social disadvantage (Baron-Cohen et al., 1985). Politeness studies, likewise, indicate that PB (saying how much you liked the payasam you were offered at a temple in India when in fact you found it cloyingly sweet and suspected its hygiene) is critical in directing pro-social choices. Clearly, the practice and appreciation of PB is both cognitively and culturally ubiquitous. In this chapter, I therefore return to Wittgenstein’s

‘Intending to Mean, Pretending to Be’

149

question (why is pretending always taking place?) by seeking to relate PB to lies and literary behaviour. I do so by taking further a hypothesis I first advanced almost two decades ago.1 My question at the time was not dissimilar to his: Why is storytelling always taking place? Why is narrative a discourse universal found in every known human culture and why is it so emotionally arousing? I noted that our palms tend to sweat, our pulse rates go up, our throats go dry, our eyes fixate, at moments of textual crisis. These are attested physical reactions. That we experience them in response to ‘mere words’ seemed to suggest that stories, poems, plays, indeed language games in all their guises, offer humans beings, a species held in thrall to language, the cheapest, not to say the safest, epistemic means to instil in us hard-won community wisdom. Literature works, in effect, like a flight simulator. One does not need to climb that cliff or love that bounder to feel the thrill of danger or the agonies and joys of falling in love. Indeed, if we actually did need personal experience to learn these things, it would be quite costly for the species as a whole, putting every individual in a great deal of physical danger. Literature saves us from ourselves, prevents harm and panic in this very real fashion, causing us to implicitly rely on it as a saviour. It is in this respect that literary formats constitute invaluable lessons in beguilement and—yes—pretence. Of course, they are not the only means by which we learn to ‘read’ our fellow beings but they are among the foremost tools our cultures provide to coach and coax us to gain expertise in gauging the nuances of social ‘evidence’ that might seem ‘imponderable’ to a less trained eye. Children learn the ‘truths’ of their cultures through such creative falsehoods. They also learn their first disappointments. As John Steinbeck, whose work we’ll briefly engage with in the last section of this chapter, puts it: when a child first catches adults out—when it first walks into his grave little head that adults do not have divine intelligence, that their judgments are not always wise, their thinking true, their sentences just—his world falls into panic desolation. (Steinbeck, 1952, p. 19) Panic desolation: This is not a condition known only to children. We continue to require ever more elaborate forms of make-believe as we move into adult life to confront emotions such as these, for which we have endless synonyms ranging from sadness to tension to depression. Many go, as we know, to extreme lengths of PB to conceal their unhappiness from others—no wonder, then, that it makes up ‘a very special weave’ in the design of human life. As much as language, PB constitutes a measure of our humanness, our infinite capacity to derive meanings from contexts and to don protective cloaks of perlocutionary effect. Our current global fascination with the theatrical show of Donald Trump’s intrigues and

150

Rukmini Bhaya Nair

counter-intrigues is a case in point. This animated drama of folly gives us the chance to interpret, over-interpret and under-interpret (Eco, 1986), at a suitably safe distance, the fusillades of pretence directed at us daily by choric reportage. Were there no PB, it would result, as Wittgenstein avers, in a major loss of subtlety, we might even say subjectivity, a loss so profound that we might never really transit to becoming the complex selves that we have the capacity to be. Precisely because the evidence is imponderable in instances of pretence, it brings into play those myriad possibilities of interpretative engagement that Toolan talks of and that language gives us for free. Tone, glance, touch, taste and so forth belong to the realm of the physical but their ‘pretended’ presence in literature gives us the chance to rehearse what we hardly have a chance to assess in the quick online flow of everyday life. The ‘safe space’ of literature, it could be argued, has been evolutionarily designated for this purpose. In the next sections, I analyse the similarities and contrasts between PB and what I will call lying behaviour (LB) and literary behaviour (LB2).

2. To Protect or Conceal or Disguise: ‘Words Are Still a Façade to Hide the Crime’ And first for that goddess fair and free [. . .] divinest Etymology. In Latin, never far from the literal Prae-tendere strays the meaning of holding or stretching on things in form of another in order to protect or conceal or disguise in: even in such a figurative use as that in Ovid’s praetenden culpae splendida verbatuae, the words are still a façade to hide the crime [. . .] it seems clear that it is still an important feature of pretending, in classic cases if not all, that the pretender is concealing or suppressing something. (Austin and Anscombe, 1958, p. 267) Hemingway’s short story [‘Cat in the Rain’] renders a few brief scenes from a marriage [.  .  .] My particular interest in the story is the seemingly emotional and temperamental gulf between the couple and the way this is conveyed and reflected by their stilted uncooperative talk . . . The wife is not merely ‘making conversation’ [. . .] But it may be that George [the husband] pretends to treat these utterances as defective informatives, therefore ones he need not attend to. (Toolan, 1985, pp. 202–203; italics in original)

In their pioneering essay on the subject, Austin and Anscombe lay out three features of PB. Such cases, they say, ‘typically’ involve: (PB) The pretence-behaviour, the actual public performance gone through in pretending, indulged in, as of course it is for the sake of dissembling

‘Intending to Mean, Pretending to Be’

151

(Rd) The reality-dissembled, about which the audience is to be hoodwinked. This may on occasion include in part, or be wholly identical with (RBd) Some real behaviour dissembled as for instance when I am really engaged in biting the carpet but disguise this fact by pretending to be kissing it. (1958, p. 268) The exuberantly bizarre nature of Austin and Anscombe’s examples cannot be gainsaid (how many people, after all, go around biting carpets and then pretend to be kissing them—although, to be fair, this illustration is in the context of a game where the people in question are pretending to be panthers, but then that’s odd enough in itself!). However, the points Austin and Anscombe make are very plausible: namely, that PB cannot ever be conflated with the behaviour that it is mimicking (Rd); and that there is always a physical action involved (RBd) in every case of PB. These conditions imply that ‘there is necessarily involved in pretence or shamming the notion of a limit which must not be over-stepped’ (Austin and Anscombe, 1958, p. 261). They admit that this limit is ‘vague’ but insist that it exists. PB differs in this analysis from ‘genuine behaviour’ (or GB) in that the former has been characterised from the earliest times, as ‘divinest Etymology’ tells us, by an essential element of dissimulation or disguise. In the present chapter, my aim is to ask whether the features Austin and Anscombe attribute to PB bear any resemblance to LB (lying behaviour, where there is also an intention to deceive although language is the means used in contrast to physical action); and to LB2 (literary behaviour, where there is, in the general case, no intention to deceive but where, like lying, there is language use as well as an indubitable element of Rd or ‘real behaviour dissembled’ (1958, p. 268), since it is not ‘real behaviour’ but a simulation of it that a literary text presents). Specifically, in the next section, I raise the question: Are there in fact GBs that distinguish the super-genres of poetry and prose, so foundational and so ubiquitous across cultures? What intentions, conventions, emotions and expectations distinguish these genres and how does one make less ‘vague’ the oppositions between them? Is there a tipping point at which one actually becomes the other? (Austin and Anscombe, for example, express their puzzlement over whether one can effectively ‘pretend’ to hole a putt in golf without actually holing it.) Can poetry, then, pretend to be prose or vice versa without falling into textual disarray? If so, how much vice in the sense of ‘bad faith’ and how much versa in the sense of ‘opposition’ may be said to characterise such a process of generic makeover? Or, is this just an absurd inquiry, a sort of tilting at windmills? Given the short span of this chapter, I cannot pretend to convincingly address this slew

152

Rukmini Bhaya Nair

of questions, but I hereby undertake to take a stab in the next section at illustrating, via a couple of literary examples, the sorts of dilemmas they raise. Before I do so, however, I want to look a bit more closely at an initial contrast: that between pretending to do something (PB) and lying about something (LB). LB seems to share with PB the three elements of public performance, insulation from reality and hoodwinking that Austin and Anscombe identify. In addition, researchers believe a lie must have three epistemic characteristics: (i) it must assert something false; (ii) the speaker must know that she is asserting a falsehood; and (iii) she must intend to deceive. Given this definition, here are some examples that I have devised that might aid a discussion about which ‘counts as’ (Searle, 1975) the most prototypical lie as well as the least prototypical lie, with various shades of semi-lie in between. A. A working woman gets tickets to her favourite matinee and informs her office that she is sick. As it happens, she can’t go in the end because she develops a headache. Has she lied? B. A little girl tells her mother that one and one add up to 11. Is she lying? C. A man is invited out to dinner by his boss and insists politely at the end of a boring evening that he has had a wonderful time. A lie? D. A boy says to his hero, the local cricketer, who he knows is nowhere near the best in the Ceat cricket ratings—‘You are the best cricketer in the world, Ramu!’ Is he a liar? E. A young man insults his rival calling him a donkey. Is he lying? F. Yudhishthira has, as you know, the reputation of being the most truthful of the Pandavas in the epic Mahabharata. One day, in order to upset Dronacharya who is leading the Kauravas and whose son is called Ashwathama, he says to Drona, ‘Ashwathama hatah (Ashwathama is dead)’. He then adds very softly, ‘iti gaja (I mean, Aswathama the elephant’. Has Yudhisthira lied? If we were to assign ‘weights’ based on informant judgements (say, 3 points for intention; 2 for knowledge; and 1 for the action of uttering something false), we could easily arrive at a rank ordering of these ‘lies’. But what would such an ordering show? Well, at least one feature that is noteworthy about answers I’ve informally recorded from responses to this list is that people seem eager to make distinctions among them with regard to their knowledge of cultural context, language nuance and moral culpability. For example, whenever I have given this task to students in India, they have always ranked F the ‘worst’ or ‘heaviest’ lie and B the ‘least’ or ‘lightest’ lie, once again bearing out Toolan’s point about the influence of language, cultural ideologies and stereotypes on social perception.

‘Intending to Mean, Pretending to Be’

153

Lies are ‘facades to hide crimes’ all right, just as in the case of pretending, but not all such crimes seem to be equally condemnable. Austin and Anscombe’s PB paradigm, which discusses an exciting array of instances—a magician pretending to saw a girl in half, a window cleaner pretending to clean windows, a man pretending be a panther and so forth—makes no distinctions, that is, in terms of the severity with which a particular PB act is socially viewed. In contrast, the cognitive apparatus that assesses LB appears to normatively require such assessments. As I say, my own informal observations have been that, in discussing these instances among themselves, participants in this sort of language game animatedly defend their judgement that C, for instance, counts as a polite excuse or ‘white lie’ and thus is less blameworthy than, say, the lies in A and F. They are able, too, to quickly distinguish the exaggerated heroworship in D from the metaphoric hyperbole in E, and so on. The linguistic performance of LB, in short, summons up moral, even emotional, evaluations in a way that PB does not seem to, at least in the Austin-Anscombe analysis, despite the fact that, intuitively speaking, one might speculate that there is indeed such a ranking. For example, a man pretending to clean the windows but actually peeping into peoples’ bedrooms is very probably adjudged more culpable than a man pretending in jest among friends to be a panther. One might thus, in further analyses, use this additional, and seemingly widespread, notion of an ethical reckoning that I have just introduced to draw further parallels between PB and LB. If a child, for instance, clutches her stomach and rolls about pretending to have a terrible stomach ache, is it, or is it not, on par with her verbally declaring she has one when she doesn’t? In this case, moreover, she could easily perform both actions together, leading to confusions between the definitions of PB and LB that will need to be teased out. However, not all cases of PB would so neatly merge with LB. It might thus be rewarding to study some of these intersections between PB and LB cross-culturally especially if one is interested in ‘total speech’, to borrow the felicitous title of one of Toolan’s books on the integrationist linguistics.

3. Against the Imagination Police: ‘What We Think of as Genre Boundaries’ Is it possible that what we think of as genre boundaries are things that have been invented fairly recently by the publishing industry? I can see there is a case for saying there are certain patterns and you can divide up stories according to these patterns, perhaps usefully. But I get worried when readers and writers take these boundaries too seriously [. . .] I would like to see things break down a lot more. I suppose my essential position is that I am against any kind of imagination police, whether they’re coming from marketing reasons or from class snobbery. (Ishiguro, 2015)

154

Rukmini Bhaya Nair Those texts are literary that we take to be literary—in talking about them, our readings and the authors in certain distinctive ways (playing the language game of literary appreciation). If we grant [. . .] that literary linguistics may be a procedure for confirming interpretations or reconstructing meanings, rather than a method of constructing meanings, it may be appropriate to note the sort of achievements possible with ‘weak theory’ (or rule of thumb) stylistics in practice. (Toolan, 1990, p. 25)

Fiction may be accounted a form of pretence insofar as it mimics or simulates the speech act of assertion, as Searle suggested long ago in the ‘Logical Status of Fictional Discourse’. While careful to say that he was analysing ‘fiction’ and not ‘literature’, neither of these ‘uses’ of language, Searle admitted, could be distinguished by any unique set of linguistic features. How one determined literary genre, in short, would depend on context and the inclinations that readers with different cultural assumptions brought to the text. This seems close enough to Toolan’s position in Stylistics that ‘those texts are literary that we take to be literary’ and his subsequent alluring advocacy of a ‘rule-of-thumb’ stylistics. But, if it is indeed the case that we all do our own (sometimes ‘all thumbs’!) things when we read, producing and reproducing what Eco has called ‘unlimited semiosis’, why bother to discuss the norms and ‘rules’ of literature at all? Why separate the genre of fiction from poetry or autobiography? Why invent theoretical categories like ‘free indirect verse’ to analyse point of view in fiction and ‘free verse’ to describe a particular form of modernist poetry? At the very end of his essay, Searle acknowledges, in as many words, that this is a relevant concern—his answer serendipitously anticipating Ishiguro’s worry 40 years later about ‘policing the imagination’ via strenuous attempts to guard and maintain genre boundaries. The preceding analysis leaves one crucial question unanswered: why bother? That is, why do we attach such importance and effort to texts which contain largely pretended speech acts? [. . .] Part of the answer would have to do with the crucial role, usually underestimated, that imagination plays in human life, and the equally crucial role that shared products of the imagination play in human life. (Searle, 1975, p. 332) In this section about ‘pretence’ in literary behaviour (or LB2), I try to gauge what happens in cases of generic crossover between prose and poetry. This is an interest that grows out of a kind of obsession in some of my earlier work (Nair, 2002a, 2002b, 2009a, 2009b,2011a, 2011b, 2015, 2016 etc.). There, I sought to argue that creating hybrid genres could have major interpretative consequences in situations of prolonged historical crisis, such as under the regimens of colonialism and

‘Intending to Mean, Pretending to Be’

155

post-colonialism. Laying out the theory I call ‘Sensuous Theory’ (Nair, 2002b, pp. 166–170), I tried to outline seven or eight features that comprised the stylistic ‘rules of thumb’ used by postcolonial writers to take over the roles of historians and critics. In these cases, the tendency was to try and redraw the boundaries between history and political commentary and literature. In such contexts of perceived ‘harm’, a writer might deliberately seek to reconfigure the boundary conditions of literary ‘safe spaces’, that is, using certain undercover tools that LB2 offers. Seducing the reader via the pretence of ‘just telling a simple story’, she might in fact engage in doing something far more complicated. Postcolonial writers may, for example, try and mirror the epistemic violence of an illegitimate regime by produce forms of linguistic ‘unreadability’ within their texts that mimic the impenetrable mindset of a ruling oligarchy. A couple of tiny examples from Salman Rushdie: In his 1983 novel Shame, widely understood to be a satire on military rule in Pakistan, the generals are called Raddi, Phisaddi and Bekaar—but you would have to know Urdu-Hindi to understand that these names mean Rubbish, Laggard and Useless! The absence of the other ‘native’ tongue here literally blocks off access to meaning and satirical nuance; it pushes out the monolingual English reader just as surely as dictatorial regimes marginalise some folk. Rushdie’s latest, 2017 novel, The Golden House, deploys a similar set of tricks; here, for example, a dread Mumbai gangster’s luxurious sailboat is called the ‘Kipling’. This strategy economically indicates in far fewer words than a historian might need not just that the corruption of postcolonial societies is a colonial inheritance but also that the intellectual hegemonies of the past can be brutally upturned in present times. I now turn to a completely different cultural context and a different set of genres—not history or literature at all but prose and poetry. The two American writers whom I want to involve in this impertinent, and very likely illegitimate, thought-experiment were both born in 1902. The resemblance almost seems to end there, however. No two writers—the one a novelist and the other a poet—could, to my mind, present a greater study in contrasts. They lived in different parts of the United States; their life-styles were completely at odds; both were famous in their own lifetimes but it is unlikely that they ever met. There are two other things they might have had in common, though. First, both were oddly incapable of PB or LB in real life,2 yet both were great masters with unmistakably distinctive styles when it came to LB2. Second, both were interested in human folly. The poet drew attention to the myriad minor flaws that men and women displayed in everyday transactions, their unreflective capacity to self-deceive, their persistent self-interest and their tendency to always keep a beady eye on their neighbours. The novelist similarly declared that there was only ‘one story in the world [. . .] Humans are caught—in their lives, in their thoughts, in

156

Rukmini Bhaya Nair

their hungers and ambitions, in their avarice and cruelty, and in their kindness and generosity too—in a net of good and evil [. . .] There is no other story’. Let us now read samples of the work of these authors.

Prose Sample A How pleasant to sit on the beach, on the beach, on the sand, in the sun, with ocean galore within reach. And nothing at all to be done! No letters to answer, no bills to be burned, no work to be shirked, no cash to be earned. It is pleasant to sit on the beach with nothing at all to be done! How pleasant to look at the ocean, democratic and damp; indiscriminate. It fills me with noble emotion to think I am able to swim in it to lave in the wave, majestic and chilly. Tomorrow I crave but today it is silly. It is pleasant to look at the ocean. Tomorrow, perhaps, I shall swim in it. How pleasant to gaze at the sailors as their sailboats they manfully sail with the vigour of vikings and whalers in the days of the vikings and whale. They sport on the brink of the shad and the shark. If it’s windy, they sink; if it isn’t, they park. It is pleasant to gaze at the sailors, to gaze without having to sail. How pleasant the salt anaesthetic of the air and the sand and the sun. Leave the earth to the strong and athletic and the sea to adventure upon. But the sun and the sand no contractor can copy. We lie in the land of the lotus and poppy. We vegetate, calm and aesthetic, on the beach, on the sand, in the sun.

Sample B To the red country and part of the gray country of Oklahoma, the last rains came gently, and they did not cut the scarred earth. The ploughs crossed and recrossed the rivulet marks. The last rains lifted the corn quickly and scattered weed colonies and grass along the sides of the roads so that the gray country and the dark red country began to disappear under a green cover. In the last part of May the sky grew pale and the clouds that had hung in high puffs for so long in the spring were dissipated. The sun flared down on the growing corn day after day until a line of brown spread along the edge of each green bayonet. The clouds appeared, and went away, and in a while they did not try any more. The weeds grew darker green to protect themselves, and they did not spread any more. The surface of the earth crusted, a thin hard crust, and as the sky became pale, so the earth became pale, pink in the red country and white in the gray country.

‘Intending to Mean, Pretending to Be’ Poetry Sample A To the red country And part of the gray country Of Oklahoma, the last rains Came gently And they did not cut the scarred Earth. The ploughs crossed And re-crossed the rivulet Marks The last rains lifted the corn quickly And scattered weed colonies And grass along the sides of the roads So that the gray country And the dark red country Began to disappear under A green cover In the last part of May, the sky grew Pale, and the clouds that had hung In high puffs for so long in the spring Were dissipated The sun flared down on the growing corn Day after day, until a line of brown Spread along the edge of each Green bayonet The clouds appeared, and went away And in a while they did not try any more The weeds grew darker green To protect themselves And they did not spread any more The surface of the earth crusted A thin hard crust, and as the sky Became pale So the earth became pale Pink in the red country And white in the gray country

157

158

Rukmini Bhaya Nair

Sample B How pleasant to sit on the beach, On the beach, on the sand, in the sun, With ocean galore within reach, And nothing at all to be done! No letters to answer, No bills to be burned, No work to be shirked, No cash to be earned, It is pleasant to sit on the beach With nothing at all to be done! How pleasant to look at the ocean, Democratic and damp; indiscriminate; It fills me with noble emotion To think I am able to swim in it. To lave in the wave, Majestic and chilly, Tomorrow I crave; But today it is silly. It is pleasant to look at the ocean; Tomorrow, perhaps, I shall swim in it. How pleasant to gaze at the sailors. As their sailboats they manfully sail With the vigour of Vikings and whalers In the days of the Vikings and whale. They sport on the brink Of the shad and the shark; If it’s windy they sink; If it isn’t, they park. It is pleasant to gaze at the sailors, To gaze without having to sail. How pleasant the salt anaesthetic Of the air and the sand and the sun; Leave the earth to the strong and athletic, And the sea to adventure upon. But the sun and the sand No contractor can copy; We lie in the land Of the lotus and poppy; We vegetate, calm and aesthetic, On the beach, on the sand, in the sun. When the prose and poetry samples were given separately, my informal interaction with students familiar with literature in English but unfamiliar

‘Intending to Mean, Pretending to Be’

159

with these writers showed that they displayed similar patterns of ‘behaviour’. When the students were handed the passages as ‘prose’, they felt that passage A was a riff, prosaic in its sentiments but playing a rhythmic game that replicated the lazy pattern of lying on the beach with nothing to do but gaze at the sea and observe, without requiring to emulate, those ‘manful’ sailors striving far-off. Prose passage B they read as the description of a landscape, also rhythmic, but with a ‘deeper’ meaning that captured the passage of time. What they did not do was dispute the genre of either passage, even when they noted the presence of rhyme in the first. Something similar happened when other students encountered the same passages as ‘poetry’. They accepted that both passages A and B were poems, finding the second jocular and the first ‘beautiful’ and evocative. It is high time now perhaps that we finally drop the pretence that we do not know who these writers are. The immersed-in-literature readers of this volume will have guessed already, especially as I planted a clue to the identity of one of them in the very first section of this essay. John Steinbeck (1952) is the author of the lyrical prose passage B that translates so easily into the ‘poetry’ of A, while Ogden Nash (1995) is the author of the charming ‘prose’ in sample A, originally written as the poem in B. Without the ‘imagination police’ around, it seems, ordinary readers are by and large willing to dispense with genre boundaries while still taking pleasure in language. Conventional markers of a ‘poem’ such as shape and rhyme matter—but not as much as we might assume. Nash (1995, p. 668), indeed, suggests that we make too much of such devices. One thing that literature would be greatly the better for would be a more restricted employment by the authors of simile and metaphor. Authors of all races, be they Greeks, Romans, Teutons or Celts, Can’t seem just to say that anything is the thing it is but have to go out of their way to say that it is like something else. (Nash, 1995, p. 668) Steinbeck ([1945], 1995) approaches the matter via a very different route, focusing on the propensities of human beings in general rather than writers but appears to arrive at a similar conclusion: Most people do not like themselves at all. They distrust themselves, put on masks and pomposities. They quarrel and boast and pretend and are jealous because they [. . .] cannot see themselves well enough to form a true liking, and since we automatically fear and dislike strangers, we fear and dislike our stranger-selves. (1995, p. 273) To form a true liking and, perhaps, true likenesses: This might well provide a stylistic ‘rule of thumb’ definition of what LB2 is. It is to understand

160

Rukmini Bhaya Nair

our ‘stranger-selves’. LB2, in short, is the realm Steinbeck intuits when, for example, he famously describes ‘Cannery Row’ as ‘a poem, a stink, a grating noise, a quality of light, a tone, a habit, a nostalgia, a dream’ (1945, p. 1), which seem to be just those ‘imponderables’ that Wittgenstein spoke of. Ishiguro, too, suggests that without the pressures of the market and class snobberies, we may well be more liberal not only in our notions of what genres mean but also in our intentions as people. LB2 not only teaches us that poetry can pretend to be prose and prose, poetry; but that, in the free-speech arena of literature across cultures, it is a shared characteristic that you have the licence to pretend to be me, an absolute stranger. The major genres of prose and poetry are, in this respect, emblematic of all verbal modes in that they ‘insulate us from reality’ only so that we may, paradoxically, come closer to it, at once expanding and embracing it. To return at the end, then, to Searle’s fundamental question: Why bother? Why attach such importance to texts ‘insulated from reality’ containing largely pretended speech acts? Perhaps we can think of this essay, in the spirit of genre makeovers, as a miniature three-act play, the first act of which begins with a Wittgensteinian mystery. What, if anything, does ‘imponderable evidence’ accomplish? Is Sir Imponderable the villain of the piece or the hero? The answer offered in this chapter is that an imponderable, by definition, places in doubt the very concept of definition. It trains us, instead, in the nuances of cultural interpretation by forcing us to accept that an element of ambiguity inheres in all human action. Enraptured audiences of such messages as we are, we can never be entirely sure of ‘the truth’ as it is played out before our eyes. Nevertheless, our skills at reading these everyday mysteries of language and behaviour establish us as bona-fide members of our cultures, equipped to cope with its various crises. A double purpose is thereby served—the better we are at interpreting subtleties of language and body language, the better we are at protecting ourselves from social harm. This first act also introduces Narrative as a sophisticated and independent heroine, heading up a vast army of literary devices, all devoted to defending the far-flung territories of interpretation and imagination. Lady Narrative and Sir Imponderable are here seen to be able allies (perhaps even romantically entangled). The second act then involves a confrontation between this redoubtable pair and the forces of Pretence from the neighbouring kingdom of Austinia. At first, there are such striking differences between the Wittgensteinians and the manners and customs of the people of Austinia that it seems that they will remain sworn enemies forever. In Austinia, insistent on rank and regalia, everyone wears medals that mysteriously proclaim PB, RD, RBd, LB, etc. on their chests, while the Wittgensteinians appear waveringly uncertain of the meanings of even well-established words like ‘story’ and ‘lying’. Gradually, though, these warring natives learn to decode each other’s language games and come to love hitherto unfamiliar styles of speech and gesture. Thus, in the third and final act, it becomes possible to stage a moral denouement of sorts, wherein various parties,

‘Intending to Mean, Pretending to Be’

161

not excluding budgies and parrots, come together to establish a large demilitarised zone called LB2. Here, crossovers and alliances of all sorts are encouraged between the denizens of traditionally divided generic territories such as prose and poetry, drama and essay, or colonial hegemony and postcolonial emancipation. In line with Moliere’s Monsieur Jourdain who famously notes that he has been speaking prose his whole life without knowing it, within this ethical space of LB2, at least a partial admission of the life-sustaining power of our multifarious illusions seems finally within reach. Have we perhaps been part of an estranged community of stylisticians, if not poets, all our lives? An imponderable.

Notes 1. Michael Toolan has been a dear friend for more than 35 years and although I speak of ‘my’ work, in truth, his insights and his intellectual generosity have been priceless. To this day, I would be at a loss without them. In the references that I have cited from Michael, I have in a sense paid tribute to those exciting early years of discussion and engagement in Singapore when we first became friends. My focus on Steinbeck and Nash in this essay is also a tribute to Michael Toolan’s longstanding interest in William Faulkner and other American writers. 2. On the incapacity for PB in everyday life, see James Michie (1983): ‘You feel that in one sense Nash never pretended: his poetic impulse is genuine, his trivial, quotidian subjects are to him serious: whether, for instance, it is better to bathe before shaving or the other way round, or which is worse, visiting a friend in hospital or being visited in hospital by a friend’. Likewise, Benson (1984) records that Steinbeck, ‘when asked on the day of the announcement if he deserved the Nobel, replied: “Frankly, no”’.

References Atran, S. and Norenzayan, A. (2004) ‘Religion’s evolutionary landscape: Counterintuition, commitment, compassion, communion’, Behavioral and Brain Sciences, 27, pp. 713–770. Austin, J. L. and Anscombe, G. E. M. (1958) ‘Pretending’, in The Aristotelian society, Supplementary Volume XXXII. London: Harrison and Sons Ltd, pp. 261–294. Baron-Cohen, S., Leslie, A. M. and Frith, U. (1985) ‘Does the autistic child have a “theory of mind”?’, Cognition, 21, pp. 37–46. Benson, J. J. (1984) The true adventures of John Steinbeck, writer. New York: The Viking Press. Eco, U. (1986) Semiotics and the philosophy of language. Bloomington, IN: Indiana University Press. Ishiguro, K. (2015) ‘Let’s talk about genre: Kazuo Ishiguro in conversation with Neil Gaiman: Genre’, The New Statesman, 4 June. Available at: www.new statesman.com/2015/05/neil-gaiman-kazuo-ishiguro-interview-literaturegenre-machines-can-toil-they-can-t-imagine (Accessed 28 March 2018). Michie, J. (1983) ‘Ogden Nash’, Observer Archive, p. 23. Available at: http:// archive.spectator.co.uk/article/15th-october-1983/23/ogden-nash (Accessed 28 March 2018).

162

Rukmini Bhaya Nair

Nair, R. B. (1984) ‘Fictional selves, empire’s fictions: The poets of John Company’, in Nicholson, C. E. and Chatterjee, R. (ed.) Tropic crucible: Self and theory in language and literature. Singapore: Singapore University Press, pp. 193–218. Nair, R. B. (1985) ‘Telling lies: Some literary and other violations of Grice’s maxim of quality’, Nottingham Linguistic Circular, 14 (Special Issue on Pragmatics), pp. 53–71. Nair, R. B. (1988) ‘Rewarding risks: Clines of metaphoricity’, with Ron Carter and Michael J. Toolan, Journal of Literary Semantics (April), pp. 20–40. Nair, R. B. (1989) ‘Text and pre-text: History as gossip in Rushdie’s novels’, Economic and Political Weekly, May, pp. 994–1000. Nair, R. B. (1991a) ‘From Freud to Rushdie: Notes on the interpretation of dreams’, Seminar, 348 (July), pp. 32–36. Nair, R. B. (1991b) ‘Dissimilar twins: Language and literature’, in Sundar Rajan, R. (ed.) The lie of the land: English literary studies in India. Oxford: Oxford University Press, pp. 260–275. Nair, R. B. (1992a) ‘Gender, genre and generative grammar: Deconstructing the matrimonial column’, in Toolan, M. J. (ed.) Language, text and context: Essays in contextualised stylistics. London: Routledge, pp. 227–254. Nair, R. B. (1992b) The hyoid bone. New Delhi: Viking and Penguin Books. Nair, R. B. (1995a) ‘The pedigree of the White Stallion: Postcoloniality and literary history’, in Brown, M. (ed.) The uses of literary history. Durham, NC: Duke University Press, pp. 159–186. Nair, R. B. (1995b) ‘Why does Rushdie offend?’ Seminar, 437 (December), pp. 103–109. Nair, R. B. (1997) Technobrat: Culture in a cybernetic classroom. New Delhi: Harper Collins. Nair, R. B. (1999) The Ayodhya cantos. New Delhi: Viking and Penguin Books. Nair, R. B. (2002a) Narrative gravity: Conversation, cognition, culture. New Delhi: Oxford University Press. Nair, R. B. (2002b) Lying on the postcolonial couch: The idea of indifference. Oxford: Oxford University Press. Nair, R. B. (2002c) Translation, text and theory: The paradigm of India. New Delhi, Thousand Oaks and London: Sage. Nair, R. B. (2003a) ‘The aesthetics of the ordinary: Evoking narrative wonder within the linear grammar of modernity’, Evam: Forum on Indian Representations, 2 (1–2), pp. 266–288. Nair, R. B. (2003b) ‘Sappho’s daughters: Postcoloniality and the polysemous semantics of gender’, The Journal of Literary Semantics, 32 (2), pp. 113–135. Nair, R. B. (2004) Yellow hibiscus: Poems. New Delhi: Penguin Paperback. Nair, R. B. (2006) ‘Implicature and impliculture’, in Maybin, J. and Swann, J. (eds.) The art of English: Everyday creativity. London: Palgrave Macmillan, pp. 99–102. Nair, R. B. (2008) ‘Language, youth culture and the evolution of English’, in Kachru, B. B., Kachru, Y. and Sridhar, S. N. (eds.), Language in South Asia. Cambridge: Cambridge University Press, pp. 466–494. Nair, R. B. (2009a) Poetry in a time of terror: Essays in the postcolonial preternatural. New Delhi and New York: Oxford University Press. Nair, R. B. (2009b) ‘Learning to write: Integrational linguistics and the Indian subcontinent’, in Toolan, M. J. (ed.) Language teaching: Integrational linguistics approaches. London and New York: Routledge, pp. 47–72.

‘Intending to Mean, Pretending to Be’

163

Nair, R. B. (2010) ‘Yudhishthira’s lie: The fiction of India’, Poetiques Comparatistes/ Comparative Poetics, pp. 227–242. Nair, R. B. (2011a) ‘The nature of narrative: Schemes, genes, memes, dreams and screams!’ in Geertz, A. W. and Jensen, J. S. (eds.) Religious narrative, cognition and culture: Image and word in the mind of narrative. London: Routledge, pp. 117–146. Nair, R. B. (2011b) ‘Thinking out the story box: Creative writing and narrative culture in South Asia’, TEXT, 10, pp. 1–22. Nair, R. B. (2012) ‘Philological angst: Or how the narrative of vensus, vaste and race in India still informs the discourse of the 21st century’, in Messling, M. and Ette, O. (eds.) WortMachtStamm: Rassismus und Determinismus in der Philologie 18./19. Jh. Munich: Wilhelm Fink, pp. 55–87. Nair, R. B. (2013) Mad girl’s love song. New Delhi: Harper Collins. Nair, R. B. (2014) ‘Narrative as a mode of explanation: Evolution & emergence’, in Lissack, M. and Garber, A. (eds.) Modes of explanation: Affordances for action and prediction. Basingstoke: Palgrave Macmillan, pp. 140–154. Nair, R. B. (2015) ‘Virtue, virtuosity and the virtual: Experiments in the contemporary Indian English novel’, in Anjaria, U. (ed.) The history of the Indian novel in English. Cambridge: Cambridge University Press, pp. 251–266. Nair, R. B. (2016) ‘Narrative: Self, poetry and politics’, in Letiche, H., Lightfoot, G. and Moriceau, J-L. (ed.) Demo(s): Philosophy, pedagogy, politics. Rotterdam: Sense Publishers, pp. 35–52. Nair, R. B. and Bhattacharya, R. (1990) ‘Salman Rushdie: The migrant in the metropolis’, The Third Text (Special Issue on the Rushdie Affair), 4 (11), pp. 16–30. Nash, O. (1995) Selected Poetry of Ogden Nash. New York: Black Dog and Levanthal. Rushdie, S. (1983) Shame. London: Jonathan Cape. Rushdie, S. (2017) The golden house. New York: Random House. Searle, J. R. (1975) ‘The logical status of fictional discourse’, New Literary History, 6 (2), pp. 319–332. Steinbeck, J. (1945) Cannery row. New York: Penguin Books. Steinbeck, J. (1952) East of eden. New York: The Viking Press. Steinbeck, J. (1995) Log from the ‘Sea of Cortez’. New York: Penguin Classics (original published 1951). Talwar, V. and Lee, K. (2002) ‘Emergence of white-lie telling in children between 3 and 7 years of age’, Merrill-Palmer Quarterly, 48, pp. 160–181. Toolan, M. J. (1985) ‘Analyzing fictional dialogue’, Language and Communication, 5 (3), pp. 193–206. Toolan, M. J. (1988) Narrative: A critical linguistic introduction. London: Routledge. Toolan, M. J. (1990) The stylistics of fiction: A literary linguist approach. London and New York: Routledge. Wittgenstein, L. (1953) Philosophical investigations. Oxford: Basil Blackwell.

10 Indeterminacy and Interpretation What Is Shown and What Is Hidden in Michael Haneke’s Caché Billy Clark La complexité commence quand ce n’est pas absolument clair. (Michael Haneke)

1. Introduction Introducing the important and influential collection Language, Text and Context, Michael Toolan (1992, p. xiv) stresses the importance of context and contextualisation in understanding texts. Rejecting a ‘myth’ which sees work in stylistics as assuming binary distinctions between text and context and between linguistic description and discourse interpretation, he suggests that stylistics has been misunderstood as ‘purely formalist, treating the text as autonomous, “delimited” and so on’ (Toolan, 1992, p. xiv). Arguably, recent increased interest in pragmatic stylistics (for examples and discussion, see Black, 2006; Caink and Clark, 2012; Chapman and Clark, 2014, forthcoming; Clark, 2014a, 2014b, 2015; Rosaler, 2016; Schubert and Volkmann, 2016; Warner, 2013, 2014) can be seen as consistent with Toolan’s suggestion, helping to develop understanding both of the interaction of texts and contexts and of the complexities involved in producing, interpreting and evaluating texts. This chapter aims to contribute to this by considering how specific ideas developed in work on pragmatics, specifically within relevance theory (Sperber and Wilson, 1986, 2015), can help to explain some of the complexities involved in interpreting texts and some of the ways in which these contribute to different kinds of responses and evaluations. It does this by focusing on Michael Haneke’s 2005 film Caché (Hidden in English). Viewers of this film have been perplexed by its failure to make clear a number of things which seem to be required in order to understand events and relationships (partly) represented in the film. These gaps are often mentioned in discussions of the film and it seems that how viewers feel about the gaps in particular often plays a significant role in determining whether they respond positively or negatively to the film overall.1 The chapter gives a very brief account of selected features of Caché, indicating some of the ways in which critics and others have responded

Indeterminacy and Interpretation

165

to it. It presents some key ideas developed in relevance-theoretic work and argues that these can help us to understand the film’s effects and how varying responses come about. The relevance-theoretic ideas focused on here are general ones about indeterminacies in texts and interpretations (Sperber and Wilson, 1986, 2015) and Furlong’s idea that interpretations can be more or less ‘spontaneous’ (Furlong, 1996, 2001, 2007, 2011). The main conclusions of the chapter are that ideas from pragmatics such as these can contribute towards fuller accounts of the complex processes involved in producing and responding to texts in general, that they help us to understand Caché and different kinds of responses to it, and that they can also help to explain differences between fairly formal interpretative and evaluative work and less formal responses.

2. Caché: What Is Shown and What Is Hidden Michael Haneke’s film Caché appeared in 2005. Since then, it has had a mixed reception from both professional and non-professional reviewers. Online ratings overall have been positive. At the time of writing (March 2018), the film has an average score of 7.3 out of 10 on imdb.com (IMDb, 2006) and a metascore of 83 out of 100 from 37 critics on metacritic.com (also viewable on IMDb). A key feature mentioned in most responses to the film is that there are difficulties in understanding what has happened and what we are viewing. Some viewers have responded to this negatively. One review on imdb.com (which gave the film a score of 1 out of 10), posted on 17 August 2006 by an author with username ‘usal’, was headed ‘What was hidden was the plot . . . you never find it’. The review described the film as ‘like an old relative you never want to talk to. They ultimately spend hours talking with nothing to say’. Others have responded much more positively. Roger Ebert (2006), for example, pointed out that some critics who saw the film at the Cannes film festival had ‘deplored its lack of a resolution’. By contrast, he suggested that, in his view, ‘it works precisely because it leaves us hanging’. Here is a brief summary of the film, indicating some of the ways in which it ‘leaves us hanging’. The film focuses on a couple, Georges (played by Daniel Auteuil), the presenter of a literary review television programme, and Anne (Juliette Binoche), a publisher. They receive a series of video tapes, left on their porch, beginning with ones which run for more than 2 hours showing only the outside of their apartment building. Later videos are accompanied by childlike drawings of what seem to be violent scenes. One tape, filmed from a car, turns out to show Georges’s childhood home. We discover that the videos and drawings have something to do with an incident in Georges’s childhood. They may or may not have been sent by Majid, a now middle-aged man (played by Maurice Bénichou) who lived with Georges’s family as a child. Majid’s parents were Algerian. His father worked for Georges’s parents. Georges’s parents

166

Billy Clark

took Majid into their family after his parents failed to return from a visit to Paris to take part in protests about the conflict in Algeria. Georges was jealous of Majid and, it seems, played a nasty trick on him which led to him being sent away. Georges told Majid to kill a chicken by cutting its head off with an axe. He then told his parents that Majid had done this to scare him. The protests which Majid’s parents took part in are now referred to as ‘the Seine massacre’ or ‘the Paris massacre of 1961’.2 Parisian police officers, acting under orders from the head of police in Paris, responded violently, killing a large number of the protesters, some by drowning after they were driven into the Seine. It is likely that Majid’s parents were among the protestors who died that evening. The French government denied the killings for many years and only slowly acknowledged the extent of the police brutality and their responsibility for deaths (which is still not fully known). Georges is also in denial at the beginning of the film. He has repressed his memories of what happened with Majid, claims at first to have no idea what the tapes are about and only slowly reveals what happened in his childhood. Eventually, one of the tapes provides a clue by showing the name of an avenue. Georges uses this to find where Majid now lives and he visits him in his apartment twice. On the second visit, Majid commits suicide by cutting his own throat in front of Georges. A near-final scene shows Georges going to bed and asking Anne not to wake him when she comes in, telling her that he has taken a pill (a ‘cachet’, which is, of course, a homophone of the film’s French title). One salient feature of the film is that there are parallels between Georges’s denial of what happened in his childhood (before the tapes start arriving as well as at first after they do) and the French government’s refusal to admit what happened on the evening of the protests in 1961. Another, arguably more salient, feature is that the film is deliberately difficult to understand in a number of ways. Here are some (far from all) of them: a.

What are we viewing when the film begins? When the film opens, we are watching the first video tape, i.e. the front of the apartment building where Georges and Anne live. The image is still and we do not know that it is a video until we have been viewing it for more than a minute when we hear footsteps and a figure appears and walks down the street. The image then breaks up as Georges fast forwards through it.

b. What are we viewing in later scenes? Later scenes are also hard to interpret. The status of a scene showing a young boy with blood on his mouth (later we realise that this is young Majid, played by Malik Nait Djoudi) is unclear. The most likely possibilities are that it is a flashback or a memory but the film does not make this clear.

Indeterminacy and Interpretation c.

167

How far can we trust anything we see? More generally, we often cannot trust what the camera is showing us. We are unsure what we are viewing when we see a long shot of the outside of the school when Georges goes to pick up his son, Pierrot (Lester Makedonsky). The camera never moves during the scene. Among the possibilities here are that it is intended to be understood as a scene from the perspective of the director (i.e. as most of the film), that it is footage shot by whoever sent the videos, or even that it is from a CCTV camera. Again, this is not made clear.

d. Who sent the videos? We never do know who sent the videos and other things. The options seem to be Majid (who expresses disgust towards Georges), Majid’s son (who visits Georges at work after Majid’s suicide, also expressing negative views and saying he ‘wanted to see your guilty face’) or Pierrot (who seems withdrawn as adolescents often are and who expresses anger at his mother, accusing her of having an affair with a friend Pierre). e.

What happened in the past? Arguably, we know the story of Majid and Georges by the end of the film but of course we are relying here on what Georges tells us and his memory. The film makes clear that neither are reliable. So, we cannot be sure of exactly what happened.

f.

What is happening in the final scene? The final scene is another immobile long shot of the front of the school. We see pupils gathering on the steps of the school before heading home. With no guidance from the camerawork, viewers may or may not notice one possibly significant detail. Towards the top left of the screen, high on the steps, Pierrot and Majid’s son greet each other and chat briefly. We have had no indication previously that they know each other. This could suggest that they have worked together to prepare and send the videos. We receive no indication of whether we can conclude this or not.

One kind of interpretation of the film would focus on the consequences of Georges’s actions and of him refusing to acknowledge what happened. Another would focus on the unreliability of memory and our interpretations of events. In connection with both, we might think of how Georges’s situation and behaviour with regard to what happened with Majid resemble that of the French police and government and the events of the Seine massacre. This also invites us to think about other kinds of repressed or denied memories by governments, institutions and individuals. It is important to notice that the gaps in the film and our inability to understand certain things both play roles in these kinds of

168

Billy Clark

interpretations. To some extent, we can argue that the film is ‘showing’ rather than ‘telling’ us about some of the effects of failure to acknowledge things. On the other hand, some viewers (like the online reviewer ‘usal’ mentioned earlier and the Cannes critics mentioned by Roger Ebert) decide that there is no point putting effort into understanding the film since there are things we need to know which it does not reveal. This chapter argues that ideas from relevance theory can help us to understand the richer interpretations and to understand why responses vary in these ways.

3. Indeterminacy and Caché Clearly, Caché is a complex text which asks a lot of its viewers. Accordingly, accounting for interpretations of the film will require a complex framework. Work in relevance theory has aimed to reflect the complexity of interpretations in a number of ways, including that implicatures (intentionally communicated conclusions) can be more or less strongly communicated, that there is often no clear distinction between implicatures and non-communicated implications (implicatures are the subset of implications which are intentionally communicated), that it is not always clear whether a particular act is intentionally communicative and that communication can be understood in terms of adjustments in the salience of sets of assumptions rather than the definite presence or absence of particular conclusions within what is communicated (for recent discussion, see Sperber and Wilson, 2015). These ideas can help to account for some of the complexities involved in interpreting and responding to Caché. Here is a brief indication of how each of these can play a role in this. This section illustrates with examples from elsewhere before moving on to apply them in discussing Caché. An influential idea in relevance theory is that implicatures can be more or less strongly communicated. If I say (1) to a new friend who has just made me a cup of tea: (1) I Love Tea When It’s Brewed Properly I might strongly implicate that I am enjoying this cup of tea. I also implicate other things with varying degrees of strength, e.g. that I think my friend is a good host, knows how to make good tea and so on. I also provide some evidence for weaker implicatures, e.g. that we will get along well, that I like her in general and so on. More generally, utterances tend to provide evidence for a range of conclusions rather than just one or two. Utterances also vary with regard to how many implicatures they support and how strongly. An utterance which suggests a fairly wide range of fairly weak implicatures is said to give rise to ‘poetic effects’ (Sperber and Wilson, 1986, pp. 217–224; Pilkington, 2000).

Indeterminacy and Interpretation

169

To take one example, the line ‘And no birds sing’ at the end of the first and last stanzas of John Keats’s ‘La Belle Dame Sans Merci’ (like many lines in that poem and, arguably, the whole poem overall) suggests a wide range of potential implicatures, including implicatures about the mental and emotional state of the character in the poem and about the poet. For many of these, readers cannot confidently assume that they are intentionally communicated but the poem nevertheless provides some evidence for them. Part of the enjoyment and interest of the poem, and of many artistic texts, lies in thinking about these potential implicatures and considering to what extent the text supports them. When we finish viewing a film, we often think about its overall meaning, sometimes explicitly in discussion with others, and this conversation helps us to find further implicatures. As mentioned, some viewers of Caché are frustrated when looking for its implicatures. Closely related to the idea that implicatures can be communicated more or less strongly is the idea that there is no clear cut-off point between weakly communicated implicatures and conclusions which are not evidenced by the communicative act and so count as non-communicated implications. Further, utterances can vary with regard to how clear a sense communicators themselves have of what they intend to communicate, i.e. a communicator might not have a clear sense of exactly what range of implicatures they intend to convey. Even for fairly straightforward utterances, a communicator need not have thought through the details of each assumption that their utterance provides evidence for. Finally, it is not always clear whether a particular act, or part of an act, is intended to be seen as communicative. An example often used to discuss this is one where an individual is speaking in a croaky voice or stifling a cough. It might not be clear whether the person with the croaky voice or cough is intentionally trying to communicate things which follow from this, e.g. that they have a cold. Each of these ideas can play a role in accounting for the complexity of interpretations of Caché. There are implicatures which vary in strength in fairly typical ways, e.g. when Georges tells Anne that he has taken pills towards the end of the film, just after asking her not to wake him when she gets home, he is strongly implicating that the pills will make it hard to wake him. He less strongly (but still strongly) implicates that he hopes to have a good night’s sleep. And so on. Then there are individual conclusions for which we find it hard to be sure how much evidence there is to support them and, in some cases, whether they are implicatures or only non-communicated implications. In the same exchange, when Georges asks Anne not to wake him, Georges speaks very calmly and quietly. At the level of character interaction, he could be intentionally implicating that he is being calm and has become more resigned to his situation. Or this behaviour could simply follow from his mental and emotional state

170

Billy Clark

with no intention to communicate these things. We can, of course, also discuss the varying strengths of these conclusions at the level of interaction between filmmakers and audience members. In films and other fictional texts in general, there is a kind of ‘layering’ where characters are seen interacting and therefore implicating things to each other, while at another discourse level, the implied or explicit narrator is telling or showing us these interactions (for an early discussion of this, see Leech and Short, 1981, pp. 237–254). Some things which are implicated at one level are not implicated at another. Another example in Caché concerns Georges’s refusal to acknowledge what he knows about the tapes and the events of his childhood. These provide evidence for negative conclusions about his character. Georges is not implicating (intentionally communicating) these but whomever we take to be the communicator/s of the film (perhaps Haneke? Perhaps the team collectively responsible for the film?) is implicating these assumptions. In this film, there are particularly salient examples of this. As mentioned earlier, we do not know the exact status of what we are looking at in several situations. At first, we do not know that we are viewing a tape at the beginning of the film. We do not know whether the school pickup scenes during and at the end of the film are to be taken as scenes shown by the film director/production team or something else (e.g. CCTV footage or tapes made by whoever made the tapes sent to the apartment). There are scenes which we are likely to assume are flashbacks but which we cannot be sure about. In each case, we cannot be certain of what we are viewing. So, there are special complexities here which make interpretations less clear and encourage us to think for a longer period of time about what we have seen and what it indicates. Interpretation processes might extend for some time after viewers have seen the film. More generally, relevance theory assumes that interpretation is not about deriving one or two, or a small set of, implicatures, but rather can be seen as involving adjustments to the ‘manifestness’ of a range of assumptions. An assumption is taken to be manifest to an individual to the extent that they are capable of representing it as true or more generally evidenced. This notion is much weaker than notions assumed in other approaches. An assumption can be manifest to an individual even if they have never entertained it. An utterance can make particular assumptions more likely to be entertained without this leading to their actually being entertained right away. To take a simple example, any utterance provides evidence that the speaker has not suffered physical injuries which prevent them from communicating. We are unlikely to infer this, though, unless there is a particular reason to. This notion can be useful in accounting for films and other relatively long texts where increased manifestness of particular assumptions resulting from encounters with one particular part of a text can be exploited later when they are needed in order to derive implicatures. When Anne meets Pierre in a café, for example, we might

Indeterminacy and Interpretation

171

not think that their behaviour shares features with the behaviour of a couple who are having an affair. We can construct these representations very quickly later, when Pierrot accuses his mother of having an affair, because the earlier scene made them more likely to be entertained. This film encourages viewers to carry on thinking about what the film might mean after their first viewing. One thing which contributes to this is the final scene where some viewers notice and many miss the interaction between Pierrot and Majid’s son. Discussing this is likely to be a catalyst for further inferences about the film, particularly since the film does not provide us with enough evidence to decide whether or not this indicates that the two boys are involved in recording and sending the tapes. One of many examples is the discussion in a ‘Cinematic’ film blog post (Ckckred, 2014). The critic Mark Cousins reports ‘an immediate hubbub’ among critics after they came outside from the premiere in Cannes: ‘The screening had caused a build-up of pressure in the auditorium: pressure to talk’ (Cousins, 2007, p. 223). As we have seen, some viewers find all of this complexity, alongside the failure of the film to provide clear answers, too much and they refuse to engage very far, deciding that it is not worth putting in further effort in trying to understand what they have seen. Viewers who go further find their efforts rewarded, as evidenced by discussions from viewers on IMDb (2006), critics on metacritic.com (also viewable on IMDb) and very rich interpretations developed by film scholars (including Chion, 2010; Khanna, 2007; Kline, 2010; Speck, 2010; Walker, 2018). Significantly, some interpretations build on some of these uncertainties, arriving at interpretations which see the film as being about interpretative uncertainties and about how these are added to when we repress or deny things. Haneke himself indicates this line of interpretation in the interview quoted earlier (available as an extra on the DVD version). He suggests that we often fail to acknowledge things and that when we hide things from ourselves this often leads to difficulties. In another interview with Solomons (2006), Haneke says, ‘if you come out wanting to know who sent the tapes, you didn’t understand the film. To ask this question is to avoid asking the real question the film raises, which is more: how do we treat our conscience and our guilt and reconcile ourselves to living with our actions?’ Thinking in this way leads some viewers to notice parallels between viewers of and characters in the film. Both are faced with interpretative difficulties with varying access to important contextual assumptions. As well as thinking about how individuals sometimes hide things from themselves and each other, we can think about parallels with larger communities or nations. As indicated earlier, the film mentions the Seine massacre of 1961, notorious for the events themselves and for the government’s denial and refusal for many years to acknowledge police involvement and responsibility. We can also go on to think about similar situations involving other

172

Billy Clark

countries (in some scenes, television news footage of ongoing conflicts around the world plays behind Georges and other characters). This approach, then, can help us to understand the complexities involved in interpreting the film, how interpretations can develop over time, how these complexities can lead to relatively rich interpretations and how uncertainties about interpretations can contribute to further interpretations. While the discussion so far indicates ways in which the relevancetheoretic framework accounts for indeterminacy, it might also be seen to raise a puzzling question. If utterances can generally be indeterminate in a range of ways and the indeterminacies in Caché are not significantly different from other kinds of indeterminacies, and if the indeterminacies indicated earlier contribute to interesting interpretations of Caché, then why do so many viewers respond so negatively? The next section suggests one way to begin to account for this.

4. Spontaneousness and Caché In considering what might be the nature of a ‘literary’ interpretation, Furlong (1996, 2001, 2007, 2011) suggests that interpretations can vary with regard to how far interpreters go beyond the fairly spontaneous kinds of interpretations typical of brief everyday exchanges. Assuming a relevancetheoretic framework, she suggests that we sometimes reach an acceptable interpretation fairly quickly and do not expend further effort thinking about it. By contrast, we sometimes go beyond the first interpretation we find and consider further possibilities, thinking fairly explicitly about evidence for possible interpretations. This latter process, she suggests, is typical of how we interact with what we think of as ‘literary’ texts. In an everyday context, if you and I are sitting eating together and I ask you to pass the milk, you are likely to infer that I would like the milk and pass it to me. You are unlikely to spend much more time and effort wondering what I could have meant beyond that. By contrast, a reader of Keats’s poem ‘La Belle Dame Sans Merci’ is likely to go much further than simply inferring what has happened in the poem. Some readers, including literary critics, spend a huge amount of effort gathering evidence for different kinds of interpretations, assessing possibilities and so on. Furlong would say that the first case (about passing the milk) is fairly spontaneous while the latter (interpreting Keats’s poem) is much less spontaneous. The idea that interpretations can be more or less spontaneous can help us to understand Caché in several ways. First, we can understand part of what Caché does to its viewers as encouraging relatively nonspontaneous interpretation. Second, it helps us to characterise different kinds of responses, since viewers who have very negative responses and encourage others not to waste time thinking about it can be understood as refusing

Indeterminacy and Interpretation

173

to pursue fairly nonspontaneous interpretations, partly because they do not think that the effort required for this will be rewarded. We can also understand some more formal and detailed interpretations as being extremely nonspontaneous. Roger Ebert (2010) spent great effort looking at different scenes to understand them. Chion (2010) focuses on the unusual use of music and other sounds in this film, including the birdsong in the opening and closing scenes which is clearly superimposed on images and not realistic (see also Walker, 2018, pp. 125–150). I would further argue that looking at texts like this helps us to understand more general features of pragmatic interpretations. In general, we expect to be able to find a reasonable interpretation of a communicative act. Sometimes, our expectations are not met and we give up. This presumably happens because we do not believe that our efforts will be rewarded with a reasonable interpretation. Our expectations about what texts such as films, prose fiction, poems and others will convey are often different from those for other texts. We often expect them to give rise to richer interpretations than those from everyday exchanges, emails, etc. In relevance-theoretic terms, we expect them to be more relevant. This is partly because of cultural assumptions about texts like these. We could also argue that asking others to interact with texts which take considerable time and effort should also raise expectations. If a film, novel or poem says something that could be expressed in a short utterance, then it would go against relevance-theoretic principles not to use the shorter, simpler utterance (artists often say something like this when resisting invitations to discuss their work). Looking at texts like these should be able to help us to understand what kinds of factors encourage viewers to put in more effort and what factors make it more likely that they will give up. There is no space to discuss this further here but there is one more point about the film which is worth noting. While individuals vary with regard to how much effort they are willing to put in to access an interpretation, it is likely that the salient gaps in Haneke’s film lower expectations that extra effort will be rewarded. Specific responses to Caché provide evidence that individuals do not like expending effort which is unrewarded. At least at first glance, it is fairly easy to build a case for the view that Haneke is being wilfully obscure and is not interested in helping viewers to find an interpretation that justifies the effort required. Haneke makes some interesting comments on this, suggesting that frustrated efforts to find particular kinds of interpretations are key to understanding the film. In one interview, in response to the question from the critic Jason Solomons (2006), ‘Do you enjoy deliberately frustrating people?’ Haneke says, ‘I look at it as productive frustration. Films that are entertainments give simple answers but I think that’s ultimately more cynical, as it denies the viewer room to think. If there are more answers at the end, then surely it is a richer experience’.

174

Billy Clark

There is a tension in Furlong’s work between the general relevancetheoretic assumption that interpreters follow a path of least effort in looking for an interpretation and then stop when they find one and the clear evidence that people often spend a large amount of effort looking for further interpretations. Evidence from various kinds of practice, including responses to Caché, suggests that we often make inferences about the likelihood of rewarding interpretations from different sources and that this influences our interpretative behaviour. We clearly also make inferences about the interpretative practices of others and this plays a role in some negative attitudes towards particular interpreters, including critics and academics, who are seen as expending too much effort and are often assessed negatively as pretentious as a result. One final question we might ask is about what happens to individuals who have given up if we then explain to them what we think are worthwhile interpretative paths. To what extent are they likely to reassess their  views? There is no doubt that individuals sometimes do reassess their interpretations of particular texts, and sometimes this is influenced by hearing the opinions of others. These opinions can come from more or less formal sources. Arguably, sharing ideas which can lead to different kinds of interpretations is one of the key functions of criticism. An important question for future research concerns the extent to which ideas from pragmatics can help us to understand the processes involved in criticism and how this affects others.

5. Conclusion Clearly, pragmatic inference plays a role in the production, interpretation and evaluation of texts. Given this, ideas from pragmatic theories can play a role in accounting for these processes. Recent developments in pragmatics have made significant steps in moving towards a fuller account of the complex processes involved in producing and responding to texts. This chapter has argued that relevance-theoretic assumptions about interpretative indeterminacies and spontaneousness can play important roles in this and that they help us to understand Caché and different kinds of responses to it. They also help to explain differences between fairly formal interpretative and evaluative work and less formal responses.

Acknowledgements I am very grateful to Michael Toolan both for his many important contributions to understanding how texts are produced, understood and evaluated, and for all of the support and encouragement he has given to me and others working in these areas over the years. I am also grateful to the editors of this volume for very helpful comments and suggestions.

Indeterminacy and Interpretation

175

Notes 1. Of course, varying responses can also be partly explained by reference to genres and expectations about them. This is not the focus of this chapter. 2. An estimated 30,000 protestors took part in the protests. The French government denied the killings until 1998 when it acknowledged 40 deaths (there are higher estimates suggesting maybe as many as 300 victims). In 2001, the then mayor of Paris, Bertrand Delanoë, unveiled a plaque commemorating the events near the Pont St-Michel. In 2012, President François Hollande acknowledged the events. A 2005 drama-documentary about the events, called Nuit Noire and directed by Alain Tasma, presents a version of the events.

References Black, E. (2006) Pragmatic stylistics. Edinburgh: Edinburgh University Press. Caché (English title: ‘Hidden’) (2005) (DVD). Michael Haneke (director). Available at: Sony Pictures Home Entertainment. Caink, A. and Clark, B. (eds.) (2012) ‘Inference and implicature in literary interpretation’, Special Issue of Journal of Literary Semantics, 41 (2). Chapman, S. and Clark, B. (eds.) (2014) Pragmatic literary stylistics. Basingstoke: Palgrave Macmillan. Chapman, S. and Clark, B. (eds.) (Forthcoming). Pragmatics and literature. Amsterdam: John Benjamins. Chion, M. (2010) ‘Without music: On Caché’, in Grundmann, R. (ed.) A companion to Michael Haneke. Oxford: Wiley-Blackwell, pp. 161–167. Ckckred. (2014) ‘Examining the ending of Caché’, June 27. Available at: https:// cinematicfilmblog.com/2014/06/27/examining-the-ending-of-cache/. Clark, B. (2009a) ‘Salient inferences: Pragmatics and the inheritors’, Language and Literature, 18 (2), pp. 173–213. Clark, B. (2009b) ‘ “The place near the thing where we went that time”: Pragmatics, stylistics and pragmatic stylistics’, Topics in Linguistics, 3, pp. 4–11. Clark, B. (2012) ‘Beginning with one more thing: Pragmatics and editorial intervention in the work of Raymond Carver’, Journal of Literary Semantics, 41 (2), pp. 155–174. Clark, B. (2014a) ‘Before, during and after Chekhov: Inference, literary interpretation and literary value’, in Chapman, S. and Clark, B. (eds.) Pragmatic literary stylistics. Basingstoke: Palgrave Macmillan, pp. 55–69. Clark, B. (2014b) ‘Pragmatics and inference’, in Stockwell, P. and Whiteley, S. (eds.) Cambridge handbook of stylistics. Cambridge: Cambridge University Press, pp. 300–314. Clark, B. (2014c) ‘Stylistics and relevance theory’, in Burke, M. (ed.) The Routledge handbook of stylistics. London: Routledge, pp. 155–174. Clark, B. (2015) ‘“What do you want me to tell?” The inferential texture of Alice Munro’s Postcard’, Études de Stylistique Anglaise, 8, pp. 99–120. Cousins, M. (2007) ‘After the end: Word of mouth and Caché’, Screen, 48 (2), pp. 223–226. Ebert, R. (2006) ‘Caché movie review’, Available at: www.rogerebert.com/reviews/ cache-2006 (Accessed 2 April 2018). Ebert, R. (2010) ‘Caché: A riddle, wrapped in a mystery, inside an enigma’, Available at: www.rogerebert.com/rogers-journal/caché-a-riddle-wrapped-in-a-mysteryinside-an-enigma (Accessed 2 April 2018).

176

Billy Clark

Furlong, A. (1996) ‘Relevance theory and literary interpretation’, Unpublished PhD Thesis. University College London. Furlong, A. (2001) ‘Is it a classic if no one reads it?’ Proceedings of the 24th Annual Meeting of the Atlantic Provinces Linguistics Association (APLA). Moncton: Université de Moncton. Furlong, A. (2007) ‘A modest proposal: Linguistics and literary studies’, Canadian Journal of Applied Linguistics, 10 (3), pp. 325–347. Furlong, A. (2011) ‘The soul of wit: A relevance-theoretic discussion’, Language and Literature, 20 (2), pp. 136–150. IMDb (2006) ‘Caché (2005)’, Available at: www.imdb.com/title/tt0387898/ (Accessed 2 April, 2018). IMDb (no date) ‘Caché (2005): Critic reviews’, Available at: www.imdb.com/title/ tt0387898/criticreviews (Accessed 2 April 2018). Khanna, R. (ed.) (2007) ‘The Caché dossier’, Special Issue of Screen, 48 (2), pp. 211–249. Kline, T. J. (2010) ‘The intertextual and discursive origins of terror in Michael Haneke’s Caché’, in Grundmann, R. (ed.) A companion to Michael Haneke. Oxford: Wiley-Blackwell, pp. 551–562. Leech, G. and Short, M. (1981) Style in fiction. London: Longman. Nuit Noire, 17 octobre, 1961 (2005) Directed by Alain Tasma, Canal+ and France 3 Cinema. Pilkington, A. (2000) Poetic effects. Amsterdam: John Benjamins. Rosaler, R. (2016) Conspicuous silences: Implicature and fictionality in the Victorian novel. Oxford: Oxford University Press. Schubert, C. and Volkmann, L. (eds.) (2016) Pragmatic perspectives on postcolonial discourse: Linguistics and literature. Cambridge: Cambridge Scholars. Solomons, J. (2006) ‘We love Hidden but what does it mean?’ The Observer, February, Available at: www.theguardian.com/film/2006/feb/19/worldcinema (Accessed 2 April 2014). Speck, O. (2010) Funny frames: The filmic concepts of Michael Haneke. London: Continuum. Sperber, D. and Wilson, D. (1986) Relevance: Communication and cognition. Oxford: Wiley-Blackwell. Sperber, D. and Wilson, D. (2015) ‘Beyond speaker’s meaning’, Croatian Journal of Philosophy, 15 (44), pp. 117–149. Toolan, M. (ed.) (1992) Language, text and context: Essays in stylistics. London: Routledge. Walker, E. (2018) Hearing Haneke: The sound tracks of a radical auteur. Oxford: Oxford University Press. Warner, C. (2013) The pragmatics of literary testimony: Authenticity effects in German social autobiographies. London: Routledge. Warner, C. (2014) ‘Literary pragmatics and stylistics’, in Burke, M. (ed.) The Routledge handbook of stylistics. London: Routledge, pp. 362–377.

11 Reliability, Unreliability, Reader Manipulation and Plot Reversals Strategies for Constructing and Challenging the Credibility of Characters in Agatha Christie’s Detective Fiction Catherine Emmott and Marc Alexander 1. Introduction One of the key properties of detective fiction is that readers are manipulated to believe wrong accounts of events to prevent them from guessing the solutions to crimes. In this chapter, we examine how readers are steered towards believing characters’ erroneous accounts by the writer’s rhetorical strategies. We look at selected stories by Agatha Christie and consider reliability and unreliability in relation to how information is primed for believability. We build on our previous work (e.g. Emmott and Alexander, 2010, 2014) to show how explicit statements are used to construct characters as credible, positioning their accounts of the crime as reliable and, conversely, how statements may challenge characters’ credibility, making their accounts seem unreliable. Since the overall aim of detective fiction is to deceive the readers, these stories frequently include surprise reversals where characters who were previously presented as reliable are revealed as unreliable, and vice versa. As Toolan (2009) points out, such surprises may ‘cause us to re-assess much of the narrative whose shape and direction we thought we had already understood’ (p. 172, also Toolan, 2001, pp. 99–103 and 2016, pp. 30–36). There is already a considerable amount of research on (un)reliability in narrative studies on a range of texts (e.g. Booth, 1991 [1961]; Phelan, 2005; Brütsch, 2014; Nünning, 2015). Our objective here is specifically to discuss reliability and unreliability in relation to plot manipulations in Christie’s detective fiction. Previous narratological discussions of Christie’s work (e.g. Zunshine, 2006) have often focused on her notorious use of the device of the narrating murderer who is eventually revealed to be unreliable.1 Our main aim in this article is not to revisit this classic topic, but to look at how reliability and unreliability are constructed in relation to a broader range of characters, particularly witnesses giving

178

Catherine Emmott and Marc Alexander

evidence, and to observe this in various stylistic modes (e.g. direct speech and thought representation). We will look particularly at whether readers are really likely to ‘suspect everybody’, which Zunshine (2006, pp. 121–133) suggests to be a key feature of reading detective fiction (basing this suggestion on the comment of one of Christie’s detectives2). This may seem like a useful dictum for reading Christie’s work since it is undeniable that many aspects of her detective worlds turn out to be unreliable, including, in some stories, all the suspects, the body and even the detective.3 Nevertheless, we argue that the whole point of the genre is to encourage false belief and a lack of suspicion of the key person and method, so that the eventual disclosure of the murderer and the explanation at the reversal stage come as a surprise. As we will see, we may be steered by the author towards placing too much faith in the credibility of particular characters, hence causing us to ‘under-suspect’ that they are in fact unreliable. Conversely, we may sometimes be pushed to ‘over-suspect’ characters, failing to see that they may actually be telling us the truth. Overall, detective fiction is generally successful when it manages to play on the cognitive limitations of readers, manipulating them in order that they miss the point of the clues provided and fail to guess who did the crime and how it was done.

2. Constructing Apparent Credibility: Observational Reliability and Its Limitations in Witness Statements In this section, we examine two cases where credibility is highlighted, with readers being steered by the author to have faith in specific characters and encouraged to have an ‘over-belief’ in their reliability. In ‘Sing a Song of Sixpence’ (a), the character discussed is only partially reliable, with key facts deliberately withheld. By contrast, in The Murder at the Vicarage (b), observed information is fully reliable in itself, but unreliable conclusions are drawn.4 (a) ‘Sing a Song of Sixpence’: Partial Observational Reliability Hiding Plot-Crucial Unreliability Agatha Christie’s short story ‘Sing a Song of Sixpence’ presents a key witness, a family maid, as highly credible. The story puts strong emphasis on the maid’s relatively convincing evidence which provides alibis for the family. Reliability attributions are made by other characters in the story to enhance her credibility. In certain respects, the maid’s evidence is indeed reliable (the family are innocent), but not in one plot-crucial respect—she is shielding the murderer, her son. Her generally reliable account disguises the fact that some very important aspects of her story are unreliable and, hence, that there is only partial observational reliability overall. She is in a difficult position in terms of giving evidence

(Un)reliability, Manipulation and Plot Reversals

179

because she is loyal to the family and does not want to incriminate them, but also wants to protect her son, an outsider to the house. This story presents the household murder of Miss Lily Crabtree. Four family members are obvious suspects since they were all present in the house at the time of the crime, most of them had arguments with the dead woman on the day of her murder and they are equal beneficiaries of her will, each due to inherit a substantial fortune. Before the maid, Martha, is introduced, the plot puzzle is set up. One of the family members, Magdalen Vaughan, enlists the help of an eminent retired barrister, Sir Edward Palliser, and presents the case to him as a choice between an outsider and a member of the family as the murderer, presenting the former as unlikely. ‘They know nobody came to the house that night.’ ‘You mean—?’ ‘That it’s one of us four. It must be. They don’t know which—and we don’t know which .  .  . We don’t know. And we sit there every day looking at each other surreptitiously and wondering. Oh! if only it could have been someone from outside—but I don’t see how it can . . .’ (p. 66; Christie’s italics and ellipses) Christie’s italicised strong epistemic modal verbs (know, must) in this example strengthen this view that responsibility lies not with an outsider but with the family. The maid, Martha, is then mentioned briefly by Magdalen, ‘they’ve questioned us all, and Martha again and again’ (p. 66), before a repetition of the point about the family, ‘they don’t know which [. . .] It’s one of us four—it must be’ (p. 66) and then a description of the other family members. This introduces Martha, but by wedging this mention of her between statements about the suspicion of the family, the focus of suspicion does not fall on Martha herself. Martha’s evidence is vital, although it creates a problem since she is saying that no-one from outside or inside, according to her knowledge, committed the crime. Her account eliminates the possibility of an outsider, since Martha had told the police that nobody came to the door. However, as we will see, she also provides evidence that the family members in the house were unable to commit the crime. Where there is no obvious perpetrator, Martha’s credibility in relation to the reliability of her evidence becomes crucial. Magdalen provides ‘reliability vouching’ (Emmott and Alexander, 2010) for Martha: ‘You trust Martha?’ ‘Oh, absolutely. She was with Aunt Lily [the murdered woman] for—oh—thirty years, I suppose. She’s as honest as the day.’ (p. 70)

180

Catherine Emmott and Marc Alexander

The construction of her credibility is also achieved by the assertiveness of her direct speech, as seen in her evidence about the three members of the family who were upstairs (she later also provides strong evidence for the fourth). Negative statements are repeated and there are strong modal expressions (know, should), with a comment on the speaker’s own experience (‘I know what I’m talking about’): ‘None of them did it [. . .] None of them could have done it without my knowing. [. . .] I know what I’m talking about [Her explanation that three members of the family were upstairs and that the stairs creak loudly] and if any one of those three had come down the stairs I should have known it. And they didn’t!’ (pp. 73–74; Christie’s italics) There is also a ‘mental attribution of reliability’ by Sir Edward immediately after this evidence: She spoke with a positive assurance which impressed the barrister. He thought: ‘A good witness. She’d carry weight.’ (p. 74) In addition to this, the reader is steered towards recognising Martha’s attachment to her murdered mistress by her own statements, by a narratorial comment on the warmth of her emotions and by Sir Edward’s further ‘mental attribution of reliability’ regarding her sincerity: ‘You were attached to Miss Crabtree?’ A warm glow—genuine—unmistakable—came into her face. ‘Yes, indeed, I was, sir.’ [She describes how Miss Crabtree stood by her when she ‘got into trouble’.] ‘I’d have died for her—I would indeed.’ Sir Edward knew sincerity when he heard it. Martha was sincere. (p. 74) Sir Edward positively evaluates Martha’s evidence. The reader is being presented with the script (Schank and Abelson, 1977) of legal experience, which may lead us to view his opinion as reliable. He has been ‘one of the most eminent criminal barristers of his day’ (p. 64), thinks of himself as an ‘innate criminologist’ (p. 69) and his opinion is respected. This prompts the use of the strong epistemic modals (know, known) in Magdalen’s description of him below: ‘You’re used to questioning people. And with all your experience you must know.’ ‘Know what?’

(Un)reliability, Manipulation and Plot Reversals

181

‘Whether they’re innocent or guilty.’ [. . .] He flattered himself that on the whole he usually had known! (pp. 68–69; Christie’s italics) Although Sir Edward wonders at one point whether Martha is shielding someone, this thought is wedged between statements about the possibility of a member of the family being guilty, as shown in the following, culminating in a strong ‘mental attribution of reliability’ regarding her supposed honesty. This deflects attention away from the fact that she may be shielding someone else (actually her son): Was that it? The four people in the house—one of them guilty? Did Martha want to shield that guilty party? Had the stairs creaked? Had someone come stealthily down and did Martha know who that someone was? She herself was honest—Sir Edward was convinced of that. (p. 75; Christie’s italics) Overall, Martha’s general credibility makes her evidence largely believable, since she appears to have been genuinely attached to her dead mistress, and Sir Edward’s conviction that she is honest also points to her innocence. There is nothing to prove that she is unreliable until the denouement. Then there is a sudden moment of discovery on the part of the investigator leading to the reversal stage. In this case the examination of the contents of the murdered woman’s handbag contradicts some of Martha’s earlier evidence, hence exposing the fact that some of her evidence is unreliable. She says to Sir Edward, ‘I tried to tell you no lies when you was asking me questions’ (p. 79) but there is a fundamental inaccuracy in her account (she did hear the doorbell ringing, albeit not ‘properly’ (p. 79)) and she has failed to mention that her son had entered the house and that she witnessed him commit the murder. Although ‘suspect everybody’ may seem like a reasonable strategy for solving a crime, it is difficult to ‘suspect everybody’ when we have almost no awareness that the guilty character—Martha’s son in this case—even exists. There has been a previous mention that Martha has ‘got into trouble’ (which could be interpreted to mean that she had been pregnant) as a girl, but this is rather different from discussing the existence of a child or the current presence of an adult son. The discourse focus previously has been entirely on how Martha’s employer has helped her in this situation, justifying her attachment to her employer. When the discourse focus is elsewhere, the reader may not even add the possibility of a relative into the inventory of characters in the story. This ‘shallow processing’ (A.J. Sanford and Sturt, 2002; A.J.S. Sanford et al., 2006) is a natural part of reading, but it can also be exploited by detective writers to bury information which later becomes important.

182

Catherine Emmott and Marc Alexander

(b) The Murder at the Vicarage: Observational Reliability and Plot-Crucial Deductive Unreliability Witnesses can also tell the truth but may nevertheless inadvertently mislead the reader. In The Murder at the Vicarage, Miss Marple, the detective, acts as a witness—her credibility is strongly endorsed and she is in fact reliable in terms of what she says she has observed. However, her strong emphasis on a specific point and the false assumptions she draws from this detract from the real solution to the crime. In some respects, this is similar to ‘Sing a Song of Sixpence’, where the emphasis on the suspicion of the family detracts from the role of an outsider, but in ‘Sing a Song of Sixpence’, the maid is deliberately deceptive in a key respect, whereas that is not the case for Miss Marple in The Murder at the Vicarage. This story plays on the social script which stereotypes elderly village spinsters as being nosey about what is happening in their neighbourhood. This accounts for Miss Marple’s ability to notice details, making her inherently credible according to this script. She is also credible because she is assertive in making her observations and, in addition, there is reliability vouching for her by the vicar and other characters. Miss Marple has encountered Mrs Protheroe, the wife of the murdered man, as she approached the scene of the crime at around the time of the murder. Mrs Protheroe has a strong motive for the crime, but apparently could not have been responsible since she was not carrying the weapon: ‘What did she shoot him with?’ ‘A pistol.’ ‘Where did she find it.’ ‘She brought it with her.’ ‘Well, that she didn’t do,’ said Miss Marple with unexpected decision. ‘I can swear to that. She’d no such thing with her.’ ‘You mightn’t have seen it.’ ‘Of course I should have seen it.’ ‘If it had been in her handbag.’ ‘She wasn’t carrying a handbag.’ ‘Well, it might have been concealed—er—upon her person.’ Miss Marple directed a glance of sorrow and scorn upon him. ‘My dear Colonel Melchett, you know what young women are nowadays. Not ashamed to show exactly how the creator made them. She hadn’t so much as a handkerchief in the top of her stocking.’ (p. 55) Here, the decisive statements made by Miss Marple, including her forceful contradictions, use of strong modality (of course, should) and dismissal of Colonel Melchett’s ideas, serve to make her evidence credible. Indeed, Melchett, the Chief Constable, recognises this, with the added consensus of the vicar:

(Un)reliability, Manipulation and Plot Reversals

183

Melchett admitted that grudgingly. She was a valuable witness— particularly valuable from Mrs Protheroe’s point of view. ‘I suppose there’s no doubt about what she says, eh?’ ‘If Miss Marple says she had no pistol with her, you can take it for granted that is so,’ I said. ‘If there was the least possibility of such a thing, Miss Marple would have been on to it like a knife.’ ‘That’s true enough.’ (p. 57) Throughout the text, there is repeated reliability vouching which endorses the fact that Miss Marple is attentive to local detail and accurate in her observations (if not always accurate in her conclusions). Many of these judgements come from the I-narrator, but there are similar comments from other characters and Miss Marple also reinforces this view herself. Some of these evaluations are as follows: [vicar] Miss Marple always sees everything. (p. 15, main narration) [vicar] ‘I think she is quite dependable [. . .] That is, in so far as she is talking of what she has actually seen. Beyond that, of course, when you get on to what she thinks—well, that is another matter.’ (p. 51, speech) [Colonel Melchett to Miss Marple] ‘you seem very good at observing’ (p. 54, speech) Miss Marple’s own assertiveness about her evidence and these reliability endorsements of her by other characters serve to build up a strong alibi for Mrs Protheroe, even though Mrs Protheroe might otherwise be a major suspect. Subsequently, the reversal stage prompts Miss Marple herself to re-evaluate what she has seen. Her observation has in fact been fully reliable in itself and, indeed, the criminals have deliberately played on the fact that Miss Marple is ‘a noticing kind of person’ (p. 180) to establish Mrs Protheroe’s alibi. Nevertheless, Miss Marple’s conclusions about Mrs Protheroe’s innocence are wrong since, until the reversal stage, Miss Marple has overlooked the possibility of an alternative source for the weapon. The gun has previously been planted at the scene of the crime by an accomplice and has in fact been used by Mrs Protheroe to shoot her husband, so Miss Marple has been unreliable in arguing that she is innocent. In this story, the reader is steered away from ‘suspecting everybody’ by the emphasis on Miss Marple’s ‘noticing’ ability and the reliability endorsements of other characters. There is even some external validation since the staff of several shops are also said to notice the same thing (p. 103). Only after the reversal stage does Miss Marple place some doubt

184

Catherine Emmott and Marc Alexander

on Mrs Protheroe’s behaviour, mentioning a point that she never made at the time, perhaps now to justify her new interpretation of this event: ‘a very curious thing (though no one happened to think of it that way)—Mrs Protheroe took no handbag with her. Really a most unusual thing for a woman to do.’ (p. 180; Christie’s italics) In providing an alternative version of events, Miss Marple, in her detective role, is again supported by the repeated reliability endorsements of the other characters. The vicar’s thoughts endorse Miss Marple’s revised story: [vicar] She spoke with such certainty that we both felt that in this way and in no other could the crime have been committed. (p. 181) There are also explicit verbal reliability endorsements during her explanation and when she has finished, such as: [Colonel Melchett] ‘Your solution is a very plausible one, Miss Marple’ (p. 183) The credibility of the detective’s argument style is important when the solution to the crime is presented, since otherwise there might be doubt about whether the solution really is the correct one.

3. Challenging Credibility: Constructing Partial Unreliability as a Means of Hiding the Truth We now move on to discussing cases where credibility is challenged, with readers being steered by the author to have an ‘under-belief’ in their reliability, even though sometimes these characters are actually telling the truth. This strategy can be used to diminish the importance of comments made by characters so that crucial plot-relevant details are given low prominence until the reversal stage (a ‘burying’ strategy; see Emmott and Alexander, 2014). In A Murder Is Announced (a), characters’ statements are largely dismissed because of the negative presentation of their personal characteristics, but this disguises the fact that there is an element of truth in what they say. In Lord Edgware Dies (b), a witness’s evidence is seen to be faulty in a detail and hence unreliable, but she nevertheless has made a correct identification of the murderer.

(Un)reliability, Manipulation and Plot Reversals

185

(a) A Murder Is Announced: Apparent Personal Lack of Credibility Christie sometimes utilises stereotypes to discredit characters, tapping into prejudices that may have been prevalent at the time, including negative representations of foreigners and of characters with mental illness symptoms. These stereotypes may be used as a smokescreen by the author to manipulate the reader to fail to see the elements of truth in what these characters say. In A Murder Is Announced, for example, some of the characters make explicit negative comments about foreigners and doubt their truthfulness (e.g. pp. 39–40). Mitzi, a maid, suffers prejudice as a refugee, but her employer, Miss Blacklock, at least partly questions the stereotype and suggests that although she believes that Mitzi has a tendency to exaggerate, she does not suspect her of the dishonesty of being a criminal: ‘I believe you police have got an anti-foreigner complex. Mitzi may be a liar, but she’s not a cold-blooded murderer.’ (p. 86; Christie’s italics) Ultimately, Mitzi assists the detectives by laying a trap for the murderer. She says publicly that she knows the truth about who was responsible for the crime and names Miss Blacklock (p. 195)—although she is again generally ignored as a liar (the detectives deliberately pretend to ignore her). The murderer herself, who is actually Miss Blacklock, knows that this is the truth and is caught when subsequently trying to attack her to silence her (p. 197). The final twist is that although Mitzi says what is true (as instructed by Miss Marple and the police sergeant), she did not directly witness this herself (and hence could not testify) since it was physically impossible for her to see what she claims to have seen at the scene of the crime (p. 215)—the trap is nevertheless successful in forcing Miss Blacklock to confess. Overall, it is plausible for Miss Blacklock to have (wrongly) believed that Mitzi was telling the truth on this occasion since she has previously said the following about her: ‘Please don’t be too prejudiced against the poor thing because she’s a liar. I do believe that, like so many liars, there is a real substratum of truth behind her lies.’ (p. 50) A Murder Is Announced also constructs partial unreliability in its presentation of the elderly character Dora Bunner, who is the old schoolfriend of Miss Blacklock and who is increasingly suffering confusion (presumably due to some form of early dementia) and is sometimes dismissed as being ‘Scatty’ (p. 50). Nevertheless, as we are told after the reversal

186

Catherine Emmott and Marc Alexander

stage, she has different degrees of incapacity and capability for different cognitive/perceptual processes: ‘The interesting thing about Dora Bunner was that she was quite unreliable in repeating things she had heard—she always used her imagination to exaggerate or distort them, and she was usually wrong in what she thought—but she was quite accurate about the things she saw.’ (p. 207; Christie’s italics) Also Dora is still able to register emotions (e.g. recognising the fear of her friend on the day that she committed her first murder) (p. 206). In addition, her confusion prevents her from fully remembering to keep hiding the fact that her friend is an imposter, so, as her health deteriorates, she starts occasionally to let slip her friend’s original name (p. 204). This prompts her friend to murder her in order to silence her—in this respect, it is Dora’s mental incapacity which ironically leads her to tell the truth, albeit with unfortunate consequences. In this novel, although Christie produces characters who may largely conform to social stereotypes of her time in order to suggest unreliability, this unreliability may sometimes be rather mixed, serving the purpose of facilitating plot twists. (b) Lord Edgware Dies: Discrediting the Evidence of an Initially Credible Witness The characters discussed in the previous section are presented as lacking in personal credibility in the (sometimes prejudiced) eyes of those observing them. If a character who appears initially credible is to be shown to be apparently unreliable, then the alternative to challenging their personal credibility is to discredit their evidence (see von Arnauld and Martini (2015) for these two strategies in law courts). Lord Edgware Dies provides a plot where there are contradictory statements from two sets of witnesses. Both cannot be right, so the solution to the crime depends on establishing which witnesses are reliable or unreliable. The strategy of finding a contradiction in the evidence is the same as in ‘Sing a Song of Sixpence’, but here it discredits the witness’s evidence at a relatively early stage in the story, making her seem unreliable during the puzzle stage. When the reversal comes, the surprise is that in fact the witness made a correct identification, even though she had seemed over-confident in giving evidence and was careless about the detail of what she saw. In this story, Lord Edgware is murdered and his estranged wife, Lady Edgware, is under suspicion. Witnesses in two locations, Lord Edgware’s home (the scene of the crime) and a private dinner party, provide statements to say that Lady Edgware was seen in both those locations at the

(Un)reliability, Manipulation and Plot Reversals

187

time of the crime. If she was at the house, she seems likely to have been the murderer, but if she was at the dinner party, she has a robust alibi. Hercule Poirot, the detective, interviews the staff at Lord Edgware’s home. The butler has worked there only since Lady Edgware left her husband and so just knows her from media coverage not directly, so his evidence is useful only in confirming that someone who looked like her was there. The secretary seems more likely to be reliable in identifying Lady Edgware since she ‘really knew Lady Edgware’ (p. 53; Christie’s italics) and is ‘absolutely positive’ it was her (p. 52). The narrator, Hastings, provides an initial positive assessment of the secretary, conveying her apparent credibility since she is ‘efficient-looking’, with ‘shrewd’ eyes and a ‘clear business-like voice’ (p. 55). Hastings thinks that Poirot is impressed with her too. The secretary, in her witness statement, tries to convey conviction, using strong epistemic expressions (e.g. absolutely, of course, sure and certainly): ‘Are you absolutely certain that it was Lady Edgware who came here last night?’ ‘That’s the third time you’ve asked me. Of course I’m sure. I saw her.’ ‘Where did you see her, Mademoiselle?’ ‘In the hall.’ [. . .] ‘And where were you?’ ‘On the first floor—looking down.’ ‘And you were positive you were not mistaken?’ ‘Absolutely. I saw her face distinctly.’ ‘You could not have been misled by a resemblance?’ ‘Certainly not. [Lady Edgware’s] features are quite unique. It was her.’ (p. 55) By a trick, Poirot subsequently undermines the credibility of this statement, showing that the secretary cannot have seen Lady Edgware’s face from the position where the secretary was standing. This contradiction leads to a reversal (still within the puzzle stage) in which he loses faith in the reliability of her evidence, subsequently stating: ‘You observe, Hastings, that the secretary is a dangerous witness. Dangerous because inaccurate.’ (p. 59) Both the secretary and Hastings point out that even though she had erroneously mentioned seeing Lady Edgware’s face, recognition can still result from considering a person’s voice and gait. Poirot, however, argues that these features can easily be mimicked and suggests that this

188

Catherine Emmott and Marc Alexander

could have been done by a well-known actress, Carlotta Adams, who has recently impersonated Lady Edgware on stage (a direct connection between the two is later found). Poirot subsequently further discredits the secretary in conversation to Hastings, suggesting that it is difficult to trust her evidence because she is over-positive and hence cannot be depended upon generally (he classes her as ‘one of those honestly inaccurate persons rather than a deliberate liar’ (p. 96) since she has no obvious motive for lying): ‘Well, what does it matter if she saw her face or not—it was [Lady Edgware]. And so with any other question. She knows. And so she answers questions in the light of her knowledge, not by reason of remembered facts. The positive witness should always be treated with suspicion, my friend. The uncertain witness who doesn’t remember, isn’t sure, will think a minute—ah! yes, that’s how it was—is infinitely more to be depended upon!’ ‘Dear me, Poirot,’ I said. ‘You upset all my pre-conceived ideas about witnesses.’ (pp. 95–96; Christie’s italics) Set against the secretary’s largely discredited evidence is the evidence of the guests at the dinner party, providing an apparently unshakeable alibi for Lady Edgware. As Inspector Japp says: ‘And unless fourteen people are lying, she didn’t do it.’ (p. 51) Later in the book, these assumptions are radically over-turned at the reversal stage. The dinner party evidence is shown to be unsound so that the alibi no longer holds, and it is discovered that Lady Edgware was in fact the woman who visited Lord Edgware’s home on the night of the crime and murdered him. In some respects, the discussions between Poirot and Hastings prepare us for this because Poirot has said only that the secretary’s evidence cannot be relied upon, not that it is actually wrong overall. Although a detail of what she has said has been inaccurate, that does not mean that her statement is untrue in its key point and hence an apparent lack of reliability (in the sense of providing robust evidence) may deflect readers from realising that this witness’s statement holds the truth about the presence of the murderer at the scene of the crime.

4. Conclusion Overall, this chapter provides fresh insights into how Agatha Christie’s work manipulates readers and this research is also relevant to understanding how misleading accounts are created in detective fiction generally and

(Un)reliability, Manipulation and Plot Reversals

189

in any heavily plotted work of this type. Control of credibility statements is a key strategy used by Agatha Christie to direct the assumptions of readers about the reliability and unreliability of witness statements. Such assumptions are nevertheless subjective and may turn out, at the reversal stage, to be inaccurate or irrelevant to the main case. One of the skills of detective writers is to make the contrasting representations of specific characters as reliable or unreliable appear realistic, both before and after the reversal. As Toolan (2009, p. 172) has pointed out, ‘Surprises are most effective when they are felt to be in no way absurd or inexplicable, but reasonable, possible, and even probable’.

Notes 1. Many discussions mention The Murder of Roger Ackroyd, commenting on the I-narrator’s unreliability in withholding key information about the crime. 2. Zunshine (2006) discusses Poirot’s comment to this effect in The Murder of Roger Ackroyd. See Kazmer (2016) for an examination of scepticism in the Miss Marple novels. 3. For example, in Murder on the Orient Express, And Then There Were None and Curtain: Poirot’s Last Case. 4. See Phelan (2005, pp. 49–53) for a model of different types of unreliability, including misreporting (in this story there seems to be accurate reporting) and mis-evaluating.

References Booth, W. C. (1991) [1961] The rhetoric of fiction. Harmondsworth: Penguin Books. Brütsch, M. (2014) ‘From ironic distance to unexpected plot twists: Unreliable narration in literature and film’, in Alber, J. and Krogh Hansen, P. (eds.) Beyond classical narration: Transmedial and unnatural challenges. Berlin: Walter de Gruyter, pp. 57–79. Christie, A. (1953) [1950] A murder is announced. Glasgow: Fontana. Christie, A. (1954) [1933] Lord Edgware dies. Glasgow: Fontana. Christie, A. (1957) [1926] The murder of Roger Ackroyd. Glasgow: Fontana. Christie, A. (1959) [1934] Murder on the Orient Express. Glasgow: Fontana. Christie, A. (1961) [1930] The murder at the vicarage. Glasgow: Fontana. Christie, A. (1970) [1934] ‘Sing a song of sixpence’, in The Listerdale mystery. London: Pan Books, pp. 64–80. Christie, A. (1976) [1975] Curtain: Poirot’s last case. London: Book Club Associates. Christie, A. (2003) [1939] And then there were none. London: Harper Collins Publishers. Emmott, C. and Alexander, M. (2010) ‘Detective fiction, plot construction and reader manipulation: Cognitive-rhetorical misdirection in Agatha Christie’s Sparkling cyanide’, in McIntyre, D. and Busse, B. (eds.) Language and style: In honour of Mick Short. Houndmills: Palgrave Macmillan, pp. 328–346. Emmott, C. and Alexander, M. (2014) ‘Foregrounding, burying, and plot construction’, in Stockwell, P. and Whiteley, S. (eds.) The handbook of stylistics. Cambridge: Cambridge University Press, pp. 329–343.

190

Catherine Emmott and Marc Alexander

Kazmer, M. M. (2016) ‘“One must actually take facts as they are”: Information value and information behaviour in the Miss Marple novels’, in Bernthal, J. C. (ed.) The ageless Agatha Christie: Essays on the mysteries and the legacy. Jefferson, NC: McFarland, pp. 114–129. Nünning, V. (ed.) (2015) Unreliable narration and trustworthiness: Intermedial and interdisciplinary perspectives, Berlin: Walter de Gruyter. Phelan, J. (2005) Living to tell about it: A rhetoric and ethics of character narration. Ithaca: Cornell University Press. Sanford, A. J. and Sturt, P. (2002) ‘Depth of processing in language comprehension: Not noticing the evidence’, Trends in Cognitive Sciences, 6 (9), pp. 382–386. Sanford, A. J. S., Sanford, A. J., Molle, J. and Emmott, C. (2006) ‘Shallow processing and attention capture in written and spoken discourse’, Discourse Processes, 42 (2), pp. 109–130. Schank, R. C. and Abelson, R. P. (1977) Scripts, plans, goals and understanding. Hillsdale: Lawrence Erlbaum. Toolan, M. (2001) Narrative: A critical linguistic introduction. London: Routledge. Toolan, M. (2009) Narrative progression in the short story: A corpus stylistic approach. Amsterdam: John Benjamins. Toolan, M. (2016) Making sense of narrative text: Situation, repetition, and picturing in the reading of short stories. New York: Routledge. Von Arnaud, A. and Martini, S. (2015) ‘Unreliable narration in law courts’, in V. Nünning (ed.), pp. 347–370. Zunshine, L. (2006) Why we read fiction: Theory of mind and the novel. Columbus: Ohio State University Press.

12 Metaphoric Interpretations of a Short Story by J. D. Salinger A Reader-Response Study Laura Hidalgo Downing

1. Introduction The present chapter explores metaphoric interpretations of the short story ‘A Perfect Day for Bananafish’, by J. D. Salinger, by real readers. As pointed out by research on Salinger, the short story is obscure, and lends itself to symbolic and metaphoric interpretation (Gwynn and Blotner, 1962, p. 110). Among the issues which are the object of discussion regarding the story, it is worth noting the possible interpretation of the bananafish tale, which gives the title to the short story, and the nature of the protagonist Seymour’s unease as motivated by either external or internal troubles (Wiegand, 1962, pp. 123–125). The story tells about the last hours of Seymour Glass, during a holiday at the seaside with his newly wed wife, Muriel, before he commits suicide. At the beach, Seymour meets a young girl, Sybil, to whom he tells his invented bananafish tale, which is open to interpretation. Following scholarship on Salinger’s stories (Grunwald, 1962, p. xiii), it is argued in the present chapter that the bananafish story is metaphoric of Seymour’s inner troubles rather than of external factors. In order to explore the possible metaphoric interpretations of the bananafish story and of the whole story itself in relation to the protagonist, Seymour, an analysis is carried out of readers’ responses to questions regarding the ending of the story and the metaphoricity of the bananafish tale and the story itself. With regard to the metaphoric interpretation of the whole story, previous scholarship argues that ‘A Perfect Day for Bananafish’, like the rest of the stories collected in the volume in which it was published, Nine Stories, can be interpreted to carry a symbolic meaning, related to the presence of a closing epigraph, a quotation of the well-known Zen koan (Gwynn and Blotner, 1962, p. 110): We know the sound of two hands clapping. But what is the sound of one hand clapping? The stories are thus seen as posing riddles involving some underlying problem or conflict in the characters. The idea of symbolic application

192

Laura Hidalgo Downing

is taken up in the present chapter as an investigation into the possible metaphoric interpretations of the bananafish story and the title of the short story itself with reference to the life and death of the protagonist, Seymour. The present study is inspired by Toolan’s work on narrative in the short story (2009, 2016) and takes up his suggestion that further studies on reader-response are necessary to check the intuitions which emerge from the observation of textual clues. The structure of the story is represented in Table 12.1 as divided into four scenes. As can be observed, the story takes place mostly in two longer scenes which happen simultaneously (1 and 2), followed by a brief encounter between Seymour and a woman in an elevator (3) and the abrupt ending with the suicide (4). The number of words and frequency distribution in each scene in the table show the extension of each scene, and allow the reader to see why the final scene, which is noticeably shorter than the previous ones, may be perceived as an abrupt ending. Because the title of the short story is ‘A Perfect Day for Bananafish’, the reader can anticipate that the bananafish story will play an important role in the interpretation of the story as a whole. The questions asked in the present study focus on the readers’ interpretation of the story ending, on the meaning of the title within the story as a whole and on the metaphoric interpretations of the bananafish story. In this way, an attempt is made to establish relations between the readers’ understanding of the abrupt ending to the story with Seymour’s suicide, on the one hand, and the bananafish theme which is a prominent one in the story, on the other.

2. Theoretical Concepts Michael Toolan’s publications have contributed enormously to the advancing of the study of how context shapes the process of interpreting a narrative text, in particular, of short stories (Toolan, 1992, 2009, 2013, 2016). The present study addresses two of the features of the relation between text and context dealt with by Toolan in his works on narrative, namely, the role of metaphor as a feature of (and potential challenge to)

Table 12.1 Number and Distribution of Words per Scene in the Short Story Scene

Number of Words

Percentage

Muriel talking on the phone to her mother in hotel Seymour on the beach with Sybil (Seymour tells her the story of the bananafish) Seymour in the elevator Final paragraph with Seymour’s suicide

1,912

47.5

1,766

43.9

238 84

5.9 2

Metaphoric Interpretation of Salinger Story

193

narrative coherence and the role of readers’ responses in the (metaphoric) interpretations of a short story. With regard to the role of metaphor, Toolan considers that this trope, together with narrative, is one of ‘the twin pillars of verbal art’, and that they manifest ‘divergent exploitations of a common foundation, namely the idea of repetition. Narrative entails repetition with change, and metaphor entails underlying repetition with overt, thought-stimulating (nonrepetitive) difference’ (2016, p. 26). The idea of underlying repetition may be interpreted as involving the perceived similarity between the two incongruous domains that are metaphorically mapped, which, however, are of course different. Metaphor may pose a challenge to the perception of narrative coherence, however, and ‘a reader’s ability to interpret superficially unconnected entities or processes as metaphorical can enable the recognition of coherence’ (Toolan, 2013, np). It may be argued that this effort of recognition of coherence by interpreting the bananafish tale as metaphoric within Salinger’s short story and as providing the title to the story itself is one of the efforts of interpretation required from the reader. In this process, the interpretations of the role of the title and of the ending of the story play crucial roles. The significance of titles in the understanding of the meaning of a work of fiction has been pointed out by Rabinowitz (2002, p. 302), who argues that titles guide the reader’s search for an interpretation of the narrative by focusing on a specific theme or character. Similarly, endings are considered to occupy privileged positions in short stories, as they play with the reader’s defeat (or fulfilment) of expectations built up during the reading process and often require the reader to carry out a reinterpretation of the story (Rabinowitz, 2002, pp. 303–304; Toolan, 2016, p. 219). The study of the metaphoric interpretation of literary texts has been the object of analysis of much research in stylistics and has been addressed from numerous perspectives, ranging from discourse cognitive stylistic analysis (Cameron and Low, 1999; Semino, 2008) to corpus-based approaches (Deignan, 2005) and, more recently, to reader-response-based studies (Sanford and Emmott, 2012; Whiteley, 2016). Semino defines metaphor as the ‘phenomenon whereby we talk and, potentially, think about something in terms of something else’ (2008, p. 1). In Conceptual Metaphor Theory, metaphor is a process in which the properties of a source domain are mapped onto a target domain, thus enabling the speaker to understand a more complex or abstract concept in terms of a more basic or familiar experience (Lakoff and Johnson, 1980). Metaphors are traditionally described as involving a relation of similarity across domains, though they have the underlying propositional form X is Y (as in ‘Juliet is the sun’), with no overt marking of a relation of similarity as necessary for the metaphoric meaning to be present (Gibbs, 1994). Metaphor is in the present chapter understood in a broad sense

194

Laura Hidalgo Downing

to include also other cross-domain phenomena, namely, similes. Thus, metaphoric meaning may be expressed or cued by means of comparative expressions such as ‘like’ and ‘as if’ (see Semino, 2008, p. 16; Sanford and Emmott, 2012, p. 65; Caracciolo, 2013, p. 62). In these cases, the expression introduced by the comparative expression is not metaphoric, but it introduces a source domain in terms of which the target is understood, as in ‘Juliet is like the sun’. Toolan, similarly, points out that comparative expressions motivate metaphoric and figurative interpretations, and he explains in detail how the form ‘as if’ induces figurative interpretation and invites the reader to enter a different world, to engage in imagination (2016, p. 232): Like all implicitly comparative constructions, they invite the reader to contemplate a different situation or world [. . .] embedded within the world and conditions of the currently portrayed world. (2016, p. 235) Considered from the perspective of Text World Theory, metaphors are world shifts from the main text world which create an alternative world in which we see something in terms of something else (Werth, 1999; Gavins, 2007). Metaphor, thus, has world-building properties, in the sense that it invites the reader to engage in a process of ‘double vision’, whereby one event is simultaneously seen and interpreted in terms of another (Werth, 1999, p. 318; Toolan, 2016, p. 237). Metaphor can thus be argued to have the power to evoke other realities, worlds or scenarios, to demand from the reader an engagement with an imaginative act, with what Toolan calls ‘double telling’ (2016, p. 237). Werth explains how metaphor works by providing ‘a sub-text which sheds light on the topic of discourse’ and ‘by opening up an area of experience in terms of which the discourse topic can be (partially) interpreted’ (1999, pp. 319–320). It may be argued that in ‘A Perfect Day for Bananafish’, the reader is invited to carry out a metaphoric world shift when the bananafish story is introduced. In this sense, the metaphoric nature of the tale contributes to the overall coherence of the short story by providing the link between Seymour’s problems, as set out in the first scene of the story, and the story resolution in the character’s suicide. The bananafish story metaphorically evokes the inner troubles of the protagonist and the possible events which may be the cause for his unease and lead to his tragic end. The present study explores the identification of figurative language, on the one hand, and of metaphors of the bananafish story, on the other, in students’ responses. Comparative expressions are explored as indicators of metaphoric interpretations in the students’ responses. For this purpose, the view of metaphor adopted is the one which is found in current discourse-based approaches to Conceptual Metaphor Theory (see, for example, Semino, 2008; Sanford and Emmott, 2012; Hidalgo Downing,

Metaphoric Interpretation of Salinger Story

195

2016; Whiteley, 2016). The present study makes use of the source and target domain mapping approach to metaphor, so that a given target domain may be accounted for by means of one or more source domains (Lakoff and Johnson, 1980, p. 10). The possibility that more than one source domain may yield an interpretation for a target is particularly significant in the present study, which stresses the importance of the fact that some readers may highlight one metaphorical scenario and one interpretation, while others may highlight another (Semino, 2008, p. 10). This provides evidence for the richness and complexity of the short story under study and shows the capacity of metaphor to provide interpretations of complex textual features. As pointed out earlier, the second dimension of the relation between text and context addressed in the present chapter is the reading process itself as carried out by real readers. Most studies on reader-response in the past have not focused on real readers’ responses, as observed by Peplow and Carter (2014). However, this has changed in the past 20 years, which have seen an increase in Experimental Studies of Literature (ESL), such as Emmott (1997), Miall and Kuiken (1998), Lohafer (2003), Sanford and Emmott (2012) and Miall (2014) among others, and in the Naturalistic Studies of Reading (NSR), which are based on an ethnographic perspective (Whiteley, 2016, also see Peplow and Carter, 2014, pp. 441–442). The present study lies somewhere in between the two main strands of reader-response studies as outlined by Peplow and Carter. It consists of a questionnaire with five open-ended questions about significant aspects of the short story. The aim of the study is qualitative, as in Naturalistic Studies of Reading, and not quantitative, unlike most Experimental Studies of Literature. For practical reasons, it was necessary to restrict the possible responses of the readers in time and space. This is why responses were elicited by means of a written questionnaire. The limitations of this approach are pointed out by Toolan (2009, pp. 1–2), who nevertheless admits that the benefits of the method go beyond its possible limitations.

3. Data and Method The data consists, first, of the short story ‘A Perfect Day for Bananafish’, by J. D. Salinger (1953). Second, it consists of the answers to five questions in questionnaires on the short story completed by 36 third-year Spanish students of the English Studies Degree at Universidad Autónoma de Madrid, 2016/2017. The reading experiment is thus performed on readers who use English as a language of instruction in their undergraduate studies on English Linguistics and Literature. Students had a background in North American twenieth-century literature, but they had not read the short story before. They were familiar with Salinger, in most cases as being the author of The Catcher in the Rye.

196

Laura Hidalgo Downing

Students were handed the short story to read in class and were asked to answer the following questions. They were given 2 hours, after which they handed in their written responses to the teacher. The questions posed to the students were the following: 1. 2. 3. 4.

Why does Seymour Glass commit suicide? Why does it make sense in terms of the story development? Is there any figurative language, like metaphors, in the story? Is the story of the bananafish metaphorical or just a story? If metaphoric, what is it a metaphor of? 5. Why do you think the story is called ‘A Perfect Day for Bananafish’? Following methodology on the difficult task of how to design questionnaires (Rasinger, 2008), the five questions aim to identify, first, the interpretations of the story ending, Seymour’s suicide, and second, the possible metaphoric interpretations of the bananafish tale and the story as a whole. It is expected that a link will be established between the metaphoric interpretations and the story ending. With regard to the metaphoric interpretations, I wished to make sure that students were consciously identifying figurative language, hence questions 3 and 4, which explicitly address figurativity. Question 3 checks whether students are able to identify specific figurative expressions in the short story, while question 4 requires the possible metaphorical interpretation of the bananafish story. By eliciting the interpretation of the bananafish story, the questionnaire requires students to identify the target domains of the metaphor. In question 5 students are asked to explain the title of the story. As observed in the ‘Theoretical Concepts’ section, titles are considered to play a significant role in the processing of coherence and the interpretation of a piece of fiction. Question 5 may elicit target domains for the metaphorical interpretation of the whole story in the light of the bananafish tale. The difference between questions 4 and 5 has to do with the discursive level: While question 4 refers to the story of the bananafish told by Seymour to Sybil, question 5 refers to the title of the short story. Both questions involve establishing a connection between the bananafish story and the overall themes of the short story. The bananafish story tells of the life and death of imaginary creatures, ‘bananafish’. The tale could be interpreted literally as a story of the life habits of bananafish, just a tale told by Seymour to a child. However, the natural search for coherence, within the short story and between the story and the title, leads the reader to find further meaning in the tale, as being related to the problem set out in the beginning of the story, Seymour’s troubles, and the abrupt solution or ending to this problem in his suicide. In their search for a meaning to the story, students are expected to elicit rich metaphorical interpretations. Responses were coded with a number and the letters ‘F’ for female and ‘M’ for male.

Metaphoric Interpretation of Salinger Story

197

4. Results and Discussion The first two questions concern the ending of the short story, in which the protagonist, Seymour, commits suicide. Table 12.2 collects the main answers to question 1: Why does Seymour Glass commit suicide? The most frequent reason provided for Seymour’s suicide is his mental disorder, followed by his experience at war and by personal problems in communication. One student answers, ‘I don’t know’. The reference to mental illness and war as the main reasons for the suicide of the protagonist provides a clue as to which themes mentioned early in the story are considered relevant for the further development of narrative coherence. Thus, both Seymour’s mental instability and the fact that he has been to war are referred to by Muriel’s mother in her telephone conversation with her daughter. The references to these two facts, however, remain superficial and often take place in incomplete sentences during the conversation, leaving it to the reader to infer the possible relevance of one or both of the events in the subsequent development of the story. In order to complement the responses to question 1 and explore the motivations in the textual development of the story, students are asked question 2: Why does it [i.e. Seymour’s suicide] make sense in terms of the story development? The answers are collected in Table 12.3. Most students establish a connection between Seymour’s suicide and the fact that his possible mental instability is mentioned by Muriel’s mother on repeated occasions in scene 1. This indicates that among the possible clues set out for the reader in scenes 1 and 2, students perceive the theme of Seymour’s possible mental illness as the most likely Table 12.2 Number of Responses per Theme in Question 1 Theme

Number of Responses

Mental illness War Problems in life and communication Other

14 11 7 4

Table 12.3 Answers to the Question: Why Does the Ending Make Sense in Terms of Story Development? Answer

Number of Responses

It is mentioned explicitly in scene 1 and/or reference to mental disorder Structure of the short story I don’t know, it doesn’t make sense Other

21 5 4 3

198

Laura Hidalgo Downing

to motivate the ending in suicide. This tendency confirms the intuition set out in the introduction of this chapter, that Seymour’s troubles are of an internal nature, rather than external (namely, the superficiality of his wife, the pressure of his mother-in-law). Five students explain that the ending of the story forms part of the conventions and expectations regarding the short story itself, as in examples (1) and (2): (1) Because this short story tries to shock you, and to do that in those times, death and suicide was the best option to do. (M13) (2) Because the story needed to have an important change, and a deeper meaning. (F27) These replies show that perception of coherence in a short story depends, not only on the clues that are interpreted on reading the text, but also on the previous knowledge of twentieth-century short stories as a genre. Considered together, responses to questions 1 and 2 reveal that there is a correspondence between the observation of textual clues in the story (question 2) and the interpretation of the end made by readers (question 1). Question 3, whether there is figurative language in the story, elicits positive answers from all students except four, though only a few provide examples of metaphors. Among these, most mention the bananafish as the possible source for a metaphoric interpretation and present it as a source domain for Seymour’s mental illness as target. Interestingly, a student mentions Muriel’s reaching of puberty as a metaphor ‘for her lack of interest in others’ (F5) and points out the significance of the semantic domain of clothes and fashion, which, the student argues, is said to represent the materiality of the world shared by Muriel and her mother. Only this student focuses on a possible metaphoric interpretation of Muriel’s character. This perspective would align itself with readings of the short story as a criticism of the materialistic American post-war society and Muriel’s superficiality (Wiegand, 1962). When asked questions 4 and 5, regarding the metaphoricity of the bananafish story and the meaning of the title within the whole story, all students except two provide metaphorical explanations. The responses to questions 4 and 5 are examined first in order to identify overt markers of metaphoricity. The cues for metaphoricity used by the students are collected in Table 12.4. Because question 4 asked explicitly about the metaphoricity of the story, most students answered, first, with expressions such as ‘it is a metaphor (for/of)’. Many students additionally used other expressions to indicate patterns of metaphoric similarity, such as ‘like’ and ‘stands for’, or equivalence, such as ‘means’ and ‘refers to’. Three students mentioned directly the play with words made by the girl Sybil, ‘see more glass’, as referring to the protagonist in a metaphorical way. This view links to the

Metaphoric Interpretation of Salinger Story

199

Table 12.4 Cues for Metaphoric Explanations in Students’ Responses Cue for Metaphoric Explanation

Question 4

Question 5

It is a metaphor (for) Like Refer/make reference to Mean Feels identified, feels the same Stand for Represents Has to do with Explain Is a sign, symbolises The common thing between See more glass Something else, opposite, irony Imaginary world

32 5 3 2 2 2 1 2 1 1 1 3 0 0

5 6 3 0 0 1 3 0 0 0 0 0 3 5

criticism on Salinger which highlights his characters’ search for spiritual enlightenment (Wiegand, 1962, p. 127). Example (3) illustrates the use of the comparative form ‘like’ to connect the source and target domains in the bananafish metaphor: (3) The metaphor of the bananafish is like war, soldiers enter a hole, they kill people and go mad, so they are not able to come back home because most of them die, or have mental problems. (F30) While in replies to question 4 there was a greater variety in expressions of similarity and equivalence than in responses to question 5, replies to question 5 provided more complex information regarding the nature of the metaphoric interpretation. Thus, in the case of question 5, three students mentioned that the title in the short story refers to ‘something else’, which may be interpreted as opposite, sarcastic or ironic. Six students provided interesting explanations regarding the imaginary nature of the bananafish and the bananafish story, and how this may be relevant to the interpretation of the story. I now turn to the analysis of the source and target domains elicited in the responses to questions 4 and 5. With regard to question 4, whether the story of the bananafish is metaphorical, the answers provided by most of the students contain a great amount of detail. What is most relevant in the responses is which aspects of the story are highlighted by the choice of particular metaphorical interpretations. The choices show that two specific scenarios draw the students’ attention onto the interpretation of the bananafish story: the war scenario and the mental illness scenario. With regard to the war scenario, some students explicitly identify mappings between aspects of the bananafish story and the experience of

200

Laura Hidalgo Downing

soldiers at war, more specifically, World War II. Thus, students establish mappings between the bananafish going into a hole and eating too many bananas and dying as source domains for the targets, soldiers going into a trench and killing at war. This is illustrated by examples (4) and (5). (4) I think fish swimming into a hole and dying is a metaphor in which fish are soldiers and the hole is a trench/war. (M11) Some students identify mental illness as a result of the war experience. Thus the image of the bananafish getting stuck and dying is interpreted as explaining how soldiers develop mental problems or die during the war. Some students develop details of the metaphorical scenarios by expanding on the potentiality of the source domains. Thus, students (F3) and (M11) make reference to the process of change of the bananafish and the men who go to war. As these students explain, the bananafish are apparently normal, though they lead tragic lives once they enter the hole which they cannot leave. This is understood as explaining the tragic life of soldiers going to war. More specific interpretations for details provided are the number of bananafish that are eaten as representing the number of men killed by a soldier, the banana fever as representing depression or mental illness and the bananafish behaving like pigs as representing soldiers killing other men, as in example (5). Here, numerous cues for metaphor can be identified, the comparative form ‘just like’ and the equivalence formula X could be Y: (5) It’s a metaphorical story. The bananafish could be soldiers, since their lives are ‘tragic’. And just like the bananafishes, they look ordinary, but once they join the Army they become someone else. The bananafish that has eaten seventy-eight bananas could be a soldier that has killed seventy-eight men, and once you kill a human being, you are not the same any more, you enter a hole that you can never leave, so you develop a post traumatic disorder, or depression, maybe both, just like the bananafish get banana fever. And then you die, because nobody really knows how to help you and the disease ends up taking over you. (F3) One student specifies that the war that is referred to is World War II, as in example (6), and another student explains that the bananafish story is a metaphor of the Holocaust. Some students expand on the source domains of entering the hole and not being able to get out as representing the way in which, after the war, soldiers are not able to adapt to society. Student (F18) explains explicitly that the bananafish that are not able to fit through the door are understood as the soldiers who cannot fit in society after the war. In these metaphorical interpretations, the fact that bananafish enter a hole and cannot leave because they do not fit through

Metaphoric Interpretation of Salinger Story

201

the door is interpreted in terms of the container metaphor. Thus, war is conceptualised as a container for traumatic experience which prevents soldiers from recovering their normal lives. More specifically, (F2) in example (7) expands on this metaphor in order to suggest that it is not the physical war but the long-lasting experience of it which becomes a container for trauma. F2 uses the mind is a prison metaphor to explain the soldiers’ traumatic experience: (6) It is a metaphor. Of WW2. The fish are the men, the soldiers. Bananas can stand for victims or violence as a general concept. They see and inflict so much of it they never can ‘get out’, they never can overcome the trauma of war and are stuck in this instable state of mind until their death, which is due to they not managing to ‘fit through the door’: fit into society once more, and be a part of it. (F18) (7) I think it could also be a metaphor for the soldiers. They are regular men that go to the war to fight, and when they come back, if they do, they cannot escape their mind, which is now their prison because of all they’ve seen. (F2) It can be argued that the overarching conceptual metaphor in this scenario is the story of the bananafish is the story of men at war. The mappings between source domains and target domains developed by the students are shown in Table 12.5. With regard to the mental illness scenario, there are various degrees of specificity in the development of the metaphorical explanation. Starting from the more general interpretations, some students interpret the story of the bananafish as being a metaphor of life, more specifically of moments in life which are critical or in which there is a loss of control. Within this scenario, most students interpret the story as metaphorical of Seymour’s life, as in (8): (8) Yes, Seymour died at the end like the bananafish. It’s metaphorical. It’s a metaphor of Seymour’s life. (F22)

Table 12.5 Sources and Targets for the war Metaphorical Scenario Sources

Targets

Bananafish Hole Eating bananas Behaving like pigs Getting out of the hole Getting stuck in the hole Fit through the door

Soldiers War, trench, their mind, traumatic experience of the war Killing other soldiers Killing human beings Overcoming traumatic experience Developing a mental disorder and/or dying Fit into society after the war experience

202

Laura Hidalgo Downing

Some focus specifically on the idea of death and suicide. Other students expand on mental illness, more specifically, on Seymour’s mental illness. The underlying metaphor highlighted by these examples is the story of the bananafish is the story of Seymour’s life/death/suicide and mental illness. The sources and targets identified by the students are collected in Table 12.6. Finally, some students combine the war and mental illness scenarios in order to explain how the bananafish story helps understand Seymour’s traumatic experience during the war and how this influences the later development of his life and his death. This interpretation is the most complex one and provides an approach to the short story as highly coherent, since the two scenarios are seen as connected. Example (9) explains that the bananafish dying after eating bananas is interpreted as Seymour having killed other human beings and then committing suicide because of the shame. This student identifies the telling of the bananafish story to Sybil as an anticipation of what he is going to do and also mentions shame as the motivation for Seymour’s suicide. (9) It may stand for the war experience. The man says: ‘They’re very ordinary looking fish but once they get in they become like pigs’ and then ‘I’ve known some bananafish to swim into a banana hole and eat as many as . . . ‘. Maybe eating means killing. The man says that after ‘eating’ bananafish die, so maybe it’s because of the shame (that is, suicide). Maybe he’s just telling the little girl (indirectly) what he’s about to do. (F35) The relation between the two connected scenarios and the sources is shown in Table 12.7. I now turn to the answers provided to question 5. The themes identified by students as related to the title of the short story are shown in Table 12.8. Table 12.6 Sources and Targets for the mental illness Metaphorical Scenario Sources Life and death of the bananafish

Targets

Human life and death and life and death of Seymour Bananafish Humans, Seymour Bananafish Mental illness (bananafish do not exist) Hole War experience, Seymour’s mental illness Banana Obsession, mental illness Eating bananas Losing control, becoming depressed, mentally ill Getting stuck in the hole and dying Becoming mentally ill, committing suicide

Metaphoric Interpretation of Salinger Story

203

Table 12.7 Sources and Targets for the Seymour’s traumatic war experience and suicide Metaphorical Scenario Sources

Targets

Bananafish Eating bananas Bananafish’s incorrect behaviour Story of bananafish

Seymour Killing men at war, committing suicide Seymour’s incapacity to readjust to society Seymour’s intention to commit suicide

Table 12.8 Answers to the Question: Why Do You Think the Story Is Called ‘A Perfect Day for Bananafish’? Answer

Number of Responses

Death and suicide The role of imagination Irony The scene on the beach Insanity Other

8 8 3 3 2 2

Two students say they do not know what the title refers to. However, it is worth noting that these students provide answers to the questions on the metaphorical interpretation of the story. Of the interpretations, the one that the title may refer to death and suicide, insanity and war is related to the metaphorical interpretations provided in the previous questions. More specifically, students answering this question explain the significance of the expression ‘it’s a perfect day for bananafish’ as meaning ‘it’s a perfect day to die’, as in examples (10) and (11): (10) Because he feels like a bananafish, and maybe for him is the perfect day to die. (F23) (11) Here bananafish is a metaphor for suicide. Seymour decided it is a perfect day to commit it. (F24) Three students identify the title as ironic, even though they do not identify irony as figurative language in the previous questions on the story: (12) Because the author wants to reflect a ‘normal story’ but it is the opposite. (M8) (13) It is sarcastic because it would be a perfect day but it turns out to be the other way. (F19) (14) It’s irony, it diminishes the importance of the story himself or the bad sides in order to you to take the ‘fun part’, the fun side of Seymour. (F20)

204

Laura Hidalgo Downing

Among the answers which do not provide a metaphorical interpretation, some students highlight the importance of the scene on the beach, which is explained as follows: (15) It may refer to a perfect day to be with Sybil. (F25) (16) Because it’s the most ‘innocent’ thing in the story. I think it also makes the reader blink and realise how weird Seymour’s behaviour is. (F31) The theme of innocence is mentioned by another student, who also points out the important role of imagination in the telling of the bananafish story. Imagination is mentioned by several students, either with reference to Seymour’s capacity to create a story or as evidence of his mental illness. (17) Maybe the imaginary creature that he has created ‘the bananafish’ keeps him close to the real world. He shares it with a child, so is an innocent creature that he can share just in a child world. So, it’s always perfect for a bananafish because it is not a real figure of this world which he can communicate. (F5) Student (F18) explains the process by which the title influences and guides the reader’s attention and interpretation of the story till the ending is reached. (18) It puts emphasis, before the last line, on the bananafish, so the reader knows that the conversation with Sybil is important, that it will help make sense at the ending, which can only be ‘appreciated’ in parallel to the bananafish story. (F18)

5. Conclusion The questions posed to the readers focused on the interpretation of the abrupt ending in the story, its relation with the title of the short story and the metaphorical interpretations provided by the readers. The results of the questionnaire showed that students interpret the ending of the short story in relation to clues they identified in the first two scenes, even if for many the ending was unexpected. With regard to the metaphorical interpretations of the bananafish story, most students were able to provide detailed metaphorical scenarios to account for the role of the bananafish story within the short story itself and as a title to the short story. Most students highlighted the theme of Seymour’s alleged mental illness as a justification for the ending and as an underlying metaphorical theme for the bananafish story. However, other scenarios were highlighted as well, such as the possible traumatic experience of Seymour at war and his

Metaphoric Interpretation of Salinger Story

205

problems in communicating with other people. These results suggest that students made use of textual clues provided in the first scene regarding both Seymour’s alleged mental illness and his experience at war, even if these facts were presented in a backgrounded way, as part of incomplete reported thoughts and speech of Muriel’s mother. These results confirm the intuition set out in the introduction to this chapter, that Seymour’s troubles will be perceived as being of an internal nature, rather than external. Only one student mentioned the possibility of a metaphorical interpretation of the story as a criticism of materialism in American society. No references were made to other possible metaphorical interpretations of the story which may have arisen from familiarity with Salinger’s works as a whole and criticism of his stories. The metaphoric interpretations thus show that students take up threads introduced by textual clues early in the story, together with background knowledge and expectations regarding the genre and conventions of twentieth-century short stories. With regard to the ability of students to explain metaphoric meanings, their responses have shown that they use a variety of comparative forms and paraphrases in order to introduce the mappings between source and target domains.

References Cameron, L. and Low, G. (1999) Researching and applying metaphor. Cambridge: Cambridge University Press. Caracciolo, M. (2013) ‘Phenomenological metaphors in readers’ engagement with characters: The case of Ian McEwan’s Saturday’, Language and Literature, 22 (1), pp. 60–76. Deignan, A. (2005) Metaphor and corpus linguistics. Amsterdam: John Benjamins. Emmott, C. (1997) Narrative comprehension: A discourse perspective. Oxford: Oxford University Press. Gavins, J. (2007) Text world theory: An introduction. Edinburgh: Edinburgh University Press. Gibbs, R. W. Jr. (1994) The poetics of mind: Figurative thought, language and understanding. Cambridge: Cambridge University Press. Grunwald, H. A. (ed.) (1962) Salinger: The classical critical and personal portrait. New York and London: Harper. Gwynn, F. L. and Blotner, J. L. (1962) ‘A slight case of incest’, in Grunwald, H. A. (ed.) Salinger: The classical critical and personal portrait. New York and London: Harper, pp. 86–114. Hidalgo Downing, L. (2016) ‘Metaphor and metonymy’, in R. Jones (ed.) The Routledge handbook of language and creativity. London: Routledge, pp. 107–128. Lakoff, G. and Johnson, M. (1980) Metaphors we live by. Chicago: University of Chicago Press. Lohafer, S. (2003) Reading for storyness: Preclosure theory, empirical poetics and culture in the short story. Baltimore, MD and London: The John Hopkins University Press.

206

Laura Hidalgo Downing

Miall, D. S. (2014) ‘Emotions, feelings and stylistics’, in Stockwell, P. and Whiteley, S. (eds.) The Cambridge handbook of stylistics. Cambridge: Cambridge University Press, pp. 424–438. Miall, D. S and Kuiken, D. (1998) ‘The form of reading: Empirical studies of literariness’, Poetics, 22, pp. 327–341. Peplow, D. and Carter, R. (2014) ‘Stylistics and real readers’, in Burke, M. (ed.) The Routledge handbook of stylistics. London: Routledge, pp. 440–454. Rabinowitz, P. (2002) ‘Reading beginnings and endings’, in Richardson, B. (ed.) Narrative dynamics. Columbus: The Ohio State University Press, pp. 300–313. Rasinger, S. (2008) Quantitative research in linguistics: An introduction. London and New York: Continuum. Salinger, J. D. (1953) ‘A perfect day for bananafish’, in Nine stories. Boston, MA: Little, Brown and Company. Sanford, A. J. and Emmott, C. (2012) Mind, brain and narrative. Cambridge: Cambridge University Press. Semino, E. (2008) Metaphor in discourse. Cambridge: Cambridge University Press. Toolan, M. (1992) Language, text and context: Essays in stylistics. London: Routledge. Toolan, M. (2009) Narrative progression and the short story: A corpus stylistic approach. Amsterdam: John Benjamins. Toolan, M. (2013) ‘Coherence’, Paragraph 24, in Hühn, P. et al (eds.) The living handbook of narratology. Hamburg: Hamburg University. Available at: www. lhn.uni-hamburg.de/article/coherence (Accessed 3 March 2018). Toolan, M. (2016) Making sense of narrative text. London: Routledge. Werth, P. (1999) Text worlds: Representing conceptual space in discourse. London: Longman. Whiteley, S. (2016) ‘A cognitive poetic approach to researching the reading experience’, in Otterholm, K., Skjerdingstad, K. I., McKechnie, L. E. F. and Rothbauer, P. (eds.) Plotting the reading experience: Theory, practice, politics, Waterloo: Wilfred Laurier Press, pp. 99–114. Wiegand, W. (1962) ‘The cures for banana fever’, in Grunwald, H. A. (ed.) Salinger: The classical critical and personal portrait. New York and London: Harper, pp. 115–136.

Section 4

Projection and Positioning

13 Winnie-the-Pooh Goes Paralinguistic . . . Manuel Jobert

1. Introduction Winnie-the-Pooh (1926) and The House at Pooh Corner (1928) are undoubtedly among the most widely read children’s books in English.1 Although quite a lot is known about the writing of the Pooh stories and the autobiographical backdrop (Thwaite, 1990), the books have received relatively little critical attention. Yet, the general public’s interest is yet to waver and, leaving aside the Walt Disney adaptation, a grand exhibition was devoted to Winnie-the-Pooh at the Victoria and Albert Museum entitled ‘Winnie-the-Pooh: Exploring a Classic’ in 2017 and, the same year, the biopic Goodbye Christopher Robin came out. Winnie-the-Pooh (WP) and The House at Pooh Corner (HPC) are stories to be read both to and by children (Toolan, 2001, pp. 178–179). In this respect, the traditional dichotomy between scripta and verba seems to be cancelled (Manguel, 1997, p. 45; Piaget, 2002).2 This is all the more striking in the case of the Pooh stories as several audio versions exist (including Winnie-the-Pooh—The Collection read by Alan Bennett in 1984 and the dramatisation of The House at Pooh Corner in 2010).3 Orality seems indeed to be a central element: Chapter 1 of WP dramatises the reading-to-a-child discourse situation. Both WP and HPC are packed with dialogue—making the texts closer to the mimetic end of the continuum than to the diegetic one—and the characters’ voices and manners of speaking are often commented upon. All these contribute to the success of the Pooh stories. Chris Powling (quoted in Thwaite, 2017, p. 156) explains: The permanence of the Pooh books has nothing whatever to do with their psychological depth or the sharpness of their social comment or their status as morality. These don’t matter a lot. What’s important through and through, is their success as storytelling. And this is a triumph. It survives Christopher Robin. It even survives that odd tone-of-voice which, for all Milne’s simple language, never quite settles for a child.

210

Manuel Jobert

The success of the Pooh books in terms of storytelling is based on the overall oral component of the story as well as the presence of narrative comments that make the characters’ voices real, bridging the gap, as it were, between reality and fiction. This study will be concerned with the characters’ voices, more specifically, with the way emotions and attitudes are expressed and conveyed to the reader/listener. The first part of this chapter presents paralinguistic vocal features (PVFs) and their textual correlates, paralinguistic vocal markers (PVMs), and provides a workable descriptive framework showing how they can be used in stylistic analysis. The second part is devoted to some paralinguistic analyses of the Pooh stories.

2. Using PVFs in Stylistic Analysis 2.1 Paralanguage and Kinesics PVFs are not, as yet, part of the traditional stylistic toolbox.4 Indeed, PVFs apply to spoken discourse, and although spoken texts clearly fall within the remit of stylistic analysis, they tend not to take pride of place. Still, PVFs are also relevant to written texts. Before tackling this subject, it seems important to define PVFs. The term ‘paralanguage’, from which the adjective ‘paralinguistic’ is derived, was coined by Hill (1958) and popularised by Trager (1958). The 1962 ‘Conference on Paralinguistics and Kinesics’ held at the University of Indiana marked the beginning of paralinguistic studies. The main contention about paralinguistic phenomena is the place kinesics. Some scholars, including Abercrombie (1968), believe it is an integral part of paralanguage; others, like Crystal (1969), think the opposite. The latest definition put forward by Crystal (1997, p. 227) is worth quoting in full: Paralanguage: A term used in suprasegmental phonology to refer to variations in tone of voice which seem to be less systematic than prosodic features (especially intonation and stress). Examples would include the controlled use of breathy or creaky voice, spasmodic features (such as giggling while speaking), and the use of secondary articulation (such as lip-rounding or nazalization) to produce a tone of voice signalling attitude, social role or some other language-specific meaning. Some analysts broaden the definition of paralanguage to include kinesic features; some exclude paralinguistic features from linguistic analysis. This definition is a good summary of the problems surrounding paralanguage ranging from an umbrella term including almost everything and proving, therefore, rather useless, to its very exclusion from the realm of linguistics proper. Only the vocal aspect of paralanguage will be

Winnie-the-Pooh Goes Paralinguistic . . . 211 considered in this study and kinesics will be considered as a separate semiotic network. Indeed, no direct or systematic mapping with language has yet been established with certainty. 2.2 Paralinguistic Vocal Features (PVFs) Trager (1958) and Crystal (Crystal and Quirk, 1964; Crystal and Davy, 1969; Crystal, 1969) have offered competing analytical frameworks for paralinguistic phenomena. The main difference between these two linguists is their theoretical stance rather than the detail of their respective analytical frameworks. Trager suggests clear-cut divisions between (i) voice set, i.e. the speaker’s personal characteristics, (ii) paralanguage, (iii) language and (iv) kinesics. Although his model is clear, it raises questions, in particular with regard to prosodic features, left hanging, as it were, between language and paralanguage. On the other hand, Crystal (1975, pp. 60–61) advocates an intra-linguistic approach based on what he dubs ‘a scale of linguisticness’: [I]t is suggested that a more appropriate model of the non-segmental sound system is to introduce the concept of SCALE of linguisticness, ranging from ‘most’ to ‘least’ linguistic. At the ‘most linguistic’ polarity would be classified those features of utterance most readily describable in terms of closed systems of contrasts, which have a relatively clear phonetic definition, which display evidence of a hierarchical structure (particularly syntax), e.g. tone-unit boundaries, nuclear tone type and placement, and other ‘prosodic’ features. [.  .  .] At the other, ‘least linguistic’ end would be placed those features of utterance which seem to have little potential for entering into systemic relationships, which are relatively indiscrete, and which have a relatively isolated function and little integrability with other aspects of language structure, e.g. breathy or raspy vocal effects—effects which are often confused with voice quality characteristics on first hearing. Vocal effects lacking any semantic force would then be considered non-linguistic, and would thus fall under the heading of either voice quality or physiological reflexes (such as coughing). The different analytical frameworks put forward by Crystal evolved but the idea of a continuum between the ‘least linguistic’ and the ‘most linguistic’ remains central to his theory. In this study, I shall adopt the following framework (Jobert, 2014) based on Crystal’s ‘scale of linguisticness’: Both Trager and Crystal mention the individual norm as an essential factor in paralinguistic analysis, i.e. the basis from which paralinguistic deviation is calculated. Yet, Brown (1990) is the first to define PVFs in connection with the speaker’s norm in a systematic way. Her descriptive

212

Manuel Jobert

1. Para-prosodic features Loudness Pitch Tempo Rhythm Pause 2. Vocal qualifiers Types of tension —articulatory precision (precise/slurred) —articulatory setting (tense/lax) —lip setting (smiling/pursed) Types of phonation —falsetto voice —whispered voice —creaky voice —harsh/husky voice —breathy voice Peripheral modes —culturally coded onomatopoeia (‘tsk’, ‘tut’, etc.) —intentional cough/throat clearing, etc. 3. Vocal qualifications Laughter Crying Sighing Figure 13.1 Descriptive Framework for Paralinguistic Vocal Features

framework is simpler than those of Trager and Crystal as she selects 11 PVFs only. She introduces the notion of ‘markedness’, defined as a deviation from the speaker’s norm. The speaker’s norm is a complex notion and three main planes need to be considered. i. The speaker’s physiological norm based on sex, age, size and weight as these characteristics determine the basic phonation type of individuals as well as their pitch and loudness. ii. The spatial (variation in accents) and sociological characteristics (variation in register, for instance).

Winnie-the-Pooh Goes Paralinguistic . . . 213 iii. The pragmatic norm, essentially studied by Politeness Theory. Proxemics (Hall, 1984) also plays a major role as you indeed adjust your loudness to the space between you and your interlocutor. It appears that determining a speaker’s norm involves taking into account a wide array of variables, and that is precisely what we do on an everyday basis when interacting with people. Yet, Lyons (1972, pp. 72–73) points to another difficulty when he claims that speakers can always add a PVF to their speech to the extent that it becomes permanent and may be mistaken for part of the individual’s norm. 2.3 From Paralinguistic Vocal Features to Paralinguistic Vocal Markers From the start, scholars were interested in the way writers used PVFs in fiction. Trager (1958, p. 20) notes: ‘In literary works, the acute author will say, “Away” was said in a muffled tone [.  .  .] or the like’. Brown (1990) looks for correlations between PVFs and verbs or adverbs used in the reporting clauses of direct speech. Of course, she does not suggest there is a straightforward mapping between PVFs and emotions. Intuitively, we know that anger may be realised with extreme loudness, or with extreme softness. She suggests that emotions trigger certain PVFs. Some terminological clarification is needed here. Of course, PVFs do not appear in written texts as such. I suggest the term ‘paralinguistic vocal markers’ (PVMs) to refer to the phrases and expressions used to present PVFs. When dealing with PVFs, writers can choose to encode either a phonetic feature or an attitudinal feature. Prototypically, two types of comment may be used: Rabbit said in a loud voice5 ‘In you go, Roo!’ (WP, p. 91; phonetic feature encoded) ‘Oh!’ said Roo rather sadly (HPC, p. 61; attitude encoded) Both types can of course be associated: ‘It’s Piglet!’ cried Pooh eagerly. (HPC, p. 39) The first type of PVMs, referring to certain phonetic features, will be called direct PVMs, whereas those indicating attitudes will be called indirect PVMs. In either case, a certain amount of inference is needed and the interpretation of these comments is heavily context-dependent. In the case of direct PVMs, the reader has to infer which emotion triggered this

214

Manuel Jobert

phonetic realisation by looking at the content of direct speech and the immediate context. In the case of indirect PVMs, readers tend to adjust their reading by supplying the adequate PVFs, whether they are reading aloud or not (Brown, 1990). The aim of this brief presentation was to show that PVFs and their written counterparts, PVMs, are worth taking on board in stylistic analysis. The contrast between, say, Ernest Hemingway and Henry James can easily be apprehended by the way they use, or do not use, as the case may be, PVMs.6

3. PVMs in the Hundred Acre Wood Before studying PVMs per se, it is important to understand the narrative situation, starting with the autobiographical dimension. WP represents the storytelling sessions A. A. Milne had with his son. The real characters (the father and the son) are dramatised, which constitutes a first step towards fictionalisation. The second step is when Christopher and his cuddly toys become characters in the stories. Throughout the process, everything is done to facilitate the identification of the readers/listeners. The ‘narrating voice’ as well as the vocal effects play a crucial role in the fictionalisation process. 3.1 The ‘Narrating Voice’ and Deictic Shifts Quite aptly for a children’s narrative, WP starts with the dramatisation of bedtime-reading, thus foregrounding the spoken voice. The reading-to-achild discourse situation is brought about with the first-person narrating voice interacting directly with Christopher Robin and commenting on a previous episode that supposedly took place outside the fictional world: When I first heard his name, I said, just as you are going to say, ‘But I thought he was a boy?’ ‘So did I’, said Christopher Robin. (WP, p. 1) The story that is going to be told will be about Winnie-the-Pooh and his friends, and the audience (Christopher and his Teddy-bear Winnie) will be the main characters. The transition from the discourse situation to the stories proper is progressive. The tale begins with the traditional incipit ‘Once upon a time’. The fairy-tale world, however, as soon as it is conjured up, is conflated with the time frame of the discourse situation:7 Once upon a time, very long time ago now, about last Friday, Winniethe-Pooh lived in a forest all by himself under the name of Sanders. (What does ‘under the name’ mean? asked Christopher Robin.

Winnie-the-Pooh Goes Paralinguistic . . . 215 ‘It means he had the name over the door in gold letters and lived under it.’ ‘Winnie-the-Pooh wasn’t quite sure,’ said Christopher Robin. ‘Now I am,’ said a growly voice. ‘Then I will go on,’ said I.) (WP, p. 2) The first lines accurately capture the first minutes of bedtime-reading when children are slowing slipping into the fictional world but are still very much aware of the presence of the reading parent. By dramatising this moment of transition, when the child-audience hovers between reality and fiction, the author manages to fictionalise the discourse situation, making it part of the fictional world.8 The explanation provided regarding the meaning of ‘under the name’ is immediately confirmed by the illustration of Winnie-the-Pooh sitting on a fallen tree and looking up, with, above his head, a notice saying ‘SANDERZ’. The purpose is to incite the child-audience to enter the fictional word even more deeply. The narrating voice, the illustration and the fact that the scene should take place in a forest9 all contribute to the child’s ever-growing cognitive implication in the fictional world. Finally, this transitory passage gives voice to the Teddy-bear sitting next to Christopher Robin as the introductory clause ‘said a growly voice’ indicates. Precisely because readers are not told which of the two human beings present (Christopher and his father) speaks in the bear’s stead, the floor is clear for a world in which cuddly animals speak and interact with humans. In the context of an oral narrative, it comes as no surprise that the bear’s voice should be marked paralinguistically in order to suggest its ‘bear-like’ quality (growl). A first blip in this deictic shift into the fictional world occurs when Christopher Robin appears as a character, along with Winnie-the-Pooh. When he discovers another enactor of himself, a deictic POP10 occurs: And the first person he thought of was Christopher Robin. (‘Was that me?’ said Christopher Robin in an awed voice, hardly daring to believe it. ‘That was you.’ Christopher Robin said nothing, but his eyes got larger and larger, and his face got pinker and pinker.) (WP, p. 7) Christopher Robin asks for confirmation of his fictional status. He is overwhelmed by his emotions and is ready to enter the fictional world for good. The way this narrative POP works seems rather paradoxical: It attracts the reader-listener’s attention to the deictic shift while, in the end, ensuring a willing suspension of disbelief on the child-audience’s part.

216

Manuel Jobert

The final instance of interaction between the narrating voice and Christopher Robin occurs at the end of the story and finishes on the narrating voice promising to recount other adventures. Here again, this narrative POP is an invitation for further voyages into the fictional world. In this respect, the title of the first story of WP is explicit. The story indeed begins and the discourse world is somewhat left behind despite the different narrative POPs that pinpoint the slow sliding into the fictional world. After chapter 1, such deictic shifts become relatively rare except when something upsetting happens. In chapter 6, Christopher Robin is worried not to have given Eeyore a birthday present. The situation is a serious one and his speech is marked paralinguistically, making his concern explicit and justifying the narrative POP: ‘And didn’t I give him anything?’ asked Christopher Robin sadly. ‘Of course you did,’ I said. ‘You gave him—don’t you remember— a little—a little’ ‘I gave him a box of paints to paint things with.’ ‘That was it.’ (WP, p. 80) Quite cleverly, the narrating voice incites Christopher to invent the story himself, prompting him to imagine what present he gave to Eeyore, turning him into a willing accomplice in the creation of the fictional world. Finally, in order to facilitate the shift from Christopher Robin, the childaudience, to Christopher Robin, the enactor in the Hundred Acre Wood, the second-person personal pronoun is used by the narrating voice in all the introductory clauses related to Christopher of chapter 1: ‘“Did I miss?” you asked’ (WP, p. 16). In the other stories, the third-person pronoun is used consistently. Not only is the transition from the discourse world to the fictional world eased, but the cognitive and emotional implication of the listening-reading child is reinforced. 3.2 Psychos and Philosophers All the characters of WP and HPC have become household names and certain personality traits are readily associated with them: Winnie is greedy, especially when it comes to honey, Eeyore is a tad pessimistic, Piglet is panicky and Tigger is rather intense. A team of mental health researchers published a tongue-in-cheek analysis of WP and concluded that all the characters suffer from objective mental illnesses. For instance, Winnie is diagnosed with Attention Deficit Hyperactivity Disorder, Piglet suffers from Generalised Anxiety Disorder and Eeyore ‘would benefit greatly from an antidepressant’ (Shea et al., 2000, p. 163). In a different vein, Vanin (2014) links each character to a particular philosophical trend (empiricism, scepticism, rationalism, sophism, etc.). Yet, the

Winnie-the-Pooh Goes Paralinguistic . . . 217 psychological diagnoses do not take us very far in terms of interpretation and the philosophical approach, for all its worth, fails to be entirely convincing for lack of textual evidence. In both studies, though, the characters are apprehended as types. Culpeper (2001, p. 215) pleads for a textual approach to characterisation based on PVFs: There is a strong relationship between certain voices and certain personality types. The notion of vocal stereotypes, the idea that particular vocal characteristics are conventionally associated with particular personality traits, is a well-established finding.11 Although I am not concerned with a strict quantitative approach, certain tendencies seem to emerge when we observe how PVMs are allocated. I shall focus on ‘paralinguistically marked’12 reporting clauses introducing direct speech (DS). Concerning Christopher Robin, the most frequent adverb used to describe his speech is ‘carelessly’ (five occurrences out of ten paralinguistically marked reporting clauses13), which tends to go contrary to the idea that Christopher is caring and helpful with his cuddly friends. However, the interpretation of this adverb is a difficult one as it may mean ‘not caring’ or ‘free from anxiety’. More often than not, though, these comments are used in contexts where he wants to appear in control, adult-like and not subject to panic. Of course, Christopher Robin tends to appear when a problem has to be solved and he is given fewer turns than several other characters. Roo ‘squeaks’ a lot (six occurrences out of nine paralinguistically marked reporting clauses). Sometimes, the verb is accompanied by an adverb: ‘[s]queaked Roo excitedly’ (WP, p. 113) with double paralinguistic information (direct and indirect). This raises the question of the paralinguistic status of his ‘squeak’ that seems to be a characterisation device referring to his voice set (individual norm) rather than a local marker of attitude. A similar question can be asked regarding Eeyore for whom indirect PVMs appear to be very consistent and out of 18 paralinguistically marked reporting clauses, 14 are related to ‘sadness’: five involve the adverb ‘gloomily’, and five others include ‘sadly’ (p. 79), ‘in his melancholy way’ (p. 109), ‘grumbled’ (p. 110), ‘in a terrible voice’ (p. 107) and ‘almost breaking down’ (p. 69). The difference with Roo is that most of these PVMs are indirect and refer to personality traits (see ‘his melancholy way’) rather than the description of a particular voice set. The case of Kanga, Roo’s mother, is yet different as it is hard to determine whether PVMs are used to characterise her or if they are mainly attitudinal. Indeed, we find ‘in a thoughtful voice’ (p. 94), ‘soothingly’ (p. 94), ‘rather severely’ (p. 95), ‘anxiously’ (p. 111) and ‘reproachfully’ (p. 142). The emotions are varied and could therefore be context-dependent. Yet, they are all easily associated with ‘parenting’. These PVMs have a hybrid

218

Manuel Jobert

status and should be regarded as both characterisation devices and attitudinal features. Other characters exhibit a wider range of possibilities. Rabbit, though less present than other characters, says things ‘politely’ (p. 23), ‘sternly’ (p. 25), ‘cheerfully’ (p. 26), ‘carelessly’ (pp. 83, 104), ‘kindly’ (p. 83), ‘solemnly’ (p. 85) and ‘proudly’ (p. 86). Similarly, in the case of direct PVMs, we have ‘Rabbit said in a loud whisper’ (p. 89) and ‘Rabbit said in a loud voice’ (p. 91). In his case, it is obvious that the PVMs change with the different situations and are not characterisation devices. This is also the case for Owl who runs a great gamut of emotions with indirect PVMs such as ‘a little surprised’ (p. 47), ‘a little anxiously’ (p. 74), ‘carelessly’ (p. 75), ‘a little crossly’ (p. 108), ‘sulkily’ (p. 138) and ‘in a kind way’ (p. 139). Only ‘said Owl in his superior way’ (p. 108) may be taken as a characterisation device. Pooh and Piglet deserve a special mention as they are the two characters with the highest number of paralinguistically marked reporting clauses associated with them. Pooh is associated with no fewer than 19 different PVMs. The top three are related to surprise: ‘very much surprised’ (p. 22), ‘in surprise’ (p. 40), ‘in great surprise’ (p. 69). Solemnity is also present with three instances: ‘solemnly’ (pp. 42, 47), ‘in a solemn voice’ (p. 52). We also find three instances of excitement: ‘excitedly’ (pp. 54, 79, 103).14 What is striking is that PVMs are so varied that they do not enhance character recognition, something which could have been expected in a children’s story. Rather, the diversity seems to indicate that Pooh lives many adventures and goes through different states of mind: He remains, after all, the eponymous hero of the story. Piglet is less of a paralinguistic animal than Pooh even if nine types of different indirect PVMs are associated with him. The first one, with three instances, is ‘anxiously’ (pp. 63, 93, 103). Other PVMs include ‘with excitement’ (p. 32), ‘bravely’ (p. 60), ‘happily’ (p. 80), ‘impatiently’ (p. 83), ‘comfortingly’ (p. 85), ‘earnestly’ (p. 104) and ‘severely’ (p. 108). The PVM ‘said in as brave a voice as he could’ (p. 94) is interesting as it indicates a case of near “paralinguistic failure”. Indeed, this phrase implies that the different PVFs associated with bravery are present but somewhat not sufficiently firmly expressed to be entirely convincing. As for direct PVMs, loudness is well represented with five ‘cried’ (pp. 34, 63, 64, 95, 96), two ‘shouted’ (pp. 76, 94) and two ‘very loudly’ (pp. 77, 94). ‘Sniffle’ is used twice (pp. 70, 84). Other PVMs include ‘quickly’ (p. 36), ‘huskily’ (p. 87), ‘sputtered’ (p. 95) and ‘whispered’ (p. 142). In the introduction, Piglet is reported to say something ‘in his squeaky voice’, indicating his voice set, and ‘squeak’ is used again but with post-modification: ‘a squeak of excitement’ (p. 32). This is difficult to interpret in phonetic terms as it implies that an emotional ‘squeak’ is superimposed on a ‘squeaky voice’, making the entire utterance rather . . . squeaky and difficult to render. This clearly shows the limitations of realism in terms of paralinguistic encoding.

Winnie-the-Pooh Goes Paralinguistic . . . 219 This first approach to PVMs shows that although some phonetic features or specific emotions associated with speech sometimes play a part in characterisation, especially direct PVMs, most of them can be properly interpreted only in context. Indeed, references to Eeyore’s sadness may simply refer to his normal state of mind or to a contextually motivated feeling. In the case of Kanga, I argued for a hybrid type of feature inbetween characterisation devices and PVMs. This clearly suggests that an exclusively quantitative approach would simply lead to collecting uninterpretable data. 3.3 Pooh and Piglet as Facetious Vocal Exponents In chapter 3 of HPC, Pooh and Piglet fall into a pit and anticipate meeting a Heffalump. I shall focus on three passages to study PVMs in context. The first one relates Pooh’s fall into the pit. In the second one, Pooh tells Piglet what he will say to the Heffalump, and in the third part, Piglet imagines his own interaction with the frightening animal. In the pit episode, we see how PVMs work in context. After his fall, Pooh mistakenly believes he is speaking unwittingly. A clear progression is at work: (i) Onomatopoeia with PVM and a non-defined utterer: ‘“Ow!” squeaked something’ (p. 39). (ii) Articulated language with PVM not addressed: ‘“Help!” said a small, high voice’ (p. 39). (iii) Articulated language not addressed with no PVM: ‘Help—help!’ (p. 39). (iv) Articulated language with PVM and addressed: ‘“Pooh!” squeaked the voice’ (p. 39). (v) Articulated language addressed with PVM and a defined utterer: ‘“You,” squeaked Piglet, “Get up”’ (p. 39). There is a shift from the least identified to the more specific: The first squeak is not identified (i), the noise is then recognised as a voice (ii), the voice is determined through the use of the definite article and, finally, the speaker is identified (v). Although there is no objective PVM in (iii), the previous ones (squeaked and small, high voice) are still active and are carried over by the reader onto the next utterance in the absence of indication to the contrary (another PVM or a contextual change). Furthermore, Pooh recognises Piglet only when he is addressed directly (iv). When the recognition has occurred, the PVMs seem to chime and the two characters are attuned paralinguistically: ‘Pooh!’ squeaked the voice; ‘It’s Piglet!’ cried Pooh eagerly. ‘Where are you?’ (p. 39)

220

Manuel Jobert We have high-pitched voices that respond to each other. The same is true of the following exchange that concludes the pit episode: ‘You fell on me,’ said Piglet, feeling himself over. ‘I didn’t mean to,’ said Pooh sorrowfully. ‘I didn’t mean to be underneath,’ said Piglet sadly. (HPC, p. 39)

The mutual apologies are accepted and sealed by this vocal reciprocity. This type of paralinguistic harmony is frequently encountered in the Pooh stories as different characters are intrinsically nice to one another: ‘But this is Me!’ said Bear, very much surprised. ‘What sort of Me?’ ‘Pooh Bear.’ ‘Are you sure?’ said Rabbit, still more surprised. (WP, p. 22) Even when the situation is potentially tense, the vocal order is usually swiftly restored: ‘It all comes,’ said Pooh crossly, ‘of not having front doors big enough.’ ‘It all comes,’ said Rabbit sternly, ‘of eating too much . . .’ Christopher Robin . . . said, ‘Silly old Bear,’ in such a loving voice that everybody felt quite hopeful again. (WP, p. 25) Clearly, A.A. Milne plays with paralinguistic conventions, stretching the reader’s cooperation: ‘It’s Piglet!’ cried Pooh eagerly. ‘Where are you?’ ‘Underneath,’ said Piglet in an underneath sort of way. (HPC, p. 39) This shows that paralinguistically marked introductory clauses are a potentially fertile locus for stylistic inventiveness and humour (see Kohn and Yarbrough, 2010). The second part of chapter 3 revolves around the interpretation of ‘Ho-ho!’ expected to be uttered by a heffalump. It is the expressivity15 of the exclamation which is at stake here as its semantic content is rather meagre. As previously, the progression in the use of PVMs seems more significant than the particular PVFs referred to. [B]ut the more [Pooh] thought, the more he felt that there is no real answer to ‘Ho-ho!’said by a Heffalump in the sort of voice this Heffalump was going to say it in. (HPC, p. 43)

Winnie-the-Pooh Goes Paralinguistic . . . 221 The least that can be said is that the phonetic rendering of ‘Ho-ho!’ is not made clear by the comment. As Pooh imagines the unfolding of the story and ponders the conversational implications, things become clearer: ‘Because when you say “Ho-ho!” twice, in a gloating sort of way’ (HPC, p. 43) As the outline of the forthcoming conversation becomes clearer in Pooh’s mind, it also becomes more definite in paralinguistic terms. Pooh is confusedly aware that there is no real answer to ‘expressivity’ without clear ‘expression’ and he resorts to a clever trick he masters very well: humming, which is another way of conveying expressivity without expression. Pooh’s purpose is clearly stated: ‘That will upset him’ (HPC, p. 43) and ‘I shall go on humming. That will Unsettle him’ (HPC, p. 44). Piglet, convinced though he is, imagines another scenario in which he would do the talking with the Heffalump. The third passage clearly represents another narrative PUSH in the world of fiction as Piglet himself imagines a conversation in which he is the hero. This mise en abyme of the narrating process is packed with PVMs. They read like the stage-directions of a dramatic piece. They are therefore visually foregrounded and clearly separated from direct speech proper. A specific PVM is ascribed to each turn and the overall effect is to clarify Pooh’s conversational strategy. This should come as no surprise as Piglet himself feels he has ‘more brain than Pooh’ (p. 44). The whole affair is the dramatisation of Piglet’s bluff with the Heffalump. Quite strikingly, the various PVMs associated with Piglet seem rather out of character. He speaks ‘carelessly’ (which seems to be A. A. Milne’s trump PVM when he wants to indicate that a character is doing his best to look in control) and answers ‘politely’ (45). Even when he expresses surprise, it is fake. When the Heffalump questions his theory about the trap, Piglet soldiers on: HEFFALUMP: ‘Oh!’ (nervously): ‘I—I thought it was a trap I’d made to catch Piglets.’ PIGLET (surprised): ‘Oh, no!’ (HPC, pp. 44–45) Piglet voices his surprise at the heffalump’s misconception but his tone of voice does not betray any sign of anxiety. The PVMs associated with the heffalump provide an interesting outline of the animal’s deteriorating confidence. He shifts from ‘gloatingly’ to ‘surprised, not quite sure of himself’ to ‘greatly disappointed’, ‘nervously’, ‘apologetically’ to conclude meekly with ‘eager to please’ (pp. 44–45). This vignette clearly dramatises Piglet’s fantasised conversational triumph. Piglet’s fortitude is tested by Christopher Robin himself: ‘Ho-ho!’ said Christopher Robin loudly and suddenly. (HPC, p. 47)

222

Manuel Jobert

Although he is shocked, Piglet manages to play his part but is unsettled by the fact that Christopher does not say the expected lines and imitates Pooh’s voice. Interestingly, the end of this conversational game is decided by Christopher when he resorts to ‘his ordinary voice’ (p. 49) thus underscoring the link between vocal-play and imagination. All these dialogues, the voice changes and the play-acting clearly foreground the vocal dimension of the Pooh stories but are also bound to be challenging material for readers-aloud and listeners.

4. ‘Contradiction’ This study has established the importance of PVFs and PVMs in stylistic analysis. The notion of deviation from the individual norm appears as a determining factor. Analysing PVMs in WP and HPC has raised a series of issues. Teasing apart the vocal notations that determine the speaker’s norm from those that are truly paralinguistic requires minutia as the same phrases are sometimes used for both types. It has become clear that PVMs need to be interpreted in context and that purely quantitative approaches are inadequate or at least insufficient. It has also been seen that PVMs, like many other tools, are not autonomous and must be used alongside other stylistic tools. Finally, despite their oral/vocal dimension, PVMs are part and parcel of the ‘texture’ of a text—albeit of a spoken text—and the ‘paralinguistic dynamics’ present in a passage are more important than the actual mapping of a given graphic marker with a given phonetic feature. In order to be complete, some follow-up work on the realisation of PVMs by readers-aloud should be pursued. This study has focused on how fictitious (i.e. non-real) paralinguistic phenomena were represented by an author and calls for another study showing how these phenomena may be rendered vocally, or not, as the case may be. The existence of various audio books of both WP and HPC will also make it possible to compare various performances.

Notes 1. In a 2014 BBC poll, Winnie-the-Pooh was named ‘the best-loved children’s book of the last 150 years’. 2. See also Ong (1982) on the tension between orality and literacy. 3. On the complex issue of reading aloud to children, see Ben Soussan et al. (2006). On the notion of textual ‘authority’, see Toolan (1996). 4. Although some research has been done on PVFs from a stylistic perspective, standard textbooks such as Leech and Short (2007), Toolan (2001), Simpson (2014), and Jeffries and McIntyre (2010) do not mention them. They are also absent from NØrgaard, Montoro and Busse (2010). 5. Throughout this chapter, boldface indicates my emphasis. 6. For a more comprehensive description of PVMs, see Jobert (2014). 7. See Perbet (2014) for an in-depth analysis of the phrase ‘Once upon a time’ and its temporal implications.

Winnie-the-Pooh Goes Paralinguistic . . . 223 8. Manguel (1997) pp. 109–110) explains this experience very accurately: ‘I would settle down . . . and, propped up high against my pillows, listen to my nurse read the Grimm’s terrifying fair-tales. Sometimes her voice put me to sleep; sometimes, on the contrary, it made me feverish with excitement, and I urged her on in order to find out, more quickly than the author had intended, what happened in the story. But most of the time I simply enjoyed the luxurious sensation of being carried away by the words, and felt, in a very physical sense, that I was actually travelling somewhere wonderful and remote, to a place I hardly dared glimpse on the secret last page of the book’ (my emphasis). See also Ben Soussan, P. et al. (eds) (2006) for empirical evidence. 9. The forest is a traditional fairy-tale trope (see Little Red Riding Hood) symbolising both a source of sylvan terror and a promise of exciting adventures. 10. ‘The means by which we shift around the deictic field of a text are described by the terms PUSH and POP. A PUSH is defined by Galbraith as a movement “from a basic level to a less available deictic plane, such as episodic memory . . . fictional story world . . . or fantasy” . . . Conversely, the term POP refers to movement out of a particular deictic field. According to deictic shift theory, we PUSH into deictic fields and POP out of them’ (Jeffries & McIntyre, 2010, p. 160). Although the relevance of the term ‘POP’ has been questioned (McIntyre, 2006, pp. 110–111), it seems here to correspond to the reading experience. 11. The problem with Culpeper’s standpoint is that the distinction between the speaker’s norm and PVFs is somewhat blurred: ‘here I shall confine myself to paralinguistic features, and examine their stereotypical relationships with aspects of personality and character’ (Culpeper, 2001, p. 216). Still, it can be argued that a bundle of PVFs used permanently may become part of the speaker’s norm (see Lyons, 1972, in the first part of this study). 12. DS with no reporting clause or with a reporting clause including reporting verbs of the type he said, he answered will be considered as unmarked. DS introduced by a reporting clause of the type he said sadly or he shouted will be considered as marked. 13. Direct PVMs include ‘shouted’ (one occurrence), ‘laughed’ (one occurrence), ‘whispered’ (one occurrence) and ‘cried’ (two occurrences). 14. Some PVMs are used only twice and include ‘proudly’ (pp. 29, 128), ‘anxiously’ (pp. 12, 26), ‘humbly’ (pp. 45, 83) and ‘kindly’ (pp. 68, 70). Other adverbs and adverbial clauses include ‘carelessly’ (p. 24), ‘crossly’ (p. 24), ‘gloomily’ (p. 26), ‘mysteriously’ (p. 32), ‘dreamily’ (p. 45), ‘in a thoughtful way’ (p. 55), ‘earnestly’ (p. 74), ‘impressed by it’ (p. 75), ‘eagerly’ (p. 101), ‘happily’ (p. 102) and ‘helpfully’ (p. 115). Although there are fewer direct PVMs, these are quite varied as well, with ‘loudness’ being represented several times: ‘shouted’ (p. 12), ‘very loudly’ (p. 22), ‘in a loud voice’ (p. 45) and ‘with one loud cry’ (p. 123). Almost the opposite, ‘whispering’ is used three times: ‘deep whisper’ (p. 10), ‘loud whisper’ (p. 13) and ‘whispered’ (p. 108). Other PVMs include ‘quickly’ (p. 23), ‘sniffling slightly’ (p. 26), ‘murmured’ (p. 46), ‘a deep sigh’ (p. 56) and ‘a growly voice’ (p. 128). 15. See Jobert (2014) on Guillaume’s difference between ‘expression’ and ‘expressivity’.

References Abercrombie, D. (1968) ‘Paralanguage’, British Journal of Disorders of Communication, 3 (1), pp. 55–59. Ben Soussan, P. et al. (eds.) (2006) Lire à haute voix à des tout-petits. RamonvilleSaint-Agne: Editions érès.

224

Manuel Jobert

Brown, G. (1990) Listening to spoken English. London: Longman. Crystal, D. (1969) Prosodic systems and intonation in English. London: Longman. Crystal, D. (1975) The English tone of voice. London: Edward Arnold. Crystal, D. (1997) A dictionary of linguistics and phonetics. 4th ed. London: Blackwell Publishers. Crystal, D. and Davy, D. (1969) Investigating English style. London: Longman. Crystal, D. and Quirk, R. (1964) Systems of prosodic and paralinguistic features in English. London: Mouton. Culpeper, J. (2001) Language and characterisation. London: Pearson Education. Hall, E. T. (1984) Le langage silencieux. Paris: Editions du Seuil. (The Hidden Language, (1959). Trans. J. Mesrie and B. Niceall). Hill, A. (1958) Introduction to linguistic structure. New York: Harcourt Brace and World. Jeffries, L. & McIntyre, D. Stylistics. Cambridge: Cambridge University Press. Jobert, M. (2014) ‘Phonostylistics and the written text’, in Stockwell, P. and Whiteley, S. (eds.) The Cambridge handbook of stylistics. Cambridge: Cambridge University Press, pp. 231–248. Kohn, L. and Yarbrough, W. (2010) ‘Construction a pedagogy of comedy: Sarcasm and print codes as social literacies in Winnie-the-Pooh’, Journal of Language and Literacy Education, 6 (1), pp. 58–74. Leech, G. & Short, M. (2007) [1981] Style in fiction. Harlow: Longman Pearson. Lyons, J. (1972) ‘Human language’, in Hinde, R. (ed.) Nonverbal communication. Cambridge: Cambridge University Press, pp. 49–85. Manguel, A. (1997)] [1996] A history of reading. London: Flamingo. McIntyre, D. (2006) Point of view in plays. Amsterdam and Philadelphia: John Benjamins. Milne, A. A. (1926) Winnie-the-Pooh. London: Methuen and Co. Inc. Milne, A. A. (1928) The house at Pooh corner. London: Methuen and Co. Inc. Nørgaard, N., Montoro, R. & Busse, B (2010) Key terms in stylistics. London: Continuum International Publishing Group. Ong, W. (1982) Orality and literacy. London: Routledge. Perbet, H. (2014) ‘La temporalité dans le conte: Quelles attentes? Le cas de la locution adverbiale “Once upon a time”‘, Revue interdisciplinaire ‘Textes et Contextes’, 4, no pag. Available at: http://revuesshs.u-bourgogne.fr/textes&contextes/ document.php?id=2410 ISSN 1961–1991X (Accessed 6 November 2017). Piaget, A. (2002) ‘Interrelations entre oralité, écriture et culture’, Revue Tranel, 36, pp. 25–45. Shea, S. et al. (2000) ‘Pathology in the hundred acre wood: A neurodevelopmental perspective on A.A. Milne’, Canadian Medical Association Journal, 163 (12), pp. 1557–1559. Simpson, P. (2014) [2004] Stylistics. Oxon: Routledge. Thwaite, A. (1990) A.A. Milne: His life. Basingstoke and Oxford: Faber and Faber. Thwaite, A. (2017) Goodbye Christopher Robin. London: Pan Books. Toolan, M. (2001) [1988] Narrative: A critical linguistic introduction. Oxon: Routledge. Toolan, M. (1996) Total speech: An integrational linguistic approach to language. Durham, NC and London: Duke University Press. Trager, G. (1958) ‘Paralanguage: A first approximation’, Studies in Linguistics, 13, pp. 1–12. Vanin, L. (2014) Les aventures du moi: Les voix philosophiques de Winnie the Pooh. Nice: Les éditions Ovadia.

14 Quotation and Overhearing in Austen Joe Bray

Among the many ground-breaking chapters in Michael Toolan’s edited collection Language, Text and Context: Essays in Stylistics (1992) is Anne Waldron Neumann’s ‘Free Indirect Discourse in the EighteenthCentury English Novel: Speakable or Unspeakable? The Example of Sir Charles Grandison’. This essay transformed understanding of free indirect discourse (FID), demonstrating amongst other things that, contrary to much critical opinion which had placed its origin at the start of the nineteenth century in the fiction of Jane Austen, the technique is widespread in the eighteenth-century novel too, specifically those of Austen’s favourite novelist Samuel Richardson. Neumann also showed that eighteenth-century varieties of free indirect speech (FIS) in particular require a broader definition of FID than those which were common in the critical literature, one which allows for any unattributed quotation of isolated words and phrases. In this essay I expand on her insights, arguing that Austen too makes use of this quotative kind of FIS, especially in situations when a character is overhearing others. As the overhearer’s thoughts are themselves filtered through third-person narrative a further layering is possible. Examples from Emma (1816) in particular suggest that the complexity of speech and thought representation in Austen’s fiction can go far beyond the narrator-character interaction which has dominated traditional accounts of FID. In her chapter, Neumann makes a number of important and influential observations about earlier uses of FID. She observes for example that what she calls its ‘words-and-phrases’ variety, specifically in the form of FIS, is very common in Samuel Richardson’s The History of Sir Charles Grandison (1753–1754), as the novel’s letter-writers continually quote snippets of each other’s speech. For her the ‘variety of forms of quotation and degrees of attribution’ in the novel is indicative of the way Richardson expected his work to be read, remembered and discussed (1992, pp. 133–134). The prevalence of this form of FID leads her to a broader conception of the style than that favoured by those who have tended to focus on what she calls ‘whole-sentence’ free indirect thought (FIT). She suggests that FID be defined as ‘any sentence containing words not

226

Joe Bray

explicitly attributed as quotation (or at least not as quotation from a specified source) but likely to originate with a character rather than with the narrator, or with some character other than the quoting character’ (p. 114). Although she admits that there are degrees of definiteness with which quoted words can be thus identified, she argues that ‘if a sentence without attribution can plausibly be read as indirectly quoted discourse, then according to my definition it is FID: that is, I call it FID’ (p. 115). Crucial to Neumann’s definition then is the observation that FID may involve a combination not just of a character’s and a narrator’s voices, but also those of a character and another character. She thus endorses the view of the technique as involving ‘double-voicing’, while noting that ‘Grandison demonstrates incontrovertibly that the double-voice theory of reported discourse does not depend on a narrator for its second voice’ (p. 118). This is the case for her not just in epistolary novels such as Richardson’s, which obviously lack a consistent third-person narrative presence. For Neumann this sheds light on where the technique may have come from. She claims that the fact ‘that the characters of eighteenth-century fiction use FID to quote each other’s speech as often as narrators do in itself strongly suggests that FID is possible in everyday speech and may have originated there’ (p. 116). Furthermore, FID in the eighteenth-century novel reflects for her wider social and cultural practice: ‘characters in eighteenth-century fiction who quote and discuss each other’s discourse become models for—if not mirrors of—how contemporary readers could—or did—quote and discuss their own conversations and reading’ (p. 134). Critics of FID in Jane Austen’s fiction have tended to focus on, in Neumann’s terms, the ‘whole-sentence’ variety. Austen has often been identified as the first novelist in English to, in David Lodge’s words, ‘give the reader intimate access to a character’s thoughts without totally surrendering control of the discourse to that character’ (1990, p. 126). This emphasis on the innovative ways in which character psychology is developed in her novels has led to a concentration on FIT, to the relative neglect of FIS (though see Shimazaki, 2015). In fact, the representation of spoken words within her third-person narratives can be as subtle and sophisticated as that of unspoken thoughts. As the rest of this chapter will demonstrate, ‘words-and-phrases’ FIS of the kind that Neumann identifies in the work of Richardson, who was singled out by Austen’s brother Henry in his ‘Biographical Notice of the Author’ after her death as Jane’s favourite eighteenth-century novelist (2006, p. 330), finds its way into her writing too, where it is capable of generating a wide and complicated range of effects. Furthermore, as Neumann observed of eighteenth-century fiction in general, it is not just the relationship between characters and narrators which can be represented in this style. Attention to the speech of one of Austen’s most distinctive and idiosyncratic characters demonstrates how her skilful handling of FIS can indicate complex relationships and tensions between characters.

Quotation and Overhearing in Austen

227

Before turning to the novel and the character in question, a recap of recent work on FID, and FIS in particular, is appropriate. Leech and Short’s (1981) model of speech and thought presentation distinguished five categories of speech (with their equivalents in thought): DIRECT SPEECH (DS): He said, ‘I’ll come back here to see you tomorrow.’ INDIRECT SPEECH (IS): He said that he would return there to see her

the following day. FREE DIRECT SPEECH (FDS):

I’ll come back here to see you again

tomorrow. NARRATIVE REPORT OF SPEECH ACTS (NRSA): He promised to return. FREE INDIRECT SPEECH (FIS): He would come back there to see her

again tomorrow. (1981, pp. 318–325)1 This model has proved hugely influential both within stylistics and beyond, and has been used widely by others (see for example Simpson, 1993). It has also inspired various projects which have applied the model to a corpus of written fictional and non-fictional texts, as well as a corpus of spoken English. As Leech and Short report in their second edition of Style in Fiction, the work on the written corpus ‘has shown that, by and large, the original [. . .] model works well when applied to non-fictional narratives as well as to the fictional narratives it was originally developed for’ (2007, p. 302). The corpus studies are summarised fully in Semino and Short (2004), along with the revisions to the model for which they have established the need (see also Semino et al., 1997; Short et al., 2002; and Short, 2003). Perhaps the most radical revision is the addition of a completely new scale, writing representation, to parallel those of speech and thought. In relation to the speech representation scale, the focus of this chapter, the most significant change is the inclusion of a new category, Narrator’s Representation of Voice (NV) to capture instances in which, in Semino and Short’s words, ‘we are informed that someone engaged in verbal activity, but we are not given any explicit indication as to what speech acts were performed, let alone what the form and content of the utterances were’ (2004, p. 44). This form of speech representation is thus ‘even more minimal that that captured by Leech and Short’s NRSA category’ (p. 44), and it consequently appears to the left of NRSA in the following revised diagrammatic representation of the speech scale given by Semino and Short (2004, p. 49):2 [N]

NV

NRSA

IS

FIS

DS

(FDS)

A further revision involved an addition to the main category of NRSA. As Semino and Short note, ‘Leech and Short’s NRSA category captures those cases where the narrator presents the illocutionary force of a particular utterance, with little or no indication of its content’ (p. 52). They

228

Joe Bray

observe however that their data included frequent examples of ‘long and extremely detailed NRSAs’ (p. 52), in which ‘a considerable amount of information is provided regarding the propositional content of the utterance’ (p. 53). Rejecting the classification of such instances as a condensed form of IS, due to the absence of a distinction between a reporting and a reported clause, Semino and Short propose a new category NRSAp, where ‘p’ stands for ‘topic’, which ‘would capture all those cases where the report of the speech act is accompanied by an explicit indication of the subject matter/topic of the utterance or utterances in question, but where there is no separate reported clause’ (p. 53). Throughout Austen’s fiction the subtle interplay between these categories of speech representation is a vital means of characterisation. Consider for example the contrasts between the two sisters’ speech in Sense and Sensibility when Marianne expresses in raptures her excitement at Willoughby having given her a present of a horse: ‘He intends to send his groom into Somersetshire immediately for it,’ she added, ‘and when it arrives, we will ride every day. You shall share its use with me. Imagine to yourself, my dear Elinor, the delight of a gallop on some of these downs.’ Most unwilling was she to awaken from such a dream of felicity, to comprehend all the unhappy truths which attended the affair; and for some time she refused to submit to them. As to an additional servant, the expense would be a trifle, mama she was sure would never object to it; and any horse would do for him; he might always get one at the park; as to a stable, the merest shed would be sufficient. Elinor then ventured to doubt the propriety of her receiving such a present from a man so little, or at least so lately known to her. This was too much. ‘You are mistaken, Elinor,’ said she warmly, ‘in supposing I know very little of Willoughby.’ (2006, pp. 68–69) The passage opens with Marianne’s excited DS before switching in the second sentence of the second paragraph into a free indirect representation of her speech. From ‘As to an additional servant’ onwards this appears to be FIS, with the personal form of address ‘mama’, the dismissive ‘merest shed’ and the italics of ‘him’ clearly indicating her actual words and her tone of voice. Her dominance in the exchange is apparent from the way that Elinor’s tentative reservations are expressed in the less vocal NRSAp: ‘Elinor then ventured to doubt the propriety of her receiving such a present from a man so little, or at least so lately known to her’. The representation of part of Marianne’s speech here in FIS, rather than DS, suggests the presence of an alternative perspective, that of the narrator, as represented by the third person. As a result of this mixture of

Quotation and Overhearing in Austen

229

two points of view an effect of distancing, or of irony, is created, which critics have often identified as a typical characteristic of FIS in particular. Leech and Short claim that because the norm for speech presentation is DS ‘the irony arises because FIS is normally viewed as a form where the authorial voice is interposed between the reader and what the character says, so that the reader is distanced from the character’s words’ (2007, p. 268). Having given an example of FIS to represent Mr Shepherd’s speech in Persuasion they comment that ‘this ability to give the flavour of the character’s words but also to keep the narrator in an intervening position between character and reader makes FIS an extremely useful vehicle for casting an ironic light on what the character says’ (p. 262).3 Certainly, there are many instances of FIS being used to create this kind of irony throughout Austen’s fiction. In Northanger Abbey Isabella reassures Catherine when Henry Tilney is suddenly and inexplicably absent from Bath: Isabella was very sure that he must be a charming young man; and was equally sure that he must have been delighted with her dear Catherine, and would therefore shortly return. She liked him the better for being a clergyman, ‘for she must confess herself very partial to the profession;’ and something like a sigh escaped her as she said it. (2006, p. 29) ‘Charming’ and ‘dear’ are suggestive of Isabella’s exaggeratedly sentimental style, and the use of quotation marks together with the deontic modality clearly indicates a representation of Isabella’s actual words, though with the third person retained (compare ‘I must confess myself very partial to the profession’). The distance created by FIS here thus helps to present Isabella as insincere, addicted to the clichés of her sentimental reading. A similar use of FIS within quotation marks occurs when she attempts to denigrate the Tilneys when Catherine blames herself after her first, disappointing visit to Milsom-street: Isabella, on hearing the particulars of the visit, gave a different explanation: ‘It was all pride, pride, insufferable haughtiness and pride! She had long suspected the family to be very high, and this made it certain. Such insolence of behaviour as Miss Tilney’s she had never heard of in her life! Not to do the honours of her house with common good-breeding!—To behave to her guest with such superciliousness!—Hardly even to speak to her!’ (2006, pp. 131–132) Again, FIS captures the insincerity of Isabella’s exaggerations, repetitions and exclamations, with the quotation marks adding to the directness of the representation of her voice. Catherine can only reply tentatively

230

Joe Bray

‘But it was not so bad as that, Isabella; there was no superciliousness; she was very civil’ (p. 132). The character from the later novels whose speech most resembles Isabella’s clichéd, hyperbolic style is Mrs Elton in Emma. FIS, again often within quotation marks, is similarly key to capturing her shallowness and insincerity. Emma’s first, negative impression of Mrs Elton is soon confirmed on her first visit to Hartfield, when she ‘had a quarter of an hour of the lady’s conversation to herself’: The very first subject after being seated was Maple Grove, ‘My brother Mr. Suckling’s seat’—a comparison of Hartfield to Maple Grove. The grounds of Hartfield were small, but neat and pretty; and the house was modern and well-built. Mrs. Elton seemed most favourably impressed by the size of the room, the entrance, and all that she could see or imagine. ‘Very like Maple Grove indeed!—She was quite struck by the likeness!—That room was the very shape and size of the morning-room at Maple Grove; her sister’s favourite room.’—Mr. Elton was appealed to.—‘Was not it astonishingly like?—She could really almost fancy herself at Maple Grove.’ (2005, p. 294) Mrs Elton’s speech here is represented in a complex mixture of ways. The report that the first subject of the conversation is Maple Grove, or rather ‘a comparison of Hartfield to Maple Grove’, is NRSA, with only a broad outline of the illocutionary force given. Enclosed within this is a fragment of DS, ‘My brother Mr. Suckling’s seat’, which gives a first direct representation of Mrs Elton’s voice. This is followed by what seems to be narrative report, but could in fact be a free indirect representation of Mrs Elton’s words (transposed to the past tense), since it is unlikely that the narrator would give a bland introductory description of Hartfield at this point. What follows within both sets of quotation marks appears to be FIS, with the person and tense changed from Mrs Elton’s actual utterance (compare ‘I am quite struck by the likeness!’ and ‘I can really almost fancy myself at Maple Grove’). The dashes, exclamations and intensifying adverbs (‘very’ twice, ‘quite’ and ‘really’) are reminiscent of the earlier examples of Isabella Thorpe’s speech in Northanger Abbey. In each case while the quotation marks allow some immediate access to Mrs Elton’s and Isabella’s voices respectively, the retention of the thirdperson perspective in FIS casts doubt on the sincerity of their words. In the example from Emma the use of FIS suggests a forced fakeness behind Mrs Elton’s initial attempts to ingratiate herself within Highbury society, and with the heroine in particular. As their acquaintance continues, Emma is especially annoyed by Mrs Elton’s partiality for Jane Fairfax, with whom she often considers herself in competition. When the ladies retreat to the drawing-room after a

Quotation and Overhearing in Austen

231

dinner at Hartfield, Emma is seething as ‘with so much perseverance in judging and behaving ill did Mrs. Elton engross Jane Fairfax and slight herself’ (p. 323). The result is that ‘she and Mrs. Weston were obliged to be almost always either talking together or silent together’ (p. 323). This is not enough, however, to prevent her catching the drift of what the other two are talking about: though much that passed between them was in a half-whisper, especially on Mrs. Elton’s side, there was no avoiding a knowledge of their principal subjects:—The post-office—catching cold—fetching letters—and friendship, were long under discussion; and to them succeeded one, which must be at least equally unpleasant to Jane— inquiries whether she had yet heard of any situation likely to suit her, and professions of Mrs. Elton’s meditated activity. (2005, p. 323) The fact that even though Jane and Mrs Elton are speaking in a halfwhisper Emma is able to glean the main topics of their conversation might suggest that her overhearing is not entirely involuntary, as might the ambiguous ‘there was no avoiding a knowledge of their principal subjects’, which may be Emma’s justification rather than narratorial report. The dashes similarly suggest a certain straining on Emma’s part, and it seems to be her somewhat too self-satisfied opinion that discussion of Mrs Elton’s efforts to arrange a situation for her friend ‘must be at least equally unpleasant to Jane’. In her study of eavesdropping in the novel from Austen to Proust, Ann Gaylin claims that ‘Emma is the only Austen novel that does not contain some form of eavesdropping as an integral part of its narrative structure or thematic concerns’ (2002, p. 188). Yet throughout the novel Emma frequently, perhaps suspiciously frequently, finds herself in the position of overhearer, as in the previous example. As the party which has been discussing arrangements for the ball at the Crown breaks up, Emma, whom Frank has ‘positively secured for the first two dances’, overhears Mr Weston whisper to his wife, ‘“He has asked her, my dear. That’s right. I knew he would!”’ (p. 276). At the ball itself, once Miss Bates’s monologue on entering the room is over, Emma ‘found herself necessarily overhearing the discourse of Mrs. Elton and Miss Fairfax, who were standing a little way behind her’ (p. 350). Again, there is a suggestion in ‘found herself necessarily overhearing’ that Emma, even if she has not quite manoeuvred herself into a position to overhear the conversation between the two, is not exactly in a hurry to remove herself into a position from which she cannot. On this occasion she is not the only one overhearing. As the exchange between Mrs Elton and Jane segues from the former giving and soliciting compliments to her praise of Frank and his dancing, Frank himself, who is alongside Emma, ‘began talking so vigorously, that

232

Joe Bray

Emma could not but imagine he had overheard his own praises, and did not want to hear more’ (p. 351). Emma is correct of course that Frank is uncomfortable at overhearing this conversation, though wrong as to the reasons why. His desire not to hear what Jane in particular will say about him contrasts with Emma’s willingness to overhear others throughout the novel, especially Mrs Elton and Jane, despite her claims to the contrary. Emma’s overhearing is especially acute during the trip to Donwell Abbey in volume III. After the group finds seats ‘tolerably in the shade’ Emma once again finds herself ‘obliged to hear what Mrs. Elton and Jane Fairfax were talking of’: —A situation, a most desirable situation, was in question. Mrs. Elton had received notice of it that morning, and was in raptures. It was not with Mrs. Suckling, it was not with Mrs. Bragge, but in felicity and splendour it fell short only of them: it was with a cousin of Mrs. Bragge, an acquaintance of Mrs. Sucking, a lady known at Maple Grove. Delightful, charming, superior, first circles, spheres, lines, ranks, every thing—and Mrs. Elton was wild to have the offer closed with immediately.—On her side, all was warmth, energy, and triumph—and she positively refused to take her friend’s negative, though Miss Fairfax continued to assure her that she would not at present engage in any thing, repeating the same motives which she had been heard to urge before.—Still Mrs. Elton insisted on being authorized to write an acquiescence by the morrow’s post.—How Jane could bear it all, was astonishing to Emma.—She did look vexed, she did speak pointedly—and at last, with a decision of action unusual to her, proposed a removal.—‘Should not they walk?—Would not Mr. Knightley show them the gardens—all the gardens?—She wished to see the whole extent.’—The pertinacy of her friend seemed more than she could bear. (2005, p. 390) Again, the conversation between Mrs Elton and Jane is represented in a variety of ways. At first the style appears to be a mixture of IS and NRSAp, with discussion again of a potential situation for Jane, and the fact that ‘Mrs. Elton had received notice of it that morning, and was in raptures’. The repetition in ‘A situation, a most desirable situation’ is a hint though of Mrs Elton’s actual words, and the sentence beginning ‘It was not with Mrs. Suckling, it was not with Mrs. Bragge, but in felicity and splendour it fell short only of them’ appears to be a free indirect version of what she says next on where the position is, representing her words closely, yet still retaining the past tense of narration. The following sentence is more direct still, picking out particularly fashionable words characteristic of Mrs Elton’s speech before summarising, again with a hint of her diction: ‘Mrs. Elton was wild to have the offer closed with

Quotation and Overhearing in Austen

233

immediately’. Her conversational partner’s contributions to the exchange are represented relatively indirectly, with the report of her ‘negative’ followed by a rather formal, stilted mixture of IS and NRSAp: ‘Miss Fairfax continued to assure her that she would not at present engage in any thing, repeating the same motives which she had been heard to urge before’. Throughout the conversation there is also of course a third presence: the overhearing Emma. The reports of speech are filtered through her perspective; it is she who judges that Mrs Elton ‘was in raptures’, and that ‘on her side, all was warmth, energy, and triumph’. It is she, not the narrator, who picks out the keywords from Mrs Elton’s speech: ‘Delightful, charming, superior, first circles, spheres, lines, ranks, every thing’, with the choice of affected and vacuous words and the casually dismissive ‘every thing’ hinting at her attitude towards the speaker. There is an explicit reminder of her presence in her reflection ‘How Jane could bear it all, was astonishing to Emma’. It is Emma to whom Jane ‘did look vexed’ and her judgement that she ‘did speak pointedly—and at last, with a decision of action unusual to her, proposed a removal’. What follow are Jane’s words in FIS within quotation marks, again filtered through Emma’s overhearing perspective. In contrast to her antagonism towards Mrs Elton, Emma appears more sympathetic towards Jane here. She of course does not know just how much Mrs Elton’s insincere concern and her over-enthusiastic attempts to find her a situation are hurting Jane, given the current precarious nature of her secret attachment to Frank. The FIS within quotation marks suggests her finally, somewhat plaintively, making her voice heard amidst Mrs Elton’s monologue, and thus earning some respite for both herself and the overhearing Emma. The bond between the two in Emma’s mind is apparent in the paragraph’s final sentence, since both of them have been suffering from Mrs Elton’s ‘pertinacity’. This passage thus represents more than simply the narrator’s quotation of the character’s words within the third person and past tense. It is rather the overhearing Emma who is straining to pick out what she can from the conversation, with her annoyance at Mrs Elton’s tiresome selfaggrandisement becoming ever more obvious. The example suggests that Neumann’s observation that ‘characters of eighteenth-century fiction use FID to quote each other’s speech as often as narrators do’ (1992, p. 116) is applicable to the early nineteenth-century novel too, as Austen adapts the widespread practice of unattributed quotation found in her favourite eighteenth-century novels. Emma’s quotations from Mrs Elton’s speech demonstrate in Neumann’s terms how ‘the subjective and evaluative expressions of one character can interweave with those of another’ (1992, p. 116). The fact that Emma’s overhearing, unspoken quotations are then themselves contained within the third-person narrative in FIT suggests a complex layering of perspectives, which advances Neumann’s theory of double-voicing further. The mixture of the three characters’

234

Joe Bray

and the narrator’s voices in this passage reveals the potential of FID, and speech and thought representation in general, for indicating the attitudes of several characters towards each other. Emma’s ‘necessary’ overhearing of many of Mrs Elton’s and Jane’s conversations is a reminder then that even when not speaking, she is present at all of the novel’s conversations, and that many of the judgements on characters’ speech that these include are hers, rather than the narrator’s. Her class-based, morally loaded opinion of Mrs Elton in particular, entrenched even before she enters Highbury society, pervades representations of the latter’s speech. Earlier during the trip to Donwell Abbey, as the group collects around the strawberry beds, Mrs Elton ‘in all her apparatus of happiness, her large bonnet and her basket, was very ready to lead the way in gathering, accepting, or talking’: —strawberries, and only strawberries, could now be thought or spoken of.—‘The best fruit in England—every body’s favourite—always wholesome.—These the finest beds and finest sorts.—Delightful to gather for one’s self—the only way of really enjoying them.—Morning decidedly the best time—never tired—every sort good—hautboy infinitely superior—no comparison—the others hardly eatable— hautboys very scarce—Chili preferred—white wood finest flavour of all—price of strawberries in London—abundance about Bristol— Maple Grove—cultivation—beds when to be renewed—gardeners thinking exactly different—no general rule—gardeners never to be put out of their way—delicious fruit—only too rich to be eaten much of—inferior to cherries—currants more refreshing—only objection to gathering strawberries the stooping—glaring sun—tired to death—could bear it no longer—must go and sit in the shade.’ (2005, pp. 389–390) It is not just the selection of topics here, many of which are typical of Mrs Elton’s speech (especially of course the inevitable leading back to ‘Maple Grove’), nor the characteristic exaggerated and affected language (‘Delightful to gather for one’s self’, ‘hautboy infinitely superior’, ‘delicious fruit’), which indicate Emma’s impatient annoyance with Mrs Elton, but also the way her speech is represented. The quotations from her speech, hopping somewhat haphazardly from one opinion on strawberries to another, suggest both garrulousness and a towering self-absorption, as though Mrs Elton is only interested in proving the superiority of her own judgements and experience. This at least is Emma’s view here; the dashes between Mrs Elton’s words give the reader space in which to infer her strongly negative reactions as the latter rambles on. There is a hint too of Emma’s perspective in the past tense of ‘could bear it no longer’, a suggestion that much of what comes before may be FIS rather than DS: Mrs Elton’s words filtered through Emma’s perspective. In other

Quotation and Overhearing in Austen

235

words, again it is Emma, rather than the narrator, who is quoting Mrs Elton here. Within the dashes and white spaces between the latter’s words the reader can imagine the overhearer’s irritation and exasperation. Anne Waldron Neumann’s insightful essay in Language, Text and Context opened the door to a broader conception of FID, one that recognises its flexibility in the representation of speech as well as thought. Her penetrating analysis of The History of Sir Charles Grandison can be extended to Jane Austen’s practice, and the ways in which her characters, like those of her favourite eighteenth-century novelist, quote and evaluate each other. The heroine of Emma is particularly attentive to the speech of others, especially on the frequent occasions she is obliged (she claims) to overhear them. Her strongly negative attitude towards Mrs Elton, and her annoyance at her friendship with Jane Fairfax, can be inferred from the subtle use of FID in these scenes. The ways in which Emma Woodhouse’s overhearing is represented hint too at the overhearer’s own selfobsession and sense of her own importance. Through a complex mixture of different forms of speech and thought representation Austen suggests that the gap between the heroine and her perceived rival in Highbury society may not be as great as either wishes.

Notes 1. As Leech and Short note, other versions of all these categories are possible, including FIS (a variety with proximally deictic ‘here’ in place of ‘there’ would also be acceptable, for example). 2. ‘N’, which stands for Narration, is, in Semino and Short’s words, ‘enclosed within square brackets because it is not a category on the presentational scales’, while ‘FDS’ is in round brackets as a result of their belief that ‘the free direct categories are best seen as a sub-type, or variant, of the respective direct categories’ (2004, pp. 49). 3. According to Leech and Short FIT has a contrastive effect, representing a move ‘into the active mind of the character’ (1981, pp. 345).

References Austen, H. (2006) [1818] ‘Biographical notice of the author’, in Todd, J. and Blank, A. (eds.) Persuasion. Cambridge: Cambridge University Press, pp. 326–332. Austen, J. (2005) [1816] Emma. Edited by R. Cronin and D. McMillan. Cambridge: Cambridge University Press. Austen, J. (2006) [1811] Sense and sensibility. Edited by E. Copeland. Cambridge: Cambridge University Press. Austen, J. (2006) [1818] Northanger Abbey. Edited by B. M. Benedict and D. Le Faye. Cambridge: Cambridge University Press. Gaylin, A. (2002) Eavesdropping in the novel from Austen to Proust. Cambridge: Cambridge University Press. Leech, G. N. and Short, M. H. (1981) Style in fiction: A linguistic introduction to English fictional prose. Harlow, Essex: Longman.

236

Joe Bray

Leech, G. and Short, M. (2007) Style in fiction: A linguistic introduction to English fictional prose. 2nd ed. Harlow, Essex: Pearson Education. Lodge, D. (1990) After Bakhtin: Essays on fiction and criticism. London and New York: Routledge. Neumann, A. W. (1992) ‘Free indirect discourse in the eighteenth-century English novel: Speakable or unspeakable? The example of Sir Charles Grandison’, in Toolan, M (ed.) Language, text and context: Essays in stylistics. London and New York: Routledge, pp. 113–135. Semino, E. and Short, M. (2004) Corpus stylistics: Speech, writing and thought presentation in a corpus of English writing. London and New York: Routledge. Semino, E., Short, M. and Culpeper, J. (1997) ‘Using a computer corpus to test a model of speech and thought presentation’, Poetics, 25, pp. 17–43. Shimazaki, H. (2015) ‘Free indirect speech in the work of Jane Austen: The previously unappreciated extent and complexity of Austen’s free indirect speech and its development from the eighteenth-century fiction’, Unpublished PhD Thesis. University of Southampton. Short, M. (2003) ‘A corpus-based approach to speech, thought and writing presentation’, in Wilson, A., Rayson, P. and McEnery, T. (eds.) Corpus linguistics by the lune: A festschrift for Geoffrey Leech. Frankfurt am Main: Peter Lang, pp. 241–271. Short, M., Semino, E. and Wynne, M. (2002) ‘Revisiting the notion of faithfulness in discourse presentation using a corpus approach’, Language and Literature, 11(04), pp. 325–355. Simpson, P. (1993) Language, ideology and point of view. London and New York: Routledge.

15 Suppression, Silencing and Failure to Project Ways of Losing Voice While Using It Chris Heffer

1. Introduction On returning to Michael Toolan’s influential 25-year-old edited collection Language, Text and Context (Toolan, 1992), what I find immediately striking is the absence of any agency in the title. The title was progressive at the time: It was saying we have gone beyond language as a purely abstract cognitive system (Chomsky, 1968), beyond looking at text simply as a coherent and cohesive sequence of words (Halliday and Hasan, 1976), and on to the relation of that text with the world around it—its context. Yet my immediate uptake of the title now is that ‘text’ comes first and it exists in a broad ‘context’ but the actual agents of language production and reception, the speaker/writer and hearer/reader without whom there would be no text, are merely subsumed by the ‘context’. I recognise this is a misreading: The edited collection itself addresses questions of reader uptake in some of its chapters. But I am aware that my response to the title today is different from the response I would have had in the late 1990s and, in a sense, the likely authorially intended perspective no longer projects successfully to me. This altered reader-response is a small hint of the ‘hazardous’ nature of communication that underlies Toolan’s ‘integrational’ perspective on language. For Toolan (1996), as for Harris (1981) before him, ‘there really is no such thing as the context, even to the extent that we can think of certain things as texts: there is only a recurrent activity of contextualizing’ (Toolan, 1996, p. 4). Another way of putting this today is that discourse is an indeterminate, socially embedded practice engaged in by speakers and hearers with various and often divergent goals. Or, as Austin (1962, p. 147) put it, and Toolan (1996) entitled it, we are interested in the ‘total speech act in the total speech situation’. One of the primary goals of communication, whatever else it involves, is to get one’s perspective across to the hearer/reader (hereafter ‘hearer’). That perspective might be in the form of an idea or belief but also an identity or style that we wish to project. When that perspective ‘carries’ to the hearer, when the hearer grasps more or less what you are saying/

238

Chris Heffer

expressing in more or less the way you intended (without necessarily agreeing with it), then we can say that the speaker ‘has voice’. But when the hazards of communication or subversive intent prevent one’s perspective from carrying to the hearer, then one can say that the speaker ‘loses voice’. The notion of voice as linked to empowerment was introduced by Hymes (1996), who made a distinction between freedom of speech and ‘freedom to have one’s voice heard’ and considered the latter as much a fundamental linguistic freedom as the former (Hymes, 1996, p. 64). Blommaert has interpreted Hymes’s understanding of voice as ‘the capacity to make oneself understood in one’s own terms, to produce meanings under conditions of empowerment’ (Blommaert, 2008b, p. 17). In this chapter I consider three ways of losing this metaphorical voice in the process of using one’s physical voice. This work draws on the Voice Projection Framework (Heffer, 2013, 2018), which attempts to account for the multifarious factors affecting whether someone’s voice will ‘carry’ in a given communicative context, but the three ways of losing voice are proposed for the first time here. After a discussion of uptake and voice, I apply the Voice Projection Framework to three different forensic contexts: cross-examination, native land right hearings and the judge’s legal instruction of juries. First, I re-analyse a short section of the crossexamination of Ann Mercer in the William Kennedy Smith rape trial (Matoesian, 2001; Awcsgo 2011) to demonstrate the deliberate suppression of another’s voice in the institutional context of cross-examination. I then note, through examples from the submissions of First Nations speakers in a land dispute hearing in Canada (CEAA, 2015), how voice can be silenced, but not necessarily deliberately, through the process of entextualisation (the transformation of total speech into divisible text) in the transcription of culturally ‘other’ speakers. Finally, I consider, through an analysis of Judge Sweeney’s communication to the jury in the English trial of the economist Vicky Pryce (R v. Huhne and Pryce, 2013), how a voice can simply fail to project even when conveyed by a powerful speaker such as a judge. The overall argument is that an understanding of voice, both structurally and agentively, is central to an understanding of communicational uptake.

2. Intention, Uptake and Voice Until the twentieth century, when agency was considered with respect to text, it was primarily in the form of speaker intentionality: What did the speaker mean by saying x? Semiotics sought to suppress that agency, in favour of empirically observable structure, by locating meaning in autonomous ‘signs’, or signal-signification pairings that required only community consensus but not individual intention (Saussure, 1983[1916]) and so could be ‘read off’ independently of authorial intention. But as Toolan (1996, p. 128) points out, ‘the positive condition of the emergence of a

Suppression, Silencing, Failure to Project

239

(signifying) mark is that an intelligent actor takes notice of it, as a purposeful part and exponent of a purposeful activity—that is, that intentionality is attributed to its source’. Grice brought intentionality back into the picture in the late twentieth century with his notion of ‘implicature’, a technical term to distinguish what is pragmatically intended by the speaker from what is semantically said. But implicatures still need to be inferred by the hearer and ‘[u]ptake is far more probabilistic than Grice’s account would suggest’ (Toolan, 1996, p. 204). Toolan (1996) is concerned primarily with the uptake of implicature in figurative language such as metaphor and irony. However, the notion that uptake, rather than being a rational calculus (Sperber and Wilson, 1986), is somewhat hazardous can be extended to communication more generally. We can see the hazardous nature of uptake quite clearly in courtroom settings. Much work on linguistic disadvantage in court has focused on the asymmetrical speaking rights of legal and lay participants (e.g. Cotterill, 2003). Witnesses, for example, are believed to suffer ‘narrative inequality’ (Hymes, 1996) because they do not have the same rights to speak and to control the interaction as the lawyer examining them. However, when witnesses do have free rein to narrate as they wish, as in small claims courts (Conley and O’Barr, 1990), they often fail to get across to the judge, who misconstrues their central affective message about wrongdoing as legally ‘immaterial’ verbiage that falls short of formally assigning blame. In other words, these lay litigants, left to their own discursive devices, fail to achieve what Bakhtin called ‘responsive understanding’ (Bakhtin, 1981), a ‘take up’ of discourse that involves grasping what is said to the point of being able to respond affirmatively or critically to that meaning. It is important to appreciate that responsive understanding is not agreement. We can responsively understand what someone says (‘I hear what you are saying’) but completely disagree with it (‘you are quite wrong’). This occurs very frequently in academic discussion, intelligent debate and even amongst politicians in parliament. However, the ability to achieve responsive understanding in hearers is an unequally distributed resource. It is this agentive capacity effectively to achieve responsive understanding that Hymes called ‘voice’. Blommaert (2008a, p. 427) conveys Hymes’s notion more clearly and critically in an article on Bernstein: People use language and other semiotic means in attempts to have voice, to make themselves understood by others. This process is complex and only partly predictable, because whatever is produced is not necessarily perceived or understood, and having voice is therefore an intrinsically social process—that is a process with clear connections to social structure, history, culture, power. That the process of having voice is only partly predictable ties in with Harris’s and Toolan’s integrationalist notion of the hazardous nature of

240

Chris Heffer

communication. However, critical work on voice has tended to fix on the power dimension, on voice as an unequally distributed resource that a speaker has more or less of in a given context. It is a vitally important insight that institutions tend to ‘“freeze” the conditions for voice’ so that if you do not speak or write in the normatively imposed way, you will not be heard (Blommaert, 2008a, p. 428) and this insight has helped reveal serious discursive inequalities in many different contexts (e.g. Hymes, 1996; Blommaert, 2008b, 2009, Bartlett, 2012). However, this critical concept of voice that links it ineluctably with power also ‘freezes’ the conditions for analysing voice. Voice tends to be analysed in terms of powerless groups lacking voice and thus the capacity to be heard and understood. It is generally assumed, though, that powerful figures who have plenty of voice in power terms will be heard and understood. But we shall see here that having voice in power terms does not automatically lead to being heard. We shall see powerless First Nations speakers losing voice but also a discursively powerful crossexaminer and judge losing voice. We need a way of analysing voice, then, that does not just account for the presence or absence of discursive power but also analyses the extent to which it might lead to understanding in a given context. In other words, we need an analysis of how voice projects to an audience in a given context. We shall see that there is a significant disconnect between what I call textual voicing and contextual hearing and between authoritative discourse and comprehensible discourse. Indeed, any speaker can find that they ‘may lose voice even while he or she is using it’ (Blommaert, 2005, p. 233).

3. The Voice Projection Framework The Voice Projection Framework (Heffer, 2013, 2018) provides a heuristic for analysing what can go wrong in the process of conveying one’s perspective (whether in the form of ideas, identities or styles) to a hearer. The notion of ‘voice projection’ is a metaphorical extension from the physical notion of an actor throwing their voice out to the back of an audience. If an actor succeeds in projecting their voice to the back of an audience, their lines will be understood; if a speaker succeeds in projecting their metaphorical voice to the hearer, their perspective (ideas, identities, styles) will be responsively understood by that hearer. At the level of speech act, this is effectively the speaker’s intended perlocutionary effect but the notion of perspective is much broader than speech act: Ideas, identities and styles all go beyond locutions. The four central elements of the framework are voicing, hearing, authority and accommodation. Voicing, or giving semiotic expression to one’s perspective, involves both animating (physically sounding or graphically conveying one’s words—see Goffman, 1981) and styling (discursively choosing one’s words—see Coupland, 2007). It also involves

Suppression, Silencing, Failure to Project

241

indexing (Silverstein, 1976), or guiding audience interpretation by pointing to socioculturally significant features, and highlighting, or emphasising to the audience the salience of certain elements (see Goodwin, 1994). Hearing concerns the immediate conditions of reception. It involves both physical listening and cognitive comprehending. It also involves important interpretative practices, notably schematic framing (Tannen, 1993) and selective focusing (Goodwin, 1994), that affect how meaning, and thus voice, carries. Working in conjunction (or disjunction) with a communication axis of voicing and hearing is a commitment axis of authority and accommodation. Here there is a tension between commitment to authoritative language (Bourdieu, 1991) and a commitment to communicating with the audience. Authority, the source of institutional power, constrains the possibilities for both voicing and hearing (and thus projection) by imposing discursive and other norms through the practices of centring (i.e. working centripetally towards increasing normativity— see Silverstein, 1998; Blommaert, 2005) and authorising (i.e. conferring authority and legitimacy on speakers and discourse—Bourdieu, 1991). Since authority constrains communicative possibilities, it can often hinder successful projection, particularly in cases where speaker and hearer do not belong to the same institutional community. In tension, then, with authority is accommodation, or the extent to which the speaker adapts their speech to the audience (Giles and Powesland, 1975). I note here practices of converging, or adapting to the communicative norms of the speaker, and persuading, or being rhetorically efficacious. In the remainder of this chapter I shall consider the three types of loss of voice—deliberate suppression, silencing and a general failure to project— and, given Toolan’s connection with stylistics, I shall focus on how styling interacts with the other elements of voice projection.

4. The Suppression of Voice in Cross-Examination Critical work on voice (Bernstein, 1990; Blommaert, 2008a; Bartlett, 2012) tends to see it as arising from structure rather than agency. Thus, we could say, as is often said, that the forensic genre of cross-examination is institutionally structured in such a way that it gives cross-examiners an immense amount of power, and thus voice (Drew, 1990; Cotterill, 2003). However, that view seriously underplays the strategic dimension of crossexamination. The institutional expectation in cross-examination is not just that the lawyer will control the discourse but that they will deliberately suppress the witness’s voice (Stone, 2009). From an adversarial legal perspective, the witness’s voice (to the extent that it parallels the lawyer’s case) is expected to be heard during the initial examination-in-chief by the lawyer who called the witness. Once the sceptre passes to the opposing lawyer, though, the legal goal is to thoroughly test the witness’s story through leading questions. Accordingly, it is effectively expected that the

242

Chris Heffer

witness will cease to have a voice as the cross-examiner puts their case to them and attempts to dismantle their story (Heffer, 2005; Cotterill, 2003). However, the simultaneous projection of one’s own voice and suppression of the witness’s voice are something that needs to be achieved through interactional strategy. A ‘successful’ strategy of simultaneous voice projection and suppression can be seen in Roy Black’s cross-examination of Ann Mercer, the friend of the complainant in the William Kennedy Smith rape trial (Matoesian, 2001). The cross-examination can be found on YouTube (Awcsgo, 2011). Black is taking Mercer through her arrival at the Kennedy mansion after her friend called her to say that she had just been raped: BLACK: So you went into the house, is that correct? MERCER: Yes. BLACK: Into the house where the rapist is, right? MERCER: I guess you could say yes. BLACK: It’s dark in there MERCER: Yes. BLACK: You go through the kitchen, right? MERCER: Yes. BLACK: Into this little hallway. MERCER: Yes. BLACK: It’s dark in this hallway, isn’t it? MERCER: Right. BLACK: You meet up with this man who your friend

says is a rapist, isn’t

that correct? In terms of voicing, Black animates in clear diction, he keeps the pace up to retain attention and maintains an apparently calm manner. He also removes all affect from his voice to downplay the gravity of the reported situation. His styling is rhetorically effective. He uses, for example, what Matoesian calls ‘resumptive repetition’ (2001, p. 79), which modifies, expands and elaborates on the previous turn. This helps maintain the narrative pace and is also highly persuasive as it follows an apparently innocuous question with an extremely value-laden one: BLACK: So you went into the house, is that correct? MERCER: Yes. BLACK: Into the house where the rapist is, right?

In his ‘detailing to death’ Mercer’s actions, he is indexing (Silverstein, 1976) sociocultural myths about rape that he evidently believes will be held by his audience: that being alone and in the dark with a suspected rapist is somehow dangerous in itself. To help that indexing, he highlights by actively foregrounding the noun ‘rapist’ through repetition.

Suppression, Silencing, Failure to Project

243

However, in addition to projecting his voice through rhetorically effective animation, styling, indexing and highlighting, he is also, as is typical in cross-examination, actively suppressing Mercer’s voice. He does this partly through his animation choice of a fast pace of interrogation, allowing the witness no time to expand on her answers, but mainly through his styling choices. Exploiting the institutional requirement for the witness to answer his questions, he uses either tag questions (‘isn’t it?’, ‘right?’) or declaratives (‘It’s dark in there’) to restrict the witness to confirmatory answers. And, as if pulling puppet strings, he leads Mercer back through the scene using the immediacy of the historic present. As long as Black uses leading questions, it is extremely difficult for Mercer to resist the insinuations. Her own voice is suppressed and replaced by a surrogate ‘Mercer’ projected by Black. It is not possible to know whether the jurors will frame the narrative in the way indexed by Black: During deliberations, jurors should counterbalance Black’s suggested framing against the framing suggested by Mercer and the prosecution team under direct examination. Nor is it clear whether the jurors will focus on the points highlighted by Black. What is indubitable, though, is that Mercer’s voice has been successfully suppressed. For better or worse, this is the strategic goal of any experienced cross-examiner. Voice suppression, though, is not something that necessarily happens due to the institutional structure of cross-examination. As the following example shows, it needs to be strategically accomplished. When the prosecuting attorney (PA), Moira Lasch, cross-examines the defendant (WS), who is a young doctor, she fails to suppress the witness’s voice: PA:

Doctor Prostko stated in her medical opinion (1.5) she observed a rib contusion on (Patty Bowman) on March 30, 1991 (0.7) What is your explanation of HOW SHE SUSTAINED THAT INJURY? WS: Doctor Prostko (1.0) eh-eh-you’re asking me my explanation. As I said (0.2) all I can tell you is what happened that night. (0.5) PA: OK WS: Now—(0.5) Doctor Prostko testified (1.0) that she had a rib contusion. (1.0) I also recall (1.7) that early (.) on (0.8) before charges were filed (1.9) before my name was even released I believe (1.6) that there were some medical reports (1.6) released from that hospital (1.6) which indicated (1.2) that Patty Bowman had a broken rib (1.5) PA: Well WS: I HAVE heard (0.5) two people testify (1.0) that she did not have a broken rib (1.0) I also heard (1.2) her own orthopaedic surgeon (1.0) testify (1.3) that she had (0.6) bilateral tenderness. (1.5) If Patty Bowman had a rib contusion (0.9) on March 30 I would expect that finding (0.4) to be noted by her orthopaedic surgeon one week later. (0.7) A rib contusion does not become bilateral rib tenderness in one week. (Matoesian, 2001, pp. 172–173)

244

Chris Heffer

In styling her question to the defendant, Lasch fails to follow the ‘golden rule’ of cross-examination: that you never ask an open question. She also fails to regain control in her following turns. The mean turn length of witness responses in cross-examination is very short (I found it to be 9.2 words over 23,670 turns in English jury trials [Heffer, 2005, p. 99]). Smith effectively produces a virtually uninterrupted explanation of 131 words. Presumably Lasch thought that her question was unanswerable without being self-incriminatory, but by opening up the interactional space she allows a socially and institutionally powerful speaker to project his own voice. Kennedy Smith does this by successfully projecting his professional identity as a doctor. He animates his turns with confidence. He first projects rational reasonableness by admitting that there were ‘some medical reports’ indicating a broken rib. He then styles his following turn to index his medical expertise through knowledge not only of medical terms (‘bilateral tenderness’, ‘rib contusion’) but also of formulaic phrases associated with medical register (‘I would expect that finding to be noted’, ‘an x does not become y in z time’). By successfully projecting an expert medical identity, he assumes a more authoritative voice than the cross-examiner with respect to the medical theme under discussion and he derails her line of attack. We can thus conclude that Lasch, despite her institutionally given discursive power, exceptionally loses voice in her own crossexamination and gives over voice to the defendant. Voice suppression, then, is interactionally achieved rather than being institutionally given. Cross-examiners are certainly institutionally empowered through asymmetrical speaking rights to control the witness’s discursive contribution: They ‘have voice’ in Hymes-Blommaert terms. But having voice does not automatically lead to projecting one’s voice or suppressing another’s successfully.

5. The Silencing of Native Voices More frequent than cases of deliberate suppression of another’s voice are cases where lay participants’ voices are institutionally silenced without this necessarily being a deliberate strategy. In recent work, Toolan has described the silencing of class and wealth inequality in the media (Toolan, 2016): Despite exponential growth in wealth inequality in the past few decades, class, at least to the extent it was lexicalised in the lexeme ‘class’, was talked about much less in the media in 2013 than it was in 1971. A similar type of silencing has occurred with regard to falling crime rates: Despite recorded crime falling for decades (Tonry, 2014), the public perception is that crime is rising because of the excessive media attention to serious crime. If perspectives are simply not voiced, in the physical sense of being animated, then there can clearly be no uptake, no responsive understanding, for example, to class and wealth inequality; if they are overly voiced, as in the case of tabloid promotion of crime

Suppression, Silencing, Failure to Project

245

panics, this can lead to a responsive but distorted understanding. Similarly, if minority ethnic groups are not allowed to actively voice their concerns in public fora, then their perspectives will literally go unheard. But even when they are allowed to voice their concerns, as in the case of native land dispute hearings in such countries as Australia and Canada, their voices can be lost in cultural misunderstanding. The Tsilhqot’in First Nations people of British Columbia went to the Canadian Environmental Assessment Agency (CEAA) to oppose the opening of a copper mine on their land (CEAA, 2015). A number of public hearings were held in which the Tsilhqot’in people were able to present their evidence in a formal Western context and with strict time limits but without restrictions on the type of oral evidence they could present. At one level, this is a context in which a minority people are being given the freedom, in everyday terms, to ‘voice their concerns’. At the same time, this is a context in which the minority people inevitably lose voice. Of the many ways in which the Tsilhqot’in people lose voice in these hearings (Payan, 2015),1 I would like to focus on the single issue of the way their oral evidence is entextualized (Bauman and Briggs, 1990) in the official transcripts of the hearings (CEAA, 2015). Firstly, the Tsilhqot’in people lose voice due to impaired listening and/ or comprehension: The first thing that they did was to make a map of the Tsilhqot’in territory. I think that’s the one called the (Native being spoken) on a paper. That is the first map that they made of the Tsilhqot’in people and (inaudible due to extraneous noise) lands. (Anges Heller, pp. 170–171, Aug 13, 2013, Volume 15) While ‘inaudible due to extraneous noise’ simply reflects non-optimal listening conditions, ‘Native being spoken’ is the standard transcription in the archive whenever the Tsilhqot’in language is being spoken. Very often, as here, the native language is used for place names or cultural artefacts for which there are no English translations. However, the elision of the native language can be much more dramatic. This occurs particularly in the case where speakers submit their evidence in the form of song. In an oral culture, song is one of the most effective and powerful ways of projecting one’s voice. However, sung submissions appear as follows in the official transcript: We’re going to sing a song for our water. —Song performed. (Nations Youth Rising, p. 192, Aug 13, 2013, Volume 15) The entire submission is the song but all that remains in the transcript is the metapragmatic comment ‘Song performed’.

246

Chris Heffer

The transcription practices here almost certainly do not constitute a deliberate suppression of the witnesses’ voices. It is standard for nonEnglish languages not to be transcribed in legal transcripts and while, in formal courts, the evidence would be interpreted and the English version transcribed, the transcripts here do not have legal status. The lack of transcription most likely reflects both the transcriber’s lack of knowledge of the local language and a belief that having an expert transcribe and translate the oral evidence would be an ‘inefficient’ use of funds. There might be a case here of selective focusing in the environment agency’s hearing. Focusing is the aspect of hearing which makes cognitively salient certain aspects of the voicing and backgrounds others (Heffer, 2018; Goodwin, 1994). It is the hearing equivalent to the speaker’s highlighting, but what is cognitively salient to the hearer is not necessarily what the speaker intentionally highlighted. In this case, the song is clearly highlighted by the Tsilhqot’in witnesses as being central to their case, but it is likely to be perceived by legal participants as emotive noise that recedes into the cognitive background as they focus on legally relevant information. However, whatever the level of conscious awareness, these transcription practices do effectively silence the voice of these witnesses. The ‘native being spoken’ in the listing of place names and plants, in the singing of oral histories, is much more than an exotic embellishment of the evidence: It is an integral part of the ‘native’ perspective, the Tsilhqot’in identity. Even in the case of submissions delivered entirely in English, the form of the transcription can result in a significant loss of voice. Tedlock (1983) recommended that the speech of Native Americans should be transcribed in poetic lines rather than in prose. The same could be said of Tsilhqot’in speech. Thus, one small section of a community leader’s submission might read: I’m here to help the band members voice out. I’m helping people, let them know they have the strength and we have strong voices and I’ve been helping people getting ready for this hearing for here and other communities and I have with me two letters and a third one is just simple questions. (Carla Jean Billyboy Cuts) Displayed as a poem with lines and layout, the submission seems to be carefully styled with distinct rhythms and effective structure to help project a powerful voice that mirrors the content about voicing out and

Suppression, Silencing, Failure to Project

247

having strong voices. But in the official transcript the speech is conveyed as follows: I’m here to help the band members voice out. I’m helping people, let them know they have the strength and we have strong voices and I’ve been helping people getting ready for this hearing for here and other communities and I have with me two letters and a third one is just simple questions. (Carla Jean Billyboy Cuts, p. 207, Aug 13, 2013, Volume 15) Transcribing in prose rather than poetic lines creates a schematic frame in which the words are construed as providing informational content rather than affective engagement. Furthermore, the careful poetic styling risks being framed in the Western reader’s mind as a careless run-on sentence by an ignorant speaker. Far from being in the powerful style suggested by the words, it is likely to be construed by the Western reader as being in a powerless style, a style that has been empirically observed to result in unfavourable judgements of competence and intelligence (O’Barr, 1982).

6. A Failure to Project Judicial Voice So far, we have seen loss of voice through deliberate strategic suppression by another speaker and through interventions such as transcription that are not deliberately intended to suppress a speaker’s voice but have the net effect of silencing that voice. The last type of loss of voice is the one I was principally thinking about when I developed the Voice Projection Framework (Heffer, 2018): the common case where a speaker’s voice, for a multitude of reasons that I try to capture in the framework, simply fails to project to the hearer and results in a lack of responsive understanding. At the end of jury trials in common law systems, judges explain to the jury the legal framework into which they need to fit the evidence they have heard in the course of the trial. Responsive understanding in this context must mean that the jurors understand the legal instructions and are able to apply them to the evidence. Occasionally, the jury will ‘nullify’ those instructions by deliberately failing to apply them, as has occurred in cases of euthanasia and the revealing of state secrets. In this case the jury has responsively understood the law but disagrees with it and so has decided to resort to common sense justice (Finkel, 2001). However, in the vast majority of cases, the risk is that the jury has not actually understood the legal instructions and so is not able to apply them fairly. It is theoretically imperative, then, that judges should project their voice successfully when communicating with the jury. The tension between authority, the source of institutional power that imposes discursive and other norms, and accommodation, the extent to which the speaker adapts their speech

248

Chris Heffer

to the audience (Heffer, 2018), is clearly central in judges’ instructions to juries and very significantly determines both the styling and the degree to which judges can accommodate to jurors. In many US jurisdictions, the judge can merely animate ‘pattern’ legal instructions authored by legal committees (Dumas, 2000). In England and Wales, though, judges have considerably more discretion to style their own instructions and the recent move has been away from pattern instructions rather than towards them. Judges are, though, still highly constrained by authoritative discourse (Bakhtin, 1981, p. 342) and, as in the US, there is still the risk that the wording they choose will be criticised by the ‘centring institutions’ (Silverstein, 1998; Blommaert, 2005) of the higher courts. There is accordingly a discursive tension in judges between merely animating authoritative legal discourse and using their authorised voice (Heffer, 2013) to style the instructional texts. Individual styling is essential for projection to a lay audience since it permits the judge to accommodate to that audience and thus facilitate comprehension. The importance of accommodation in judicial instruction can be seen in the Vicky Pryce case, involving an economist who had fraudulently taken her Cabinet minister husband’s driving penalty points. The case was held up by both the tabloids and the broadsheets in the UK as showing the rotten nature of a legal system that hinges on the ‘stupidity’ of jurors. That ‘stupidity’ was premised on a list of questions about the legal framing of the case that the jury sent the judge, which Judge Sweeney infamously dismissed as showing ‘absolutely fundamental deficits in understanding’. However, rather than deficits in understanding, I argue that we have a case of failed projection of both the judge and jurors’ voices (Heffer, 2018). One example of the judge’s failure to actively project his voice in a way that can be responsively understood by the lay jurors can be seen in his key legal instruction on the rare (and now abolished) defence of marital coercion, which Pryce was claiming. The various forms of mark-up in the following text all relate to legal register: The law recognises, via the defence of marital coercion, that a wife is morally blameless if she committed an offence only because her husband was present and coerced her—that is put pressure on her to commit the offence in such a way that, as a result, her will was overborne (in the sense that she was impelled to commit the offence because she truly believed that she had no real choice but to do so). One could perhaps forgive a jury for not understanding this legally inflected instruction. But one might also, at first blush, think one can excuse the judge because one might expect at least this definitional paragraph to be dominated by the authoritative legal discourse of statutes and judicial judgements. However, only the words in bold come from

Suppression, Silencing, Failure to Project

249

the 1925 Act that established marital coercion in law. The lack of definition of ‘coercion’ in the Act leads to the inclusion of three key terms (in italics) from judicial judgements relating to the defence: ‘pressure’, ‘will was overborne’ and ‘impelled’. None of the other words in the text are authorised by law. Instead, they are words that stylistically conform to legal register: either legal terms (underlined) such as ‘commit an offence’ or the pseudo-legal ‘morally blameless’ or, more commonly, formulaic phrases (dotted underline) that are not exclusive to legal register but are typically found in legal discourse across common law countries and in particular textual environments (Heffer, 2018). Furthermore, the paragraph conforms to the typical definitional syntax of legislation (Bhatia, 1994): It is one long sentence with multiple complex embedding that, as psycholinguists have shown, can cause problems in comprehension. The crucial point, though, is that, in England and Wales, judges both are permitted to converge with everyday non-legal language and indeed often do so (Heffer, 2005, pp. 157–181). Diverging from ordinary language, as in this case, can thus be seen as a failure of the judge to project his voice to the jury. By not converging discursively with the jury, he is decreasing the chances of their responsively understanding his instructions, as demonstrated by the question they subsequently put to him: Q1. You have defined the defence of marital coercion on page 5 of the jury bundle and also explained what does not fall within the definition by way of examples. Please expand on the definition, provide examples of what may fall within the defence, specifically ‘will was overborne’ and does the defence require violence or physical threat? What is extraordinary about this question from a supposedly ‘dumb’ jury is both that it shows convergence with legal linguistic norms (e.g. ‘jury bundle’, ‘fall within the definition’) and that it picks up on a substantive error in instruction (legal precedence establishes that the judge should have provided examples of what might fall within the defence). Interestingly, in his Ruling on Marital Coercion, intended for fellow legal professionals, the judge glossed the comparatively rare terms ‘overborne’ and ‘impelled’ with the ordinary terms ‘overcome’ and ‘forced’. He did not gloss them, though, for the lay jury either in his original instruction or in his response to their question about it. What this case briefly shows (a full analysis can be found in Heffer, 2018) is that even the most powerful speaker in court can fail to project their voice to an audience. The objection often made at this juncture is that the judge is focusing on a different audience: the appellate judges that might review his summing-up. But legal authorities in the UK have stressed time and again that the primary audience is the jury and this has led to radical linguistic reforms in instruction, such as the recommended paraphrase of the legal term ‘beyond reasonable doubt’ with the everyday

250

Chris Heffer

‘be sure’. However powerful institutionally, the judge can still lose voice whilst using it.

7. Conclusion There are three different ways in which one might lose one’s metaphorical voice while using physical voice. In the case of cross-examination in jury trial, the cross-examiner deliberately tries to suppress the witness’s voice by tightly controlling their answers. There is a strong correlation in this case between the freedom to speak and the capacity to be heard, as can be seen when a cross-examiner accidentally opens up discursive space to a powerful witness. In the case of the Tsilhqot’in land hearings, on the other hand, the native speakers have freedom to say or sing what they like but Western legal transcription practices effectively silence their voices by treating as superfluous what the speakers construe as central to their argument. Finally, in the case of jury instruction, the judge’s voice simply fails to project to the jury because he is focusing on a different audience: his colleagues in the appeal courts. In terms of the relation between voice and power, we have seen powerful witnesses in their own communities lose voice and become powerless before a Western audience; powerful cross-examiners both succeeding and failing to suppress the witness’s voice; and powerful judges failing to project their voice to the jury. Consequently, voice does not correlate directly either with speaking or with power but must be considered in its own right. The potential impact of this critical notion of voice is that it does not correspond to the act of speaking in itself. It is perfectly possible to be ‘given voice’, in the sense of being allowed to have one’s say about things that matter to us, but without actually ‘having voice’, in the sense of being able to achieve responsive understanding. While some witnesses might affectively be satisfied with ‘having their say’ in court, in the interests of justice, witnesses should not just be ‘given voice’ but be figuratively ‘heard’ by the rhetorical audience of jury and/or judge. Their evidence should be taken up and deliberated on by the factfinders. And the judge’s legal instructions also need to be taken up and applied to that evidence. These are the ideal conditions for deliberation of the evidence by judge or jury but they are conditions that often fail to be satisfied, potentially with tragic results. Returning to the general question of context and agency, the likelihood of successful voice projection is certainly constrained by the context: The witness’s voice in cross-examination will generally be suppressed because the cross-examiner has interactional control and it is in their interests to suppress it; the ethnic minority speaker’s voice is likely to be silenced in a Western formal institutional context partly through practices of legal transcription; and judicial instruction in many jurisdictions is constrained by regulation, preventing accommodation with the jury.

Suppression, Silencing, Failure to Project

251

However, in each case we have also seen the effect of the speaker’s own agency: Black succeeds while Lasch fails to exploit their institutionally empowered discursive control to suppress the witness’s voice; the CEAA fails to invest in transcription practices that would allow native voices to carry a little more; and Judge Sweeney fails to exploit his jurisdictional discretion to linguistically converge with the jury. Losing voice, then, can happen irrespective of power or discursive context, though the personal consequences for the rape complainant or the native land rights activist are far graver than for the professional judge.

Note 1. I would like to thank my former MA student Nicole Payan for raising my awareness of the relevance of voice projection in a native land rights context. The examples discussed are hers.

References Austin, J. L. (1962) How to do things with words. Oxford: Oxford University Press. Awcsgo. (2011) Mercer Cross part 1 of 3.wmv, Available at: www.youtube.com/ watch?v=H-V7qmra0nI (Accessed 2 April 2018). Bakhtin, M. (1981) The dialogic imagination. Austin: University of Texas Press. Bartlett, T. (2012) Hybrid voices and collaborative change: Contextualising positive discourse analysis. New York: Routledge. Bauman, R. and C.L. Briggs (1990) ‘Poetics and performance as critical perspectives on language and social life’, Annual Review of Anthropology, 19, pp. 59–88. Bernstein, B. (1990) Class, codes and control: The structuring of pedagogic discourse (vol. 4). London: Routledge. Bhatia, V. K. (1994) ‘Cognitive structuring in legislative provisions’, in Gibbons, J. (ed.) Language and the law. London: Longman, pp. 136–155. Blommaert, J. (2005) ‘Bourdieu the ethnographer: The ethnographic grounding of habitus and voice’, The Translator, 11 (2), pp. 219–236. Blommaert, J. (2008a) ‘Bernstein and poetics revisited: Voice, globalization and education’, Discourse & Society, 19 (4), pp. 425–451. Blommaert, J. (2008b) Grassroots literacy: Writing, identity and voice in Central Africa. London: Routledge. Blommaert, J. (2009) ‘Language, asylum, and the national order’, Current Anthropology, 50 (4), pp. 415–441. Bourdieu, P. (1991) Language and symbolic power. Cambridge: Polity Press. CEAA. (2015) ‘New prosperity gold-copper mine project’, Canadian Environmental Assessment Agency. Canadian Government. Chomsky, N. (1968) Language and mind. New York: Harcourt, Brace and World. Conley, J. M. and O’Barr, W. M. (1990) Rules versus relationships: The ethnography of legal discourse. Chicago: University of Chicago Press. Cotterill, J. (2003) Language and power in court: A linguistic analysis of the O.J. Simpson trial. Basingstoke: Palgrave Macmillan.

252

Chris Heffer

Coupland, N. (2007) Style. Cambridge: Cambridge University Press. Drew, P. (1990) ‘Strategies in the contest between lawyer and witness in crossexamination’, in Levi, J. N. and Walker, A. G. (eds) Language in the judicial process. Boston, MA: Springer. Dumas, B. (2000) ‘US pattern jury instructions: Problems and proposals’, Forensic Linguistics, 7 (1), pp. 49–71. Finkel, N. J. (2001) Commonsense justice: jurors’ notions of the law. Cambridge, MA: Harvard University Press. Giles, H. and Powesland, P. (1975) ‘A social psychological model of speech diversity’, in Giles, H. and Powesland, P. (eds.) Speech style and social evaluation. New York: Harcourt, Brace, pp. 154–170. Goffman, E. (1981) Forms of talk. Philadelphia: University of Pennsylvania Press. Goodwin, C. (1994) ‘Professional vision’, American Anthropologist, 96 (3), pp. 606–633. Halliday, M. A. K. and Hasan, R. (1976) Cohesion in English. London: Longman. Harris, R. (1981) The language myth. London: Duckworth. Heffer, C. (2005) The language of jury trial: A corpus-aided analysis of legal-lay discourse. Basingstoke: Palgrave Macmillan. Heffer, C. (2013) Projecting voice: Towards an agentive understanding of a critical capacity’’, in Cardiff Papers in Language and Literature. Cardiff: Cardiff University Press. Heffer, C. (2018) ‘When voices fail to carry: Voice Projection and the case of the “dumb” jury’, in Durant, A. and Leung, J. (eds.) Language, power and law: The invisible exercise of power through language. Cambridge: Cambridge University Press, pp. 207-235. Hymes, D. (1996) Ethnography, linguistics, narrative inequality: Toward an understanding of voice. London: Taylor & Francis. Matoesian, G. M. (2001) Law and the language of identity: Discourse in the William Kennedy Smith rape trial. Oxford: Oxford University Press. O’Barr, W. M. (1982) Linguistic evidence: Language, power and strategy in the courtroom. New York: Academic Press. Payan, N. C. (2015) ‘“The least I can do is speak out”: Projecting voice through aboriginal oral traditions’, Unpublished M.A Thesis. Cardiff University. Saussure, F. D. (1983) [1916] Course in general linguistics. Trans. R. Harris. London: Duckworth. Silverstein, M. (1976) ’Shifters, linguistic categories and cultural description’, in Basso, K. and Selby, H. A. (eds.) Meaning in anthropology. Albuquerque: University of New Mexico Press, pp. 11–55. Silverstein, M. (1998) ‘Contemporary transformations of local linguistic communities’, Annual Review of Anthropology, 27, pp. 401–426. Sperber, D. and Wilson, D. (1986) Relevance: Communication and cognition. Oxford: Wiley-Blackwell. Stone, M. (2009) Cross-examination in criminal trials. Hayward Heath: Tottel. Sweeney, J. (2013) ‘Summing-up in R v Vasiliki Pryce’, Available at: www.crime line.info/news/pryce-jury-summing-up. (Accessed 27 March 2018). Tannen, D. (ed.) (1993) Framing in discourse. Oxford: Oxford University Press. Tedlock, D. (1983) Spoken word and the work of interpretation. Philadelphia: University of Pennsylvania Press.

Suppression, Silencing, Failure to Project

253

Tonry, M. (2014) ‘Why crime rates are falling throughout the western world’, Crime and Justice, 43, p. 1. Toolan, M. (1992) Language, text and context: Essays in stylistics. London: Routledge. Toolan, M. (1996) Total speech: An integrational linguistic approach to language. Durham, NC: Duke University Press. Toolan, M. (2016) ‘Peter Black, Christopher Stevens, class and inequality in the Daily Mail’, Discourse and Society, 27 (6), pp. 642–660.

16 ‘Hey YouTube’ Positioning the Viewer in Vlogs Stephen Pihlaja

1. Introduction In the 25 years since Language, Text and Context (Toolan, 1992) first appeared, the development of the Internet has created new ways in which these three elements (language, text and context) of communication have come to be understood. In particular, online video has provided new affordances of media production, new genres and new ways of storytelling. Driving online video production and consumption, YouTube has also changed the way media is consumed and produced online. The site YouTube, established in 2005, now hosts videos produced by a variety of celebrities, production companies, communities and brands. One innovation the site has popularised is a new short form video genre called the ‘vlog’, a portmanteau of video and blogging, which is itself a portmanteau of ‘web’ and ‘log’, shortened to ‘blog’. A vlog blends traditional film-making techniques, the text-based genre of blogging, the selfie and reality television, and has grown as a dominant genre. Initially, the vlog was spurred by a lowered cost of webcams, which allowed users to record videos from their computers and post them online. In this configuration, vlogging was simply users talking directly to webcams placed on the tops of their computer monitors. However, as point-and-shoot cameras developed to take high-quality video and mobile technology improved upon cameras in smartphones, the vlog became increasingly varied. ‘Daily vlogging’ has grown in popularity, wherein users record their daily lives in a variety of settings, and mix commentary and direct interaction with non-present users and with people in the physical location where a vlog is recorded. These videos often include regular audience address and pragmatic cues similar to face-to-face conversation (Frobenius, 2014). They also can include the audience as a character in the story of the vlog. Using the notion of ‘positioning’ (Bamberg, 1997) and applying it to online video, the chapter will consider how ‘context’ and ‘text’ interact, particularly in deictic shifts, or shifts in address, in vlogs.

‘Hey YouTube’

255

2. Background Daily vlogging blends elements of documentary film-making and reality television with the stylistic device of audience address; that is, speaking directly to the viewer through the camera. Address can be marked in a variety of ways, through greetings, pronominal reference to the viewers or gaze. Although vlogs may differ in their exact form, a consistent part of any vlog is address of the viewer through the camera. As in other forms of storytelling, users frequently greet the non-present audience in habitual ways, with phrases such as ‘Hey YouTube’. Bell and Ensslin (2011) analyse how addressing the non-present audience as ‘you’ occurs in literary texts, including Interactive Fiction found in digital contexts. In digital fiction, address of ‘you’ provides a variety of functions, which Bell and Ensslin present using Herman’s (2004) typology of the functions of ‘you’ in fiction. They point out the address of ‘you’ in Interactive Fiction and video games, unlike other forms of print narratives, means that the reader is ‘somewhat obligated’ (Bell and Ensslin, 2011, p. 318) to take on the position of ‘you’ because Interactive Fiction and video games engage the reader or player or viewer as a character within the fictional world of the game or text. In digital fiction, the address of ‘you’ and the affordance of choice within text, including the use of hypertext, creates novel ways of storytelling and reading, giving agency to the reader regarding the direction in which a particular story might progress. In vlogging, the address of the viewer throughout the vloggers’ narration of their ‘real life’ also potentially offers novel ways for the viewer to interact with the vloggers. Interaction, through the apparatus of social media, is not only possible, but encouraged, and contact between users as both video makers and viewers has long been a part of YouTube community life (Pihlaja, 2014). The viewer may engage both with the vlog page by ‘liking’ and commenting on a video, and with the ongoing story of the vlog by subscribing to the channel to follow the story. They might also interact with vloggers they like in other contexts, such as Twitter, Instagram or forums such as Reddit. Unlike in other forms of passive media consumption such as reality television shows, the addressed viewer is often explicitly positioned as an active agent, with the viewer being asked to 'give the video a thumbs up' (YouTube’s apparatus for rating a video), commenting or subscribing. This positioning is possible because of the focus on interaction among users of social media (Page, 2013), and the notion of mutual access. However, while interaction between any two users on YouTube is possible, YouTube’s revenue-sharing programme wherein users are paid to create content through advertising has led to a proliferation of popular, professional content creators. Consequently, much daily vlogging has become similar to a more traditional media model, with video producers

256

Stephen Pihlaja

and audiences, unlike other social media sites where users are more likely to be both producers and consumers of content. Still, the social media affordances of YouTube remain and the vlogs often include at least a notional interaction between producer and viewer in the development of narratives in vlogs. To provide a better understanding of the role of the viewer in vlog storytelling, this chapter will employ positioning analysis. The relationship between storytelling and the construction of the social world was popularised by Labov’s work looking at narrative in the inner-city and the role of narrative in identity construction and maintenance (Labov, 1972). Since Labov’s work, ‘narrative’ has been used in a variety of different contexts and frameworks (De Fina and Georgakopoulou, 2012). Although Labov’s notion of ‘narrative’ focused on ‘narratives of personal experience’, the kinds of narrative and the contexts in which they occur have developed further, particularly in analysis of interaction among speakers in conversations. For example, Bamberg and Georgakopoulou’s later work on positioning (2008) also includes ‘narratives-in-interaction’ and shows how narrative positioning in interaction can involve more than one speaker (see also De Fina and Georgakopoulou, 2012). In these narratives, important relational work is done and the ‘identity work’ of narrative is co-produced. Narratives, of course, may still have one single speaker as the teller, with an audience, but they may also emerge in interaction, with speakers constructing narrative accounts of their experiences together (Ochs and Capps, 2011). The interaction in and around a story can be analysed on a variety of levels and Bamberg (1997) and colleagues (Bamberg and Georgakopoulou, 2008) describe positioning in storytelling as occurring on three levels. • • •

The interaction of characters within individual stories or narratives (Level 1); The interaction of storyteller and audiences (Level 2); and A storyteller’s identity in relation to the larger social environment (Level 3).

By examining these different levels of positioning, Bamberg suggests that how, why and to what effect a certain narrative is being told becomes clearer. The analysis can reveal why certain narratives are told in certain contexts, how storytellers and audiences view one another and what the stories reveal about the social conditions in which the narratives are told. The focus of the analysis in this chapter is the development of context in Levels 1 and 2. In YouTube videos, the context of the storytelling and hearer and the position of the audience is problematised given the lack of an explicit fictional world. While in Bell and Ensslin’s (2011) work on digital fiction, the ‘you’ remains within the fictional world of Level 1, the vlogs I analyse in this text are not fictional. Ontologically, the viewer

‘Hey YouTube’

257

and the vlogger inhabit the same world, but unlike in interactive contexts where both the listener and hearer may be characters in the narratives while also being present at the telling, as in stories told in face-to-face interaction (De Fina and Georgakopoulou, 2012), vlog viewers are not physically present and cannot interact with the storyteller at the time of the telling. Instead, the address of the audience as ‘you’ creates a position for the viewer as simultaneously the viewer of the video, but often also a character within it. The ‘you’ of the vlog is the audience more generally, an imagined audience from the perspective of the vlogger, but also particular users who have agency and can theoretically interact with the storyteller in the comments or even the real world. The effect of this positioning of the viewer in vlog narratives will be the focus of this analysis, by looking at how the audience is addressed and what agency is afforded them. I present a case study of the influential YouTuber Casey Neistat, who enjoyed considerable popularity as a daily vlogger from 2015 to 2016. Neistat’s videos frequently involved audience address in a dynamic way, directly speaking to the viewer and making reference to them as ‘you’. The aim of the chapter is to describe and analyse where and when shifts in address between people in the physical world of the vlog and the viewer take place and to what effect. Looking closely at two videos and the subsequent comments, I analyse both technical features of address in vlogs and the subsequent consequences on the social interaction between the vlogger and the audience, and among members of the viewing audience.

3. Method In 2017, the popular YouTuber Casey Neistat had nearly eight million subscribers to his channel, and according to the social media statistics aggregating website Socialblade (Socialblade, no date), his videos regularly received more than one million views and he ranked in the top 300 most subscribed video makers on YouTube. Neistat’s videos generally fall into three categories. First, he creates vlogs in which he regularly addresses the viewer as a part of an ongoing narrative, speaking directly to the camera while physically interacting with others in his day-today life. Second, Neistat less frequently makes videos in which he only addresses the viewer, keeping the camera in a static position and talking directly to the audience. This style tends to feature when Neistat presents himself as having something important or difficult to say to the viewer, like when he announced that he was voting for Hillary Clinton in the 2016 US Presidential election, or when he announced he was no longer going to vlog on a daily basis. Third, and most rare, are videos in which audience address does not explicitly occur. These videos can be seen in Neistat’s older work as a film-maker, and in some of the more carefully edited videos he posts on the channel.

258

Stephen Pihlaja

Given the limited scope of this chapter, a close analysis of two videos was chosen for the case study to analyse Neistat’s positioning of the viewer, one in which he addresses viewers as a part of a daily vlog and one in which he discusses the topic of his relationship with viewers. The details of the videos as they are published on YouTube are as follows: Video 1 Title

STAGE FRIGHT!!

Date

Published on 14 Feb 2016

Length 7 minutes 20 seconds URL

www.youtube.com/watch?v=vzCmK9xmwM8

Video 2 Title

i’m not that happy

Date

Published on 12 Jan 2017

Length 8 minutes 33 seconds URL

www.youtube.com/watch?v=sVnI_2JgXGY

NB Video titles and addresses were accurate at the time of data collection in September 2017.

The focus of the analysis that I present is on the talk in the videos. The transcriptions for the videos were adapted from YouTube’s automatic whole word transcripts, and were checked for accuracy before being formatted as intonation units (Chafe, 1988). The transcripts totalled 6,212 words. I then employed Bamberg’s (1997) notion of positioning to look specifically at moments of audience address. In the first video, this included specific analysis of shifts between the first and second level of positioning: from places where a story is being told to the viewer and places in which the viewer is explicitly a part of the story. In the second video, the analysis focused on how Neistat spoke explicitly about his position and that of the audience. I also searched the comments for references to Neistat and second-person address, looking for stretches of discourse in which users either attempted to interact with Neistat or interacted with one another around the issues of audience as character or their own position in relation to Neistat, either as fans, community members or viewers.

4. Analysis Neistat’s videos typically follow a ‘daily vlogging’ format, in which he records and broadcasts his daily life, including many different elements of life, such as a static shot of him speaking directly to the camera,

‘Hey YouTube’

259

reminiscent of Burgess and Green’s (2008) definition of a vlog as ‘a talking head, a camera, and some editing’ (p. 6). These moments occur often in his vlogs, but occasionally Neistat records full-length videos in this style, often with caveats that he has something important to speak about. This chapter looks at two videos, one of each style. In the video ‘STAGE FRIGHT!!’ published on 14 February 2016, Neistat vlogs during a trip with his family to Baltimore where he is meant to speak at a convention. The video starts with Neistat speaking to a crowd of young people, waving to them while they cheer and telling them that he needs to go back to meet his wife. At the end of the video, it becomes clear this scene comes from later in the day. After an interlude of music and footage of New York City, in the next scene Neistat is in a hotel room, with his wife and toddler daughter in the background sitting on the bed. The shot juxtaposes the more intimate setting with the first very public scene. Neistat addresses the audience directly, with the shot focused on his face and several moments of silence before the video begins with Neistat taking a drink from a coffee cup and finally addressing the viewer directly, saying, It’s morning in Baltimore I’m here today to speak uh at an event for kids not kids teenagers pretty excited about it Structurally, Neistat uses audience address to foreshadow what follows in the video and add context to the shot that came before and the story of the vlog. Neistat positions himself as both the storyteller and the main character of the narrative. The audience is positioned as both the story hearer and as a companion with Neistat on the trip, having access to both the private location of his hotel room, and also Neistat’s feelings about the day. Neistat then deictically shifts his address to his wife, Candice, and daughter sitting behind him, saying, Hey what are you doing today are you are you gonna come with me Although his wife responds, her words are not audible and Neistat addresses the viewer again, saying that he doesn’t know what his family is doing for the day. In the exchange between Neistat and Candice, the viewer is still positioned as central to the scene, downplaying Candice

260

Stephen Pihlaja

and Neistat’s conversation and emphasising the viewer’s presence. The physical positioning of Candice and their daughter in the background and the relative absence of Candice’s voice further establish the viewer as Neistat’s companion in a story from his perspective. Neistat’s role as the storyteller is evident in his choosing and ordering of events, but also in how he implicitly structures the story through his editing of the vlog which works to include the audience at key points in the story. Audience address, marked in this vlog primarily through gaze, at private moments then provides a common link between two positions that Neistat takes up as a character in his own story. In the first, Neistat positions himself more generally as a celebrity who is very popular, with crowds cheering for him at several different times in the vlog. In the second, Neistat positions himself as a father and husband, through his actions of caring for his family. These are two distinct parts of the vlog with different positions for the characters within them, but the viewer is equally given access to both through the mediation of the vlog, providing a sense of assumed omniscience on their part. The vlog continues with Neistat showing footage from the day and engaging with the various people helping him prepare for his talk. Throughout these scenes, he speaks directly with fans and staff members, all while recording the interaction himself. At one moment, he addresses the video’s audience, saying, That so that’s fun they’re letting the kids in now so we’re back here hiding until they let the kids in then we’re gonna go out there and say hi The address here functions in a similar way to the earlier example, structuring the narrative by explaining what is happening and giving context for what has just been shown and what will come (‘they’re letting the kids in now’). It also includes an evaluative remark ‘that’s fun’ positioning the viewer as Neistat’s confidant, someone who is separated from the crowd and accompanies Neistat in private places. The juxtaposition between the public and private is particularly highlighted, with the large groups of fans he films and individual fans asking, ‘Can I be in the vlog?’, a recurring comment that is captured by Neistat as he films fans when they meet him in public places. The deictic choices, however, position the viewer as taking Neistat’s perspective with him, experiencing the same things as Neistat without having to make a request. For example, the use of we (‘we’re gonna go out there and say hi’) exploits the referential ambiguity of the first-person plural pronoun and appears to refer to the

‘Hey YouTube’

261

people physically present with Neistat, but the reference can be heard ambiguously, as referring also to the Neistats and the wider audience of viewers watching along through the YouTube channel. The vlog then includes footage of Neistat with his wife and daughter. First, they are shown eating at a restaurant where he again addresses the wider YouTube audience, describing where he and his family are and what they are eating. Next, Neistat, walking alone, addresses the viewer, explaining that he is leaving to find medicine for his daughter who is sick and to buy food for his family. As in the previous cases, the context, narrated to the audience, does important identity work for Neistat, bringing together the parts of the vlog involving his family, and positioning Neistat as a father, as prioritising his family’s needs. After acquiring the medicine and food, he returns to the convention centre to deliver his talk and there is a brief scene of Neistat being cheered by fans. The vlog ends with the same footage of Neistat saying goodbye to a group of fans that the vlog began with. However, the scene continues past where the opening footage ended, with Neistat turning the camera away from the crowd and back to himself as he walks away, saying, Okay that was fun now I’m going home and going to bed This final address of the audience serves both the purpose of ending the story and adding Neistat’s own evaluation of the day as ‘fun’. The viewer is positioned as having spent the whole day with Neistat, having experienced the same things that he has. Neistat does not refer to anything that has occurred that he has not filmed (although, of course, many such instances are un-narrated and not included in the video). Instead, the entirety of the video is presented as a telling of Neistat’s experience in one day. What is and is not included in these kinds of daily vlogs is the topic of the second video in this analysis ‘i’m not that happy’. Posted on 12 January 2017, the video follows a more traditional vlog style, with a static shot of Neistat speaking directly to the camera without his iconic sunglasses. In this vlog, Neistat responds to popular YouTuber PewDiePie’s video called ‘Forced Positivity on YouTube’ in which PewDiePie criticises a culture of video-making in which vloggers produce content in which they are only happy and positive and avoid negative aspects of life. Neistat discusses the negative implications of this both for the producer and for the viewers. Neistat does not explicitly agree or disagree with PewDiePie’s claim that this is particularly negative for ‘younger viewers’, who might get the wrong impression about the lives of YouTubers. Instead, Neistat presents his own ‘thesis’ about the relationship between

262

Stephen Pihlaja

vlog content and the viewer’s position. The vlog comes at a time when Neistat has stopped daily vlogging and begins with Neistat explaining that he is ‘giving an update’: Back when I was daily vlogging I used to say that talking to the camera is a little bit like talking to a therapist I mean I don’t know who you are under— who most of you are but just like a therapist you don’t really talk back much and you give me really considerate feedback in the comments Neistat refers to the content he creates as ‘talking to the camera’, providing insight about his understanding of vlogging and the larger social context in which he makes the video. Through the ambiguity of the second-person pronoun ‘you’ he positions himself as communicating with everyone and no one at the same time. This shifts in the next part of the statement where the ‘camera’ comes to stand in for his audience when he follows up his point by saying, ‘I don’t know who . . . most of you are, but just like a therapist, you don’t really talk back much’. Neistat’s sarcastic comment that viewers ‘give me really considerate feedback in the comments’ also suggests that the interactivity YouTube offers is not something that he necessarily views as a positive thing. Indeed, the negativity of YouTube comments has long been well-documented (Lorenzo-Dus, Garcés-Conejos Blitvich and Bou-Franch, 2011; Moor et al., 2010) and Neistat is no exception in attracting ‘hate’ from viewers. To assess the extent to which comments on Neistat’s videos are actually negative rather than positive is beyond the scope of this analysis. However, comment threads do provide possibilities for conversational-style interaction between the YouTube audience and Neistat. In the comment threads on the two videos, there is evidence of users making explicit attempts to engage in a dialogue with Neistat about the video, but no instances where he responds to them directly in the comments. There are instances when discussion occurs, but these are between different commenters around the topic of the video. This lack of engagement from Neistat is likely the consequence of his own popularity, with the two videos in this chapter garnering 2,123 and 17,642 comments respectively, creating a context in which it would be impossible for Neistat to respond to every comment.

‘Hey YouTube’

263

Neistat describes his own experiences, and particularly the stress involved in daily vlogging, saying he works to ‘pluck’ happy and fun parts out of his day to produce videos that make the viewer feel good. Neistat refers to ‘you the audience’ and ‘you viewers’ in the video, suggesting that the videos are to manage and maintain his audience. Neistat describes waking up and thinking, ‘What’s today’s vlog episode gonna be’ and then ‘throughout the day trying to create it’. Daily vlogging, for Neistat, is a particular genre of video that is intentionally ‘shaped’ (as Neistat describes editing) to create content that makes the audience ‘feel good’. This ‘thesis’ for creating content is specifically about increasing viewership for the videos because ‘no one wants to watch a video of someone grovelling about how lame their life is’. Neistat’s suggestion in the title ‘i’m not that happy’ becomes then a statement about authenticity in vlogging, rather than an intimate confession. The focus on being ‘authentic’ in talking about vlogging draws attention to the medium of social media and the value placed on authenticity (Leppänen et al., 2015). However, Neistat, in attempting to appear authentic, simultaneously exposes the constructedness of his vlogs. He explains that viewers should not perceive vlogging as an accurate portrayal of a life, but instead as entertainment, produced for viewers to feel good and enjoy themselves. Neistat’s positioning of the viewer suggests that understanding YouTube, and particularly YouTube celebrities and their fans, in terms of a peer-to-peer social media ‘community’ is problematic. For Neistat, the vlog is a genre of film-making, one in which the viewer’s position is made explicit and exploited within narrative structure, but it does not suggest a meaningful shift in the actual role of the viewer, and so is unlike the deictic address of Interactive Fiction presented by Bell and Ensslin (2011). The vlog is a one-to-many way of broadcasting content. Neistat’s social positions, first, in relation to the actual members of his audience, and second, as he presents the viewer in his videos, are substantially different positions. In this video, the viewer is conceived of as consumer of the vlog, someone who watches for entertainment and to feel ‘positive’ about their lives, rather than to have access in a meaningful way to Neistat’s private life. Neistat makes clear that the content presented in the vlog is a produced, if not performed, collection of moments edited together to present a narrative that must be positive, regardless of the actual circumstances of his life. The vlog as a developing genre poses challenges to models of storytelling in which there is either a clear teller and hearer as in Labov (1972) or Bamberg’s (1997) earlier work or an integrated, co-constructed narrative as in Bamberg and Georgakopoulou’s (2008) later work. The analysis here has shown the ways in which this perception of integrated co-tellership may not always apply to mediated contexts of narration. Although some forms of social media may offer access to the private lives of different celebrities and content creators, the extent to which this

264

Stephen Pihlaja

access is genuine or viewers are actually interacting with vloggers must be questioned. Neistat’s position as both the storyteller and the main character in the vlog is a well-established stylistic device in first-person narratives. The vlog potentially innovates on the viewer’s position, where the viewer is both a presence within the ongoing narrative and a ‘story hearer’. Close analysis, however, has shown that audience address is still primarily a stylistic choice. Despite their positioning, the viewer watching the YouTube video is not actually a participant within the world of Neistat’s video in a meaningful way. Neistat uses audience address to maintain narrative continuity and a sense of intimacy, but the address of the audience is a means to the storytelling end. The address of the viewer is a form of synthetic personalisation (Fairclough, 1989) common in broadcast talk, meant to create a connection with the audience, and is, at least in Neistat’s case, a performed intimacy. This positioning does not mean, however, that the audience is actually participating in the vlog. This is problematic, however, in that Neistat, as a social media celebrity, encourages interaction through commenting on and liking his videos. Doing this is not the same as interaction with Neistat himself as a person. Rather, the likes and comments on the videos help drive his popularity, not necessarily the audience’s proximity to him. Audience address and the perception of intimacy have consequences for how his viewers might perceive their relationship with and knowledge of him. Questions of ‘authenticity’ aside, there is discrepancy in Neistat’s vlogs which make the viewer a central focus in his life, while he also says, ‘I don’t know who you are’. The statement exemplifies the consequences of social media celebrity, which is simultaneously built on the sense that the viewer is a part of the celebrity’s social network, but also on the sense that the celebrity is growing so popular that they cannot meaningfully engage with a majority of users. Some users may mitigate this alienation by speaking about and to their audience as a community around their shared interest, often conceived of as being larger than the individual celebrity at the centre (like Hank and John Green, the Vlogbrothers, referring to their fans as ‘Nerdfighteria’). However, Neistat’s model of YouTube engagement contrasts to that of other users because it seems to eschew this explicit community engagement and insists instead on his own position as a video maker with an audience.

5. Conclusion This brief analysis of two vlogs has shown how audience address and viewer positioning structure narratives. They also create a sense of intimacy and agency, particularly when videos are posted on social media sites where users have apparently direct access to the producers (which in this case is available through ranking the video or adding comments). ‘Context’ then includes the potential for dynamic interaction between

‘Hey YouTube’

265

the social media video and the physical world of which it is a representation. This intimacy and agency, however, is to some extent simulated rather than genuine. The positioning of the viewer as participant need not include direct participation in the content production. ‘Vlogging’, despite its other novel properties, has come to mean a recognisable style of video production, rather than a radical change to media production and consumption. Although the apparatus of social media does provide the potential for co-production and interaction, video producers may see themselves as primarily producing content for audiences and using social media as a route for distribution without a genuine shift in how audiences and users interact.

References Bamberg, M. (1997) ‘Positioning between structure and performance’, Journal of Narrative and Life History, 7 (1–4), pp. 335–342. Bamberg, M. and Georgakopoulou, A. (2008) ‘Small stories as a new perspective in narrative and identity analysis’, Text & Talk, 28, pp. 377–396. Bell, A. and Ensslin, A. (2011) ‘ “I know what it was. You know what it was”: Second-person narration in hypertext fiction’, Narrative, 19 (3), pp. 311–329. Burgess, J. and Green, J. (2008) Agency and controversy in the YouTube community. In Proceedings IR 9.0: Rethinking Communities, Rethinking Place - Association of Internet Researchers (AoIR) Conference, IT, University of Copenhagen, Denmark. Chafe, W. (1988) ‘Punctuation and the prosody of written language’, Written Communication, 5 (4), pp. 395–426. De Fina, A. and Georgakopoulou, A. (2012) Analyzing narrative: Discourse and sociolinguistic perspectives. Cambridge: Cambridge University Press. Fairclough, N. (1989) Language and power. London: Longman. Frobenius, M. (2014) ‘Audience design in monologues: How vloggers involve their viewers’, Journal of Pragmatics, 72 (Supplement C), pp. 59–72. Herman, D. (2004) Story logic: Problems and possibilities of narrative. Lincoln, Nebraska: University of Nebraska Press. Labov, W. (1972) Language in the inner city. Philadelphia: University of Pennsylvania Press. Leppänen, S., Møller, J. S., Nørreby, T. R., Stæhr, A. and Kytölä, S. (2015) ‘Authenticity, normativity and social media’, Discourse, Context & Media, 8 (Supplement C), pp. 1–5. Lorenzo-Dus, N., Garcés-Conejos Blitvich, P. and Bou-Franch, P. (2011) ‘Online polylogues and impoliteness: The case of postings sent in response to the Obama Reggaeton YouTube video’, Journal of Pragmatics, 43 (10), pp. 2578–2593. Moor, P. J., Heuvelman, A. and Verleur, R. (2010) ‘Flaming on youTube’, Computers in Human Behavior, 26 (6), pp. 1536–1546. Ochs, E. and Capps, L. (2011) Living narrative. Cambridge, MA: Harvard University Press. Page, R. (2013) Stories and social media: Identities and interaction. London: Routledge.

266

Stephen Pihlaja

Pihlaja, S. (2014) Antagonism on YouTube: Metaphor in online discourse. London: Bloomsbury. Socialblade. (n.d.). ‘caseyneistat YouTube statistics’, Available at: https://socialblade. com/youtube/user/caseyneistat (Accessed xxx, please insert last access date). Toolan, M. (ed.) (1992) Language, text and context:Eessays in stylistics. London: Routledge.

Section 5

Politics

17 The Value of Intertextual Associations How GM Technologies Are Given Value Through Association Alexanne Don This chapter discusses how ‘science’ and genetically modified organisms (GMOs) are associated and given value by proponents of genetic modification (GM) in the news media, at the same time as these clusters of positive attributes are contrasted with stances opposed to the use of genetic modification of plants and animals. The discussion focuses on how meaning-making practices in the recontextualisation of scientific research, i.e. in news media pieces, act to position certain actors and institutions in the community. Processes of recontextualisation (see Figure 17.1 later in the chapter; Bernstein, 2003) occur when texts and the ideas they contain are taken from their original context and reproduced in another, often leading to changes in the original meanings. Under systemic functional linguistics (SFL), meanings of wordings are only potential in any text, and are entirely dependent on the context in which meanings are activated. This makes textual analysis problematic, since not only are texts de-contextualised for the purposes of analysis, but their ideas are also the products of recontextualisation, sometimes with long intertextual histories. Any textual analysis, linguistic or otherwise, is thus only capable of claiming that certain meanings are potential in any text, with a three-way trajectory of intertextual resources—ontogenesis, phylogenesis and logogenesis (cf. Halliday and Matthiessen, 1999, p. 17)—taken into account. Text analysts are generally able to focus on only one of these aspects of meaning activation, whether it is the linear development of meanings and attitudes within specific texts (logogenesis), the development of meanings within languages and social fields of language practices (phylogenesis) or the development of meaning-making ‘reservoirs’ (Bernstein, 2003) in the human individual (ontogenesis—see, for example, Hasan, 1996). With this in mind, we step gingerly into the discussion of the language, texts and contexts of pro-GM technology. While the analysis of the articles selected here demonstrates that arguments regarding GMOs have not substantially changed since research conducted by Cook (2004) was published, the chapter will focus specifically on how the main actors and institutions in the texts analysed have been positioned (e.g. van Langenhove and Harré, 1993) and how

270 Alexanne Don attitudes, both negative and positive, towards these actors are invoked by repeated association of ideas of science and technology with certain actors and their actions. The discussion is illustrated by reference to the Appraisal framework (Martin and White, 2005) and with examples from a number of recent news articles with a pro-GM stance, some of which explicitly positively evaluate the development of GM food and its proponents, which act at the same time to position opponents negatively. I argue that repeated positioning of this kind contributes to the potential for invoking a negative attitude towards opponents. Under Appraisal, evaluative language is categorised under three areas: attitudinal meanings in which positive and negative evaluations of targets are expressed; engagement resources in which the relationships between speaker/writer and audience are implied; and graduation resources which are concerned with the up- or down-scaling of attitudes. The main purpose of this chapter is to show how what is termed ‘invoked’ or implicit attitudes under Appraisal are entirely dependent on their repeated use in related contexts, together with the ability of language users to retrieve the associations accruing to those wordings. My claims are not novel and derive from the Bakhtinian (e.g. 1986) perspective that all texts are written with a view both backwards to already encountered texts and forwards to potential readings by common language users. The Appraisal framework labels those wordings which are explicitly evaluative inscribed attitude, but it needs to be acknowledged that even these evaluatively stable wordings gain or lose their evaluative ‘charges’ over time, and many can also have their negative or positive values ‘flipped’ in certain co-textual environments (see e.g. White, 2016 for discussion). Many of the evaluative stances and positionings in the articles discussed in this chapter rely not on explicit wordings, but on what Maton (2014) has termed ‘axiological charging’ of specific ideas linked with wordings in specific contexts: ‘the condensing of meanings (from affective, aesthetic, ethical, political and moral stances)’ (Maton, 2014, p. 153); resulting in constellations of meaning around particular ideas. Working with Maton’s framework, Sieborger and Adendorff (2017, p. 5) explain how this type of evaluative charging is effected: ideas in a constellation [. . .] appear to be clustered together from the speaker’s viewpoint. Once they have been clustered and constellated, these ideas are charged: that is, they are assigned a particular value, as being positive, neutral or negative. Often this process works in binaries, so that a positively-charged constellation is contrasted with a negatively-charged constellation. While many of the evaluative stances in these articles are explicit, or ‘inscribed’, many position the anti-GM advocates by co-reference and association, without explicitly evaluating them, what Appraisal terms

The Value of Intertextual Associations

271

‘invoked’ attitude. While intra-text reference (e.g. meanings built up on the autonomous plane of discourse, or logogenetic meaning-making, cf. Sinclair, 1993) accounts for the reading of negative or positive evaluation of groups and individuals within discrete texts, what is of interest here are the implications that carry over in readers’ minds from one piece to the next, and enable a reader to discern whether, without conscious thought, the writer of the text is either pro-GM or anti-GM based solely on the use of certain wordings (i.e. from both an ontogenetic and phylogenetic perspective). In these texts, invoked attitude thus works by reference to intra-, inter- and extra-textual associations which co-opt the positive connotations ‘science’ enables, while also acting to subvert other anti-GMO stances as lacking scientific ‘objectivity’, and as lacking an ethically motivated attitude. The discussion is necessarily circumscribed within the domain of the pro-GM lobby and the media reports and commentary pieces used by advocates to imply that anyone proposing anti-GM policies can be grouped as negatively evaluated by their association with attitudes such as anti-science, old-fashioned, destructive, criminal, lacking in knowledge and emotional. These stances towards environmentalists are thus contrasted with their opposites, such as scientific, modern, creative, rational and able to feed the world. The Appraisal framework can be used to systematise and account for analysts’ reading of evaluative stances in texts, as it provides a network of categories of semantic and evaluative meanings that analysts can use to build up a picture of the stances towards targets that are brought into play in any stretch of text or corpus of texts. In the case study here, the motivation was to discover how newspaper texts published in Australia and New Zealand during one short period indicated their stance towards those opposing the uses of GM crops and foods. While the findings supported those of Cook (2004), the main purpose of the study was to investigate the associations made within these texts and how such associations positioned the identities mentioned and the readers themselves, through stances towards both groups of people and ideas. One of the issues with Appraisal analysis is the difficulty analysts sometimes have with invoked or implicit attitude, and selecting an appropriate category for the evaluative wordings noted in the text(s). In their seminal work on the Appraisal model, Martin and White (2005, p. 25) use the notion of a cline of instantiation to explain how wordings may gain their meanings in context by association with all instances of their usage— from the level of wording to the level of context of culture. Seen from another perspective, this means that meanings—and evaluative meanings in particular—are dependent on intertextuality in all its forms for interpretation (see Don, 2016 for further discussion). For this study, the question is more closely related to how readers are able to interpret negative or positive attitudes towards people and institutions through a repeated association of them with particular ideas and attributes. This perspective

272 Alexanne Don was of interest to Davies and Harré (1990, p. 48) in their early discussion of the notion of positioning, who stated: the significance of an initial act of single attribute positioning comes about through the association of or embeddedness of that attribute within a culturally well established cluster of attributes, called up by the positioning. The corpus used for the study comprised a collection of 22 news articles and views pieces positive towards GM technology, taken from Australian and New Zealand newspapers published between 2012 and 2013, using the online news corpus Factiva. AntConc 3.2.3 was then used to obtain a frequency list and typical collocates of frequent wordings, along with keywords for each text based on a contrasting reference corpus whose topics were not GM, science, or technology: This comprised a random collection of 20 newspaper articles on dress and identity, and ten journal articles on aspects of discourse analysis. Newspaper articles were chosen not only for their availability, but also since they form a public marketing arena for the reporting of scientific research and the production of technologies arising from their research. In this way they recontextualise the research done by science and provide the public and other authorities with a way of positioning the institution of science, the academy and the technological products, and their associated actors. In Figure 17.1, the way in which knowledge is circulated through various social fields of activity is represented for science and technological

Figure 17.1 Newspapers as a Field of Recontextualisation (after Maton, 2014)

The Value of Intertextual Associations

273

innovation around the discipline of genetic engineering. In this representation, production fields involve scientific research and the technological use of these findings to create new lifeforms and associated chemicals, while ‘recontextualisation fields’ encompass the reporting of these findings and innovations to, first, fellow scientists and biotech engineers, and, second, the general public through the subsequent digestion and reporting of such findings in popular science journals and the news media. These recontextualisation fields in turn impinge on decisions made by the public—especially in terms of government policy and farming practices— in reproduction fields where the knowledge is reproduced in trials, discussion, policy discussions and actual use. Many of the articles in the corpus focused on a specific actor or institution—as either spokesperson or target of reporting, as revealed by keyword analysis. Table 17.1 shows ranked keywords for the whole corpus in the first column, as well as keywords for seven individual articles, with actors or institutions italicised. It is not surprising that most articles have a high frequency of actors or institutions, since news articles and comment pieces use reference to such representatives to support their own arguments. It needs stating at this point that while there was no distinction made in the corpus between ‘hard news reports’ and ‘comment pieces’ (cf. White, 1997), news reports writers may evaluate explicitly only through spokespersons, but comment pieces can contain highly evaluative language and obvious positioning by the writer of the piece themselves. At the same time, authority to speak and be cited was linked to certain areas of expertise in these articles, especially news reports where such actors were positioned as important and given extra column inches. This again relates to axiological charging in what Maton (2014) describes as a hierarchy of knowers, where moral and ethical value is condensed and given value by association with specific actors. Again, unsurprisingly, Table 17.1 also shows that the reporting verb ‘said’ appears with high frequency (position 9 in the overall corpus), and that the most commonly mentioned group is farmers. As an example, in an editorial (Marlborough Express, N.Z., 2012; see Table 17.1: Doc1, position 3), one specific actor, William Rolleston, was quoted or cited in 28 out of its 30 sentences, and in one of these Rolleston also cites another ‘higher authority’, ‘OECD Studies’. Rolleston was also named several times, with labels such as Dr Rolleston, the Science and Innovation Ministry Board Chairman and a member of the Science Board and Federated Farmers vice-president William Rolleston. In this way, he is positioned to speak with authority on matters pertaining to both science and agriculture. Rolleston was also credited as speaking at ‘an international conference on agricultural biotechnology in Rotorua’ and was grouped with others who ‘attended a biotech conference’ and ‘were assigned personal bodyguards’ due to a perceived ‘real threat’ from protesters. In this way,

Corpus

gm crops food genetically australia farmers * technology modified said canola biotech crop environmental agricultural agriculture marsh billion

Rank

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

gm food rolleston science dr conference genetically zealand production modified attended said public clean environmental farmers * federated

Doc1 seralini gm rats corn cent climate pollution roundup tumours warming public air deaths death emissions per dawley

Doc2 greenpeace environmental wwf australia activists appeasement boswell certification fanatical food senator sustainability useless court activist crops csiro

Doc3

Table 17.1 First 17 Keywords of Corpus and Seven Individual Articles

gm wa canola crops facts technology farmers * genetically grown science support credible highlights modified campaign australia views

Doc4 ge rice crops genetically engineered food monsanto seeds vitamin golden protests years children blindness cent farmers * glyphosate

Doc5

food australia farmers * innovation sector dairy said engagement manufacturing profits volatility challenge industry exports helou invest premium

Doc6

australia food genetically cotton crops kruithoff modified farmers * farming monsanto sector market asia varieties local fallen investors

Doc11

The Value of Intertextual Associations

275

although protesters were not mentioned in any other capacity, such actors were thus identified as anti-GM, and by association, both violent and anti-science. Antaki and Widdicombe (1998, p. 3; emphasis in original) summarise such positioning as ‘having an identity’: for a person to ‘have an identity’—whether he or she is the person speaking, being spoken to, or being spoken about—is to be cast into a category with associated characteristics or features. Of course, it is then what Dr Rolleston said, shared, told conference attendees, agreed with, or warned that become evaluatively charged, and in the following excerpts it can be observed that he uses ‘science’ to positively charge the message that GM technology is a force for good: Sentence6: we can use science to increase our productivity and decrease our environmental footprint. SE7–8: New Zealand’s clean and green image did not sell products overseas, but food quality, safety and price did, he said. We can use science to improve these things. SE14: ‘. . . we need to have access to all the scientific tools available and GM is part of that,’ Dr Rolleston said. SE26: He also warned scientists there that New Zealand risked being left behind the rest of the world. SE27: He continued to stress the need for the Government to invest more in science and technology to improve food value and production against decreasing land use. SE28: OECD studies show there’s a link between spending on science and GDP . . . SE29: That has been recognised by world leaders around the various countries to invest in science. In this short article it can also be observed that supporters are contrasted with other groups who do not necessarily support a pro-GM food stance, and who may need educating about its positive attributes; this closely parallels the findings by Cook (2004), who discusses three main groups: those speaking, those spoken to and those spoken about. He (2004, p. 123) observes that the public are portrayed, indeed positioned by GM proponents, either as passive recipients of expert knowledge and wise decisions or as active participants through dialogue in the decision-making process.

276 Alexanne Don As an illustration of this, in another longer article (Neales, 2013a, 53 paragraphs, DocNeales1), the public are indeed portrayed as passive recipients of expert knowledge (e.g. from the President of Science and Technology Australia, and ‘we scientists’) which has not been disseminated adequately, thus leading to ‘hysteria and fear’: PAR44: Science & Technology Australia’s president Michael Holland agrees with Ludwig, saying scientists have to accept some of the blame for community hysteria and fears. PAR45: But he hopes the pendulum is swinging back—a consequence, he believes, of greater community acceptance and understanding of the benefits of other complex gene biotechnologies, such as stem cell medical therapies. In the following paragraph (PAR46) the cited authority encapsulates previous ‘community hysteria and fears’ in terms of scientists failing to recontextualise adequately—since the benefits of GM were ‘so self-evident and obvious’ to them: PAR46: ‘But on GM, we as scientists were appallingly bad at explaining the benefits of this new technology to the public because we thought it was so self-evident and obvious’, Holland says. Those spoken about, if not merely ignorant, are generally ‘opponents’ of GMOs, who ‘have become, by association, not just worriers, but ignorant and destructive opponents of progress’ (Cook, 2004, p. 18). Furthermore, as Harré, Brockmeier and Mühlhäusler (1999, p. 52) point out: The use of science as a rhetorical device presupposes an implicit contrast with the irrationality of other ways of looking at the world. (my emphasis) This means that ‘other ways of looking at the world’ are thus associated with those groups who behave ‘irrationally’ and are, in turn, associated in many of the articles with ‘protesters’, ‘activists’ and Greenpeace. Some particularly egregious, and not so implicit, examples appear in an editorial from Queensland Country Life (2012, Doc3): SE1: The sewer-like depths to which environmental activists will sink was brought into the light of Australia’s judicial system this week with judgement passed on those responsible for the wanton destruction of genetically modified crops. SE13: The message to Australia’s farming community is that negotiation and appeasement of these types of hard line, fanatical ideologues is useless. SE14: It didn’t work for Neville Chamberlain in the 1930s trying to stymie Adolf Hitler’s Nazi menace and neither would it work

The Value of Intertextual Associations

277

against the Taliban in Afghanistan, who, like Greenpeace, have no compunction in also sending women and children into the frontline. As Cook (2004, p. 108) observes: Through metaphors and comparisons, GM has become entangled with themes of warfare, terrorism, intercultural conflict and religious difference. In other articles, the evaluative charging of GM proponents and anti-GM groups is less overt, but at the same time the links are obviously made. In the previously mentioned article by Sue Neales in The Australian (2013a, DocNeales1), a word frequency analysis provides lynas, scientists, wheat, greenpeace and gates, making the main actors clear—especially if the reader is aware that Lynas was a well-known UK activist for the Soil Association, and that ‘gates’ refers to Bill Gates and his philanthropic foundation. In keeping with journalistic practice, Neales appears to aim for balance in reporting the words of spokespersons on several sides of the issue, but those she appears to favour say, agree with, admit or are convinced, while Greenpeace is the only actor to exhort: PAR11: ‘It’s the precautionary principle: that where the results of a new technology are still unknown, or where there is a lack of scientific knowledge or consensus regarding its safety, it’s smarter not to use it,’ Greenpeace exhorts. The Appraisal framework considers reporting verbs such as say, admit, tell—and exhort—under engagement: attribution. Work in this area, especially with respect to the use of claim in newspaper reports, indicates that reporting verbs can act to distance the reporter from both the quoted speaker and the quoted material (Martin and White, 2005, pp. 112–116; White, 2012), especially in co-texts where readers have been aligned with the propositions throughout the article in question. Martin and White note that many reporting verbs such as say, believe, states, tell, etc. are neutral in their dialogic effect, which the engagement framework classes as ‘acknowledge’. On the other hand, they point out, this type of acknowledgement of speakers who are not authority figures in the field or who have low social standing can invoke a negative assessment of the cited material (2005, p. 116)—in a similar way to that argued earlier where the citing of an actor with high social standing (e.g. a professor, president of a company, a recognised scientist) can imbue a statement with positive value. In the case of uncommon reporting verbs such as ‘exhort’, common contexts of use and (non)frequency of appearance in co-text, I argue, are also capable of invoking attitudes towards speakers or their cited words, and this is certainly the case with the reporting verb ‘exhort’, whose usual context appears to be religious in flavour as the following corpus queries attest.

278 Alexanne Don For example, a short survey of the lexical item ‘exhort’ using Google reveals that six of the 12 instances on each of the first three screens of results refer to biblical or evangelical contexts, an underlying intertextual association which still seems to cling to the term. Further reference to a larger corpus (Collins Wordbanks) revealed the same typical context for this reporting verb. For example, Table 17.2 shows that in the typical domains Wordbanks defines, the word ‘exhorts’ appears with the highest relative frequency in religion. Wordbanks also provides a ‘wordsketch’ for the word ‘exhorts’ showing common subjects and objects for the verb (see Table 17.3) which again demonstrates that religious and political contexts are frequent. In one sense then, we can claim that, in contexts or disciplinary fields where infrequently used, or non-neutral reporting verbs (as well as other lexical items and phrases) appear, they carry with them intertextual associations which can potentially trigger either negative or positive associations in the reader. This also applies logogenetically in the local co-text—in a text where many voices are reported through such neutral verbs as say, tell or state, and one voice is represented as delivering their opinion using a locally novel lexical item, the potential is high for foregrounding that speaker and the cited material as noteworthy in some respect. In the Neales article already cited (DocNeales1) this foregrounding, or ‘priming’ (see Hoey, 2005), is extended when, in the following paragraph, through deft anaphoric substitution, such exhorted views are associated with destructive behaviour: PAR12: It was this attitude towards GM crops that prompted two Greenpeace activists in July 2011 to climb over a fence at CSIRO’s plant research centre in Canberra and whipper-snip an entire trial plot of pioneering new wheat varieties bred using genetic engineering techniques. Table 17.2 Frequency of Use of ‘Exhorts’ in Specific Wordbanks Corpus Domains doc.domain Frequency

Rel

[%]

biog business culture fiction lifesci medicine natsci news religion to_be_sp tv_radio

5 8 5 2 7 1 1 51 14 3 3

162.7 288.2 119.3 19.4 147.2 74.4 98.3 97.7 589.7 91.4 27.0

The Value of Intertextual Associations

279

Table 17.3 Wordsketch of Subjects and Objects of ‘Exhort’ in Rank Order of Frequency Object

Subject

faithful Slovenia Pakistani believer Christian congregation electorate follower Iraqi crowd listener viewer citizen employee public audience

hoarding loudspeaker preacher slogan poster banner article editorial sign aide speech priest message executive head campaign

It also needs to be acknowledged at this point that articles in the corpus, while including both tabloid and broadsheet publications, were selected for their pro-GM stance and are likely to also support a positive economic-technological stance. The Australian newspaper from which several of the articles are drawn, although a broadsheet publication, is notably a Murdoch institution and most of its editorial policy is unsurprisingly right of centre. This study is not designed to reveal any underlying ideological stances since these are part of the criteria for selection. What is in focus is the ways in which language is used to imbue entities and actors with negative or positive values within the circumscribed recontextualisation fields in which GMOs are viewed as positive. This type of evaluative association accruing to terms in a specific field was discussed in a short study of the language of wine appreciation, focusing on the style of the reviews by a specific wine critic (see Hommerberg and Don, 2015). From another perspective, Maton (2014, p. 155) argues for the concept of ‘constellations’of meanings, in which positive lexical items at the same time imply negative value for terms which are constrasted, giving the example of ‘student-centred learning’ which is ‘invariably associated by proponents with a series of stances contrasted to “teachercentred” ones’. It is worth quoting at length what Maton (2014, p. 163) argues such constellating of meanings allows: Terms can also become used as a dog whistle: without needing to be made explicit, a position can be implied, and valorization or

280 Alexanne Don disapprobation evoked, through the use of other constellated terms. Thus, using the term ‘teaching’ (or, worse, ‘transmission’) to refer to pedagogic relations may be viewed as advocating behaviourism, positivism, authorative imposition, disempowerment of students, disengagement from learners’ experiences, and conservativism. This has implications for our identities as writers and readers, since once we read a certain phrase in the context of our field of interest, it comes entailed with its associations with other texts, making it clear that language can never be entirely neutral in context. In the following diagram (Figure 17.2), four main recurring actors or concepts—science, pro-GM, Greenpeace and anti-GM—were paired with phrases, other actors or attributes in the corpus to graphically illustrate clusterings of associations accruing to each. In this set of clusters, in many cases a direct link was noted, e.g. ‘Greenpeace’ with ‘exhort’. In other cases, the implication of an action was summarised, e.g. ‘Greenpeace’ with ‘destructive’. With an Appraisal analysis, such invoked attitude is not summarised in this way—instead the invocation of an attitude is categorised. In this case, the actions of the institution or group ‘Greenpeace’ would have been classed as negative judgement: propriety, since destructive behaviour incorporating whipper-snipping ‘an entire plot of pioneering new wheat varieties’ would be judged negatively under social sanction (cf. Martin and White, 2005, p. 52), and indeed did involve litigation in the specific instance reported. For the same reason, although the term ‘anti-GM’ was not common in the corpus, attitudes and actions were implied to be anti-GM. For example, referring again to DocNeales1, PAR12, Greenpeace activists are reported to have destroyed a plot of wheat varieties that had been ‘bred using genetic engineering techniques’, thus implying their negative stance towards GM technology. This type of diagram (called a ‘force-directed graph’) visually illustrates the contrasts in stances towards these ideas in the corpus. Those nodes in the diagram which appear larger are also those which have been repeated—the more instances, the bigger the nodes—leading to the representation of constellations around the main ideas, e.g. pro- and anti-GM. The diagram makes it obvious that both science and pro-GM stances are strongly linked, while Greenpeace is not associated with science through any link. Instead, pro-GM stances are linked with lexical items and ideas which are generally given an inscribed positive attitude under the Appraisal framework: revolutionary, essential, productive, future, pioneering, innovative, benefits, choice, research; while anti-GM stances are linked with terms or ideas such as destructive, ignorant, outdated attitudes, emotional, oppose change, activists and improper science. One area of interest for me here is the linking of activists with these negative attributes, while also being associated with Greenpeace. Putting aside the inscribed negative terms explicitly linked with those activists

Figure 17.2 Clusters of Associations of Four Main Signifiers in the Corpus

282 Alexanne Don who were part of the Greenpeace protests described in the article cited earlier, this association of activists with the cluster of attributes and terms on the right side of the diagram makes it clear that activism and science are opposed in the pro-GM world. Mention of activists and activism is thus liable to evoke negative associations in the context of articles on GM technology. In the small corpus, there were also other institutions mentioned in association with anti-GM stances, such as the WWF (World Wildlife Fund) and NGO (non-government organisation), which also brings these groups under suspicion, since in at least one instance NGOs were protrayed as being against farmers and involved in cash-producing schemes, with WWF in particular ensnaring agriculture. On the other hand, farmers and the CSIRO (Commonwealth Scientific and Industrial Research Organisation) are strongly linked to science, with the CSIRO credited with providing evidence, research and authority. It’s therefore likely that articles citing such groups will at the same time act to evoke association with negatively charged positions, through such intertextual relations. In Don 2016, I argued that all lexical items were inevitably ‘meaningful’ and evaluatively charged as a product of their use—their phylogenetic history—activated in context, i.e. specific instances of interaction with readers or hearers. Readers, in turn, engage with the evaluative charging through their own ontogenetic resources (life experiences). At the more evoked or implicit end of the spectrum (cf. Don, 2016), it is often difficult to pinpoint the means by which phrases in context come to have the evaluative charges they seem to carry, and my purpose here has been to propose that repeated use of certain attributes in relation to certain actors tends to make attitudinal readings more likely in similar contexts. In this chapter my concern has been to suggest that evaluative meanings are not inherent in lexical items themselves, but are largely dependent on context and the associations each level of context carries. We may refer to the context of the text itself and the logogenesis of meanings within that text to pinpoint what potential meanings a word or phrase may evoke in compliant readers’ minds, or we may refer to readers themselves and their likely values, knowledge and identity associations as a means of accounting for potential readings of a text within a circumscribed group or discipline. This in turn must take account of the phylogenetic resources of each language-using group, and the texts which instantiate that language. With this in mind, advocating for the increased use of corpora in seeking potential associations between contexts and wordings in texts seems more than justified. Toolan (1992, p. 162) refers to ‘the principle of “operationalizability” which always underlies stylistics’ where, in ‘their contextualized reading of a text, stylisticians seek out accounts of linguistic form, of the

The Value of Intertextual Associations

283

lexicogrammar of texts in general, which may serve as useful diagnostic instruments in uncovering patterns, tendencies, textual omissions, or silences or foregroundings (in short, style)’. If we are to uncover such patterns and tendencies, together with likely contextualised readings, then the larger the sample we can use to support our own readings, the better we will be able to provide such accounts of textual styles. In addition, we need to attend to matters of systematic linguistic analysis if we are to make claims as to the ‘usefulness of our diagnostic instruments’, and my hope is that this chapter has provided a small example of how this might be achieved.

References Antaki, C. and Widdicombe, S. (eds.) (1998) Identities in talk. London, Thousand Oaks and New Delhi: Sage. Bakhtin, M. M. (1986) Speech genres and other late essays. Trans. V. W. McGee. Austin: University of Texas Press. Bernstein, B. B. (2003) Class, codes and control volume IV: The structuring of pedagogic discourse. London: Routledge. Cook, G. (2004) Genetically modified language: The discourse of arguments for GM crops and food. London: Routledge. [Kindle Edition]. Davies, B. and Harré, R. (1990) ‘Positioning: The discursive positioning of selves’, Journal for the Theory of Social Behaviour, 20 (1), pp. 43–63. Don, A. (2016) ‘“It’s hard to mesh all this”: Invoking attitude, persona and argument organisation’, Functional Linguistics, 3 (9), pp. 1–26. Halliday, M.A. K. and Christian M. I. M. Matthiessen. (1999) Construing experience through meaning: A language-based approach to cognition. London and New York: Continuum. Harré, R., Brockmeier, J. and Peter Mühlhäusler. (1999) Greenspeak: A study of environmental discourse. Thousand Oaks, CA, London and New Delhi: Sage. Hasan, R. (1996) Ways of saying: Ways of meaning. London and New York: Cassell. Hoey, M. (2005) Lexical priming: A new theory of words and language. London: Routledge. Hommerberg, C. and Don, A. (2015) ‘Appraisal and the language of wine appreciation: A critical discussion of the potential of the Appraisal framework as a tool to analyse specialised genres’, Functions of Language, 2 (22), pp. 161–191. Martin, J. R. and White, P. R. R. (2005) The language of evaluation: Appraisal in English. London: Palgrave Macmillan. Maton, K. (2014) Knowledge and knowers: Towards a realist sociology of education. London: Routledge. [Kindle Edition]. Siebörger, I. and Adendorff, R. D. (2017) ‘We’re talking about semantics here: Axiological condensation in the South African parliament’, Functions of Language, 24 (2), pp. 196–233. Sinclair, J. McH. (1993) Written discourse structure’, in Sinclair, J. McH., Hoey, M. and Fox, G. (eds.) Techniques of description. London: Routledge, pp. 6–31.

284 Alexanne Don Toolan, M. (ed.) (1992) Language, text and context: Essays in stylistics. London: Routledge. [Kindle Edition]. Van Langenhove, L. and Harré, R. (1993) ‘Positioning in scientific discourse’, in Harré, R. (ed.) Reason and rhetoric: Anglo-Ukrainian studies in the analysis of scientific discourse. Lewiston, NJ: The Edwin Meller Press, pp. 1–20. White, P. R. R. (1997) ‘Telling media tales: The news story as rhetoric’, Unpublished PhD Thesis. Sydney University. White, P. R. R. (2012) ‘Exploring the axiological workings of “reporter voice” news stories: Attribution and attitudinal positioning’, Discourse, Context & Media, Elsevier, 1 (2–3), pp. 57–67. White, P. R. R. (2016) ‘Evaluative contents in verbal communication’, in Rocci, A. and de Saussure. L. (eds.) Verbal communication. Berlin and Boston, MA: Walter de Gruyter Mouton, pp. 77–96.

List of Articles Used Acknowledgement: all articles retrieved from Factiva (ProQuest LLC: www. proquest.com) via UNSW Library Services ‘Appeasement is useless in the face of fanatical activist’, (2012) Queensland Country Life, 22 November. Connor, S. (2013) ‘Europe’s GM stance denies Africa the right to feed itself, warns leading academic’, The Independent, 3 June. Dyer, G. (2013a) ‘GE crops our only solution’, The Nelson Mail, 8 April. Dyer, G. (2013b). ‘“Golden rice” hailed as aid to preventing blindness . . . ‘, The Daily Post, 17 April. Dyer, G. (2013c). ‘Rice and shine: GE crops are the answer’, Taranaki Daily News, 9 April. Flint, N. (2013) ‘Anti-GM attitudes are harming the hungry’, Sydney Morning Herald, 3 July. GM. (2012) ‘Key to lifting food production’, The Marlborough Express, 3 October. ‘GM demands clear facts’, (2013) Donnybrook—Bridgetown Mail, 4 March. Godwin, S. (2013) ‘Man of the Mallee’, The Weekly Times, 15 May. Kitney, D. (2013) ‘Innovation to feed farm exports new approach needed to grasp opportunity’, The Australian, 19 April. Lomborg, B. (2012) ‘Scare tactics tarnish scientific debate’, The Australian, 17 October. Neales, S. (2013a). ‘An inconvenient truth’, The Australian, 18 January. Neales, S. (2013b). ‘Sowing seeds of discord—global food forum’, The Australian, 19 April. no author. (2013a) ‘GM’s seeds of disquiet’, Waikato Times, 4 May. no author. (2013b) ‘FOOD shortages and soaring prices might be big concerns’, Queensland Country Life, 30 May. no author. (2013c) ‘Media has responsibility to clear air about GM fears’, Stock Journal, 4 April. no author. (2013d) ‘When backing global biotechnology crop research’, The North Queensland Register, 6 June. [NoTitle] (2013) The Courier, 6 April.

The Value of Intertextual Associations

285

Sampson, A. (2013) ‘View of genomic future is rosy’, The Weekly Times, 5 June. Smyth, B. (2013). ‘Consider bigger picture on GM crops’, Bega District News, 27 May. Toohey, P. (2013). ‘Our great northern frontier’, The Advertiser, 1 April. Urban, R. (2013) ‘Farmers not Asia-ready: Monsanto boss—EXCLUSIVE—’, The Australian, 15 April.

18 Public Women Power, Gender and Semiotic Representations Carmen Rosa Caldas-Coulthard

What distinguishes man from woman is his access to representation, to cultural symbolization, the power of naming, in which he uses women, along with other silent animals, as symbols, as objects for representation. (Susanne Kappeler, 1986, p. 81)

1. Introduction Notions of power in society are inseparable from questions of gender and sexuality. Although there has been some improvement in the ways women in positions of power are described in post-modernist societies, their representation in media discourses is often very dismissive, especially in the Brazilian context. The semiotic resources used to represent them tend to rely on mythical discourses of motherhood, domesticity, beauty and youth and not on the functional roles they perform. Their professional contribution is therefore frequently undervalued. In this chapter, I intend to discuss how women in positions of power are described and categorised in public narratives with particular focus on the Brazilian press.1 My primary data consists of current media (including newspapers, magazines and web materials). By engaging with journalistic discourse, my focus is on the importance of semiotic resources in generating epistemological claims about women and femininity. Stemming from a feminist perspective, I have drawn on the concept of critical discourse and narrative analysis for my analysis following authors such as Toolan (1988), Fairclough (1992, 2003, 2010), Hall (1997), Caldas-Coulthard and Coulthard (1996), Caldas-Coulthard and Moon, R. (2010) and Wodak (2000). I use multimodal tools of analysis from Kress and Van Leeuwen (2001), Kress (2010), Van Leeuwen (1996, 2005, 2008), Machin (2007) and Machin et al. (2016), since I believe that gendered identities are mainly constructed through multiple semiotic modes. My overriding interest is in the ways in which texts and images produce evaluative effects, which subtly condition behaviour and attitudes in relation to powerful women. I will demonstrate how female professionals emerge as gendered subjects and how their identity is construed not according to their function in the workforce, but mainly through their domestic roles and their

Public Women

287

physical appearance. My main aim is therefore to make visible the ways by which sexist ideologies are materialised in semiotic resources that produce unacceptable stereotypes in relation to women in power, in the hope that this will inspire closer scrutiny of the media content, leading to an informed critique and transformation.

2. The Question of Gendering My first hypothesis, when examining the question of how powerful women are represented in media discourses, is that there are processes of gendering that differentiate powerful women from powerful men semiotically. I define gender: . . . along two key dimensions. First, social relationships (and representations) are infused with assumed differences between the sexes. These assumptions are ideological and can be shown to be constructed in language and other forms of representation and in social practices in institutions. . . . Second, gender imbalance never operates alone but intersects with other axes of inequality. (Machin et al., 2016, p. 306) Gendering therefore refers to discourses oriented by the ideational function proposed by Halliday (1978, 1985) or how we codify the world. We use semiotic systems, language being the main one, to organise, understand and express our perceptions of reality. Identities are construed therefore according to the semiotic resources people use to interact with each other. In the representation of powerful women, gendered discourses are used to communicate ideas and practices, which emphasise professional women’s gender to the detriment of other aspects of their identity, functions and roles in society. Because sexual roles are given prominence, women and men are positioned differently in public discourses although they might be performing the same actions. Female identity representations are unfavourable to women since their main focus is on appearance and sexuality (Mills, 1995). My second hypothesis is that texts and images produce negative evaluative effects that influence attitudes in relation to powerful women.

3. Some Facts It is interesting to consider some worldwide statistics in terms of numbers of women in positions of power, in areas such as government, politics and business. According to the Pew Research Centre, based on the World Economic Forum, in 2017 there were 15 female world leaders currently in office, eight of whom were their country’s first woman in power (10% in terms of percentage in relation to male leaders). While the number of current

288

Carmen Rosa Caldas-Coulthard female leaders—excluding monarchs and figurehead leaders—has more than doubled since 2000, these women still represent fewer than 10% of 193 UN member states. (Geiger and Kent, 2017)

The world average of women in National Parliaments is only 23.5% according to International Parliamentary Union data in December 2017 (Women in National Parliaments, 2018). This under-representation is generalised and very few countries, with the exception of some Scandinavian countries, have gender parity. In business, according to the American Financial Magazine Fortune, the representativeness of women in CEO positions has fallen worldwide. Overall, women continue to be under-represented in the top role at Fortune 100 companies. Last year, 10 such firms appointed new CEOs. Only one was a woman. That makes a total of only seven female CEOs among the 100 largest U.S. companies based on revenue. (Bloomberg, 2018) At the beginning of 2018, pay disparity between men and women existed in all realms of economic life, even in the ‘Gig economy’: Female Uber drivers make 7% less per hour than their male counterparts—even though the algorithms that determine pay for the ride-hailing service are gender blind, according to a multi-year study. (Korosec, 2018) The lower level of female salaries is an issue that professional women are continuing to struggle against. Globally speaking, therefore, the number of women in top jobs is still very small and female leadership continues to be a very problematic issue. According to recent ‘More men named John’ memes and images that circulated on sites like Skillshare International (Charity Organization at Skillshare), there are more CEOs in major American firms called ‘John’ than women CEOs!

4. The Brazilian Situation In Brazil, according to the organisation More Women in Power (Mais mulheres no poder, Plataforma, 2016), women are 51% of the population and 52% of the electorate. And according to the National Ministry for Women’s Affairs, in 2014, women also had a much higher educational level and were more economically active than men (Politicas para Mulheres, 2014). Nevertheless, Brazil, occupies one of the lowest positions in the Interparliamentary Union ranking (Women in National Parliament,

Public Women

289

2018), with a mere 9% of women in the Lower and Upper Houses of the Legislature Government. In the executive branch of government,2 female under-representation is a given fact—from city and state councils to federal government. Although Brazil recently had a woman president elected twice, Dilma Roussef (2010, 2014—and latterly impeached for corruption allegations), the number of female candidates for executive posts is very small, compared to other countries in Latin America, such as Chile and Argentina. The most recent presidential cabinet has no women ministers. In the judiciary, data from the Association of Brazilian Judges (Associação dos Magistrados Brasileiros—AMB, April 2010) shows that in the base of social power, 50.5% of solicitors in Brazil are women, but as one moves up the hierarchy, the proportion goes down. In the Supreme Court, for example, there are three women out of a court of 11 judges. We can see therefore, based on the information given in terms of numbers and percentages, that there is a culture of sexual division of work, prejudice and inferiority that makes it difficult for women to have autonomy and presence in crucial decisions in the political, legislative and judicial arenas. These facts also demonstrate the difficulties women in executive and legislative decision-making domains have to disrupt conservative conceptions that associate and reduce their field of work to gendered stereotypes of domestication and the private. Consequently, women’s insertion in areas of work (administrative, economic and technological) associated with the public and the political, culturally reserved to men, is more difficult to be attained. And the maintenance of governmental, legislative and judiciary gender inequality is an unequivocal fact. These distortions make Brazilian democracy not as representative and participatory as are other segments of society. The figures presented so far were a starting point for my research into the representation of women in power in the press. And the initial questions were: What are the discursive resources used by journalists to describe Brazilian women in positions of power? What are the ideological implications of these representations?

5. Discursive Representation Van Leeuwen (1996) suggests, in his discussion of how social practices are transformed into discourses, that there is an ‘array of choices’ or a system network, from which we choose to categorise or name other social actors and represent them in texts. To start with, people can be included in or excluded from discourses. Van Leeuwen distinguishes three major forms of ‘categorization’. One is ‘functionalization’, defining people’s identity ‘in terms of an activity, in terms of something [people] do, for instance an occupation or role’ (Van Leeuwen, 1996, p. 54).

290

Carmen Rosa Caldas-Coulthard

A second is ‘identification’, categorising people ‘not in terms of what they do, but in terms of what they, more or less permanently, or unavoidably, are’ (ibid.). He then distinguishes three types of ‘identification’: ‘classification’, ‘relational identification’ and ‘physical identification’. In the case of ‘classification’, people are defined ‘in terms of the major categories by means of which a given society or institution differentiates between classes of people’ (ibid.)—in other words, what we are: our age, gender, provenance, class, ethnicity, sexual orientation and so on. ‘Relational identification’ categorises people’s relations to each other (e.g. kinship, work, friendship and general social connections). Finally, ‘physical identification’ has to do with physical characteristics. The last category, appraisement, has to do with the ways social actors are evaluated in discourse. Table 18.1 Categorisation Framework Following Van Leeuwen (1996) Functionalisation Identification

Classification Relational Physical Personal

Appraisement

Occupation, role, function in society Age, gender, provenance, race, ethnicity, sexuality, class, wealth, religion, politics. Kinship, work relationship, personal relationship Size, colouring, appearance, clothing, etc. Attractiveness Emotional state, behavioural traits, intellect, morality, etc. General evaluatives and affectives

Applying this framework to my data proved very fruitful. The following examples come from an investigation of verbs of saying (Caldas-Coulthard, 1995), where I wanted to investigate the differential representation of women and men ‘speaking’ in the press. Using the collocational profiling tool of the Bank of English Corpora3 I made use of concordance lists of verbs of ‘saying’ from a five-million-word corpus of The Times and from a 20-million-word corpus of the BBC World Service. While pursuing this investigation, coincidently I discovered how men and women had different access to voice (men are represented as speaking more in the press than women) and were also categorised in different ways. I found a cline of modification ranging from the personal name of the speaker through the simple terms of address (Mr, Mrs, Miss or a title—Dr, Lord) to highly complex nominal groups. The following examples (from simple to complex nominal groups) illustrate this point. Examples of male categorisation: • • •

Mr Maxwell Dr Bartell Lord Deborough

Public Women • • • • • • • • • • •

291

Jades Camel, the broker Keith Wafter, medical director of Cilag Mr Paul Davie, economist Mr Michael J. Fuchs, the chairman of Warner Mr David Hill, chairman and managing director of the Australian Broadcasting Corp Mr Michael Portillo, the Secretary of State for Employment Prof. Patrick Minford, of the University of Liverpool, a monetarist and supply side economist Denis Giffod, the founder of ACE (the Association for Comics Enthusiasts) and owner Dr Jan Pentreath, chief scientist of the government authority Sir Charles Tidbury, former chairman of Whitbread brewers The prominent conservative activist Paul Weyrich, Clinton, front runner for the Democratic presidential nomination, the Arkansas Governor

Women, on the other hand, are categorised and identified differently: • • • • • • • • • • • • • •

Jane Grigson Mrs Reagan Miss Hilary Campbell, of Edinburgh 23-year-old Nicole Stewart His grandmother, Mrs Barbara Wilkinson Mrs Frances McDaid, his mother Ursula Vaughan Williams, widow of the composer Ralph Vaughan Williams Tricia Howard, 48, the woman with whom the Liberal Democrat leader dallied in 1986 Mrs Clasper, a mother of two and part time charity worker Hillary, Mr Clinton’s politically attuned wife of the front-running Democratic presidential contender, Bill Clinton The 18-year-old Miss Black America beauty pageant contestant Miss Ann Widdecombe, Conservative MP Sara Keays, the colonel’s daughter who once hoped to marry the then Conservative Party chairman Cecil Parkinson and become an MP Miss Keays, aged 44, left with an epileptic eight-year-old daughter

Male speakers tend to be categorised as speakers by their professional designations and titulation, or by their position in the government or in some kind of public institution, in other words, according to their function in society (Dr, Lord, Sir, the broker, the medical director, the economist, the chairman, the chief scientist, etc.). Women speakers, by contrast, are ‘identified’ according to names, their social relations (mother, widow, wife, grandmother), age and provenance or social relationships (‘the

292

Carmen Rosa Caldas-Coulthard

woman with whom the Liberal Democrat leader dallied in 1986’, for example). In these examples, even the women in the most known positions of power did not have their honorifics or titles (the lawyer, Dr, etc.) mentioned. Hillary Clinton is an obvious case—she is identified only as a wife—or Ann Widdecombe, introduced as a ‘Miss’, not even Ms, although she is described as a ‘conservative MP’. Women therefore, according to the corpus examples (we have to remember, however, that the concordances give us only a glimpse of the text, not the whole text), are categorised as private ‘characters’ in relation to their male partners as wives, girlfriends, sisters, widowers, etc. The dependency on a male partner seems to be a constant reinstatement of their dependent roles in society. Women in powerful positions are represented in very similar ways in the press texts I examined. As Bulawka (2012) well demonstrates in her study of female politicians in the Polish press, identity types centred discursively in different domestic behavioural properties, corresponding to dominant cultural readings of femininity. Some of these readings refer to archetypal models, which construe specific female identity positions (the mother, the witch, the seductress, the princess—discussed later). Baxter (2017, p. 24), like Bulawka, in her research on stereotypical newspapers’ construction of women leaders, quotes Kanter (1993) who suggests that women leaders are ‘tokenised’ in male-dominated organisations because ‘they pose a threat to the dominant order, they are forced into gendered subject positions or “role traps”—“the iron maiden”, “the seductress”, “the mother” and “the pet”’ (Baxter, 2017). These, like Bulawka’s roles, are also based on historical archetypes of women powerful positions. An important theoretical issue for my analysis is the question of semiotic evaluation, which is the attaching of values to people, things and actions, which can be ‘positive’ or ‘negative’, ‘good’ or ‘bad’. ‘Evaluation pervades much more deeply into the way human beings orient themselves to the world around them shaping and giving meaning to life’ (Kiernan, 2008, p. 112). As Volosinov (1986, p. 10) suggests, semiotic systems are always systems of evaluation. For him, every sign is subject to the criteria of ideological evaluation. . . . The domain of ideology coincides with the domain of signs. They equate with one another. Wherever a sign is present, ideology is present, too. Everything ideological possesses semiotic value. Martin and White (2005) in their detailed description of the systems of appraisal and attitude give the label ‘evaluative disposition or stance’ to the ways we access the world around us. One of the main functions of evaluative language is attitudinal, in other words, ‘we use language to assign values of “praising” and “blaming,” with meanings by which

Public Women

293

writers/speakers indicate either a positive or negative assessment of people, places, things, happenings and states of affairs’ (White, 2001). For Martin and White (2005), there are three types of attitude: affect, which is evaluation which indicates emotional states of affairs (likes and dislikes, for example—I love women politicians); judgement, which is the assessment of human behaviour with reference to rules, social acceptability, systems of ethics, legality, etiquette and social norms (women politicians talk too much); and finally appreciation, which is the evaluation of human artefacts as well as human individuals (women politicians look terrific). My claim is that most semiotic systems (images, written and oral language) create evaluation and in doing so become a resource for the construction of identities. And it is through the system of evaluation, especially through judgements and aesthetic appreciations, that powerful women are represented, most of the time negatively. Together with linguistic evaluations, as Bulawka (2012) notes, at the discursive level, the symbolic adaptation and recodification of fairy tales offer multiple benefits for producers and consumers of media texts. Therefore, the framing of public women in terms of categories derived from fairy tales is a ‘hanger’ for media narratives (Toolan, 1988), which then promotes familiar contexts with new participants. These participants are then legitimised as feminine dominant types and evaluated as such. Therefore, the princess, the witch, the seductress and the mother are very frequent in the discourse of the press, suggests Bulawka (2012, p. 213). She also points out that as the arrival of women on the government scene is a relatively recent phenomenon, the evocation of the well-known identity models allows the creators of the political news to maintain an ‘interpretative control’ of the somewhat novel circumstances that gave rise to female professional authority. The canonical image of belligerent women gets a mythical signification as a symbol of women who have the supreme power. They are ugly, old, demonic or seductive. They are evil constructors. They signify ambition and revenge, and in the professional world, especially in politics, they are threatening. For Van Leeuwen (2008, p. 111), comparisons in discourse serve predominantly a (de)legitimating role by triggering moral evaluations. As well as invoking intertextual analogies, the appraised content in the different depictions relies also on ‘discursive strategies of storytelling and symbolization—the overdetermination of social actors by associating them with fictional characters or practices’ (ibid., p. 117). Some recent linguistic examples follow from the international media which delegitimise important professional women.

294

Carmen Rosa Caldas-Coulthard

A reference to Julia Klöckner and Annegret Kramp-Karrenbauer, as possible successors for Angela Merkel, in the German Government: ‘Crown princess’ or ‘mini-Merkel’? The fresh faces of German politics who could replace Angela Merkel (Huggler, 2018) Christine Lagarde, the managing director of the International Monetary Fund: Profile: Christine Lagarde, ‘rock star’ head of the IMF Synchronised swimmer Poised, chic and a fluent English speaker, Ms Lagarde, 60, is the first woman to head the IMF. But then, throughout her career, she has become familiar with the ‘first woman to . . .’ tag. (BBC, 2016) Angela Merker, prime minister of Germany: Angela Merkel, Germany’s mother. The German chancellor has somehow become a ‘Mutti’ to the nation. Victory beckons again. (Knight, 2013) Mary Beard, the British historian: Mary Beard’s hair gets people going. Even her admirers refer to her ‘grey hair, wild as a witch’. (Grase, 2013) Australian elections 2013: Tony Abbott says female candidate has sex appeal. Would-be MP Fiona Scott is young, feisty and has ‘a bit of sex appeal’, according to the opposition leader. (Jabour, 2013)

6. Visual Representation As in language, images have the potential to transmit an evaluative stance to represented actors and add negative or positive meanings to the semiotic constructions.4 Through multimodal resources (Kress and Van Leeuwen, 2006; Machin, 2007) in photos, colour, distance between the represented participants and viewers, focus, vectors and size, an imagemaker can make us believe that what we see is reality and therefore true.

Public Women

295

According to Hall (1968, pp. 83–108), however, images show particular events, people, places and things and they have a point of view. They ‘document’, or in semiotic terminology, they denote. Other images, however, depict people, places, things and events to get general or abstract ideas across. They connote ideas and concepts. When we communicate through the spoken or written modes, we demand or offer information or goods and services (Halliday, 1985, p. 69). In very similar ways, the exchange of information also happens through the visual mode and through particular semiotic resources. Kress and Van Leeuwen (2006), based on the Hallidayan dialogic theory, developed a system to describe what happens in visual dialogue. Through gaze (Van Leeuwen, 2005, p. 120) and vectors (direction of the gaze), represented participants offer to and demand from viewers. If the represented participant looks at viewers, s/he is demanding some action or reaction from them (therefore called ‘demand’ pictures). If the represented participant does not look directly at the viewer, the producer of the visual sign is simply giving the viewer information about some state of affairs (therefore called ‘offer’ pictures). Representations of beautiful people, smiling, placed in interesting and attractive contexts tend to draw us to the represented participants and our reaction is typically positive. If physical features are distorted or people are placed in odd contexts, this can produce the opposite effect. Both positiveness and negativeness are always construed through semiotic resources which produce reactional effects from viewers, as we will see. According to Machin (2007, p. 25) asking what an image connotes is asking: what ideas and values are communicated through what is represented, and through the way in which it is represented? Or, from the point of view of the image-maker: How do I get general or abstract ideas across? How do I get across what events, places and things mean? What concrete signifier can I use to get a particular abstract idea across? For Barthes (1977, p. 23), connotation is realised through the choice of poses; objects (inducers of ideas—book case equals an intellectual person); settings (landscapes, river, mountains, the sea, etc. used as metaphors for changes in life, passing of time, peace); ‘photogenia’ (framing, distance, focus, illumination). And, of course, the same object or pose can have different connotations for different people. A woman in a short skirt can be viewed as ‘modern’ in Western cultures, for example, but could be arrested for immorality in the Middle East. Visual choices, nevertheless, like their linguistic counterparts, are not neutral but always presented through an ideological point of view being thus always ideologically motivated. The other important point is that images, as Machin (2004) suggests, do not record reality, but put forward ideas about certain people

296

Carmen Rosa Caldas-Coulthard

or types and ‘convey particular kinds of scripts, values and identities’ (p. 781). They are a powerful heuristic tool. Using the tools of Kress and Van Leeuwen’s visual grammar (2006), I examined pictures of female Brazilian politicians, and found that the same lexical derogatory meanings are found in their visual representation. A photo of Graça Foster—ex-president of the largest oil company in Brazil, Petrobras—which was published in a blog post by Litoralmania (Brasil, 2015), is a picture in which Foster is presented with a downturned mouth and sucking in her top lip whilst frowning (see Figure 18.1). The ‘offer’ picture on this affords negative interpretations. Her distorted mouth connotes archetypal representations of toughness, even aggression. The viewer, not interacting with her gaze, reacts and evaluates the company director therefore as a scary, ugly witch, or as Kanter (1993) suggests, an ‘iron maiden’. Interestingly, in a cartoon published on Twitter by the account Political Cartoon (2018), the cartoonist of the British newspaper The Times represented Ms May, the prime minister, as a witch-like figure incongruously, wearing a ‘little black dress’ trying to ‘please’ President Trump in Davos (2018)! In fact, this witch-like representation of Ms May is not atypical and can be found in many British cartoons published on this Twitter account. In the case of this particular cartoon, Ms May’s exaggerated nose is intended to signify ‘witchery’ while her low cut ‘black dress’ connotes sexuality. A speech balloon confirms the sexual allusion (‘time for a quickie’) and undermines the prime minister even more. Among the male figures also in the image, the witch sexual figure has just a little bit of Trump’s attention, so the roles of witch and seductress conflate here.

Figure 18.1 CEO of Petrobras

Public Women

297

My next example is of three main Brazilian ministers of state who are represented in a cartoon on a blog post by the Brazilian quality newspaper Folha de Sao Paulo (2011) as the characters ‘Super Hero Girls’ and are named as such—including the word ‘girl’ in the headline ‘Dilma exchanges three Little Pigs for Super Hero girls’. The image (which is currently available at the newspaper’s blog post listed in the references) shows the three politicians caricatured, with snouts instead of noses and presented with diminutive bodies and large eyes, similar to many Disneystyle characters, but in a decidedly unflattering light. The text in the cartoon continues, ‘The Super powerful ministers in charge of the political articulation programmes, Belchior, Gleisi and Ideli dominate the government’ (Folha de Sao Paulo, 2011). It is worth noting that apart from the first reference, only the first names are used to introduce the ministers (their names are Miriam Belchior, Gleisi Hoffmann and Ideli Salvati), and with no titulation or honorifics, although all of them were also senators and very important ministers. Another ironic text, which refers to the same powerful ministers/ senators, corroborates the fictional representation of cartoons and fairy tale stories. Women’s Republic ‘I think I will have to plait my hair’ [reference to Rapunzel] said minister Gilberto Carvalho, the last male of the species with some power in the Government. He was in the first work meeting after the arrival of the ministers Gleisi Hoffmann (Home Office) and Ideli Salvati (Institutional Relations).5 (Revista Época, 2011) This quote not only diminishes the only three female ministers in the government at that time, alluding to Rapunzel’s plaits and her decorative but powerless state, but also suggests that they too are powerless and simply decorative, like Rapunzel. And the titles of both quotes—‘The Super Powerful girls’ and ‘Women’s Republic’ are extremely ironic, because the ministers are constructed not only as powerful in their position but also as immature and therefore incompetent (reference to childhood). They are also delegitimised by their body representation—they are not real, but overdetermined cartoonish bodies. In fact, the real force of these identifications emanates from high degrees of concentration of what Barthes calls ‘text and image connotaters’ since they are loaded with cultural meanings (Barthes, 1977, p. 50). Connotaters are ‘general, global, and diffuse; it is, if you like, a fragment of ideology’ (ibid., p. 9 1). In this way, the archetype behaves as a stereotype and becomes naturalised. In news coverage of the Brazilian politician and presidential candidate, photographs of the ex president Dilma Rousseff circulated in 2010 which

298

Carmen Rosa Caldas-Coulthard

showed her holding her newborn grandchild (e.g. see Sofia News Agency, 2010, and also Folha de São Paulo). Here, the president of the Republic is represented as ‘the mother’, a constant reference for Merkel too and other women leaders, as pointed out earlier. The role of the mother is one of the most socially accepted symbols of femininity, but also serves as a model of authority. The transference from the role of mother to the professional or political spheres produces, however, effects of disempowerment, since there is a constructed incompatibility between the roles (a woman cannot be a housewife and a president at the same time). Finally, let us consider the seductress. Beauty and seduction have also been constant criteria to justify the selection of women appearing in the news. To be sexy is one of the most semantically inflated and stereotypical images of femininity. Women classified as seductress are evaluated mainly in terms of physical looks and recognised by their capacity to seduce men (Baxter, 2010, p. 33). The images of sexualised bodies, fragmented and as ‘offer’ images (interpellating the viewer), are very common in the Brazilian press, where sexuality is linked to youth as a cultural capital (CaldasCoulthard, 2010; Caldas-Coulthard and Moon, 2016). A notable example of this is the Brazilian lawyer Denise Rocha who in the media coverage has repeatedly been reduced to a sexual persona. For example, in the image which accompanied the Noticias blog post which dubbed her a hurricane (2016) a still of an intimate video in which she featured was reproduced. In the image, Rocha wears a close fitting T-shirt and denim ‘hot pants’, with long blonde hair falling over her shoulder as she looks directly into the camera whilst posing with gestures that draw attention to her body in a sexualised manner. Beautiful women have always been part of the news everywhere, but in the Brazilian context this is particularly enhanced. In this example, the lawyer and ex parliamentary adviser Denise Rocha became known (derogatively) as the Parliamentary Investigative Commission’s ‘hurricane’ after the leaking of an intimate video with her boyfriend. In this ‘demand’ picture, her direct gaze interpellates the viewer and constructs her as seductive. The choice of her pose and photogenia (her naked legs, her clothes, her stereotypical blonde hair) and the framing of the body induce ideas (Barthes, 1977) of sexuality. In this representation, therefore, the lawyer is not news because of her intellectual capacities, but simply because she is a seductress. And this, of course, clearly demonstrates gendering at work. Through sexualised images, the underlying ideology is that female professionals break social norms and should be condemned. To sum up the analysis, in visual representations in news as narrative texts, powerful women tend to be represented as one of a small number of stereotypical types, which are demeaning. Because they are women, they are not afforded the importance that would be given to a male in their role. Although powerful men can be represented in the press as fathers (for example, Barack Obama was constantly photographed with

Public Women

299

his daughters), in general there is not counter representations for a male ‘witch’ or a ‘grandfather’ and, with the exception of Putin, powerful men do not appear in the press as sexualised as women do.

7. Final Remarks By attending to the gendered dimension of political representations I hope to have demonstrated how the many linguistic and semiotic mechanisms used by journalists encode the acquired notion of female difference. In my analysis, female professionals emerge as gendered subjects in the sense that their identity can be construed not according to their function in the workforce, but according to their domestic roles or their physical appearance or sexual attractiveness. Women are frequently judged by naturalised stereotypes that place them in private contexts in opposition to the public contexts where they perform professionally. The main linguistic and semiotic categories used by journalists to represent powerful women point to family relations and physical identification. The intertextual constructions of women linked to classic female narrative roles (the mother, the witch, the princess, the seductress) realised by linguistic or visual categorisation and identification point to private contexts instead of the public spheres where powerful women act. The semiotic rendering therefore inscribes powerful women in traditional concepts of femininity. Women in this sense are construed linguistically and semiotically as a separate category, which downgrades them in terms of hierarchies of power. From a feminist viewpoint, the negative portrayal of women as witches or as sex objects creates ideological effects. It sends people a negative message against women’s accession into masculine spheres of privilege and authority. These and other discriminatory feelings should not happen in the context of women’s growing social independence. The visual analysis has suggested that the representation encourages readers to draw misleading conclusions about female professionals, especially in the Brazilian context where abuse against women is an enormous social problem. The serious stereotypes constructed through different modes of representation produce sexism, derogation and exclusion. In the representation of powerful women, gender imbalance continues to operate with other axes of inequality. My critical reading therefore aimed at destabilising these axes.

Notes 1. All the translations from Portuguese into English are my own. 2. In Brazil, there is a distinction between the executive, which governs the country (although headed by elected people like the president, governors and city mayors—the majority of the executives are appointed), and the legislature, composed entirely of elected representatives: members of parliament at

300

Carmen Rosa Caldas-Coulthard

a federal level (divided in two houses), deputies at state and councillors at municipality level. The judicary is again separate and independent power and composed of appointed judges. The three powers are all independent of each other. 3. The Bank of English Corpora was created by COBUILD at the University of Birmingham and comprised at the time of investigation around 450 million words (approximately 71% British English, 21% North American English, 8% Australian English), mainly drawn from 1990s or early 2000s written sources but also including more than 20 million words of spoken interaction and 40 million words of transcribed radio broadcasts. 4. Unfortunately, for publishing reasons, most of my visual examples could not be reproduced in my final text. Readers however can access some of them through the links provided. 5. A Republica das Mulheres ‘Acho que vou ter de arrumar umas tranças no cabelo’, disse na manhã da segunda-feira passada o secretário-geral da Presidência da República, ministro Gilberto Carvalho, último espécime masculino com algum poder no Palácio do Planalto. Ele estava na primeira reunião de trabalho da Presidência desde a chegada das ministras Gleisi Hoffmann (Casa Civil) e Ideli Salvatti (Relações Institucionais).

References Barthes, R. (1977) Image, music, and text. London: Fontana. Baxter, J. (2010) Positioning gender in discourse: A feminist methodology. London: Palgrave Macmillan. Baxter, J. (2017) Women leaders and gender stereotyping in the UK press: A poststructuralist approach. London: Palgrave Macmillan. BBC (2016) Christine Lagarde, ‘rock star’ head of the IMF. Available at: https:// www.bbc.com/news/business-13452436 Brasil, Agencia (2015) ‘CPI da Petrobras convoca Graca Foster para depor na quinta fera. Litoralmania’, Available at: www.litoralmania.com.br/cpi-da-petrobrasconvoca-graca-foster-para-depor-na-quinta-feira/ (Accessed 11 April 2018). Bloomberg (2018) Available at: http://fortune.com/2018/02/07/women-executivejobs/?iid=sr-link1 Bulawka, H. M. (2012) ‘Gender representations in the Polish press: A feminist critical discourse study’, Unpublished PhD Thesis. University of Birmingham. Caldas-Coulthard, C. R. (1995) ‘Man in the news: The mis-representation of women speaking in news-as narrative discourse’, in Mills, S. (ed.) Language and gender: Interdisciplinary perspectives. London: Longman, pp. 226–239. Caldas-Coulthard, C. R. (2010) ‘Women of a certain age: Life styles, the female body and ageism’, in Holmes, J. J. and Marra, M. (eds.) Femininity, feminism and gendered discourse. Newcastle upon Tyne: Cambridge Scholars, pp. 21–40. Caldas-Coulthard, C. R. and Coulthard, M. (eds.) (1996) Texts and practices: Readings in critical discourse analysis. London: Routledge. Caldas-Coulthard, C. R. and Moon, R. (2010) ‘Curvy, hunky, kinky: using corpora as tools in critical analysis’. Discourse and Society 21(2): 1–35. Caldas-Coulthard, C. R. and Moon, R. (2016) ‘Grandmother, gran, gangsta granny: Semiotic representations of grandmotherhood’, Gender and Language, 10 (3), pp. 309–339.

Public Women

301

Fairclough, N. (1992) Language and social change. Cambridge: Polity Press. Fairclough, N. (2003) Analysing discourse: Textual analysis for social research. London: Routledge. Fairclough, N. (2010) Critical discourse analysis: The critical study of language. Prentice London: Sage in association with The Open University. Folha de Sao Paulo, Focando a notícia. (2011) ‘Dilma troca três porquinhos por meninas super poderosas’, Focanda a Notica. Available at: www.focan doanoticia.com.br/dilma-troca-tres-porquinhos-por-meninas-superpoderosas/ (Accessed 11 April 2018). Folha de São Paulo (2011) Dilma se destacou no governo Lula como a mãe do PAC’, (Programa de Aceleração do crescimento. Available at: www1.folha.uol. com.br/poder/817529-.shtml. Geiger, A. and Kent, L (2017) Number of women leaders around the world has grown, but they’re still a small group. Available at www.pewresearch.org/ fact-tank/2017/03/08/women-leaders-around-the-world. Grase, A. (2013) The haters of Mary Beard's grey hair. Available at: https://www. theguardian.com/commentisfree/2013/jan/25/mary-beard-long-grey-hair Hall, E. (1968) ‘Proxemics’, Current Anthropology, 9 (2/3), pp. 83–108. Hall, S. (1997) (ed.) Representation: Cultural representations and signifying practices. London: Sage. Halliday, M. A. K. (1978) Language as social semiotic. London: Edward Arnold. Halliday, M. A. K. (1985) An introduction to functional grammar. London: Edward Arnold Huggler, J. (2018). Available at: www.telegraph.co.uk/news/2018/02/13/crownprincess-mini-merkel-rising-star-fresh Jabour, B. (2013) Tony Abbott says female candidate has sex appeal. Available at: https://www.theguardian.com/world/2013/aug/13/tony-abbott-candidatesex-appeal Kappeler, S. (1986) The pornography of representation. Cambridge: Polity Press. Kanter, R. M. (1993). Men and women of the corporation (2nd ed.). New York, NY: Basic Books. Kiernan, P. (2008) ‘Deconstructing narrative identity in ELR: An analysis of teacher interaction in Japanese and English’, Unpublished PhD Thesis. University of Birmingham. Knight, B. (2013) Angela Merkel, Germany’s mother. Available at: https://www. theguardian.com/commentisfree/2013/sep/20/angela-merkel-germany-mother Korosec, K. (2018) Available at: http://fortune.com/2018/02/06/uber-gender-paygap-study/?iid=sr-link3 Kress, G. (2010) Multimodality: A social semiotic approach to contemporary communication. New York: Routledge. Kress, G. and Van Leeuwen, T. (2001) Multimodal discourse: The modes and media of contemporary communication. London: Edward Arnold. Kress, G. and Van Leeuwen, T. (2006) Reading images: The grammar of visual design. 2nd ed. London: Routledge. Machin, D. (2004) ‘Building the world’s visual language: The increasing global importance of image banks in corporate media’, Visual Communication, 3 (3), pp. 316–336. Machin, D. (2007) Introduction to multimodal analysis. London: Hodder Education.

302

Carmen Rosa Caldas-Coulthard

Machin, D., Caldas-Coulthard, C. R. and Milani, T. (2016) ‘Doing critical multimodality in research on gender, language and discourse’, Gender and Language, 10 (3), pp. 301–308. Mais mulheres no poder (2016). Available at: www.spm.gov.br/central-de-conteudos/ publicacoes/2016/plataforma-mais-mulheres-no-poder-2016 Martin, J. R. and White, P. R. R. (2005) The language of evaluation: Appraisal in English. New York: Palgrave Macmillan. Mills, S. (1995) Feminist linguistics. London: Routledge. ‘More men named John’, Available at: https://me.me/i/more-men-named-johnrun-big-companies-o-than-all-2894463 (Accessed 11 April 2018). Noticias. (2016) ‘Belas mulheres fizeram parte do noticiário ao se destacarem no meio político brasileiro. Available at: https://noticias.r7.com/brasil/fotos/ primeira-dama-do-turismo-roubou-a-cena-mas-outras-musas-ja-abalarama-republica-brasileira-confira-26042016#!/foto/11. Política para mulheres (2014). Available at: www.spm.gov.br. Political Cartoon on Twitter (2018) ‘Twitter’, 25 January. Available at: https:// twitter.com/Cartoon4sale/status/956647500043243522). Sofia News Agency. (2010) ‘Bulgarian-descended Dilma Rousseff said to outshine Merkel, Clinton’, Available at: www.novinite.com/articles/120543/ Bulgaria-Descended+Dilma+Rousseff+Said+to+Outshine+Merkel%2C+Clin ton (Accessed 11 April 2018). Revista Época (2011) A república das mulheres. Available at: http://revistaepoca. globo.com/Revista/Epoca. Toolan, M. (1988) Narrative: A critical linguistic introduction. London: Routledge. Van Leeuwen, T. (1996) ‘The representation of social actors’, in Caldas-Coulthard, C. R. and Coulthard, M. (eds.) Texts and practices: Readings in critical discourse analysis. London: Routledge, pp. 32–70. Van Leeuwen, T. (2005) Introducing social semiotics. London: Routledge. Van Leeuwen, T. (2008) Discourse and practice: New tools for critical discourse analysis. New York: Oxford University Press. Volosinov, V. N. (1986) Marxism and the philosophy of language. Cambridge, MA: Harvard University Press. White, Peter. R. R. (2001) ‘Appraisal: An overview’, Available at: www.grammatics. com/appraisal/index.html (Accessed 19 March 2018). Wodak, R. (2000) Discourse and discrimination. London: Routledge. Women in National Parliaments (2018). Available at: http://archive.ipu.org/wmne/classif.Htm

19 The Citizen Caught Between Dialogue and Bureaucracy Wolfgang Teubert

I am not a client, a customer, nor a service user. I am not a shirker, a scrounger, a beggar, nor a thief. . . . I am a man, not a dog. As such, I demand my rights. I demand you treat me with respect. I, Daniel Blake, am a citizen, nothing more and nothing less. (I, Daniel Blake, 2016)

1. Responding by Rebelling In Ken Loach’s film bearing the name of the protagonist, Daniel Blake rejects being treated by those in charge, in this case, the jobcentre administration, in a form of discourse I call ‘indexical’,1 i.e. a way of communicating that precludes him from challenging the way he is interrogated or from entering a negotiation with his caseworker. Bureaucratic discourse is one of its manifestations. What the persona Daniel Blake demands instead is a form of communication I will call dialogic discourse. It is a discourse where participants have equal rights, including the right to challenge what has been said, the meaning of words, but also propositions, suppositions and demands. As a human being, and as a citizen, in particular, he believes he is entitled to this form of communication. Citizenship, as he (or Ken Loach, signing responsible for the script) understands it, implies equality in the sense that all citizens enjoy the same inalienable rights, among them the right to be heard and responded to in a free, unregulated dialogue including all citizens. Michael Toolan has chronicled the 50 years of gradual disappearance of the notion of equality from contemporary public discourse. This is, for him, a deeply worrying phenomenon. In his article ‘Peter Black, Christopher Stevens, Class and Inequality in the Daily Mail’ (Toolan, 2016),

304 Wolfgang Teubert he notes that in Christopher Stevens’s contributions to the Daily Mail in 2013 ‘class has disappeared from the discursive agenda of contemporary Britain, like black and white film. It is only mentioned in relation to Britain in the past or elsewhere’ (2016, p. 657f.). Is contemporary society really made up of equals since reference to class has all but disappeared from public discourse? Only occasionally the topic re-emerges, often in the guise of outsider observations. In the New York Times of 27 April 27 2012, we read about Britain: ‘From the food people eat to everyday vocabulary and that ultimate giveaway, the accent, class remains a powerful identifier on this side of the Channel in a way that can seem quaint or even almost mediaeval to continental Europeans’ (Bennhold, 2012). In Britain, however, those in power had been denying the reality of class and its distinctions since the days of Margaret Thatcher, who is said to have declared in 1992, ‘Class is a communist concept’ (Jones, 2014). Later, in 1999, Labour Prime Minister Tony Blair himself rang the death knell on the traditional notion of class, when he said: ‘I want to make you all middle class’ (Hanley, 2015). Is class in Britain, as far as it still exists, no longer marked?2 Not to talk about upper, lower or middle class is to invite the illusion that wealth, family background and a privileged education are irrelevant in view of an encompassing citizenship. Toolan mentions some of the labels which in our days seem to have replaced class, such as ‘[b]lacks, lesbians, Catholic priests, madrassah, or junior doctors’ (Toolan, 2016, p. 568). Talking about class means talking about inequality. Talking about diversity seemingly does away with inequality, by assigning everyone their fair place in society, whether rich or poor. The London borough of Havering offers this definition of diversity on its website: ‘[Diversity] means recognising, valuing and taking account of people’s different backgrounds, knowledge, skills, needs and experiences. It is also about encouraging and using those differences to create a cohesive community and effective workforce’ (ScienceCouncil, 2016). Was it perhaps the class wars of the past that prevented a cohesive society and economic growth? In his forthcoming book The Language of Inequality in the News, Michael Toolan asks, ‘How did increased inequality become so “normal” and accepted as reasonable?’ (Toolan, forth.). I fully concur with his conclusion: I believe that it did so partly by gradual changes in public discourses, changes in the ways of discoursal representation of those with power and those without, so that this state of affairs became understood to be normal, reasonable, and ordinary. A naturalising and legitimizing within public discourses of kinds and degrees of inequality, previously less accepted as reasonable and ordinary ways to describe people and their circumstances, took hold. (Toolan, forth., 9f)

Citizen Caught Between Dialogue, Bureaucracy

305

In this contribution, I first want to sketch the narratives endorsing and rejecting the link between citizenship and equality. I will then try to demonstrate how what I call dialogic discourse is based on the arbitrariness of language signs, while what I call indexical, or, in this case, bureaucratic, discourse treats what is said as non-negotiable, disallowing those addressed to challenge what they are told. In the last section, I will argue that there is little evidence to suggest dialogic discourse is on the rise again in the twenty-first century. The task I have set myself is interpreting the reality confronting us, which is not so much the world out there as the world shaped and made meaningful by what has been said about it, by others and by ourselves. As this discourse is plurivocal, this (discursive) reality is presented in and can be viewed from different, often mutually exclusive, perspectives.3 As there is no external vantage point, there will never be a true or final interpretation. My contribution is an invitation to question what I claim, for my hope is that a dialogue worth its name will lead, for its participants, to new levels of reflexion.

2. Why Citizenship Means Different Things and Why Equality Should, or Should Not, Be a Value While citizenship was much discussed among the luminaries of the enlightenment era, people began to value it only when they saw themselves elevated from the status of subjects to that of citizens. This found its early expression in the Declaration des droits de l’homme et du citoyen (26 August 1789), particularly in article 6: La Loi est l’expression de la volonté générale. Tous les Citoyens ont droit de concourir personnellement, ou par leurs Représentants, à sa formation. Elle doit être la même pour tous, soit qu’elle protège, soit qu’elle punisse. Tous les Citoyens étant égaux à ses yeux sont également admissibles à toutes dignités, places et emplois publics, selon leur capacité, et sans autre distinction que celle de leurs vertus et de leurs talents. [The law is the overt expression of the general will. All the citizens have the right to contribute personally, or by their representatives, to the formation of law {to the legislative process}. The law must be the same one for all, either that it protects, or that it punishes. All the citizens, being equal in its eyes, are also acceptable by all dignitaries, in all places and in all measure of public employment, according to their capacity and without other distinction than that of their virtues and their talents.]4 (Larocheusa, 2002) The French historian and co-founder of the Annales movement, Lucien Febvre, characterised this concept of citizenship in these words: ‘The

306 Wolfgang Teubert Revolution makes a group of subjects, vassals and members of restricted communities into the body of citizens of one and the same state’ (Febvre, 1973, p. 213). For the ruling class in Britain, alarmingly, this notion quickly found many followers on their side of the Channel, too, made popular, among others, by Thomas Paine, by then emigrated to America. William Pitt the Younger expressed the sentiments of the establishment in his speech in parliament on 31 January 1799 in these words: ‘[T]his false and dangerous mockery of the sovereignty of the people is in truth one of the chief elements of Jacobinism, one of the most favourite impostures to mislead the understanding, and to flatter and inflame the passions of the mass of mankind’ (Pitt, 1799, column 1287). It was however less the concept of citizenship than that of equality which became the key target and thus had to be denounced at all cost. A pamphlet by Hanna More mocks this idea in a dialogue between a mason (Tom) and a blacksmith (Jack): TOM: JACK:

What is Equality? For every man to pull down every one that is above him, till they’re all as low as the lowest. (as cited from Herzog, 1998, p. 131)

The same sentiment is still expressed today, for instance, by Alexander Boot, columnist of the Daily Mail: True equality can only exist in heaven; in earth, the belief that all men are created equal is wishful thinking. For men are created unequal in strength, intelligence, character—well, in everything. Earthly inequality is thus a natural order of things, and it can only be distorted by unnatural means. Even then it won’t disappear; it’ll be replaced by a worse type of inequality or else camouflaged by demagoguery . . . Equality of result can indeed be achieved by leveling downwards (the only direction in which it’s ever possible to level) . . . [P]rison is the epitome of egalitarian aspirations, the ideal towards which they strive. (2011) The French concept of citizenship never became really popular in Britain. While in France as well as in the United States equality in political and legal terms is regarded as an inalienable right, in Britain this right can be abrogated if one doesn’t do as told. In Theresa May’s keynote speech to the 2016 Tory conference, the value of equality has been replaced by fairness. Her notion of citizenship is as far removed from the French concept as the abyss a no-deal Brexit would create between the continent and Britain. In her view, citizens have to respect fixed rules, imposed by others: [W]e also value something else: the spirit of citizenship. That spirit that means you respect the bonds and obligations that make our

Citizen Caught Between Dialogue, Bureaucracy

307

society work. . . . [I]f you believe you’re a citizen of the world, you’re a citizen of nowhere. You don’t understand what the very word ‘citizenship’ means. (May, 2016) In her speech, the word ‘equal’ does not occur once. Her key term is ‘fair’, occurring 24 times. Theresa May is not alone in her views. We find them also expressed in the Citizenship Review, published on behalf of its convenor Lord Goldsmith under the title ‘Citizenship: Our Common Bond’ in March 2008. There, the British notion of citizenship is defined as ‘the statement of a reciprocal relationship under which the individual offers loyalty in exchange for protection’ (2008, p. 10). This is recapitulating Thomas Hobbes’s view of citizenship, summed up by Derek Heater in these words: ‘Men effect their transition from a state of nature to a civil society by surrendering their natural rights to a sovereign in return for protection, hence becoming citizens’ (Heater, 2006, p. 56). According to Goldsmith, the relationship ‘between citizen and the state’ could be ‘described as a contract’ (2008, p. 92). For him and all those subscribing to his views, it is not, like in the continental enlightenment discourse, the citizens forming the state by covenant, but the everlasting state endowing citizens, as they come and go, with certain privileges in exchange for their loyalty. This is why the Conservative Party supports the ‘concept of a British Bill of Rights and Responsibilities’ (2008, p. 93). It is not alone. According to The Guardian of 25 October 2007, Labour Prime Minister Gordon Brown and his ministers ‘believe a British bill of rights and duties could provide an explicit recognition that human rights come with responsibilities for every citizen’ (Travis and Wintour, 2007). Those who do not play by rules made by those in power will lose their entitlement. Admittedly, there are other voices. John Redwood, for instance, a prominent Conservative, has published, on his website under the date of 7 September 2016, an article with the title ‘The Sovereignty of the People’ (Redwood, 2016). What he says, though, is very much a minority view: I believe in the sovereignty of the UK people. As a democrat, I believe that the people exercise their rights and freedoms by choosing representatives for their Parliament, and dismissing them at general elections if they cease to please. Between elections the sovereign people can either let their Parliament get on with the job they were elected to do, or the people can argue, lobby, press, campaign for their Parliament to vary its plans. (Redwood, 2016)

3. Dialogic vs. Indexical Discourse For Ferdinand de Saussure, linguistic signs, words, do not refer to a discourse-external world but to other words. Together with all that

308 Wolfgang Teubert distinguishes one word from the rest, they create the meaning of a text. Words are arbitrary, for what they stand for ‘has no natural connection with the signified’ (Saussure, 1966, p. 69). But for him, the meaning of words is fixed: ‘[T]he arbitrary nature of the sign is really what protects language from any attempt to modify it’ (Saussure, 1966, p. 73). This is what I cannot agree with. On the contrary, for me, arbitrariness implies that the meaning of a linguistic sign is the (always provisional) result of contingent negotiations between the members of a discourse community. Every time someone reacts to something said, by accepting or rejecting it or by rephrasing it, intertextual links from this new utterance to these earlier contributions will emerge in the form of the same or closely related lexical items. What such a lexical sign (token) means depends on the contexts in which it occurred previously and the way the current context differs from them. The meaning of a lexical item thus is contained in the entirety of contexts in which it has occurred. This definition of meaning may be more philological than interpretive, but one that is at least in parts operational. From a more interpretive perspective, each occurrence of a word can reveal its interdiscursive content, by the discourse participant detecting what is appended and what is omitted compared to the way it was used previously. Thus, each occurrence of a word will add something to what has already been said about it. This is what habitually happens when people engage in a discussion. In such a dialogic discourse, not just words but also propositions and suppositions are arbitrary and thus subject to negotiation, often implicit in the way they are used, or explicit in the form of a paraphrase. What we never know beforehand in a dialogue of equals is where this discussion leads. Will what one utterance stipulates be picked up by subsequent utterances? Meaning, in this sense, is always contingent and not reducible to rule-driven processes. In a dialogue, whether in a live conversation or a written exchange, people exchange arguments pro and con a given paraphrase. And every speaker is keen to make an impact on what will be said by others in ensuing contributions. An apt example is the case of ‘excellence’ pitted against or reconciled with ‘diversity,’ particularly in the field of higher education. Already in 1993, a Center of Excellence for Diversity in Health Education and Research had been established at the Pennsylvania Perelman School of Medicine (Penn Medicine, no date). The majority view evolving in English-speaking countries has been that a diversity of students does not negatively impact excellence. The discussion, however, is far from over. An opinion piece by Margaret Wente in a Canadian daily on 4 November 2017, offers this argument: Diversity [not excellence] is the point—not diversity of thought or intellect, but diversity of race, gender, ethnicity and sexual orientation . . . The unexamined notion behind the diversity craze is that the

Citizen Caught Between Dialogue, Bureaucracy

309

more diverse the team or institution, the better the performance. In fact, there is no real evidence for this. Another fallacy is the assumption that skills, desires, preferences and motivation are evenly spread across all groups in society .  .  . But the argument that equal outcomes are the one true measure of equality is corrosive. It means we’re doomed to see people through the prism of race and gender instead of talent and achievement. (Wente, 2017) Soon after there was a response by the vice-chancellor of Regina University, Vianne Timmons, in the journal University Affairs (17 November 2017): Margaret Wente argues that diversity and excellence cannot exist together at Canadian universities. I strongly disagree. Diversity and excellence are not at odds in our universities. Nor are they at odds in the seven principles on equity, diversity and inclusion .  .  . Ms. Wente’s implication that we cannot achieve excellence by committing to equity for groups such as women, Indigenous people, and persons with disabilities is ludicrous . . . Taking steps to ensure that conscious and unconscious bias do not play a role in hiring and promotion does not mean that excellence cannot exist in universities and other institutions. (Timmons, 2017) What do the seven commenters have to say about Timmons’s stance? Unsurprisingly, most of them support her. But there is one contribution that brings in a new perspective: One point that rarely seems to make it into these discussions is that it is patently false to suggest . . . that we know how to recognize excellence ahead of time .  .  . [I]n practice our view of what constitutes ‘excellence’ is limited to the small range of traditional metrics and situations with which we have experience, and blind to the actual challenges any candidate will face. (Rader, 2017) Dialogic discourse can challenge what has been said so far. It does it by paraphrasing the topical items, in this case excellence and diversity. Part of their meaning is what has been said about their relationship. Wente (2017) pits excellence against diversity, arguing that excellence measured only in quote-determined outcomes does not reflect that some groups may indeed do better than others. Timmons (2017) claims equity for minority groups does not prevent excellence from happening. The commenter brings in a new aspect: As we cannot measure excellence in advance, we can never predict individual outcomes.

310 Wolfgang Teubert What is not disputed is that ‘diversity of race, gender, ethnicity and sexual orientation’ (as Wente, 2017, puts it) is an issue in need of being addressed in the context of higher education. What is ignored, however, is the elephant in the room, namely class. Being born rich facilitates access to higher education institutions bearing the badge of excellence, and thus one’s individual outcome. A rare exception to this obliviousness is a contribution to the blog of the Centre for Labour and Social Studies (CLASS) of 23 October 2017, with the headline ‘Oxbridge diversity is important, but inequality in higher education runs much deeper’. The first sentence reads: ‘The revelation last week that Oxbridge continues to recruit the majority of its students from white middle class families in the south east of England comes as no surprise’ (Gamsu and Donnelly, 2017). The attribute ‘in the south east of England’ clarifies that what is meant is the top segment of the middle class, which could also be called upper class if this expression were not tabooed. In most cases, though, class is not talked about. It is this pervasive absence of the class argument that Michael Toolan sees as an indicator for the fact that contemporary media discourse avoids talking about inequality. The confrontation of arguments in a dialogic discourse like the one on excellence vs. diversity will typically engender new ideas, and thus promote the evolutionary adaptation of discourse. Some contentions will come to feature more prominently than others; some will hardly or never be referred to and thus have no impact on meaning. It is this kind of dialogue that makes discourse develop in unpredictable directions. Perhaps in ten years’ time, no one will still talk about excellence pitted against diversity, or maybe the other way around. It is therefore most regrettable that corpus linguistics has not paid enough attention to the temporal dimension of discourse. There is a reason why I have taken this example for dialogic discourse not from the realm of politics, but from the kind of discourse we (still) find in the global academic republic of letters. Its citizens are, at least in theory, equal. What matters is not whether a case is stated by a postdoc researcher or a tenured professor, but how the peer community reacts to it, whether they choose to ignore it or keep referring to it. In reality, however, as research evaluations tend to demonstrate, academic discourse is perhaps not quite as classless as it pretends. The notion of dialogism and its role in discourse is by no means new; it has become, over the last decades, an established element of discourse studies. It was Julia Kristeva who introduced Roland Barthes and other French intellectuals back in the 60s of the last century to the concept of intertextuality in her exegesis of Bakhtin’s terms ‘dialogue’ and ‘ambivalence’: ‘[A]ny text is constructed as a mosaic of quotations; any text is an absorption and transformation of another’ (Kristeva, 1986, p. 39). V. N. Volosinov, identical or not with Bakhtin, formulates it in these words: ‘Any utterance . . . makes response to something and is calculated to be

Citizen Caught Between Dialogue, Bureaucracy

311

responded to’ (1986, p. 72). From a different perspective, this is also the mantra of Jacques Derrida’s Limited Inc, where he discusses the fact that each token/occurrence of a word (type) has a different meaning, because it occurs in a different context: ‘Every sign, linguistic or non-linguistic, can be cited, put between quotation marks; in doing so it can break with every given context, engendering an infinity of new contexts in a manner which is absolutely illimitable’ (Derrida, 1988, p. 12). Or, in other words, ‘Iteration alters, something new takes place’ (p. 40). Dialogic discourse undergoes never-ending transformations, and with its evolution the meaning of lexical items or text segments will always be provisional. Every new contribution offers a new perspective, a new paraphrase and interpretation of what has been said previously. Without dialogic discourse, nothing new can be said. This is what distinguishes it from what I call indexical discourse. I readily admit that my take on Charles Peirce’s gradually evolving exploration of iconic, indexical and symbolic signs is somewhat idiosyncratic.5 For me, iconic and indexical items are not so much signs in the sense that their user is reflecting on or interpreting their meaning, but rather cues or prompts. They make the addressee respond to them in a specified way, without the freedom to modify their reaction. Controlled language, as it is used for instance in certain machine translation applications, consists of lexical elements whose signification is fixed, not context-dependent, and not subject to interpretation. The user of an indexical item is not in a position to reparaphrase it. Any deviation from the rules leads to an abortion of the communicative act. Technical terms are another example of indexical language. Once terms are standardised by the relevant authority, they express a strictly defined concept not subject to negotiation. The origin of terminology is engineering. A production process has to be laid down by the engineer in such a way that the workforce will carry it out exactly as intended, avoiding any ambiguity. Terms denote all essential elements of the process, things and the parts they consist of, states, measurements, processes and actions. Workers are not supposed to engage their own common sense, but to follow strictly instructions point by point. Manufacturing processes of this kind depend as much on indexical language as on hierarchy. Those in charge hand out the instructions; those working on the assembly lines must not question them. Dialogic discourse would disrupt production. We find indexical discourse at one end of a cline, with dialogic discourse at the other end. To be a citizen, in the continental sense of the word, means reclaiming one’s voice from one’s overlords, and implies citizens engage in open discussions with each other on the basis of equality. In Immanuel Kant’s famous first sentence of his essay asking, ‘What is enlightenment?’ (commonly translated as ‘Enlightenment is man’s emergence from his self-incurred immaturity’; Goodreads, www.goodreads. com/quotes/246637-enlightenment-is-man-s-emergence-from-his-

312 Wolfgang Teubert self-imposed-immaturity-immaturity-is, no date), immaturity stands for German Unmündigkeit, meaning ‘the state of having no voice’, either as an infant or here, in this context, in legal terms (‘being of age’). The French Revolution’s aim was to put an end to this state. Citizens were promised the capacity to jointly decide the fortunes of their compact, by forming a civil society based on liberty, equality and solidarity. Citizenship in this sense spells out the abolition of class structure and the provision of equal rights for all. However, those who had been pulling the strings all along did not give up so easily. While the newly won democracy, in its parliamentarian variety, first gave some, then more, and finally almost all people of age the right to vote, citizens were reduced to voters, thus every few years delegating their voice to representatives selected by political parties, and these representatives had to hand it over, to a large extent, to the party whip. Only in theory were people given the right to have their say. Whether they also were heard is a different matter. Newspapers provided the main platform of public discourse, a discourse, though, that could hardly be dialogic, as citizens were denied control over the press. How ordinary citizens can make their voice be heard is, as we will see, still an unresolved question. This is perhaps the main reason why communication between authorities and the people whom they are to serve appears less dialogic and more indexical. As objects of authority, people are commonly treated like the industrial workforce. Just like the engineer tells the foreman, and the foreman tells the worker in unambiguous terms what they have to do, in the interaction between officialdom and the people this discourse is shrouded in bureaucracy, as indexical as that of the production process. The citizen has become, once again, the subject. What is euphemistically called a service offered by the state becomes an exercise of power, and bureaucratic discourse is its instrument. It re-establishes class and makes short shrift with equality. Max Weber, in his posthumously published collection of papers Economy and Society (1979 [1922]), has devoted much of his research to the phenomenon of bureaucracy. This is how Patty Mulder sums up Weber’s notion: ‘Bureaucracy is an organisational structure that is characterised by many rules, standardised processes, procedures and requirements, number of desks, meticulous division of labour and responsibility, clear hierarchies and professional, almost impersonal interactions between employees’ (Mulder, 2017). Already in 1894, in the fifth edition of Meyers Konversations-Lexikon (encyclopaedia), we find this astonishingly progressive definition: [Bureaucracy is the t]erm for a short-sighted and narrow-minded organisation of civil servants which does not understand the practical needs of the people . . . The foundation of bureaucracy is the rule of absolutism. Bureaucratic rule marks the era of the police state, of

Citizen Caught Between Dialogue, Bureaucracy

313

police censorship of the minds of the people in the 19th century. The formation of constitutional governance . . . the importance of a free press for debate of public affairs . . . are accomplishments precluding a bureaucratic form of administration in our times. (Meyers Konversations-Lexikon, 1894; my translation) In a bureaucratic discourse between state institutions and the public, for instance, the discourse of jobcentres, caseworkers instruct their clients to follow the rules. It is they who ask the questions, design the forms to fill in and process the clients’ responses in accordance with rules made by those in charge. As the clients must choose between what is permitted as a response to the questions they are asked, they cannot point out when such a question is, due to the specifics of a given case, inapplicable. While this creates an inequality between the institution (the jobcentre) and its clients, the caseworkers themselves are equally bound by the rulebook and have no space for reasoned manoeuvring. This is even more the case as its instruments are increasingly outsourced to commercial companies. The Work Capability Assessment (WCA), determining if a client is fit to work, to take an example, was first run by Atos, and later, after complete failure, by Maximus. The assessment process is organised in the way of a computer programme. This is how Wikipedia describes it: The process uses fixed criteria and a system of points in an attempt to differentiate between people claiming ESA [Employment and support allowance] who have no significant disabilities and those who, because of the type and severity of their disability, do face significant barriers to work . . . The way the test has been carried out has been strongly criticised. The semi-structured interview style and the use of decision-making software led to a suspicion that the WCA was merely a Whitehall box-ticking exercise: a checklist of standard questions and a ‘computer says no’ attitude to eligibility, leading to large numbers of disabled people wrongly being declared ‘fit for work.’ 6 (‘Work Capability Assessment’, 2018; emphasis in the original) In Loach’s film, the protagonist’s capability to work is assessed in a telephone interview. As he has recently suffered a heart attack, he has not only lost his job, but he is also under doctor’s order not to work. However, in order to apply for ESA, his only option for any kind of financial support is to undergo the prescribed work capability assessment. Here is a brief extract of the dialogue between his assessor and himself from the film script: Good morning, Mr Blake. My name’s Amanda. I’ve got a couple of questions here for you today to establish your eligibility for Employment Support Allowance.

314 Wolfgang Teubert . . . Can you raise either arm as if to put something in your top pocket? Can you raise either arm to the top of your head as if you are putting on a hat? I’ve telt you, there’s nowt wrong with me arms and legs. Could you just answer the question, please. Well, you’ve got me medical records. Can we just talk about me heart? D’you think you could just answer these questions? Okay. So was that a yes, that you can put a hat on your head? Yes. . . . Yes. Yes, it’s me fucking heart. I’m trying to tell you but you’ll not listen. If you could just answer the question, please. (I, Daniel Blake, 2016; my emphasis) It is easy to see what is happening in such interviews, very similar to the myriad of call centres run by public institutions or commercial companies to provide ‘service’ to citizens or customers. The interviewer/ caseworker sits in front of a screen and scrolls the checklist down from top to bottom, reading out the questions and noting down the answers, and prompting the interviewee to come up with a response fitting one of the choices offered. Whatever is not in the script cannot be dealt with. There is no room for negotiation. It is this kind of bureaucratic discourse that strips people of their rights as citizens. There does not seem to be much research in discourse studies on this sort of institutional interaction. Much of what can be found comes, perhaps not quite surprisingly, from Scandinavian countries, and it was Elisabet Cedersund and Roger Säljö who first drew attention to the phenomenon of ‘bureaucratic discourse’ in the context of the welfare state (Cedersund and Säljö, 2009). From the papers assembled in the recently republished volume Listening to the Welfare State, edited by Michael Seltzer, Christian Kullberg and Illmari Rostila (2017, first published in 2001), the overall impression is that in spite of bureaucratic attempts to strictly regulate communication between caseworkers and their clients, dialogic discourse still allowed for some however limited negotiation of positive outcomes, which caseworkers then were required to retranslate into the checklist form. But we must remember that the research covered here was carried out many years before the recent policy of austerity had been implemented. I found one research paper dealing with the kind of institutional discourse we currently find in the UK, by Vanesa Fuertes and Colin Lindsay: ‘Personalization and Street-Level Activation: The Case of the UK’s Work Programme’ (2017). It pictures scenery indeed far removed from what seemed to be the case in Scandinavian countries but is exemplified by the

Citizen Caught Between Dialogue, Bureaucracy

315

film I, Daniel Blake. The authors found little evidence for any attempt at dialogic discourse and the quest for a meaningful solution for a problem. Instead, the authors found that frontline practices and street-level interactions .  .  . have become increasingly standardized, rather than delivering tailored, personalized services . . . [S]treet-level practice was tightly constrained by an organizational context that demanded standardized, work-first interactions and imposed severe resource constraints on caseworkers . . . [S]tandard operating procedures, narrowly focused programme content and severe resource limitations appear to have imposed a stricter form of standardization . . . [C]aseworker–client interactions were often functional and focused on the completion of repetitive tasks. The relationship between these standardized processes and the sought outcome of sustainable employment often seemed unclear (or sometimes immaterial) to both caseworker and client . . . [What we saw was an] activation regime that long ago ‘locked in’ mechanisms to minimize the costs of supporting those on benefits while imposing compulsory activities designed to move people into work as quickly as possible. (439f.) The persona Daniel Blake wants to be treated as a citizen, not as a subject. He wants to engage in a meaningful dialogue with the jobcentre, supposedly there to ‘serve’ citizens like him in need of support. Bureaucracy in the sense described here, however, embraces the closely connected concepts of inequality and hierarchy. Relying on communication in the form of indexical discourse, it becomes an insurmountable obstacle to a functioning civil society, a society able to adapt to the needs of its citizens, as Jessica Studdert writes in The Guardian on 12 February 2018: [S]uccessful organisations around the world, as well as in the UK, recognise that in an increasingly complex and networked society, sticking to a rigidly hierarchical approach limits their impact .  .  . In practice, this means shifting away from traits that characterise hierarchy—iron control, rigid sign-off and processes that imply wisdom only resides with senior management . . . [A] more fundamental shift away from rigid hierarchies will be required, to adopt cultures that enable creativity, collaboration and autonomy. (Studdert, 2018)

4. Civil Society, Citizenship and Equal Rights in the Twenty-First Century? For Jürgen Habermas, the public sphere is the social space in which citizens come together to debate the common good. With the arrival of social media, Google, Facebook and Twitter, one should think that for the first

316 Wolfgang Teubert time in history, a true civil society engaging in dialogic discourse would have become a real option. The social media provide a public sphere, a virtual space, a platform where citizens of a whole country or even beyond can meet in order to share and exchange their views of common affairs, in ways unthinkable just a few decades ago. Before the arrival of social media, public discourse across the whole nation remained largely confined to newspapers, radio and television. People were talked to but had little chance to talk back, or to discuss matters among themselves, unless they entered an institutional framework, for instance, a political party, subjecting them to its rules. Habermas seems to put much trust in the traditional media. For him, ‘[m]ediated political communication in the public sphere can facilitate deliberative legitimation processes in complex societies only if a selfregulating media system gains independence from its social environment, and if anonymous audiences grant feedback between an informed elite discourse and a responsive civil society’ (Habermas, 2006, p. 412). Equal rights for all citizens, however, are not easily reconciled with an ‘elite discourse’ demanding a ‘responsive civil society’. And strangely enough, Habermas does not endorse the new social media as empowering civil society. In his view, ‘people who more extensively use the electronic media, and consider them an important source of information, have a lower level of trust in politics and are more likely to take a cynical attitude towards politics as a consequence’ (2006, p. 422). This chimes in nicely with the current blaming of social media for providing a platform for hate speech and fake news, and the steadily growing call for censorship. It almost seems as if those who are in charge have only a limited interest in being discussed and challenged by those they are in charge of (cf. also Teubert, 2012). I am not sure if for Habermas such ‘a self-regulating media system’ that is ‘independent from its social environment’ would also entail independence from entrepreneurial intervention by its owners. In Britain, media are seen these days less as a precondition of a civil society than as a commercial good. The Communications Act, passed in 2003 by Tony Blair’s New Labour government, regulates the British media industry very much as a market, not as a requisite for a civil society. In the almost 400 pages of the Act, the word ‘citizen(s)’ comes up just three times. Instead we find more than 70 occurrences of ‘consumer(s)’. Indeed, this word seems to encapsulate the new relation between the people’s representatives and the people. In the twenty-first century, the government, one could say, has become part of the service industry. It is little wonder then that communication between the state as a service provider and the people as its clients frequently takes the form of focus groups, often consisting of a mere dozen carefully selected participants (Devault, 2018). These interactions often reveal themselves to be less a consultation and more an exercise in appeasement. Lisa Featherstone’s

Citizen Caught Between Dialogue, Bureaucracy

317

grim essay ‘Talk Is Cheap: The Myth of the Focus Group’, published in The Guardian on 6 February 2018, begins with these words: ‘Focus groups make us feel we matter—but no one in power cares what we think’ (Featherstone, 2018). A focus group is a virtual gathering of people who participate in highly regulated question-answer sessions intended to elicit people’s perceptions about a given topic. Instead of an open discussion, there is only bureaucratic discourse. While most focus groups are run by industry, some are also conducted for the government. In Featherstone’s view, focus groups have become part of a process in which citizenship is reduced to consumerism [. . .] The process looks like democracy in action, and most people enjoy participating [. . .] The story of the focus group is a story of the relationship between elites and the masses . . . Policy-makers now regard ‘citizen engagement’ as so important that they routinely hire specialised private firms or non-profit organisations to manage public debate about local policy decisions [. . .] Corporate elites are, so far, doing an excellent job of using the listening process to attract our votes and our spending. We the people are, meanwhile, struggling to be heard in ways that actually change anything, because giving voice is not the same as taking power. (Featherstone, 2018) A civil society made up of citizens enjoying equal rights and participating in an unregulated dialogue, empowering them to deal with the problems confronting them, remains a utopian goal, it seems, also in the twentyfirst century.

Notes 1. Following loosely Charles Peirce, I call language signs indexical that do not allow communication partners to negotiate their meaning. An indexical sign implies a fixed correlation between the sign and what it stands for, a correlation that is either innate or the result of trial and error or taught or, as in Daniel Blake’s case, imposed by authority. The opposite of iconic or indexical signs are what Peirce calls ‘symbolic’ and Ferdinand de Saussure calls ‘arbitrary’ signs, i.e. signs without a fixed meaning and open to interpretation, and, I would add, dialogic negotiation (cf. Peirce 1992/1998, on his different definitions, and, as a criticism of Peirce’s and de Saussure’s sign concepts, Roy Harris (1996). A useful discussion of the development of Peirce’s sign theory is offered by the Stanford Encyclopedia of Philosophy (https://plato.stanford. edu/entries/peirce-semiotics/). 2. As I do not accept there is a principal difference between object language and meta language, I try to resist the temptation to explicitly distinguish between class as an object (class), as a concept (‘class’) or just as a word in a text (class), unless there is a really good reason. 3. Totalitarian regimes, religious authorities and increasingly also social media often find reasons to regulate (indexicalise) this discourse.

318 Wolfgang Teubert 4. French original and English translation available at http://larocheusa.org/dec laration.htm. 5. A useful discussion of the development of Peirce’s sign theory is offered by the Stanford Encyclopedia of Philosophy (available at https://plato.stanford.edu/ entries/peirce-semiotics/). 6. One could argue that this Wikipedia entry (https://en.wikipedia.org/wiki/ Work_Capability_Assessment) is biased, even though references are provided for claims underlined here. But this would presuppose a concept of objectivity to which I do not subscribe.

References Bennhold, K. (2012) ‘Class war returns in new guises’, The New York Times, 26 April. Available at: www.nytimes.com/2012/04/27/world/europe/27iht-let ter27.html (Accessed 9 April 2018). Boot, A. (2011) ‘The destructive myth of equality’, Daily Mail Online, 29 December. Available at: www.dailymail.co.uk/debate/article-2079543/The-destructivemyth-equality.html (Accessed 9 April 2018). Cedersund, E. and Säljö, R (2009) ‘Bureaucratic discourse, conversational space, and the concept of voice’, Semiotica, 97 (1–2), pp. 79–102. Derrida, J. (1988) Limited Inc. Evanston: Northwestern University Press. Devault, G. (2018) ‘What is a market research focus group?’ The Balance, 12  March. Available at: www.thebalance.com/what-is-a-market-researchfocus-group-2296907 (Accessed 9 April 2018). Featherstone, L. (2018) ‘Talk is cheap: The myth of the focus group’, The Guardian, 6 February. Available at: www.theguardian.com/news/2018/feb/06/talk-ischeap-the-myth-of-the-focus-group (Accessed 9 April 2018). Febvre, L. (1973) ‘Frontière: The word and the concept’, in Burke, P. (ed.) New kind of history: From the writings of Febvre. London: Routledge. Fuertes, V. and Lindsay, C. (2015) ‘Personalization and street-level practice in activation: The case of the UK’s work programme’, Public Administration, 94, pp. 526–541. Gamsu, S. and Donnelly, M. (2017) ‘Oxbridge diversity is important, but inequality in higher education runs much deeper’, Class for Labour and Social Studies, 23 October. Available at: http://classonline.org.uk/blog/item/oxbridge-diver sity-is-important-but-inequality-in-higher-education-runs-muc (Accessed 9 April 2018). Goldsmith, L. (2008) ‘Citizenship: Our common bond’, Available at: http://image. guardian.co.uk/sys-files/Politics/documents/2008/03/11/citizenship-report-full. pdf (Accessed 9 April 2018). Goodreads. (no date) ‘Immanuel Kant’, Available at: www.goodreads.com/quotes/ 246637-enlightenment-is-man-s-emergence-from-his-self-imposed-immaturityimmaturity-is (Accessed 9 April 2018). Habermas, J. (2006) ‘Political communication in media society: Does democracy still enjoy an epistemic dimension?’ Communication Theory, 16, pp. 411–426. Hanley, L. (2015) ‘Please, Liz Kendall, stop talking about “white working-class communities”‘, The Guardian, 3 June. Available at: www.theguardian.com/ commentisfree/2015/jun/03/liz-kendall-stop-talking-about-white-workingclass-communities (Accessed 9 April 2018). Harris, R. (1996) Signs, language and communication. London: Routledge.

Citizen Caught Between Dialogue, Bureaucracy

319

Heater, D. (2006) Citizenship in Britain: A history. Edinburgh: Edinburgh University Press. Herzog, D. (1998) Poisoning the minds of the lower orders. Princeton: Princeton University Press. I, Daniel Blake (2016) ‘Movie script’, Available at: www.springfieldspringfield. co.uk/movie_script.php?movie=i-daniel-blake (Accessed 9 April 2018). Jones, O. (2014) ‘We’re not all middle-class now: Owen Jones on class in Cameron’s Britain’, NewStatesman, 29 May. Available at: www.newstatesman.com/ culture/2014/05/we-re-not-all-middle-class-now-owen-jones-class-cameron-sbritain (Accessed 9 April 2018). Kristeva, J. (1986) The Kristeva reader. Edited by T. Moi. Oxford: Wiley-Blackwell. Larocheusa. (2002) ‘Déclaration des droits de l’homme et du citoyen du 26 Août 1789’, Available at: http://larocheusa.org/declaration.htm (Accessed 9 April 2018). May, T. (2016) ‘Theresa May’s conference speech in full’, The Telegraph, 5 October. Available at: www.telegraph.co.uk/news/2016/10/05/theresa-mays-conferencespeech-in-full/ (Accessed 9 April 2018). Meyers Konversations-Lexikon (1894) ‘Bürokratie’. 5th ed. Leipzig: Bibliographisches Institut. Mulder, P. (2017) ‘Bureaucratic theory by Max Weber’, Available at: www.toolshero. com/management/bureaucratic-theory-weber/ (Accessed 9 April 2018). Peirce, C. S. (1992, 1998) The essential Peirce, selected philosophical writings. Volume 1 (1867–1893), 1992, in Houser, N. and Kloesel, C. (eds.) and Volume 2 (1893–1913), 1998, Peirce Edition Project (eds.). Bloomington and Indianapolis, IN: Indiana University Press. Penn Medicine (no date) ‘Center of excellence for diversity in health education and research’, Available at: www.med.upenn.edu/inclusion-and-diversity/pcedher. html (Accessed 9 April 2018). Pitt, W. (1799) Hansard’s Parliamentary Debates. 3rd series. Commencing with the Accession of William IV, Vol. II (1831) London: Hansard. Rader, S. (2017) ‘Comments on Equal opportunities bring excellent outcomes: A response to Margaret Wente’, University Affairs, 17 November. Available at: www.universityaffairs.ca/opinion/in-my-opinion/equal-opportunities-bringexcellent-outcomes-response-margaret-wente/ (Accessed 9 April 2018). Redwood, J. (2016) ‘The sovereignty of the people’, John Redwood’s Diary, 7 September. Available at: http://johnredwoodsdiary.com/2016/09/07/thesovereignty-of-the-people/. Saussure, F. (1966) Course in general linguistics. New York: McGraw-Hill. Science Council (2016) ‘Diversity glossary’, Available at: https://sciencecouncil. org/web/wp-content/uploads/2016/01/Diversity-Glossary.pdf (Accessed 9 April 2018). Seltzer, M., Kullberg, C. and Rostila, I. (2017) Listening to the welfare state. Abingdon: Routledge. Studdert, J. (2018) ‘An obsession with hierarchy is undermining social services’, Guardian 12 February. Available at: https://www.theguardian.com/publicleaders-network/2018/feb/12/change-councils-beat-austerity-demoralised (Accessed 9 April 2018). Teubert, W. (2012) ‘Democracy and web-based dialogue’, in Cooren, F. and Létourneau, A. (eds.) (Re)presentations and dialogue. Amsterdam: John Benjamins, pp. 99–123.

320 Wolfgang Teubert Timmons, V. (2017) ‘Equal opportunities bring excellent outcomes: A response to Margaret Wente’, University Affairs, 17 November. Available at: www.university affairs.ca/opinion/in-my-opinion/equal-opportunities-bring-excellent-outcomes-response-margaret-wente/ (Accessed 9 April 2018). Toolan, M. (2016) ‘Peter Black, Christopher Stevens, class and inequality in the Daily Mail’, Discourse and Society, 7, pp. 642–660. Toolan, M. (forthcoming) The language of inequality in the news. Cambridge: Cambridge University Press. Travis, A. and Wintour, P. (2007) ‘Brown: We need bill of rights as well as Human Rights Act’, The Guardian, 25 October. Available at: www.theguardian.com/ politics/2007/oct/25/humanrights.constitution (Accessed 9 April 2018). Volosinov, V. N. (1986) Marxism and the philosophy of language. Cambridge, MA: Harvard University Press. Weber, M. (1979 [1922]) Economy and society. Oakland, CA: University of California Press. Wente, M. (2017) ‘Equal outcomes have replaced equality of opportunity’, The Globe and Mail, 4 November. Available at: www.theglobeandmail.com/opinion/ equal-outcomes-have-replaced-equality-of-opportunity/article36825364/ (Accessed 9 April 2018). ‘Work Capability Assessment’. (2018) Wikipedia. Available at: https://en.wikipedia. org/wiki/Work_Capability_Assessment (Accessed 9 April 2018).

Contributors

Marc Alexander is Professor of English Linguistics at the University of Glasgow and is the third Director of the Historical Thesaurus of English. His research primarily focuses on the study of words, meaning and effect in English, often using data from the Thesaurus or approaches from the digital humanities. He also works on oratory and style in Parliamentary discourse from 1803 to present, the linguistics of statutory and constitutional interpretation and the cognitive and persuasive rhetoric of detective fiction. Joe Bray is Professor of Language and Literature at the University of Sheffield. His work encompasses literary stylistics, the eighteenth- and nineteenth-century novel and experimental literature. He is the author of The Epistolary Novel: Representations of Consciousness (2003), The Female Reader in the English Novel: From Burney to Austen (2009) and The Portrait in Fiction of the Romantic Period (2016). His editorial work includes Ma(r)king the Text: The Presentation of Meaning on the Literary Page (2000), Mark Z. Danielewski (2011) and The Routledge Companion to Experimental Literature (2012). He is currently completing a book on the language of Jane Austen for the Palgrave Language, Style and Literature series, which will be published in 2018. Michael Burke is Professor of Rhetoric at Utrecht University. He is also Dean of the Honour College and responsible for the honours programmes across all faculties within Utrecht University. His fields include classical rhetoric, stylistics and cognitive poetics. His is the author of Literary Reading Cognition and Emotion: An Exploration of the Oceanic Mind (2011) and editor of The Routledge Handbook of Stylistics (2014) and Stylistics: Critical Concepts in Linguistics (2017). He is a former Chair of the Poetics and Linguistics Association (PALA). Beatrix Busse is Professor of English Linguistics at Heidelberg University. Her research interests include the history of English, English historical linguistic and Shakespeare studies, stylistics and corpus linguistics. She

322

Contributors

is the co-editor of the series Discourse Patterns (de Gruyter Mouton) and reviews editor of the International Journal of Corpus Linguistics. One of her current research projects is about language in urban space and patterns of discursive place-making in Brooklyn, New York. Carmen Rosa Caldas-Coulthard is Professor of English Language and Applied Linguistics at the Federal University of Santa Catarina, Brazil, and Senior Research Fellow in the English Department at the University of Birmingham, UK, where she worked from 1996 to 2012. She has published widely in the areas of critical discourse, media and gender studies. Her research interests are in social semiotics and gender representation, visual communication and identity politics. She is the author of Language and Sex (Atica, Brazil, 1991), News as Social Practice (UFSC, Brazil, 1997) and co-editor with Malcolm Coulthard of Texts and Practices: Reading in Critical Discourse Analysis (Routledge, 1996) and Translation: Theory and Practice (UFSC, Brazil, 1991). She is also the co-editor with Michael Toolan of The Writer’s Craft, the Cultures’ Technology (Amsterdam: Rodopi, 2005) and with Rick Iedema of Identity Trouble: Critical Discourse and Contested Identities (London: Palgrave, 2008). She is co-editor of the journal Gender and Language (Equinox). Billy Clark is Professor of English Language and Linguistics at Northumbria University in Newcastle. His research and teaching interests cover a wide range of topics in linguistics and stylistics, with a particular focus on semantics and pragmatics. This has included work on lexical and syntactic meaning, semantic change, phatic communication, prosodic meaning, multimodality and pragmatic processes involved in the reading, writing and evaluation of texts. He has a longstanding interest in connections between work at school and at university. He is a member of the UK Linguistics Olympiad committee and, with Marcello Giovanelli and Andrea Macrae, coordinates the Integrating English project (http://integratingenglish.org). Matthew Collins is an AHRC final-year PhD candidate at the University of Birmingham, from where he obtained an MA in Literary Linguistics in 2015. His thesis focuses on narrative cohesion in the fifteenth-century Morte Darthur by Thomas Malory. He was formerly Researcher in Residence at the Royal Shakespeare Company and a Visiting Fellow at the University of Warwick. His research interests include corpus linguistics, historical pragmatics and narratology. Alexanne Don (University of Birmingham, 2007) is an independent researcher teaching at the University of New South Wales, Sydney, Australia. Her research interests concern applications of Systemic Functional Linguistics—especially the use of the Appraisal framework in the investigation of social media, identity and argument structure.

Contributors

323

She is presently involved in collaborative research theorising the language of identity in dress, lyrics and environmental communication. Her previous employment includes 11 years as an English instructor at schools and universities in Fukuoka-ken, Japan. She also has qualifications in TEFL and Art Education. Her artwork continues in garden creation, and in driving a Canon 7D. Catherine Emmott is Reader in English Language and Linguistics at the University of Glasgow. Her research interests mainly focus on the cognitive study of narrative texts, using approaches from linguistics, stylistics, literary studies, psychology and artificial intelligence. She is author of the books Narrative Comprehension: A Discourse Perspective (Oxford University Press, 1997) and Mind, Brain and Narrative (with Anthony J. Sanford, Cambridge University Press, 2012). She has also published articles on cognitive stylistics, anaphora and referential form, medical life stories and cognitive and rhetorical aspects of detective fiction. Mel Evans is a Lecturer in English Language and Linguistics at the University of Leicester. Her research broadly explores language variation and change in literary and non-literary texts, addressing questions of identity construction, genre and pragmatics, specialising in English from the early modern period (1500–1700). She is presently a Co-I on the AHRC project ‘Editing Aphra Behn in the Digital Age’, analysing Behn’s literary style and investigating texts of questionable authorship using stylistic methods, to inform the new scholarly edition of Behn’s works. Chris Heffer is Director of MA Programmes, including the MA in Forensic Linguistics, in the Centre for Language and Communication Research at Cardiff University and Co-Founder of Cardiff Language and Law (CaLL). He has published articles in linguistic and legal journals on various theoretical and communicational aspects of the trial process and jury instruction and is the author of The Language of Jury Trial: A Corpus-Aided Analysis of Legal-Lay Discourse (Palgrave, 2005) and co-editor of Legal-Lay Communication: Textual Travels in the Law (Oxford University Press, 2013). He has a particular interest in communicational breakdown between judge and jury and he has advised judicial studies boards on this. He is currently completing a book on untruthfulness: Lying and Bullshit in Life and Law: The TRUST Framework (Palgrave, forth.). He has also been working for several years on a monograph Rhetoric and Rights: A Theory of Forensic Discourse (Oxford University Press, forth.). He sits on the Editorial Board of the OUP Language and Law series. He is possibly the only forensic linguist to have testified in a trademark case in the UK. Laura Hidalgo Downing is Associate Professor at the Universidad Autónoma of Madrid. Her research interests include discourse analysis,

324

Contributors

metaphor, stylistics and cognitive poetics. She is the author of Negation, Text Worlds and Discourse: The Pragmatics of Fiction (Ablex, 2000). She is the co-editor, together with Blanca Kraljevic, of the Special Issue on Metaphorical Creativity Across Modes (Metaphor in the Social World 13:2, 2013) and is the co-author, together with Robert Cockroft, Susan Cockroft and Craig Hamilton, of the third revised edition of Persuading People: An Introduction to Rhetoric (Palgrave, MacMillan, 2014). Lesley Jeffries is Professor of English Language and Linguistics at the University of Huddersfield. A former Chair of the Poetics and Linguistics Association, her research interests span stylistics and critical discourse analysis, and her major publications include Stylistics (Cambridge University Press, 2010; with Dan McIntyre), Critical Stylistics (Palgrave, 2010) and Opposition in Discourse (Bloomsbury, 2009). Her most recent book is Key Words in the Blair Years (Bloomsbury 2017; with Brian Walker). At Huddersfield she is Director of the Language in Conflict project and is currently editing The Routledge Handbook of Language in Conflict (Routledge, forth.). Manuel Jobert is Professor of English Language and Linguistics at the University of Lyon (Jean Moulin Lyon 3—France) where he specialises in pragmatics, stylistics and phonology. He is Visiting Professor at the University of Huddersfield (UK) and the former Chair of the French Society for English Stylistics (Société de Stylistique Anglaise). He co-edited a book on linguistic impoliteness (Aspects of Linguistic Impoliteness, Newcastle upon Tyne, Cambridge Scholars Publishing) in 2013 and one on euphemisms (Empreintes de l’euphémisme—tour et Détours, Paris, L’Harmattan) in 2010. He co-authored a textbook on English phonetics (Transcrire l’anglais britannique et américain, Toulouse, Presses Universitaires du Mirail) in 2009. His book, The Pragmatics of Irony and Banter (John Benjamins), co-edited with Sandrine Sorlin, was published in 2018. Marina Lambrou is Associate Professor at Kingston University, UK. She is also the Head of the Department for Linguistics and Languages. Her teaching and research areas are interdisciplinary, taking both stylistic and narratological approaches to personal stories and fictional narratives, and she has also published on media discourses and pedagogy. Marina is researching personal narratives based on traumatic experiences to understand and gain insights on how individuals structure and evaluate their extraordinary experiences which will eventually be published in book form. She is also currently writing a book entitled Disnarration and the Unmentioned in Fact and Fiction (forth.) which explores a dimension of narrative storytelling across a range of texts from oral narratives told by children to media stories and several works

Contributors

325

of literary fiction. Marina has published widely and is the co-editor of Contemporary Stylistics (Continuum, 2007; with Peter Stockwell), coauthor of Language and Media (Routledge, 2009; with Alan Durant), editor of the special edition on ‘Narrative’ for the journal Language and Literature (Sage, February 2014; Vol. 23, No. 1) as well as published numerous book chapters that include ‘Dialogism in Journalistic Discourse: Ian McEwan’s Response to 7/7’ in Mildorf, J., and Thomas, B. (eds.), Dialogue Across Media (John Benjamins, 2007); and ‘Stylistics, Conversation Analysis and the Cooperative Principle’ in Burke, Michael (ed.), The Routledge Handbook of Stylistics (Routledge, 2014). Michaela Mahlberg is Professor of Corpus Linguistics at the University of Birmingham, UK, where she is also the Director of the Centre for Corpus Research and the Director of Research and Knowledge Transfer for the College of Arts and Law. Michaela is the editor of the International Journal of Corpus Linguistics (John Benjamins) and together with Wolfgang Teubert she edits the book series Corpus and Discourse (Bloomsbury). One of her main areas of research is Dickens’s fiction and the sociocultural context of the nineteenth century. Her publications include Corpus Stylistics and Dickens’s Fiction (Routledge, 2013), English General Nouns: A Corpus Theoretical Approach (John Benjamins, 2005) and Text, Discourse and Corpora: Theory and Analysis (Continuum, 2007; co-authored with Michael Hoey, Michael Stubbs and Wolfgang Teubert). Michaela was the Principal Investigator on the AHRC-funded project CLiC Dickens: Characterisation in the Representation of Speech and Body Language From a Corpus Linguistic Perspective which led to the development of the CLiC web app. Dan McIntyre is Professor of English Language and Linguistics in the Department of Linguistics and Modern Languages at the University of Huddersfield, where he teaches stylistics, corpus linguistics, audiovisual translation and the history of English. His major publications include Stylistics (Cambridge University Press, 2010; with Lesley Jeffries) and Language and Style (Palgrave, 2010; co-edited with Beatrix Busse). With Brian Walker he is currently writing a book called Corpus Stylistics, for Edinburgh University Press. At Huddersfield, he is also Director of Language Unlocked (languageunlocked.co.uk), a linguistic consultancy service that has provided corpus-based language analysis for such clients as the BBC, the UK Green Party and Unions 21. He is editor of the journal Language and Literature (Sage) and co-editor of Babel: The Language Magazine (babelzine.com). He also edits the book series Advances in Stylistics (Bloomsbury) and Perspectives on the English Language (Palgrave; with Lesley Jeffries). His latest book is Applying Linguistics: Language and the Impact Agenda (Routledge, forth., 2018; co-edited with Hazel Price).

326

Contributors

Rocío Montoro is Senior Lecturer in English Language at the University of Granada, Spain. Her research interests include the stylistics of popular fiction, questions related to authorial intention as well as corpus and multimodal stylistics. She is the author of Chick Lit: The Stylistics of Cappuccino Fiction (2012) and the co-author (with Nina Nørgaard and Beatrix Busse) of Key Terms in Stylistics (2010); she has also authored a number of articles in journals and chapters in edited collections. She is joint editor (with Paul Simpson) of the series Language, Style and Literature (Palgrave) and assistant editor of Language and Literature. Emma Moreton is a Senior Lecturer in the School of Humanities at Coventry University where she teaches undergraduate and postgraduate modules in corpus stylistics and discourse analysis. Her current research uses corpus and computational methods of analysis to examine the language of historical correspondence data, with a particular focus on nineteenth-century letters of migration. Emma was project manager on two JISC-funded projects which explored the use of visualisation tools with corpus data and she was the Co-I on an AHRC research networking project (Digitising Experiences of Migration: The Development of Interconnected Letter Collections) which explored issues surrounding the transcription, digitisation and annotation of migrant letter collections. Rukmini Bhaya Nair is Professor of Linguistics and English at the Indian Institute of Technology, Delhi. She received her PhD from the University of Cambridge and has since taught at universities ranging from Singapore to Stanford. Awarded an honorary doctorate by the University of Antwerp in 2006 for her work on narrative pragmatics, Nair has authored more than 100 papers and articles as well as nine books of which Lying on the Postcolonial Couch: The Idea of Indifference (2002, Minnesota University Press), Narrative Gravity: Conversation, Cognition, Culture (Routledge, 2003) and Poetry in a Time of Terror (Oxford University Press, 2009) are somewhat representative. Recipient of several awards such as the J.N. Tata Scholarship, the Hornby and Charles Wallace grants, etc. from the time she was a student Nair’s latest award was a Professorial Fellowship at the Centre for the Study of Developing Societies in 2016. Currently, she is a member of the Society of Fellows, Indian Institute of Advanced Study, Chair of the Advisory Board on Scholarships of the Open Society Foundation, New York, as well as on the Consultative Board of the International Pragmatics Association (IPRA). Consulting editor at Biblio: A Review of Books, Nair serves on the editorial boards of international journals such as Text Matters, Language and Dialogue and Literary Semantics. Her latest major grants from the Indian government are to conduct basic research in cognition on ‘Language, Emotion and Culture’ (2010–2014) and to work on the ‘Capabilities Approach to Education’

Contributors

327

(2013–2016). Nair also won the All India Poetry Society First Prize in 1989 and has since published three volumes of poetry with Penguin, been translated into languages such as Chinese, Swedish and Italian and read her poetry at venues from Struga and Sydney. Her first novel, Mad Girl’s Love Song (2013), was on the ten-book list for the prestigious DSC Prize. Selected as a ‘Face of the Millennium’ in a national survey of writers by India Today, Nair has been called ‘India’s first significant postmodern poet’. The entry on her in the standard reference volume Oxford Companion to Modern Poetry (OCMP, ed. Ian Hamilton and Jeremy Tod, 2014), which contains entries for poets across the world during the past century (1910–2010), including iconic names such as T. S. Eliot and Pablo Neruda, says of her work that it ‘has been widely admired by other poets and critics for its postmodern approach to lyrical meaning and feminine identity’. Nair’s writings, creative and critical, have been included in courses at Chicago, Delhi, Harvard and Oxford. According to Nair, she does research for the same reasons that she writes poetry—to discover the possibilities and limits of language. Nina Nørgaard is Associate Professor of Applied Linguistics and Director of the Centre for Multimodal Communication at the University of Southern Denmark. Her research interests lie within the fields of stylistics, multimodality, multimodal stylistics, critical discourse analysis and the semiotics of architecture. She has a keen interest in combining work and insights from the (traditionally separated) fields of stylistics and multimodality. Her major publications include Key Terms in Stylistics (2010; co-edited with Beatrix Busse and Rocío Montoro) and Multimodal Stylistics of the Novel: More Than Words (forth., 2019) as well as articles and book chapters on multimodal stylistics. Ruth Page is a Senior Lecturer at the University of Birmingham in the Department of English Language and Applied Linguistics. She spends her time teaching and writing about storytelling, especially focusing on the stories that people tell in social media contexts. She has written several books including Stories and Social Media (Routledge, 2012) and Narratives Online: Shared Stories and Social Media (Cambridge University Press, forth.), and was the lead author of the student textbook Researching the Language of Social Media (Routledge, 2014). Her publications in journals and edited collections have explored narratives in a wide range of contexts, from print literature and the mainstream media to conversational stories and stories published in online contexts such as blogs, social network sites and Wikipedia. She has been the editor of Discourse, Context & Media, and convened BAAL’s Special Interest Group for Language and New Media. Stephen Pihlaja is a Reader in Stylistics at Newman University, Birmingham. He holds a PhD from the Open University and is Senior Fellow of the Higher Education Academy. He is the author of two

328

Contributors

monographs: Antagonism on YouTube (Bloomsbury, 2014) and Religious Talk Online (Cambridge University Press, forth.) and has edited special issues of Language and Literature (2016) and Metaphor and the Social World (2017). His interests include discourse analysis, metaphor in online discourse and talk about religion and religious identity. Dan Shen is Changjiang Professor of English Language and Literature at Peking University. She is on the advisory or editorial boards of the American journals Style and Narrative, the British Language and Literature and the European JLS: Journal of Literary Semantics. In addition to six books and more than 100 essays in China, she has published Style and Rhetoric of Short Narrative Fiction with Routledge and numerous essays in North America and Europe in stylistics, narrative studies and translation studies. Wolfgang Teubert held the Chair in Corpus Linguistics at the University of Birmingham from 2000 to 2013. He had studied German and English linguistics at Heidelberg University. Before he moved to Birmingham, he was Senior Fellow at the German Language Institute in Mannheim, among other things responsible for corpus research in the framework of European multilingual language resources co-operation. His focus has always been on meaning, the meaning of lexical items, of text segments and of whole texts. He wanted to establish corpus linguistics as a field with its own unique theoretical foundation. Language, for him, is not a mechanical system but exists in the form of discourse, i.e. a discourse consisting of arbitrary signs which do not refer to a discourse-external reality but mean only what has been said about them. A world without discourse has no meaning. Discourse has a diachronic dimension. New utterances keep interpreting what has been said, and by endorsing, modifying or rejecting it, they give birth to new ideas. Now that he is fully retired, Teubert’s aim is to establish hermeneutics (i.e. the art and craft of interpretation) firmly as the theoretical framework of discourse analysis, with corpus linguistics as its methodology. Viola Wiegand is a Research Fellow on the CLiC Dickens project at the University of Birmingham. Her main research interest is the study of textual patterns with tools and frameworks from corpus linguistics, stylistics and discourse analysis. Viola’s current research focuses on surveillance discourses and nineteenth-century fiction. She is assistant editor of the International Journal for Corpus Linguistics.

Index

action process 21–22, 26 actors, represented 294 affect 68, 140, 241, 293 appraisal 12, 270–71, 277, 280, 292, 322 appreciation 148, 154, 279, 293 assertion (as a speech act) 154 assumption 169–70, 174, 309 attractors 106 Austen, J. 12, 225–29, 231, 235 Austin, J. L. 148, 150–53, 237 Bakhtin, M. 239, 248, 310 behavioural context 4 biographical 17, 46, 226 blend 105–6, 110–11 body language 128, 160 Bremain 104–6, 110–14 Bremond, C. 36–37 Brexit 11, 103–7, 110–21, 306; Brexiteer 107–10; Brexiter 107–10 British National Corpus (BNC) 10, 65, 107, 131, 134 British National Corpus of Contemporary British English (BNC2014) 107 caricatures 34 categorization 75, 289–90, 299 character 34–39, 56, 72, 125–28, 131, 134, 169–70, 178, 181, 185, 193, 198, 215–18, 221–23, 225–27, 229–35, 254–55, 257–60, 264, 306; character psychology 226 characterization 34, 217–19, 228, 325 Chazelle, D. 30, 34, 37–41 Chick Lit Corpus 65–6, 71–74 citizenship 303–7, 312, 315, 317–19 classification 46, 48, 55, 228, 290 cluster 126, 132–34, 137, 140, 272, 282

cognitive linguistics 6, 9 cognitive stylistics 2 collocates 107–10, 118, 272; collocation 117, 126; semantically negative 109–10 colloquialization 68–69, 75 concordance 67, 71–72, 124–30, 134–37, 290, 292 connotation 118–20, 271, 295 contexts of situation 147 contextualization 1, 5, 17, 72, 164 conversation analysis 6, 129 corpus linguistics 124–28, 310 Corpus of Early English Correspondence (CEEC) 10, 43, 35 Corpus of Global Web-Based English (GloWbe) 11, 110 Corpus of Historical American English (COHA) 107–8 Corpus of Online Registers of English (CORE) 110 corpus stylistics 2, 10, 63–65, 68, 74, 123–24, 127, 131, 138 corpus tools 79, 89, 124–26 corpus-based analysis 43, 71, 79, 100, 103–4, 193, 325 counterfactual 30, 32–40; divergence 30, 32–36, 39–40; plot 33; thinking 35 covert progression 17, 19, 21–22, 24–27 critical linguistics 3 Davies, B. and Harré, R. 6 deixis 79, 81, 93, 96, 99; deictic 11, 79, 93, 94, 96, 99–100, 214–16, 223, 254, 260, 263 ‘demand’ picture 295, 298 denotation 113 densification 68–70, 75 detective fiction 10–12, 177–78, 188 deviation 54, 125–27, 140, 211, 222, 311

330

Index

dialogic 277, 295, 303, 305, 307–12, 314–17 digital fiction 7, 255–56 digital humanities 124, 127, 140 digital technology 7 direct speech (DS) 131, 178, 180, 213–14, 217, 221 discourse 1, 4–5, 9, 11–14, 26, 29, 33, 43–46, 48, 50, 52, 55–58, 79, 87–88, 100, 104, 106, 110, 129–30, 149, 154, 164, 170, 181, 193–94, 209–10, 214–16, 225–26, 231, 237, 239–41, 248–49, 258, 271–72, 286, 290, 293, 303–5, 307–18, 321–28; gendered 287; public 287, 303–4 disnarration 29–30, 32, 39–40, 324 diversity 9, 11, 218, 304, 308–10 Duranti, A. 4 Early Modern English 46 Emma 12, 225–26, 230–35 entextualisation 5, 11, 238 epistolary 43–48, 50–57, 226, 321 equality 303–6, 309, 311–12 evaluation 20, 48, 51–53, 153, 164, 174, 183, 261, 270–71, 292–93, 310, 322 experientiality 43–45 Experimental Studies of Literature (ESL) 9, 195 female difference 299 figurative language 194, 196, 198, 203, 239 film 11–12, 29–35, 37–38, 40–42, 164–73, 303–4, 313, 315; genres 30–32 foregrounding 10, 25, 103, 106, 110, 125, 214, 242, 278 forked 30, 32, 35–40 free direct speech (FDS) 227 free indirect discourse 26, 225 free indirect speech (FIS); ‘words-andphrases’ FIS 225–26 free indirect style 18 free indirect thought (FIT) 25, 225; ‘whole-sentence’ FIT 225–26 functionalization 289–90 Furlong, A. 165, 172–74 gap 35, 37, 82, 210, 235 gaze 5, 147, 156, 158–59, 255, 260, 295–96, 298

gender 3, 8, 11–12, 45, 46, 48–49, 53, 55, 286–90, 299, 308–10, 322; imbalance 287, 299; inequality 289; parity 288 gendered 8, 127, 286–87, 289, 292, 299; gendering 287, 298; identities 286 genre of fiction 154 Golden House, The 155 grammar of film 30 graphemic structure 104–5 Grexit 112, 121 Halliday, M. A. K. 50, 105, 237, 269, 287, 295 Haneke, M. 164, 170, 171, 173 heteroglossia 5 historical context 1, 7, 17, 93 History of Sir Charles Grandison, The 225–26, 235 Hollywood 30–34, 40; Golden Age of 30, 32, 40 hyperbolic style 230 ideational function 287 identification 45–46, 87, 126, 131, 134, 184, 186, 194, 214, 290, 297, 299; physical 290, 299; relational 290 identity politics 8 identity positions, female 292 ideology 26, 31, 292, 297–98; ideological context 1, 11; ideological effect 104–5, 111, 299; sexist 287 illocutionary force 148, 227, 230 implication 2, 9, 48, 54, 129, 140, 168–69, 215–16, 221–22, 261, 271, 280, 289, 309; non-communicated 168–9 implicature 168, 239 indeterminacy 164–65, 168, 172–74 indexical 303, 305, 307, 311–12, 315, 317 Indirect Speech (IS) 227 indirectly quoted discourse 226 interactional context 4 interdiscursive contexts 5 interpretation 170–74; interpretative uncertainty 171; ‘literary’ 172; negative 296; pragmatic 79, 173; process 170 Irish migrant, female 80–81 irony 199, 203, 229, 239

Index judgment 17, 21–22, 24, 149 Kafka, Franz 18, 21; The Judgment 17, 21–22, 24 keyword 12, 103, 111, 113, 120, 124–26, 128, 233, 272–74 Kress, G. and van Leeuwen, T. 5, 286, 294–96 La La Land 31–34, 36–39 Language, Text and Context 2–5, 9, 11–12, 17, 29, 104, 164, 225, 237, 254, 269 letters 10–12, 43–44, 46–57, 79–82, 87, 89–91, 93, 96–97, 99–102, 105, 125, 156–58, 196, 215, 231, 246–47, 310, 326 lexical innovation 104 linguistic: anthropology 4, 9; creativity 103, 110; sign 307–8 literariness 123 literary: literary behaviour 149–51, 154; text 3–4, 6, 10–12, 17, 30, 46, 123–24, 126–27, 131, 151, 172, 193, 255, 323 Loach, K. 303, 313; Daniel Blake 303, 314–15 Lord Edgware Dies 184, 186–89 Lough letters 81, 91 Lying behaviour 150–51 Mansfield, K. 25–27; The Singing Lesson 24, 26–27 media: discourse 286–87, 310, 324; narrative 293 mediated discourse analysis 5 mental process 19, 23 metadata 2, 7–8 metaphor 25, 71, 87, 105, 159, 192–96, 198–203, 239, 324, 328 modal expressions 180 modality 21, 40, 182, 229; deontic 229; epistemic 179–80, 187 Modern Vampire Corpus (MVC) 67 morphology 42, 104, 110 multimodal stylistics 2, 326–27 multimodal resources 294 Murder at the Vicarage, The 178, 182 musical 30–33, 40–41, 108 naming 105, 112, 286 narrative 3, 6, 12, 17–18, 24, 29–36, 39–63, 125, 131, 148–49, 160, 177,

331

192–93, 197, 210, 214–16, 221, 225–27, 230–31, 233, 239, 242–43, 255–57, 259–60, 263, 286, 293, 298–300, 305, 322–28; progression 63, 125; public narrative 286 narrative report 50, 230 narrative report of speech acts (NRSA) 227–28, 230–33 narrativity 43–45, 48, 52–53, 56–57; high 44–45, 52–53, 56–57; low 44–45 narrator, omniscient 18, 25 narrator’s representation of voice (NV) 227 Naturalistic Studies of Reading 9, 195 neologism 104, 106–7 Neumann, A. W. 225–26, 233–35 newspaper discourse 104 nominalisation 105 Northanger Abbey 229–30 noun phrase 64, 69–70, 105–6 NOW Corpus 107, 110–11, 113–15, 117, 121 ‘offer’ picture 295–96 opposition 117–19, 121, 151, 294, 299, 324 overdetermination 293; overdetermined 297 overhearing 225, 231–35 parallelism 54 paratextual 2, 7 perspective, narrative 25, 48, 228, 230, 233–38, 240, 244, 257 Persuasion 229 plot 17–19, 22, 24–25, 30–41, 165, 177, 179, 186, 278, 280; development 17–19, 22, 24–25, 30; manipulation 177 poetry 1, 151, 154–55, 157–61, 326–27 point of view, ideological 295 postcolonialism 2, 155, 326 power 3, 7–8, 12, 63, 92–93, 110, 112, 194, 239–41, 244, 247, 250, 286–89, 292–93, 297–99, 304, 307, 312–17 pragmatic 1–2, 12, 79, 129, 164, 173–76, 213, 254, 322; inference 174 pragmatic stylistics 1–2, 67 pragmatics 4, 12, 145, 164–65, 174–76, 322–24, 326 pretend behaviour 12, 148

332

Index

progression 11, 15, 17, 19, 21, 22, 24–28, 63, 125, 219, 220 ProQuest International Newsstream database 111–15, 121 prose 1, 52, 64, 65, 151, 154–56, 158–61, 173, 246–47 questionnaire 195–96, 204 race 290, 308–10 readers 1, 4, 6–7, 9, 11–12, 17, 33, 36, 63–64, 70, 109, 133, 153–54, 169, 172, 177–78, 184, 188, 191–93, 195, 198, 204, 214–15, 226, 271, 277, 280, 282, 299 recontextualisation 5, 269, 272–73, 279 referent 103, 107, 111, 113, 115, 117, 120; referential vagueness 112 register 44, 50, 69–70, 74, 76, 113, 133, 186, 212, 244, 248–49 relational process 19–20 Relevance Theory 164, 168, 170 reliability 12, 177–84, 187–88; endorsements 183–84; vouching 179, 182–83 Remainer 107, 110 response 17, 53, 56, 112–13, 137, 147, 149, 173, 237, 249, 309–10, 313–14, 325; reader-response 12, 191–92, 195, 237 reversal stage (of a plot) 178, 181, 183–84, 188–89 Richardson, S. 225–26 romantic comedy 30–34, 38 Rushdie, S. 155 Salinger, J. D. 11–12, 191, 193, 195, 199; ‘A Perfect Day for Bananafish’ 11, 191, 195 script 38, 108, 180, 182, 303, 313–14 semantic 79, 87, 91–93, 99, 103, 107, 109–10, 120, 124–25, 198, 211, 220, 239, 271, 298, 322, 326, 328; domains 79, 91–92, 99; prosody 109 semiotic 1, 2, 5, 10, 211, 239–40, 286–87, 292–95, 299; evaluation 292; modes 1, 10, 286; resources 2, 5, 286–87, 295 Sense and Sensibility 228 sequential blend 104 sexuality 14, 286–87, 290, 296, 298 shifting referent 103 showing 12, 33, 38, 47, 87, 136, 139, 165–68, 170, 187, 210, 222, 248, 260, 278

silent movie genre 32–33, 40 Sing a Song of Sixpence 178, 182, 186 Sketch Engine 79, 91, 93, 99, 111, 121 social actor 289–90, 293 social context 17, 26, 56, 148, 262, 325 social history 81 social semiotics 5, 322 speech clusters 131–32, 134, 137, 140 speech presentation 131, 229; speech and thought presentation 227; speech and thought representation 178, 225, 234 Steinbeck, J. 149, 159–61 structuralism 30; structuralist 36 Style in Fiction 227 stylistics 1–7, 9–11, 14, 29–30, 40–42, 63, 65, 68, 74, 101, 104, 123–25, 127, 129, 131, 138–39, 154, 164, 193, 225, 227, 241, 282, 321–28 superdiversity 8–9 surprise 6, 10, 39, 43–46, 48, 50, 53–57, 77, 177–78, 186, 215, 218, 221, 310 suspense 10–11, 39, 43–46, 48, 50–56, 58; suspension 52, 128–29, 215 systemic 5, 123, 147, 211, 269, 322; functional linguistics 5, 269, 322 telling 3, 5, 18, 20, 39, 46, 50–56, 71, 100, 155, 166, 168, 170, 178, 184–85, 194, 202, 204, 257, 259, 261 text types 1, 57, 68 Text World Theory 127, 194 Toolan, M. 1–10, 14, 17, 28–30, 42, 44–46, 50, 52, 56, 63, 71, 74, 104, 123–25, 140, 147–48, 150, 152–54, 164, 174, 177, 192–95, 209, 222, 237–39, 241, 244, 254, 282, 286, 293, 304, 310, 322 topic identification 87 unreliability 12, 167, 177–78, 182, 184–86 variationist 3 voice 9, 12, 26, 209–10, 217, 220–21, 226, 230, 234, 244–48, 250, 278, 307; double-voice theory 226; double-voicing 226, 233 word formation 69, 104 writing representation 227 YouTube 2, 7, 242, 254–58, 261–64, 328