Literary and Linguistic Approaches to Feminist Narratology

Literary and Linguistic Approaches to Feminist Narratology Ruth E. Page Copyright material from www.palgraveconnect.co...

Author: Ruth E. Page

103 downloads 1348 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Literary and Linguistic Approaches to Feminist Narratology Ruth E. Page

Copyright material from www.palgraveconnect.com - licensed to Universitetsbiblioteket i Tromsoe - PalgraveConnect - 2011-03-09

Literary and Linguistic Approaches to Feminist Narratology

10.1057/9780230286665 - Literary and Linguistic Approaches to Feminist Narratology, Ruth E. Page



This page intentionally left blank

Ruth E. Page



Literary and Linguistic Approaches to Feminist Narratology

© Ruth E. Page 2006

No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted her right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2006 by PALGRAVE MACMILLAN Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N.Y. 10010 Companies and representatives throughout the world PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd. Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries. ISBN-13: 978–1–4039–9116–4 hardback ISBN-10: 1–4039–9116–2 hardback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Page, Ruth E., 1972– Literary and linguistic approaches to feminist narratology / Ruth E. Page. p. cm. Includes bibliographical references and index. ISBN 1–4039–9116–2 (cloth) 1. Feminism and literature. 2. Narration (Rhetoric). 3. Discourse analysis, Narrative. I. Title. PN56.F46P34 2006 808.3′93082—dc22 2005044657 10 15

9 14

8 13

7 12

6 11

5 10

4 09

3 08

2 07

1 06

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham and Eastbourne



All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission.


For Gavin, Isi and Toby





List of Figures, Tables and Boxes

ix

Acknowledgements

xi

1 Feminist Narratology in Context Narratives about theory Developments in narratology Developments in feminism Postmodern feminist narratology About this book

1 1 2 6 11 17

2 The Question of Gender and Form Psychoanalysis, literary theory and feminism Female/feminine/feminist alternatives A linguistic approach: Narrativity and gender Narrativity and gender in Flesh and Blood Narrativity and gender in Beloved Narrativity and gender in Une Glossaire/a Glossary Narrativity in Pale Fire and Invisible Cities Conclusion

20 20 22 25 27 31 35 40 43

3 The Question of Gender and Context Lanser’s call for feminist narratology Revising theories of plot Marking of climactic peak Alternative patterns of Peak marking Conclusion

45 45 49 58 61 70

4 The Importance of Similarities Moving towards sociolinguistics Data sample and analytic framework Evaluation as a structuring device Evaluation as a means of constructing solidarity Summary and explanation

73 73 75 78 82 89

5 Questioning Intra-category Variation Feminism and reader response Hypertext, gender and reader response

94 94 96

vii



Contents

viii Contents

100 101 107 111 114

6 Media Narratives of Success and Failure On the limits of performativity Critical linguistics and media texts Data sample and narrative framework Narratives of progression Representation: Naming over time Narratives of stability and regression Conclusion

116 116 117 119 122 127 138 142

7 Gender, Age and Narrative Development Age as a sociolinguistic variable Literacy in New Zealand schools Data sample and methodology Storytelling characteristics at seven years of age Storytelling characteristics at ten years of age Cultural context and gender Conclusion

144 144 145 146 148 156 168 170

8 Beyond Feminist Narratology? Synthesis: Weaving the strands together The impact on narrative theory Concepts of gender The importance of context Looking to the future The question of questions A last word

173 173 173 176 177 181 185 188

Notes

189

Bibliography

193

Index

207



Data sample and methodology Reading pathways: Degrees of linearity Narrativity and the storyworld Gender and characterization in the storyworlds Conclusion

Figures 3.1 Longacre’s model of plot (adapted from Longacre, 1983: 22) 3.2 Outline of the Problem–Solution and related patterns (adapted from Hoey, 2001: 166) 3.3 Predictable patterns in Sir Gowther 3.4 Predictable patterns in Sir Isumbras 3.5 Parallel predictable pattern elements in Emare 3.6 Interrelated Problem–Solution patterns in The Tale of the Monk who Received Bishamonten’s Aid 3.7 Predictable patterns in The Deeply Jealous Woman 6.1 Overarching Goal–Achievement pattern in Booth’s narrative of progression 6.2 Interrelated Goal–Achievement patterns in Rodham Clinton’s narrative of progression 6.3 Problem–Solution patterns in Booth’s narrative of regression

54 55 60 60 62 67 69 123 125 141

Tables 5.1 Percentage of linear steps taken by readers (by gender) 5.2 Percentage of linear steps taken by readers (by academic discipline) 5.3 Process types, their meanings and key participants (Halliday 1994: 143) 5.4 Proportions of process types in the storyworlds (expressed as a percentage) 5.5 Process types in the story strands of Fishnet 6.1 Reporting of the prime minister’s spouse in the Guardian and Times 6.2 Naming of Cherie Booth/Blair by newspaper (in %) 6.3 Naming of Hillary Rodham Clinton by newspaper (in %) 6.4 Correlation between naming and topic for Cherie Booth/Blair (in %) 6.5 Correlation between topic and naming choice in the Washington Times and Washington Post (in %)

102 103 107 108 110 120 129 130 132 135

ix



List of Figures, Tables and Boxes

x List of Figures, Tables and Boxes

137 147 148 148 150 156 156 157 161 162

Boxes 7.1 Personal narrative subdivided into episodes 7.2 Schematic outline of Fishing for Octopus by Feana Tu’akoi

159 163



6.6 Naming of Hillary Rodham Clinton in Washington Times over time (in %) 7.1 Ethnic origin of New Zealand children 7.2 Narrative components used by 7-year-olds: Retellings 7.3 Narrative components used by 7-year-olds: Personal experiences 7.4 Structural features of Narratives, Anecdotes and Recounts 7.5 Use of Result and child’s age (Personal narratives) 7.6 Use of Result and gender of child (Personal narratives) 7.7 Use of Result and nationality of child (Personal narratives) 7.8 Use of Result and child’s age (Retellings) 7.9 Use of Result and gender of child (Retellings)

I have to confess to a secret interest in reading acknowledgements sections, feeling that they give some glimpse of the stories that lie behind the emergence of the book itself. I have looked forward to expressing my appreciation to a number of people who have been part of the narrative behind this particular work. Beginning with the most formal, this book would not have been completed at this particular juncture without research leave funded by the Arts and Humanities Research Board (reference number: BRL/AN11563/APN18912) and by the University of Central England. I am grateful to both for the opportunity to work on the project in an uninterrupted manner. There are many colleagues who have been helpful in stimulating ideas that have shaped my thinking in important ways. Michael Toolan, Catherine Emmott and Carmen Rosa Caldas-Coulthard have all been consistently generous in their advice and support. David Roberts, Howard Jackson and Charles Owen also contributed suggestions for comparative texts that have found their way into various chapters. Some of the chapters emerged from papers given at the annual conferences of the Poetics and Linguistics Association. I benefited from the ensuing critiques offered by my fellow scholars there. I would particularly like to thank Marina Lambrou for reading through chunks of this manuscript and helping me to clarify my thoughts on numerous points. Needless to say, all errors remain the responsibility of the author. Special thanks also go to Jamila Hakam, Marie McGowan-Griffen, Ray Hosking and Debbie Eagle who either facilitated or helped collect the data used in several chapters. More personal thanks go to the many friends who helped in so many ways behind the scenes while I have been working on this book. If I do not name you all in person, it does not mean my thanks are any less heartfelt. But there are three people who deserve extra-special mention. Thanks go to Gavin Page – you have helped in so many practical respects, from excel spreadsheets to your hard work in keeping our family life running smoothly. Also to Isi Page, whose story lies in part behind the childbirth narratives discussed in Chapter 4 and who is now old enough to read and tell great stories of her own, and Toby Page, whose story is just beginning. You bring enormous pleasure to my life.

xi



Acknowledgements




1

Narratives about theory It is now almost commonplace to begin writing a book about narratives with some reflection on the pervasive nature of storytelling. Indeed, the extensive presence of stories that we tell and that get told to us manifests itself in a myriad of ways. From high brow fiction to soap opera, conversational stories to those disseminated in the media; from private contexts to public domains including education, the law or medical environments, narratives appear in many guises. They perform multiple functions in creating and sustaining our understanding of ourselves, others and the world around us. Likewise, gender acts as a primary axis, which from birth onwards plays vital roles in processes of identification, social relationships and in various contentious ways might be implicated in macro-level structures of power. In recognition of their prominence, both areas have become increasingly influential as subjects of academic interest and theorization. In recent decades the development of narratology and of feminism have accrued status, so that while by no means unproblematic, both must be considered as significant intellectual forces at the end of the twentieth century. Given the wide-ranging and sometimes overlapping terrain covered by both, it is perhaps unsurprising that points of intersection between the two should be found. Narratives are human activities, and the assumptions and procedures involved in their telling and analysis are human constructions, which a feminist would argue must entail a consideration of gender. ‘Feminist narratology’ is the umbrella term which embraces the exploration of narrative from this point of view. More specifically, Warhol defines this as ‘the study of narrative structures and strategies in the context of cultural constructions of gender’ (cited in 1



Feminist Narratology in Context

Mezei, 1996: 6). The terms used in feminist narratology may be familiar: ‘narrative’, ‘gender’, ‘feminism’, ‘context’ and so on. But these are not transparent, or universally understood within or between different fields of study. In isolation, and even more so in combination, defining what is meant by ‘feminist’ or ‘narratology’ is not a straightforward matter, for both have been vigorously debated and are characterized by increasing pluralism. One of the central tenets of feminist narratology is its insistence on contextualization as a means of understanding the interplay between gender and narrative form. From the outset, then, it is all the more important to contextualize what might be understood as feminist narratology. I begin by tracing the developments in narrative theory and in feminism over the last three decades, showing how significant trends in both fields helped contribute to the emergence of feminist narratology and have subsequently shaped its landscape.

Developments in narratology Feminist narratology began in the mid-1980s, as its name suggests, from within the domain of narratology. Histories of narratology abound. In a simplified form, these highlight overarching trends which chart the movement from its structuralist origins which became academic currency in the 1970s, to a postclassical critique and evaluation in the 1980s, through to a postmodern diversification of both theoretical stance and interdisciplinary application from the 1990s onwards. Glancing through these stories, those surveying the development of narrative theory in these decades reiterate time and again its expansion across boundaries and its diversification (Gibson, 1996; Bal, 1997; Currie, 1998; Kreiswirth, 2000; Richardson, 2002). It would seem that while Barthes confidently asserted the all-encompassing nature of narrative as ‘international, transhistorical, transcultural’ (1977: 79) with a universalizing assumption that has since been challenged, narratology’s transdisciplinary application does indeed contribute to the illusion of narrative being everywhere. The analysis of narratives in their various forms and functions has been given as evidence of this, including those studies which have been done in the fields of visual arts, history, philosophy, theology, psychoanalysis, sociology, legal studies, pedagogy and the natural sciences. Not only has it been applied widely, narrative theory has itself been the subject of interdisciplinary revision as it has drawn upon theories from outside its original home in literary studies, including those as wide-ranging as postcolonialism,



2 Literary and Linguistic Approaches to Feminist Narratology

cognitive science, hypertext, and most pertinent to this study: gender theory (Richardson, 2000b: 169). Feminist narratology is also part of the reconceptualization of narratology that has taken place in these years, a process that has been summarized as a move from ‘coherence to complexity’ (Currie, 1998: 2). This is expressed as a transition from early structuralist studies of narrative, which approached texts as objects that could be analyzed in a ‘scientific’ and systematic manner to a postmodern position where different kinds of questions are asked of the analysis, with the recognition that both the text and the analysis thereof cannot be ‘neutral’ but may be ideologically loaded in some way (Kreiswirth, 2000: 296). Currie locates the period of transition as centering around the 1980s, exactly the period in which feminist narratology began. The move away from an abstract, apparently universal narratology can be seen in the co-existence of various definitions of the field offered by eminent theorists who published key texts in this decade. On the one hand, the ‘narrow’ view continued to hold prominence. Prince, who distinguishes clearly between the concerns of narratology, criticism and interpretation, exemplifies this. At the opening of his Narratology (1982), he defines the project as follows: Narratology examines what all narratives have in common – narratively speaking – and which allows them to be narratively different. It is therefore not so much concerned with the history of particular novels or tales, or with their meaning, or with their esthetic value, but rather with the traits which distinguish narrative from other signifying systems and the modalities of these traits. (pp. 4–5) However, alongside this, ‘broader’ understandings of narratology also came into use. For example, Bal’s Narratology defines the remit of narratology as including interpretation, with narratology ‘conceived as a set of tools, as a means to express and specify one’s interpretive reactions to a text’ (1997: x, quoting from the Preface to the 1985 edition). Universalism is eschewed, as she goes on to write that ‘the theory presented here holds no claim to certainty’ (Bal, 1997). Tensions between these narrow and broad positions continued through the following years. It is not that one superseded the other (for example, Prince (1995a and b) continues to work from the same stance), but by the end of the 1980s, some critics at least were expressing dissatisfaction with this ‘narrow’ perspective and implying that a change of emphasis was needed. The turbulence implied in these debates is reflected in the



Feminist Narratology in Context 3

two collections of articles, ‘Narrative Revisited’ Volumes I and II, published in Poetics Today (1990). Amongst these papers, similar concerns were aired as the writers reconsidered narratology. Structuralist approaches were critiqued for their assumed universalism (Brooke-Rose, 1990: 283) and there was recognition of the need to take account of contextual factors (Chatman, 1990: 309). Narratology, it seemed, was undergoing an identity crisis. It was from within the midst of these debates and changes that the concept of feminist narratology emerged with Lanser’s seminal paper (1986) bearing that name. Despite earlier precursors (Brewer, 1984; Warhol, 1986), it is this work by Lanser that is credited as the principal impetus for integrating feminism with narratology. In proposing the interchange between these two theoretical paradigms, her approach is in many ways typical of the wider expansion of narratology at this time, which began to be ‘energized by a variety of perspectives’ (Herman, 1997: 1049), for example, including psychoanalysis (Brooks, 1984; De Lauretis, 1984), philosophy and sociolinguistics (Pratt, 1977). More specifically, Lanser’s central argument for a two-way, mutually beneficial dialectic between feminism and narratology illustrates important theoretical trends within narratology. She proposed that ‘narratology . . . might be altered by the understandings of feminist criticism and the experience of women’s texts’ (1986: 342), calling into question the corpus of texts from which the models of narratology had been derived and so adding her voice to those which challenged structuralist universalism, especially from a contextualist point of view. Significantly, Lanser’s critique politicized the terms of this debate, suggesting that the assumption of universalism was not neutral, but founded on an androcentric bias. Just as universalism was argued to be abstract, undesirable, even untenable, this became replaced by an interest in ideology which was felt increasingly through narrative studies and debated in literary criticism more generally (Currie, 1998: 4–6). If Lanser’s goal to use feminism to challenge narratology reflected and contributed to these wider currents in the revisions of narratology in the 1980s, it brought to a head debates about the very definition and purpose of narratology itself. The contrast between the narrow and broad views of narratology posed a dilemma for those responding to Lanser’s call: What exactly was entailed by feminist narratology, and as a concept was it even plausible? For those in the ‘narrow’ camp, where narratology remained an abstract system of categorizing narratives qua narratives, feminist narratology was rejected out of hand as either irrelevant (Diengott, 1988) or at the least something to be seriously




questioned (Prince, 1995a). For those working with a ‘broad’ approach, who argued that narrative theory also embraced matters of meaning, context and evaluation, feminist narratology was welcomed and proved a fruitful means of discussing both ‘women’s texts’ and narratology itself. To some extent, this debate now seems somewhat redundant in the acknowledgement that the line between ‘interpretation’ and ‘analysis of narrative form’ is fuzzy, if present at all (Bal, 1997: 10–11). Indeed, the proof of the pudding, as it were, is clearly evident in the wealth of research that put feminist narratology to work (summarized by Richardson (2000b) and discussed in more detail in Chapter 3). Yet the debate remains important for clarifying the project of feminist narratology. Writing more recently, Warhol reviews feminist narratology, observing that ‘feminist narratologists never tried to replace the structuralists’ systems with alternative systems of their own’ (2003: 24–25). Feminist narratology is not then a separate set of feminist narrative models, but is better understood as the feminist critique of narratology (in its broad sense) which operates on the basis of feminist applications of narrative theory to a range of texts that goes beyond the corpus originally drawn upon by the early structuralist work.1 The relationship between ‘narratology’ and ‘feminist narratology’ is a complex and uneven one. The latter cannot be said to have substantially altered narratology in its narrow sense, and Lanser notes that ‘narratology “proper” ’ (which I take to be roughly synonymous with what I have termed here as a narrow view of narratology) has remained disinterested in feminism (1999: 168). In many ways, feminist narratology is typical of the revisionist work in postclassical narratology, which did not necessarily reject the models of structuralist narratology wholesale, but integrated them with other theoretical perspectives. Paradoxically, while feminist narratology provides an essential critique of narratology (in both its broad and narrow senses), it has also illustrated the very strength of narrative theory as this has been developed, and in some cases revised, through the analysis of a range of texts of interest to feminist critics. Thus in both its origins and development, feminist narratology need not be seen in opposition to narratology. The superficial separation of the two overlooks the point that in some senses, feminist narratology is a product of ‘narratology proper’ as a revisionist response and in carrying to logical conclusion its expressed intention of examining narrative models against an ever-increasing range of texts. This feminist critique does not declare the end of narratology. Quite the opposite, the challenge is vital, that is both necessary and life-giving as it sustains and continues to transform narrative studies.





Looking back, now some twenty years later, the development of feminist narratology is clearly embedded in the debates and complex transitions within narratology of the 1980s. But if feminist narratology has needed clarification, what the modifier ‘feminist’ means also deserves careful consideration. As a political movement, feminism, or more appropriately feminisms, has never been unified and whose development is full of controversy, reversal and change (Bryson, 1999: 5). Throughout the last three decades, work that has gone under the name of ‘feminism’ has been characterized by tensions regarding the object of feminisms’ attention, its goals and how these might be expressed. At the turn of the millenium, ironically, within this diversity, there is common agreement that now ‘one cannot even attempt generalizations about the state of feminist theory in the West, let alone in the rest of the globe’ (Braidotti, 2001: 381). This multiplicity and ambivalence is far-reaching and presents itself in a range of ways. From an internal perspective, there is an ongoing argument which sets up an opposition between feminist theory, a focus on discourse, a search for universals on one hand, and activism, feminist practice and experience on the other (Fraser and Nicholson, 1990; Walters, 1996; Hanssen, 2001).2 The relative prominence and specifics of these terms has varied according to time and context, and, as we shall see, has particular importance for determining what feminist narratology might mean. Indeed, feminists have been divided over the contrasting values and meanings associated with the central terms of their struggles, such as equality, difference and diversity. As Guerrina notes, these have been developed in specific social and political contexts, sometimes with quite differing aims and outcomes (2001: 35). Looking at feminisms from a disciplinary perspective, matters are equally complex. While women’s studies have now become institutionalized, Harding and Norberg suggest that its influence has expanded across intellectual areas with few exceptions (2002: 943). At the same time, feminisms have also been characterized by drawing upon and adapting a whole spectrum of theoretical paradigms, notably marxism, psychoanalysis, philosophy and deconstruction. Since Lanser’s (1986) paper, time has moved on, and, as Kavka puts it, that ‘Feminism ain’t what it used to be’ (2001: 1). Precisely how that change is articulated, however, is fraught with difficulty. Various terms have been suggested to indicate the contrast between feminisms at the beginning of the new millenium and that which preceded them some thirty years or so earlier. ‘Third wave’, ‘post feminism’ and ‘postmodern



Developments in feminism

feminism’ have all gained currency, but at least the first two of these have been fiercely contested. Given the potential to interpret ‘post feminism’ as implying the demise of feminism itself, (an illusion I would not want to perpetuate), I choose not to use this term here. Despite the potential racism that might be invoked by the wave model (Springer, 2002), I will use ‘second wave’ to refer to period of feminist development in the 1960s and 1970s. This is not because I do not want to acknowledge the important contribution that women of colour have made to feminisms’ various projects, but because feminist narratology emerged out of the context usually delimited by this term. However, I am not convinced that ‘third wave’ is the most useful way of referring to the most recent shifts within feminist thought and activity. It seems to me that the developments from the late 1980s into the 1990s (approximately speaking) are less a separation from the work immediately preceding it and more of a continuation. To describe these as a separate wave suggests a value-laden attempt to contrast the changes that have taken place, where the earlier ‘feminism’ is viewed nostalgically as a period of consensus, which is either lauded for its political edge or berated as outmoded in the light of postmodern theory that followed it. As even this brief discussion has shown, in attempting some meta-narrative of feminisms’ developments, the ideological significance of exclusion or imposing a false sense of uniformity of experience is felt acutely. However, it is inescapable that significant changes have taken place between the mid-1980s and the beginning of the new millenium. Kavka summarizes it as A shift from explaining women’s subordination in terms of a single constraining system – whether we call it capitalism, patriarchy, biology, or even language – to focusing on the discursive, material and cultural differences that make up the being or becoming of women. (2001: xiii) The corollary of this was that the category of ‘woman’ and even ‘gender’ itself was radically reconceptualized in the light of poststructuralist theorizing. Simultaneously, groups within these overarching categories rightly drew attention to the diversity of their own experiences, especially as these were refracted through the lenses of race and/or sexuality. While the view of second wave feminism as unified is undoubtedly illusory, the increasing pluralism within feminism is signalled by the language that was used to identify individuals, groups and theories. Books from the earlier period used the descriptor, ‘feminist’,




but by the 1990s this had diversified into a proliferation of modified terms: ‘black feminism’, ‘lesbian feminism’, ‘radical feminism’, ‘liberal feminism’, ‘queer feminism’, ‘black male feminism’, ‘French feminism’, ‘American feminism’, ‘global feminism’, ‘ecofeminism’ and so on. Because these developments are both derived from and symptomatic of postmodern influences, the label I choose to refer to the later period is the last of the three: ‘postmodern feminism’. In the face of this internal and diachronic multiplicity, what did feminist narratology mean in the 1980s, and what could it mean now? The emergence of feminist narratology in the 1980s is clearly influenced by trends in feminism from this period. First, it is perhaps no surprise to find the analysis of narratives as a potential subject for feminist critique, given the increasing prominence of literary and cultural theory in feminism during these years (Kavka, 2001: xiv). Hand in hand with this shift into focusing on the discursive as a site of gender politics, feminism in academia followed the second wave exposure of sexism by means of challenging the notion of academic objectivity itself (Nicholson, 1990: 3). Together, these two dynamics prepared the way for a feminist critique of the apparent neutrality of narrative theory, which in turn could be used to articulate the feminist interpretation of texts. Importantly, feminist narratology as it is described in Lanser’s early paper does not deal with just any kind of discourse, but is situated within the domain of literary studies. So when Mezei (1996) questions whether feminist narratology refers to feminist literary theory as opposed to ‘feminism(s) – a movement’ (p. 6), I would answer that its origins align it more closely with the former, for Lanser hoped that her paper might influence ‘American departments of literature’ (p. 342), and is informed by the literary interests of feminists such as Showalter (p. 343). Feminist narratology thus has similar emphases to strands of the feminist literary criticism of this time, with its insistence that ‘all interpretation is political’ (Belsey and Moore, 1989: 1) and a desire to draw attention to the work of women authors that had been neglected in earlier studies. Within the diversity of American feminist literary criticism in the 1980s, the influence of French theorists such as Cixous, Irigaray and Kristeva also began to be felt (and contested) as their works were translated and disseminated. This is less obvious in Lanser’s (1986) paper, but the vision of a specifically feminine discourse as envisioned through psychoanalytically derived models of language, desire and sexuality also informed later feminist analyses of narrative theory and literature (Winnett, 1990). While there are important differences between these currents of feminist literary criticism, both gynocriticism




and écriture feminine imply a model of analyzing ‘feminine’ texts as different from ‘masculine’. This opposition is interpreted within a hierarchical framework of power relations, where the recovery of ‘feminine’ alternatives is seen as a form of emancipation and cultural revision. The feminism of 1980s feminist narratology is thus very much a product of its time, imbued with a second wave agenda of overturning inequality towards women, but framing this within a specifically literary context. By the end of the 1990s, concepts of gender had radically altered under the influence of poststructuralist theorizing, especially that laid out by Judith Butler (1990, 1993). No longer was the focus on a binary model of difference but on the ways in which gender might be performed in various ways and in different contexts. This change in theoretical stance is also reflected in developments within feminist narratology, which likewise embraced queer theory in a more fluid understanding of the ways that gender might be implicated in narrative structures and response (Lanser, 1995, 1999; VanHoosier-Carey, 1997; Warhol, 2002, 2003). However, even though the concepts of gender might have changed, the nature of feminism found in feminist narratology has still retained its theoretical emphasis. Because the discussions of cultural context and of women’s ‘experiences’ are filtered through the analysis of texts (usually literature), feminist narratology has had little to do with application in feminist practices. We might then question how to define the politics of the ‘feminism’ in feminist narratology. Mezei’s distinction between the feminism of ‘feminist literary criticism’ and ‘feminism(s) – a movement [. . .] a political position’ suggests a telling separation. While I think that it is too strong to deem the feminist interpretation of texts to be apolitical, there are certain limitations to its political potential, or at least its politics need to be understood in particular ways. In the case of feminist narratology, this is rooted in the points of compatibility and difference in the goals of feminism and narratology. One of the fundamental principles of feminism is that it should analyze and interrogate gender relations (Flax, 1990: 40). From the early days of feminist narratology onwards, the tools of narrative theory have proved useful as a means of achieving this. Although it may appear that narratives are everywhere, they are by no means universally expressed. Issues of representation, patterns of desire and agency, the right to speak and how that speech is evaluated are all critical matters for both gender relations and narrative analysis. However, feminism is not content with simply describing such matters. It also seeks to change those arrangements through political action (Flax, 1990: 40; Bryson, 1999: 5). Herein




lies the heart of the dilemma, for narratology makes no such claim. It is true that narratology has become politicized insofar as it is now concerned with ideology in important ways, but there are restrictions on the ways we might understand this as political action. Mitchell writes about a similar quandary emerging from the integration of feminism and psychoanalysis. She says, As a political practice in search of a political theory, it [feminism] can use concepts and arguments from elsewhere to analyze its own object – the position of women – in the relevant contexts, but it cannot convert these concepts and arguments into political ones in and of themselves. (2001: 16) Similarly, neither can feminist narratology necessarily transform narrative analysis into political activism. In part, the problem lies with the object of study. It is not that the analysis of literary texts is devoid of attention to material and social relations. However, feminist narratology is concerned with texts and how these might relate to their various contexts, whereas ‘feminism(s) as a movement’ (to borrow Mezei’s distinction again) might be said to be more intent on altering the contexts themselves. Indeed, analyzing gender as it is important to narrative theory, for example in terms of the characterization, plot structure or voice, does not necessarily neatly translate into an understanding of how the oppression of women operates in its various contexts and how this might be challenged. Related to this is the separation that occurs between academic disciplines and environments outside these. As Hartmann et al. argue, certain disciplines tend to be aligned with particular activities and occupations, and therefore perhaps the kind of theory and analysis coming out of departments of literature should not be expected to inform other domains such as public social practice (1996: 938). More generally, the institutionalization of feminism has led to a situation where discussion of feminist theory may become isolated from practice and activism as the audience becomes limited to professional academics.3 Thus the location of feminist narratology within academia might well limit its potential to operate as a form of social change. Currie expresses similar doubts about the political potential of narrative theory more generally. He writes, ‘If the role of the intellectual is to speak the truth to power, as Gramasci formulated it, the evidence suggests that power is not listening’ (1998: 8). Do these limitations mean that the project of feminist narratology should be abandoned, and the metaphorical baby thrown out with the




bath water of political activism? It is with the desire to answer in the negative that this book has been conceived. I hope that the preceding discussion has made it abundantly clear that the concept of feminist narratology is not without limitations. But this does not mean that it is of no use. Indeed, with the burgeoning of narrative studies and the recognition that feminism remains critically important both in theory and in practice, there is perhaps a greater need than ever to take account of the ways in which these two important fields of study might contribute to one another.

Postmodern feminist narratology The aim of this book is to affirm the potential of feminist narratology, not only to recognize the valuable work that has been carried out in this field over the last two decades, but also to understand the ways in which these are situated within specific historical and theoretical contexts. This review forms the basis for reconceptualizing the interchange between feminism(s) and narrative theory in order to argue for its ongoing usefulness to both fields. In particular, I want to suggest that feminist narratology must take into account the changes that have taken place in both fields as a result of postmodern influences. As the meta-narratives outlined earlier suggest, both share a similar progression away from a concern with universalizing categories into diversification and intersectionality where revisionist narratology questioned the assumptions of an abstract system and in feminism, post-structuralists challenged the construction of universal categories of womanly experience. The problems of universalism were countered by the inclusion of multiple perspectives that resulted in the fracturing of narratology and feminism into increased pluralism that went hand in hand with a focus on localized contexts. It has become obvious that no one set of data, model of analysis, methodology or theory alone can achieve the goals of feminism or narratology. Rather each will have particular strengths and limitations. For example, the localized studies of gender and language use developed through Eckert and McConnell-Ginet’s (1998) concept of Communities of Practice are extremely helpful for looking at micro-level interchanges but are less concerned with exploring macrolevel forces that may also shape gendered behaviour. Likewise, the analysis of plot structure may be useful for looking at the global arrangement of a text, but may tell us little about other important aspects of its organization, such as focalization or characterization. Moreover, this diversification is politically important in drawing attention




to the specific experience of groups or individuals who might appear otherwise excluded. However, plurality in itself is not without its problems and in turn the limitations of an atomistic approach have now begun to be raised. In both feminism and narrative studies there are increasing demands for integration between areas, so that this multiplicity might be useful (Eckert and McConnell-Ginet, 1999; Richardson, 2000b; Herman, 2001). At the beginning of the new millenium, feminist narratology now faces the challenge of responding to this dual call for both diversification and integration. This challenge has implications for the feminist theory, narratology and narrative texts that are brought together in the continuing work of feminist narratology. This book responds to the call for a more holistic approach to feminist narratology that not only looks beyond its existing boundaries in terms of data, theory and discipline but also brings this multiplicity together in the belief that synthesis can be productive. The approach I take is essentially comparative. There are several benefits that arise from this. First, as Porter Abbott puts it, the operation of ‘suddenly seeing your old familiar terms from a new disciplinary perspective can be salutary precisely because the differences of field are so great’ (2000: 260). Such a shift may allow us to re-evaluate the meaning of terms that become used so readily as to be taken for granted, such as ‘gender’, ‘context’, ‘text’. Second, this comparison opens the possibility of dialogue between researchers working in different areas. At its most basic, this would include the acknowledgement of other relevant findings and arguments, which in their points of commonality and difference may help to clarify or alter our own perspective. Perhaps more contentious and far-reaching is the integration that transposes models, methodology or theory across boundaries. This may be useful, but should always be handled sensitively so that distortions that may arise from decontextualization are at best minimized, or at least acknowledged. The integration that I am proposing is between feminist narratology and the work on narrative that has taken place in language and gender studies. The interdisciplinary combination of literary and linguistic studies is in many senses nothing new, as the expansive field of stylistics (or literary-linguistics, as it has come to be known) indicates. The rationale for such a synthesis has been laid out by many scholars who have gone before me (summarized recently by Simpson, 1997; and Toolan, 1990, 1998, 2001). In general terms, my work proceeds from the premise that the study of literary texts using frameworks from linguistics, using those literary texts in the process of formulating linguistic models or comparing different text types and perspectives





Given the fact that language, the object of study in linguistics, is also the fabric all literary texts are made of, linguistics and literary criticism should be almost natural allies in the endeavour to develop approaches to literary texts that do justice to the multifaceted nature of literary texts and to their dependence on language as a medium. (2002: 7) While there are important differences between the two, the early desire to equate literary and linguistic approaches with a corresponding subjectivity and objectivity has been replaced by an acknowledgement that ‘interpretation and persuasion’ are common to both, and the belief that the possibilities suggested by this way of reading are not only valid but rewarding (Toolan, 1990, Chapter 1). The benefits of integrating findings from linguistic research, applying linguistic models and drawing in non-literary texts are manifold and of specific relevance to feminist narratology. In its evolution from narratology (in its narrow sense) to the more wide-ranging ‘narrative theory’, the focus has moved away from privileging literature and gone on to include a variety of other narrative texts (Genette, 1990). It is then perhaps all the more surprising that feminist narratology has largely adhered to its roots in narratology proper and feminist literary criticism and mostly kept to analyzing literary texts. There are exceptions to this, such as Bal’s important work on the visual arts (1990, 1997, 1999) and Warhol’s engagement with cultural studies (1999, 2003). Nonetheless, feminist narratology has remained a literary concern, without taking much notice of parallel research that has been carried out in other fields. Yet considerable work in sociolinguistics and Critical Discourse Analysis (henceforth CDA) has also interrogated similar issues to those posed by feminist narratology, such as whether or not gender makes a difference in the storytelling styles used by groups of women and men (Johnstone, 1990; Abney, 1994; Georgakopoulou, 1995; Coates, 1996) and the ways in which sexism may continue in the representation of women (Clark, 1992; Mills, 1995; Caldas-Coulthard, 1996). Incorporating approaches from these perspectives is an important extension of feminist narratology. On the one hand, it widens the type of data that should be analyzed to include narratives not conventionally considered as ‘literature’ such as conversational stories, media reports and writing of various kinds



may prove mutually beneficial. As Gymnich, Nünning and Nünning point out,

by children. In so doing, it moves feminist narratology closer to its original project of testing narrative theory against a fuller range of texts produced by, and about, women. These ‘non-literary’ narratives are also important for providing alternative examples of what ‘context’ might entail, as they function in a range of ways in relation to their ‘real world’ environment. Perhaps in part because of this, sociolinguistic analyses of narratives have tended to be more aware of the multiple ways in which gender might intersect with a range of other variables. In turn, this may also have significant implications for the way in which ‘gender’ is treated in the analysis of literary texts. Finally, studies in sociolinguistics and CDA have tended to place greater emphasis on the dialectical relationship between the narrative and its social or ideological context than has happened in the text-immanent focus of literary feminist narratology. If the apparently diffuse relationship between social context and literary analysis limits its feminist potential, then these more contextually sensitive approaches to narratives in use may alleviate this in some measure. The integration of work from linguistics also complements the methodology traditionally used in feminist narratology. One of the great strengths of feminist narratology has been its application of close reading. However, it is important not to generalize about the wider implications of gender relations on the basis of individual texts. Thus Prince writes, In postclassical as well as in classical narratology, there has been little extensive empirical or experimental (cross-cultural) exploration of these or similar problems and I think that we – ‘literary’ narratologists – are too often inclined to take locally suggestive and persuasive arguments about understanding and responses for generally true statements. (2001: 233) It is precisely this inclusion of empirically oriented methods (for example, using corpus-based approaches or analyzing wider samples of data) that may be of use to feminist narratology. Of course, these should not be valourized as superior to close reading. It is clear that empiricism in itself may be used to oversimplify and reach abstract generalizations just as readily as other methods (Bing and Bergvall, 1998: 503; Eckert and McConnell-Ginet, 1999: 194). Nonetheless, they provide an important addition to feminist narratology, enabling us to gain a greater sense of what ‘actually is’ and so building a wider picture




of the ways in which gender might be important in the telling and reception of narratives. Feminist narratology has also yet to take on board fully the ramifications of the changes in feminist theory. The implications of reconceptualizing gender as a fluid, performative notion rather than as a fixed category may be felt in far-reaching ways. The very terms of the analysis have been destabilized. While second wave feminists made a distinction between biological sex (male/female) and socially constructed gender (masculine/feminine), later theorists, notably Judith Butler, have collapsed this distinction, arguing that sex itself is socially constructed and is better understood as a continuum of possibilities rather than as a binary pair. Throughout this book, I have continued to use the term ‘gender’ to refer to the ways in which individuals may identify themselves or be identified (mostly) as women or men, and to describe the behaviour and beliefs that enable these norms to be reproduced or contested. However, I also recognize that other gendered alternatives may be possible that go beyond this two-way distinction, for example, through gender blending or in cyborg imagery. More recent studies in feminist narratology have gone beyond the binary model of gender used in earlier years (Lanser, 1995, 1999; Warhol, 2003), but it is notable that the corpus of texts that has attracted attention has stayed within Lanser’s early directive that feminist narratology should analyze texts by women, mostly British or American authors. It seems less easy now to pose ‘texts by women’ as a simple, homogenous category, and important differences between women need also to be recognized. Typically, these other differences have highlighted alternative sources of oppression or exclusion, such as race, class or sexuality, and may also include others such as age, historical period or academic discipline. While the data that I look at is still influenced by the eurocentric origins of feminist narratology, I have tried to include examples that go beyond this and come from other cultures and periods. Thus the narratives I look at include work by Afro-American author Toni Morrison, medieval Japanese monogatari, and stories by Maori children in New Zealand. Where the data remains British, I have wherever possible signalled its cultural specificity rather than assuming it to be universally representative. Moreover, to study ‘texts by women’ assumes that masculinity is itself a homogeneous concept to which femininity is opposed, so I have also included narratives by or about men in order to examine points of similarity and difference within and between categories. Most fundamentally, reconceptualizing gender means that the kinds of research questions asked by and of feminist narratology




should be revised. Bing and Bergvall provocatively highlight the many ways in which traditional questions about language and gender may reinforce a binary model of gender relations (1998). Their critique might also usefully be applied to much of the feminist narratology that has gone on in the last 20 years, which at its very inception was founded upon the premise that women’s narratives might be different from men’s. Looking at the situation now, it may be time for feminist narratology to question its questions. Because the assumption of difference has been so influential, much of this book’s thrust is to challenge that in various ways. This includes testing hypotheses of difference against various data and demonstrating the restrictions of categorizing narratives, their tellers and readers in this way. I go on to look at the ways in which stereotypes and power relations based on gender difference continue to be important in narratives of different kinds. So while difference remains a central issue for feminist narratology, the questions we ask about it need to be of a different nature now and avoid perpetuating false oppositions and preconceptions. Feminist narratology cannot be understood as a single, unified entity. In many ways, it never has been. The original concept was forged from two distinctive fields and the subsequent studies have been described as ‘eclectic’ (Mezei, 1996: 7). However, at this point in its development, the multiplicity it might embrace is more expansive and increasingly necessary. In this pluralistic re-envisioning, feminist narratology has become thoroughly postmodern, and in line with the postmodern feminist theory projected by Fraser and Nicholson (1990). They describe this as ‘comparativist rather than universalizing, attuned to changes and contrasts rather than covering laws’, and which ‘would replace unitary notions of woman and feminine gender identity with plural and complexly constructed conceptions of social identity, treating gender as one relevant strand amongst others’ (pp. 34–35). However, despite the political motivation that impels such multiplicity, this must also guard against the fragmented separatism that ignores possible points of commonality between groups. So the postmodern feminist narratology that I want to argue for is not only plural, but also integrative. Like the metaphor of a tapestry used by Fraser and Nicholson (1990: 35), postmodern feminist narratology is made up of many strands which may not only be differently coloured but also have points at which they cross and combine. It is only as they are woven together that the overall picture of feminist narratology can begin to be formed.





My description of the strands in this tapestry is organized around particular questions that feminist narratology might consider. I begin in Chapter 2 by interrogating the very basis of feminist narratology: problematizing the ways in which gender might be related to narrative form. The focus of the discussion is the feminist narrative theory that has been derived from psychoanalytic theory, especially the feminist revisions of Brooks’ (1984) work on narrative desire. I critique the binary opposition of Brooks’ ‘male plot of ambition’ and the subsequent feminist alternatives that were suggested on the grounds of their essentialist and universalizing tendencies. Instead, I review these gendered plot types in the light of current debates about narrativity and examine how the stylistic features that contribute to varying degrees of narrativity may be found in work by both male and female authors. I also question the eurocentric focus of this strand, and suggest that gender cannot be treated in an abstract manner, and should not ignore other important influences like that of race, as seen for example in the work of Toni Morrison. This opens the debate concerning the extent to which a specific narrative feature might be interpreted as ‘feminist’. Because of its landmark status, Lanser’s (1986) paper is reviewed in detail in the next chapter. Here I acknowledge the ways in which it is influenced by the characteristics of second wave feminism and begin to explore the limitations of a binary model of gender for feminist narratology. Following on from the previous chapter, I provide an alternative feminist revision of plot analysis, this time drawing on the work of Michael Hoey’s (2001) culturally popular patterns of organization. Again, the analysis and discussion is cross-cultural as I compare a selection of texts from the medieval period that prescribe gendered patterns of moral behaviour. These include Middle English romances and Japanese monogatari. Lanser’s work stresses the importance of contextualization for feminist narratology, but here I question how easy it is to examine this relationship between text and context, and how this might operate within certain restrictions for literary texts. Chapter 4 provides an alternative perspective on the influence of context and shifts to a sociolinguistic perspective. The aim of the study in this chapter is to test the claim that women and men might tell stories differently against some empirical data. Labov’s (1972) highly influential work on ‘danger of death narratives’ has also been critiqued for its gender specific bias. My work uses the Labovian framework to compare the oral narratives told by women and men about an equally



About this book

personal and emotive topic: that of childbirth.4 What is striking about this set of narratives is not so much the points of difference between the stories told by the women and men but the over-riding similarity in their use of an anecdotal structure. Thus I question the extent to which gender might be considered as a salient variable, and how feminist narratology might need to take into account the fine-grained ways in which this intersects with other contextual parameters. Chapter 5 addresses an area that has been given relatively little attention in feminist narratology: that of reader response. The study looks at a group of readers’ interactions with a fictional hypertext. In so doing, I also raise the need to take account of the influence of a narrative’s medium, an issue that has been taken for granted in the primacy of print culture. The analysis revisits the concepts of narrativity and plot structure outlined in earlier chapters, but here uses these to examine the strategies that the readers used for processing this transitional narrative. While the discussion provides some much needed empirical data regarding women’s and men’s actual reading practices, it also highlights the dangers of assuming difference based on a simplistic, binary categorization of ‘women’ and ‘men’. In particular, I argue for the need to examine intra-category variation, which might be explained for reasons that have little to do with gender at all. The following two chapters examine cases where assumptions about gender norms continue to be important in the way narratives are used and evaluated. Chapter 6 considers the ways in which power relations between women and men continue to be of importance, particularly in terms of media reports. Using perspectives from CDA, I examine the ways in which the British and American press has constructed narratives of success, failure and stability for two particularly high-profile figures: Cherie Booth/Blair and Hillary Rodham Clinton. In my discussion of these narratives, I argue that feminist narratology should go beyond the description of texts alone, and pay greater attention to the ends to which these narratives might be put. In the reporting of Booth/Blair and Rodham Clinton, isolated narratives might suggest that these two women are feminist role models, but examining the wider context reveals the extent to which these narratives serve patriarchal ends of building their respective husband’s reputation and so political power. In applying a corpus-based approach, I argue for the need to complement the close reading of feminist narratology with analyses of a much broader number of texts. Chapter 7 turns to another area which has not yet been embraced fully by feminist narratology and examines the relationship between




storytelling style, age and gender in a cross-cultural sample of narratives told by British and New Zealand children. Once again, the consequences of storytelling style are of considerable importance here, where the ability to conform to culturally specific patterns of narrative prescribed within educational contexts may have social and academic consequences for the children involved. Again I argue for a contextsensitive approach to the analysis of both gender and narrative form. The findings discussed in this chapter provide an important challenge to feminist narratology’s assumptions that it is women writers who might diverge from normative patterns, leading to questions about where such assumptions might have been derived from. Brian Richardson points out that even while feminism has had a far-reaching transformation on narrative theory, the subsequent reconceptualizations ‘continue to be debated and refined’ (2000b: 168). In this book, the vision of postmodern feminist narratology is one in which these debates may take place across rich and varied terrain, exposing sometimes surprising areas of similarity and difference. At the beginning of the new millenium, I argue that in its eclectic and postmodern plurality, feminist narratology is still very much a going concern, relevant to those who consider themselves to work with narrative theory from a wide range of perspectives. The following chapters demonstrate some of the ways in which this might be so.




2

Psychoanalysis, literary theory and feminism The strand of feminist narratology with which I am concerned in this chapter began to emerge from points of intersection between the crosscurrents of feminism and psychoanalysis within literary criticism in the 1980s. The rising influence of psychoanalysis within literary studies generally, and more specifically in the study of narrative, was in part a response to the limitations of structuralism (Amiran, 2000: 212). Peter Brooks’ highly influential work on narrative dynamics (Reading for the Plot) begins by asserting this position, describing it as ‘an attempt to move beyond strict allegiance to the various formalisms that have dominated the study of narrative in recent decades’ (1984: xiii). Like many others, Brooks’ work attempts to go ‘beyond’ these limitations and engages with psychoanalytic models of dynamics and desire, which Clayton suggests became ‘one of the master tropes of contemporary criticism’ (1989: 34). Brooks’ study of narrative is ambitious, combining the analysis of ‘how narratives work on us, as readers’ (p. xiii) with a study of the ‘text itself as a system of internal energies and tensions, compulsions, resistances, and desires’ (p. xiv), resulting in what he describes as an ‘erotics of art’ (p. xv). Throughout, he explicitly draws on the work of Freud (and to a lesser extent, Lacan). Thus he describes the trajectory of plot structure as ‘a form of desire’ which is likened both to Freud’s notion of Eros and, in its move towards closure, the death instinct. There is something intuitively attractive about Brooks’ arguments in which we might recognize the ‘pull’ of a narrative during the act of reading. However, his analysis is not unproblematic and has been critiqued for a range of reasons, not least of which is its apparent gender bias, a critique that should be understood within its wider context. 20



The Question of Gender and Form

The feminist revision and appropriation of psychoanalysis has also been seen as part of a reaction to formalism in literary criticism (Gallop, 1987: 314). Writers such as Hirsch (1989) and Chodorow (1989) argued for the use of psychoanalysis in achieving the feminist ends of understanding how ‘we become gendered subjects’ (Hirsch, 1989: 18). They challenged the phallocentrism in male psychoanalytic theorists’ work, asking how women’s desires and pleasure might be articulated and reconstructing the narratives of absent female figures from the Oedipal myth (De Lauretis, 1984; Yaeger, 1988; Hirsch, 1989). This formulation of feminine ‘alternatives’ (a term used by both Hirsch, p. 2, and Yaeger, p. 3) can be understood within the second wave feminist work of recovering texts by women who had previously been excluded from the literary canon (for example, Showalter, 1978; Spender, 1989). It is within this practice of exposing sexism in literary criticism that Brooks was critiqued by those who had interests in either psychoanalysis or narrative theory or both (for example, Clayton, 1989; Hirsch, 1989; Winnett, 1990). Hirsch is the most exacting and lengthy in her discussion, arguing that Brooks’ work is ‘gender blind’ (p. 53), founded upon a ‘sexual and psychological model which is exclusively male’ and analyzing ‘no novels written by women’ (p. 54). Brooks’ description of plot is indeed gender biased, and constructed as male. I quote the following passage at length, as his gendering of plot is most explicit here. The ambitious hero thus stands as a figure of the reader’s efforts to construct meanings in ever-larger wholes, to totalize his experience of human existence in time, to grasp past, present and future in a significant shape. This description, of course, most obviously concerns male plots of ambition. The female plot is not unrelated, but it takes a more complex stance toward ambition, the formation of an inner drive toward the assertion of selfhood in resistance to the overt and violating male plots of ambition, a counter-dynamic which, from the prototype Clarissa on to Jane Eyre and To the Lighthouse, is only superficially passive, and in fact a reinterpretation of the vectors of plot. The ambitious heroes of the nineteenth-century novel [. . .] may regularly be conceived as ‘desiring machines’ whose presence in the text creates and sustains narrative movement through the forward march of desire. (1984: 39) This description of plot raises some problematic issues. First, Brooks elides the content of the narrative, ‘the ambitious hero’, with the cognitive process of ‘the reader’s efforts’. Not only does this association lack any



The Question of Gender and Form 21

empirical evidence, it is unclear who exactly is referred to as ‘the reader’ (my emphasis). Instead, in Brooks’ equation, this universalized figure is assumed to be male, like the hero. Second, the content of the narrative (its hero and its theme of ambition) also seems to be the basis for Brooks’ description of the plot as ‘male’. This does not explain why ambition should be constructed as ‘masculine’ or what it is about the plot structure itself that should be gendered. Finally, Brooks glosses over the question of the ‘female plot’, its characteristics and relationship to the ‘male plot of ambition’. This is consigned parenthetically to a footnote, described in negative terms as ‘an endurance: a waiting (and suffering) until the women’s desire can be a permitted response to the expression of male desire’ (p. 330). His focus ‘of course’ and ‘obviously’ is the ‘male plots of ambition’. In so doing, he fails to address the questions raised by Hirsch, who asks, ‘How do women writers, and how do women’s stories, relate to this conception of plot so fundamental to nineteenthcentury realism? Can this plot pattern accommodate a female subject? Is there space for female difference here?’ (1989: 54).

Female/feminine/feminist alternatives In answering those questions, feminist writers have proposed alternative narrative patterns that are derived from the analysis of texts by women writers, mostly from the nineteenth and twentieth centuries. These narrative patterns are described in explicitly gendered terms, often celebrating specifically female experience grounded in the body. This includes Hirsch’s outline of female revisions of oedipal plot patterns, Anderson’s description of a female orgasmic plot pattern (1995) and Winnett’s use of birth and breastfeeding as metaphors for narratives of female pleasure (1990: 509). Common to all of these is the assumption that female alternatives are characterized by narrative difference from the ‘male plot’, and that these differences are manifest in the sequential ordering of the narrative patterns. This results in a binary opposition between the male and the female alternatives. In summary, these represent the ‘male plot’ as linear, with a trajectory of rise, peak and fall in narrative tension ending with a defined point of closure. In contrast to this, the ‘female plot’ is non-linear, repetitive and resistant to narrative closure (Winnett, 1990: 509); contains multiple climaxes or none at all (Anderson, 1995); and is likened to the lyric form which is organized by pre-oedipal timelessness (Wallace, 2000: 177). As Homans (1994) writes in her eloquent and perceptive essay, this takes for granted that the sequential narrative I have described as the ‘male plot’ has considerable restrictive




potency for the representation of women and ‘changing traditional narrative, or finding alternatives to it, can have emancipatory effects for women’ (p. 6). The site of feminist struggle, then, is not only the exposure of phallocentrism in literary theory but becomes transferred onto the narrative form itself. This desire to ‘break the sequence’ (to borrow again Woolf’s much cited phrase) finds echoes in écriture feminine, a set of writings also influential in feminist literary criticism in this period. Here it is not only narrative sequencing, but also language ‘as a universal structure’ (Belsey and Moore, 1989: 14) that was taken to be the focus of feminist resistance in the Anglo-American adoption of the writings of Cixous, Irigaray and Kristeva. Like the feminist theories of narrative outlined here, it is also rooted in psychoanalysis, and articulates an alternative feminine language that continuously contests boundaries. For example, in ‘Sorties,’ Cixous links (although does not define) ‘a feminine practice of writing’ with excess (p. 108), non-linearity (p. 110), resisting closure (p. 113), the female body (p. 113) (including specifically lactation, p. 112), and going beyond linguistic boundaries, ‘running away with syntax, breaking the famous line’ (p. 115). Irigaray’s description of feminine language is also associated with the body in her description of female autoeroticism, and is similarly non-linear in ‘her language in which “she” goes off in all directions. Contradictory words seem a little crazy to the logic of reason, and inaudible for him who listens with readymade grids, a code prepared in advance. In her statements – at least when she dares to speak out – woman retouches herself constantly’ (1981: 103). The parallels between the alternative practices of writing described by these French writers and the characteristics of the ‘female plot’ in some cases led those working on narrative to draw upon these theoretical works. This was not uncritical, but retained an emphasis on the positive valuation of female alternatives as a form of feminist resistance. For example, Yaeger (1988) situates her work in response to ‘French feminist writers’ but supplements this with an account of how ‘real, historical women’ have used unconventional narrative strategies to challenge literary traditions (p. 20). Hirsch, likewise, positions herself critically in response to the limitations of ‘French feminism’, but provides a reading of the Demeter myth that resonates with many features of écriture feminine (1989: 35). Wallace draws on Kristeva’s notion of ‘Women’s time’ in her questioning of desire, narrative and gender (2000). The influence of these French theorists’ writings in this strand of feminist narratology is perhaps unsurprising, given the shared interests in psychoanalysis, feminism and literary texts. However, this




attempt to forge an alternative analysis of ‘female plots’ is problematic for a range of reasons. First, in its rejection of formalism and their ‘move beyond’ the ‘identification of minimal narrative units and paradigmatic structures’ (Brooks, 1984: xiii), these psychoanalytically derived theories of narrative also lose a degree of precision that might be considered preferable from a linguistic perspective. As a result, the description of the plot types becomes highly metaphorical. This is found in the modifying labels given to the plot types themselves, for example ‘of ambition’, ‘endurance’ (Brooks, 1984), ‘female orgasmic plot pattern’ (Anderson, 1995: 109); and ‘blood narratives’ or ‘milk narratives’ (Daemmrich, 2003: 217). Similarly, the use of concepts of ‘desire’, ‘arousal’ and the more graphic images of birth and lactation given by Winnett (1990) to describe narrative sequences may retain some intuitive validity but fail to operate as adequate descriptions of narrative form. As Cameron and Kulick point out, the metaphor of desire is itself difficult, and the psychic processes and ‘inner states’ with which it is associated are not in themselves observable phenomena (2003: 98). This would not be problematic, were these descriptions not associated with terms like ‘plot’ and ‘narrative’ which within narratology and everyday usage also carry with them meanings that entail structure as well as content, such as ‘action structure’ (Ronen, 1990) and ‘connections between events’ (Concise Oxford Dictionary). Dealing in metaphor alone raises the question of what criteria are being used to distinguish between these plot types. Even with descriptions that appear to be more concerned with elements of narrative structure, such as ‘multiple climaxes’ and ‘cyclical time’, it is not clear precisely what it is about the plot the critic is referring to. This difficulty stems in part from confusing the form of the narrative and its thematics. For example, Brooks’ description of the male plot of ambition given earlier is prefaced by the statement that ‘Ambition provides not only a typical novelistic theme, but also a dominant dynamic of plot’ (1984: 39). Likewise, the narratives that Wallace deals with foreground desire as a theme (2000), and Anderson’s discussion of plot is explicitly linked to texts that focus on heteronormative sexual activity (1995). It seems then, that the theme is simplistically equated with the ‘plot’ or with the sequencing of the narrative described as its ‘dynamics’ or ‘trajectory’. The slippage between thematics and form is then exacerbated when labelling the plots in this way is abstracted from its original context. Thus Winnett responds by rightly pointing out the sexism inherent in Brooks’ descriptions of narrative desire and dynamics, but does so by concentrating on its underlying




metaphor of physiological experience, rather than dealing with the details of Brooks’ specific thematic content (1990: 508). Consequently, the alternative she offers is at one remove from the content of actual texts and based on a metaphor of female experience, which she extends to ‘meet the Masterplot on the terrain it has staked out for itself – that of form’ (p. 509). The consequence of this interplay between gendered thematics (either in explicitly biological terms or as socially constructed) and these descriptions of plot is that it is difficult to ascertain whether it is the form, thematics or both that is considered to be gendered.

A linguistic approach: Narrativity and gender The first step in disentangling form and theme is to provide a more precise way of analyzing the texts. This requires a replicable set of parameters against which narrative forms might be compared. As Mills (1995) points out in her critique of ‘French feminists’, the linguistic analysis of texts is a vital step in elucidating the validity of claims to gender difference in language, which here includes narrative structure. In order to do this, I reframe the description of ‘male’ and ‘female’ plots through the lens of narrativity. ‘Narrativity’ is an issue that remains central to questions about the very status of narrative itself, as Prince’s interrogative list demonstrates (2000). However, McHale’s (2001) discussion of this topic indicates that what precisely constitutes ‘narrativity’ is open to challenge. ‘Is narrativity a matter of kind or degree?’ he asks (p. 165). I follow Herman (1997) in distinguishing between narrativehood as that which concerns the criterial factors necessary to classify a text as a narrative (a matter of kind) and narrativity as a scalar notion related to ‘factors that allow narrative sequences to be more or less readily processed as narratives’ (p. 1048) (a matter of degree). I am interested in the second of these. I am not trying to argue that the male plot of ambition is a narrative and that the female plot is not. Instead, I am suggesting that these typologies can be compared by examining the forms of narrativity that they invoke. The perception of narrativity may be understood as a complex relationship between linguistic features ‘in’ the text and extra-linguistic features ‘outside’ the text, such as the reader’s world knowledge that may be shaped by specific cultural contexts in various ways (Herman, 1997; Hoey, 2001). These linguistic features ‘in’ the text include those traditionally studied in narratology. This incorporates the way in which temporal order is marked and organized, the type of characterization employed, the status of a text’s narrator and/or focalizor, and the global




organization of the text, for example as indicated by the presence of Problem-Solution or Goal-Achievement patterns and arranged into a teleological progression.1 These factors should not be regarded as a definitive list (criterial to a text’s status as a narrative as opposed to some other text type) but rather as an open set, which may be drawn upon in combination to create narrativity of various kinds and degrees. Implicit in Herman’s definition of narrativity is the understanding that any given text may be perceived as having a higher or lower degree of narrativity and also that the reader is able to recognize these degrees as being more or less like ‘narrative’. From this we might infer that certain combinations of these linguistic features occur in recognizable patterns which, as they then become grounds for processing a text ‘as narrative’, are perceived as high in narrativity. What is striking about this in relation to feminist narratology is that there seems to be a correlation between degrees of narrativity and the stereotypical gendering of plot where the ‘male’ and ‘female’ plots exhibit strong and weak narrativity respectively. The narrativity of the ‘male plot of ambition’ can be characterized by the presence of the following factors. The temporal sequence is clearly defined and does not disrupt the perception of ‘real world’ chronological order, enabling the hero and reader ‘to grasp past, present and future in a significant shape’ (Brooks, 1984: 39). The characterization is consistent and coherent, typically focused through a central protagonist who is usually male and expresses and achieves their desires. Brooks’ description of ambition as ‘a force that drives the protagonist forward, assuring that no incident or action is final or closed in itself until such a moment as the ends of ambition have been clarified, through success or else renunciation’ (p. 39) parallels the progression of the Problem–Solution and Goal–Achievement patterns, which reach closure when the problems have been satisfactorily resolved and goals achieved. Finally, the teleological progression of this plot with its characteristic rise, climax and fall in tension can be described in linguistic terms using Longacre’s (1983) description of the surface marking used in the ‘anatomy of plot’. This reaches a climax, which is indicated in the textual detail of the narrative using factors that Longacre describes as ‘Peak marking’. This comprises heightened vividness, a shift in orientation (either spatially or in the focalizing figure), a more or less dramatic form of speech presentation, or a change in the size of the textual units being used (1983: 26–38), summarized by Marley as ‘disruption of the norm’ (1987: 83). The narrativity of the ‘female plot’ may contrast with these narratological features in a range of ways. For example, the degree of precision with which any factor is used may change, or a typical pattern is disrupted in




some way (for example, time sequence might be disrupted by anachronies, problems are left unresolved, there are multiple passages marked as Peak and so on). These characteristics are not fixed in a stable configuration, but are unified in their contrast with the narrativity of the ‘male’ plot. Because the features of the ‘female’ plot diverge from the well-known pattern, it may then also be perceived as lower in narrativity. Why this equation of narrativity and plot types should occur is an important question. One possible response is to note that the ‘male plot of ambition’ has strong similarities with profiles such as Labov’s (1972) narrative outline. As is now well recognized, these classical models might be re-evaluated in terms of potential gender bias in the corpora from which the pattern is derived (see Prince, 1995a, 1996; Cazden, 1997). Thus it could be argued that our sense of what constitutes a ‘well-formed’ narrative (that is, high narrativity) is conditioned by a body of stories told by male speakers about stereotypically masculine experiences (in the case of Labov’s 1972 study, fight narratives told by male adolescents). More generally, Hoey’s approach to textual organization foregrounds the cultural situation of each pattern (2001). Given that gender may be understood as socially constructed as opposed to biologically determined, it might also be possible that cultural values of gender have some relationship with patterns of organization used in telling stories. For example, in certain situations it might be that the ability to solve problems or achieve goals is restricted to privileging one gender above another, as in stereotypically represented heroes and heroines where it is only the male characters that are allowed to follow quest-like progressions. Of course, this would be limited to specific genres and cultural contexts, but nonetheless the possible correlations between gender and predictable patterns remain a factor that may also contribute to the alignment of high narrativity with ‘male plots’ and the relegation of ‘female plots’ to weak narrativity. However, while I have initially transposed the models of ‘male’ and ‘female’ plots onto a polarity of strong and weak narrativity, this binary pairing deserves much closer and more critical re-examination. The following analysis of some contemporary fiction provides a further test case for exploring the relationship between narrativity and gender.

Narrativity and gender in Flesh and Blood The fiction of contemporary author Michèle Roberts is the focus for the analysis and discussion of these issues. Her work is of particular interest for two reasons. First, it is rich in gender-related themes and imagery




including identity, sexuality, religion, the woman as artist and the mother–daughter relationship. In its interrogation of these issues, it has been categorized as ‘feminist’ (Rowland, 1999; Falcus, 2003). Second, it contains examples of interesting experimentation with narrative form, which as Roberts herself comments is closely related to the content of her novels (1994: 171). As such, it might be expected that her narratives provide interesting points of connection between the two areas. This is not to claim that Roberts is representative of all feminist women writers, or that there is an unequivocal relationship between form and meaning. Rather, her work provides an appropriate example that might form the starting point for exploring issues of gender and narrativity. I begin with her novel, Flesh and Blood (1995). Like her other short stories and novels, Flesh and Blood returns to many of the themes described above, but it is the piece where the relationship between gender and identity is most explicitly foregrounded and deconstructed in a form of ‘gender blending’ (Neubert-Köpsel, 2001: 176). Roberts challenges the ways in which gender is marked through surface appearance (especially clothing), biological features, sexuality, naming and social practices, doing so in a manner that creates narrative uncertainty. The novel takes the form of a series of short narratives, each of which is named after a character and arranged symmetrically like a hall of mirrors around a central section headed ‘Anon’. These chapters construct a quasi-history of mothers and children. However, rather than forming a straightforward genealogy, these chapters are linked in ways that reflect upon one another in inverse and contradictory ways that blur fantasy and reality, raise unanswered questions and critically destabilize the notion of a fixed gendered identity. The narrativity of Flesh and Blood is weakened in ways that support the thematic focus on gender and identity. The most striking disruption occurs in the construction of character. While not criterial to narrativehood in the same manner as temporal succession, characterization is an important means of narrative coherence. As Longacre puts it, ‘You must keep track of who does what to whom or the story falls flat on its face’ (1974: 362). In Flesh and Blood, not only do the cast of characters change from chapter to chapter but also the reader’s sense of narrative coherence is challenged by the shifting and ambiguous description of the characters’ gender, disrupting the frames of reference that would enable the reader to process the text as a coherent narrative. The ‘same’ character narrates the opening two chapters, headed ‘Fred’ and ‘Freddy’, which are linked to the closing chapter ‘Frederica’. In the first chapter Fred is described as male: ‘I saw myself: one of those men, framed in




black, on a police poster; hollow-eyed, wild-haired, staring’ (p. 1). On the run, the criminal ‘Fred’ takes refuge in a dress shop where he disguises his male identity in a feminine dress. ‘My large cotton handkerchief and my discarded tie, well stuffed in, did duty for breasts. I smoothed ballooning skirts, put my hands up to check my hair, drew my licked fingertips along my eyebrows. I kicked off my shoes: too big and heavy’ (p. 5). Even though the shopkeeper assumes ‘Fred’ to be a woman, the reader is not led to believe that Fred has undergone a biological change in sex. In the next chapter, ‘Freddy’, the narrator appears to be the same character, narrating ‘his’ childhood. However, some ten pages later, it is clear that ‘Freddy’ is biologically female, marked by the onset of menstruation (p. 15) and now named ‘Frederica Stonehouse, in a bulky tunic and pullover, and with a stash of sanitary towels in a brown paper bag in my briefcase’ (p. 17). As Neubert-Köpsel and her students comment, this results in a sense of narrative disruption. My own experience of reading this text with groups of undergraduate students reinforces this. Without exception, these students also commented on the disorientation they felt when discovering the gender change of what at this stage in the novel appears to be the central narrator and main protagonist. Fred/erica is not the only character given a dual gender representation. In the chapter called ‘Felicite’, the reader is introduced to a character who is coded as male by the choice of name (George Mannot), dress, textual references with masculine pronouns ‘he’, ‘his’, ‘him’, and by biological definition as the narrator implies that George is responsible for Felicite’s defloration (p. 54). However, in the reciprocal chapter, ‘Georgina’, ‘George’ is revealed to be a woman in disguise. I’m sure that G. could point to either George or Georgina, even as it concealed them [. . .] she embarked upon a daring masquerade, like something in one of the modern novels we know she read and loved, in which she [. . .] allowed herself two selves, two lives, or was it three? Her life as a woman in London, her life as a man in France, his/her experience at the moment of crossing over from one to the other and back again. (pp. 155–156) With both of these instances, Roberts uses the reflective pairing of the chapters and the ambiguous characterization to challenge the reader’s notion of gender as a stable, defining feature of a character’s identity. Instead, the binary gendering of Fred/erica and Georg/ina as exclusively male or female is shown to be illusory, shifting according to the interpretation of dress codes, naming and sexuality.




The thematic focus on gender and identity is supported by alterations to the teleological progression of the text. As discussed earlier, the correspondence between surface Peak marking and a climactic moment of crisis is typical of the high narrativity found in the ‘male plot of ambition’. At first sight, Flesh and Blood does have a chapter that appears to be a narrative Peak, in that the narrative style switches from standard prose into a poetic form without use of conventional syntax. Mamabebe love you are here with you together us now over and over so non-stop mamabebe so wanting you born this love us so close skinskin talking heartbeat belonging with you allowed love home flesh my mamabebe our body singing to you so beautiful love listen mamabebe listen (p. 109). As a break from the established textual norm, this chapter does indeed seem to be like a ‘zone of turbulence’ (Longacre, 1983: 25). Critically though, this chapter does not seem to correspond to the apex of a trajectory of problematic tension that marks a move into resolution. Instead, in content, it forms the ultimate point of gender deconstruction, describing a paradise of pre-Oedipal union with the mother that stands in contrast with the absence or critique of mothers in both preceding and subsequent chapters. In this sense, it forms the two-way interface against which all the other chapters are arranged (in structure) and reflect (in content). As such, the plot of Flesh and Blood is not unidirectional, moving towards a single resolution. Rather, the individual chapters provide multiple reflections on the mother–child relationship and the possible constructions of gender. The narrator hints at this multidirectional nature of the narrative in the first chapter, saying, ‘I’d always had a gift for inventing stories. I shuffled them like a pack of cards in my head’ (p. 1), where the chapters, like the cards in the pack, do not have to be followed in a fixed linear sequence. The privileging of the centre section contrasts with the silencing of the ‘death blow’ that closes the ‘male plot of ambition’. Indeed, Flesh and Blood resists closure. Each of the chapters ends with the narrator giving an Abstract (cf. Labov, 1972) of the next story, which conventionally opens a narrative sequence rather than closing it. The final chapter is no exception, ending with the words, ‘So we walk through Soho and into the next story’ (p. 175). Theoretically, it would be possible to read this as an invitation to begin rereading the text, treating the first chapter as ‘the next story’ and so restarting the sequence towards the central section again. The opposition to the ‘death blow’ in this resistance to




closure is supported by the narrator’s equation with storytelling and physical survival. In the first chapter, ‘Fred’ says, ‘Scheherezade had told stories, night after night, to save her life [. . .] I wasn’t sure if my stories could save my life or my mother’s.’ (pp. 6–7). Not only is this central section metaphorically associated with surviving execution, its content celebrates the beginning of life with the union between child and mother. Here the stylistic features emphasize the ‘lyric timelessness’ that is associated with women’s bodies in feminist theories of narrative (Wallace, 2000: 177). The language in ‘Anon’ corresponds to many of the features characterizing Cixous’ description of écriture feminine. It is described as literally written on the body, ‘on your blissful skin the hands of the masseuse play a writing game’ (p. 108), and specifically within the female body ‘swimming in our waters’ (p. 109). It avoids syntactic closure at the level of both the word and the sentence, we is one whole undivided you/me broken now mended you/me restored mamabebe our body of love pickedup putbacktogether. (p. 109) In this way, the language itself mirrors the contesting of boundaries, be they psychic or socially constructed along gendered lines, and explicitly associates this with a celebration of the female body in the figure of the mother. In summary, both in its lack of coherent characterization and multidirectional association between chapters and in the use of stylistic features parallel to those suggested by écriture feminine, the narrativity of Flesh and Blood contrasts with that of the ‘male plot of ambition’. These narrative features might be interpreted as particularly appropriate to the feminist content of this text, which explores the mother–daughter relationship and challenges the binary definition of gender as either male or female. However, I would strongly resist arguing that this interpretation defines a fixed ‘female plot’. This simplistic equation of form and gender is deeply problematic for reasons examined in the following discussion.

Narrativity and gender in Beloved The first criticism I want to make in relation to the labelling of narrative alternatives as ‘male and female’ is that it universalizes the categories ‘female’ and ‘male’ and obscures other important differences that might




lie within groups of women and men. This tendency might be taken as symptomatic of feminist theory in the late 1960s to the mid-1980s, which, as Nicholson argues, ‘tended to reflect the viewpoints of white, middle-class women on North America and Western Europe’ (1990: 1). The feminist narratology derived from psychoanalytically derived theory has been implicated in this limited perspective and rightly labelled as ‘elitist’ and ‘Eurocentric’ (Homans, 1994: 8). For example, Winnett prefaces her analysis of an alternative to the ‘masterplot’ of narratology with the caveat that ‘it does not even hold up as a model for all ‘female’ narrative’ (1990: 508), but what she analyses is the writing of white, British women, without any discussion of what influence these parameters might have. Hirsch acknowledges similar limitations in her study (1989: 9). However, both these critics, like others who draw on the psychoanalytic framework of desire, go on to use language which slips into universalism, even when attempting to resist this. For example, Winnett attempts to reconceive narrative ‘in terms of an experience (not the experience) [her emphasis] of the female body [my emphasis]’ which she relates to an unqualified ‘Female experience’ (p. 509). In using metaphors derived from apparently neutral, but biologically oriented experiences, these theorists further divorce their analysis from the particularities of potentially differing groups of women. Within feminism generally, this replication of universal structures led to women from marginalized groups arguing for the recognition of Black feminism (Collins, 1990: 116) and of a womanist perspective (Mori, 1999: 2), both of which despite their differences seek to give value to the experiences of women of colour. Homans (1994) points out that outside the context of white, eurocentric literature, the trajectory of the ‘male plot of ambition’ need not be given a negative value or assumed to be restrictive. In her analysis of Zora Neale Hurston’s narratives, she shows how this African-American writer puts this apparently ‘male’ form to positive feminist use. I want to reinforce Homan’s argument for the inclusion of race in an analysis of feminist narratology by turning to the example of Beloved (1987), by Toni Morrison. Like Michèle Roberts’ work, Beloved explores the relationship between mother and daughter.2 Morrison’s work has also been analyzed for its characteristic use of non-linear narrative structure (Cutter, 2000: 61). However, to simply synthesize these two aspects and interpret this as another example of the ‘female plot’ is to overlook a critical aspect of Morrison’s work: its relationship to historical context and Morrison’s status as an African-American writer. To disregard the importance of race would be to ignore Morrison’s attempt to narrate the absent history of slavery in America, an absence that exists at personal, social




and national levels (Matus, 1998: 103–104). This ignorance would be to perpetrate that racist absence at an academic level, an act that would (or at least should) now be almost unthinkable in the light of the work of Black feminist writers and much greater appreciation of diversity within feminist theory and practice. It would fail to address the political dimension of Morrison’s writing, in which she critiqued the ‘early image of Women’s Lib [as] an elitist organization made up of upper-middle-class women . . . not paying much attention to the problems of most black women’ (Morrison, 1971: 16, cited in Mori, 1999: 2), and her argument that African-American literature should not be judged ‘solely in terms of its referents to Eurocentric criteria’ (Morrison, 1989: 22, cited in Cutter, 2000: 63). I turn now to compare the narrativity in Beloved with that in Flesh and Blood and the ‘male plot of ambition’. The characterization in Beloved, like Flesh and Blood, is destablized and hence might be understood as weakening the perceived narrativity. This disruption takes place in the central character of Beloved. Instead of altering her gender, the text continually offers multiple interpretations for Beloved’s identity, failing to resolve realist interpretations (she is an abused girl, escaped from a white man) with the fantastic (she is the ghost of Sethe’s murdered daughter) and the symbolic (she represents the ‘sixty million and more’ to whom the book is dedicated). When Paul D. asks Denver if Beloved is her sister, Denver replies ‘At times. At times I think she was – more’ (p. 266). Although this multiplicity contrasts with the ‘male hero’s’ drive of ambition to ‘construct meanings in ever-larger wholes’ (Brooks, 1984: 39), it is difficult to argue that this characterization is a feminist feature. Rather, it would seem that this ambiguity makes possible the more open interpretation of the text that incorporates confronting not only Sethe’s personal past but also a national history. The manipulation of time in Beloved also resists straightforward linearity. Sethe describes this elision of the past and present, saying ‘I was talking about time. It’s so hard for me to believe in it. Some things go. Pass on. Some things just stay’ (p. 35). Throughout the text this is realized through the analepsis that continually interrupt the present. In the first chapter alone, the time moves from the narrative present (1873) when Baby Suggs is dead, back to when Baby Suggs is still alive (the next paragraph), forwards with a proleptical trace alluding to the arrival of Beloved ‘if she’d come, I could make it clear to her’ (p. 4), further back with an analepsis of the memory of Sethe copulating with the engraver to pay for Beloved’s gravestone (p. 5) and then further back in time to memories of Sweet Home (p. 6). This oscillation between the




narrative present and analeptical fragments of the past might be interpreted as another contrast with the ‘male plot of ambition’ in which ‘past, present and future’ is grasped in a ‘significant shape’ (Brooks, 1984: 39). However, to interpret this simply as a ‘female’ alternative again is too limiting. As Matus writes, this temporal discontinuity ‘conveys the nature of the traumatic past’ and forces the reader to return to the text for further readings (1998: 111), and in so doing achieves the goal of repeatedly telling the story. This narrative survival has a particular meaning, which rather than the abstract return to the maternal space in Flesh and Blood does not allow the unspeakable past of black slavery to be forgotten. Beloved and Flesh and Blood share one other similar narrative feature. This is the central section, which radically breaks with the established textual style in the rest of the text. Like ‘Anon’, Beloved’s monologue avoids conventional syntax and is radically discontinuous. Likewise, it is possible to interpret the passage as a description of the pre-oedipal child and its mother. For example, Beloved says, ‘I am not separate from her there is no place where I stop her face is my own and I want to be there in the place where her face is and to be looking at it too a hot thing’ (p. 210). Superficially, there are similarities between this and the metaphorical fluidity of écriture feminine, or the semiotic with its absence of linear time, ‘all of it is now it is always now’ (p. 210). However, to stop at this interpretation is to deny the alternative readings that relate the fragmentation to the traumatic experience of those who made the Middle passage, or to the possible identity of Beloved as a sexually abused woman. Moreover, Skinner (1995) suggests that the stylistic features used here are a deliberate move on Morrison’s part to extend the African-American oral tradition by mimicking the griot’s performance (pp. 63–64). The point we have reached is that both Flesh and Blood and Beloved have features of narrativity that contrast with the unilinear chronology, coherent characterization and Problem–Solution patterns of the ‘male plot of ambition’ described by Brooks. However, I have argued that a similarity in form should not lead to imposing a similar interpretation of the texts, particularly one which reduces those points of compatibility to a universal ‘female’ alternative. To do so does not acknowledge the potential variation in the experiences and narratives of different groups of women, and the relationship this might have with other significant factors such as historical context or race. I suggest that this issue is a problem inherent in the use of psychoanalysis within narrative theory, which has the potential to derive narrative patterns from Greek myths




as if they were a ‘timeless psychic order’ (McCormick, 1992: 187). This is true for both the narrative theorists who are charged for their sexism and those who attempt to address this from a feminist standpoint, for (with the exception of Homans, 1994) none of the theorists who have been surveyed in this chapter have related their analysis to the specific experiences of women in particular cultures. In abstracting ‘female’ qualities, or reconstructing the story of absent ‘mothers’, there needs to be greater attention to the potential cultural variability of these terms.3 The interrelationship between form, content and interpretation is a complex one of interdependence. As Homans (1994) has shown, the same form can be interpreted as patriarchal restriction in relation to one context or feminist emancipation in another. I have shown how the same features of narrativity in Flesh and Blood and Beloved provide some evidence for relating these stylistic features to feminist concerns, but there may be additional interpretations that point to interrelated themes, such as the oppression of race and the recovery of national identities. The discussion so far has challenged the universal nature of the opposition of ‘male’ and ‘female’ plots. I now take this interrogation of form and content further and question the extent to which gender can be related to the narrativity of different plot types, if at all.

Narrativity and gender in Une Glossaire/a Glossary I begin by considering another text by Michèle Roberts, Une Glossaire/ a Glossary (henceforth Une Glossaire), which is found at the close of her collection of short stories, During Mother’s Absence (1993), and breaks more radically still from the narrativity used in the ‘male plot of ambition’. Une Glossaire is striking as a short story because, as the title suggests, it is structured around sections following an alphabetical list where a French term is then glossed with an English interpretation. Like conventional glossaries, this text is placed at the close of the volume of short stories, but unlike its conventional equivalents it contains substantial passages of continuous prose that may be read without reference to the rest of the anthology and is not just a supplementary list. Indeed, as Toolan (2001) writes, we may debate whether or not Une Glossaire should be classed as a narrative at all (pp. 8–9) and this has important consequences for the way readers might then interpret it. Even if we conclude that the narrativehood of Une Glossaire can be established as positive, the very act of challenging its status suggests that its degree of narrativity is also questionable.




A range of narratological features may be listed that either weaken or strengthen the narrativity of this text. Perhaps most obviously, the structuring framework of the glossary itself weakens the degree of perceived narrativity. Glossaries would normally be classified as a text type identified by Hoey as a discourse colony. His definition of this genre is based on an analogy with the natural world where mainstream texts are like people and colony texts like beehives or anthills. As he writes, the crucial organizational feature of colony texts is that the ‘component parts do not derive their meaning from the sequence in which they are placed’ (2001: 74–75). Typical examples of colonies include texts as diverse as dictionaries, address books, examination papers and shopping lists. As Hoey’s description suggests, this genre can be understood as diametrically opposed to mainstream discourse such as narrative.4 In terms of the schemas and world knowledge evoked by the structural similarities and labelling of Une Glossaire as a glossary, the patterns of organization do not correspond to anything that might suggest high narrativity. On the contrary, the allusion to a discourse colony might be interpreted as an almost anti-narrative impulse. Une Glossaire’s similarity with this non-narrative form is apparent when it is compared with the nine characteristic properties of a colony text. As the following summary indicates, it is found to match six or arguably seven of these, including the definitive categories (1), (2) and (5).

The nine properties of a discourse colony

Une Glossaire

1. 2. 3. 4. 5. 6. 7. 8. 9.

YES YES YES NO YES YES NO YES YES

Meaning is not derived from sequence Adjacent units do not form continuous prose There is a framing context No single author and/or anon One component may be used without referring to the others Components can be reprinted or reused in subsequent works Components may be added, removed or altered Many of the components serve the same function Alphabetic, numeric or temporal sequencing

Source: Adapted from Hoey, 2001: 88

Given Une Glossaire’s colony-like attributes, it would seem that there is ample evidence to argue for a case of extremely weak narrativity. The following passage gives two of the glossary entries (headed ‘Calvados’ and ‘Camembert’) that exemplify this tendency.




Calvados Calvados Brandy made from cider apples. Sipped, in tiny glasses, after Sunday lunch. Made on the Spriets’ farm, and named after Falstaff, who is supposed to have once stayed there. Camembert Camembert The round cheese with a rough chalky coat which Brigitte scrapes off before serving it. We eat it for breakfast, often also at lunch and supper. When you buy a Camembert, Brigitte instructs us: you must smell it and press it to see if it is really ripe. The inside gushes out, yellow and runny, over the plate. You just eat it. You don’t fiddle about with it. None of this Robert Carrier nonsense about deep frying it in breadcrumbs. (pp. 150–151). What these entries also demonstrate, and as Toolan concludes in his brief discussion of this text, this is clearly not a genuine discourse colony: An alphabetized list is sharply non-narrative. But since this is such an annotated and embellished list, one in which the whole becomes clearer as the later items on the list are reached, we still have grounds for saying that here is a narrative, albeit of a non-traditional kind (p. 9). In addition to this, we might suggest that it is not just the embellishment of the entries that make this colony unusual. Rather, there are specific features that activate what Porter Abbott (2000) terms the ‘narrative motor’, which then causes us to interpret the text as a story as opposed to some other text type. The first of these is the references to time and transition found in many (although not all) of the entries which may be used by the reader to construct a linear chronology into which the various memories of the narrator may be slotted. This allusion to temporality is perhaps the most fundamental property that allows the text to be processed as narrative, being, as Ricoeur (2000) puts it, the ‘ultimate referent’ for narrativity. However, the very form of the temporal markers in Une Glossaire is interesting and suggestive of weak narrativity. There are relatively few dates that anchor the chronology to a specific time span (1967: p. 161; 1979, 1986: p. 132). Instead the transition of time is described through the biological and social changes in the individuals remembered. Many of these relate to the development of womanhood in the narrator; for example in the entry headed ‘Cinema’ the narrator says,





Similar transition is described in terms of the physical decline of various family members, the ultimate mortality of the beloved aunt, grandmother and grandfather, for example ‘Grandmère suffers increasingly from heart disease . . . this plump, capable, white-haired person suddenly becomes a chronic invalid, her head poking forward and down, her belly swelling, her whole body thickening’ (p. 161). Alongside these physical markers of time are references to familial and religious festivals that mark rites of passage, described in entries such as Assomption, Moisson, Noces and Noces de diament. As such, the temporality appears to be rooted in human experience rather than abstract, and to be cyclical rather than purely linear in character. Further stylistic features such as the use of present tense, the absence of temporal references from many of the entries and the deliberate evocation of ‘timelessness’ particularly in relation to food, ‘these timeless meals’ (p. 147), might also be interpreted as weakening the chronology and hence narrativity of this text. The second feature that contributes to the narrativity of Une Glossaire is the pervasive references to a cast of characters. In Une Glossaire, these references to individuals exist both within and between entries and form a network of cohesive identity chains that, like cohesion more generally, ‘encourage a hearer or reader to interpret the combined utterances as belonging together in some way’ (Hoey, 1991: 10). In addition to this agent orientation, we might also include the spatial and geographical references to setting, which in some cases become elided with the identity of particular characters. For example, the narrator says, I think my grandmother is the house. The storehouse of food and plenty. The cupboards stocked with sheets and tablecloths, tea and dinner services, boxes of silver cutlery and black-handled knives brought out for best occasions. She is still there. (p. 161) These references to character or setting are greater in number than those of temporality, for while many entries are achronological, agent orientation or setting appears in all. As such, this might be taken as the primary means by which narrative coherence is facilitated and the



After the age of ten, after the onset of menstruation, breasts, frizzy hair and spots, I appear less on film. . . . All through my childhood and adolescence I draw obsessively: pictures of lovely naked women. I still do; but now they are allowed to have big swinging breasts, curving thighs, genitals, furry triangles of hair. (p. 153)

desire to read the text as an organic unity set against the non-narrative dynamic of reading the colony entries in isolation. These features of high narrativity are found in greater and lesser extent in the various glossary entries. Those containing a dense concentration of narrativity might be thought of as core entries that establish narrative coherence while those that are weak are like satellites that are only loosely tied to the reconstructed narrative frame. This distinction between core and satellite entries once more destabilizes the degree of narrativity where movement towards a narrative pattern in the core entries is interrupted by the satellites. Overall then, even while there are crucial stylistic features in Une Glossaire that suggest we are in the presence of narrativity (McHale, 2001), the disruption of these features in various ways weakens or subverts this narrativity significantly. The question then remains of whether or not this weak narrativity is associated with gender. Superficially, Une Glossaire may be aligned with the characteristics of écriture feminine that contrast with the linear teleology of the ‘male plot of ambition’, described in detail earlier. Roberts herself seems to hint at the connection between language, the body and the mother–child relationship. For example, in the entry ‘Français’ the narrator describes her native tongue as ‘My mother’s tongue. My mother-tongue, that I take in along with her milk. . . . My tongue lapping at pleasure’ (p. 157). The implied fluidity of this description is perhaps mirrored by the non-linear flow of the text, organized around the headings expressed in the language of the ‘mother tongue’ and deliberately marked as feminine by Roberts’ alteration of the grammatical gender of ‘glossaire’.5 The importance of the female body in marking temporal progression, along with the cyclical references to temporality might be interpreted as a way of ‘writing the body’. From a rather different stance, the use of agent orientation and spatial reference as the primary means of creating narrative coherence might be seen as a feminist move too. The significance of space is articulated prominently in the opening entry of Une Glossaire where the narrator chooses not to create a history but instead to ‘write a sort of geography. To reclaim the past’ (p. 133, emphasis added). As Homans (1994) notes, this spatializing move towards image rather than linearity is found in many texts by women. In particular, the allusions to a house as a structuring metaphor might in Cosslett’s (1996) terms seem like a gesture towards matrilinealism where the patriarchal dynasty of narrative forms is challenged by alternative kinship structures, placing the ‘Daughters of the House’6 at the centre of this narrative form. However, I would argue that there are fundamental and far-reaching problems with this overly simplistic mapping of narrative form and




feminism. From another standpoint, we might argue that the degree of narrativity is not really to do with gender at all. Rather, the weak and anachronous temporality and privileging of agent orientation as a mode of narrative coherence are simply appropriate to the telling of a story that celebrates human relationships in order to avoid the narrative-like changes that lead to death. The going beyond boundaries, cyclical and arbitrary ordering which defeats the ‘death drive’ of the (male) plot dynamics has less to do with a feminist emancipation of a narrative form and more to do with the specific content of this text, which if it reached its final conclusion would express the point of final (mortal) absence that the narrator states she wishes to avoid. Indeed, the mapping of narrative form and gender per se might be seen as problematic and the view inherent in the psychologically derived feminist theory that the structure as well as content is gendered (Homans, 1994: 9) should be challenged. Considering work from feminist linguistics, there is ample evidence that to gender a linguistic form of any kind is misguided. Georgakopoulou restates this important point, saying, ‘language and gender research has shown that isolating certain linguistic forms and trying to map them with “male” or “female” as correlations between devices and social meaning is, as a rule, a futile exercise’ (1995: 461). In common-sense terms alone, the application of this principle to literary narratives would seem to bear out the same conclusion. If it was weak narrativity itself that was ‘feminine’ and used by female writers alone, then this poses a problem for the wealth of avant-garde and experimental writing composed by male writers and conventional narratives written by women (see texts discussed by Homans, 1994; Fludernik, 1996a; McHale, 2001). Each of the features of weak narrativity used in the writing of Michèle Roberts that I have contrasted with the ‘male plot of ambition’ has also been used in texts written by men, with male protagonists whose thematic concerns do not focus on contesting gender ideology in any clear way.

Narrativity in Pale Fire and Invisible Cities An example of this is Nabokov’s Pale Fire (1962). Like Une Glossaire, this can be analyzed as a hybrid narrative which also draws on the formal features of the discourse colony (Marley, 1987). As Marley’s detailed analysis shows, the construction of the narrative in the form of poem with annotated footnotes and index as commentary also fits Hoey’s analysis of the non-narrative discourse colony.




The nine properties of a discourse colony

Pale Fire’s footnotes

1. 2. 3. 4. 5. 6. 7. 8. 9.

? YES YES NO ? NO NO YES YES

Meaning is not derived from sequence. Adjacent units do not form continuous prose. There is a framing context. No single author and/or anon. One component may be used without referring to the others. Components can be reprinted or reused in subsequent works. Components may be added, removed or altered. Many of the components serve the same function. Alphabetic, numeric or temporal sequencing.

Source: Adapted from Hoey, 2001: 88; Marley, 1987: 262

Pale Fire also resists closure, for ‘Shade’s’ poem is left incomplete at line 999, and the final entry of the commentary takes us back to first line of the poem, headed ‘Line 1000: [= Line 1: I was the shadow of the waxwing slain]’ (p. 229). This final entry also ends with the narrator projecting a future replaying of the plot, But whatever happens, wherever the scene is laid, somebody, somewhere, will quietly set out – somebody has already set out, somebody still rather far away is buying a ticket, is boarding a ship, a plane, has landed, is walking towards a million photographers, and presently he will ring at my door – a bigger, more respectable, more competent Gradus. (p. 236) However, as with Une Glossaire, there are features in Pale Fire which do prompt narrativity, notably the use of precise chronological references (for example, ‘July 1’, ‘1915’, p. 62; ‘July 2’, ‘1920’, p. 64). As with the agent orientation in Une Glossaire, these time markers construct a network of cohesive chains, which encourage the reader to process the individual footnotes as a coherent narrative. Invisible Cities (1997 [1974]) by Italo Calvino is another text showing ‘weak’ narrativity quite unlike the ‘male plot of ambition’. Although this text is not a discourse colony as such, the organization of the chapters does not rest upon chronological sequence. Rather, the text relies on a metaphor of spatialization as Marco Polo recounts to the Kublai Khan the various cities that may or may not exist in the extent of his kingdom. These descriptions lack precise referents in time and place. Instead of temporal references which would facilitate the reader constructing chronological order, the cities are described using iterative




frequency, for example ‘Each year in the course of my travels I stop at Procopia and take lodgings in the same room in the same inn’ (p. 146). Along with the use of the present tense, the temporality that should provoke high narrativity instead is weak and anachronous. Past, present and future are blurred into one, rather than organized into a ‘significant shape’ as with Brooks’ male plot. The narrator puts it thus: ‘[Marco] cannot stop; he must go on to another city where another part of his pasts awaits him, or something perhaps that had been a possible future of his and is now someone else’s present. Futures not achieved are only branches of the past: dead branches’ (p. 29). Even the framing narrative of the conversation between Khan and Polo lacks temporal order, and instead is characterized by poetic and inconclusive exchanges, in some cases including repetition of earlier conversations. For example, in ‘Thin Cities 1’ we read, Newly arrived and totally ignorant of the Levantine languages, Marco Polo could express himself only with gestures, leaps, cries of wonder and of horror, animal barkings or hootings, or with objects he took from his knapsacks – ostrich plumes, pea shooters, quartzes – which he arranged in front of him like chessmen. (p. 21) And then again in ‘Trading Cities 1’ a similar passage occurs: Newly arrived and quite ignorant of the languages of the Levant, Marco Polo could express himself only by drawing items from his luggage – drums, salt fish, necklaces of warts hogs’ teeth – and pointing to them with gestures, leaps, cries of wonder and horror, imitating the bay of the jackal, the hoot of the owl. (p. 38) Invisible cities is a fragmentary text, without a series of problems that are resolved or a clearly defined point of closure. Its temporality is weak and is characterized instead by points of repetition, and fluid images that emphasize reflection, illusion and stasis. All of this might well be taken as a description of the ‘female plot’ proposed by Winnett or Anderson, and has points of similarity with the apparently feminine state of ‘lyric timelessness’ associated by some with women’s writing, and indeed with the non-sequential nature of écriture feminine. However, for both Pale Fire and Invisible Cities, I would argue that there are few grounds, if indeed any, for suggesting that these formal features constitute a ‘female’ alternative to the ‘male plot of ambition’. Instead, these texts can be interpreted as examples of experimental




writing, which as Rampton (1993) and Couturier (1993) suggest, can be located within postmodern trends. The challenging of narrative coherence through the features of weak narrativity allows the narrators of both texts to reflect on the decentering of language. Nabokov’s Kinbote creates a commentary that has little to do with the original text and Marco Polo’s communication with Khan remains elusive, ‘The Great Khan deciphered the signs, but the connection between them and the places visited remained uncertain’ (p. 22). Moreover, to argue that these texts are ‘female plots’ might by extension lead to a corresponding attempt to gender postmodern writing more widely. As Hutcheon (1996) points out, although there are some intersecting concerns between feminism and postmodernism, the two are quite separate enterprises with differing political emphases. It is this absence of feminist political investment in Pale Fire and Invisible Cities that makes it all the more dubious to label their formal features as ‘female’. In fact, the content of both might be critiqued from a feminist perspective as reinforcing patriarchal values. In Pale Fire, academia and authority structures are almost exclusively represented as masculine. In Invisible Cities, the empire is the possession of a male figure (Khan), a territory explored by a male character (Polo) and coded as female by the use of feminine names for the cities (for example, Despina, Olivia, Phyllis). This leads me again to question the basis upon which theorists have labelled their paradigms ‘male plots’ or ‘female plots’. The analysis here has shown that there is no empirical evidence to suggest that the alternative strategies of narrativity are used exclusively by women. Rather, male authors can use these forms too, without any relationship to feminist content or interpretation, and in some cases to reinforce patriarchal values. It is quite clear that based on the evidence of this analysis, it is not the narratological form itself that is gendered. To label narrative strategies as ‘male plots’ or ‘female plots’ is thus misleading, for it conflates the narrative form with the content of the narrative or its interpretation. This is not to say that the form in which a story is narrated cannot contribute to a reader’s interpretation of a text’s meaning, which may involve feminist politics of some kind. But it is not the form itself that is female, feminine or feminist.

Conclusion In this chapter, I have raised three main criticisms of this type of feminist narratology. First, categorizing the various plot types in metaphorical terms lacks the support of empirical data and detailed linguistic




analysis. Second, labelling the alternatives as ‘male’ and ‘female’ illustrates a universalizing and essentialist assumption that all male and female experience is the same and is biologically determined. Finally, the slippage between narrative content and plot structure results in a simplistic correlation of narrative structure and gender where it is the linguistic form itself that appears to be gendered. These problems are to a certain extent inherent in the theoretical models from which this strand of feminist narratology is derived. Broadly speaking, psychoanalysis has been criticized for its potential universalism and tendency towards decontextualized abstraction (McCormick, 1992), as has the writing of Cixous described as écriture feminine (Mills, 1995). Indeed, the appropriation of Cixous, Irigaray and Kristeva as ‘French feminists’ by Anglo-American scholars (including those discussed in this chapter) has been described as a dangerous form of decontextualization (Winter, 1997; Delphy, 2000; Moses, 2003) whereby their work is falsely taken as representative of feminism in France and ignores a more diverse situation. This principle of abstraction is problematic from both a feminist and a narratological point of view. Various writers have argued that when theoretical arguments and paradigms are divorced from their actual contexts, then a discussion of feminist principles has the potential to become apolitical. Once the discussion shifts from the particular into the abstract, it becomes difficult to ask vital feminist questions, such as to whom the differences of gender matter and what might be done about them. From a narratological perspective, the abstraction has also been shown to be dubious, for the gendering of alternative plot models when analyzed in terms of their narrativity simply cannot hold up against the analysis of a wider range of particular texts. This debate raises questions concerning how far the separation of form and content can be taken, how this relates to matters of context and the parts that each of these might play in the process of interpretation as a reader might make sense of a narrative. In this chapter, I have argued that gender cannot be attributed to the narrative form itself. In the next chapter, I discuss how these difficult questions may be approached from the perspective of contextualist narratology.




The Question of Gender and Context

Lanser’s call for feminist narratology ‘Postclassical narratology’ is a term used by David Herman to describe the broadening field of narrative theory and analysis (1997). As the term ‘post’ indicates, this work not only follows the classical work of structuralist narratology in time but also adopts a critical relationship towards these earlier models, challenging and augmenting them in a variety of interdisciplinary ways. Feminism is amongst the range of perspectives being used to ‘energize’ this field (Herman, 1997: 1049). This is particularly associated with the work of Susan Lanser (1986, 1988, 1995, 1999), which is significant because of its seminal status and because some twenty years on from its inception, it can be understood as a product of its time. In 1986, Lanser’s insightful and provocative paper ‘Toward a feminist narratology’ appeared in Style. Like the feminist response to Brooks’ work surveyed in the previous chapter, her ensuing critique of narratology challenged the existing body of narrative theory for its androcentric bias on the interrelated grounds of corpus and criteria. She asked, upon what body of texts, upon what understandings of the narrative and referential universe, have the insights of narratology been based? It is readily apparent that virtually no work in the field of narratology has taken gender into account, either in designating a canon or in formulating questions and hypotheses. (p. 343) Moreover, she pointed out that this gender bias went unacknowledged in the work of narrative theorists such as Propp (1968 [1928]), Greimas (1966), Iser (1978), Genette (1980), Barthes (1977) and Todorov (1977) 45



3

who she cited as ‘evident examples of the ways in which the masculine text stands for the universal text’ (p. 343). Lanser’s imperative to address this situation was to extend the corpus of texts studied to include those written by women (pp. 342, 343, 345), proposing that the analysis of this would form the basis from which the existing narrative theory could be tested and potentially ‘change substantially’ (p. 344). Since Lanser’s paper, considerable work has attempted to fulfil this imperative to rewrite narratology. Richardson writes in summary, Feminism, arguably the most significant intellectual force of the second half of the twentieth century, has (as should be expected) utterly and fruitfully transformed narrative theory and analysis in many ways. Virtually every component of or agent in the narrative transaction has been subjected to sustained examination. (2000b: 168) This reworking of narrative has been particularly far-reaching and includes the work of Lanser (1995, 1999) and Fludernik (1999) on narrative voice, Bal (1990) and Warhol (2001) on focalization, Case (2001) and Homans (1994) on plot structure, Prince (1995a, 1996) on the influence of corpora and Mezei’s (1996) significant and wide-ranging collection to name just a few. As this work so readily demonstrates, the integration of feminist criticism and narratological analysis has indeed been fruitful. However, the proposal was not uncontroversial at the time, and is worth reviewing in the light of changes in narratology and feminism in the last two decades. Lanser’s work, like that of the psychoanalytically derived narrative theory reviewed in Chapter 2 is clearly embedded in the priorities of second wave feminism. This is characterized by a recurring use of a binary model of gender that emphasized difference. Lanser explicitly aligns herself with the majority of US feminist critics of the period, ‘whose main interest is the “difference or specificity of women’s writing” ’ (p. 343) going on to claim that feminist narratology Could provide a particularly valuable foundation for exploring one of the most complex and troubling questions for feminist criticism: whether there is indeed a ‘woman’s writing’ and/or a female tradition, whether men and women do write differently. For given the volatile nature of the question, the precision and abstraction of narratological systems offers the safety for investigation that more impressionistic theories of difference do not. (p. 346, my emphasis)




The data that Lanser proposes as the basis of feminist narratology is also typical of feminist criticism at this time with its desire to recoup a previously absent corpus of literature by women into the canon. However, this focus on difference as demonstrated through the writing of women might now be seen as problematic. First, there is a tendency to conflate the narrator with the author. Despite Lanser’s expressed suggestion that the revision of narratology include examining ‘women as both producers and interpreters of texts’ (p. 343), the very text she analyzes and takes as representative of ‘women’s writing’ is in fact anonymous. She writes, ‘No indication is given of the letter’s source, authenticity, or authorship . . . I make no assumptions about the author’s sex’ (p. 346). We might then ask what criteria Lanser is basing her category of ‘women’s writing’ on. Is this derived from the gender of participants in the text, or of the biological sex of the author, or on some assumptions about the stylistics of ‘women’s language’? None of these is satisfactory. If it is based on the represented gender of a narrative participant (such as the narrator), then this might mean that the so-called women’s writing could be authored by a man (in which case Lanser’s critique of narratology’s corpus as androcentric appears irrelevant). On the other hand, if only women can author ‘women’s writing’, then this implies that gender is biologically determined rather than socially constructed – an assumption that might be critiqued on essentialist grounds. Even if women’s writing is defined on the basis of stylistic features, Lanser’s starting point of equating this with the features of ‘women’s language’ outlined in the work of Kramarae (cited p. 348) would now be open to question. Subsequent sociolinguistic studies have shown these features to be stereotypical assumptions that lack empirical evidence, a controversy that Lanser only alludes to in a footnote (p. 359). Second, Lanser’s description of this alternative corpus of ‘women’s writing’ has a tendency to construct the category ‘women’ as if it were a universal group. Hence throughout the essay there are references to ‘women’s texts’ (pp. 342, 345, 357), ‘women’s works’ (p. 343), ‘women and literature’ (p. 346), ‘women’s narratives’ (p. 350) and ‘women’s language’ (p. 350). Although Lanser makes passing reference to ‘narratives of other dominated peoples’ (p. 350) and ‘texts by writers of Asia and Africa’ (p. 358), her own work and indeed most of that which followed from it has for the most part focused on American or European authors.1 This is ironic, given that Lanser intends feminist narratology to ‘study narrative in relation to a referential context that is simultaneously linguistic, literary, historical, biographical, social and political’ (p. 345). Such an aspiration is laudable. It is unfortunate, then, that



The Question of Gender and Context 47

Lanser’s analysis, interesting as it is, has moments of globalizing from a white, euro-american perspective without detailed reference to this wider context. For example, when talking about the polyphony in her chosen text, she writes that it is, ‘a sharper version of the polyphony of all voice and certainly in visible ways, of the female voices in many women’s narratives’ (p. 349). This may be so, but Lanser’s argument would be stronger, had this been grounded in the historical and social specifics of the particular women- and male-dominated societies she goes on to gesture towards. As it stands, this aspect of her work again seems to exemplify the white elitism that Homans (1994) critiques feminist narrative theorists for, and, as I wrote in relation to the psychoanalytically derived feminist narratology of Chapter 2, is typical of this stage of second wave feminism generally. Not only might this tendency be criticized from a Black feminist or womanist perspective, it also needs to take into account the variations in context that Lanser herself itemized in the list given earlier, recognizing that women are not a homogenous group, but have both experiences and narratives that may vary according to diachronic, social, ethnic and other situations. Lanser’s proposal for feminist narratology might also be understood within the emergence of what Chatman labelled ‘contextualist’ narratology (1990). As I discussed in more detail in Chapter 1, this provoked considerable debate and brought into sharp focus questions about the definition and purpose of narratology itself. Feminist narratology served (and still serves) as a central source of examples for this dispute (Diengott, 1988; Prince, 1995a; Darby, 2001, 2003). Diengott’s arguments, for example, stated unequivocally that ‘feminism’ had nothing to do with ‘narratology’, as the ‘system of poetics’ by its very definition was separate from ‘interpretation’ (pp. 43–45). Even Prince (1995a) who is more conciliatory towards Lanser’s stance warns against giving in to the ‘interpretive temptation’ (p. 82). Lanser’s response to Diengott addresses these concerns, arguing convincingly that the system itself cannot be abstract (and hence indifferent to gender) as it is formulated from and then tested against the experience of actual texts (1988: 55). More generally, during the last two decades there has been a wider acceptance that the analysis of the system and interpretation of texts need not be understood as mutually exclusive, with one displacing the other. As Richardson argues, ‘the choices available to the narrative theorist include the option of “both/and” ’ (2000b: 171). However, the contextualist debate is important, for it highlights the delicate difficulties involved in talking about gender in relation to narrative analysis. Does feminism relate to form, content, context or interpretation? As




I have argued in Chapter 2, to assume that a narrative form is itself gendered is at the least dubious. But this is not to dismiss a feminist critique of form altogether. Rather, in this chapter, I question the ways in which gender might play a part in the interrelationship between context, content and narrative structure, and to what extent these parameters can be treated as separable from one another. I do this by following Lanser’s suggestion that the corpus of narratives studied should be extended in order to test the validity and, if appropriate, modify aspects of classical narratology. However, in the analysis and discussion that follows, my work extends Lanser’s project in vitally different ways. First, there is a contrast in the data that I use. I am not concerned with establishing or describing an alternative tradition of ‘women’s writing’, an enterprise that poses various problems for its universalizing and essentialist potential, as I have indicated above. Instead, the narratives that I look at are deliberately chosen because of their anonymity. Thus the analysis shifts to matters of representation (as opposed to the author’s status) and how this is realized through textual forms. This is based on the assumption expressed in CDA that no representation (be it fictional or non-fictional) is ever neutral but instead will be related to ideological values of extra-textual context in a complex and dialectical way. The narratives that I analyze are a selection of Middle English romances and Japanese medieval narratives (in translation). As such, they also broaden the range of data that has been considered in postclassical feminist narratology, both in diachronic and in cultural context. Because these texts are anonymous, the feminist questions that I ask of narrative theory and analysis reframe those posed by Lanser. First, I continue to examine how well classical narratology can account for the organization of these texts and what this analysis might suggest about the cultural specificity of the theoretical models. Secondly, rather than asking how this might delineate ‘women’s writing’, I debate the relationship between form and content and how this might be related to particular cultural constructions of gendered behaviour, thereby emphasizing the socially constructed nature of gender rather than treating it as a biological given.

Revising theories of plot In her 1986 paper, Lanser singles out theories of plot as an aspect of narratology that needs ‘radical revision’ (p. 356). Lanser is not alone in drawing attention to this. Indeed in Chapter 2, it is the notion of ‘plot’ that stands at the crux of the feminists’ debate with Brooks’ psychoanalytic




revision of formalist narratology. Although there are certainly grounds for the feminist critique of plot, I argued that the psychoanalytically derived alternatives were weakened on the grounds of their decontextualized universalism. At first glance, it would seem that contextualist narratology is an apt alternative that would address these flaws as it retains an attention to context and offers ‘the precision and abstraction of narratological systems’ (p. 346) as replicable parameters with which to compare a range of texts. It is then all the more ironic that Lanser’s reflections on plot are offered as, at her own admission, ‘sketchy’ (p. 357) first moves towards this. It is unclear to me from Lanser’s discussion exactly which narratological concept of plot she is objecting to. She seems to rely on the analysis of content, in which she contrasts (male) ‘textual actions’ (p. 356) with (female) ‘stasis’ (p. 355), yet objects to ‘the notions of plot that have followed from Propp’s morphology’ (p. 357) without clearly demonstrating how her own text does (not) conform to his theory of functions. I agree with Lanser’s suggestion that theories of plot should be revised, but I differ in the way I propose this revision might take place. First, I would argue that it is necessary to be more explicit about which specific theory of plot is under scrutiny. Second, I contend that any description of an alternative should do precisely what Lanser suggests earlier in her paper, and use ‘the precision . . . of narratological systems’ (p. 346), even when this is combined with interpretation and discussion of contextual influences. I want to return to one further point that Lanser makes about plot in relation to a feminist revision of narratology. She says, until women’s writings, questions of gender and feminist points of view are considered, it will be impossible even to know the deficiencies of narratology. It seems to me likely that the most abstract and grammatical concepts (say, theories of time) will prove to be adequate. On the other hand [. . .] theories of plot may need to change substantially. (p. 344) Like Deingott, we might then ask why it is theories of plot that may need to change but not other parts of narratology (1988: 45). While I disagree with Diengott’s claim that this statement indicates that narratology need not change at all, I do think it is telling distinction and points to both the complexity and the relevance of plot theory when addressing feminist concerns. My own tentative answer involves exploring the contentious uses of ‘plot’ in narrative theory and literary criticism.




The analysis of plot has formed an important area in narrative theory. However, as Ronen (1990) points out in her paper surveying this area, the apparent centrality of plot accompanies a complicated situation where the term is left undefined by some narratologists and used in differing ways by others (p. 322). For example, discussions of plot may rest on the description of the content of a narrative, a position summarized by Chatman when he says, ‘The events in a story are turned into a plot by its discourse, the modus of presentation’ (1978: 43). It is this concept of plot that is often found in literary criticism and is related to the specific story content of a given text, rather than accounting for its structural features.2 As Chatman goes on to argue, a content-driven approach to plot theorization (such as Aristotle’s or Northrop Frye’s) is unhelpful as the criteria for defining terms tend towards endless proliferation and are difficult to replicate (1978: 89). In contrast to this, structuralist definitions of plot attempt to outline a schematic relationship between those story events in terms of an abstract system. Examples of this include the well-cited work of Propp (1968 [1928]), Todorov (1977) and Greimas (1966), along with less known work such as that by Robert Longacre (1983). Such structuralist accounts have also been found wanting. As Ronen (1990) goes on to discuss, there is little agreement over whether plot should have a criterial status within narrativehood or be regarded as an element of narrativity (for example, Brooks uses ‘plot’ as a universally defining feature of narrative (1984: 5) whereas other narratologists would argue that only temporal sequence is needed to distinguish the genre, for example Labov and Waletzky, 1967). Moreover, there has been widespread recognition of the dangers of universalism, and that the recognition of a plot is determined by culturally specific factors. Chatman writes perceptively on this point, saying Perhaps the best way to understand taxonomies [of plot] is to treat the historian or critic as ‘native speaker,’ a user proficient in a code. It is his behaviour as much as the work itself that we need to examine. . . . In short, the characterization of plot into macrostructures and typologies depends upon an understanding of cultural codes and their interplay with literary and artistic codes and codes of ordinary life. (1978: 94–95) However, to examine plot in a way that integrates extra-textual factors, in particular the ‘behaviour’ of the critic or reader is less than straightforward. As Toolan writes, the notion of literary competence upon




which Chatman’s suggestion is founded is itself ‘markedly culture specific’ and that ‘when we talk of “readers’ abilities” we have to keep clearly in mind that we are talking only of the acquired and developed ability of a group of readers, rather than some universal mental ability’ (2001: 29). Moving outside the text requires the analyst to address matters which are by their very nature difficult to pin down, sometimes appearing ‘fuzzy and provisional’ (p. 36). In summary, approaches to plot theory recognize in differing and problematic ways the need to analyze content, structure and contextual factors. It is this controversy which has increasingly foregrounded the contextualized nature of theory, analysis and interpretation that makes plot theory particularly of interest to feminist narratology. As Bergvall summarizes, the questions that have predominately recurred in language and gender studies include not just ‘whether there is gender differentiation in language use’ and ‘what forms it takes linguistically’ but also ‘whence it arises’ and ‘what effects it has in society’ (1999: 274). If feminist narratology is to attempt answers to more than just the first two of these questions, it needs a theory of plot that can also take into account extra-textual factors such as ideological influences on textual production and reception. This is all the more important when gender is understood to be socially constructed and open to variation across and within cultures, rather than as a monolithic construct that can unequivocally define a concept like ‘women’s writing’. However, as the preceding discussion indicates, no single plot typology has yet satisfactorily accounted for this. Therefore, in the revision of plot that I offer here, I adopt a postclassical practice of synthesizing aspects of analysis derived from different paradigms. Longacre’s work on the anatomy of plot provides a structuralist starting point. This is complemented by a more recent development in written discourse analysis of Michael Hoey’s work on culturally recognized patterns of text organization (2001). My aim in the remainder of this chapter is twofold. The first is to use these models in narratological criticism and indicate how particular formulations of gender ideology are significant in both the interpretation of form and the content in these medieval texts. The second is perhaps more far-reaching, and is to suggest how this analysis might in turn be used to modify the theoretical frameworks of ‘plot’ and in doing so re-evaluate the criteria used in narratology both past and present. I begin by providing a brief introduction to the models. Longacre’s (1983) scheme for analysing plot as the ‘notional’ (deep) structure of narrative discourse would seem to be precisely the kind of theory that




Lanser objected to. It lays claim to a universal status (pp. 20, 29), yet although Longacre provides a large range of examples to support his framework including Mixe and Ga’Dang folktales, he only includes two small examples from female authors amongst the some 14 other male authors cited in much greater detail. Longacre’s outline is clearly structuralist in orientation, separating out a series of categories for the notional (deep) structure as realized by a universal set of surface forms (p. 29). For ease of reference, I present Longacre’s schema in summary below (Figure 3.1). Alongside the problems of its apparent universalism and limited corpus of examples, this model approach might also be objected to from a poststructuralist perspective. Fundamentally, the abstraction of a deep structure at all is dubious at the least (Toolan, 2001: 16). Further questions might be asked about how the audience distinguishes between elements of this deep structure. A crucial example of this is the distinction between the semantically contrasting elements of Climax and Denouement, either, both or neither of which, Longacre proposes, can be textually realized as surface Peak, thus calling into question how the pivotal (and in Longacre’s definition, criterial) components of the plot may be identified. Despite these critical difficulties, Longacre’s categories may still retain some use. In the following analysis I focus on the occurrence of the Peak markings. I use this to compare the textual patterns across the narratives both as a means of examining what kinds of content get narrated in this way and second to test how far Longacre’s schema of a climactic Peak can account for a different range of texts about women and men. Hoey’s work on written discourse analysis provides a useful comparison with Longacre’s structuralist approach. Hoey’s analysis is wide ranging and attempts to account for a variety of text types. The aspect of his research that I am interested in here is what he describes as ‘culturally popular patterns of organization’ (2001: 122). Like recent developments in cognitive narratology (Herman, 2002), Hoey’s approach is underpinned by the assumption that the text is a site of interaction between writer and reader (2001: 35). As such, the analysis of the patterning ‘in’ the text is contextualized, understood in relation to extra-textual world knowledge. However, rather than articulating this world knowledge in terms of schemata or scripts (for example, as employed by Herman (1997)), Hoey points to the more specific set of expectations a reader might have as ‘a more generalized set of expectations that are shared across a range of texts’ (p. 122). These patterns act as a kind of ‘template’ for the reader, and follow patterns that occur so




Title

Aperture

Stage

Pre-peak episodes

Peak

Peak’

Post-peak episodes

Closure

Finis

Notional Structure

Surface only

Surface only

1. Exposition lay it out

2. Inciting Moment get something going 3. Developing Conflict keep the heat on

4. Climax knot it all up proper

5. Denouement Loosen it

6. Final Suspense Keep untangling

7. Conclusion wrap it up

Surface only

Figure 3.1 Longacre’s model of plot (adapted from Longacre, 1983: 22)


Copyright material from www.palgraveconnect.com - licensed to Universitetsbiblioteket i Tromsoe - PalgraveConnect 2011-03-09

54

Surface Structure

often as to become ‘predictable’. The patterns that Hoey goes on to describe (Problem-Solution, Goal-Achievement, Desire Arousal and so on) are akin to the action structures that readers draw on when processing narrative (Giora and Shen, 1994) and parallel aspects of structuralist plot models. A schematic outline of Hoey’s Problem–Solution pattern is presented in Figure 3.2. According to Hoey, the various components of each pattern may be lexically signalled in the text itself, although recognition of such signals is dependent on the reader’s (or listener’s) world knowledge. A second means of spelling out the relations between the components is to project the text into a dialogue. For the Problem–Solution pattern, this would be as follows:

Projected question

Element of pattern

What is the situation? What aspect of that situation is problematic? What response was made? What was the result? How successful was this?

Situation Problem Response Result Evaluation

Situation (optional) Aspect of situation requiring a response For example

Problem

Goal

Opportunity

Desire Arousal

Gap in Knowledge

Response

Negative Evaluation and/or Result

or

Positive Evaluation and/or Result

Figure 3.2 Outline of the Problem–Solution and related patterns (adapted from Hoey, 2001: 166)




This is a highly simplified sketch of the bare bones of Hoey’s work in this area. He goes on to discuss the complex ways in which multiple interrelated sequences of the patterns may occur in texts. While it is by no means without contentions, Hoey’s work is a useful heuristic for probing the possible relationships between narrative form, content and context, one to which we will return in future chapters. From the point of view of plot structure, it is interesting that when Hoey’s prototype is set alongside Longacre’s outline, approximate points of compatibility emerge. Longacre’s ‘Plot’

Hoey’s ‘Problem–Solution’ outline

Exposition Inciting moment Developing conflict Climax Denouement Final suspense Closure

Situation Aspect of situation requiring a response (i.e. Problem) Response (Positive) Result Positive Evalution

By overlaying the two perspectives, Hoey’s model augments these earlier paradigms, providing an alternative, more contextualized means of conceiving of these sequential semantic patterns. Throughout his discussion, Hoey goes to great lengths to stress that the Problem–Solution pattern is not a universal structure (2001: 122–123), but appears within particular cultural contexts, and is one of a potentially infinite range of patterns that might occur. In the discussion that follows, Hoey’s model is a useful starting point both for narratological criticism and for reflecting on narratology itself. It provides the means by which the analyst may discuss particular texts, for example, considering what situations are constructed and understood as ‘problematic’, which participants are able to articulate desires, achieve goals, solve problems and so on. Perhaps more generally, a feminist perspective might take Hoey’s observation that certain patterns occur with great frequency while others do not and then ask why this might be so and what this might reflect about the ideological values in a given culture. The texts that I analyze are taken from the medieval period. Because these texts are less known than those that have been examined in most postclassical narratology, I will provide a brief description of them first. The three Middle English romances are Sir Isumbras, Sir Gowther and Emare.3 There are strong thematic similarities between the three narratives




which suggest that it is possible to view them as a distinct group, belonging to the subgenre of poems described as ‘religious romances’ (Hopkins, 1990: 12–13). Each story involves a pattern of loss and separation where, after a period of initial stability, the central protagonist is forced to undergo a series of hardships. Following the conclusion of their suffering, the protagonists are restored to their original status or reunited with their family. Their suffering is placed within an explicitly Christian context where the pious behaviour of the hero or heroine conveys a didactic message, and while Hopkins (1990) quite rightly does not include Emare in her discussion of penitential romances, the medieval concept of penance plays a crucial role in facilitating the resolution of each narrative too. Within this threesome, there are important differences in content. One of the most obvious is the gender of the central protagonist, for Sir Isumbras and Sir Gowther follow the activities of a hero, whereas Emare focuses on the experience of a heroine. Also, while the suffering of Gowther and Isumbras takes the form of a penitential process, Emare’s hardships do not. So while these romances have some thematic material in common, they tell stories about human suffering in different forms and from alternative perspectives. The Japanese stories are taken from the collection known as Konjaku Monogatari Shu (Times of tales past, edited and translated by Ury (1979)). This is an extensive collection of narratives which is assumed to give insight into the popular beliefs of twelfth-century Japan and to reflect the popularized influence of Buddhism at the time (Mills, 1972; Ikeda, 1979; Kato, 1979). Like the religious romances, these narratives are didactic, and are associated with public Buddhist preaching. As such, they too might be considered as conveying messages about morally acceptable behaviour within this particular context. The collection includes stories about women and men from many walks of life. The content of these tales reflects the multiple contesting views about femininity and women’s spiritual status within Buddhism found also in other collections such as Kankyo no Tomo (A Companion in Solitude) (Pandey, 1995). This includes examples of sexually alluring women who are transformed into oni (demons) and foxes,4 but these exist alongside examples of the superlative pious behaviour of nuns. The tales that I analyze in detail are that of the pious nun (7.18), ‘How a monk through Bishamonten’s aid begot gold and obtained a means of support’ (17.44) and ‘How a deeply Jealous Woman Turned into a Demon while still Alive’ (Kankyo no Tomo 2.3). These tales all centre on points





Marking of climactic peak Each of the Japanese and English medieval narratives considered in this chapter contain examples of what Longacre describes as Peak marking. However, the form, distribution and function of these Peaks indicate stylistic variation between the texts and suggests grounds for modifying Longacre’s framework. According to Longacre’s outline, Peak marking should predictably occur in an episode-like unit that corresponds to the Climax or Denouement in the notional structure of plot (1983: 24). This marking includes rhetorical underlining (especially through repetition), concentration of participants, heightened vividness (for example, through the dramatization of speech), change in pace or a change in orientation. Peak thus demarcates the pivotal transition in an Aristotelian plot progression where the ‘knot’ becomes ‘untied’ and resolution begins to unfold. In the narratives considered here, the typical pattern of a single episode of Peak marking occurs in three of the narratives: Sir Isumbras, Sir Gowther and The Pious Nun. Sir Isumbras and Sir Gowther are both considerably longer narratives than The Pious Nun. As might be expected in more complex texts, the Peak marking is found not only in the central section, but also in clusters at other points in the narrative. However, as I have argued elsewhere, these smaller clusters are a result of the hierarchical organization of the texts and can be distinguished qualitatively and quantitatively as lesser turning points (Page, 2000). In all three narratives, the Peak-like episodes are used to narrate problematic events that form the central crisis. In Sir Isumbras and Sir Gowther these are ‘set piece’ battles where the hero confronts and kills his enemy, the heathen king. In The Pious Nun, this is where the unscrupulous monk opens the borrowed scroll and finds that the Nun’s copy of the sutra has been miraculously erased. Examples from The Pious Nun and Sir Isumbras are given below which show rhetorical underlining (repetition and parallelism) and concentration of participants (an increase in the number of characters present). These are underlined in the following extracts from the texts. When he opened the scrolls, he saw nothing but yellow paper, and not a single character. He thought this strange and opened



of transformation and the Buddhist principles of cause and effect, but show differing perspectives on how femininity might be implicated within this spiritual context.


Whenne he was on that gode steede He spronge forth as sparke on glede And thowghte to smyte sore Hit was seene ther his stede yode The knyghte slewe all that byfor hym stode Bothe lesse and more Till he come to the hyghe mountayn The hethenn kynge then hath he sleyn And all that with hym were. (Sir Isumbras, lines 445–453) As Longacre’s model suggests, these Peaks do serve to mark the high point of the complications in the narrative events. In each case, after this climactic crisis, resolution becomes possible. Isumbras and Gowther receive absolution and are restored to their former status and riches. The nun receives her scrolls back and after her repentance the scriptures are restored. This plot structure of a single, pivotal high point is particularly appropriate to the content of these narratives, all of which focus on some kind of spiritual transformation (either the spiritual conversion of the hero or the transformation of the scriptures). The importance of transformation is also supported by the Problem–Solution patterns found in these texts. In all six narratives considered in this chapter, several culturally popular patterns of organization occur, interconnected by multilayering, where the response to one problem gives rise to another. In Sir Isumbras and Sir Gowther, the problem of the heroes’ sinful state (Isumbras is proud; Gowther, more dramatically, is the son of a devil) is responded to by a penitential process. In both cases this involves conflict with a heathen enemy, which may be understood as a penitential indulgence and therefore part of the means of the hero achieving his forgiven state. A summary of these patterns is given in Figures 3.3 and 3.4 As these diagrams suggest, the patterns in these narratives contribute to an overarching progression which indicates change of some kind. This change is both spiritual (conversion) and worldly, as



the other scrolls. All seven alike were totally devoid of writing. Astounded, he showed them to the congregation, and everyone in the congregation saw exactly what Fa-tuan had seen. Fa-tuan and the congregation as well became fearful and ashamed, and they sent the sutra back to the nun. (Ury, 1979: 64–65)

60

Response: Duchess prays Has encounter with fiend Result: Gowther is born

Problem: Gowther is the son of a fiend

Response: To undergo penitential process

Sinful state

=

Goal Forgiveness Means Penance

Including:

[Problem Threat from Sultan Response Three-day battle Result Sultan is killed]

Final Result Forgiveness achieved

Figure 3.3

Forgiven

Predictable patterns in Sir Gowther

Problem: Isumbras is proud

Sinful state

Response: Penance is required = Goal: Divine forgiveness Response: Isumbras undergoes penance

[Including Problem: Heathen king abducts Isumbras’ wife Response: Isumbras to kill Heathen King in three-day battle

Positive Result Forgiveness granted

Figure 3.4

Forgiven Positive Result Reunited with wife]

Predictable patterns in Sir Isumbras



Problem: Duke and Duchess cannot Conceive a child


Thus syr Gwother coverys is care That fyrst was ryche and sython bare And efte was ryche ageyn. (Sir Gowther, lines 739–741) Ofte was he welle and wo But nevur so well as he was tho (Sir Isumbras, lines 769–770) So both the Problem–Solution patterns and the Peak marking of a single Climax structure the narratives around a pattern of transformation. As such, they follow the basic principles of aristotelian plot from which Longacre’s model is derived, founded upon ‘reversals and recognitions’ (Halliwell, 1987: 38) and a move from equilibrium to disequilibrium and back again.

Alternative patterns of Peak marking Significantly, not all the narratives follow this straightforward pattern. In Emare, there are five passages that might be classed as Peak. Unlike Sir Isumbras and Sir Gowther, where the smaller clusters differ to the Peak, in Emare all five passages contain a similar amount and range of rhetorical underlining, heightened vividness, concentration of participants and change of vantage point. Each of the passages also corresponds to a moment of crisis for a character within the text. This is summarized below: 1. Emare’s father and the crowd grieve as the heroine is cast adrift at sea (lines 241–309). 2. The Old Queen forbids the King of Galys from marrying Emare (lines 433–456). 3. The King of Galys grieves over the false letter sent by the Old Queen (lines 541–564). 4. Emare is cast adrift at sea again (lines 601–648). 5. The King of Galys discovers the Old Queen’s treachery (lines 742–804). The similarity in the Peak marking in these passages is exemplified in the following extracts. This includes the direct speech, use of exclamations and repetition.



indicated by the comparisons encapsulated at conclusions to the narratives:

When he of sowynyng kovered was Sore he wept and sayde, ‘Alas, For my dowghter dere! Alas that Y was made man, Wrecched kaytyf that I hyt am!’ (Emare, lines 289–293) The kyng sayde, ‘By Goddys name, Syr Kadore, thou art to blame For thy fyrst tellynge! Thow sholdest fyrst have tolde me Of my ladye Egare I love most of all thyng!’ (Emare, lines 757–762). So both in terms of the linguistic forms used and in terms of the content of these passages (they all occur as moments of crisis) it appears that the plot of Emare does not correspond to Longacre’s paradigm but instead follows an alternative pattern characterized by multiple climaxes. Likewise, the culturally popular patterns in Emare do not follow a single pattern of transformation. While the Problem–Solution patterns of separation and reunion occur in a temporal sequence, what is striking is the similarity between these events resulting in the duplication of the patterns. This is summarized as a diagram in Figure 3.5. Furthermore, while the heroes underwent moral and physical transformation, the description of Emare throughout the narrative emphasizes her constancy. For example, she remains ‘whyte as lylye flour’ (lines 205, 946) symbolizing her moral purity. In contrast to the agency of the heroes who are represented as affecting change in the circumstances, Emare is cast in the more passive role as a desired object. This is highlighted in the predictable patterns, which take the form of

Problem: Emare is cast into the sea by Artyus

Response: Artyus goes to Rome To seek penance

(Loss)

(Reunion)

Problem: Emare is cast into the sea by the Old Queen

Response: King of Galys goes to Rome to seek penance

(Loss)

(Reunion)

Figure 3.5 Parallel predictable pattern elements in Emare





desire-arousal patterns and like the overarching sequence of loss and reunion are duplicated. The first of these patterns is spelt out in detail here.

Who within this situation is particularly attractive? The fayrest wommon on lyfe (line 222) What effect did this have on the male character? That all hys herte and all hys thowghth Her to love was y-browght (lines 223–224) What did the male character do about this? So he was anamored hys thowghtur tyll Wyth her he thowghth to worche hys wyll And wedde her to hys wyfe. (lines 226–227) What was the result? Then sayde [Emare] Nay syr, God of heven hyt forbede That ever do so we shulde! (lines 250–252)

Situation

Object of Desire Desire Arousal

Attempt at desire fulfilment

Negative result

Therefore, it is not Emare herself who is active and involved in undergoing moral change. Instead, she is acted upon by others. The content of the romance places value in her staying the same over time, maintaining her virtue in the face of hardship. This is reflected in some of the structural features of this alternative plot of stability that contrasts with the plot centred on transformation. For example, while a transformation (especially of conversion) is expected to occur only once and is mirrored by the single Peak passages of Sir Isumbras and Sir Gowther, constancy is perhaps best exemplified by enduring repeated trials or moments of conflict, as demonstrated in the duplicated predictable patterns and the multiple Peak passages. On one level, I have explained these differences in the plot structure in terms of stylistic variation that relates to the general themes in these narratives. Thus the narratives about transformation are well accounted for by Longacre’s monoclimactic outline while the narrative which emphasizes constancy does not. I would go further than this, and argue that these values and the plot structures need to be understood in terms of how they reflect and function within particular contexts. From a social constructivist perspective, Gergen writes that



What is the situation? ‘The mayden that was of semblant swete Byfore her owene fadur sete (lines 220–221)


Thus neither the content of the narratives, that is the values of ‘transformation’ or ‘constancy’, nor the structures used to articulate these tales are abstract but exist and operate within particular cultural contexts. Therefore I propose to offer a feminist interpretation of these narratological contrasts but one that does not map the difference in forms reductively onto a binary and universalized pairing of ‘male’ and ‘female’. This is not to dismiss gender from the equation altogether. Rather, gender is a significant influence on both the content and the structure of these stories, but in an indirect and contextualized way. This can be explained by examining the particular importance of ‘transformation’ and ‘constancy’ within particular cultural traditions. The three Middle English romances may be treated as a group showing certain contextual similarities. All three stories place value on aristocratic individuals who are restored to positions of authority at the close of the narratives. Likewise, the influence of Christian beliefs is clear in all three narratives where God is represented as the ultimate (male) figure of authority and the means by which narrative resolution is achieved. Despite these similarities, there is an obvious difference between the romances. This is that the patterns of constancy and transformation are assigned to female and male characters respectively, emphasizing different patterns of moral behaviour for these characters. This leads me to suggest that the values of constancy and transformation (and hence the plot structures which help support these meanings) intersect with certain gendered values. As Blamires writes, constancy as a ‘feminine virtue’ would seem to be part of a broader contemporary view of women where ‘stabilitas’ became ‘the supreme criterion for assessing women in a culture obsessed with feminine “weakness” ’ (1997: 168). In addition, both the heroes undergo a transformation which is not only spiritual but also involves a loss of status and worldly goods, whereas Emare does not. This particular transformation was not possible for medieval women due to their socio-economic position as ‘possessions’ of their husbands. Caroline Walker Bynum discusses this in more general terms, commenting that the reversals and transformations of male saints’ conversions were not a feasible option for their female contemporaries. She says, Despite the fact that both chastity and marital status were more central themes in women’s lives than in men’s, male saints were far



Narrative tellings do more than create conversational realities; they are themselves constituents of ongoing institutionalized patterns of societal conduct. In this sense, they function so as to generate and sustain (and sometimes disrupt) cultural tradition. (1998: 8)


Hence the differences in the linguistic features of the monoclimactic plot which centred on transformation and the multi-climactic plot emphasizing stability have parallels in the ideological assumptions about masculinity and femininity current in medieval Christianity and in the social and economic details of the ‘real world’ experience of men and women from this time. The relationship between form, content and gender is complex. What I am suggesting is that ideological assumptions about gender that are significant in western, medieval Christianity are important in shaping the content of the three religious romances. Further, the way that this content is structured as a plot is not isolated from this but instead can be interpreted as helping to carry the gendered meanings of these tales. However, I do not want to suggest that gender is somehow intrinsically bound up in either the moral values or the plot structure: both of these are culturally constructed. Indeed, the evidence indicates such mutability. In other narratives, such as The Pious Nun, it is the female character’s repentance that generates the central transformation. As Pandey argues, the Buddhist attitude to the status of women ‘was by no means monolithic and unchanging’ (1995: 325). While various canonical texts present ‘women’ as inferior to men and debarred from certain forms of rebirth, other collections, like the Companion to Solitude, provide examples of ‘women from different social classes who are actively involved in their quest for salvation and for the most part are successful in doing so’ (1995: 330). Clearly, spiritual transformation may take on different meanings, gendered and otherwise, in alternative belief systems and varying across culture and time. More generally, transformations might be of many kinds, not just spiritual and the way that these are narrated need not be associated with gender ideology in such a prominent way (if at all). Thus my central point is that gender may influence the use that elements of plot structure are put to, but any examination of this cannot be carried out in abstract terms. Instead it should be rooted in the analysis of particular texts and the values of specific contexts. The second point I want to emphasize is that I am not trying to propose a binary paradigm of plot types where the mono-climactic plot is ‘male’ and the multi-climactic plot with its repeated Problem– Solution patterns is ‘female’. As I discussed at length in Chapter 2,



more likely to undergo abrupt adolescent conversions, involving the renunciation of wealth, power, marriage and sexuality. The point I am making here is an obvious one. Women could not take off their clothes and walk away from their husbands, as did Francis. Simple social facts meant that most women’s dramas were incomplete. (1991: 43)

attempting to map this contrast in form onto analogies with sexual metaphors simply does not hold up in the face of wider empirical analysis. Furthermore, from a feminist point of view it is worth noting again that constructions of gender may vary both within and between cultures and over time. Likewise, from a narratological perspective, a two-way split between single and multiple climactic patterns might be regarded as reductive. Instead, the contrasts in the narratives patterns may include other possible variations and manipulations both in Peak marking and in the use of predictable patterns of organization. The two other Japanese tales show further types of variation from Longacre’s outline. The tale of the monk who received Bishamonten’s aid is intriguing, and problematic from a narratological point of view. The events in the narrative are as follows. A poor monk seeks aid from the Buddhist deity, Bishamonten. On his return from the temple, he encounters a youth whom he assumes to be a temple boy. The ‘boy’ seeks the monk’s aid as a travelling companion, but then is revealed to be a young woman. Despite his misgivings, the monk is unable to part from her, makes love to her and she becomes pregnant. However, at the birth of the child, the boy/woman and baby disappear, miraculously leaving behind a lump of gold which the monk sells to become prosperous. The narrative contains a single passage of Peak marking, distinguished by the shift into direct speech and the use of repetition and exclamations as forms of rhetorical underlining and heightened vividness. This occurs when the monk becomes suspicious of the ‘boy’s’ gender. The following example comes from the monk’s interrogation of the ‘boy’. But this is so odd! It’s not at all like being in bed with a boy. I don’t know why, but my heart simply melts towards you. Can it be that you’re a woman? Is that so? Tell me the truth. From the moment I first saw you, I haven’t felt able to part from you for an instant; and this seems even stranger and harder to understand.’ (p. 118) However, it is difficult to argue that this passage corresponds to the Climax where the events of the narrative are either at an ultimate crisis or about to be resolved. Indeed, later, more problematic events which might form the Climax and Denouement, that is the ‘boy’s’ giving birth and subsequent disappearance are not narrated using Peak marking. The monk did as the boy told him, laying out a mat in a storage shed. The boy was there only a short time, it seemed, and then the baby was born. The mother took off her own robe to cover the child and was lying with it in her arms; suddenly she was no longer there. (p. 119)





Frame pattern Problem (A) Monk is very poor ‘desperately poor ’ ‘unable to obtain a residence’ Embedded pattern Response Seek Bishamonten’s aid ‘pray . . . to relieve him’

Result? Monk meets ‘temple boy’

Goal Temple boy’s desire ‘let me go with you Positive result /Goal achieved ‘I’ ll take him with me’ Question: ‘Can it be that you are a woman’

Projected response ‘Put her out’

= Problem (B) Buddha will be angry People will gossip

Actual response Made love to her Result She becomes Pregnant

= Problem (C) ‘what will monk do?’ Response ‘Do nothing’

Positive result: Monk becomes ‘prosperous in all things’

Result ‘boy’ and baby disappear, leaving rock of gold

Evaluation Miraculous provision ‘This is an example of the marvels wrought by Bishamonten’

Figure 3.6 Interrelated Problem–Solution patterns in The Tale of the Monk who Received Bishamonten’s Aid



This mismatch between the Peak marking and the climactic structure might be explained by the ambiguous construction of Problem–Solution patterns in this narrative. An outline of the patterning is presented in Figure 3.6.

This diagram shows the relationship between the framing pattern A (the monk’s poverty which is transformed into prosperity) and the Problem–Solution pattern B embedded within the ‘Result’ stage. It is the interrelationship between these patterns that reveals a paradoxical view of female sexuality which is focused in the Peak marking. On the one hand, the woman’s body was considered to be a site of impurity and an embodiment of vice, one of which was their capacity to arouse sexual desire in men (Pandey, 1995: 326). With this view of female sexuality, the temple boy’s true identity as a woman would indeed be problematic for the monk and put him at risk from spiritual and social admonition, ‘What the Buddha himself will think – that I dread!’ It is this negative aspect of sexuality that is highlighted through the use of Peak marking in pattern B which suggests that it is the ‘boy’s’ gender that is the crisis point for the narrative as a whole. However, on the other hand, the very instability of women’s bodies leads to them functioning as ‘the ideal trope for all the transformation tales in setsuwa’ (Pandey, 1995: 329). This often takes a negative form as they became deceitful foxes or oni5 but also had the potential to be positive on occasion. It is this second possibility that is used in this narrative, for the woman is transformed into gold, a gift from Bishamonten. I suggest that in order for the logic of the framing pattern A to operate, it is vitally important that the passage marked by Peak is not the main climax for the narrative. Instead of posing a problem, the boy’s feminine gender becomes the means by which the monk is able to receive aid. The framing pattern thus constructs the appearance of the boy and the subsequent events as a ‘result’ following the monk’s prayer to Bishamonten for aid. This legitimizes the events that follow and means that not only does the monk escape the negative consequences he projects for himself at the Peak passage but is actually rewarded for his incorrect behaviour of giving in to his desire. Clearly the content of this narrative is rich in examples of gender inequality. Not only does the woman become the object of and fulfil the monk’s illegitimate desire, she is blamed for this as ‘she was adorable and seductive’, and is then removed from the text and transformed into his literal possession and means of wealth. What is striking is the way in which the skewing of the plot structure through the interaction of the Peak marking and the Problem–Solution patterns operates to legitimize and express approval of this. The narrative of The Deeply Jealous Woman presents a further variation on the positioning and function of Peak marking. This story follows the transformation of an abandoned lover, the lady of Milo. In response to her abandonment, she disguises herself as a demon and runs away,




hiding in a broken temple. The local villagers hold the ‘demon’ responsible for the death of several of the young men and decide to burn down the temple. The ‘demon’ reveals her true identity to them and then burns herself to death in the fire. The Climax is not positioned in the middle of the narrative as a crisis that is resolved but forms a didactic conclusion to the tale, culminating in the woman’s plea: I beg all of you to get together and with all your might copy out the Lotus Sutra in one day and perform religious services on my behalf. Also, those among you who have wives and daughters, do not fail to tell them my story and forbid them to let feelings such as I had to arise in their hearts. (Pandey, 1995: 340) This alternative placement of the ‘Peak’ might be explained on two grounds. First, of all the narratives discussed in this chapter, The Deeply Jealous Woman is the only one that closes without a positive resolution. The predictable patterns reveal this, as indicated in Figure 3.7. It is feasible to argue that the Peak marking comes at the conclusion of this narrative because the ‘knot’ of the woman’s jealousy is never ‘untied’. The content of the narrative does not allow her the opportunity of repentance and transformation back to humanity and this is mirrored in the structure that does not contain a ‘final suspense’ as such, or at least only is accounted for by a single, brief sentence. Second, the Peak marking in this narrative does not serve only a Problem The woman’s feelings of jealousy

Response Dresses like a demon, leaves home, kills deserting lover

Result Cannot regain former physical appearance and suffers terrible internal burning

Negative evaluation/Final result = irretrievably negative Uses testimony as religious homily and then burns to death

Pattern ends Figure 3.7

Predictable patterns in The Deeply Jealous Woman




structural function but also corresponds to what Longacre labels as a didactic Peak (1983: 24). Longacre’s description of this structure is not particularly satisfactory. He says, ‘A didactic Peak is a special elaboration of some episode which precedes or follows the action Peak’ (p. 24). However, the principle of using Peak marking to highlight a thematic focus is interesting, and has parallels in Labov’s analysis of Evaluation. Both serve the structural purpose of highlighting the apex or the highpoint of the narrative events and demarcate the central turning point between problematic events (Complication) and their final outcome (Resolution). Labov goes on to note that Evaluation also serves a social, interpersonal function of foregrounding that which is ‘tellable’. It is ‘the means used by the narrator to indicate the point of the narrative, its raison d’être: why it was told and what the narrator was getting at’ (Labov, 1972: 366), all of which is motivated by the need to make the story relevant for its audience. In this case, both the Peak marking and the Problem–Solution pattern reinforce the moral import of the tale, which reiterate a specific gender ideology. This is that jealousy is a ‘quintessentially feminine failing’ (Pandey, 1995: 328, Tonomura, 1994: 138–140), with the woman herself becoming an embodiment of that sin. Thus it is the woman’s feelings of jealousy that are constructed as the ‘problem’ to be resolved, not the sexual promiscuity of the male lover, and it is for this that she is punished and serves as a warning to other women. While placing the Peak at the close of the narrative serves the purpose of reinforcing the gendered moral of the text, I do not wish to suggest that the form is itself gender specific. Instead, it points to the social function of the narrative as a moral exemplum, which is not a gendered form, but here is being used to support the gender ideology being emphasized in the text.

Conclusion The discussion of these medieval narratives has shown how a narratological analysis of plot structure can be useful in drawing attention to the ways in which various virtues and vices are constructed in relation to particular gender ideologies. Notably, the analysis has revealed a range of morals relating to femininity which are part of specific cultural contexts rather than a unified or abstract vision of the virtuous woman. The Japanese narratives also suggest variation within a given culture with contesting representations of pious women who can be agentive (The Pious Nun), jealous women who are unredeemed and an unresolved tension in how women’s sexuality should be treated. The study




of narratives from a feminist perspective does indeed, as Lanser suggests, benefit from the application of narratological frameworks. However, this is only one-half of feminist narratology’s directive. What remains is to consider the question of how narratology itself might be altered in response to this analysis. The first, and perhaps most obvious, observation we might make is that the study of these narratives supports Lanser’s claim that structuralist narratology ‘has been mistaken in trying to arrive at a single definition and description of plot’ (p. 357). The universalism in models like Longacre’s is untenable in the face of the alternative structural patterns that have been analyzed earlier in this chapter. If narratology is to take account of these variations, it is important that theories of plot are not formulated as a rigid typology, for example determined on the basis of the number of Peak passages present. I have argued that this kind of classification might be reductive and fail to account for alternative patterns that blur or skew narrative categories. Lanser proposed that a narratology for feminist criticism would ‘favor flexible categories over fixed sets’ (p. 345). My analysis endorses this and suggests taking a set of linguistically oriented parameters as a starting point for providing a replicable point of comparison for differently structured plots. I have derived this from Longacre’s description of Peak marking and Hoey’s work on predictable patterns, but these are not the only parameters that might be used. I have argued throughout for the dangers of abstracting a narrative structure from its specific content and context. In line with contextualist narratology more generally, the linguistic features of a text are not seen as isolated from but as functioning within particular social contexts. Thus the analysis I have offered in this chapter has revised structuralist approaches to plot by recognizing that the overarching framework and features such as Peak marking are not structural alone or just related to the content of their given text but rather also entail a social dimension. They may be shaped by ideological assumptions from their cultural context and may in turn be used to sustain, challenge or expose points of contradiction in those beliefs. Again, I reiterate that in doing so I want to avoid any unequivocal correlation between gender, form and function, but to acknowledge that the same form might serve very different purposes in alternative settings. Thus the emphasis on transformation is not exclusively male, and indeed this mono-climactic structure might be used to tell amoral or immoral stories that chart different types of change (gendered and non–gender specific) in other contexts.




I began this chapter by asking how we might explain differences in narrative form in relation to gender. The answer I have given is to provide an analysis which relates the patterning of plot to the content of these stories and their distinctive cultural context. On the basis of this, gender may be seen as an important influence not only on the content of a narrative but on the way in which that content is structured, insofar as that helps to reinforce the meaning(s) of the text. However, this relationship is an indirect one and the boundaries between context, content and form are fuzzy. For example, I have emphasised content and form as products which are acted upon by the ‘context’ but storytelling is a dynamic process which in turn may have influence in reinforcing certain views of gender behaviour. Further evidence from contemporary sources would be needed to gauge the extent and form of this. The narratological criticism of these medieval texts is not limited to matters of interpretation. Instead, it has been used to reformulate a theory of plot which is context sensitive, recognizes that there might be all kinds of narrative patterns at work and suggests that narrative structure may also serve social functions. Clearly, the patterns and functions of Peak marking have only covered a limited number of what might be an open-ended range of variations. Feminist narratology has more work to do yet in extending its breadth of analysis. It is then that it might begin to answer questions of why it is that certain narrative patterns become so popular and eventually normative in particular contexts.




4

Moving towards sociolinguistics The belief that literary language should not be isolated as mutually exclusive from language more generally is inherent in the contextualist narratology into which Lanser’s feminist narratology was launched (Pratt, 1977; Chatman, 1990). This position argues that narratology should broaden its boundaries and might benefit from taking into account linguistic models that have been derived from other types of data and that the kinds of narratives analyzed need not be confined to those from canonical literature. Labov and Waletzky put it like this: In our opinion, it will not be possible to make very much progress in the analysis and understanding of these complex narratives until the simplest and most fundamental narrative structures are analysed in direct connection with their originating functions. (1967: 12) While this stance has been widely adopted in stylistics, it is also important to bear in mind that narratives from different media and genres bring with them particular conventions, reader expectations and may function in various ways that need not map onto one another in a simplistic way. As the research which has adopted such an interdisciplinary approach has shown, moving outside the domain of literature can be useful not only in exposing points of similarity with other text types, but also in defining a given narrative’s characteristic distinctions (Fleischman, 1997; Fludernik, 2000b). In line with this, Lanser’s (1986) paper does indeed draw on linguistic research concerning the analysis of women’s speech styles (p. 348). Since then, a vast amount of research in the field of feminist linguistics has gone on to explore the ways in 73



The Importance of Similarities

which women and men use language, some of which has challenged the findings used by Lanser. It is perhaps then surprising to find that in much of the narratological work that has sprung from Lanser’s study there is so little cross-reference to the parallel research that has been carried out in linguistics, especially that concerned with gender and narrative. In order to address this lack, the discussion moves in a sociolinguistic direction here. In Chapter 3, I have argued for the importance of analyzing the relationship between gender and narrative structure from a contextual point of view. Here the relationship between text, context and gender is reconsidered from an alternative perspective, drawing upon a collection of oral narratives of personal experience. There are obvious differences between the contextual situation of literary narrative (where the author and audience might be removed from one another in space and time), and the oral narratives (where speaker and audience are present together in the immediate speech situation). However, while the discussion of context might need to take into account different types of factors, neither oral nor written narratives may be separated from the wider social context in which they operate. This comparison is particularly useful in reflecting on the ways in which gender and context may be formulated in feminist narratology’s discussion of literary texts. Just as Lanser argued that narratology might benefit from extending its corpus to include women’s texts (p. 343), so now feminist narratology should be challenged to widen its corpus to include texts other than literature in order to further define its understandings of narrative texts and theory. I do not propose that the findings should be the same. Nonetheless, the examination of gender as a potential influence on narrative form must take into account empirical findings of how women and men do tell stories. It is with this in mind that I turn to the collection of narratives told by women and men about the birth of their children. Research concerning the possible relationships between gender and language is wide and diverse. One stream of studies, ranging from the popular to the academic, has explored the differences in the ways that various groups of men and women use a range of linguistic forms. At its most fundamental, the study of difference establishes opposing categories of ‘masculine’ and ‘feminine’ linguistic behaviour, stereotypically presented as the division between competitive/cooperative styles which in turn is mapped onto a value system of hierarchy/intimacy (see Tannen, 1990; Coates, 1995). This identification of difference may also be found in studies that centre more specifically on storytelling styles, exemplified in the work of Johnstone (1990), Coates (1996) and




Meinhof (1997). While later studies have placed more emphasis on speakers’ ability to adopt a variety of speech styles and on the influence of context, there does appear to be some empirical evidence to support the suggestion that men and women may make use of language in differing ways (Holmes, 1998b; Pilkington, 1998; Coates, 1996, 2000, 2001). On the basis of these studies, there would seem to be grounds for further investigation of the ways in which various men and women tell stories. However, the study of difference is not unproblematic. The criticisms levelled at this school of thought are well documented (see Davis, 1996; Johnson, 1997; Cameron 1998a,d; Eckert and McConnell-Ginet 1998, 1999; Bergvall, 1999; Mullany, 2000). Amongst other arguments, Cameron makes the well-known but crucial point that all analysis and interpretation is partial and selective and that ‘where the object of observation and analysis has to do with gender it is extraordinarily difficult to subdue certain expectations’ (1998d: 270). Thus the cultural scripts of stereotypical gender difference may lead to the correlation of certain features with a particular gender while obscuring other linguistic features or possible interpretations so that similarities between speakers or differences that run counter to stereotypical trends are overlooked. Hence my exploration of the narratives in this study attempts to recognize both their similarities and dissimilarities and despite the essentialist difference inherent in the nature of the experience being recounted (that is, it is only women who can actually give birth to a child, whereas both men and women can participate in the experience as birthing partners) I do not intend to make any essentialist claims about the use of linguistic resources. Any tentative suggestions that may be made concerning the connections between the content and the form of the narrative and contextual factors such as gender remain precisely that and should not be taken as signalling a definitive correspondence. This is not only because of the nature and size of the data sample considered here but also because these narratives are situated in a precise social and cultural context from which it is at best undesirable to abstract universal principles.1

Data sample and analytic framework Previous sociolinguistic studies exploring the relationship between gender and storytelling have highlighted the contrasting topics and self characterization in the narratives by women and men. They suggest a pairing that parallels the stereotypical binaries of heirarchy/intimacy and competitive/cooperative styles where male speakers are represented



The Importance of Similarities 75


The oral narratives of male speakers described in classic accounts by sociolinguists and folklorists are about danger and violence, conflict and conquests. . . . The personal narratives told by women in conversation with their friends are not about dangerous or death-defying events. (1996: 99) Therefore, if the storytelling of men and women is to be explored more fully then narratives about a range of topics, including those that are not about activities encoded as stereotypically masculine, need to be analyzed. The birth stories in this study offer one such alternative. The data sample consists of nine pairs of narratives where women and men on separate occasions recount the birth of their children and of five narratives where women talk about their experiences as birthing partners, responding to a prompt question such as ‘Tell me about what happened when X was born’. The various practices and discourses surrounding childbirth have particular gendered values. There is a sense in which the experience may be regarded not only as centred on a female experience (biologically speaking) but also as taking part within a socially gendered context, as the comments made by both male and female speakers explicitly indicate (Examples 4.1 and 4.2). Example 4.1: Speaker (W) 1. I felt really sorry for him 2. ‘cause he walked in to this all female environment. Example 4.2: Speaker (H) 1. I think at first N. felt that a lot of the staff at the hospital were – what’s the word – a bit sexist, if you know what I mean 2. The other way to what you think of as being sexist 3. Thinking this is a female environment. Clearly, the subject matter is not one that readily lends itself to the male speakers representing themselves in a ‘heroic’ manner, and although one speaker did in a rather ironic way (Example 4.3), many of the male speakers commented on their feelings of helplessness, as Examples (4.4) and (4.5) demonstrate.



as heroic in stories that promote contest while female speakers are often shown to be foolish or embarrassed in stories where community is important (Eggins and Slade, 1997: 251). This difference has important implications for the study of narrative more generally. As Coates points out,


Example 4.4: Speaker (D) 1. I couldn’t do anything 2. Basically all I could do was be with O 3. The doctors were coming in every hour or so 4. Just checking her out 5. Making sure everything was ok 6. And I couldn’t be of any assistance 7. I couldn’t do anything 8. I was basically just lying on the floor trying to get some rest. Example 4.5: Speaker (B) 1. But I knew there was nothing I could do 2. Nothing I could say 3. No way I could comfort her to help her deal with it 4. Just allow her to get through it. As such these stories are interesting in that they provide examples of a very different kind of scenario from those considered by previous studies of gender and narrative, and while the nature of the experience is not gender neutral, for both women and men, being involved in the birth of a child is deeply personal and often emotional as the moving nature of many of these stories attests. The narrative framework used to analyze the stories is that developed by Labov and Waletzky (1967), Labov (1972, 1997) with a particular interest in the use of evaluation devices. As mentioned in Chapter 3, Evaluation has two important and interrelated dimensions. First, the location of the evaluation focus serves a structural purpose, for once this has been identified, then the global outline of the entire narrative may be categorized. Second, evaluation emphasizes the interpersonal nature of storytelling, for it highlights the material that the speaker is presenting as ‘tellable’ or ‘most reportable’, thereby making their story interesting for their audience. Thus the examination of evaluation devices provides a framework for exploring the ways in which different tellers give structure and meaning to their experience. On the basis of earlier research investigating ‘difference’ in speech styles, we might expect certain contrasts in the use of evaluation to emerge with women using more devices than the



Example 4.3: Speaker (J) 1. I was also suffering terrible elbow pains ‘cause I was having to rub her back 2. And I didn’t know if I’d be able to keep the momentum up all night 3. But anyway I did, being a true hero.


Evaluation as a structuring device A preliminary survey of the narratives indicates that in very general terms, this quantitative contrast in evaluation holds true. While on average the male speakers told longer stories than the women (2128 words compared with 1728 words), the women used more evaluation devices than the men (89 evaluation devices/100 words and 67 evaluation devices/100 words respectively). However, of more interest is a more detailed account of the distribution and function of evaluation. In his 1967 paper, Labov posits the ‘normal form’ for narratives of personal experience, that of AbstractÔrientation^ComplicationÊvaluation^Resolution^Coda. Clearly the birthing stories in this corpus are much longer than the minimalist narratives studied by Labov and need not include complicated or dangerous events (although some certainly do). However, given the climactic nature of the experience of giving birth, it might be expected that the narratives would contain a similar structural profile which increases in tension and then decreases after the actual birth of the child. Earlier studies suggest that stories told by women might differ from those told by men in terms of how closely they follow normative models such as that described by Labov. Gergen writes that Men are far more likely to accommodate themselves to prevailing criteria for ‘proper story telling’ than women. Women’s autobiographies are more likely, for example, to be structured around multiple end points, and to include materials unrelated to any particular end point. (1998: 6) Based on this statement, one might expect the structural patterning of evaluation devices to support the initial suggestion of a gendered difference in the storytelling. However, a close study of the birthing stories shows that neither the stories told by the women or by the men follow the outline proposed by Labov. Instead, all the narratives contained a wealth of evaluation distributed in a complex manner. To a certain extent, this is as Labov predicts when he says



men, for if the division of intimacy/hierarchy mapped on to cooperative/competitive and feminine/masculine styles is to be believed, then the use of evaluation to create an affectual relationship between speaker and listener might be higher in the stories told by women. These issues are now explored in relation to the analyses of the narratives.


The significant difference here is that the distribution of the evaluation devices in the birthing stories is not ad hoc but rather appears in multiple clusters as opposed to being concentrated in a single, clearly identifiable high point. In this respect, it is not just the women but also the men who depart from the ‘proper’ model and instead structure their stories around multiple points of focus. These evaluation clusters, like Labov’s Evaluation Section, occur when a speaker breaks away from narrating the main events of the story and provides some kind of commentary, explanation or opinion. The following example shows this pattern. It is found early on in the narrative told by this speaker where she talks about identifying the onset of labour. Example 4.6: Female speaker (M) 1. and then on the Monday I woke really early in the morning 2. and decided I would go swimming 3. and for the first time when I was swimming I actually had Braxton Hicks contraction 4. and I was like ‘Oh ok’ 5. but I didn’t really pay much attention to it 6. I suppose I didn’t really want to get too excited 7. cause I’d been waiting by this stage so long 8. that I was totally fed up with the whole thing 9. so I came back home 10. and just spent the day-

CA

Eval

CA

The separation of the evaluative commentary from the narration of events is emphasized here by the use of discourse markers. The narrative events in lines (1–4) are all connected by and, indicating the continuity in the sequence of actions. This repeated use of and also establishes a textual norm. As Schiffrin (1987: 131) points out, switching between the regular use of and and other connectives may signal a change in the discourse unit. This is precisely what happens here. Following the embedded Evaluation in (4), the connective changes from and to but, indicating not only the referential contrast in the speaker’s hope that labour has started and her restraining disbelief but also a contrast in the narrative units Complicating Action and



The evaluation of the narratives forms a secondary structure that is concentrated in the evaluation section but may be found in various forms throughout the narrative (1972: 369).

Evaluation. The lines (6–8) provide an evaluative explanation for the speaker’s disbelief in (5) which comes to an end when the speaker returns to the narrative events in (9). This return is marked by so, which in addition to its causal meaning functions here as the means of topic management. As with the earlier passage of Complicating Action, the use of and as a connective is resumed. While not all of the evaluation clusters in the corpus use this precise pattern of discourse markers, they all share the similar movement away from the event line into some kind of aside. Their widespread distribution across the narratives suggests the need for their closer consideration. One explanation for the multiple clusters of evaluation devices might be that in fact the birth stories need to be analyzed at a more delicate level and the clusters represent the apexes of mini-narratives or episodes. However, the evaluation clusters here do not seem to function in quite this way. First, there does not appear to be a clearly hierarchical arrangement of units with one section constituting the Peak (cf. Longacre, 1983) as in the case of the medieval narratives in Chapter 3. Second, the events that are marked out by the evaluation need not be problematic or some kind of crisis at all. For example, in some narratives, the speakers reflect positively on various aspects of the experience such as the quality of care received, seeing the child born or discovering the gender of their new baby, making considerable use of evaluation devices. The following extracts are typical of this. Example 4.7: Female speaker (M) 1. and I just said ‘what is it?’ 2. and they said ‘it’s a girl’ 3. and I couldn’t – I just was so surprised 4. I didn’t realize I had thought about it so much 5. that I had thought so much it would be a boy 6. because secretly I had wanted a little girl 7. and I was so amazed Example 4.8: Female speaker (P) 1. So then I sort of expected her to say 2. ‘Oh well, I’ll go home now and come back in a few hours time. 3. You just get on with it.’ 4. because I was having a home birth 5. so I didn’t have to go anywhere which was nice 6. but she said ‘Oh no, I’m going to stay now, now I’m here’





Example 4.9: Female speaker (V) (birthing partner) 1. So that was really special 2. Seeing his head come out 3. And thinking ‘Gosh, you know, I actually saw him before L. and X. did’ 4. You know and that was an incredible feeling 5. Really amazing. The corollary of this is that using the evaluation devices to locate the Complication and Result of the narrative framework becomes extremely problematic. Indeed the labels ‘Complication’ and ‘Result’ seem inadequate as a means of describing the nature of the events in these stories, because they are not necessarily focused on a central moment of crisis. This leads me to question how the pattern of evaluation in these texts might be interpreted. One possible answer may be found in the narrative genres discussed by Martin and Plum (1997).2 In their discussion, they identify a type of narrative known as the Anecdote. They describe this structure as an account of a remarkable event which is intended to produce some kind of shared emotional response between the speaker and the listener. The typical structure of this form is (Orientation)^Remarkable Event^ Reaction^(Coda). I would suggest that these birthing stories are better understood as a series of linked Anecdotes where the speaker evaluates certain events or aspects of the experience as particularly remarkable. This interpretation of narrative structure has much in common with Labov’s discussion of evaluation as that which coincides with the ‘most reportable event’ (1997: 406) but is more helpful in understanding these narratives in that it does away with the notion of a move from the classic plot structure’s progression from Complication to Resolution. Indeed, it highlights the importance of understanding narrative structure in relation to the nature of the story being told. Crucially, this acknowledges the limitations of abstracting Labov’s system of labelling from the data originally analyzed. As others have noted, the move from Complication to Resolution is inherent in the nature of the ‘danger of death stories’.3 However, in spite of the cultural valency of this pattern, not all narratives of personal experience need entail problems that get



7. and it was lovely 8. ‘cause she was just really relaxed and chilled out 9. and she had this, I don’t know, this really calming feel about her.

resolved nor does a remarkable event necessarily involve a negative scenario as implied by Eggins and Slade’s description of the Anecdote as a narrative centred on a crisis (1997: 237, 243). The very strength of Martin and Plum’s definition of the Anecdote as a narrative based on a ‘remarkable event’ is the way in which this takes account of the relative nature of what might be classed as ‘remarkable’ for individual speakers within particular social and cultural contexts. As such, interpreting these narratives of childbirth in terms of Anecdotes is valuable in recognizing that speakers will construct narratives of a wide range of experiences in very different ways. The equation of evaluation with the reaction to a remarkable event also explains the diversity in the placement and quantity of evaluative clusters and asides in the narratives studied here. Across the stories there did not appear to be any constraints on the number of evaluation clusters found in any one narrative, other than that there was always more than one. Neither did they consistently appear in the same location within the whole framework and might be found near the outset, closure nor any point in between the two. If ‘the most reportable event’ is considered to be that which ‘has the greatest effect upon the needs and desires of the participants in the narrative’ (Labov, 1997: 402) then this diversity may be understood. Even within the small number of participants interviewed in this study, a wide variety of birthing experiences are represented ranging from a water birth at home to elective and emergency caesarean sections. In addition, like the women who took part in Freed’s (1999) study, these participants had very different expectations of and hopes for childbirth. Given this diversity and the highly personal, individual nature of the experience it is not surprising that what is found to be ‘remarkable’ varies from speaker to speaker.

Evaluation as a means of constructing solidarity Despite the diversity in the placement of the evaluation clusters, they share a similar function. As Martin and Plum propose, they generate an affectual response shared by listener and speaker and hence negotiate solidarity (1997: 301). This interpersonal dimension of evaluation constitutes another area in which similarities and differences between women’s and men’s use of language might be explored. Studies focusing on the differences in speech styles points to the construction of solidarity between speakers as an area of gender contrast. This is summarized by Holmes when she says:





Contrary to this reiteration of the binary equation of women and men with cooperative and competitive speech strategies, other linguists have explored the ways in which men do in fact construct solidarity within single sex groups, although sometimes by very different means to those used by women. Studies such as Johnson and Finlay (1997), Pilkington (1998), Kuiper (1998) and Cameron (1998d) emphasize that the linguistic forms used by a group of speakers may vary according to context and that to suggest a mutually exclusive pattern of solidarity as opposed to status seeking speech styles is simply not tenable. In the narratives considered in this study, the evaluation used by both men and women results in a similar pattern of shared emotional response. This is indicated in various ways. First, and most obviously, the listener might explicitly express their reaction. This is seen in the following example. Example 4.10: Male speaker (G) 1. So I honestly thought at that point – I thought we’d uh lost the baby 2. I really did 3. I thought we’d lost the baby 4. 5. I was – it was horrendous An alternative indication of emotional response is laughter, often shared between speaker and listener. This was a common occurrence and can be found in all but one of the narratives. The analysis of humour is complex, and, as Holmes (2000) indicates, manifestations of humour may be ambiguous and multifunctional. Indeed, laughter itself is not straightforward and might signal a range of emotions or reactions ranging from ridicule to recognition. The instances of humour marked by shared laughter in these narratives covered a range of topics. Using the framework developed by Hay (2000), these topics fall into two main categories. The first of these is ‘coping humour’. This seems to be a face-saving device whereby the speaker and listener use laughter to distance the anxiety or embarrassment inherent in the material being narrated. The following extracts are all examples of this.



Women tend to interact in ways which will maintain and increase solidarity, while (especially in formal contexts) men tend to interact in ways which will maintain and increase their power and status. (1998b: 472)


Example 4.12: Female speaker (P) 1. By that time I was kind of ‘Yeah I pretty much must be in labour 2. ‘Cause it hasn’t stopped 3. But you know, I’m still ready to be embarrassed and her just say 4. ‘Oh no, it’s a bad case of bowel trouble’ (both laugh) Example 4.13: Male speaker (B) 1. And there’s blood and water everywhere 2. And I don’t know where to stand or what to do (all laugh) 3. ‘I don’t know what to do’ (sing song) 4. So I just decided on patting Rachel’s brow (all laugh) The second strategy that is used in the narratives is ‘sharing humour’. This occurs when the speaker reveals something about themselves ‘and lets the audience know them better’ (Hay, 2000: 718). Both men and women make use of this strategy, which serves a function of creating solidarity. However, within these instances there is a distinction in the type of material that prompts the laughter. In Coates’ discussion of women’s narratives, she notes that the listeners show their identification with the speaker by laughing when the speaker expresses her inner thoughts (1996: 107). Similarly, in these women’s narratives, as well as laughing at events, speaker and listener often respond to the women sharing their internalized expectations or evaluation of events. This happened in 23 cases (that is, on average 1.6 instances/narrative). The following examples are fairly typical. In the first, the speaker and listeners share amusement at the speaker’s reaction to her arrival in hospital, while in the second they laugh about the speaker’s expectations of mess. Example 4.14: Female speaker (M) 1. Well we arrived at hospital 2. and I was about one and a half centimetres dilated (all laugh) 3. I was so gutted 4. ‘cause in my head I’d always assumed like I’m going to wait at home 5. and they’re going to tell me I’m seven centimetres (all laugh) 6. I was like ‘One and a half?’



Example 4.11: Female speaker (O) 1. I’d always wanted it to be wrapped before it was put on me belly 2. ‘Cause I just didn’t want all the slime and the gore (both laugh) 3. And I was just thinking ‘Thank God they’re not going to be able to do that now’

Example 4.15: Female speaker (V) (Birthing partner) 1. very often you know like I’ve heard dads say ‘Oh the mess you know and the blood and guts and everything that comes out – Oh yuck’ 2. so I was sort of expecting you know needing galoshes or something (all laugh) 3. and you know ‘is that it?’ In contrast to this, in the men’s narratives, the shared laughter occurs only once when the speaker is relating his thoughts (that is, on average 0.1 instances/narrative). On all other occasions, the laughter is a response to an external incident, as in the following example. Example 4.16: Male speaker (G) 1. So I was in the outside lane of the dual carriageway 2. and I did my lights as if I was pretending to be a police car (all laugh) 3. flashing 4. and it was amazing 5. everyone just got out of the way So while shared laughter in response to evaluation is a common strategy for creating solidarity in the storytelling of both the men and the women, there does appear to be some difference between the evaluated material that prompts this reaction. This contrast might be interpreted in terms of the stereotypical characterization of masculine discourse as more concerned with informative meaning (Holmes, 1998b: 464), for it is primarily the events that are presented as amusing by the men whereas it is both the events and the speaker’s internal reaction that is recognized by the women.4 However, this contrast between external and internal material might also be seen in the light of the speakers’ varying participation in the practices and discourses surrounding childbirth. Many of the male speakers in this study quite clearly had become involved in the preparation for the birth of their children and talked about their anxieties and expectations of the process. However, in general, women tend to undergo much greater preparation, for example, through classes, antenatal appointments, reading magazines and so on. This asymmetrical pattern appears to be a general trend, as indicated by Greenhalgh et al. who comment that





To a certain extent, the women’s foregrounding of internalized reactions might reflect this difference. However, the female birthing partners who did not necessarily attend classes or become involved in the preparation for the birth in the same way that the prospective mothers might also presented their expectations in a way that prompted an affectual response of laughter. Therefore, there do seem to be some grounds here for tentatively arguing for a correlation between an emphasis on external events as opposed to internalized reactions based on the gender of the speaker, although this needs to be set against other possible complicating contextual factors that will vary from speaker to speaker. Further support for this interpretation may be found in the speakers’ differing use of lexical detail. When relating the events and participants of the narratives, the male speakers tended to be more precise in their accounts, often providing names or more detailed descriptions. This can be seen in the following examples where the subsequent male and female speakers form reciprocal pairs. Example 4.17: Male speaker (E) 1. and then the doorbell went at eleven o’clock 2. and of course it was all homegroup were on the doorstep 3. there was Tracey, Sandra, Andrea Jones and Mary from next door was there as well 4. and they were all saying ‘Ooh what’s happening? What’s happening?’ Example 4.18: Female speaker (P) 1. so E. went to the door to let her in 2. and it was all the neighbours standing on the front doorstep 3. going ‘What’s happened? What’s happened? We’ve heard we’ve heard-’ Example 4.19: Male speaker (B) 1. as we walked up the road 2. we’ve got friends down the road, Peter and Wendy and Michael and Julie, Simon and Liz



It is rare for a man to attend classes without his partner. Therefore, other factors, such as the woman’s desire to attend or not attend classes and the nature of the classes available, may play some role in determining fathers’ attendance. (2000: 178)

3. 4. 5. 6. 7. 8. 9. 10.

so we stopped in each of their houses and told them what was going on we said ‘We think it’s all happening’ (laughs) it was really good in fact it was one of the most poignant parts of it really ‘cause we’ve had this community thing in the road for so long so to just be able to drop in and say – for them to be able to participate in it all was just really really nice

Example 4.20: Female speaker (M) 1. so we just waited around a bit 2. and told everybody in the road that finally things might be happening Example 4.21: Male speaker (K) 1. I think if I remember rightly there must have been four baths in the Jacuzzi bath Example 4.22: Female speaker (W) 1. I went in the spa pool for a bit on and off, on and off Interestingly, the male speaker’s use of more precise orientation provides an alternative to Johnstone’s claim that Women use more detail about people because their stories tend to take place in a social world, defined by people, whereas men’s stories take place in the material world of objects and events. (1990: 94) While the men’s use of detail is perhaps more factually precise and so more akin to a testimonial account, their narratives quite clearly are part of a social world where the relationships between people are of great significance, at least for some, as Speaker B’s comments about the value of belonging to a local community suggest (Example 4.19). In contrast to this orientation material, when the speakers refer to medical procedures or personnel, the female speakers tended to use more precise terminology than the males who employed more generalized or euphemistic terms, often hedged in some way. The following examples are typical of this comparison.





Example 4.23: Male speaker (A) 1. after L. was whatever it was cut

Example 4.25: Male speaker (C) 1. she was just an assistanty helper type person 2. the guy, the surgeon guy, or whatever he is Example 4.26: Female speaker (N) 1. two student midwives 2. you know John the surgeon Example 4.27: Male speaker (K) 1. and they called in the paediatrician? Obstetrician who came in with her salad servers Example 4.28: Female speaker (U) 1. so basically they gave me an episiotomy and delivered her by forceps These contrasts in the choice of vocabulary signal more than a simple concern with conveying information, also indicating speakers’ alternative perspective on and participation in events. In particular, the female speakers’ use of medical terminology and the male speakers’ generalizations might be understood as examples of overlexicalization and underlexicalization respectively.5 Rather than suggesting that the male speakers do not have ‘access to the concepts concerned’ (Fowler, 1996: 216), the use of vocabulary might indicate the differing ways in which the men and women interact with the various groups and medical discourses that surround childbirth. While women ‘learn new ways of talking about their physical needs’ and become involved in these communities, albeit temporarily and partially (Freed, 1999: 258–259), as mentioned earlier, men’s participation remains considerably more peripheral. Clearly the degree of involvement will vary from individual to individual, but in general terms at least, this distinction is exemplified at a socio-political level in the asymmetrical maternity and paternity rights current in the UK at the time of writing which give the prospective mother but not the prospective father leave from work to attend antenatal appointments and classes.



Example 4.24: Female speaker (L) 1. I didn’t need an episiotomy


The point we have reached is that there are various similarities and dissimilarities in the way the women and men narrate their experiences. The various lexical choices along with the differing material that prompts shared laughter suggests a subtle contrast where the men’s storytelling seems to be more concerned with conveying factual information than the women’s. The question then remains of whether or not these features may be interpreted as the result of gender difference. On the surface, it is tempting to propose that the differences are indeed evidence to support the gendered pattern discussed earlier that men are focused on informative content whereas women tend to be more concerned with affectual meaning, a binarism rooted in the metaphorical pairing of logic and emotion. However, even if this gendered pairing is accepted (and which I argue against below) further questions might, and some would argue should, be asked such as where the origins of the division of men’s concern with information and women’s emphasis on emotion comes from. One possible explanation for this corresponds to the cultural explanation summarized and critiqued by Holmes (1998b: 477) where it is claimed that women and men are socialized into different subcultures where the well-rehearsed terms of masculine hierarchy and feminine intimacy are set against one another. The basis of this pattern might be interpreted in terms of social and economic factors. Commenting on the nature of women’s friendship, Johnson and Aries write The character of close female friendship obviously owes much to women’s historic social position. The quintessence of these friendships is deeply rooted in sex-segregation and its division of male and female spheres. Women who devote most of their time to family nurturance and homemaking usually find it convenient and desirable to seek relationships with women in similar positions; they find in these relationships many bases for sharing and many opportunities for support and comfort. (1998: 223) While for any number of reasons women continue to make other choices than to prioritize homemaking, it is notable that the asymmetry in the maternity rights available to speakers in this study reinforces this pattern. This is reflected in choices made by the speakers in terms of their occupations, where only one woman had chosen to return to paid employment on a full-time basis and only two of the



Summary and explanation

men work flexi-hours in order to accommodate daytime childcare needs. According to this paradigm, exchanging information might be considered more fitting to the public sphere. However, explaining linguistic differences in terms of gendered behaviour is not only reactionary in so far as it reinforces essentialist notions of masculinity and femininity, its explanatory power is also limited and overlooks the significance of the macro-level similarity between the narratives: that both women and men construct their experiences as a series of linked Anecdotes rather than as more conventional, plot-driven stories. At a basic level, the corollary of this is that a mutually exclusive categorization of storytelling style according to the gender of the speaker is simply untenable. When considering the reasons for a speaker’s use of an Anecdote structure, it is readily apparent that gender is not a variable that can be isolated but must be considered in relation to a nexus of potentially influential factors that might be broadly described as contextual. As both Martin and Plum and Eggins and Slade note, the characteristic social function of the Anecdote is to promote solidarity between speaker and listener. This emphasis on solidarity is highly pertinent to the elements of context in which these narratives are situated. One such factor is the relationship between the speaker and the listener which in the case of these narratives is one of friendship where agreement to take part in the interview came in a spirit of cooperation, partly expressing that friendship. As such, the choice to use Anecdotes is not that surprising, for as Eggins and Slade suggest, the choice of storytelling genre appears to be related to the ‘degree of familiarity or contact between the participants’ (1997: 266), where the occurrence of Anecdotes is associated with close contact between speaker and listener(s). This is easily understood, for using a structure which promotes a shared emotional response to events might be considered as a highly appropriate means of telling a story between friends where status and power are not at stake, compared with a more ‘plotted’ version of events that might be told in more formal contexts and for different purposes.6 Of course, the strength and nature of individual friendships is almost impossible to quantify with each relationship entailing a unique personal history. More generally, there is variation between speakers in this study in that some interrelate to form subgroups of smaller, denser social networks. For example, speakers B, E, M, P and W all belong to the same local community, living in the same street, attending the same church and sharing leisure activities together.




Other, less dense networks include those between speakers (A, C, L, J, N, T) and (D, G, O, S). These factors clearly complicate a simplistic correlation between the use of Anecdotes and the friendship between speaker and listener. However, I would argue that at a broad level, the co-operative motivation for storytelling, that is the need to maintain good social relations, would seem to supersede these individual differences. The possible influence of other communities of practice is worth mention here. Eckert and McConnell-Ginet describe a community of practice (CofP) as An aggregate of people who, united by a common enterprise, develop and share ways of doing things, ways of talking, beliefs and values – in short practices. (1999: 186) As Freed (1999) points out in her paper on this topic, the grouping together of women purely on the grounds of a shared experience of pregnancy does not constitute a CofP. In the same way, the speakers of this study, while they share the common factor of having at least one child, cannot be said to form a single community in the sense of the definition given above. Nonetheless, they have all participated in some of the ante/postnatal support groups7 that do form communities with their own distinctive practices. While individuals will no doubt have varying responses to such groups, one important characteristic of these organizations is that they aim to foster support between group members, often through the sharing of experience. Indeed recounting the experience of childbirth is often part of the initial reunion of antenatal groups. It seems possible to argue that the mutually supportive nature of these communities has some influence on the data collected in this study, not only because of the currently perceived relationship between speaker and listeners but because of past experiences and the pre-rehearsed nature of these particular narratives. As Eckert and McConnell-Ginet write: Styles and frameworks developed in a CofP, which have been experientially very important to individuals in their previous histories, may be carried into interactions with strangers, or into communities of practice new to them. (1999: 189) Further research including narratives from speakers who had not taken part in such groups would be needed before a more definite conclusion could be asserted.




Precisely how the gender of the speaker might interact with these factors is not a straightforward matter to assess. As discussed earlier, the men’s involvement in the CofP surrounding childbirth is different from that of women’s, and will also vary from speaker to speaker. Similarly, the gender of the speaker might be seen as influencing the nature of the friendship between speaker and listeners. The data in this study provides a contrast between the women’s single-sex conversations and the mixed sex of the women’s and men’s. This parallels a difference in the friendships too. While the women would often meet up together either on a one-to-one basis or in bigger groups, it was far more common for the men and women to meet up together in a larger social group and on a less frequent basis. Thus the single-sex friendships between the women might be regarded as closer than that of the mixedsex relationships, not least because of the increased time spent together. So in general terms, the friendship between speaker and listener might well explain why expressing solidarity through using Anecdotes is important to both men and women. However, the differences between the narratives may be understood in terms of the relative closeness of the friendships, for to disclose and respond to an individual’s inner thoughts (as in the women’s narratives) is perhaps more personal than to simply give an account of what happened, mirrored in the more precise, factual detail provided by the men. To conclude, a close study of these narratives revealed that there are some differences in the ways that the female and male speakers tell their stories, but that there are also important similarities too. While gender difference does correspond with the contrast in some linguistic features, it is not possible to state definitively that ‘gender is the most salient variable’ (Cazden, 1997: 186) for these narratives. Indeed, a range of other factors appears influential, not least of which is the particular context in which the narration of these experiences is situated. Once again, the caveat against decontextualizing results deserves reiteration. Further research extending the contexts and topics of narration is required before more certain conclusions can be drawn concerning different women and men’s uses of language in storytelling. However, the analysis of these oral narratives has significant implications for the study of literary texts in feminist narratology. First, and perhaps most importantly, it reinforces the argument I have put forward in Chapter 2: that gender and narrative form should not be equated simplistically with a binary model of difference. Feminist narratology must take account of the changes in theorizing that have subsequently drawn attention to the problems with such a model, summarized in Eckert and




McConnell-Ginet (1999). This includes radically revising the notion of gender as a fixed, stable category and recognizing the potential that speakers have to ‘perform’ their gender in variable ways (cf. Butler, 1990). Alongside this, it is also important not to obscure variation within categories, and, as this study has shown, also to recognize the points of similarity in stories told by, about and for women and men. The sociolinguistic approach that I have adopted in this chapter has also shown the fine-grained and complex way in which gender might interact with a number of other variables, and that the telling of these tales is embedded within the particularities of a specific context. This observation might be usefully applied to the study of literary narratives. Even while the contextual factors may differ in their immediacy, the study of literary texts must also guard against abstraction and should bear in mind the potential influence of factors that include beliefs and practices that relate not only to gender but also to ethnicity, class, historical period, mode of delivery and numerous interconnected others. In so doing, feminist narratology’s study of texts both literary and otherwise must avoid perpetrating stereotypes and false dichotomies rooted in a metaphorical basis of unqualified ‘difference’. To make a more convincing discussion of the ways in which gender might intersect with the characteristics of narrative form, considerably more empirical data are needed, both about the tellers and the receivers, as well as the tales themselves. It is with this broader picture of narratives from different media and cultures that the analysis of texts given the status of ‘literature’ might be more fully understood.




Questioning Intra-category Variation

Feminism and reader response In Lanser’s (1986) paper, she suggested that structuralist narratology be rewritten in a way that ‘takes into account the contributions of women as both producers and interpreters of texts’ (p. 343, emphasis added). As I have indicated in earlier chapters, some considerable work has gone into recouping the value of texts written by women. In comparison, relatively little analysis of the role of gender in the reception of texts has taken place in feminist narratology. The feminist critique of reader response theory that runs contemporary to Lanser’s paper did do just this, arguing that gender made a difference to the way in which texts were read (summarized in Caughie, 1988). The apparently universal nature of the various readers constructed by Barthes (1977), Iser (1978) and Fish (1980) were exposed as models of reading that dealt in ‘male terms’ and favoured erotic paradigms derived from masculine metaphors (Rivkin, 1987: 11–12). In finding alternatives to these, the identity of the reader as a woman and/or as a feminist was then interpreted as a political practice. However, I question whether these studies could be classed as narratological as such, for they did not really deal with the interaction between gender, reader response and the formal operations of narrative texts, nor did they reflect on narratological theory. Indeed, the inclusion of women (and men) as interpreters of texts might be regarded as problematic for narratology per se. The study of a reader’s response to a narrative is at one step removed from, although indisputably connected to, its formal characteristics and so less closely aligned with structuralist narratology’s goal of producing a systematic study of the features of narrative texts. Indeed, studying readers’ responses is notoriously more difficult than focusing on the text. The act of reading itself is less than 94



5

easy to scrutinize, being so laden with familiar conventions as to become almost invisible. As Jane Yellowlees Douglas puts it, ‘In a reading environment crowded with conventions, cues and decorum, we can never hope to clarify precisely what happens during the transaction between reader and text’ (1992: 2–3). A reader’s expressed interpretation of text is also susceptible to question, ridden as it is with subjectivity. Even if this is given the ‘reality test’ of consensual validation (Rogers, 1982: 42), a range of contextual factors may be brought into play that make these comparisons difficult to replicate. Despite this, there is some recent evidence that the interpretation or response to a narrative may be influenced by the gender of the reader/ viewer. Fludernik’s (1999) study of narratives where the biological sex of the protagonists is indeterminate is supported by the interpretation of a number of actual readers. She hints at the possibility of the reader’s reactions being influenced by their biological gender (p. 156), and draws attention to the gender-specific nature of schemata that readers (regardless of their biological sex) make use of when interpreting these texts. Warhol’s (2003) recent, more extensive study has the rather different focus of examining readers’ affective responses to popular culture narratives. Her work is important, not least because of her rewriting of effeminacy that dissolves binary gender classifications. Reflecting postmodern gender theory, she argues that the readers’ emotional responses are not determined by their biological gender, but rather the sensations and affect which she terms ‘effeminacy’ are ‘performative’ and open to men and women, regardless of sexual orientation. Warhol’s book opens with a series of vignettes, taken to be representative of actual readers’/viewers’ responses. However, the detailed attention to actual readers is not sustained throughout. Perhaps because of the difficulty of quantifying emotional response, Warhol chooses instead to focus ‘not on the “real” feelings of (all) “real” readers, but on generic textual patterns that structure the feelings of fans of specific popular genres’ (p. 22). While this is understandable, in my opinion, Warhol’s work would have been stronger still had she included more data from these ‘real readers’ and thus tested out the affective power of her textual analyses. Dealing with the practices and interpretations of actual readers may be more difficult to analyze, but I suggest that this is an area that needs to be more closely addressed in feminist narratology. It is on the basis of such evidence that the correlation between gender, patterns of interpretation and narrative can be more clearly examined, and in so doing refute theoretical stereotypes of gendered behaviour as both Fludernik and Warhol begin to do.



Questioning Intra-category Variation 95

Such an analysis should question the ways that practices of reading are gendered (if they are at all), and how these might be related to the structural features of narrative texts. My own response to this is to investigate the responses of real readers as they interact with an experimental form of digital storytelling: a multi-stranded narrative presented in hypertext. I have several reasons for choosing this as a means of looking at gender, narrative and reader response. The first is that the medium of hypertext enables us to put the practices of reading under scrutiny in a way that print narratives do not (Douglas, 1992: 13). One of the persistent problems for feminist readers’ response theory is how to measure the similarities and differences in the ways individuals approach a print text. The readers’ navigated pathways and their evaluative commentary on this provide two such measures for analyzing the reading of this hypertext narrative. Second, the process of reading a hypertext narrative involves a measure of textual construction and interaction unlike that of conventional print narratives, thus the reading patterns are directly related to the formal characteristics of the narrative in a clear and overt way. Third, it has been claimed that what is read influences how it is read. The emergent forms of hypertext narrative have yet to be considered by feminist narratology. In this way, my discussion offers a new perspective on the potential influence of the narrative’s medium. This is important, given that hypertext has been theorized as gendered in contradictory and contested ways (Brady Aschauer, 1999). The results of the experiment that I discuss in this chapter thus have two-way implications of relevance not only to feminist narratology but also to the developing criticism of hypertext fiction.

Hypertext, gender and reader response The figure of the reader is a prominent one in theoretical discussions about the emergent forms of storytelling that make use of hypertext technology. The reader of hypertext is said to be active (Liestol, 1994: 99), empowered with creative status as a reader/author (Landow, 1997: 195), potentially experiencing agency (Murray, 1997), immersion (Ryan, 2001) and interaction in the process of reading (Douglas, 1992), although some have gone on to question exactly how liberating this empowerment actually is (Murray, 1997; Gaggi, 1998; Miall, 1999). Generally speaking, the formal characteristics of experimental literary hypertext (such as Joyce’s Afternoon and Moulthrop’s Victory Garden) are said to lead to ‘new practices of reading’ (Liestol, 1994: 87), where, to




use Jane Yellowlees Douglas’ terms, readers become ‘inner directed’ and adapt to the differing demands of hypertext narrative (1992: 108). This includes using multiple readings in order to make sense of the text (Selig, 2001: 643) and building a mental map that encompasses the multidimensional possibilities of the hypertext narrative world, rather than seeking a single, linear sequence that reaches an unequivocal point of closure (Douglas, 1992: 99). However, how far these ‘new practices’ intersect with patterns of reading driven by print technology or are influenced by other literary schemas remains a point of debate (Ryan, 2001). The picture of the hypertext narrative reader is thus a complex one with many unresolved questions. Critically, with the exception of Douglas’ work, this is further problematized by the theoretical and universalized figure of the ‘reader’ in much of the discussion. As Douglas and Miall have pointed out, there is still relatively little evidence of how actual readers do respond to literary hypertexts. While Douglas’ work is a provocative and well-designed study that does address this issue, even her discussion does not pay attention to the potential influence of variables such as age, gender or race in the reading practices of the students who took part in the experiments. This study is a small step in extending this area of research further, particularly focusing on the possible influence of the reader’s gender. Gender has also featured in theoretical discussions of hypertext as an emergent form, where the opposition between the features of hypertext and older technologies has been mapped onto gendered values. Thus the linear argumentation of forms previously found in printed media (for example, academic discourse or, to take a literary example, Aristotelian plot) set alongside the assumption that the author is the centre of creative power have been inflected with patriarchal values. Hypertext has been contrasted with this. It is said to be non-linear, operating on associative principles; it apparently empowers the reader; in its multilinearity it offers the possibilities for alternative voices or discourses to be expressed alongside others. Both in terms of non-fictional texts (for example, the writing of academic papers, the construction of web pages and discussion fora) and literary fiction, the hypertextual experience provides possibilities for exposing, crossing and sometimes deconstructing boundaries and opening up new spaces in which alternative voices and identities may be expressed (Odin, 1997: 599). Despite the association of technology and masculine domination, at a superficial level at least, hypertext seems to realize some of the characteristics of écriture feminine proposed by French theorists and summarized by Jane





In fact, like Patchwork Girl, Quibbling makes us wonder whether hypertext fiction and, indeed, all hypertext is in some way a feminist sort of writing, the electronic embodiment of that l’écriture feminine for which Hélène Cixous called several decades ago. (1997: 206, my emphasis) As just mentioned, there are clear parallels between hypertextual forms and the formal features of communication called for by some feminist theorists. Indeed, these have been effectively used in hypertext narratives that might be interpreted from a feminist perspective. Many of the pieces housed at Assemblage, the Women’s New Media Gallery curated by Carolyn Guertin, could be cited as examples of this, such as Caitlin Fisher’s These Waves of Girls. In this piece, the creation of feminine identity and exploration of female sexuality (lesbian and heteronormative) is aptly supported by the hypertext form which is cyclical, multimedia and associative. The structure of this hypertext narrative is like a series of waves, (both refractive and cyclical) and in their metaphorical fluidity fulfil Cixous’ call for writing which ‘bursts partitions, classes, and rhetorics, orders and codes, must inundate, run through, go beyond the discourse with its last reserves, including the one of laughing off the word “silence” that has to be said’ (1989: 113). As such, it might be considered as highly appropriate for writing about the female body, with its association of rhythm, cycles and fluidity. However, I would argue that to claim that ‘all hypertext writing is in some way a feminist form of writing’ is too strong. In common-sense terms alone, it would seem perfectly possible to envision hypertext writing that in its content might be quite contrary to the feminist aims of ‘destabiliz[ing] oppressive hierarchies and cultural hegemonies’ (Clark, 2002), and instead might reinforce gendered hierarchies. Can Joyce’s Afternoon or Moulthrop’s Victory Garden be said to do this? And what of the virtual reality games described by Douglas (1996) that force the reader/viewer to adopt the perspective of a straight male character? Murray’s work on hypertext does not explore gender as such, but briefly mentions an example of individuals



Love as ‘disruption, resistance, multiplicity, emotional texture, nonlinearity, formal experimentation’ (2002). It is perhaps not surprising to find this gendered equation of hypertext and feminism reiterated in Landow’s discussion of fictional narratives written in hypertext form. He writes,

who chose to reinforce gender stereotypes in their interactions with a virtual reality game Sim City (1997: 88). Just because a text is formally transgressive, it does not entail that its representation of gender will be so too. Thus it is not hypertext as a form that is inherently feminist in some way, any more than any linguistic form can itself be seen as such. Indeed, I would go further still and question exactly the ways in which hypertextual writing or practices of reading can be seen as gendered at all. One such exploration draws on the concept of narrativity. Various critics have described the ways in which the narrativity of hypertext stories may be disrupted by unconventional reconfigurations of story and discourse time, multiple storyworlds or seemingly incompatible versions of the same storyworld, fragmentation of character and hypertext’s apparent resistance to closure (Douglas, 1992; Landow, 1997; Brady Aschauer, 1999; Selig, 2001). These are the very features that Landow and the feminist critics have taken as the basis for arguing that hypertext be gendered as feminine (Sullivan, 1999; Love, 2002). By extension, this reiterates the pairing of narrativity and gender that I problematized in Chapter 2. Once again, I challenge this theoretical equation, this time by investigating what kinds of narrativity strategies the readers draw upon when interacting with the hypertext. This concentrates on two areas. First, the pathways that the readers navigated through the hypertext were analyzed in terms of their linearity. Second, the readers’ creative extensions of the original hypertext were examined, looking at the actions and characters that were constructed. As I discussed in Chapter 2, those factors that enable a text to be processed as a narrative are many and varied. Neither linearity nor action is the only index of narrativity. However, they are both implied in criterial definitions of narrativehood. For example, Labov writes, ‘We define narrative as one method of recapitulating past experience by matching a verbal sequence of clauses to the sequence of events which (it is inferred) actually occurred’ (1972: 359–360, my emphasis). The content of a text, that is how ‘event-like’ the experiences might be and the form in which these are relayed, that is what kinds of sequences are constructed can thus be regarded as particularly high-level factors involved in distinguishing degrees of narrativity here. I should stress again that I am not presupposing that either parameter is inherently gendered. Rather, I wish to use these terms to test the variability of reader response and so in turn throw further light (or caution) on the equation of hypertext, reading practice and gendered values.





As Landow writes, hypertext fiction comes in many different forms and genres, and it would be mistaken to assume that all narratives written using hypertext share the same non-linear features. Like print narratives, genre and form can be manipulated in many ways with different effects. The text that I chose is an amateur piece of writing by Charles Sundt called Fishnet. Fishnet might best be described as transitional writing rather than as native hypertext narrative. It consists of eight story strands about the same set of characters that were originally written as conventional short stories and then integrated together into one overarching web. As such, Fishnet does not contain many of the subtleties and characteristics of more complex, experimental hypertext narratives. For example, it retains a high degree of linearity, it is not multimedia (it contains no image, sound or animation), and it does not have multiple endings or multiple retellings of the same sequence. Although there are discussion pages linked to the story, the reader can only comment on rather than contribute to the story content. The stories of Fishnet are divided into short lexias amounting to about 280 text chunks in total, connected by around 600–800 links. The hyperlinks that connect the chunks are either associative (highlighted lexical terms) or follow a linear pattern (a sequence of page numbers given at the bottom of the lexia). Apart from this, the reader is given no guidance on how they might find their way around the text. The links between stories via the associative terms are rarely based on logic or coherence. Rather the author designed the links purposefully to ‘distract’ the reader, linking scenes primarily on the basis of how exciting they were (personal email communication). The underlying principle is that the reader treats Fishnet like a series of stories being screened simultaneously that they can switch between, following a given story strand if they become particularly interested in it. The rationale for choosing Fishnet was not that it was an example of fine hypertext writing. Rather, in its simplicity, it enables the reader to achieve a reconstruction of a conventional narrative, or to incorporate a more experimental way of reading. This could be measured relatively easily by tracking the individual pathways through a reading log. Seventy readers took part in the experiment, of whom thirty-five were male and thirty-five female. However, to categorize the readers simply on the basis of their biological gender is too crude. As we have seen in the earlier chapters, such variables should not be abstracted, but grounded in the specifics of their local context. Influential writers Eckert and



Data sample and methodology

McConnell-Ginet (1999) rightly point out that studies that focus on gender categorization as an explanation of difference may overlook important intra-category variation. To that end, I want to draw attention to the potential influence of one particular difference between the groups of women and men readers. This is that while all were students at UK universities, they were studying for degree programmes in different academic disciplines. The first group (42 students) was based in School of English, within a humanities discipline. The second, slightly smaller group (28 students) were studying within a faculty of Computing, Engineering and Technology. The influence of disciplinary background may be of particular importance in this study. On one level, the constituents of these groups reflects the gender asymmetry of students in these fields, where male students predominate in the computing course and the female students outnumber the male students in the School of English. These patterns may well influence the interactions of individuals within educational settings.1 From the point of view of the data being considered here, these are perhaps of less importance than the differing knowledge base and skills that are emphasized in these disciplines, both of which may be influential in the way the readers approach the hypertext narrative. I do not want to suggest that computing studies embody a ‘male’ perspective or literary studies a ‘feminine’ one.2 Rather, on the one hand, the computing students have a more detailed knowledge of the technology and information systems involved in building hypertext. On the other, the students in the School of English may well have been trained to approach the narrative with a degree of critical distance (Fialho, 2004). While all the readers inevitably bring with them the experience of prior texts and real world schemata, the critical theory and range of literature required by the curricula in the School of English may well be of specific influence too.

Reading pathways: Degrees of linearity One way of exploring the narrativity re-created in the reading of hypertext fiction is by examining the pathways navigated by the readers. These were recorded in a reading log as the readers worked through Fishnet. The pathways were then analyzed in terms of what kinds of links the readers made between the different text chunks, particularly whether they tried to re-establish a linear sequence between them.3 Within Fishnet, the connections the reader might make between the lexias fall into two main types. The reader might elect to follow consecutive links within the same story strand either by following the numerical sequence of links at the bottom of the page or by clicking the ‘next’




button. In following this pathway, the reader may construct a single, chronologically ordered narrative sequence. This type of connection is high in linearity as it emphasizes a straightforward time sequence. In turn, these may be described as exhibiting a high degree of narrativity. In contrast to this, the reader might follow the lexical hyperlinks found in the main body of Fishnet’s textual chunks. Instead of constructing a chronological sequence, these switch the reader into another story strand, which may appear unrelated other than by some metaphorical association. These links are low in linearity as they do not necessarily encompass temporal order and may disrupt the coherence of a storyworld. Consequently, these appear lower in narrativity. Of course, the two options represent extremes and it is quite possible for a reader to use a combination of these, reading part of one story strand and then moving on to another. The pathways taken by the readers in this study were analyzed, counting the number of linear steps that were taken and then calculating these as a percentage of the total steps taken by each reader. The results of this are summarized in Table 5.1 below. Several conclusions might be drawn from these statistics. First, looking at the overall averages for the male and female readers we might note the similarity in the scores. On the basis of this evidence, it would seem that the biological gender of the reader does not make a difference to the reading pathway that they took. However, aggregate scores can be misleading. In fact, there was extreme variation in the reading pathways of the students. No two students read exactly the same version of Fishnet’s storyweb. There were six male readers who took only linear steps, and eight who took no linear steps at all. Likewise, there were three women who took only linear steps and one who took none at all. The remainder of the readers used a combination of the two patterns, pursuing some consecutive steps through a narrative strand but then switching over into a number of different storyworlds. I hypothesized that the intra-category variation might be due to an influence of the different academic disciplines. The scores were also averaged to compare the patterns of the Computing and the School of English students (Table 5.2). Comparing the averages on this basis suggested a slight difference between the two groups, with the Computing students taking more Table 5.1 Percentage of linear steps taken by readers (by gender) Male readers (overall average)

Female readers (overall average)

35.67%

33.40%




Questioning Intra-category Variation 103 Percentage of linear steps taken by readers (by academic discipline)

Computing students (overall average) School of English students (overall average)

37.8% (Male: 37.8; Female: 50.9) 31.9% (Male: 41.9; Female: 25.7)

linear steps than did those in the School of English. A closer examination suggested that while the male readers from both subject groups used relatively similar proportions of linear steps, the women in School of English were distinctive in that they were the group whose pathways had the highest proportion of associative links, with only one of the women following an entire linear sequence. The question then remains of whether these differences can be related to gender in a meaningful way. Superficially, the high narrativity, linear pathway shares some characteristics with the chronology of Brooks’ ‘male plot of ambition’, discussed at length in Chapter 2. It corresponds to Carolyn Dinshaw’s metaphorical definition of ‘reading like a man’ as providing a ‘single, univalent textual meaning fixed in a hierarchical structure’ (1989: 28–29). In comparison, the lower narrativity of the associative pathway with its fragmentary open-endedness is suggestive of écriture feminine. However, these metaphors ultimately rooted in psychoanalytic paradigms of binary gender difference do not explain why the readers should adopt these patterns, and why they are in fact open to readers of either gender. The motivation for choosing either pathway might be influenced by any number of factors, and the evaluative commentary that the readers kept as they undertook the task is helpful here. With only two exceptions, the readers were negative in their response to Fishnet, frequently deeming it ‘dull’ or ‘confusing’. Regardless of their gender or their academic discipline, many readers commented on the disorientation they experienced when moving from one story strand to another via the hyperlinks, for example: Example 5.1 I find it confusing. It seems to be a jumble of disjointed stories. Using for the first time, I kept believing I had lost my way. (Male reader, English student) Example 5.2 I thought it was confusing. Being used to reading books, I find it hard to click from one word to another. It seems that you never really get into the story. (Female reader, English student)



Table 5.2

These readers echo the disorientation experienced by those in Douglas’ study (1992: 106) when confronted with a new medium that challenged their preconceptions of what a narrative could do. Moreover, in line with Miall’s predictions, the sense of disorientation and the fragmentary nature of the text chunks inhibited their engagement with the text (1999: 166). As one reader put it, Example 5.3 Not as interesting as a whole book. There seems to be less to commit to, like a 600 page book, so texts seem unimportant (Male reader, English student). Thus, the experience of these readers supports Gaggi’s argument that hypertext’s apparent empowerment of the reader may be specious (1998: 105). The freedom to navigate between different chunks of text does not in itself equate to ‘empowerment’ nor lead to the reader becoming more ‘active’, if ‘active’ is taken to include their engagement or desire to pursue reading a given text. It should be noted that the reader’s disengagement and disorientation are not simply the product of the medium in which the narrative is presented. Here the academic discipline seemed to have some influence on the criteria the readers used to evaluate the experience. The Computing students critiqued the hypertext for the creator’s application of technology, for example: Example 5.4 The story can be very confusing as there are way too many links. (Male reader, Computing student) Example 5.5 Confusing, hyperlinks not strictly relevant to the content. (Male reader, Computing student) Example 5.6 Limitations [are] trying to write as many stories as this and putting the links in the right places in order to construct it properly. (Male reader, Computing student) In contrast, the students from the School of English focused their evaluation on the content of the narrative that was used in this study. Several





Example 5.7 The plight of this interactive narrative was not helped by the fact that the stories that make up the site are poor to say the least. (Male reader, English student) Example 5.8 There is little chronology from one microcosm to another, producing an incompetently intertwined series of unrelated ‘narrative-bites’, whereby the reader is left uncommitted to the story (I certainly felt so!). (Male reader, English student) Thus, it is too broad to claim that it is the hypertext alone that leads to the disorientation. In the case of Fishnet, the disorientation is increased by the absence of signposts or a map that would guide the reader through the experience and by the lack of clear coherence between the story strands as connected by the hyperlinks. The readers’ comments also suggested some of their motivation for navigating a particular pathway. Despite their disorientation, like those in Douglas’s study, the readers here sought to find some kind of coherence in their reading experience. Winter describes this as an assumption of relevance, saying, ‘The moment you put together any two sentences for a purpose, your listener or reader looks for a sensible connection between their topics’ (1994: 49). This led both the male and the female Computing students and some of the male English students to take a linear route through the texts. However, the Computing students tended to phrase this as a positive choice initiated from the start of the reading process, whereas the English students expressed this as a less meaningful decision brought into play part way through. Example 5.9 I didn’t click on any of the hypertext links because this distracts me and I lose track, for which reason I did not select any. (Female reader, Computing student) Example 5.10 The overall lack of coherence is disconcerting, alienating the reader from any sense of integration with the text. To compensate for this lack



readers commented that they could see the potential of the form, but were disinterested in this particular text that they deemed to be poorly written and lacking cohesion.


The readers’ expectations of the medium were also given as reasons behind their choice of reading strategy. While in 1992 Douglas wrote that interactive narrative ‘has no established conventions and is set in an altogether alien, electronic reading environment’ (p. 13), it appears that even as little as a decade or so later, hypertext narratives bring with them certain expectations. One of the female students from the School of English explained her associative pathway, saying, Example 5.11 If I just read through the story continuously I would also feel like I was missing out on the point of the hypertext. However, it was not necessarily because the readers were unfamiliar with hypertext as a medium that they chose to adopt a linear pathway. Several male readers from the School of English wrote in detail about their frustration in reading Fishnet because it did not match up with the possibilities of interaction. I quote one of them in full here. Example 5.12 I have never read a hypertext narrative of this type before but I have read hypertext novels that are truly interactive. By this I mean that the storyline branches out into several different routes after each section, the reader being able to dictate the course of events or choices that the characters make by clicking on different hyperlinks. In most cases these are also open ended on certain routes enabling the reader to add their own page to the interactive novel if they so desire. This type of truly interactive narrative demonstrates the wonderful diversity and opportunity that is afforded by this medium. . . . Fishnet, in my opinion, constitutes an abuse of the medium. Thus it is not that these male readers were acting as Douglas describes ‘other directed readers’ and branding hypertext narratives per se as inadequate because they failed to function as print narratives (p. 108). Rather, it was because Fishnet did not exploit the possibilities offered by hypertext that they stopped using medium-specific reading strategies of the lexical hyperlinks to navigate the text and switched instead to the default of using the numerical sequence of links or the ‘next’ button.



of direction, I often felt tempted to merely clip ‘next’ to fully understand the chronology of an interesting aspect of a particular story. (Male reader, English Student)


An alternative way of looking at the students’ responses to the hypertext is to analyze the creative writing that they carried out subsequently. I was interested in the storyworlds that they constructed, particularly with regard to the types of actions that predominated and whether the characterization reflected gendered behaviour in some way. Here I am adopting Herman’s definition of the term ‘storyworld’, as referring to a mental model of what happened, with whom and within what kinds of contexts (2002:13). Herman’s work is a thorough and eclectic study, drawing on a wide range of research traditions. This includes his use of Halliday’s (1994) model of transitivity. This familiar tool from the tradition of systemic functional linguistics provides a framework for charting ‘who does what to whom and in what circumstances’. It is expressed as a system of processes and participants (see Table 5.3). Halliday’s model has been widely applied across a number of linguistic domains including stylistics, critical linguistics and narrative theory. Within the last of these, it has proved a sensitive means of exposing ideological positioning (Toolan, 2001), with potential for distinguishing between narrative subgenres (Herman, 2002) or media (Herman, 2004). Here I am interested in the use of transitivity as an index of narrativity. The different process types might be taken as implying varying degrees of kinetic transfer. At one extreme, Material processes may be highly action-centred, whereas, on the other, processes

Table 5.3 Process types, their meanings and key participants (Halliday 1994: 143) Process type

Category meaning

Participants

Material: Action Event Behavioural

‘doing’ ‘doing’ ‘happening’ ‘behaving’

Actor, Goal

Mental: Perception Affection Cognition Verbal

‘sensing’ ‘seeing’ ‘feeling’ ‘thinking’ ‘saying’

Sayer, Target

Relational: Attribution Identification Existential

‘being’ ‘attributing’ ‘identifying’ ‘existing’

Carrier, Attribute Identified, Identifier Existent

Behaver Senser, Phenomenon



Narrativity and the storyworld

of being (Relational and Existential) may be less so. As already noted, criterial definitions of narrative foreground a temporal sequence of actions (for example, as in the work of Labov and Waletzky, 1967; Labov, 1972. Therefore we might hypothesize that texts with a high proportion of Material processes might, likewise, appear to be higher in narrativity. This analysis thus builds on Herman’s analysis of storyworlds, moving in the direction signalled by Emmott (2003) and examining the possible correlation between storyworld process types, narrativity and gender. The creative writing produced by the students showed a sharp contrast between the students in the School of English and those studying Computing. Of the 28 Computing students, only 14 completed this part of the exercise, and those who did wrote only very short texts, consisting of a few sentences at the most. Because of the small size of that part of the data sample, I will concentrate on the more extensive writing done by those in the School of English in the remainder of this chapter. I analyzed the process types in every clause in each of the texts. As might be expected, each of the students used a range of process types in their storyworld representations. In line with the variation in the topics they wrote about, there was considerable variation in the precise proportions of the different processes by individual students. Despite the intra-category variation, the analysis of the process types also offers the possibility to consider broader trends in the writing of these students. First, looking across the differing process types, on average the relative proportions used indicate a ranking in order of frequency as follows: Material>Mental>Intensive>Verbal (>Existential>Behavioural) This was the same for both the female and the male students. However, a comparison based on gender difference gives the following results (Table 5.4). At first glance, these averages indicate points of similarity and difference between the types of action represented by these students’ storyworlds,

Table 5.4 Proportions of process types in the storyworlds (expressed as a percentage)

Male Female

Material

Behavioural

Mental

Relational

Existential

Verbal

51.9 39.8

0.7 1.3

20.9 29.7

15.5 15.8

2.9 3.1

8.8 10.3




where the processes of ‘being’ (Relational and Existential) and ‘communication’ (Verbal) are very close to one another. In contrast, there is some variation in the relative proportion of Material and Mental processes used, where the male students used more Material processes than the female students whereas the female students used more Mental processes than the male ones. Comparing these differences, we might suggest that the male students’ storyworlds presented a greater proportion of external action whereas the representation in the female students’ storyworlds indicated a greater proportion of internal action (especially as realized by thought, perception and emotional response). Examples 5.13 and 5.14 illustrate this difference. Material processes are underlined and Mental processes italicized. Example 5.13 Nothing. As he opened his eyes he expected to feel pain, taste blood, scream. But he didn’t. As he watched the scene unfold he felt like he was trespassing, watching the final shots of an 8 mm film – after the brutality had finished and all that was left was the horror of destruction. But it was his own destruction. No sound, no colour, just black and white images of ambulances, blank soul barren faces, torn and twisted metal. He couldn’t believe it was true – all those Sunday mornings forced to listen to the drone of the parish priest: heaven and hell; faith, hope and charity . . . were these even right? He couldn’t remember the rest. He supposed that he would soon find out. (Female reader from the School of English) Example 5.14 Carly now lives next to the beach on the Mediterranean with her husband and two children. The snowglobe sits on the mantelpiece of her real fire that they light up every night while drinking warm chocolate drinks. She has got her holidays booked in Egypt for the next ten years and has not seen the snow for the last ten years either. (Male reader from the School of English) In terms of the parameters of narrativity, the male students’ storyworlds seem higher in narrativity than the female students’ storyworlds, at least as signalled through the proportion of Material processes used. We then might go on and consider how this contrast might be explained. Superficially, the gendered equation of high and low narrativity with male and female values seems to echo the stereotypical binaries of male




110 Literary and Linguistic Approaches to Feminist Narratology Table 5.5

Process types in the story strands of Fishnet

Story strand

Process types (as a %)

Recurring Dreams (2 male characters) (Sci-fi thriller)

63.6

3.5

14

12.6

Snakes (1 female, 2 male characters) (Romantic thriller)

56.7

5.2

24.6

5.6

Snowman (3 female, 2 male characters) (Short story)

48.6

0.8

21.5

11.3

0.6

5.6

0

7.9

0.6

17.1

action versus female inaction. Although this crudely simplifies the findings, this pairing of gender and action is similar to the representation of characters in the original storyweb, where the male-centred strand (Recurring Dreams) contains a higher proportion of Material processes than do the strands which contain female characters (summarized in Table 5.5). However, the evidence from the students’ storyworlds does not support such a simple correlation, for the female students used Mental processes of cognition and affection in their representation of both male and female characters, regardless of which story strand they were extending. In the story strands, the choice of process type, and hence degree of narrativity, does not seem to be related to gendered stereotypes in the storyworld representation. In the absence of clear evidence about the students’ previous reading of print narratives, which might also have influenced their responses to the text, I suggest that the narrativity of the storyworlds coincides with the readers’ preferences for navigating through the hypertext. Regardless of their biological gender, the readers who followed more linear pathways seemed to carry this desire for traditional narrative coherence into creating storyworlds with a higher degree of narrativity. The one female reader from the School of English who had taken an exclusively linear



Material Behavioural Mental Relational Existential Verbal

path through Fishnet created a storyworld with proportions of Material and Mental processes that matched those constructed by the men who took similarly linear paths. Thus, while on a superficial level there seemed to be a contrast in the narrative strategies used by the overall groups of women and men, relying on the aggregate scores based on gender difference might obscure the behaviour of individuals within those groups and other points of shared commonality.

Gender and characterization in the storyworlds The readers’ expectations of technology affected not only their reading pathways, but also the ways in which they characterized the participants of the storyworlds. This included the assumption that technology in the form of virtual reality offered unprecedented opportunity for establishing ambiguous identities that exploited the deconstruction of fixed gender boundaries. Earlier studies have considered the ways in which individuals might go beyond ‘sex’ as a ‘static condition of the body’ (Butler, 1993: 2), whether that be in fictional realizations of science fiction (Graham, 2001) or in the use of MOOs or MUDs (Roberts and Parks, 2001). Indeed, virtual reality seems to be an environment par excellence where issues of identity, particularly gendered identity may be scrutinized. In a context where alternative realities are constructed, as Salih and Butler puts it, the reality of gender is also put into crisis: it becomes unclear how to distinguish the real from the unreal. And this is the occasion in which we come to understand that what we take to be ‘real,’ what we invoke as the naturalized knowledge of gender is, in fact, a changeable and revisable reality. (2004: 100) So while by far the majority of students created storyworlds with characters who could be unambiguously identified as male or female, a few went beyond this to challenge a fixed definition of gender. For example, one male student picked up on the science fiction thread in Fishnet where a character was repeatedly reborn, and forced to relive the same experiences (Example 5.15). However, this reader complicated the storyworld by altering the assumed gender of the central character. Example 5.15 When ‘Joseph’ hit puberty things started to go very wrong. You see poor Joseph had been christened Georgina and now had to cope with the





What is interesting about this extract is the tension between the possibility of deconstructing the boundaries of gender identity and the governing norms that enable gender identification (or in Butler’s terms, the practices that materialize gender). As discussed in Chapter 2, these practices of naming, gender-specific codes of dressing and heterosexual paradigms can all be used to interpret the gender identity of a character within the categories of ‘male’ or ‘female’. On the one hand, this characterization opens up the fluidity of gender identity (the protagonist is not confined to one gender), yet the piece is also shot through with conventional gender values. The gender switching here implies a binary model of either masculine or feminine identity rather than exploring transgendered or intergendered possibilities. Moreover, the regulatory power of this categorization is indicated by the ambivalent evaluation that is given to shifting between genders, for example, as something that had ‘started to go very wrong’ and resulted in ‘embarrassing practices’. Even while gender is superficially fluid, this fluidity is prohibited rather than to be desired. Another male student took the deconstruction of identity further still and created a storyworld where the boundaries of the body were challenged in a cyborg-like union between human and machine. Like Haraway’s cyborg imagery, this storyworld seems to be one that is ‘postgender’ (1991: 150) where the permeability of the human body in its assimilation of technology throws into question categories of human identity. I quote from his writing in Example 5.16. Example 5.16 Lobe by lobe, the brains of the experimenters were extracted and transferred to the mainframe of the global network server. From vile to phile. Gender was now genuinely an attitude and penises were ‘add ons’ available at the ‘online’ peripheral outlet. The e-penises would be attached to the printer part of the hard drive and with the advent of materialized drives, the digital devas could happily fuck each others brains out. Although this storyworld is one where gender is apparently adopted as an ‘attitude’, I find it notable that the explicitly gendered imagery is



advances of boys at school. Not to mention all the embarrassing practices a man suffers when taking his wife shopping. It didn’t matter what poor Georgina did, whenever her mother would take her to buy a new bra Georgina’s face would turn a shade of pure crimson.

unequivocally masculine. This may be a result of the word play and metaphor associating technology and male sexuality (for example, ‘hard drive’/penis), but still from a feminist point of view I would ask what option there is for femininity in this destructive dystopia? And this deconstruction (both theoretically and figuratively) need not work in favour of women. As Green writes, Submission to or penetration by the machine’s apparatus might signify the redefinition and expansion of the body by information technologies. When that body is gendered feminine, however, and acts as a woman’s body, such disciplines have inescapable resonance with genres such as pornography. (2001: 168) For both examples, it would seem that the deconstruction of gender boundaries does not necessarily lead to a position where gender ceases to matter. Rather, a feminist might argue that in these cases, the adoption of femininity as something ‘very wrong’ and its exclusion from the text altogether are instances where gendered categories, even in (or especially because of) their absence, might matter very much indeed. The postmodern gender theory of Judith Butler (1990, 1993) and Donna Haraway (1991) seem particularly apt as a means of interpreting the narrative storyworlds where the medium of technology has been thematized, and where the virtual environment opens up the fluidity of identification in increasing measure. Within feminist narratology, this theorizing is important for constantly drawing the analyst back to examine the gendered frames of reference and schemata used in processing storyworlds and recognizing that these should not be treated as naturalized ‘givens’. However, I would also suggest that the alignment of postmodern gender theory with fiction and with virtual reality also serves to indicate its possible limitation. The theoretical discourse, use of metaphor and drawing on fictional sources for examples have led some to demand a closer attention to the ways in which the fluidity implied in a performative understanding of gender is both constrained and enabled within real-world contexts (Brady Aschauer, 1999: 16). Indeed, as Roberts and Parks (2001) note, gender switching in virtual environments may be far less radical and extensive than has been supposed, and need not be carried over into an ‘enduring expression of sexuality, personality or gender politics’ (p. 266). This serves as a cautionary note to feminist narratology when examining the relationship between gender, text and context to be careful in the emphases that might be assigned to textual interpretation and not to extrapolate




from fictional to ‘real’ contexts in a mimetic fashion. Indeed, further questions about performativity may continue to be asked, for example, refining our understanding of agency and constraint in the process of ongoing gender formation. In order to address these questions, even if only in small part, it is necessary to look at more narratives, fictional and non-fictional, and examine the ways in which versions of gendered behaviour are employed both in the representation and in the reception of texts.

Conclusion I close this chapter by restating the main points that have emerged from this study of hypertext reading and response. First, the analysis has provided little evidence to support the imposition of a binary model of gender difference on the reading responses of these students. Instead, the overall trend indicates points of similarity between some of the men and some of the women. Importantly, I have stressed the danger of simplified categorization of ‘men’ and ‘women’ and suggested that differences within those groups also be accounted for. In attempting such an explanation, it is clear that the activity of reading should not be decontextualized. From this perspective, gender is just one of many interconnected variables that may influence the world knowledge and specific practices that a reader brings to the text. The readers in this study were all university students, and the reading took place in relation to that educational situation. Therefore, I have hinted at the role academic disciplines may play in shaping subject knowledge and approaches to reading. There is some evidence from the readers’ comments that this knowledge, particularly their experience and expectations of hypertext as a medium, influenced the way that they read and evaluated this text. It would seem, then, that what is read does indeed bear critically upon how it is read. These constraints include not only the influence of the medium, but also the specific structure of Fishnet itself. As a transitional text, it retains a strong dependence on the high narrativity often found in traditional print narratives. Combined with a lack of signposting and poor design of the hyperlinks, this may have contributed to the readers’ preferences for a linear, high narrativity response to the text. This contextualized approach points to the importance of avoiding an abstract, binary model of gender when examining the readers’ responses and hypertext. The transitional nature of Fishnet blurs the distinction between the gendered pairing of masculine/ print culture and feminine/hypertext structure reviewed at the opening




of this chapter. Some texts, like Fishnet, contain elements of both. In the light of this study, Landow’s gendered claims for the feminist potential of hypertext need to be modified. Moreover, given the increasing use of technology in education at all age levels, it may be that readers’ expectations of the medium will alter in perhaps unpredictable ways. Given the ethical dimensions involved in access to such technology and its application in education and other settings, this is an issue that should continue to be debated both within feminist narratology and beyond. The analysis of gender and reader response here reinforces the argument made by Eckert and McConnell-Ginet: that discussions of gendered behaviour should be supported by localized studies and not abstracted from these contexts. The disadvantage of this is that it is difficult to examine wider trends. Clearly, this study is only able to offer a very partial perspective and there is considerable scope for feminist narratology to examine the questions relating to gender, narrative and reading practices further. In so doing, it might contribute to a bigger picture of the ways in which gender is, and is not, significant in reading practices in a range of contexts. As with the study of narratives by women (and men), the questions that narratology might ask should reflect the change in theoretical perspective from the early influence of second wave feminism. Based on the small study I have carried out, perhaps the questions should no longer be whether or not there is gender difference in reading, but why difference should continue to be important in this area.




Media Narratives of Success and Failure

On the limits of performativity In earlier chapters, I have challenged feminist narratology’s focus on identifying gender difference, arguing for the importance of recognizing similarities between the storytelling of women and men and also highlighting the complications of intra-category variation. This supports some of the conclusions of recent gender theory: principally that gender need not be treated as a stable, fixed ‘given’ which is then used as the basis for an abstract, binary system of difference (Nicholson, 1990; Bergvall, Bing and Freed, 1996; Bronfen and Kavka, 2001). This antifoundationalist position has been expressed most acutely by those feminists drawing on philosophical postmodernism (particularly associated with Butler, 1990, 1993). However, Butler’s theory of performativity, and the theorizing that has stemmed from this has not been without critique. In its emphasis on the diversity of individuals within particular contexts (an important corrective to the universalizing assumptions of ‘difference’), feminist theories should not lose sight of the ways in which stereotypical metaphors of difference and gender-specific power relations might continue to hold influence. As Salih and Butler (2004) argue, it is not that performativity should be understood as a quasi-theatrical process undertaken at the will of the individual. Rather, gender is ‘the effect rather than the cause of a discourse which is always there first’ (p. 91) and one which is not unlimited in its fluidity. Butler herself indicates this, asking how are we to understand the limits of such production, the constraints under which such production occurs? Are these social and political limits on the resignificability of gender and race, or are these limits that are, strictly speaking, outside the social? (1993: 20) 116



6

However, Butler’s suggestion that there are limits ‘outside the social’ is symptomatic of the tendency towards abstraction in this kind of theorizing, and the need to take greater account of how gendered discourses operate in social and temporal contexts. Without this attention to the actual ways in which gender is constituted and experienced, the deconstructive approach to gender threatens to become apolitical, merely the flip side of the universalism it critiques. With this in mind, I locate the analysis here firmly within the social domain, specifically exploring the influence of the media as a form of ‘metadiscursive control’ (Walsh, 2001: 15). My focus is the representation of certain women, shaped by specific ideological and politically shifting contexts. In returning to non-literary texts again, the discussion addresses a significant weakness in feminist narratology. That is, the need to go beyond the narratological description alone of texts by and about women, and to extend the contextualist perspective to examine the possible social function(s) of these texts. As Cameron writes about the matter of gender difference, the identification of patterns is not enough to achieve feminist ends. We need to ask not just how men and women differ, nor even just why they differ (i.e. ‘because of what local conditions’), but also for what larger purpose they differ. (1996: 43) If this is an issue for language and gender studies more generally, then it is all the more so for feminist narratology, with its text immanent approach and early focus on canonical literary texts. While I do not wish to suggest that literary texts are decontextualized or free from ideological influence, they have been seen as limited and to some extent elitist (Mills, 1998). In line with the broadening of feminist cultural studies and narratology more generally, widening the boundaries of feminist narratology to include media texts about women is intended to redress this balance and to question the ways in which existing stereotypes and patterns of representation might serve gender-specific ends.

Critical linguistics and media texts Toolan argues that all narratives are ‘a kind of political action’ (2001: 206), and media texts are no exception to this. Indeed, it is now commonly accepted in studies of the media from a critical linguistic perspective that news reports do not transparently represent ‘facts’ (Fowler, 1991; White, 1997), but, like language more generally, are seen as ‘ideological’



Media Narratives of Success and Failure 117

(Cameron, 1998c: 161) and related to the values, beliefs and practices of their social context in various ways. This relationship with social reality is complex and dialectical. At one level, as a cultural artefact, media discourse is part of social reality itself. However, the relationship is not static, but the discourse operates within particular social contexts, and is said to be both affected by and able to affect the power relations embedded therein (Fairclough, 1995: 12), although precisely how this interchange takes place is not an easy matter to describe. One set of influential power relations are those based on gender difference. Previous studies have suggested that despite its apparent ‘feminization’ (van Zoonen, 1998), media discourse is affected by masculinist concepts and practices. Ross and Sreberny-Mohammadi write, ‘most media professionals and certainly all the owners and controllers of media institutions are men and the way in which politics is reported is significantly determined by a male-oriented agenda’ (1997: 106). The impact of this is felt in a number of ways, including in certain contexts the assumption of an ideal reader as male (Walsh, 2001: 21), and representation of women and men in asymmetrical ways reflected in features such as naming choices (Fowler, 1991; Clark, 1992), uses of evaluation (Caldas-Coulthard, 1996) and in broad terms through the articulation of gender stereotypes (Coltrane and Adams, 1997; Seidman Milburn et al., 2001). While the extent to which such representation influences the self-perception and social behaviour of actual readers is debatable (see also Tuchman, 1996; Muncer et al., 2001), it is clear that the media has a particular potency as a ‘gatekeeping power’ that extends over localized domains to such a degree that Bergvall (1999) proposes that its ability ‘to shape general assumptions – to create ascriptions of “appropriate” gender roles – must be examined by those studying the interrelationship of language and gender’ (p. 285). The analysis that follows combines a narratological approach with perspectives that belong to CDA. CDA understands language (including narrative) to be a form of social practice and, as such, shares feminist narratology’s contextualist stance. Crucially, CDA promotes itself as a three-way research tool that encompasses not only the study of the text itself, but also the processes by which the text is produced and sociohistorical conditions that these processes operate within. In its insistence that the analyst must consider the interconnections between these areas, CDA proves to be a fruitful means of extending feminist narratology and suggesting answers to the question of what effect gender values might have in these news narratives about women. I should stress that in no way do I want to propose a mimetic relationship between text




and context. Rather, I approach this via Fairclough’s concept of textual heterogeneity (1995: 60). This works on the premise that textual analysis ‘can give access to the detailed mechanisms through which social contradictions evolve’ (p. 15) and that media texts can operate as ‘a sensitive barometer of sociocultural change’ (p. 52). Thus the features in a text are understood to signal changes in social context. In particular, he suggests that the linguistic tensions and contradictions that manifest in forms of hybridity and intertextuality ‘can be seen as a materialization of social and cultural contradictions’ (p. 60). This hybridity might involve macro-level factors, such as the co-presence of multiple discourse styles or the blending of genres. Also, it might operate at a micro-level in the juxtaposition of contesting patterns of representation (say, for example in transitivity choices or evaluative statements). Given that women’s roles in the public sphere have changed considerably in recent decades (Davidson and Cooper, 1992; Bradley, 1999; Hakim, 2000) but that the media continues to operate from masculinist assumptions, it might be predicted that the representation of high-profile female figures will act as a site of such contradiction. I do not presuppose that this will mirror a simplistic transition, for example, that representation has become more or less sexist. Instead, I am interested in the way in which media narratives might exhibit linguistic contradictions and might serve multiple and perhaps contesting ends.

Data sample and narrative framework The study is focused on two high-profile figures: Cherie Booth/Blair and Hillary Rodham Clinton. Both of these women are interesting and complex individuals who have been reported with considerable intensity and constructed in relation to number of different roles that originate in the private and public domains. The prominence of these women is undoubtedly due to their status deriving from their marriages (Cherie Booth is married to the British prime minister at the time of writing, Tony Blair, and Hillary Rodham Clinton is former First Lady, married to Bill Clinton), thereby becoming elite figures and so newsworthy (Bell, 1991: 158). Indeed, as might be expected, both women are presented in soft news reports in the traditional roles for such a position.1 For example, they have been seen as an escort for their husbands, as leading figures in fashionable and social events and as a ‘noblesse oblige’ with charitable works (Winfield, 1997). However, both women have also departed from these traditional roles in significant ways and been reported in relation to ‘hard news’ stories connected with their




independent professional status. In the case of Hillary Rodham Clinton, this relates to her political career, exemplified by her leadership of health care reform in the early years of the Clinton administration and her later role as Senator for New York. Cherie Booth is a leading human rights lawyer and QC, which has led to her being reported in relation to public policy and in certain controversial cases where she has opposed her husband’s government. Neither woman is the first prime minister’s or presidential spouse to have a career and identity separate from her husband. Bostdorff catalogues a series of presidential spouses who exercised political influence (1998: 201–203). In the UK, recent examples of prime minister’s spouses include Mary Wilson, a poet; Dennis Thatcher, a businessman; Norma Major, an author. However, the media interest in Cherie Booth/Blair and Hillary Rodham Clinton is considerably more intense than in that of their predecessors. Scharrer and Bissell report that Rodham Clinton was reported approximately twice as frequently as Nancy Reagan and over three time more frequently as Barbara Bush in the first two years of their husbands’ administration (2000: 67). A brief examination of news reports in the UK reveals a similarly stark picture for Cherie Booth/Blair. The Guardian and Times newspapers were searched for headlines concerning the previous two prime minister’s spouses. The results of this are summarized in Table 6.1. The media’s attitude to Cherie Booth/Blair and Hillary Rodham Clinton is also considerably more contradictory and negative than towards their predecessors. Whilst both women have been constructed as iconic role models, they have also attracted considerable censure. Whilst this is undoubtedly for various reasons, studies of Rodham Clinton have related this to her non-traditional political activity (Burrell, 2000; Scharrer and Bissell, 2000). Thus both women can be regarded as controversial feminist figures. Their profiles are made ambiguous, first because their news value is dependent upon their husband’s status rather than on theirs alone, and second because of the ambivalent and contradictory ways in which they have been evaluated. The data sample used to examine the profiles of these women is an electronic text collection of news reports available on Lexis Nexis. The Table 6.1 Reporting of the prime minister’s spouse in the Guardian and Times Prime Minister’s Spouse Time Span Number of times named in a headline Dennis Thatcher Norma Major Cherie Booth/Blair

1982–1990 1990–1997 1997–2004

2 66 386




time span for each sample was determined by the year in which their spouse came to political power and then was extended to 2004. The analysis focuses on reports where the woman was named in the headline. The US reports on Hillary Rodham Clinton run from 1992 to 2003 and are taken primarily from the Washington Post and the Washington Times but are supplemented by a smaller sample taken from the New York Times, Bulletin’s Frontrunner and Chicago Sun-Times. The number of reports for Rodham Clinton total 788. The UK reports on Cherie Booth/ Blair run from 1997 to 2004 and are taken primarily from the Guardian and the Times newspaper. The sample also contains a lesser number of reports from the Sunday newspapers associated with both of these groups (the Observer and the Sunday Times), the Independent and the Daily Mail. The number of reports totals 674. Given that quality news reports project an image of being both ‘serious’ and ‘non-gender marked’ (Caldas-Coulthard, 1995), it might be predicted that this sample of texts has the potential to represent these women in a similarly balanced way. However, perhaps because of their news value as notable paradigms (Fowler, 1991: 15), the media have tended to personalize their representation of these women, treating them as iconic feminist figures who stand as role models for career women in debates about women’s activity in the public sphere.2 This personalization and the ideological consequences that result from it are a key factor in understanding the complex ways in which the news narratives about them operate. As Fowler writes, personalization is important, for ‘the obsession with persons, and the media’s use of them as symbols, avoids serious discussion and explanation of underlying social and economic factors’ (1991: 16). This strategy is particularly evident in the profiles that the media construct that centre on the women’s personalities and achievements, interweaving biographical background with current affairs. The profiles, while giving a superficial appearance of authenticity through the reference to biographical detail, are highly selective and give prominence to certain events through repetition, evaluation and their arrangement around the nucleus of the headline and a value-laden end point. Gergen uses the interpretation of the narrative end point to formulate a three-way typology of narratives of progression, regression and stability (1998). This framework is based on a semantic distinction, where narratives of progression signal some kind of improvement, narratives of progression end with decline and narratives of stability exhibit constancy in value over time. Interpreting a narrative according to this typology is clearly reliant on the audience’s world knowledge and perspective (what might be improvement in one context need not




be interpreted as such in another), and thus is highly culture specific. The distinction between the three types may be redrawn more robustly with reference to Hoey’s model of culturally popular patterns of organization. As outlined in Chapter 3, patterns such as the Problem–Solution or Goal–Achievement paradigms may conclude with either a positive or a negative outcome. Although not exclusively so, there is some correlation between patterns which close with a negative result or evaluation and narratives of regression. In contrast, patterns that close with a positive result or evaluation may be structured as narratives of progression. Thus the lexical cues and explicit evaluation of these predictable patterns can be used to classify Gergen’s narrative types in the following discussion. The media reports of Cherie Booth/Blair and Hillary Rodham Clinton have been patterned around these different possibilities and demonstrate their potential for various social uses.

Narratives of progression In the early years of Tony Blair’s appointment as prime minister, the reporting of Cherie Booth hailed her as a success story. An example which shows this in detail is ‘From daughter of someone to wife of someone’ by Michael Gove which was published in the Sunday Times, 2 November 1997. Like many other reports from the time, Cherie Booth/ Blair was presented in glowing terms.3 The report discusses Cherie Booth’s successful ability to maintain the two ‘personas’ of ‘Booth’, the lawyer, and ‘Blair’, the wife of the prime minister, exemplifying these in a narrative of her recent activities running up to the annual Labour party conference, and ending with a statement of positive evaluation that ‘Cherie Booth is an example to all those who believe that they should at least be allowed to try’ to ‘have it all’. As such, the report forms a narrative of progression, a pattern that emphasizes the increasing potential of Booth/Blair’s achievements and influence. This is represented diagrammatically in Figure 6.1. The narrative episodes contained in the report that form the ‘Means’ of the narrative pattern are not represented in a single temporal sequence. Instead they are used to exemplify Booth/Blair’s personal qualities. Thus the account of her academic achievements demonstrates her intelligence and career potential, and the narrative of her recent events as ‘Booth’ and ‘Blair’ is taken as evidence of her tenacity (‘she’s a great getter-onner’). However, the third narrative-like episode illustrates not Booth/Blair’s professional capacity but rather her relinquishing of a political career in order to support her husband and family. These





Means: 1. Succeeds in academic career 2. Relinquishes political aspirations, supporting husband’s career instead 3. Attends her godchild’s christening 4. Launches breast cancer campaign 5. Chairs Bar conference 6. Attends Labour Party conference

Result/Positive evaluation: She is ‘an example to those who want to “have it all” ’ Figure 6.1 Overarching Goal–Achievement pattern in Booth’s narrative of progression

episodes set up a contrast between her traditional role of supportive wife and the independent career achievements within the male-dominated world of the law. It seems that while the report wants to promote Cherie as a ‘have it all’, it simultaneously cannot reconcile the values of these two versions of femininity. The tension between the traditional and non-traditional is made even more explicit in the contradictory evaluation that is used about her. On the one hand, she is positively evaluated, but the very qualities that are applauded also put her at risk of critique. In her professional capacity, the report praises her as an ‘intelligent and articulate woman’ who ‘lives on eloquence’, but this appears incompatible with the traditional role of ‘mute and adoring wife’. Instead, as wife of the prime minister, her verbal skill might lead to her downfall, with the ‘danger that she will make a gaffe’. The contradiction between these positions is represented as an inescapable predicament, for even if she adopts the silencing equated with the traditional role, such ‘softening produces ridicule’. The evaluation thus constructs the two roles as mutually exclusive, and undercuts her professional success by foregrounding her ultimate dependence on her husband’s position articulated in the headline: from the media’s point of view, she is only the ‘wife of someone’. The dichotomy between an identity derived from public achievements and one that focuses on personal relationships (for example, marriage) also runs throughout the press reports of Rodham Clinton. Scholars have charted the media’s changing representation of Hillary Rodham Clinton and her relative popularity during the Clinton administration.



Goal: To successfully combine personae of ‘Booth’ and ‘Blair’

They outline a meta-narrative where after an initial and brief period of approval, Rodham Clinton became demonized in the media but recovered from this by 1998–1999 to rise in popularity and run successfully for Senate (Burrell, 2000; Parry-Giles, 2000; Anderson, 2002). These changes have been related to the representation of Rodham Clinton in non-traditional political roles (which is negatively evaluated) and more conventional female roles that evoke sympathy and approbation (for example, as the ‘wronged wife’ in relation to the Clinton-Lewinsky scandal or as an ambassador for women’s rights on the international stage). However, this meta-narrative of successful recovery is complicated by a feminist evaluation of Rodham Clinton’s strategies, which is echoed in the academic scholarship and in the media itself. As Parry-Giles puts it, Within the coverage of Rodham Clinton, a troubling depiction resonates: we are to fear women with power, yet admire women with the status of victim. Such images reward women who do not challenge the vows of marriage regardless of the circumstance while sending an ominous message to women who aspire to a position of power. (2000: 221) This ambivalent and contradictory evaluation is found even in reports that suggest her future potential and so, according to Gergen’s terms, would count as a narrative of progression. These media narratives of progression occurred around the time that Rodham Clinton began to launch her campaign to run for Senate in 1999. An example of this is ‘The Hillary Dilemma’, Washington Post, 21 March 1999. This discussion piece documents Rodham Clinton’s ambiguous status as a feminist role model. On the one hand, the article lists her various achievements of which feminists might approve: her beginning as a ‘smart, super-educated young woman’, leading to a legal career begun in the Rose Law firm, her advocacy for children, her highprofile speeches on women’s rights and recent rise in the national poll ratings which leave her ‘hyped as the runaway favorite in a New York Senate race’. On the other, the article also includes events which construct a narrative of Rodham Clinton’s personal choices within her marriage: to give up her career, to relinquish her maiden name and above all to ‘stand by’ her husband a second time in the face of highly public charges of infidelity. These may be summarized as two distinct but interrelated Goal–Achievement patterns, represented in Figure 6.2. The conflict between these two facets of Rodham Clinton’s identity is manifest in the textual hybridity where the commentary slips between the




Personal narrative

Political narrative

Goal: To be with Clinton

Goal: To build political career

Means: Relinquishes job in law firm to move to Arkansas Relinquishes use of maiden name Defends Clinton against charges of infidelity

Means 1: Takes position in Clinton’s administration Result: Health reform fails Means 2: Speaks out for human rights on international stage

Result: Stays married to Clinton who declares ‘I love you’

= Means 3

Positive Evaluation Rodham Clinton’s poll ratings and popularity rise

Figure 6.2 Interrelated Goal–Achievement patterns in Rodham Clinton’s narrative of progression

language of a romance novel and a political resumé. Both in this and earlier reports, the Clintons’ marriage is described in terms of personal emotion. For example, Hillary’s choice to move to Arkansas was because ‘she loved Bill Clinton deeply’. However, this romance of ‘happy endings’ is interwoven with an account of Bill Clinton’s oscillating career, and Rodham’s own political achievements. This hybrid of the romance and the resumé places Rodham Clinton in an apparently contradictory position. In the romance plot she is cast as the heroine, which objectifies her as the goal of Clinton’s desires and conventionally presents the point of closure as a happy marriage. In the political resumé, Rodham Clinton is the ambitious hero(ine), setting out on the quest to achieve political power. As romance heroine, she gains public support, for example, by changing her name and more contentiously by ‘stoically enduring’ charges of her husband’s infidelity. Yet these are the very choices that are problematic from a feminist perspective. As the ambitious heroine forging a political career, the evaluation reverses. While feminists might have approved her position as heading up health care reform, the research indicates that this led to a decline in her popularity with the general public. Thus the apparent ‘success’ narratives of Cherie Booth/Blair and Rodham Clinton are deeply contradictory and characterized by




incompatible statements of dissent and approval. One explanation of these linguistic tensions might lie in Fairclough’s concept of textual heterogeneity (1995: 60). At a simplistic level, the evaluation within these stories might be taken as an indication of the ideological conflict between conservative views of the First Lady/prime minister’s spouse and socio-cultural changes relating to women’s roles in the workplace. Research has suggested that working patterns for women have changed in various ways with subsequently increased participation in the public sphere, optimistically summarized by Leonard (2001) when she states that ‘there has been a steady increase of women among the higher paid sectors of the economy’ (p. 69). These studies go on to indicate that the positioning of women within these domains of employment, be they management (Davidson and Cooper, 1992), politics or the church (Walsh, 2001) more generally, is riddled with a range of gendered inequalities and tensions. This is especially true for women who combine roles in the private domain (for example, as mothers) with paid employment in the public (Garnsey and Rees, 1996; Hakim, 2000; Leonard, 2001). Hence, this is a contradictory and complex situation that is still in transition, and I do not want to imply that social or economic power has unequivocally shifted in favour of women. As an ongoing site of discursive struggle, the social changes surrounding women’s roles might then go towards explaining the linguistic contradictions in the ‘success’ stories of Booth/Blair and Rodham Clinton as high profile, work-centred women. However, the feminist interpretation of Cherie Booth and Rodham Clinton as emergent role models is somewhat constrained. While superficially the narratives of progression might be taken as valourizing the achievements of these women, both of whom have entered allegedly sexist professions, this optimistic interpretation must be understood as limited, for the personalization of their biographical profiles fails to address the discriminatory practices and values encountered by (at least some) women in the workplace. Nor does the personalization recognize that in reality the experience of women within the public sphere is considerably diverse (see for example, Hakim (2000), on the distinction between work-centred and adaptive women). I would argue that the potential for interpreting Booth and Rodham Clinton as feminist figures is further problematized when these ‘success’ stories are placed in the context of a much larger sample of reports. Both the frequency with which the women are reported and the identities that are assigned to them reflect that as public figures their representation is firmly entrenched in the masculinist values of a politically influenced agenda.





The choice of name used to identify the woman may be important here. As Clark writes, ‘Naming is a powerful ideological tool .. . Different names for an object represent different ways of perceiving it.’ (1992: 209). Alongside the powerful connotations that might accompany the lexical terms used in naming, the choice of surname and title may be significant too. Indeed, this has formed an important element in debates about non-sexist uses of language (see Mills (2003) for a full summary), and is seen as critically related to women’s identity (Arichi, 1999: 411) and their right to be known on their own terms, rather than in relation to someone else (Ehrlich and King, 1998: 167). However, as Ehrlich and King go on to argue, non-sexist intentions do not necessarily result in non-sexist language use or interpretation. Instead, the meanings attached to terms, here naming choices, may be governed by macro-level political structures (McConnell-Ginet, 1998: 209) and at a localized level are subject to continual renegotiation and reinterpretation, based on not only an individual’s political affiliations but their position within a range of communities of practice (Mills, 2003: 14). Therefore, when examining these naming choices used by the media, certain limitations must be taken into account. First, the names used to identify either woman may not be a matter of their own choice. Much has been made of the apparent restrictions influencing Hillary Rodham Clinton when she took her husband’s surname in 1982. While Cherie Booth has indicated the significance of keeping her original name, saying that it is ‘Very important, as I became a QC as Cherie Booth, not Cherie Blair’ (Independent, 3 October 2001), as Walsh (2001) notes, the media ‘are under no obligation to accept the self-selected titles of women in the public eye’ (p. 47). Indeed, the Guardian’s guidelines for naming Cherie Booth/Blair reflect this potential discrepancy. On the one hand, the editorial stance is that ‘the Guardian believes that women, like men, should have their own identity . . . where possible, we try to follow what the woman herself wishes to be called’ but simultaneously ‘we do not regard ourselves as wholly bound by what people style themselves’ (David Marsh, assistant editor, personal email communication). Furthermore, readers might interpret the choice of name assigned to the women in a variety of ways, depending on context and their own position. As such, the names used to identify Hillary Rodham Clinton or Cherie Booth/Blair cannot be taken as a simplistic measure that representation has changed in either sexist or non-sexist directions. Rather, the range and frequency of naming options used might be



Representation: Naming over time

taken as indicative of the emphases that the media assign to their professional or marital identities. This is most relevant to Cherie Booth/ Blair. There is a clear distinction between the use of her maiden name, which she continues to use in her independent legal career, and the use of her married surname, which the media use when she is reported as Tony Blair’s wife. The choices for naming Hillary Rodham Clinton do not contrast so sharply or map onto a distinction between professional/ marital identity, first, because she does not use her maiden name alone and, second, because her professional career in politics has been so closely intertwined with that of her husband. The starting point for examining this is to consider the headlines in which Cherie Booth/Blair and Hillary Rodham Clinton are mentioned. Given that the headline occupies a position of textual and evaluative prominence in the news report (Bell, 1991: 152; White, 2000: 384), it might be predicted that the representation of her in these fragments would be of particular importance and indicative of the emphases assigned to the identities associated with the various naming choices. Each of the headlines was analyzed according to the name attributed to woman, the date and topic of the news report and the newspaper in which the report was published. A summary of the names used for each of the women, along with the frequency of each name is given in Tables 6.2 and 6.3. Even at a somewhat superficial first glance, the range and frequency of these titles is telling. For both women, there are a greater number of names used to identify them in relation to their marital status than those that relate to their independent employment. Thus for Cherie Booth/Blair there is a contrast between (Blairs, Cherie Blair, First Lady/wife, prime minister’s partner/wife) and (Booth, Cherie Booth). For Hillary Rodham Clinton, the use of her surname ‘Rodham’ or ‘Rodham Clinton’ or her identification as Senator may be grouped together as distinct from the five variations which use ‘Clinton’ alone, or foreground her marital status as ‘wife’ or ‘First Lady’. This might be understood in terms of lexical variation where the number of marital identifiers is taken as an example of overlexicalization, at least in comparison to the professional identifiers in the headlines from this sample. As Fowler (1996) notes, this may have social significance as ‘a proliferation of terms in some semantic field indicates an unusual preoccupation with a part of the culture’s, or the writer’s experience’ (p. 219). The relative frequency of the various names is in line with this. While Cherie Booth/Blair is most often identified by First Name alone (44.5%), the titles that associate her with Blair represent 42.8% of the remaining headlines while those that identify her as Booth represent only 12.7%. The relative




Naming of Cherie Booth/Blair by newspaper (in %)

Name

Telegraph Guardian Independent Daily mail Mail on Sunday

Observer

Times

Sunday Times

Misc

BBC News Total

Blairs Booth Cherie Cherie Blair Cherie Booth Cheriegate First Lady Wife Mrs Blair Ms Booth PM’s consort PM’s partner PM’s wife PM’s family QC

0.3 0.0 0.3 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0

4.4 1.1 9.0 8.1 1.6 1.0 0.5 0.2 1.1 0.2 0.2 0.0 1.3 0.0 0.0

0.5 0.0 1.1 1.3 1.5 0.2 0.2 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0

0.0 0.0 2.6 0.3 0.0 0.0 0.2 0.2 0.5 0.0 0.0 0.0 0.0 0.0 0.0

0.2 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.3 0.0 4.2 1.5 0.3 0.2 0.0 0.2 0.0 0.0 0.0 0.2 0.5 0.2 0.0

3.7 1.8 15.2 7.4 1.9 0.3 1.5 0.5 0.6 0.3 0.0 0.0 0.2 0.2 0.2

1.6 0.0 6.0 0.8 0.3 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0

0.2 0.0 4.8 0.3 1.1 0.3 1.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0

0.0 0.0 1.0 0.3 0.3 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

11.1 2.9 44.5 20.3 7.1 2.1 3.4 1.0 2.3 0.5 0.2 0.2 4.0 0.3 0.2

Total

1.1

28.5

5.3

3.7

0.5

7.4

33.7

9.5

8.4

1.8

100.0

129



Table 6.2

Naming of Hillary Rodham Clinton by newspaper (in %) Washington Times

Washington Post

New York Times

Bulletin’s Frontrunner

Chicago Sun-Times

USA Today

Clinton Clintons First Lady Hillary Hillary Clinton Hillary Rodham Hillary Rodham Clinton Mrs Clinton Other Clinton Rodham Wife Senator

0.00 4.94 3.02 23.87 3.70 0.14 0.14 1.78 0.14 0.27 0.41 0.14

0.82 9.88 16.87 6.72 16.46 0.00 0.27 0.69 0.00 0.41 0.69 1.10

0.00 0.14 0.82 0.41 1.92 0.00 0.00 0.27 0.14 0.00 0.27 0.00

0.82 0.27 0.27 0.69 0.14 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.14 0.27 0.00 0.41 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.27 0.14 0.00 0.14 0.00 0.00 0.00 0.00 0.00 0.00 0.00

1.65 15.64 21.40 31.69 22.77 0.14 0.41 2.74 0.27 0.69 1.37 1.23

Total

38.55

53.91

3.98

2.19

0.82

0.55

100.00


Total


130

Table 6.3

frequency of the names used for Hillary Rodham Clinton in the headlines presents an even starker picture. The names which include her maiden surname, Rodham, or identify her as ‘Senator’ amount to only 2.06% of the sample. Instead, while First Name alone, ‘Hillary’, is still the most frequently used choice (31.69%), the remainder of the other names (approximately 67%) foreground her married identity. Thus in terms of both range and frequency, the naming choices used in the headlines suggest that for both women, the representation related to their married identity is of greater cultural significance, and it is from this that their news value is primarily derived.4 These admittedly generalized findings would seem to support the masculinist trends for naming women proposed by Fowler. That is, the most frequent form of identification is by First Name alone, and the naming that reflects domestic status is of considerably greater range and quantity than naming as a professional (1991: 101–104). The interpretation of these kinds of statistics must remain provisional and the effect of any given item must be considered within its individual context. A closer examination of the headlines in which the women are named suggests a complex picture. This is one of contrast. In the case of Cherie Booth/Blair, the representation of her as the celebrity wife of the prime minister is juxtaposed with that of her as a high profile QC. An initial step in distinguishing between these two patterns of representation is to compare the correlation of naming choices and subject matter covered in the reports. This is summarised in Table 6.4. The figures given earlier in Table 6.2 suggested that the two most dominant papers in the sample (the Guardian and the Times) made use of both the surnames ‘Blair’ and ‘Booth’, and that on average these alternatives were relatively equally selected. However, the figures in Table 6.4 indicate that there is some difference in the identities linked with each name. As might be expected, the topics associated with the name choice ‘Blair’ (preceded by either ‘Mrs’ or First Name) tend to be those of the conventional prime ministerial consort (for example, to do with her as celebrity or a fashion report). This also includes the reports in which Cherie Booth/Blair was implicated in a scandal. 1. Cherie Blair at awards (Times, 28 May 2001) 2. Cherie Blair at the NEC in Birmingham (Times, 26 March 2001) 3. Women: A snap at £2000: Why did Cherie Blair spend a fortune taking her own hairdresser to New York? (Guardian, 29 September 2001) 4. Cherie Blair takes the blame: No. 10 kept in the dark on links with fraudster (Guardian, 6 December 2002).




Correlation between naming and topic for Cherie Booth/Blair (in %) Name Blairs Booth Cherie Cherie Blair Cherie Booth First Lady Mrs Blair Ms Booth PM’s wife Cheriegate QC

Topic Celebrity Charity Fashion Law Politics Scandal Role model for women

7 0 0.4 0 1.5 3 0

0.3 0.4 0 1.9 0.2 0.2 0.2

15 1.5 3 4.5 5 13 2.5

7 0 1.5 0.5 1.2 9 2

1.5 0.7 0 4 0.2 0 1.2

1.7 0 0.4 0 0.8 0.2 0.5

0.5 0 1.5 0.4 0.2 0.8 0.2

0 0 0 0.4 0 0 0.2


0.9 0.5 0 0.4 1 0.8 0.7

0 0 0 0 0 2.2 0

0 0 0 0 0 0.2 0


132

Table 6.4


5. Cherie Booth engaged to test law on compulsory annuities (Independent, 14 March 2001) 6. Cherie Booth’s firm in crisis talks to ‘rethink brand’ (Independent, 31 March 2001) 7. Children failed by social services, says Booth (Times, 12 November 2002) This pattern is not in itself particularly remarkable. However, it is worth noting that the most frequently used naming choice, First Name only, is associated considerably less with the reporting of legal matters (only 4.5% of the reports) and is more often correlated with the topics conventionally expected of a prime ministerial consort. This adds further support to the claim that the high profile figure of ‘Cherie’ constructed by the media is one founded upon Cherie Blair’s marital status rather than her independent, professional identity. Again, this is perhaps as might be expected, given the masculinist trends that persist in the reporting of politics (Walsh, 2001). As Fairclough writes, not all representations are equal (1995: 47). While the marital status of Cherie Blair forms the basis for the more prominent representation, the emphasis on ‘soft news’ topics in relation to this influential identity might be considered somewhat negative, trivializing women and their achievements (Ross and Sreberny-Mohammadi, 1997: 104). The media’s apparent fascination with Cherie Booth/Blair’s appearance is a good example of this. This not only includes commentary on her fashion sense but also debates her use of alternative holistic therapies (see Examples 8–13). 8. Cherie is all ears to combat stress (Times, 2 October 2001) 9. Cherie’s acupuncture (Independent, 6 October 2001) 10. Pressure point: Cherie Blair has been wearing an acupuncture needle in her ear this week (Guardian, 4 October 2001) 11. The alternative Mrs Blair (Guardian, 8 November 2001) 12. Avuryeda, mystery medicine (Times, 2 November 2001) 13. Smart enough to know better (Observer, 11 November 2001). In these reports, her use of ‘alternative’ treatments is treated with scepticism, for example, reported as a belief in ‘mumbo jumbo’ (‘Smart enough to know better’, Observer, 11 November 2001). In this way, she is



In contrast, the reports where she is named with the surname ‘Booth’ include her charity work, but are primarily focused on legal matters.

not only objectified but also made to appear irrational and credulous, characteristics at odds with the intelligence and clear judgement which she has been attributed with in her professional career (‘Court of Appeal’, Observer, 29 April 2001). Not only is Booth/Blair’s professional identity by far in the minority (in terms of frequency), the potential this might have to represent her as a successful professional is undercut by the more prevalent preoccupation of the media with these more conventional (and negative) feminine stereotypes. The correlation between name, topic and newspaper suggests the influence of a political agenda in determining the news value of Hillary Rodham Clinton. A closer look at the naming choices in Table 6.3 shows a variation according to newspaper, particularly between the Washington Post and the Washington Times. The Washington Times strongly favours the First Name alone (24%), whereas the Washington Post uses this much less and instead refers to her more frequently as ‘First Lady’ (17%) or ‘Hillary Clinton’ (16%). Despite the fact that none of these include her maiden surname, the reports that occur in relation to these naming choices are most frequently hard news reports about her political activity, clearly constructing her as a political figure. In the Washington Times, the reports where she is named as ‘Hillary’ also include nearly as many that discuss her alleged involvement in the scandals that dogged the Clintons throughout their administration. This is summarized in Table 6.5. 14. Former staffer’s suit says Hillary had active ‘Filegate’ involvement (Washington Times, 16 September 1999) 15. Hillary’s name turns up at trial of McDougals (Washington Times, 15 April 1996) 16. Hillary’s voice to be heard at meeting (Washington Times, 17 November 1992) 17. Hillary Clinton urges public to support foreign aid (Washington Post, 17 June 1997) 18. First Lady’s first task breaks new ground (Washington Post, 27 January 1993) 19. First Lady treads carefully through Egypt’s politics (Washington Post, 24 March 1999). The precise interpretation of any given naming choice will be open to debate. The use of First Name alone is a case in point. On the one hand, it might suggest intimacy and hence empathy for the named individual, typically associated with women in the British press (Fowler, 1991: 99).




Correlation between topic and naming choice in the Washington Times and Washington Post (in %) Clinton Clintons First Lady

Washington Post Celebrity Fashion Politics Scandal Women Washington Times Celebrity Fashion Politics Scandal Women

Hillary Hillary Hillary Hillary Clinton Rodham Rodham Clinton

Mrs Clinton Rodham Senator Wife Total

0.00 0.00 1.52 0.00 0.00 1.52

6.09 0.25 5.33 6.60 0.00 18.27

6.60 0.25 13.71 6.85 3.55 30.96

2.54 0.76 4.82 1.52 2.03 11.68

5.33 0.76 14.97 5.84 4.57 31.47

0.00 0.00 0.00 0.00 0.25 0.25

0.25 0.00 0.00 0.25 0.00 0.51

0.25 0.00 0.51 0.25 0.00 1.02

0.00 0.00 0.00 0.00 0.51 0.51

0.51 0.00 1.27 0.00 0.00 1.78

0.00 0.00 1.27 0.51 0.25 2.03

21.57 2.03 43.40 21.83 11.17 100.00

0.00 0.00 0.00 0.00 0.00 0.00

2.86 0.00 3.57 6.43 0.00 12.86

0.71 0.36 2.86 1.43 2.50 7.86

6.43 1.43 28.93 22.14 3.57 62.50

1.79 0.00 4.64 1.43 1.43 9.29

0.00 0.00 0.36 0.00 0.00 0.36

0.36 0.00 0.00 0.00 0.36 0.71

0.00 0.00 1.79 2.14 0.71 4.64

0.00 0.00 0.00 0.00 0.71 0.71

0.00 0.00 0.00 0.71 0.00 0.71

0.00 0.00 0.36 0.00 0.00 0.36

12.14 1.79 42.50 34.29 9.29 100.00

135



Table 6.5

Here the use of First Name puts this assumed proximity to political uses. There is a contrast between the more personal naming of ‘Hillary’ as a scandalized political figure in the Washington Times (Examples 15–17) and the rather more distant and less scandalized identity given in the Washington Post (examples 18–20). The use of ‘Hillary Clinton’ and ‘First Lady’ perhaps infers a more objective distance in the Post’s representation and a more neutral focus on her professional status rather than the negative accounts of her apparent wrongdoing in the Whitewater scandal. This might be explained by the ideological positioning of these two papers. While both might be described as ‘right wing’, the neoconservative activities of the Washington Times and its owner’s overt support of George Bush have been described as particularly anti-Clinton (Laurent, 2004). This might then explain the Washington Times’ more prevalent use of negative reporting focused on scandal. The Washington Times’ representation of Rodham Clinton over time shows how the intensity of the media interest is determined by a political agenda. As the figures in Table 6.6 show, the scandal-centred reports about ‘Hillary’ are most frequent in 1992, 1996 and 1999, the years in which a political election took place. Given that Hillary Clinton’s public roles have included campaigning on behalf of her husband as well as more recently being elected for Senate, it is not surprising to find that the electoral cycle increases her news value in terms of relevance (Fowler, 1991). However, it does lead me to question how the personalized narratives of regression, stability or progression might function within this political context. One answer may lie in the way they contribute not to only her own reputation, but also, significantly, to that of her husband’s. This is true for both Booth and Rodham Clinton. Thompson (2000) describes the potential influence of reputation in terms of Bourdieu’s concept of symbolic capital, which an individual may use in the exercise of symbolic power. He writes, Anyone who wishes to acquire political power or to exercise it in a durable and effective fashion must also use symbolic power to cultivate and sustain the belief in legitimacy. (p. 98) The representation of Cherie Booth/Blair and Rodham Clinton is particularly important as a resource for their respective husbands’ symbolic power, given that their image is embedded within a gendered framework described by Walsh (2001) as ‘the male MP and his wife [. . .] who together represent a promotional package’, which, she argues, is




Naming of Hillary Rodham Clinton in Washington Times over time (in %)

Clinton Clintons First Lady Hillary Hillary Clinton Hillary Rodham Hillary Rodham Clinton Mrs Clinton Rodham Senator Wife Total

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Total

0.0 1.4 0.4 7.1 3.9 0.0 0.0 1.1 0.0 0.0 0.4

0.0 1.1 1.8 5.7 1.4 0.0 0.7 0.0 0.7 0.0 0.0

0.0 2.5 1.4 3.5 1.1 0.0 0.0 0.4 0.0 0.0 0.0

0.0 1.4 1.1 3.9 0.0 0.0 0.0 0.4 0.0 0.0 0.0

0.0 2.1 1.1 10.6 1.1 0.4 0.0 1.8 0.0 0.0 0.0

0.0 0.7 0.7 3.9 0.7 0.0 0.0 0.4 0.0 0.0 0.0

0.0 1.1 1.1 5.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 0.4 0.4 12.4 1.1 0.0 0.0 0.7 0.0 0.4 0.0

0.0 1.4 0.0 6.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0

0.0 0.0 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 0.4 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 0.4 0.0 1.8 0.0 0.0 0.0 0.0 0.0 0.7 0.0

0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 12.8 7.8 62.1 9.6 0.4 0.7 4.6 0.7 1.1 0.4

14.2

11.3

8.9

6.7

17.0

6.4

7.4

15.2

7.8

0.7

1.1

2.8

0.4

100.0

137



Table 6.6

‘easy to sell’ (p. 77). The construction of this ‘package’ also intersects with what Thompson describes as the evolution of a ‘politics of trust’ where it is the personality (including the private life) of the politician that comes under scrutiny, rather than the nature of their party’s political policies. At a general level, this has resulted in a UK media representation of Tony Blair that has been described as a strategic performance incorporating a sense of ‘authenticity’ through the construction of Blair as a ‘normal person’ (Fairclough, 2000: 117–118). Thus the specific nature of the narratives and the contradictions contained therein cannot be understood in relation to the women’s identity alone. Instead, these must also be seen in relation to changing attitudes towards their husband’s popularity as a political leader which become of heightened significance during periods of electoral campaigning.

Narratives of stability and regression Gergen suggests that one of the social uses of narratives of progression can be to ‘invest [a] relationship with increased value and promise for the future’ (1998: 8, 16). It is not surprising, then, that narratives of progression should be found during the ‘honeymoon period’ immediately following Tony Blair’s election in 1997, when Cherie Booth’s own success was used to consolidate that of her husband. Likewise, the ‘success story’ of Rodham Clinton’s political achievements, especially within the domain of women’s rights can be seen as building her own symbolic capital in the months preceding her campaign for Senate. However, when viewed diachronically, the accounts of Rodham Clinton’s ongoing commitment to her husband, reiterated in reports across time, repeated within those reports and through the double sex scandals in which Bill Clinton was implicated (involving Gennifer Flowers in 1992 and Monika Lewinsky in 1998) also construct a meta-narrative of stability for Rodham Clinton. There are many possible social uses of narratives of stability. There are two that are particularly relevant here. The first is that it may serve to construct an image of Rodham Clinton’s authenticity that she ‘is what she seems’ and so is trustworthy. This is critical, given that her repeated defence of Bill Clinton was an important means of protecting his reputation and so symbolic capital. If Bill Clinton’s alleged indiscretions which led to the second-order offence of lying served to attack the image of him as ‘trustworthy’, Rodham Clinton’s support may be taken of evidence to the contrary: that if she trusted him, so could the public. The second function is that the plot of stability exemplifies Rodham Clinton’s tenacity and her ability to endure repeated trials. As I argued in




Chapter 3 in relation to the medieval romance, Emare, ‘stabilitas’ was an important feminine virtue. It would seem that constancy remains a significant virtue for Rodham Clinton at the end of the twentieth century, for it is this that led, at least in part, towards generating positive public opinion in her favour (Burrell, 2000: 533). From a feminist point of view, this use of the stability narrative might be regarded as troublesome. It would seem to limit Rodham Clinton’s agency, for it constructs her virtue as staying the same over time, rather than being able to change herself or challenge aspects of the situation that might interpreted as wrongdoing against herself. More troubling still is the construction of her husband’s indiscretions as a means of proving Rodham Clinton’s worth, which hints at an ongoing double standard towards women’s and men’s sexual behaviour. This narrative of stability thus stands at odds with the narrative of progression. While both construct apparently positive values, the narrative of progression emphasizes Rodham Clinton’s potential to improve both her own value and those of women and children around her (and so activate change) whereas the narrative of stability instead plays on her commitment to remaining the same. If the narratives of progression and stability have social uses of building symbolic capital, narratives of regression are a crucial way in which it might be undercut. These narratives are characterized by a move into problematic events and close with negative evaluation, providing a macro-level contrast with the narratives of progression and stability. For both Booth and Rodham Clinton, these narratives of regression centre on their alleged involvement with scandalous events. The media’s increasingly prominent reporting of political scandal is well documented. As Thompson’s (2000) illuminating study outlines, this is embedded in social changes, particularly the development of a ‘politics of trust’ discussed earlier and the increased mediatization of politics. This is part of the more general marketization of the media where scandal’s apparent ability to ‘sell’ newspapers heightens the commodification of these public figures. Various studies have indicated that this is a key trend in US political coverage (Astroff and Nyberg, 1992: 11), and the reporting of Cherie Booth would suggest that these features are salient to the UK media too. Indeed, the media’s attention on Booth’s role in what came to be known as ‘Cheriegate’ is so intense as to provoke interest in itself, accounting for a startling 29% of the reports from the seven-year period considered here, despite the fact that the nucleus of the scandal lasted for only a month. Like the narratives of progression and stability, these narratives are highly personalized. In the case of the US reports, the language may




construct Hillary herself as the central problem for her husband. For example, in ‘Hillary Clinton, trying to have it all; Lawyer, author, activist, strategist, mother, political wife . . . and now, political problem?’ from Washington Post, 10 March 1992, the chronological end point is that ‘She [Hillary] has gone from being Bill Clinton’s heat shield to potentially singeing him’ (emphasis added). Similarly, ‘First Lady’s woes mount on Whitewater, health care’, published in the New York Times, 6 March 1994, ends with the negative evaluation: She [Rodham Clinton] is finally where she wanted to be all her life. And she’s not enjoying it. Her father died. Vince Foster died. And she feels her credibility has been badly hurt by the focus on her role in Whitewater. Again, the negative evaluation is personal, ‘her credibility has been hurt’, set in the context of personal bereavement, which in the reference to her father has little to do with current political affairs. In so doing, the reports not only fail to address the social and economic issues of government policy but also make Rodham Clinton’s reputation the basis for not only her own but also her husband’s symbolic power, a basis which can all too easily be destroyed through mediatized scandal. Furthermore, through the use of linguistic forms such as nominalization, the reports obscure the role of the media or para-media organizations in contributing to this situation, for it is unclear who is responsible for generating ‘the focus on Whitewater’ that is so damaging to Hillary Rodham Clinton. Following the scandal nicknamed ‘Cheriegate’, which centred on Cherie Booth’s alleged financial involvement with fraudster Peter Foster at the close of 2002, many reports took a more negative stance towards her. The report referred to in detail here is typical of this, ‘How did Cherie manage to blow it so badly?’ by Gillian Bowditch which was published in The Scotsman, 16 September 2003.5 No longer are Booth’s academic credentials and career prowess included, but instead the report constructs a narrative of her faux pas and apparent inadequacies, particularly focusing on the potentially negative consequences of her failed relationship with former ‘lifestyle adviser’, Carole Caplin. The report is structured around a series of interrelated Problem–Solution patterns summarized in Figure 6.3. Again, the problem, according to the report, is highly personalized and described in terms of a character flaw, ‘Mrs Blair does seem to lack the essential qualification – judgement’ and to her personal relationship




Pattern A Situation: Horoscope warns Blairs Media and ministers give warning About Caplin Problem: Caplin is destined to bring Cherie ‘nothing but grief ’ Response: Tony Blair ousts Caplin ‘in most cack-handed way’ Result: = Problem 1 for Blairs Caplin plans book on Blairs Book could inflict damage ‘to set the record straight’ by exposing political division and details of Cheriegate Result Embarrassment for Blairs Negative evalution A strong and popular PM should survive, but …

= Problem 2 for Blairs Blair is no longer a strong and popular PM

Negative result Cabinet ministers do not ‘rally round’

Predicted result 1 Minister will not support Blair

Predicted result 2 Court injunction

Negative Evaluation The damage has been done Pattern B Situation: Mrs Blair had goodwill and sympathy Problem She blew it through lack of judgement, e.g. her relationship with Caplin Negative result She lost goodwill

Figure 6.3

Response Blairs must read writing on the wall

Implied negative evaluation End of Tony Blair ’s political power

Problem–Solution patterns in Booth’s narrative of regression

with Caplin, who is destined to bring her an emotion, ‘nothing but grief’. However, the result of these personal problems is a political situation, implying the prime minister’s loss of power, ‘the damage has already been done . . . The Blairs should stop reading the stars and concentrate instead on the writing on the wall’. As with the representation of





Conclusion The point we have reached is that the narratological analysis has highlighted points of contradiction in the news narratives about both women. First, this includes contradictions within the reports themselves, indicated through conflicting evaluations of approbation and censure and through the hybridity of discourse. Second, there are also contrasts between the reports, particularly those that are structured as narratives of regression and progression. I have suggested that these contradictions can be explained in various ways. On one level, the tensions within the narratives can be seen as examples of textual heterogeneity and thus as evidence of the ongoing conflict between social changes relating to women’s roles in the public sphere set against more traditional expectations of a conventional, supportive political spouse. However, this does not account for the contrasts between the narratives, for it is not that women’s participation in the public sphere has changed that radically within the period in question. Instead, these contrasts need to be understood in relation to the social function that these narratives fulfil in relation to the symbolic power of Tony Blair and Bill Clinton, where the women’s narratives of success and stability build their husband’s reputation and the narratives of scandalous regression contaminate it. The overwhelming evidence that the women’s news value is derived from their marital identity and determined by the political status of their husband (for example, shown in the naming choices and the frequency of reports over time) throws into doubt my suggestion that the textual heterogeneity signals social change for women. Instead, the values derived from this masculinist agenda suggest that perhaps not so much has changed at all. The analysis and discussion of these media texts has some important implications for feminist narratology more generally. First, I have argued that to look at single texts in isolation can be misleading. For example, if only the ‘success’ narratives from the beginning of Blair’s government had been analyzed, then the picture of Cherie Booth as a woman who portrayed career potential would have appeared much stronger. Without the wider analysis of other narratives and the much



Rodham Clinton, the personalization clearly functions within an ideological context, depoliticizing the situation and avoiding serious discussion of matters of public policy and international world events (in the UK, this included the Hutton Inquiry and the war in Iraq) which might have given rise to Blair’s loss of public popularity.

broader analysis of the headlines, the extent to which this feminist interpretation might be curtailed could not be seen. Literary narratives might also benefit from these wider comparisons, perhaps testing the ways in which patterns of representation are sustained or vary within the work of a particular author, within genres or historical periods. Second, the processes of production and dissemination are also important. Here I have alluded to the importance of marketization as influencing the personalized representation of the women, especially in the increase of interest surrounding scandal. In the US, these processes of production are strongly politicized, and it has been suggested that the news reporting, especially smear campaigns, may be used to serve the interests of particular groups (Brock, 1993). The origins of feminist narratology also reflect concern in the production of literary texts, particularly in rediscovering texts by women. However, further questions might be asked about the influence and prominence of various publishing houses in categorizing texts as ‘feminist’, and the dissemination of texts through curricula in the education system at various levels. Finally, it is clear that the identities of Cherie Booth and Hillary Rodham Clinton constructed in the media reflect the ongoing constraints of gender ideology. While their identities may be shifting and reflect competing versions of femininity, such as the conventionally supportive political wife, the successful lawyer or politician and a feminist spokesperson, they do indeed operate within social and political limitations. The extent to which these representations influence the behaviour of actual readers is beyond the scope of this study, but the US research has hinted at the ways in which the more or less conventional representation of Rodham Clinton affected public opinion (Kaplan, 2001; Anderson, 2002). In its adoption of poststructuralist gender theory, feminist narratology should not neglect the ways in which gender stereotypes continue to be of relevance and operate to support agendas of various kinds. While fictional worlds may enable gender categories to be deconstructed more readily, it is clear that these categories cannot be abandoned and remain of ‘real world’ significance.




Gender, Age and Narrative Development

Age as a sociolinguistic variable This chapter turns to another situation in which the role of gender as a variable in storytelling style continues to be debated and may be of significant social and political consequence. This is the storytelling of children, told in an educational context. Children’s storytelling is an important area, for, as of yet, feminist narratology has primarily analyzed stories by adult tellers or for an adult audience. However, research indicates that the stories told by children may vary considerably compared with those by adult tellers (Eaton Collis and Lewis, 1999; Lambrou, 2005). Therefore children’s stories provide an important area of contrast against which the assumptions and models of narratology may be explored from a feminist perspective. This analysis once again takes feminist narratology beyond the description of narrative form alone into questioning the reasons for and consequences of variation in narrative, for children’s storytelling is intertwined with matters of literacy, which in turn is crucially related to power. Various studies have shown that within educational settings, certain forms of discourse become privileged while others may be undervalued or even excluded (Scollon and Scollon, 1984; Michaels, 1991; McDonald and McNaughton, 1999). As narrative is an important resource for expressing cultural values and a means by which an individual may perform their identity (which incorporates aspects of gender, ethnicity, class and so on), this privileging or exclusion is by no means neutral and may be ideologically significant. The outcome of such distinctions may have long-term consequences not only for an individual’s sense of identity but also for their academic achievement. Whether this involves isolated individuals or wider groups of people distinguished on the 144



7

basis of their gender, race or cultural background, such variation deserves closer attention. During the last two decades, there has been a wealth of research that has examined the various ways in which children acquire narrative skills, indicating a wide range of influential factors. This includes sociolinguistic parameters of gender (Barwood, 2001; Prochnow et al., 2001; Dutro, 2003) and cultural background (Michaels, 1981, 1991; Wang and Leichtman, 2000; Harkins and Ray, 2004). Narrative performance may also vary according to type of text (Rosen, 2001), the topic of the narrative (McCabe and Peterson, 1990; Tenenbaum et al., 2005) and the situation in which the narrative gets told (Aldridge and Wood, 1999; Eaton, Collis and Lewis, 1999). These studies readily demonstrate that narrative is not universal in its form, nor in the ways that it might be acquired. The place of gender in this picture is no less complex. Some of the now classic accounts of children’s narrative development do not differentiate the children’s narrative abilities according to gender at all (for example, Kernan, 1977; Peterson and McCabe, 1991). Where gender is included as a variable, the literature indicates some debate as to whether or not this makes a significant difference, as summarized in Rosen (2001: 7–8). Some studies indicate more points of similarity than difference (Worthy, Moorman and Turner, 1999; Prochnow et al., 2001), while others point to important contrasts (Tenenbaum et al., 2005). As Rosen’s (2001) wide-ranging and thorough survey of literacy achievement indicates, the difference in narrative abilities according to gender varies from country to country. So while the role of gender remains of critical importance in studies of narrative and literacy (Dutro, 2003: 467), it is clear that it cannot be treated in an essentialist way and mapped onto linguistic behaviour in a one-to-one fashion. If feminist narratology is to explore the narratives told by young women and girls, it must do so in a way that remains sensitive to this multicultural variation.

Literacy in New Zealand schools New Zealand is one country in which gender does seem to make a significant difference to the literacy achievements of its school students. Wilkinson (1998) reports on the International Association of Educational Achievement (IEA) survey of 1990–1991. He notes that among all the countries who participated, New Zealand ‘showed the second largest difference in achievement between girls and boys in primary school’ for their reading literacy (p. 145). Alongside this, New Zealand



Gender, Age and Narrative Development 145

also showed ‘the largest difference in achievement between those students learning in their home language and those who were not’ (p. 145). In summary, girls outperformed boys, and ‘Pakeha’ children (the Maori term for New Zealanders of European descent) achieved higher scores than non-Pakeha children. This difference became even more pronounced in ‘low decile’ schools, with composite classes.1 Recognizing the greater migration in recent years (particularly from the Pacific Islands and Asia) and the substantial natural increase amongst the Maori, considerable concern has been raised about how to meet the educational needs of an increasingly diverse student population. However, studies of the students’ literacy skills have dealt with gender and with ethnicity as separate issues (see Prochnow et al., 2001; Barwood, 2001 on gender; and McDonald and McNaughton, 1999; Phillips et al., 2004 on ethnicity). Given the clustering of gender, race and socioeconomic status, it seems possible that there may be some interrelationship between these influences. Narrative skills are an important focus for this pedagogic concern, as it is in relation to these macro-level text structures that the gaps in achievement were most acutely felt (Wilkinson, 1998: 147; McNaughton et al., 2003). The analysis that I present here is a small-scale study that compares the storytelling abilities of a multicultural group of children in a New Zealand school with that of a mono-cultural group of children in a UK school. Many of the pedagogic and sociolinguistic studies cited earlier are impressive longitudinal and large-scale surveys. My analysis here does not claim to be such an extensive empirical overview but rather provides an initial context from which a more detailed qualitative analysis might be understood. Clearly, these findings alone cannot be taken as anything other than indicative, and should not be interpreted as representing some kind of universalized narrative development. On the contrary, the multiple intersections between gender, age and ethnicity point towards the existence of differing expressions of masculinity and femininity as performed through these narratives, and which may vary from child to child. However, the analysis that I present signals some important contrasts that need to be understood in relation to broader socio-cultural factors, particularly as these are embedded within patterns of power.

Data sample and methodology I took a data sample of stories written by 66 children. These fell into two age groups: 33 children were aged 7–8, the other 33 were aged 10–11.




Of the whole sample, approximately half were from a UK primary school (34 children). All of the children in this group would be categorized as white middle class. The remaining 32 children attended a low decile school in Auckland, New Zealand. As is typical of such schools, the group of New Zealand-based children was culturally diverse, including Pakeha, Maori, Pacific Island and Asian ethnicities (summarized in Table 7.1). I took samples of two different types of narrative writing that was carried out by the children as part of their regular curriculum. The first type was a retelling of an existing story. This was a common activity in both UK and New Zealand schools, where the class teacher read a story aloud and the children then retold this in their own words. These retellings were thus derived from narrative patterns which had been ratified by the educational setting as appropriate ways of telling a story. The second type of narrative writing was a personal account of the child’s own experience, for example as an account of recent events or, for the older UK children, an autobiographical episode. These two types of storytelling illustrate the multiple narrative styles that children may acquire during their schooling. Significantly, they also demonstrate the potentially different ways in which the students might negotiate official discourse or express their individual voice through the use of alternative narrative patterns. The model used to analyze these narratives is that of Labov (1972, 1997), which will be familiar from the discussion in Chapter 4. It has proved an effective means of charting children’s emerging narrative competence (Kernan, 1977; Eaton, Collis and Lewis, 1999). In this chapter, I build on this earlier work, with a particular interest in the children’s manipulation of the prototypical global structure outlined by Labov (AbstractÔrientation^Complicating ActionÊvaluation^Result^ Coda). This can be used to examine the extent to which children of different ages, gender and cultural background produce ‘complete’ narrative sequences, that is narratives that conclude with a recognizable Result or Coda. As I have alluded in Chapters 2 and 3, early feminist Table 7.1

Ethnic origin of New Zealand children Asian Maori Pacific Island Pakeha

7-year-olds 10-year-olds Total

4 7

3 4

1 3

3 7

11

7

4

10




narratology, like feminist literary criticism from the period, weighted the issue of closure with ideological significance. The so-called complete narratives were assumed to be normative and the male prerogative. In contrast, attempts to ‘break’ or go beyond ‘the sequence’ were considered to be a form of feminist emancipation (DuPlessis, 1985). The analysis here revisits and challenges this gendered equation once again, considering the various reasons that might explain a range of narrative styles.

Storytelling characteristics at seven years of age The stories told by the younger children (7–8-year-olds) considered in this study support the findings of earlier research that characterize children’s storytelling as brief and incomplete (Aldridge and Wood, 1999). For the most part, the narratives did not contain all six of the Labovian stages. The varying constructions used by the children are summarized in Tables 7.2 and 7.3. Significantly, the incomplete nature of these narratives did not correlate with a pattern of gender difference. That is, they were not the preserve of the female children, but were told by both boys and girls. This was true of the small sample of highly fragmentary texts that consisted of only one narrative component, either Complicating Action Table 7.2 Narrative components used by 7-year-olds: Retellings 7-year-olds

Orientation only

CA only Or, CA (eval)

Or, CA, All Eval, Result 6 components

UK Boys NZ Boys UK Girls NZ Girls

1 1 0 0

0 0 0 1

10 1 7 5

1 3 3 0

0 0 1 0

Total

2

1

23

7

1

Table 7.3 Narrative components used by 7-year-olds: Personal experiences 7-year-olds

Orientation only

CA only Or, CA (eval)

Or, CA, All Eval, Result 6 components

UK Boys NZ Boys UK Girls NZ Girls

1 1 0 0

1 1 1 0

10 3 10 4

0 0 0 2

0 0 0 0

Total

2

3

27

2

0




alone or Orientation alone. Indeed, the presence of texts like these might be explained by the age of the children, as indicated by earlier research. Peterson and McCabe (1983) suggest a profile of children’s emerging narrative development where the earliest stages are characterized by narratives of Complicating Action only. Likewise, Lambrou (2005) comments that it was the youngest age group from her data sample that produced what she called ‘almost narratives’ which failed to develop into recognizable story patterns. The small number of these narrative-like fragments cannot be used to make any generalized statements about the specifics of this stage in narrative development, but signal the ongoing and potentially uneven ways in which children’s storytelling might progress for both boys and girls. What is perhaps most striking about the results in Tables 7.2 and 7.3 is the overwhelming tendency for the children to follow a pattern of Orientation^Complicating Action, with evaluative comments where present dispersed throughout either or both of these. At first glance, it would seem that the majority of children have grasped the ability to formulate the beginning and middle of a story, but do not necessarily follow this through to a clearly marked stage of resolution or closure. This was true for both the retellings and the children’s personal accounts. In both types of writing, the analysis here does not support a hypothesis of gender difference, for both the girls and the boys used this same structural pattern. Instead, the apparently ‘incomplete’ nature of these texts might be motivated by other, non-gender-related reasons. The omission of a ‘Result’ section is most prominent in the personal writing, where only 2 of the 34 children included this (that is, approximately 6% of the 7-year-olds). One explanation for this might be that the children simply have not reached the age when they become skilled at incorporating these narrative elements of closure. However, Peterson and McCabe (1983) found that from the age of 6, children could produce narrative sequences that continued to resolution and conclusion. Thus an age-based explanation is not entirely satisfactory, leading me to question what other reasons might lie behind the prevalence of this form. Another possible answer might be found in the relationship between topic and the story genre being used. As I discussed in relation to the childbirth stories in Chapter 4, the six-part prototype outlined by Labov was derived from a corpus of stories about troublesome events, inherent in the concept of ‘danger of death’. Stories that do not necessarily centre on such problematic scenarios may be structured in rather different ways. Again, the range of story genres proposed by Martin and Plum (1997) is helpful here. Eggins and Slade (1997) elaborate on




Martin and Plum’s work, and suggest semantic, formal and functional distinctions between these different story genres. In their terms, an Anecdote focuses on a crisis of some kind and is told to entertain or amuse whereas Recounts simply retell chronological events and ‘do not necessarily deal with a problem’ (p. 237). The structural differences between Narratives, Anecdotes and Recounts are summarized in Table 7.4. I suggest that rather than producing incomplete Narrative sequences, the personal experiences told by the children might be better understood as examples of Anecdotes and Recounts. In some cases, the reports that the children gave of their experiences did not centre on events that were out of the ordinary. This included accounts of everyday activities as the example below illustrates. Example 7.1 1. Yesterday my brother came to my aunties home. 2. When he came I was at school. 3. When I came home he was watching TV 4. then I took off my clothes 5. and I put on my night soot 6. then I watch tv 7. when I finish it was night 8. then we eat dinner 9. then we sleep and sleep (New Zealand Asian girl, age 7)2

Orientation

CA

In this narrative, there is no external evaluation to indicate the tellability of any of the events included. Instead, the constant use of ‘then’ as a connective in lines 4, 6 and 8 indicates the predominance of chronology as

Table 7.4 Structural features of Narratives, Anecdotes and Recounts Beginning

Middle

End

Narrative

(Abstract)Ôrientation

(Coda)

Anecdote


Recount


^ Complication Êvaluation^Resolution ^ ^ Remarkable EventÊvaluation ^ ^ Record of Events^Reorientation^

(Coda) (Coda)

Source: Adapted from Eggins and Slade, 1997: 236




an organizing principle, as might be expected for a Recount. The mundane nature of the events being recorded here is clearly repetitive rather than remarkable, as the switch from the simple past into the present tense from line 6, suggesting an iterative rather than singular frequency of these routines. This is echoed stylistically in the child’s use of repetition not only of the connective ‘then’, but also of ‘sleep’ (line 9). While it might be argued that this repetition is indicative of the child’s restricted vocabulary, it is also highly effective in creating a sense of the endlessly mundane nature of the events, culminating, perhaps most appropriately, with the repetition of sleeping inactivity in line 9. It is not, then, that Recounts are inadequate Narratives. Instead, even when ‘simply’ retelling events, they may be skilfully constructed so as to emphasize the characteristics of a particular worldview. By way of contrast, other personal accounts in this data sample did include events which the tellers evaluated in some way. These might be better classed as Anecdotes. In some cases, the remarkable event is troublesome, but this is not always the case. The event evaluated might be of a positive nature, such as a trip to the beach (Example 7.2) or seeing a butterfly hatch (Example 7.3). Example 7.2 1. In the holidays I went to see my uncle 2. and we had lots of exciting things to eat at the beach. 3. When we got them we made a castle. 4. Then we went home to cook them. 5. When we ate it my mum said it was beautiful. (New Zealand Maori boy, age 7) Example 7.3 1. Last week we saw a catterpillar turn into a chrysalis 2. a couple of weeks after the chrysalis hatched. 3. At first the butterflys body was fat 4. but the metamorphosis made it skinny. 5. About Twenty minutes later it was pumping its wings 6. then it flew away. 7. When it flew away I felt lucky to See It happen. (New Zealand Pakeha girl, age 7)

Orientation CA Evaluation

Orientation CA

Evaluation

In both examples, the stories are structured as typical Anecdotes which begin with Orientation, are followed by a series of narrative




events and then close with Evaluation. Here the evaluative comments are clearly marked by the initial subordinate clause, which relates the evaluation to a particular event in a localized manner. In Example 7.3, this is eating the food (line 5), in 7.4 the butterfly’s first flight (line 7). The relative absence of the Result stage in these stories of personal experience does not mean that the stories the children are telling are failed or incomplete narratives. Rather, it is the result of electing to use alternative story genres such as a Recount or an Anecdote, neither of which necessitates a Result section to complete them. In fact, both the girls and the boys in this sample were able to produce effective stories using these genres, as the examples given above attest. The reasons behind the prevalence of Anecdotes and Recounts might also be related to the influence of particular models of literacy. Previous research has identified the divergent literacy practices that children may experience outside school (Heath, 1982), but within the educational setting, children are exposed to and engage with multiple literacies too. One significant practice in the early years of schooling is what Michaels (1981) describes as ‘sharing time’. During this time, the children are encouraged to describe to the rest of the class some recent activity or event they have been involved in. Both the schools from which the samples were taken regularly employed this practice, particularly in the early primary years. Michaels argues that this practice functions as a ‘kind of oral preparation for literacy’ (p. 425) and goes on to describe the discourse features of the white American children’s discourse style as ‘topic-centred’. The examples that she uses to illustrate this style might also be analyzed as Recounts, for they provide a brief Orientation followed by a sequence of narrative events. In itself, this is unremarkable, given that Recounts, like Anecdotes, have usually been identified as conversational genres. However, it seems plausible that in this younger age group, the children are ‘taking’ this knowledge of conversational genres to build narrative schemas that are then transferred to other types of literacy practices, such as writing stories. This adaptation is not so problematic when the children are narrating personal experiences, which in their written form are perhaps one of the most closely allied storytelling practices to their oral counterparts in ‘sharing time’. However, when retelling existing stories, this incompleteness appeared to be less acceptable. Overall, the retellings constructed by the children were still dominated by the pattern of Orientation^Complicating Action, with evaluation dispersed throughout. However, compared with the personal stories,





Example 7.4 1. Once upon a time there lived a donkey called Chaki. 2. He lived at a hot desert. 3. The boy named Ali was very thirsty 4. So Chaki went of into the desert. 5. On the way he found a man who was a king. 6. He took the donkey to his pal[ace]. 7. The king has a magician. 8. He gave Chaki lots and lots of water. 9. Then his village never ran out of water again. (New Zealand Pakeha boy, age 7)

Orientation

CA

Evaluation Result

Example 7.5 (This example continues from a sequence of Orientation and Complicating Action.) 1. The next day the loin went out hunting. CA 2. He stepped on a strong brown piece of rope. 3. Suddenly as soon as he had sepped on the rope a very strong brown net fell on him. 4. The lion struggled and pulled, pulled and struggled. Evaluation 5. But he could not get free. 6. He rored and rored 7. Hus cries were heard all over the jungel 8. The mouse heard it from far away. Result 9. The mouse rushed to find the lion 10. She found him stuck in the net 11. The mouse chewed and crunched with her little sharp teeth 12. When the large lion was free out of the huge net he was delited 13. He said ‘Thank you mouse. 14. You were right after all, sorry’ 15. It’s all right. 16. Little friends are offen great friends. Coda (UK girl, age 9)



there was a greater likelihood of the children producing narratives that included a Result section (24.5% of retellings had Results compared with 2.9% of the personal narratives). The following examples show the closing sections of New Zealand and UK children’s stories containing Results.

At this age, the use of the narrative pattern and style associated with the retellings is not gender specific. An almost equal number of boys and girls did not include a Result (11 and 12 children respectively). Similarly, of the seven children who did include a Result, three were girls and four were boys. The factors that might promote the inclusion of a Result do not appear to be related to gender in any clear way. First, story topic is of importance. Unlike the personal narratives, the retellings lend themselves more readily to being structured as a progression from Complicating Action to Result as they are underpinned by Problem–Solution patterns. In his analysis of these culturally popular patterns, Hoey argues that once a ‘Problem’ has been signalled as such in a text, there is a very strong expectation that the pattern be completed by a satisfactory response (2001: 124). Predictably, in both the Examples 7.4 and 7.5, the patterns are completed. The village gets water and the lion escapes. At a superficial level, the content of the retold stories might then explain why there is a greater tendency for Result to occur in these narratives. Second, the retellings show the influence of written models of literacy. Although the children experienced the stories aurally (that is, the stories were read aloud to them), the source texts were themselves in the written mode. The use of evaluation is of interest in this respect. In Examples 7.4 and 7.5, the evaluation not only is used to express an external judgement about specific narrative events, but can be used in other sophisticated ways to support the structure of the Problem–Solution patterns. For example, in Example 7.4, line 3, the intensifier ‘very’ is used to highlight the problematic thirst to which the Donkey responds, marked by the causal connective, ‘so’ in line 4. Likewise, in Example 7.5, the lion’s distress is marked by the intensifying repetition in lines 4 and 6, underscored again by the exaggerated quantifier ‘all’ in line 7. The problem itself, ‘But he could not get free’ is signalled as a negative situation through the use of comparators. This includes the modal negation, ‘could not’ and adversative conjunction ‘but’, all of which create a moment of narrative crisis, or in Barthes’ terms, a nucleus or a ‘real hinge-point of the narrative’ (1977: 93). Similarly, evaluation devices reinforce the positive outcomes. For example, there is ‘lots and lots’ of water, which ‘never’ runs out again (lines 8 and 9). This use of evaluation is not unique and may be found in other narratives, particularly more complex, literary texts (Page, 2000). What is notable here is that this contrasts with the use of evaluation found in the children’s Anecdotes and Recounts. Even in personal narratives that contain a problematic situation which is resolved, the evaluation found in Anecdotes does not structure them as plot-like sequences. Instead, the external evaluation is





Example 7.6 1. One sizzling hot day me, my sister Elizabeth and my dog fudge were going to the park on our bikes with the dog on his lead what was atached to the handle bar of my bike 2. But when we got the park trouble struck. 3. My sister was hit by a boy on his bike too 4. My sister was right buy the poind when the boy hit her. 5. Oh no cried me (UK boy, age 7)

Orientation CA

Evaluation

This suggests that in the two different types of writing, the children are in some measure drawing on alternative models of narrative, and that it is this that motivates the relative inclusion or absence of a Result section, not some abstract division of masculine complete sequences and feminine incompleteness. While the data does not support a model of gender difference in storytelling style at this age, there is some evidence to suggest that the use of this narrative pattern is culturally specific. Of the seven children who did include a Result section, only one child was not European or of European descent, suggesting an association between this ‘complete’ narrative style and a western pattern. Stevens (2003) characterizes this outline as when the narrator ‘delve[s] into a thorough description of the problem and its solution, yielding a generic message which allows the interlocutor to identify and/or create repeated applications onto it’ (p. 178). This is precisely what happens in Examples 7.4 and 7.5, where the narrator solves the problems and may use this as the basis of concluding a generalized moral, such as ‘Little friends are offen great friends’ (Example 7.5, line 16). The teacher’s feedback to the students in the UK sample indicates that this is the model being endorsed and to which they should conform. For example, when a story did not have a Result, the teacher would often write some encouragement of the child’s opening, such as ‘A great start’ and might follow this up with ‘What happened next?’ prompting the expectation that the narrative should continue further. Another child was told to ‘separate the beginning, middle and end’. In this way, the children are encouraged to recognize the tripartite structure of this particular model of a ‘complete’ narrative,



used primarily to make explicit, localized judgements about specific events in the sequence, as in line 5 of Example 7.6.

and where necessary to modify their future writing accordingly. I do not wish to suggest in any way that the individual teacher’s feedback was restrictive or inappropriate. In all cases, the advice was encouraging and constructive. What I am trying to draw attention to is the specificity of the narrative model being promoted in this part of the children’s literacy curriculum. As Scollon and Scollon (1984) write, the power dynamics of a school situation are such that this kind of endorsement leaves little scope for students from backgrounds whose literacy styles do not match up with the official model. This might pose significant academic, social and political issues for such students. Evidence for this kind of disparity will be discussed in relation to the older children’s narratives.

Storytelling characteristics at ten years of age The use of the structural pattern Orientation^Complicating Action^ (Evaluation) continued in the personal stories told by the 10-year-old children, including texts that might be classed as Anecdotes, and occasionally Recounts. However, it is notable that the proportion of children who told these ‘incomplete’ sequences decreased significantly from 94% to 57%. Instead, many more of the 10-year-olds structured their personal experiences so as to include a Result. This is summarized in Table 7.5 below. This is as might be predicted, for previous research indicates that as the children become older, so they become more skilled at incorporating an increasing number of narrative elements. Once again, there is insufficient evidence to support a clear pattern of gender difference in this development (see Table 7.6). An equal number of boys used Table 7.5

Use of Result and child’s age (Personal narratives)

No Result/Coda Result/Coda

Table 7.6

7-year-olds

10-year-olds

94% 6%

57% 43%

Use of Result and gender of child (Personal narratives)


Female

Male

33% 19%

24% 24%




Gender, Age and Narrative Development 157 Use of Result and nationality of child (Personal narratives) UK No Result/Coda Result/Coda

New Zealand

5% 38%

52% 5%

a Result as those who did not, whereas the girls were slightly more likely to omit the Result. However, because of the size of the data sample (the difference is generated by two children), it is difficult to use this slight contrast in the girls’ narratives to make any substantial claims. However, there was marked difference in the nationality of the children who used these forms, where the majority of children who completed their personal narratives with Result or Coda were from the UK, whereas those who did not were from New Zealand. This is summarized in Table 7.7. The personal narratives written by the UK children were now far more likely to include the closing elements of Result and Coda. This was equally true for boys and girls. An example of one such ‘complete’ narrative is given in full (Example 7.7). Example 7.7 1. I suppose I could say I have been to many theme parks but haven’t been the one to go on scary rides 2. It was October the 5th 2003 when it all happened. 3. It will never be forgotton. 4. It was a beautiful sunny day 5. the heat was blasting on our foreheads as we were queing up for tickets for Wild Rides theme park. 6. Fineally, we got through the barrier 7. and entered the amazing theme park 8. You could hear the Run away train whizing through our ears. 9. I was so excited 10. I just couldn’t stand still. 11. But then it all happened. 12. The big ride staring me in the face the fabulous apocalypse 13. should I go on it, I thought.

Abstract Orientation

CA (1)

Evaluation CA (2)

Evaluation



Table 7.7


I was shaking but my mom was raring to go. But I plucked up courage and me and my mom joined the quew we kept seeing people going up and after a minute or two they were back down again. 20. This was the big bit 21. It was our turn. 22. Me and my mom sate in the seat 23. as the bar slowly raised above our heads I was very frightened 24. and nearly fell out. 25. Anyway, it started slowly 26. it truged up getting higher and higher 27. you could see the whole park from this hight 28. then the camera came down 29. and I knew we were at the top. 30. Just then like rain from the sky it shoot down mega fast 31. and then stopped at the bottom. 32. I was so pleased I did it 33. and went on several times after. 34. I have learnt that big rides are really good fun and not all that scary (UK girl, age 10)

CA (3)

Evaluation CA (4)

Evaluation Result Coda

This narrative, typical of the sample, may be usefully compared with the Anecdotes of the younger age group. Perhaps most obviously, the 10-year-old’s narrative is far more complex than the younger children’s anecdotal stories. This complexity is manifest at a number of levels. First, more narrative components are present, including an opening Abstract, then the closing Result and Coda. Second, the Complicating Action itself contains a wealth of evaluation, including both external and internal forms. The external evaluations punctuate the events, moving away from the time line and providing the narrator’s judgements or reactions (see lines 9, 13, 20 and 32). In this respect, the evaluations seem to function in a similar way to that in the Anecdotes and structure the Complicating Action as a series of episodes, so that the text can be subdivided as in Box 7.1.



14. 15. 16. 17. 18. 19.

Gender, Age and Narrative Development 159 Personal narrative subdivided into episodes

CA (1): Entry into the theme park (l.6–8) CA (2): Encountering the Apocalypse Ride (l.11–12) CA (3): Waiting in the queue (l.15–19) CA (4): Taking the ride (l.22–31)

There is also considerable use of internal evaluation. These functioned in a similar fashion to the evaluation in the younger children’s retellings. This includes the demarcation of narrative nuclei. For example, the modality in ‘Should I go on it’ (line 13) projects the hypothetical possibilities of taking the ride or not. The culmination of internal evaluation in episode four ‘taking the ride’ uses comparators, ‘higher and higher’ (line 26), and intensifiers, ‘the whole park’ (line 27). These also occur in combination ‘like rain from the sky it shoot down mega fast’ (line 30) to create narrative suspense and so dramatize the climactic nature of the experience. I have argued that while these functions are by no means exclusively literary in style, these could be aligned with the western style of written narratives. As these occurred with much greater frequency in the narratives of the older children, it would seem that by the age of ten, these UK children have adapted their narrative style much more closely to that of this westernized, literary model. The evaluation is also used frequently to make a globalized comment on the narrative as a whole, again, characteristic of the westernized model of storytelling. The inclusion of a framing Abstract and Coda is particularly important here. These were found in all except one of the UK personal narratives by 10-year-olds. Examples of these are given as follows: Example 7.8 Abstract I suppose you could say I’m one of the ones always getting into trouble but in a laughing way. But one laugh in my life really stands out. On February 18th 2003 I did something I will never forget. Coda In addition, I have learned to read lisence plate before I get in a car and I will memorise mine, it is P466AFV!!! (UK boy, age 10)



Box 7.1


Coda It’s really quite funny to think about it now. Although I was scared, I didn’t give up and this has turned me into a good swimmer and boosted my confidence. So much so that I’m now going for gold! (UK boy, age 10) The Abstract and Coda are often regarded as the most peripheral and least essential components of Labov’s framework. Abstracts function not only to provide a brief summary of the tale, but in the case of oral narratives may be a means of bidding for the floor and securing an extended turn in the discourse. As these are written narratives with a guaranteed audience (the teacher), the Abstract might indeed be considered as somewhat redundant. However, in these narratives, the Abstract and Coda provide the components in which the children focus their globalized evaluation of their experiences. In the case of the narrative 7.7 considered in detail here, this encapsulates the main theme of the story where the internal evaluation highlights the transformation in the narrator. The narrator’s personal experiences are then used as the basis for making a generalized statement in the Coda, which reinforces the theme of the story once again. The evaluation in these lines is underlined in Example 7.10. Example 7.10 1. I suppose I could say I have been to many theme parks but haven’t been the one to go on scary rides 34. I have learnt that big rides are really good fun and not all that scary

Abstract Coda

Thus the inclusion of a Coda and Abstract is not just an indication of increased structural sophistication in the storytelling of these UK children, but is clearly aligned with the model of western storytelling that promotes the use of global evaluative statements. Previous research has related this shift from localized to global uses of evaluation to the child’s age (Eaton, Collis and Lewis 1999: 704), but it is important to remember that the children do not naturally acquire these additional styles as some innate ability. This development must be understood as



Example 7.9 Abstract I’m not afraid of many things, but the thought of jumping into a deep pool was enough to make me jump out of my skin. Well, that was what it felt like when I was 5 and just about to have my first swimming lesson.

taking place within a specific cultural setting where particular forms of storytelling are put forward as the dominant, prescribed model. Within the UK school, the influence of National Curriculum requirements and the need to prepare children for Standard Assessment Tests may go towards explaining the strength with which this model is endorsed. The match between the children’s ethnicity and the mainstream model perhaps contributes to the uniformity with which this style was adopted, even in their personal writing. Thus it is not that the New Zealand children have not developed their narrative style at the same rate as the UK children. Rather, the difference might be due to the predominance of the western model being taught in a matching mono-cultural setting in the UK school, whereas the New Zealand sample was highly multicultural and needed to accommodate a number of divergent narrative styles. The cultural specificity of the western pattern becomes more evident still in the retellings produced by the 10-year-olds in this study. Within the multicultural sample of New Zealand students there was some evidence to suggest that these children were adapting their storytelling to authorized models of narrative in sometimes quite markedly different ways. This was made apparent in their manipulation of the narrative’s macro-level organization. The most common pattern for the retellings was that of the ‘complete’ sequence comprising Orientation^Complicating ActionÊvaluation^Result. Three quarters of the retellings were now structured in this way, reversing the trend of the 7-year-olds’ retellings (see Table 7.8). Perhaps this is not surprising, given the alignment of the conventional Problem–Solution type subject matter in these stories and the children’s increasing narrative sophistication, as demonstrated in relation to their personal narratives. Both boys and girls chose to use this structure more frequently than to omit the Result section. This is summarized in Table 7.9. These figures also indicate that there is some difference in the relative proportion of boys and girls who did not use this conventional pattern. A greater number of boys did not adapt their storytelling to this norm

Table 7.8

Use of Result and child’s age (Retellings)


7-year-olds

10-year-olds

76% 24%

25% 75%




162 Literary and Linguistic Approaches to Feminist Narratology Use of Result and gender of child (Retellings) Female

Male


6% 47%

19% 28%

Total

53%

47%

(three times as many boys as girls). This reverses the claim made by Gergen cited earlier in Chapter 4 that ‘Men are far more likely to accommodate themselves to prevailing criteria for “proper story telling” than women’ (1998: 6). What is significant here is that this tentative evidence for a pattern of gender difference is not abstract, but must be understood as intersecting within specific cultural patterns. Within the New Zealand sample, the most marked differences occurred between the narratives told by the Pakeha girls and the Maori boys. The Pakeha girls’ stories were most like and the Maori boys’ stories were most unlike the western pattern adopted by the UK children. I should stress that this cultural difference is not an exclusive categorization, for one Maori girl did follow the conventional pattern that closed with Result and two of the children who did not close their stories with a defined result were Pakeha (one boy, one girl). However, the differences between the storytelling of the Maori boys and the Pakeha girls are so extreme as to warrant further analysis and comment. One story that the children retold was called Fishing for Octopus by Feana Tu’akoi (2001). The class teacher had read this story to the children over a period of several days. The fictional story concerns the activities of a young Tongan boy, Sione, and his father who go fishing for octopus, using the traditional Tongan practice of baiting the octopus with a shell on a stick. The structure of the story is complex and contains multiple levels, where the father narrates an embedded hypodiegetic narrative of the Octopus and Rat, a Tongan myth that explains the use of a shell as octopus bait. Due to the length of the original story and for copyright reasons, I am unable to reproduce the full story here. Instead, I provide a skeleton analysis that facilitates a fuller comparison of two children’s retellings considered later (Box 7.2). One of the themes of the story is the contrast between Tongan and New Zealand ways of life. The juxtaposition of two cultural traditions extends beyond the story content (how to catch an octopus) and incorporates aspects of differing storytelling styles. On the one hand, the embedded Octopus and Rat myth is structured as an interactive dialogue, culminates in a Tongan saying and is used in an associative manner to explain



Table 7.9


Complicating Action Ep 1: Aunty Lupe laughs at Sione and his father for wearing lifejackets to go fishing. Ep 2: Sione and his father paddle out to sea Ep 3: Dad explains that they will use the shell as a fake rat to bait the octopus. Sione asks why this would work. Ep 4: In response, the father tells an embedded narrative Orientation: In the old days, the Rat and Octopus were friends Octopus would carry Rat on his head while they swam. Complicating One day, the Octopus swam further and Action The Rat went to the toilet on his head. Evaluation Sione says ‘Gross!’ Result Rat does not tell Octopus what he has done until they get back to shore. Then he taunts the Octopus. Rat hides in a hole because Octopus is angry Coda Octopus still hate rats today and will go crazy to catch one. Ep 5: Dad catches an octopus using the shell as bait Ep 6: Sione sees the octopus head Ep 7: Sione upsets the canoe when startled by the slimy octopus. Ep 8: Dad keeps Sione safe, but the octopus escapes. Result: Dad solves the problem by deciding to catch an octopus the ‘New Zealand way’. This is buying a fresh octopus from the market in town.

Tongan practices. But while the traditions of Pacific Island culture are being celebrated here, the narrative contains many features similar to the westernized pattern described earlier. This includes the use of Goal– Achievement patterns, with the goal being to successfully catch the Octopus. Even though the story begins in medias res, it is rich in orientation detail and contains a wealth of evaluation. The original Fishing for Octopus thus contains a blend of narrative styles, where the oral Tongan story is contained within a westernized framework, thereby offering the children a range of narrative models that they might draw upon.



Box 7.2 Schematic outline of Fishing for Octopus by Feana Tu’akoi

The most lengthy and complete of the retellings was narrated by a Pakeha girl, Melanie. Melanie is an extremely competent storyteller, well able to re-create the stylistic features of the western model of narrative. This was exhibited in many of the retellings she produced, but is most impressive in her handling of the narrative complexity of this tale. At a micro-level, Melanie’s telling reiterated the detail of the original, as seen in the examples below, taken from the opening of each text. Example 7.11 (Opening to the original version) 1. ‘Eh, hee-hee-hee!’ 2. Auntie Lupe’s laugh was loud and musical. 3. ‘Why are you wearing those things?’ she cried. 4. Dad grinned. 5. ‘They’re called life jackets,’ he said, 6. and we’re wearing them to keep ourselves safe.’ 7. ‘You’ve been living over in New Zealand for too long,’ said Auntie Lupe. 8. ‘You look like a couple of tourists!’ 9. ‘But I still fish like a Tongan,’ said Dad. 10. ‘You watch – we’ll have octopus for dinner!’ 11. Auntie Lupe shook her head and went back up to the house, still laughing Example 7.12 (Opening to Melanie’s retelling) 1. ‘ha ha ha ha he’ 2. Aunty lupes laugh was loud and scary. 3. She said ‘What are those stupid things your wearing’ 4. Siones dad said ‘They are life jackets 5. They keep us safe’ 6. Aunty lupe laughed even louder than before 7. And said you have been living in New Zealand for to long. 8. Then she started walking up the golden soft sand back to the house. There are obvious parallels between the two versions, but it is notable that Melanie’s version emphasizes the use of descriptive detail. For example, she mirrors the double adjective in line 2, but introduces further adjectives in lines 3, 6 and 8 (underlined above). The inclusion of such detail is heavily prescribed in the UK story writing. For example, in one of the UK sample texts analyzed here the children were explicitly instructed to ‘include ten adjectives’, and when marked, the work was




awarded credit for the incorporation of such description. Melanie’s storytelling shows similar tendencies to conform to the characteristics of this discourse style. The structural sophistication in her retelling is also remarkable. Not only is Melanie skilled in replicating the lexical detail of the original, she is also well able to negotiate the shift in diegetic levels. This includes the embedding of the Octopus and Rat story, structured as a dialogue between Sione and his father. Example 7.13 1. Sione’s dad got out a shell 2. And said This is what we are going to catch octopus with. 3. Sione said What do octopuses eat. 4. Shouldn’t we be using that. 5. No my son 6. I will tell you a story 7. Its about a mouse an a octopus

Complicating Action

Abstract

Melanie was not alone in this, and the other Pakeha girls all included this shift in hypodiegetic level too. However, she was the most successful at incorporating the events found in Complicating Action episodes 1–2, 5–8 and Result of the original. This is important because it is these episodes that critique and then legitimize the ‘New Zealand way’ of fishing that frame the central Tongan story. The wearing of life jackets mocked in Episode 1 appears a sensible precaution when the canoe nearly capsizes (Episodes 7 and 8) and buying an octopus, rather than catching one, is the ‘Means’ by which the ‘Goal’ of providing an octopus for tea is successfully achieved. On one level, this demonstrates Melanie’s ability to reproduce the westernized patterns of Goals and Achievement, structurally realized as a ‘complete’ sequence that ends with a Result stage. On another, in terms of the specific content on this story, these are used to validate her own cultural position as that of Pakeha New Zealander too. The retelling produced by a Maori boy, Nathan, was quite different to this. I reproduce his version in its entirety (Example 7.14). Example 7.14 1. Sione and his dad are at Tonga 2. And they go fishing for octopus. 3. His dad is a cheerful, cheeky and strong man.

Orientation





Sione is a scared and happy boy. Octopus is a angry octopus. Sione and his dad used a shell because the octopus is angry of rats and mice. The octopus got mad and wanted to eat the mouse because the mouse left his droppings on his head.

Complicating Action

Several observations may be made about this text. It is brief, and does not contain the multiple diegetic levels of either the original or the Pakeha girls’ retellings. While there is a clear Orientation (lines 1–5) followed by a report of events (6–10), there is no explicit Evaluation, Result or Coda. The events that are included do not seem to be centred on a recognizable predictable pattern, such as the Goal–Achievement or Problem–Solution sequences used in the original. The narrator simply states that ‘they go fishing for octopus’ (line 2) rather than expressing this as intended goal, as in the original ‘You watch – we’ll have octopus for dinner!’ Indeed, Nathan’s version never makes it clear whether the octopus is ever caught. Even within the Complicating Action, there are no temporal markers which might organize the events into a chronological sequence. Instead, Nathan’s retelling can be characterized as an extremely succinct summary which draws together certain aspects of the original telling. This summarizing takes place in both the Orientation and the Complicating Action. As would be expected, in the Orientation section, the narrator gives us information about the characters and their location. This is lexically detailed, containing a number of adjectival attributes, for example the father is ‘cheerful, cheeky and strong’ (line 3); while Sione is ‘scared and happy’ (line 4). However, these distil the separate attributes the characters exhibit throughout the narrative into single clauses, without indicating the chronological separation of the states. Thus Sione is ‘scared’ in Episode 7 and ‘happy’ in the Result of the original, but in Nathan’s retelling, this appears to be a simultaneous (and contradictory) combination (line 4). The events in the Complicating Action also restate some of the main events of Episodes 3 and 4 from the original story. However, this is not organized as a temporal sequence. Instead, lines 5–10 are organized around associative principles. The central theme ‘Octopus is a angry octopus’ acts as a semantic nucleus around which the explanations of the uses for and reasons behind this anger in lines 6–7 and 8–10 are arranged as satellites. It is



4. 5. 6. 7. 8. 9. 10.

not then that Nathan’s retelling is ad hoc, but rather that it uses different structural bases for its organization. These differing characteristics share similarities with the discourse style used by Maoris in conversational narratives. Holmes (1998a) describes this pattern as showing ‘A tendency to omit or truncate story components’, and more specifically ‘A number of Maori stories have no explicit resolution or coda and as a result, from a Pakeha perspective, the story seems to have been left unfinished’ (p. 32). This is a most apt description of Nathan’s storytelling. From a Pakeha perspective it does seem incomplete and the audience is left to infer the temporal connections that would provide chronological coherence to the narrative. Again, this is typical of Maori discourse style in general, where meanings are left implicit and apparently undeveloped (Holmes, 1998a: 35). These differences in storytelling style may be related more generally to Maori values and literacy practices. These contrast with that of mainstream literacy and can be indicated by a comparison of the backgrounds of the children in this study. Melanie’s home literacy experiences are typical of mainstream patterns. From an early age both mother and grandmother read to her regularly, usually at bedtime. In the preschool attended before her current school, mainstream practices were in place with story times and a book corner. Melanie continues to participate in these practices and in turn now reads regularly to her younger siblings. The predominate model here is one familiar to those who have shared such experiences. Heath (1982) notes that this structures the interaction as between a teller who controls the narration of the story and an audience who learn to ‘listen and wait’ (p. 53). Interaction is minimalized. In contrast, Nathan has few memories of being read to at home. Instead, his experiences of narrative are dominated by ‘church literacy’, which Phillips et al. (2004) associate with recitation, a practice which may not transfer easily to the kinds of literacy being promoted in school (p. 310). There is some evidence to suggest that the influences of oral discourse style shape the Maori children’s perceptions of storytelling as a literacy event. Unlike the mainstream pattern, Maori storytelling is an ongoing process of interaction (Holmes, 1998a: 35), one in line with the more general Maori emphasis on group relations rather than individualism (Hohepa et al., 1996: 30). Thus the practice of retelling a story is understood as a dialogue between narrator and audience. This sense of responding to an interlocutor was explicitly spelt out in one of the 7-year-old Maori children’s retellings, illustrated in Example 7.15.




Example 7.15 1. The great escape. 2. Who lived in the cage? 3. A hamster 4. How did’t he get out of the cage. 5. I think he got out of a hole 6. I wonder who did’t it 7. I do not know 8. I thinkak tanden did’t it (New Zealand Maori girl, age 7, retelling) In Nathan’s version of Fishing for Octopus, this assumed interaction is not overtly expressed. However, if his retelling is understood as a response to the hypothetical question ‘tell me what happened?’ then the summarized response becomes much more appropriate. The detail of the retelling is then redundant, for the child predicts, rightly, that his audience (the teacher) already knows the full content of the story and can infer the missing information. However, reproducing this pattern of assumed interaction does not help meet the ‘demand for explicitness which characterizes western education’ (Holmes, 1998a: 52).

Cultural context and gender Given these cultural differences in discourse style and literacy practices, it is perhaps no surprise that Nathan’s storytelling does not match that of the mainstream western model. The values of Maori culture may also go some way towards explaining the gender asymmetry in the use of narrative styles in this group of 10-year-olds. Holmes’ study found that young Maori men’s storytelling was more strongly influenced by Maori culture than that of their female counterparts. She relates this to the gender roles within Maori society, characterized by male dominance where ‘Maori men typically derive more status from affiliation with Maori culture’ and within which women might be ‘excluded from traditional areas of power’ (1998a: 51). Similarly, even though this is a smallscale study it is notable that none of the Maori boys (from either age group) produced stories with a Result, and tended to reproduce Maoristyle narratives, whereas at least one Maori girl chose to adapt to the mainstream conventions. Based on the evidence from this study, this gender difference is age specific. It is pronounced in the writing of the 10-year-olds, but not in




that of the younger age group. One reason for this might be that the narratives of the 7-year-olds were chiefly characterized as ‘incomplete’ (that is, without a Result or Coda), and thus the parallel features of the Maori style do not appear anomalous in this context. It might also be a result of the education system. While I have attempted to stress the differences in narrative style in a non-pejorative manner, it is important to remember that not all differences are equal. Within the New Zealand context, it is the Pakeha who are the dominant social group, and that it is the westernized narrative style associated with this that is recognized and promoted within mainstream educational contexts. Earlier research has indicated that the synchronization of culturally oriented literacy practices within the home and those within the school can influence children’s ability to adapt to mainstream schooling demands (Heath, 1982; Phillips et al., 2004). Where there is a similarity, then the children have few problems adapting to literacy models of school. Conversely, where there is a mismatch, there is the potential for the children to be misunderstood and less successful in academic terms. This issue escalates as the children progress through the education system (McDonald and McNaughton, 1999: 349), in a ‘rich get richer’ syndrome. Those able to accommodate the mainstream discourse style gain credit for this. In contrast, if little value is ascribed to alternative patterns, then the student may become disadvantaged. This may be manifest as inadequate understanding or support from within the education system (Michaels, 1991), or at the least, demotivation on the part of the student. While these gender differences are not widespread, I have drawn attention to them because it is possible that these might lead to interethnic communication problems within an education system, and result in specific social and political consequences. Exposing and analyzing these points of difference in narrative style is important, for it is when such areas of mismatch are described that action can be taken towards changing the situation. Indeed, there is optimistic evidence that such action has been initiated within the New Zealand education system more generally, proving that such differences are ‘neither necessary nor immutable’ (Phillips et al., 2004: 320).3 This is laudable, but the adoption of such changes must be handled with sensitivity. For example, in enabling children to achieve educational standards, must this come at the cost of eradicating diversity and reinforcing existing power relations that pose the dominant group as the norm? The evidence in this study has hinted that it is the Maori boys who are particularly at odds with mainstream narrative norms. However, while




identifying a specific group on the basis of gender or ethnicity as nonnormative or ‘at risk’ can also perpetrate stereotypes. This may be unhelpful and lead to pejorative labelling and over zealous remediation, potentially obscuring students outside the category who might also have divergent discourse styles in need of recognition and value (Prochnow et al., 2001). I agree with Dutro when she argues that ‘it is crucial to understand how gender functions in classroom contexts and affects children’s experiences’ (2003: 467), but this kind of analysis must balance an understanding of macro-level influences with the performances of the individual child.

Conclusion In this chapter, I have reported the findings of a cross-cultural study of children’s written storytelling. The results indicated that for the younger age group there was no substantial evidence to indicate gender difference in the children’s narrative skills. Neither boys nor girls produced narratives that were ‘complete’ according to the six-part Labovian framework. The narratives of the older children suggest that this ‘fully formed’ model may be culturally specific, aligned with a westernized model of story. This model was heavily prescribed in the UK school curriculum, and both older boys and girls from this mono-cultural context told similarly ‘complete’ narratives. However, the stories written by the New Zealand children showed extreme variation, polarized by the Pakeha girls and Maori boys. I have argued that this gender variation in storytelling style is inextricably linked to cultural contrasts between Maori and Pakeha discourse style, rather than abstract or universalized. The findings support earlier research which reported that girls achieved higher literacy scores than boys and that Pakeha children outperformed non-Pakeha. However, the stories told by the New Zealand children did not uniformly support this categorization, and the relative correlation between Pakeha boys and non-Pakeha girls is not clear. Further research with a larger data sample would be needed to explore this in more detail. These findings have important implications for the wider project of feminist narratology. Primarily, it challenges once again the universalized assumption that gender and form can be correlated in a simplistic, binary manner. In particular, there is no evidence to suggest a pairing of normative patterns with a ‘masculine’ style, hierarchically juxtaposed with a non-normative ‘feminine’ alternative. First, the results show many points of similarity between the boys’ and girls’ stories. Where




the narratives are incomplete and so apparently non-normative, this might be motivated by a number of factors rather than an essentialist discourse strategy like écriture feminine. Here these sociolinguistic factors include the age of the child, the choice of story genre in relation to topic, cultural practices and discourse styles. Where the differences did intersect with the child’s gender, this was shown to be age specific and overlaid with cross-cultural contrasts. Significantly, it was boys who did not adopt mainstream structures, reversing the assumption that the normative models of narrative are ‘male’ in some way. Indeed, from this study it is difficult to argue that the mainstream literacy models are unequivocally gendered in any explicit manner. Instead, the intersection of normative patterns and socio-political power is a complex one which may shift according to context and involve a number of variables, not just gender alone. Within feminist narratology, it remains critical to distinguish between the axes that power (itself a diffuse and slippery concept) operates within. This network is multidimensional, entailing gender relations, but not exclusively so as some hierarchy where all men have power over all women. As earlier chapters have endorsed, it is clear that a specific narratological feature such as global structure cannot be simplistically mapped onto ideological values such as ‘gender’. However, the use of such forms may be valued in differing ways which have social and political consequences for different groups, including here those distinguished by ethnicity and gender. The analysis presented in this chapter has suggested the potential for a group of boys to be disadvantaged, a position diametrically opposed to the devaluing of women’s texts underpinning the genesis of feminist narratology in the 1980s. This reversal might be explained by closer attention to the educational and academic contexts from which feminist narratology emerged. Second wave feminism was reacting against patriarchal influences both in the construction of the canon, and more generally in higher education academic curricula and environments. These are quite different contexts from those in which the children in this study are situated, and to that extent it is to be expected that different patterns of gendered (non-)normativity might emerge. However, feminist narratology might well ask further questions about the uses to which narratives get put in differing educational contexts and the consequences this might have for differing groups, in terms of both the future success and the value of both individual and their narratives. Significant changes have occurred in the development of women’s studies since the 1980s. Although much has been achieved in terms of challenging





As long as the current system of rewards and privileges dominates, certain talents and commitments will continue to be undervalued and the present system of values will continue to have gendered implications. (2001: 355) As both narrative analyses and gender studies continue to be integrated across these academic contexts, new territory is provided in which the relationship between narrative and gender may be examined not only in terms of the subject knowledge incorporated in the curricula, but also as a means of questioning the potential role of gender in shaping our norms of what constitutes ‘narrative’ itself.



norms and establishing gender studies as a critical academic field in its own right, the success of feminism has not come without potential drawbacks (Martin, 2001). Critiquing the current climate, Martin goes on to write,

8

Synthesis: Weaving the strands together The strands of feminist narratology brought together in this volume are certainly diverse in terms of their disciplinary orientation and subject matter, ranging across media, historical period and theoretical terrain. In this last chapter, I want to reflect on the connections that might be made when these strands are woven together, the ways in which they might enhance and challenge one another. The diversity of postmodern feminist narratology need not be understood as ad hoc. Instead, there are common assumptions that run through many of the strands. The first, and perhaps most obvious of these, is that postmodern feminist narratology eschews universalism. This move away from universal categories and structures is apparent in many ways. Paradoxically, while it consistently underpins the use of narrative theory and the understanding of gender throughout each of the chapters, this unifying principle also opens up the possibility of multiple variations.

The impact on narrative theory Reviewing the achievements of both early and postmodern feminist narratology phases, a key area in which the move away from universalism has been felt is the revision of narrative theory. Feminist narratology’s rewriting of this can be understood as part of the process of evolution common to postclassical narratology more generally (Sommer, 2004: 4).1 Taking into account texts by and about women has not proved the need to reject existing models and establish radically different theoretical paradigms. Instead, examining these frameworks against a wider body of actual narratives of many varieties, not just differentiated on the 173



Beyond Feminist Narratology?

basis of gender, has indicated that the ‘deficiency’ of structuralist narratology might lie less with the models themselves and more its universalizing assumptions. Thus the narrative theory that is applied within feminist narratology is not of the structuralist, quasi-scientific bent ascribed to the genesis of narratology itself (Prince, 1995b: 110). Rather, as Lanser (1986) anticipated, the narrative theory that I have applied throughout tends to ‘favor flexible categories over fixed sets’ (p. 345) and has refined existing models by ‘add[ing] to the narratological distinctions that already exist’ (p. 344). This has taken the form of reworking classical narrative models so that they are not seen as definitive outlines, but rather as paradigms open to variation. Thus Longacre’s (1983) anatomy of plot is enhanced by its alignment with the more fluid and plural framework of Hoey’s culturally popular predictable patterns (2001). Labov’s (1972) six-part ‘fully formed’ narrative is understood as one of various conversational story genres identified by Martin and Plum (1997). The concept of narrativity may be probed in a number of ways, which includes the use of tools familiar from classical narratology (such as Genette’s work on temporal relations) and also from text linguistics, such as Hoey’s work on discourse colonies (2001) and Halliday’s system of transitivity (1994). That feminist narratology has not developed an alternative system of narrative theory should not be regarded as a failing. On one hand, this is symptomatic of shifts in narratology more generally in recent years (Fludernik, 2000a: 2). More importantly, in framing its revision as a reworking of the existing categories, postmodern feminist narratology avoids constructing another set of hierarchical oppositions, for example, between ‘feminist narrative theory’ and ‘masculine narrative theory’. This might have reinforced a false, binary pairing and worked in the disservice of different groups of women. An important outcome of this more flexible approach is that while this narrative theory does not claim a universal status for itself, it is able to mediate across texts and perspectives. This use of narrative theory acts like a ‘lingua franca’, bridging analyses and discussion that have often remained in separate but parallel paths (Sommer, 2004: 6). For example, the concern with teleology intersects with both the literary concerns of psychoanalytically derived theories of plot and with empirically oriented sociolinguistic studies of conversational narrative or children’s storytelling. Indeed, many of the narrative frameworks have proved their ongoing use in application to both ‘literary’ and ‘non-literary’ texts as a means of drawing out interpretive conclusions, often of importance from a feminist perspective. The identification of Problem– Solution, Goal–Achievement and other related patterns has been an




effective means of examining romance literature from the medieval period, media reports from the turn of the millennium and stories told by children of different cultures. In each case, the analysis of these patterns is useful for asking questions about the relationship between content and form. For example, who or what is represented as the ‘problem’ to be resolved, or ‘goal’ to be achieved may be highly pertinent in terms of the gender relations being constructed in the text, and indicative of the social context from which the text is derived. In the examples I have covered, this ranges from the problematic issue of feminine sexuality in Japanese monogatari (Chapter 3) to the contentious interpretation of Hillary Rodham Clinton and Cherie Booth/Blair as feminist role models (Chapter 6). The moral values associated with these narrative patterns show the remarkably robust nature of gendered stereotypes that may travel across time and media. An example of this is the feminine virtue of stability, highlighted by the medieval romance heroine’s repeated survival of trials in a Problem–Solution pattern, echoed centuries later in the evaluation of Hillary Rodham Clinton’s response to her husband’s alleged infidelity. While this is clearly not a universal attribute, it is striking to see its reappearance, pinpointed with such clarity by the narrative analyses. The predictable patterns are also salient from a structural point of view. However, the interpretation of such narrative features might vary considerably. The issue of closure is a case in point. In the discussion of media texts, the ideological value associated with the end point of positive or negative evaluation was used to differentiate between narratives of success and failure for the high-profile figures of Cherie Booth/Blair and Hillary Rodham Clinton. The social uses to which these narratives were put in building and undercutting their husband’s symbolic capital illustrated the ways in which narrative forms can operate as a means of serving patriarchal ends. However, the discussion of the children’s inclusion of a ‘Result’ section that correlated with a positive outcome was understood as rooted in mainstream educational practices (Chapter 7). The ability to adapt to this narrative pattern could be of social significance (for example, affecting academic achievement), but this did not necessarily reflect a patriarchal agenda. In fact, it was a group of boys who were disadvantaged by the privileging of these narrative norms. Another framework that has adapted flexibly across different types of narrative is that formulated by Labov (1972). The variation from this norm has been explained in part by the use of alternative story genres, especially Anecdotes and Recounts (Martin and Plum, 1997). Again, this has been useful for examining narratives of very different kinds, ranging from



Beyond Feminist Narratology? 175

conversational narratives about childbirth to the stories told by children in schools, but the relationship between these narrative forms, functions and gender is not univalent. In the case of the childbirth narratives, the use of the Anecdotal structure was seen as motivated by interpersonal needs for solidarity, shared in varying degrees by both women and men. This social dimension is less prominent in the discussion of the children’s narratives, where the influence of oral models of literacy may be related to the prevalence of Anecdotal structures in the younger children’s writing, and which developed into gender-specific patterns over time. So while the same narrative framework may be used to identify similar patterns in different texts, the meaning associated with that pattern, or part of that pattern, is not singular, but embedded with particular contexts and functions.

Concepts of gender The impact of integrating these literary and linguistic approaches to narrative is a heightened awareness that gender cannot be understood as a universal concept. Time and again I have reiterated the need to move away from the monolithic model of gender presupposed in the second wave emphasis of early feminist narratology. To that end, the analyses that I have offered have tried to indicate at least some of the potential multiplicity implicated in gendered experiences. Categories such as ‘women’ and ‘men’ have been shown to be considerably diverse, where ‘gender’ intersects in multiple ways with other variables including historical period, cultural context, age, ethnicity and academic background and could include others such as sexuality or class. Regardless of whether the narratives are ‘literary’ or ‘non-literary’, the identification of gender cannot be treated in an abstract way. Where questions of gender have been raised in relation to various stories, I have emphasized the need to ground these in the specifics of a localized context. Thus the subtle differences in the oral stories told by women and men about the birth of their children are attributed to the dynamics of the specific interpersonal relationships that existed within that social network. The variations in the representation of Hillary Rodham Clinton are interpreted in the light of the political stance taken by particular newspapers, which may vary over time. The writing of Michèle Roberts and Toni Morrison is understood as inextricably bound up with their desire to articulate narratives that foreground the importance of racial as well as gendered identity. The differences between the older New Zealand children’s storytelling is embedded in the context of differences




between Maori and Pakeha culture and their use of different discourse styles. From the perspective of the postmodern feminist narratology drawn on here, gender is quite clearly not an essentialist concept that can be regarded as an abstract category. It is always interrelated with an openended number of other factors that may vary from context to context. With more far-reaching implications, the treatment of gender has also suggested ways in which the categorization of ‘men’ and ‘women’ is not itself fixed. In the case of Michèle Roberts’ fiction, this included examples of gender-blending that deconstructed a single gendered identity of exclusively ‘male’ or ‘female’ for her characters. In the creative extensions of the hypertext fiction written by university students, the categories of gender were further disturbed by the use of cyborg imagery merging human and machine. However, this deconstruction has co-existed with much evidence to reinforce the ongoing significance and efficacy of the norms of sexual difference. Examples include the readers’ use of a default heteronormative paradigm to establish gendered identities for characters in texts, even when texts (like Roberts’ Flesh and Blood) challenge this. Likewise, it is salutary to note the extent to which even supposedly high profile feminist figures such as Hillary Rodham Clinton are co-opted within masculinist agendas of the media. Clearly, destabilizing the norms of what it means to talk about ‘women’ and ‘men’ does not mean that postmodern feminist narratology can dismiss questions about the power that gendered norms and stereotypes continue to exert in various stories and with differing consequences.

The importance of context The outcome of this move away from universalism into localized perspectives is that the issue of context assumes an increasing level of priority and attention. Indeed, the importance of context is a further assumption that weaves throughout many of the strands of postmodern feminist narratology. This was inherent in early feminist narratology, indicated by Lanser’s assertion of its importance for ‘determining meaning’ (1986: 343) and typical of wider trends towards contextualism in narratology at that time. However, in the integrative approach of postmodern feminist narratology, contextualism has entered a new phase. In its early conception of context, feminist narratology tended to treat this in a somewhat diffuse manner. I have already spent some time indicating the implications this had for understanding elements of context, like ‘gender’ in potentially abstract terms. In contrast, the second wave of contextualism understands context in more detailed and localized




terms, an approach heavily influenced by postmodern feminist theorizing and sociolinguistic perspectives. Context always has been a multifaceted concept, understood in different ways by researchers working in different paradigms. Within postmodern feminist narratology there is an increasing recognition of the indispensible but plural meanings this might entail, especially when attempting to account for the role of gender in the analysis of narratives. Without argument, the relevance of any given contextual factor will vary according to the narrative in question and fluctuate according to text type. For example, factors important in face-to-face interaction (as in the oral narratives considered in Chapter 4) may be much more immediate than in the separation of teller and audience involved in written narratives (such as literary texts). Thus whether the narrative is told in a public or a private setting may be highly significant in oral storytelling, and create particular gendered relationships between teller and audience. In contrast, the physical setting in which a literary narrative is written or read may bear little on the interpretation of gender relations. Instead, matters of producing and consuming written narratives may involve a rather different array of parameters. I have touched on some of these earlier in the book, as in the case of the political agenda that motivates (at least in part) the stories that get told in the media about Cherie Booth/Blair and Hillary Rodham Clinton. Literary narratives are not exempt from similar restrictions. Michèle Roberts comments on the influence of market demand and mainstream publishing requirements in terms of her own and Jeanette Winterson’s representation of lesbian characters and their stories (Bratton, 2002). In its postmodern plurality, it may be that some contextual elements are more closely related to gender than others, and the relationship that this bears on any given text is not uniform. As Chatman writes, the contextualist trend in narrative studies is provocative and begs a number of questions that are not easily answered (1990: 314–315). Indeed, moving into the analysis of extra-textual features does not transfer comfortably onto textual study. Language is not simply mimetic, but exists in a dialectic relationship with social reality. But while that relationship is difficult to pin down in clear terms, it cannot be severed either. Because the role of gender in these real-world contexts remains of such importance and in many varied ways, the relationship between narrative and context is one that will remain central to feminist narratology. However, it is one that must be treated with sensitivity and may involve incorporating modes of analysis that lie beyond structuralist narratology. As the various studies in this book have brought out, building such a




picture of contextual influences is an ongoing project which will require many more studies grounded in specific situations to avoid abstraction and help us understand the multiple ways in which gender might be important in the telling of stories. In its exploration of the possible relationships between content, form, context and gender, postmodern feminist narratology has not reached any universalized conclusions. Instead, it has become apparent that there is no unequivocal relationship between narrative form and gender. Lanser (1986) posed the question of ‘whether women and men do write differently’ (p. 346). The hypothesis of difference underpins the very premise that initiated feminist narratology, exemplified by its critique of an androcentric corpus and subsequent focus on texts by and/or about women. This led to an extrapolated assumption where the difference between ‘normative’ patterns and ‘non-normative’ variations were inflected with gendered values. Examples of this include the gendered plot types (male ambition contrasted with female alternatives) critiqued in detail in Chapter 2. Yet the evidence does not point to a clear pattern of difference in the narrative strategies used by women and men. Richardson’s (2000b) study of linearity pertinently demonstrates that unconventional narrative forms are by no means the exclusive privilege of feminist writers. Instead, authors typically cited as employing experimental narrative techniques for feminist ends, such as Woolf, Jean Rhys and Angela Carter, did not necessarily reject stylistic features such as linearity out of hand. The analysis presented in many of the chapters here supports this conclusion. My own study of Michèle Roberts, Toni Morrison, Nabokov and Calvino indicated the many points of stylistic similarity between these authors’ works, regardless of the gendered content of these stories or of the author’s identification. Extending this to include women and men’s engagement with narratives of many different kinds, the case against a gendered pattern of difference becomes stronger. Across the different strands of postmodern feminist narratology, there is an almost shocking amount of similarity in the storytelling of women and men. For example, both women and men who talked about the birth of their children used Anecdotal structures, there was significant similarity in the amount of linearity used by the male and female readers of the hypertext narrative, and within the youngest age group of the children and across all groups of the UK children there were parallel patterns of narrative development. In the face of this evidence, the equation of male-normativity contrasted with female-non-normativity simply does not hold.




Given the challenge to this kind of gendered pairing, it is important to examine the bases of normativity further. The kind of feminist narratology laid out by Lanser is quite right to ‘scrutinize its norms to be sure of what they are normative’ (1986: 345). However, within the strands that are present in this book, the reasons behind normative and non-normative uses of narrative do not always seem to be gender specific. To refer again to Labov’s six-part model, feminist critics have drawn attention to the gender bias in his original corpus (see Chapter 2). From this point of view, structural norms are equated with masculinity (if not in the biological gender of the male teenagers, then in the social value ascribed to the topic of the fight narratives). However, as I have already indicated, it is not just female storytellers who used alternative forms. Moreover, the women’s and men’s use of Anecdotes (rather than the Labovian model) seemed to be derived from the social function of these stories to construct solidarity, not by some kind of feminist or intrinsically feminine impulse. Likewise, the children’s uses of apparently incomplete sequences (lacking the final stages of Result and Coda) could be explained by a number of reasons, such as the age of the child, the choice of story topic and the features of Maori discourse style. In the similar case of the hypertext narrative, the readers’ decisions to reconstruct conventional patterns of narrative through their pathways was motivated at least in part by their expectations of the medium and constrained by the transitional nature of the text being read. There was little to hint that this was explicitly gendered in any way. Thus non-normativity (in relation to the Labovian framework) is not some gendered reflex against a patriarchal norm in these cases. However, I do not want to suggest that Labov’s model (or indeed any other) is universal or can be treated in an abstract way. Indeed, from the various studies that I have brought together here, it is less the masculine nature of Labov’s corpus that emerges and more its alignment with the characteristics of a westernized style of storytelling (at least in comparison to Maori storytelling, for example). As narratives from other cultures continue to be studied, other discourse styles and narrative patterns may emerge that refine our sense of what constitutes the ‘norm’ and which may or may not correlate with gender values in various ways. The result of this plurality is that the relationship between narrative form and gendered issues does not map on simplistically to a binary pattern of difference, nor can it be divorced from the specifics of its context. Some have gone as far as to claim that narrative structure cannot be associated with ideology at all. Richardson writes,





I agree with Richardson insofar as form and gender cannot be correlated in an abstract sense. Indeed, much of the work in feminist linguistics has stressed this point about the relationship between language and gender. As Johnson writes, ‘there are very few generalizations which can be made about formal, structural aspects of language of one sex as opposed to the other’ (1997: 22). But this is not to say that questions about gender are irrelevant to narrative analyses. Although linguistic forms (including narrative structure) are not themselves meaningful, they can contribute to the way in which a text is interpreted. As the well-worn cliché points out, it is not just what you say but the way it is said that matters. Those ‘ways’, which include the use of narrative form, employ linguistic conventions. Although these are not gender specific in an exclusive or fixed sense, those working in critical linguistics would also argue that these can be powerful in the selective way that they represent the world (Cameron, 1998c: 161). The relationship between gender and narrative structure is thus an indirect and mediated one, inextricably part of the interplay between content, form and context that culminates in the practices of interpreting meanings from a given text.

Looking to the future Looking at the picture built so far, we might ask how far the feminist narratology envisaged back in the 1980s has achieved its goals. If Lanser’s (1986) paper was entitled as a gesture ‘Toward a feminist narratology’ (my emphasis), has it now arrived? Feminist narratology in both its earlier and postmodern forms has indeed contributed to the reworking of narrative theory and has proved to be an effective way of analyzing and interpreting a range of literary and non-literary texts. In addition, the influence of feminism within the field of narrative theory may be felt in other more diverse ways. Structuralist narratology has been critiqued as a ‘white male tradition’ (Fludernik, 2000a: 2), but as new developments emerge in postclassical narratology, it is notable that women are participating in their formation. In fact, Fludernik rightly points out that structuralist narratology had founding mothers as well as fathers (citing Kate Hamburger and Dorrit Cohn). To these female



The quest to find an easy symmetry between form and ideology should be abandoned, along with other comparably simplistic and undialectical claims made in the name of ideological purity or commitment. (2000b: 693)

narrative theorists, we might now add Fludernik herself, who has contributed to a linguistically oriented revision of narratology (1993, 1996b); Ryan’s (1991) and Ronen’s (1994) work on possible worlds; Emmott’s development of cognitive frames in narrative processing (1997); Murray’s work on hypertext (1997); Fleischman’s work on tense and temporality (1990), to name but a few influential examples. None of these (with the exception of Fludernik, 1999) has engaged directly with feminist concerns, yet their presence indicates a change in the body of scholars who are shaping the face of postclassical narratology. Likewise, while the wider developments in postclassical narratology go far beyond a feminist agenda, I no longer think that it is true that narratives by or about women are completely neglected as sample texts. For example, Herman’s (2002) work on storyworlds uses narratives by Edith Wharton. Toolan (2001) uses short stories by Katherine Mansfield in his influential introduction to narrative analysis. However, there is still some further work to be done here to accomplish a more balanced, multi-cultural corpus, especially by including authors who lie outside the existing canon. The increased recognition of feminist narratology’s status is reflected in its citation in authoritative works of reference. In Prince’s (1995b) entry ‘Narratology’ for The Cambridge History of Literary Criticism, feminist narratology gets only a passing mention (p. 128). Yet the Routledge Encyclopedia of Narrative Theory (2005) and the recent second edition of Elsevier’s Encyclopedia of Language and Linguistics (2006) contain separate, lengthy entries for the term. More generally, research literature surveying recent developments in narrative theory recognizes feminist narratology as a major (although not uncontentious) orientation (Darby, 2001; Fludernik, 2000a; Richardson, 2000b). Given these achievements, it would seem that we have indeed moved some way along the road ‘toward’ feminist narratology. The question then remains of how far feminist narratology is still a relevant concern for postclassical narratology and beyond. In the remaining sections of this chapter, I want to look ahead to some of the future directions that feminist narratology might take. The work that I have brought together in this book has widened the scope of feminist narratology in some measure, but this is only partial and it would be naïve to assume that it had exhausted the potential for further developments. The narratives I have looked at are not the only source texts available, nor is linguistics the only theoretical paradigm that might usefully combine with feminist narratology. One reason for the ongoing relevance of feminist narratology is the increasing range of data that has yet to be considered. Generally speaking, there would




seem to be few limits on the texts that could be analyzed. The concern to recognize the role of narrative in texts and disciplines that have not traditionally engaged with narratology is still increasing, with a diversity ranging from computerized databases to house sale details.2 However, certain areas might be particularly salient at this point in time. The first of these is the narratives that appear in newly emergent forms of media, such as computer games, virtual reality, digital storytelling, and to which we might add the more established transmedial possibilities of music (for example, in song lyrics), image and film. This is being signalled as an important new area of research for postclassical narratology (Ryan, 2004), and given the stereotypical gender associations that are related to some of these media, this would seem to be an important point of intersection for feminism with this transmedial narratology. A second area to interrogate is the narratives situated within domains from which the definitions of gender are derived. Bing and Bergvall (1998) suggest the importance of both the church and science in formulating the basis of gendered distinctions and relationships. The role of narrative in constructing these theological and scientific discourses may well be an important means by which we further scrutinize the ways in which we understand the term ‘gender’ itself. Finally, fields that have been conventionally dominated by one gender, but are undergoing change are worth further attention. It is within such sites of transition that the work of stories in reconstructing and challenging social roles may be felt in acute terms. Examples from professional contexts spring to mind, including the use of narratives within the law and medicine. Leisure pursuits may also be an interesting source of data, as in the commentaries of male-dominated sports which women are now participating in, such as football (Meân, 2001) or body building (Bing and Bergvall, 1998: 504).3 Feminist narratology might well find further useful partnerships with other disciplines and theoretical frameworks. Although the alliance that I have forged between linguistics and feminist narratology has been little pursued before, it is perhaps not an unsurprising one. The emergence of narratology itself has been identified as one of the results of the ‘linguistic turn’ within contemporary theory (Kreiswirth, 2000), and the practices of narratology rely heavily on linguistic evidence (Gymnich, 2002: 61). But the contributions of linguistics cannot account for all dimensions of narrative form and function. Certainly in the case of narratives which rely less heavily on language as a medium and instead employ other semiotic codes, there may be other paradigms that may be more useful when dealing with such texts and extra-textual factors




such as reader response. Some such interdisciplinary cross currents have already been hinted at in feminist narratology’s more recent work. Both Bal (1999) and Warhol (2003) argue for feminist narratology’s use within cultural studies. Indeed, there are obvious points of compatibility between narratological concerns such as focalization and theories of the gaze, spectatorship and audience that are of such relevance within the related subfield of feminist film theory. Toolan suggests that film is the ‘contemporary narrative genre with possibly the greatest cultural significance’ (2001: 103). As such, there is much more that feminist narratology might contribute to our understanding of the narrative forms and functions within films and other similar media, such as developments within television, soap opera and so on. Postmodern feminist narratology has stressed the need to understand gender as a concept inseparable from other influential variables. Dealing with the precise relationship between these interconnected variables may also benefit from drawing on other theoretical sources. For example, the relationship between gender and sexuality is highly contentious and has led to confusion within literary studies (Warhol, 2003: 2–3). If the need to disentangle these parameters has been a vexing problem within feminism and literary criticism, it is no less so for feminist narratology. It is here that queer theory may be of use, which itself is dominated by debates about normativity, gender and sexuality, but is crucial in helping the analyst ‘to see beyond the limitations of the homo/hetero opposition’ (Walters, 1996: 831). Gestures towards a queer narratology have already emerged, in particular Roof’s work (1996). These have been incorporated already in recent studies by Lanser (1999) and Warhol (2003), but given the dominance of this theoretical lens, I predict that its influence may continue for some time yet. A second, oft-cited combination is that of gender and race. Understanding the relationship between the two may be assisted by the insights of postcolonialism. Given that minority literatures remain an underused resource within feminist narratology, this kind of theoretical integration is pertinent and in line with recent developments in postclassical narratology (Doyle, 1994; Gymnich, 2002). These kinds of integrations are vital, but are not uncontentious and raise further questions about the relative status of gender within such discussions. For example, does one variable assume priority over the other? I have proposed that postmodern feminist narratology is characterized by diversity, including multiplicity along the lines of culture, ethnicity, sexuality and so on. Others have argued that these other factors are at least as important as gender (Bing and Bergvall, 1998: 498), in which case the primacy of gender as a ‘master’ status variable (Bergvall, 1999: 274) is





thrown open to debate. This in turn might have significant implications for the status of feminist narratology.

One set of implications relates to the kind of questions that postmodern feminist narratology might go on to ask. Bal points out that ‘to ask meaningful questions’ (1999: 22) is foundational to the future of narratology, be it feminist or otherwise. With that in mind, I return to Bing and Bergavall’s provocative review of the research questions traditionally asked in language and gender studies. They ask 1. Why are questions that strengthen the female–male dichotomy so frequently asked, while those that explore other types of variation evoke much less interest? 2. How much of this apparent dichotomy is imposed by the questions themselves? (1998: 497) The kinds of questions posed in early feminist narratology, and to which I have sought answers in this book, have often been of the type that have inadvertently reinforced both a false concept of gender difference and a binary categorization of ‘women’ and ‘men’. However, in the face of the evidence I have provided in many of the different analyses, the significant cross-gender similarities would seem to throw further doubt on the viability of exploring narratives from this point of view. But this does not mean that asking questions about the relationship between gender and narrative form should no longer be formulated. Rather, this serves as an important caveat not to perpetrate false dichotomies in the face of empirical evidence to the contrary. There may be good reason to draw attention to the anomaly rather than a wider trend, for example, in order to render marginalized individuals or groups more visible when they might be obscured or dismissed as intra-category ‘noise’. That said, it is salutary to notice quite how much interest is generated in the possibility of gender difference even when this is derived from relatively small points of contrast. This returns us to the first of Bing and Bergvall’s questions. A tentative answer might lie in the power of gender polarization and the inequalities of representation, experience and opportunity that continue to work against groups of women in different contexts. As many would contend, the cause of feminism is not over yet (Bryson, 1999). In this respect, questions about the female–male dichotomy need to be raised as a means of moving towards changing such situations. But in the case of



The question of questions

feminist narratology these need to be framed as questions not so much about how women and men might differ as to how stories might function to reinforce or challenge gendered inequalities. There are many more questions that feminist narratology might ask about narratives that come from literature and from other domains. These could take the work I have begun here in different directions. Critically, they could be useful in negotiating between localized perspectives and wider trends. Attention to the particulars generated by understanding gender as a diverse phenomenon can make it difficult to observe the ways that these might function within a bigger picture. It is this broader perspective that may be helpful in addressing the predominance of gender difference that Bing and Bergvall challenge (1998). As Bergvall questions in a later paper, ‘How can we explain the strong beliefs in binary sex/ gender that arise and are spread across wide stretches of communities and cultures? How do we talk about widely shared beliefs that do not seem to be globally discussed?’ (1999: 280). One way to begin this process is to do as Bergvall goes on to suggest and follow the practice of formal linguistics by formulating claims that can then be tested for their possible crosscultural applicability (p. 281). Postmodern feminist narratology can no longer talk in terms of binary opposition. Rather, questions about gender and narrative need to be framed in terms of ranges of possibilities that may be strategically drawn upon in specific contexts. With that in mind, feminist narratology might revisit some of the universal claims that have emerged from within sociolinguistic studies. One of these is the assertion that women are stylistically more flexible than men (Bergvall, 1999: 281). If Lanser began feminist narratology by asking questions about gender difference, we might now question whether gender makes a difference in the amount of stylistic variation individuals employ in their storytelling. This should be examined against a range of data that might strengthen the project of feminist narratology further still. For example, answering this question might entail a diachronic perspective. There is much more that needs to be known about the creative, experimental and conventional ways in which writers have used narratological forms of many different kinds. It is useful not simply to gauge how these vary across stories about or by writers who identify themselves in gendered ways but whether (or not) these have been theoretically associated with gendered values in earlier historical periods, and how this compares with recent discussions of gender and narrative form. Eckert and McConnell-Ginet predict that the questions and generalizations to be explored in feminist linguistics will go beyond differentiation of content. They write,





Developments in postmodern feminist narratology need to follow suit. It is not enough to identify degrees of stylistic variation in narratives. Although it is always vital not to confuse correlation and causality, feminist narratology should consider the functions these narratives perform in generating, sustaining or renegotiating gender identities and social relationships. Again, this raises the question of feminist narratology’s relationship with context. On the one hand, feminist narratology has always claimed to be contextualist. But this is more of an account of the assumed relationship between text and context, rather than an attempt to intervene in that context. The extent to which feminist narratology will, or even should, be able to function not just as an ideologically oriented theory but also practice is yet unresolved. This is an ongoing issue that remains fraught with debate within allied fields of critical linguistics and studies of language and gender. Fairclough closes his (1995) study with the searching but unanswered question, ‘What can be done about this text?’ (p. 205). Others have argued that exposing the ideological nature of representation (which might include those found in narratives) is an important first step towards that recognizing that ideological inequalities can (and perhaps should) be reconstructed (Cameron, 1998c). This kind of dialectical relationship between theory and practice is by no means straightforward. There are plenty of studies of sexist language that show that simply changing the terms of the text does not equate with changing social relationships between women and men, which will vary according to social context, individual preference and over time (Cameron, 1998c; McConnell-Ginet, 1998; Mills, 2003). Moreover, there is no guarantee that studies of gender and language use, be this in feminist narratology or elsewhere, might not be co-opted within popular culture for purposes to which feminists might object. Well-known examples of this include Tannen’s work on gender difference in speech styles (Cameron, 1996) and Lakoff’s descriptions of women’s language (Cameron, 1995; Bergvall, 1999). It is true that these examples indicate just how powerful the myths of gender difference can be and the undesirable ends to which such false assumptions can be put, even in the face of empirical evidence to the contrary. Still I would argue that feminist narratology does have something to offer beyond an academic



Whatever new generalizations about language and gender may ultimately be made, they will shift the focus – away from attributes and toward actions, strategies, and values; away from properties that women and men might have, and toward their social practices and social relations, both same sex and cross-sex. (1999: 198)

exercise in itself. The socially oriented approach to narrative analysis I have moved towards might well be useful to those working outside of academia. To touch on examples related to the studies I have covered in this book, those engaged with literacy practices in educational contexts or those working in the media might be prompted to reconsider the gendered consequences of narratives about, told and read by women and men, boys and girls, lesbian, gay, straight, intersexed and transgendered individuals. A future goal for feminist narratology might be to erode the boundaries between academic and non-academic studies and uses of narratives, integrating audiences and practitioners in order to better appreciate how narratives both shape our experiences and how, in turn, these might be recrafted with positive social ends.

A last word Feminist narratology in its initial and postmodern phases has accomplished a great deal, but the kinds of research that might be done and the questions that might be asked under the heading of this umbrella term are far from over. As new theoretical paradigms appear in postclassical narratology and yet more narrative texts are analyzed, gender needs to remain a vital concern in such practices. The ongoing work of maintaining gender’s place in such discussions (be that in terms of sample texts, further domains of application, or so on) remains relevant, for once its visibility is obscured then it becomes too easy for marginalization to take place and for the naturalization of an inevitably partial perspective to reassert itself. But the future work of feminist narratology is more than a form of political correctness. It stands in recognition of the continued and multiple ways in which gendered inequalities continue to exist for individuals and groups across the world (Bryson, 1999: 3–4). I close by returning again to Fraser and Nicholson’s tapestry metaphor (1990: 35), from which my multistranded description of postmodern feminist narratology has been derived. I have brought together just a few strands of literary and linguistic narrative analysis in this book. It is my hope that it points towards further opportunities for synthesis and integration where the body of postmodern feminist narratology may grow as strands of different colours, and textures are increasingly drawn in. The resulting cloth might go on to serve many purposes and tell the stories of feminist narratology for some time yet to come.




1

Feminist narratology in context

1. Feminist narratology is not alone in its emphasis on application rather than the formulation of narrative systems. Lanser (1999) and Bal (1997) note that the turn towards application is a widespread trend in narrative theory in recent years. 2. Hanssen’s (2001) insightful review of these debates signals the ways in which the opposition between ‘theory’ and ‘practice’ is in many ways a false one. However, because of the way in which these terms have been contested, the relationship between the two dimensions of feminist activity remains a controversial one which often remains polarized (Walters, 1996; Hartmann et al., 1996). 3. I do not wish to imply that academics cannot be or are not activists of any kinds. Rather, I am trying to indicate the broader tendency that has been seen in at least some quarters that isolates academic theory and dislocates it from practice. The case of Butler’s work on queer theory is a case in point. Whilst it emerged from Butler’s activist engagement with queer politics, it has subsequently been redeployed and critiqued for its apolitical tendencies. 4. The oral narratives (Chapter 4), readers’ responses (Chapter 5) and children’s stories (Chapter 7) have all been anonymized. Details of names and specific locations are altered in order to protect the identity of the individuals concerned. In all cases, consent has been given to reproduce the data here.

2

The question of gender and form

Chapter 2 is a revised version of R. E. Page, ‘Feminist narratology? Literary and linguistic perspectives on gender and narrativity’ in Language and Literature, 12 (1):43–56 (© Sage Publications, 2003), reprinted by permission of Sage Publications Ltd. 1. To avoid confusion, structural terms used in narratology have been capitalized so as to differentiate them from their everyday usage. I have followed this practice throughout the narrative analyses in all later chapters. 2. See Hirsch (1989) for a reading of this relationship from a psychoanalytic perspective. 3. See Collins (1990) and James (1993) on the distinctive practices and conception of motherhood in African-American contexts. 4. This polarization is an oversimplification of the discussion presented by Hoey. In fact, hierarchical arrangements of embedding between discourse types are possible, and there are ways in which colonies and mainstream texts may be read in similar ways. 5. I am grateful to Anne George who originally pointed this out to me. 6. This is the title of Michèle Roberts’ novel that succeeds During Mother’s Absence. 189



Notes

190

The question of gender and context

1. An exception to this is Warhol (2001) who writes about Alice Walker’s work. 2. For a recent example of this approach, see Merrill (1999). 3. All three of these romances may be found in Mills (1973). All quotations and line references are from this edition. 4. See for example, Volume 27, Chapter 41, ‘How the Fox of Koyagawa turned into a Woman and rode on Horses’ Croups’ in Times of Tales Past (Ury, 1979) 5. For examples of this in the Konjaku Monogatari see ‘How Kaya no Yoshifuji, of Bitchu Province, became the husband of a fox and was saved by Kannon’; ‘How the Fox of Koyagawa turned into a woman and rode on horses’ croups’; ‘About the Two wet-nurses in the House of Middle Captain Masamichi who looked exactly alike’. All of these are contained in Ury (1979).

4

The importance of similarities

Chapter 4 is reprinted by permission of Sage Publications Ltd from R. E. Page, ‘Evaluation in childbirth narratives told by women and men’ in Discourse Studies, 4 (1):99–116 (© Sage Publications, 2002). 1. Obviously the narratives considered here do not take account of all women and men, and the topic of childbirth should not be taken as a universally common denominator that defines gender. For many and various reasons both women and men may not participate in this experience. 2. Plum’s four-way division of narrative genres is restated in Eggins and Slade (1997). The definition and description of the Anecdote are taken directly from Plum’s work. 3. See Bruner (1997) on the importance and limitations of ‘trouble’ in Labov’s narratives. 4. See also Hay (2000) for a similar discussion on gender difference and humour. 5. I am grateful to Paul Simpson who originally suggested this connection to me. 6. To a certain extent this oversimplifies the complex dynamics of the individual relationship between speaker and listener. However, at a broad level, the cooperative motivation for storytelling, that is the need to maintain good social relations, would seem to supersede these individual differences. 7. In the UK, these include groups organized by the National Health Service, the National Childbirth Trust, Active Birth or informal friendship groups.

5

Questioning intra-category variation

1. See Bergvall (1996) on the shifting ways in which female engineering students negotiate discourse patterns in a male-dominated educational setting. 2. Although neither subject is in itself gendered, stereotypical associations of this sort continue to be influential. An example of this is found in the parent– child interactions discussed by Tenenbaum, et al. (2005). 3. The concept of ‘linearity’ is worth further consideration. Following Aarseth (1994), linearity may be understood as a property of the text as a material artefact. As such, Fishnet may be classed as a non-linear text because it ‘is not



3

Notes

191

simply one fixed sequence of letters, words, and sentences but one in which the words or sequence of words may differ from reading to reading because of the shape, conventions or mechanisms of the text’ (p. 51). While Aarseth sets to one side the ‘poetic unit’ or ‘meaning’ of the text, in addition to the physical characteristics of the text, linearity also relates to the interpretation of the text too, albeit in a rather more metaphorical sense. From this stance, the text is not just a physical collection of words but understood as a site of interaction between author, writer, audience and reader (Hoey, 2001). As such, the linearity of the physical text interacts with the connections that the reader makes between its constituents.

6

Media narratives of success and failure

Chapter 6 is a revised version of R. E. Page, ‘ “Cherie: lawyer, wife, mum”: contradictory patterns of representation in media reports of Cherie Booth/Blair’ in Discourse and Society, 14 (5):559–580 (© Sage Publications, 2003), reprinted by permission of Sage Publications Ltd. 1. The UK does not have a position equivalent to the ‘First Lady’ in the US. It is important to note that the wife of the prime minister is not expected to have a political role. Cherie Booth/Blair has repeatedly refused such a title, and has been chastised when the media has felt that she has overstepped the boundaries and influenced political matters in some way or other. Paradoxically, the media have continually used the term ‘First Lady’ to refer to the wife of prime minister, both in the case of Cherie Booth/Blair and the former prime minister’s wife, Norma Major. 2. This may be seen in the repeated reporting of the women as iconic figures. Examples of this include ‘Cherie Booth tops female icon poll’ (BBC News, 2 May 2000), and ‘The ivillage top 20 – Cherie Booth Blair’ (accessed 24 January 2002) which describes her as one of ‘the most powerful women in Britain today’. 3. Other examples include ‘The woman of substance who remains an enigma: The airbrushed image of Cherie Booth masks an extraordinary personality’ (Independent, 3 May 1997); ‘Cherie Blair; Joan Smith on the QC, working mother and “Spouse Girl” who won’t be pigeon-holed’ (Independent on Sunday, 4 May 1997); ‘50 most powerful women in Britain: Cherie Booth; She’s the so-called First Lady, mother, barrister – and under constant press surveillance’ (Guardian 25 May 1997); ‘Cherie – middle England’s new Queen of Hearts; She’s the ideal modern woman – and she sells papers’ (Independent, 2 November 1997). 4. The newspaper reports that are considered in detail are part of two larger text collections which also include reports from the period where the woman in question is not named in the headline. This constitutes a corpus of 2, 431, 993 words for Cherie Booth/Blair and a corpus of 3, 085 004 words for Hillary Rodham Clinton (counted using Monoconc version 1.1 (Barlow, 1996)). In order to assess whether the frequency of the naming choices in the headline is representative of these women more generally, the use of the surnames in the headlines was compared with that in the text collection as a whole. The result of this are summarized below.



Notes

Notes

Name Hillary Clinton Mrs Clinton Hillary Rodham Clinton Senator Clinton Cherie Blair Mrs Blair Cherie Booth Ms Booth

No. of occurrences 2630 1635 1335 66 2731 1813 1168 304

These figures indicate a picture broadly in line with that found in the headlines: that is, the use of the marital surname is far more frequent than the professional surname, or profession indicated through title, as in the case of ‘Senator’. 5. Other examples include: ‘BOA risk credibility over Cherie Blair’ (Daily Telegraph, 6 June 2003) No 10 aides fear Cherie ‘has lost way’ (Sunday Times, 4 May 2003) No win for Cherie (Sunday Times, 3 August 2003) It may be sour grapes, but Cherie can only lose by the bedroom photo-shoot (Daily Telegraph, 5 August 2003)

7

Gender, age and narrative development

1. New Zealand schools are ranked on a decile scale according to the student population’s demographic profile. Schools with a high proportion of students of low socioeconomic status and with a high percentage of multicultural students are deemed ‘low decile’ (scoring between 1 and 3). At the time of this study, the school from which the New Zealand stories are taken was ranked on scale 3. 2. I have laid out the children’s stories by clause in order to facilitate ease of reference. In all other respects, the writing remains that of the children. Spelling, punctuation and grammar were not corrected. 3. This is becoming disseminated more widely. Since taking part in the study, the Auckland school attended by these children has begun implementing reforms that gave greater value to the non-Pakeha students’ culturally distinctive literacy practices.

8

Beyond feminist narratology?

1. Recent examples include Fludernik (2001), Jahn (2001) on voice and focalization. 2. The role of narrative in these diverse examples was discussed at the fourth Media in Transition conference held at MIT, Cambridge, MA, 5–9 May 2005. 3. The contradictions exposed in challenging gender stereotypes in such situations has been popularized in recent mainstream films. The Adventures of Priscilla, Queen of the Desert (1994), The Full Monty (1997) and Billy Elliot (2001) all centre around crises of masculinity in relation to men who dance. Bend it Like Beckham (2002) follows the fortunes of a young British Asian girl who aspires to play football (soccer) professionally.



192

Aarseth, E. J. (1994). Nonlinearity and literary theory. In G. P. Landow (ed.) Hyper/text/theory, pp. 51–68. Baltimore: John Hopkins University Press. Abney, L. (1994). Gender difference in oral folklore. The SECOL Review, 18 (1): 62–79. Aldridge, M. and Wood, J. (1999). Telling it how it was: A comparative analysis of children’s evidential and non-evidential narrative accounts. Narrative Inquiry, 9 (2): 257–277. Amiran, E. (2000). After dynamic narratology. Style, 34 (2): 212–226. Anderson, A. S. (1995). Gendered pleasure, gendered plot: Defloration as climax in Clarissa and Memoirs of a Woman of Pleasure. Journal of Narrative Technique, 28 (2): 108–138. Anderson, K. V. (2002). Hillary Rodham Clinton as ‘Madonna’: The role of metaphor and oxymoron in image restoration. Women’s studies in Communication, 25 (1): 1–24. Arichi, M. (1999). Is it radical? Women’s right to keep their own surnames after marriage. Women’s Studies International Forum, 22 (4): 411–415. Astroff, R. J. and Nyberg, A. K. (1992). Discursive hierarchies and the construction of crisis in the news: A case study. Discourse and Society, 3 (1): 5–24. Bal, M. (1990). The point of narratology. Poetics Today, 11 (4): 727–753. ——. (1997). Narratology: Introduction to the Theory of Narrative, 2nd edition. Toronto, Buffalo, London: University of Toronto Press. ——. (1999). Close reading today: From narratology to cultural analysis. In W. Grünzweig and A. Solbach (eds) Transcending Boundaries: Narratology in Context, pp. 19–40. Tübingen: Gunter Narr Verlag Tübingen. Barlow, M. (1996). MonoConc for Windows, Version 1.1, Athelstan software. Barthes, R. (1977). Image-Music-Text, translated by Stephen Heath, London: Fontana. Barwood, G. (2001). Boys and reading: Clues to gendered social practice. New Zealand Journal of Educational Studies, 36 (1): 71–79. Bell, A. (1991). The Language of News Media. Oxford: Blackwell. Belsey, C. and Moore, J. (eds) (1989). The Feminist Reader: Essays in Gender and the Politics of Literary Criticism. Basingstoke: Macmillan. Bergvall, V. L. (1996). Constructing and enacting gender through discourse: Negotiating multiple roles as female engineering students. In V. L. Bergvall, J. M. Bing, and A. F. Freed (eds) Rethinking Language and Gender Research: Theory and Practice, pp. 173–201. London and New York: Longman. ——. (1999). Towards a comprehensive theory of language and gender. Language in Society, 28 (2): 273–293. Bing, J. M. and Bergvall, V. L. (1998). The question of questions: Beyond binary thinking. In J. Coates (ed.) Language and Gender: A Reader, pp. 495–510. Oxford: Blackwell. Blamires, A. (1997). The Case for Women in Medieval Culture. Oxford: Clarendon Press. 193



Bibliography

Bostdorff, D. M. (1998). Hillary Rodham Clinton and Elizabth Dole as running ‘mates’ in the 1996 campaign: Parallels in the rhetorical constraints of First Ladies and vice presidents. In R. E. Denton Jr (ed.) The 1996 Presidential Campaign: A Communicative Perspective, pp. 199–228. Westport, Connecticut and London: Praeger. Bradley, H. (1999). Gender and Power in the Workplace: Analysing the Impact of Economic Change. London: Macmillan. Brady Aschauer, A. (1999). Tinkering with technological skill: An examination of the gendered uses of technologies. Computers and Composition, 16: 17–23. Braidotti, R. (2001). Becoming -woman: Rethinking the positivity of difference. In E. Bronfen and M. Kavka (eds) Feminist Consequences: Theory for the new Century, pp. 381–413. New York: Columbia University Press. Bratton, M. (2002). Winterson, Bakhtin and the chronotope of a lesbian hero. Journal of Narrative Theory, 52 (2): 201–226. Brewer, M. (1984). A loosening of tongues: From narrative economy to women writing. Modern Language Notes, 99 (5): 1141–1161. Brock, D. (1993). The Real Anita Hill: The Untold Story. Oxford, New York: Free Press, Maxwell Macmillan. Bronfen, E. and Kavka, M. (eds) (2001). Feminist Consequences: Theory for the New Millenium, New York: Columbia University Press. Brooke-Rose, C. (1990). Whatever happened to narratology? Poetics Today, 11 (2): 283–293. Brooks, P. (1984). Reading for the Plot: Design and Intention in Narrative. Cambridge, London: Harvard University Press. Brown, K. (ed.) (2006). Encyclopedia of Language and Linguistics, 2nd edition. London: Elsevier. Bruner, J. (1997). Labov and Waletzky thirty years on. Journal of Narrative and Life History, 7 (1–4): 61–68. Bryson, V. (1999). Feminist Debates: Issues of Theory and Political Practice. Basingstoke and London: Macmillan. Burrell, B. C. (2000). Hillary Rodham Clinton as First Lady: The people’s perspective. The Social Science Journal, 37 (4): 529–546. Butler, J. (1990). Gender Trouble: Feminism and the Subversion of Identity. New York and London: Routledge. ——. (1993). Bodies that Matter: On the Discursive Limits of ‘Sex.’ London: Routledge. Bynum, C. W. (1991). Fragmentation and Redemption: Essays on Gender and the Human Body in Medieval Religion. New York: Zone Books. Caldas-Coulthard, C. R. (1995). Man in the news: The misrepresentation of women speaking in news-as-narrative discourse. In S. Mills (ed.) Language and Gender: Interdisciplinary Perspectives, pp. 226–239. London and New York: Longman. ——. (1996). ‘Women who pay for sex. And enjoy it’: transgression versus morality in women’s magazines. In C. R. Caldas-Coulthard and M. Coulthard (eds) Texts and Practices: Readings in Critical Discourse Analysis, pp. 250–270. London: Routledge. Calvino, I. (1997 [1974]). Invisible Cities. Trans. W. Weaver. London: Vintage Press. Cameron, D. (1995). Verbal Hygiene. London: Routledge. ——. (1996). The language-gender interface: Challenging co-optation. In V. L. Bergvall, J. M. Bing, and A. F. Freed (eds) Rethinking Language and Gender Research: Theory and Practice, pp. 31–53. Harlow: Longman.



194 Bibliography

——. (1998a). Gender, language and discourse: A review essay. Signs, 23 (4): 945–973. ——. (1998b). The Feminist Critique of Language, 2nd edition. London: Routledge. ——. (1998c). Lost in translation: Non-sexist language. In D. Cameron (ed.) The Feminist Critique of Language, 2nd edition, pp. 155–163. London: Routlegde. ——. (1998d). Performing gender identity: Young men’s talk and the construction of heterosexual masculinity. In J. Coates (ed.) Language and Gender: A Reader, pp. 270–284. Oxford: Blackwell. Cameron, D. and Kulick, D. (2003). Introduction: Language and desire in theory and practice. Language & Communication, 23: 93–105. Carter, C., Branston, G. and Allan, S. (eds) (1998). News, Gender and Power, London: Routledge. Case, A. (2001). Authenticity, convention, and Bridget Jones’s Diary. Narrative, 9 (2): 176–181. Caughie, P. L. (1988). Women reading/reading women: A review of some recent books on gender and reading. Papers in Language and Literature, 24: 317–335. Cazden, C. B. (1997). Speakers, listeners and speech events in issues of universality. Journal of Narrative and Life History, 7 (1–4): 185–188. Chatman, S. (1978). Story and Discourse: Narrative Structure in Fiction and Film. Ithaca and London: Cornell University Press. ——. (1990). What can we learn from contextualist narratology? Poetics Today, 11 (2): 309–328. Chodorow, N. (1989). Feminism and Psychoanalytic Theory. London: Yale. Cixous, H. (1989). Sorties: Out and out: attacks/ways out/forays. In C. Belsey and J. Moore (eds) The Feminist Reader: Essays in Gender and the Politics of Literary Criticism, pp. 101–116. London: Macmillan. Clark, C. L. (2002). Hypertext theory and the rhetoric of empowerment: A feminist alternative. Kairos, 7.3. Available from: http:english.ttu.edu/kairos/7.3/ binder2.html?coverweb/clark/page1.html. [Accessed 19 July 2004]. Clark, K. (1992). The linguistics of blame: Representations of women in the Sun’s reporting of crimes of sexual violence. In M. Toolan (ed.) Language, Text and Context, pp. 208–224. London: Routledge. Clayton, J. (1989). Narrative and theories of desire. Critical Inquiry, 16: 33–53. Coates, J. (1995). Language, gender and career. In S. Mills (ed.) Language and Gender: Interdisciplinary Perspectives, pp. 13–30. Harlow: Longman. ——. (1996). Women Talk. Oxford: Blackwell. ——. (2000). Small talk and subversion: The female speakers backstage. In J. Coupland (ed.) Small Talk, pp. 241–263. Harlow: Pearson. ——. (2001). ‘My mind is with you’: Story sequences in the talk of male friends. Narrative Inquiry 11 (1): 81–101. Collins, P. H. (1990) Black Feminist Thought. Knowledge, Consciousness and the Politics of Empowerment. London: Routledge. Coltrane, S. and Adams, M. (1997). Work-family imagery and gender stereotypes: Television and the reproduction of difference. Journal of Vocational Behaviour, 50: 323–347. Cook-Gumperz, J. and Green, J. L. (1984). A sense of story: Influences on children’s storytelling ability. In D. Tannen (ed.) Coherence in Spoken and Written Discourse, pp. 201–218. New Jersey: Ablex. Cosslett, T. (1996). Feminism, matrilinealism and the ‘House of Women’ in contemporary women’s fiction. Journal of Gender Studies, 5: 7–18.



Bibliography 195

Couturier, M. (1993). Nabakov in postmodern land. Critique, 34 (4): 247–260. Currie, M. (1998). Postmodern Narrative Theory. Basingstoke: Macmillan. Cutter, M. J. (2000). The story must go on and on: The fantastic, narration and intertextuality in Toni Morrison’s Beloved and Jazz. African American Review, 34 (1): 61–75. Daemmrich, I. G. (2003). Paradise and storytelling: Interconnecting gender, motif and narrative structure. Narrative, 11 (2): 213–233. Darby, D. (2001). Form and context: An essay in the history of narratology. Poetics Today, 22 (4): 829–852. ——. (2003). Form and context revisited. Poetics Today, 24 (3): 423–437. Davidson, M. J. and Cooper, C. L. (1992). Shattering the Glass Ceiling: The Woman Manager. London: Paul Chapman Publishing Inc. Davis, H. (1996). Theorising women’s and men’s language. Language & Communication, 61 (1): 71–79. De Lauretis, T. (1984). Alice Doesn’t: Feminism, Semiotics, Cinema. London: Macmillan. Delphy, C. (2000). The invention of French feminism: An essential move. Yale French Studies, 97: 166–197. Diengott, N. (1988). Narratology and feminism. Style, 22 (1): 42–51. Dinshaw, C. (1989). Chaucer’s Sexual Poetics. Wisconsin: University of Wisconsin Press. Douglas, J. Y. (1992). Print Pathways and Interactive Labyrinths: How Hypertext Narratives Affect the Act of Reading. Unpublished PhD thesis, New York University. Ann Arbor: UMI. ——. (1996). Virtual intimacy and the male gaze cubed. Sorry, we ran out of space – so it’s just a guy thing. Leonardo, 29 (3): 205–215. Doyle, L. (1994). Bordering on the Body: The Racial Matrix of Modern Fiction and Culture. New York: Oxford University Press. DuPlessis, R. B. (1985). Writing Beyond the Ending: Narrative Strategies of TwentiethCentury Women Writers. Bloomington: Indiana UP. Dutro, E. (2003). ‘Us boys like to read football and stuff’: Reading masculinities, performing boyhood. Journal of Literacy Research, 34 (4): 465–500. Eaton, J. H., Collis, G. M. and Lewis, V. A. (1999). Evaluative explanations in children’s narrative of a video sequence without dialogue. Journal of Child Language, 26: 699–720. Eckert, P. and McConnell-Ginet, S. (1998). Communities of Practice: Where language, gender and power all live. In J. Coates (ed.) Language and Gender: A Reader, pp. 484–494. Oxford: Blackwell. ——. (1999). New generalizations and explanations in language and gender research. Language in Society, 28 (2): 484–494. Eggins, S. and Slade, D. (1997). Analysing Casual Conversation. London: Cassell. Ehrlich, S. and King, R. (1998). Gender-based language reform and the social construction of meaning. In D. Cameron (ed.) The Feminist Critique of Language, 2nd edition, pp. 164–179. London: Routledge. Emmott, C. (1997). Narrative Comprehension: A Discourse Perspective. Oxford: Clarendon Press. ——. (2003). Narratology: Bridging the old and the new. Symploke, 11 (1–2): 227–232. Fairclough, N. (1995). Media Discourse. London: Edward Arnold.



196 Bibliography

——. (2000). New Labour, New Language? London: Routledge. Falcus, S. (2003). Her odyssey, herstory in Michèle Roberts’s Fair Exchange. Critique, 44 (3): 237–250. Fialho, O. (2004). Foregrounding and refamiliarization: Understanding readers’ response to literary texts. 24th Annual Conference of the Poetics and Linguistics Association, New York, 25–29 July 2004, unpublished conference paper. Fish, S. (1980). Is There a Text in This Class?: The Authority of Interpretive Communities. London, Cambridge, Mass.: Harvard University Press. Flax, J. (1990). Postmodernism and gender relations in feminist theory. In L. J. Nicholson (ed.) Feminism/Postmodernism, pp. 39–62. London: Routledge. Fleischman, S. (1990). Tense and Narrativity. From Medieval Performance to Modern Fiction. London: Routledge. ——. (1997). The ‘Labovian model’ with special consideration of literary narrative. Journal of Narrative and Life History, 7 (1–4): 159–168. Fludernik, M. (1993). The Fictions of Language and the Languages of Fiction: The Linguistic Representation of Speech and Consciousness. London: Routledge. ——. (1996a). Distorting language at its roots. Sprachkunst: Beitrage Zur Literaturwissenschaft, 27 (1): 109–125. ——. (1996b). Towards a ‘Natural’ Narratology. London: Routledge. ——. (1999). The genderization of narrative. GRAAT, 21: 153–175. ——. (2000a). Beyond structuralism in narratology: recent developments and new horizons in narrative. Anglistik, 11 (1): 83–96. ——. (2000b). Genres, text types, or discourse modes? Narrative modalities and generic categorization. Style, 34 (2): 274–292. ——. (2001). New wine in old bottles? Voice, focalization, and new writing. New Literary History, 32: 619–638. Fowler, R. (1991). Language in the News. London and New York: Routledge. ——. (1996). Linguistic Criticism, 2nd edition. OUP: Oxford. Fraser, N. and Nicholson, L. J. (1990). Social criticism without philosophy: An encounter between feminism and postmodernism. In L. J. Nicholson (ed.) Feminism/Postmodernism, pp. 19–38. London: Routledge. Freed, A. (1999). Communities of Practice and pregnant women: Is there a connection? Language in Society, 28 (2): 257–271. Gaggi, S. (1998). From Text to Hypertext. Decentering the Subject in Fiction, Film, the Visual Arts, and Electronic Media. Philadelphia: University of Pennsylvania Press. Gallop, J. (1987). Reading the mother tongue: Psychoanalytic feminist criticism. Critical Inquiry, 13 (1): 314–329. Garnsey, E. and Rees, B. (1996). Discourse and enactment: Gender inequality in text and context. Human Relations, 49 (8): 1041–1064. Genette, G. (1980). Narrative Discourse. Trans. J. Lewin. Ithaca: Cornell University Press. ——. (1990). Fictional narrative, factual narrative. Poetics Today, 11 (4): 755–774. Georgakopoulou, A. (1995). Women, men and conversational narrative performances: Aspects of gender in Greek storytelling. Anthropological Linguistics, 37 (4): 460–486. Gergen, K. J. (1998). Narrative, Moral Identity and Historical Consciousness: A Social Constructionist Account [online] Available from: http://www.swarthmore.edu/ SocSci/kgergen1/text3.html [Accessed 12 May 1998].



Bibliography 197

Gibson, A. (1996). Towards a Postmodern Theory of Narrative. Edinburgh: Edinburgh University Press. Giora, R. and Shen, Y. (1994). Degrees of narrativity and strategies of semantic reduction. Poetics Today, 22 (1): 447–458. Graham, W. (2001). Cyborgs or goddesses? Becoming divine in a cyberfeminist age. In E. Green and A. Adam (eds) Virtual Gender: Technology, Consumption and Identity, pp. 302–322. London and New York: Routledge. Green, N. (2001). Strange yet stylish headgear: Virtual reality consumption and the construction of gender. In E. Green and A. Adam (eds) Virtual Gender: Technology, Consumption and Identity, pp. 150–192. London and New York: Routledge. Greenhalgh, R., Slade, P. and Spiby, H. (2000). Fathers’ coping style, antenatal preparation and experiences of labor and the postpartum. Birth, 27 (3): 177–183. Greimas, A. J. (1966). Structural Semantics: An Attempt at a Method. London and Lincoln: University of Nebraska Press. Guerrina, R. (2001). Equality, difference and motherhood: The case for a feminist analysis of equal rights and maternity legislation. Journal of Gender Studies, 10 (1): 33–42. Guertin, C. (2004). Assemblage: The Women’s New Media Gallery [online] Available from: http://trace.ntu.ac.uk/traced/guertin/assemb_g-l.htm [Accessed July 2004]. Gymnich, M. (2002). Linguistics and narratology: The relevance of linguistic criteria to postcolonial narratology. In M. Gymnich, A. Nünning, and V. Nünning (eds) Literature and Linguistics: Approaches, Models and Applications: Studies in Honour of Jon Erickson, pp. 61–76. Trier: Wissenschaftlicher Verlag Trier. Gymnich, M., Nünning, A. and Nünning, V. (eds) (2002). Literature and Linguistics: Approaches, Models and Applications: Studies in Honour of Jon Erickson. Trier: Wissenschaftlicher Verlag Trier. Hakim, C. (2000). Work-Lifestyle Choices in the 21st Century: Preference Theory. Oxford: OUP. Halliday, M. A. K. (1994). An Introduction to Functional Grammar, 2nd edition. London: Edward Arnold. Halliwell, S. (1987). The Poetics of Aristotle. Chapel Hill: University of North Carolina Press. Hanssen, B. (2001). Whatever happened to feminist theory? In E. Bronfen and M. Kavka (eds) Feminist Consequences: Theory for the New Century, pp. 58–98. New York and Chichester: Columbia University Press. Haralambos, M. and Holborn, M. (1991). Sociology: Themes and Perspectives, 3rd edition. London: Collins. Haraway, D. J. (1991). Simians, Cyborgs and Women: The Reinvention of Nature. London: Free Association Books. Harding, S. and Norberg, K. (2002). New contexts, new issues. Signs, 27 (4): 943–946. Harkins, D. A. and Ray, S. (2004). An exploratory study of mother-child storytelling in East India and Northeast United States. Narrative Inquiry, 14 (2): 347–367. Hartmann, H., Bravo, E., Bunch, C., Harstsock, N., Spalter-Roth, R., Williams, L., and Blanco, M., (1996). Bringing together feminist theory and practice: A collective interview. Signs, 21 (4): 917–951. Hay, J. (2000). Functions of humour in the conversations of men and women. Journal of Pragmatics, 32: 709–742.



198 Bibliography

Heath, S. B. (1982). What no bedtime story means: Narrative skills at home and school. Language in Society, 11: 49–76. Herman, D. (1997). Scripts, sequences, and stories: Elements of a postclassical narratology. PMLA, 112 (5): 1046–1059. Herman, D. (2001). Story logic in conversational and literary narratives. Narrative, 9 (2): 131–137. ——. (2002). Story Logic. Problems and Possibilities of Narrative. Lincoln and London: University of Nebraska Press. ——. (2004). Toward a transmedial narratology. In M-L. Ryan (ed.) Narrative across Media. The Languages of Storytelling, pp. 47–75. Lincoln and London: University of Nebraska Press. Herman, D., Jahn, M. and Ryan, M-L. (eds) (2005). The Routledge Encyclopedia of Narrative Theory. London: Routledge. Hirsch, M. (1989). The Mother/Daughter Plot. Narrative, Psychoanalysis, Feminism. Bloomington and Indianapolis: Indiana University Press. Hoey, M. (1991). Patterns of Lexis in Text. Oxford: OUP. ——. (2001). Textual Interaction: An Introduction to Written Discourse Analysis. London: Routledge. Hohepa, M., McNaughton, S. and Jenkins, K. (1996). Maori pedagogies and the roles of the individual. New Zealand Journal of Educational Studies, 31 (1): 29–40. Holmes, J. (1997). Struggling beyond Labov and Waletzky. Journal of Narrative and Life History, 7 (1–4): 91–96. ——. (1998a). Narrative structure: Some contrasts between Maori and Pakeha storytelling. Multilingua, 17 (1): 25–57. ——. (1998b). Women’s talk: The question of sociolinguistic universals. In J. Coates (ed.) Language and Gender: A Reader, pp. 461–483. Oxford: Blackwell. ——. (2000). Politeness, power and provocation: How humour functions in the workplace. Discourse Studies, 2 (2): 159–185. Homans, M. (1994). Feminist fictions and feminist theories of narrative. Narrative, 2 (1): 3–16 Hopkins, A. (1990). The Sinful Knights. A Study of the Middle English Penitential Romance. Oxford: Clarendon Press. Hutcheon, L. (1996). Negotiating Postmodernism and Feminisms. In K. Mezei (ed.) Ambiguous Discourse: Feminist Narratology and British Women Writers, pp. 262–267. Chapel Hill: University of North Carolina Press. Ikeda, D. (1979). On the Japanese Classics: Conversations and Appreciations. Trans. B. Watson. New York and Tokyo: Weatherhill. Irigaray, L. (1981). This sex which is not one. In E. Marks and I. de Coutrivon (eds) New French Feminisms: An Anthology, pp. 99–106. New York and London: Harvester Wheatsheaf. Iser, W. (1978). The Act of Reading: A Theory of Aesthetic Response. London: Routledge and Kegan Paul. Jahn, M. (2001). Narrative voice and agency in drama: Aspects of a narratology of drama. New Literary History, 32: 659–679. James, S. M. (1993). Mothering: A possible Black feminist link to social transformation? In S. M. James and A. P. Busia (eds) Theorizing Black Feminisms. The Visionary Pragmatism of Black Women, pp. 44–54. London and New York: Routledge. Johnson, S. (1997). Theorising language and masculinity. In S. Johnson and U. H. Meinhof (eds) Language and Masculinity, pp. 8–26. Oxford: Blackwell.



Bibliography 199

Johnson, F. L. and Aries, E. J. (1998). The talk of women friends. In J. Coates (ed.) Language and Gender: A Reader, pp. 215–225. Oxford: Blackwell. Johnson, S. and Finlay, F. (1997). Do men gossip? An analysis of football talk on television. In S. Johnson and U. H. Meinhof (eds) Language and Masculinity, pp. 130–143. Oxford: Blackwell. Johnstone, B. (1990). Stories, Community and Place. Bloomington & Indianapolis: Indiana University Press. Kaplan, C. (2001). Hillary Rodham Clinton’s Orient: Cosmopolitan travel and global feminist subjects. Meridians: Feminism, Race, Transnationalism, 2 (1): 219–240. Kato, S. (1979). A History of Japanese Literature: The First Thousand Years, Trans. D. Chibbett. London and Basingstoke: Macmillan Press Ltd. Kavka, M. (2001). Introduction. In E. Bronfen and M. Kavka (eds) Feminist Consequences: Theory for the New Century, pp. ix–xxvi. New York: Columbia University Press. Kernan, K. T. (1977). Semantic and expressive elaboration in children’s narratives. In S. Ervin-Tripp and C. Mitchell-Kernan (eds) Child Discourse, pp. 91–102. New York, San Fransisco and London: Academic Press, Inc. Kindt, T. and Muller, H. (2003). Narratology and interpretation: A rejoinder to David Darby. Poetics Today, 24 (3): 414–421. Kreiswirth, M. (2000). Merely telling stories? Narrative and knowledge in the human sciences. Poetics Today, 21 (2): 293–318. Kuiper, K. (1998). Sporting formulae in New Zealand English: Two models of male solidarity. In J. Coates (ed.) Language and Gender: A Reader, pp. 285–293. Oxford: Blackwell. Labov, W. (1972). Language in the Inner City. Philadelphia: University of Pennsylvania Press. ——. (1997). Some further steps in narrative analysis. Journal of Narrative and Life History, 7 (1–4): 395–415. Labov, W. and Waletzky, J. (1967). Narrative analysis: Oral versions of personal experience. In J. Helm (ed.) Essays on the Verbal and Visual Arts: Proceedings of the 1966 Annual Spring Meeting of the American Ethnological Society, pp. 12–44. Seattle: University of Washington Press. Lambrou, M. (2005). Story Patterns in Oral Narratives: A Variationist Critique of Labov and Waletzky’s Model of Narrative Schema. Unpublished PhD thesis, Middlesex University. Landow, G. P. (1997) Hypertext 2.0, Revised and Amplified Edition. London, Blatimore: John Hopkins University Press. Lanser, S. (1986). Toward a feminist narratology. Style, 20 (3): 341–363. ——. (1988). Shifting the paradigm: Feminism and narratology. Style, 22 (1): 52–60. ——. (1995). Sexing the narrative: Propriety, desire and the engendering of narratology. Narrative, 3, 85–94. ——. (1999). Sexing narratology: Toward a gendered poetics of narrative voice. In W. Grünzweig and A. Solbach (eds) Transcending Boundaries: Narratology in Context, pp. 167–183. Tübingen: Gunter Narr Verlag Tübingen. Laurent, E. (2004). Bush’s Secret World. London: Polity Press. Leonard, M. (2001). Old wine in new bottles? Women working inside and outside the household. Women’s Studies International Forum, 24 (1): 67–78.



200 Bibliography

Liestol, G. (1994). Wittgenstein, Genette, and the reader’s narrative in hypertext. In G. P. Landow (ed.) Hyper/text/theory, pp. 87–120. Baltimore: John Hopkins University Press. Longacre, R. E. (1974). Narrative versus other discourse genre. In R. M. Brend (ed.) Advances inTagmemics, pp. 357–376. Amsterdam and London: North Holland Publishing Company. ——. (1983). The Grammar of Discourse. New York and London: Plenum. Loutfi, M. F. (ed.) (2001). Women, Gender and Work. Geneva: International Labour Office. Love, J. (2002). Elecriture: A course in women’s writing on the Web. Kairos, 7 (3). Available from: http://english.ttu.edu/Kairos/1.3/archival/3.html. [Accessed July 2004]. Luckhurst, R. (1996). ‘Impossible mourning’ in Toni Morrison’s Beloved and Michèle Roberts’s Daughters of the House. Critique, 37 (4): 243–260. McCabe, A. and Peterson, S. (1990). What makes a narrative memorable? Applied Psycholinguistics, 11: 73–82. McConnell-Ginet, S. (1998). The sexual (re)production of meaning: A discoursebased approach. In D. Cameron (ed.) The Feminist Critique of Language, 2nd edition, pp. 198–210. London: Routledge. McCormick, R. W. (1992). Politics and the psyche: Feminism, psychoanalysis and film theory. Signs, 18 (1): 173–187. McDonald, S. and McNaughton, S. (1999). Features of children’s storytelling on entry to school. New Zealand Journal of Education Studies, 34 (2): 349–354. McHale, B. (2001). Weak narrativity: The case of avant-garde narrative poetry. Narrative, 9 (2): 161–167. McNaughton, S., Phillips, G. and MacDonald, S. (2003). Profiling teaching and learning needs in beginning literacy instruction: The case of children in ‘low decile’ schools in New Zealand. Journal of Literacy Research, 35 (2): 703–730. McQuillan, M. (ed.) (2000). The Narrative Reader. London and New York: Routledge. Marley, C. N. (1987). Stylistic Implications of Discourse Analysis with Special Reference to R. E. Longacre’s Theory of Narrative Analysis. Unpublished MPhil thesis, University of Birmingham. Martin, B. (2001). Success and its failures. In E. Bronfen and M. Kavka (eds) Feminist Consequences: Theory for the New Century, pp. 353–380. New York: Columbia University Press. Martin, J. and Plum, G. A. (1997). Construing experience: Some story genres. Journal of Narrative and Life History, 7 (1–4): 299–308. Matus, J. (1998). Toni Morrison. Manchester: Manchester University Press. Meân, L. (2001). Identity and discursive practice: Doing gender on the football pitch. Discourse and Society, 12 (6): 781–815. Meinhof, U. H. (1997). ‘The most important event of my life!’ A comparison of male and female written narratives. In S. Johnson and U. H. Meinhof (eds) Language and Masculinity, pp. 208–228. Oxford: Blackwell. Merrill, R. (1999). Raymond Chandler’s plots and the concept of plot. Narrative, 7 (1): 3–21. Mezei, K. (ed.) (1996). Ambiguous Discourse: Feminist Narratology and British Women Writers. Chapel Hill and London: University of North Carolina Press.



Bibliography 201

Miall, D. S. (1999). Trivializing or liberating? The limitations of hypertext theorizing. Mosaic, 32 (2): 157–171. Michaels, S. (1981). ‘Sharing time’: Children’s narrative styles and differential access to literacy. Language in Society, 10: 423–442. ——. (1991). The dismantling of narrative. In A. McCabe and C. Peterson (eds) Developing Narrative Structure, pp. 303–351. Hillsdale New Jersey, Hove and London: Lawrence Erlbaum Associates. Mills, D. E. (1972). Medieval Japanese tales. Folklore, 83: 287–301. Mills, M. (ed.) (1973). Six Middle English Romances. Dent and Sons: London. Mills, S. (1995). Feminist Stylistics. London: Routledge. ——. (1998). Post feminist text analysis. Language and Literature, 7 (3): 235–253. ——. (2003). Caught between sexism, anti-sexism and ‘political correctness’: Feminist women’s negotiations with naming practices. Discourse and Society, 14: 87–110. Mitchell, J. (2001). Psychoanalysis and feminism at the end of the millennium. In E. Bronfen and M. Kavka (eds) Feminist Consequences: Theory for the New Century, pp. 3–17. New York and Chichester: Columbia University Press. Mori, A. (1999). Toni Morrison and Womanist Discourse. New York: Peter Lang. Morrison, T. (1971). What the Black Woman thinks about women’s lib. New York Times Magazine, 22 August, 1971. ——. (1987). Beloved. Picador: London. ——. (1989). Unspeakable things unspoken: The Afro-American Presence in American Literature. In H. Wonham (ed.) Criticism and the Color Line: Desegregating American Literary Studies, pp. 16–29. New Brunswick: Rutgers U.P. Moses, C. G. (2003). Made in America: ‘French feminism’ in academia. In R. Celestin, E. Dalmolin and I. de Coutrivon (eds) Beyond French Feminism, pp. 261–284. Palgrave: London. Mullany, L. (2000). The application of current language and gender theory to managerial meeting discourse. Nottingham Linguistic Circular, 15: 1–16. Mulvey, L. (1975). Visual pleasure and narrative cinema. Screen, 16 (3): 6–18. Muncer, S., Campbell, A., Jervis, V. and Lewis, R. (2001). ‘Ladettes,’ social representation and aggression. Sex Roles: A Journal of Research, 44 (1/2): 33–44. Murray, J. H. (1997). Hamlet on the Holodeck: The Future of Narrative in Cyberspace. Cambridge, Massachusetts: MIT Press. Nabokov, V. (1962). Pale Fire. London: Penguin. Neubert-Köpsel, I. (2001). Deconstructing duality: Utopian thought in the concept of ‘gender blending’ in Michèle Roberts’s ‘Flesh and Blood.’ In B. Neumeier (ed.) Engendering Realism and Postmodernism: Contemporary Women Writers in Britain, pp. 172–179. Amsterdam, New York: Rodopi. Nicholson, L. J. (1990). Feminism/Postmodernism. New York and London: Routledge. Odin, J. K. (1997). The edge of difference: Negotiations between the hypertextual and the postcolonial. Modern Fiction Studies, 43 (3): 598–630. Page, R. E. (2000). A Narrative Analysis of Three Middle English Romances. Unpublished PhD thesis, University of Birmingham. Pandey, R. (1995). Women, sexuality and enlightenment: Kankyo no Tomo. Monumenta Nipponica, 50 (3): 325–356. Parry-Giles, S. J. (2000). Mediating Hillary Rodham Clinton: Television news practices and image-making in the postmodern age. Critical Studies in Media Communication, 17 (2): 205–226.



202 Bibliography

Peterson, C. and McCabe, A. (1983). Developmental Psycholinguistics: Three Ways of Looking at Child’s Narrative. New York: Plenum Press. ——. (1991). Linking children’s connective use and narrative macrostructure. In A. McCabe and C. Peterson (eds) Developing Narrative Structure, pp. 29–53. Hillsdale, New Jersey, Hove and London: Lawrence Erlbaum Associates. Phillips, G., McNaughton, S. and MacDonald, S. (2004). Managing the mismatch: Enhancing early literacy progress for children with diverse language and cultural identities in mainstream urban schools in NZ. Journal of Educational Psychology, 96 (2): 309–323. Pilkington, J. (1998). Don’t try and make out that I’m nice! The different strategies women and men use when gossiping. In J. Coates (ed.) Language and Gender: A Reader, pp. 254–269. Oxford: Blackwell. Porter Abbott, H. (2000). What do we mean when we say ‘Narrative Literature?’ Looking for answers across disciplinary borders. Style, 34 (2): 260–273. Pratt, M. L. (1977). Toward a Speech Act Theory of Literary Discourse. Bloomington: Indiana University Press. Prince, G. (1982). Narratology: The Form and Function of Narrative. Berlin: Mouton. ——. (1995a). On narratology: Criteria, corpus, context. Narrative, 3: 73–84. ——. (1995b). Narratology. In R. Selden (ed.) The Cambridge History of Literary Criticism, Volume 8, From Formalism to Poststructuralism, pp. 110–130. Cambridge: Cambridge University Press. ——. (1996). Narratology, narratological criticism and gender. In C-A Mihailescu and W. Hamarneh (eds) Fiction Updated: Theories of Fictionality, Narratology and Poetics, pp. 159–164. Toronto and Buffalo: University of Toronto Press. ——. (1999). Revisiting Narrativity. In W. Grünzweig and A. Solbach (eds) Grenzüberschreitungen: Narratologie im Kontext, pp. 43–52. Tübingen: Gunter Narr Verlag Tübingen. ——. (2000). Forty-one questions on the nature of narrative. Style, 34 (2): 317–318. ——. (2001). A commentary: Constants and variables of narratology. Narrative, 9 (2): 230–233. Prochnow, J. E., Tunmer, W. E. and Chapman, J. W. (2001). A longitudinal study of early literacy achievement and gender. New Zealand Journal of Education Studies, 36 (2): 221–236. Propp, V. (1968 [1928]). The Morphology of the Folktale. Austin: University of Texas Press. Rampton, D. (1993). Vladimir Nabokov. Basingstoke and London: Macmillan. Richardson, B. (2000a). Linearity and its discontents: Rethinking narrative form and ideological valence. College English, 62 (6): 685–695. ——. (2000b). Recent concepts of narrative and the narratives of narrative theory. Style, 34 (2): 168–175. ——. (ed.) (2002). Narrative Dynamics: Essays on Time, Plot Closure and Frames. Columbus: Ohio State University Press. Ricoeur, P. (2000). Narrative time. In M. McQuillan (ed.) The Narrative Reader, pp. 255–261. London: Routledge. Rivkin, J. (1987). Resisting readers and reading effects: Some speculations on reading and gender. In J. Phelan (ed.) Narrative Poetics: Innovations, Limits, Challenges, pp. 11–22. Columbus: Ohio University Press. Roberts, M. (1993). During Mother’s Absence. London: Virago.



Bibliography 203

——. (1994). Post-script. In S. Roe et al., The Semi-Transparent Envelope, pp. 171–175. London: Marion Boyars. ——. (1995). Flesh and Blood. London: Virago. Roberts, L. D. and Parks, M. R. (2001). The social geography of gender-switching in virtual reality environments on the Internet. In E. Green and A. Adam (eds) Virtual Gender: Technology, Consumption and Identity, pp. 265–285. London: Routledge. Roe, S., Sellers, S., Ward Jouve, N. and Roberts, M. (eds) (1994). The Semi-Transparent Envelope, London: Marion Boyars. Rogers, R. (1982). Amazing reader in the labyrinth of literature. Poetics Today, 3 (2): 31–46. Ronen, R. (1990). Paradigm shift in plot models: An outline of the history of narratology. Poetics Today, 11 (4): 817–842. ——. (1994). Possible Worlds in Literary Theory. Cambridge: Cambridge University Press. Roof, J. (1996). Come as You Are: Sexuality and Narrative. New York: Columbia University Press. Rosen, M. (2001). Gender differences in reading performance on documents across countries. Reading and Writing: An Interdisciplinary Journal, 14: 1–38. Ross, K. and Sreberny-Mohammadi, A. (1997). Playing house – gender, politics and news media in Britain. Media, Culture and Society, 19 (1): 101–109. Rowland, S. (1999). Michèle Roberts’ virgins: Contesting gender in fictions, re-writing Jungian theory and Christian myth. Journal of Gender Studies, 8 (1): 35–42. Ryan, M. -L. (1991). Possible Worlds, Artificial Intelligence, and Narrative Theory. Bloomington: Indiana University Press. ——. (2001). Narrative as Virtual Reality: Immersion and Interactivity in Literature and Electronic Media. Baltimore and London: John Hopkins University Press. ——. (ed.) (2004). Narrative across Media: The Languages of Storytelling. Lincoln and London: University of Nebraska Press. Salih, S. and Butler, J. (2004). The Judith Butler Reader. Oxford: Blackwell. Scharrer, E. and Bissell, K. (2000). Overcoming traditional boundaries: The role of political activity in media coverage of First Ladies. Women and Politics, 21 (1): 55–83. Schiffrin, D. (1987). Discourse Markers. Cambridge: CUP. Scollon, R. and Scollon, S. B. K. (1984). Cooking it up and boiling it down: Abstracts in Athabaskan children’s story retellings. In D. Tannen (ed.) Coherence in Spoken and Written Discourse, pp. 173–200. Norwood, New Jersey: Ablex. Seidman Milburn, S., Carney, D. R. and Ramirez, A. M. (2001). Even in modern media, the picture is still the same: A content analysis of Clipart images. Sex Roles: A Journal of Research, 44 (5/6): 277–294. Selig, R. L. (2001). The endless reading of fiction: Stuart Moulthrop’s hypertext novel Victory Garden. Contemporary Literature 41 (4): 642–660. Showalter, E. (1978). A Literature of their Own: British Women Novelists from Bronte to Lessing. London: Virago. Simpson, P. (1997). Language through Literature. An Introduction. London and New York: Routledge. Skinner, B. L. (1995). Jazz Reading: Toni Morrison’s Erotics of the Text. Unpublished PhD Thesis, Howard University.



204 Bibliography

Sommer, R. (2004). Beyond (classical) narratology: new approaches to narrative theory. European Journal of English Studies, 8 (1): 3–11. Spender, D. (1989). Women and literary history. In C. Belsey and J. Moore (eds) The Feminist Reader: Essays in Gender and the Politics of Literary Criticism, pp. 21–34. Basingstoke and London: Macmillan. Springer, K. (2002). Third wave black feminism. Signs, 27 (4): 1059–1082. Stevens, A. (2003). Bilinguals as narrators: A comparison of bilingual and monolingual Hebrew and English narratives. Narrative Inquiry, 13 (1): 151–191. Sullivan, L. L. (1999). Wired women writing: Towards a feminist theorization of hypertext. Computers and Composition, 16: 25–54. Sundt, C. (2004). Fishnet. Available from: http://www.angelfire.com/ny5/fishnet/ index.html. [Accessed July 2004]. Tannen, D. (1990). You Just Don’t Understand: Women and Men in Conversation. New York: Ballantine. Tenenbaum, H. R., Snow, C. E., Roach, K. A. and Kurland, B. (2005). Talking and reading science: Longitudinal data on sex differences in mother-child conversations in low-income families. Journal of Applied Developmental Psychology, 26: 1–19. Thompson, J. B. (2000). Political Scandal: Power and Visibility in the Media Age. Cambridge: Polity. Todorov, T. (1977). The Poetics of Prose. Oxford: Blackwell. Tonomura, H. (1994). Black hair and red trousers: Gendering the flesh in Medieval Japan. The American Historical Review, 99: 129–154. Toolan, M. (1990). The Stylistics of Fiction: A Literary-linguistic approach. London: Routledge. ——. (1998). Language in Literature: An Introduction to Stylistics. London: Arnold. ——. (2001). Narrative: A Critical Linguistic Introduction, 2nd edition. London: Routledge. Tu’akoi, F. (2001). Fishing for Octopus. In School Journal, 2 (1): 8–15. Tuchman, G. (1996). Women’s depiction by the mass media. In H. Bauer and A. Gray (eds) Turning It On: A Reader in Women and Media, pp. 11–15. London, Arnold. Ury, M. (1979). Tales of Times Now Past: Sixty-Two Stories From a Medieval Japanese Collection. Berkely, Los Angeles and London: University of California Press. VanHoosier-Carey, K. A. (1997). Not Fitting the Framework: Gender Performativity and Narrative Experimentation by British Women Writers, 1880–1930. Unpublished PhD Dissertation, University of Texas at Austin. Van Zoonen, L. (1998). One of the girls? The changing gender of journalism. In C. Carter et al. (eds) News, Gender and Power, pp. 33–46. London: Routledge. Wallace, H. M. (2000). Desire and the female protagonist: A critique of feminist narrative theory. Style, 34 (2): 176–187. Walsh, C. (2001). Gender and Discourse: Language and Power in Politics, the Church and Organisations. Edinburgh and London: Pearson. Walters, S. D. (1996). From here to queer: Radical feminism, postmodernism and the lesbian menace (or why can’t a woman be more like a fag?). Signs 21 (4): 830–869. Wang, Q. and Leichtmen, M. D. (2000). Same beginnings, different stories: A comparison of American and Chinese children’s narratives. Child Development, 71 (5): 1329–1346.



Bibliography 205

Warhol, R. (1986). Toward a theory of the engaging narrator: Earnest interventions in Gaskell, Stowe and Eliot. PMLA, 101: 811–818. ——. (1989). Gendered Interventions: Narrative Discourse in the Victorian Novel. New Brunswick: Rutgers University Press. ——. (1999). Guilty cravings: What feminist narratology can do for cultural studies. In D. Herman (ed.) Narratologies: New Perspectives on Narrative Analysis, pp. 340–355. Columbus: Ohio State University Press. ——. (2001). How narration produces gender: Femininity as affect and effect in Alice Walker’s The Color Purple. Narrative, 9 (2): 182–187. ——. (2002). Queering the marriage plot: How serial form works in Maupin’s Tales of the City. In B. Richardson (ed.) Narrative Dynamics: Essays on Time, Plot, Closure and Frames, pp. 229–248. Columbus: Ohio State University Press. ——. (2003). Having a Good Cry: Effeminate Feelings and Pop-Culture Forms. Columbus: Ohio State University Press. White, P. (1997). Death, disruption and the moral order: The narrative impulse in mass-media hard news reporting. In F. Christie and J. R. Martin (eds) Genres and Institutions: Social Processes in the Workplace and School, pp. 101–133. London: Cassell. ——. (2000). Media objectivity and the rhetoric of news story structure. In E. Ventola (ed.) Discourse and Community: Doing Functional Linguistics, pp. 378–397. Tübingen: Gunter Narr Verlag. Wilkinson, I. A. G. (1998). Dealing with diversity: Achievement gaps in reading literacy among New Zealand students. Reading Research Quarterly, 33 (2): 144–167. Winfield, B. H. (1997). The making of an image: Hillary Rodham Clinton and American Journalists. Political Communication, 14: 241–253. Winnett, S. (1990). Coming unstrung: Women, men, narrative and principles of pleasure. PMLA, 105: 505–518. Winter, B. (1997). Misrepresentations: What French feminism isn’t. Women’s Studies International Forum, 20 (2): 211–224. Winter, E. (1994). Clause relations as information structure: Two basic text structures in English. In M. Coulthard (ed.) Advances in Written Text Analysis, pp. 46–68. London: Routledge. Wirth, L. (2001). Women in management: Closer to breaking through the glass ceiling? In M. Fetherolf Loutfi (ed.) Women, Gender and Work, pp. 239–249. Geneva: International Labour Office. Worthy, J., Moorman, M. and Turner, M. (1999). What Johnny likes to read is hard to find in school. Reading Research Quarterly, 34 (1): 12–27. Yaeger, P. (1988). Honey-mad Women: Emancipatory Strategies in Women’s Writing. New York: Columbia University Press.



206 Bibliography

Abstract, 30, 158, 159–60 academic disciplines, 10, 101, 102, 104, 114 age, 144–5, 149 Anecdote, 81–2, 90, 92, 150, 151–2, 154, 158, 175–6, 179–80 Bal, Mieke, 3, 13, 184 Barthes, Roland, 2, 154 birth stories, 76–7 black feminism, 32 Booth, Cherie, 18, 119–21, 122–3, 127, 131–3, 175, 178 Brooks, Peter, 20, 21, 22 Butler, Judith, 9, 15, 111, 112, 113, 116–17 Calvino, Italo, 179 Invisible Cities, 41–3 characterization in Beloved, 33 in Flesh and Blood, 28–9 in hypertext, 111–12 self characterization, 75–6 Cheriegate, 139, 140 children’s storytelling, 144–5 Cixous, Hélène, 8, 23, 98 close reading, 14 closure, 148, 149, 175 co-optation, 177, 187 Coda, 159–60, 166 Communities of Practice, 11, 91 contextualist narratology, 48, 73 contextualization, 2, 4, 17, 114, 117, 177 corpora, 4, 27, 46, 47, 74, 180, 182 Critical Discourse Analysis, 13, 14, 49, 118 cultural studies, 184 cyborg, 112, 177 desire, 20, 24 difference (as a feminist term), 6

discourse colony, 36, 40–1 diversity (within feminism), 6, 7 écriture feminine, 9, 23, 31, 39, 44, 97–8, 103, 171 Emare, 61–4 empirical approaches, 14, 17, 174 ethnicity, 146, 161, 171 eurocentrism, 15, 32, 47–8 Evaluation (as narrative component), 70, 77–81, 150, 154, 159, 160, 166 factual meaning, 89 see also informative meaning Fairclough, Norman, 118, 187 female plot, 22, 23, 26–7, 43 feminism, 2, 6–8, 9, 10, 32, 185 feminist linguistics, 40, 73, 181 see also language and gender studies feminist literary theory, 8 First Name only, 131, 133, 134–6 Fisher, Caitlin, 98 Fludernik, Monika, 95, 181, 182 focalization, 184 Fraser, N. and Nicholson, L. J., 16, 188 French feminism, 23, 44 Freud, 20 gender, 2, 12, 14, 47, 52, 102 gender blending, 28, 177 gender difference binary model of, 9, 18, 46, 92, 179, 180 criticism of, 75, 101, 116–17, 185–6 as sociolinguistic variable, 13, 89–90, 92, 145 in speech styles, 25, 46–7, 74–5, 82 Gergen, Kenneth, 121, 138, 162 Halliday, M. A. K., see transitivity Haraway, Donna, 112, 113 Hirsch, Marianne, 21, 22 Hoey, Michael, 17, 36, 52, 53, 55–6, 154 Homans, Maragaret, 32, 35, 48 207



Index

humour, 83–5 hypertext and gender, 97 as a medium, 104, 114, 115 narratives, 18, 96, 179 readers of, 96–7 informative meaning, 85 Irigaray, Luce, 23 Japanese monogatari, 57–8, 175 Kristeva, Julia, 23 Labov, William, 27, 73, 77, 78, 99, 108, 147, 148, 170, 174, 175, 180 language and gender studies, 16, 52, 187 see also feminist linguistics Lanser, Susan, 4, 17, 45–6, 47, 48, 50, 71, 94, 177, 181 lexical detail, 86, 87, 164 linearity, 33, 97, 98, 99, 101, 102, 179, 190 linguistics, 14, 183 literacy, 144, 145, 152, 154, 156, 167, 188 Longacre, Robert, 26, 28, 52–3, 56, 174 male plot of ambition, 21, 22, 24, 25, 26, 35, 43, 103 Maori boys, 162, 165, 170 culture, 168 discourse style, 167, 169 Material processes, 107, 108, 109 matrilinealism, 39 media texts, 117–18 Mental processes, 109 Middle English romances, 49, 56–7 Morrison, Toni, 179 Beloved, 32–4 Nabokov, Vladimir, 179 Pale Fire, 40 naming, 127–31 narrative of progression, 121, 138, 139 narrative of regression, 121

narrative of stability, 121, 138–9 narrative theory, 2, 8, 173–4 narratives of personal experience, 78, 147, 152 narrativity, 17, 26, 27, 99, 102, 107, 109–10, 174 definition of, 25 narratology definitions of, 3 histories of, 2 New Zealand, 145–6, 162 education system, 169 normativity, 180 oral narratives, 74 other directed reader, 106 overlexicalization, 88, 128 Pakeha, 164 girls, 162 Peak marking, 26, 53, 58–9, 66, 68, 70 performativity, 114, 116 personalization, 121, 126, 136, 139, 140, 142 plot, 24, 49 definitions of, 51 types, 179 politics of trust, 138 postclassical narratology, 45, 182 postcolonialism, 184 postmodern feminism, 6–8 postmodern gender theory, 95, 113, 143 postmodernism, 11, 43 poststructuralism, 9 Prince, Gerald, 3, 48 Problem-Solution pattern, 55, 56, 62, 68, 122, 154, 174 psychoanalysis, 20, 21, 34, 44 queer theory, 9,184, 189 reader response, 18, 94 reading pathway, 102, 104, 110–11 Recount, 150, 151, 152 Result, 149, 152, 154, 157, 175 retellings, 147, 152 revisionist narratology, 11



208 Index

Index 209

scandal, 139, 140 second wave feminism, 7, 46, 48, 171 sexist language, 119, 127–8, 187 sharing time, 152 social networks, 90 soft news reports, 119, 133 solidarity, 82, 90, 176, 180 stability, 63, 64, 139, 175 story genre, 81,149, 152, 174, 175 story topic, 149, 154 storyworlds, 107 structuralist approaches, 4, 51 structuralist narratology, 181 stylistics, 12, 73 Sundt, Charles, 100

symbolic capital, 136, 138, 139, 175 see also symbolic power symbolic power, 136, 140, 142 technology, 113 textual heterogeneity, 119, 126, 142 textual hybridity, 124–5 third wave feminism, 6 Tongan (nationality), 162 topic, see story topic transitivity, 107, 174 universalism, 6, 11, 32, 51, 71, 94, 117, 170, 173, 177, 186 virtual reality, 111, 113, 183 Warhol, Robyn, 1, 13, 95, 184 Washington Post, 134, 136 Washington Times, 134, 136 western pattern of narrative, 155, 160, 161, 162, 163, 165, 169, 180



Roberts, Michèle, 27–8, 178, 179 Flesh and Blood, 28–31 Une Glossaire, 35–40 Rodham Clinton, Hillary, 18, 119–21, 123–5, 131, 138–9, 175, 177, 178

Literary and Linguistic Approaches to Feminist Narratology

Literary Approaches to Biblical Interpretation

Cognitive and communicative approaches to linguistic analysis

Cognitive and Communicative Approaches to Linguistic Analysis

An Introduction to Narratology

Psyche and the Literary Muses: The contribution of literary content to scientific psychology (Linguistic Approaches to Literature)

Encyclopedia of feminist literary theory

Narratology Beyond Literary Criticism: Mediality and Disciplinarity (Narratologia)

Narratology Beyond Literary Criticism: Mediality and Disciplinarity (Narratologia - Volume 6)

Directions in Empirical Literary Studies: In Honor of Willie Van Peer (Linguistic Approaches to Literature)

A History of Feminist Literary Criticism

Reconfigured spheres: feminist explorations of literary space

Literary-Critical Approaches to the Bible. An Annotated Bibliography

A Dialogue of Voices: Feminist Literary Theory and Bakhtin

Medieval Marriage: Literary Approaches, 1100-1300

Language Teaching: Integrational Linguistic Approaches (Routledge Advances in Communication and Linguistic Theory)

Style and Social Identities: Alternative Approaches to Linguistic Heterogeneity (LPSP 18) (Language, Power and Social Process)

Style and Social Identities: Alternative Approaches to Linguistic Heterogeneity (LPSP 18) (Language, Power and Social Process)

Narrative Counselling: Social And Linguistic Processes Of Change (Discourse Approaches to Politics, Society and Culture)

Towards A Natural Narratology

Classification and Modeling with Linguistic Information Granules: Advanced Approaches to Linguistic Data Mining (Advanced Information Processing), 1st edition, 2004

Introduction to Literary Studies

Approaches to numerical relativity

Theoretical Approaches to Turbulence

Approaches to Intentionality

Approaches to Fundamental Physics

Approaches To Quantum Gravity

Critical Approaches to Literature

The Cambridge Companion to Feminist Literary Theory (Cambridge Companions to Literature)

Approaches to Paleoecology

Approaches to fundamental physics

Literary and Linguistic Approaches to Feminist Narratology

Literary Approaches to Biblical Interpretation

Cognitive and communicative approaches to linguistic analysis

Cognitive and Communicative Approaches to Linguistic Analysis

An Introduction to Narratology

Psyche and the Literary Muses: The contribution of literary content to scientific psychology (Linguistic Approaches to Literature)

Encyclopedia of feminist literary theory

Narratology Beyond Literary Criticism: Mediality and Disciplinarity (Narratologia)

Narratology Beyond Literary Criticism: Mediality and Disciplinarity (Narratologia - Volume 6)

Directions in Empirical Literary Studies: In Honor of Willie Van Peer (Linguistic Approaches to Literature)

A History of Feminist Literary Criticism

Reconfigured spheres: feminist explorations of literary space

Literary-Critical Approaches to the Bible. An Annotated Bibliography

A Dialogue of Voices: Feminist Literary Theory and Bakhtin

Medieval Marriage: Literary Approaches, 1100-1300

Language Teaching: Integrational Linguistic Approaches (Routledge Advances in Communication and Linguistic Theory)

Style and Social Identities: Alternative Approaches to Linguistic Heterogeneity (LPSP 18) (Language, Power and Social Process)

Style and Social Identities: Alternative Approaches to Linguistic Heterogeneity (LPSP 18) (Language, Power and Social Process)

Narrative Counselling: Social And Linguistic Processes Of Change (Discourse Approaches to Politics, Society and Culture)

Towards A Natural Narratology

Classification and Modeling with Linguistic Information Granules: Advanced Approaches to Linguistic Data Mining (Advanced Information Processing), 1st edition, 2004

Introduction to Literary Studies

Approaches to numerical relativity

Theoretical Approaches to Turbulence

Approaches to Intentionality

Approaches to Fundamental Physics

Approaches To Quantum Gravity

Critical Approaches to Literature

The Cambridge Companion to Feminist Literary Theory (Cambridge Companions to Literature)

Approaches to Paleoecology

Approaches to fundamental physics

Recommend Documents