(Routledge Interdisciplinary Perspectives on Literature) Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, Frans Mäyrä-Narrative Theory, Literature, And New Media_ Narrative Minds and Virtual Worlds-Routl (1)

Narrative Theory, Literature, and New Media

Offering an interdisciplinary approach to narrative, this book investigates storyworlds and minds in narratives across media, from literature to ­digital games and reality TV, from online sadomasochism to oral history ­databases, and from horror to hallucinations. It addresses two core questions of ­contemporary narrative theory, inspired by recent cognitive-scientific developments: what kind of a construction is a storyworld, and what kind of mental functioning can be embedded in it? Minds and worlds become essential facets of making sense and interpreting narratives as the book asks how story-internal minds relate to the mind external to the storyworld, that is, the mind processing the story. With essays from social scientists, literary scholars, linguists, and scholars from interactive media studies answering these topical questions, the collection brings diverse disciplines into dialogue, providing new openings for genuinely transdisciplinary narrative theory. The wideranging selection of materials analyzed in the book promotes knowledge on the latest forms of cultural and social meaning-making through narrative, necessary for navigating the contemporary, mediatized cultural landscape. The combination of theoretical reflection and empirical analysis makes this book an invaluable resource for scholars and advanced students in fields including literary studies, social sciences, art, media, and communication. Mari Hatavara is Professor of Finnish Literature at the University of ­Tampere, Finland. Matti Hyvärinen is Professor of Sociology at the University of Tampere, Finland. Maria Mäkelä is Senior Lecturer of Comparative Literature at the ­University of Tampere, Finland. Frans Mäyrä is Professor of Information Studies and Interactive Media at the University of Tampere, Finland.

Routledge Interdisciplinary Perspectives on Literature

  1 Mobile Narratives Travel, Migration, and Transculturation Edited by Eleftheria Arapoglou, Mónika Fodor, and Jopi Nyman   2 Shipwreck in Art and Literature Images and Interpretations from Antiquity to the Present Day Edited by Carl Thompson   3 Literature, Speech Disorders, and Disability Talking Normal Edited by Chris Eagle   4 The Unnameable Monster in Literature and Film Maria Beville   5 Cognition, Literature and History Edited by Mark J. Bruhn and Donald R. Wehrs

  9 Trauma in Contemporary Literature Narrative and Representation Edited by Marita Nadal and Mónica Calvo 10 Contemporary Trauma Narratives Liminality and the Ethics of Form Edited by Susana Onega and Jean-Michel Ganteau 11 The Future of Testimony Interdisciplinary Perspectives on Witnessing Edited by Jane Kilby and Antony Rowland 12 Literature and the Glocal City Reshaping the English Canadian Imaginary Edited by Ana María Fraile-Marcos

  6 Community and Culture in Post-Soviet Cuba Guillermina De Ferrari

13 Apocalyptic Discourse in Contemporary Culture Post-Millennial Perspectives of the End of the World Edited by Monica Germanà and Aris Mousoutzanis

  7 Class and the Making of American Literature Created Unequal Edited by Andrew Lawson

14 Rethinking Empathy through Literature Edited by Meghan Marie Hammond and Sue J. Kim

  8 Narrative Space and Time Representing Impossible Topologies in Literature Elana Gomel

15 Music and Identity in Postcolonial British SouthAsian Literature Christin Hoene

16 Representations of War, Migration, and Refugeehood Interdisciplinary Perspectives Edited by Daniel H. Rellstab and Christiane Schlote 17 Liminality and the Short Story Boundary Crossings in American, Canadian, and British Writing Edited by Jochen Achilles and Ina Bergmann 18 Asian American Literature and the Environment Edited by Lorna Fitzsimmons, Youngsuk Chae, and Bella Adams 19 Transnational Feminist Perspectives on Terror in Literature and Culture Basuli Deb 20 Children’s Literature, Domestication, and Social Foundation Narratives of Civilization and Wilderness Layla AbdelRahim 21 Singularity and Transnational Poetics Edited by Birgit Mara Kaiser 22 National Poetry, Empires and War David Aberbach 23 Technologies of the Gothic in Literature and Culture Technogothics Edited by Justin D. Edwards 24 Global Ecologies and the Environmental Humanities Postcolonial Approaches Edited by Elizabeth DeLoughrey, Jill Didur, and Anthony Carrigan

25 Theoretical Schools and Circles in the Twentieth-Century Humanities Literary Theory, History, Philosophy Edited by Marina Grishakova and Silvi Salupere 26 Gender, Race, and American Science Fiction Reflections on Fantastic Identities Jason Haslam 27 Space and the Postmodern Fantastic in Contemporary Literature The Architectural Void Patricia García 28 New Directions in 21st-Century Gothic The Gothic Compass Edited by Lorna Piatti-Farnell and Donna Lee Brien 29 Latin American and Iberian Perspectives on Literature and Medicine Edited by Patricia Novillo-Corvalán 30 Institutions of World Literature Writing, Translation, Markets Edited by Stefan Helgesson and Pieter Vermeulen 31 Narrative Theory, Literature, and New Media Narrative Minds and Virtual Worlds Edited by Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, and Frans Mäyrä

This page intentionally left blank

Narrative Theory, Literature, and New Media Narrative Minds and Virtual Worlds

Edited by Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, and Frans Mäyrä

First published 2016 by Routledge 711 Third Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2016 Taylor & Francis The right of the editor to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging in Publication Data Narrative theory, literature, and new media : narrative minds and virtual worlds / edited by Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, and Frans Mäyrä. pages cm. — (Routledge interdisciplinary perspectives on literature ; 48) Includes bibliographical references and index. (ebook) 1. Narration (Rhetoric) 2. Discourse analysis, Narrative. 3. Storytelling in mass media. 4. Literature and society. 5. Online authorship. I. Hatavara, Mari, editor. II. Hyvärinen, Matti, editor. III. Mäkelä, Maria, editor. IV. Mäyrä, Frans, editor. P96.N35N43 2015 302.2301’4—dc23 2015005158 isbn: 978-1-138-85414-7 (hbk) isbn: 978-1-315-72231-3 (ebk) Typeset in Sabon by codeMantra

Contents



List of Figures and Tables

xi

Introduction: Minds in Action, Interpretive Traditions in ­Interaction

1

M A R I H ATAVA RA, MATTI H YVÄ RIN E N , M A R I A M Ä K E L Ä , A N D FRAN S MÄYRÄ

SECTION I   1 Texts, Worlds, Stories: Narrative Worlds as Cognitive and Ontological Concept

11

M A R I E - L AU R E RYAN

  2 Storyworlds and Paradoxical Narration: Putting Classifications to a Transmedial Test

29

L I V I U L U TA S

 3 The Charge against Classical and Post-Classical Narratologies’ “Epistemic” Approach to Literary Fiction

50

G R E G E R A N D E RSSO N

SECTION II   4 How You Emerge from This Game Is up to You: Agency, ­Positioning, and Narrativity in The Mass Effect Trilogy

67

H A N N A - R I I K K A RO IN E

  5 Playing the Worlds of Prom Week B E N SA M U E L , DYL AN L E DE RL E - E N SIGN , MIKE T R EANOR , N OA H WA R D R IP - FRUIN , JO SH Mc COY, AA RO N R EED, A N D M I C H A E L MATE AS

87

viii Contents  6 Scripting Beloved Discomfort: Narratives, Fantasies, and Authenticity in Online Sadomasochism

106

J. TU O M A S H A RVIAIN E N

  7 Storyworld in Text-Messages: Sequentiality and Spatialisation

122

AG N I E S Z K A LYO N S

SECTION III   8 Defending the Private and the Unnarratable: Doomed Attempts to Read and Write Literary and Cinematic Minds in Marguerite Duras’s India Cycle

147

TY TTI R A N TAN E N

  9 Of Minds and Monsters: The Eventfulness of Monstrosity and the Poetics of Immersion in Horror Literature

167

G E RO B RÜ M ME R

10 Narrative Conventions in Hallucinatory Narratives

187

TO M M I K A K KO

11 Narrative and Minds in the Traditional Ballads of Early Country Music

205

A L A N PA L M E R

SECTION IV 12 Mind Reading, Mind Guessing, or Mental-State Attribution? The Puzzle of John Burnside’s A Summer of Drowning

223

M ATTI H Y VÄ R IN E N

13 Mind as World in the Reality Game Show Survivor

240

MARIA MÄKELÄ

14 Performing Selves and Audience Design: Interview Narratives on the Internet J A R M I L A M I L DO RF

256

Contents  ix 15 Documenting Everyday Life: Mind Representation in the Web Exhibition “A Finnish Winter Day”

278

M A R I H ATAVA RA



Afterword: A New Normal?

295

B R I A N M cH A L E

List of Contributors Index

305 307

This page intentionally left blank

List of Figures and Tables

FIGURES   1.1   1.2   1.3   2.1  2.2  2.3  5.1   5.2

  5.3   5.4

  7.1   7.2   7.3   7.4

The scope of storyworlds. Story-prominent vs. world-prominent narratives. A two-dimensional representation of Figure 1.2. Noise in Sagan om Karl-Bertil Jonssons julafton. © PennFilm Studio 2014. Sagan om Karl-Bertil Jonssons julafton. © PennFilm Studio 2014. Le tableau. © Blue Spirit Animation / Be Films / Blue Spirit Studio / Sinematik / France 3 Cinema / Rezo Productions / RTBF (télévision belge) / 2011. Prom Week. Oswald taunting Doug for one of his past actions, namely walking Jordan home after school. Learning this backstory between Doug and Jordan may help inform the player’s future playthroughs of the level. A player has successfully gotten Oswald closer to achieving his romantic goals by flirting with Nicholas. Play trace graph showing how often each distinct path through Simon’s story was traversed (shown by the number associated with each node, emphasized with color). The large band of nodes seen at the top of the diagram represents approximately one third of the total size of the complete graph. The cutout shows a section of the map in detail including examples of social exchanges (like “pickup line” and “confide in”) that appeared in more than one play trace. The majority of play traces are unique. Event sequence in (1). Event sequence in (2). Connectors and counterparts in mental spaces. Mental spaces in example (2).

14 23 25 38 38 43 88

90 96

99 126 132 134 136

xii  List of Figures and Tables 13.1 Russell Hantz giving a confessional in the twelfth episode of Heroes vs. Villains.243 13.2 Parvati Shallow giving the immunity idol to Jerri Manthey in the tenth episode of Heroes vs. Villains.251 13.3 Heroes Rupert Boneham, Amanda Kimmel, Colby Donaldson, and Candice Woodcock expressing their ­ disappointment due to Parvati Shallow’s blindside in the tenth episode of Heroes vs. Villains.252 15.1 Photograph 1. Lahti City Museum, Picture Archive. 282 Foto Tiina Rekola. 15.2 Photograph 2. Lahti City Museum, Picture Archive. 283 Foto Tiina Rekola. 15.3 Photograph 3. Lahti City Museum, Picture Archive. 283 Foto Tiina Rekola. TABLES   1.1 Evaluating narratives on three criteria.   2.1 Devices of paradoxical narration.   6.1 Typical, simplified narrative structure of a ­sadomasochist ­session. Note that massive variance exists between sessions and individual players, and that individuals’ internal ­narratives may deviate strongly from this template.   7.1 Narrative structure in (1).   7.2 Narrative structure in (2). 14.1 Excerpt 1: Interview with Rudy Autio. 14.2 Excerpt 2: Woman Caring for Her Partner. 14.3 Excerpt 3: Love Story.

26 37

117 127 133 264 267 270

Introduction Minds in Action, Interpretive Traditions in Interaction Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, and Frans Mäyrä Today’s interdisciplinary narrative studies are inspired by the broader scientific and philosophical trend of regarding minds and worlds to be ­ ­thoroughly interconnected. The constructivist take on this relationship perceives the mind to be the ultimate subject of meaning-making, while the world—as the mentally constructed model of a “universe”—is its object. However, the dominant post-Cartesian emphasis on the embodied and the distributed mind reminds us of the fact that the mind as an object of study can only emerge through interaction with its environment, be that the material world or other mental subjects. A mind in action is a mind emerging in dialogue with a world. However, both “mind” and “world” are such large-scale notions that specific narrative-related questions involving them must be formulated in order to establish their descriptive and methodological potential. The interest in minds and worlds comes together in the development of cognitive-theoretically informed narratology, which has taken great strides toward understanding the processes of immersion and readerly orientation within the storyworld and the perceptual positioning on the levels of storyworld, narration, and the actual reading process. However, potential affordances and—particularly—discrepancies between different media are areas of research that have been insufficiently covered by narrative studies. Furthermore, the theoretical and methodological exchange between the different fields of narrative studies has often remained superficial, largely neglecting the empirical basis and semiotic sensitivity fundamental to interand transdisciplinary research. In this volume, researchers from literary studies, social studies, language studies, and game studies discuss, apply, and test narratological theories of world and mind construction in diverse media, ranging from literature to digital games to reality TV, from online sadomasochism to oral history ­databases, and from horror to hallucinations. The broad range of disciplines allows us to investigate the roots of the overwhelming interest in the mind that is particularly persistent in literary studies. In social research, for ­example, the whole concept of the mind tends to be excluded because of the reluctance of discourse theorists to speculate on what happens “within people’s heads.” Literary theorists, for their part, do not face similar constraints: they do not need to intrude into anyone’s head but only to enter the worlds

2  Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, and Frans Mäyrä of literary texts, films, or games. The research materials of literary and media studies give these disciplines a highly privileged position in discussing and theorizing minds. For example, while reading Clarissa ­Dalloway’s thoughts, we are not “inside her head,” but within Virginia Woolf’s novel Mrs Dalloway. More than sixty-five years ago, the British language philosopher ­Gilbert Ryle (1949) made a valiant effort to banish the Cartesian body-mind ­dichotomy from scientific discourse, claiming that the role given to the mind was often nothing more than a series of category mistakes. For Ryle (1949, 49), any description of the workings of a person’s mind would only be a description of “the ways in which parts of his conduct are managed.” It is clear that the contemporary “storyworlds”—understood as mental and complex c­ onstructions that receivers of narrative texts presumably create— would not fit into Ryle’s narrow interpretation of the mental as an element of conduct management. Although Ryle correctly argues that “mind” and “body” ­cannot be understood as equal and independent systems, he eagerly subscribed to the theory of psychological behaviourism, which later became infamous for its incapacity and reluctance to thematize human sense-making, culture, or humans as meaning-making beings. According to Jerome Bruner (1990, 1), it was the early “Cognitive Revolution” of the 1950s that brought the mind back into the human sciences “after a long cold winter of objectivism.” Unfortunately, at the time Bruner wrote his words, there was little cause for celebration. The whole issue of the mind, in Bruner’s understanding, had already been technicalized and dehumanized by the emergent cognitive science. This new cognitive psychology engaged in a fierce conflict with the discursive and narrative versions of psychology, leading to a widespread rejection of the term “mind” in social research. Bruner’s work has remained one of the most-cited classics of narrative psychology, but the concept of the mind, let alone theorizing about mind-related issues, has been disregarded in social studies on narrative (see Clandinin 2007). In literary studies, the story has been entirely different. Dorrit Cohn’s Transparent Minds (1978) raised the issue of representing characters’ minds, making it one of the most enduringly debated issues in narratology. Of course, as later critics demonstrated, fictional minds are ultimately not transparent—if they are at all present—in fictional discourse. In this introduction, we do not need to reproduce the debate about whether certain modes of representing minds are distinctive to fiction, let alone to take a definite position on the matter (see, e.g. Cohn 1999; Palmer 2004, 2010; Herman 2011, 2013). For us, it is of greater importance to stress Cohn’s incisiveness in exhibiting and elaborating the vital historical role fictional literature has in producing minds that are textualized and represented and are thus made readable—at least in the concrete sense. Literary fiction has provided modern culture with experimental, possible, weird, dull, complex, and simple minds to be surveyed, encountered, recognized or simply discarded.

Introduction  3 Fictional minds offer discursive material relevant both to the theoretical discussion of minds and to analytical approaches to mind representation in new media. Furthermore, the conventions of mind representation in fictional and real-world environments interact: the interpretative skills and strategies exercised in fictional contexts are correspondingly utilized and evolved in encounters with minds outside fiction (cf. Herman 2011, 10). In the title of his recent book, David Herman (2013) tellingly uses the broad term “the sciences of mind,” importantly including the approaches of the philosophy and psychology, etc., of the mind. This is the broader mixture of disciplines we want to draw from in this volume in order to investigate how minds interact with and within worlds. The contributors of the book ask how generic and media-specific possibilities and limitations shape the represented worlds and the minds within them. Are there stories that rely more on “worldness” than others? What is included and what is excluded in textual or audio-visual representations of a world, and what motivates these representational choices? Why do fictional narratives so often blur the boundaries of allegedly separate ontological domains with metalepses? And how is it possible to create virtual worlds that are translatable from one narrative medium to another, as happens in contemporary transmedial storytelling able to turn imagined worlds into commercially successful franchises (Star Wars, Lord of the Rings, etc.)? This volume places more emphasis on interpretive than on ontological worlds: instead of asking the what of modal logic, we are more interested in the how and the why of literary and cultural studies and the social sciences. However, philosophy considered worldmaking a crucial interpretive activity well before the cognitive turn, with Kant being the most obvious authority for the claim against “unmediated” reality. For Nelson Goodman, the writer of the 1978 classic Ways of Worldmaking, worlds are, first and foremost, products of symbolic activity, of ordering, overlooking, and supplementation: “We are confined to ways of describing whatever is described. Our universe […] consists of these ways rather than of a world or of worlds.” (Goodman 1978, 3) Thus, following Goodman, Bruner (1987) also uses the concept of world to refer globally to our interpretive efforts in making sense of our environment. Yet if perception itself is a symbolic process, the construction of a world that is based on representation is symbolic to the second degree. Juxtaposed with this relativist and anti-realist critique of ready-made worlds, some recent narratological theories on embodied readerly immersion in storyworlds seem to emphasize the easy accessibility of textual universes. Representational variation across media and genres has been downplayed in favour of searching for the cognitive universals of experiencing represented worlds. According to David Herman, a leading theorist in the field, narrative worldmaking is a readerly process of mapping and framing that results in a mental model of the storyworld (see, e.g. Herman 2013, 103–43). There seems to be nothing symbolic—and thus no risk of misinterpretation—in

4  Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, and Frans Mäyrä the process where the reader simply experiences an embodied “deictic shift” (Herman 2002, 5, 14, 271) to the temporal and spatial coordinates of the textual world. As Marie-Laure Ryan, the other leading theorist in the field, poignantly notes in her self-reflexive critique of the aims and methods of cognitive narratology, a brain scan demonstrating that readers mentally simulate bodily actions is, after all, a circuitous way of introducing nothing else but a truism to the field of narratology (Ryan 2010, 472). Thus, the nexus of minds and storyworlds remains, at heart, an interpretive question for interdisciplinary narratologists. In this collection, true interdisciplinary exchange takes place not so much between the humanities and the hard sciences as between different interpretive traditions and methodologies. Games and interactive media are a good case in point. Game studies has already pursued an understanding of the form and experience of games and their playing for several decades, and this line of inquiry has led to its own debates and schisms, sometimes parallel to those of literary studies and similar disciplines, but often driven by the somewhat unique characteristics of games as media and performance. The “narrativist” and “ludologist” positions that emerged in late 1990s and early 2000s are defined by their difference regarding the storytelling potentials of digital games: whereas Janet Murray (1997), for example, is a believer in the future of cyberdrama and “constructivist stories,” Jesper Juul (1999) has argued that it is “the strength of the computer game that it doesn’t tell stories.” One of the most famous episodes in this debate was Markku Eskelinen’s critique of Murray’s reading of Tetris, which she interpreted as a having “dramatic content” as an enactment of the contemporary overtasked lives of Americans (Murray 1997, 144). Eskelinen calls this reading an “interpretative violence” and argues that rather than focusing on stories (the “uninteresting ornaments” or “marketing tools,” as he says), attention should be directed at the actual game mechanisms and spatio-temporal dynamics that are unique to games: while narratives require story and discourse times, games operate between user and event times (Eskelinen 2001). One does not need to agree with Eskelinen or this line of “radical ­ludology” (Klevjer 2002, 191–92) in order to see the importance of differentiating between configurative and interpretative modes, or between rules, goals, and (re)presentation when discussing storyworlds in a transmedial context. The typical gaming situation is driven by the player’s performance, which is aimed at solving or successfully accomplishing the challenges and puzzles presented by the game: the player must identify the available ingame elements that she can manipulate or otherwise interact with, and then determine what kind of gameplay dynamics the rules of games allow in this environment (cf. Järvinen 2008, 30–31). This, in turn, creates rather specific conditions for the design and experiencing of storyworlds that relate to games, and it also explains why such concepts as “fiction” or “event” carry rather specific meanings in the context of game studies. As MarieLaure Ryan (2009, 45) suggests, we need to differentiate “narrative games”

Introduction  5 at one end of the interactive fiction spectrum from “playable stories” at the other. The most reasonable approach is, perhaps, to place abstract games like ­Tetris outside of the scope of interactive fiction altogether. While we acknowledge and want to maintain that not everything is narrative, the borders are porous both conceptually and in relation to the objects of study. Therefore, in investigating narrative minds and virtual worlds, this book explores the limits of both narrative objects and narrative studies. The emphasis on theoretical reflection vis-à-vis close analysis of case studies varies from chapter to chapter, but the intention is to closely scrutinize the interpretive practices of humanist and social research and studies on ­interactive media. In this interdisciplinary dialogue, one of the greatest ­challenges is finding a sufficiently unified language to be shared by the different disciplinary traditions. For this reason, the use of such concepts as “fiction” varies from chapter to chapter. We have chosen not to supress the distinctiveness of each field of study; instead, we have preserved the ­original use of the concepts in each chapter. However, the chapters have cross-­references explaining the differences in conceptual frameworks to aid the reader in navigating through and across the disciplinary field(s). After this introduction, the book is divided into four sections, each highlighting a different facet of the main issues at hand, namely, intermedial and interactive mind and world construction in different narrative environments. Section I primarily comprises theoretical contributions that focus on the boundaries of texts, stories, and narrative levels. It offers comprehensive insights into readerly engagement with narrative-representational levels and their transgressions. Marie-Laure Ryan’s chapter elaborates on two trends in approaching narrative worlds, the cognitive approach, which regards the storyworld as a spatio-temporal totality whose evolution readers or spectators simulate mentally, and the ontological approach, which, inspired by possible worlds theory, focuses on the notion of the “fictional world,” and regards “worlds” as defined over sets of propositions with mutually compatible truth values. Ryan outlines several possible relations among text, world, and story: texts telling stories that spread over many worlds, texts presenting worlds that contain many stories, and worlds targeted by many different texts, often from different media. Liviu Lutas’s chapter investigates media differences in the use of narrative devices that transgress and distort the limits of narrative universes, putting metalepsis and syllepsis to the transmedial test. Lutas’s cinematic examples show how logocentrism has led to seemingly inappropriate classifications and conclusions that ought to be reworked in postclassical narratology. The final chapter of the first section, by Greger Andersson, comments on and evaluates the discussion of readers’ interaction with narrative fiction. Andersson analyses the roles allocated to lifelikeness and composability in narrative theory, while evaluating the “epistemic” and the “separatist” approaches to fiction, the previous regarding fictional narratives as a secondary variant of nonfictional narratives, the latter highlighting their distinctiveness.

6  Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, and Frans Mäyrä Section II brings together scholars working in diverse narrative environments to address the problem of world generation in their specific ways. Hanna-Riikka Roine starts the section by addressing the ever-prevalent issue of narrativity in games. Roine takes an intermediary stand between narratologists and ludologists by suggesting that the emergent narrativity of digital games does not stipulate narrative communication; in her interpretation of The Mass Effect Trilogy, she anchors the emergent narrative and its meanings in the player’s experience of acting as the role-playing character. Roine’s chapter is followed by another game studies contribution by Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, Noah Wardrip-Fruin, Josh McCoy, Aaron Reed, and Michael Mateas. In this chapter, the team discusses the interactive storytelling game Prom Week to demonstrate the vast potential for generating possible worlds in computer games, not just as mutually exclusive outcomes of gameplay but as actualized and meaningful versions of the same fictional world. Furthermore, Prom Week, a game developed by the authors themselves, is an experiment that ought to raise interest among social scientists, since it is based on a social artificial intelligence system inspired by Erwin Goffman’s “dramaturgical” approach to understanding social life. In the third chapter of Section II, J. Tuomas Harviainen analyses a narrative environment hitherto unexplored by narrative theorists: sadomasochist play. He focuses on the relationship between predefined scripts such as desires, restrictions and conventions, and actualized narratives. Harviainen’s approach is intermedial as he studies the effect of platform—which can be the physical world, chat messages, or a virtual world played with avatars— on the forming and transforming of scripts, as well as on the possible worlds emerging in the games. Agnieszka Lyons closes Section II with her pioneering study of text messages as potential narratives. In her linguistically oriented analysis, Lyons focuses on temporal sequences and storyworld generation, thus creating a theoretically illustrative juxtaposition between casual everyday communication and intentionally crafted stories. Section III focuses on mind representation and discusses minds that are determined by the genre and minds that are difficult to access or deny access. In addition, this section addresses the (in)capability of language and narrative to represent and to make sense of experience. Tytti Rantanen’s chapter opens the discussion on represented minds with a notoriously challenging example, ­ esisting Marguerite Duras’s India Cycle, which consists of both text and film. R the cognitivist assumption about the general accessibility of other people’s minds, Rantanen demonstrates how the failures in reading another mind can be built into an aesthetic principle that synthetizes Duras’s novelistic and cinematic expression. Gero Brümmer’s chapter approaches the problem of mind and experience from the generic viewpoint of horror literature. In Brümmer’s analysis, the motive of the indescribable monster proves to be a paradoxical vehicle for immersing the reader in experiencing the storyworld because the traumatic encounter is typically conveyed through narrative distancing. Tommi Kakko continues on the theme of unreadable minds and ineffable experiences

Introduction  7 in his chapter on the tradition of hallucinatory narratives, taking his examples from the nineteenth and twentieth centuries. Kakko d ­ emonstrates how drug literature’s radical experiential otherness can challenge the canonized narratological notions of naturalization and narrativization. The section closes with Alan Palmer’s chapter, which introduces a pioneering study on country and western music of American and British origin. Palmer’s analysis of song lyrics focuses on two types of narrative minds: the narrating instance or the protagonist as an experiencing individual, and the minds of other characters as they are attributed by these primary mental subjects. Section IV compares social and literary models for mind-reading, mind attribution, and mind construction, and their potential for accessing, portraying, or documenting other minds. The emphasis lies on the mutual exchange of analytical methods and theoretical insights between the study of fictional and documentary materials and between literary and social ­studies. Matti Hyvärinen’s chapter introduces a literary case study that both illustrates and challenges contemporary theories of mind attribution: the unreliable narrator of John Burnside’s A Summer of Drowning (2011) uses mind-reading in all its complexity to shield herself from actual, potentially difficult, and painful encounters with others. As Hyvärinen points out, this fictional novel challenges the prevailing optimistic notion of the social and the distributed mind as the optimal mental state that we would all benevolently strive for. Maria Mäkelä’s chapter tests the notion of mind attribution in the context of the reality game show Survivor. Reality TV is first and foremost about the performing individuals’ feelings and their feelings about others’ feelings, yet Mäkelä argues that the excess of expressivity in the confessional interviews—as well as the excess of speculation about other players’ mental states and intentions—creates generic distortion in this media laboratory that purports to reflect authentic social dynamics. The problem of the medium’s effect on the represented mind or experience is further addressed in Jarmila Mildorf’s chapter, which provides a genuinely interdisciplinary approach to conversational storytelling as recorded in oral history Internet databases. By using analytical methods from sociolinguistics, narratology, sociology, and psychology, Mildorf examines how the Internet as a medium shapes the narrative identity formation of the interviewee, the communicative situation of the interview, and the ways in which the speaker tries to engage the minds of the assumed audience. Mari Hatavara’s chapter probes the line between documentary and fictionalized experientiality in the web exhibition “A Finnish Winter Day.” Hatavara’s intermedial analysis of text and image reveals that the recording of everyday life by museum professionals uses techniques of mind representation that, in fact, fictionalize the documented everyday experiences. The use of fictionalized experientiality indeed gives the reader the feel of “what is it like” for the interviewees to go about their daily life, but at the same time it risks the clarity of intentions. In the final chapter, Brian McHale offers a commentary on the book’s theoretical and analytical contributions.

8  Mari Hatavara, Matti Hyvärinen, Maria Mäkelä, and Frans Mäyrä REFERENCES Bruner, Jerome. 1987. “Life as Narrative.” Social Research 54.1: 11–32. ———. 1990. Acts of Meaning. Cambridge: Harvard University Press. Clandinin, D. Jean, ed. 2007. Handbook of Narrative Inquiry. Thousand Oaks, ­London, New Delhi: Sage. Cohn, Dorrit. 1978. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton, NJ: Princeton University Press. ———. 1999. The Distinction of Fiction. Baltimore and London: The Johns Hopkins University Press. Eskelinen, Markku. 2001. “The Gaming Situation.” Game Studies 1.1. http://www. gamestudies.org/0101/eskelinen/. Goodman, Nelson. 1978. Ways of Worldmaking. Hassocks: The Harvester Press. Herman, David. 2002. Story Logic: Problems and Possibilities of Narrative. Lincoln: University of Nebraska Press. ———. 2011. “Introduction.” In The Emergence of Mind, edited by David Herman, 1–40. Lincoln: University of Nebraska Press. ———. 2013. Storytelling and the Sciences of Mind. Cambridge, MA: The MIT Press. Järvinen, Aki. 2008. Games without Frontiers: Theories and Methods for Game Studies and Design. Acta Electronica Universitatis Tamperensis 701. Tampere: University of Tampere. http://urn.fi/urn:isbn:978-951-44-7252-7. Juul, Jesper. 1999. “A Clash between Game and Narrative.” Copenhagen: ­Copenhagen University. http://www.jesperjuul.net/thesis/. Klevjer, Rune. 2002. “In Defense of Cutscenes.” In Proceedings of Computer Games and Digital Cultures Conference, edited by Frans Mäyrä, 191–202. Tampere: DiGRA & University of Tampere. http://www.digra.org/wp-content/uploads/ digital-library/05164.50328.pdf. Murray, Janet Horowitz. 1997. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. New York: Free Press. Palmer, Alan. 2004. Fictional Minds. Part of the Frontiers of Narrative series, edited by David Herman. Lincoln and London: University of Nebraska Press. ———. 2010. Social Minds in the Novel. Columbus: The Ohio State University Press. Ryle, Gilbert. 1949 [1966]. The Concept of Mind. Harmondsworth: Penguin Books. Ryan, Marie-Laure. 2009. “From Narrative Games to Playable Stories: Toward a Poetics of Interactive Narrative.” StoryWorlds: A Journal of Narrative Studies 1.1: 43–59. ———. 2010. “Narratology and Cognitive Science: A Problematic Relation.” Style 44.4: 469–95.

Section I

This page intentionally left blank

1 Texts, Worlds, Stories Narrative Worlds as Cognitive and Ontological Concept Marie-Laure Ryan

Every period in literary theory (or should I say in the humanities) has its favorite concepts. In the fifties and sixties, under the influence of Saussure’s linguistics, we witnessed a so-called “language turn” that inspired structuralism, semiotics, New Criticism, and deconstruction and placed the notion of text or textuality at the center of attention. With its emphasis on the signifier, at the expense of the signified, this movement regarded the literary text as the gate to a meaning that was absolutely unique to it; it assumed (more or less tacitly) that if you changed a single word, the entire meaning was changed. When the term world was used, the textualist school meant some kind of infinite sum of meanings that could not be paraphrased (a favourite battle cry of New Criticism was indeed: “The heresy of paraphrase”)1. It follows from these positions that the text was the only mode of access to its world. Because textualism is reluctant to isolate a narrative level of meaning from the global textual world, it implicitly adhered to a strict formula: 1 text—1 world—1 story. After the linguistic turn came the narrative turn of the eighties, and ­“narrative” or “story” became prominent. One of the effects of the narrative turn was a shift of focus from the signifier to the signified. While stories are transmitted by discourse, which means by text, they remain inscribed in our mind long after the signifiers have vanished from memory. This means that a story is a cognitive rather than a linguistic construct. The fact that stories can be summarized, adapted, and translated, and that they can be told by various media, emancipates them from language and makes them somewhat independent from the particular signs through which they are transmitted. The structuralist idea that Cinderella and a Chinese folk tale can be versions of the same story, which was heretic for textualism (Smith 1981), becomes very acceptable for a narratologist. Instead of 1 text, 1 world, 1 story, one could now have the possibility of many texts—1 world—1 story. As narratology has expanded from literature to other disciplines and media, we have seen the emergence of yet another theoretical concept, the concept of “world.” In earlier days, “world” was a totality of meanings associated with authors or with genres. Critics would speak of “the world of Proust” or “the world of Kafka” or even “the world of epic poetry,” meaning by this a distinctive set of values, themes, or objects of thought. In its new

12  Marie-Laure Ryan narratological use, “world” is no longer the world of an author or of a genre, but rather the world of a story—literally a “storyworld.” It combines a spatial dimension, the setting, and a temporal dimension, the narrative events. The new theoretical prominence of the concept of world further weakens the formula 1 text—1 world—1 story. Contemporary culture, whether popular or highbrow, practices an aesthetics of proliferation that implements the full range of possible relations among text, world, and story. This proliferation can take several forms. In this chapter, I will concentrate on the first two. • • •

Narrative proliferation: a world with many stories. Ontological proliferation: a story (or a text) with many worlds. Textual and medial proliferation: many different texts targeting the same world, especially texts of different media. This is the phenomenon currently known as transmedia storytelling (Jenkins 2006).

As the Czech narratologist Jirí Koten (2010) observes, the narratological concept of world can be traced back to two lines of ancestry. When we speak of storyworld the influence comes mainly from cognitive approaches to narrative (Herman 2009), while when we speak of fictional world the influence comes from schools and disciplines interested in the ontological status of imaginary entities: philosophy of language, formal semantics, and more particularly possible worlds theory (Pavel 1986; Doležel 1998; Ryan 1991). Yet the association between storyworld/cognitive approach and fictional world/ontological approach should not be taken in an exclusive sense, for storyworlds can raise ontological issues, and the recognition and evaluation of fictional worlds involve cognitive operations. The concern of the cognitive perspective is self-evident: It asks how the mind constructs stories and their world(s), either as an encoding, productive activity or as a decoding, interpretive activity. While the first of these questions has received in the past a lot of speculative attention by authors and philosophers interested in the nature of creativity, especially under the influence of Romanticism and its cult of genius, our far more empirically minded period has overwhelmingly focused on the second, because it is much more amenable to experimentation, or at least to self-examination. Few critics are creators, but all of them can ask: What goes on in my mind when I read (watch, play with) a narrative text? In contrast to the cognitive perspective, which focuses, at least in principle, on operations that every interpreter performs, the ontological perspective asks theoretical questions that go far beyond the concerns of ordinary people. These questions ­concern the nature, or mode of existence, of creations of the imagination. We can read/watch/play stories, especially fictional ones, without asking ourselves about the ontological status of the characters (or objects) that occupy our thoughts or about the relations between their world and the world we live in; yet insofar as stories can contain many worlds, and worlds with

Texts, Worlds, Stories  13 various existential modalities (material, imagined, dreamed, feared, desired, ­anticipated, etc.), the recognition of these ontological differences and of the borders that separate them (borders occasionally transgressed) is an integral part of the cognitive processing of stories. Ontological questions, therefore, can dovetail with cognitive ones, and they are not necessarily abstruse metaphysical issues raised for the pure pleasure of theoretical debate. STORYWORLDS The concept of world is intuitively very accessible. The nine definitions in the Oxford English Dictionary reveal two dominant themes: world as a planet (preferably the planet Earth, but there are also extra-terrestrial worlds), and world as a totality of things, as “everything that exists.” In this second sense “world” becomes synonymous with “universe.” Of these two conceptions the second is more useful to narratology, since a theoretical concept of world should apply equally well to a narrative of space travel, such as Star Wars, and a narrative that focuses on a small area, such as Flaubert’s Madame Bovary (1857). The totality conception of world is particularly useful in the case of storyworld. A storyworld is not just the spatial setting where a story takes place; it is a complex spatio-temporal totality that undergoes global changes. Put more simply, a storyworld is an imagined totality that evolves according to the events in the story. To follow a story means to simulate mentally the changes that take place in the storyworld, using the cues provided by the text. However, this rough definition leaves some questions unanswered. For instance: Does the concept of storyworld apply to all narratives or only to fictional ones? For nonfiction, could one simply say that the storyworld is the world of the text, this is to say, the real world? I believe that a distinction should be made between storyworld and reference world. A text of nonfiction describes the real world, but it may do so more or less accurately and always incompletely. Imagine that a text of nonfiction presents a distorted, false, or deliberately inaccurate version of reality—in other words, that it tells lies. In this case the audience may be capable of making a distinction between the world projected by the text—the storyworld—and the world that serves as referent. When a story is told as fiction, however, the storyworld cannot be distinguished from the reference world, since the story creates its own world. While in nonfiction the storyworld provides information that can be integrated into our representation of reality, at least if we believe it; in fiction, we construct the storyworld largely for its own sake. Another problematic issue concerns what kind of information belongs to the storyworld and what kind does not. Extending Gérard Genette’s typology of narrators (1972, 256), we can distinguish two types of narrative elements: intradiegetic elements, which exist within the storyworld, and extradiegetic elements, which are not literally part of the storyworld

14  Marie-Laure Ryan but play a crucial role in its presentation. Storyworlds are larger than what is directly shown in the text, larger than the narrative “here” and “now.” Let’s take the example of drama. It frames a certain time span, and it shows events that take place in a specific location, the location that occupies the stage, but the storyworld also includes events that precede the beginning of the action, as well as events that do not take place on stage but are narrated by the characters. The same distinction can occur in novels, for instance in Virginia Woolf’s Mrs. Dalloway (1925): The novel takes place over a single day, and its setting is London, but it references many events that precede that day and that take place elsewhere, for instance the traumatic World War I experience of Septimus Smith. In other words, storyworlds encompass not only the story per se, but also the backstory, and sometimes the afterstory (such as the later life of the protagonists, as represented in epilogues), and not only the scene of the story, but all the places that characters think or talk about. Storyworlds should therefore be divided into an inner circle occupied by the events that constitute the focus of the story and an outer circle that represents a larger spatial and temporal frame (Figure 1.1). Storyworld

Extradiegetic music Comments, evalutions, digressions Camera work

Speech bubbles and frames in comics

Back story

Actions and speech of characters

After story Third person narrator’s discourse

Remote locations First person narrator’s discourse

Figure 1.1  The scope of storyworlds.

Let’s now return to the difference between intradiegetic and extradiegetic elements. A good example of this difference is the sound track of movies. Film theorists have long been aware of the distinction between diegetic music—music that originates inside the storyworld and is perceived by

Texts, Worlds, Stories  15 the characters—and extradiegetic music, which controls the expectations and emotions of the spectator but does not exist within the storyworld. In drama, the objects on the stage are (normally) part of the storyworld, but the stage directions are not; they are rather instructions by the author on how to put together a storyworld. In literary narrative, the speech of characters is clearly part of the storyworld, but the status of the discourse of the narrator is more problematic. Here we can distinguish three possibilities. First, there is what Genette calls intradiegetic narrators (1972, 256), such as Emma Dean in Wuthering Heights (1847). These narrators are individuated characters who tell about their own life or about events they have witnessed. Their discourse is not the main support of the narration; rather, it is quoted by the main narrator. Since their act of telling is witnessed by other characters and may influence future events, it is clearly part of the storyworld. The discourse of these intradiegetic narrators is just an extended case of character speech. Another type of narrator that can be assimilated to the case of the intradiegetic narrator is the letter writers in epistolary novels. Their act of narration is clearly an event within the storyworld, since their letters can be intercepted by other characters and influence future events. Second, there is the case of regular first-person narrators—narrators who are individuated characters, who appear as characters in the story, and whose discourse is the main support of narration, such as Robinson Crusoe in the eponymous novel by Daniel Defoe. These narrators exist as individuals within the storyworld, but their discourse is not perceived by the other characters, and it has no effect on the evolution of the storyworld. This is why Genette (1972, 256) calls these narrators extradiegetic. Their act of narration is what makes it possible for the reader to imagine the storyworld, but it is not an event within the story, and most of the time it does not imitate a distinct nonfictional (“natural,” some narratologists would say) type of discourse, such as diary, testimony, or autobiography. Quite often in first-person narration we cannot tell if the narrator is speaking, writing, or just thinking. If it sounds paradoxical to regard standard first-person narrators as extradiegetic, compare their discourse with the camera in film. The camera in Saving Private Ryan (1998) is what makes it possible to see the landing on Omaha Beach, but we certainly do not imagine that there was a camera and a cameraman on the Omaha Beach of the storyworld. The paradox of a narrator who exists in the storyworld, while his discourse is not part of the story, can be resolved by the well-known narratological distinction between the experiencing I, the narrator as character, and the narrating I, the narrator as camera. Another way to handle this paradox is to regard the narrator’s discourse as ontologically part of the extended time frame, while the events represented by this discourse belong cognitively to the narrow time frame. The narrator, consequently, is situated at the outer edge of the outer circle. The third case concerns impersonal third-person narrators, who often narrate from an omniscient perspective. These narrators clearly do not exist

16  Marie-Laure Ryan in the storyworld, since they are not individuated, and if they do not exist as individuals, neither does their discourse. My personal inclination is to consider such narrators as disembodied entities whose function is that of a logical placeholder: They vouch for the (fictional) truth of their assertions; in fact they guarantee it, since they possess the highest degree of narratorial authority (Doležel 1998), but they do not exist as flesh and blood persons.2 The position I am defending stands halfway between the non-narrator theories of third-person narration, proposed by Ann Banfield (1982), Richard Walsh (2007, chapter 4) and Sylvie Patron (2009), and theories that regard these narrators as individuals with distinct genders and personalities and the supernatural ability to read other people’s minds and move freely within the storyworld, allowing a given object to be represented from various spatial points of view. I do not see how a narrator could both present individuating human properties and non-human abilities,3 but this paradoxical combination of features is entailed by the position of those critics who regard third-person omniscient narrators as embodied persons. Moreover, I do not believe that it is necessary to assign the same ontological status to the narrator throughout a text. A good example is the narrator of Madame Bovary, who starts out as a schoolmate of Charles Bovary, but then disintegrates into an impersonal omniscient narrator. All in all, the mode of existence of the third-person narrator is typical of the kind of ontological question that fascinates theorists (cf. the passionate debates raised in the eighties by ­Banfield’s proposal, recently revived, with equal passion, by Patron), but that ordinary readers do not have to consider. FICTIONAL WORLDS While the concept of storyworld transcends the distinction fiction/­nonfiction, the concept of fictional world is constituted by its difference from the real world, a difference that lies in its mode of existence, or ontological status. The main source of inspiration for capturing this ontological status has been the philosophical concept of possible world. For possible worlds theory (also known as modal logic), a world is defined over a set of mutually compatible propositions. One way to conceive of the mode of existence, or more precisely the coming-into-being of possible worlds, is to associate them with future states of the real world. Out of a common matrix of truth values that defines the world of the present, different future worlds can be created by changing the value of one or more propositions. In accordance with the central tenet of possible worlds theory—which claims that there can be only one actual or real world from a given point of view—one of these worlds will become actual, while the others will remain unrealised possibilities. This conception of possible worlds is fundamental to strategic planning, since the basis of rational action is the computation of the various states (=worlds) that can result from the planned action, as well as from other behaviours.

Texts, Worlds, Stories  17 Another explanation for the existence of possible worlds situates their origin in an act of the mind, such as imagining, dreaming, hallucinating, … or producing fictions. If one applies this conception of possible worlds to narrative fiction, fictional worlds will be created by the mind of authors for the benefit of audiences. Readers, spectators, or players relocate themselves in imagination into these worlds, pretending that they are actual (Ryan 1991). In the best cases, this game of pretense results in an experience of immersion in the fictional world. An issue that could receive different answers depending on whether one takes a cognitive or an ontological approach to fictional worlds is the question of what exists in them. Consider the minimal narrative proposed by E.M. Forster as an example of plot: “The king died, then the queen died of grief.”4 Does this story project a world that extends beyond the objects mentioned in the text, a world that contains other existents and events, or are the facts mentioned in the text all there is to the storyworld? In other words, is the narrative an incomplete description of a full storyworld, or is it a full description of an incomplete storyworld? For logicians, an incomplete object is an object that presents ontological gaps, which means that there are properties that this object neither has nor does not have, in violation of the principle of the excluded middle. One could represent an incomplete entity as a wheel of Swiss cheese: It has gaps that stand no chance of ever being filled, because they concern information that simply does not exist. From a cognitive point of view, the answer to the fullness of fictional worlds depends on the texts’ power of immersion. If a text manages to make a world present to the mind, inviting the user to imagine much more than its signs can describe or imply, then it projects a full world. What is not known about this world is not treated as ontological gaps, but as missing information. On the other hand, a text like “The king died” hinders immersion, because all it does is to ask readers to consider a set of propositions and their implications. While a full world can always yield more discoveries, a set of propositions is easily exhausted. We know all there is to know about “The king died” after a quick reading. No critic will write a long essay trying to interpret its world. A purely ontological approach to the question of worldness, by contrast, wants an objective answer valid for all fictional texts, rather than a variable answer relying on a criterion as subjective as power of immersion. (You may find a text immersive while I find it exceedingly boring.) Partisans of incompleteness (e.g., Doležel) will argue that since texts are finite, but the propositions that describe a full world infinite in number, fictional texts can never list all the properties of a world or of its inhabitants, and they cannot, consequently, create ontologically full worlds. On the other hand, partisans of completeness will argue that if the properties of Forster’s queen were those and only those specified by the text, she would be only a partial human being, able to experience love and grief over the loss of a loved one but no other emotion. The properties of being beautiful, nice, jealous, cruel,

18  Marie-Laure Ryan or of having had children would be like holes in the Swiss cheese of her being since the text says nothing about these features, and she would be a very strange creature indeed, one that we cannot really imagine. To avoid this counterintuitive view, fictional characters and fictional worlds could be regarded as intentional objects of thought, brought to the reader’s mind through a command to the imagination issued by the author. I doubt that it was Forster’s intent to have the reader imagine a creature who is, on one hand, a human being but on the other hand lacks determination on most of the properties normally associated with human beings. Since ­Forster’s text differs from realistic types of fiction in the number of objects it asks readers to imagine and in the quantity of information that it provides about these objects but not in the quality of this information, the ontological question must receive the same answer for Forster’s story and for War and Peace (Voina i mir, 1869). The choices are: (1) All fictions that project a world project a complete world, because worlds are by definition complete; (2) fictions do not project worlds, but rather, assert a certain number of propositions on the basis of which readers form incomplete mental representations of pseudo-existents; (3) fictions project “small worlds” (Eco 1990) consisting of relations among a limited number of existents, but within their perimeters, fictional worlds are imagined as ontologically complete. Applied to Forster’s story this last solution (to which I am partial) yields: There is a world with a king and a queen who must be imagined on the model of real-world royalties, but there is nothing more in this world than the scenes described in the story. This solution ascribes a unified ontological status to all fictional worlds—they are all small, compared to the real world, but since their size varies with the amount of information provided by the text, they differ cognitively. WORLD AS COGNITIVE VS. WORLD AS ONTOLOGICAL CONCEPT To illustrate the distinction between a cognitive and an ontological conception of world, I propose to look at two fictions (or rather three). The first two are the novel Cloud Atlas (2004, by David Mitchell), which was made into a film in 2012, directed by Lana and Andy Wachowski and Tom Tykwer. Both the novel and the film consist of six separate stories, which take place at different times and in different locations. The first one concerns the journey of an American lawyer sailing from the South Pacific back to California in 1850; the second one is about a young musician who writes down scores for a famous composer in the 1930s in Belgium (though the film sets the story in England); the third is set in California in the 1970s, and tells about a young reporter who investigates the attempts by a nuclear power company to cover up the dangerous flaws of the plant it is building; the fourth follows the misadventures of an elderly, eccentric book publisher who is committed

Texts, Worlds, Stories  19 against his will to a nursing home in contemporary England; the fifth tells about a dystopic future society in Seoul, South Korea, where human beings can be cloned, and the clones are used as slaves; and the sixth, set in Hawaii in a very distant future, depicts how mankind has regressed to a primitive state after a mysterious event called The Fall. From an ontological point of view, Cloud Atlas projects a (nearly) unified world. The six stories take place in different places, at different times, and involve different characters, but they do not represent mutually exclusive possibilities. Even though they are not linked to each other by relations of causality, we can imagine that the stories correspond to various moments in the history of the same global world, strung together like the beads of a necklace. The only exception is story 3, which is revealed in story 4 to be a novel and not an account of real events,5 but when we read it for the first time, we take it as factual account. Indeed, the kind of events that it reports could very well happen in the same world as the other stories. The ontological connection of the stories is hinted at by the dominant themes of the narrative: the repeated claim that “everything is connected” and the presence of an identical birthmark on the shoulder of the main character of each story, which suggests that these characters are reincarnations of the same individual, despite their widely different personalities. In the movie, the theme of reincarnation is reinforced by the fact that the same actors play different roles in different stories, a device that would not be possible in a novel. The unity of the text as a whole is further established by a system of embedding that locates each story as a material object within the next story. For instance, the text of the first story is the diary of a character named Adam Ewing. The hero of the second story, Robert Frobisher, discovers and reads this manuscript. Frobisher composes a musical work titled “Cloud Atlas,” and writes a series of letters to his lover. Both of these media objects fall into the hands of Luisa Rey, the reporter of the third story. The Luisa Rey novels are read in story four. The memoir written by Timothy Cavendish in story five is filmed, and viewed by Sonmi-451, the heroine of story five. The confession she makes before being executed becomes the sacred text of the religion of story six. The fact that story three is a novel and consequently a fiction within a fiction rather than a representation of the same world as the other stories creates a breach in this neatly symmetrical pattern. If Luisa Rey is a character in a novel, how could she meet Sixtus Sixsmith, the lover of the now dead Robert Frobisher, both of whom belong to the fictionally real world, and how could she read in story three the letters addressed by Frobisher to Sixsmith in story two? Two answers can be proposed: (1) Frobisher and Sixsmith have counterparts in the fictional world of Luisa Rey, and the letters that Luisa discovers belong to these counterparts; or (2) thanks to a collapse of ontological boundaries, the fictional Luisa Rey is able to communicate with real individuals. This collapse would be a case of metalepsis (defined below). Given the self-reflexive, postmodern slant of the whole novel, the metaleptic interpretation is the more satisfactory solution,

20  Marie-Laure Ryan though it creates a logically impossible situation. On the other hand, the fact that Timothy Cavendish, the protagonist of story four, has access to the Luisa Rey novels is not a paradox, since real people do read novels that transport them into fictional worlds. Characters of a lower ontological level have epistemic access to worlds of higher levels through books and stories (though they can only enter these worlds in imagination), but characters of a higher level have no knowledge of the worlds of a lower level. While from an ontological perspective all but one of the stories of Cloud Atlas belong to the same world, from a cognitive perspective each story projects its own storyworld. When the readers or spectators pass from one story to the next, they experience a world where nothing is familiar: neither setting, nor characters, nor social environment, and they must construct the storyworld from an almost blank state. The defamiliarization that takes place with every new story explains why readers, upon encountering story three, do not think of it as a novel. The organization of the text on the discourse level does little to alleviate the cognitive burden of constructing six different storyworlds. In the novel, the stories are divided into two parts (except for the sixth story), and these parts are presented in the sequence 1-2-3-4-5-6-5-4-3-2-1.This pattern actualizes a structure known in computer science as a stack (Ryan 1991): the various elements are piled upon each other, and they are processed according to the principle “first in, last out.” The stack principle means that when readers reach level six, they must keep five half-told stories in the back of their minds. Only story six unfolds as an uninterrupted whole. Once story six is completed, the text returns to story five, which is still reasonably fresh in memory. But as the reader is sent back to older levels, it becomes more and more difficult to remember what the story was all about. It is fortunate that the medium of the book allows readers to return to earlier pages and to refresh their memory. The spectators of the movie do not have that luxury. In the film the symmetrical stack structure is replaced with a chaotic organization. The stories are fragmented into many more elements than in the novel, and these fragments, which tend to become shorter and shorter as the film progresses, are presented in a seemingly random order. For spectators who see the film without having read the novel, and this was my case, it is very difficult to reconstitute the plot. When I left the theatre I was totally confused, and the first thing I did when I got home was to look up the Wikipedia article to make sense of the film. While the various storyworlds of Cloud Atlas differ cognitively, but are logically compossible and could therefore be part of the same global world, we find the reverse situation in Tom Tykwer’s earlier film, Run, Lola, Run (1998). The film represents a genre that David Bordwell (2002) calls ­“forking path” narratives. These narratives focus on a decision point, out of which several different futures develop depending on the character’s deliberate choice of action or on random coincidence. In Run Lola Run the decision point is a phone call to Lola from her boyfriend, Manni, who has lost a

Texts, Worlds, Stories  21 large sum of money he owes to a crime boss. He will face dire consequences if the money is not delivered within 20 minutes. The film explores three forking paths in which Lola tries different courses of action to get the money in time. In the first “run” Lola tries to borrow the money from her father, a banker, but he refuses. Then she helps Manni rob a supermarket, but she gets shot and apparently dies. In the second run she robs her father’s bank, but Manni is hit by a car as he runs toward her to get the money. In the third run, Lola wins the money at the casino, but in the meantime Manni has recovered the money he lost, so everything is for the best in the best of all possible worlds. The three worlds that fork out of the common decision point are clearly different from an ontological point of view, since they contain incompatible events, but in all three branches, Manni has the same problem, Lola has the same goal, the setting is constant, and the network of interpersonal relations remains unchanged. As the clock is rewound and a different alternative is explored, the spectator is taken back to a familiar situation, and no additional cognitive effort needs to be devoted to the construction of the background. OTHER TYPES OF PROLIFERATION Cloud Atlas and Run, Lola, Run illustrate two basic forms of proliferation: a world that includes many stories for Cloud Atlas and a story (or text) that includes many worlds for Run, Lola, Run. The case of a world with many stories is found in many different genres and media: for instance, in TV soap operas, which represent the interleaved destinies of many characters and follow multiple plot lines; in novels of magical realism, which often consist of many little stories taking place in the same setting rather than of a unified narrative arc; or in a film like Babel (2006), which presents three different stories, one located in Mexico, another in Morocco, and the third in Japan. The spectator knows that these stories take place in the same world because they present common elements. Another example of a world with more than one story comes from a structure that may be called non-ontological narrative embedding, i.e. the embedding of a story that refers to the same world as the framing story and extends its representation, rather than transporting the reader into a new world. For instance, in “Sarrasine,” the short story by Balzac that was made famous by Roland Barthes’ S/Z (1970), the narrative begins with the description of a lavish reception in the Parisian house of a rich family. Among the guests is a withered old man who awakens the curiosity of a marquise. The narrator tells the story of the old man to the marquise in exchange for a night of love—which he does not get in the end, because the marquise is too upset by the tale to keep her promise. Since the embedded and embedding stories refer to the same world, they complement each other,

22  Marie-Laure Ryan and passing from one to the other does not require the crossing of an ontological boundary. In contrast to “Sarrasine,” works like The Canterbury Tales, The Decameron or The Arabian Nights are not worlds with many stories, but rather texts with many worlds. These texts feature a framing story and many embedded ones, told by the characters of the framing story. Insofar as the embedded stories are presented as fictions, they do not refer to the same world as the framing story. For instance, the characters in Ali Baba and the 40 Thieves or Aladdin and the Magic Lamp are not part of the world where Scheherazade tells stories to the Sultan to postpone her execution, and there is no chance that Scheherazade could meet Aladdin, except in a postmodern parody. These examples illustrate the case of ontological proliferation: a text that sends its readers into many other worlds than the primary fictional world, where the embedding story takes place. (See Ryan 1991, chapter 9 on the two types of embedding.) In most examples of ontological proliferation, the ontological borders of the worlds are respected. By this I mean that these worlds do not bleed into each other, that their casts of characters remain distinct, and that the events of the embedded worlds have no direct effect on the embedding world. (They can of course have an indirect effect, as when Scheherazade is saved by the spellbinding stories she tells the Sultan.) But some narratives engage in a ­playful transgression of ontological boundaries. This is known as metalepsis (Pier 2011), and it is a common phenomenon in postmodernism. ­Metalepsis creates ontological paradoxes by staging interactions between a character from the world that the fiction presents as real and a character from a representation that exists within this world. A prototypical example of metalepsis is Julío Cortázar’s short story “Continuity of Parks” (1967), in which a reader is murdered by the character of the book he is reading. Through metalepsis, the worlds of embedded stories become part of the embedding storyworld. Another way for stories to entangle several ontologically distinct worlds is by creating an alternative to the cosmology of the standard version of possible worlds theory. Here I am thinking of narratives that rely on the socalled many-worlds cosmology, also known as multiverse. While in the standard cosmology only one of many possibilities can be realised, and only one world can be actualized, in the many-worlds cosmology, all possibilities are realised, and there are countless parallel actual worlds that come into being whenever certain conditions are met. Some physicists have endorsed this cosmological model to explain the weird behavior of subatomic particles, such as the fact that they seem to be in several places at the same time. The many-worlds cosmology has been enthusiastically adopted by science fiction writers, and there is a number of novels based on the idea of a multiverse where characters have counterparts in other worlds (Ryan 2006). To turn this cosmology into a unified plot characters are made to travel to parallel worlds and meet their counterparts, which can lead to some rather intriguing narrative situations.

Texts, Worlds, Stories  23 But the most common way for a story to encompass several worlds is when its basic reality is split into ontologically distinct sub-realities, such as the realm of the gods vs. the realm of the humans in Homer, England vs. Narnia is C.S. Lewis’ Narnia chronicles, or, in The Matrix (1999), the “free” world of Neo and Morpheus vs. the virtual reality created by the machines. WORLD VS. STORY AS SOURCES OF INTEREST While worlds are spaces that contain stories, stories are mental constructs that imply a world. Yet texts may be more or less “world centered” and “story-centered,” and there is a full continuum of intermediary forms between these two poles (Figure 1.2). Prominence of world

Prominence of story

“The king died, then the queen died of grief”

Jokes Tragedy

Fantasy Science fiction

Micronations: Talossa, Bergonia, etc.

Figure 1.2  Story-prominent vs. world-prominent narratives.

From a strictly logical point of view, the case of a story that does not project a world is very problematic. The core constituents of narrative are events, but there cannot be events without existents, and since existents are objects with spatial extension, they must exist somewhere. This means that there must be a world that contains them. But this world may be left largely implicit. Consider again “The king died, then the queen died of grief.” If this is a story, it offers little to the imagination. Readers will register the information that something happened in some abstract fictional world, but they will not be tempted to visualize the scene and to fill in the blanks in the story. A narrative genre that minimizes world-creation is jokes. Not only are jokes too short to display a rich storyworld, the same joke can be told about different kinds of people. For instance, jokes that used to be told of certain ethnic groups are now told about blondes (assumed to be stupid), and jokes about lawyers are recycled by musicians into conductor jokes.

24  Marie-Laure Ryan The fact that the butt of certain jokes can be easily transformed from one category of people to another demonstrates that the appeal of these jokes lies in some properties of the story that transcend the particular embodiment of the characters. Another narrative genre that privileges plot over world is tragedy. The genre downplays particular social circumstances, to focus on a network of personal relations that could happen anywhere, anytime. This is why Greek tragedy is often performed on a bare stage with no distracting props. The case of world without story is much more feasible that the case of stories without worlds. A good example is the phenomenon of the micronation. The Internet contains many imaginary countries created for the pure pleasure of playing God. They have names like Bergonia and Talossa, and they are brought into being by documents that represent an encyclopedic sum of knowledge. The creators of these micro-nations can play as many roles as they want: ethnographer, geographer, political scientist, linguist, cartographer, historian, and climatologist. But one role they do not play is that of novelist. For this reason, visiting these countries is like reading all the descriptions in a novel and skipping the action parts. While jokes and tragedy come the closest to the story pole, science fiction and fantasy are the closest genres to the world pole. In these genres, the plot serves as a trail that takes the audience through the storyworld and provides a glimpse into its distinctive natural features and cultural institutions. As noted science-fiction author Philip K. Dick said in an interview: “The first thing is the idea. A pure idea. The next thing is characters who will be confronted with the environment based on that idea. … In other words I translate an idea into a world. Then you need people who live in that world” (Dick 1984). Or, as Henry Jenkins writes on his blog: “It’s long been a charge directed against science fiction works that they are more interested in mapping complex environments than in telling compelling stories. Many of my favorite SF novels—Snow Crash, for example—break down into near incoherence by the end, yet they offer us richly realised worlds that I would love to be able to explore in greater detail than any one narrative allows.” The greater the distance of a fictional world from ordinary reality, the more the interest of the reader or spectator will be directed toward the world, at the expense of the plot, because the invention of a world that differs from reality is a true feat of the imagination that rivals the creative power of God (Wolf 2012). It does not take a very original plot to make a narrative successful when the world attracts a lot of attention. But the one-dimensional schema of Figure 1.2 is misleading, because it suggests that the more prominent the world, the less interesting the plot, and vice-versa, the more indeterminate the world, the more interesting the plot. This is certainly not the case, as we can see from the example of “The king died, then the queen died of grief.” Here there is hardly any world, but this lack of worldness does not add to the appeal of the plot. The story is very boring. On the other hand, there are great works of literature—here

Texts, Worlds, Stories  25 I think especially of the great novels of the nineteenth century—where plot and world are both very developed and none takes second seat to the other. To represent this situation we need a two-dimensional diagram, where the y axis represents “worldness” and the x axis “plotness,” or “tellability.” On this diagram, “The king died” will be low on both counts. Figure 1.3 transposes Figure 1.2 in two dimensions. + Fantasy, Science fiction

Interest In world

War and Peace

Micronations

Tragedy Jokes

The king died, etc. _ _ Interest in plot

+

Figure 1.3  A two-dimensional representation of Figure 1.2.

But even this grid does not do justice to how audiences evaluate narratives, because it lacks a third dimension, the interest taken in the medium itself, whether it is attention to the form and style of the writing in languagebased narrative, to the work of the camera and the play of actors in film, or to the quality of the art and the arrangement of the frames on a page in comics. Some texts that score rather low on worldness or tellability, such as postmodern experimental fiction, may be rehabilitated for intellectual audiences by a high score on “innovative use of the medium.” It would take a cube with three dimensions to evaluate texts in terms of world, plot and medium,6 and this cube could not be represented on a two-dimensional page, but we can transpose it into tabular form, a type of representation that can accommodate any number of artistic dimensions. Table 1 represents my very personal scores on some famous narratives:7 first Forster’s made-up story (which nobody would bother to tell for its own sake), then Aldous

26  Marie-Laure Ryan Huxley’s Brave New World, which stands for science fiction in general, then War and Peace as an example of a great realistic novel, then Dan Brown’s Da Vinci Code, a popular thriller that topped the best-seller lists in the U.S. for almost a year, then David Mitchell’s Cloud Atlas, an experimental postmodern novel, then James Joyce’s Finnegans Wake, a text in which language draws so much attention to itself that the construction of a world and of a story becomes virtually impossible. The ultimate artistic narrative would score the maximum on all three criteria, but this kind of narrative is probably a utopia, because human attention is limited, and intense focus on the medium distracts people from the story and from the world. Table 1.1  Evaluating narratives on three criteria. Interest in story Interest in world Interest in medium 0 0 0 2 3 1 3 3 1 3 1 0 1 2 2 0 ? 3

“The king died” Brave New World War and Peace Da Vinci Code Cloud Atlas Finnegans Wake

Literary theory has long been dominated by a textualist attitude that locates the aesthetic value of texts in the dimension I call medium and that regards the interest taken in story and in world as symptomatic of popular, ­“low-brow” literature. It is time to abandon this view and to recognize that the ability to create compelling stories and worlds is no less a form of art than writing in a way that draws attention to language. While narratology has acknowledged for quite a while the pleasure provided by well-crafted stories as aesthetic, it is only recently that worldmaking has been recognized as a legitimate artistic activity. James DiGiovanna goes as far as claiming that worldmaking is a form of conceptual art in which the fictional world is the concept (2007, 115). By multiplying texts around worlds, worlds within stories, or stories within worlds, the aesthetics of proliferation bears testimony to the fascination that worlds exercise on the imagination. NOTES 1. Title of chapter 11 of Cleanth Brooks’ The Well Wrought Urn (1947). 2. Following Bordwell (1985) against Chatman (1990), I would not postulate a narrator for visual types of fiction such as film or drama, except in cases of voice-over narration, a phenomenon that remain to be satisfactorily theorized. It could be objected that postulating a narrator for language-based fiction but not for visually based media prevents a unified account of fiction, but visual and verbal fiction can be brought under a common denominator by invoking

Texts, Worlds, Stories  27 the  notion  of make-believe. In both cases the audience takes something (the author’s utterance, the actors’ play) for something that it is not: a narrator’s representation of true facts, the behavior of real people. A theory such as B ­ anfield’s that postulates a narrator for some kinds of language-based fiction (first person), but none for third-person narration or for film and drama creates a discontinuity within the medium of language, and it has a hard time explaining how there can be noncommunicative forms of language, i.e., language without an utterer. As for theories that postulate a narrator for all fictions, they must pull a narrator out of nowhere like a magician pulls a rabbit out of a hat for visual media. In language there is a mediating instance who reports the story, but in visual media the spectator can pretend to see and hear unmediated events. 3. This could happen in fantastic texts but not in realistic novels. 4. If the fact that the text was not composed as a story to be enjoyed for its own sake but rather as an example demonstrating the conditions of plot creates a problem, consider instead this minimal, much more moving short story attributed (probably apocryphally) to Hemingway: “For sale: baby shoes, never worn.” 5. By contrast, all the other stories mimic nonfictional genres: Story one is a diary, story two a series of letters, story four a written autobiography, story five responses to an interviewer, and story six an oral narrative of personal experience told by the protagonist to his grandson. Stories 1, 2, 4, 5 and 6 belong to a first-degree ontological level, while story three, as a fiction within a fiction, belongs to a second-degree level. I am indebted to Brian McHale for pointing out to me the fictional status of story three. 6. What about an axis for “characters?” I was asked during an oral presentation of an earlier version of this chapter. My answer is that characters are an integral part of the plot: Events cannot be separated from their participants. On the other hand, if the believability of characters, the vividness of their representation, and their ability to arouse emotions transcend the plot, then a fourth ­column could be easily added to Figure 1.3 to represent these features. 7. The criteria for the evaluations of “use of the medium” are as follows: 0, sloppily written; 1, elegantly written, but the writing does not attract excessive ­attention to itself; 2, consciously experimental writing; 3, a new kind of ­language, extremely self-referential and difficult to decode.

REFERENCES Babel. 2006. Directed by Alejandro González Iñárritu. France, USA, Mexico: ­Paramount Vantage. Banfield, Ann. 1982. Unspeakable Sentences: Narration and Representation in the Language of Fiction. London: Routledge and Kegan Paul. Barthes, Roland. 1970. S/Z. Paris: Seuil. Bordwell, David. 1985. Narration in the Fiction Film. Madison: University of W ­ isconsin Press. ———. 2002. “Film Futures.” SubStance 31.1: 88–104. Brooks, Cleanth. 1947. The Well-Wrought Urn: Studies in the Structure of Poetry. New York: Harcourt, Brace and World. Chatman, Seymour. 1990. Coming to Terms: The Rhetoric of Narrative in Fiction and Film. Ithaca: Cornell University Press.

28  Marie-Laure Ryan Cloud Atlas. 2012. Directed by Andy Wachowski, Lana Wachowski, and Tom ­Tykwer. Germany, USA, Hong Kong, Singapore: Cloud Atlas Productions. Cortázar, Julio. 1967. “The Continuity of Parks.” In Blow-Up and Other Stories, translated by Paul Blackburn, 63–65. New York: Pantheon. Dick, Philip K. 1984. “‘The Mainstream that Through the Ghetto Flows,’ an ­Interview with Philip K. Dick.” The Missouri Review 7.2: 164–85. DiGiovanna, James. 2007. “Worldmaking as Art Form.” The Journal of the Arts in Society 2.1: 115–22. Doležel, Lubomír. 1998. Heterocosmica: Fiction and Possible Worlds. Baltimore, MD: Johns Hopkins University Press. Eco, Umberto. 1990. “Small Worlds.” In The Limits of Interpretation, 64–82. ­Bloomington: Indiana University Press. Forster, E.M. [1927] 1990. Aspects of the Novel. Hammondsworth: Penguin. Genette, Gérard. 1972. Figures III. Paris: Seuil. Herman, David. 2009. Basic Elements of Narrative. Malden, MA: Wiley-Blackwell. Jenkins, Henry. 2006. Convergence Culture: Where Old and New Media Collide. New York: New York University Press. ———. “Transmedia Storytelling 101.” Confessions of an Aca-Fan (blog), March 22, http://henryjenkins.org/2007/03/transmedia_storytelling_101.html. Koten, Jirǐ. 2010. “Fictional Worlds and Storyworlds: Forms and Means of ­Classification.” In Four Studies of Narrative, edited by Bohumil Fort, Alice Jedličková, Jirǐ Koten, and Ondrej Sládek, 47–58. Prague: Institute of Czech Literature. Mitchell, David. [2004] 2012. Cloud Atlas. New York: Random House. Patron, Sylvie. 2009. Le Narrateur: Introduction à la théorie narrative. Paris: Armand Colin. Pavel, Thomas. 1986. Fictional Worlds. Cambridge, MA: Harvard University Press, 1986. ­ arratology, edited Pier, John. 2011. “Metalepsis.” In The Living Handbook of N by  Peter Hühn. URL: http://www.lhn.uni-hamburg.de/article/metalepsis-revisedversion-uploaded-12-may-2014. Run Lola Run (Lola rennt). 1998. Directed by Tom Tykwer. Germany: X-Filme ­Creative Pool, Westdeutscher Rundfunk (WDR), Arte. Ryan, Marie-Laure. 1991. Possible Worlds, Artificial Intelligence and Narrative ­Theory. Bloomington: University of Indiana Press. ———. 2006. “From Parallel Universes to Possible Worlds: Ontological Pluralism in Physics, Narratology and Narrative.” Poetics Today 24.7: 633–74. Smith, Barbara Herrnstein. 1981. “Narrative Versions, Narrative Theories.” In On Narrative, edited by W.J.T. Mitchell, 209–32. Chicago: University of Chicago Press. Walsh, Richard. 2007. The Rhetoric of Fictionality: Narrative Theory and the Idea of Fiction. Columbus: Ohio State University Press. Wolf, Mark J.P. 2012. Building Imaginary Worlds: The Theory and History of ­Subcreation. New York: Routledge.

2 Storyworlds and Paradoxical Narration Putting Classifications to a Transmedial Test Liviu Lutas The main goal of this chapter is to put the classification of the so-called paradoxical narrative devices to a transmedial test. This classification is the result of a collaboration that took place between 2001 and 2004 within the framework of the Interdisciplinary Center for Narratology in Hamburg. The project was led by emeritus professor Klaus Meyer-Minnemann, and involved Nina Grabe, Sabine Lang, and Sabine Schlickers. The main results of this project have been published in Spanish, in the anthology La narración paradójica; ‘Normas narrativas’ y el principio de la ‘transgresión’ in 2006. According to this classification, there are four paradoxical narrative devices: metalepsis, syllepsis, epanalepsis, and hyperlepsis. The size of this chapter will only permit me to focus on metalepsis and syllepsis, which I have already analyzed in other contexts. This limitation should however not cause any major problem, since my goal here is not an exhaustive overview of the classification. I will rather examine whether some of the classification premises change when the two devices are considered in the way they appear in media other than verbal. Besides, the classification has not been used as frequently as expected in literary studies after its publication in 2006, probably because it is too detailed and complex. Moreover, as I show below, some of the grounds on which the classifications were made are rather arguable. Still, the effort to make a comprehensive typology of devices that have in common the fact that they break the doxa of the narrative discourse, more precisely the fact that they break the limits between different narrative levels (Meyer-Minnemann and Schlickers 2010), is welcome. Indeed, it suffices to look at the great number of attempts to classify the different types of metalepses to realize that there is a need to systematize them. The choice to focus particularly on metalepsis and syllepsis is also motivated by the fact that the two devices are very closely related to one another. According to the Interdisciplinary Center for Narratology, both of these devices break the limits between the world of the teller and the world of what is told. The only essential detail that differs between them is the way in which the breaking occurs: transgression in the case of metalepsis and levelling, or bringing together, in the case of syllepsis. Moreover, narrative metalepsis is arguably at the basis of the whole typology developed by the Interdisciplinary Center for Narratology in Hamburg, since it was in

30  Liviu Lutas relation to its definition by French narratologist Gérard Genette that the breaking of the limits was first mentioned in relation with the typological criteria. Genette’s first definition of narrative metalepsis, that is “any intrusion by the extradiegetic narrator or narratee into the diegetic universe (or by the diegetic characters into a metadiegetic universe, etc.), or the inverse” ­(Genette 1980, 234–35), was actually followed just two pages further down by his highlighting of the fact that the breaking of the frontiers between worlds was essential in the case of metalepsis: “to overstep in defiance of verisimilitude a boundary that is precisely the narrating (or the performance) itself: a shifting but sacred frontier between two worlds, the world in which one tells, the world of which one tells” (Genette 1980, 236). One important aspect in this chapter is the relation between these narrative devices and the storyworlds of the fictional works in which they appear. An issue that can be discussed in such a context is that paradoxical narrative devices, by drawing the attention to the artificiality of the fictional work, hinder the reader’s ability to construct a world into which he or she could immerge. This is an issue that has been extensively debated lately by theorists like Werner Wolf (2005), Monika Fludernik (2005), or Sonja Klimek (2009), having shown that metalepsis does not necessarily have the anti-illusionistic, disruptive effect that other theorists, such as Marie-Laure Ryan (2010), Frank Wagner (2002), or Debra Malina (2002) found inevitable. Ryan thinks for instance that the ability to immerge into the world of fiction can be blocked by the use of paradoxical devices of the self-reflexive kind: “Many postmodern texts try to block immersion through the use of self-referential devices that remind the reader of the constructed nature of the fictional world” (Ryan 2010, 14). Wagner makes a similar point when he states that metalepses draw the reader’s attention to the existence of two different levels in a representational process, signaling thus the constructed nature of the story (Wagner 2002, 39). According to Malina, the self-reflexive dimension of the narrative metalepsis produces automatically an anti-illusionistic effect on the recipient. Indeed, narrative metalepsis for Malina is a “breach in narrative structure that undermines the narrative’s illusions” (Malina 2003, 138). As I have shown in an article on the relationship between metalepsis and participation (Lutas, forthcoming), metalepsis can sometimes, on the contrary, have an illusionistic effect and can even be used in order to actually represent the immersion in the world of fiction. For instance, in the short story “Umney’s Last Case,” written by Stephen King in 1993, a writer enters the world of a detective story he has written and switches places with the main character, who is actually his own creation. In such an example, the metaleptic intrusion can be interpreted as the epitome of fictional immersion, quite the contrary of the blocking mentioned by Wagner, Ryan, and Malina above1. However, as I argue below, there are other aspects than immersion related to storyworlds on which a consideration of paradoxical narrative devices could cast a new light. This goes in the other direction too: Viewing

Storyworlds and Paradoxical Narration  31 paradoxical narrative devices not only on the basis of narrative levels but also through the lenses of interactions between ontologically distinct worlds could, in some cases, permit reconsidering certain issues that seem problematic in the typologies. Another point in this chapter is the importance of the medium for the concepts of paradoxical narrative devices. Indeed, the definitions of these devices are based on the written medium, stemming principally from literary theory. A number of studies—from Brian McHale’s ­Postmodernist Fiction (1987) to more recent ones such as Werner Wolf’s “Metalepsis as a ­Transgeneric and Transmedial Phenomenon” (2005), the anthology ­Metalepsis in ­Popular Culture edited by Karin Kukkonen and Sonja Klimek (2011), and Jan Alber and Alice Bell’s Ontological Metalepsis and Unnatural Narratology (2012)—all have in common the tendency to enlarge the application field for metalepsis to representational arts in general. This tendency has not been taken into account in Sabine Lang’s typology, where the main criterion is the difference between discourse and story (discurso and historia in the Spanish original, Lang 2006, 32). In my opinion, a look at the way in which metalepsis appears in other media, a transmedial perspective in other words, could contribute to a better understanding of the device itself and of its relation to the other paradoxical narrative devices. PARADOXICAL NARRATION AND TRANSMEDIALITY Let me start by considering the last point mentioned in the introductory part above. Could paradoxical narrative devices be considered as transmedial phenomena? According to intermediality theorist Irina Rajewsky, transmediality can be defined as “media unspecific phenomena that can be employed in various media” (Rajewsky 2002, 13, my translation), exemplified by rhythm, sound, temporality and so on. Werner Wolf uses the term “transmedial devices” for the same kind of phenomena—meaning devices whose use is not restricted to one single medium (Wolf 2005). Meanwhile, the concept of transmediality according to this definition has been put into question by other theorists. In most of the cases this kind of criticism is motivated by an essentialist view of the different media, rightly criticized by Jørgen Bruhn who finds the idea of a pure, distinct medium as a “historical and ontological illusion” (Bruhn 2008, 26, my translation). However, even in a recent attempt at coming to terms with the concept of transmediality, Regina Schober seems to have difficulties avoiding the trap of essentialism: But what if there are no such things as “media unspecific phenomena”? What if every mediated phenomenon is automatically bound to its mediality, its cultural context and its perceptional circumstances? If it is true that implicit and explicit connections are not only inherent

32  Liviu Lutas but indispensable prerequisites in media adaptations, we may have to abandon, or at least modify, such a notion of transmediality that assumes a universal ontological realm independent of, and thereby disconnected from, its medial expression. (Schober 2013, 93) Schober might be right in assuming that every phenomenon has a specific appearance in each medium, but I find it difficult to agree with her suggestion that transmediality could be dispensed with. On the contrary, I am of the opinion, expressed for instance by Werner Wolf, that the transfer of concepts between different forms of art or media makes it possible to “highlight […] formal, functional and historical similarities” between the concepts themselves (Wolf 2005, 104). Besides, where should the borderline be drawn in order to conclude that a certain phenomenon is bound to a certain medium? Let me illustrate this point with a simple example: A subjective camera can be considered as a media-specific device—since it belongs either to the film medium or to the photographic medium—but the question is whether a comparison with the literary device of internal focalization wouldn’t contribute to a better understanding of both devices. As a matter of fact, one of the important aspects of internal focalization in literature is that it has been considered not only as a phenomenon of perception but also of cognition: Seeing the events through the eyes of a character would mean also having access to the character’s thoughts and feelings. François Jost criticizes this conflation of two different things, insisting on the importance of the perceptual dimension in focalization. However, at the same time, Jost applies the concept of focalization to film, keeping only its cognitive dimension. He therefore introduces a new term, “ocularization,” which would only have to do with perception: In order to differentiate visual point of view, on the one hand, which once again is not a metaphor in the cinema but rather a narrative reality, and cognitive point of view on the other, I would propose the following terminology: ocularization has to do with the relation between what the camera shows and what the characters are presumed to be seeing; focalization designates the cognitive point of view adopted by the narrative, with the equalities or inequalities of knowledge expressed at their full strength. (Jost 2004, 74) What Jost’s reflection shows most clearly is how the comparison between the cinematic and the written media helps bringing some clarity into the very complex concept of focalization. Theorist Lars Elleström proposes a solution to the problem of the essentialist comprehension of media when he suggests that different media could

Storyworlds and Paradoxical Narration  33 be considered according to the modalities through which they come to expression and reception. According to him, “the modalities are the essential cornerstones of all media without which mediality could not be comprehended and together they build a medial complex integrating materiality, perception and cognition” (Elleström 2010, 15). There are four elementary modalities: the material, the sensorial, the spatiotemporal, and the semiotic. It is these modalities, according to Elleström’s later development of the subject, that form a transmedial basis on which more complex features can be constructed. These features at a second level, called “compound media characteristics” by Elleström (2013), cannot be transmedial in the same way as the elementary modalities, since indeed they are closely linked to a specific medium. To explain this in a simple way, let me get back to the same example as above: The device of the subjective camera is a compound media characteristic linked to a medium where a camera is necessary. Still, at a level below, the level of elementary modality, a subjective camera has to do with the sensorial modality, more exactly with the visual mode, which belongs to the transmedial basis. It is exactly by acknowledging such properties of lower complexity that one can discover certain similarities between devices from different media, which seem completely incompatible at a first look. Paradoxical narrative devices are examples of compound media characteristics. As such, they could hardly be considered as fully transmedial. Not even narration in general can be considered to be a trait that can be found in all media, despite the so called “narrative turn,” according to which everything is narrative (see for instance Richardson 2000 and Hyvärinen 2006). But this is not a good reason to reduce the applicability of the paradoxical narrative devices to only one medium: the verbal one. I find for instance the insistence on the need of a narratorial voice for the metalepsis to occur inconsistent with the initial aim of Genette. Admittedly, metalepsis as a narrative device was classified under the heading of “voice” by Genette, but as it appears clearly in his definition already quoted above, the narratologist did take into account the possibility of metalepsis to occur even in performance media, even though he insisted on the act of telling: “a boundary that is precisely the narrating (or the performance) itself: a shifting but sacred frontier between two worlds, the world in which one tells, the world of which one tells” (Genette 1980, 236 my emphasis.). “Telling” in this case seems to stand in contrast with “performing,” so Genette’s addition of the term “performance” was important for the subsequent opening of the concept to all representational arts. Narratologist John Pier is of the opinion that “metalepsis is not a media-specific phenomenon,” and as such “has a significant role to play in the transmedial narratology [...] and in intermediality, although to date this connection remains largely unexplored” (Pier 2009, 200). I partly agree with Pier’s first argument, about metalepsis not being media specific, and will apply metalepsis to the film medium in this chapter. Still, it is not my aim here to investigate whether metalepsis can exist in all media, not

34  Liviu Lutas even in all representational media. What I completely subscribe to is Pier’s second argument, namely that its transmedial potential should be better explored, especially in order to cast a new light on some theoretical and classificatory aspects. Such a transmedial approach should of course take into account different forms of metalepses in film, something that cannot be done in a comprehensive manner within the frame of this chapter. Still, I would like to stress that a transmedial approach should go beyond the simple intermedial imitation (cf. Wolf 2002) of a narratorial voice, in the form of the device of a voice-over, which indeed is practically a verbal device, since the filmic medium has the ability to incorporate verbal media in different ways. It is thus of little interest to investigate the instances of metalepses from a film like Marc F ­ orster’s Stranger than Fiction (2006), where the main character hears the voice of a narrator who is invisible to him and to the film viewers. Such instances do not really cast a new light on a theory that is largely built on examples of verbal media. It is in other scenes from the film, for instance when an actual meeting between the character and the narrator eventually takes place, that the filmic medium can free itself from the verbal constraint, showing up some interesting, if not really ingenious, solutions to the meeting between different levels. In the sections below, I will have a closer look at both literary extracts and film scenes that can be used in order to explore some of the issues regarding the categorization of paradoxical narrative devices. TRANSGRESSIVE OR LEVELLING DEVICES The following is an extract from Honoré de Balzac’s Les illusions perdues: “While the venerable churchman climbs the ramps of the Angouleme, it is not useless to explain the network of interests into which he was going to set foot” (quoted by Fludernik [2003]. In the French original: “Pendant que le vénérable ecclésiastique monte les rampes d’Angoulême, il n’est pas inutile d’expliquer le lacis d’intérêts dans lequel il allait mettre le pied”). This is an often used example in the theory of metalepsis, used in order to support different categorizations. But this is also an example of a very commonly used literary device in works from the nineteenth century in general and has earlier been labeled as authorial intrusion or romantic irony already by Karl ­Wilhelm Friedrich Schlegel at the beginning of the nineteenth century. ­However, theorists belonging to the so-called classical narratological ­paradigm2 have insisted on not taking the real author into consideration in such cases, since author and narrator should be different entities in fictional narrative (cf. Genette 1990). According to such a view, the voice uttering these words is not the voice of the author, Balzac, himself. It is the voice of a heterodiegetic and extradiegetic narrator that is uttering these words, meaning a narrator who is not literally part of the storyworld, which is why the narratological concept of narrative metalepsis can be applied to this quotation.

Storyworlds and Paradoxical Narration  35 But what seems to happen here is that the narrator gives the impression that he or she is part of that world anyway, and that he or she has to give some background information in a hurry, while the main event is going on3. This of course is nothing but a trick, a way of giving the reader the illusion that the narrated world has a real existence independent of the narrator’s will. Another example, also from a well-known French nineteenth century novel, Alexandre Dumas’s The Count of Monte Cristo, can illustrate how the reader can get the illusion that the narrator can narrate one event at the same time another event is happening: We will leave Villefort on the road to Paris, travelling — thanks to trebled fees — with all speed, and passing through two or three apartments, enter at the Tuileries the little room with the arched window, so well known as having been the favorite closet of Napoleon and Louis XVIII, and now of Louis Philippe. (Dumas, chapter 10) Genette had actually differentiated these kinds of narrative devices from metalepses and called them narrative syllepses. Dorrit Cohn, John Pier, and Monika Fludernik chose to ignore this when they discussed the Balzac quotation above. Cohn placed the quotation in the category of “discursive metalepsis” (“métalepse discursive” in the French original), which for her is the opposite of “diegetical metalepsis” (“métalepse de l’histoire” in the French original), and found it much more “harmless” (“inoffensive” in the French original) (Cohn 2005, 122). Pier calls this example a “minimal metalepsis” (“métalepse minimale” in the French original) (Pier 2005, 249), considering that the transgression of the frontiers is only suggested in such a case, not real. Fludernik preferred to call it “rhetorical metalepsis” (Fludernik 2005, 79–81) as opposed to the “ontological metalepsis,” according to a classification made previously by Marie-Laure Ryan. According to Ryan, ontological metalepsis “opens a passage between levels that result in their interpenetration, or mutual contamination,” while rhetorical metalepsis only “opens a small window that allows a quick glance across levels, but the window closes after a few sentences, and the operation ends up reasserting the existence of the boundaries” (Ryan 2006, 207). After having expressed an initial doubt as to whether to consider rhetorical metalepses as metalepses at all, Fludernik admits that When I realized that the projected simultaneity metaphorically moves the narrator into the realm of the fictional world that I started to see where the boundary crossing might be located. In order to be able to talk while the cleric is climbing the stairs, the extradiegetic narrator would have to be located in the story, otherwise the while cannot link the same kind of temporality. (Fludernik 2003, 388 my emphasis)

36  Liviu Lutas The emphasized word above, “simultaneity,” reinforced by other terms related to temporality further down in the argumentation—“meanwhile,” ­“isochrony between the telling of the story and the time moving on the plane of the narrated world, the synchronization of narrating time and narrated time”—should have made Fludernik consider this case as a syllepsis. The same goes for Pier, who also mentioned the temporal dimensions of the minimal metalepsis in Balzac’s example, calling it “a minimal metalepsis, where the time of the story and the time of the narration live together” (Pier 2005, 249 my translation). Indeed, it is hard not to see the parallels to ­Genette’s definition of the paradoxical device of narrative syllepsis: “The fact of taking together those anachronic groupings governed by one or another kinship (spatial, temporal, or other)” (Genette 1980, 85). The temporal dimension is even clearer in ­Genette’s next comment on the device: “Syllepsis […] affects sequence (since by synthesizing ‘similar’ events it abolishes their succession) and duration (since at the same time it eliminates their time intervals)” (Genette 1980, 155). As mentioned above, Genette’s rather short and incomplete definition of narrative syllepsis focuses on events at the same narrative level, being rather of a horizontal type. The definitions did not take into account the possibility of bringing together temporalities from different narrative levels, for instance from the diegetic and the extradiegetic levels. This, together with the fact that narrative syllepsis was not studied under a chapter of its own4 as was metalepsis, might explain why Fludernik did not observe its applicability in the case of the Balzac quotation, which she considered to be a rhetorical metalepsis instead. In 2006, the narrative syllepsis received renewed theoretical attention. In her definition of syllepsis, Sabine Lang emphasized the importance of narrative levels for the device, especially by the example she chose to illustrate it. Indeed, according to Lang, “syllepsis is a procedure of simultaneisation of what is not simultaneous,” as in “while the character does this, I, the narrator, explain/will explain this other thing to you” (Lang 2006, 33 my translation)5. The simultaneisation of the two temporalities, the one of the extradiegetic narrator’s world and the other of the diegetic world, is exactly what happens in Balzac’s quotation, which had been considered as a metalepsis by all the three narratologists who analyzed it: Genette, Cohn, and Fludernik. However, while I agree with Lang in her categorization of syllepsis as a device distinct from metalepsis, I disagree with her and with MeyerMinneman and Schlickers as to the grounds of the classification. Indeed, I think that narrative syllepsis should not be classified according to the type of the infraction. In a heavy critique against Genette, the Interdisciplinary Center for Narratology in Hamburg chose to disregard the temporal aspect of the syllepsis and to emphasize its levelling dimension. They considered syllepsis as a levelling device, meaning a device that effaces the limits, and placed it together with epanalepsis in a class that stands in contrast

Storyworlds and Paradoxical Narration  37 to metalepsis and hyperlepsis, as it appears in the table below (cf. MeyerMinnemann and Schlickers 2010, 93 or Lang 2006, 31):

Table 2.1  Devices of paradoxical narration. I Levelling devices (Procédés nivelleurs - according to Meyer-Minnemann and Schlickers) (Procedimientos de anulación de límites - according to Lang) (= devices which efface the limits. My translation.) SYLLEPSIS EPANALEPSIS

II Transgressive devices (Procédés transgresseurs - according to Meyer-Minnemann and Schlickers) (Procedimientos de transgresión de límites - according to Lang) (= devices which transgress the limits. My translation.) METALEPSIS HYPERLEPSIS

As I have written elsewhere (Lutas, forthcoming), in my opinion the differentiation between metalepsis and syllepsis should not be made on the basis of the processes of levelling and transgressing the limits, but rather on the temporal dimension, as Genette did originally. Indeed, there seems to be a transgressive dimension in all syllepses and analogously a levelling dimension in all metalepses. This could be more obvious when considering cases of syllepses in other media, such as film. Indeed, narrative syllepsis is frequently used in films and often in ways that, being slightly different from their literary counterparts, could contribute to an enlargement and a sharpening of the theoretical perspective. As in the case of the metalepsis from the film Stranger than Fiction that I mentioned above, the most obvious kinds of syllepsis in film would be those of verbal type. Indeed, the use of voice-over makes it possible to employ the same kinds of narrative devices as in the verbal medium. A syllepsis of verbal kind is used for instance when the extradiegetic narrator of Sagan om Karl-Bertil Jonssons julafton (Christopher’s Christmas Mission in the E ­ nglish translation), a Swedish animated film from 1975 shown every Christmas Eve on Swedish television, says “let us now stop everything for a little moment and ask ourselves a couple of questions” (Danielsson 1975 my translation), when Karl-Bertil’s father has fallen asleep in front of the TV set. Thanks to the device, the viewer has the impression that the narrator stops the course of the plot in order to enact events that are supposed to happen at the same time Karl-Bertil’s father is sleeping. What is interesting anyway is that even filmic devices were used in order to mark the starting and the ending of the syllepsis. Thus, just before the words “let us now stop everything for a little moment” are uttered, the film shows the screen of a TV set right at the moment when the program ends, and the random dot pattern of the so called “noise” when no transmission signal is obtained by the analog TV set appears on the screen

38  Liviu Lutas (see Figure 2.1 below). This not only enforces the effect of the words uttered, underlining the fact that what is about to succeed happens during a pause, but is used also when presenting the events, which are projected against the background of the noise on the TV set (see Figure 2.2 below). The effect is one of a simultaneous representation of the world of Karl Bertil’s father, who is sleeping in front of a TV set showing nothing but noise, and the world outside, the world of the poor who are about to open the gifts they received from Karl-Bertil. Admittedly, the events from the world outside might be the product of virtual narration, to use Marie-Laure Ryan’s term (1995), i.e. nonfactual, or presented in a virtual mode, as they could have happened. But this would not diminish the sylleptical dimension of the scene. On the contrary, the sylleptical encounter between the two worlds has an even more disturbing edge if the worlds are ontologically separated.

Figure 2.1  Noise in Sagan om KarlBertil Jonssons julafton. © PennFilm Studio 2014.

Figure 2.2  Sagan om Karl-Bertil Jonssons julafton. © PennFilm Studio 2014.

As a matter of fact, the use of a voice-over in such a situation is rather unnecessary in film, since film viewers have grown accustomed to most of the techniques of montage and especially cutting. But what the narrator’s voice reveals is something that is actually implicit in all transitions between scenes where a dissolve is involved. A dissolve is a gradual transition between two images, made in such way that the images merge seamlessly into one another during one or two seconds (see for instance Fielding 1985, 152). Indeed, even in a transition between events separated by less than a second, a transition motivated sometimes only by the need of changing the camera perspective, there is a temporal jump that has a sylleptical dimension. This is what is done as early as 1940, in Charles Chaplin’s The Great Dictator, in the transition between two frames in the scene where the barber, who is mistaken for Adenoid Hynkel (Hitler’s parodical alter ego played by Chaplin himself) goes up the stairs in order to make the final speech of the film: For a short moment, so short that the eye of a seasoned film viewer could even disregard the effect, the same characters appear simultaneously on the screen in double versions, thanks to the technique of double exposure, as if two temporalities have merged into one. Twenty-four years later, the same technique is used in Guy Hamilton’s James Bond movie Goldfinger (1964). Thanks to the fact

Storyworlds and Paradoxical Narration  39 that the public has grown more accustomed to this technique, the time for the double exposure is here longer than in The Great Dictator, and the characters in the scene are more clearly double exposed in the transition between two scenes during the golf game between Goldfinger and James Bond. The sylleptical dimension of the transitions discussed above has become so conventional that were it not for the words uttered by a voice-over, as in the case of Christopher’s Christmas Mission, it would probably pass unnoticed. But in some transitions, the sylleptical dimension can be more salient. This could be because these transitions are not supposed to pass unnoticed, but are rather a meaningful part of the film’s theme. Thus, in Steven Spielberg’s The Adventures of Tintin (2011), the transition between the scene when Captain Haddock and Tintin wander in the desert and the scene enacting the story told by Captain Haddock during this wandering is made in a clearly sylleptical way, according to Genette’s first definition of syllepsis, quoted above: “groupings governed by one or another kinship (spatial, temporal, or other).” The kinship here, where the ship appears as if sailing in the desert, is visual, since the dunes actually look like waves in the sea. This transition is not just a case of a simple dissolve, since it is longer than two seconds and consequently less conventional. Besides, the theme of the past haunting the present of captain Haddock is so important in the film that even small formal details can be interpreted from the point of view of their contribution to that theme. Similar sylleptical transitions are frequently used in Robert Lepage’s La face cache de la lune (2003), where the connection between scenes from the main character’s present and his past is made through the zooming in on objects in the present that suddenly appear in the past when the camera zooms out, without any cut having occurred. In this case, too, the inability of the character to free himself from his traumatic past is of central importance, the syllepsis becoming thus more than a gratuitous formal experiment. There is another cinematic technique that can be studied using the concept of syllepsis. The interpolation of normal time and so called ­“bullet-time,” that is events filmed at variable speed, such as frozen time, dead time, flow motion, slow motion, or time slice (cf. Stephanie Argy 2001). Such techniques often used for instance in The Matrix Series – The Matrix (1999), The Matrix Reloaded (2003), and The Matrix Revolutions (2003)—seem specific to the film medium. However, it does not seem too far-fetched in such cases to apply Fludernik’s reflections on rhetorical metalepsis—which correspond rather to the narrative syllepsis as I argue above—especially when she mentions Laurence Sterne’s device of freezing the plot. ­According to Fludernik, the narrator “additionally freezes the actions on the story level in order to gain time for his discourse, thus interfering with the story” (­ Fludernik 2003, 389). All the examples analyzed above show how difficult it can be to distinguish between syllepsis and metalepsis, both on the basis of the dichotomy levelling/ transgressive devices and on the grounds of the temporal dimension. Such a difficulty shouldn’t however lead to an abandon of the concepts altogether. Even if a case study based on this distinction does not necessarily result in a

40  Liviu Lutas final and unquestionable categorization, the reflections on the different aspects related to the concepts can highlight certain details in the analyzed works that would have passed unnoticed otherwise. In the next part, I investigate another problematic issue in relation with the typology of paradoxical narrative devices: the hierarchy of levels and the direction of the transgressions. HORIZONTAL AND VERTICAL TRANSGRESSIONS I will start with a film example this time, namely the final scene from Andy De Emmony’s TV movie God on Trial (2008). The movie is about a group of Jewish prisoners in a concentration camp who put up a rabbinical court in order to decide whether God has broken his own laws when he made them go through such an ordeal. But at the same time, there is a frame narrative running along, occurring more than sixty years later, in the present of the film viewer, about a group of people visiting the camp under the guidance of a young woman who is informing them about the atrocities that took place there during the Holocaust. The two narratives run along in parallel and do not interfere until the final scene. In this scene, the two groups from the different time periods suddenly appear in the same frame, inside the most atrocious of all places in the camp: the gas chamber. In a long and slow, completely wordless, panning shot, the prisoners from the past appear standing among the visitors from the present, staring at the film viewer right through the camera, their nakedness contrasting with the visitors’ comfortable and warm clothes. Certainly, this example could be analyzed as another case of narrative syllepsis, since here too there is a paradoxical meeting between two worlds separated in time. Such a conclusion is supported by a visitor’s comment right at the end, after the scene in the gas chamber, an elderly man who could actually have been a prisoner in a concentration camp himself in his youth. “They are still with us,” he says, as if to confirm that the sylleptical meeting actually took place and was not just a narrative device. The example is well suited to highlight also the difficulty of deciding whether this is a case of levelling or transgression of levels. But what is particularly interesting in the context of this chapter is whether this is a case of a vertical or horizontal infringement of the frontiers between the different worlds. The type of the infringement, vertical and horizontal, is actually one of the other important criteria used in the typologies of paradoxical narrative devices. It has mainly been applied to narrative metalepsis, but it can be used just as well in relation to narrative syllepsis. As a matter of fact, Sabine Lang has applied it in an analogous manner to the other three devices in her typology, even though I have difficulties following the logic of applying a vertical infraction to something that she calls a levelling device. The possibility of different types of metalepses was introduced already in the original definition by Genette. His addition of the words “or the inverse” in relation to the direction of the intrusion was to be reemployed at a later

Storyworlds and Paradoxical Narration  41 stage: “Any intrusion by the extradiegetic narrator or narratee into the diegetic universe (or by the diegetic characters into a metadiegetic universe, etc.), or the inverse” (Genette 1980, 234–35 my emphasis. See also above.) It was reemployed first by Genette himself in his book from 2004, where he coined the term “antimetalepsis” for the cases when the characters left their fictional world and entered the ontologically separated world of their creator; then by John Pier, who coined the terms ascending metalepsis and descending metalepsis, depending on the direction of the transgression (Pier 2005). Both ascending and descending metalepses infringe upon the limit between narrative levels, but this presupposes that they are considered products of verbal media. Indeed, the concept of narrative levels is one of the compound media characteristics that seems to have a hard time losing its bonds to the verbal medium, even though its hypernym, narration, has been more or less accepted at least in representational arts from other media. The concept of narrative levels seems to be such an important detail that theorist Sonja Klimek takes the precaution of specifying that she works with “levels” not “worlds,” taking it for granted that “levels” only concern ­“written and not visual fiction” (Klimek 2011, 25). It is probably the same precaution that makes Klimek formulate what she calls “a plea to respect Genette’s initial definition” (Klimek 2011, 26) and to restrict the use of metalepsis only to cases where there is a vertical infringement, that is between “the world of the cre­ onsequently, ator” and “the world that is created within the artefact” (ibid.). C Klimek criticizes the introduction of what she considers to be a third class, but what really is a second class, of metalepsis, since it is opposed to the vertical one: the horizontal metalepsis. First coined by Meyer-Minnemann and Schlickers (2005) and Lang (2006), the horizontal metalepsis is a transgression between two different worlds situated at the same narrative level (Meyer-Minnemann and Schlickers 2005, 140)6. According to Klimek, “If we include transgressions between two parallel worlds under the term metalepsis, we must give up Genette’s definition, in which the transgressed frontier must be between the world of representation and the world which is represented— a definition that clearly excludes horizontal jumps” (Klimek, 2011, 25). I think that Klimek’s criticism is based on a logocentric perspective on the frontier between the world of representation and the world that is represented. As mentioned above, Klimek has difficulties accepting the existence of narrative levels in visual fiction. Admittedly, it is easier to consider vertical metalepses as discourse based and horizontal metalepses as world based, but I do not think that discourse and worlds are hermetically separated entities. The example from the movie God on Trial is well suited to explain what I mean. Indeed, the two worlds in the movie, the world of the visitors and the world of the prisoners, are situated at the same narrative level, being the product of an act of narration situated hierarchically above them. The meeting of the two worlds can thus be seen as a case of a horizontal metalepsis— or a horizontal syllepsis if one prefers to focus on the temporal dimension. However, a very close look shows that the worlds are not only temporally

42  Liviu Lutas but also diegetically dissociated. However odd this may seem, the guide in the frame narrative can be interpreted as a kind of narrator who is creating the other narrative. Such a hypothesis can be supported among other things by the fact that she, the guide, even encourages the visitors to “imagine” the events that actually take place in the parallel narrative. In such a case, the meeting of the two worlds would be a vertical transgression, since it would violate the frontier between two hierarchically separated narrative levels. The example above, of the film God on Tial, is a suitable reminder of how unstable typologies can be, among other things depending upon the criteria applied. As shown above, a transmedial approach helped me mitigate the logocentric bias in order to consider the meeting between two worlds in the same work of fiction as a horizontal metalepsis. Another film example will help me get even further in my investigation of the essence of horizontal paradoxical narrative devices. Jean-François Laguionie’s Le tableau (The Painting) from 2011 is an animated film that can be considered as a metalepsis from the start to the end. Already in the first scene a vertical metalepsis occurs both visually and verbally. Indeed, while the camera is zooming in, giving the viewer the impression that the frame is left behind and that the limit established by the canvas is transgressed, a character in the painting, a girl called Lola, comes to life and addresses the viewers with the words: “Voilà ! Vous venez de pénétrer dans le tableau” (“Here you are! You just entered the painting”). Such a comment could be interpreted as a metalepsis of the discursive kind according to the typology mentioned above, the “penetration” of the painting being in that case a metaphor for the act of reception. ­Actually, such a comment is quite close to the Balzac quotation above, or even closer to the Dumas quotation, where the narrator addresses the narratee with the words, “We will […] enter at the Tuileries the little room with the arched window” (see above). But beyond such classificatory problems, the fact remains that the transgression also has to do with a representational art from a non-verbal medium. Nevertheless, a verbal comment in such a case only underlines that the phenomenon is similar: a transgression of the frontier between “the world in which one tells, the world of which one tells,” with Genette’s own words (Genette 1980, 236, see also above). The fact that it is an intrusion into an ontologically different world appears more clearly in a movie, thanks to the visual dimension, but this does not mean that there are no narrative levels involved in this transgression. The world in which one tells corresponds to the world of the camera before it penetrates through the frame into the world of the painting, after which the character of Lola can enter into direct dialogue with the viewer. To be more precise, this example corresponds to Genette’s first definition of metalepsis, “any intrusion by the extradiegetic narrator or narratee into the diegetic universe” (see above), since it is the narratee, meaning the recipient of the narration, who is entering the diegetic universe of the painting. Following John Pier’s dichotomy, this would be a descendant metalepsis, of the kind that Pier exemplifies with an extract from George Eliot’s Adam Bede (1859), when the narratee and the narrator together transgress the demarcation line between the narrative levels (Pier 2005, 250).

Storyworlds and Paradoxical Narration  43 There are however even more interesting examples of metalepses in this movie, especially of the kind that could be analyzed as horizontal. Indeed, after the metaleptic start, the film represents the events in the world of the painting in a traditional way, without several levels or worlds involved, making the viewer forget the initial metaleptic intrusion. All of a sudden, Lola, who has abandoned her ability to communicate with an extradiegetic narratee and has become a character inside the world of the painting, arrives at the frontier of her world, the canvas of the painting, and jumps out of it, followed immediately afterward by two other characters. They enter an extradiegetic world, an ontologically different world, the world of the creator, the painter, a world that could be the same as the initial world of the narratee, the viewers’ world. This is of course a hierarchically different world, so the transgression is undoubtedly vertical. According to John Pier’s typology, this would be the case of an ascending metalepsis, which he exemplifies with an extract from Mark Twain’s The Adventures of Huckleberry Finn (1883), more precisely when Huck goes up one level and enters Mark Twain’s world (Pier 2005, 252). This transgression is so important in the movie Le tableau that the exact moment of the jump is used on the DVD cover (see Figure 2.3 below).

Figure 2.3  Le tableau. © Blue Spirit Animation / Be Films / Blue Spirit Studio / Sinematik / France 3 Cinema / Rezo Productions / RTBF (télévision belge) / 2011.

44  Liviu Lutas What happens next is very interesting from a theoretical point of view. When Lola jumps into her creator’s world, she lands on the frame of another painting. Shortly after that, the second painting falls down from the wall because of Lola’s weight, and Lola falls through the canvas into the world of this second painting, representing a war. She becomes a character inside that world, able to interact with the soldiers there. This example of metalepsis could be analyzed as an ontological metalepsis, as opposed to the more “rhetorical” or “discursive” cases studied above. Indeed, Lola literally travels among different worlds in a way that could be analyzed through the lenses of the possible world theories. According to Alber and Bell’s attempt to apply these theories to metalepses (Alber and Bell, 2012), such travel between different ontological worlds would correspond, at least partly, to Daniel Lewis’s concretist philosophical position. According to the Concretists, as opposed to Abstractionists such as N ­ icholas Rescher, Saul Kripke, Jaakko Hintikka, or Alvin Plantinga, an individual cannot exist simultaneously in more than one world. Thus, when Lola penetrates the world of the painter, she disappears from the world of the painting. But what is more important is that this is a horizontal metalepsis, since Lola actually crosses the frontier between two worlds situated at the same level: the worlds of two paintings hanging on the same wall. It is this kind of metalepsis that Sonia Klimek finds incompatible with Genette’s original definition, since the transgression is between worlds, not between narrative levels. But what this movie example shows is that in order to move from one painting to the other, Lola must necessarily pass through her creator’s world, the world “in which one tells.” One could argue that this example cannot be applied to literature, where the movement between two parallel worlds could occur without a necessary transition through a higher narrative level. But does the fact that such a transition is not represented imply that it does not occur? Isn’t it rather that the transition is implicit and does not have to be spelled out explicitly? My answer would be that this depends. In some cases, such as when a character from a fictional work appears in a later fictional work, it is probably better to talk about what Marie-Laure Ryan calls “transfictionality,” meaning “the migration of fictional entities across different texts” (Ryan 2013). But in the cases when such a migration is actually represented, when a character is represented when he or she literally leaves the storyworld, the paradoxical dimension is more obvious. However, even if I cannot provide a definitive answer as to the nature of horizontal metalepsis in the context of this chapter, a transmedial perspective is necessary in order to better understand this narrative device. CONCLUSION This short overview of some classification aspects of two paradoxical narrative devices, metalepsis and syllepsis, has highlighted that there is more

Storyworlds and Paradoxical Narration  45 work to be done in this field. It is actually a little paradoxical, if I can use such a term, that these devices, which are used more and more frequently and ingeniously in today’s cultural productions, haven’t yet received the theoretical attention needed to achieve enough consensus for a typology to be established. The typology established by the Interdisciplinary Center for ­Narratology in Hamburg, with four different classes of such devices, is a laudable attempt but has not succeeded in imposing itself. Even if there are reasons for this failure, as argued above, this attempt has shown that it is necessary to go beyond the monological focus on narrative metalepses. Admittedly, as Marie-Laure Ryan rightfully claims, metalepsis “has become one of the favorite toys of postmodern culture” (Ryan 2006, 439). This goes also for theory, where narrative metalepsis has been the object of a rapidly growing interest from the beginning of this millennium. However, the transmedial perspective that was applied on some of the classification aspects above has highlighted that many of the underclasses of metalepses—such as the rhetorical metalepsis, heterometalepsis, minimal metalepsis, etc.—should be considered as narrative syllepses, according to the long-forgotten definition by Gérard Genette, and later recycled by the Interdisciplinary Center for Narratology in Hamburg and myself (cf. Lutas 2012). Typologies might seem, as Werner Wolf had to admit, “unfashionable, since they seem too reminiscent of outmoded structuralist approaches with their notorious emphasis on static, ahistorical models” (Wolf 2002, 29). They can also seem “heuristically barren and generally problematic, since they inevitably create artificial boundaries and are flawed by borderline cases or the possibility of multiple classification” (ibid.). However, I ­completely agree with Wolf when he warns against the disparaging of all classificatory endeavours, and when he concludes that typologies “at least chart a field and make us see its diversity and […] sometimes reveal the very existence of phenomena hitherto neglected” (Wolf 2002, 30). This has clearly been the case in the analyses above, for instance in the case of the concept of narrative syllepsis, which thanks to the temporal dimension that differentiates it from metalepsis has proved to be useful when studying extracts from different media. By looking for instance at filmic examples of transitions between different scenes through the lenses of narrative syllepsis, it has been possible to highlight how such apparently harmless and conventional devices can contribute to the work’s overall meaning. Likewise, by employing the underclasses of vertical and horizontal metalepses to extracts from two film examples, it has been possible to show that those extracts are much more meaningful than simple experiments. However, the most important point of this chapter is that typologies, which were initially constructed on the basis of a single medium, can gain a lot in precision and clarity if they are put to a transmedial test. Essentialist views on paradoxical narrative devices as traits of verbal media alone may not only lead to incomplete, and sometimes erroneous, classifications but may

46  Liviu Lutas also reduce their relevance. If the view of media as pure and hermetically isolated entities is to be abandoned, as Jørgen Bruhn rightfully recommends, then of course it is necessary to loosen up some of the old classifications by taking other media into consideration. NOTES 1. See also Jean-Marie Schaeffer (2005, 325) and Jean Bessière (2005) who have highlighted this paradoxical aspect of metalepsis as an allegory of fictional immersion. 2. Classical narratology had its heyday during the structuralist period, in the 1960s and 1970s, and is reputed, sometimes wrongly, for trying to disregard the external context. For a definition of the dichotomy classical–postclassical narratology, see David Herman (1999). 3. This situation of an extradiegetic narrator who wants to give the impression that he or she is a part of a storyworld is similar to “the paradox of a narrator who exists in the storyworld, while his discourse is not part of the story,” analyzed by Marie-Laure Ryan in her contribution to this volume. 4. Actually Genette mentioned syllepsis in two different chapters: “Order” (­ Genette 1980, 85) and “Frequency” (Genette 1980, 155). 5. In the Spanish original: “la silepsis resulta ser un procedimiento de simultaneización de lo no simultáneo,” como en “mientras el personaje hace esto, yo, el narrador, les explico/les voy a explicar a Uds. Aquello” (Lang 2006, 33). 6. Admittedly, Frank Wagner’s concept of intertextual metalepsis (2002) and Sophie Rabau’s concept of heterometalepsis (2005) were precursors of horizontal metalepsis, since they made reference to cases when characters from distinct fictional worlds meet in the same world. But I am of the opinion that such cases should be studied apart, probably under the better-suited term of transfictionality used by Marie-Laure Ryan. This concept is discussed further down in this article.

REFERENCES The Adventures of Tintin. 2011. Directed by Steven Spielberg. USA: Columbia Pictures. Alber, Jan, and Alice Bell. 2012. “Ontological Metalepsis and Unnatural ­Narratology.” Journal of Narrative Theory 42.2: 166–92. Argy, Stephanie. 2001. “Frozen f/x still in action: There’s less love for morph.” Variety, January 2001. http://variety.com/2001/film/news/frozen-f-x-still-in-action1117792382/. Bessière, Jean. 2005. “Récit de fiction, transition discursive, présentation actuelle du passé, ou que le récit de fiction et toujours métaleptique”. In Métalepses. Entorses au pacte de la représentation, edited by John Pier and Jean-Marie Schaeffer, 279– 94. Paris: Édition de l’EHESS. Bruhn, Jørgen. 2008. “Intermedialitet: Framtidens humanistiska grunddisciplin?” Tidskriften för litteraturvetenskap 38.1: 21–38.

Storyworlds and Paradoxical Narration  47 Cohn, Dorrit. 2005. “Métalepse et mise en abyme.” In Métalepses. Entorses au pacte de la représentation, edited by John Pier and Jean-Marie Schaeffer, 121–30. Paris: Édition de l’EHESS. Danielsson, Tage. 1975. Sagan om Karl Bertil Jonssons julafton.http://www.angel fire.com/bc/densturrebrandt/karl-bert.html. Dumas, Alexandre. The Count of Monte Christo. Pennsylvania State University: The Electronic Classic Series. http://www2.hn.psu.edu/faculty/jmanis/dumas/ countmc1.pdf. Elleström, Lars. 2010. “The Modalities of Media: A Model for Understanding Intermedial Relations.” In Media Borders, Multimodality and Intermediality, ­ edited by Lars Elleström, 11–48. London: Palgrave Macmillan. ———. 2013. “A Theoretical Approach to Media Transformation.” In Translatio: ­Transmédialité et transculturalité en littérature, peinture, photographie et au ­cinema, edited by Alfonso de Toro, 97–106. Paris: L’Harmattan. La face cachée de la lune. 2003. Directed by Robert Lepage. Canada: FCL Films. Fielding, Raymond. 1985. The Technique of Special Effects Cinematography. Oxford: Focal Press. Fludernik, Monika. 2003. “Scene Shift, Metalepsis, and the Metaleptic Mode.” Style 37.4: 382–400. ———. 2005. “Changement de scène et mode métaleptique.” In Métalepses. Entorses au pacte de la représentation, edited by John Pier and Jean-Marie Schaeffer, 73–94. Paris: Édition de l’EHESS. Genette, Gérard. [1972] 1980. Narrative Discourse. Ithaca, NY: Cornell University Press. ———. 1990. “Fictional Narrative, Factual Narrative.” Poetics Today 11: 755–77. God on Trial. 2008. Directed by Andy De Emmony. UK: Hat Trick Productions. Goldfinger. 1964. Directed by Guy Hamilton. UK: EON Productions. The Great Dictator. 1940. Directed by Charles Chaplin. USA: Charles Chaplin Productions. Herman, David. 1999. “Introduction. Narratologies.” In Narratologies. New ­Perspectives on Narrative Analysis, edited by David Herman, 1–30. Colombus: Ohio ­University Press. Hyvärinen, Matti. 2006. “Towards a Conceptual History of Narrative.” In The Travelling Concept of Narrative, Studies across Disciplines in the Humanities and Social Sciences 1, edited by Anu Korhonen and Juri Mykkänen, 20–41. Helsinki: Helsinki Collegium for Advanced Studies. Jost, François. 2004. “The Look: From Film to Novel. An Essay in Comparative Narratology.” In A Companion to Literature and Film, edited by Robert Stam and Alessandra Raengo, 71–80. Maiden: Blackwell Publishing. Klimek, Sonja. 2009. “Metalepsis and Its (Anti-)Illusionist Effects in the Arts, Media and Role-Playing Games.” In Metareference across Media, edited by Werner Wolf, 169–87. Amsterdam/New York: Rodopi. ———. 2011. “Metalepsis and Fantasy Fiction.” In Metalepsis in Popular Culture, Narratologia 28, edited by Karin Kukkonen and Sonja Klimek, 22–40. Berlin/ New York: De Gruyter. Lang, Sabine. 2006. “Prolegómenos para una teoría de la narración paradójica.” In La narración paradójica; ‘Normas narrativas’ y el principio de la ‘transgresión’, edited by Nina Grabe, Sabine Lang and Klaus Meyer-Minnemann, 21–47. Madrid: Iberoamericana.

48  Liviu Lutas Lutas, Liviu. 2012. “Sur la syllepse narrative – Un concept théorique négligé.” In Poétique - revue de théorie et d’analyse littéraires 168: 445–66. Paris: Seuil. ———. Forthcoming. “Metalepsis and Participation in Games of Make-Believe.” Malina, Debra. 2002. Breaking the Frame: Metalepsis and the Construction of the Subject. Colombus: Ohio State University Press. McHale, Brian. 1987. Postmodernist Fiction. New York: Routledge. Meyer-Minnemann, Klaus, and Sabine Schlickers. 2010. “La mise en abyme en narratologie.” In Narratologies contemporaines: approches nouvelles pour la théorie et l’analyse du récit, edited by John Pier and François Berthelot, 91–108. Paris : Éditions des archives contemporaines. Pier, John. 2005. “Métalepse et hierarchies narratives.” In Métalepses. Entorses au pacte de la représentation, edited by John Pier and Jean-Marie Schaeffer, 247–62. Paris: Édition de l’EHESS. ———. 2009. “Metalepsis.” In Handbook of Narratology, edited by Peter Hühn, John Pier, Wolf Schmid, and Jörg Schönerl, 190–203. Berlin: de Gruyter. Rabau, Sophie. 2005. “Ulysse à côté d’Homère. Interprétation et transgression des frontières énonciatives.” In Métalepses. Entorses au pacte de la représentation, edited by John Pier and Jean-Marie Schaeffer, 59–72. Paris: Édition de l’EHESS. Rajewsky, Irina O. 2002. Intermedialität. Tübingen: Francke. Richardson, Brian 2000. “Recent Concepts of Narrative and the Narratives of N ­ arrative Theory.” Style 34.2: 168–75. Ryan, Marie-Laure. 1995. “Allegories of Immersion: Virtual Narration in P ­ ostmodern Fiction.” Style 29.2: 262–86. ———. 2004. “Metaleptic Machines.” Semiotica, Journal of the International ­Association for Semiotic 150: 439–69. ———. 2006. Avatars of Stories. Minneapolis: University of Minnesota Press. ———. 2010. “Fiction, Cognition, and Non-Verbal Media.” In Intermediality and Storytelling, edited by Marina Grishakova and Marie-Laure Ryan, 8–26. Berlin/ New York: De Gruyter. ———. 2013. “Transmedial Storytelling and Transfictionality.” Poetics Today, 34.3: 361–88. Sagan om Karl-Bertil Jonssons julafton (Christopher’s Christmas Mission in the ­English translation). 1975. Directed by Per Åhlin, script by Tage Danielsson. ­Sweden: Sveriges radio. Schaeffer, Jean-Marie. 2005. “Métalepse et immersion fictionnelle.” In M ­ étalepses. Entorses au pacte de la représentation, edited by John Pier and Jean-Marie Schaeffer, 323–34. Paris: Édition de l’EHESS. Schober, Regina. 2013. “Adaptation as Connection–Transmediality Reconsidered.” In Adaptation Studies: New Challenges, New Directions, edited by Jørgen Bruhn, Anne Gjelsvik, and Eirik Frisvold Hanssen, 89–112. London: Bloomsbury Academic. Stranger than Fiction. 2006. Directed by Marc Forster. USA, Columbia Pictures. Le tableau. 2011. Directed by Jean-François Laguionie. France: BE-Films. Wagner, Frank. 2002. “Glissements et déphasages: Notes sur la métalepse narrative.” Poétique, 130, 235–53. Paris, Éditions du Seuil. Wolf, Werner. 2002. “Intermediality Revisited: Reflections on Word and Music ­Relations in the Context of a General Typology of Intermediality.” In Word

Storyworlds and Paradoxical Narration  49 and Music Studies, edited by Suzanne M. Lodato, Suzanne Aspden, and Walter ­Bernhart, 13–34. Amsterdam: Rodopi. ———. 2005. “Metalepsis as a Transgeneric and Transmedial Phenomenon: A Case Study of the Possibilities of ‘Exporting’ Narratological Concepts.” In Narratology Beyond Literary Criticism, edited by Jan Christoph Meister, 83–107. Berlin/New York: De Gruyter.

3 The Charge against Classical and Post-Classical Narratologies’ “Epistemic” Approach to Literary Fiction Greger Andersson The aim of narratology—whether in its classical or post-classical version— has always been to find definitional criteria that will determine what “all and only narratives have in common” (Prince 2003, 66). It has generally been assumed that the theory—a theory that also concerns how readers make sense of narratives—should be valid for both what are called natural (factual) narratives and narrative fiction. However, this assumption has been challenged by theoreticians advocating a “separatist” approach. They argue that the production and reception of literary fiction presuppose sets of rules that differ from those that govern factual narratives. The distinction between these approaches and especially their putative contradictory views on how readers approach literary fiction is the main concern of this chapter. When David Herman discusses two approaches to literary fiction in The Emergence of the Mind, he uses the term “exceptionalist” to describe the approach I call “separatist” (Herman 2011). Herman, who advocates the alternative approach, maintains that readers understand fiction as they understand nonfictional narratives or perhaps, although this is not totally clear, even situations they come across in their daily life. The one d ­ ifference is that readers of fiction are always conscious that they are not presented with a falsifiable version (Herman 2002, 16; Herman 2011, 33; cf. Ryan in this volume). This approach has been described as “epistemic” by a ­“separatist” like Lars-Åke Skalin (forthcoming). Skalin distinguishes between (factual) narratives, which are supposed to have an informative purpose, and literary fiction. The former variant, which is assumed in general definitions of narrative such as “someone telling someone else that something happened,” generates, he argues, an “epistemic stance,” according to which readers or listeners attend to a narrative to gain information. A “separatist” or “exceptionalist” approach, on the other hand, assumes, as I have already mentioned, that the production and reception of literary fiction are activities that differ from the production and reception of informational discourses and that literary fiction must be understood on its own terms (Skalin, forthcoming). But what does this “on its own terms” imply and what are the salient differences between the two approaches? In order to be able to address these issues it is necessary to distinguish among at least three “separatist” approaches. Skalin seems, for example,

The Charge against Classical and Post-Classical Narratologies’  51 to advocate a different “separatist” approach than Richard Walsh (2007) even though there are common elements in the two approaches. Walsh argues, as does Skalin, that common versions of narratology are based on mistaken theories of fiction. They also agree that common concepts such as the pre-existing story or the obligatory narrator are ad-hoc hypotheses that should be reconsidered from the perspective of a more adequate theory of fiction. Even so, Skalin argues, and in this he appears to deviate from at least some applications of Walsh’s theories (cf. Brix Jacobsen et al. 2013; Krogh ­Hansen et al. 2013) that literary fiction should be distinguished not only from natural narratives but also from common uses of fictionality since it has a specific raison d’être. Literary fiction (or at least some literary fiction) is thus, according to Skalin, not only serious communication—fiction can be used in all kinds of forms and mixtures—by a communicator on the same ontological level as the readers, but art. Herman, however, appears to implicate neither Walsh nor Skalin but theorists advocating what has been called “unnatural narratology” when talking about “exceptionalists.” He hence assumes that “exceptionalists” base their case on the proposition that fiction may include elements that cannot be found in nonfiction. As an example he chooses Dorrit Cohn’s (1978) criterion that it is only in fiction that we can have access to other people’s minds. Herman counters this by pointing out that minds are not always transparent in fiction and that even readers of fiction are engaged in mind reading and that we also read minds and assume that we can understand them in real life. I am not convinced by Herman’s reasoning and do not think that his suggested distinction between his own approach and that of the exceptionalists covers the different approaches I discuss.1 The salient issue is, in my view, not whether there are elements in fiction that would be unnatural in nonfiction but rather the “separatists’” suggestion that fiction has another purpose and is apprehended as a different kind of communicative act than is nonfiction.2 Skalin objects, for example, to the “epistemic stance” and argues that readers are expected to approach at least some literary fiction with an attention that can be described as an appreciation of aesthetic effects. This kind of attention resembles the appreciation of art forms like painting, music, etc. Accordingly, when he distinguishes between the “epistemic” and the ­“separatist” approach he points to the difference between the telling of things and events as real referents independent of the telling and the construction of motifs that will be functional elements in an aesthetic composition. And, when it comes to the reading of fiction, he argues that this is not a pretending to make sense of “facts” but rather a reacting to aesthetic effects generated by constituents of the artwork. It could therefore be argued that a ­“separatist” like Skalin assumes that he has another perception of the semantics of literary fiction than do theoreticians who assume the “epistemic” approach.3 In the remainder of this chapter I will discuss the “epistemic” and “separatist” approaches, mainly Skalin’s variant, to literary fiction and refer to

52  Greger Andersson how readers make sense of fictional narratives, if, and if so, how they draw inferences and fill in gaps and, finally, to readers’ sense of “presence” and involvement in fiction; that is the peculiar kind of attention people often refer to when reading literature by the use of metaphors like ­“transportation,” “immersion,” “lost in a book,” etc. In this discussion I refer to examples from my studies of biblical texts and biblical scholarship and to a short story by the Swedish author Stig Dagerman. The biblical examples are from 1–2 Samuel, and from interpreters who aspire to use a literary or narrative approach in their study of the Bible. Dagerman’s short story, “To Kill a Child” [“Att döda ett barn”], was first published in 1948. It was commissioned by a Swedish organization for traffic safety. The story tells how a child is killed by a car on an ordinary day. A couple travelling to the sea have to pass through three small villages. In the third village a child is sent to neighbors to borrow sugar for the family’s morning coffee. The story oscillates between the two scenes, in the car and in the kitchen, until the accident puts an end to the characters’ short-term plans and changes their lives forever. The central theme of the story can be found in the following long sentence: Because life is constructed in such a merciless fashion, even one minute before a cheerful man kills a child he can still feel entirely at ease, and only one minute before a woman screams out in horror she can close her eyes and dream of the sea, and during the last minute of that child’s life his parents can sit in a kitchen waiting for sugar, talking casually about the child’s white teeth and the rowing trip they have planned, and that child himself can close a gate and begin to cross a road, holding in his right hand a few cubes of sugar wrapped up in white paper, and for the whole of that minute he can see nothing but a clear stream with big fish and a wide-bottomed boat with silent oars. (Dagerman 2013, 19–20) Very little is specified in this story. An unspecified couple in an unspecified car is travelling on an unspecified road on their way to an unspecified sea, passing through three unspecified villages, while an unspecified family is preparing for breakfast, etc. The story is thus in a sense about “everyman” and shows, as was obviously the intention of the organization that commissioned the story, that cars and traffic can be a deadly threat that we must handle responsibly. Yet, it is also a story about the mercilessness of life in general. It could thus be said, if we were to accept Skalin’s reasoning that, as a piece of art in distinction from a nonfictive narrative report, the short story is constructed to give form to a certain theme.4 In this sense, what Dagerman achieves is a work of art that can be described as a tragedy that affects his readers as tragedies do. But, and this is the question we must turn to now, do theoreticians whose approach I have called “epistemic” really assume that clauses in Dagerman’s

The Charge against Classical and Post-Classical Narratologies’  53 short story like “holding in his right hand a few cubes of sugar wrapped up in white paper” are apprehended by readers as units of information rather than as a motif of innocence that resonates with our premonition of the car crash and portrays the mercilessness of life, and do they assume that readers are affected by the story because they are relocated to the fictional world and not because of the genre of this composition and the notion conveyed that life is “constructed in a merciless fashion”? WORLD-MAKING AND READERS’ DRAWING OF INFERENCES AND GAP-FILLING A tenet in current attempts to improve and elaborate the theorizing in line with the “epistemic” approach to narratives has been to refer to versions of the idea of a fictional world to which readers are supposed to relate (see for example Marie-Laure Ryan’s chapter in this volume). Accordingly, the reading of fiction is understood as a reader’s construction, cued by elements in the text, of a mental model of a world (cf. Schaeffer and Vultur 2005, 238). Moreover, readers are supposed to “relocate” to this world (cf. Emmott 2005, 351). Herman says for example that he uses storyworlds “to suggest something of the world-creating power of narrative, its ability to transport interpreters from the here and now of face-to-face interaction, or the spacetime coordinates of an encounter with a printed text or a cinematic narrative, to the here and now that constitute the deictic center of the world being told about” (2002, 14). The reading of fiction and nonfiction would thus share basic properties: Both depend on the construction of mental models. These “worlds” are “global mental representations enabling language users to draw inferences about items and occurrences either explicitly mentioned or else implicitly evoked in a discourse” (Herman 2005, 569). How are we supposed to understand this talk about readers making models of a world, relocating to this world, etc.? (1) Are these descriptions to be taken only as metaphors that remind us that a text needs to be interpreted and that this act demands some creative processes of our minds? Or (2) Do these descriptions really denote that readers attend to stories from an internal perspective, and, if so, does such a perspective imply that readers are supposed to be located in a fictive world listening to a narrator who tells a story that is true in that world or that they are confronted with the “items and occurrences” of the world itself with no mediating teller? These questions relate to the putative difference between the “epistemic” and the “separatist” approaches and their distinct assumptions regarding how readers are affected by their apprehension of the purpose or genre of a narrative. Consider for example the following issues (a) and (b) below that concern readers’ understanding of Dagerman’s short story. (a) Adherents of what I have called the “epistemic” approach generally assume that readers ponder about whose voice they hear in a sentence like

54  Greger Andersson the one quoted above from Dagerman’s short story, and to which diegetic level this voice belongs (cf. Ryan in this volume). The common answer is that the voice is extradiegetical and hence that it does not belong in the storyworld (the setting and actions) readers construct in their imagination when reading the story. The sentence hence breaks the illusion, and readers are either transported back to reality or to an imagined situation in which they are listening to a fictional narrator telling a story about an accident. They might even consider whether or not this voice interprets the events correctly. This reasoning, which is based on a “two (or more)-leveled” model of stories, assumes that readers go between a level of immersion into a storyworld and a level of narration. Skalin, on the other hand, would argue that stories (literary art) are one-leveled. He would hence claim that the sentence quoted is apprehended by readers as being on the same level (in the sense of being on the canvas) as all the other parts of this work of art. (b) Christer Johansson, who discusses Dagerman’s “To Kill a Child” in his doctoral dissertation, points out that many things that can be left unspecified in the semiotic system of a short story cannot be unspecified in the semiotic system of a film (Johansson 2008, 98–104). Accordingly, the child in Dagerman’s short story must have a gender and an age in the film. The car must have a color and be of a certain model, and so on. It could be assumed that theoreticians who advocate an “epistemic” approach would argue that readers have already constructed a model of these things in their minds. But according to Skalin’s aesthetic approach this is incorrect. The point is that the story has a raison d’être according to which the filling in of these gaps is not significant. It could even be suggested that gaps in this story have an aesthetic function (or effect) simply as gaps. Dagerman (as well as the film director) is displaying motifs such as good people in a car on their way to the sea, good people in a little village with a little child, everyday dreams about good things, and so on, but these motifs can come to have a different function if they find expression in specific cars, specific faces, etc. The very genre or theme of the story may thus be lost. However, the question about how we are to understand the approach of world-theoreticians not only concerns the distinction between the “epistemic” approach and the aesthetic approach of for example a “separatist” like Skalin but also the more general issue of the relationship between “lifelikeness” and “narrative wholeness” in narrative theory. I use “life-likeness” both to denote that literature can imitate life and to denote that readers understand fiction using the same rules of interpretation as they do when facing situations in real life. It is commonly claimed, both by adherents of the “epistemic” approach and by “separatists,” that a major value of fiction is its ability to simulate real-life situations. When readers confront these situations their perspective is widened, and this offers them an opportunity to learn something about what it might mean to be a human being in this world. The assumption that fiction simulates possible factual situations seems at times to be linked to a principle of equivalency. According to this

The Charge against Classical and Post-Classical Narratologies’  55 principle, readers relate to and interpret situations displayed in fiction as they relate to and interpret situations in real life. This could be taken to be in line with the “epistemic” approach since some “world-theoreticians,” as we have noted, suggest that readers are transposed mentally to a fictional world and interpret this world from an internal perspective. Yet, there is a certain friction in the reasoning of at least some theoreticians advocating the “epistemic” approach, between the concept of simulation and the principle of equivalency, on the one hand, and narrative wholeness, on the other. I use the expression “narrative wholeness” to refer to the suggestion that a narrative is comprehended as a unit whose interpretation and values are given by the composition. This suggestion implies that readers are supposed to “follow” a narrative and accept its offered values. The displayed situations with “items and occurrences” are, according to this reasoning, not first and foremost particulars that are interpreted from an internal perspective but “functions” (motifs) in a compositional wholeness, that is, both the narrative structure and the genre or intent of the narrative. Gap-filling can be used as an illustration of the above-mentioned friction in the “epistemic” approach. It is a common notion that since authors cannot tell everything they will depend on their readers’ ability to draw inferences and fill in gaps. But does this notion suggest that readers are supposed to only fill in gaps in the authors’ compositions in order to follow the stories (or appreciate the artwork), or that they are supposed to fill in supplementary facts about the world to which they are relocated? Some theoreticians could, as I have already mentioned, be taken to assume that readers fill in gaps in the world the text is supposed to refer to. This gap-filling would then be based on the principle of equivalency. However, many if not all theoreticians assume, on the other hand, that the process of gap-filling is governed and restricted by some kind of relevancy (Walton 1990; Sternberg 1985; cf. Walsh 2007, 18). This implies that the gap-filling they refer to must be relevant in relation to a notion of intent, genre, or a compositional wholeness. This, and this is my point, implies that their reasoning, after all, is not really about the filling out of a world but rather about readers’ ability to make necessary inferences to “follow” and “understand” a story. This could be taken to imply that reading (1) above is the correct one, that is that common descriptions of readers’ involvement with fiction are to be taken only as metaphors that remind us that a text needs to be interpreted and that this act demands some creative processes of our minds. The lack of clarity concerning the meaning of readers’ assumed drawing of inferences and gap-filling can be found not only in theoretical discussions but also in empirical studies like Bortolussi and Dixon’s (2002). To be able to examine how readers draw inferences and construct meaning these authors distinguish between features in texts and readers’ constructions and examine the relationship between features and constructions via a manipulation of features and a study of how these manipulations affect readers’ constructions. Yet, both when they formulate issues and questions and

56  Greger Andersson when they interpret their results, Bortolussi and Dixon appear to assume the “epistemic” approach to literary fiction. When they assert that readers respond to characters as real people (2002, 134) and that “the mental representation of the narrator is analogous to the representation that people construct of other participants in a conversation” (2002, 191–92), we seem to come across the same kind of ambiguous propositions we see in other descriptions that assume an “epistemic” approach. It could be argued that the ambiguities I have tried to point out imply that the metaphors used by adherents of the approach I have called ­“epistemic” should not be taken literally and hence that the separatists’ critique is exaggerated. Common descriptions are thus, as I have already suggested several times, to be taken only as metaphors that remind us that a text needs to be interpreted and that this act demands some creative processes of our minds. Yet, and this is the issue I will turn to now, is it not possible that if readers, as Herman and others suggest, interpret “items and occurrences” (2005, 569) using for example folk-psychology then they will often come up with “strange” interpretations of literary texts that are not in line with the intent or purpose of the story? DISQUIETING INTERPRETATIONS A basic criticism made by Skalin, Sten Wistrand, and me of the “epistemic” approach is that some critics who apply this kind of narratology come up with disquieting interpretations. They are termed “disquieting” because we hold that readers who come across such interpretations realize that they are based on a different rule system than the one they themselves have assumed. This is a serious criticism, at least if one accepts Paul Ricoeur’s suggestion that theories like narratology “simulate a narrative understanding that is always prior to them” (1991, 23). The most common problem is that critics, based on what can be described as the fallacy of narrativization or naturalization, change their frame of interpretation and start to discuss a fictional narrative as if it were some kind of factual information and as a consequence direct their interpretative focus to the world rather than to the story. Critics can thus, as Skalin points out, analyze Prince Hamlet as if he were a real person (1991), discuss what kind of insect Gregor Samsa actually was transformed into, or what snake it was that caused the fatality in Conan Doyle’s The Adventure of the Speckled Band (Wistrand 2012), etc. The problem is often related to mind reading. Let me take an example from a venerable literary tradition, the Old ­Testament. In 1 Samuel 1, we hear about the barren Hannah and her bullying childbearing rival Peninnah. When Hannah cries, their husband Elkanah, tries to comfort her with these words: “Hannah, why do you weep? Why do you not eat? Why is your heart sad? Am I not more to you than ten sons?” (1 Sam 1:8 NRSV) J. P. Fokkelman ponders, when commenting on

The Charge against Classical and Post-Classical Narratologies’  57 this passage, about what Elkanah was thinking (mind reading). However, his comment does not concern the meaning or function of the motif in the narrative, he rather asks for Elkanah’s “real motives.” Fokkelman can thus explain that Elkanah is envious, longing for Hannah’s love. Fokkelman even informs us that we “get through to his subconscious when we observe that the Elkanah who wants to be victorious puts himself in a row of children. This inadvertent association, which ignores the generation gap, I interpret as a sign that Elkanah is not at all sure of his excellence, and in his heart wants to be told by her that he is her darling little boy” (1993, 30). It is very common among biblical scholars who assert that they read the biblical texts as literature to reason as Fokkelman does and to speculate about the persons and events they assume that they have been informed about. These scholars apparently regard the narratives as “windows to a world” and interpret the “items and occurrences” of this world as if they had been informed about them in another form. What then is the problem with these interpretations? Most readers would probably find it awkward to reason like Fokkelman about a fictional story. And they find it silly (cf. Walton’s “silly questions,” 1990) to speculate about Hamlet’s childhood or Sherlock Holmes’s snake. The problem with a certain kind of drawing of inferences and gap-filling would thus be related to the fact that these stories are made up. Accordingly, it could be argued that biblical scholars like Fokkelman do not make the same Quixotic mistake the interpreters of Shakespeare or Doyle, Skalin and Wistrand refer to, since these scholars assume that the biblical narratives aspire to be historical. Critics who assume that they are informed about something can isolate a sentence in this information and examine its reference. Such an examination is not restricted by the function and meaning of the sentence in its context. The critic can, like Fokkelman interpret the referent using his or her knowledge of psychology, for instance. There are simply no irrelevant inferences or suggested gap-fillings if narratives are read in this way. But if, on the other hand, we were to approach 1 Samuel 1 as a story, we could protest against ­Fokkelman’s interpretation and argue, first, that he has assumed a different frame and genre than we have and, second, that the function of the Elkanah motif is not to display a character in desperate need of psychotherapy. Elkanah is instead portrayed as a good but powerless man who must admit that he cannot give his wife the child she is longing for (this is a rather common biblical motif). Hannah then turns to God, and a miracle occurs. It could probably still be objected that Fokkelman’s reading is not problematic since it does not completely distort the meaning of the text. This is perhaps correct, but my point is that the kind of oscillation between interpretative frames that Fokkelman can be taken to illustrate tends to veil rather than to reveal the meaning and function of the texts. It can thus be concluded that the “separatists” do not relate the silliness of certain inferences or gap-fillings solely to the question of whether or not a

58  Greger Andersson narrative is fictive but also to readers’ notion of the intent, genre, or compositional wholeness of a narrative, regardless of whether it is made up or not. They suggest, as we have noted, that if a narrative is apprehended as a story—story is then not understood as a “world” but as a literary structure offered to readers by an author for purposes other than information—then its sentences are not about particulars but rather display motifs in a compositional wholeness. The biblical texts would, according to this reasoning, be a kind of “borderliners,” not only in the sense of being made up or not, but also in the sense of being literary stories or historical narratives (i.e., information). The problem with scholars like Fokkelman would then be that they claim to read these narratives as stories and yet often change stance and interpret the historical events they assume they are being informed about. I must repeat though that my point is not that readers do not draw inferences and fill in gaps when reading fiction, since it is obvious that authors like Dagerman in “To Kill a Child” assume that readers understand the meaning of the series of sentences they present. It could even be claimed that the very effect of Dagerman’s story is based on his ability to provide cues that trigger premonitions that jar the idyllic tableaus he first depicts. The salient issues are instead whether inferences and gap-filling are restricted to a “following” of a narrative and how the assumed criteria of relevance relate to “world-theories.” IMAGINATION VS. IMAGERY Another issue that concerns the understanding of “world-theories” and such metaphoric expressions as “transportation” and “immersion” is that they could be taken to explain a semantic process, psychological reactions, or both. Theoreticians who advocate a putative “epistemic” approach would thus claim that these theories both clarify how readers make sense of “To Kill a Child” and why the story affects them. A major objective of those studying narrative comprehension is to understand the nature of reading. In literary studies, the mental processes involved in converting texts into rich and complex cognitive representations are often taken for granted. However, from a linguistic and psychological point of view, there is a significant amount still to be learned about the way in which readers move from perceiving mere strings of words on the pages of books to the sensation of being so immersed in different worlds that they feel as if they are witnessing events and experiencing the emotions of characters […]. These worlds have been called “narrative worlds,” “texts worlds,” and “storyworlds” by different researchers. (Emmott 2005, 351)

The Charge against Classical and Post-Classical Narratologies’  59 Ryan explains (see her chapter in this volume) the psychological process as a game of pretense: If one applies this conception of possible worlds to narrative fiction, fictional worlds will be created by the mind of authors for the benefit of audiences. Readers, spectators or players relocate themselves in imagination into these worlds, pretending that they are actual. In the best cases, this game of pretense results in an experience of immersion in the fictional world. (Ryan 1991) Theoreticians who describe the reading of fiction in terms of world-making, as Emmott and Ryan do in the passages quoted, could be taken to suggest either that the construction of a mental model and experiences of “presence” are two sides of the same coin or that the construction of a world is a necessary prerequisite for feelings of presence. Anežka Kuzmičová, who claims in her doctoral dissertation that it is a misunderstanding of the theory of immersion to assume that theoreticians imply that readers are “following the narrated events from ‘within’ the storyworld, exclusively and consistently, throughout the entire text” (Kuzmičová 2013, 70), could perhaps be taken to assume the latter option. Kuzmičová suggests that we should distinguish between “imagery” and “imagination.” She studies “imagery” from the perspective of what is called second generation cognitive studies, and focuses on “embodied cognition.” Assuming that there is a connection between the unconscious neurological or muscular reactions (simulations) that occur when we come across phenomena in the real world and what we experience when reading fiction, she uses “imagery” to denote “simulations” that reach our consciousness so that we can reflect about them. Yet, “imagery” is also, as some kind of spontaneous near-sensory phenomena, separated from voluntary evoked “imaginations.” Kuzmičová distinguishes between two domains of imagery, each comprising two variants. The domain “referential imagery” consists of “enactment imagery” and “description imagery,” “verbal imagery” of “speech-imagery” and “rehearsal-imagery” (2013, 29ff.). “Enactment imagery” is described as short sensations of vicarious experience. A prerequisite for this kind of imagery is that there is an experiencer in the narrative. Kuzmičová refers to the study of mirror neurons and suggests that “enactment imagery” occurs when characters are said to have sensorimotor experiences, preferably in direct interaction with their physical environment (2013, 31). “Description-imagery” is a “semi-autonomous experience” that is mediated via a verbal filter. The imager’s embodied mind is in this case “situated outside the storyworld” (2013, 32). Kuzmičová challenges the common notion that description causes experiences of presence. She asserts instead, as we have noted, that readers have

60  Greger Andersson certain sensorimotor simulations when characters interact with objects in their environment. Descriptions, on the other hand, do not correspond to any kind of perceptual experience in the real world, and they are thus not perceptually mimetic. Yet, this kind of mimesis can occur if there is an experiencer in the fiction through whom readers’ may, as it were, perceive. Kuzmičová claims that this suggestion is confirmed by the fact that readers tend to skip long descriptions or read them fast and forget them. If, after all, we “see something” it can be described as voluntary imagination, as for example when we try to imagine a particular car model (2013, 90). However, not even this is common in reading, since many descriptions in fiction are too complicated and long-winded. The second domain, “verbal imagery” comprises, as we have seen, “speech-imagery” and “rehearsal imagery.” In the former variant we have the sensation of hearing a voice that is not our own but the voice of a character, while the latter variant resembles inner speech since we hear our own voice reading the words of the text. It is, Kuzmičová asserts, possible that these simulations occur constantly but that they only impinge on our consciousness and become imagery occasionally. Studies such as Kuzmičová’s could be taken to explain what metaphors like transportation, relocation, and immersion refer to. Yet, I am not sure that Kuzmičová and “world-theoreticians” or scholars like Bortolussi and Dixon are talking about the same thing. This relates to the ambiguity, which I have already referred to, of the theories of “world-making” and metaphorical terms such as transportation, for instance. They could, firstly, indicate that readers construct mental models and that they relocate to these models in their minds. This is a semantic suggestion. It does not necessarily mean that readers who construct storyworlds or draw inferences “feel as if they are witnessing events and experiencing the emotions of characters” (Emmott 2005, 351). The terms could secondly refer to “imagery” in the sense of sensations of presence. This distinction is important, because a “separatist” could argue that one does not need to accept the semantic theory to be able to explain readers’ experiences of presence. Kuzmičová’s psychological suggestions are thus not incompatible with an approach that holds that readers are affected by the aesthetic impact of a literary work of art. Adherents of such an approach can very well find it reasonable to suggest that readers experience “presence” when reading “To Kill a Child” although so much is unspecified—and remains unspecified—since readers do not need to construct a voluntary imagination of the child, the car, etc., because characters (experiencers) constantly interact with items in their environment. They open car doors, feel the sun in their eyes, shave, set the breakfast table, open gates, hold pieces of sugar in their hands, imagine what it will be like to row a small boat in the river or swim in the sea, etc. Yet, they would probably point out that the mimetic effect that Aristotle refers to is not created by these effects but by the genre and the conveyed notion that life is “constructed in a merciless fashion.”

The Charge against Classical and Post-Classical Narratologies’  61 It could even be argued that Kuzmičová’s study—I am not sure she would agree about this—can be taken to challenge the “epistemic” approach, since the “imagery” to which she refers appears to be occasional sensations rather than effects based on readers’ transportation to mental models of a world. Moreover, the sensations she describes imply that readers appreciate literary fiction with a peculiar and specific attention. They thus tend to see or hear things when these sensations are cued either by mirror effects or by semantic effects like when the voice as it were of the narrator or of a character really becomes (or functions as) a “voice,” that is a voice that holds a certain perspective. CONCLUSION: PROBLEMATIC METAPHORS I have in my discussion of two approaches to narrative fiction assumed that the major difference between those who advocate an “epistemic” approach and “separatists” is that the latter claim that it is a mistake to theorize about fiction and the reading of fiction as if it were a secondary version of a primary nonfictional form. Instead, fiction is to be regarded as a (or several different) communicative acts. A central concern in my treatment of the theoretical approaches has been the tenet adopted to explain and develop the “epistemic” approach via references to versions of the idea of readers’ world-making. According to the “separatist” reasoning I have discussed, neither the semantic nor the psychological argument sustains the necessity of the “epistemic” approach. Yet, a major issue underlying my reasoning has been whether theoreticians who suggest that readers construct “worlds” really assume an “epistemic” approach or if such an interpretation is a misreading of the metaphors they use. How, for example, are we to understand their reasoning about “gap-filling” and the drawing of inferences? If we were to assume a “separatist” approach, we could argue that these terms only refer to the reader’s quest for meaning required by a particular story. “Story” is then, as I have mentioned, not understood as a “world” but as a literary structure offered to readers by the author for aesthetic purposes and not for information. If this is what “world-theoreticians” suggest (the assumption that fiction cannot be falsified; the discussion about relevant gap-filling; and Bortolussi and Dixon’s and Kuzmičová’s critiques of earlier empirical studies could all be taken to imply this), then there is no real tension between the two approaches at all. The “disquieting” interpretations that I have suggested (along with critics such as Skalin and Wistrand) would in that case merely be “vulgarizations” based on too literal a reading of certain metaphorical expressions. It could be argued though that if this apprehension of the “world-­ theoreticians” is correct, they should not use those problematic metaphors. They should instead try to come up with theoretical concepts that

62  Greger Andersson do justice to the relative closedness of fiction, to its artwork properties, its transformation of items and occurrences into motifs, its construction of themes, and its functions in order to provide more accurate descriptions of how we read literary fiction. The alternative interpretation is that “world-theoreticians” actually assume that readers are informed about a world and that they direct their interpretative focus to this world. It could even be suggested that these theoreticians hold that readers relocate to a world and interpret items and occurrences as “matter.” Readers could in such cases be expected to interpret the world as they would interpret situations they come across in their daily life. If so, then there is a real tension between the two approaches, and critics who represent the “separatist” approach could argue that the “epistemic” approach will generate disquieting interpretations, since interpreters, so to speak, change stance or rule-system and read a fictional story as facts. They would then run the risk of both pursuing a non-aesthetic sense-making and neglecting the compositional wholeness of the story to focus on and examine particulars instead. This is a salient issue since scholars who advocate a “separatist” approach argue that fictional narratives are compositions consisting of motifs. The mistake of the interpreters they criticize would then not only be that they provide irrelevant inferences and gap-filling and so tend to produce new versions but that they miss instead the very raison d’être of fiction. NOTES 1. I am not convinced that theoreticians working with “unnatural narratology” (Krogh Hansen 2011), whom Herman views as exceptionalists, are ­“separatists.” The very designation “unnatural narratology” could be taken to imply that they assume that there are “natural fictional narratives” while “separatists” such as Skalin assume that all literary fiction is “unnatural” if the rules of nonfiction are taken as a norm (Skalin 2011). 2. It could though be argued that unnatural elements are a problem for the ­“epistemic” approach if readers do not perceive them as unnatural, since that would imply that readers do not construct representations of fiction in the way the theoretical framework implies. 3. Ryan argues in her chapter in this volume that the study of literature has gone from a focus on the text (the signifier) to the world (the signified). It could be suggested that the “separatists” represent a rhetoric or pragmatic turn in which the focus is yet again on the text. Both Skalin and Walsh thus focus, regardless of certain differences, neither on the story (in the sense of a fictitious content) nor on the world but rather on the text as an act of communication and the effects it generates. And they would probably argue that world-theoreticians who do not consider the function of the text risk starting to read the text as a referential act and thus to miss the intended frame of interpretation. 4. I am speaking of a theme as in art forms like music and painting and not of a thesis.

The Charge against Classical and Post-Classical Narratologies’  63 REFERENCES Bortolussi, Marisa, and Peter Dixon. 2002. Psychonarratology: Foundations for the Empirical Study of Literary Response. New York: Cambridge University Press. Brix Jacobsen, Luise, Stefan Kjerkegaard, Rikke Andersen Kraglund, Henrik Skov Nielsen, Camilla Møhring Reestorff, and Carsten Stage. 2013. Fiktionalitet. Frederiksberg: Samfundslitteratur. Cohn, Dorrit Claire. 1978. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton, NJ: Princeton University Press. Dagerman, Stig. “To Kill a Child” [Att döda ett barn]. 2013. In Sleet: Selected S­ tories, translated by Steven Hartman, 17–20. Jaffrey, NH: David R. Gordiner. Emmott, Catherine. 2005. “Narrative Comprehension.” In Routledge Encyclopedia of Narrative Theory, edited by David Herman, Manfred Jahn, and Marie-Laure Ryan, 351–52. London: Routledge. Fokkelman, J. P. 1993. Narrative Art and Poetry in the Books of Samuel: A Full Interpretation Based on Stylistics and Structural Analysis. Volume IV: Vow and Desire: I Sam. 1–12. Assen: Van Gorcum. Herman, David. 2002. Story Logic: Problems and Possibilities of Narrative. Lincoln NB: University of Nebraska Press. ———. 2005. “Storyworld.” In Routledge Encyclopedia of Narrative Theory, edited by David Herman, Manfred Jahn, and Marie-Laure Ryan, 569–70. London: Routledge. ———, ed. 2011. The Emergence of Mind: Representations of Consciousness in Narrative Discourse in English. Lincoln, NB: University of Nebraska Press. Johansson, Christer. 2008. Mimetiskt syskonskap: en representationsteoretisk undersökning av relationen fiktionsprosa-fiktionsfilm. PhD. diss., Stockholms universitet. Acta Universitatis Stockholmiensis. Krogh Hansen, Per, ed. 2011. Strange Voices in Narrative Fiction. Berlin: De Gruyter. Krogh Hansen, Per, Henrik Skov Nielsen, and Stefan Kjerkegaard, eds. 2013. “­Fiktion og Fortæling.” K&K: 115. Kuzmičová, Anežka. 2013. Mental Imagery in the Experience of Literary Narrative: Views from Embodied Cognition. PhD. diss., Stockholms universitet. Prince, Gerald. 2003. A Dictionary of Narratology. Rev. ed. Lincoln, NB: University of Nebraska Press. Ricoeur, Paul. 1991. “Life in Quest of Narrative.” In On Paul Ricoeur: Narrative and Interpretation, edited by David Wood, 20–33. London: Routledge. Schaeffer, Jean-Marie, and Ioana Vultur. 2005. “Immersion.” In Routledge Encyclopedia of Narrative Theory, edited by David Herman, Manfred Jahn, and MarieLaure Ryan, 237–39. London: Routledge. Skalin, Lars-Åke. 1991. Karaktär och perspektiv: att tolka litterära gestalter i det mimetiska språkspelet. Uppsala: Univ. Uppsala. ———. 2011. “How Strange Are the ‘Strange Voices’ of Fiction?” In Strange Voices in Narrative Fiction, edited by Per Krogh Hansen, 101–26. Berlin: De Gruyter. ———. Forthcoming. “Appreciating Literary Stories: Expectation Schemata and the Artwork as Performance.” In Expectations: Reader Expectations and Author Intentions in Narrative Discourses, edited by Stine Slot Grumsen, Per Krogh ­Hansen, Rikke Andersen Kraglund, and Henrik Skov Nielsen. Copenhagen: Medusa. Sternberg, Meir. 1985. The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading. Bloomington: Indiana University Press.

64  Greger Andersson Walsh, Richard. 2007. The Rhetoric of Fictionality: Narrative Theory and the Idea of Fiction. Columbus: Ohio State University Press. Walton, Kendall L. 1990. Mimesis as Make-Believe: On the Foundations of the ­ Representational Arts. Cambridge, MA: Harvard University Press. Wistrand, Sten. 2012. “Time for Departure.” In Disputable Core: Concepts of ­Narrative Theory, edited by Göran Rossholm and Christer Johansson, 15–44. Bern: Peter Lang.

Section II

This page intentionally left blank

4 How You Emerge from This Game Is up to You Agency, Positioning, and Narrativity in The Mass Effect Trilogy Hanna-Riikka Roine Over the past twenty years, narratological research has shifted its focus toward understanding narrative as a transmedial phenomenon. Consequently, interest has grown in the role of stories in different media. Digital games in particular have been celebrated as a democratic art form, a medium allowing the user to “actively engage” in a story rather than “passively receive” it (as a recent example, see Holmes 2012). In this chapter, I use BioWare’s Mass Effect Trilogy (2007–2012)—one of the most acclaimed science-fiction sagas in the digital role-playing genre—as a case study. The trilogy immerses the player in the adventures of Commander Shepard, a playable character who strives to defeat a race of sentient machines called the Reapers who threaten to purge the galaxy of all organic life. The crux of engagement in these kinds of games is often pinpointed as the act of making choices. In one of the missions in Mass Effect, for example, the player must address a potentially dangerous situation involving Wrex, one of Shepard’s crewmates, on the planet Virmire. Depending on how the player has played the game up to this point, she can either talk Wrex into remaining loyal, gun him down, or allow another character to kill him. The decision of whether Wrex lives or dies has serious repercussions for later events in the trilogy. Such choices are identified as having the potential to create new types of stories in digital games. Interestingly, they are also regarded as a feature that eliminates stories altogether, since narrativity is seen to require a communicative source “behind” the representations (such as a clearly definable author or narrator). At a minimum, the sense of the story’s being related is often deemed a necessary condition for something to be considered a story (e.g.. Abbott 2012). In my opinion, these arguments are problematic in the study of stories in role-play and in the digital game medium in general. The proposed arguments reduce the user’s or player’s activity to choicemaking. This overplays choices as a part of the stories and gameplay in such interactional environments, and it obscures other significant aspects, such as the player’s overall experience of the game. In the abovementioned mission on Virmire, the player must engage in a range of activities besides making dramatic one-off decisions, such as combating repeated swarms of enemies, driving a vehicle, and upgrading or modifying her characters’ equipment.

68  Hanna-Riikka Roine Concentrating on designer-led choices is also prone to obscuring the more creative and expressive sides of player activity and meaning-making. Furthermore, looking for “new stories” in games can be misleading. As Ruth Page (2012, 186) found in her recent study on stories in social media, there is rarely a straightforward dichotomous contrast between the stories in old and new forms of technology. Instead, we should focus on the new ways of authoring and using digital artefacts and keep in mind that they are “new” in comparison to the text-oriented, literary versions of narrative and narrativity. This is especially important if we aim to understand the narratologically challenging features of digital games: some of these features follow from the fact that the games are realised digitally, whereas others arise from the game medium itself. I attend to these challenges using the example of digital roleplay because this genre brings together achievement- and goal-oriented gameplay and more narratively motivated player activities in a unique way. A very concise account of the game events following Shepard’s decision on Wrex’s fate illustrates this combination: “[L]eading the strike team, Shepard reaches the facility, but discovers its main purpose is not to breed krogan, but to study Sovereign’s horrific indoctrination effect.”1 Here, leading the strike team and reaching the facility are realised through strategic, combat-centric gameplay, whereas discovering the main purpose of the facility is depicted by means of conversations and cutscenes. If describing repetitive, tactical aspects of digital role-playing games (RPGs) verbally is difficult, capturing their more imaginative dimensions and the experience of play as a whole in purely formal terms is even more challenging. This is why RPGs are not always considered their own separate genre within the spectrum of digital games: the nature of the role-playing experience easily eludes definitive accounts. It may be true that they share many formal features and elements with other digital games (e.g., puzzles, combat mechanics), but much of their uniqueness stems from acting in a fictional game world via playable character (see Hitchens & Drachen 2009, 7). In this chapter, I concentrate on this unique feature: the player’s experience of acting as a role-playing character inside the game world. As a starting point, the role-playing character not only provides the player with the means to grasp the goals and various possibilities of the game, it also enables her to become an integral part of the game world; the character thus functions as a medium for player creativity and expression. In my analysis, I focus on discussing how stories and narrative devices in general are used as resources in this process. Obviously, such devices do not play such an intrinsic role in all game genres, but RPGs like the Mass Effect Trilogy are designed around creating and developing the character while furthering the events and goals of what is often called the “backstory,” such as investigating the activities of the villain of the game and eventually stopping his plans. These two elements are closely related: the player’s actions and choices both develop the character and affect the events and

How You Emerge from This Game Is up to You  69 opportunities inside the game world. In the study of narrative in the digital game medium, it is important to recognise that stories in digital RPGs cannot be regarded as a one-way report or communication from the author (or designer) to the audience (or player). Instead, stories are used as part of the agency and positioning in reciprocal gameplay. My aim in this chapter is therefore to provide some important insights for both narrative study and RPG research. GAMES AND THE DIGITAL GAME MEDIUM: AUTHORING AND CONTROL Gameplay and stories are generally seen as a complicated combination. In the early 2000s, scholars of game studies (baptised ludologists) argued quite forcefully that digital games in particular display a unique formalism that defines them as a different genre from narrative, drama, and poetry (e.g., Eskelinen 2001; Frasca 2003). Thus, they argued, the “proper” study of games must focus on the analysis of their formal qualities, such as their rules. Perhaps the great emphasis on these qualities led ludologists to form an alliance with classical narratology in viewing narrative as a mode of presentation defined by certain textual features. As Marie-Laure Ryan has observed, the ludologists “rallied around an implicit battle cry […] ‘Games are games, they are not narratives’” (Ryan 2006, 183). Although the “blood feud” (see Jenkins  2004,  118) between ludologists and narrativists has abated somewhat in recent years, its legacy is substantial. The more recent interpretations of narrative and narrativity have not been applied to games to replace the older, traditional models that concentrated on recounting events communicated by the narrator to the narratee (e.g., Prince 1987, 58). RPG researchers Michael Hitchens and Anders Drachen (2009, 15), for example, point out that “a sequence of (typically) related events” that contribute to the “narrative element” of RPGs can be “portrayed” by digital games. In other words, the elements of digital RPGs cannot be termed narrative, because they lack authorial power in the traditional sense. For his part, game designer Greg Costikyan (2008, 6) has stated that a “story is [a] linear, controlled experience,” whereas in games, the players cannot be “constrained to a linear path of events, unchangeable in order,” else they feel that they are “being railroaded through the game, that nothing they do has any impact, that they are not playing in any meaningful sense.” According to this perspective, interesting gameplay is based on freedom and choices, whereas an interesting story must rest on linearity and control. Nevertheless, as Costikyan (ibid.) notes in his article, there are innumerable game styles that combine gameplay and stories successfully, and they do so in ways that evidently appeal to wide audiences. It is not impossible to meet the demand for linearity: gameplay, like any human experience unfolding in the course of time, is linear.

70  Hanna-Riikka Roine The more immediate problem for existing theories on narrative seems to be the one of authorial control and communication. “Perfectly these events,” as Costikyan (ibid.) puts it, cannot be attributed to the author or narrator. Here, the creative or expressive aspects of gameplay—such as the player’s ability to interact with the fictional world, navigate it more or less freely, and make choices (such as deciding whether a fictional character lives or dies)— are seen to reject effective storytelling. Richard Bartle (2009, 107) has tried to resolve the situation by suggesting that “all stories are interactive, in that they’re written for an audience,” describing gameplay as “interacting with the designer,” that is, being an audience to the designer’s story. However, Bartle’s contribution does not really tell us how the designer’s story can be encountered. His view on the interaction occurring between the player and the designer is also misleading, since it ignores the affordances of the digital game medium. Overall, one might be justified in stating, like Miguel Sicart (2011), that “players don’t need the designer—they need a game, an excuse and a frame for play.” The quotes above offer just a few examples of the contributions that attempt to describe the nature of stories and meaning-making in digital games. Partly, the challenges arise from the concept of the game, which can be used in more than one sense. A primary distinction of two elementary “layers,” namely the core—or the game as gameplay—and the shell—or the game as representation and sign system—makes the structure of the concept slightly easier to grasp (see Mäyrä 2008, 17). Gameplay—the abstract rule-set governing everything the player can do—is typically named the core because it defines the games in digital media as games, not digitally realised films or animations, for example. However, the representative or semiotic shell implementing the rule-set plays an especially major role in digital games, and player activity is involved in both the core and shell (ibid. 53). This should be kept in mind when discussing stories: their workings should not be simply placed at the level of the shell and viewed as represented sequences of events added on top of the core by the designers. Some insights into the affordances of the digital games medium are useful at this juncture for understanding both the interplay between the core and the shell and the challenges faced by existing theories of narrative. It is important to understand that the primary representational property of the computer is the codified rendering of responsive behaviours. In a digital game, this means that the player’s configurative actions and choices are tightly interwoven into the representative dimension. In other words, the events “portrayed” to the player in the Mass Effect Trilogy, for example, are essentially produced by the act of play. From a storytelling viewpoint, then, digital RPGs do not excel in presenting or evoking predetermined, unchangeable sequences of events. On the other hand, in legacy media such as literature and film, representation is static or unconditional. This does not mean, however, that narratives are eliminated altogether in the digital games medium.

How You Emerge from This Game Is up to You  71 Attempts to locate the authorial power and the differences in authoring fiction in old and digital media have often concentrated on interactivity. In my opinion, Bartle is on the right track when he suggests that interactivity in itself is insufficient for making distinctions, since “all stories are interactive” in the very elementary sense that no story exists unless there is someone to receive it. Concepts such as “interactive narrative,” then, are devoid of meaning. In addition to the concept of the game, interactivity is another term that is used both extensively and vaguely. Ryan provides one example of this in the context of “storytelling in digital age”: she considers interactivity to be the greatest difference between old and new media. She suggests that interactivity does not facilitate storytelling because “narrative meaning presupposes the linearity and unidirectionality of time, logic and causality, while a system of choices involves a nonlinear or multilinear branching structure” (Ryan 2006, 99). However, this approach does not provide us with an adequate understanding of modern digital media, and Ryan herself has recently moved on to the study of storyworlds (see Ryan 2013). Nevertheless, the approach—along with the concept of interactive narrative—is still quite common, and this is why I use it as a baseline for my discussion. The most significant difference between the computer and earlier media of representation (such as literature and film) is not the ability to contain structures, such as trees or rhizomes or to enable the user to navigate a structure by clicking a mouse or pressing the buttons of a controller. After all, such interactivity can be generated by means of gamebooks such as Choose Your Own Adventure. As Janet Murray (2011, 66) puts it, a more important property is the computer’s procedurality—its ability to represent and execute conditional behaviours. Likewise, Ian Bogost (2007, 4) defines “the core practice of software authorship” in terms of procedurality, maintaining that “to write procedurally, one authors code that enforces rules to generate some kind of representation, rather than authoring the representation itself.” Procedurality is thus understood not just as an ontological marker of digital games, but also as the specific way in which these games can build discourses of ethical, political, social, and aesthetic value. Bogost’s argument on procedurality, alongside his concept of “simulation fever,” attempts to formalise how the game as a system contains meaning and how the player experiences it. However, as Sicart points out in his critical essay “Against Procedurality” (2011), proceduralists such as Bogost seem to believe that the meanings of the game and play evolve from the way the game has been created and not how it is played. Sicart states that meaning should not be viewed as having been procedurally created or generated by the computer; rather, meaning is played (ibid.). I agree with Sicart, and I return to the topic of played meaning later in this chapter. Other problems emerge if we take structures like trees or rhizomes as the primary models for story construction in games. Ryan (2006, 99) concludes that narrative meaning, as “the product of the top-down planning of a storyteller or designer,” clashes with the “bottom-up input from the user.” With

72  Hanna-Riikka Roine its hierarchy, this approach strongly reminds me of literary analyses that utilise the story/discourse distinction and hold to it as if it were a law of nature. This distinction, called by James Phelan (2011, 58) “the mother’s milk of narratology,” is based on distinguishing the story—the “what” of narrative, consisting of events and existents—from the discourse—the “how,” including the structure and medium. In my opinion, this distinction is put to the best use if it is seen as part of the sensemaking and interpretative process. While Ryan’s conceptualisation of “interactive narrative” addresses interaction as “behaving in the audience,” it fails to address the reciprocity digital media can provide. It is also important to note that modern digital RPGs have moved away from hard-coded systems that presented players with a series of challenges with a sole solution; this move has brought emergent complexity into play (cf. Costikyan 2008). Next, I examine reciprocity in relation to authoring and player activity. Noah Wardrip-Fruin (2009, 7) aptly notes that although it is common to compare the creation of media with writing text, composing images, or arranging sound in the digital environment, “one must think of authoring new processes as an important element of media creation” (original emphasis). The authoring of processes in digital media can be illustrated with the classic example of choices, which features both in Ryan’s phrasing of “a system of choices” and in many classic definitions of games that emphasise the freedom and the enjoyment of problem solving, such as Sid Meier’s claim that “a [good] game is a series of interesting choices” (Rollings & Morris 2000, 38). In digital RPGs, choices typically stand out as an element of character development. The endless number of choices in Mass Effect, the first instalment in the trilogy, contributes to the repetition, which is a substantial part of the gameplay experience in any digital RPG, even though the fictional worlds represented in the games can be extremely wide, diverse, and detailed. Although the contents and contexts of the choices in Mass Effect differ, their execution is identical: a dialogue wheel presents “paragon choices” in blue, “renegade choices” in red, and neutral choices in white. The mechanics—namely the dialogue tree—are based on computational processes used by the game engine, which Wardrip-Fruin (2009, 46) has also called “operational logics.” The numerous missions completed by Shepard and his/her crew resemble each other because the game engine’s mechanism for tracking players’ progress is virtually identical in every mission or quest. At the end of Mass Effect, the player, through Commander Shepard, faces the “final choice.” A lone Reaper called Sovereign is attacking the Citadel, the space station at the heart of galactic civilisation, with the support of the villain of the game, the Reaper-indoctrinated Saren. The objective of the game’s last mission is to stop Saren from transferring control of the Citadel to Sovereign and to find a way to finally destroy the Reaper. If the player is victorious in various battles during the final mission, Shepard manages to gain control of all of the Citadel’s systems; he/she then opens the communications channel. The first message he/she receives is a mayday call from

How You Emerge from This Game Is up to You  73 the Citadel council’s flagship, the Destiny Ascension. Next, Joker, Shepard’s pilot, contacts the commander and says that he is with the entire Earth ­Alliance fleet and that they can save the Ascension if Shepard opens the mass relays—devices allowing interstellar travel—to the Citadel. Shepard must choose whether to tell the Alliance to save the Ascension and the council or to wait until the space station’s protective arms open and then concentrate on destroying Sovereign. The first option is the paragon choice while the second is the renegade choice. Both choices have significant ramifications, not only for the ending of the first game, but also for the next two game instalments. If Shepard chooses to ignore the Destiny Ascension’s pleas for help, the flagship will be destroyed and the council will die with it. This results in humanity’s greater prominence in the galactic order, but it also affects how Shepard is seen by other races in the galaxy. I wish to make two important points here. First of all, the work of creating digital RPGs such as Mass Effect cannot be viewed simply as authoring “a system of choices” that make up tree- or rhizome-like structures and the sequences of related events that are navigated through. For genuine agency to emerge, the digital environment must be meaningfully responsive to user input. As Murray (1997, 128) importantly points out, activity alone is not agency: the actions must be chosen and the effects related to the player’s intentions. For this reason, the game’s designers must author processes that enable the player—not the story or the representation as such—to make choices. However, this should not be taken to mean that players are just “activators” of the processes that set the meanings contained in the game in motion; nor are they just “derivers” of meaning from the system (see Sicart 2011). As I mentioned above, the unique feature of RPGs is the player’s capacity to act in a fictional game world via a playable character. Some important parts of creating meaningful responsiveness in Mass Effect are, for example, the processes and other elements that position the player as Commander Shepard. The devices used are diverse: the player is addressed in the second person (“You have followed Saren to the Citadel”), the abovementioned mechanics for making choices—along with other audio-visual representations of the player’s ability—are utilised to control the character inside the game world, and the character is developed through the use of resources. Overall, the authored processes facilitate the player’s participation in the game world, provide her with a frame for play, and engage her in a reciprocal action instead of allowing her to “complete” meanings that are prearranged prior to the act of playing the game. These processes can be seen as enabling the dialogue and as part of communicating possibilities in a way that is comprehensible to the players, since the human mind naturally does not work in the same way as a computer processor. Communication is therefore not excluded from the digital environment because the elements of the game system can be understood in their representational capacity. In this capacity, they not only stand for elements of the system, but they are

74  Hanna-Riikka Roine also designed to do so, and this makes them into communicative acts (Walsh 2011, 79). I next turn to the communication between the computer and the player from this perspective. FROM SYSTEM TO PLAYER: AN INVITATION TO A POSITION The active role of users is usually highlighted in discussions on digital games. Unlike in literature and film, for example, the most immediate question in games is usually not “What does this mean?” but “What can I do, and what should I do?” This results in different affordances in relation to work of authoring, and it reminds us that meaning is not procedurally created or generated by the computer; rather, meaning is played (see Sicart 2011). Murray (2011, 70) notes that a large part of digital design in general is “selecting the appropriate convention to communicate what actions are possible in ways that the human interactor can understand.” One aspect of this is the need to “script the interactors.” The earliest successful example of this was Joseph Weizenbaum’s natural language processing program, Eliza (1964–66), and its simulation of a psychotherapist (DOCTOR). In addition to facilitating the machine in processing users’ responses to scripts, Weizenbaum’s program succeeded in scripting or positioning the interactor as a participant in the highly conventionalised and familiar scenario of a therapy session. The program therefore guided the participant’s behaviour in a way that made meaningful action possible within the confines of a relatively primitive system. In light of this example, one could argue that the player must be scripted in digital RPGs in order for the player to understand what kind of behaviour is possible or desirable in the position of the playable character. In other words, a frame of this kind enables the player to comprehend what kind of actions she can perform and what kind of choices she can make in relation to the goals of the game. The reciprocity of the digital games medium, coupled with the fact that many video game developers (such as BioWare) are intent on attracting new types of players to their games, contributes to the importance of recognising scripting and positioning. In the case of the Mass Effect Trilogy, for example, the interactor is obviously scripted as a player, which enables her to use the game as a frame for play. However, as player actions (such as the order of completing quests) in digital RPGs can have multiple motivating factors, player activity is not solely about competition and winning. The designers, for their part, can anticipate these factors in addition to the multiple routes and chains of action the players can “concretely” choose. For example, the more freedom the player has regarding when, how, and whether to take on and complete quests, the more the events of different quest strands must be considered in relation to each another for the system to respond “meaningfully” to the player’s actions. In other words, the effects of the chosen action must be related to the player’s intentions.

How You Emerge from This Game Is up to You  75 There are other kinds of aspects to this freedom, which is usually hailed as one of the generic features of RPGs. While different user functions have already been recognised in relation to cybertexts (see Aarseth 1997), RPG researchers have discussed the players’ varying approaches to the game elements in more detail. The so-called Threefold Model distinguishes three player attitudes or styles: gamists, dramatists, and simulationists (e.g., Kim 2004). Gamists, for example, focus on the game’s challenges and the optimal strategies to overcome them, whereas dramatists are more concerned with the narrative qualities of the game, such as the nuances of meaning or the exploration of themes. Simulationists, in turn, are concerned with the internal consistency of the events that unfold in the gameworld and ensuring that they are only caused by in-game factors. In this view, digital RPGs such as the instalments in the Mass Effect Trilogy can be played with various goals in mind, such as beating the game as quickly as possible or creating and enjoying dramatic scenes. It is worth noting that Mass Effect 3 makes this feature quite explicit, as it provides the player with two campaigns entitled “Action” and “Story” alongside the “classic” RPG experience that combines these two aspects. Designers also need to consider players’ other varying needs and expectations: a topical issue related to this is the discussion of how gender and romances are depicted in games. In the Mass Effect Trilogy, for example, the third instalment introduced the possibility for pursuing a same-sex relationship. These issues are especially important for digital RPGs, which aim to give the players as full a range of character choices as is possible. So-called open world RPGs, such as Bethesda Game Studios’ Elder Scrolls V: Skyrim (2011), go to great lengths with the freedom of this kind, and even the Mass Effect Trilogy does not base its character creation on the modelling of readymade heroes. On the one hand, role-playing characters like Commander Shepard cannot be too ready-made, because as such they would be too distinct from the player. On the other hand, they need to offer a clear role with certain characteristics—otherwise it would be just a case of player herself acting in the game world. Due to this ambivalent nature, digital role-playing characters have been placed between two more extreme options, namely those of avatar and actor. An avatar is usually defined as a representation or manifestation of the player’s presence within the game world, while the actor is seen as a character distinct from the player with his/her own personality, characteristics, and, to some extent, mind. These two options are, of course, a simplification or a blueprint for game designers, but they can offer us an interesting viewpoint on the question of the scripting and mental models the playable characters can evoke. Such mental models were first highlighted by Donald Norman’s seminal study The Design of Everyday Things (1988), which brought attention to the models formed by the user based on the appearance and behaviour of the object. As Murray (2011, 60) notes, effective mental models are particularly important for machines whose workings are usually hidden from

76  Hanna-Riikka Roine us. She continues: “[i]n digital design we must ask ourselves what mental models the interactor will bring to the object” (ibid., 61). In the case of Shepard, the player can perceive him/her as her representer or avatar in the fictional game world and as Commander Shepard, a fictional entity. It is important to note that both the rule-based gameplay (the core) and the representative layer (the shell) contribute to these attitudes. As an avatar, Shepard is an audio-visually represented manifestation that, in a sense, embodies the player’s agency—that is, the player’s means to act in the game world and interact with it. For example, the avatar-Shepard belongs to a certain character class that enables the player to access certain active powers in combat, to increase health, shield and melee damage (measured in points), and to influence Shepard’s reputation or morality. Actor-Shepard, however, can be perceived as an imagined, fictional character with a past, present, and a potential directed toward the future, but as such, Shepard is still dependent on player input. Role-playing characters in digital form should be viewed, along with any computational objects, as variables that can have multiple instantiations, and not as predetermined entities. When a role-playing character is created, then, it is not created as a single version of an object (or a person), but as many possible versions with many variations (cf. ­Murray 2011, 53). This also forms a frame or model for the player’s creative freedom. Why does scripting and positioning matter in the discussion of narrative devices in digital games? After all, the narrative elements of digital games are traditionally considered those during which the player is passive, such as cutscenes. In past debates, this point has been raised, especially by ludologists, to argue that “games are not narratives” (e.g., Juul 2001). Arguments like this, however, illustrate why it is crucial to distinguish between two senses of narrativity as a concept: the quality of being narrative and the quality of inviting narrative interpretation. In his recent article, Walsh analyses the ways in which a simulation can generate behaviour that is narratively legible. He defines narrativity in its second sense as follows: To invite narrative interpretation means something more specific [than an event being susceptible to narration]; such a form of behaviour is necessarily communicative, since it implies an awareness of some narrative paradigm (some convention, genre, masterplot, or stereotype) that makes the representation narratively intelligible; and it assumes the mutuality of that awareness between the sender and the receiver of the communicative act. (Walsh 2011, 81) This account of narrativity opens up a new viewpoint on the question of why stories matter in digital RPGs and how they can be used as a part of the communication with the player. Digital RPGs make use of various narrative paradigms, but probably the most popular of these in the genres of

How You Emerge from This Game Is up to You  77 fantasy and science fiction is the heroic quest. However, understanding these paradigms in digital RPGs is not a case of analysing how the game mechanics are “wrapped” around some genre and masterplot or examining how the playable character is modelled on some effective stereotype. Rather, one must focus on how the paradigms can be used to shape player engagement and frame the act of play. In relation to this, Murray (1997, 89–90) brings rhetorics into discussion: “Just as we now know how to think about what made Tolstoy propel Anna Karenina in front of that train […], we need to learn to pay attention to the range of possibilities offered us as interactors in the seemingly limitless worlds of digital narrative.” Murray’s example bears a resemblance to Phelan’s rhetorical understanding of narrative: “Somebody telling somebody else on some occasion and for some purpose(s) that something happened” (Phelan 2007, 3 my emphasis). Phelan’s understanding of the rhetorics of narrative does emphasise communication, but it obviously cannot be extended from literary fiction to digital RPGs as such. Instead of looking for a “somebody” who authors the meanings of digital RPGs or asking to what kind of narrator-like figure meanings can be traced back to, I suggest we focus on the processes and cues shaping the act of play. When acting or making any kind of decision in the Mass Effect ­Trilogy, the player must think about the experience of play resulting from the game in actuality—the game-as-process—both in terms of representation and as a dynamic system evolving according to certain rules (cf. Sihvonen 2009, 146). In digital RPGs, the processes and elements enabling the player to role-play the character lie at the heart of this experience of play, as both the game’s progression structure and the meanings this structure potentially evokes are linked to the playing of the character. This results in an ambivalent situation, which Espen Aarseth (1997, 127) captures by describing the playable character as “an empty body, a contested ground zero.” In relation to shaping the act of play, this body is contested or empty precisely because it is simultaneously the player’s tool for constructing a story (or “secret plot,” as Aarseth calls it) and an integral part of that story. There are also two kinds of player attitudes inherent in this position. The option of perceiving Shepard as a possibly existing person is a feature that digital RPGs share with more traditional forms of fiction, such as literature and film. However, as RPGs are, first and foremost, games, the more imaginative dimensions of character construction are directed toward certain goals and are closely linked to configurative activity. Using an example from Mass Effect 2, I next illustrate how this close linkage can be considered as a method of communicating the potential of the game and its fictional world to the player. One of the important storylines running through all three games relates to the genophage, a biological weapon that was deployed against a warrior race called the krogans by two other races, the salarians and the turians. The weapon was designed to severely reduce krogan numbers by infecting the species with a genetic mutation, but it was not intended to completely eradicate them. In Mass Effect 2, Shepard recruits Mordin Solus, a salarian

78  Hanna-Riikka Roine scientist who participated in the modification of the genophage. As Mordin is still coming to terms with the consequences of these past actions, one of the so-called loyalty missions in the game is closely related to genophage. The mission “Mordin: Old Blood” initiates when Mordin informs Shepard that a group of mercenaries has captured Maelon, his former protégé, who helped Mordin on the genophage project. Maelon’s capture raises possible security concerns, meaning that he must be saved from the krogan home planet Tuchanka. It is possible to reduce the mission to a list of gamist challenges and options and to document it in a walkthrough guide, area by area, such as the “Hospital Approach” or “Hospital Interior”: Entering the hospital you only have one way to go down the stairs to your left. At the first landing you find a dead human body. Selecting this starts a conversation with Mordin and you can learn about the experiments being conducted […] ending the conversation with the options can earn a few morality points.2 In addition to showing us how the walkthrough guides describe player activity, the account above tells us something crucial about narrativity in digital RPGs. Alexandra Georgakopoulou’s work (2007) on stories in interactional environments, for example, has revealed the limitations of viewing stories as full-fledged, teller-led performances that unfold within a single event from beginning to end. Instead of communicating the inevitable “big story” of the genophage and Maelon’s research, the mission on Tuchanka unfolds sequentially in the moment-by-moment interaction between the player and the system. These sequences are realised through the player’s actions via the playable character, and in this, Commander Shepard can be seen as the player’s tool for answering the question, “What should I do?” in accordance with certain goals, such as “winning,” ­“making up a dramatic story,” or “developing the character.” However, if a digital RPG is well designed, the two layers work together quite seamlessly. At the end of the mission, for example, Maelon is found safe and sound. When Mordin confronts him, Maelon reveals that—out of guilt—he is working with the Krogans to undo the genophage. Shepard is finally given the decision of what to do with Maelon’s research data; these options are presented to the player through four different dialogue options. In short, Shepard can convince Mordin to either destroy or save the data. ­Choosing to destroy the data will earn the player Paragon points and contributes to the commander’s positive moral status. Alternatively, saving the data increases the Renegade score and contributes to a more ruthless or ­calculating Shepard. The example above illustrates how inviting the player into the position of the playable character participates in the meaning-making processes, but positioning does not create or communicate these processes to the player as such. In a sense, positioning provides a unifying frame that makes the game,

How You Emerge from This Game Is up to You  79 its workings, and its goals much easier to understand. One can also argue that the narrative coherence of digital RPGs is not based on the idea that some kind of instance or agent is responsible for organising the material into a cohesive or meaningful whole. Instead, the functions traditionally linked with the narrator (or implied author)—such as maintaining coherence or constructing the work as a whole—are based on the processes that enable the player’s positioning in the character role. In strategy games, for example, there is no need to invite the player into a position inside the fictional game world. Instead, the player is given the ability to indirectly control game units under her command, and, to facilitate this, a godlike view of the world. The position inside (or outside) the world is not the object of representation, but rather an indirect consequence of representing other objects (such as the fictional game world with its existents). This is also reflected in the usage of the second person in RPGs: “you” are positioned vis-à-vis various objects and events, as the user of various resources. In other words, when the user is invited into a position, she is positioned in relation to the existents and events in the fictional world and as a participant in configurative activity. In the final section of this chapter, I examine player performance in relation to communication. FROM PLAYER TO SYSTEM: PERFORMING IN AN EMERGENT STORY It should be clear by now that stories in digital games cannot be viewed as cases of somebody communicating something predetermined to somebody else or as players completing meanings that are determined prior to the act of playing the game. I have analysed above how narrativity can emerge in games in relation to directing and informing the player’s action and discussed the role of scripting and positioning in this emergence. Compared with previous work done on “interactive narrative,” my approach aims to take the reciprocity of digital media into account, and this is why I next highlight the meaning of performance as a communicative act. This is especially significant in the context of RPGs, because one of the defining features of role-play is the player’s performance of who she is in the game world as the playable character. One can ask whether the game system in digital RPGs is designed in a way that enables the player to position herself as the user of multiple resources (e.g., various game elements, familiar storylines, character traits, and other similar elements) in order to actively perform as a character such as Commander Shepard. The player’s performance is important not only for the sake of enjoyment or immersion, but also for communicative purposes. The question remains, however, how freely can the player play her role and not be railroaded along a path of some kind toward Shepard’s inevitable destiny? I examine this question of freedom from the perspective of emergent narration.

80  Hanna-Riikka Roine In order to be a truly emergent phenomenon, Commander Shepard’s character story must be a recursive product of the interaction between the player’s own representational strategies and the input of the game system itself. Walsh (2011, 81) has already named the Sims series a “relatively weak form of emergent narrative” since its instalments include some overt narrative prompts (such as the alien abduction scenario), but they are not integral to the functioning of the simulation. According to him, emergent narrative requires the simulative system to provide integrated narrative cues that are specifically responsive to the user’s input. Without cues of this kind, the system lacks the communicative intent that is deemed crucial for narrative to emerge: the elements of the system must be designed to stand for something in their representational capacity. Once again, however, it is important to recognise the reciprocity of digital games and the way meanings are created by play. Walsh (ibid., 82), for example, has argued that genuinely emergent narrative is possible in RPGs but only if the player maintains the duality of her performance: “playing a role must be simultaneously action and communication— behaviour as representation.” He discusses live action role-playing in relation to this idea and concentrates on the way the players perform for one another. The duality of performance in the digital environment is perhaps not as obvious a phenomenon as one sees when observing two people engaged in face-to-face role-play, but it is nevertheless crucial. As already mentioned, the p ­ rimary representational property of the computer is the codified rendering of responsive behaviours: in a sense, this is a very concrete example of behaviour as representation. In addition to this, it is important to realise that scripting, for example, is not only a case of communicating to the player what kind of behaviour is possible or desirable inside the fictional game world, but also a way of enabling her to communicate her actions and intentions to the computer. Sicart (2011) phrased this aptly: “[W]hen a player engages with the game, we enter the realm of play, where the rules are a dialogue and the message, a conversation.” What kinds of narrative cues or devices do digital RPGs use in order to evoke emergent narratives, and how are they related to gameplay? Well, one principal way of introducing such cues and devices in the Mass Effect ­Trilogy is by framing Shepard as a larger-than-life space opera hero and inviting the player to play him/her accordingly. A simple example illustrates this: at the beginning of the games, the player creates her playable character (unless she wishes to play either the default male or female Shepard or downloads her custom character from earlier games in the cases of the ­second and third instalment). This is done by means of a system called ­“Profile Reconstruction,” which is fashioned after a military service record. In this process, the narrative cues are most apparent in making a choice concerning Shepard’s pre-service history and his/her psychological profile. Nine different combinations are available, and one of the possibilities includes

How You Emerge from This Game Is up to You  81 combining the pre-service history of “Colonist” and the psychological profile of “Sole Survivor”: You were born and raised on Mindoir, a small border colony in the Attican Traverse. When you were sixteen slavers raided Mindoir, slaughtering your family and friends. You were saved by a passing Alliance patrol, and you enlisted with the military a few years later. During your service, a mission you were on went horribly wrong. Trapped in an extreme survival situation, you had to overcome physical torments and psychological stresses that would have broken most people. You survived while all those around you fell, and now you alone are left to tell the tale.3 These short introductions are similar to character backstories in many game genres, and they definitely show the effectiveness of stereotyping. They are especially important to RPGs because one of the primary goals of such games is not just using the playable character as a tool for achieving victory, but also developing the character as extensively as possible and adopting a dual attitude to the character in the process. The selection of Shepard’s pre-service history, for example, affects gameplay directly because it determines how many bonus Paragon or Renegade points the Commander has at the start of the game. The points mechanically open up special “good” or “evil” dialogue options that, in addition to other actions, grant the player more points. Compared with media such as literature and film, it is crucial to recognise that the process of representing meaning in a digital medium should be thought of as a process of abstracting objects and behaviours as efficiently as possible (see Murray 2011, 54). In a sense, RPGs accomplished this before being transferred to the digital format. The first edition of classic pen and paper RPG, Dungeons & Dragons (1974 Gygax & Arneson), was influential in its abstraction of the human personality into a set of attribute scores and presenting the concept of personal growth through life experience as the accumulation of experience points and “levelling up.” Other notable examples of such abstractions in digital RPGs obviously include the mechanics enabling quests and the choices available in conversations. While Profile Reconstruction in the Mass Effect Trilogy initiates the player’s scripting of the character role with certain effective stereotypes, these abstractions enable her further to “perform” what kind of a hero she wishes Shepard to be. The mechanics keep track of the points, for example, and translate or communicate this to the system. The importance of being able to create and perform different types of Shepard is further highlighted in the Mass Effect Trilogy because there is no purely gameplay-driven motivation for avoiding a particular type of action. The games keep track of the Paragon and Renegade points separately instead of representing them on a single axis. A good action will therefore not simply make up for an evil one, or vice versa. In Mass Effect 3, the points

82  Hanna-Riikka Roine have a significant impact on the resolution of another major storyline— the relationship between the geth and the quarians. The geth are a race of networked artificial intelligences, created by the quarians as labourers and tools of war. In the past, the geth became sentient and questioned their masters, which led to the quarians’ attempt to exterminate them, and, finally, war. The geth won the war and reduced the quarians to a race of nomads. During the trilogy, the geth are often fought against, as many of them have chosen to co-operate with the Reapers. In Mass Effect 3, they have allowed themselves to be controlled by Reaper code, believing the loss of their free will to be an acceptable price to pay to avoid extinction. The geth and the quarian forces finally meet at Rannoch, the original quarian home planet, and Shepard can try to broker peace between the two races. This description of the relationship between the geth and the quarians highlights one of the features of the nature of worldness in digital RPGs. The Mass Effect Trilogy creates its sense of a vast fictional world most efficiently by positioning the player as an active partner or agent in the crisis between the two races, and in a sense the player’s (or Shepard’s) past actions come back to haunt her at Rannoch. For example, if, during Mass Effect 2, Shepard’s quarian crewmate Tali was exiled from the Migrant Fleet—the quarians’ mobile home consisting of fifty thousand ships —the likelihood of achieving peace is greatly reduced. Conversely, taking time to save one of the quarian admirals on Rannoch in Mass Effect 3 adds to the chances of success. Therefore, it is not simply a case of the designers’ communicating the history of the geth and the quarians to the player; the player also creates a history of her own with them. In addition to performing favourable past actions, the player must manage to accumulate sufficiently high Paragon or Renegade scores for her character. This history weighs heavily on Shepard’s attempt to broker peace—war would result in one or the other of the two races being obliterated in the conflict. It is important to note that choosing to broker peace (or gaining the ability to make it) is neither simply a question of creating a happy ending for the war in the Mass Effect universe nor a chance to gain experience or reputation points for the character’s development. The way dramatic situations of this kind are tightly intertwined with more gamist goals in digital RPGs can be further illustrated with a comparison to digital games more inclined toward the genre of interactive drama, such as Heavy Rain (2010, Quantic Dream). In Heavy Rain, there are no significant motivations for the choices made during dialogues, for example, except for their contribution to the story, even though the game is operated by adventure game mechanics. In Mass Effect 3, on the other hand, one of the main objectives is to accumulate war assets, which in turn indicate the galaxy’s readiness to produce Effective Military Strength (EMS). The war assets have a direct effect on the consequences of Shepard’s final choice in the trilogy: for example, a very low EMS score results in the vaporisation of Earth if the player chooses to destroy the Reapers instead of controlling them or merging organic and

How You Emerge from This Game Is up to You  83 synthetic life forms into one. As both the geth and the quarian fleets are valuable war assets, brokering peace between the two obviously has a gameplaydriven motivation. However, this is one of the points where the way the player has chosen to play her character—that is, how she has played the game—will influence events. Abstractions and goal-oriented processes obviously have consequences in game fictions, and many researchers remain somewhat sceptical on the matter. For example, Wardrip-Fruin (2009, 79) states that there is a mismatch between the great variety of situations in which digital games are expected to “perform a fiction” and the simple model of fiction and character embodied in their processes. According to him, this mismatch results in a tendency toward a breakdown that takes a shape determined by the underlying processes, and is it therefore uninteresting. In my opinion, recognising these shapes does not mean that game fiction has failed—instead, they can be used as resources in the player’s performance and immersion in the game. The enjoyment of fiction is seen too often as a “total” experience, and this is why the recognition of the structures or shapes is viewed as negating the immersion in the fictional reality. However, as Frans Mäyrä and Laura Ermi (2005, 7) have argued, gameplay immersion is a multi-faceted phenomenon with different aspects that can appear and be emphasised differently in the individual cases of different games and players. Likewise, Lisbeth Klastrup (2009) has suggested that “worldness” in the case of a game world includes both the experiences of being in-the-world and the experiences of and about the world. The immersive aspects of games cannot therefore be simply viewed as a case of imaginative absorption into the game world. As mentioned above, stories in interactional environments in particular should not be viewed as full-fledged, teller-led performances that unfold within a single event from beginning to end (cf. Georgakopoulou 2007). In the Mass Effect Trilogy, the story of Commander Shepard is not monolithic; it is a result of many small fragments used in the performance of the character as a part of the communication or dialogue between the player and the system. Overall, recognising structures and shapes is important because the player needs to be able to move from the general principles of the system—such as how quests are structured—to smaller details—such as how they will contribute to the development of the playable character. As digital RPGs must abstract the devices and elements of, for example, quest narrative, the player who recognises the shape can very consciously “tweak” the stories to her liking while playing the game. This attitude is both gamist and narrativist in nature because it can be used simultaneously to beat the game and to perform as a certain kind of fictional character. Finally, it is noteworthy that the participatory nature of digital media has already profoundly affected the usage and reception of all media: they now appear to us as phenomena to be cut, pasted, reassembled, and distributed with ease. Nevertheless, there are operations that the player cannot subvert: the only

84  Hanna-Riikka Roine character you can play in the Mass Effect Trilogy is Commander Shepard. Positioning the player as the generic hero of a space opera and acting according to more or less stereotypical moral standards brings its own limitations as well. Despite their negative connotations, limitations and rules are also possibilities in the shape of a frame, and this sheds light on another important aspect of character stories in digital RPGs. The fictional Mass Effect universe can be perceived simultaneously as being a possibly existing world that is experienced from the inside—as if the events were happening to the player— and as being blatantly constructed according to certain strategies and, perhaps, from familiar building blocks. This is obvious in one of the essential themes of the trilogy: the relationship or boundary between human and nonhuman, an archetypal element of science fiction. In the act of creative play, where storylines and other similar elements can be used as resources, the Mass Effect Trilogy engages players in a dialogic relationship with the game system, enabling them to act or perform in a fictional world they enjoy. CONCLUSIONS AND NEW OPENINGS Perhaps one of the greatest challenges for future narrative study is to acknowledge the profound effect of digital media on the ways stories are told and used today, and to recognise the fact that this development is exponential in nature. My analysis on how narrative cues and devices are used as resources in the act of role-playing a character in a fictional game world shows the limitations of viewing stories as linear, narrator-led performances. I have also discussed the inherent problems of existing theories of interactive narratives, since they have long ignored both the reciprocal or dialogic nature of digital media and the importance of player agency. In my view, these problems have originated not only in the traditional model of narrative communication developed in the literary environment, but also in the more formalist approaches to games that ignore players as co-creators and performers. Recognising the communication or conversation between the player and the system has highlighted their emergent role in meaningmaking during the act of play and resulted in the abandonment of the view that stories have trees or rhizomes of some kind that are determined by designers. For its part, positioning opens up new viewpoints on digital role-play as a phenomenon that can combine gamist goals and narrativist ambitions quite seamlessly. The challenge for future RPGs is to harness the creativity of the players as extensively as possible but to offer them enjoyable frames for play at the same time. Of course, there are already creative approaches, such as participatory design that engages the users in the design process, and free and lowcost development platforms are now available, facilitating the explosion of so-called user-generated content. In recognising media content as something to be cut, pasted, and reassembled, instead of looking at monolithic story

How You Emerge from This Game Is up to You  85 structures, an interesting case study for further study would be to examine the importance of (familiar) fictional worlds for user enjoyment. Like games, phenomena such as fanfiction and fan editing are interesting examples of this, since they very self-consciously and openly use repeated storylines, stereotyped characters, and other such elements for communally exploring and creating spaces of enjoyment. NOTES 1. This account is from the article “Storyline” in the Mass Effect wiki (http:// masseffect.wikia.com/wiki/Storyline). 2. This walkthrough can be found in the Mass Effect wiki at http://masseffect. wikia.com/wiki/Mordin:_Old_Blood. 3. All options for these combinations of pre-service history and psychological profile can be read in the Mass Effect wiki: http://masseffect.wikia.com/wiki/ Commander_Shepard.

REFERENCES Aarseth, Espen. 1997. Cybertext: Perspectives on Ergodic Literature. Baltimore: Johns Hopkins University Press. Abbott, H. Porter. 2012. “H. Porter Abbott.” In Narrative Theories and Poetics: 5 Questions, edited by Peer G. Bundgård, Henrik Skov Nielsen, and Frederik Stjernfelt, 1–9. Copenhagen: Automatic Press, VIP Inc. Bartle, Richard A. 2009. “Alice and Dorothy Play Together.” In Third Person: ­Authoring and Exploring Vast Narratives, edited by Pat Harrigan and Noah Wardrip-Fruin, 105–18. Cambridge, MA: MIT Press. Bogost, Ian. 2007. Persuasive Games: The Expressive Power of Videogames. Cambridge, MA: MIT Press. Costikyan, Greg. 2008. “Games, Storytelling, and Breaking the String.” In Second Person: Role-playing and Story in Games and Playable Media, edited by Pat ­Harrigan and Noah Wardrip-Fruin, 5–13. Cambridge, Massachusetts, MA: The MIT Press. Ermi, Laura, and Frans Mäyrä. 2005. “Fundamental Components of Gameplay Experience: Analysing Immersion.” In Proceedings of DiGRA 2005 Conference: Changing Views – Worlds in Play. Vancouver: DiGRA. http://www.digra.org/wpcontent/uploads/digital-library/06276.41516.pdf. Eskelinen, Markku. 2001. “The Gaming Situation.” Game Studies 2.1. http://www. gamestudies.org/0101/eskelinen. Frasca, Gonzalo. 2003. “Simulation Versus Narrrative: Introduction to Ludology.” In Video/ Game/Theory, edited by Mark J.P. Wolf, and Bernard Perron, 221–36. London: Routledge. Georgakopoulou, Alexandra. 2007. Small Stories, Interaction and Identities. ­Amsterdam: John Benjamins Publishing Company. Hitchens, Michael, and Anders Drachen. 2009. “Many Faces of Role-Playing Games.” International Journal of Role-Playing 1: 3–21.

86  Hanna-Riikka Roine Holmes, Dylan. 2012. A Mind Forever Voyaging: A History of Storytelling in Video Games. Marston Gate: Amazon.co.uk, Ltd. Jenkins, Henry. 2004. “Game Design as Narrative Architecture.” In First Person: New Media as Story, Performance, and Game, edited by Pat Harrigan, and Noah Wardrip-Fruin, 122–30. Cambridge, MA: The MIT Press. Juul, Jesper. 2001. “Games Telling Stories: A Brief Note on Games and Narratives.” Game Studies 1.1. http://www.gamestudies.org/0101/juul-gts. Kim, John H. 2004. “Immersive Story. A View of Role-Played Drama.” In Beyond Role and Play: Tools, Toys and Theory for Harnessing the Imagination, edited by Markus Montola and Jaakko Stenros, 31–38. Helsinki: Ropecon ry. Klastrup, Lisbeth. 2009. “The Worldness of EverQuest: Exploring a 21st Century Fiction.” Game Studies 9.1. http://gamestudies.org/0901/articles/klastrup. Mass Effect. 2007. BioWare; Electronic Arts & Microsoft Game Studios. Mass Effect 2. 2010. BioWare; Electronic Arts. Mass Effect 3. 2012. BioWare; Electronic Arts. Murray, Janet H. 1997. Hamlet on the Holodeck: The Future of Narrative in ­Cyberspace. New York: The Free Press. ———. 2011. Inventing the Medium: Principles of Interaction Design as a Cultural Practice. Cambridge,MA: MIT Press. Mäyrä, Frans. 2008. An Introduction to Game Studies: Games in Culture. Los ­Angeles: Sage Publications Ltd. Norman, Donald 1988. The Design of Everyday Things. New York: Doubleday. Page, Ruth. 2012. Stories and Social Media: Identities and Interaction. Routledge: New York & London. Phelan, James. 2007. Experiencing Fiction: Judgments, Progression, and the Rhetorical Theory of Narrative. Columbus: The Ohio University Press. ———. 2011. “Rhetoric, Ethics, and Narrative Communication: Or, from Story and Discourse to Authors, Resources, and Audiences.” Soundings 94.1–2: 55–75. Prince, Gerald. 1987. A Dictionary of Narratology. Lincoln: University of Nebraska Press. Rollings, Andrew, and Dave Morris. 2000. Game Architecture and Design. Scottsdale, AZ: Coriolis. Ryan, Marie-Laure. 2006. Avatars of Story. Minneapolis: University of Minnesota Press. ———. 2013. “Transmedial Storytelling and Transfictionality.” Poetics Today 34: 362–88. Sicart, Miguel. 2011. “Against Procedurality.” Game Studies 11.3. http://gamestudies. org/1103/articles/sicart_ap. Sihvonen, Tanja. 2009. Players Unleashed! Modding The Sims and the Culture of Gaming. Turku: University of Turku. Walsh, Richard. 2007. The Rhetoric of Fictionality. Narrative Theory and the Idea of Fiction. Columbus: The Ohio State University. ———. 2011. “Emergent Narrative in Interactive Media.” Narrative 19.1: 72–85. Wardrip-Fruin, Noah. 2009. Expressive Processing: Digital Fictions, Computer Games, and Software Studies. Cumberland: MIT Press.

5 Playing the Worlds of Prom Week Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, Noah Wardrip-Fruin, Josh McCoy, Aaron Reed, and Michael Mateas

If, as Eco argues, a literary text is “a machine for producing possible worlds” (Eco 1984, 246), including the many worlds the audience ­imagines, then one way of looking at computer games is as machines for producing such machines. As audiences play games, they not only imagine possible worlds from the fiction1 experienced thus far, but they also imagine what range of possible worlds the game system is capable of producing through play (through interaction with its operations) and may actively experiment—not only in attempting to shape the next stages of the fiction through employing different strategies, but also in attempting ­“counterfactual” playthroughs of parts of the fiction already experienced (and then perhaps returning to a saved game that restores the previously experienced fiction as the “actual world” of their play). Each of the experienced states and imagined possibilities of the game system is also a point from which a multitude of fictional possibilities could be imagined. This has rich potential for the experience of fiction that, unfortunately, most computer games almost entirely squander. If we look at the player experience of many story-focused computer games, we see a form in which possible worlds are constantly produced. But these have a very particular structure. The player character (e.g., Lara Croft, Guybrush Threepwood) attempts to do something over and over, creating many possible worlds of failure. Finally, one projected world results in success, and the fiction moves on to the next scenario. As this happens, the player’s effort essentially results in the reconstruction of the successful fictional world defined by the game’s design, or one of a handful of such worlds, and all of the previously produced worlds are simply degenerate preliminaries to each successful stage. In short, while gameplay can create many possible configurations from which to imagine possible worlds, most game fictions have an embedded “textual actual world” (Ryan 1991, 24), and most game players are aware of this. Marie-Laure Ryan proposes an exploratory/ontological spectrum (Ryan 2001) to describe how much power the player has in affecting the ­“textual actual world” of a game; interaction in games in which change can be enacted is considered ontological, while interaction in games with largely static stories waiting to be discovered and read is exploratory. But even if the appearance of a wide variety of audience-driven possible worlds in most forms of computational narrative is an illusion—with the audience tasked either with uncovering and interpreting a single fiction in the past or with performing the

88  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al prescribed actions of the game’s actual world in the present—there are other forms of computational narrative that do produce a wide variety of possible worlds. Perhaps chief among these are story generation projects in artificial intelligence, as discussed by books ranging from Ryan’s 1991 Possible Worlds, Artificial Intelligence, and Narrative Theory to Noah Wardrip-Fruin’s 2009 Expressive Processing. Rather than the single successful world embedded in many game designs, these systems are capable of producing many worlds— with significant variations in the systems’ areas of dynamism—both as final outputs and in the process of generating these outputs. The possible stories that could be produced by these systems are varied and vast and are generated through sophisticated models of characters and authors alike. But in an experience that Wardrip-Fruin has termed the “Tale-Spin effect” (Wardrip-Fruin 2009, 146), this rich possibility is never translated into an audience experience. Instead, the output of most AI storygeneration systems does less to prompt the imagination of further possible worlds in the audience than a middling plain text fiction. This chapter discusses a project that combines aspects of both narrative games and story generation and, in so doing, creates an experience of exploring possible fictional worlds that is—thus far—unique. Prom Week (McCoy et al. 2012) is a game in which even a single successful playthrough results in the creation of a range of meaningfully different versions of the same fictional world. Driven by a social simulation, the gameplay of Prom Week is explicitly about imagining possible worlds of relationship among its characters and projecting how social actions could bring different types of worlds about. In other words, Prom Week not only attempts to deliver on the promise of games as machines for generating, through play, machines for producing possible worlds, it also is constructed so that engaging in this second-level thinking about fictional possibilities is key to successful strategy in the game.

Figure 5.1  Prom Week.

Playing the Worlds of Prom Week  89 In the sections that follow, we will briefly describe the fiction of Prom Week (set in the week leading up to the year’s final dance at a U.S. high school), its gameplay (which “rewinds” the week at the start of each level, inviting the player to produce a new possible week, focused on the concerns of a different character), and the underlying social simulation (which is inspired both by how social life functions in storytelling and by social science theory that is itself inspired by storytelling practices). We then describe some of what we have learned from this experiment, including the responses we have heard from audiences and what we have seen in examination of the “traces” left by tens of thousands of players. POSSIBLE WORLDS IN GAMES Possible worlds theory has been used in several discussions of player immersion in game worlds. For example, Jan Van Looy has adapted the concept of “recentering” in a fictional world (so that, for example, indexical terms refer to the fictional world) proposing a “virtual recentering” (Looy 2005), in which the player is immersed in a virtual world and reorients herself the same way a reader will when reading a work of fiction. Lisbeth Klastrup, in a paper about the “worldness” of online gameworlds such as EverQuest (Klastrup 2009), discusses some limits of player immersion. In these massively shared play spaces, Klastrup observed the way players perceive the fiction. Noting that “players... have a very conscious and instrumental approach to the world, occasionally treating it and talking about it as just a piece of software,” Klastrup found that the players’ discourses about their own play freely switched registers between actual world events and fictional world events. Klastrup’s work can be read in contrast to Van Looy’s assertions of player recentering, by calling attention to player awareness of these boundaries between worlds. In addition to her extensive work on possible worlds theory, Marie-Laure Ryan has discussed current challenges and limits for different genres of digital fiction in a 2009 article “From Narrative Games to Playable Stories” (Ryan 2009). Ryan puts forth two terms to describe approaches to interactive narrative: “narrative game, in which narrative meaning is subordinated to the player’s actions, and the playable story, in which the player’s actions are subordinated to narrative meaning” (Ryan 2009, 45). These categories exist along a continuum, but the space of digital playable stories has been significantly less explored than digital narrative games. One approach to creating responsive playable experiences is to build gameplay around a simulation, most commonly physics. An aspect of simulation-driven games is their ability to present the player with numerous choices and respond to a far greater degree than any human author could reasonably predict. This immensely expands the space of possible play traces through the game. Prom Week adopts this approach but uses a social simulation to make these diverse traces narratively significant.

90  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al PROM WEEK’S FICTION The fiction of Prom Week revolves around the social lives of 18 characters at a U.S. high school in the week before their senior dance. Inspired by wellknown high school movies, the game parodies the intense social jockeying of a memorable week for many soon-to-be graduates. Though the narrative of a Prom Week story is highly dynamic—driven by a social simulation that is in turn leveraged by the player to determine the fates of the characters— the characters themselves have statically defined backstories that determine their individual character traits and their starting social relationships with their fellow students (the “source world”). The makeup of the social landscape is always available for the player to review, but the backstory—though never changing—is not explicitly revealed to the player. Instead players learn the backstory through dialogue between the characters, who may reference past favors and grievances as motivations for their present actions. Through repeated plays of the game, players simultaneously create a new possible social and story world for the characters even as their knowledge of the source world of Prom Week

Figure 5.2  Oswald taunting Doug for one of his past actions, namely walking Jordan home after school. Learning this backstory between Doug and Jordan may help inform the player’s future playthroughs of the level.

Playing the Worlds of Prom Week  91 grows. As this knowledge grows, their ability to successfully predict how best to manipulate the social state to create a desirable destination world grows as well. PROM WEEK’S GAME Prom Week is a game about social dynamics. Players interact with the world of Prom Week solely by having characters engage in social exchanges with each other. These exchanges represent patterns of social behavior, such as having one character Ask Out another character on a date, or having two characters Reminisce about past experiences. When players select a social exchange, they are presented with a short scene consisting of animations and partially generated English dialogue (incorporating templates so that character names and the details of their current social situation can be dynamically filled in based on who is performing the exchange). The performance changes the social state of the world: the characters are now dating and they like each other more. Not every attempt at a social exchange is successful, however. A lonely romantic might pine for a date with another, but if the two do not share any common interests, the romantic is likely to be shot down. Since every action, be it successful or otherwise, furthers the narrative by affecting the relationships of the characters, gameplay is an exploration of a story generator’s outputs. The social exchanges available to the characters are determined by the social simulation system. This will be covered in more detail below, but briefly, it makes use of more than 5,000 social and cultural considerations or rules to ensure characters behave believably based on their current relationships, their personal character traits, and the previous exchanges logged in the social history. For example, characters who are friends are likely to be pleasant with each other while characters who are enemies are likely to be hostile. We say “likely” because it is rare for characters in Prom Week to have relationships as simple as being only friends or enemies; it is almost always the case that there are multiple factors informing how characters interact with each other. Sometimes these factors come from characters having a multifaceted relationship, such as concurrently being friends and enemies with each other, or “frenemies.” Other factors might involve interactions with a third party; a soured friendship due to someone asking his best friend’s girlfriend out on a date is one example. The game is divided into 10 campaigns or levels, each of which focuses on a different character—and each of which, as it begins, returns to the “source world” for Prom Week. These levels are then further divided into stages, each representing a different day of the titular week leading up to the big dance. Anything the player does in earlier stages of a level carries through to the end of the week; if a player has a character cheat on his date on Thursday, this betrayal will still be remembered on Friday. That is to say,

92  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al while each level returns to the source world, the progressive stages within a level create a new, internally consistent, possible world. Learning the consequences of this social history plays a central role in succeeding in the game, as players can begin laying the groundwork for future relationships early in the week. Having characters engage in small talk or light flirting on Monday can lead to them becoming loyal friends or doting couples by the week’s end. The cumulative nature of the social history means that longer levels become more complex due to an accruing social state. To help ease new players into the game, earlier levels are shorter than a full week (i.e., a level’s week may begin on Thursday, but the prom is always on Friday). Every level provides the player with a set of goals to potentially complete. These goals are framed in relation to the character who is the focal point for the level. For example, one of the earlier levels focuses on the character Zack, who has several aspirations. One goal is to find a date. Another is to ruin a notorious bully’s chances of becoming Prom King, because this is emotionally charged for Zack. The story of Prom Week continues even if goals are not met, but the goals provide players a guide to some of the narratively significant directions in which that level’s fictional world could move. Goals can be satisfied through an open-ended set of solutions discovered through interaction with the characters and social state. For example, to work toward the goal of getting a date, the player could have Zack form a friendship with a popular character over a shared interest. This friendship itself could eventually blossom into a romance. Alternatively, since Zack’s new friend is popular, other members of the student body could perceive Zack as a member of the popular clique as well. Popularity is admired by many characters in Prom Week and consequently would lead to certain characters who otherwise would be disinclined to acknowledge Zack’s existence suddenly becoming interested in him, in the hopes of elevating their own social status. That said, other characters—perhaps those who identify themselves as outcasts—might become less likely to accept Zack’s advances and could even view his induction into popularity as a sign of betrayal. All of these possibilities are driven by the social simulation system, and no one path is inherently more correct than another. Upon completing a level, an ending sequence is performed based on the combination of goals achieved and how the player went about completing them. Through analogy with the popular physics game Angry Birds (2009), just as there is no single correct way to hurl the birds to knock down the towers, there is no one way to create a goal social state through action. As Angry Birds is a physics puzzle game, Prom Week could be considered a “social physics” puzzle game. Once the player finishes a level and completes a sufficient number of goals, she can play the week over again with a different character. This projects another possible world with the same source world as the story that was just concluded, but one that is framed for the player in terms of the

Playing the Worlds of Prom Week  93 concerns, situations, and hopes of a different character. Play in this new world is informed by the player’s experiences in previous levels: her knowledge of the backstory, her intuition about the desires of characters, and her growing ability to successfully manipulate the social state. PROM WEEK’S SOCIAL SIMULATION Prom Week is driven by the social artificial intelligence system Comme il Faut (McCoy et al. 2014), or CiF, and is inspired by Goffman’s ­“dramaturgical” approach to understanding social life (Goffman 1959). Comme il Faut is a French phrase that translates to “Being in accord with conventions or accepted standards.” CiF is a model of social state, a collection of processes that can reason over that social state, and a framework for defining actions that can alter the social state. Though CiF is a powerful tool for social reasoning, CiF in and of itself is not a playable experience. Rather, it is intended to be used as a component of a game or system that wishes to leverage social dynamics, such as Prom Week. CiF reports what actions characters would like to take, but it is up to the game using CiF to interpret how that should be manifested to the player. Due to the emphasis in CiF on social norms and how they guide social exchanges, the representation of each character is thin. What makes characters rich and unique is their relational situation in the social world and their interconnected history. This is a direct artifact of the sociological base of CiF; the model of characters is inspired by the concept of semiotic self, where the myriad factors of history, experience, future predictions, and social forces define a malleable self that is not lost in larger societal collectives (Bakker 2011, 187–206). The system determines the most salient social influences for a character by considering a full context of social norms, ­history, and current circumstance. Every social exchange authored for a CiF story world has a single primary intent, or intended change to the social state. The intents, and thus the social exchanges, a character wants to pursue are recalculated after every social exchange in a process called volition formation. For each pair of characters, volition formation ranks all possible intents and exchanges based on a hand-authored set of social influence rules. Each rule has a weight value that adjusts volition for either a specific social exchange or for an intent (and thus a set of social exchanges) either positively or negatively. Rules are domain specific and in aggregate allow characters to behave appropriately within a specific storyworld’s social context. It is worth noting that, while the characters of a CiF world will all have their own volitions, there is no explicit encoding of theory of mind. The characters themselves do not project possible worlds based on what they believe the other characters are thinking. Similarly, there is no support for hidden information: once a social exchange transpires and relationships

94  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al change, all members of the story world will instantly take this new state into account when determining their volitions. Still, even with perfect information and no modeling of the minds of others, the potential space for possible worlds is vast. The social exchanges a character wants to engage in can be thought of as the speculations and hopes that character has for the future (e.g., “I really want this person to date me” or “I really hope I can make peace with my rival”). Each pair of characters will have its own set of desired social exchanges, and these exchanges are not reciprocal. For example, a sycophant may have a strong volition to praise a braggart, while the braggart is content to ignore his admirer. This leads to a very large branching factor in potential stories generated from CiF storyworlds; in Prom Week, there are usually dozens of choices for social exchanges at any given moment in the game. Each social exchange changes the story and the social state in different ways, which in turn leads to different sets of possible social exchanges. Prom Week uses this social simulation to model a media-derived stereotype of U.S. high school life, with some notable exceptions. The rules that determine the volitions of Prom Week’s characters were derived from an ethnographic analysis of media depictions of high school life, such as Mean Girls (2004), Twilight (2005), and Saved by the Bell (1989), though there is undoubtedly also influence from the authors’ own memories of their personal time in high school. Using these sources for inspiration, many Prom Week characters highly value things such as popularity (if characters have a status of popular, they are generally more likely to be admired by the student body), romance (characters frequently have goals revolving around getting a date for the prom), and loyalty (if a character cheats on her date, the social fallout for all three parties involved is often swift and terrible). The authors of Prom Week do not claim that this is how real teenagers preparing for their actual prom are likely to behave, but rather that the system ensures that the agents in Prom Week will remain true to character within the world of the game, which is consistent with these other media depictions of American high schoolers. Though creating a system that accurately depicts high school students would have ample applications, the stereotypes in Prom Week serve important functions themselves. For instance, they help bootstrap play through Wardrip-Fruin’s notion of the SimCity Effect. As Wardrip-Fruin writes in Expressive Processing (Wardrip-Fruin 2009, 301), players bring their own (often incomplete) understanding of the domain when they play a game, use this to inform initial play, and then expand and correct this understanding given the game’s feedback. SimCity, the simulation game that inspired the theory, places a player in the role of an all-powerful mayor. It is likely that most players have not had the opportunity to actually be the mayor of a city or devote much time to urban planning and development. However, if they find that their virtual citizens are complaining about pollution, they may recall from their own lives and knowledge that trees clean the air and

Playing the Worlds of Prom Week  95 proceed to fill their towns with parks and forests. If citizens then stop complaining about air quality, players will have learned something about the simulation framed by knowledge of the real world. With respect to Prom Week, humans are complex creatures whose actions can differ greatly from culture to culture and from context to context. To be able to gauge their values and manipulate their social standing without any past knowledge is a tall task. Thus, the authors trust players to leverage their knowledge of high school stereotypes to instantly get a sense of the general relationships the characters are likely to have and how they are likely to respond to each other. However, this also provides a telling contrast when players come to realize which stereotypes are not encoded or are inverted. For example, homophobia is not encoded in Prom Week; there are no social influence rules that take the sexes or genders of a couple into account. Though the exact qualities of a perfect romantic partner differ for every character (as individual character traits will influence the choice of a date) characters in Prom Week are generally attracted to those with shared experiences, who engage in playful flirting. As characters flirt with each other, their latent romantic affection toward each other increases, and as they engage in more (positive) social exchanges with each other, the characters will have more history to draw from when deciding how they want to interact with the other. All of these situations can happen with any character; with enough massaging of the social state, any character can become popular and thus win hearts more easily; any character can develop a crush on any other character, and thus be more willing to take the plunge and ask him or her out on a date. Although the few pre-existing relationships authored in the backstory of the world are heterosexual, if a male character is attracted to another female character, that attraction comes from his character traits and current relationship, as well as their shared social history together. If that same history and those same traits existed in a male character instead of female, the attraction would remain just as strong. We found this is counter to some players’ worldview and particularly their view of high school culture, where any deviation from heteronormativity can be met with ostracization and derision. Because of this, some Prom Week players struggle with a goal in the campaign of the character Oswald. Oswald has the goal of getting a date for the prom, and his stages include several female characters and a male character, Nicholas. Many players immediately begin by having Oswald attempt to woo the females in the level. Although nothing is impossible in Prom Week with the right social manipulation, the female characters chosen for that level have backstories that make them disinclined to be romantically interested in Oswald, making any attempts at coupling an uphill battle. By that same token, Nicholas and Oswald’s history with each other makes them naturally inclined to start dating; the player just needs to (potentially) broaden his worldview about what is possible and refine his understanding of the procedures that inform the possible worlds of the game.

96  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al

Figure 5.3  A player has successfully gotten Oswald closer to achieving his romantic goals by flirting with Nicholas.

In some sense this leveraging—and subverting—of the player’s understanding of high school stereotypes is foregrounding the principle of minimal departure (Ryan 1991, 48) through play and serving as an example of Ian Bogost’s simulation fever. In Unit Operations, Bogost writes “simulation fever is the struggle between the omissions and inclusions of a source system and the player’s subjective response to those decisions” (Bogost 2006, 132). In our case, “media representations of high school social life” is the source system and players make decisions in Prom Week based on assumptions learned from these other media. Players come into the possible worlds of Prom Week assuming that these worlds operate similarly to their own world and the high school worlds they are familiar with from other media. For the most part the game conforms to this understanding, but when it doesn’t—such as not restricting same-sex relationships that are absent in many media depictions of high school— players are forced to reevaluate their understanding of Prom Week (and its underlying social model in CiF). At the very least this can change her readings of the Prom Week worlds she has encountered thus far, but can possibly also inspire reflection on the real world by how it differs from this fictional incarnation of it.

Playing the Worlds of Prom Week  97 THE WORLDS OF PROM WEEK Play in Prom Week involves traversing an extremely large space of possible worlds and making social maneuvers within them. This manipulation of the social space is the primary story content in the kind of high school narrative we wanted to make playable. In a very real sense, the gameplay is the story. Every action the player takes advances the game’s narrative and sends ripples throughout the internal social state, which in turn affects which actions are available in subsequent turns. CiF is a partner for the player, providing the gameplay with narrative meaning and shape. This is in contrast to a “sandbox” game2 in which gameplay may be the story, but the story is formed only in the mind of the player and not understood or reasoned over by the system. We have characterized CiF as a “social physics” engine, which simulates a stylized social world the same way video game physics engines simulate a version of Newtonian physics. One could claim that due to the variety of ways players can solve a “platformer” level, it has just as many possible worlds as a CiF-driven game. There are several key differences that mark Prom Week as something new. The primary one is that while a record of a player’s jump heights and positions when traversing a standard Mario (1985) level could be considered a narrative, it is a terribly boring one. The fiction in such games is discussed at a higher level of granularity not in the individual jumps. In contrast, the individual moves in a session of Prom Week do constitute material for building a meaningful narrative. For example, one Prom Week story might begin with the character Zack flirting with the character Chloe, on whom Doug has a crush. Threatened by Zack’s advances, Doug might develop a feeling of jealousy toward him, which could result in the two of them becoming enemies. Alternatively, if Zack had at first become friends with Doug, he might be less inclined to woo Chloe, as he knows it will upset his new friend. However, since Zack has the trait “arrogant” (and presumably still is just as attracted to Chloe as he was in the original scenario), he may still have a small volition to ask her out. This would have an even greater angering effect on Doug, since he’s not only viewing his crush be taken away from him, but is being betrayed by someone that he viewed as a friend. Though, like the record of Mario jump heights and positions, any given turn in Prom Week can be reduced to a dry, logical/numeric description of the state (e.g., Doug and Zack are friends is true, Doug and Chloe are dating is false), we hope it is clear that the playthroughs presented here have the potential to have significant narrative differences from each other. Additionally, physics engines keep track of the current state of the world but not the moves that led to it. Prom Week does track history, and this gives the system greater power to construct a meaningful and distinct possible world with each social move. While the player may remember the exciting narrative from the playthrough of a Mario level, the game’s system does

98  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al nothing to incorporate it. Prom Week can meaningfully reference previous events within this play trace. This referentiality within the play trace creates a possible world that is narratively significant. THE STORIES OF PROM WEEK The nature of Prom Week, as a game played in a web browser, gave us the opportunity to store the stories Prom Week’s players were producing in the form of play traces. A Prom Week play trace is composed of the chronological order of social exchanges the player engaged in while playing the game, annotated with pertinent information such as the characters that were involved. Since, as previously discussed, the source world is consistent at the start of every play of the game, and character volitions are computed completely deterministically, the complete narrative the player experienced can be reconstructed through these catalogued play trace files. We have analyzed these play trace files in two ways, both of which revealed qualities of the system that we as the creators could not have predicted. One analysis approach is distant, looking at the range of ways the fictional worlds were shaped. The other analysis approach is close up, looking at how a particular exchange of dialogue can play different roles in the fictional worlds of different players. The former is looking at the shape of play traces, while the latter examines how particular dialogue appears in context. Both are ways of trying to understand the different worlds Prom Week creates through audience interaction. STORIES AT A DISTANCE Analyzing play traces generated from real play situations enables evaluating the impact players have on possible worlds projected through their unfolding stories. Even with the large amount of variation supported by CiF in a story world as content-rich as Prom Week, players could potentially be exploring only a small area of the potential story space. To gain a better understanding of the variation in stories that players experience through Prom Week, a holistic and detailed understanding of the play traces is useful. To get a sense of how CiF’s simulation and Prom Week’s gameplay impact the actual choices presented to the player, level traces were analyzed and visualized using the Façade Log Analysis and Visualization Tool (Sali and Mateas 2011; Sali 2012) a visualization tool that aims to enhance the current toolset for studying interactive narratives. This tool helped in forming an understanding of how players were interacting with the released version of Prom Week. Even though the player has many options of social exchanges to choose from, it is not clear without evaluation that there are enough paths through the story space to create meaningful reflections of the play of each individual player.

Playing the Worlds of Prom Week  99 Furthermore, story goals, level casts, and the desires of the characters themselves may restrict the options available in such a way that many players will be forced down a narrow few paths in their pursuit of story goals. We found there was a very large degree of variation in the possible worlds players were able to create through Prom Week. We took a sample of 263 play traces of the final stage of the character Simon’s level and discovered that no two were exactly alike; the space was rich enough to allow for an entirely unique play trace per player. Though several play traces of that stage began similarly, it only took any given player about four social exchanges to create a world experienced by no one else, unique to her.

confide confide in in 40 40

share share interest interest 21 21

co

share interest 4 1

4+

share interest 1

confide in 3

flirt 1

idolize 1

share interest 1

confide confide in in 11

woo 1

idolize 1

pick-up line 1

confide in in 11 confide

woo 1

insult friend of 1

give advice 1

bicker 1

Figure 5.4  Play trace graph showing how often each distinct path through Simon’s story was traversed (shown by the number associated with each node, emphasized with color). The large band of nodes seen at the top of the diagram represents approximately one third of the total size of the complete graph. The cutout shows a section of the map in detail including examples of social exchanges (like “pickup line” and “confide in”) that appeared in more than one play trace. The majority of play traces are unique.

Even traces with subtle differences in gameplay actions (for example, the sequence of social actions Reminisce, Confide in, Ask Out as opposed to Confide in, Reminisce, Ask Out) can result in two remarkably different possible worlds, as the ordering of these social exchanges could potentially drastically alter the actions the characters want to take. The general trend of play traces becoming unique for players held true across all levels of the game and is particularly evident in the more difficult later levels. Though play traces with similar actions in different orderings have the potential to lead to very distinct worlds, intuitively they are likely to be less distinct than two play traces that involve entirely different social exchanges and

100  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al characters. Some work has been done to automate the process of determining how dissimilar two play traces are from each other (Osborn et al. 2014). In other words, steps have been taken to chart the universe of Prom Week’s possible worlds, determining how close or distant any two worlds are from each other. Mapping this space is valuable for several reasons. Seeing the clustering, if any, of Prom Week’s possible worlds is interesting in and of itself and highlights whether there are certain high-level strategies that are employed by many players. Finding the outlier worlds—worlds that greatly differ from any other—and analyzing their corresponding play traces can reveal insights into particularly unique approaches to the game. This in turn can help answer important game design questions, such as whether Prom Week enables creative or subversive play. Comparing and contrasting the completed destination worlds generated by players is one way to illuminate the expressivity of the system. However, additional details about the possible worlds of Prom Week can be gleaned by narrowing our focus down to the level of the dialogue generated by the system. STORIES UP CLOSE To illustrate the range of roles that an element of Prom Week can play in the construction of different possible worlds, the following section provides interpretations of play traces of actual players involved in a single social exchange outcome (of which Prom Week has more than 800). This is an example of the “story sampling” approach to understanding widely varying narrative systems (Samuel et al. 2014). While it cannot be said for certain that the player that generated the gameplay interpreted the sequence of social exchanges in the ways described, the provided interpretations are indicative of the dramatic social drama Prom Week is able to create because story content is directly linked to the social state managed by CiF. The example is a possible outcome of the social exchange Ask Out, in which the responder tragically refuses to cheat on the person he or she is already dating—despite sharing feelings for the initiator of the social exchange. This scene of dialogue will be referred to below as the “tragic rejection.” The uninstantiated dialogue for the tragic rejection is: INITIATOR:  RESPONDER,

I have a proposition for you. Hear me out. I know you’re dating DATING_RESPONDER, but you need someone who really understands you and cares about you, too. Someone … like me. RESPONDER:  Oh, INITIATOR. I would be lying if I said I didn’t sometimes wonder whether we could be together. But I just can’t right now. I’m with DATING_RESPONDER, and I couldn’t cheat on PRONOUN_OF_RESPONDER. It just wouldn’t be right.

Playing the Worlds of Prom Week  101 INITIATOR:  But,

we’re so right for each other! Can’t you see we were meant to be together? RESPONDER:  Stop it. Just stop. We can’t. We just can’t, okay? You should go. INITIATOR:  I’ll always be here if … if you change your mind. When put into the context of particular characters and varying social history, this scene of dialogue takes on a variety of narrative roles and meanings. One interpreted story from a play trace involves the characters M ­ onica and Nicholas (who are dating) and Cassandra. The story begins with ­Nicholas trying to break up with Monica and Monica talking him out of it. Next, Monica, feeling rejected, makes several romantic moves toward Cassandra, to which Cassandra reacts poorly. Nicholas, feeling jealous ­ and irritated at Monica, chews her out for flirting with Cassandra. Now, the tragic rejection: Cassandra, given some time to think about Monica’s advances (which were confusing to her at first), asks Monica out. Monica, feeling bad for making Nicholas angry, has decided she doesn’t want to be a cheater, and refuses. In this story, the tragic rejection has taken on narrative meaning that neither the system nor the authors anticipated, though is consistent with the characters and the choices the player has made. In the context of the story, Monica’s torn refusal to date Cassandra implies that Monica felt remorse for her previous flirtatious behavior. It also paints her earlier actions with Cassandra as baiting Nicholas into caring about her. Once Nicholas showed her he cared, by getting angry with her, her desire to get together with ­Cassandra was lessened. This interpretation of the tragic rejection reveals Monica to be manipulative and gives us reason to pity Cassandra. Another play trace involves the characters Doug and Jordan (who are dating) and Chloe (who is friends with Jordan). It begins with Doug and ­Jordan having a tender moment where Jordan reveals something embarrassing about herself, and they talk about how they need to trust one another. Next, Chloe flirts with Doug, and he responds politely. Next, Jordan goes to confront Chloe about this, and she can’t bring herself to be mad at her friend. Then, right after Jordan compliments Doug, Chloe tries to date Doug, and the tragic rejection plays out (where he admits to having feelings toward her but ultimately rejects her). In this case, the tragic rejection can be interpreted as revealing both Doug’s weakness for romantic attention and his loyalty to his present romantic partner. The space of possible worlds in most single-player video games is highly limited by the amount of pre-authored narrative content. As discussed above, CiF enables a very wide degree of narrative responsiveness during play. Play trace data collected from Prom Week suggests that these possible worlds not only exist, but they are also being explored. While an individual stage might present only a few initial options, the possibilities branch out

102  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al extremely quickly. After just a few moves, players are almost all in their own unique world. Given the narrative significance of these differing worlds, CiF enables numerous playable stories. CONCLUSION We may be seeing the emergence of a new kind of experience of fiction, based on play with a story generation system that projects possible worlds. As we write, we believe Prom Week is the first example of this. The second, Ice-Bound, is forthcoming—it makes story authoring the metaphor for this form of play and is co-created by Aaron Reed (Prom Week’s lead author) and Jacob Garbe. Other projects are in the works. While only time will reveal the impact of this approach, our experiences with Prom Week are already giving us a sense of some of what may be significant. We have suggested a few aspects above. For example, how making each level of Prom Week a new version of the same week creates a fictional experience that foregrounds possibility spaces (rather than, say, foregrounding fate). Or how the exploration of possibility spaces slowly, and incompletely, allows audiences to build up an understanding of the underlying rules by which these fictional worlds are created (which may be at odds with their understanding of the genre or the world). In this final section we would like to discuss one additional experience some players have through the combination of play and story generation. We will begin with play. In many forms of play, as players understand their situation (consciously or not), they take actions that they feel are their own, rather than fully dictated by the game’s designers. This could be executing a long, deep Go strategy or lunging for the ball in Tennis. Naturally, players of Prom Week feel that the choices they make about how characters will interact are similarly their own. But in Prom Week this can combine with experiences that are more common with fiction. An audience member may feel empathy with a character’s situation or speculate about how one character might react if a second character takes some action with a third character, exercising her own theory of mind abilities in the fictional context. And then Prom Week’s underlying social situation enables the creation of a fiction, based on the audience member’s actions, that reflects insights gained through these kinds of engagements with the fictional world. As a result, we have seen a new kind of fictional experience reported by Prom Week players: a feeling of responsibility. Reports have come from a variety of players. For example, as Craig Pearson wrote on the games website Rock, Paper, Shotgun: I presumed I’d need to be nasty, but that route got me nowhere. Not that it wouldn’t have worked, and horribly it makes me want to see

Playing the Worlds of Prom Week  103 if I could destroy Buzz, but I won the game by accidentally being nice and friendly. So now I feel bad and impressed, and want to play it all over again ... Next time I’ll be looking at more upbeat solutions, because the alternative, frankly, is hating myself. (Pearson 2012) And similarly, as Alastair Stephens, co-host of the Storywonk podcast, wrote: The complexity of these relationships is absolutely, intricately mechanical—but like all successful stories, it swiftly moves beyond the mechanical, beyond the ludic, to the personal and emotional. The temptation to manipulate these characters is enormous, but crossing that line feels… wrong… In the end, I stopped playing Prom Week because I didn’t like the person I felt like when I played it, and I can think of no greater compliment than that. But I’ll be back tomorrow, Simon. You and me, buddy. You and me. (Stephens 2012) It might seem odd to conclude this chapter by quoting two people saying they felt bad about themselves after experiencing Prom Week. But it points to something important about the playful projection of possible worlds as an experience of fiction. We might feel bad after watching The Bicycle Thief (1948) or The Wire (2002). We might feel bad because of the interconnected empathy we have with the characters and understanding we develop of a social system. But we don’t feel a sense of personal responsibility for how we decided to engage with that system and shape the lives of those characters. Audiences do feel those things with Prom Week. Perhaps this is most obvious when they feel bad about their choices. But it also operates when they feel joy—as one player reported to us when he successfully inverted the high school’s popularity structure and saw the results of his inventive strategy in characters’ lives. If it turns out that Prom Week is the first example of an experience of fiction that will grow and diversify with time, we believe this potential for a feeling of responsibility may be a key reason. NOTES 1. Fiction here refers to the non-real components of a game—such as its setting, characters, and plot—with which the player interacts through the very real rules the game imposes (i.e., gameplay). Elsewhere in this volume, particularly in ­Hatavara’s chapter, fictionality is discussed in a different, fine-grained manner. For more detail on how the term fiction has been used in game studies, see Jesper Juul’s Half-Real (Juul 2005). 2. “Sandbox” games de-emphasize linear goals in favor of players’ free movement through a simulated world. Also known as “open-world” games.

104  Ben Samuel, Dylan Lederle-Ensign, Mike Treanor, et.al REFERENCES Angry Birds. 2009. Created by Jaakko Iisalo. Rovio Entertainment. Bakker, J I. 2011. “The ‘Semiotic Self’: From Peirce and Mead to Wiley and Singer.” The American Sociologist 42.2–3: 187–206. doi:10.1007/s12108-011-9140-3. The Bicycle Thief. 1948. Directed by Vittorio de Sica. Italy: Ente Nazionale Industrie Cinematografiche. Bogost, Ian. 2006. Unit Operations: An Approach to Videogame Criticism. ­Cambridge, MA: MIT Press. doi:10.1111/j.1467-954X.2007.00687_7.x. Eco, Umberto. 1984. The Role of the Reader: Explorations in the Semiotics of Texts. Bloomington: Indiana UP. Façade. 2005. Created by Andrew Stern and Michael Mateas. Portland, OR: ­Procedural Arts. Goffman, Erving. 1959. The Presentation of Self in Everyday Life. Garden City, NY: Doubleday, Anchor. Juul, Jesper. 2005. Half-Real: Video Games between Real Rules and Fictional Worlds. Cambridge, MA: MIT Press. Klastrup, Lisbeth. 2009. “Game Studies - The Worldness of EverQuest: Exploring a 21st Century Fiction.” Game Studies 9.1. http://gamestudies.org/0901/articles/ klastrup. Looy, Jan Van. 2005. “Virtual Recentering: Computer Games and Possible Worlds Theory.” Image & Narrative, no. 12. http://www.imageandnarrative.be/inarchive/ tulseluper/vanlooy.htm. McCoy, Joshua, Mike Treanor, Ben Samuel, Aaron A. Reed, Michael Mateas, and Noah Wardrip-Fruin. 2014. “Social Story Worlds With Comme Il Faut.” IEEE Transactions on Computational Intelligence and AI in Games 6.2: 97–112. doi:10.1109/TCIAIG.2014.2304692. Mean Girls. 2004. Directed by Mark Waters. United States. Meyer, Stephenie. 2005. Twilight. New York: Little, Brown and Company. Osborn, Joseph C, Ben Samuel, Joshua McCoy, and Michael Mateas. 2014. ­“Evaluating Play Trace (Dis) Similarity Metrics.” In Proceedings of AIIDE 2014. Raleigh, NC. http://games.soe.ucsc.edu/sites/default/files/glz-eval-FINAL.pdf. Pearson, Craig. 2012. “Impressions: Prom Week.” Rock Paper Shotgun. http://www. rockpapershotgun.com/2012/02/16/impressions-prom-week/. Prom Week. 2012. Created by Josh McCoy, Mike Treanor, Ben Samuel, Aaron Reed, Noah Wardrip-Fruin, and Michael Mateas. Center for Games and Playable Media. http://promweekgame.com. Ryan, Marie-Laure. 1991. Possible Worlds, Artificial Intelligence, and Narrative Theory. Bloomington & Indianapolis: Indiana University Press. ———. 2001. “Beyond Myth and Metaphor - The Case of Narrative in Digital Media.” Game Studies 1.1. http://www.gamestudies.org/0101/ryan/. ———. 2009. “From Narrative Games to Playable Stories: Toward a Poetics of ­Interactive Narrative.” StoryWorlds: A Journal of Narrative Studies 1.1: 43–59. doi:10.1353/stw.0.0003. Sali, Serdar. 2012. “Playing with Words: From Intuition to Evaluation of Game ­Dialogue Interfaces.” PhD. diss., University of California, Santa Cruz. Sali, Serdar, and Michael Mateas. 2011. “Using Information Visualization to ­Understand Interactive Narrative: A Case Study on Façade.” In Proceedings of the Fourth International Conference on Interactive Digital Storytelling. Vol. 7069. Vancouver, Canada. doi:10.1007/978-3-642-25289-1.

Playing the Worlds of Prom Week  105 Samuel, Ben, Josh McCoy, Mike Treanor, Aaron A. Reed, Michael Mateas, and Noah Wardrip-Fruin. 2014. “Introducing Story Sampling: Preliminary Results of a New Interactive Narrative Evaluation Technique.” In Foundations of Digital Games. Ft. Lauderdale, FL. http://www.fdg2014.org/papers/fdg2014_wip_18.pdf. Saved by the Bell. 1989. Created by Sam Bobrick. United States: Rysher Entertainment. Stephens, Alastair. 2012. “Prom Week.” http://alastairstephens.com/prom-night/. Super Mario Bros. 1985. Created by Shigeru Miyamoto and Takashi Tezuka. Nintendo. Wardrip-Fruin, Noah. 2009. Expressive Processing: Digital Fictions, Computer Games, and Software Studies. Cambridge, MA & London: MIT Press. The Wire. 2002. Created by David Simon. USA: HBO.

6 Scripting Beloved Discomfort Narratives, Fantasies, and Authenticity in Online Sadomasochism J. Tuomas Harviainen

This chapter discusses narratives found in sadomasochist activities. ­“Sadomasochism” is an umbrella concept (as are its many synonyms with varying connotations, e.g., BDSM). Under that umbrella are found various practices that deal with pain, restraint, humiliation, and so forth in a consensual manner (Nordling 2009). They are based on a temporary power exchange, in which the practitioners’ normal social roles are replaced by those of a fictional situation and in which some participants (one or more) are given power over others for either a short period of time (a “session”) or, rarely, for a longer, undefined duration (Dancer et al. 2006). The purpose of such exchanges is pleasure, which for many is of a sexual nature (Brandhurst 2011) but does not always have to be (Newmahr 2011). Individual practitioners, however, have highly individualised desires, starting from whether they prefer dominant or submissive roles, or both, and then moving on to which of the activities they find pleasurably discomforting and which just unpleasant. For example, one submissive may enjoy even very harsh pain but hate all humiliation, whereas another prefers verbal and social disgrace but is not into pain (Harviainen, in press). According to an often-cited cluster definition by Weinberg, Williams, and Moser (1984), five key components are usually present in sadomasochism. Not all of them are needed for an activity to constitute sadomasochism, but they are typically found together. These five are: 1 The appearance of dominance and submission; the appearance of rule by one partner over the other. 2 Role playing. 3 Consensuality, that is, voluntary agreement to enter into the interaction. 4 Mutual definition, i.e., a shared understanding that the activities constitute SM or some similar term. 5 A sexual context, though the concept that SM is always sexual is not shared by all participants (Weinberg, Williams, and Moser 1984, 380–81). The umbrella activities likewise tend to be clustered in probability (Alison et al. 2001). This means that certain things are more likely to be found with certain others or learned consecutively. For example, whipping and

Scripting Beloved Discomfort  107 spanking tend to be practiced by the same people, and it is likely that people who are into anal fisting have also at least tried anal sex. To an external eye, sadomasochism resembles violence and shares many of its traits and acts, but in a fiction-based, consensual manner (Harviainen 2011). This transforms the normal logic and moral of things—pain is for example “given” to a partner, as it is welcome and pleasurable, rather than “inflicted,” and not to hit one’s partner during a session may be the ethically wrong thing to do (same as with boxing, for example). Rosemary Jackson (1981) calls this situation “paraxis”: a transformation of the ordinary path into a dislocated replacement of the normal, familiar logic of things, to a new temporary world of fantasy. In sadomasochism, practitioners voluntarily play with transgressive themes, ones that outside of the proper context would constitute violence (Jenks 2003). This is the same kind of difference as that between rape and consensual sex. Just like most of those who may enjoy rape fantasies do not actually wish to be raped (Zurbriggen and Yost 2004; Stear 2009), consensual sadomasochists do not find non-consensuality arousing (Newmahr 2011). One’s inclinations furthermore do not escalate from consensuality into the non-consensual, despite several popular myths claiming so (Nordling 2009). Therefore, by its very nature as an exemplarily playful, transgressive paraxis to violence, sadomasochism inevitably takes place in a virtual world. In this chapter, I analyse the way in which the nature of such v­ irtual worlds, as they range from text-based imagination to physical role-play ­fictions, affects the way its narratives are created and experienced by its practitioners. The analysis is based on a review of existing literature, as well as two decades of ethnographic research I have conducted within the F ­ innish ­sadomasochist scene (see e.g., Harviainen, in press, for details). NARRATIVE SEMI-SCRIPTING As noted by Carol Siegel (1995), sadomasochism is innately narrativistic. This goes beyond the human mind’s habit to construct narratives in retrospect out of even unrelated elements (see Gazzaniga 1998). Games and play, sadomasochist sessions included, do consist of narratives that are properly perceived only in retrospect but which may also be consciously constructed during play (Myers 2010, 93). As event-action sequences by active agents, which are in their way “written,” and can be later “read” (Ricoeur 1988), narratives build up as the play takes place, and end at the point where the activity ceases (and may, at that time, become parts of other, larger narratives). Following Herman (2002), this chapter understands sadomasochist narratives to be both the construction of a perceived continuity from the events that take place in relation to—before, during, and after—a sadomasochist encounter, as well as the communication of that experience to (at least) one’s session partner.

108  J. Tuomas Harviainen Practitioner preferences determine how explicitly a narrative desire is stated out before play, and it may vary even for the same individuals on a case-by-case basis (Weinberg 1978). The structure of a session—a ­physical or a virtual one—is supposed to turn the activity into a set of narratives, a process that is emphasised by the fact that the activity is set apart from mundane life and has a clear beginning and an end, usually followed by a semi-phase of aftercare and discussion (see e.g., Newmahr 2011). An underlying fantasy is what creates any game-like play, by being the narrative framework of challenges that the activity presents. In sadomasochism, that fantasy is the ability to give and receive consensual pain, in a fantasy setting, where a semi-scripted narrative-purpose-defined role works as a safe filter for bracketing facets of one’s and one’s partner’s empathy, selfvalue, or even humanity (Stear 2009). There is also an obviously theatrical quality to sadomasochism at even a superficial glance, as noticed by both those who have observed it from the outside, as well as those who have immersed themselves in it (see Deleuze 1967, and Mains 2002, respectively, for examples). Its practitioners play accentuated roles, often using very visual paraphernalia; the play has a clear beginning and end; and so forth. Most importantly, the activities have set parameters and often pre-planned stages, very much like that of a loose theatrical script. A current of skilled improvisation flows through a sadomasochistic session. This does not mean that the activity would be completely free-form. On the contrary, as Alperson (2010, 273–74) states, ­“Improvisation is not completely free or autonomous activity. I­mprovisation depends fundamentally on routines, rituals, and practiced activities of all sorts. The point is rather that, in improvisational activity, freedom seems to be on display in the spontaneity of the activity.” To improvise a performative action is to be goal-directed yet make decisions that alter the activity as it is being performed. Paraphrasing Herman (2002), the semi-scripting of a sadomasochist session can, in somewhat simplified terms, be considered to be like a recipe. First, the participants choose a dish (say, pizza) to cook then discuss potential toppings (and the toppings that they definitely do not want to have included). Then, the cooking itself takes place, maybe with some more discussion of options taking place during it. Finally, there is a cool-down period, after which the participants enjoy the results of the action, together. Likewise, session participants choose a setting, elements that they want to include or to exclude, and then improvise based on that outline. Afterwards, they share in the aftercare and the afterglow. With experience, one is also able to guess what one’s partner would like that session to contain, similarly to improvising with a cooking recipe (see e.g., Ortman and Sprott 2013). As noted by Herman (2002, 88–89), a recipe is not a narrative, but rather a set of imperatives and instructions, yet its results can form narratives. Given the theatricality, however, I believe the concept of a semi-script to be more accurate than a “session recipe.”

Scripting Beloved Discomfort  109 Intriguingly, as one notices that the similarities with drama go beyond the surface, one also starts noticing subtle differences. First and foremost, the scripting and improvisation system of sadomasochism is different from both more traditional, scripted drama as well as from improvisational theatre. As noted by one of its key thinkers, Keith Johnstone (1981), improvisational theatre has sought to move away from the script in order to focus on the structure of play. The umbrella nature of sadomasochism, combined with individual preferences and limitations (e.g., preferred acts, whether the session may include penetration, and so forth), however, forces the participants to stick to parameters that create at least somewhat coherent narratives out of the activity. Otherwise, the play feels neither safe nor enjoyable. Because of this, Mains (2002) has suggested that instead of the theatrical, we should rather compare sadomasochist sessions to psychodrama. however, one compares the acts’ structures to, say, the short semi-­ If, ­ improvisational etudes of Stanislavski (2010), the narrative similarities are still more obvious than those of a directed psychodrama. Likewise, we can find apt comparisons to the narratives of sadomasochism in games (­Harviainen 2011) and in semi-improvised, semi-scripted forms of television, especially some types of reality TV: in both, participants act as if they had freedom of activity (excluding, of course, the potentially controlled state of a submissive), but are actually bound by goals, expected story arcs, rules, and so forth (see Torner 2012). I, however, believe that the closest narrative analogy can be found elsewhere: in music. While the participants of a sadomasochist session indeed perform theatrical acts, often viscerally, the narratives through which they do so are closest to those of improvisational jazz, played together. As ­Alperson (2010, 276) puts it, “Improvisers demonstrate skills in creating musical passages that have both short and long form interest and coherence. That is, there are skills involved in creating both interesting phrases and longer, extended musical statements in which the shorter passages are thematically integrated.” This is how the semi-scripts of a sadomasochist session function: some of them bring consistency and continuity into the main narrative of the session, while others (e.g., a singular instance of whippings, before moving on to spanking) form the “interesting phrases” (or, maybe, rather “phases”). The fictional and the actually experienced blur in that process. Paraphrasing Bruner (1990, 44), the sadomasochist narrative is determined by the sequence of events, not the truth or falsity of its parts. Likewise, present is the social aspect of giving enough room for each player’s contribution. To not do so marks one as a selfish person, an unfit partner for further play. An escapist root has furthermore been suggested to both activities, but little evidence exists to support that conclusion, at least in a generalizable manner (see Carvalho 2010, and Baumeister 1988, respectively, for examples of these hypotheses). Rather, while some practitioners may indeed seek escape from themselves or their daily pressures, most sadomasochists appear to have more transgression-pleasure oriented goals, those

110  J. Tuomas Harviainen of playful, artificial challenges (see e.g., Nordling 2009). Because of this, they stay aware of the semi-scripts. In a typical session, multiple semi-scripts are at play, each forming facets of the narratives that are constructed. Some of them exist as contradictions. At the core, for example, are two seemingly mutually exclusive semi-scripts: the essential fact that everything takes place voluntarily and consensually and the essential fact that the participants need to emphasise the fact that one side has power over the other. The latter, in some cases, may also include semi-scripts of pretending non-consensuality (e.g., interrogation or rape fantasies). The other difficult pair is the way that the activity is obviously performative, given the presence of roles and the altered social dialectics, yet it has to feel authentic. Therefore, enough scripting has to exist to support the roles and the power hierarchy, but not so much that the activity starts to feel like theatre. Certain sessions form an exception to this rule: psychodrama-like and often not sexually oriented, in them the submissive participant seeks to work out some trauma or stress through the play, leading to just the dominant partner’s being in an accentuated role (since if he or she were not, the situation would potentially constitute abuse). Likewise, in most sadomasochist sessions are present multiple s­exual scripts (see e.g., Alison et al. 2001; Brandhurst 2011). These types of “scripts” are behavioural patterns and narrative expectations that people bring into social situations (in this case sexualised ones), not formalised scripts in the theatrical sense. They affect behaviour but are not fixed and are adjusted according to the way events develop. For example, a lot of people have a script in their minds of how an optimal date situation would play out. As sadomasochism is fantasy play, people bring also into those encounters these semi-ready expectations and presumptions. Those combine with the persons’ sexual preferences, limitations, and so forth, creating a mesh of limits, hopes, and desires—in other words, strong narrative expectations. Going against another participant’s preferences in a session causes a breach in both the expected and the perceived narratives. If a personal limit is broken (accidentally or not), the fictional world of the play is shattered. In such a case, the activity has slipped into violence, although potentially unintentionally so. Nevertheless, sadomasochists play with those borderlines, to keep up the transgressive nature of the activity, at least in their own minds (Harviainen, in press), yet they rarely cross them intentionally. Given this multitude of semi-scripted narrativity, it may seem difficult to accept the fact that sadomasochistic encounters may nevertheless have a central narrative. If we however see the play as forming a set of possible worlds, bound together as a temporary system, it starts making more sense. Those inside the system pretend that it is intact and its boundary strong, because a part of that fiction is to pretend that it is not fiction, i.e., that there is no boundary to another, more real, world. The boundary is nevertheless porous, and what leaks in to the fiction are the semi-scripts that set constrictions to the narrative (see Harviainen 2012). The actors pretend to

Scripting Beloved Discomfort  111 be free in action, but they are not. This is further emphasised by the fact that the system is allotelic, i.e., directed toward an external goal: the transgressive pleasure of the participants themselves, not just that of the roles that they play. The way the system’s narratives and possible worlds are constructed is easiest to understand through two concepts taken from art theories. First of all, the session’s groundwork, including its semi-scripting, forms through what Stanislavski (2010) calls “given circumstances”: the setting of a performative piece. Some of those given circumstances are formally explicated (“you are the slave, and I am the dominant”), others can be implicit (such as a participant’s sexual scripts, unless they are discussed beforehand). From these story seeds grows the activity, with other semi-scripts guiding to what directions it grows. The second key concept is Allan Kaprow’s (1966) idea of Chance. Chance artworks are things that are allowed to change (or even be changed by people who come in contact with them), and whatever is the result of the process is to be treated as if that were what the artwork was supposed to become all along. In narrativist terms, a Chance piece is given its primary narrative only post de facto, but that narrative is to be seen as fully intentional and pre-planned. In a similar fashion, a typical sadomasochist session (physical or digital) is only given its meaning and story (actually several, as will be discussed later in this chapter) after the activity itself has been concluded. The platform where one practices sadomasochistic activities has some impact on this, however, due to which we now turn to examine some key types of sadomasochist play that takes place in the digital realms. PLATFORM VARIATION As discussed above, sadomasochist sessions, regardless of where they take place, share some traits. Central to those are the voluntary nature, the presence of pretence, and the semi-scripted basis of narrative formation. Several others exist as well: the participant expectation that the fiction be preserved as much as possible, the focus on an embodied experience, and a fantasy basis. In hermeneutical terms, the primary framework for interpreting the activity at hand (like in other forms of role-play; see Harviainen 2008) is the fictional setting, even as the physical body may exert force, say the words, experience pain, and/or feel excited. This has been confirmed by several studies, also outside of sadomasochism. For example, Australian trainee firemen interviewed by Lloyd ( 2007) stated that no matter how realistic a physically simulated fire is, they know the setting to be artificial and therefore do not react to the stimuli the same way as they would to a real fire. In the same manner, knowing the situation to be visceral yet fundamentally fictional, sadomasochists in physical play do not react to acts that they perform or experience in the way that they would to similar ones outside of play.

112  J. Tuomas Harviainen The further one goes from a direct, unmediated physical play (as described by Moser 1998 and Newmahr 2011, for example) into the realm of the virtual, the more things change. Reliance on pre-planned scripts increases and the activities take on a more admittedly performative nature. With the mediation comes also an unavoidable emotional distance from the events that are taking place. The advantage to this is that it makes bracketing parts of one’s partner’s humanity—an essential component to being able to enact the sadomasochistic power dialectic—easier. The disadvantage is that instead of immediate, visceral feedback, one gets only mediated pleasure and pain, experienced with imagination rather than as directly embodied. As the activity is still interactive, the difference is not as drastic as that between, say, ­sexual intercourse and watching pornography, but it does have a direct impact on things such as arousal and gratification. This is especially obvious in sadomasochistic scenes conducted online via chat, forums, emails, or messages. As Cross and Matheson (2006) found out, such play closely resembles table-top role-playing games such as Dungeons & Dragons. Through shared narration, participants imagine the events that are taking place. There is even a kind of shared game mastering function present, because participants have by social contract executive scripting power relating to parts of the shared narrative’s creation (Harviainen 2011). One’s presence in the activity is dissociative, but the effects of personally interpreting the shared narrative can still be felt in an embodied manner, especially if the role-play is limited to power dialectics instead of to ­fictional characters—or if the characters match the participants’ preferences suitably: for example, a persistent anecdote says that if one sees two lesbian strangers playing in a sadomasochist chat room, they are both more likely than not actually men, off-line, but never admit to that fact online. Chat and other text-based forms of sadomasochist play are simultaneously allotelic (performed for an external purpose) and autotelic (selfrewarding). In other words, they have purposes and goals both outside and inside of their fiction. One of the key purposes is the construction of a believable, intriguing shared narrative containing the right elements, as that is the thing that gives the most pleasure to the participants outside of the fiction. The exact type of narrative changes depending on the motives of the participants. If, for example, the online play is enacted as a substitute for physical play due to not being able to meet the partner for some reason, for the sake of staying safe with a person one does not want to risk meeting in real life, or for staying without physical marks, the narrative is likely to resemble that of a physical session, just in a mediated fashion. In turn, if the interaction takes place online (or through text messages, etc.) as a form of foreplay, the narrative may be much more fragmented, consisting more of promises, friendly threats, and other elements that build up the atmosphere for an intended physical session in the future. Visible logs, especially on shared virtual spaces such as community forums, can furthermore be used by certain practitioners to establish expert status within sadomasochist

Scripting Beloved Discomfort  113 circles. By showing the complexity of their written play, combined with its contents, they seek to point themselves out as skilled and in-the-know, the same way practitioners of many other types of play do (see Harviainen, Gough, and Sköld 2012). Expectations of future sessions can also be perceived in another type of play, that of sadomasochists’ online dating. While at first glance maybe not obvious as play, dating potentially includes so many game-like elements that it is fruitful to analyse those, because the conventions carry over into relationships between sadomasochists and the way they conduct their sessions. Pick-up artists do not call what they do a “game” by accident (see Strauss 2005), and the same applies, I believe, to sadomasochist dating. This is due to the presence of power dialectic related preferences and the corresponding role-accentuation. According to my observations of online profiles and interviews I conducted for an earlier work (see Harviainen, in press), sadomasochists, in their dating, often strongly play up their preferences, in order to arouse interest and to stand out from the crowds. This basically follows the principles of “narrative identity,” the way one presents oneself differently in varying situations, to the level where one’s very identity can be perceived to change based on contextual narratives (Holstein and Gubrium 2000). What makes the identity here a game-like one is that the system is designed to function like a non-competitive ping-pong match, hopefully creating a shared narrative: focus is on the self-presentation itself instead of on the presentation’s being a tool for handling a social situation. The “game” has an allotelic, fiction-external goal, that of a fruitful rendez-vous, a potential relationship, or both. The problem area with this particular type of online identity play is that whereas it may be common to present oneself in a better light in typical online dating situations, sadomasochists are particularly keen on honesty and authenticity (see Mortensen 2003, 223–25, and Ortmann and Sprott 2013, 19–20, for examples), because of safety. The activity carries a social stigma (see Goffman 1963) due to its association with non-standard sex and (wrongly) with violence. Because of this, many practitioners do not want to share their real names or contact information. When this semi-anonymity is combined with the very tangible risks of physical, possibly sexual play with a ­ randhurst new acquaintance, honesty and authenticity become crucial (see B 2011 for examples of risks). Likewise, as practitioners value skills and experience, because those ensure both safety and a higher probability of an enjoyable ­ arviainen 2012). encounter, honesty is in high demand (Newmahr 2011; H However, if one tries to push the expected narrative and the identity too directly into the self-presentation, problems arise. For example, coming off directly as a commanding male dominant is more likely to lead into being filed under “creep” than to a good session that flows from the already-roled initial contact. Rather, one needs to playfully present oneself as a potential commanding male dominant. Likewise, responding in direct submission at

114  J. Tuomas Harviainen once to a stranger sets off its own set of warning signals. This is why the presentation needs to play with the edge of the fiction—this time from outside the possible worlds, the future creation of which it promises. It is also the reason its semi-scripts need to stay loose. In a way, the online sadomasochist dating game is the polar opposite of a chat session, in the way in which it plays with narrative and authenticity. SADOMASOCHISM IN VIRTUAL WORLDS A third variation can be found in virtual worlds, such as Second Life and World of Warcraft, where sadomasochists have both clandestine encounters and virtual communities. In them, the character of the avatar becomes a medium for the person’s creativity, expression, and experiences (see Roine this volume). While lacking the immediacy of a physical encounter, such spaces offer a significant number of advantages. For example, one is able to change one’s presented gender, race, species, and so forth, and thus enjoy encounters that would be impossible in the real world. Likewise, the environment may allow magic (including a regular favourite: healing potions, used for both extending play and for aftercare), flight, or other altered physics. Furthermore, of particular interest to certain practitioners, virtual worlds grant access to playing themes that would be highly unethical to explore in the physical realm, including rape fantasies and even death scenes (Brown 2013). As these worlds are persistent, i.e., exist and develop even when one is personally not using them, they also provide a sense of continuity and of emerging, new opportunities for further play. On one hand, sadomasochist play in such worlds requires additional preparation in the form of more thoroughly explicated semi-scripts or plans. This is particularly true in the case of environments like World of Warcraft, where the end-user license agreement forbids erotic role-play. There, players have to clandestinely choose a location and conduct the activity via symbolic gestures and through private messages, effectively turning the situation into a kind of chatplay, as described above. What, despite the rules, drives people to play out sadomasochist scenes in World of Warcraft is primarily the fact that they play it anyway and want to use the same, familiar platform. Another influence is the fact that, again, playing against the license agreement is by itself a transgressive act. In contrast, communities in Second Life can openly engage in their preferred activities and even construct settings for that very purpose (see Frank 2013). For example, Sixma (2009) describes a specific setting where people act out male-dominance fantasies based on the books of John Norman. In addition to peer support and suitable visuals, such communities may also allow people to live according to transgressive morals: whereas male sadomasochists normally tend to be more pro-equality than men on average (Cross and Matheson 2006), in that particular community, certain so-called

Scripting Beloved Discomfort  115 “lifestylers” used the area to enact a male-dominance culture that they believe is the “natural order” (Sixma 2009). This shows that a virtual setting can also function as a tool for narrative offline identity, hiding one’s potential stigma (in the case of the lifestylers, also from other sadomasochists). With the fictional setting come fictional possibilities and the option for even more fictional scripts, as well as the possibility to emphasise power dialectics through elements such as avatar size. The play, even when less drastic fantasy is involved, may be very strong in both the fiction-internal and the fiction-external (player-arousing) sense (i.e., it is both autotelic and allotelic in nature). The presence of visible virtual bodies allows for a clearer perception of unfolding, shared narrative. When combined with explicit pre-scripting and/or semi-scripting of certain elements, and the possibility to document the play through the use of special software, the environment’s offered narratives become very attractive. This is especially true of those who spend a lot of time in the same worlds for other purposes, as well as those who have fantasies they cannot or do not want to enact in the physical world (whether for physical, social, or ethical reasons) and/or those who want to preserve visual documentation of their play. It also allows for greater narrative control than a physical session, if desired, as the characters’ responses are system-based and thus more predictable. This brings forth a very interesting factor: as extensively documented by Brown (2013), those who in the long term engage in particularly “rough” play (at least by real-world terms) in virtual worlds tend to give narrative explanations for their decisions. So even when things to an outside eye look like the player just wanted to play an extreme fantasy scene, such as the rape, torture, and death of her character, for pornographic purposes, the player, as described by Brown (ibid.), may believe (or claim) that it was done solely for a fitting narrative closure. Like Mildorf (this volume) notes of interview narrative, sadomasochists too may consider several audiences when describing their practices and an “implied self” presented through the discourse. As a result, their answers are affected by expectations of social norms, even if that fact or an implied wider audience is not mentioned at all. Similar statements are given about activities in physical sessions and online dating: “it seemed like the right thing to say (or do) at that time.” Intriguingly, we have no way of knowing which of the numerous possible explanations, if any, are true in any of the cases. That openness to interpretation is at the heart of fantasy narratives, which can range from vicarious pleasures (or agonies) to the purely metaphoric, even in the case of something as fundamentally embodied as sadomasochism. UNDERSTANDING SADOMASOCHIST NARRATIVES While the narratives of sadomasochist sessions, regardless of platform, do contain a type of the personal “doubles” claimed by Jackson (1981) as

116  J. Tuomas Harviainen necessary for fantasy, their approach is less than fantastic. The activities, despite taking place in fictional worlds, have a visceral goal to them, that of gratification through transgression, control and endurance. The goal in itself is allopoietic—one that resides at least partially outside of the fiction, in the participants themselves (Harviainen 2011). It is usually of a sexual nature but does not have to be (Harviainen, in press; Newmahr 2011). In essence, while the played roles within the fictional world act out the things that take place during each session, the participants themselves feel the results of those actions. In a virtual session, the effects are mediated into imagined situations and possible arousal, but in a physical encounter each lash of the whip and each thrust of a penetration is experienced by both the persons inside the fictional setting and the people performing those fictional personas. The result of this situation, as far as narratives are considered, is a factual splintering. In Marie-Laure Ryan’s (2001) terms, the process creates a multitude of simultaneous “possible worlds.” Some of these exist solely within the fiction; others within the minds of each participant; and finally, one in the shared discourse of the session participants. As noted by Montola (2003) and Harviainen (2008), people experience shared virtual worlds in a multi-layered manner. Their asynchronous fictions mesh, because those are similar enough to function together, but their narratives are by no means identical. In a two-person dominant/submissive session, for example, we can usually perceive the presence of no less than six narratives of varying complexity and integrity. The first of these is the pre-planning narrative or semi-script: the articulation of what is to be expected. As noted before, this consists of things such as limits and desires, expressed beforehand like a recipe, so that the play itself need not be disturbed by meta-discourse. After that, the participants start to experience narratives of their own as things progress. These may or may not be similar, but they are never identical, due to the personal nature of participation in a session. As the session ends, during the part that sadomasochists tend to call “aftercare” (practiced by most but not all), the two participants, through discourse, construct a shared, “official version” of what happened. In game studies, this is called a “Lehrskovian reduction,” the attribution of a condensed formal story to sum up individual experiences (based on Lehrskov 2007). This is not, however, an element actually shared by the participants. Instead, they take that story, combine it with the ones they constructed during the session, and form out of the parts one more narrative, one more possible world, each: that of a “my take on the official version.” Additionally, if the participants were role-playing more or less fully described fictional characters, for example an evil officer and a captured spy, two more narratives and possible worlds will exist—those that the characters would have experienced had they been real persons inside their fictional world.

Scripting Beloved Discomfort  117 Table 6.1  Typical, simplified narrative structure of a sadomasochist session. Note that massive variance exists between sessions and individual players, and that individuals’ internal narratives may deviate strongly from this template. Stage Pre-planning Threshold Ordeals 1-(n) Virtual resolution (Physical resolution, if any) Aftercare Story attribution

Narrative Processes explicated semi-scripts; discussion of personal preferences and restrictions; setting discussion; character creation (if a strongly role-play based session); unsaid wishes and plans session begins; fictional world is established and treated as primary reference; liminality consecutive segments of activity (caning, whipping, piercing, intercourse, verbal humiliation, etc.) or combinations of several at once ending of the activity in the fictional world, though a fictioninternal release of some kind; switch of primary focus back to the real world if takes place at all, is likely to be concurrent with (or near) virtual release (e.g. fiction-internal dominant and the physical dominant experiencing orgasm at the same time in physical play); in case of online play, may precede the virtual end, in which case the fiction is then wound down in a suitable manner discussion of the events that took place; making sure no lasting harm was done (physically or mentally); possible intimacy turning the aftercare discourses into an unpolished consensus of what took place

Furthermore, participants take this gratifying, if at times unpleasant, activity very seriously, even as it is for them a kind of play. It is for them more akin to pushing the proverbial envelope on both personal limits and social norms than an act of performance, despite the presence of performative elements (Harviainen 2012). The study of narratives has in cases of rolereversal and performative role-accentuation typically relied on Mikhail Bakhtin’s (1973) idea of the carnival to explain such phenomena. In the case of sadomasochist narratives (excluding some stage performances intended for titillating the audience) that explanation fails, because for Bakhtin, such activities are inherently satirical of the current social and/ or religious situation (Jackson 1981). While this may well be true of the origin of many orgiastic practices (see Frank 2013), it does not hold true in the case of sadomasochistic sessions. Under most circumstances, sadomasochism is by no means satirical, but is on the contrary a set of possible worlds intended to feel as real as possible, while still preserving its inherent, mandatory consensuality. In the same way, it is not allegory or a substitute, as for example Deleuze (1967) claims, because to be normalised as allegory or symbolism is to be deprived of the power of the activity itself (Jackson 1981). A much more fitting definition is found from Georges Bataille (e.g., 1957). Fantasy is not irrational, but rather anti-rational, a setting where

118  J. Tuomas Harviainen contradicting ideals can simultaneously exist (Bessière 1974). First and foremost of these is the fact that a sadomasochist session can and almost always will be both voluntary and based on at least one participant’s being at the mercy of another. This is explained by Bataille’s (1957) idea that humans seek simultaneously to feel the obliteration of the boundaries of their selves and to sustain their individual existences. Sadomasochism provides its practitioners with the ability to do just that. Whether dominant or submissive, whether dealing with humiliation or pain, their narratives from that activity will deal with being in control and at the same time relinquishing control: the submissive to the dominant and the dominant to the submissive, who could with just one utterance of a safe word completely re-define the activity that is taking place. Effectively, the participants play with the authenticity of the situation itself, as well as with the boundaries of the fiction. This is both mandatory and playful, and many intentionally flirt with the borders of the possible worlds they create for the sake of teasing and transgression. CONCLUSIONS In this chapter, I have discussed the narratives involved in sadomasochist play, itself a very wide-ranged concept. Its practitioners work with skilled improvisation, conducted with the help of and restricted by semi-scripts that participants bring with them to the fictional world. Out of the fiction arise several simultaneous possible worlds, where the events are given slightly different narrative meanings. Practitioners often do attempt to share the main points and construct a kind of official version of the things that took place during the session, but this, while allowing them to discuss the events more freely later on, inevitably leads to the construction of more possible worlds: fictional, potential realities that arise from the personal interpretations of that official truth. In Mildorf’s terms (this volume), whereas the “experiencing personas” were present in the actual session—virtual and thus mediated or physical—the shared official version then invokes their “narrated personas,” slightly different versions of which exist in each of the possible new worlds. The platform used for the activity has a mediating effect on the session and its narratives, if it takes place in an online environment. This opens up new possibilities but also distances the experience from the practitioners. Some seek that distance intentionally, for the sake of safety. Online play, whether in the form of dating or chat messages, can furthermore in itself function as semi-scripting for other sadomasochist activities. In sadomasochism, we can see a particularly intriguing way in which personal, first-person narrative experiences fuse with narrative identities, scripting, post de facto story attribution—and even unstated wishes.

Scripting Beloved Discomfort  119 ACKNOWLEDGEMENTS The author thanks the editors for their highly useful commentary, and Dr. Jonas Linderoth for his insightful comments regarding the topic of this chapter when its first drafts were being written. REFERENCES Alison, Laurence, Pekka Santtila, N. Kenneth Sandnabba, and Niklas Nordling. 2001. “Sadomasochistically Oriented Behavior: Diversity in Practice and ­Meaning.” Archives of Sexual Behavior 30.1: 1–12. Alperson, Philip. 2010. “A Topograhy of Improvisation.” Journal of Aesthetics and Art Criticism 68.3: 273–80. Bakhtin, Mikhail. 1973. Problems of Dostoevsky’s Poetics, translated by R. W. ­Rotsel. Münster: Ardis. Bataille, Georges. 1952. L’erotisme. Paris: Les Editions de Minuit. Baumeister, Roy F. 1988. “Masochism as Escape from Self.” Journal of Sex Research 25.1: 28–59. Bessière, Irene. 1974. Le récit fantastique: La poétique de l’incertain. Paris: Larousse. Brandhurst, Christoph. 2011. Kinky Sex: Die etwas härtere Nummer. München: Wilhelm Heyne Verlag. Brown, Ashley. 2013. “Sex between Frames: An Exploration of Online and Tabletop Erotic Role Play.” PhD diss., University of Manchester. Bruner, Jerome. 1990. Acts of Meaning. Cambridge: Harvard University Press. Carvalho, John. M. 2010. “Repetition and Self-Realization in Jazz Improvisation.” Journal of Aesthetics and Art Criticism 68.3: 285–90. Cross, Patricia A., and Kim Matheson. 2006. “Understanding Sadomasochism: An Empirical Examination of Four Perspectives.” In Sadomasochism: Powerful Pleasures, edited by Peggy J. Kleinplatz and Charles Moser, 133–66. New York: Haworth. Dancer, Peter L., Peggy J. Kleinplatz, and Charles Moser. 2006. “24/7 SM Slavery.” In Sadomasochism: Powerful Pleasures, edited by Peggy J. Kleinplatz and Charles Moser, 81–101. New York: Haworth. Deleuze, Gilles. 1967. Présentation de Sacher-Masoch - Le froid et le cruel. Paris: Les éditions de Minuit. Dungeons & Dragons. 1973. Created by David Arneson and Gary Gygax. Frank, Katherine. 2013. Plays Well in Groups: A Journey through the World of Group Sex. Plymouth: Rowman & Littlefield. Gazzaniga, Michael S. 1998. The Mind’s Past. Berkeley: University of California Press. Goffman, Erving. 1963. Stigma: Notes on the Management of Spoiled Identity. New York: Simon & Schuster. Harviainen, J. Tuomas. 2008. “A Hermeneutical Approach to Role-Playing ­Analysis.” International Journal of Role-Playing 1: 66–78. ———. 2011. “Sadomasochist Role-Playing as Live-Action Role-Playing: A TraitDescriptive Analysis.” International Journal of Role-Playing 2: 59–70. ———. 2012. “Ritualistic Games, Boundary Control and Information Uncertainty.” Simulation & Gaming 43.4: 506–27.

120  J. Tuomas Harviainen ———. In press. “Information Literacies of Self-Identified Sadomasochists: A Case Study.” Journal of Documentation. Harviainen, J. Tuomas, Richard D. Gough, and Olle Sköld. 2012. “Information ­Phenomena in Game Related Social Media.” In Social Information Research, edited by Gunilla Widén and Kim Holmberg, 149–71. Bradford: Emerald. Herman, David. 2002. Story Logic: Problems and Possibilities of Narrative. Lincoln: University of Nebraska Press. Holstein, James A., and Jaber F. Gubrium. 2000. The Self We Live by: Narrative Identity in a Postmodern World. New York: Oxford University Press. Jackson, Rosemary. 1981. Fantasy: The Literature of Subversion. London: Methuen. Jenks, Chris. 2003. Transgression: Key Ideas. London: Routledge. Johnstone, Keith. 1981. Impro: Improvisation and the Theatre. London: Eyre Methuen. Kaprow, Allan. 1966. Assemblage, Environments & Happenings. Text and Design by Allan Kaprow. With a Selection of Scenarios by 9 Japanese of the Gutai Group, Jean-Jacques Lebel, Wolf Vostell, George Brecht, Kenneth Dewey, Milan Knízák, and Allan Kaprow. New York: Harry N. Abrams. Lehrskov, Ulrik. 2007. “My name is Jimbo the Orc.” In Lifelike, edited by Jesper Donnis, Morten Gade, and Line Thorup, 159–63. Copenhagen: Projektgruppen KP07. Lloyd, Annemaree. 2007. “Learning to Put Out the Red Stuff: Becoming Information Literate through Discursive Practice.” Library Quarterly 77.2: 181–98. Mains, Geoff. 2002. Urban Aboriginals: A Celebration of Leather Sexuality. 3rd ed. Los Angeles: Daedalus Publishing. Montola, Markus. 2003. “Role-Playing as Interactive Construction of Subjective Diegeses.” In As Larp Grows Up: Theory and Methods in Larp, edited by Morten Gade, Line Thorup and Mikkel Sander, 82–89. Copenhagen: Projektgruppen KP03. Mortensen, Torill E. 2003. “Pleasures of the Player: Flow and Control in Online Games.” PhD diss., University of Bergen. Moser, Charles. 1998. “S/M (Sadomasochistic) Interactions in Semi-Public Settings.” Journal of Homosexuality 36.2: 19–29. Myers, David. 2010. Play Redux: The Form of Computer Games. Ann Arbor: ­University of Michigan Press. Newmahr, Staci. 2011. Playing on the Edge: Sadomasochism, Risk, and Intimacy. Bloomington: Indiana University Press. Nordling, Niklas. M. 2009. “Many Faces of Sadomasochistic Sex.” PhD diss., Abo Akademi University. Ortmann, David M., and Richard A. Sprott. 2013. Sexual Outsiders: Understanding BDSM Sexualities and Communities. Lanham: Rowman & Littlefield. Ricoeur, Paul. 1988. Time and Narrative, Volume 3, Translated by Kathleen McLaughlin and David Pellauer. Chicago: University of Chicago Press. Ryan, Marie-Laure. 2001. Possible Worlds, Artificial Intelligence and Narrative ­Theory. Bloomington: Indiana University Press. Second Life. 2003. Linden Research. Siegel, Carol. 1995. Male Masochism: Modern Revisions of the Story of Love. Bloomington: Indiana University Press. Sixma, Tjarda. 2009. “The Gorean Community in Second Life: Rules of Sexual Inspired Role-Play.” Journal of Virtual Worlds Research 1.3. doi: http://dx.doi. org/10.4101/jvwr.v1i3.330.

Scripting Beloved Discomfort  121 Stanislavski, Konstantin. 2010. Näyttelijän työ: Oppilaan päiväkirja, edited and translated by Kristiina Repo. Helsinki: Tammi. Stear, Nils-Hennes. 2009. “Sadomasochism as Make-Believe.” Hypatia 24.2: 21–38. Strauss, Neil. 2005. The Game: Penetrating the Secret Society of Pickup Artists. New York: Regan Books. Torner, Evan. 2012. “Kid Nation: Television, Systemic Violence and Game Design.” In Immersive Gameplay: Essays on Participatory Media and Role-Playing, edited by Evan Torner and William J. White, 127–46. Jefferson: McFarland. Weinberg, Martin S., Colin J. Williams, and Charles Moser. 1984. “The Social ­Constituents of Sadomasochism.” Social Problems 31.4: 379–89. Weinberg, Thomas S. 1978. “Sadism and Masochism: Sociological Perspectives.” The Bulletin of the American Academy of Psychiatry and the Law 6.3: 284–95. World of Warcraft. 2005. Blizzard Entertainment. Zurbriggen, Eileen L., and Megan R. Yost. 2004. “Power, Desire, and Pleasure in Sexual Fantasies.” Journal of Sex Research 41.3: 288–300.

7 Storyworld in Text-Messages Sequentiality and Spatialisation Agnieszka Lyons

The focus of narrative studies has greatly expanded over the years following the oft-discussed narrative turn, or—as Hyvärinen (2012) argues—a number of narrative turns. Literary texts, which came under scrutiny in classical narratology, are now analysed alongside a wide array of other types of texts—in the Hallidayan understanding of the term (Halliday and Hasan 1985)—in a range of disciplines within the social sciences including anthropology, law, and linguistics (see Czarniawska 2004 for the discussion of the use of narrative in social science research). For example, short forms referred to as “small stories” (cf. Bamberg 2006; Georgakopoulou 2007, 2013; Page 2010) and described as open-ended and fluid, occurring “in the small moments of talk” (Page 2012, 426), are being explored as situated tools for identity construction while narrative analysis is used to understand the social context of health and social care (e.g., Stephens 2011). At the same time, technology development has brought the question of (digital) media into narrative and narrative processing and led to the popularisation of inter-disciplinary approaches to narrative analysis. The framework is being applied to a wider range of texts than ever before, e.g., cartoons (Baldry and Thibault 2006), digital audio files (Page 2012), Facebook ­status updates (Page 2010), and opera (Hutcheon and Hutcheon 2010), as well as a range of narrative texts illustrated in this volume, ­including video games (e.g., Mäyrä, and Roine, both in this volume), reality TV (Mäkelä, this volume), and virtual worlds (e.g., Harviainen, this volume). While to the best of my knowledge not yet analysed for its narrative features, a variety of amateur fiction composed and read on mobile phones and resembling textmessages—keitai shousetsu (mobile phone novels)—has gained popularity in Japan since 2000 (Coates 2010). This chapter brings text-messaging (not to be confused with mobile phone novels, whose analysis falls outside the scope of this chapter)—which has often been treated mainly as a carrier of degenerate language (with a few notable exceptions)—to the discussion and establishes it as bearing narrative features and worthy of further analysis in terms of its narrative potential. I discuss texters’ use of specific discursive tools to (re)construct storyworlds and position themselves and the intended recipients of their text-messages within them. In what follows, I draw on deictic shift theory

Storyworld in Text-Messages  123 (Duchan, Bruder, and Hewitt 1995), cognitive narratology (Herman 2004, 2010), and semantics (Fauconnier 1985) to theoretically ground the discussion. After providing information about the modality of texting in the context of its narrative potential in the following section, I proceed to discuss two of the characteristics associated with narrativity in text-messages: sequentiality (understood as a temporal sequence of events) and storyworld construction and present two case studies to illustrate the applicability of these concepts to texting, before drawing some concluding remarks. NARRATIVE POTENTIAL OF TEXTING Often viewed as communicatively impaired and allowing for only limited expressive possibilities, texting has been analysed as either a (socio)linguistic curiosity (with its abbreviated spelling and cryptic meaning-making) or from a functional perspective, for example, as providing a tool for microcoordinating during the day. While there is currently a significant number of scholarly studies devoted to texting, this form of communication is yet to be recognised as allowing for expressive possibilities comparable in range with those of oral and literary narratives, poetry, or art (but see Tagg 2013). The analysis presented here is based on nearly two thousand text-­messages collected from English and Polish native speakers living in ­London and ­Warsaw, respectively. The data collection, the detailed account of which can be found in Lyons (2014), involved the friends-of-friends network approach, where groups of friends were asked to contribute their text-messages to the study and recommend other friends who might be interested in participating. As a result, the data collected include a large percentage of texting interactions in friend-friend dyads, which resulted in a frequent adoption of a playful texting pattern between interactants. Although this study is not intended as comparative in focus, the use of two samples from comparable groups of users in two different cultural and linguistic environments ensures that the observed phenomena are not language- or culture-specific. It can be expected, however, that the modality employed for communication will impact on the content of messages. Two of the features often mentioned as characterising texting are portability and constant reachability (e.g., Licoppe 2004, 2012; Baron and Hård af Segerstad 2010). Resulting from these characteristics are the unique properties of texting, including those related to establishing location and drafting a temporal frame in interactions. Texters often comment on an unfolding situation, report “breaking news” (Georgakopoulou 2007, 2013) in their lives, or plan the immediate or near future. Above all, their communication is usually strongly embedded in the interactive context and based on the mutual familiarity of the interactants with each other and with the technology used. In this sense, texting may not immediately strike us as bearing narrative potential, but I argue that a closer look at the content of text-messages

124  Agnieszka Lyons allows us to place them alongside other narrative-like texts (cf. Page 2012). While not claiming that text-messages should be classified as prototypical Aristotelian narratives, likened to Propp’s Russian folk stories, or Labovian personal experience narratives, I follow Page (2010), who analyses the narrative character of Facebook status updates, in accepting that in the face of the constantly changing character of modern communication, it is more accurate to talk about features of narrativity whose presence makes a text more or less narrative-like. I support Page’s (2010) point that the episodic narrativity found in a variety of everyday stories told using electronic media (Facebook status updates, emails, tweets, etc.) suggests that the traditional understanding of what characterises a prototypical narrative may not be fully representative of the current narrative practice. Consequently, new approaches are being developed to account for less conventional narratives (e.g., unnatural conversational narratives discussed in Mildorf 2013), and evidence of narrative features is provided in media and platforms not traditionally associated with the concept of narrativity and the domain of narrative studies (some examples include video games, Facebook status updates, and cartoons). Concurrent with Page (2012), some forms of narratives, especially the small everyday narratives transmitted by electronic media, may display only some of the features traditionally identified as characterising narratives of personal experience (Labov and Waletzky 1967). Page argues that, rather than being dismissed as ephemeral and fragmentary, Facebook status updates should be treated as a fertile territory for the analysis of new ways in which temporality and sequence are constructed in everyday small narratives. I propose a further advancement to this approach through the analysis of text-messages as discourse containing narrative features. In what follows, I focus on two of the properties associated with narratives: sequentiality and storyworlds, commenting on the applicability of existing frameworks to the analysis of narrativity in text-messages. SEQUENTIALITY In its Labovian understanding, narrative involves retelling past events in such a way that the clauses that recount the event are in the same order as the original events and include at least one temporal juncture (Labov and Waletzky 1967). As observed by Schiffrin (2009a, 2009b) and evident from the recent studies of narratives (e.g., Montoro 2010; Gibbons 2010; Ensslin 2010, and studies in this volume), the scope of narrative analysis has widened in the recent years. Studies have shown that the Labovian mapping of verbal sequences of clauses to the sequence of events (Labov 1972) does not reflect the human experience of time (Mishler 2006). It was proposed that narratives can take a fragmentary form, be co-constructed in interactions (e.g., Bamberg 2004, 2006; Georgakopoulou 2007), and in the social media context take the form of “networked narratives” (Page, Harper, and Frobenius 2013).

Storyworld in Text-Messages  125 It has been proposed also that narrative discourse can be activated in a text on the basis of genre-related considerations, such as those listed by Baldry and Thibault (2006, 14). For example, scenes in cartoons, contrary to appearances, can not only represent a single moment in time, but rather a sequence of events that can be deduced from the visual information available. Cartoon participants who take part in a sequence of events must maintain their identity from one event to another in the sequence, despite the fact that the transition between events entails a change in event participants. In the following section, I demonstrate that a single text-message can, similarly, imply not only a single event or state, but rather a whole sequence of events or states. Sequential relationships are established based on lexically and grammatically realised indications of temporal relations between reported actions and states, and—at times—a change in location within a storyworld, similar to those discussed in the case studies presented in this chapter. CASE STUDY 1: SEQUENTIALITY THROUGH CHANGE OF LOCATION The text-message discussed in this section (example (1)) was written by a young male at a time of his pre-planned date with the recipient. Due to a  misunderstanding, the sender had driven a long way from his house to the recipient’s area, where he found the recipient unavailable. At the time of sending the text-message, the sender was physically located in his car, which was parked outside the recipient’s house, and he had already attempted phoning the recipient twice. In his text-message, in which the original spelling and grammar are preserved, the sender refers to a sequence of actions involving the change of his location. Through the choice of linguistic resources, reference is made to three individual locations within the real world—either explicitly mentioned or implied—and movement (completed and intended) between them. (1) Thank You very much. I have come all dis way and u r no were to be seen or heard. Goodnite babe. Im gona go home. Cheer up The time of sending the message (which I describe as NOW) marks the temporal deictic centre of the sender (DC) and divides the time span referred to in the text-message into two parts: (1) the finished period of time until NOW and (2) the future time which starts at NOW. Temporal reference in texting, like in other forms of asynchronous (although often, arguably, near-synchronous) communication, is subject to a potential time-lag between the moment a text-message is created and the moment it is read by a recipient. This relationship is represented in Figure 7.1, which illustrates the event sequence in example (1) from the sender’s vantage point. It is also to this deictic centre that the recipient has to shift in order

126  Agnieszka Lyons to understand the conveyed meaning when she reads the message, which could be at a much later point (the question of deictic shift is discussed later in this chapter). S = sender R = recipient DC = S’s deictic centre /completed/

/intended/ NOW

S not HERE

S on way to DC (in motion) R at unknown location

S at DC R not at DC

S on way HOME R at unknown location

Figure 7.1  Event sequence in (1).

Temporal sequence can be indicated through a number of lexical and linguistic resources, including sequencers, such as first, then, after that, following this, later, finally, etc., and the choice of tenses (simple past or historical present tense in conventional narratives). The sequence of events in oral narratives is usually represented by simple past tense or historical present (cf. Schiff rin 1981) while in the analysed text-message its construction is achieved using more varied tense forms. The sender employs the verb come in present perfect tense to indicate a completed action whose results are evident at the time of texting. Through the use of I have come the sender implies that he was at a different location, i.e., not here, before and subsequently, after having been on the way, is currently at his deictic centre, i.e., here. At the same time, the recipient is no were to be seen or heard “nowhere to be seen or heard” at the sender’s spatial and temporal deictic centre. The sender indicates a plan to go home, expressed through the use of Im gona go “I’m going to go.” Although the use of going to marks a current intention, it gives the reader a window into the sender’s future action, which will constitute a further element in this narrative sequence of events. The reconstructed sequence of events entails continuity of participant roles (sender and recipient) throughout the narrative sequence and a change in one of the participants, namely a change in the sender’s location. An analysis of narrative event structure in example (1) is presented in Table 7.1. The table shows the particular phases of the narrative event in (1), identifies participants and actions performed by them, and provides linguistic evidence from the analysed example. I employ abbreviated forms to refer to the sender (S), recipient (R), and deictic centre (DC).

Storyworld in Text-Messages  127 Table 7.1  Narrative structure in (1). Phase [implicit/past] S is at an unspecified distant location.

Participants S R (absent)

2

[implicit/past] S is on his way to the venue of meeting with R. He is travelling a long distance.

S

3

Having arrived, S is located at DC. R is absent, which makes S upset and leads him to the intention to go home; R not at DC.

S R (absent)

4

[intended/future] S is S on his way home.

1

Action S stationary, located AT NOT HERE

Evidence all dis way … quantifier all for a large amount or quantity I have come all dis S in motion, directional vector way … Present Perfect towards DC, implying arrival; of COME for R at DC (sender’s recently completed action with evident mental space) consequences; proximity to DC through dis S stationary at … u r no were to DC; be seen or heard. R at NOT HERE Thank You very much. correctly spelt sentence with capitalised You for marked pronunciation with emphasis on the personal pronoun S (intended) in Im gona go home. motion GOING TO for intentions; GO directional vector away from DC and towards home, not implying arrival

It is clear from the above example that the choice of verbs in the analysed text-message has an added communicative aspect. Not only do the tenses and the change-of-location verbs signal a sequence of events, but they also indicate the direction of movement of the sender (come and go), with other discursive means additionally signalling the distance travelled (all dis way), which—in turn—implies the existence of a spatialised storyworld within which the characters move. Motion is inseparable from and dependent on the existence of a world within which it occurs. It can be the physical world around texters who micro-coordinate or a mentally created world, e.g., in the context of creating joint communicative space between remotely located participants. Similar to those occurring in literary and oral narratives, storyworlds in texting are represented as “having a specific spatial structure” (Herman 2004, 264). Events in a story tend

128  Agnieszka Lyons to occur within a mental model constructed by the reader from the text as it is encountered at the active space-time location to which the reader has been directed by the syntax and semantics of the text. The space in which this narrative unfolds has to have dimensionality and identifiable landmarks with reference to which the sender orients him- or herself and the recipient. The following section looks at the spatial aspect of a discursively constructed storyworld, a feature discussed in earlier studies of both literary and oral narratives (e.g., Herman 2001, 2004, 2010), but unexplored in the context of texting. As suggested at the beginning of this chapter, with its remotely located interactants and the on-the-go character of communication, texting offers a rich field for the investigation of space and location in the context of storyworld construction, an aspect the following section aims to explore in some detail. SPACE AND STORYWORLD Apart from a clear temporal and sequential focus, narratives can also be comprehended as constituting “verbal and visual cues prompting their readers to spatialise storyworlds into evolving configurations of participants, objects, and places” (Herman 2004, 263). The question of spatial mapping of unfolding narratives in text-messaging is particularly interesting due to the portable character of mobile phone devices and the ensuing complexity of deictic reference in communication between remotely located interactants. A similar situation occurs in other forms of text-only mobile interactions, for example those made possible by mobile phone applications for instant messaging such as WhatsApp. When placing themselves in a situation, interactants adopt a certain spatial, temporal, and psychological point from which they experience it (Zwaan 1999). Similarly, readers of literary narratives and listeners of oral narratives adopt a certain vantage point that conditions their comprehension of the story (see, for example, the discussion of focalisation in Katherine Mansfield’s story “Bliss” and the applicability of the term to conversational narratives in Mildorf 2006). It is assumed that comprehenders (i.e., readers, hearers, or viewers, depending on the type of narrative) place themselves, to some extent, within the narrated situation and from that vantage point construct a situation model: a mental representation of, among other things, locations, objects, characters, and events in the story (Zwaan 1999). The story in a fictional narrative is, like all mental representations, partially a construction of the comprehender (reader or listener) but also based on text of the author/speaker. In contrast, the storyworld is mostly a mental construct of the reader/listener. In their interpretations, compehenders can import knowledge of the everyday world and of other possible worlds into the current storyworld; this provides the listeners/readers with the illusion

Storyworld in Text-Messages  129 of mentally inhabiting a fully specified and coherent world. ­Consequently, they become active participants (Busselle and Bilandzic 2008) and “writer[s] of [their] own version of the story” (Oatley 2002, 43). In accordance with the Deictic Shift model, which I discuss in the following section, fictional narration results in readers’ imagining deictic fields as lifted from their physical locations and shifted into storyworlds. These are created and experienced according to the linguistic make-up of texts, creating a world that fits the words used to describe it, following a world-to-word direction of fit (Searle 1983). As the following section demonstrates, a similar shift can take place in text-messages, which serves as another indication of their narrative potential.

Deictic Shift As a cognitive structure, the deictic centre lends coherence to a text and allows the reader to correctly localise events and other aspects of the story when they are not explicitly indicated in the text. In order to correctly interpret and experience the text, comprehenders need to refer to their general knowledge, logical and pragmatic constraints, and special stances toward it (Segal 1995, 16). Readers or listeners use the available semiotic cues to construct more or less detailed representations of the worlds to which they deictically shift in the process of interpreting narratives. From their adopted vantage point, they (re)construct the space-time configuration of narrated events, entities forming part of the narrated worlds, and relations between them (Herman 2010, 80). In the case of oral narratives in which authors build the context of the story from their own perspective (see Herman 2010), it is possible to interpret a given narrative from the perspective of a deictic centre located at the author’s physical location (Zubin and Hewitt 1995). In literary narratives, on the other hand, authors usually distance themselves from the storyworld presented in their fiction, and it is understandable (particularly in thirdperson narratives) that the author and the speaker are not equivalent. In the case of both literary and oral narratives, reacting to available cues, readers/listeners build associations with the story characters and are transferred into the storyworld, which they view in the same way as they would view the here and now of their physical environment. Such an interpretation is possible based on their knowledge of the world and prior experiences with deixis in real life, and such shift is made possible by transferring the linguistic reference of deixis from the speech situation and shifting it to the locations and characters of the storyworld (Zubin and Hewitt 1995). According to deictic shift theory (DST), readers/listeners and authors shift their deictic centre from the real-world situation to an image of themselves at a location within the storyworld, i.e., a world as constructed through a narrative. Such a shift from an actual situation to a described

130  Agnieszka Lyons situation has been termed deictic shift (Duchan, Bruder, and Hewitt 1995). It results in people and objects in a particular narrative being more accessible to the comprehender and, in consequence, to potential transportation, defined as a phenomenological experience in which a reader’s (or recipient’s, more generally) mental processes are fully focused on the events occurring in the narrative1 (Green and Brock 2002), and the loss of self-awareness. Green and Brock’s Transportation-Imagery Model can be applied to any kind of text that evokes measurable images, i.e., mental contents that possesses sensory qualities (Dadds, Bovbjerg, Redd, and Cutmore 1997). Transportation, then, represents the extent to which a reader becomes absorbed in constructing both situation and storyworld mental models. Readers are able to experience transportation into and move within the unfolding narrative storyworld thanks to such linguistic elements as deictic pronouns and verbs (including tenses and aspects). In fact, they are motivated to perform deictic shifts because it is only from the deictic centre within the storyworld that deictic words make sense (Galbraith 1995). To allow for such transportation, it is also not unimportant for readers to maintain a certain mindset, which has been described in earlier research as “willingness to suspend disbelief.” It assumes natural human scepticism when dealing with mediated content. It is difficult to say whether disbelief in such situations needs to be suspended and whether such suspension is a conscious decision. It is possible that no such (semi-)conscious suspension is necessary and, especially as societies become more digitalised and digitally literate, that disbelief is not the default approach to mediated input. Franks (2013, personal communication) prefers to refer to the “willingness to believe” instead. Deictic shift into and within a storyworld in texting, while demonstrably possible (see example (1)), is characterised by certain differences when compared with the deictic shift seen in literary narratives. First, the latter type of storyworld is assumed to be distinct from that of the speaker/writer’s reality. Text-messages are sent between two easily identifiable interactants who usually know each other, which makes them more similar to the situation encountered in oral narratives. There is thus no automatic assumption that the world presented in text-messages is fictional and that it is likely to evoke imagery required for transportation (Green and Brock 2002). Another major difference is found in the number of possible locations of deictic centres in texting and in literary narratives, with text-based mobile communication allowing for a wider range of possible reference frames and deictic centres. Taking into account modality affordances and established conventions within electronically mediated communication, it can be expected that the discursive tools employed by texters to construct representations of storyworlds may differ from those found in conventional literary and oral

Storyworld in Text-Messages  131 narratives. The case study discussed in the next section extends the discussion of sequentiality in texting to involve cases where deictic shift into a virtually created communicative location is discursively performed using accepted modality conventions. In the discussion, I draw on the concept of mental spaces (Fauconnier 1985) and the Identification Principle (Nunberg 1978) to further theorise construction and spatialisation of a texting storyworld. CASE STUDY 2: DEICTIC SHIFT THROUGH TEXTING CONVENTIONS The text-message to be analysed in this section was sent by a female graduate student to her friend who was supposed to join the sender and her other friend at their hairdresser’s that evening. The recipient was unable to make it but had not let the other two girls know. At the time of sending the text-message, the sender and her other friend were already at the hairdresser’s and were waiting to be seen. Both the sender of the textmessage and the recipient were frequent texters, and a large part of their communication throughout the day was conducted in this way. They had a number of things in common, studying at the same university and being involved in a number of projects together. They also shared familiarity with digital communication and both used the Polish instant messenger Gadu-Gadu. In example (2), there is no mention of change of location or motion, in which it differs from example (1), but there is a clear reference to a number of locations, both having their equivalents in the real world and constructed solely in the virtual domain. (2)

Slyszalam  ze mialas byc u naszego fryzjera a   I’ve heard that you were supposed to be at our hairdresser’s and jakos jest babo?!   Cie tu nie       widzimy! Gdzies somehow you here not we see! Where in the world are you hag?!  ;) ;) “I heard that you were supposed to be at our hairdresser’s but we can’t see you here somehow! Where are you, woman?! ;)”

Here again, a series of sequentially organised events can be identified and, just like in example (1), participants maintain their identity throughout the text. The complexity of the storyworld in example (2) is related to the presence of additional participants and of a particular case of deictic shift. The sequence of events in (2) is represented in Figure 7.2.

132  Agnieszka Lyons S = sender R = recipient AI = another individual UI = unspecified informant DC = S’s deictic centre JCS = joint communicative space /completed/ NOW

mental space 1

S receives information from UI R at unknown location

mental space 2 (imagined)

S at DC with AI R not at DC S’ hits R’ at JCS

Figure 7.2  Event sequence in (2).

Similar to the previous case study, the time of sending the text-message marks the temporal deictic centre (NOW), and the use of past tense helps identify past events. There is also, however, a clearly specified spatial deictic centre (tu “here”), and it is possible to spatially place other locations with reference to it. Before discussing mental spaces in the analysed text-message (see Figure 7.2), I provide a closer account of the narrative sequence in (2) in Table 7.2. In addition to the abbreviations used in Table 7.1, reference is made to another individual (AI), who is different from the recipient and who accompanies the sender to the hairdresser’s. Reference is also made to an unspecified informant (UI), who can but does not have to be the recipient (R) or the individual who is physically co-located with the sender (AI). The table includes information about the narrative phases represented in the text-message in question, participants, and actions taken by them. Discursive evidence from the analysed text-message is provided. Similar to example (1), this text-message carries information about the existence of narrative events in a sequential order. The sender refers to having been given information which, verified at the time of composing the message, does not prove correct. Rather than stated explicitly, the event that constitutes the first phase is implied in the content of the text-message. The context suggests that the sender had received information about the recipient’s plan to join her (the sender) at the hairdresser’s. The second phase is temporarily anchored in the sender’s temporal deictic centre, i.e., the time when the textmessage was sent. It is clear that the recipient is not present at the physical location where the sender expected her to be (her hairdresser’s). The final phase is more difficult to temporally locate in the unfolding narrative. As linear, the text-message has a clear reading path, which follows the order of

Storyworld in Text-Messages  133 words in the text (Kress 2003). If we were to take Labov’s sequential narrative structure as a model, it would follow that the order of words determines the reading of the text and meaning-making with reference to the sequential order of events. Following on from this, the final phase of the narrative in (2) would be seen as taking place in a sequential order after the preceding one. The employment of angle brackets to delimit the third narrative phase (cf. Nunberg 1990; Lyons 2014) and the grammatical shift from first-person narration to third-person narration lead to the need to consider other possibilities in the temporal and spatial interpretation of the text-message (2). Table 7.2  Narrative structure in (2) 1

2

3

Phase [implicit/ past] S receives information from an unspecified informant (possibly R) that R will be at S’s hairdresser’s later. S is at her hairdresser’s (DC) with an unspecified person (not R). R is not at the S’s hairdresser’s, which makes S dissatisfied.

Participants S UI

S AI

[enacted S’ discursively/ R’ present] S is hitting R with a frying pan.

Action S receives information from UI

Evidence Slyszalam ze mialas byc u naszego fryzjera ‘I’ve heard you were supposed to be at our hairdresser’s.’ perfective form of the verb SŁYSZEĆ for accomplished actions … mialas byc … S with AI a jakos Cie tu nie stationary at widzimy. ‘You were HERE, supposed to be … and R at NOT we can’t see you here HERE in somehow.’ motion, content expressing the directional difference between the vector expected (‘you were towards DC, supposed to be’) and implying the factual (‘we can’t arrival; R at see you’); DC (sender’s mental space) indicator of confusion and annoyance (‘somehow’) S stationary at ‘‘ angle brackets for a deictic shift to another (virtual) mental space; present tense for actions happening at the moment

As established earlier, the sender and the recipient were not co-present in the same physical space at the time the text-message was sent. It is therefore unlikely that the first two phases discussed in Table 2 and the action of hitting in the third phase constitute a sequence of events in the same unfolding

134  Agnieszka Lyons narrative. It is logically impossible for the interactants to be explicitly not co-present (jakos Cie tu nie widzimy) and physically co-present to allow for the act of hitting (wali “hit-3.SG”), which requires physical proximity. Consequently, a shift into another conceptual location has to be performed to make sense of the event sequence in (2). The notion of mental spaces and reference to the current discussions in the field of unnatural narratology are helpful in explaining this phenomenon.

Mental Spaces Mental space is a theoretical construct that contains an idealised cognitive model of reality (rather than its faithful representation). Fauconnier (1985, 16) describes mental spaces as “constructs distinct from linguistic structures but built up in any discourse according to guidelines provided by the linguistic expressions.” Mental spaces function as entirely distinct from each other and can be established by space-builders, i.e., certain linguistic expressions, such as prepositional phrases (in John’s mind), adverbs (probably), connectives (if … then …), or subject-verb combinations (Mary hopes …). He notes that spacebuilders will always establish mental spaces as included in their ­parent spaces, although this inclusion does not have to be expressed explicitly2. ­Mental spaces, created by space-builders, must be connected to their parent spaces by means of connectors that link triggers and targets in these spaces.

x1

M

x2

M’

Figure 7.3  Connectors and counterparts in mental spaces.

In this scenario element x1 belonging to mental space M has its counterpart x2 in mental space M’, and mental space M is a parent space of mental space M’. x1 ∈ M x2 ∈ M’ M’ ⊂ M Conceptual separation of an entity (x1) and its counterpart (x2) and the existence of multiple mental spaces may occur also in electronically mediated

Storyworld in Text-Messages  135 communication, with its use of increasingly complex technologies and, as in the case of texting, remotely located participants. In example (1), for instance, the sender’s home is assumed to exist in reality (mental space M), but in the text-message itself, it constitutes only a textual representation of the image the sender has of his home. The complexity does not end there. Since meaning-making is a two-stage process that involves encoding and decoding, the meaning of home the recipient constructs is likely to differ from that of the sender. Hence the use of the word home can trigger at least two different representations in two separate mental spaces constructed at two distinct temporal and spatial deictic centres. Representation can be realised here in one of two main ways: visual and textual. The visual dimension of representations (which falls outside the direct scope of this chapter and therefore will not be analysed in detail) includes 2D- or 3D-avatars in computer games and virtual environments and icons in Internet forums (see discussion in other chapters in this volume and, for a discussion of the employment of autonomous avatars in electronic communication and their expressiveness, Cassell and Vilhjálmsson 1999). The textual dimension, which is the focus of this chapter, involves representations of storyworld elements by means of discursive tools, which serve as links through which correspondence between entities in separate mental spaces is achieved. Language, among other tools, is used to construct an image that corresponds to its real-life referent. Pragmatic function (Nunberg 1978) serves to establish a link between a referent and a referee, and the use of language, including names, definite descriptions, and pronouns, facilitates shifts between these corresponding entities. Reference applies not only to the link between real-life objects, like in the case of personal pronouns directly referring to people, but also to mental images, the latter being of interest here. Following on from this, a concrete entity (e.g., person or object) will differ from its corresponding mental representation, as in the case of home in example (1) and interactants’ counterparts in example (2), which I discuss below. Correspondence between a concrete entity and its equivalent is established based on the Identification Principle (Nunberg 1978): if two objects (A and B) are linked by a pragmatic function (F), a description of one of them, the trigger (A) can be used to identify its counterpart, the target (B). B = F(A) The Identification Principle allows the target to be identified through the description of the trigger by means of a connector that maps the image onto reality. The links between these representations therefore become links between two (or more) distinct mental representations located in two or more distinct mental spaces. The mental space delimited by angle brackets in example (2) ( “hit-3.SG the hairstyle with a frying pan”) contains an imagined, discursively constructed, or virtual reality, in which the sender’s and the recipient’s counterparts

136  Agnieszka Lyons are physically co-present and the sender’s counterpart is hitting the recipient’s counterpart. Despite the fact that the interactants are not located in each other’s immediate vicinity, the final part of the message suggests that a physical action of hitting is taking place in real time and space. The action of hitting must therefore be conducted in a space distinct from reality (M), in a separate mental space (M’). It is performed not by the sender (S) and the recipient (R) themselves, but by their counterparts (S’ and R’, respectively) which are constructed in that mental space (M’) (see Figure 7.4). Therefore, the following ensues: M’ ⊂ M S, R ∈ M S’, R’ ∈ M’ S – Sender R – Recipient

S

S’

R

R’

M (= Sender’s reality)

S’ – Sender’s counterpart in M’ R’ – Recipient’s counterpart in M’

M’ (= Mental space in which hitting takes place)

Figure 7.4  Mental spaces in example (2).

The deictic shift into this space is marked in two ways. First, there is a change in narration style: the first-person narration in the first two phases (slyszalam “I’ve heard”) is replaced by third-person narration in the final part (wali “hit-3.SG”), indicating also a change in positioning within the narrative. Second, the sender employs conventions accepted in texting and other forms of electronically mediated communication to signal this shift. Following Fauconnier’s (1985) transcription conventions, the action in example (2) will be represented as: wali (S’, R’) where the action (wali – “hit-3.SG”) is performed by the sender’s alter persona (S’) on the recipient’s alter persona (R’), which is being reported as close to the interactacts’ perceptions of their communicative partners. The sender of (2) reported she had in her mind the image of herself playfully hitting the recipient who was trying to avoid repeated blows while smiling. The

Storyworld in Text-Messages  137 text-message, she claimed, worked based on the assumption that the recipient, like the sender herself, adopted the play frame in reading the message. Here, the interactants’ familiarity with the conventions employed in electronically mediated communication proves vital. The convention of using angle brackets in the analysed text-message comes from the Polish instant messenger Gadu-Gadu, with which both the sender and the recipient were familiar. Enclosing certain (usually single) words in third person singular (e.g., wali “hits”) in angle brackets results in the occurrence of an animated icon in the conversation window in Gadu-Gadu. The sender creatively appropriated this convention in her texting—a text-only form of communication where animated icons do not appear—and thus referred to the sender’s and recipient’s common ground. She confirmed in a later interview that the reason for the employment of this strategy was to evoke in the recipient associations with Gadu-Gadu conventions. As a result, it was hoped, the action of hitting would be imagined. This type of content resembles stage directions, i.e., implicit voices of authors of plays that provide cues to the theatrical and dramatic effects of the scenes they control and are an important vehicle of meaning in early theatre (McJannet 1999; Dessen 2009). The convention for stage directions in plays is to print them in italics, which would not be possible in text-messages. Instead, conventions used in text-only forms of electronically mediated communication include the use of certain grammatical forms, as well as text-category indicators (Nunberg 1990; Lyons 2014), i.e., sets of printable characters that are employed to represent enactment, for example, hugs can be represented in chatrooms by means of parentheses around a user’s screen name (del TesoCraviotto 2004). Another convention commonly found in virtual worlds, chatrooms, and IRC involves the use of verbal glosses of gestures and movements characteristic of face-to-face communication. The convention here is to enclose a given action in asterisks (Werry 1996) or angle brackets (del Teso-Craviotto 2004; Crystal 2006). In IRC, which he analyses, Werry (1996, 60) finds textual representations, or symbolic enactment, of “[h]ugs, kisses, offers of coffee, yawns, shaking hands, and the popping of champagne.” Similarly, Crystal (2006, 42) mentions the use of such kinesic effects as and . In example (2), interaction taking place in an alternative mental space (M’) is enclosed in angle brackets, which clearly mark the beginning and end of interaction through the sender’s and recipient’s counterparts (S’ and R’ in space M’). As such, this exchange is separate from interactions taking place in the sender’s reality (M), but maintains a link with this reality through specific connectors. Thanks to such shifts between mental spaces, texters (or narrators) are able to construct storyworlds that appear incoherent, impossible, or illogical. This brings to mind unnatural narratives, the study of which focuses on the ways projected storyworlds are different from those that can be explained by our knowledge of the real world (Alber, Iversen, Nielsen, and Richardson 2010). For example, experimental or hallucinatory narratives (see Kakko’s chapter in this volume) cross the boundaries of the natural

138  Agnieszka Lyons in constructing storyworlds that are characterised by logical impossibilities in their represented spatial and temporal organisation (Alber, Iversen, Nielsen, and Richardson 2010). The analysis presented in this chapter makes it evident, yet again, that conventional approaches to narratives—be they literary or oral—with their orderly temporal sequence and anthropomorphic narrator, account for a large part of but not all possible spatiotemporal combinations. At the same time, text-messages—although often portrayed as small and purely transactional—offer rich expressive possibilities both in terms of the logically coherent and the impossible. Playful shifts from one mental space to another, be it to hit the recipient with a frying pan, as was the case in case study (2) or to create the feeling of intimacy, as is the case in sexting3, are a spontaneous creation of regular mobile phone users, rather than the careful work of writers or artists. Since the human mind is bound by its cognitive possibilities, interpreting the unnatural has to be done through reference to and manipulation of existing frames and scripts to create “new cognitive parameters that transcend our real-world knowledge” (Alber, Iversen, Nielsen, and Richardson 2012, 376). The question of interpreting and making sense of the “unnatural” in a texting narrative involves not only reshuffling familiar frames in narrative processing, but also taking into account the modality of texting, its affordances, and constraints, which tie in with the importance of genre awareness in processing post-classical narratives. Familiarity with the accepted conventions of electronically mediated communication and with one’s texting partner (cf. the notion of audience design proposed and developed by Bell 1984, to which Mildorf refers in her chapter in this volume) helps draw conclusions as to the most plausible interpretation of both natural and unnatural texting storyworlds. CONCLUSION: NARRATIVITY IN TEXT-MESSAGES Instances discussed in this chapter clearly demonstrate that there is a need for a new approach to the analysis of texting that would recognise phenomena beyond the surface linguistic curiosities of abbreviated spelling and atypical punctuation on the one hand and the functional application of ­texting for maintaining contact throughout the day and purely transactional uses on the other. I have argued that reading and analysing text-messages should not differ from reading and analysing other texts, such as literary or oral narratives, for which we have a wealth of past research. The case studies discussed in this chapter have shown that texters construct imagined locations and joint communicative spaces that facilitate more vivid exchanges. Texters may to some extent “lose themselves” in their texting interactions, smile to their mobile phone screens, and/or even report the feeling of arousal (as in the case of sexting) as a result of communicating with their remotely located partner. It can be concluded that texters experience a certain level of situatedness, defined as “the predicament of

Storyworld in Text-Messages  139 being in a world” (Rank and Petta 2005), in the discourse-based realm of a text-message storyworld, which they create through the choice of verbs of motion, deictic words, and other linguistic devices (see also Schiff rin 2009a, who shows that narratives can evoke the nexus of place, time, and identity, rather than just experience in time). In these worlds, communicators exist as discursive counterparts of their physical real-life selves and move from place to place, their direction reflected in the choice of verbs of motion. In this constant flow (Breslow 2013), texters adopt different deictic centres and reference frames through their choice of discourse, which enable a deictic shift similar to that described in literary and conversational narratives. Based on the analysis of sequentiality and storyworld construction, I have shown that narrativity in texting is a productive and yet unexplored field for investigation. The features of texting, particularly its limited buffer size, portable character, and personal character of communication, mean that existing frameworks for narrative analysis may have to be revisited to account for the range of parameters that need to be considered in the analysis of texting storyworld construction. The portable character of texting adds an on-the-go aspect to space construction, significant when considered against the distinction between exophoric and endophoric storytelling. Endophorically evoked worlds, as when tellers direct recipients to shift from the temporal deictic centre to other spatiotemporal coordinates, need to be explored in terms of deictic reference and deictic shift and compared to those evoked exophorically, where reference to the features of the current context is made through the use of deixis. An analysis of texting interactions can also shed some light on unnatural storyworld co-construction in unfolding narratives during communicative exchanges by means of textmessages. Additionally, there emerges a need to explore the question of narrative positioning in texting interactions and negotiation of self- and other-positioning within a storyworld. As has become clear, there is a need to adopt an interdisciplinary approach to the study of texting in order to fully grasp its expressive power. Narratology, cognitive linguistics, and communication studies, as well as interactional sociolinguistics and multimodal discourse analysis, can help us understand how storyworlds are constructed in everyday digital lives. While this chapter focused on a very limited range of narrative-related questions, I have indicated a need for further exploration and opened a line of inquiry to be pursued in future studies. NOTES 1. Concepts similar to transportation include flow (Csikszentmihalyi 1990), absorption (Tellegen and Atkinson 1974), and experiential mode (Epstein 2003). 2. Some examples of explicit and implicit embedding can be found in Fauconnier (1985).

140  Agnieszka Lyons 3. The term sexting refers to the act of sending sexually explicit messages by mobile devices. It has been found to evoke the feeling of transportation or being there (Ijsselsteijn and Riva 2003) among texters.

REFERENCES Alber, Jan, Stefan Iversen, Henrik Skov Nielsen, and Brian Richardson. 2010. “Unnatural Narratives, Unnatural Narratology: Beyond Mimetic Models.” ­ ­Narrative 18.2: 113–36. ———. 2012. “What Is Unnatural about Unnatural Narratology? A Response to Monika Fludernik.” Narrative 20.3: 371–82. Baldry, Anthony, and Paul J. Thibault. 2006. Multimodal Transcription and Text Analysis: A Multimedia Toolkit and Coursebook with Associated On-line Course. London/Oakville: Equinox. Bamberg, Michael. 2004. “Narrative Discourse and Identities.” In Narratology Beyond Literary Criticism, edited by Jan C. Meister, Tom Kindt, and Wilhelm Schernus, 213–37. Berlin/New York: Walter de Gruyter. ———. 2006. “Stories: Big or Small: Why Do We Care?” Narrative Inquiry 16.1: 139–47. Baron, Naomi S., and Ylva Hård af Segerstad. 2010. “Cross-Cultural Patterns in Mobile-Phone Use: Public Space and Reachability in Sweden, the USA, and Japan.” New Media & Society 12: 13–34. Bell, Allan. 1984. “Language Style as Audience Design.” Language in Society 13: 145–204. Breslow, Harris. 2013. “When Space and Place Begin To Flow: The Spatial Envelope of Portable Place.” Paper presented at 4th Global Conference: Space and Place, Oxford, September 9–12. Busselle, Rick, and Helena Bilandzic. 2008. “Fictionality and Perceived Realism in Experiencing Stories: A Model of Narrative Comprehension and Engagement.” Communication Theory 18: 255–80. Cassell, Justine, and Hannes Vilhjálmsson. 1999. “Fully Embodied Conversational Avatars: Making Communicative Behaviors Autonomous.” Autonomous Agents and Multi-Agent Systems 2: 45–64. Coates, Stephanie. 2010. “The Language of Mobile Phone Novels: Japanese Youth, Media Language and Communicative Practice.” Paper presented at 18th Biennial Conference of the Asian Studies Association of Australia, Adelaide, July 5–8. Crystal, David. 2006. Language and the Internet. 2nd ed. New York: Cambridge University Press. Csikszentmihalyi, Mihaly. 1990. Flow: The Psychology of Optimal Experience. New York: Harper & Row. Czarniawska, Barbara. 2004. Narratives in Social Science Research. London: Sage. Dadds, Mark R., Dana H. Bovbjerg, William H. Redd, and Tim R. H. Cutmore. 1997. “Imagery in Classical Conditioning.” Psychological Bulletin 122: 89–103. del Teso-Craviotto, Marisol. 2004. “Virtually There: Creating Physicality in Dating Chat Rooms.” Texas Linguistic Forum 48: 73–82. Dessen, Alan C. 2009. “Stage Directions and the Theatre Historian.” In A Handbook on Early Modern Theatre, edited by Richard Dutton, 513–27. Oxford: Oxford University Press.

Storyworld in Text-Messages  141 Duchan, Judith F., Gail A. Bruder, and Lynne E. Hewitt, eds. 1995. Deixis in N ­ arrative: A Cognitive Science Perspective. Hillsdale, NJ: Lawrence Erlbaum Associates. Ensslin, Astrid. 2010. “Respiratory Narrative: Multimodality and Cybernetic Corporeality in ‘Physio-cybertext’.” In New Perspectives on Narrative and ­ ­Multimodality, edited by Ruth Page, 155–65. New York: Routledge. Epstein, Seymour. 2003. “Cognitive-experiential Self-theory of Personality.” In ­ComprehensiveHandbook of Psychology, Volume 5: Personality and Social P ­ sychology, edited by Theodore Millon and Melvin J. Lerner, 159–84. Hoboken, NJ: Wiley & Sons. Fauconnier, Gilles. 1985. Mental Spaces: Aspects of Meaning Construction in ­ atural Language. Cambridge/London: The MIT Press. N Franks, Anton. 2013. Personal communication. Galbraith, Mary. 1995. “Deictic Shift Theory and the Poetics of Involvement in ­ arrative.” In Deixis in Narrative: A Cognitive Science Perspective, edited by N Judith F. Duchan, Gail A. Bruder, and Lynne E. Hewitt, 19–59. Hillsdale, NJ: Lawrence Erlbaum Associates. Georgakopoulou, Alexandra. 2007. Small Stories, Interaction and Identity. ­Amsterdam & Philadelphia: John Benjamins. ———. 2013. “Storytelling on the Go: Breaking News As a Travelling Narrative Genre.” In The Travelling Concepts of Narrative, edited by Mari Hatavara, LarsChrister Hydén, and Matti Hyvärinen, 201–24. Amsterdam: John Benjamins. Gibbons, Alison. 2010. “‘I Contain Multitudes’: Narrative Multimodality and the Book That Bleeds.” In New Perspectives on Narrative and Multimodality, edited by Ruth Page, 99–114. New York/London: Routledge. Green, Melanie C., and Timothy C. Brock. 2002. “In the Mind’s Eye: Transportation-­ imagery Model of Narrative Persuasion.” In Narrative Impact: Social and C ­ ognitive Foundations, edited by Melanie C. Green, Jeffrey J. Strange, and ­Timothy C. Brock, 315–42. Mahwah, NJ: Lawrence Erlbaum. Halliday, MAK, and R. Hasan. 1985. Language, Context and Text: Aspects of ­Language in a Social-semiotic Perspective. Burwood, Vic.: Deaken University Press. Herman, David. 2001. “Story Logic in Conversational and Literary Narratives.” Narrative 9.2: 130–37. ———. 2004. Story Logic: Problems and Possibilities of Narrative. Lincoln and ­London: University of Nebraska Press. ———. 2010. “Word-image/Utterance-gesture: Case Studies in Multi-modal S­ torytelling.” In New Perspectives on Narrative and Multimodality, edited by Ruth Page, 78–98. New York: Routledge. Hutcheon, Michael, and Linda Hutcheon. 2010. “Opera: Forever and Always Multimodal.” In New Perspectives on Narrative and Multimodality, edited by Ruth Page, 65–77. New York: Routledge. Hyvärinen, Matti. 2012. “Prototypes, Genres, and Concepts: Travelling with N ­ arratives.” Narrative Works 2.1: 10–32. Ijsselsteijn, Wijnand, and Riva, Giuseppe. 2003. “Being There: The Experience of Presence in Mediated Environments.” In Being There: Concepts, Effects and ­Measurement of UserPresence in Synthetic Environments, edited by Fabrizio Davide Giuseppe Riva and Wijnand Ijsselsteijn, 3–16. Amsterdam: Ios Press. Kress, Gunther. 2003. Literacy in the New Media Age. New York: Routledge. Labov, William. 1972. Language in the Inner City. Philadelphia, PA: University of Pennsylvania Press.

142  Agnieszka Lyons Labov, William, and Joshua Waletzky. 1967. “Narrative Analysis.” In Essays on the Verbal and Visual Arts, edited by J. Helm, 12–44. Seattle: University of Washington Press. Licoppe, Christian. 2004. “‘Connected’ Presence: The Emergence of a New Repertoire for Managing Social Relationships in a Changing Communication T ­ echnoscape.” Environment and Planning D: Society and Space 22.1: 135–56. ———. 2012. “Understanding Mediated Appearances and Their Proliferation: The Case of the Phone Rings and the ‘Crisis of the Summons.’” New Media & Society 14.7: 1073–91. Lyons, Agnieszka. 2014. Self-presentation and Self-positioning in Text-messages: Embedded Multimodality, Deixis, and Reference Frame. PhD diss., University of London. McJannet, Linda. 1999. The Voice of Elizabethan Stage Directions: The Evolution of a Theatrical Code. Newark, NJ: University of Delaware Press. Mildorf, Jarmila. 2006. “Sociolinguistic Implications of Narratology: Focalization and ‘Double Deixis’ in Conversational Storytelling.” In The Travelling Concept of Narrative, edited by Matti Hyvärinen, Anu Korhonen, and Juri Mykkänen, 42–59. Helsinki: Helsinki Collegium for Advanced Studies. ———. 2013. “‘Unnatural’ Narratives? The Case of Second-person Narration.” In The Travelling Concepts of Narrative, edited by Mari Hatavara, Lars-Christer Hydén, and Matti Hyvärinen, 179–200. Amsterdam: John Benjamins. Mishler, Elliot G. 2006. “Narrative and Identity: The Double Arrow of Time.” In HCI Models, Theories, and Frameworks: Toward a Multidisciplinary Science, edited by Anna De Fina, Deborah Schiffrin, and Michael Bamberg, 30–47. Cambridge: Cambridge University Press. Montoro, Rocío. 2010. “A Multimodal Approach to Mind Style: Semiotic Metaphor vs. Multimodal Conceptual Metaphor.” In New Perspectives on Narrative and Multimodality, edited by Ruth Page, 31–49. New York/London: Routledge. Nunberg, Geoffrey. 1978. The Pragmatics of Reference. Bloomington, IN: Indiana University Linguistics Club. ———. 1990. The Linguistics of Punctuation. Stanford: CSLI. Oatley, Keith. 2002. “Emotions and the Story Worlds of Fiction.” In Narrative Impact: Social and Cognitive Foundations, edited by Melanie C. Green, Jeff rey J. Strange, and Timothy C. Brock, 39–69. Mahwah, NJ: Lawrence Erlbaum. Page, Ruth. 2010. “Re-examining Narrativity: Small Stories in Status Updates.” Text & Talk 30.4: 423–44. ———. 2012. Stories and Social Media: Identities and Interaction. New York/­ London: Routledge. Page, Ruth, Richard Harper, and Maximiliane Frobenius. 2013. “From Small ­Stories to Networked Narrative: The Evolution of Personal Narratives in Facebook S­ tatus Updates.” Narrative Inquiry 23.1: 192–213. Rank, Stefan, and Paolo Petta. 2005. “Appraisal for a Character-Based StoryWorld.” In IVA 2005 – Intelligent Virtual Agents – 5th International Working C ­ onference, edited by Themis Panayiotopoulos, Jonathan Gratch, Ruth Aylett, Daniel ­Ballin, Patrick Olivier, and Thomas Rist, 495–96. Berling/Heidelberg/New York: Springer. Schiffrin, Deborah. 1981. “Tense Variation in Narrative.” Language 57.1: 45–62. ———. 2009a. “Crossing Boundaries: The Nexus of Time, Space, Person, and Place in Narrative.” Language in Society 38: 421–45.

Storyworld in Text-Messages  143 ———. 2009b. “Tales of Transgression: Negotiating the Moral Order in Oral N ­ arratives.” StoryWorlds: A Journal of Narrative Studies 1: 61–77. Searle, John R. 1983. Intentionality: An Essay in the Philosophy of Mind. ­Cambridge: Cambridge University Press. Segal, Erwin M. 1995. “Narrative Comprehension and the Role of Deictic Shift T ­ heory.” In Deixis in Narrative: A Cognitive Science Perspective, edited by Gail A. Bruder, Judith F. Duchan, and Lynne E. Hewitt, 31–49. Hillsdale/Hove: ­Lawrence Erlbaum Associates. Stephens, Christine. 2011. “Narrative Analysis in Health Psychology Research: ­Personal, Dialogical and Social Stories of Health.” Health Psychology Review 5.1: 62–78. Tagg, Caroline. 2013. “Scraping the Barrel with a Shower of Social Misfits: Everyday Creativity in Text Messaging.” Applied Linguistics 34.4: 480–500. Tellegen, Auke, and Gilbert Atkinson. 1974. “Openness to Absorbing and Slf-­altering Experiences (‘Absorption’), a Trait Related to Hypnotic Susceptibility.” Journal of Abnormal Psychology 83: 268–77. Werry, Christopher C. 1996. “Linguistic and Interactional Features of I­nternet Relay Chat.” Computer-Mediated Communication: Linguistic, Social and Cross-­Cultural Perspectives, edited by Susan C. Herring, 47–63. Amsterdam/­ Philadelphia: John Benjamins. Zubin, David A., and Lynne E. Hewitt. 1995. “The Deictic Center: A Theory of Deixis in Narrative.” In Deixis in Narrative: A Cognitive Science Perspective, edited by Judith F. Duchan, Gail A. Bruder, and Lynne E. Hewitt, 129–15. H ­ illsdale, NJ: Lawrence Erlbaum Associates. Zwaan, Rolf A. 1999. “The Mental Leap into Imagined Worlds.” Current Directions in Psychological Science 8.1: 15–18.

This page intentionally left blank

Section III

This page intentionally left blank

8 Defending the Private and the Unnarratable Doomed Attempts to Read and Write Literary and Cinematic Minds in Marguerite Duras’s India Cycle Tytti Rantanen

INTRODUCTION: EXCEPTIONAL, UNREADABLE, OR UNNARRATABLE MINDS? Ten years ago, when I was first immersed in Marguerite Duras’s India Cycle, an ensemble of three novels and three films, I became anxiously haunted by its mysterious characters: the abandoned somnambulist Lol V. Stein, the decadent femme fatale of colonial Calcutta Anne-Marie Stretter, and the awkward vice consul Jean-Marc de H., a persona non grata suspended for an irrational shooting incident. Within the same texts there were other minds attempting to get hold of these oddities but ending up with a collection of uncertain fragments, rumours, or mere fabrications. Literary or cinematic, these narrativizing minds still did not seem able to get access to the whole story or to reveal the genuine essence of their targets, the Durassian unreadable and unnarratable minds. In his article on “unreadable minds,” H. Porter Abbott (2008, 448) encourages us to embrace “that peculiar combination of anxiety and wonder” that gains its full strength when we accept the unreadability of certain literary minds. Indeed, part of the fascination of fiction—and art in general—may be its ability to exceed our grasp. Although Duras touches upon heavy issues such as desire, death, madness, and trauma (both personal and transhistorical), the setting or events of her works are not necessarily so extreme. As readers or spectators, we do not face the unearthly ineffability caused by incredible monsters or hallucinations (see contributions by Brümmer and Kakko in this volume). Rather, not only anxiety but also excitement is aroused by both the unnarratability of Durassian minds and the impossible spatio-temporal storyworlds that, from one work to another, toy with the tension between writing and film, concrete and abstract. The vast quantity of the scholarship on Duras approaches this tension—both thematic and formalist—from a perspective of trauma theory, gender studies, or post-colonial studies. All of these approaches are relevant points of view. I am also keen to examine the relationship between Duras’s poetics and political activism. However, I suggest that we should not hasten to remove all the strangeness, but instead enter

148  Tytti Rantanen the Durassian void and treat the private unnarratable area sketched in her works not as a crossword puzzle to be stubbornly solved but as a sanctuary from stuffy assumptions and flimsy mind reading. In this chapter, I concentrate on two novels from the India Cycle, Le ravissement de Lol V. Stein (1964, The Ravishing of Lol Stein 1966) and Le Vice-consul (1966, The Vice-Consul 1968). In Le ravissement de Lol V. Stein (from now on, Lol V. Stein, and RLVS/RLS in citations), the character-­ narrator Jacques Hold delves into reconstructing the initial rejection trauma of Lol V. Stein. Lol’s fiancé has left her for another woman during a ball night. Ten years after Lol starts stalking her old friend Tatiana Karl and Tatiana’s lover, Jacques Hold. In Le Vice-consul (LVC/TVC in citations), there is no such central narrating consciousness as in Lol V. Stein. It is mostly the collective “on,” the white society of the colonized Calcutta, who gossips about the French Vice-Consul at Lahore, Jean-Marc de H., and the French ambassadress, Anne-Marie Stretter. Meanwhile, Peter Morgan is working through his white man’s burden by writing a novel about an anonymous mad Cambodian beggar woman. Later on, in the novel L’amour (1971) and in the three films of the cycle (La femme du Gange 1973, India Song 1974, and Son nom de Venise dans Calcutta desert 1976), Duras first stretched literary narration to the utmost and then detached sound from image, all the while working on the elements of the same eerie storyworlds. Toward the end of this chapter, I will examine briefly how the dynamics of unnarratability are elaborated in Duras’s cinematographical experiments. There are also other levels of complex narrative spatiality in the India Cycle, ­especially in the films, that operate simultaneously—but separately—on multiple diegetic levels. Most often the works are experiments with various kinds of ­“impossible” or “abstract” spaces. Duras was not the only author associated with the loose label of nouveau roman to test the limits of mental representations. In her essay “De Dostoïevski à Kafka,” Nathalie Sarraute describes the need for deciphering the deepest secrets of other fictive minds that gnaws away at the souls of Dostoyevsky’s characters: It is this continual, almost maniacal need for contact […] that attracts all these characters like dizziness and incites them on all occasions to try, by any means whatsoever, to clear a path to the “other,” to penetrate him as deeply as possible and make him lose his disturbing, unbearable opaqueness; in their turn, it impels them to confide in him and show him their own innermost recesses. (Sarraute 1990, 33) This tendency to speculate on other minds is obviously deeply rooted in reading and interpreting fiction, both at the level of the actual reader and that of a character inside the diegesis. It has also been—and still is—the very central topic of current narratological discussion. A variety of cognitive and metacognitive speculations can be found from realist to the most

Defending the Private and the Unnarratable  149 experimental narratives, but one cannot take for granted that this urge would gain any consummation. In many cases, the metacognitive efforts for mind-construction rather seem to underline the impotence and clumsiness behind human communication or to accentuate the uncanny effects of which fiction is capable. David Herman posits himself and the contributors to his anthology The Emergence of Mind (2011, 8–9, see also Andersson in this volume) in opposition to what he calls the “Exceptionality Thesis,” i.e., the premise that readers experience fictional minds on a different basis than those in the real world, for it is only in fiction where we possibly could have access to another mind. This ethos can be traced back to Dorrit Cohn’s classical study Transparent Minds (1978), and, according to Herman, further applications are localizable in the research of “unnatural narratologists.”1 He argues that fictional narratives do not afford any clear-cut view of others’ minds and therefore the everyday cognitive and folk psychological processes of meaning-making and mind-reading become applicable to reading fiction as well. I wholly agree with Herman’s reservations regarding the transparency of fictive minds. Our paths differ at the point where in my view it seems a more fruitful and interesting solution to abandon folk psychology and proceed bravely beyond the “Exceptionality Thesis” (or “separatist approach,” as Andersson calls it), to examine how fiction, slick in its swerves, may turn down even the most artful efforts to penetrate other minds. It seems that the recent emphasis on mind in post-classical narratology sometimes too eagerly embraces the idea of constructing, reading, and interpreting other (fictive) minds as a proof for the overall benevolence of “natural” human interaction. If we are to follow Herman’s ethos (which Stefan Iversen calls the “Similarity Thesis” 2013, 99) and lump fictive minds together with actual minds, or a more general tendency of cognitive narratology to renaturalize impossible narrative elements, we risk taming the “affective power” or missing the “essential dynamics” between a work of verbal art and real-life experientiality, to which “separatists” such as Iversen (ibid., 96) and Maria Mäkelä (2013, 130) alert us. Taken to its extreme, this mind-reading may lead to flagrant overinterpretations and narrative exploitation. If we ceaselessly aim at verbalizing each other’s hidden inner worlds, what kind of space is left for the individual freedom of not being “connected” every time everywhere? In this chapter, I aim to demonstrate how Duras’s works abandon the belief that constructing other minds is always an authentic and decent form of human interaction—at least in narrative fiction.2 I will also discuss briefly how Duras elaborated this rejection of coherent communication in her India Cycle films by separating sound from image. The more frantic the efforts for mind constructing, the more ambivalent the result (if there even is a result). By highlighting this impossibility, these novels introduce a space for readings that lean more on the “negative way” of knowing than on clear-cut cognitive reasoning.

150  Tytti Rantanen José García Ángel Landa (2011, 437) argues that narrative may provoke even claustrophobic feelings in a mind longing for the openness of unplotted reality, free from any “manipulative and vicious” narrative pattern. Thus, the resistance to narration is a battle for space and certain limits for privacy. Moreover, in the introduction to their anthology Beyond Narrative Coherence (2010, 7), Matti Hyvärinen, Lars-Christer Hydén, Marja Saarenheimo, and Maria Tamboukou pose the question of whether narrative coherence might even be a harmful phenomenon, as it intertwines with the problematics of power, idealization, and marginalization. The mistrust of narrative structures as a tool for parsing experience and social environments is not a new phenomenon. This concern has its echo in an overall post-war tendency to regard Western metaphysics and reason as dubious.3 As Sirkka Knuuttila (2011, 155) points out, Duras’s work can be positioned in the continuum of the antipsychiatry movement of the 1960s and the 1970s, as the author revises the age-old imagery of madwomen, letting them “flee from all control of reason” in a way that appears to be a “strong emancipatory vehicle.” Facing or trying to surmount one’s narrative inadequacy is one of the central features of the India Cycle. Peter Morgan wants to “take the misery of Calcutta” and tries to process the exotic, horrifying “unheimlich,” the poverty and the leprosy, by writing. For him it seems to be a therapeutic act. Jacques Hold also gets truly obsessed with reconstructing Lol’s mind and eventually turns this absent-minded woman into a projection of his own (auto)erotic fantasies. We do not get “authentic” access to Lol’s consciousness; in the novel, there are sequences that are narrated in free indirect discourse by a third-person narrator, but in the middle of the novel, it turns out that Jacques Hold is actually both the third and the first-person narrators. Thus, we can hear Lol’s own voice only in the direct discourse, in the dialogue. Otherwise Lol stays in her private unnarratable area. I see this kind of narrative tension as a struggle for space: the narrator, Jacques Hold, is trying to penetrate into this space, and Lol tries to defend it. THE URGE FOR AND THE FAILURE OF READING—AND WRITING—MINDS Jacques Hold does not content himself with passing around ten-year-old rumours about the somnambulist-like Lol, but rather takes an active part in producing the story and participating in its events, the little scenes Lol keeps witnessing. Mutual stalking creates a peculiar liaison between Lol and the man. She actively chooses Jacques Hold to take part in her almost ritualistic re-enactments of the initial rejection trauma, but Hold aims at taking full narratorial agency in reconstructing Lol. If we are to believe his own testimony, he gets closer to Lol than old small-town rumours or Tatiana’s ­faux-semblant memories can. Still, what else are Jacques Hold’s own fabulations than faux-semblant—false fabrications creating an illusion

Defending the Private and the Unnarratable  151 of authenticity? Jacques Hold is so persuasive a narrator that as Sirkka Knuuttila (2011, 118) critically points out, even in numerous studies on Le ravissement de Lol V. Stein the representations of Lol’s emotions and reflections are taken as having direct access to her mind, her feminine desire. It is deceptively easy to ignore the fact that everything except Lol’s dialogue is mediated through the obsessive janus-faced intradiegetic narrator. Maria Mäkelä (see 2006) has examined a great deal of fictive characters who, in order to fabulate other (possible or even imaginary and fantasized) minds, seize narrative capacity conventionally considered unattainable to them. Still, she doubts whether these literary minds could be aware of their own literariness and constructedness (Mäkelä 2013, 148). In the case of Lol V. Stein, Jacques Hold, in somewhat flexible fashion, both repeatedly foregrounds his role as a fabricator (“j’imagine, j’invente”) to the extent that his own mind nearly turns unreadable and is occupied by fantasies— trying to occupy Lol’s mind in turn. At least one thing is certain in this never-ending, hypothetical, embedded game of hide and seek: even if Lol had an unreadable and unnarratable mind, that of Jacques Hold is a highly narrativizing one. Cognitive narratology tends to take for granted that the model for narrative minds is a benevolent and uncomplicated “everyman,” even ­ though it is not always the most interesting choice. Quite the contrary, as Mäkelä (2006, 257) has earlier argued, it seems that it is the excessive, artistic, obsessive, paranoid mind that is similar to techniques of modernist narration—not an “ordinary” human mind resorting to the frames of everyday experientiality. Lol is prejudged to be insane because of her i­ ntroverted and absent-minded roaming, but Jacques Hold himself does not come across as Prince Valiant, innocently worried about her sanity. Rather, his stalkertype obsessiveness matches up with Mäkelä’s description of “creepy” modernist narrators. This could be an alternative way of understanding the ­“Exceptionality Thesis”: fictive minds might be something exceptional, perverted, and extraordinary, had we not good reason to suspect that even actual minds are quirkier and more complicated than the nice, communicative default bonhomme of folk psychology.4 Alain Robbe-Grillet also emphasizes the movement from realistic representation toward more twisted mental landscapes in modernist fiction: [It is] as if the false—that is at once the possible, the impossible, lies, hypothesis, etc.—had become one of the privileged themes of modern fiction; a new kind of narrator is born: no longer a man who describes the things he sees, but at the same time a man who invents the things around him and who sees the things he invents. Once these hero-narrators begin ever so little to resemble “characters,” they are immediately liars, schizophrenics, or victims of hallucinations (or even writers, who are creating their own story). (Robbe-Grillet 1965, 162–63)

152  Tytti Rantanen I do not believe that Robbe-Grillet wanted to lump modernist narration and insanity together. Rather he seems to declare that the ideal and standard for “new realism” emerges no more from the spotless mind of a benevolent Everyman, but, as Mäkelä suggests, from the obsessive and artistic features of these creeps. Like Jacques Hold, they may get carried away by their own creative potential. Part of the critique addressed to the usability of folk psychology and mind reading stems from the understanding of “our normal everyday stance toward the other person” not as a third-person spectatorial, clinical observation, but as a second-person interaction (Gallagher 2008, 164; Iversen 2013, 100–101). In the real-life context, this undoubtedly is a healthy adjustment. But in terms of twisted fictions, the suspicious act of mind reading seems to happen in a highly fabulatory first-person context, in the realm of fantasies and projections that might have very little to do with the genuine mind of their object. Interaction can easily turn into (self-)deception with distorted or selfish motives. Impeding the access to fictive minds can turn into an ethical choice, at least if we echo the fury with which Nathalie Sarraute discusses the psychologizing characters who endeavour to penetrate each other’s minds, regardless of the fact that their assumptions are formed within the limits of their own consciousness. Thus, the clairvoyant tendency in literature is not a privilege of “those who are enlightened by Christian love but of all these dubious characters, […] these larvae who continue to dig and stir in the very dregs of the soul and sniff with delight its nauseous slime.” (Sarraute 1990, 36–37). In the India Cycle, most of the rumors, legends, and collective ­speculations circulated by the gossiping middle class or colonial elite are bound to this never-ending hunger for exploitative excitement. Jonathan Culler (1986, 134) describes the perpetual balancing act between art and communication: the power of literature is in its nature “something other than ordinary communication.” Still, the reading process is tempted to reduce and recuperate its strangeness in the process of naturalization, bringing it “within our ken.” Could the process be two-fold? Literature (among other narrative arts) is also capable of portraying the shortcomings of “natural” interaction. In Duras’s œuvre, it is a recurrent pattern that one character or a collective of characters craves for information and interpretations of another, mysterious character. As little is known and even less understood, characters like Lol V. Stein, the Vice-Consul, or Anne-Marie Stretter cease to be their own autonomous personalities and become stimuli for different kinds of fantasies and projections. Their image is built upon mundane details and rumours that begin to have a narrative presence of their own. Thus, this interpretive project is something more aestheticized than mere everyday folk-psychology based interaction—it is more about trying to narrativize these intriguing outsiders on the basis of what is told of them. Alas, the final result of this process is not a genuine personality in all its nuanced diversity but rather a quasi-mythical caricature. As Landa (2011, 419–20) remarks, narrativity often involves reworking previous experiences in order to produce new ones in new articulations. In

Defending the Private and the Unnarratable  153 the intratextual nexus of variants the India Cycle forms, no events or motifs are repeated as identical, nor do they cumulate. Each variant is gnawing away at the possible original event—or perhaps there is no original ­version, as we never get direct access to the whole picture. When different elements of the storyworld—or the India Cycle mythos—are being repeated, they become more and more stripped down. One event that is repeated in every narrative in the Lol V. Stein storyline (including—in addition to Le ravissement de Lol V. Stein—the novel L’amour and the film La femme du Gange) is the trip to the initial “crime scene,” the municipal casino of T. Beach. Here is the first variant from Le ravissement de Lol V. Stein: Lol regardait. Derrière elle j’essayais d’accorder de si près mon regard au sien que j’ai commencé à me souvenir, à chaque seconde d’avantage, de son souvenir. […] J’ai entendu les fox-trot d’une jeunesse sans histoire. Une blonde riait à gorge déployée. Un couple d’amants est arrivé sur elle, bolide lent, mâchoire primaire de l’amour, elle ignorait encore ce que ça signifiait. […]Nous avons entendu le déclic d’un commutateur et la salle s’éclaire de dix lustres ensemble. Lol pousse un cri. […] L’homme éteint. […] –Il y a longtemps ? demande-t-il. –Oh, dix ans, dit Lol. –J’étais là. Il change d’expression, reconnaît mademoiselle Lola Stein l’infatigable danseuse, dix-sept ans, dix-huit ans, de la Potinière. Il dit : –Pardon. Il doit savoir le reste de l’histoire aussi, je le vois bien. Cette reconnaissance échappe complètement à Lol. (RLVS, 180–82 emphasis added) Lol was looking. Behind her, I was trying to accord my look so closely to hers that, with every passing second, I began to remember her memories. […] I heard the fox trots of an uneventful youth. A blonde was roaring with laughter. A couple—two lovers—came toward her, a slow-moving comet, the primary maw of love, she still didn’t realize what it meant. […] We heard the faint click of a light switch, and the ballroom’s ten chandeliers light up together. Lol gives a cry. […] The man turns out the lights. […] “Has it been a long time?” he asks. “Oh, ten years,” Lol says. “I was here.” His expression changes, he recognizes Miss Lol Stein, the indefatigable dancer—seventeen years old, eighteen—of the Potinière. He says: “I’m sorry.” He must know the rest of the story, too. I can see he obviously does. Lol hasn’t the slightest inkling that he knows. (RLS, 170–71 emphasis added)

154  Tytti Rantanen At first, as usual, Jacques Hold tries to absorb Lol’s memories into his own imagination. He seems to reach flickers of moments just before the young girl became the mysterious woman he now knows (or purports to know). Once again he reconstructs a simulation of Lol’s mind. But as the moment is so heavy with emotion, this is not enough. When the caretaker switches on the lights, Lol seems to have such a vivid recollection that she reacts the same way as she did ten years ago: she gives out a cry. Excited, Jacques Hold extends his mental interpretations to the cognition of the caretaker as well. On the basis of the facial expression of the caretaker (only vaguely described), Jacques Hold assumes that the man must recognize Lol V. Stein and that “[h]e must know the rest of the story too; I can see he obviously does.” But it is just as possible that the caretaker is only confused in front of this weird, crying woman. Hold does not specify what he means by “the rest of the story,” but he probably refers to Lol’s mental collapse and seemingly serene marriage shortly after the ball night. For Hold himself, this is not “the rest of the story” but only the beginning of his own narrative project. WRITING THE OTHER MIND, PERFORMING THE SELF According to Emma Kafalenos, when faced with confusing situations, fictive minds tend to react just like actual readers: if we are lacking essential information, we try to recuperate those situations as chains of events, first chronologically, then causally. These chains form narratives in turn, objects of interpretation (Kafalenos 1999, 35–36). Of course in fiction, events do not form sound and intact chains, and when characters trying to understand each other’s actions and the motives behind them form these chains of information, any lacunae may prove to be crucial for a coherent interpretation (ibid., 55). This of course needs not be a problem but is rather a poetic asset. Like the obsessive Jacques Hold, Peter Morgan, a young colonial officer, puts his soul into fabulating the beggar woman’s progressive insanity and stream of consciousness in Le Vice-Consul. Still, as soon as he puts his pencil down, he admits that he really does not know much about the actual creature that has given him the inspiration: Elle est là, devant la résidence de l’ex-vice-consul de France à Lahore. A l’ombre d’un buisson creux, sur le sable, dans son sac encore trempé, sa tête chauve à l’ombre du buisson, elle dort. Peter Morgan sait qu’elle a chassé et nagé une partie de la nuit dans le Gange, qu’elle a abordé les promeneurs et qu’elle a chanté, c’est ainsi qu’elle passe ses nuits. Peter Morgan l’a suivie dans Calcutta. C’est ce qu’il sait. […] Peter Morgan est un jeune homme qui désire prendre la douleur de Calcutta, s’y jeter, que ce soit fait, et que son ignorance cesse avec la douleur prise. (VC, 29)

Defending the Private and the Unnarratable  155 There she is, opposite the residence of the former Vice-Consul of France in Lahore. In the shade of an overhanging bush, her dress of coarse ­sacking still sopping wet, she lies asleep. Her bald head is shaded by the bush. Peter Morgan knows that she has spent part of the night swimming and hunting for food in the Ganges, and the rest accosting passers-by in the streets and singing. This is how she spends her nights. Peter Morgan has followed her through the streets of ­Calcutta. This is what he knows. […] Peter Morgan is young. He wants to shoulder the misery of ­Calcutta. He wants to plunge into its depths. He wants to do it now, to get it over, so that wisdom may start to grow out of bitter experience. (TVC, 18) Peter Morgan thus indulges in writing in order to process the unsayable horrors of the exotic country. He still manages to keep his consciousness separated from that of the unknown woman who speaks a language he is not able to understand, even his authorial role remains visible in most of the mise-en-abîme passages: “Elle marche, écrit Peter Morgan” (RLVS, 9). Jacques Hold, too, repeatedly admits that he is inventing and imagining things he does not know or find satisfactory, but still the boundaries become notably more blurred than in Le Vice-consul. In Le ravissement de Lol V. Stein as well as in the other parts of the India Cycle, the role of the narrator as a seeing, perceiving, and experiencing entity becomes foregrounded in many ways. The narrating pronoun and temporal constructions of retrospection and simultaneity can alternate even inside the same paragraph, which makes it difficult to draw clear lines to separate narrating, perceiving, and recalling. I am not claiming that Jacques Hold narrates the events as they occur (as this is clearly not the case), but through narration he relives these moments time and again, develops new theories and conclusions about Lol’s mental state, and evaluates and re-evaluates himself and his behaviour—especially in those sections, where the narration “zooms out” representing Jacques Hold in the third person. Thus, the narrator inevitably becomes the object of his own narration. Every first person narrator is unavoidably subjective as regards his or her own story—let alone that of someone else. What we do get more or less straight from Lol is what she herself tells in the dialogues: she for example admits having stalked Jacques Hold and Tatiana Karl from the rye field behind the Hôtel des Bois. In this regard, Jacques Hold is not necessarily a dishonest narrator though he should not have access to Lol’s mind in his position. Lol’s mind might be as dizzy as Jacques Hold represents it. In his recent article on “naturalizing and unnaturalizing reading strategies,” Henrik Skov Nielsen aims at clarifying Genette’s concept of focalization. Nielsen suggests that the unspoken keyword to a more dynamic comprehension of Genettean focalization is “restriction,” for focalizaton

156  Tytti Rantanen can be defined as “a restriction of access to point of view” (Nielsen 2013, 75). Interestingly enough, constructed focalizations in the India Cycle abandon this route: while writing his tale of a mad beggar woman, Peter Morgan tries to immerse himself in the point of view of a strange woman he has seen on the street in order to adopt her experiences and write them out. At the same time, he of course becomes an author in his own right inside Duras’s narrative. When it comes to Jacques Hold and Lol, we could label their narrative power relation as an expansion of access to a point of view—except that this point of view is after all only hypothetical. According to Genette (1972, 213), paralepsis, the focalizing character’s impossible or unnatural knowledge of events and other characters, should not be confused with the reader’s interpretations of small clues insinuated by the narrator, as these clues can easily surpass the limited comprehension and reasoning of a character. But just as it is not relevant to explain the uncanny reporting by Jacques Hold to be mere fantasy, Genette is not that interested in discussing whether the unnaturalness of paralepsis emerges from physical impossibility or psychological implausibility. He would rather like to examine what kind of impact it has on the overall coherence and narration of a text, or to put it simply: what kind of poetic weight it has (ibid., 221–22). Rüdiger Heinze (2008, 282) comes to the same conclusion (though via ­Jonathan Culler) that paralepsis is a more precise label for phenomena that often are described as a first-person (or character) narrator’s “omniscience.” Both Le ravissement de Lol V. Stein and Le Vice-consul as well as Duras’s other writings foreground the intricacy of literary perception and epistemology. Henrik Skov Nielsen (2004, 133, 136) has studied peculiar first-person narratives, where the narrating I is reporting information on things he simply has not been able to experience or observe himself or, as often is the case with classic literature, repeats discussions and details to an extent that would be impossible for a “natural” mind. Nielsen suggests establishing a new narrative category, that of the “impersonal voice in first-person fiction.” This category is able to transgress the limits of the ordinary and to give the narrating I access to the area that has traditionally been privileged to the extra-diegetic third-person narrator (ibid., 139–40). Le ravissement de Lol V. Stein evokes confusion about whether we should take Lol’s mental representations for granted, even if they are only Jacques Hold’s constructions and every single piece of information (or interpretation) comes from the narrator who himself inhabits and is deeply involved in the storyworld. This obsessive involvement, a cognitive struggle, is hard to leave aside, as it grows to be the dominant element of the novel. Instead of arguing whether the first-person narrator is reliable or unreliable, Nielsen (ibid., 143) would prefer taking notice and accepting that the limit between the first and the third person narration could be more vague than traditionally considered. I find this insight valid, as the narration in Le ravissement de Lol V. Stein in any case ceaselessly shifts between different narrative pronouns. The limit between different fictive minds is rather a question of

Defending the Private and the Unnarratable  157 interpretation than something to be traced carefully from different linguistic signifiers. On the whole, one of the greatest merits of Nielsen’s model (and many other models created in post-classical narratology) is its aim to extend the narrow space established for each separate case of narrative voices. Substantially less artfully than Jacques Hold in Lol V. Stein, the high society in Le Vice-consul also tries to understand Jean-Marc de H. by eagerly scouting every minor detail of his childhood and adolescence, even though the vice consul’s aunt tries to implicate in her letter that the key to the mystery might lie somewhere else: “La conduite insensée de mon neveu à Lahore ne témoigne-t-elle pas en fin de compte de quelque secret état de l’âme, de quelque chose qui nous échappe mais qui n’en est peut-être pas pour autant tout à fait indigne? Avant que d’être tout à fait blâmée, cette conduite ne devraitelle pas être considérée avec attention, peut-être dans son principe? Pourquoi remonter à l’enfance pour expliquer sa conduite à Lahore ? Ne faudrait-il pas chercher aussi à Lahore?” –“Je préfère qu’on en reste aux conjectures habituelles, qu’on cherche dans l’enfance, dit l’ambassadeur.” (LVC, 42) “Admittedly, my nephew behaved like a madman in Lahore, but, when all is said and done, could this not perhaps point to some hidden emotional disturbance which eludes us, but which may not be entirely discreditable? Before condemning his conduct out of hand, would it not perhaps be advisable to go into it further, to probe in depth? Why go back to his childhood for an explanation of his conduct in Lahore? Should not enquiries also be made on the spot?” “I prefer to follow the normal procedure, and investigate the childhood background,” says the Ambassador. (TVC, 29) Perfectly aware of the enormous interest focused upon him (and probably enjoying every single moment), the Vice-Consul takes advantage of these “habitual conjectures” and makes sure his only companion, the alcoholic director of the European Circle, passes on every puzzling confession he makes. The habitual conjectures, usual suspects, childhood traumas, and all manner of folk psychologizing seem after all to lead the keen explorers far away from the original enigma. One case of interpretational fallacy occurs when the straightforward wife of the Spanish ambassador is engaged in small talk with the Vice-Consul Jean-Marc de H. and (incorrectly) thinks she has triumphed over his undefinable strangeness, his “impossibility,” naturalizing it to be mere fear of leprosy: –C’est à dire … il y en a … rarement remarquez, mais cela arrive … la femme d’un secrétaire, chez nous, ou consulat d’Espagne, elle devenait

158  Tytti Rantanen folle, elle croyait qu’elle avait attrapé la lèpre, il a fallu la renvoyer, impossible de lui enlever cette idée de la tête. […] –Mais au fait, avait-elle la lèpre cette femme ? Alors elle s’écarte et, tout en évitant de le regarder, elle se rassure, croit avoir découvert enfin quelque sentiment familier chez le viceconsul : la peur. (LVC, 112–13) “Well, I must admit, actually, there aren’t many, but it does ­happen. There was the wife of one of our own Secretaries at the Spanish C ­ onsulate; she went out of her mind. She was convinced she had contracted leprosy. Nothing would persuade her otherwise. She had to be sent home.” [—] “But had this woman, in fact, contracted leprosy?” She moves away from him and, although not looking at him, feels reassured, believing that she has at last awakened a normal response in the Vice-Consul: fear. (TVC, 87–88) This “natural” interpretation proves to be false: the Vice-Consul’s decadent unease cannot be explained away with one single diagnosis. Nor is his obsession with Anne-Marie Stretter a simple outburst of heterosexual erotic desire. Yet among the narrow-minded circles of the white Calcuttan élite a “normal response” (or, actually, “a familiar feeling”) is fear. False recognition of this shared fear would assimilate the unpredictable man as part of “normal procedures” and choreographies of reason, good manners, and harmless chit-chat. If we compare the white society of Calcutta and Jacques Hold as Lol’s “readers,” we can see an analogy with Culler’s (1986, 114) depiction of literary (in)competence: an incompetent reader can linguistically piece together the phrases formed by words, but after that he is baffled and does not know “what to make of this strange concatenation of phrases,” as he does not master the grammar of literature. A lazy and incompetent reader does not see anything special behind the “language” of Lol’s or the Vice-Consul’s behaviour and is satisfied with mundane gossiping, whereas Hold is—or thinks he is—capable of reconstructing a nuanced interpretation of Lol’s fascinating inner world. The analogy is apt, because as the narrator of the whole novel Hold is not constructing Lol’s mind out of philanthropic or medical interest, but cultivates the whole process as a part of his artistic creation, compensating for “the lack of cold, hard facts about her life.” As a self-assertive narrator, Jacques Hold is not only interpreting, but rather producing Lol. This bears a distant resemblance to Peter Morgan’s narrative project in Le Vice-consul, but even if he might project his own feelings of discomfort and nausea to his narrative of a mad beggar woman, he does not subsume himself to be a part of the text.

Defending the Private and the Unnarratable  159 THE NEGATIVE WAY OF KNOWING It is difficult if not impossible to reach a conclusion about what Lol can finally see from her stalking point in the rye field behind the hotel. Once again, Jacques Hold takes liberties to function as Lol’s vision with his imaginative attempts to occupy her consciousness. Thus Lol is only seemingly the subject of these voyeuristic acts. Jacques Hold imagines Lol’s seeing things she really cannot see, at least if we are to believe in her own account: –Ce qui est passé dans cette chambre entre Tatiana et vous je n’ai pas les moyens de connaître. Jamais je ne saurai. Lorsque vous me racontez il s’agit d’autre chose. (RLVS, 136) “I have no way of knowing what went on in that room between Tatiana and you. I’ll never know. When you tell me, it’s something else.” (RLS, 125) For Lol, falling asleep is a chance to flee Jacques Hold’s narrative desire into nothingness. Interestingly enough, even though Jacques Hold is imagining this “autre chose,” the things Lol could see or emotions she could feel, he does not tell us what she is dreaming about when she’s sleeping. Falling asleep is one shady corner of Lol’s private unnarratable area, where Jacques Hold loses his narrative hold. In essence, Lol V. Stein and Jean-Marc de H. are alike as characters. For the people around them they are both fascinating and disturbing obsessions. They are always seen through their “madness”; all external properties cease to be individual traits. They are interesting only as the evidence or anticipations of the dominant insanity. Both in India Song and Le Vice-consul this exploitation reaches a point where Jean-Marc de H.’s performance is faced with cold ignorance: “Excusez-nous, le personnage que vous êtes ne nous intéresse que lorsque vous êtes absent.” (LVC, 147, “Forgive me, but a man of your sort is only interesting in his absence” TVC, 116.) Is Lol V. Stein as well only interesting for Tatiana Karl, Jacques Hold, or even for us readers in her absence? The absence however seems to give Lol more room, a private space she can master by falling asleep or wandering around the streets of S. Tahla. Even in her presence Lol remains mysterious and does not articulate her motives at all or only occasionally. She stays in the background. For Jean-Marc de H., the blunt remark is an insult, as he particularly wants to be present. He even starts the whole performance in order to generate something “public” between him and Anne-Marie Stretter. The trip to the initial “crime scene,” the municipal casino of T. Beach, seems to mark a turning point in Lol’s traumatic enactment and in her and Jacques Hold’s twisted relationship: Elle me parle de Michael Richardson sur ma demande. Elle dit combien il aimait le tennis, qu’il écrivait des poèmes qu’elle trouvait beaux.

160  Tytti Rantanen J’insiste pour qu’elle parle. Peut-elle me dire plus encore? Elle peut. Je souffre de toutes parts. Elle parle. J’insiste encore. Elle me prodigue de la douleur avec générosité. Elle récite des nuits sur la plage. Je veux savoir plus encore. Elle me dit plus encore. Nous sourions. Elle a parlé comme la première fois, chez Tatiana Karl. La douleur disparaît. Je le lui dis. Elle se tait. C’est fini, vraiment. Elle peut tout me dire sur Michael Richardson, sur tout ce qu’elle veut. (RLVS, 190) At my request, she talks to me about Michael Richardson. She tells me how great a tennis buff he was, says that he used to write poetry that she found beautiful. I urge her to talk about it. Can she tell me anything more? She can. Each word is a shaft of pain wracking my whole body. She talks on. Again I urge her to continue. She lavishes pain with generosity. She tells me about nights on the beach. I want to know still more. She tells me still more. We are smiling. She’s been talking the way she did that first night, at Tatiana Karl’s. The pain vanishes. I tell her so. She says no more. It’s over, truly over. She can tell me anything, whatever she wants to about Michael Richardson, about anything. (RLS, 180) Hold feels unease with Lol’s opening up: If Lol becomes strong enough to find the absent words, Hold loses his narrative hold, useless. But does the recuperation and release truly take place? Is Lol using the whole range or verbalization and narrativization of her own life, emotions, and experiences? What is she talking about when she is talking about Michael Richardson? A bunch of dispensable anecdotes and details, the same kind of scattered material the Vice-Consul is so eagerly feeding the Club Secretary. Jacques Hold’s association of Lol having “been talking the way she did that first night, at Tatiana Karl’s” does not engender an idea of profound and selfanalysis, as Tatiana remarked on that same “first night” that Lol is recounting her life “as a book.” Even Jacques Hold admits his inevitable incompetence, but he however takes it as a possibility, a new way to approach Lol through negation. For him, it is a way to stick out from the ignorant mass, “the perverters of the truth”: En ce moment, moi seul de tous ces faussaires, je sais: je ne sais rien. Ce fut là ma première découverte à son propos: ne rien savoir de Lol était la connaître déjà. On pouvait, me parut-il, en savoir moins encore, de moins en moins sur Lol V. Stein. (RLVS, 81) Now, I alone of all these perverters of the truth know this: that I know nothing. That was my initial discovery about her: to know nothing

Defending the Private and the Unnarratable  161 about Lol Stein was already to know her. One could, it seemed to me, know even less about her, less and less about Lol Stein. (RLS, 72) Thus, to know less and less about Lol V. Stein seems an insightful strategy for coping with the unbearable paradox of understanding as a violation. Turning the habitual and predictable upside down is an efficient strategy to confuse the restrictions of a petit bourgeois consumer society, refusing the unspoken laws of comme il faut. As David Herman (2011, 8) summarises, “folk-psychological rules of thumb are what people use to characterize their own and others’ reasons for acting in the ways they do.” When these rules no longer apply, the predictable gives way to the unpredictable. Thus, making ready-made assumptions and too convenient mind-reading exercises ridiculous is also a political act. It is interesting to notice, that the 1968 riots mark a change of ­dynamics in Duras’s œuvre. Trauma and madness do remain the essential Durassian topoi from her early works till her death. And yet, before 1968 her novels and scripts are more likely to portray a singular deranged character in a more or less “normal” world, whereas from Détruire, dit-elle (1969) onwards, a “negative way of knowing” and letting go of conventional reasoning start to characterize not only hapless individuals, but turn into a collective, poetic, and political asset. “Normal” people, stuck to the invalid rules of folk-psychology become lost outsiders in Duras’s kingdom of freaks. Either they face an overwhelming yet inquieting “love” as Elisabeth Alione’s rigid husband in Détruire … or just a strong, empty stare as the unsuspecting young salesman in Duras’s 1972 film Nathalie Granger. Thus, contrary to what Abbott (2008, 465) argues, the immediate experience of an unreadable fictional mind does not necessarily preclude all empathy, for while it abandons the “presumption of a readable mind,” it does enable facing the twisted storyworld with an open mind—not open to be read, but open to respect before mental sovereignty or, in Abbott’s words, “the human unknowable.”

FROM VOICE OF GOD TO VOICE OF RUINS Before concluding, I will take a brief glance of unnarratable space in the India Cycle’s films. In her post-1968 exhaustion, Duras abandoned the novel for some 15 years, producing minimalist “récits,” drama and films. Her films, including those belonging to the India Cycle, continue to confuse the borders of expression. With La femme du Gange (1973), Duras made her most long-lasting cinematic innovation: the separation of sound from image. She developed this innovation further with her following film, India Song, which confounds the process of mind construction even more. First of all, we can hear the main characters like Anne-Marie Stretter or the

162  Tytti Rantanen Vice-Consul speak at the same time as we can see them on the screen—but there is no synchronized lip movement. Second, in the soundtrack we hear disembodied voices that seem to comment on the events on the screen or try desperately to construct the story from vague recollections and hesitations. The third group of voices is those of the white elite, the gossiping guests at the Ambassadress’ reception, but we never see those people or the orchestra playing 1930s dance music. With her next film, Duras went even further. For the exterior scenes of India Song she used an abandoned, almost ruined mansion, the Rothschild villa in Bois de Boulogne near Paris. Haunted by the villa, Duras came back and filmed the ruined palace from cellar to attic. This footage became Son nom de Venise dans Calcutta désert (1976). In this film we see almost nothing but ruins, and in a way, this film is the ruination of its predecessor. Apart from the last two or three minutes, the film shares the same soundtrack with India Song. We hear the same voices talking and gossiping, the same orchestra playing the same tango, but except for couple of brief, seemingly unrelated scenes, there are no bodily referents visible on the screen, not even to the extent of India Song. The strongest degree of narrativity can be found in the soundtrack, if at all. Likewise, even if we were familiar with the main characters from India Song, the content of their minds remains more or less as enigmatic as in Duras’s works in general. It is revealing that in many scenes we see first the reflection of a character in a mirror, and only then the “real” character. Duras’s cinematic experimentations inspired Deleuze to embrace a new kind of autonomy between the two components, sound and image—this is how modernist cinema, emerging from the domain of complex “timeimage” instead of more conventional “movement-image,” gains the essential possibilities of being audio-visual. As a result, the voice-over loses its omnipotence, it becomes doubtful, uncertain, and ambiguous—but this leads to a new kind of autonomy. The sound image, the auditive image, is born from the rupture with the visual image (Deleuze 1985, 326–27). In Michel Chion’s audio-visual theory, the use of voice-over creates different kinds of “acousmêtres,” “acousmatic characters whose relationship to the screen involves a specific kind of ambiguity and oscillation.” Usually these acousmêtres are omniscient, “the voice of God”-like entities, as in the most typical use of voice-over, but in Duras’s case, they are rather partial or paradoxical acousmêtres, as they do not master the visual content on which they are commenting (Chion [1990] 1994, 129–30). Thus, the two female voices in the first sequence of India Song are not violently imposing any selfenclosed narrative. Although they are repeating the common speculations, they reveal more about themselves and their mutual passion than about Anne-Marie Stretter. They approach her mystery with tender and empathizing questions: V2: V1:

Ne souffre pas, n’est-ce pas? Non plus … Une lèpre … du cœur.

Defending the Private and the Unnarratable  163 […] Non, ne supporte pas … les Indes … ne supporte pas … (Duras 1979, 220.) V1:

She doesn’t suffer, does she? Not anymore … a leprosy … of the heart. […] V1: No, she can’t stand … India … can’t stand … V2: V1:

There is something sensual in the way these voices are tenderly trying to weave together the worn-out, almost mundane fragments they are left with. They are full of cognition and emotion—in a way, with these voices we are left with nothing but narrative minds. Like Jacques Hold, the voices aim at constructing their own version of the characters’ destinies, but the fulfilment and succession of narrative desire remains ambiguous. As radical as the solution to detach the image from sound is Duras’s tendency to leave film and text interwoven. From published scripts we can read abstract mergings of storyworlds practically impossible to realize. In Le camion (1977), we have a film of Duras and Gérard Depardieu reading a script that could have been a film called Le camion. These spatio-temporal and ontological violations invite us to extend the limits of a cinematic work toward the other films, back to a printed text, again apart from it, to be watched, listened, read, and reflected on. In Transparent Minds (1983, 7), Dorrit Cohn claims that (literary) narrative fiction is “the only kind of narrative” able to portray “the unspoken thoughts, feelings, perceptions of a person other than the speaker.” In the same year, Bruce F. Kawin classified varying images and impulses that do not necessarily relate to any cinematic character’s consciousness as the projections of a “mindscreen,” the narrating offscreen intelligence. According to him, a film is an artificial system without its own consciousness. Nevertheless, it is capable of imitating a consciousness the locus of which is this mindscreen, a medium for reflecting the problems of “being-within-limits,” facing the limits of self-definition and self-deception (Kawin 1978, 55, 115, 187). The question whether Cohn’s condition “other than the speaker” is actualized in the cinema is tricky, for the whole ontology of “speaker,” “shower,” or “narrator” in cinema is disputable. In highly problematic cases like this, one welcomes readily the newly risen interest in the concept of the author in narrative studies. While discussing “inventing authors,” Henrik Skov Nielsen (2013, 85) is ready to abandon the narrator as “the main agent of telling,” for “the source of information […] is not the unknowing character but the world-creating author.”5 When it comes to Duras, she is rather world-destroying than world-creating as an author, with her fascination of ruins and devastation. In the field of film studies, the question of the author has not been such a taboo as in literary criticism, owing to the 1950s tradition of critique des auteurs. David Bordwell even counts the element of auteurism as a part of the semiotic

164  Tytti Rantanen system of the art film mode. Doing this, he does not mean that the director would be relevant as an individual as such, but as a part of the overall spirit and the organization of a film (Bordwell 2008, 154). In Duras’s case, one might doubt whether she was even interested in organizing anything. More likely, she wanted to shake up her literary-cinematographical kaleidoscope, wanting to mix the same story elements, characters, and obsessions to new visions. CONCLUSION While objecting to Herman’s assimilating tendencies, Nielsen stresses the rhetorical consequences of returning narrative agency to the author—if we assume that narratives are fictional inventions fabulated by the author, we need not follow the regularities of the real world: “If we interpret a narrative as fiction, we interpret it as creating (aspects of) a fictional world.” Keeping the rhetorical interests and purposes of narratives in mind, Nielsen does not want to move toward Abbottian “incomprehensiveness, mysteriousness, or noncommunication” (Nielsen 2013, 86–87). I hope to have shown the rhetorical dimensions of Durassian ­“private unnarratable space” and ways of defending it, as their ostensible “noncommunication” can be a significant poetic asset. Their strength is in their a­ bility to flee from normal procedures, ordinary and predictable. Immersing himself in the story of a mad beggar woman may, for Peter Morgan, be an attempt to toy with the same bestial and disquieting freedom the Vice-Consul represents. Still, we never get to learn the “real” experiences and emotions of Peter Morgan’s muse, as no one among the white society understands her language. Neither is Le ravissement de Lol V. Stein a typical case history, as it foregrounds the contradictory act of reproduction instead of allowing free access to Lol’s mind. As Shira Wolosky describes Samuel Beckett’s art, “the language of selfnegation finally fails to silence itself, and in so doing proves fecund”; the ineffability of language gives way to “reproductive and inventive energy” (Wolosky 1995, 132). The artistic effect of Duras’s writings and films is to maintain the tension: when there is no closure and when the secrets do not get unveiled explicitly, the same elements, characters, and literary-cinematic topoi in all their imperfection may continue their free floating from work to work, toward other virtual worlds. Jacques Hold’s artful fantasies and the passionate partiality of India Song’s acousmêtres become the poetic surplus of their own futility, just as the ruins of a palace may be even more fascinating than the palace itself in its days of splendour. NOTES 1. The “unnatural” trend, which can be traced back to Brian Richardson’s study Unnatural Voices (2006) has later spawned manifestos and anthologies, while it

Defending the Private and the Unnarratable  165 is not a uniform, strict methodology, but rather a reading strategy (see Nielsen 2013). 2. This statement has been made by Galen Strawson in his 2004 outburst “Against Narrativity,” but whereas Strawson touches upon questions of how to ­construct—or not—a personal identity and narrative self-comprehension, my take here is more concerned with the ethos and the limits of the poetic creation. 3. For the connection between the post-war doubt toward the narrative and the development of French literature from Robbe-Grillet’s Dans le labyrinthe (1959) to Tournier’s Le roi des aulnes (1970), see Meretoja 2014. 4. For recent applications and critique of folk psychology in narrative studies, see Herman 2011 and Iversen 2013. 5. Richard Walsh, particularly, is involved in summoning the author back to life in his study The Rhetoric of Fictionality (2007).

REFERENCES RLS = Duras, Marguerite. [1964] 1986. The Ravishing of Lol Stein, translated by Richard Seaver. New York: Pantheon Books. RLVS = Duras, Marguerite. [1964] 2004. Le ravissement de Lol V. Stein. Paris: Gallimard,. LVC = Duras, Marguerite. [1966] 2006. Le Vice-consul. Paris: Gallimard. TVC = Duras, Marguerite. The Vice-Consul. [1966] 1987, translated by Eileen ­Ellenborgen. New York: Pantheon Books. Abbott, Porter H. 2008. “Unreadable Minds and the Captive Reader.” Style 42.4: 448–67. Bordwell, David. 2008. Poetics of Cinema. New York and London: Routledge. Chion, Michel. [1990] 1994. Audio-Vision: Sound on Screen, edited and translated by Claudia Gorbman. New York: Columbia University Press. Cohn, Dorrit. [1978] 1983. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton, NJ: Princeton University Press. Culler, Jonathan. [1975] 1986. Structuralist Poetics: Structuralism Linguistics and the Study of Literature. London: Routledge. Deleuze, Gilles. 1985. Cinéma 2. L’Image-Temps. Paris: Les éditions de minuit. Duras, Marguerite. 1979. “India Song. Son nom de Venise dans Calcutta désert. Découpage intégral et texte in extenso par M.C. Ropars-Wuilleumier.” L’Avant-scène 225: 14–65. Gallagher, Shaun. 2008. “Inference or Interaction  : Social Cognition without ­Precursors.” Philosophical Explorations 11.3: 163–74. Genette, Gérard. 1972. Figures III. Paris: Editions du Seuil. Heinze, Rüdiger. 2008. “Violations of Mimetic Epistemology in First-Person ­Narrative Fiction.” Narrative 16.3: 278–98. Herman, David. 2011. “Introduction.” In The Emergence of Mind: Representations of Consciousness in Narrative Discourse in English, edited by David Herman, 1–40. Lincoln, NE: University of Nebraska Press. Hyvärinen, Matti, Lars-Christer Hydén, Maria Tamboukou, and Marja ­Saarenheimo. 2010. “Beyond Narrative Coherence: An Introduction.” In Beyond Narrative Coherence, edited by Matti Hyvärinen, Lars-Christer Hydén, Maria Tamboukou, and Marja Saarenheimo, 1–15. Amsterdam: John Benjamins Publishing Company.

166  Tytti Rantanen Iversen, Stefan. 2013. “Unnatural Minds.” In A Poetics of Unnatural Narrative, edited by Jan Alber, Henrik Skov Nielsen, and Brian Richardson, 94–112. ­Columbus: The Ohio State University Press. Kafalenos, Emma. 1999. “Not (Yet) Knowing. Epistemological Effects of Deferred and Suppressed Information in Narrative.” In Narratologies: New Perspectives in Narrative Analysis, edited by David Herman, 33–65. Columbus: The Ohio State University Press. Kawin, Bruce F. 1978. Mindscreen: Bergman, Godard, and First-Person Film. ­Princeton, NJ: Princeton University Press. Knuuttila, Sirkka. 2011. Fictionalising Trauma. The Aesthetics of Marguerite Duras’s India Cycle. Frankfurt am Main: Peter Lang. Landa, José Ángel García. 2011. “Narrating Narrating: Twisting the Twice-Told Tale.” In Theorizing Narrativity, edited by John Pier and José Ángel García Landa, 419–52. München: Walter de Gruyter. Mäkelä, Maria. 2006. “Possible Minds. Constructing-and Reading–Another ­Consciousness as Fiction.” In FREElanguage INDIRECTtranslation DISCOURSE­ narratology: Linguistic, Translatological and Literary-Theoretical Encounters. Tampere Studies in Language, Translation and Culture, Series A, Vol. 2, edited by Pekka Tammi and Hannu Tommola, 31–60. Tampere: Tampere University Press. ———. 2013. “Cycles of Narrative Necessity: Suspect Tellers and the ­Textuality of ­Fictional Minds.” In Stories and Minds: Cognitive Approaches to ­Literary N ­ arrative, edited by Lars Bernaerts, Dirk De Geest, Luc Herman, and Bart ­Vervaeck, 129–51. Lincoln and London: University of Nebraska Press. Meretoja, Hanna. 2014. The Narrative Turn in Fiction and Theory: The Crisis and Return of Storytelling from Robbe-Grillet to Tournier. Basingstoke: Palgrave Macmillan. Nielsen, Henrik Skov. 2004. “The Impersonal Voice in First-Person Narrative F ­ iction.” Narrative 12.2: 133–50. ———. 2013. “Naturalizing and Unnaturalizing Reading Strategies. Focalizations Revisited.” In A Poetics of Unnatural Narrative, edited by Jan Alber, Henrik Skov Nielsen, and Brian Richardson, 67–93. Columbus: The Ohio State University Press. Richardson, Brian. 2006. Unnatural Voices: Extreme Narration in Modern and ­ ontemporary Fiction. Columbus: The Ohio State University Press. C Robbe-Grillet, Alain. 1965. “From Realism to Reality.” (Du réalisme à la réalité, 1955 & 1963.) In For a New Novel : Essays on Fiction (Pour un nouveau roman, 1963.), translated by Richard Howard, 157–68. New York: Grove Press Inc. Paris: Gallimard. Sarraute, Nathalie. 1990. “From Dostoievski to Kafka.” (“De Dostoïevski à Kafka,” 1947.) In The Age of Suspicion: Essays on the Novel (L’ère du soupçon. Essais sur le roman, 1956), translated by Maria Jolas, 9–50. New York: George Brazillier. Strawson, Galen. 2004. “Against Narrativity.” Ratio 17.4: 428–52. Walsh, Richard. 2007. The Rhetoric of Fictionality. Columbus: The Ohio State U ­ niversity Press. Wolosky, Shira. 1995. Language Mysticism: The Negative Way of Language in Eliot, Beckett, and Celan. Stanford: Stanford University Press.

9 Of Minds and Monsters The Eventfulness of Monstrosity and the Poetics of Immersion in Horror Literature Gero Brümmer

The monster is often held to be the central element and its discovery the plot-motivating goal of horror literature (cf. Carroll 1990, 99). However, the monster rarely appears alone. Its appearance is anticipated, reacted to, and eventually resisted by the fictional characters. The audience often takes on the role of observer in this encounter. From a narratological perspective, horror literature thus revolves not only around the monster but also around the event of encountering the monster. It is this encounter with the monster that is then ultimately meant to lead to an emotional reaction in the audience, which is considered one of the goals of horror literature (cf. Carroll 1990, 14; King 2010, 25–26; Lovecraft 2000, 23). It is during the encounter with the monster, particularly in the moment of its presentation, that horror literature can fail by trying to achieve two goals that, while not exactly at odds with each other, work on different levels: (1) the presentation of an interesting and convincing monster, judged primarily by the standards of the fictional ontology, and (2) provoking an emotional reaction in the reader, situated outside the world of the fictional text. Horror literature’s reliance on the monster as an object that exists on the level of the storyworld but is inevitably judged in terms of the reader’s own world makes it a particularly fruitful subject for the research on eventfulness, narratability, and immersion. The repeated presentation of affective reactions by fictional characters evaluates the encounter as eventful and simultaneously raises readers’ expectations, while the immersive nature of the focus on the characters and their consciousness has the potential to create a false sense of equivalency between the impact the encounter had in the storyworld and the impact its eventual presentation will have on the reader. In this paper, I will look at how the dynamics of immersion allow readers to make sense of the events in the storyworld and their impact on characters on both a conceptual and an emotional level. One goal of this analysis is to show that the construction of eventfulness revolves around the entities within the storyworld involved in the event, in this case the monster and the characters. In addition, I want to highlight how certain strategies of narrative distancing, which at first glance may appear to hinder immersion, can actually be immersive on a different level, thus making it necessary to closely look at the conceptual and emotional components of immersion.

168  Gero Brümmer I will first look at the connection between the concepts of eventfulness, narratability, and immersion and their significance for the readers’ emotional response to the events of narrative fiction. Afterward I will give a brief introduction to horror fiction, focusing on the presentation of the monster in particular. As the challenges faced in the presentation of the monster are the same challenges any narrative faces when it comes to emotionally affecting the audience, it will be useful to show how horror literature attempts to deal with overcoming the gap between what the audience expects from the monster and what the text can possibly deliver. In doing so, I want to focus on the form this presentation takes, in which I will distinguish between evaluation and description as two different ways to provide readers with access to the monster. Building on this distinction, I will show that horror literature uses various distancing strategies in order to suggest rather than to present the monster and to avoid the danger of presenting a monster that readers cannot relate to. The use of these distancing strategies also serves another function, namely the fostering of immersivity in two different forms, which I refer to as conceptual and emotional immersion. I will close with a reading of Arthur Machen’s novella The Great God Pan and the use of distancing strategies within that text. The purpose of that analysis is to show how different distancing strategies serve to foreground the fictional mind of those who encounter the monster rather than the monster itself. EVENTFULNESS, NARRATABILITY, AND IMMERSION Narrative analysis often foregrounds the importance of events as a defining feature of narratives. Any change of state (and any action) in a narrative can be considered an event, but some theorists suggest the distinction between different types of event (cf. Hühn 2011; Schmid 2003). My analysis in this chapter will only concern itself with those events, sometimes referred to as type II events, which require additional evaluative components beyond a simple change of state. Type II events are characterized by their relevance, unpredictability, persistence, irreversibility, and non-iterativity (Schmid 2003, 26–29). These events can also be generally described as “breach[es]” of “canonicity” (Bruner 1991, 11) or deviations from the “normal, expected course of things” (Hühn 2011, Paragraph 2). A change of state does not need to exhibit all of these features in order to be considered an event. Instead, the features that make an event eventful are “gradational and can be realized to varying degrees” (Schmid 2003, 24), which means that, depending on the distribution of these features, an event can be considered more or less eventful. As these events need to deviate from what is considered canonical or normal, their eventfulness is tied to the degree of this deviation as well (cf. Hühn 2011, Paragraph 34). An event that appears, even in comparison to other events, to be an exceptional instance of deviation would also be considered more eventful.

Of Minds and Monsters  169 As is implied by the use of terms like deviation, normality, and canonicity, eventfulness is considered to be context-sensitive (Hühn 2011, Paragraph 25; Hühn 2008, 143–44). For narrative fiction, the features introduced by Schmid concerning the relevance or predictability of an event, for example, are first of all viewed in the context of the fictional world. An event can be extremely relevant for one character but insignificant for another. More importantly, this discrepancy can be expanded to the reader as well and thus necessitates the inclusion or at least the consideration of additional contexts, both within and outside the text (Hühn 2008, 143–44). As Hühn explains, “[w]hat for a hero is an unpredictable event can for the reader be a central part of a genre’s script” (Hühn 2011, Paragraph 28) and what in the context of the story can be a breach of canonicity can be par for the course for the reader, thus leading to a discrepancy in what either party considers eventful (cf. Hühn 2008, 148). As one consequence of this context-sensitivity, eventfulness is often discussed in connection with the “tellability” (sometimes also ­“narratability”) (Prince 2008) of a narrative. Tellability concerns the question of what makes a story worth telling by asking what the “point” of a given narrative is. ­Tellability is not solely determined by narrative events, of course, but a ­story’s eventfulness can be a factor in determining whether it is considered to have a point (cf. Hühn 2011, Paragraph 5). The importance of an event in a text can be foregrounded through “evaluative devices.” These include the reflections of characters, the repetition of events, and the “use of disnarrated events—explicitly referring to what did not take place but could have,” in order to underscore, foreground, or emphasize certain aspects of the n ­ arrative (Prince 2008, 23–24). Just like eventfulness, narratability is considered to be context-sensitive (cf. Prince 2008, 23–25). Whether an audience considers the breach of canonicity in the narrative eventful may also influence whether the audience considers the story itself worth telling. But due to its context-sensitivity, this evaluation may not translate to another context and the events foregrounded in a text as eventful may not be considered tellable by their audience (cf. Hühn 2008, 148). As Prince points out, “claiming that (sequences of) events are unusual, extraordinary, bizarre, unfortunately does not suffice to make them so” (Prince 2008, 24). One approach to mitigating the gap between the evaluative framework of the fictional world’s ontology and the readers’ own ontology is the reinforcement of those qualities of a text that foster the audience’s immersion in a text. Immersion (Ryan 2003), or transportation (Gerrig 1993; Hogan 2013) refers to the ability of a reader to become “lost in a book” ­(Gerrig 1993, 3). Readers are transported to a narrative world as part of the process of reading a text. They leave their own world behind and “adapt willingly to local conditions,” as Gerrig calls them (Gerrig 1993, 9). This adaptation functions by way of a process that Ryan calls “recentering,” in which ­“consciousness relocates itself to another world” and in which the virtual

170  Gero Brümmer world of the fiction becomes the actual world for the reader (Ryan 2003, 103). Both metaphors describe the bridging of a distance between the reader’s world of origin and the fictional world of the text. This bridging of the distance is made easier by a process of projection that revolves around what is called the “principal of minimal departure”: “[W]e will project upon the world of the statement everything we know about the real world,” only making “those adjustments which we cannot avoid” and only when texts actively discourage this projection (Ryan 1980, 406; Gerrig 1993, 13). The concept of transportation is based on the process of mental simulation as established by cognitive research (cf. Hogan 2013, 56; Ryan 2003, 110). Simulation as it pertains to literary texts and literary transportation involves a “hypothetical imagination” stimulated by the “words of the storyteller” (Hogan 2013, 56–58). Through the interaction of the text and the imagination of the readers, guided by the principal of minimal departure, the story world that readers are transported to is based on a combination of the text and their own ideas, experiences, and schemata that they project into the world on the basis of the “presumed identity” of the storyworld and their own world (Hogan 2013, 62–63). Simulation thus needs to be considered in connection with the assumed continuity between what Hogan calls a “presupposed” world, a world shared by authors and readers, and the world of the text. There is an ongoing interaction between the two, and it is through this interaction that fiction can affect the readers’ sense of the presupposed world, for example by calling into question the norms and values that dominate that world (Hogan 2013, 63). A successful immersion or transportation thus appears to be a good way of overcoming the context-sensitivity of eventfulness. If readers are successfully transported into a text and recenter themselves with regard to the norms and values of that world, the significance of the events in the storyworld should be clear to them, and they should be more likely to agree with their eventfulness. Conversely, the evaluative devices that foreground eventfulness can aid in the process of immersion, for example, through the generation of a general interest in subsequent events in the narrative. ­Furthermore, since the evaluation of the significance of an event depends on the “consciousness of a perceiving subject” (Hühn 2008, 148), the presentation and evaluation of an event can bring readers closer to the characters in the story. The evaluation of eventfulness can also help to explain the world of the text to readers, thus making it easier to accomplish a recentering with regards to that world. By highlighting and explaining how a currently ongoing event in the story is a breach of canonicity and a deviation from the norm, the text can provide information on what is normal and canonical within the storyworld, thus potentially bridging the distance between the readers’ ontology and the fictional ontology. With these preliminary considerations out of the way, it is now time to take a brief look at the genre of horror literature and the status of events within the genre.

Of Minds and Monsters  171 COMMUNICATING THE EVENTFULNESS OF MONSTROSITY IN HORROR LITERATURE: BETWEEN DESCRIPTION AND EVALUATION There are different ways to define horror literature, but for the purpose of this paper I will limit my analysis to those examples of horror literature that present a monster. In many regards, I will build on what Noël Carroll has called an “entity-based” definition of horror, which “involves essential reference to an entity, a monster, which then serves as the particular object of the emotion of art-horror” (Carroll 1990, 41). It should be noted that in his work, Carroll makes a distinction between entity- and event-based theories of horror. While this might at first seem like it would lead to problems when it comes to discussing the eventfulness of horror, this distinction does not mean that Carroll does not take events into account at all, since he considers the discovery and the presentation of the monster a central element of multiple prototypical horror plots (cf. Carroll 1990, 99). The encounter with the monster becomes an important event in horror literature. In this section, I will look at the entities involved in this event, its structure, and its relationship to the emotional reactions of the audience. In doing so, I will adopt many of the elements that Carroll outlines in his book The Philosophy of Horror, since it provides a very concise explanation of how the encounter between the monster and the characters of the fictional world is structured. However, while Carroll uses these structures to provide an explanation for the emotional responses of the audience to horror fiction (and fiction in general), I will focus more on the potentially immersive qualities of these structures. First, a word on monsters. Monsters, in a general sense, are creatures that are abnormal or unnatural. Carroll bases part of his explanation of monstrosity on the concept of impurity outlined in Mary Douglas’ study Purity and Danger. Impurity is an evaluative attribution that derives from an object’s deviation from what is considered normal or natural in a given ontology. Many (nonfictional) monsters are hybrid in nature and are the result of what Carroll calls “category mistakes” (Carroll 1990, 31) or what Douglas describes as “confusion[s]” (Douglas 1966, 53). They are the result of a “breaking of that which should be joined or joining of that which should be separate” (Douglas 1966, 113). In general, monstrous beings are those that “are imperfect members of their class, or whose class itself confounds the general scheme of the world” (Douglas 1966, 55). As a result, monsters are viewed as “abnormal, as disturbances of the natural order” (Carroll 1990, 16). Beyond being abnormal, disgusting, and threatening, the very existence of monsters calls into question the legitimacy or at least the stability of prevailing notions of normality and canonicity. This destabilizing function of the monster in its historical and socio-cultural contexts has given rise to an independent field of monster studies (for examples, see Cohen 1996b;

172  Gero Brümmer Geisenhanslüke and Mein 2009). Focusing on the danger the monster poses due to its situatedness at the borders between different categories, many studies highlight the cataclysmic change the monster’s existence causes in the order of a given system. As Cohen states, “[t]he monster is difference made flesh, come to dwell among us” (Cohen 1996a, 7). This positioning of the monster in multiple contexts allows for considerable versatility in the creation of monsters. This versatility is reflected in the works of many horror authors. For instance, research on H.P. Lovecraft has highlighted many of the instances in which Lovecraft uses monsters to destabilize different ideas of canonicity by drawing on the schemata and conventions of, just to name a few examples, evolutionary biology (Burleson 1990, 134–46 ­passim), mathematics (Hull 2006) and sexuality (Wisker 2013). As Cohen points out, the monster can enter any of these systems, and when it does it is quite difficult to remove it (cf. Cohen 1996a, 4–5). In a way, monsters thus become the embodiment of an event, or events made flesh, in that they carry many of the features of eventfulness. I will only consider some of them here: their status as things-that-should-not-exist certainly points to their relevance, their existence is unpredictable because they defy the order of things on which that predictability was based, and, most important for horror literature, they are certainly persistent, at least for the characters who meet them, who often undergo dramatic changes during and after their encounter with the monster. In connection to the latter feature of monsters, it is important to explain the function of character reactions in horror fiction. Character reactions play a significant role in evaluating the eventfulness of encounters with monsters and the monstrous (cf. Carroll 1990, 18). Carroll points out that there is often a “theme of visceral revulsion” that we are witness to when we read horror literature (Carroll 1990, 19). He mentions Jonathan Harker’s ­shudder at the touch of Dracula in the novel of the same name as one of many examples of this type of affective response (Carroll 1990, 17). For Carroll, these emotional character responses are an important part of horror fiction. They evaluate the creature by providing a set of instructions as to which response to the creature is appropriate (cf. Carroll 1990, 18). ­Furthermore, the audience’s responses are “meant, ideally, to parallel those of characters. Our responses are supposed to meet (but not exactly duplicate) those of the characters; like the characters we assess the monster as a horrifying sort of being (though unlike the characters, we do not believe in its existence)” (Carroll 1990, 18). The presentation of the monster in horror literature relies on two different modes: evaluation and description. I will use a passage from M.R. James’ short story “Canon Alberic’s Scrap-Book” to illustrate this point: At first you saw only a mass of coarse, matted black hair; presently it was seen that this covered a body of fearful thinness, almost a ­skeleton, but with the muscles standing out like wires. The hands were of a

Of Minds and Monsters  173 dusky pallor, covered, like the body, with long, coarse hairs, and hideously taloned. [...] Imagine one of the awful bird-catching spiders of South America translated into human form, and endowed with intelligence just less than human, and you will have some faint conception of the terror inspired by this appalling effigy. (James [1904] 1984, 8 emphasis added) The creature the narrator sees is described as some form of hybrid creature, a mixture of the shapes of a human being and a spider. Its hybrid nature confounds the distinction between animals and humans. It is also described as a beast, but it has “intelligence just less than human,” in contrast to its otherwise beast-like features. This description, in addition to the monster’s talons and its coarse, matted hair, can then lead to an evaluation of the creature as both threatening and disgusting, the two defining features of monsters that Carroll proposes in his approach (cf. Carroll 1990, 27). What is more important for the present analysis, however, is that in addition to the description of its features, the presentation uses the language of emotions to evaluate the effect the creature has on the observer: it is “hideous,” ­“appalling,” and inspires “fear” and “terror.” IMMERSION AND DISTANCING IN THE PRESENTATION OF THE MONSTER By engaging with the monster through the consciousness of characters in the story, readers also become invested and immersed in the storyworld. Through the evaluation of the encounter with the monster as eventful, readers learn the significance of this encounter, while through the description of those features that are evaluated as deviations or breaches they learn about the norms and values of the fictional world. More than that, the description of the monster can provide readers with detailed information that can aid in their recentering in that world. Writing on the topic of the stranger from the perspective of translation studies, Horst Turk suggests that when it comes to categorization the distinction is made between two types of “belonging”: the alien and the alter. The alien (alienus), is that which “belongs to another” (“dem Fremden zugehörige”) while the alter describes the “other of two”—one that does not belong to another but is not the “same” either (Turk 1990, 10–11). Both terms can essentially be used to describe the same being: the alien clarifies that it does not belong to one’s own group while the alter positions that being in relation to another member of the other group (Turk 1990, 11). Alterity and alienity pose different challenges to a person’s ability to categorize and analyze others, calling their own categories into question by showing their limitations. The problem is only exacerbated when they then try to communicate their evaluation to another person, since in the translation of

174  Gero Brümmer their experience into language they are limited by what Heinke calls the collective linguistic and cultural experience of their “reference system” (Heinke 2005, 98), or the notions of normativity, the schemata and scripts they are familiar with. It is here that the context-sensitivity of the monster proves useful for immersion. If the fictional world is considered as its own self-contained world, then the monster is an object within this world (cf. Ryan 2003, 103–105). At the same time, part of the monster itself remains virtual in the sense that the context-sensitivity of monstrosity means that there are many possible ways the monster will deviate from the norm, all of which remain open as long as the text does not imbue the monster with specific features. As soon as the text gives the monster a feature that excludes other possible ­features, this feature is actualized and the other previously possible features are discarded. In H.P. Lovecraft’s “The Dunwich Horror,” the text positions the monstrous creature as having features of both the alien and the alter: “The limbs, save for their black fur, roughly resembled the hind legs of prehistoric earth’s giant saurians, and terminated in ridgy-veined pads that were ­neither hooves nor claws” (Lovecraft [1929] 2008, 279 emphasis added). The legs are described as (“roughly”) resembling those of dinosaurs, while the narrator fails to find an adequate way of explaining the appearance of the creature’s feet. Too alien in nature, the only recourse is to draw analogies via negation, to state what the feet do not look like. By not committing to a clear description of the monster and leaving certain features in the realm of the virtual, the exact deviation of the monster thus remains open to the interpretation of the reader. The encounter between characters and monsters thus becomes the most significant immersive event of horror literature. More than that, however, it gains additional significance because it can also be the moment in which the audience encounters the monster as well. Consequently, it is during the main event of horror literature, the presentation of the monster, that the risk inherent in the structure of horror literature becomes clear. The text can present the encounter with the monster as an event, foreground and emphasize its eventfulness, and can make the reader understand why meeting the monster would be considered eventful. However, because of the context-sensitivity of both eventfulness and monstrosity, there is no guarantee that the reader will be affected in a manner similar to that of the characters. The discrepancy between how readers and characters evaluate the same event can make it necessary to create a certain distance between the evaluation and the description. This distance can take different forms. One possibility is to create a temporal distance between the evaluation of the monster and its description in the structure of the narrative text. This is a common occurrence in horror literature, where the presentation of the monster is often preceded by a focus on character reactions as a form of evaluation.

Of Minds and Monsters  175 An example of this can be seen in “Canon Alberic’s Scrap-Book,” where the paragraphs that precede the description of the monster provide a taste of the eventfulness of what is yet to come: On the pavement before the throne were grouped four soldiers, surrounding a crouching figure which must be described in a moment. A fifth soldier lay dead on the pavement, his neck distorted, and his eyeballs starting from his head. The four surrounding guards were looking at the King. In their faces, the sentiment of horror was intensified; they seemed, in fact, only restrained from flight by their implicit trust in their master. All this terror was plainly excited by the being that crouched in their midst. I entirely despair of conveying by any words the impression which this figure makes upon anyone who looks at it. I recollect once showing the photograph of the drawing to a lecturer on morphology—a person of, I was going to say, abnormally sane and unimaginative habits of mind. He absolutely refused to be alone for the rest of that evening, and he told me afterwards that for many nights he had not dared to put out his light before going to sleep. However, the main traits of the figure I can at least indicate. (James [1904] 1984, 7–8) “Canon Alberic’s Scrap-Book” thus provides the promise of a deeply emotional event within the storyworld, while also setting up the expectations of the audience. The audience knows of the danger posed by the creature through the mention of the dead soldier, who presumably fell victim to the creature. The characters in this scene exhibit sentiments of horror and terror, while a more remote witness, one who only saw the photograph of the drawing of this scene, “absolutely refused to be alone for the rest of the evening.” This witness is also provided with some characteristics that let readers gain an insight into his mind and lets them understand part of the reference system of that character, who is abnormally sane, unimaginative, and—as a lecturer at a university—presumably also rational. The description here is more concerned with the experiencing subject than with the monster, of which there is no description yet. It is only referred to as the “figure” and the “being,” giving no indication of its form or its features. The passage from James’s text contains another distancing strategy, when the narrator points out that he cannot convey “by any words the impression which this figure makes upon anyone who looks at it”—a very clear evaluation of its eventfulness—and can only “indicate” the “main traits” of the figure, thus distancing himself from the following description. ­Addressing the inability to provide a reason for why an extreme emotional reaction takes place and the related impossibility of accurately describing monstrous features is a common distancing strategy, often related to a limitation of language as a medium or the impossibility of communicating

176  Gero Brümmer intense experiences in general (see also Kakko, this volume, for an analysis of hallucinatory narratives). This notion of ineffability is often invoked by H.P. Lovecraft as well. In “The Dunwich Horror,” for example, the detailed presentation of a monster is prefaced by the statement that, while it “would be trite and not wholly accurate to say that no human pen could describe it,” it is still an experience impossible to effectively convey to another human being (Lovecraft [1929] 2008, 279). Another strategy is to posit a possible source for an emotional reaction without committing to it or presenting it clearly. One such instance occurs in Dracula, in a passage that Carroll also cites as an example of how character reactions structure the readers’ own emotional responses: As the Count leaned over me and his hands touched me, I could not repress a shudder. It may have been that his breath was rank, but a horrible feeling of nausea came over me, which, do what I would, I could not conceal. (Stoker [1897] 2003, 25) In this passage, Harker cannot accurately describe any feature that caused his reaction and instead can only speculate that “[i]t may have been that his breath was rank” (emphasis added) and that this led to his irrepressible reaction (Stoker [1897] 2003, 25). Similarly, in Robert Louis Stevenson’s The Strange Case of Dr. Jekyll and Mr. Hyde, one of the characters tries to describe Mr. Hyde, but is unable to give a clear description, instead repeatedly referring to his emotional reaction as the only evidence for monstrosity: “I never saw a man I so disliked, and yet I scarce know why. He must be deformed somewhere; he gives a strong feeling of deformity, although I couldn’t specify the point” (Stevenson [1886] 2006, 7–8). Hyde is evaluated as ugly, as being repulsive, and as giving a “feeling” of deformity, but there is no description of deformity, not even clarity on whether Hyde is actually deformed in any way. Consequently, readers are not given enough information to decide if they agree with this assessment. CONCEPTUAL AND EMOTIONAL IMMERSION This distortion of the readers’ access to the storyworld can best be described with Werner Wolf’s explanation of the level of mediation as a “pane of milk glass” with varying degrees of opacity (Wolf 1993, 374). When readers are denied access to the story, they are likely to focus on the mediator instead. What functions as a strategy useful in delaying the presentation also serves another very specific function in horror literature: providing an on-going evaluation and a form of immersion that is not interrupted (and consequently undermined) by presentations of discrepant features. In horror literature, recentering is necessary in order to understand why the monster

Of Minds and Monsters  177 is considered horrifying. Through descriptions and evaluations, readers can make sense of the monster in the fictional ontology conceptually. ­Conversely, by identifying and understanding the monster, readers can make inferences about the norms and nature of the fictional cosmos (or at least about an individual character’s understanding of the same). I would call this type of focus conceptual recentering or conceptual immersion, as it allows readers to better understand the characters and their individual reference systems. However, horror literature aims to do more than just explain why the monster is meant to be horrifying. It also needs to evoke an emotional reaction in the reader. The affective reactions that fiction can cause, intends to cause, or may fail to cause have given rise to many different approaches to answering the question of why audiences are frightened (or otherwise affected) by fictional events. One explanation, as outlined above with regards to eventfulness, lies in simulation theory and transportation (Hogan 2013; Gerrig 1993). These theories then give rise to the question of how far this simulation can go and how “lost” audiences can become in a fictional world, whether there are limits to their ability to recenter themselves or whether it is possible to go so far as to call what fiction creates a form of “illusion” (cf. Wolf 2011, ­Paragraph 1). Horror literature tries to close the gap by immersing readers in the world and allowing them to recenter themselves with regard to the fictional ontology, slowly appropriating its concepts of normality and monstrosity. By letting readers make inferences about the monster based on character reactions, the text does not need to actualize any of the monster’s features. By shifting the focus from the monster to the experiencing subjects in the story, the text uses the language of emotions as a means to suggest a more immediate access to the characters. By reading the affective reaction of the characters without a description of the monster, readers are encouraged to draw parallels between the emotions of the fictional characters and their own, allowing for what I would call emotional immersion or emotional recentering. In contrast to conceptual immersion, this aspect of immersion focuses more on the evaluative aspects of mediation. When audiences read the emotive responses of the fictional characters they serve as a “set of instructions” (Carroll 1990, 17) for them and lead them to assume that whatever the characters are confronting would lead to a similar ­ arroll 1990, 17; Hogan 2011, 3). More than reaction of their own (cf. C that, ­Carroll claims that readers assume a direct correlation between the emotive responses and the features of the monster: “we regard the description or depiction of them as unsettling virtue of the same kind of qualities that revolt someone like Jonathan Harker” (Carroll 1990, 17). Reading Harker thus allows the audience to read Dracula. Through concepts like the principle of minimal departure and the presumed identity between fictional ontology and extra-fictional ontology, emotional immersion fosters conceptual immersion as well. By focusing on the evaluative aspect of monstrosity, however, readers are left to supply the descriptive elements themselves,

178  Gero Brümmer which is very advantageous for horror literature. They are free to project their own ideas of monstrosity, their own schemata and experiences, so long as they are not actively discouraged or countered by the text. This is aided by the focus on the emotional reactions of the characters, which are themselves often left open to interpretation, since, as researcher like Patrick Colm Hogan have pointed out, the expressions of emotions themselves are context-sensitive and based on prototypes (Hogan 2011, 4). That is why another effective distancing strategy is “disnarration” (cf. Prince 1988; 2008), which can be the refusal or inability of characters to narrate an event at all. Here, readers are often given no description of the event and can only gain an insight into what happened through their attempts at reading “unreadable minds” (cf. Abbott 2008). These minds may then exhibit signs of overwhelming emotional experiences, often in the form of insanity, which in turn hints at the extremeness of the event, the horror of the monster, etc. The more unreadable a mind becomes as a result of the encounter with the monster, the more readers want to gain insight into this mind, since it could potentially hold the key to the most effective presentation of monstrosity imaginable (for a more critical look at the ethics involved with the sometimes almost obsessive need to take part in another’s experience, see the essay by Rantanen in this volume). In the following section I will focus on the structures of immersion and the strategies of distancing employed by Arthur Machen’s 1894 novella The Great God Pan in order to show how horror literature can position the encounter with the monster as a highly immersive event without always needing to commit to specific features of this monstrosity. Toward the end of that reading I would also like to suggest alterization and alienization as strategies that the text can use to distance itself from statements already made in the narrative, creating a sort of ambiguous distance between the world of the text and the world of the reader. THE POETICS OF IMMERSION AS STRATEGIES OF NARRATIVE DISTANCING The plot of The Great God Pan focuses on a woman named Helen Vaughan and the effect her actions have on those who encounter her. It is slowly revealed that Helen is the offspring of a human woman and the god Pan, who managed to gain entrance into this world through an experiment originally intended to break through the limits of human perception. Touching on issues of sexuality and the nature of reality, the text is presented as a series of narrative accounts by different characters in the story who either had direct contact with Helen or the god Pan or who instead recount the stories of those who have been unfortunate enough to interact with either one of these entities directly. What makes The Great God Pan interesting for this analysis is that it contains two monstrous creatures, Pan and Helen, with very different presentations.

Of Minds and Monsters  179 Throughout the story, there are very few descriptions of Helen and even fewer of Pan, leaving readers to gather information about them in a more mediated form, usually by looking at the carnage their actions have wrought. The majority of what readers learn about Helen’s actions is mediated through the evaluation of those who encountered her, which themselves are also given in vague terms. At one point, a character’s own uncertainty is posited as the reason for his inability to describe Helen: “[h]e positively shuddered as he tried to describe the woman, but couldn’t tell why” (Machen [1894] 2006, 31). Of another of Helen’s victims, readers learn from a physician that he “died of fright, of sheer, awful terror; I never saw features so hideously contorted in the entire course of my practice, and I have seen the faces of a whole host of dead” (Machen [1894] 2006, 30). The more intense the reaction, the more powerful the evaluation, and the less descriptive the information about Helen which is made available to the reader, instead positioning another mediating instance (the people who describe the victims) between the reader and Helen. By doing this, the text stresses the importance of the encounter with the monsters over the monsters themselves. By positioning different narrators with different reference systems as mediators, The Great God Pan manages to simultaneously draw on different sources of horror, the two major ones being sexuality and myth, or phrased with regard to the boundaries being crossed, the border between what the text indicates is normal and deviant sexuality and that between the natural and the supernatural. Furthermore, these transgressions are themselves caused by another crossing of boundaries, that between science and nature. The story starts with an experiment in which Dr. Raymond operates on a young girl, Mary, in order to allow her to see the god Pan, thus opening a door to Pan’s domain. Mr. Clarke, his friend, is a witness to this experiment and one of the main narrators and reference points for the audience. At one point, Clarke attempts to put the consequence of Pan’s existence into words: It is too incredible, too monstrous; such things can never be in this quiet world, where men and women live and die, and struggle, and conquer, or maybe fail, and fall down under sorrow, and grieve and suffer strange fortunes for many a year; but not this, Phillips, not such things as this. There must be some explanation, some way out of the terror. Why, man, if such a case were possible, our earth would be a nightmare. (Machen [1894] 2006, 23) This statement is both lucid and to the point and gives a good explanation for the impact the existence of the monsters has on this world, but it is quite removed from Clarke’s more visceral reactions to the experiment he witnessed. As the experiment proceeds, Dr. Raymond starts to make surgical incisions on the girl’s (Mary’s) body. Clarke, unable to watch the proceedings, turns away “shudderingly” and feels “sick and faint” (Machen [1894]

180  Gero Brümmer 2006, 16). Despite the reader’s awareness of Clarke’s internal state, there is nevertheless a distance between the reader and Clarke since it is not made explicitly clear whether it is the physical reality of the medical procedure that affects Clarke or the ethical deliberations on the nature of the experiment. If the reader feels a certain distance from Clarke, the distance to Mary and her experiences is even greater. Sometime after the procedure is finished, Mary’s demeanor changes significantly: Suddenly, as they watched, they heard a long-drawn sigh, and suddenly did the colour that had vanished return to the girl’s cheeks, and suddenly her eyes opened. Clarke quailed before them. They shone with an awful light, looking far away, and a great wonder fell upon her face, and her hands stretched out as if to touch what was invisible; but in an instant the wonder faded, and gave place to the most awful terror. The muscles of her face were hideously convulsed, she shook from head to foot; the soul seemed struggling and shuddering within the house of flesh. It was a horrible sight, and Clarke rushed forward, as she fell shrieking to the floor. (Machen [1894] 2006, 16) By not giving readers access to Mary’s consciousness, their only information about what is happening is the description of Mary’s physical reaction or rather Clarke’s evaluation (“a horrible sight”) of these reactions. Mary’s reactions are evaluated and interpreted by suggesting possible causes or intentions, stating that she moved her hand “as if” to touch something and that her soul “seemed” to be struggling inside her. Readers can only infer Mary’s emotional state through her actions, the look of “terror” on her face, and her “shrieking.” The eventfulness of this encounter is intensified when the chapter ends with a brief epilogue of the experiment and its results. Readers then learn that Clarke returned three days after the experiment, only to find Mary “grinning vacantly”; a “hopeless idiot” according to Dr.  Raymond. There is no doubt about the origin of this horrifying experience when Raymond confirms that this result may have been inevitable, c­ onsidering that “she has seen the Great God Pan” (Machen [1894] 2006, 16). Later in the story, Clarke is in the process of reading an account of a young girl who became friends with Helen Vaughan, Mary’s daughter. That girl, Rachel, and another child, Trevor, directly encountered both Helen and Pan and their accounts immediately follow one another, but there is a significant difference in the description of their experiences due to the reference systems invoked. The text first establishes the eventfulness of the encounters by mentioning the consequence of the encounter for Trevor and Rachel, since before the two episodes are described, it is already mentioned that of the “persons concerned in this statement,” one has died prematurely (referring to Rachel) and another (Trevor) is now “an imbecile” (Machen 1984, 19).

Of Minds and Monsters  181 It is made clear in the first account that the young girl, Rachel, had some form of traumatic experience with the implication being that the girl has been raped: As soon as she saw her mother, she exclaimed, “Ah, mother, mother, why did you let me go to the forest with Helen?” Mrs. M. was astonished at so strange a question, and proceeded to make inquiries. Rachel told her a wild story. She said—Clarke closed the book with a snap, and turned his chair towards the fire. When his friend sat one evening in that very chair, and told his story, Clarke had interrupted him at a point a little subsequent to this, had cut short his words in a paroxysm of horror. (Machen [1894] 2006, 23) The reader is never told what happened to Rachel, and Clarke himself neither witnessed nor experienced these events, but his reactions as well as his refusal to read on allow the readers to understand the enormity of the events described in the account. Clarke makes the conscious decision not to learn more about Rachel’s experience and subsequently does not provide readers with a description of the event, only with its evaluation. The text acknowledges the limits of conceptual immersion while leaving the most powerful indicator of the horror behind Pan’s existence in Clarke’s refusal to read on and his “paroxysm of horror.” Through repeated refusal to acknowledge the sexual nature of the events (including what happened to Mary), readers learn the limits of what the characters are willing to talk about in the story. Conversely, Trevor’s account focuses more on the monstrous hybridity of Pan. The only thing that those questioning Trevor can gather is that he followed a girl, Helen, into the woods and then saw her with a “strange naked man” (Machen [1894] 2006, 20). The boy is unable to further describe what he saw. Unable to see what the boy saw, readers are left with the description of his condition by others, like his father and the narrator, in whose eyes he became “nervous and strange,” “waking in the night with cries of ‘The man in the wood! Father! Father!’” The boy slowly recovers but then suffers another episode that leads to “violent hysteria:” [...] a few minutes later, […], they were both horrified by a piercing shriek and the sound of a fall, and rushing out they found the child lying senseless on the floor, his face contorted with terror. The doctor was immediately summoned, and after some examination he ­pronounced the child to be suffering from a kind of fit, apparently produced by a sudden shock. The boy was taken to one of the bedrooms, and after some time recovered consciousness, but only to pass into a condition described by the medical man as one of violent hysteria. The doctor exhibited a strong sedative, and in the

182  Gero Brümmer course of two hours pronounced him fit to walk home, but in passing through the hall the paroxysms of fright returned and with additional violence. (Machen [1894] 2006, 22) Once again, the evaluation of the vagueness of Trevor’s condition (“a kind of fit”) and the reason behind it (“apparently produced by a sudden shock”), as well as the description of his external features (“his face contorted with terror”) in lieu of a more reliable account (like the readers’ understanding that Clarke “felt sick and faint” during the experiment) produce a form of emotional immersion that allows readers to imagine the features of the monstrous Pan even before they are presented or actualized by the text. Readers then learn that this second “shock seemed too severe for the boy Trevor, and at the present date he suffers from a weakness of intellect, which gives but little promise of amending” (Machen [1894] 2006, 22). It appears that the final reaction was triggered by the boy’s seeing a stone head of a “faun or satyr,” apparently reminding him of what he saw in the woods that day. The only description of the monster provided by the characters in these passages is that it is a man, that it is strange, and that it apparently resembles a faun or satyr. While this would simply make him some form of hybrid, the text emphasizes the insufficiency of a retrospective description in comparison to the actual experience when Clarke remarks that the source of this account told him that “he has never received such a vivid presentment of intense evil” (Machen [1894] 2006, 22). This is further emphasized by the fact that even the object in question, the stone head, is only a representation of Pan, not Pan himself, and even this representation is not described in detail. Toward the end of The Great God Pan, readers are presented with a manuscript written by Dr. Robert Matheson, a physician who conducted an autopsy on Helen Vaughan’s body. The first thing readers are told is that the doctor had died of an “apoplectic seizure” some four years after the events described in the manuscript and that the notes were “much abbreviated, and had evidently been made in great haste” (Machen [1894] 2006, 61). This prepares readers for the descriptions that follow establishing that the events about to be described may have been the reason for Matheson’s death, once again indicating the intensity of the experience before its narration. Afterwards, they are given Matheson’s first statement in which he assures his readers that he is of sound mind and that he is “suffering under no delusion,” even taking the time to calm down and steel his nerves to make sure that he was in his “real and true senses,” a statement already put into question by the information provided at the start of the manuscript (Machen [1894] 2006, 61). Matheson states that “horror and revolting nausea rose up within” him as he watched as the “firm structure of the human body” (including the flesh and muscles) before him “began to melt and dissolve” (Machen [1894]

Of Minds and Monsters  183 2006, 62). The process he witnesses goes against all of his previous knowledge, and he is unable to find an explanation within his own reference system, referring instead vaguely to “some internal force” of which he “knew nothing.” He continues on by stating: Here too was all the work by which man had been made repeated before my eyes. I saw the form waver from sex to sex, dividing itself from itself, and then again reunited. Then I saw the body descend to the beasts whence it ascended, and that which was on the heights go down to the depths, even to the abyss of all being. The principle of life, which makes organism, always remained, while the outward form changed. (Machen [1894] 2006, 62) Instead of merely dissolving, transforming the body wholly from the alter to the alien, it instead stays in the realm of alterity for a while, playing out a sort of twisted mockery of the evolutionary process. The form moves from “sex to sex” and from the “heights” of the evolutionary ladder to the bestial “depths.” All throughout, however, the “principle of life” remains, illustrating the natural laws behind it while perverting them by allowing the “outward form” to change in such an extensive manner. It is here that the two sources of horror in Pan, nature and sexuality, meet again. The effect this has on Matheson can be easily explained by the destabilization of his own scientific reference system, but of course there are more terrifying implications, like the nature of the relationships Helen Vaughan had with the people she seduced and how these have to be evaluated if she is neither truly male nor female (alterity) and, worse, not even entirely human (alienity). What he witnesses has a significant impact on Matheson, and his mind seems to have trouble comprehending such an extensive destabilization since his next entry reads like a hallucination: “The light within the room had turned to blackness, not the darkness of night” but rather, as he makes clear, “the negation of light” in which “objects were presented” to his eyes “without any medium” (Machen [1894] 2006, 62). Matheson seeks refuge in the reason of scientific discourse, attempting to analyze the effect this has on him, reaffirming that he can still see “clearly and without difficulty.” The text gradually distances itself from the reliability of Matheson’s descriptions, widening the gap between emotional and conceptual immersion. Perhaps it would be best to describe the strategies used here as strategies of alienization and alterization (cf. Turk 1990, 25). Alienization is a form of estrangement, in which something that has familiar features is pushed toward unfamiliarity and excluded from the group to which it formerly belonged. Alterization, on the other hand, refers to a process that is meant to establish familiarity even in cases where there is none by trying to conceptualize and label unfamiliar elements within a familiar reference system.

184  Gero Brümmer In the end, the spectacle seems to be over and Helen Vaughan’s body is reduced to a “substance as jelly.” However, one more transformation awaits the reader: Then the ladder was ascended again … [here the MS. is illegible] … for one instance I saw a Form, shaped in dimness before me, which I will not farther describe. But the symbol of this form may be seen in ancient sculptures, and in paintings which survived beneath the lava, too foul to be spoken of... as a horrible and unspeakable shape, neither man nor beast, was changed into human form, there came finally death. (Machen [1894] 2006, 62) The reader is made aware that Matheson has become so affected by the phenomenon he witnessed that his ability as a mediator is impeded. His writing becomes illegible and certain passages seem to be missing. Furthermore, he now actively refuses to describe what exactly he saw, instead referring to the “symbol” of this form found in sculptures and paintings, just like the one the boy Trevor encountered earlier in the story. Accompanied by the evaluation, the description itself moves further away from the realm of alterity to one of alienity. Matheson repeats the notion that the form before him is “unspeakable” or “too foul to be spoken of,” a combination both of his reluctance and his very inability to do so. Instead of describing the creature as a hybrid or a category confusion, he rather affirms the negative, stating that what he saw was “neither man nor beast,” something truly alien in the sense that it lies outside the reference system altogether, unlike the shifting from sex to sex that he explained earlier. In fact, he goes so far as to no longer call what is before him a “body,” as he did earlier. Instead, it is always described as only the “form” or the “shape,” never as body and certainly not as a creature. CONCLUSION Throughout the majority of The Great God Pan, the text makes it clear that on a purely descriptive and even evaluative level, there is no possibility for a potential audience to truly understand the intensity of this monstrosity and to share the eventfulness of the encounter, requiring reliance on the characters’ assurances and judgments. This is highlighted by the discrepancy between Clarke’s highly emotional reactions to witnessing the experiment or reading the narrative accounts of Helen’s actions and his futile attempt to put the meaning of Pan for him and his understanding of the world into words. In addition, it is implied that any attempt to describe Pan will always be insufficient, since Pan’s nature is alien to the characters’ understanding, and their descriptions will always end up as false analogies. In the final paragraph of Matheson’s account, the body of Helen herself becomes a mediator for Pan. The description of Helen, the creature of

Of Minds and Monsters  185 alterity, who at least in part is bound to the characters’ reference system, begins to draw more and more on the alien nature of Pan. Conceptually and descriptively, readers finally gain access to Pan through Helen’s transformations as witnessed by a character whose mental state gradually destabilizes. Even as the text is increasing the layers of mediation between reader and monster conceptually, the intense reactions to the monster allow the readers to imagine that at least on an emotional level they had access to Pan all along. As the analysis of The Great God Pan has shown, it makes sense to consider the poetics of immersion in the context of the entities of the fictional world and the different goals that structure its presentation. While Machen’s text might inhibit immersion on one level, the same inhibiting factor (like narrative distancing) might foster another form of immersion on a different level. With regards to horror literature in general, it will be interesting to see if the same holds true for other texts of that genre or whether this is a special feature of Machen’s text. Outside of horror literature, it might be useful to consider the status that specific objects or entities have in other events that can be considered emotionally involving, such as encounters that fill the fictional characters with feelings of extreme joy, anger, sadness, beauty, etc. REFERENCES Abbott, H. Porter. 2008. “Unreadable Minds and the Captive Reader.” Style 42.4: 448–67. Bruner, Jerome. 1991. “The Narrative Construction of Reality.” Critical Inquiry 18.1: 1–21. Burleson, Donald R. 1990. Lovecraft: Disturbing the Universe. Lexington: University of Kentucky Press. Carroll, Noël. 1990. The Philosophy of Horror, or, Paradoxes of the Heart. New York: Routledge. Cohen, Jeffrey Jerome. 1996a. “Monster Culture (Seven Theses).” In Monster ­Theory: Reading Culture, edited by Jeffrey Jerome Cohen, 3–25. Minneapolis, MN: University of Minnesota Press. ———, ed. 1996b. Monster Theory: Reading Culture. Minneapolis, MN: University of Minnesota Press. Douglas, Mary. 2005. Purity and Danger: An Analysis of Concepts of Pollution and Taboo. London and New York: Routledge. Geisenhanslüke, Achim, and Georg Mein, eds. 2009. Monströse Ordnungen: zur Typologie und Ästhetik des Anormalen. Bielefeld: Transcript. Gerrig, Richard J. 1993. Experiencing Narrative Worlds: On the Psychological Activities of Reading. New Haven, CT: Yale University Press. Heinke, Jörg. 2005. Die Konstruktion des Fremden in den Romanen von David Malouf. Würzburg: Königshausen & Neumann. Hogan, Patrick Colm. 2011. Affective Narratology: The Emotional Structure of S­ tories. Lincoln: University of Nebraska Press. ———. 2013. Ulysses and the Poetics of Cognition. London: Routledge.

186  Gero Brümmer Hühn, Peter. 2008. “Functions and Forms of Eventfulness in Narrative Fiction.” In Theorizing Narrativity, edited by John Pier and José Angel García Landa, 141–64. Berlin: Walter de Gruyter. ———. 2011. “Event and Eventfulness.” In The Living Handbook of N ­ arratology, edited by Peter Hühn, John Pier, Wolf Schmid, and Jörg Schönert. ­Hamburg: Hamburg University Press. http://www.lhn.uni-hamburg.de/article/event-andeventfulness. Hull, Thomas. 2006. “H.P. Lovecraft: a Horror in Higher Dimensions.” Math H ­ orizons 13.3: 10–12. James, Montague Rhodes. 1984. “Canon Alberic’s Scrap-Book.” In The Penguin Complete Ghost Stories of M. R. James, 11–20. London: Penguin. King, Stephen. 2010. Danse Macabre. New York: Gallery. Lovecraft, H. P. 2000. The Annotated Supernatural Horror in Literature, edited by S. T. Joshi. Hippocampus Press. ———. 2008. “The Dunwich Horror.” In Necronomicon: The Best Weird Tales of H.P. Lovecraft, edited by Stephen Jones, 264–98. London: Gollancz. Machen, Arthur. 2006. The Great God Pan; and, The Hill of Dreams. Mineola, NY: Dover Publications. Prince, Gerald. 1988. “The Disnarrated.” Style 22.1: 1–8. ———. 2008. “Narrativehood, Narrativeness, Narrativity, Narratability.” In Theorizing Narrativity, edited by John Pier and José Angel García Landa, 19–27. ­Berlin: Walter de Gruyter. Ryan, Marie-Laure. 1980. “Fiction, Non-factuals, and the Principle of Minimal Departure.” Poetics 9.4: 403–22. ———. 2003. Narrative as Virtual Reality: Immersion and Interactivity in Literature and Electronic Media. Baltimore: Johns Hopkins University Press. Schmid, Wolf. 2003. “Narrativity and Eventfulness.” In What Is Narratology?: ­Questions and Answers Regarding the Status of a Theory, edited by Tom Kindt and Hans-Harald Müller, 17–33. Berlin: Walter de Gruyter. Stevenson, Robert Louis. 2006. The Strange Case of Dr. Jekyll and Mr. Hyde. ­London: Penguin. Stoker, Bram. [1897] 2003. Dracula. London: Penguin. Turk, Horst. 1990. “Alienität und Alterität als Schlüsselbegriffe einer Kultursemantik.” Jahrbuch für Internationale Germanistik 22.1: 8–31. Wisker, Gina. 2013. “‘Spawn of the Pit’: Lavinia, Marceline, Medusa, and All Things Foul: H.P. Lovecraft’s Liminal Women.” In New Critical Essays on H.P. Lovecraft, edited by David Simmons, 31–54. London: Palgrave Macmillan. Wolf, Werner. 1993. Ästhetische Illusion und Illusionsdurchbrechung in der Erzählkunst: Theorie und Geschichte mit Schwerpunkt auf Englischem Illusionsstörenden Erzählen. Tübingen: Max Niemeyer Verlag. ———. 2011. “Illusion (Aesthetic).” In The Living Handbook of Narratology, edited by Peter Hühn, John Pier, Wolf Schmid, and Jörg Schönert. Hamburg: Hamburg University Press. http://www.lhn.uni-hamburg.de/article/illusion-aesthetic.

10 Narrative Conventions in Hallucinatory Narratives Tommi Kakko

Drug literature can be considered a literary genre in its own right. It is hardly the most respectable kind of literature, because it is often written by drugaddled eccentrics and oddball amateur philosophers. But as with any form of literature that has had an impact on generations of readers, it deserves attention from literary historians, critics, and scholars. This essay will look at some general features of the genre, its history, and the main themes discussed in two important early works. These are Thomas De Quincey’s ­Confessions of an English Opium-Eater (1821) and Fitz Hugh Ludlow’s The Hasheesh Eater: Being Passages from the Life of a Pythagorean (1857). The major themes of the two classic works are then compared to a more recent example, Terence McKenna’s True Hallucinations: Being an Account of the Author’s Extraordinary Adventures in the Devil’s Paradise (1993). Hallucinations appear in all of the works and are the main focus of the essay, because they represent a specific problem that spans the entire genre. Namely, how to describe experiences of objects that appear to the senses under the influence of certain drugs. It would be easy to dismiss the question as one that overthinks the ravings of dysfunctional minds, but this would be doing the authors a great disservice. As the works inevitably discuss the nature of perception and the human mind using theoretical tools borrowed from philosophy, a contemporary reader can gain insight into the history of epistemology and metaphysics. In addition to philosophy, they provide interesting peeks into the ways the disordered mind was conceptualized at different times. What is also revealed is that the form of drug narratives appears to have remained more or less the same since De Quincey. The modern popular drug narrative takes the form of a story of redemption. Mike Jay notes in his classic study Emperors of Dreams that this “my drug hell” confession related by a recovering drug abuser has become the standard in popular media: “By contrast, ‘how I took drugs for many years without any terrible consequences,’ though a far more common story, is one which few people are moved to write” (2000, 65). Why this has occurred, I argue, is related to the self-perpetuating formal qualities the narratives seem to share. Monika Fludernik’s concept of narrativization, “a reading strategy that naturalizes texts by recourse to narrative schemata” (Fludernik 2002, 25),

188  Tommi Kakko applies here. Stories that follow established and even schematic literary conventions facilitate the adoption of such reading strategies and, in turn, ­stories that follow them gain in popularity due to their form. If one assumes a philological and literary historical approach to the texts, it is possible to examine the connections between literary traditions and related literary conventions more closely. In this case, it becomes apparent that authors who wrote after De Quincey were able to resort to the generic conventions popularized by the Opium-Eater. The burden of the hallucinating subject was thus eased by a more or less common language that formed around the topic. Conversely, a historical perspective also means that a reader cannot approach the stories as incorrigible first-person reports of experience. One has to keep in mind that the language of past reports always affects the way in which these extraordinary experiences are reported. Despite common generic conventions, each of the authors has a special take on the form of the drug narrative. I include McKenna’s case because of his influence and contemporaneity. It is also virtually impossible to find critical assessments of his works that concentrate on the more literary qualities of his writing. As a modern storyteller, McKenna’s sphere of influence may have been marginal, but he did manage to accumulate a large number of appreciative readers around the world. More important than his influence or literary style, however, are the questions he raises. These questions, one cannot fail to note, are very much like the ones raised by De Quincey, Ludlow, and many other authors in the genre who have had to face their own hallucinatory experiences. De Quincey asks his questions during the first decades of the 1800s. Ludlow does the same in the 1850s after De Quincey. McKenna, who grew up in the sixties, asks them during the early 1990s. Finally, one must naturally ask oneself what these stories can tell readers about ways of reading the mind in literature in general and especially in cases where the mind encounters something so alien and strange that words fail to do justice to the experience. HALLUCINATORY EPISODES IN EARLY DRUG LITERATURE As with any defined genre, texts that can be considered drug literature share various narrative elements, conventional literary devices, and, of course, a history of literary works that can be singled out as paradigmatic examples. In this case such works tend to contain an important element of autobiography and candidly discuss issues pertaining to the experiences enjoyed and endured by their authors. Although stories of mystical realms and fabulous creatures encountered by the hallucinating subject have always been the stuff of literature, De Quincey was arguably the most influential modern author to include drug-induced hallucinatory episodes as a central element in his autobiography. The Romantic poets, Samuel Taylor Coleridge among them, embraced opium in a way that built a new mythology around the

Narrative Conventions in Hallucinatory Narratives  189 1

drug. Unlike Coleridge, whose opium habit finally crippled his intellect, De Quincey approached the subject analytically. As Jay points out, De Quincey placed himself at the centre of his narrative and effectively replaced the drug with the autobiographical subject as the leading character in drug literature. He did so after watching Coleridge, whom he idolized, deteriorate into an invalid. According to Jay, De Quincey thought “that the way to escape the drug’s clutches was not to curse and berate its effects but to understand and address them, and to forge them the central narrative of one’s life” (2000, 60). His stories were confessions in the Enlightenment genre made popular by Rousseau and aimed at exploring the role of the subject in the grips of the effects of opium. In the narrative template created by De Quincey—one of escalating depravity and subsequent redemption cataloguing the pleasures and inevitable pains of opium—the protagonists are debauched by the drug they love. They are redeemed by denying themselves the very pleasures that would destroy them. Visions, dreams, and hallucinations appear frequently in De Quincey’s work. The introduction of a drug distinguishes them from similar episodes in previous literature. As the visions are drug-induced, their causes are always known to the subject and while they are rarely judged anything but astonishing, they lack the divine origin one would normally find in, say, reports of the religious visions and prophecies of mystics.2 De Quincey attributed his visions primarily to an innate faculty he had possessed since childhood, one that was painfully enhanced by his opium habit. In his psychological speculations De Quincey concluded that “in some that power is simply a mechanic affection of the eye; others have a voluntary, or a semi-voluntary power to dismiss or summon them” (1985, 67). His approach was that of a scientific observer who had arrived at the uncharted territories between dreaming and consciousness fully aware of what he called the “fierce chemistry” (68) of his brain. While the same faculty was celebrated by the Romantic poets, De Quincey showed readers its dark side. He also introduced a method of introspection that would become as important to his most careful readers as his narrative template would be to the wider reading public. De Quincey’s cool and analytic take on the visionary faculty inspired the American Fitz Hugh Ludlow to produce his own stories of the drug experience, The Hasheesh Eater. Ludlow was unapologetic about his ­admiration and emulation of De Quincey’s pioneering work—De Quincey had many followers after Ludlow, but few would follow him as diligently. As his disciple, Ludlow reproduces De Quincey’s redemption narrative almost to the letter. The narratives trace a similar basic trajectory. Both authors experiment with various substances until they feel comfortable with the drug featured in their respective stories, both have a brief honeymoon with the substance, fall in love with it, experience life-transforming hallucinations, and suffer from addiction until they are able to free themselves. Despite his vivid accounts of the horrors of addiction and withdrawal, De Quincey was criticized for glamorizing opium. This is perhaps not surprising, for in De

190  Tommi Kakko Quincey’s hands opium quite often sounds like a gateway into Coleridgean lands of unearthly delights. It should be noted that neither De Quincey nor Ludlow finally resolved their substance abuse problems as well as the narrators of their books did. Despite their different drugs of choice, De Quincey and Ludlow were both struck by the strangeness of the experiences they went through under the influence. They were appalled and distraught as much as they were ­mystified, but initial feelings of horror (cf. Brümmer in this volume) usually gave way to fascination. The sublime peaks of the episodic hallucinatory experiences left them in awe, which no doubt amplified the allure of the drug for many readers. In contrast to De Quincey’s cool empiricism, Ludlow’s hyperbolic prose takes the reader to more metaphysical realms of thought. Ludlow felt restricted by the philosophies of British empiricists who, he thought, circumscribed the knowable in intolerably narrow terms and fenced him in a barren desert of objective facts.3 For Ludlow, German Idealism seemed to transcend the restrictions of empiricism and provide a language with which it was possible to describe, at least to some extent, the vast depths of the ideal realms revealed through his experiments with hashish oil. Like De Quincey, Ludlow took a great interest in language in his story, but his concerns ran much deeper. He was particularly concerned with ­finding a language powerful enough to convey the impact of the outrageous ontological and epistemological ambiguity of his drug-fuelled visions. The hallucinations revealed, to quote Ludlow’s hyperbolic report, “forms and modes of existence which, on earth, are impossible to be expressed, for the reason that no material emblems exist which even faintly foreshadow them” (2006, 112). He tries to convey his experiences to his friends and finds himself powerless: “What that truth was I strove to express to my companion, yet in vain, for human language was yet void of signs which might characterize it” (105). He cries out to God, exhibiting a cringeworthy melodramatic angst: “[G]rant me the gift of a supernatural speech, that I may, if ever I return, come to humanity like a new apostle, and tell them of realities which are the essence of their being!”(ibid.) There is no genuine scepticism about the veracity of the experience itself. Rather, the mind is filled with imagery that is simply too fantastic to relate. Both authors turned to contemporary philosophy in order to comprehend something of the otherworldly depths that challenged their notions of reality. De Quincey did so with the cool head of an empiricist scholar, Ludlow with the heated fervour of the enthusiastic idealist. They turned to philosophy, because the hallucinations they faced were too bizarre to submit to anything like the categories used to describe and explain everyday life. The fundamental problem their accounts brought to the surface was the impotence of language when it is confronted with a hallucinatory e­xperience. The epistemological question they struggled with consisted of the problem of describing places and objects that, by definition, do not exist but nevertheless appear before the baffled subject. The ontological problem this entailed

Narrative Conventions in Hallucinatory Narratives  191 involved reconciling a drug-induced and hence completely natural visionary experience with its apparently supernatural import. De Quincey, Ludlow, and others writing in the genre have all struggled with translating extraordinary private experiences into the public language of literature. In fact, they do so frequently enough to warrant the claim that the problem of description constitutes one of the main generic f­eatures of drug literature. Authors very often express their inability to relate what takes place once the drug takes hold and the experience approaches its sublime peak. While it is very difficult for the authors to provide adequate descriptions of hallucinations, they are easily introduced into the mind through suggestion. Art, music, and poetry prompted images in the minds of both De Quincey and Ludlow. Oriental imagery in particular featured heavily due to the contemporary fascination with the Orient. In other words, public ­language is effective in generating hallucinatory images in the minds of the hallucinating subject. The mechanism does not, however, work in reverse, and the hallucinating subject is more often than not unable to translate the most extraordinary features of the private hallucinatory experience into l­anguage, to extract them from the mind of the subject and place them into the public sphere, without eroding the character of the experience dramatically. It is not surprising that the genre emerged during a time when ideas of the unique position of the subject and the role it played in relation to the sublime took hold of the Romantic imagination. In a study examining the role of medicine and visions in the Romantic period, Gavin Budge writes: “Romanticism is virtually synonymous with visionary kinds of experience … visions, hallucinations and apparitions are not only frequent themes in Romantic writing, but are so woven into its texture as to blur any dividing line between the imaginary and the real” (2013, 3). However, hallucinatory narratives were different from typical Romantic poetry. The authors found themselves challenged by the task of describing a world so strange that they readily confessed to failure at the outset. What caused these visions was not related to the deepest mysteries of Wordsworthian nature. Rather than trying to catch nature in the act, authors like De Quincey and Ludlow got much more of its sublimity than they had hoped for. Having gained access to a realm of natural visions, they tried desperately to distance themselves from such mysteries when they were revealed to them. In trying to overcome the limitations of language, they turned to the fantastic: the possibility of creating objects with language, inhabiting a parallel universe or crossing over at will, inhabiting the mind of another, bringing a hallucinatory object back into the real world from the world of the hallucination, and the like. Where language failed to capture the real, fiction took over. Given the extraordinary circumstances, however, one can hardly say that the stories fail as autobiography when the fantastic enters the text. Rather, the special role of highly figurative language as a tool to describe an ersatz-reality is understood by both authors and readers. Failure to capture anything essential about the deep

192  Tommi Kakko mysteries revealed by the world of hallucinations is inevitable and therefore the authors must provide a flawed report whatever the case. Because the genre is defined by an experience that refuses to surrender to figuration and takes the author to the very limits of linguistic expressivity, the generic conventions that make up its characteristics revolve around the practical problem of a language that cannot transcribe events or objects in a way that captures their essential nature. Much like one can identify the Romantic subject as the driving force of Romantic poetry, the absurd demand for an impossible language that can communicate the hallucinatory experience can be identified as the question that motivates the genre of autobiographical drug literature in general and the narratives’ hallucinatory episodes in particular. The genre, in other words, questions the role of language as a viable means of communicating fully what one has to say. Conversely, a reader introduced to the conventions of the genre will quickly learn to expect only approximations of what actually occurred and a reading experience that is highly figurative as the author can only sketch an impressionistic ekphrasis of what is revealed on the screen before the eye of the mind. Authors who write about their hallucinations thus do not claim to represent their experiences fully or even accurately, but as best they can with the tools available to them. It is often the case that they vouch for the veracity and intensity of their experiences nevertheless. Determining whether or not they actually occurred outside the author’s mind is not always even an issue. A VISIT FROM A FLYING SAUCER Much of modern drug writing has moved to Internet forums where one can find numerous trip reports written by amateurs and so-called drug geeks.4 Trip reports differ from drug literature in that they simply report the effects of drugs without placing these episodes in a framing narrative. They thus get rid of the classic redemption narrative and have a structure of their own, one which can also be found in De Quincey and Ludlow. Curiously, the trip reports show that descriptions of experiences categorized according to specific drugs also tend to generate their own tropical and figurative language as the reports accumulate. That is, the experiences are discussed by the online communities in a way that resembles the creation of a ­literary genre in miniature. The authors take recourse to culturally available or ­tellable stories that provide a set of conventional expressions that seem apt for the situation.5 The basic form of the stories is usually similar, at least with regard to psychedelics. The set and setting are introduced, the drug ingested, the subject waits for it to take effect, the trip begins and peaks, and the aftereffects are then reported. The details of the descriptions, especially at the peak at the trip where language struggles to keep up with the visions, tend to coalesce over time into a mass of generic images. What takes place in the online communities is a creative linguistic activity centred on the

Narrative Conventions in Hallucinatory Narratives  193 very problem of description already familiar to the likes of De Quincey and ­Ludlow. That is not to say that the entire effort is one involving mere fiction or that the drug experience is only of secondary importance to the language games played by the participants, but instead to note that the search for the correct means of expression is as important to these online communities as it was for the earliest writers coming to grips with the experience. Terence McKenna is one of the most influential modern authors whose writings act as a point of reference with regard to several psychedelics. A self-styled psychedelic guru, McKenna was endowed with an exceptional facility with language, which he attributed simply to his liberal arts education. He was also an advocate of psychedelics and, in terms of his l­iterary output, a modernizer of the drug literature genre. His experience with various types of hallucinogens was vast and unlike many who dabble in the subject, he was lucid and methodical in his descriptions of their effects, chemistry, and cultural significance. McKenna and his brother Dennis are known for their many books that range from a mushroom grower’s guide to theoretical works attributing human evolutionary traits to changes in diet. McKenna was also a popular lecturer who toured extensively, and many of these lectures and talks are available to the public in online archives. McKenna often referred to his life-transforming visit to the Colombian Amazon in 1971 where he and his team of explorers hoped to study an easily ingested form of a tryptamine hallucinogen. The team was distracted by psilocybin-containing mushrooms that grew in abundance at one of their campsites and with their discovery the expedition took an unexpected turn. Plans to discover and study a new type of hallucinogen were put on hold, and the mushroom took its place, partly due to a brief psychotic episode suffered by McKenna’s brother Dennis after he had ingested the mushroom. The entire story of the expedition was finally published by McKenna as True Hallucinations in 1993. From the subtitle, Being an Account of the Author’s Extraordinary Adventures in the Devil’s Paradise, it is clear that McKenna places his tale within the same genre as De Quincey and Ludlow. As is the case with true connoisseurs, McKenna was well-read in drug literature and knew his literary history well enough to incorporate many generic traits into his own narrative. True Hallucinations is an autobiographical adventure story with several reports of wild hallucinatory episodes. McKenna’s work is indebted to De Quincey and Ludlow, but the work is atypical as drug literature in several distinct ways. The Orient as a source of exoticism has been replaced by the Amazon rainforest and its inhabitants. The cultures of the native peoples of the Amazon are naturally of great interest to the party of explorers, but they are hardly informed students of local customs. During the period between the expedition and the publication of True Hallucinations McKenna, who had already travelled extensively in Asia studying (among other things) Tibetan culture and working as a smuggler, focused his attention on the shamanic cultures of the Americas. The story itself explains why he chose to

194  Tommi Kakko do so. As the drug experience launched his studies into the subject and his subsequent career, it is fair to say that De Quincey’s redemption-template has been discarded in True Hallucinations. Adverse effects of the mushroom are reported, but the story has no real downfall of a penitent hero. There is no talk of addiction or of a deeply personal struggle to rid oneself of the drug. Quite the contrary, the book ends with the discovery of a new and joyous mystery that has been laid out in front of the author for him to explore further. The episodic reports of hallucinations are also quite unconventional, but broad lines with typically generic features can be identified. After familiarizing himself with the mushroom and a number of experimental trips at his camp in La Chorrera, McKenna is drawn to the river by his intuition, which commands him to watch the skies. Or rather, the mushroom suggests he do so as McKenna speaks of the Psilocybe cubensis mushroom as a conscious entity that was able to impart information encoded in its flesh into the mind of one who ingests it. The hallucinations reach their peak when McKenna encounters a UFO on the banks of the river flowing through their campsite. In short, he observes a cloud formation that transforms into a spaceship. As an expert in psychedelics, McKenna is familiar with the protean nature of the experience and the difficulties experienced by those who wish to remember and report it in detail. He suspects the vision will be ephemeral, tries to drink in the image with his senses, and, to his amazement, sees it very clearly: “It was a saucer-shaped machine rotating slowly, with unobtrusive, soft, blue and orange lights. As it passed over me I could see symmetrical indentations on the underside. It was making the whee, whee, whee sound of science fiction flying saucers” (1994, 158). The saucer holds a special place in the story. It is symbolic, metaphorical, and very real for McKenna. It reveals something essential about the world without being entirely ontologically true or false or indeed any sort of intergalactic vehicle. At no point should one think McKenna believes he has seen an actual spaceship filled with little green men. As the title of the book reveals, the veracity of the hallucinatory objects McKenna sees and hears is a question that interests him. He contemplates the noumenal existence of the flying saucer before him and its relationship to his merely subjective perceptions and notes that he has an advantage over most people who would encounter such a thing: Was it real in the naïve sense in which that question is asked of UFOs and tables and chairs? No one else saw this thing as far as I know. I alone was its observer. I believe that had there been other observers, they would have seen essentially what I have reported, but as for “real,” who can say? I saw this thing go from being a bit of cloud to being a rivet-studded aircraft of some kind. Was more true to itself as cloud or aircraft? Was it a hallucination? … I am familiar through direct experience with every known class of hallucinogen. What I saw

Narrative Conventions in Hallucinatory Narratives  195 that morning did not fall into any of the categories of hallucinated imagery I am familiar with. (McKenna 1994, 158–59) The question he asks about the actual objective existence of the vision is not rhetorical. The hallucination is true, as the title of the book implies, but only in a very specific sense of the word. It is ontologically ambiguous, as hallucinations tend to be, and its ambiguity manifests itself in dramatic fashion. Its quasi-being reveals a truth that cannot be communicated directly. The strangeness of the experience is palpable in McKenna’s description, but it is accompanied by an even stranger familiarity that makes a mockery of questions such as “Is it real?” What he actually recognizes turns out to be more puzzling than the fact that he has just witnessed a flying saucer take off to the skies: [A]s the saucer passed overhead, I saw it clearly enough to judge that it was identical with the UFO, with three half-spheres on its underside, that appears in an infamous photo by George Adamski widely assumed to be a hoax. I had not closely followed the matter, but I accepted the expert opinion that what Adamski had photographed was a rigged up end-cap of a Hoover vacuum cleaner. But I saw this same object in the sky above La Chorrera. … By appearing in a form that casts doubt on itself, it achieves a more complete cognitive dissonance than if its seeming alienness were completely convincing. (McKenna 1994, 159) The vision of a vacuum cleaner part impersonating a UFO seen by ­ cKenna poses yet more questions. If one grants that the description is M accurate, as the narrator assures the reader it is, one has to ask if this really is a case of an earlier representation of a UFO manifesting itself as a sense perception. That is, a sign posing as the very thing it stands for, or a signifier posing as a signified experienced as an external object. What has previously been the means of describing the experience becomes the expe­ cKenna’s account, but doubt about the reversal of the normal rience in M relationship between reference and referent remains, as McKenna is quick to point out. He writes as if there was some form of intelligence behind the vision and the way in which it appears. The intention to baffle the subject receiving the message seems clear to him even if the content of the message itself is undecipherable. Given McKenna’s knowledge and admiration of Ludlow’s Hasheesh Eater, one may read the episode as a novel take on the latter’s problem of finding a language powerful enough to report hallucinatory experiences. Language certainly breaks down often enough in McKenna’s account and gaps remain between private experience and public language, but McKenna does have an advantage over Ludlow. McKenna has the opportunity to refer

196  Tommi Kakko to, among other things, modern epistemologies informed by the linguistic turn and cultural theory that emphasize the role of the symbolic mode as a force shaping the lives of people in its sphere of influence. Ludlow’s solution to the problem of describing the ineffable was to appeal to the ways in which the world is constructed through Kantian intuitions and categories. McKenna is able to use the linguistic turn to his advantage in trying to get at the heart of the problem. In a sense, McKenna offers a solution to Ludlow’s problem of finding a supernatural language. It is unlikely, however, that Ludlow could have foreseen that such a language would be borne out of an inversion of the normal relationship between signs and objects in the world. That is, that such language would be identical to the experience itself and so solidly rooted in everyday language. RESISTING CULTURAL CONVENTIONS In his influential Structuralist Poetics, Jonathan Culler makes an often-cited point (cf. Rantanen in this volume) about the nature of literature. It is worth repeating at length: [W]e are attracted to literature because it is obviously something other than ordinary communication; its formal and fictional qualities bespeak a strangeness, a power, an organization, a permanence which is foreign to ordinary speech. Yet the urge to assimilate that power and permanence or to let that formal organization work upon us requires us to make literature into a communication, to reduce its strangeness and to draw upon supplementary conventions which enable it, as we say, to speak to us. … The strange, the formal, the fictional, must be recuperated or naturalized, brought within our ken, if we do not want to remain gaping before monumental descriptions. (Culler 1975, 134) Literary language holds a promise of some unearthly power or perhaps merely a strangeness that can be made to relate to the truly strange. Its power will, then, be converted back or naturalized into more mundane interpretations by readers who wish to make sense of the text. Poets have known this as long as there has been poetry. Their readers have known of the need to shape the fantastically strange into something more palatable as long as poetry has been read. As Gerald Bruns notes in his history of hermeneutics: “Western culture has always been deeply allegorical in its ­operations and results; it has a special genius for constructing ways of reading poetry, or any alien discourse, so as to make it consistent with its own prevailing cultural norms” (1992, 230). Poetry, as a craft and even as a practice built around the mystery of the transcendental logos, should make sense to readers and, conversely, poetry should make sense of the world.

Narrative Conventions in Hallucinatory Narratives  197 Culler’s brand of structuralist poetics also contains an additional claim according to Bruns, who ventriloquizes the stance of the structuralist: Whatever it is on the surface, poetry has a lawlike deep structure or textual logic continuous with the culture itself, whence it is but a short step to convert poetics into a general cultural analytic in which all discursive and non-discursive modes of production can be examined from within the conceptual frame of a single methodological outlook. (Bruns 1992, 231) McKenna’s prose appears to rebel against any such monolithic cultural t­heory. Simultaneously, it recognizes the theoretical impulse to create a ­single methodological outlook—one might as well call such an outlook a metaphysics or at least an epistemology. In the end, McKenna’s UFO is a way of pointing out how one is restricted by cultural norms, institutions, and even one’s language. In the experience reported by McKenna, the flying saucer assumes a shape that makes it impossible not to think that representations are ontologically prior to their associated objects. That is, it is impossible to miss the point that the representation has actually constructed the experience. The message of the vision appears to reveal something not unlike the strong thesis of social constructionism in an absurdly amplified fashion. The institution of language can after all be understood, in the manner of the sociologists Berger and Luckmann, as a set of earlier cultural conventions that have been assumed unquestioningly by the masses and which, in fact, control aspects of human thought.6 Experiencing the breakdown of language in the hallucinatory state and so realizing the ability of language to impose meaning on the world is a valuable experience for the modern subject, McKenna seems to claim. The breakdown acts as a visceral reminder of how strange nature can be outside socially conditioned human life. To the uninitiated, the fact that one’s mind is, as it were, under the control of customs, social roles, and the institution of language may be distressing in itself. The breakdown of language also makes one question whether there really can be a single epistemology that would be able to explain all cultural artefacts from Shakespeare to McKenna’s flying saucers. In this context, the need to naturalize (in Culler’s sense of the term) the fantastic can be seen as a need to understand the radical otherness of the experience. In terms of language, referring to objects language cannot reach fulfils this need. Authors like De Quincey, Ludlow, and McKenna can be read as examples of the violence one reporting an exceptional experience must do to the radically other in order to force it into a frame of reference that makes sense to readers. In McKenna’s absurd UFO encounter in particular, the symbolic and the actual merge in the bizarre image of the saucer. The experience may be absurd or even completely ludicrous to the reader, but something of one’s relationship to reality is revealed as signs are reported to become objects in the world. When the world becomes language, the power of language to refer to objects

198  Tommi Kakko outside itself is naturally destroyed or discarded. In a metaphysical scheme of this nature, language can only refer to other language, as structuralists were keen to remind their contemporaries. There is an element of caricature and even satire in McKenna’s description of the fully symbolic and simultaneously actual flying saucer, but his laughter stems from the innate absurdity of human aspirations to understand the world. Its criticism is gentle. It is not a wholesale rejection of the projects of structuralism or social constructionism. It is perhaps based on the undeniable fact that one has to be able to stand outside closed systems like structuralism or constructionism in order to understand them. The only way to do so, it seems, is to render the subject powerless to function within the rules of the system. Readers may be used to the conventions used in hallucinatory episodes in drug literature, but this does not mean that they can be fully naturalized (Culler) or narrativized (Fludernik).7 The hallucinatory object seems to be a case of the “non-natural” (Fludernik 2002, 8) that demands attention beyond the realm of signification. Genres such as drug literature communicate to readers that there are occasions when experience cannot be vocalized. The readers are left to their own devices to narrativize the silences in these episodes. Or, perhaps, they are left to fail and interpret the text through their failure to naturalize their strangeness. Working with such binary schemes, one may sometimes get the impression that narratology and hermeneutics portray the interaction between the reader and the text as a struggle to return rogue texts to the fold, but here this would be misleading. Instead of a struggle between texts and readers, one might be better off looking at hallucinatory episodes in drug literature as friendly cat-and-mouse games between the author and the reader. Authors are attracted to strange literary devices in order to communicate the ineffable, while readers indulge them knowing that language is never efficient enough to succeed in its task. Both rely on tools that are mutually recognized as insufficient to bridge the gap between the experience and the author and the reader. Not all language, then, seems to require naturalization or narrativization in the same degree. The ultimately insular phenomenological world of the subject experiencing a hallucination must remain beyond reach in order to be described and read within the confines of generic drug literature. There must be a breakdown of communication with the breakdown of language in order for the experience to be communicated. What is also needed is an active reader willing to play with the text, with fellow readers and perhaps even with the author: in other words, a discourse that is open-ended and unresolved. The activity generated thereby will bring about new conventions within a genre, but even they are conditioned by the inevitable failure of communication generated from any attempt to communicate fully the nature of the hallucinatory episode. The language of hallucinatory episodes in drug literature cannot escape Bruns’s criticism of structuralist epistemology either. Western culture may have a genius for subsuming alien discourses under its cultural norms, but as McKenna’s example shows, when

Narrative Conventions in Hallucinatory Narratives  199 the message of the discourse seems to carry with it a resolve not to submit to linguistic norms, one should allow for the possibility of discourse that will function on the basis of that resistance. Dreams of a single unified methodological outlook that could incorporate that which expressly denies subjugation to method should therefore be tempered by allowing a possibility for the radically other. A unified methodology supplemented with a footnote that points to discourses that function on a principle totally alien to it, however, can hardly maintain its status as “a general cultural analytic,” as Bruns puts it. As for claims to methodological consistency, without allowing for the existence of the alien and the strange, consistency begins to resemble methodological violence. Therefore, instead of applying a method of reading to the above literary texts, one should look at what these literary texts have to say about the methods one tries to apply to them. What seems to motivate De Quincey and especially Ludlow and McKenna is the possibility of returning to the world of the “ontic logos,” as Charles Taylor (1989, 144 et passim) has called it. This would entail a return to a world where the modern problems of alienation and the like would not affect the subject as one would simply not be able to detect the gap between one’s experience and one’s representation of it. Of course, immanence of this nature is not available to the moderns and returning to a pre-modern mode of experience is simply not possible. What is also out of reach is a language that would be anything other than a representational mode of reporting experience, a pale shadow of the immediacy of Ludlow’s dreams. Such immediacy is destroyed through a process of objectification that leads to a disengagement with the world. Taylor writes elsewhere: “If we take a domain of being in which hitherto the way things are has defined meanings or set standards for us, and we now take a new stance towards it as neutral, without meaning or normative force, we can speak of objectifying it” (Taylor 2007, 283). A vision or hallucination, a thing of ideas, is thus depleted of inherent meaning as an object. The subject is doomed to describing it with words that merely mirror the conceptual rather than the actual world. Simply put, one no longer perceives things through their meanings: “Objectification brackets those meanings, and sets them aside. They no longer animate our enquiry. We as agents-living-meaning withdraw, as it were, from this enquiry. We ­ urselves outside a certain space of meanings while examining the place o things of this domain” (ibid., 284). A hallucinating subject who relates to his or her hallucinations without bracketing the possible meanings of visions would surely be thought mad in a culture that has collectively discarded the ontic logos as a viable mode of being. They might be redeemed as, to use Taylor’s term, agents-living-meaning, but their redemption would be indistinguishable from lunacy. Yet, dreams of the ontic logos linger on. Sometimes they create a nostalgia for a time when man was more in tune with the world and not suffering from the bizarre condition of not being able to report the contents of his or

200  Tommi Kakko her mind. Nostalgia for an ancient way of being in the world in the natural state of man before the Fall would certainly explain Culler’s rhetoric of returning the literary into its natural, recuperated state. A linguistic method that would be powerful enough to encompass the entire world of human experience is not far from the linguistic immanence of the ontic logos. It is of course only natural to strive for ways of naturalizing and narrativizing the strange, the formal, the fictional, and the alien if the other option is simply to stay silent before monumental descriptions that categorically refuse to conform to prevailing cultural norms. Literary language should make sense, one feels instinctively, even when it points to the disenchantment that has severed its ties to immanence. Literary language makes sense even when it does not, one wants to say. Whether or not one wants to defend such a claim is another matter, but descriptions of experiences that are discovered to function as descriptions only if they are left in their non-natural state are hardly enough to derail the human impulse of making sense of the world. That way madness lies, quite literally. CONCLUSION As one reads thoughtful authors struggling with language, attempting to capture the peak of a hallucination, one cannot but think that their task would have been made much easier had they been able to appeal to a supernatural power as the source of their visions. De Quincey’s decision to make the subject self-contained in relation to the hallucinatory experience was perhaps more far-reaching than all the other narrative conventions he created. Ludlow dreamed of a supernatural language that would help him become a new apostle for humanity. He wanted to be able to say something about the essential being of man, because he felt he had found something hidden in human nature. It (whatever it was) demanded to be known, but simultaneously appeared in a form that resisted cultural assimilation and left Ludlow in despair. McKenna found something like a new language that might have matched Ludlow’s dream, but only by creating a paradigm where the world was simply made of language. In McKenna’s ironic solution to the problem of linguistic impotence, humanity seems to be trapped in a prison house of language from which there is no real escape. The answer he found also appeared in a form that contradicted itself. McKenna’s sublime saucer shows the reader how language conditions experience and, more importantly suggests there are experiences that are not conditioned by language. Some things, terrifying and magnificent, seem to escape its grasp. The result is a delicious paradox that can be celebrated as a reflection of the absurdity of elevating human understanding above the infinite variety of nature. The linguistic turn and the various ambitious brands of semiotics related to it seem to suffer a blow in McKenna’s attempt to understand what he ­witnessed in the skies above La Chorrera. The episode suggests that a successful pan-cultural theory based on structuralist notions of language, the

Narrative Conventions in Hallucinatory Narratives  201 kind criticized by Bruns, would enable the naturalization of culture itself as a whole. This would realize Ludlow’s dreams of a supernatural speech. It would also amount to an alchemical summoning of the philosopher’s stone, an abstract panacea for all scholarship for all time. Naturally dreams of this kind are idealizations of the perfection of understanding, a philosophical enthusiasm to be commended, but they are ready targets for ridicule by authors like McKenna. He speaks for the other, against its decimation in the hands of pan-theoretical thought determined to perfect itself at all cost. There is no place for the strange, the alien, the other in such thought. For what could a perfectly naturalized culture naturalize itself in relation to if not itself and itself alone? Oddly enough, the wild-eyed shaman in this case preaches moderation. What, then, do these stories tell us about ways of reading the mind in literature? First, they show that disenchanted modernity has revealed the inadequacy of human language to record visions that occur in the absence of the divine. The best descriptions of such visions cannot be naturalized or narrativized, because they refuse to compromise the strangeness of the experience. The language of hallucinations is one of paradoxes, metaleptic tropes, irony, satire and playfulness. Its formal elusiveness mirrors the ­elusive nature of the experience and perhaps in this sense it is in some sense mimetic and naturalizable. Beyond that, it is difficult to see how it could be “brought within our ken,” as Culler says. Second, discussion about the elusive experience is made possible by the space that the literary conventions create within the genre. That is, one might be better off not thinking of them as conventions that govern the discourse. The experience itself is of more importance. Certain conventions are recycled, and iterations of the same story are passed around, but this is done in the spirit of play. It is also done as a self-aware experiment in prose that can never reach but an approximation of the original experience. Finally, one should consider the fact that in the narratives discussed above something essential of the mind of the other is left uncovered, which is probably a close approximation of the way the minds of others appear in real life. The actual minds of others are relatively closed to us. They are often wide open in fiction. In reality, they are strange and unfathomable, sometimes almost alien in their strangeness. What little we know of them comes through in trickles of imperfect language we can never be completely sure mean what we think they mean. McKenna was fond of repeating that reality is not only stranger than one can imagine, but stranger than one can suppose. Perhaps we could say the same about the mind and in doing so be one step closer to the understanding of which we all dream.

NOTES 1. The influence of opium in particular has been the subject of great interest in modern literary criticism at least since the thirties with the publication of

202  Tommi Kakko M.H. Abrams’s The Milk of Paradise: The Effect of Opium Visions on the Works of DeQuincey, Crabbe, Francis Thompson, and Coleridge (1934). 2. Late eighteenth- and early nineteenth-century men of science remained surprisingly open to apparitions and hallucinations. John Ferriar’s An Essay Towards a Theory of Apparitions (1813) and Samuel Hibbert’s Sketches of the Philosophy of Apparitions; or, an Attempt to Trace Such Illusions to Their Physical Causes (1824) continued a line of research that had begun with more speculative works and that had been taken over by the associationism John Locke, David Hume, and others. Ferriar and Hibbert had no doubts about the authenticity of hallucinatory phenomena, but Hibbert, for example, made it clear he believed there was an element of exaggeration in reports of various types of apparitions. ­Ferriar, in turn, had little to say about any divine agency behind hallucinations and limited his study to “profane” examples. Their approaches show a reluctance to address the possibly supernatural causes of the experience, but more than that they were exclusively focused on the role played by the human mind when it perceived an apparition or a hallucination. In other words, they favoured the scientific as opposed to the philosophical or theological approach. Relying on the best scientific models of his age, so did De Quincey. 3. The empiricists would have included Locke and his protopsychological view of the mind according to which the mind is furnished by experience alone. After receiving the necessary sensory input, the theory suggests, the mind could string together ideas into new configurations to create complex ideas of things that never existed—these could manifest themselves as sense perceptions in a disordered mind. Ludlow found a means of escaping the empiricist framework in the philosophy of Kant, Fichte, Hegel, and Schelling. For discussion on Ludlow’s Kantian idealism as a reaction against Lockean philosophy, see Kakko 2010. 4. Trip reports and other related writings by enthusiasts can be found online, for example on websites such as Erowid, the Lycaeum and the Shroomery. See bibliography for links. 5. Tellability normally refers to the value of the narrative, its noteworthiness, but a slight distinction should be made here. Labov and Waletzky (1967) noticed that in an interview situation subjects tended to emphasize certain phrases and expressions in order to make the point of their story more apparent. Knowledge of the structure of oral narratives was important, according to Labov and Waletzky, for the study of “the syntax and semantics of English below the ­sentence level” (ibid., 12). Subjects were prompted to speak about emotional events in their lives in order to generate authentic samples. Tellable stories were thus stories that passed the “so what” test in their judgments. The interaction between the interviewer and interviewee is not always this straightforward when an emotional response is not prompted for the specific purpose of gathering linguistic variables but, say, information. In such cases, as Gubrium and Holstein point out, “[i]t is not uncommon to hear respondents remark that they are not sure how they feel or what they think, or that they haven’t really thought about the question or topic before, or to hear them actually think out loud about what it might mean personally to convey particular sentiments or answer in a specific way and ask the interviewer for assistance in doing so” (2001, 36). Tellability may thus refer to the point of telling a story or the specific kind of story the interviewee assumes to be the proper kind of story for a given situation. The ­latter sense of the term is more applicable to Internet forums as

Narrative Conventions in Hallucinatory Narratives  203 it is reasonable to assume that those who post their experiences study previous posts in order to compose a fitting description. 6. Social constructionism was famously formulated by Berger and Luckmann in a 1966 work when McKenna would have been a college student. For an example of a discussion contemporaneous with McKenna’s True Hallucinations addressing the topic, see Sismondo (1993) and Cetina (1993). Since then, social constructionism has been distinguished from social constructivism, a term ­ favored by Sismondo and Cetina. In short, constructivism pertains to the cognitive processes of individuals whereas social constructionism argues, following Berger and Luckmann, that social practices and institutions produce knowledge in some fields of human endeavor. 7. On conventionalization see Fludernik 2012.

REFERENCES Abrams, M. H. [1934] 1971. The Milk of Paradise: The Effect of Opium Visions on the Works of DeQuincey, Crabbe, Francis, Thompson, and Coleridge. New York: Octagon Books. Berger, Peter L. and Thomas Luckmann. [1966] 1984. The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Harmondsworth: Penguin. Bruns, Gerald L. 1992. Hermeneutics: Ancient and Modern. New Haven: Yale U ­ niversity Press. Budge, Gavin. 2013. Romanticism, Medicine and the Natural Supernatural: ­Transcendent Vision and Bodily Spectres, 1789–1852. London: Palgrave Macmillan. Cetina, Karin Knorr. 1993. “Strong Constructivism—from a Sociologist’s Point of View: A Personal Addendum to Sismondo’s Paper.” Social Studies of Science 23.3: 555–63. Culler, Jonathan. 1975. Structuralist Poetics: Structuralism, Linguistics and the Study of Literature. London and Henley: Routledge & Kegan Paul. De Quincey, Thomas. 1985. Confessions of an English Opium-Eater. Oxford: Oxford University Press. Erowid: Documenting the Complex Relationship between Humans & ­Psychoactives. http://www.erowid.org/. Ferriar, John. 1813. An Essay Towards a Theory of Apparitions. London: Cadell and Davies. Fludernik, Monika. 2002. Towards a ‘Natural’ Narratology. London: Routledge. ———. 2012. “How Natural Is ‘Unnatural Narratology’; or, What Is Unnatural about Unnatural Narratology?”. Narrative 20.3: 357–70. Gubrium, Jaber F. and James A. Holstein. 2001. “From the Individual Interview to the Interview Society.” In Handbook of Interview Research, edited by Jaber F. Gubrium and James A. Holstein, 2–33. Thousand Oaks: SAGE Publications, Inc. Hibbert, Samuel. 1824. Sketches of the Philosophy of Apparitions; or, an Attempt to Trace Such Illusions to Their Physical Causes. London: Oliver & Boyd. Jay, Mike. 2000. Emperors of Dreams: Drugs in the Nineteenth Century. Sawtry: Dedalus. Kakko, Tommi. 2010. “Hallucinatory Terror: The World of the Hashish Eater.” In Cannabis–Philosophy for Everyone: What Were We Just Talking About?, edited by Dale Jacquette, 103–13. Malden: Wiley-Blackwell.

204  Tommi Kakko Labov, William, and Joshua Waletzky. 1967. “Narrative Analysis: Oral Versions of Personal Experience.” In Essays on the Verbal and Visual Arts, edited by June Helm, 12–44. Seattle: University of Washington Press. Ludlow, Fitz Hugh. 2006. The Hasheesh Eater: Being Passages from the Life of a Pythagorean. New Brunswick: Rutgers University Press. Lycæum Entheogenic Database & Community. http://www.lycaeum.org/. McKenna, Terence. 1994. True Hallucinations: Being an Account of the Author’s Extraordinary Adventures in the Devil’s Paradise. New York: HarperCollins. Shroomery: Magic Mushrooms Demystified. http://www.shroomery.org/. Sismondo, Sergio. 1993. “Some Social Constructions.” Social Studies of Science 23.3: 515–53. Taylor, Charles. 1989. Sources of the Self: The Making of Modern Identity. ­Cambridge: Cambridge University Press. ———. 2007. A Secular Age. Cambridge, MA: Belknap Press of Harvard University Press.

11 Narrative and Minds in the Traditional Ballads of Early Country Music Alan Palmer

A regrettable gap in narrative theory is its neglect of popular song lyrics. This essay is the first stage in a project intended to help to fill this gap. The project is an analysis of the narratives to be found in 50 country and western songs from the 1920s to the present day. It involves an examination of the constructions of their storyworlds, in particular the attributions of mental states to the minds of their characters, to see how these ­narratives developed over time. I chose 50 songs that seemed to have a high level of narrativity, but I did not use specific criteria because I did not want to prejudge the evidence. My conclusions, therefore, will be bottom-up and evidence led, rather than top-down and determined from the start by the choice of material. My preliminary overall hypothesis is that attributions are sketchy, attenuated, fragmented, and elusive in early country songs but are self-consciously fuller, more detailed, and more coherent in modern and contemporary country music. This essay will focus on the first two of the six groups that make up the corpus: six traditional ballads of British and Irish origin that are still sung within the American country music scene and eight ballads of American origin, most of which arose out of actual events in American history. My initial hypothesis was that the first group would reveal a stable, settled, and fairly uniform pattern of characteristics associated with traditional ballads, while the second would be much more varied. I thought that some of the American ballads would stay close to the British model and be rather randomly thrown together in the ways that I will describe below, while others would be more carefully crafted and show signs of the initial development of the country song form that became commercially popular from the 1930s onward. As is so often the case with this sort of exercise, my unsurprising conclusion, having now studied the evidence, is that it’s a little more complicated. There are some similarities and some differences between the two groups, and it’s not always obvious why. Certainly, the second group has more in common with the first than I thought when I began this exercise. After a brief discussion of some of the general characteristics of ballads, I will consider the following narratological and cognitive aspects of the 14 song lyrics: their heterodiegetic (or third-person) and homodiegetic (or firstperson) narrators; the attribution of reasons for actions and mental states

206  Alan Palmer generally—by narrators, characters to themselves, and characters to other characters; and the presentation of emotions. Ballads are highly conventional works, and listeners will often be able to ascribe a wide range of ­mental states to characters in specific situations simply from their knowledge of the genre as a whole. What I am primarily looking at in this essay are those occasions on which motives for actions and emotions are explicitly presented in the words of the songs. FICTIONAL MINDS In Fictional Minds (2004), I outlined a theory for the study of the novel; it is, I think, equally applicable to the narratives contained in country songs. I argued that, in order to understand a novel, we have to try to follow the mental functioning of the characters operating within its storyworld. The constructions of the minds of fictional characters by narrators and readers are central to our understanding of how novels work, because readers enter storyworlds primarily by attempting to follow the workings of the fictional minds contained in them. Fictional narrative is, in essence, the presentation of mental functioning. These storyworlds are aspectual. As the philosopher John Searle explains, “Whenever we perceive anything or think about anything, we always do it under some aspects and not others” (1992, 156–57), and this is equally true of fictional characters. Like real people, characters experience the same events in different ways. A key tool for analysing the process of recovering and reassembling ­fictional storyworlds is the application of attribution theory: the study of how we ascribe states of mind to others and also to ourselves. The ability that we have to infer the mental processes of others from their behaviour is ­ ctional often referred to as theory of mind (Zunshine 2006). In relation to fi minds, attribution theory can be used to formulate tentative answers to questions such as these: how do readers attribute states of mind such as emotions, dispositions, and reasons for action to characters? How do heterodiegetic and homodiegetic narrators attribute states of mind to their characters? How do characters attribute mental states to themselves and to other characters? And, finally, with regard to the issue of characterization, how does an attribution of a mental state help to build up in the reader a sense of the whole personality of that character? Extending these ideas to song lyrics, I propose that it is only possible to understand a narrative song by following the mental functioning of the ­narrator and the other characters who inhabit the storyworld created by that lyric. In the examples discussed below, we understand the song by ­following what the narrator is telling us about the working of the minds of the characters in the song. I have found it extremely helpful to apply some narratological tools to popular song lyrics. I unearthed some difficult cases involving some

Narrative and Minds in the Traditional Ballads of Early Country Music   207 uncertainties and ambiguities, as detailed below, but they are not evidence that the application of the tools is a failure. On the contrary, I feel that bringing narrative theory to song lyrics has fulfilled an important and valuable role in unearthing these interesting and significant complexities. THE FOURTEEN SONGS Traditional ballads of British Isles origin: 1 “The House Carpenter” - Clarence Ashley (1930) 2 “Barbara Allen” - Hedy West (1965) 3 “Willie Taylor” - Uncle Earl (2005) 4 “Fatal Flower Garden” - Nelstone’s Hawaiians (1929) 5 “The Butcher’s Boy” - Buell Kazee (1928) 6 “The Wagoner’s Lad” - Buell Kazee (1928) Traditional ballads of U.S. origin: 7 “Peg and Awl” - The Carolina Tar Heels (1928) 8 “The Streets of Laredo” - Marty Robbins (1960) 9 “Ommie Wise” - GB Grayson (1927) 10 “John Hardy Was a Desperate Little Man” - The Carter Family (1930) 11 “John Henry” - JE Mainer’s Mountaineers (1936) 12 “White House Blues” - Charlie Poole and the North Carolina Ramblers (1926) 13 “Frankie and Johnny” - Jimmie Rodgers (1929) 14 “The Lawson Family Tragedy” - The Blue Sky Boys (1976) Although most of these 14 ballads exist in many variants and have been recorded by a wide range of artists, I’ve chosen to examine the lyrics of the versions that happen to be in my record collection and with which I am most familiar. The choices are certainly not definitive in any way. I have supplied the names of the artists and the recording dates for anyone who wishes to track them down. It may be helpful for readers who are not familiar with these songs to be given a brief indication of their subject matter. In “The House Carpenter,” an ex-lover tempts a woman away from her husband (the house carpenter of the title) and her children. They board a ship that then sinks, causing them to drown. Barbara Allen, after going to the bedside of sweet William, who is dying for love of her, goes home to die too. After Willie Taylor is press-ganged into the Navy, his fiancée dresses as a man to board a ship to find him. After her secret is discovered and her captain tells her of Willie’s new lover, she shoots them both and is rewarded with command of a ship. In “Fatal Flower Garden,” a boy is playing with his friends. When his ball goes into the garden of a gypsy lady, she tempts him in and murders him, although the actual murder is not described. “The Butcher’s Boy” is about

208  Alan Palmer a young woman who discovers that her lover is unfaithful to her and goes home to hang herself. “The Wagoner’s Lad” features a dialogue between lovers in which the young man is intending to leave because the young ­woman’s parents disapprove of him. She tries unsuccessfully to persuade him to remain. “Peg and Awl” is different from the others because it is about work, not love and/or death. A shoemaker is put out of work by machines that can make one hundred pairs of shoes in the time it takes him to make one. In “The Streets of Laredo,” the narrator encounters a dying cowboy who knows that he has done wrong. It is derived from British ballads such as “The Unfortunate Rake” and “The Bard of Armagh.” “Ommie Wise” is a standard murder ballad. Other examples include “The Banks of the Ohio” and “The Knoxville Girl.” It is based on a real case in which nineteen-yearold Naomi Wise was murdered in 1808 by her lover, John Lewis, who later escaped from jail. “John Hardy Was a Desperate Little Man” is also based on a historical event. John Hardy murdered a man in a game of cards, stood trial, and was executed in 1894. The historical background to “John Henry” is not as clear cut. It is likely, but no more than that, that there was a real John Henry who worked in the 1870s as a steel driver, hammering a steel drill into rock to make holes for explosives. In legend, he died after winning a race against a steam-powered hammer. “White House Blues” is about the assassination of President William McKinley in Buffalo, NY, in 1901. “Frankie and Johnny” has a similar story to “Willie Taylor.” In 1899, Frankie Baker shot Allen (also known as Albert) Britt after discovering him with Nelly Bly. Some variants of the song are called “Frankie and Albert.” “The Lawson Family Tragedy” is about a North Carolina farmer, Charlie Lawson, who, on Christmas Day, 1929, murdered his wife and children and then killed himself. BALLADS Performances of traditional ballads such as the 14 listed above are almost always variants chosen from among a large number within a particular song-group. Variants within a song-group are often known by completely different titles. The music theorists Mary-Ann Constantine and Gerald Porter (2003) show that “flux and interchangeability of stanzas” is very common: “floating verses, usually couplets or a single stanza, are the very essence” of the ballad (Constantine and Porter 2003, 119). They say of such song-groups that they often “have locatable roots in real events” but that they take us “into a territory of truly minimal narrative, evoked in an extremely indirect style” (2003, 115). According to Constantine and Porter, many ballads “are only incomplete by outside standards: to the singers and informed listeners they are perfectly sufficient” (2003, 3–4). This is because, “In ballads especially, ‘meaning’ (in contrast to a simple plot line) does not

Narrative and Minds in the Traditional Ballads of Early Country Music   209 reside primarily in a text, or in a simple performance, but rather in the minds of those in a community in which ballad performances take place” (2003, ix my emphasis). In these circumstances, it is not possible to talk of the purpose of the song as it emerged from its historical context. Composed songs are works of art that are created to inspire particular aesthetic effects in the listener, and so the construction of the minds contained in their narratives is ­relatively straightforward, despite the complexity of those effects. It is usually fairly clear how composers and performers wish the listener to receive the stories in their songs. With ballads, though, the only agency involved is the artist’s decision to record one particular version of the song and his or her treatment of it. Ballads of this sort resemble the found object such as a stone or piece of wood with which the British sculptor Henry Moore filled his studio and which, in many cases, became indistinguishable from his actual sculptures. This historical process can make mind construction by the listener very difficult. Any variant chosen from within a song-group is a fairly arbitrary selection, usually of four to eight verses, taken from a pool of up to 100 or so. It is likely that the cuts involved in the selection process will result in a jerky, gappy listening experience. Descriptions of important actions will often be omitted. Crucial to my purpose, even when those actions are described, the reasons for them will be left out. Nevertheless, it is eminently possible to treat the whole performance as an aesthetic object. The results are often elliptical, as will be illustrated below, but, as Constantine and Porter point out, in words that beautifully encapsulate the sensation of listening to a ballad, “new narrative meaning can leap across the spaces between stanzas like electricity between two points” (2003, 118). In fact, all of the 14 songs are satisfying aesthetic wholes. It is astonishing that they are able to create such vivid fictional minds despite being such tiny stories consisting of only a very few words and lasting less than four minutes in every case. Within the fluid, protean world of the traditional ballad, it is hardly ­surprising that there are several factors that complicate this comparative exercise. First, as already stated, there are nearly always minor, and sometimes substantial, differences between versions. An interesting example of minor differences is a comparison between “Fatal Flower Garden” and a variant that appears in the “Ithaca” episode of James Joyce’s Ulysses (1986, 565–67). In Joyce’s version, usually called “Sir Hugh” though not referred to as such in the novel, the boy is not anonymous but is named (Harry Hughes); the murderess is a “Jew’s daughter” not a gypsy lady; she is dressed only in green not in yellow and green; and the actual murder is described (she cuts off his head with a penknife). Similarities include a nearly-identical line (the chilling “Where no one could hear him call” and “Where none could hear him call”) and the fact that no reason for the murder is given.

210  Alan Palmer Second, I will not be referring to the tunes that accompany the lyrics because variants are often set to different melodies. Referring to the ones I happen to know would add more randomness and contingency to the exercise, and it would not be possible to consider as many as 14 songs in this way. Nevertheless, I urge anyone whose curiosity is stimulated by this essay to listen to some of these songs on such websites as YouTube. Third, the distinction between ballads of British origin and those of American origin is not a hard and fast one and should be treated with caution. All of the songs in the first group have been assimilated into the American tradition and have underdone changes on the way. “Sir Hugh” is just one example. Also, although several of the songs in the second group are clearly American in origin because they are based on real events that happened in the United States, some, such as “Frankie and Johnnie” and “The Streets of Laredo” (e.g., “Oh, beat the drum slowly and play the fife lowly”), incorporate elements from earlier ballads. Finally, ballads are regarded as anonymous because it makes no sense to try to assign a composer to a variant that exists within a large amorphous song-group. This certainly applies to all of the British ballads. However, several of the American ballads must be of comparatively recent origin because of the historical events described in them. It has always seemed quite likely that some of them originated with a single composer, but the identity of that person is not now known. I would put “The Streets of Laredo,” “Frankie and Johnnie,” and “The Lawson Family Tragedy” into this category, but there is not enough space here to justify this assertion. Even if I am right, there are two further complications: some performers (such as A.P. Carter of the Carter Family) routinely claimed to be the composers of traditional ballads; and many of the songs that might have been written by one person were taken up and modified by others, creating variants. The picture is a complex one.
 My working hypothesis was that attribution in traditional, old-world ballads was sketchy, attenuated, and fragmented but became a ­little fuller and more rounded in the new-world ballads because some of them appear to have been composed in response to an actual event. It seemed likely that composed songs would tend to be more coherent in their reasons for characters’ actions. I will now discuss the extent to which my initial assumptions were justified. NARRATION The amount of direct speech in the 14 songs is an important context for the issues relating to both narration and attribution. In summary, there is a good deal of direct speech in the first group and less, though still a lot, in the ­second. The amounts of direct speech as percentages of the total lyric in the first group are as follows: 75%, 20%, 30%, 40%, 65%, and 80%. The aggregate percentage for the six songs is 50%. The equivalent percentages

Narrative and Minds in the Traditional Ballads of Early Country Music   211 for the eight songs in the second group are as follows: 0%, 65%, 20%, 30%, 40%, 30%, 55%, and 20%. The aggregate percentage is 33%. Although there is a good deal of variation within both groups, there is a clear trend toward a reduction in use. In the first group, only two of the six are below 40% and three are above 60%; in the second group, only two of the eight are above 40% and five are 30% or below. The songs in the first group all use direct speech but contain no indirect speech. Things loosen up a little in the second group. One song (“Peg and Awl”) has no speech of any sort, and two feature a small amount of indirect speech. Surprisingly, both examples of indirect speech conclude with a line of free indirect discourse: “He told her to meet him at Adams’s spring / He’d bring her some money and some other fine things” (“Ommie Wise”); and “[Frankie and Johnny] Swore to be true to each other, true as the stars above / He was her man, he wouldn’t do her wrong.” A strikingly large number of songs, two in the first group (2 and 4) and five in the second (8, 10, 12, 13, and 14), include statements made by characters as they are dying, and in one, “The Butcher’s Boy,” there is a suicide note. I talk more about death later. All six of the British-origin ballads are narrated heterodiegetically. ­However, four of them are interestingly marginal or problematic in various ways as detailed below. In the other group, two of the eight narrators are homodiegetic, and only one of the remaining six heterodiegetic narrators can be classified as problematic. However, although there is more heterodiegetic narration in the first group than in the second, this difference is reduced to a certain extent in its effect by the large amount of direct speech, which acts as a kind of surrogate homodiegetic narration. I develop this point below. Two of the heterodiegetic narrators, one in each group, refer to themselves: “I’m sure it wasn’t three” (“The House Carpenter”); and, in an opening rhetorical flourish that is quite common in popular songs, the narrator says that “I” will tell a story about Ommie Wise. There are no narratees in the first group and only one in the second: in “White House Blues,” the narrator twice addresses McKinley’s children. The lyrics of all of the 14 ballads are, in general terms, examples of external focalization. The narratives are behaviourist—there is no sustained access to characters’ consciousnesses through which the events of the storyworld are perceived. Within this broad context, there are a few occasions on which the narrative is focalized through a particular character for very brief periods—a couple of lines or so—and I will deal with these below in the context of attributions of mental states. It might be helpful if I briefly discuss here the cases of homodiegetic narration that I regard as marginal. The Clarence Ashley version of “The House Carpenter” is structured as follows: 1 Direct speech (man) 2 Direct speech (woman) 3 Direct speech (man)

212  Alan Palmer 4 5 6 7 8 9 10

Direct speech (man) Direct speech (man) Heterodiegetic narration Heterodiegetic narration Direct speech (man) Direct speech (woman) Heterodiegetic narration

The song has a heterodiegetic narrator, but there is so much dialogue between the lovers that the narrator is effaced and the lyrics have a strongly homodiegetic flavour. This is especially so as the first five verses consist of dialogue and the heterodiegetic narration does not begin until the sixth verse. On the other hand, the first line ends with a tag (“said an old true love”) that alerts the listener to the existence of the narrator. The narrator of “The Wagoner’s Lad” is a curious blend of the intrusive and the completely effaced. The song begins with an extraordinary ­feminist manifesto by the heterodiegetic narrator, which is not at all what one might expect from a traditional ballad: “Lord, hard is the fortune of all womankind / They’re always controlled, they’re always confined.” The verse then specifies that women are first controlled by their parents until they are married and then by their husbands for the rest of their lives. However, the remainder of the song consists of totally untagged dialogue between the lovers with no heterodiegetic narration of any kind, during which it can sometimes be difficult to tell who is speaking. “Barbara Allen” and “The Fatal Flower Garden” are marginal cases of a different sort. I have classified them as heterodiegetic, even though both contain a trace of a narrator-character. The first line of the former is: “In Scarlet town where I was born,” although this “I” is not heard from again, and the rest of the song is clearly heterodiegetic in nature. The third line of the latter refers to “all the boys in our school,” but there is no other sighting of this “our” in the rest of the lyrics. The effect of both is rather like that created at the beginning of Madame Bovary (1857), which starts with what appears to be a homodiegetic narrator who soon disappears from the narrative. The single problematic case in the second group is “White House Blues.” It is structured as follows: 1 2 3 4 5 6 7 8 9

Heterodiegetic narration (including direct speech - doctor to McKinley) Heterodiegetic narration Homodiegetic narration (address to narratees - McKinley’s children) Heterodiegetic narration Direct speech (McKinley) Homodiegetic narration (address to narratees - McKinley’s children) Homodiegetic narration Homodiegetic narration Homodiegetic narration

Narrative and Minds in the Traditional Ballads of Early Country Music   213 10 Direct speech (McKinley’s wife addressing killer) 11 Heterodiegetic narration (including direct speech - doctor to horse) 12 Heterodiegetic narration (including direct speech - doctor to McKinley) Verses one, two, and four are straightforward heterodiegetic narration, but in the third and sixth verses the narrator appears to step into the storyworld in order to address McKinley’s children. Verse five consists of McKinley’s untagged direct speech (one of the pre-death statements referred to earlier). Then there are three homodiegetic verses that show the narrator within the storyworld again, this time, for some reason, racing his horse against a train. They are followed by another verse (ten) of untagged direct speech in which McKinley’s wife addresses his killer. Verses eleven and twelve return us to heterodiegetic narration. As you can imagine, it’s a bewildering mixture, and the result is both chaotic and exhilarating. The jolly, swaggering performance by Charlie Poole, who sounds ancient but is in fact only 34 at the time of the recording, is well worth listening to on YouTube. REASONS FOR ACTIONS Characters in the songs under consideration show little evidence of successful mind reading. Only a few ever attempt to find out what other characters are thinking, and they tend to be unsuccessful. The lover in “The House Carpenter” tries to find out why the woman is weeping, but every guess is wrong. In “The Butcher’s Boy,” the mother and the father can see that their daughter is troubled, but they do not know why. The only successful attempt to follow the thought processes of another is the rather obvious and straightforward case of the bartender in “Frankie and Johnnie” who knows that it will cause Frankie trouble to hear that Johnny is being unfaithful to her. The attribution of mental states by heterodiegetic narrators to characters and by characters to themselves in the 14 songs generally relate to reasons for behaviour. There is very little interest in mental states for their own sake. These are portraits of minds in action. I will discuss, within both groups of songs, first those that display what I call an attributional hole at the centre of the lyrics and then those for which the causes of important actions are clear. It is commonly said that actions in traditional ballads are entirely unmotivated. I have found this not to be the case. Nearly all of them do contain reasons for actions. The opposite impression arises because the cause of the main or central act, the subject of the song such as a murder or an elopement, is often not explained. This is why I use the term attributional hole. Most of the actions in “The House Carpenter,” such as the return of the man, his decision not to marry a king’s daughter, and the elopement, are fully and explicitly motivated through self-attribution. A spoken declaration of love (“And it’s all for the love of thee”) is given twice as a reason for the actions. There is then an explicit discussion about the causes of behaviour.

214  Alan Palmer Seeing the woman “weeping bitterly” after she has left her family, her lover suggests three possible reasons: she may be weeping for the loss of his money, or for the loss of his “store” (presumably his possessions), or for her husband, the house carpenter. In reply, she gives a fourth alternative, which is the real reason: the loss of her children who she will not see again. However, in attributional terms, there is a large gap in the middle of the song. There is no record of her decision to flee from her family with her lover and therefore of the states of mind that caused her to take that decision. One verse contains his invitation to go with him; the next verse records her farewell to her children. Compare this with the three whole verses devoted to a careful motivational analysis of her subsequent weeping. In “Barbara Allen” there is also some general causal attribution. The ­narrator specifies that the reason for Sweet William’s death is love of Barbara Allen, that men avoid her and that she hears a reproach in the tolling of the death bell because every stroke seems to say “hardhearted Barbara Allen.” (Unusually, the whole of the verse regarding the bell is focalized though Barbara Allen herself.) It is strongly implied, but not explicitly stated, that her hardheartedness is the cause of all the events described in the song. As with “The House Carpenter,” some actions are motivated, but there is still an attributional hole in the middle of the song. She tells her mother of her decision to die (“Sweet William died for me today / I’ll die for him ­tomorrow”), but she does not state the reason for that decision. Her words belong to the logic of the typical ballad storyworld but not to the real world. Does she love William or not? Is guilt the reason? We are not told. The heterodiegetic narrator begins “Willie Taylor” with an attribution that is, unusually, non-causal and so not related to motives or intentions: the couple is described as “Full of mirth and loyalty.” Self-attributions give her reasons for going to sea and dressing as a man: to search for Willie Taylor. Otherwise, as with the previous songs, the narrative is behaviourist at its core. We are not told the heroine’s motive for killing Willie Taylor and his new lover, although the listener will, of course, assume that it is jealousy. It is less easy to understand the reasons for the captain’s rather unlikely action of making her a ship’s commander. Perhaps it is an expression of his admiration for her decisiveness! “The Fatal Flower Garden” is of a similar pattern to the other songs. The attributional hole at its centre, this time relating to the murder of the child, is again surrounded by explanations of less important actions. Five verses are devoted to the reasons, given by the narrator, the boys go out (to talk and play), why they are tempted to go into the flower garden (to retrieve their ball), why they initially decide not to (they are not allowed to), and why one boy changes his mind (he is tempted not by the offer of fruit, but by a diamond). But then no reason is given for the most important action: the murder. This song is an early example of the theme of the apparently motiveless murder that is popularly associated with the traditional ballad generally and the American ballad in particular (e.g., “Ommie Wise,” “John

Narrative and Minds in the Traditional Ballads of Early Country Music   215 Hardy,” and “The Lawson Family Tragedy”). Although every song so far has had a silence at its heart, causation is often clearly implied by the context. “The Fatal Flower Garden” is far more mysterious than the others. In contrast, the remaining two songs in the first group, “The Butcher’s Boy” and “The Wagoner’s Lad,” are both extremely clear about motivation. In the former, the narrator attributes to the father of the girl the intention “to give her hope.” The girl self-attributes the reason for her depression. She says that the railroad boy who courted her has gone to London. There, in one of the most moving couplets that I have ever heard, “He takes that strange girl on his knee / And he tells to her what he won’t tell me.” In her suicide note she specifies that the purpose of the memorials that she has asked for is to warn the world that she died for love. In the latter song, the causation is also stated throughout. The man has made a decision to go away, and the reason that he gives for his intended action is that her parents don’t like him because he is poor. The girl then unsuccessfully gives her reasons for wanting him to stay. Within the second group, there are four songs that I would regard as having an attributional hole at their centre. “Ommie Wise” is rather similar to the British ballads in that we are not told John Lewis’s reasons for murdering Ommie. The narrator says that John Lewis is a liar, that she was deluded by him, and that he concluded to tell her his mind and to leave her behind. However, there is no information about why he came to this conclusion. “John Hardy” is unusual in having no reasons for actions at all: there is no explanation for why he shot the man, what he thinks about it having done it, how he feels about the consequences of his actions, and so on. Although there are some attributions (he affirms his love for his wife, and his two daughters express their love for him), none of these are mental states that, within the song, relate to the causes of behaviour. The origins of the next two songs as variants within large song-groups are obvious. Versions of “John Henry” typically contain the following e­ lements: his premonition as a child that steel-driving will lead to his death, his race against a steam hammer, his death and burial, and his wife’s reaction to these events. My version is light on the third element because, in it, he simply gets sick and goes home, rather than dies and is buried. It is appropriate for a song of this sort that the narration is completely behaviouristic. There are no attributions of mental states of any kind. There are no murders and no love element, and so the need to consider motivation in any depth is not so pressing. Such songs as these are fairly conventional exercises in which the meaning of the words is not central to the listening experience. They can sometimes develop nonsense lyrics, such as those in the well-known example of “Cotton Eyed Joe.” Less obviously, there are no real reasons for actions in “White House Blues” either. As shown above, it is very casually put together with a good deal of repetition and several non-sequiturs. The only two fleeting attributions are the narrator’s reporting that Roosevelt (McKinley’s successor) is doing his best and McKinley himself saying that the only thing that grieves him about dying is that he is leaving his poor wife behind.

216  Alan Palmer The other four songs within the second group generally feature characters who have a consistent and easily understood mental life. In “Peg and Awl,” the causation is simple and explicit: there is no need for the narrator, a shoemaker, to continue to make shoes because of the new technology. In “Streets of Laredo” there are self-attributions by the homodiegetic narrator (“We bitterly wept for we loved the cowboy”), and also by the cowboy himself: he states that he used to be “dashing” and “gay,” he reveals that his is a “sad” story, he admits that he’s done wrong, and he tells of his sorrow at his own death. It is significant that the cowboy repeatedly says that he knows that he’s done wrong because this is the first time that ethical issues have arisen in such an explicit way. Although he doesn’t say what he has done that was so wrong, the ballad-listening community (in effect, the implied listener) will know that many of the earlier British variants within this song-group imply that the protagonist dies of syphilis. “Frankie and Johnnie” is a song in which causation is closer to everyday life than in the earlier ballads. In this respect, it makes an interesting comparison with “Willie Taylor,” as their stories are so similar. In the earlier song, no explanation is given for the act of murder, although it can easily be inferred, of course, from the context. By contrast, the later song makes the motive for the murder explicit. The lyrics emphasise that she shot Johnny because he was doing her wrong. The warden then explains that she’s going to the electric chair because she shot her man. Even simple actions such as Frankie leaving her house are motivated: it’s to buy beer. Like “Streets of Laredo,” there is a moral, albeit given in a rather contradictory way. The final verse states that the story has no moral, but then goes on to supply one anyway: there isn’t any good in men. It may seem perverse to place “The Lawson Family Tragedy” here when its narrator explicitly draws attention to the impossibility of causal attribution in relation to the central motive of the song—why he killed his family: “But we’ll never know what caused him / To take his family’s life.” ­However, I would argue that it does belong here for that reason. As the murders occurred on Christmas Day, 1929, the song must have been written in the early 1930s, and it seems likely that, by then, audiences would expect more information about causation than the earlier ballad-listening community did. I think this is why the narrator would feel it necessary to draw attention to the causal absence in this case. Such a statement would certainly sound completely out of place in the traditional ballads. EMOTIONS Both groups contain songs that express very little emotion and certainly much less than might be expected from the tragic and violent events described in them. Equally, both groups have songs that are extremely emotive.

Narrative and Minds in the Traditional Ballads of Early Country Music   217 The narrator of “Barbara Allen” notes that her words to William while at his deathbed—she simply says that she thinks that he’s dying—are the only words that she says to him. It is not an obvious reaction, and more would normally be expected. Her own death is marked by a similar reticence. After she tells her mother that, as William died for her today, she’ll die for him tomorrow, the next verse records only the fact of her death. Similarly, “Willie Taylor” and “The Fatal Flower Garden” are notable for a complete lack of emotional involvement. In the former song, the opportunities to express the young woman’s grief and sense of loss when her fiancé is press-ganged, and her anger and jealousy when she sees him with another lover, are not taken. However, the emotional coldness that characterizes some songs in the first group should not be overemphasised at the expense of the emotions present in the others. In “The Butcher’s Boy,” feelings are more developed and recognizable than in the songs just mentioned. The deep concern of both the mother and the father for their daughter is made explicit. Her suicidal depression is more easily accessible to a modern audience than say, Barbara Allen’s calm decision to die, because her reasons for being troubled are so well-specified. Her resentment that her lover puts another girl on his knee is precisely the sort of detail that is missing from the earlier songs. In “The Wagoner’s Lad,” the combination of the opening statement, the girl’s initial declaration, the parents’ hostility, and the man’s intransigence, combined with the singer’s heartfelt delivery, also makes this an emotive song. In “The House Carpenter,” the woman kissing her child during her farewell and her subsequent bitter weeping are indicators of great emotion. The displays of emotion in the songs in the second group are also patchy and erratic. “Ommie Wise” is definitely an American ballad in the restrained British style in this respect. The responses to the murder are seriously deficient by everyday standards. There are no real expressions of feeling in “John Henry” either. It may seem as though emotions are more prevalent in “Frankie and Johnnie,” and I must confess that I had assumed so when I began writing this section of the essay. However, a close examination of the words shows that this is an illusion, created, perhaps, by the increased interest in causation noted in the previous section. In “Peg and Awl,” the narrator is fatalistic and accepting of the technological change. Contrary to what I think would be conventional expectations, it is not an agonized cri de coeur. In fact, because pegging shoes “ain’t no fun,” he has quite positive feelings about the new machine, saying that it is the “Prettiest little thing you’ve ever seen” and that it has set him free. In contrast, two of the American songs show substantially increased levels of emotional display. “The Streets of Laredo” is marked by expressions of deep emotion on the part of the dying cowboy. The narrator weeps bitterly because he loved the cowboy. The cowboy himself refers to his “sad” story, and he tells of his sorrow at his own death. This seems a rather obvious thing to do, but some of the characters in the early ballads do not do so. In “The Lawson Family Tragedy,” the emotions are laid on with a trowel. It is, in fact,

218  Alan Palmer mawkish and sentimental. The emotiveness apparent in this song is a long way from the restraint of the older songs, and an indication of the increased level of sentiment to come in the country songs of the 1930s and 1940s. One noticeable difference between British and American ballads is their treatment of death. The incidence of murder varies markedly between them. Although five of the six British ballads feature deaths, only two are murders. Six of the eight American ballads are about death, but every one of them is a murder. A remarkable statistic. In three of the six the murderer is gaoled, and in two he is then executed. Also, a jaunty attitude to death is a striking characteristic of several of the American ballads. For example, the approach to death in “John Henry” is remarkably casual. “John Hardy” is a song of murder and then execution that is disturbingly callous in tone. “White House Blues” adopts an astonishingly brutal and flippant attitude toward the assassination of the nation’s President. It treats McKinley’s death in a way that is almost comic: “McKinley hollered, McKinley squalled / Doc said to McKinley ‘I can’t find that ball.’” And the rest of the song continues in a similarly sprightly vein. In his book, Long Steel Rail, Norm Cohen reports how a friend of D.H. Lawrence recalled that, in 1915, Lawrence sang a variant of this ballad to her. He “set our brains jingling with an American ballad on the murder of President McKinley with words of brutal jocularity sung to an air of lilting sweetness” (2000, 417). That’s a very good description of the Charlie Poole version, too. CONCLUSION It may be helpful to summarise the most important of the similarities and differences between the two groups. First, the differences. The events in the first group are imaginary or of indeterminate origin, while many in the ­second actually occurred. Some of the songs in the second group may have been composed by an individual. There is a good deal of direct speech in the first group and less, though still a substantial amount, in the second. There is no indirect speech in the first group but a little, including some free indirect discourse, in the second. All of the narrators in the first group are heterodiegetic; two of the eight in the second are homodiegetic. I have classified four of the narrative situations in the first group as problematic or marginal, but only one in the second group. There is the beginning of a consideration of ethical considerations and also a more strongly marked emphasis on murder in the second group. However, with one exception, there will be a consistent trend in the future away from death generally and murder in particular as a subject for country songs. The exception is a renewed interest in murder in the 1950s, caused by the fashion in that decade for melodramatic narrative pseudo-ballads such as the well-known “The Long Black Veil.” Such formal issues as heterodiegetic versus homodiegetic narration, ­focalization, and the narratee acquire great significance later in the development

Narrative and Minds in the Traditional Ballads of Early Country Music   219 of the country song. There is a marked shift in emphasis in future decades toward homodiegetic narration. In fact, of the remaining 36 songs in my corpus of 50 that are not discussed in this essay, only eight are heterodiegetic. Fourteen of those 28 homodiegetic songs feature narratees, seven of whom I would define as structural narratees. That is, they are absolutely essential to the song and part of the purpose of performing it. Such songs are performative speech acts that are intended to elicit a certain response from the narrate, such as persuading a spouse or partner to excuse an intended or actual infidelity. Finally, as the form of the popular song takes shape, and focalization becomes noticeably more internal, the narrative situation becomes much less marginal or problematic. I have classified only four of the remaining 36 songs as such. The three major similarities were rather surprising to me. One is that the amount of attribution of mental states in the two groups is roughly equal. Another, more specifically, is that in both cases, several songs have what I call an attributional hole at their heart (four out of six and four out of eight, respectively). Finally, the proportion of songs in both groups that strike me as emotionally cold and surprisingly restrained is exactly half in both cases. At the beginning of this exercise I asked myself these questions: are there signs in the American ballads of the country songs to come? What features of the newer ballads foreshadow the complexity, sophistication, and depth of the country song from the 1960s onwards? Is there a sense of popular music taking shape and going in different directions? My answer is that there is much less sign of all of this than I thought there would be. The development of the American ballad into the modern country song was more gradual than I had expected. On reflection, though, it is not surprising that changes were slow and organic in nature as the American ballad tradition grew directly out of the British one. Indeed, the undoubtedly composed country songs of the late 1920s and 1930s were still heavily influenced by the tradition from which they arose. The moral is that one should not underestimate the importance of the role that inertia, the persistence of traditional customs and practices over time, plays in the development of popular music. Nevertheless, I did find that the second group of songs is more varied than the first. Some of the American ballads do stay close to the British model and are rather randomly thrown together, but others are more carefully crafted and show faint signs of the initial development of the country song form that became commercially popular between the 1930s and the 1950s. I have in mind, in addition to the formal differences listed above, the approach to mental causation in “Frankie and Johnny,” the self-consciously emotive quality of “The Streets of Laredo,” and the prominence of both of these characteristics in “The Lawson Family Tragedy.” A rather odd impression occurred to me during the writing of this essay. It was that the history of narrative in country music mirrors the history of the novel. That is to say, the ballads (both British and American in ­origin) are basically behaviourist—attributions of mental states are fairly rudimentary and can usually be easily inferred from the context. They are

220  Alan Palmer therefore rather similar to pre-rise-of-the-novel prose narratives. Composed songs from the 1920s to the 1950s show a steady increase in depth that can plausibly be compared to the maturation of the novel in the eighteenth century. In the 1960s, country music began to change, and this change corresponds to the novel of the nineteenth century to the present day. The individuals portrayed in these songs have intense and conflicted feelings about complex and ­difficult moral, social, and practical dilemmas, and the relationships in them are explored with an insight and sensitivity that is, in my view, unmatched by other popular music. REFERENCES Cohen, Norm. 2000. Long Steel Rail: The Railroad in American Folksong. Second edition. Urbana-Champaign: University of Illinois Press. Constantine, Mary-Ann, and Gerald Porter. 2003. Fragments and Meaning in ­Traditional Song: From the Blues to the Baltic. Oxford: Oxford University Press. Joyce, James. [1922] 1986. Ulysses. Edited by Hans Walter Gabler. Harmondsworth: Penguin, Palmer, Alan. 2004. Fictional Minds. Lincoln: University of Nebraska Press. Searle, John R. 1992. The Rediscovery of the Mind. Cambridge, MA.: MIT Press. Zunshine, Lisa. 2006. Why We Read Fiction: Theory of Mind and the Novel. ­Columbus: Ohio State University Press.

Section IV

This page intentionally left blank

12 Mind Reading, Mind Guessing, or Mental-State Attribution? The Puzzle of John Burnside’s A Summer of Drowning Matti Hyvärinen John Burnside’s A Summer of Drowning (2011) is a perplexing novel with perplexing representations of minds. In fiction, third-person narrators liberally exhibit thoughts, feelings, and minds of all characters, yet in Burnside’s novel, it is the first-person narrator Liv, a teenage girl who steps forward as a seasoned reader of minds. However, the novel is not a strong case for the existence of such generalized and benevolent (“folk-psychological”) capacity to read other minds that the supporters of “Theory of Mind” seem often to suggest (see Zunshine 2006; Palmer 2004). There is the unnerving problem that Liv seems to read minds too perfectly, even the minds of people she has hardly met, while failing drastically to understand the two people she knows and loves most—her mother and her surrogate father and neighbor Kyrre Opdahl. Daniel Hutto (2008, 46) suggests that “by far the best and most reliable” way of attaining knowledge about other people’s reasons is the horse’s mouth method, yet precisely this basic folk-psychological approach of conversation is interpreted by Liv as “intrusion.” To add confusion and complexity, the more unnatural the events of the novel become, the less reliable and the more paranoid the narration becomes. As an attempt at fostering some clarity, I develop an idea of “mind reading”—or more exactly mind-attribution—as a discursive-mental strategy in the service of diminishing the value of present partners of social interaction. The story takes place on the distant Norwegian island of Kvaløya, which is situated near Tromsø and the Nordic Circle, during a short summer. Liv’s mother is a prominent artist who achieved her position after withdrawing from Oslo to the solitude of the island. The teenage girl adores her mother as much as she despises all the banality of life in Tromsø and, in general, everywhere outside her home and the island. The story begins just after Liv has finished school and is considering her future. Two brothers, Mats and Harald Sigfridsson, both from the same school, about the same age as Liv, are found mystically drowned during calm weather, without any plausible explanation or even a reason for rowing a stolen boat at night. A British tourist, Martin Crosbie, comes to hire a nearby cottage, only to disappear later without a trace. Liv’s only friend, Kyrre Opdahl, has a history of telling Nordic folklore stories to Liv; he suggests now that huldra, an infamous magical female character of folklore fame, who is able to enchant gullible

224  Matti Hyvärinen men and drown them, is behind the tragedies. Even though Kyrre and other locals, and even the narrator herself, continually repeat that huldra is more of an idea than a real person, as the summer proceeds, Liv begins increasingly to identify huldra with a real girl of her own age, Maia. In addition to the incidences of drowning, a much more personal and profound turbulence shakes Liv in the form of letters she receives, hides, and wants to burn in the midsummer bonfire. A woman called Kate Thompson writes from England, informing Liv that her father is seriously ill and wants to meet her. Liv succeeds in delaying her visit so long, however, that her father dies on the night before her arrival at the hospital. There, a peculiar contrast between high-level mind reading and systematic ignorance of folk-psychological expectations characterize her meetings with Thompson. On one level, the novel offers cues to reading it as a romantic—and unmistakably unnatural—saga about the puzzling Nordic midsummer light, a great artist flourishing in the solitude of the island, the magic figure of huldra enchanting and destroying naive men, and only a single girl being perspicuous enough to see through all of this tragedy. After the summer, Liv decides to continue living with her mother and devote her life to mapping the landscape between her mother’s house and Kyrre’s now empty house. Drawing maps from stone to stone, from tree to tree, may possibly qualify as art for art’s sake; at least it completes Liv’s withdrawal from the trivia of the social world. “I have no wish to do anything, no wish to create. I am a witness, pure and simple, an unaffiliated, lifelong spy,” as she has it (Drowning, 50). From another perspective, this closure can be configured as a foreclosure of life and mind. On a darker level of reading, one can thus identify a series of phenomena that gradually abolish—or at least seriously undermine—the story of the Romantic North. The first of these is Liv’s disturbingly quick and detailed mind-attribution. I begin by analyzing some of the most blatant cases. The second theme is Liv’s recurrent strategic use of mind-attribution in protecting herself from genuine encounters in emotionally demanding situations. The third feature, in support of the previous ones, is the realization that Liv repeatedly misreads her mother and Kyrre, the two people she should actually know most thoroughly, on central issues concerning her own life. The fourth problem is Liv’s reluctance to share the facts about her visit to ­England, even with her mother and Kyrre, and finally her wild visions of Maia as huldra, including the visions of huldra’s catching and destroying Kyrre. In a rather paradoxical way, and in contrast to the invasive reporting of other minds, Liv fails to report adequately on her own tragedies of the summer, her qualia, as well as her own potential participation in Martin Crosbie’s or Kyrre Opdahl’s disappearance. In a worrisome way, she seems to resist contributing to any “social” or “intermental” mind (Palmer 2004, 2010); and equally, the “social mind” of her nearest environment resists confirming her unnatural version of the events.

Mind Reading, Mind Guessing, or Mental-State Attribution?  225 MOMENTS OF MIND READING As Dorrit Cohn (1978, 7) maintains, “[N]arrative fiction is the only ­literary genre, as well as the only kind of narrative, in which the unspoken thoughts, feelings, perceptions of a person other than the speaker can be portrayed.” Later theories of mind-attribution have challenged the validity of this “exceptionality thesis” by acknowledging the continuous sensitivity humans regularly have as regards the intentions and emotional states of their ­partners of conversation (see Zunshine 2006; Palmer 2004, 2010, 2011; Herman 2011). David Herman (2011, 11) most explicitly argues against the exceptionality of fiction by first noticing how “fictional minds are accessible but not transparent,” and second that “[e]very-day minds are not transparent, but they are accessible” (also Palmer 2010, 44). The key argument of this chapter is that these general claims do not go far enough in clarifying the issue of accessibility of minds. Instead of mere abstract claims about equal or differing accessibility, we most obviously need arguments about degrees of accessibility. Burnside’s novel (from now on, Drowning) provides us with excellent material for testing empirically credible, everyday mind-attribution taking place within the storyworld of the novel. The narrator’s own mind is far from transparent, to the point that it seems to be only minimally accessible even to the narrator herself. The whole interpretation of the novel depends on how far the reader is willing to trust in Liv’s sincerity and capacity for reading minds. The most critical point is still not about reading minds, occasionally, incorrectly, not even about resorting sometimes to wild mind guessing. The issue is about the strategic malevolence of mind-attribution, a kind of mind projection in the service of one’s own imagined world (see Tytti Rantanen, in this volume). Departing from many of the early works on mind reading (Zunshine 2006; Palmer 2004), my discussion is not based on the cognitive “Theory of Mind” (here, I largely follow Iversen 2013). One particular problem with the Theory of Mind is the obvious difficulty of drawing its limits of ­credibility. If Liv has an excellent Theory of Mind, as daughter of a major artist, on what grounds could we challenge her expert capacity to read other minds so fluently? In this chapter, I discuss the understanding of other minds from the more general perspective of folk psychology. A broad meaning of the concept of folk psychology is outlined by Jerome Bruner (1990), who uses the term to describe the script-like cultural knowledge about canonical sequences of events and the narrative means of dealing with the deviations of the expected. The more narrow meaning of the terms only refers to the understanding that people have different desires and reasons motivating their actions. Mind-attribution, in this strict meaning, also focuses on evaluating the desires and reasons of the studied person. Hutto (2007) argues, in his Narrative Practice Hypothesis (NPH), that children originally learn that human agents have desires and reasons for actions by listening to fairy tales and other narratives. To understand oneself as an agent, in other words,

226  Matti Hyvärinen is a narrative achievement for Hutto. This socio-cultural understanding of folk psychology, and the consequent capacity to understand minds, allows questioning the credibility of Liv’s mind reading. I am well aware that this reading is based on the use of my own folk-psychological knowledge, simply because there is not, and cannot be, any scientifically composed handbook of folk psychology. The purpose is to make of the mind readings topoi of argument, as something that can be argued, negated, or further elaborated. This kind of argumentation is possible because folk psychology is based on shared cultural knowledge, not on any theory package working in the mind. Alan Palmer (2004, 130–69) suggests the useful concepts of “mind beyond the skin” and “social mind,” in defending the “externalist” understanding of mind against the old idea that mind is confined inside the brain and skin. “An important part of the social mind is our capacity for intermental thought. Such thinking is joint, group, shared, or collective, as opposed to intramental, or individual or private thought. It is also known as socially distributed, situated, or extended cognition, and also as intersubjectivity,” as Palmer (2010, 41) recently argues. However, I am not primarily interested in locating such intermental units, as Palmer seems to be. In Drowning, one of the most chilly aspects of the narration is the realization of the gradual disappearance of the intermental confirmation of reality. Hutto (2004, 2008) specifically criticizes mind reading as “spectator sport,” meaning a distant third-person perspective on other minds. In most cases, he argues, the most reliable understanding of other minds is achieved from second-person encounters and actual conversation. Liv looks out of her window one night and sees the Englishman for the first time. In the light of the midsummer night, she finds something akin to a ghost-like unreality in him: “at the first glance, it seemed to me that he was a man without substance, not a ghost so much than an illusion, a phantasm in which he himself scarcely believed” (Drowning, 45 italics added). It is not merely Liv’s own perception; it is as if it were Crosbie himself having these ideas of not believing in his own existence. Looking at Crosbie, from some distance in the light night, Liv continues her appraisals: If Mother had been there, she would have said he was sensitive, or delicate, but to my mind there was more to it than that, something that had to do with my first impression of his being hurt or lost, like some animal that has strayed from its own habitat and finds itself exposed. … (Drowning, 46) The narrator takes the liberty of employing thought report (Palmer 2004, 75–86) or psycho-narration in accounting for the people she meets. Using side-shadowing (Morson 1994) and hypothetical narration (­Riessman 1990), Liv invites the image of her mother’s mind and evaluates the ­visitor and his character from her mother’s supposed perspective, before suggesting her own interpretation. The essential difference between these separate moments of

Mind Reading, Mind Guessing, or Mental-State Attribution?  227 mind reading is between the second-person and third-person perspectives, in the sense that Liv has a long and intimate second-person history of observing her mother, interacting with her, and making her observations largely but not exclusively reliable, while she neither has previous knowledge about Crosbie, nor has a history of dialogs with him. The reported thoughts, in other words, cannot carry along any layers of ­Crosbie’s own speech. Nevertheless, she is fast in appraising Crosbie already from afar. During their first and only proper conversation, Liv makes further evaluations: He laughed, rather sadly I thought. Though it was always hard to know, with Martin Crosbie, how real any of his supposed emotions were. He had worked long and hard on seeming innocent, I think. (Drowning, 56 italics added) Liv and Crosbie meet only a few times and discuss properly only this ­single time, yet the narrator knowingly uses the expression “it was always hard to know, with Martin Crosbie,” as if they had a long history of sharing thoughts. The observation concerning the difficulty of detecting real ­emotions remains distantly within the range of potential capacities even during a short encounter, by adeptly reading the embodied emotional states. The second part, “he had worked hard on seeming innocent,” is connected to the time of narration in the present tense. Nevertheless, it is such a complex statement about the history of Crosbie’s control of emotions that it is not credible without a real history of conversations and observations. How, indeed, do you tell those people who simply look innocent from those who have worked hard to achieve the ability to look innocent? Another instance of radical mind reading occurs when Liv meets a journalist who came to the island to interview her mother. The acute issues here are jealousy (her mother had better not take incidental visitors seriously) and sex (who would be worthy of her mother?). Liv stays in the adjoining room overhearing the interview, and becomes increasingly worried because of her mother’s tone and an intimacy she had not heard earlier (Drowning, 78). In other words, the mind reader already has a stance to protect, an attitude preceding the actual meeting with Frank Verne. As always, Liv is swift in her scrutiny of Verne. After a few words of tentative pleasantries, we have a powerful example of Liv’s mastery in reading minds: … and though he was smiling, I could see that he was studying me, trying to work out what I was hiding. Because I was hiding something. I had to be. Everybody had something they kept hidden and the only difference between one person and another was how long it took to figure out. That was what he was thinking. I could see that he was sure of this simple fact and the thought passed through my mind that I would either puzzle him or disappoint him, because I wasn’t hiding anything at all. (Drowning, 81)

228  Matti Hyvärinen The passage has a truly complex propositional structure, a real private t­ heory of other minds, presented by resorting to free indirect speech in the thought representation. Verne (like many others, obviously) believes that every person has a secret. Just by looking, it is possible to reveal this secret (as she thinks Verne believes). Verne believes that Liv has a secret. Verne is looking at Liv in order to find out the secret. Verne will be disappointed, because Liv does not have a secret. The crucial question here concerns the particular version of folk psychology motivating these propositions. If we think for a second about a visiting journalist who is professionally, intellectually, and erotically interested in the artist mother, what might he be ­seeking while looking at the daughter? Brunerian (1990) common sense would r­ ecommend that he—at least—is trying to figure out whether the daughter is inclined to complicate his plans or whether she already is an adult-enough person to be conversed with on equal footing. Alternatively, he may simply want to have an image of her character, to learn how to possibly converse with her in the future. How about gently comparing mother and daughter? From Verne’s point of view, of course, Liv already has a secret, the secret worry she has because of the warmth in her mother’s voice. Rather than any folk psychology, one can detect in the passage a piece of alarmingly paranoid psychology. It is, after all, Liv herself who has excelled at spying and detecting the secrets of Kyrre’s visitors. In this sense, Liv is straightforwardly projecting her own mindset onto Verne’s mind. There is one remarkable aspect in all these mind reading episodes. They never encourage a sustained exchange of ideas or any attempts at testing her interpretations. Rather, they work as excuses to finish the conversation. In total contrast to the near mystical clarity in reading other people’s minds, the narrator turns opaque in explaining her interest in “spying” ­during the summer. She used to spy on Kyrre’s visitors, because they mystified her. Here we almost receive a description of folk psychology, in the sense of Bruner (1990) or Hutto (2004), but in the negative. She affirms that she does not understand the visitors’ desires, fears, or wishes or what stories they wanted to recount to her. Rather than trying the horse’s mouth method in understanding others, Liv chooses understanding from a distance, spying with the help of technical tools. The disclosure above equals admitting that she is not, after all, very well equipped for understanding others. Shaun ­Gallagher (2007, 213–14) writes that in “most intersubjective situations” we indeed have “direct understanding of another person’s intentions” because they are “explicitly expressed in their embodied actions and their expressive behaviors.” Visual reading of embodied action and facial gestures can indeed reveal a lot about another person’s intentions, but again, the emphasis is on second-person encounters, not on the third-person reading from afar. Gallagher, of course, does not refer here to any such conceptually rich reading of another person’s ideas as Liv presented above. However, Liv seems to understand precious little about her own spying mind. She decides not to spy on Crosbie, yet she continues with it immediately. As a consequence, the narrator resorts to a weirdly negative narration, a

Mind Reading, Mind Guessing, or Mental-State Attribution?  229 technique she often uses in distancing her motivations from her actual behaviour. She proclaims that “she had no desire to watch,” and watches nevertheless, and next that she “didn’t want to know” the contents of Crosbie’s shopping bags but, a moment later, reports them in detail (Drowning, 51). At any rate, the recurrent negative explanation means that she does not need to give any proper explanations about her spying. This recurrent use of these negatives tells the reader that Liv may not be considered as the most reliable narrator, not at least when it comes to accounting for her own mind. A similar ambiguity shadows her relationship with the drowned Mats Sigfridsson. At first, curiously, Liv claims that Mats was “nothing” to her, he simply was a boy from her class at school. Then she mentions how “remote from the rest of the world” Mats was. Not that she knew him well, but she had thought that Mats might have been able to understand the way she “saw the world.” The bells start ringing when the narrator next declares that she is “not talking about a romance here” (30) and that she “wasn’t attracted to Mats” (31). Who had made such claims, and what exactly is motivating this continuous counter-argumentation? These repeated negatives invite the possibility that she, after all, was emotionally much more engaged than she was able or willing to divulge. The narrator, ten years afterward, is still working hard at speculating on the reader’s mind, in an attempt at rejecting all possible doubts about her. INTRUSION AND INTERACTION Liv had received binoculars as a 13th birthday present, and since then she had been spying on Kyrre’s solitary visitors—never families, never couples. She considered this spying to be harmless and kind; indeed, she explains that her basic motivation behind the spying was that she “wanted them to be happy” (Drowning, 27). Nevertheless, once she sees Crosbie driving away, she is immediately at his door, “feeling slightly guilty,” but after finding the door not locked, she slips in, ready to search the house (Drowning, 132). She finds Crosbie’s computer unlocked, and goes on to survey the contents. She finds photographs of normally dressed girls of about her age (­Drowning, 134). Gradually, a whole archive of high quality photographs portraying young girls is revealed, including some photos of her. There is nothing indecent in the pictures, and they were not about spying in Liv’s sense of the term. Dangerously, however, the girls were objects of desire, as Liv has it. Therefore, she interprets the photos as theft, not as innocent spying, and without a second thought she decides to destroy all the photo archives that Crosbie has on his computer. So much for only wishing those she spied on “to be happy.” The incident highlights a fundamental contradiction in Liv’s thought. She allows herself to spy on other people, enter their homes and computers, and administer corrections; this does not constitute any kind of ethically

230  Matti Hyvärinen problematic instance of intrusion. On the other hand, she herself hates to be observed by others and later has feelings of being followed. As a reference to the sexual theme of huldra, she is appalled while encountering the ­phenomenon of desire (more exactly, while meeting something she interprets as desire). It is not that she herself is merely not interested in sexuality, as she claims, she is actively and systematically against sexuality, despises it, and wants to censor other people’s attitudes and behaviour when possible. The most curious aspect of her intrusion into Crosbie’s computer is that Liv never considers the consequences of her intrusion, for example, when thinking about Crosbie’s sudden disappearance from the island. Despite the fact that Crosbie’s car and belongings all disappear at the same time, Liv does not think for a second about the effects of her own agency or the consequent embarrassment or hurt, but accuses huldra/Maia instead. Liv’s orderly time of spying and looking at picture books is interrupted when she receives a letter from Kate Thompson, informing her of her father’s illness. Against obvious folk-psychological expectations, Liv does not disclose the contents of the letter to her mother, and her mother, within the range of this somewhat exceptional family dynamics, does not ask ­anything about it either. Liv indeed praises her mother as a person who will not “intrude” on her private matters (Drowning, 65). Asking questions, showing an interest, and discussing daily concerns may thus constitute an intrusion into her life. Her mother is not at all like Thompson, who wrote the letter. However, even the narrator’s critical account informs one of the highly ­discrete way Thompson is writing, using a mode that social psychologists have called “doing delicacy” (Nijnatten & Suoninen 2014). The first letter does not ask Liv to do anything; it simply explains that her father, named Arild Frederiksen, lives in England and is seriously ill. Thompson asks Liv not to see her as an intruder, even though she is a stranger, and identifies herself as a well-intended stranger. That struck me as funny. How could this woman think it was well intentioned, to write such a letter and send it, out of the blue, to someone she did not know? (Drowning, 66) “Such a letter,” constitutes in Liv’s world a violation of protected canonicity, which is so self-evident that it needs no further explication. A letter coming from outside the closed sphere of her life without prior consent is questionable. If her mother does not speak about her father, no one else is entitled to do so. The choice of words is noteworthy. A letter informs her of the existence and illness of her father, yet the writer’s intentions sound “funny.” Ten years later, the narrator still airs her wish to have burnt all the letters in a midsummer night’s bonfire. The letter indeed had been “an intrusion”; and she only wanted “to be left alone.” A man is dying, yet Liv keeps thinking that telling her the facts “was not fair” (Drowning, 71).

Mind Reading, Mind Guessing, or Mental-State Attribution?  231 The letters themselves do not contain any such violation of folk-­ psychological canonicity, which would call for an explanation of reasons, or even motivate Liv’s harsh reaction. Rather, it is Liv’s righteous and hurt ­reaction that raises questions and calls for better explanations. For one thing, at the time of the story, she is somewhat too old to be the stereotypical, sullen, and totally egocentric teenager; for another, the narrator, a whole decade later, still does not take any more distance from this peculiar attitude. There is not a drop of empathy toward the seriously ill man or Kate Thompson; the only person needing to be pitied is Liv, so cruelly disturbed. She uses a lot of energy in vain speculations about the seriousness of the illness and her father’s role behind the invitations. In other words, she is not able to consider the invitation without her paranoid fears of inordinate manipulation. In her second letter, Thompson specifically clarifies the huge significance her visit would have. Again, in a manner of carefully “doing delicacy” (“but if you could find the time to come”), Thompson tries to adopt Liv’s perspective, even to accept her possible resistance before explaining how immensely important her visit would be to her father (Drowning, 120). Thompson urges Liv to see the situation from the perspective of the dying man and uses folk-psychological arguments in the sense Bruner (1990) uses the term. Liv, on the other hand, remains totally untouched and sees the issue firmly from the perspective of what primarily interests herself. “What surprised me even more was her assumption that I would want to see my father, that I would be curious, at the very least, to know what he was like. Yet I wasn’t curious. Not in the least” (Drowning, 120). However, only a modicum of empathy and compassion is what is requested. Liv, who in her own domestic environment is curious enough to spy on her neighbours and to break into Crosbie’s computer and reveal his secrets, declares now total indifference as regards her own father. Against all her expectations, her mother, after hearing of the contents of the letters, immediately and emphatically encourages her to go to England. Even in the case of her mother, she turns out to be rather a lousy mind reader. During their drive to the airport, they meet Liv’s father-substitute, Kyrre Opdahl. Her mother lightly announces that Liv is going to England, in order to meet her father. This announcement discloses a new, shocking misreading of her mother’s mind: That shocked me. I had assumed she wouldn’t want to talk about him. After all, she had been pretending he didn’t exist for years. (Drowning, 166 first italics added) The “shock” reveals that Liv is working with a set of assumptions about her father that her adored mother does not share at all. After returning home, Liv remains angry at her mother—not for the earlier “pretention,” but for the way she had now openly discussed the journey with Kyrre. After the journey, Liv also learns that it was not her father who had unfairly rejected her and her mother; on the contrary, the father simply did not fit

232  Matti Hyvärinen into her mother’s life dedicated to the arts. Nevertheless, learning these facts does not change anything in Liv’s thoughts; at least, there is no triggering of any self-critical reflection about her harsh reaction to her father’s death. Instead, all this belongs to the information she still wants to wipe away from her consciousness during the time of the narration. Her mother, instead, now sees no problem with her meeting with Arild Fredriksen; in contrast, she keeps telling her what a good man he was. The reader cannot miss the grave incongruence in her mother’s behaviour. Why did she not tell Liv about her father earlier; why was she so stubbornly evasive for years? Liv does not ponder on this. Instead she adopts the old position of forgetting the father as the natural stance, and a stance that binds her and her mother most safely together. Liv manages to defer her visit to the hospital so long that she never meets her father alive. While arriving during the previous night, she thinks, like a stubborn teenager, that even though she could still go to the ­hospital, one night would not make any difference. The big issue was not to rush to the hospital, because she “was tired” and “felt damp and slightly grimy” (­Drowning, 175–76). In the face of her father’s coming death, the acute issue is how she feels bodily. At the hospital, this self-righteous girl does not want to see the body of her father (again a potential threat to her world view); instead she insists on leaving the hospital straightaway. When ­Thompson does not let her escape immediately, the narrator explains that it was Thompson who “needed something more” (Drowning, 184). Liv sticks to her strong understanding about Kate Thompson as an intruder, yet Thompson manages to have a conversation with her. During this singular personal encounter with Thompson, Liv uses mind-attribution in a determined way. She has just lost her chance to meet her father alive, ignoring her father’s wish to see her at least once before his death. These are not concerns that seem to worry Liv at all, neither at the time of the story, nor at the time of narration a decade on, because she did not want to leave her island in the first place. After all, she did not invite these people into her life. Kate Thompson wants to tell her about her father, while Liv focuses on not listening. She is not the least open for human communication; instead she thinks the whole conversation only serves Thompson’s needs. In her arrogance (“I couldn’t help thinking”) she translates Thompson’s attempt at talking to her only as a sign of her loneliness, of not having anybody else to talk to about the deceased. After diagnosing Thompson’s loneliness, she proceeds into far-fetched observations about the misery of their past life: … I suddenly had an overwhelming sense of loneliness, a sense of a sad, slightly dismayed couple […] but I knew they had come together, not because of something they had shared but out of common sense their best days were over, a common feeling that whatever they had wished for in life hadn’t quite materialised. (Drowning, 194, italics added)

Mind Reading, Mind Guessing, or Mental-State Attribution?  233 Even though the existence, life, and death of her father would not have meant anything to her, the above passage is weird, because Kate Thompson’s talk would still be entirely understandable and well-motivated: even after the father’s death, Thompson wanted to connect Liv to him somehow, to give her something to remember. However, Liv’s mind works with a d ­ ifferent agenda entirely. Without any prior information, joint experience, communication, or extensive life experience in general, she is capable of attributing loneliness and a sad life to a person who has just lost her partner. Characteristically, the narrator first minimizes her own responsibility for her own thoughts with the idiom “I couldn’t help thinking.” Nevertheless the emphatic claim that she somehow knew how “they had come together,” for definitely miserable reasons and not for sharing something more profound. These are drastic claims and cannot be based on any mind reading (note that the whole episode consists in Liv’s rejection of all folk-psychological expectations of decent behaviour; now she should have perfect command of folk psychology in reading Thompson’s mind); they rather appear as malevolent mind-attribution. By attributing this sordid life to Kate Thompson (and her father), she actively closes her ears and eyes both to the actual conversation and to the death of her father, no longer needing to take these seriously. Attributing miserable histories and ideas to other minds, therefore, is for her an effective mental and discursive strategy of downplaying the relevance of the other. Desperately, she sticks to the imagined, mythical story about her mother and father and cannot attribute any value to her father and his actual life outside the sphere of her mother. DISPLACEMENTS AND PARANOIA As successful as Liv was at effectively evading her father and Kate Thompson, the trip to England was not without consequences. After leaving the hospital, a haunting feeling of being followed creeps into Liv. After all, she feels guilty, but not because of her own behaviour; she feels guilt toward her mother. She suspects that, just by listening to Kate Thompson, she has been part of an attempted betrayal of her mother (Drowning, 203). In the hospital, Liv had offered her gravely misleading explanation of Arild’s disappearance from her mother’s and her own life. Yet, Liv is not the least worried about having understood the facts of the story incorrectly, she is worried about Thompson’s (presumed and attributed) thought that her mother had misinformed her about her father (Drowning, 244). So far, she had built all her reactions to her father’s state on her fictions about her parent’s story, whereas now she only feels guilty because she was listening to a story that might compromise, somehow, the integrity of her mother and her stories. Before leaving her hotel, while having her breakfast, Liv sees, outside in the garden, a small girl lingering, despite the wet weather. At first glance, the girl’s face looks angelic and pleasant but, suddenly, “the look of her face

234  Matti Hyvärinen turned to a grimace of utter, violent hatred, not just of me, but everything and everyone” (Drowning, 220). Liv feels that the girl is somehow familiar but does not understand how. When she raises the alarm for the personnel to check out the girl, no one can see any child around. Again, the outraged and hateful little girl is out there, haunting innocent Liv, and observed only by Liv. What she did or said during her visit is not the slightest problem; the real problem is the experience of being followed, the “preposterous” ideas of Kate Thompson, and the hateful girl who comes to disturb her on the last morning. No wonder then that the next section of the novel is entitled “huldra.” Liv returns home but does not want to share her experiences with anybody. “The last thing I wanted was a meaningful conversation about Arild Fredriksen’s death” (Drowning, 229). Liv wants to remain unseen, even abandoning her spying. Yet, she happens to see Crosbie—and now with huldra. Of course, what she perceives is Martin Crosbie in the company of Maia—who is now tightly identified as huldra—and an obvious affair is blossoming between the two. Alas, this is not the only affair that comes to shock her (Drowning, 230–31). The narrator keeps insisting that her mother does not live in full solitude, because every Saturday afternoon she has an artistic tea party, gathering a small group of local artists and intellectuals. Liv’s attitude toward the group is twofold: ironically and condescendingly she calls the men “suitors,” seeing them as necessary but slightly comical pawns who witness her mother’s grandeur as an artist, thinker, and desired beauty. In this game, her mother is the one who sends the men off, lightly, and concentrates then exclusively on her artistic work. Of course, the very name “suitors,” semiconsciously, invites the image of the eventually returning Odysseus. After Arild’s death, this is of course not going to happen. In contrast Ryvold, one of the most valued regulars comes, to the house to say goodbye, purposefully when Liv’s mother is absent. By coincidence, he had met the lover of his youth, and they had decided to give a try to a new life together. Liv’s reaction to this news is intense, emotional, and negative. She is disappointed with Ryvold and blurts out: “And I always thought it was Mother you were in love with” (­Drowning, 254). Liv sees his departure as “betrayal,” not so much of her mother, but of himself, as “if he had settled for something less than he deserved” (Drowning, 255). Ryvold is committing the same mistake as Liv’s father, accepting something less valuable than her mother. In order to be honest with himself, Ryvold should have preferred this once-a-week meeting with an admirable but inaccessible woman to a real-life spousal relationship. Later, when Liv tells her mother about the visit, she curiously omits the part about the “girl,” as if it were a purported insult to her mother. On the same evening, her mother confirms that she had never wanted marriage, with Arild or anybody else. Instead of trying to think or talk through these new facts of life, Liv swiftly moves to the delusional side of her world. She wakes up in the night, looks at the meadows, and feels that a new story has started, being part of an unknown world and unknown logic. For Liv, stories seem to take place

Mind Reading, Mind Guessing, or Mental-State Attribution?  235 following their own intrinsic logic, existing before the narrators and narration. The possibility that her own mind would be the primary source of this emerging story never occurs to her. As a consequence, Liv takes her binoculars, offers her regular disclaimer on Crosbie’s boring and “tawdry romance,” and then nevertheless starts spying on him. What she sees is that Crosbie has taken out Kyrre’s boat, in the company of Maia, and is radiating unnatural happiness. In this troublesome happiness, Crosbie seems to be the double of Ryvold and her father. While Liv looks away for a second, Crosbie has disappeared and the surface of the water is calm. Liv runs to the shore, sure that Crosbie has been drowned, but without raising the alarm. Later, there is a weird encounter among Maia, Liv, and her mother, but clearly no shared account emerges of what has happened. Liv tests her story about the drowning, but her mother is not convinced. Liv realizes that her mother no longer believes her but is rather convinced that Liv is “seeing things” (Drowning, 269–70) and begins, accordingly, to treat Liv as a patient. It is worth comparing the speed with which Liv has intervened in Crosbie’s supposed drowning with her slow reaction and even resistance to her real father’s illness and dying. One ­possible way of explaining this incongruence is that the mere existence of her father, independent of her mother, seriously threatened her phantasy, while the whole “drowning” takes place within her safe phantasy world. Earlier in the day, Liv had witnesses Ryvold leaving the suitors. She hears her mother confirm Kate Thompson’s version about her separation from Arild. Her image about the small social world she mostly—and exclusively— appreciates, turns out to be based on phantasy. Despite her expert capacity to read alien minds, she has failed drastically in understanding the social world nearest to her. Instead of considering this rupture, her interest turns to the malevolent huldra, the vicious and dangerous principle of female sexuality. From this night onwards, her world grows increasingly apart from her mother’s and from everybody else’s world as well. As the narrator, at the end of the novel reveals: “I’m not crazy—I know enough, after all, not to talk about these things to the living…” (Drowning, 328). Surely, we readers do not inhabit the same world as those living in the novel. HER MOTHER’S BETRAYAL In Liv’s world, Crosbie is drowned and dead, Maia an enemy and a grave threat to her peace. While Liv is recovering from the previous encounter, resting in her room, her mother brings Maia into the house, to sit for her as a model. The narrator frames the setting like a true horror story, full of danger, having an alien, hostile presence with alien odours inside the house. This is one of the few instances when Liv is both angry and disappointed with her mother, who simply ignores Maia’s dangerousness, asking plainly if Liv does not like her (Drowning, 279). They soon enter into a discussion about

236  Matti Hyvärinen the night when Liv “saw” the drowning. The negotiations about reality are delicate and careful here. Liv’s mother agrees that something awful had happened, but she insists that it had happened to Maia as well. Liv quickly discounts this existence of diverging versions of reality by explaining that Maia must have been able to tell her own version of events earlier and more convincingly to her mother. The paranoid logic1 is watertight, and so no further argument or observation can challenge the phantasy about the vicious huldra any longer. Within her paranoid script, Liv soon decides to “reclaim her house,” that is, to drive Maia out of the house. This activity, of course, is somewhat absurd if we believe that Maia is the dangerous and powerful huldra; if she is simply a girl sitting for her mother, the activity is both needless and crude. The reclaiming of territory indeed leads to a hostile conversation in the garden. According to Liv’s phantasy, Maia is—due to the sitting—­ emotionally dependent on her mother, and she rejoices in advance of the idea of her mother nonchalantly sending Maia off. Here, the lonely daughter of the artist is probably revealing her most vulnerable point, being rejected too early and often, the experience beyond reflection. Be that as it may, the idea is a blatant case of mind-attribution, of projecting her own fears and priorities onto an entirely different mind. The encounter is presented in a way that offers entry into two, radically different worlds. Liv remains exclusively within her paranoid scheme and uses mind-attribution to keep her story intact. When the girls start to provoke each other, Maia turns out to be much more poignant: “Tell me,” she said. “Did you ever fuck anybody?” She glanced at me sideways, still smiling her sweet, practiced smile. “Or are you just as cold as your nice, cold house …?” (Drowning, 285) This is a crude question, of course, but it nevertheless gets to the point: does Liv ever touch anyone, mentally or physically, or does she exist only inside her cool detachment of a phantasy world? Liv tries to hurt Maia by asking whether her mother is still using her or not, with little influence. Instead, Maia soon hits much harder by hinting that Liv’s mother, being “a complicated woman,” has “some things she needs to think over” (Drowning, 286). With this short comment, Maia presents herself as one of the few persons not ready to join in with the unreserved adoration of her mother, airing the issue about the real woman beyond the official portrait. The comment, alarmingly, also hints at some real conversations that may have occurred during the sitting sessions. Be that as it may, this competition in ridicule again foregrounds a massive contradiction in Liv’s mind reading. Earlier, I documented her obviously unlimited capacity to “read” other, adult minds (Crosbie, Thompson). Now she meets a girl of her own age and school, and she appears totally at a loss to read Maia’s reasons and desires.

Mind Reading, Mind Guessing, or Mental-State Attribution?  237 Liv’s delusional turn finds its apex during the episode of Kyrre’s disappearance. Liv has, after her trip, realized that Kyrre is the only person besides her mother whom she really loves. This, however, does not include sharing the experiences of her journey with Kyrre. Of course, Liv does not reveal to Kyrre her intrusion into Crosbie’s computer, and does not even think about it while Kyrre, for his part, is angry about the disappearance of his visitor. In the decisive scene, Liv, Kyrre, and Maia meet in the meadow near Liv’s home. In Liv’s phantasy, Kyrre has decided to attack huldra directly and disregards the risks because he wants to protect the people he loves most in the world and lead huldra away from their house. Nevertheless, Liv is appalled by Kyrre’s enchanted politeness toward Maia, by the old man’s ridiculous offer to give shelter to the young, homeless girl, and the two of them walking away as if lovers. Liv tries to stop them, but badly imprisoned by her own mind, she tells Maia that her mother has, after all, some unfinished business with her, and she should go to meet her immediately. In her phantasy, both Kyrre and Maia are planets circulating around her mother, as she is, not independent humans with their own desires and reasons. In reaction to their disappearance, Liv has a series of most unnatural perceptions and experiences, which I consider to be psychotic. After running around the woods and meadows in heavy rain, Liv collapses immediately upon arriving home. The narrator adds one telling detail: even though she was seriously ill, only her mother nursed her during the coming weeks, and no medical expert was ever invited to examine her. The mental nature of her collapse needed no further witnesses. DISTURBING NOVEL, DISTURBING MIND On my reading, the novel is so confusing because the narrator herself is so confused and increasingly paranoid in the storyworld. This is clearly a novel with two separate worlds (see Ryan, this volume), but because of the unreliable narrator, the reader has an unending task in checking details that are somehow confirmed or solely told from within the psychotic phantasies of the narrator. The novel is fine-grained in showing the complex ­texture of natural and unnatural elements in the mind of the girl during her worsening crisis. The novel provides a number of hints about Liv’s history, about how her artist mother left her alone for days on end in order to be able to ­concentrate on her own creative work. Both with Frank Verne and Maia, Liv nurtures the pleasant thought of her mother “sending off” the visitor. This is what Liv does—she spies on people from afar, imagines their minds, and sends them off rather than make any effort to connect with them by using the folk-psychological horse’s mouth method. The psychological level of the novel is densely crafted and credible but cannot be fully explored within the range of this chapter. The novel problematizes both the benevolence and the nature of mind reading, as well as assumptions about automatically or naturally existing

238  Matti Hyvärinen social or distributed minds. The issue is not primarily about the correctness of mind reading, it is about the performative difference between mind reading (as if trying to understand) and mind projection (attaching various contents for various reasons to other minds). The novel effectively undermines the assumption about the self-evidently well-intended process of mind reading. Equally, the reader of the novel has to struggle with de-constructing the assumptions about shared, social minds and shared worlds as the story becomes increasingly supernatural and delusional. At the end of the novel, Liv’s mind is effectively distinct, not only from the minds outside the island, but also from the minds of Kyrre, Ryvold, and her mother. As I have tried to argue above, the first-person narrator’s mind reading indeed seems to require the folk-psychological checking and evaluation that we tend to practice in everyday interaction. The interpretative dilemma seems to concern Herman’s (2011, 11) second claim. Within the novel’s storyworld, Liv seems to think and behave as if other minds were more or less transparent for her. The use of such discursive forms as thought report and free indirect speech, prior to any history of actual dialogue, surpasses the credible accessibility of other minds in everyday interaction. Paradoxically, the novel foregrounds both the constant everyday mind reading, and the qualitative ­difference between representing minds in fiction and everyday situations. This suggests that there is possibly a much longer distance between the ­everyday guessing and knowing of other minds to the explicit, immediate, and verbally rich representations of the minds of fiction than is often admitted. In recent literary theory, “conventional” tends to be quasi-automatically attached to “conversational” and “natural” narratives (e.g., Richardson 2013, 16). This recurrent juxtaposition gravely simplifies the place of the unnatural in many narratives and the ways the unnatural is highly conventionalized in contemporary culture. When Liv’s understanding of her life breaks down, she discursively resorts to conventional Norwegian folktales, which provide her with a distinctively conventional and unnatural language for her paranoia and psychosis. The immersion into the mythic unnatural/ supernatural contains very little by way of life experimentation; rather it works as a shelter against other people and against both maturation and sexuality. Burnside, on the other hand, provides his readers with a rare and nuanced vision of a mind growing increasingly confused. The novel does not offer the two worlds and realities as neatly separated and opposite; instead, the reader has to struggle back and forth inside the already flawed mental map of the narrator. NOTE 1. On the diagnostics of paranoia, see Leader (2011). Leader claims, interestingly, that “[d]elusion is thus a positive rather than negative phenomenon, an attempt at healing rather than a pathology in itself” (2011, 70). This makes the timing of the delusions an even more intriguing issue.

Mind Reading, Mind Guessing, or Mental-State Attribution?  239 REFERENCES Bruner, Jerome. 1990. Acts of Meaning. Cambridge: Harvard University Press. Burnside, John. 2012. A Summer of Drowning. London: Vintage. Cohn, Dorrit. 1978. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton: Princeton University Press. Gallagher, Shaun. 2007. “Pathologies in Narrative Structure.” In Narrative and Understanding Persons, edited by Dan D. Hutto, 203–24. Cambridge, UK: ­Cambridge University Press. Herman, David. 2011. “Introduction.” In The Emergence of Mind, edited by David Herman, 1–40. Lincoln: University of Nebraska Press. Hutto, Daniel D. 2004. “The Limits of Spectatorial Folk Psychology.” Mind & ­ anguage 19.5: 548–73. L ———. 2007. “The Narrative Practice Hypothesis: Origins and Applications of Folk Psychology.” In Narrative and Understanding Persons, edited by Daniel D. Hutto, 43–68. Cambridge: Cambridge University Press. ———. 2008. “The Narrative Practice Hypothesis: Clarifications and Implications.” Philosophical Explorations 11.3: 175–92. Iversen, Stefan. 2013. “Broken or Unnatural? On the Distinction of Fiction in NonConventional First Person Narration.” In The Travelling Concepts of Narrative, edited by Matti Hyvärinen, Mari Hatavara, and Lars-Christer Hydén, 141–62. Amsterdam: John Benjamins. Leader, Darian. 2011. What is Madness? London: Hamish Hamilton. Morson, Gary Saul. 1994. Narrative and Freedom. New Haven: Yale University Press. Nijnatten, Carolus van, and Eero Suoninen. 2014. “Delicacy.” In Analysing Social Work Communication: Discourse in Practice, edited by Christopher Hall, Kirsi Juhila, Maureen Matarese, and Carolus van Nijnatten, 136–72. London: Routledge. Palmer, Alan. 2004. Fictional Minds, edited by David Herman, Frontiers of ­Narrative. Lincoln and London: University of Nebraska Press. ———. 2010. Social Minds in the Novel. Columbus: The Ohio State University Press. ———. 2011. “1945–: Ontologies of Consciousness.” In The Emergence of Mind. Representations of Consciousness in Narrative Discourse in English, edited by David Herman, 273–97. Lincoln: The University of Nebraska Press. Richardson, Brian. 2013. “Unnatural Stories and Sequences.” In A Poetics of Unnantural Narrative, edited by Jan Alber, Henrik Skov Nielsen, and Brian ­Richardson, 16–30. Columbus: The Ohio State University Press. Riessman, Catherine Kohler. 1990. Divorce Talk: Women and Men Make Sense of Personal Relationships. New Brunswick and London: Rutgers University Press. Zunshine, Lisa. 2006. Why We Read Fiction: Theory of Mind and the Novel, edited by James Phelan and Peter J. Rabinowitz. Columbus: The Ohio State University Press.

13 Mind as World in the Reality Game Show Survivor Maria Mäkelä

In the two-hour opening episode of the twentieth season of Survivor: Heroes vs. Villains, Russell Hantz, a contestant already familiar to the audience from the previous season, announces in his introductory interview that he’s going to “use the mind game” to beat the other players and win the million-dollar prize. In the end, he comes third in a game of twenty contestants, ­losing all the votes of his former co-players to the winner Sandra Diaz and the first runner-up Parvati Shallow. In the “Reunion” episode’s aftermath, Russell expresses his frustration at the fact that Sandra won the game without any social game, with no strategy other than “get rid of Russell.” He even argues that there “is a flaw in the game,” since a player can win by assuming a passive role in the socio-psychological laboratory of Survivor, where the actions focus on alliances, strategic voting, and blindsiding through fake alliances and immunity trade-offs. Yet Russell’s defeat and frustration reveal the true complexity of the game: the winner needs to be both a merciless “villain” and a popular “hero” to first eliminate his/her co-players one by one and then win back their support in the final vote. Viewers may remember how Russell was anxious to find the first hidden immunity idol of the season (in episode four)—a talisman that protects the player from votes cast against him in the tribal council—completely disregarding the consensus reached among the tribe to leave the idol hidden and unused for that round. Sandra spies on Russell and goes back to camp to report on what she saw: “He’s a stupid ass. … He sealed his own fate.” Boston Rob, who is the apparent leader of the tribe at this point in the game, announces in his confessional to the camera that he doesn’t “trust Russell’s ass at all” and will therefore make sure that his allies vote for Russell in both upcoming rounds, first to force him to play the protective idol, and second to vote him off the island. Yet Russell easily outlasts Rob in the game, with Rob being the eighth contestant to be voted off. So, who is playing the mind game, and what does this game actually entail? In this chapter, my aim is to demonstrate, using a couple of examples from Survivor: Heroes vs. Villains, how the storyworld of the show is a stage for intermental mind games1 that are given material form in the show’s confessional interviews and pseudo-mythological rituals. The nexus between mind and world in Survivor is emblematic of today’s mediated narrative

Mind as World in the Reality Game Show Survivor   241 environments, such as social media and reality TV in general, where unlimited self-expression and manipulation go hand in hand. The spirit of contemporary narrative theory is transmedial as well as transdisciplinary: narrative theorists of the cognitive paradigm such as David Herman, Marie-Laure Ryan, and Lisa Zunshine are keen to detect the same narrative universals at work in all media, especially in the ways that stories set up worlds and inhabit them with mental subjects. However, my own methodological point of departure is the assumption that different narrative genres and media have their own specific means of foregrounding mental action: in reality TV game shows, of which Survivor is one of the most popular and long-lasting formats, the internalization of storyworld e­ lements and the excessive display of intermentality are such media- and genre-specific ­elements. In the hope of reaching beyond narrative and cognitive universals, I approach the confessional interviews with the methodological framework of consciousness representation. This approach is traditionally associated with the classical narratological analysis of minds in fiction, but it is adaptable, as I will claim, to contemporary media environments (see also Hatavara’s chapter in this volume). Furthermore, the cognitive-narratological notion of storyworld construction will need to be reconfigured in order to explain the mechanics of internalization—of transforming material objects into mental content—that is emblematic of contemporary reality TV. As Marie-Laure Ryan points out in her exemplary analysis of Survivor as part of the history of narrative media, “[i]f the opposition of life and art is one of formlessness versus design […] Survivor aggressively pursues the artifice of art” (2006, 68). Ryan’s analysis captures the dynamics between “the kitschy new age spirituality” and Western capitalist materialism in ­Survivor (ibid., 71). Following Ryan’s (2006, 72–73) example, I also want to look at the show as a hyperreal spectacle that, through scripting and editing, is able to produce a reality that is better than reality, quite in the sense of Baudrillard’s simulacrum. However, my own emphasis is not on the spectacle of life in general but on the performative nature of the mind in particular. My hypothesis, which extends beyond Survivor, is that reality TV is, first and foremost, a performed intermental mind: the expressive and emotional confessionals and the overtly explicated and materialized scheming give voice to mental operations that are far more automatic, or, at best, remain implicit, in everyday social navigation (see Hutto 2008; also Hyvärinen’s chapter in this volume). THE MIND: AN EXCESSIVE DISPLAY OF INTENTIONAL STANCES Coach, a laughing stock to the other players because of his shamanistic outlook and pseudo-Asian meditative and martial art activities, contemplates the initial mental setting of the game in the opening episode of Heroes vs.

242  Maria Mäkelä Villains: “You can feel something in the air, there’s an electricity, there’s a tension, there’s a level of excitement that is coursing through everybody’s veins, especially mine.” In his opening confessional interview, Coach is thus tuning in to the level of intermental manipulation at which the actual game will be played, as well as instantiating the dominating discursive mode of the show—that of abundant display and the analysis of mental and intermental states. In her recent study on the workings of the so-called theory of mind in popular culture, Lisa Zunshine (2012) gives many examples of what she calls “the culture of greedy mind readers,” demonstrating how the application of mental states to other people is at the heart of our everyday social navigation and, what is more, even the primary source for the interest and the pleasure we take in watching cultural representations of human life. In her earlier study, Zunshine (2006) quite convincingly demonstrated how ­literary fictions in particular—thick with unreliable voices and complex consciousness representation—issue an unparalleled challenge to our capacity to track down and evaluate intentional stances. Later, however, her theoretical and analytical scope widens to cover practically all art and entertainment, even non-representational art (Zunshine 2012, 148–50) that doesn’t display “embodied transparency,” that is, representations of bodily gestures and facial expressions that would catalyse mind attribution. This expansion in her theoretical argument, although cognitivetheoretically attractive, weakens its descriptive power in the field of ­narrative theory, literary research, and cultural studies, since it loses sight of the genre-specific potential in mind construction and mind attribution in different narrative environments. In a book chapter on reality TV, ­Zunshine’s argumentation lumps together reality shows and novels, since both “build their appeal around human cognitive universals, such as our ability, need, and desire to read minds in social contexts” (Zunshine 2012, 119). The only distinction she makes is between the trained acting of emotions in fictional drama and film and the spontaneous overflow of feelings—and the failures in concealing those feelings — expressed by “ordinary people” in reality shows (ibid., 121). Yet I find it difficult to read Survivor confessionals as spontaneous expressions of mental content that would only highlight the cognitive universals of attributing intentions or mental states to oneself and others. As Zunshine (2012, 5) points out, in everyday communication our theory of mind is “fast, messy, intuitive, not particularly conscious, and mostly not verbalized.” In a reality game show, however, this is not the case. Let us look at an illustrative example of how intentional stances and their attribution are displayed in the twelfth episode of Survivor: Heroes vs. ­Villains. At this point, the remaining players in the two tribes, the “Heroes” and the “Villains,” have been merged into one tribe that the players choose to call “Yin Yang.” The twelfth episode witnesses one of the most flagrant

Mind as World in the Reality Game Show Survivor   243 yet partly unsuccessful “blindsides” of the season: Russell tries to manipulate Parvati into believing that her long-term ally Danielle is about to blindside her and vote her off, since she allegedly feels threatened by Parvati’s skill and popularity as a player. What is more, Russell also goes to Danielle and feeds her a similar story about Parvati. Russell willingly discloses his own intentions as well as his confidence about the women’s ensuing intentions (of not addressing each other about the matter) in his “confessional,” a private monologue to the camera: I’m working Danielle, I’m working Parvati. I’m telling Parvati we need to get rid of Danielle. I’m telling Danielle we need to get rid of Parvati. So, I’m working both of them, I’ve got both of them under control. Parvati is not going to tell Danielle, Danielle is not going to tell ­Parvati, because that would mess up both of their games!

Figure 13.1  Russell Hantz giving a confessional in the twelfth episode of Heroes vs. Villains.

Yet a cut to the footage from the camp proves Russell’s hypothesis wrong as Parvati immediately goes to Danielle and Russell’s scheme is disclosed. We are returned to Russell’s monologue, where he continues hypothesizing about the effects this scheme would have on Parvati: I’m a get Danielle off tonight, because it eliminates that pair. As soon as we get rid of Danielle, Parvati is going to be so scared of me that she’s going to stick to me like glue, and she’s going to vote the way I tell her to vote.

244  Maria Mäkelä Next, we hear Parvati’s confessional, introducing her genuine intentions as well as a third-person perspective of Russell’s state of mind: Russell running around double-talking is hurting his game. He’s ­getting outplayed by me and Danielle at this point. And it’s making him a little bit crazy. All this scheming, performed in discussions shot at the camp and in confessional interviews, is preparation for the tribal council, where each player will cast their vote on who should be sent from the island. The council takes place at sundown, in an exotic construction that resembles an enormous tree house, and each player carries a torch that symbolizes their “life” in the game that is snuffed out if the others vote them off. The council is also a location for shared confessionals among tribe members, since it is here the game host, Jeff Probst, interrogates the players about their emotions, tactics, and interpersonal relations. Probst’s questions typically provoke spontaneous arguments, revelations, and articulated facial expressions. This time, Probst is able to elicit the crisis that shakes the Russell-Parvati-Danielle alliance. Parvati expresses her interpretation of Russell’s scheme: “Russell is trying to test loyalty, that’s what I got from the whole thing.” Russell and Danielle enter into a heated argument over who said what, and the audience knows perfectly well that Russell is lying. The audience is offered one closeup shot after another of the surprised players’ faces. Finally Danielle breaks down and appears to lose hold of the “mind game”: DANIELLE:  The three of us are in an alliance so I don’t understand why …

I haven’t done anything to be disloyal to him … I don’t understand why he’s testing me. I don’t know [starts crying]. PROBST:  So what are the tears about right now? DANIELLE:  [Crying] I’m just so freaking … I get exhausted and I’m just like … It’s just too much for me right now … I just had a rough day. We have been in an alliance since the beginning. I don’t know why he is trying to mess with it. PROBST:  Is it feeling personal to you, is it hurting your feelings? DANIELLE:  Yeah. Danielle goes on and, as it seems, lets her guard down and accidentally reveals how strong her alliance with Parvati actually is. Russell’s immediate response to this revelation is mock-surprise: “Oh really!?” This is ­followed by an extradiegetic sound effect marking both surprise and a sense of threat—thus supporting the performed collective mental state among the players. With his words and expressions, Russell is gesturing the other players to recognize the threat that such a strong alliance between Parvati and Danielle issues to everyone else’s game. A considerable amount of facemaking among other players ensues, the camera focusing on each in turn:

Mind as World in the Reality Game Show Survivor   245 Parvati closing her eyes indicating disappointment; Russell looking at Jerri and forming the name “Danielle” with his lips; and most notably, Jerri’s stunned face as she happens to be the indecisive flip-voter in this round. There are also two cuts to the “jury,” which consist of former players who have been voted off but who are present in tribal councils to evaluate the remaining players’ actions and will finally choose the Sole Survivor (i.e., the winner) from the final three players. In these shots of the jury, we see Courtney, one of Russell’s former enemies, first “boohooing” the complaining Danielle, then agitatedly waving her hands when Russell blatantly lies. In these gestures and expressions among the jury, we see some of the seeds of Russell’s ultimate defeat in the final vote being planted: in the live broadcast from the finale, we see all jury members condemning Russell’s tactics and leaving him without any votes. It seems both analytically and theoretically insufficient—and even reductive—to say that this exposure of mind games is only an illustration of cognitive universals. The confessionals and the tribal councils are discursive and bodily performances of the mind-reading game that would, as Zunshine reminds us, remain unverbalized and implicit in everyday communication. Internal states are overly explicated, not just through expressive voice but also through exaggerated facial expressions. Furthermore, if we wish to look for a common denominator for narrative strategies in literary fiction and in reality shows (as fleetingly suggested by Zunshine), I should think that instead of repeating the cognitive-scientific lesson about the universals of mind attribution, we could study how the naturally incommunicable is communicated and materialized in both of these forms of narrative mediation. The novelistic conventions textualize and intentionalize inner states as they represent the alleged internal “voice” of fictional characters. This voice is, paradoxically, incommunicable and, at the same time, communicated, and as I have argued elsewhere, it becomes pseudo-directional as the inner voice appears as if it had rhetorical intentions. (See, e.g. Mäkelä 2011 and 2013.) Somewhat correspondingly, reality shows steer the performers into overly expressed emotions and intentions, thus perverting intermental actions into cultural representations that generate interpretive challenges beyond the everyday theory of mind reasoning. Consider how, for example, Survivor’s host, Probst, with his provocative questions at the tribal council, pushes the players for interpretations of each other’s thoughts and feelings, as in the fourth episode of Heroes vs. Villains: PROBST:  Cirie,

how worried are you tonight that you might be going home? CIRIE:  I’m always worried, always. PROBST:  Tom, do you believe that Cirie’s worried that she’s going home tonight? TOM:  I believe that Cirie is not worried that she is going home tonight.

246  Maria Mäkelä In the case of Survivor, I would even venture to claim that the ultimate point of the show is to distort intermentality by displaying and materializing it through the confessional voice, through provocation during the tribal council, and all in all, through turning it into a game of popularity, trust, and manipulation. The confessionals and the tribal council are narrative ­situations that expand on the complex webs of intentional stances and levels of intention that define the position of the players. What is more, the explication of interpersonal dynamics in Survivor takes on forms that come close to the cognitive-theoretical modelling of intentional stances (for the original formulations, see Dennett 1987): To make the discussion of mental states manageable, we make it sound neatly isolated, evenly paced, intentional, self-conscious, and fully ­verbalized, as in, “I suspect that she is thinking that they don’t realize that she is having a difficult time concentrating when they are whispering and laughing.” Still, even if we have no choice but to talk about it this way, we should remember that this is not how our theory of mind really works. It’s fast, messy, intuitive, not particularly conscious, and mostly not verbalized. (Zunshine 2012, 5) Yet it is not only in theoretical modelling that intentional stances can be laid bare in all their uncanny embeddedness; consider the title that a Survivor fan gave to a YouTube clip of Amanda’s confessional from Heroes vs. Villains Extras2: “Amanda becomes worried that her tribe is under the impression that she is aligned with Cirie after Tom eavesdropped on a conversation.” The title of the clip represents a typical audience response: individual experience and feelings are considered essential narrative information, as well as the laying out of intentional stances. In fact, a non-naturalizing reading of reality confessionals could treat them as consciousness representation, since as I indicated earlier, the overly expressive and experiential voice of the players bears some resemblance to the generic peculiarities of experiential voice in literary fiction. One particular editorial strategy especially lends support to this reading: the use of confessionals as voice-over narration when the footage shows a particularly climactic moment in the social exchange between players. The whole editing process of Survivor always takes place after the fact, so we can assume many editorial decisions foreshadow the outcome of the game; thus, the opening episode of Heroes vs. Villains already starts to develop Russell’s character as the ultimate villain. Boston Rob in particular—a player considered by many other “Villains” as the true leader of the tribe—provokes Russell into contentious confessionals: Rob, he thinks he’s the boss of the camp, he thinks he’s in control. But this is my mountain, and I’m still the king of the hill. Survivor is my

Mind as World in the Reality Game Show Survivor   247 palace, so I’m still the king unless I’m dethroned. But you know what, that ain’t going to happen. This confessional, like many others in the show, first hides its natural source as an interview and appears as a voice-over narration. As Russell starts speaking, the footage first shows Coach pledging his allegiance to Rob with Russell looking on, followed by a cut to Russell alone, leaning frustrated against a tree trunk, looking at Coach and Boston Rob bonding and having a good time outside the frame. Only after these shots do we get to see the authentic communicative situation—Russell’s interview monologue to the camera. I take this voice-over use of confessionals to be a move toward “interiorizing” the confessional voice, as turning natural communication— the interview—into internal, non-communicative discourse (cf. Nielsen 2011 on the non-communicative potential of literary fiction). Through the double exposure of visual focalization (footage showing Russell looking at Rob and Coach) and the after-the-fact confessional, the explication of social relations becomes a representation of immediate private experience. Patrick Keating (2013) analyses the narrative dynamics of the reality game show Project Runway, which also features confessional interviews and their innovative editing. For Keating, the interviews are potentially “­fallible or dissembling, creating a […] barrier to viewer understanding” (ibid., 61). Keating goes on to discuss the “temporal ambiguity” created by cuts from scenes of social anxiety or confrontation to interview footage where the player relives or “feigns or performs” (Keating’s words) his emotions, using the present tense and an expressive style but obviously speaking after the fact (ibid., 62). Keating is surely right to point out that the art of reality TV hinges on the ability to make discontinuous events seem simultaneous—this is precisely what is going on in the voice-over example d ­ iscussed earlier: an interview shot after the fact (Russell witnessing Rob and Coach bonding) is made to seem like a presenttense reflection on the situation. Yet, unlike Keating, I would not consider this reliance on interviews in terms of restricted omniscience, as limited access to mental content, or as a poignantly “deceptive” editorial move. Furthermore, as Keating (2013, 67) points out, one remarkable narrative feature of reality shows is that they can project possible future events through an artful montage of actual scenes. Keating discusses such proleptic editing as an emplotment strategy, but I would like to add that this type of emplotment is almost always infused with experientiality: a typical glimpse of what’s “coming next” is motivated through an individual player’s hopes, fears, or intentions. As Haralovich and Trosset (2004) point out in their discussion on the crucial role of unscripted chance in Survivor, the generic pact of reality shows assumes authenticity, and any proof of manipulation through scripting will result in a break in that contract and the audience’s dismissing the show. However, the overall dynamics of competitive reality TV shows rather suggest that confessional interviews are, in fact, to be taken at face value within the genre: the generic code marks the confessional interviews as consciousness representation.

248  Maria Mäkelä Then again, even the discursive genre of confession in general can be seen as circling around the missing “original emotion” that the confessor ostensibly reveals to her audience. As Peter Brooks argues, both real and fictional confessions are verbal performatives that actually create the inwardness of the person confessing (Brooks 2000, 2). Brooks’s view on confessional ­discourse finds empirical support in interview research conducted in the social sciences: as Jaber F. Gubrium and James A. Holstein (2002, 16–25) demonstrate, an interview situation set up to provide complete freedom of expression for the interviewee and to encourage him to “tell his own story in his own voice” will result in disturbingly generic and schematic life narratives because the interviewee is likely to resort to culturally available frames for narrativization. Yet such a culturally determined narrative of one’s “own” experience is not automatically misleading—rather, it is a welcome reminder of the fact that the narrativization of one’s experience is always a negotiation with conventions. With this critique of assumed “authenticity” in mind, we may consider how confessional interviews construct ­mental states, instead of reflecting or misrepresenting them. Thus, I would like to claim that in Survivor, the primary interpretive dichotomy suggested by genre dynamics is not that of expressed intentions and emotions versus the privacy of an individual’s mind. Rather, the crucial tension is built between the intermentally constructed and the confidentially expressed. Here, we can momentarily resort to structuralist thinking: if meaning is generated through difference, then in reality TV shows, the significant difference is created by opposing the pseudo-directional confessionals (as the “truth”) with the reciprocal communication between the players (as “lies”). Survivor is therefore definitely worth analysing as an emblem of contemporary media culture, where experiential discourse arises from the conflict between the private and the public. In addition to individual interviews typically shot at the deserted beach or in some other contemplative space, there is also another confessionary genre in Survivor—the voting confessional. At the tribal council, the players take turns walking across a rope bridge that leads to the voting place; here each player writes down a name on a piece of paper, and sometimes we get to hear a voting confessional as the player shows the name to the camera and whispers a message to the person he or she is voting off. The voting confessional is montaged with a close-up of the person being voted for, usually showing his or her worried face to create an uncanny feeling that this person actually hears the message. Here an iconic moment from the ninth episode of the seventeenth season of the show, S­ urvivor: Gabon, must be mentioned: Chrystal and almost all the other ­players want to get rid of Randy. Some harsh words are whispered as Randy’s name is written down, and also Chrystal starts her “private” monologue to the camera without the other players hearing and yet suddenly raises her voice and ends up shouting: “You have made my life hell!” As expected, close-ups of exaggeratedly stunned and amused faces among the players and the jury follow, suggesting that some aggressive parts of the

Mind as World in the Reality Game Show Survivor   249 private confessional actually reached the other players’ ears. Such distortion of the convention highlights the quasi-communicational nature of the confessional voice in reality TV: people “privately” either shout or whisper to the camera while the rest of the players make faces in the background. At least for me, this resonates with some of the consciousness representation techniques in the novel. Something that is originally private and incommunicable takes on a grotesquely material shape.3 Nevertheless, this does not mean that we should treat reality TV as ­fiction; nor should we consider narrative fiction and reality shows to be using exactly the same modes of narration. The analogy between literary ­fiction and reality TV is only a methodological choice that enables me to demonstrate that the genre is, in fact, capable of foregrounding and thematizing quite specific aspects of the human mind, namely the potential to construct far-reaching hypotheses about others’ intentions and also the problematic relationship between experience and expressivity, mind and matter. THE WORLD: METONYMIC EXTENSIONS OF THE MIND GAME The narrative dynamics of Survivor support Marie-Laure Ryan’s (1991, 148–74) theory that tellability results from the complexity of virtual outcomes: the more forking paths and alternative scenarios the story maps, the more point there is in telling the story. A radical view of the show would suggest that, in fact, the plot and the actions rest on nothing but virtuality: there would be nothing tellable in Danielle’s just being voted off—even if that came as a surprise to her and to some others; the genuinely tellable narrative results in all the embedded, projected, and anticipated intentions and reactions at work before and after the voting. Consequently, the “story,” told in one Survivor episode or over the entire season, is dictated by the intermental domain that has its abundant manifestations in the discourse or the narration of the show and, as I will go on to argue, also in its material storyworld. Ryan (2006, 68) argues that “Survivor openly exploits its own power to create behaviors.” She draws an analogy between reality TV and scientific experiments with artificial life (ibid., 74–75). For Ryan, the initial conditions of the system are the personalities of the characters (ibid., 75), which makes perfect intuitive sense, especially from the point of view of production, where casting plays a crucial role. Yet if we look for the initial narrative conditions of the show, including the discursive environments provided by the show—the confessionals, the tribal council, the physical challenges, the tactical negotiations at the camp—they provide the boundary conditions for “artificial life” to evolve in a narratively engaging direction. The experimental laboratory of Survivor is not only psychological but also discursive. Then again, one cannot tell psychology and discourse apart in the show, just as the material storyworld forms an elementary part of both of these phenomena.

250  Maria Mäkelä It is precisely the close interconnectedness of psychology, discourse, and the material elements of the game that shapes the show into a carefully designed laboratory of mind attribution. The holistic constructedness of Survivor’s setting and events justifies, to my mind, the use of the concept storyworld. As Ryan points out in her chapter in the present volume, within narratology there are two perspectives on storyworlds, the cognitive and the ontological. Different research questions arise from these two paradigms. Cognitively speaking, we may ask: does the viewer of Survivor understand the setting and the actions of the show to form a universe of its own, an autonomous domain (“a laboratory”) that becomes meaningful precisely because it is different from the viewer’s own experiential plane? My answer is: yes. Ontologically speaking, a reality show turns out to be a tough nut to crack; there seems to be no point in trying to demonstrate that, for example, when the contestants, citizens of the real world, appear on-screen, they become part of a fictional domain. From the point of view of generic discourse and the generic potential for thematization, the cognitive perspective on storyworlds appears to be a more workable approach than the ontological perspective. The physical reality of Survivor consists of challenges, where either two tribes or individual players compete for Immunity or Reward (which can be food, entertainment, camping gear); robinsonian activities, such as building a hut for one’s tribe; and ritualism, as seen in items such as the hidden immunity idol and the scenes such as the tribal council. Of course, the framing physical reality is that of an island far from civilization, where the “castaways” try to survive extreme conditions. The island, the ­battle and the new-age spirituality in which all the actions are wrapped, all appear as metonymic extensions of the “mind game.” Some of the c­ hallenges may single out the physically toughest or most skilled, and some challenges— such as puzzles—call for spatial and geometric understanding, but in the overall dynamics of the game, the challenges seem strikingly arbitrary. The only true motivation for the challenges is their effectiveness in dramatizing tribe dynamics and in stirring social conflict within the tribe that loses the challenge—or, in the case of individual challenges, to maximize interpersonal tension and controversy. For this reason, typical footage from a challenge—instead of showing a comprehensive view of the actual performance—focuses on the reciprocal dynamics of facial expressions that indicate determination, frustration, or disappointment. In other words, the physical dimension of the game is also harnessed to support the performance of mind attribution. The hidden immunity idols are especially effective in embodying embedded intentional stances. In the tenth episode of Heroes vs. Villains, a manifold blindside with two idols takes place: J.T. mistakenly assumes that a strong women’s alliance dominates the Villains group and he wants to save Russell with an idol he has found. However, the audience knows that it is precisely Russell who dominates the group. Russell hands the idol that J.T. has given

Mind as World in the Reality Game Show Survivor   251 to him to his ally Parvati, whom he wants to save from the vote. Russell does not know that Parvati already has an idol of her own—­information she shares only with Danielle as a demonstration of their mutual trust. At the tribal council, Probst prepares the ground for the magnificent blindside that the production crew knows is coming: PROBST:  J.T.,

has there been any talk in your group about who might be having hidden immunity idols, if there are any in the game at all? J.T.:  The idols are a big topic in this conversation. It’s a mess, and I don’t have any idea who might have an idol or who’s got one. I know I don’t have one, so I’m pretty worried about it. PROBST:  Parvati, a topic of discussion on the Villain tribe? PARVATI:  Yeah, definitely, I mean we’ve seen what idols can do. As far as on the Villain tribe, it broke completely a solid alliance and turned the game. Before the votes are counted, Parvati risks her own position in the game and hands the idols to two of her allies, Jerri and Sandra. This solution turns out to be ingenious: votes cast against the two women do not count, and the one finally voted off is J.T., the player who originally thought he was making a bold tactical move by giving the idol to Russell. Parvati proves to be the master of embedded intentions, humiliating not only J.T. but consequentially also Russell, her current ally but future enemy.

Figure 13.2  Parvati Shallow giving the immunity idol to Jerri Manthey in the tenth episode of Heroes vs. Villains.

252  Maria Mäkelä

Figure 13.3 Heroes Rupert Boneham, Amanda Kimmel, Colby Donaldson, and Candice Woodcock expressing their disappointment due to Parvati Shallow’s blindside in the tenth episode of Heroes vs. Villains.

The hidden immunity idol therefore not only represents immortality; in itself, it carries all the virtual possibilities of its strategic use and the possible consequences that follow from those uses. Yet another twist of intentions is brought into play in some Survivor seasons: a player carves a fake hidden immunity idol out of wood or a coconut in order to mislead the other contestants.4 Thus, the sheer quantity of embedded intentions embodied by the idol makes it difficult to translate these intermental dynamics into everyday human experience. This surely is not what “we all do with real people,” even though that is what Zunshine’s (2012) study on theory of mind in popular culture suggests. Yet, interestingly, Zunshine’s (2006) earlier formulation on narrative fiction’s ability to challenge our capacity for mind attribution and for tracking down several levels of intention (“Parvati wants Russell to think that she is in control although Russell is able to fool J.T. to think that Russell is scared of being voted out” and so on) seems analogous to the disturbingly manifold intentionality that is at work in Survivor. The crucial difference between literary fiction and a reality TV game show is that whereas the former is often an ambiguous mixture of authorial and figural voices that call for an interpretation of textual markers in order for us to be able to attribute voices to agents, the latter perverts, materializes, and performs this interpretive move in its own discourse. CONCLUSION: MANIPULATED REALITY, DISTORTED MIND Are the standards of what constitutes a good narrative changing as ­personal experiences and emotions are increasingly treated as information in

Mind as World in the Reality Game Show Survivor   253 contemporary media culture? At least the canonized cognitive-­narratological definitions of narrative as “mediated experientiality” (Fludernik 1996, 12–13, 28–30) or as depending on qualia (“what is it like to be X”; Herman 2007, 256–57) or the “raw feels” of an individual (Herman 2009, 152–53) seem to be synchronized with this trend. Brian McHale aptly criticizes cognitive narratology for exaggerating the connection between narrative and consciousness: “If narrative is essentially identical to mental functioning one is tempted to wonder what, if anything, is left over. Is there anything in a narrative that is not mind?” (McHale 2012, 119; cf. Palmer 2011, 202). Thus, it becomes all the more important to distinguish media- and genrespecific strategies for foregrounding and thematizing different aspects of the mind. What makes Survivor an especially pertinent test case for mindrelated intermedial narratology is that we can sincerely ask: is there anything in Survivor that is not mind? Moral concern looms large in public debate on reality TV and even in some research conducted on the genre: a fairly recent socio-psychological study, for example, demonstrates how long-term viewers of Survivor are considerably more exposed to antisocial behaviour (aggression, malevolence, etc.) than viewers of other reality-based programmes such as the news (Wilson et al. 2012). Yet in their popular collection of essays, The Psychology of Survivor (2007), several American researchers on psychology and the social sciences demonstrate how the game serves as a laboratory for family dynamics, sociopathy, and ostracism, to name but a few topics. Survivor dramatizes the more general tendency in contemporary media culture—for example, in social media—to immerse oneself in the uneventfulness of life and to turn one’s own mind and the minds of others into stages of battle and survival. Yet different narrative genres thematize and challenge our mind-reading tendencies in different ways. Reality TV, like any other type of mediation, has its own limitations and possibilities that foreground, and consequently also distort, the minds and the world presented.

NOTES 1. In literary narratology where I come from, the notions of intermental thinking or social mind have been brought to discussion by Alan Palmer. An entire issue of the journal Style (25:4, 2011), is dedicated to both supportive and critical responses to Palmer’s call for a narratology that would recognize the centrality of intermental units in novels, that is, cognitive units that extend between several individual minds. Palmer’s arguments rely heavily on the study of theory of mind (introduced later in this chapter in connection with Lisa Zunshine’s theories) and they also draw a heavy analogy between intermental thinking in actual communication and collective minds represented in literary fiction (see Palmer 2011). A relevant response for my present concerns is the one by the philosopher of mind Daniel D. Hutto, who criticizes Palmer for (1) drawing too easy an analogy between actual and fictional minds (Hutto 2011, 277; see also Mäkelä

254  Maria Mäkelä 2013) and (2) for exaggerating the complexity of mind attribution in everyday social contexts. Palmer’s approach does not leave room for media and genre specificity in the analysis of mind attribution, whereas Hutto’s more nuanced perspective allows for differentiation of automatic, effortless mind attribution and more speculative “theorizing” in complex social situations. Unfortunately any closer dialogue between Hutto’s nuanced mind-philosophical theories and Survivor falls outside the scope of the present chapter. 2. Some confessionals appeared only as extra material on the CBS website when the show was being aired. 3. In fact, a more plausible analogy can be drawn between reality confessionals and the soliloquy in drama. A comparative analysis of Shakespeare’s plays and Survivor needs to be developed in some future context. 4. A concise account of the strategic uses of idols can be found at http://survivor. wikia.com/wiki/Hidden_Immunity_Idol.

REFERENCES Brooks, Peter. 2000. Troubling Confessions: Speaking Guilt in Law and Literature. Chicago & London: The University of Chicago Press. Dennett, Daniel C. 1987. The Intentional Stance. Cambridge, MA: MIT Press. Fludernik, Monika. 1996. Towards a ‘Natural’ Narratology. London & New York: Routledge. Gubrium, Jaber F. and James A. Holstein. 2002. Handbook of Interview Research: Context and Method. Thousand Oaks: Sage. Haralovich, Mary Beth, and Michael W. Trosset. 2004. “Expect the Unexpected: Narrative Pleasure and Uncertainty Due to Chance in Survivor.” In Reality TV: Remaking Television Culture, edited by Susan Murray and Laurie Ouellette, 75–96. New York: New York UP. Herman, David. 2007. “Cognition, Emotion, and Consciousness.” In Cambridge Companion to Narrative, edited by David Herman, 245–59. Cambridge: ­Cambridge University Press. ———. 2009. Basic Elements of Narrative. Oxford: Wiley-Blackwell. Hutto, Daniel D. 2008. Folk Psychological Narratives. The Socio-Cultural Basis for Understanding Reasons. Cambridge, MA: The MIT Press. ———. 2011. “Understanding Fictional Minds without Theory of Mind!” Style 45.2: 276–82. Keating, Patrick. 2013. “Narrative Dynamics in the Competitive Reality Show.” ­Storyworlds: A Journal of Narrative Studies 5: 55–75. Mäkelä, Maria. 2011. “Masters of Interiority. Figural Voices as Discursive ­Appropriators and as Loopholes in Narrative Communication.” In Strange Voices in ­Narrative Fiction, edited by Per Korgh Hansen, Stefan Iversen, Henrik Skov Nielsen, and Rol Reitan, 191–218. Berlin & Boston: de Gruyter. ———. 2013. “Cycles of Narrative Necessity: Suspect Tellers and the Textuality of ­Fictional Minds.” In Stories and Minds: Cognitive Approaches to Literary N ­ arrative, edited by L. Bernaerts et al., 129–51. Lincoln, NE: University of Nebraska Press. McHale, Brian. 2012. “Transparent Minds Revisited.” Narrative 20.1: 115–24.

Mind as World in the Reality Game Show Survivor   255 Nielsen, Henrik Skov. 2011. “Unnatural Narratology, Impersonal Voices, Real Authors, and Non-Communicative Narration.” In Unnatural Narratives – Unnatural ­Narratology, edited by Jan Alber and Rüdiger Heinze, 71–88. Berlin: De Gruyter. Palmer, Alan. 2011. “Social Minds in Fiction and Criticism.” Style 45.2: 196–240. Ryan, Marie-Laure. 1991. Possible Worlds, Artificial Intelligence, and Narrative Theory. Bloomington: Indiana UP. ———. 2006. Avatars of Story. Minneapolis: University of Minnesota Press. Survivor: “Gabon.” Season 17, episode 9 “Nothing Tastes Better Than Five Hundred Dollars.” Prod. Mark Burnett, David Burris, Charlie Parsons and Jeff Probst. CBS broadcast Nov 20, 2008. Survivor: “Heroes vs. Villains.” Season 20. Prod. Mark Burnett, David Burris, Charlie Parsons and Jeff Probst. CBS broadcast Feb 11 – May 16, 2010. Wilson, Christopher, Tom Robinson, and Mark Callister. 2012. “Surviving Survivor: A Content Analysis of Antisocial Behavior and Its Context in a Popular Reality Television Show.” Mass Communication and Society 15.2: 261–83. Zunshine. Lisa. 2006. Why We Read Fiction: Theory of Mind and the Novel. ­Columbus: Ohio State UP. ———. 2012. Getting Inside Your Head. What Cognitive Science Can Tell Us About Popular Culture. Baltimore: Johns Hopkins UP.

14 Performing Selves and Audience Design Interview Narratives on the Internet Jarmila Mildorf

The last few decades have seen a proliferation of oral history databases that can be accessed on the Internet. Typically, people tell stories from their lives, and those stories are made available as audio files or videotaped recordings, as well as short transcripts. If one listens to these stories one may come across narrative features that make one think about the stories in relation to the minds involved in the storytelling process. For example: how are characters’ minds presented in a narrative genre where access to other people’s minds is generally not deemed possible? How do storytellers engage their interlocutors’ minds and draw them into the storyworld? And in what ways does the fact that the stories are published online have an impact on the shape of these stories (because surely there must be some influence)? Sociolinguistics and discourse studies—two disciplines where such stories are researched—do not usually ask questions about the mind (however, cf. Sanders 1987). Instead, researchers focus on “talk-in-interaction” and on what people “do” when they talk to one another. However, there seems to be a tacit understanding that we somehow “know” or are able to infer what someone thinks, wants to achieve, or “really wants to say” when saying something. Even in lines of research where scholars very adamantly try not to connect their linguistic investigations to mentalistic terminology, e.g., in discursive psychology, it is surprising to see how much “mind work” still goes into the actual analyses and interpretations of conversational interactions1. As Sanders (2005, 77) has it: Discourse studies inescapably rest on assumptions about what persons cognitively would and could do with their words in any given instance. Such assumptions are built into every observational claim we make. While we can go about our business most of the time as if this weren’t so, sometimes the only way to ground and justify, or challenge, observational claims in discourse studies is to attend to the assumptions about cognition on which they rest. In this contribution I flesh out some ideas about how a more interdisciplinary approach can contribute toward an understanding of the workings of “minds-in-interaction” in conversational storytelling. Using a

Performing Selves and Audience Design   257 blend of concepts and analytical tools from sociolinguistics, narratology, sociology, and psychology, I first offer some theoretical considerations: I reflect on the nexus of narrative and identity and on how the Internet as a new medium has influenced this nexus. Then I focus on the interview and especially the narrative interview and oral history interview as special communicative ­situations, as well as on the “mind work” speakers and recipients in such interviews undertake. Finally I present analyses of selected interview narratives from three oral history databases to explore in more detail how speakers engage the minds of their interlocutors and, by extension, of the Internet audience in and through their narratives and thus invite them to follow, understand, and empathize with their life stories. NARRATIVE, IDENTITY, AND PERFORMANCE The literature on the relationship between narrative and identity is vast and reaches across numerous humanities and social science disciplines. I will therefore limit myself to some basic questions and give special consideration to the Internet as a more recent medium where narrative identity or, indeed, “narrative identity” (Ricoeur 1988, 246), also plays a major role. McAdams, Josselson, and Lieblich (2006, 4) describe narrative identity as “the stories people construct and tell about themselves to define who they are for themselves and for others.” In psychology, the question of narrative identity is debated along three axes: 1 Unity or multiplicity: The first axis concerns the extent to which narrative identity supports a sense of unity and integration of the self rather than multiple and sometimes conflicting aspects of the self. 2 Self vs. society: The second axis addresses the question whether narrative identity is regarded as an achievement of the individual or as ­constructed in and through a psychosocial context. 3 Stability vs. growth: The third axis considers the idea of a core self that remains stable over the course of a lifetime in contrast to notions of change, development, and growth. (McAdams, Josselson, and Lieblich 2006, 5–9) To my mind, none of these axes presents an either/or dichotomy. Instead, the various aspects should be considered as points on a continuum that can vary for each individual and across a life span2. In this contribution, I am particularly interested in the situated, interactional, and performative aspects of narrative identity. After all, storytelling in everyday life is a special kind of discursive practice, and it is central to the creation of narrative identity because it offers the possibility of creating coherence, of presenting linguistically one’s experience of time, and of re-enacting previous life experiences in the here and now of the telling (Lucius-Hoene and Deppermann 2002, 53).

258  Jarmila Mildorf On a more fundamental level—and metaphorically speaking—one can also argue that stories become a protective shield for our inner selves. Charlotte Linde (1993, 121), for example, writes: Our own internal experience, if we permit ourselves to notice this, is of a self without armor—perhaps without boundaries as definite as we would like—walking around in a world of others who appear to have proper boundaries and effective armor. Hence we may perceive ourselves to be in an alarmingly vulnerable position that must be ­remedied. And the remedy is to narrate, to create a self as other, replicating our experience of the actual others we seem to experience. It is thus not surprising that frequently our stories justify or explain our actions and that we use them to present ourselves in a certain light—to others but also to ourselves. These stories of self can take various shapes. A number of studies have emphasized that narrative identity must not only be sought in elaborate and often lengthy life stories usually elicited in life story interviews but also and especially in the kinds of stories people tell one another on a daily basis. Pasupathi (2006), for example, focuses on the collaborative construction of everyday selves in conversational storytelling. Likewise, Ochs and Capps (2001) talk about “embedded stories” and Georgakopoulou (2007) about “small stories” to describe a somewhat neglected narrative genre: the sometimes fragmented, incomplete, and often collaborative pieces of stories that become part of the conversational flow. When people tell stories they first decontextualize their past experiences in order to re-contextualize them in the given storytelling situation. And they accommodate their stories to the situation at hand, reacting to their interlocutors as well as following their own agenda of self-presentation (Günthner 2005). As sociolinguistic research has amply demonstrated, speakers use stories to create professional, ethnic, and gendered identities for themselves (see contributions in Thornborrow and Coates 2005). Their stories also mark their participation in social groups such as the family (Langellier and Peterson 2004), the workplace, or other institutions (Linde 2009). In this sense, stories contribute toward people’s “performances of self,” as Erving Goffman (1959) put it. In performing their selves, people may or may not draw on existing cultural templates, and this is where the media play an increasing role; it is here that identities are presented, staged, and discussed (Lucius-Hoene and Deppermann 2002, 50). NARRATIVE IDENTITY AND THE INTERNET Digital media have changed the way narratives work. Marie-Laure Ryan (2005, 516) discusses the extent to which the very properties of digital systems (interactivity, volatility, and networking) are not conducive to ­

Performing Selves and Audience Design   259 narrative meaning in a classical sense3. Even the more mundane personal stories people tell on personal websites, in web logs and chat rooms, for example, have been influenced by the media. A recent study by Ruth Page (2012) investigates new social media such as discussion forums, social ­network sites, twitter, web logs, and wikis, among others, and draws attention to the ways in which the process of telling stories as well as the stories told have changed. Thus, many stories web users produce combine features of orality and writing, make use of several medial forms (pictures, videos, sound, and writing), resemble more the embedded or small stories mentioned above and are interactive to the extent that other web users can comment on a story. The example of personal web logs shows the hybrid nature of Internet stories: blogs draw on “online forms of communication such as e-mail and personal web pages, along with offline genres, particularly diary writing and autobiography” (Page 2012, 51). Stories in blogs thus become more personalized than “official” stories but are also less retrospective and less monolithic than autobiography, for example. Arguably, one of the main functions of storytelling on the Internet is visibility or getting attention. Psychological research on interpersonal perception has shown that the Internet with its possibilities for creating personal websites and using social network forums is a new key to impression management and to the transportation of information about website owners’ personalities (Krämer and Winter 2008; Vazire and Gosling 2004). For example, Marcus, Machilek, and Schütz (2006, 1030), who correlated self-evaluations of website owners to evaluations by unknown visitors to websites and to the websites’ designs, found out that “meaningful inferences of personality could be derived from visiting a personal Web site for just 5 minutes.” While the rhetoric of the social media industry has from the beginning emphasized positive values such as “democracy, participation, collaboration, and dynamic change” (Page 2012, 206), there have also been more critical voices. Ramón Reichert (2013), for example, stresses the fact that storytelling becomes a means for seeking attention and that it is by far not liberal and liberating because the different kinds of user software impose their own specific rules. Rather than offering individuals room for communicating who they are, Reichert (2013, 531) argues, digital storytelling formats already influence the “selves” expressed in them. Page (2012, 207) further contends that “[f]ar from a utopian, collaborative environment where all tellers and their stories are valued equally, the narrative dimensions ­ arket of social media suggest a hierarchical system based in a linguistic m where visibility and interaction are the prized values.” These considerations are relevant for the kinds of data I present below to the extent that the interviews are published on the Internet. Even though the speech format of the narrative interview basically remains the same, the outcome is different because participants are aware of the fact that the ­interview will appear online and will thus be accessible for a large audience, possibly including friends, family, and acquaintances. What happens to Labov’s

260  Jarmila Mildorf “Observer’s Paradox”4 when research subjects not only become aware that they are being systematically observed but also that what they say will be eternally out there on the world wide web? (cf. Etzersdorfer 1987, 53) THE INTERVIEW: A SPECIAL COMMUNICATIVE GENRE Interviews are not simply “conversations” or even “chats” with another person, although they do follow some basic rules of ordinary conversation (Riessman 2008, 23–24). Thus, only limited topics are covered, and the ­ ­conversation normally progresses smoothly because the participants adhere to turn-taking rules (Sacks, Schegloff, and Jefferson 1974) and because they use linguistic cues such as questions and self-repairs to clear up misunderstandings (Schegloff 1992). In contrast to ordinary conversations, the interview is not only recorded, but it is also usually guided by the researcher. As Mishler (1986, 245) contends, “[T]he aim of an interview is defined by the interviewer who also controls its shape and flow as well as the form and intent of specific questions.” However, so-called “narrative interviews” aim at providing narrative opportunities, and interviewers will therefore try and yield the floor to the interviewee for as long as possible. Oral history interviews published on the Internet are mostly secondary data in the sense that other researchers have prepared the transcripts and have decided which parts of the data to present and how to present them. This raises “interpretive issues,” as Catherine Kohler Riessman (2008, 22) points out, “including imagined audience and other contexts implicated in production.” What Riessman says about work in archives is partially also true of work with internet narratives: “Documents do not speak for themselves; decisions by the author and/or archivist have already shaped the texts an investigator encounters” (Riessman 2008, 22–23). Two of the databases I use for this contribution (healthtalk.org and StoryCorps) do not provide complete interviews but only short excerpts on their websites. Here, a thematic selection has already been made. At least these websites make recordings available where possible. The third database, the Smithsonian Archives of American Art, provides longer transcripts of interview excerpts but only snippet audio file samples. The transcripts have been tidied up, i.e., they contain no information on speech patterns or prosodic features. Furthermore—and this applies to all of these databases—not only did I not have access to the original experience that was narrativized in the interview situation (see Riessman 2008, 22), but I also did not have direct access to the interview situation in the sense that I had not conducted the interviews myself. This means that my interpretations are not once but rather twice removed. One needs to bear these points in mind when considering the data. Another aspect that is relevant in this context is audience design. In sociolinguistics, the term “audience design” initially emerged in the context of sociophonetic variationist research (Coupland 2007, 58–62). Allan Bell

Performing Selves and Audience Design   261 (1984) investigated radio news broadcasts in New Zealand and discovered that the same news readers used different phonetic styles depending on whether they read for national radio or a smaller community radio station. He used the term “audience design” to account for this finding. Duranti (1986, 243) emphasizes the importance of taking into consideration the audience for whom utterances are produced, and he even talks about “­co-authorship,” a notion that by now has widespread currency in sociolinguistics. Duranti also points out that we need to “recognize the informative function that certain utterances have with respect to hearers and bystanders—as opposed to addressees” (Duranti 1986, 243). The idea that things might be said for other people rather than the immediate interlocutor will be interesting for my examples, as one often forms the impression that interviewees tell things for the benefit of a larger Internet audience. In a similar vein, Irene Etzersdorfer (1987, 57) suggests in her theoretical considerations concerning oral history interviews that there is what she calls a “doppelter Dialog” / a “double dialogue” in the interview situation: between the historian and the interviewee or informant on the one hand and between the informant and a larger social group or society as well as its history on the other5. Informants are aware of the historical dimension of the interview, and this awareness will colour the content of their responses. Therefore, what informants say must not be taken at face value. There can also be a hidden, unconscious dimension that surfaces in what is particularly emphasized or what is left out, what is answered in response to questions never asked, what is denied, or what is distorted, and in displays of emotionality (Etzersdorfer 1987, 57). One could add to Etzersdorfer’s list more rhetorical items such as the use of metaphors and irony, for example, and of course silence as an important discursive feature. This idea of two levels of communication reminds one of Phelan’s (2005) concepts of “narrator and disclosure functions” in character-narration. THEORIZING THE COMMUNICATIVE SITUATION IN INTERNET INTERVIEWS Phelan starts out from the assumption that narrative is a means of indirection, of conveying (sometimes discrepant) messages through storytelling. He assumes that communication in character narration takes place along two “tracks,” as he puts it: “the narrator-narratee track” and the “narratorauthorial audience track.” The first track (narrator-narratee) involves the fact that “the narrator acts as reporter, interpreter, and evaluator of the narrated for the narratee.” This is what Phelan calls “narrator functions” (Phelan 2005, 12). Along the narrator-authorial audience track, narrators unknowingly communicate to the authorial audience (i.e., the readers) all sorts of things. Ultimately, it is the (implied) author who makes the n ­ arrator tell those things and thus creates a special communicative situation. Phelan

262  Jarmila Mildorf terms this “disclosure functions,” i.e., the fact that textual signals convey underlying messages that may diverge from the narrator’s ostensive messages. One may object here: How does Phelan’s narrative communication model, which is geared toward describing written narratives, map onto the kinds of interview narratives I have just outlined? And how can his model, which includes an author, an implied author, a narrator, a narratee, and an authorial audience, possibly be applied to a speech situation where we typically have two interlocutors? The point I wish to make is that speech situations in everyday storytelling are not as simple as one might expect. I already pointed to the idea that the addressee can be accompanied by other people who are overhearing the talk or audiences that are implied. In the interviews I describe here, there is usually a dyadic setup. However, some of the narrative features I discuss in my examples show that, even though the Internet audience was not physically present, speakers accommodated to that audience in what they said and how they said it. Likewise, the notion of “speaker” or “narrator” in everyday storytelling is often taken to be self-evident: the person telling his or her story is author and narrator at the same time and, in cases of stories of personal experience, also becomes a character in the story. However, Erving Goffman already pointed out in his book Forms of Talk that the notion of “speaker” can be more complex. He distinguishes between the “animator,” i.e., the person giving voice to an utterance; the “author,” who selected what is said and how it is said; and the “principal,” i.e., someone whose position or beliefs are expressed through the words that are spoken (Goffman 1981, 144). An obvious example would be a strategic political speech where a politician might read out a text that has not been scripted by himself and that expresses the values and opinions of, say, the political party behind the politician rather than his own. I think Goffman’s division can be found in more mundane examples. For instance, learners of a foreign language often use phrases and expressions they have heard and then recycle them in conversations, thus transferring passive into active vocabulary. Married couples or other people living together use words and express beliefs that originated from the other person. And one can think of situations where people may use words that express other people’s world views because they are pressurized or fear repercussions. Phelan’s model may, if suitably adapted, be useful for analysing specifically narrative speech situations. When I tell a story about myself I draw my interlocutor’s attention to a particular image of myself that may be more or less explicit. One could call that “implied self” in analogy to the literary term “implied author.” In fact, “implied self” seems to be even more appropriate because, as I already mentioned, the question of “authorship” can be a tricky one. Whether the self I create and transmit through my stories provides an image I also consciously authored or whether it is conveyed without my realizing it may not always be clear. Indeed, I may not be the

Performing Selves and Audience Design   263 author of my own story; I may instead be recycling cultural stories of self, for example, or a story someone else authored for me. And the implied self, the image of myself I create through storytelling, will vary to a certain degree from one storytelling situation to another. This process of accommodation and self-adaptation may even involve the use of “fictive” material. Thus, Ricoeur (1988, 246) contends that “the story of a life continues to be refigured by all the truthful or fictive stories a subject tells about himself or herself.” Too much variation or, put more negatively, inconsistency will result in others questioning my identity and integrity. In Phelan’s terms, one could say that interlocutors can become aware of a narrative’s “disclosure functions,” which—in the case of conversational storytelling—may be transmitted consciously or unconsciously. It also makes sense to distinguish between the “narrating persona” and the “experiencing persona” in oral storytelling, just as scholars in autobiography studies differentiate between “narrating” and “narrated I” (Smith and Watson 2010). I prefer the term “experiencing,” however, because, as we shall see, conversational storytelling affords people the opportunity to ­re-enact prior experiences rather than merely tell about them. Pasupathi (2006) calls this the “dramatic mode” in contrast to the “reflective mode,” which is anchored more strongly with the narrator. Against this background, it might even be more accurate to assume a three-partite division of the speaker’s role in conversational storytelling6: the “narrating persona,” who tells the story in a given communicative situation; the “narrated persona,” i.e., the version of the speaker within the presented storyworld; and the “experiencing persona,” i.e., the persona who experienced a situation in the past and partially re-experiences it in the present storytelling situation. Put differently, the “experiencing persona” in fact allows for a merger of narrating and narrated personae in the dramatic mode. In the following examples of interview narratives, I will focus on three areas in which narrators engage their audiences’ minds: 1. in the way they create a storyworld through focalization; 2. in the way they present their own and other characters’ minds; 3. in the way they perform their stories for an implied audience and thus offer projection screens for interpretation. DRAWING THE AUDIENCE INTO THE STORYWORLD My first example is taken from the Smithsonian Archives of American Art. Interviews in this archive are typically lengthy life story interviews, and the website makes long transcripts available. Unfortunately, the sound files offer only snippets from the interviews, sometimes even with materials being left out. The excerpt below is from an interview with Rudy Autio, a FinnishAmerican craft artist, conducted by LaMar Harrington, an art-museum curator and director7. I use Labov and Waletzky’s (1967) scheme of narrative clauses for line breaks in the transcript. The transcription code used

264  Jarmila Mildorf here and throughout the chapter follows largely the Jefferson Transcription System8. Table 14.1  Excerpt 1: Interview with Rudy Autio.

1

LH: Now we also spoke yesterday about Pete having gone to Black Mountain in 1953, and the influences on him at that time. I think we didn’t talk yesterday about the Hamada-Leach-Yanagi visit in 1952. You’ve covered that in a number of conversations, I think. RA: Yes. LH: But if you can stand to go through it again, it ... RA: Well, no problem. I may probably remember different things, in talking with you now.

2

But a lot of this is fading into the distant past now,

3

so I can’t be too clear on everything that happened there now.

4

The things that seem most memorable to me was the ease with which Hamada worked on the wheel.

5

I think this was the most beautiful part of it.

6

To watch Hamada work on the wheel was just (.) beautiful.

7

It was the economy of everything he did.

8

The feel he had for clay, which can’t ↑really be described,

9

>he would-<

10

well, for example, if, if it took one spin at the wheel to make a mark on it, that’s all that was necessary, you know, we-

11

when we throw on the wheel we sit there

12

and crank away

13

and crank and crank and crank and crank

14

and (.) hope that it- by cranking something:: great would happen.

15

But, er, Hamada didn’t need to do that.

16

And, er, then, er, the other way, other thing that I noticed (.) the casual way he handled the pieces.

17

He just ↓sliced ↑them off the (.) hump and, er,

18

if he left an accidental finger mark on it, er, which, >you know, we would- we would’ve cleaned off with a sponge, why he just left itwe went back to the doctorpretending to be an aeroplane when he’d just been told his father was deadI said the neighbours were beginning to comment< (0.4) and show concern. >And she said “Just because your neighbours think he’s got Alzheimer’s doesn’t mean he has.”< (1.3) .hh And I said (0.3) I wanted a second opinion. (0.8) And she was very angry (0.3) and she said (0.2) “You’ll get the same message from (( ))” but, yes I could see (.) a psychiatrist. (1.3) And so he saw (.) a psychiatrist (0.4) who said to me (.) he thought the problem was neur- neurological. (0.9) And I ↑read the Alzheimer’s (0.4) uhm, News and by ↑chance (0.6) there was (.) an article about Professor ((name)) clinic in ((place)). And I phoned him and said “Can I bring ((husband’s name))?” And he said “Yes” (0.6) and (.) told me how to do it through the NHS (.) and my doctor cooperated. (1.0) .hh And I’ve told that story in detail (0.2) because (0.4) I consider that was the first (0.3) unethical (1.4) thing that was done (0.3) in that that neurologist knew (0.7) he had Alzheimer’s, >knew he had a dementia< (0.8) because (.) when she wrote (.) to ((professor’s name)) she told him so. (1.4) And (1.2) had we been told the truth in the first place things would have (0.7) worked out very differently. (1.2) Uhm, I ↑can’t pass a comment on >whether it would have been better or worse< (0.4)

268  Jarmila Mildorf 28 29 30 31

but >((husband’s name)) certainly would have been here< (0.5) and we would have (.) followed a much more conventional path. (2.2) .hh Uhm (1.4) they were marvellous to me (.) at the (0.6) clinic (0.8) a::nd ((professor’s name)) talked to me at (length, er, 0.9) about (1.2) preserving life or not, >quality and quantity of life.< (0.6)

32

.hh And essentially said (0.6) “Let him take all the risks he wants to. (0.8) If he’s knocked down by a bus what does he lose? (1.2) Just (.) years of (0.6) gathering dementia, uhm, so (.) let him do as much (.) as he wants to do.” And I said (0.6) “I have thoughts about things that coroners sometimes ((laughs)) say.” And he said >“If it ever came to that I would support you.”< (0.5) And so (0.9) I (0.4) allowed and encouraged ((husband’s name)) >to ride a bicycle< (1.0) for as long as was possible (0.5) .hh uhm, and I kept him (0.6) out (0.4) of full time care (0.4) for as long (.) as was possible (0.6) because he was a ve↑ry ↓prou:d and very independent man (1.1) a:nd (0.9) I felt (.) that was what he (0.3) what he would want.

33 34 35 36 37 38

The first thing to notice when listening to the audio file is that the speaker talks in a very measured tone, weighing her words and using marked pauses to structure her sentences and to accentuate single words. The more noteworthy are moments in the narrative where the speaker’s speech is sped up, thus enlivening the narrated events for the interlocutor. This happens, for example, when the speaker in the beginning relates her visit to the neurologist (lines 1 to 14). The narrative contains another, embedded narrative, namely the story of how the speaker’s husband behaved when he heard about his father’s death (line 8). By increasing the speed of the narrative, the speaker renders it more dramatic (Pasupathi 2006; Tannen 1989). This is further achieved through the use of direct speech, e.g., when the speaker relates seemingly verbatim how the neurologist responded to her concerns (lines 7, 10, and 13). Tannen (1989) calls this “constructed dialogue,” thus taking into account the fact that the presented dialogue is as much constructed within the given speech situation as it allegedly presents an original speech situation. This is the kind of “decontextualizing” and “re-contextualizing” I referred to earlier. Günthner (1999, 704) analyses the use of prosody in constructed dialogue and assigns four functions to it: (i) to contextualize whether an utterance is anchored in the reporting world or the storyworld. (ii) to animate the quoted characters and to differentiate between the quoted characters. (iii) to signal the speech activities and the affective stance of the reported characters. (iv) to comment on the reported speech as well as on the quoted characters.

Performing Selves and Audience Design   269 The last two functions are particularly interesting for the study of minds in conversational stories (see also Mildorf 2008). In “repeating” what the doctor said to her, the speaker in our example also assumes a reproachful tone, thus enacting the doctor’s irritation for the current audience. The speaker actually says that the doctor was “very angry” (line 12) but by using constructed dialogue, she additionally signals that the doctor’s verbal reaction affected her in that situation and became memorable in a negative way. In other words, “constructed dialogue” not only fulfils what Phelan calls the “reporting” part of the narrative function here but it also communicates something of the doctor’s and the woman’s feelings at the time. On the level of the current storytelling situation, this lively rendition of the encounter also discloses the extent to which the speaker is still emotionally affected because the dramatization decreases the speaker’s distance to the events in the storyworld; compare this to a potentially more “reflective mode” in Pasupathi’s (2006) sense. More importantly, as Pasupathi (2006, 142) also points out, the dramatic mode co-opts the listener as a “partner” who is placed “in the position of simultaneously supporting the story and the proffered version of the self.” We can see this in the second example of constructed dialogue in lines 32 to 34, where the speaker reports what the professor at the specialist clinic said to her. The professor’s suggestion that the husband should be allowed to take “all the risks he wants to” (line 32) may well be viewed controversially, given that other people may have come to harm, too. In re-enacting her own scepticism at the time of the consultation (“I have thoughts about things that coroners sometimes say,” line 33), the speaker signals to her current interlocutor that she is and was aware of the potentially problematic nature of this suggestion. At the same time, however, her slightly laughing voice invites the listener to also adopt a less severe viewpoint (in case the listener’s position was more critical) and to yield to the supportive words of the professor, who, after all, spoke as a person of authority. In a way, we cannot but understand the speaker’s predicament and feel sympathetic toward her. In line 23, the speaker offers a meta-narrative comment: “I’ve told that story in detail.” It is not so much a justification for holding the floor (because this is, after all, an interview, where she is by definition given more floor space) but an opener to the speaker’s criticism of the way she had been treated by the neurologist. She even calls the doctor’s withholding of a clear diagnosis “unethical”12 and explains what ramifications this behaviour had for her life. In lines 26 to 29 the speaker uses what Gary Saul Morson (1994) calls “side-shadowing”: she reflects on what might have happened had things been different in the first place13. In doing so, she engages her audience’s mind in a special way, inviting listeners to entertain the possibility of a different life and thus conveying to them a stronger sense of her own feelings of loss and frustration. She thus also confirms in an indirect way the same “moral stance” (Ochs and Capps 2001, 50) on the related events that she openly already expressed in her critical comment. Here we can see how narrative indirection can be used to justify one’s own position and to create empathy in listeners.

270  Jarmila Mildorf PERFORMING SELVES FOR AN AUDIENCE My third example is taken from the StoryCorps website14. In this oral history project, participants are encouraged to tell stories to one another rather than to an interviewer. This partially also accounts for the fact that one finds rather “odd” scenarios where interviewees tell each other stories they presumably already know. In this example, a couple remembers how they fell in love. Speaker 1 is the husband, who worked as a road manager for a famous singer, and speaker 2 is the wife, who was a back-up singer15. The co-constructedness of the story is captured in the way the speakers’ turns are laid out in this transcript. Table 14.3  Excerpt 3: Love Story. 1

S1: The plan was for me to stay at the office ’n run the road from the office [so

2

S2: [Right.

3 4 5

S1: I would train managers ’n send them up. I would sneak out on the road to see if they were doing their ↑job ’n (.) I’d been (.) tryin’ to hit on you ’n date you for yea::rs ’n you wouldn’t give me the time a day.

6

S2:

7

S1: Right. [((laughs))

8

S2:

9 10

S1: [“Gimme my check.” That’s what you sai::d.]

I didn’t ↑trust management. [So I didn’t have nothin’ to do with you. All you di::d wa- for me was [gimme

11

S2: information. Gimme my check.] And that was it bro. [((laughs)) I’m so::rry.]=

12 13

S1: It’s lonely bein’ a manager=

14

S2: =Right.

15

S1: [It’s true.

[((laughs))]

16

S2: [But our relationship was goo::d=

17

S1: =mhm=

=I know.

18

S2:

19

S2: Well, you were my counselor when I had women problems I’d come talk to you.=

20

S1: =Right, right=

=as far as bein’ able to hang out.

21

S2:

22

S1:

23

S2: =You know and I’m like (.) “Now, can I stay?” [you know

=And you’d listen so goo::d n then to send me back to my room=

24

S1: [No.=

25

S2:

=Right.=

=“Don’t you wanna hug me=

26

S1: =No.=

27

S2: =’n hold me” [’n-

28

S1: [No.

29

S2:

and you like (.) “No:, let’s pray:: or read some scripture” (Continued)

Performing Selves and Audience Design   271 30

S1: [’n] I’m like (.) “O::h ↑no.”

31

S2: [Right.]

32 33 34 35 36

S1: Well (.) you remember I trained this ↑guy ’n (.) I heard this guy was not doin his jo:b ’n he was sittin in the audience while the show was goin on ’n doin different things ’n (.) I ↑came out on the road to check up on you guys ’n (.) as I was ↑goin through the ↑venue (.) I came upon the ladies’ dressing room (.) ’n it was just a ↑law that you don’t go in the ladies’ dressing room=

37

S2: =Right=

38

S1: =e↑specially without kno:cking=

39

S2:

40 41

S1: But the ↑door was cracked (.) ’n I ↑go in (.) and there you were. Our ↑eyes meet (.) ’n I kiss you (.) ’n you kiss me back (.) [on the lips.

42

S2:

43 44

And I’m like S1: “↑IT’S ↓OVER. ↑THAT’S ↓IT.” I gotta fire this guy, alright? [Even if this guy was-

45 46

S2: come back out on the ↓road.=

47 48

S1: =Right. Even if this guy was doin’ a great jo::b he was fired that day, okay? [↑It’s ↓over.

49

S2: [O↓ka↑y:=

50 51 52 53 54 55

= I need your job man ‘cause I wanna get ((name)) S1: alright? ((spoken with laughter in his voice)) (.) ’n the ↑next morning I gave ((laughing voice)) that guy his ticket (.) ’n (.) I’m standin’ at the bus waitin’ for everyone to come (.) ’n you ↑come to the ↓bus ’n I ↑see it on your ↓lips, you’re about to say (.) “That kiss’d meant nothing.” >’n ↑I looked you in your eyes ’n before you could say anything I said< (.) “It’s too late.”

56

S2: ((laughter))

=Exactly.

[I remember that, I remember that.

57

S1: and (.) after (.) thirteen years of marriage=

58

S2: =Yes=

[Because ↑you wanna

59

S1: =[uh

60

S2: [↑thirteen years of ↓marriage=

61 62 63 64

S1: =I just, I just ↑love what we have (.) in each other, I, I’m thankful of, to, to have you in my ↑life, I mean, you, you when I wake up ’n see you, you know (.) there’s not a ugly day as long as you’re there.

The first thing to notice is the great amount of interactivity between these two speakers. Their turns are interlaced to such an extent that they almost seem to be telling the story as one voice. Falk (1980) called this phenomenon verbal “duetting,” which is quite common in couples (Coates 2005). The couple’s involvement in this conversation can be seen in numerous overlapping turns, shared laughter and in the great amount of latching, i.e., when a turn almost, but not quite, overlaps the preceding one (indicated by equals signs in the transcript). Both speakers also frequently use backchannels, i.e., linguistic items that signal an interlocutor’s attention and support

272  Jarmila Mildorf in a conversation (e.g., “right,” “mhm,” “I know”). Nevertheless, it is mainly the woman’s role here to be supportive. The husband has far longer turns at talk. Interestingly enough, he uses short instances of you-narration, addressing the story to his wife, even though she of course knows the story, having participated in it herself. Elsewhere I argue that, while such you-narration in conversational storytelling seems to support collaborative memory work, it also contributes toward a “performance of you and I” (Mildorf 2012). Indeed, a lot of the information the couple exchanges is redundant in the sense that they both know the facts anyway. So when the husband provides background information in lines 32 to 36, for example (“Well you remember …”), this is clearly more for the audience’s benefit than for the wife’s. When the husband explains in line 44 why he had to fire one of the road managers, namely because he wanted to take his job so he could be close to his future wife, she immediately explains the real motivation behind this action in an overlapping turn (lines 45 to 46). Presumably she does this in support of her husband’s narrative, to make it clearer to an audience that of course needs to be informed about this. Another striking feature of performativity in this verbal exchange is the predominance of the dramatic mode (Pasupathi 2006). We find a lot of direct speech representation, and in lines 23 to 30 husband and wife even re-enact the kind of bantering argument they shared before they actually became a couple, each assuming his or her own proper role. When the husband eventually imitates the wife’s response (“No, let’s pray or read some scripture,” line 29) he even mimics her sweet tone of voice. Even where there is no direct speech presentation the narrative is very animated, as can be seen in the lively intonation contour including numerous rises and falls in pitch (indicated by vertical arrows), stressed words (underlined expressions), and brief pauses used to accentuate narrative clauses. The climax of this narrative, the moment of the couple’s first kiss (lines 40 to 41), is additionally foregrounded by means of extremely short narrative clauses and by present tense, which relocates the past experience to the present moment and thus makes it even more dramatic. The story is also interesting as far as mental and emotional representations are concerned. The way the husband presents his reaction at the moment of this first kiss is similar to such presentations earlier in the excerpt: he uses direct speech (“It’s over. That’s it,” line 44), and his speech becomes emphatic through shifting pitch and a loud voice. Now, one can hardly imagine that the husband spoke exactly those words at the moment when he kissed his wife. They rather seem to be used in the interview situation to re-enact the excitement the husband felt back then. Later in the story we also find an interesting example of a mind-reading activity taking place when the couple meets again after their first kiss. The husband describes how he could “see” on his wife’s lips that she was going to deny the significance of their kiss. The use of direct speech once more not only presents what she might have said but what the husband inferred must have been going on in his wife’s mind at the time. This example demonstrates how performativity in conversational storytelling is used to engage listeners’ minds and to create involvement.

Performing Selves and Audience Design   273 CONCLUSION I hope to have shown to what extent storytellers in oral history interviews conducted for Internet databases accommodate their narratives to an implied larger online audience and thus create identities for themselves and perform these selves for others. As we saw in the examples, speakers’ awareness of the special situational context led to specific (and sometimes peculiar) rhetorical moves. I used the term “audience design” to capture this. By drawing on Phelan’s distinction between “narrator disclosure functions” I tried to show how speakers functions” and “­ either deliberately or unintentionally reveal more about themselves, their feelings, and their thoughts than they communicate on a surface level. The two main points I wanted to make, however, are: first, that conversational storytelling is more complex than we may at first assume and that it is worthwhile to use more narrative-specific tools to unravel those complexities. I am not saying that conversational storytelling and literary narratives are the same, but I do think that it is desirable to explore their commonalities as well as their ­differences more systematically. The second point is that we need to factor in speakers’ minds-in-interaction when analysing conversational storytelling. Talk-in-interaction is not something that happens independent of the minds of the people involved, and we need to form a better understanding of how we are made to follow and make sense of what someone tells us, how we are made to feel sympathetic or unsympathetic toward the presented characters (see also Sklar 2013), and how we engage meaningfully and interpretively with others in conversation. NOTES 1. Cf. also the debate in the special issue of Discourse Studies 8.1 (2006). 2. There is also a debate about whether identity is necessarily narrative in nature. Galen Strawson’s (2004) critique is prominent in this context. For further discussion, see Hyvärinen (2008, 2012) and McDonald (2013). 3. This is why designers of fictional digital narratives need to “think with” the medium, as Ryan puts it. 4. The “Observer’s Paradox” refers to the fact that one tries “to find out how ­people talk when they are not being systematically observed” (Labov 1972, 209), and the only way to obtain these data is by systematic observation. 5. A similar point is made by Marta Kurkowska-Budzan and Krzysztof Zamorski, who also foreground the role of the listener in oral history: “Oral history puts the narrator in first place but allows the listener to participate in a sort of conversation with the storyteller. […] This above-described dialogue has its place in our human historicity which shifts and changes; each contact with a new experience influences our understanding of the past as a building block in human identity. Each oral history interview plays a role in an individually and socially specific way in which the past affects the partners in this conversation” (Kurkowska-Budzan and Zamorski 2009, xiv).

274  Jarmila Mildorf 6. I thank Matti Hyvärinen for pointing this possibility out to me. 7. Oral history interview with Rudy Autio, 1983 Oct. 10-1984 Jan. 28, Archives of American Art, Smithsonian Institution. The interview transcript as well as the audio file on which this analysis is based can be found at: http://www.aaa.si.edu/ collections/interviews/oral-history-interview-rudy-autio-11713. 8. For a quick overview of this transcription system, see http://homepages.lboro. ac.uk/~ssjap/transcription/transcription.htm. 9. The repetition of the word “crank” in lines 12 and 13 dramatizes those craft artists’ efforts and renders their work experience more graphic for the audience. 10. I thank Prof. Sue Ziebland, the principal investigator of HERG (Health Experiences Research Group) in the Nuffield Department of Primary Care Health Sciences, University of Oxford, for kindly allowing me to use the material from the healthtalk.org website. 11. The entire audio file can be listened to at: http://healthtalkonline.org/peoplesexperiences/nerves-brain/carers-people-dementia/interview-42. 12. This interview was conducted as part of an Oxford University study of the ethical dilemmas facing carers of people with dementia. Hence participants (including this interviewee) may have been more likely to frame issues as “ethical” (Sue Ziebland, personal communication). 13. This kind of hypothetical storytelling is not as uncommon in everyday life as one might perhaps think. In her study of how people make sense of their divorce, Catherine Kohler Riessman identified similar stories about how things could have been, which she calls “hypothetical narrative” (Riessman 1990, 76). 14. I thank StoryCorps for kindly allowing me to use the material from their website. 15. The entire audio file can be listened to at: http://storycorps.org/?p=26255.

REFERENCES Bell, Allan. 1984. “Language Style as Audience Design.” Language in Society 13.2: 145–204. Coates, Jennifer. 2005. “Masculinity, Collaborative Narration and the Heterosexual Couple.” In The Sociolinguistics of Narrative, edited by Joanna Thornborrow and Jennifer Coates, 89–106. Amsterdam: John Benjamins. Coupland, Nikolas. 2007. Style: Language Variation and Identity. Cambridge: ­Cambridge University Press. Duranti, Alessandro. 1986. “The Audience as Co-Author: An Introduction.” Text 6.3: 239–47. Etzersdorfer, Irene. 1987. “Einige Überlegungen zur Theorie von ‘Oral-History’-­ Interviews.” In Vertriebene Vernunft I: Emigration und Exil österreichischer W ­ issenschaft 1930–1940, edited by Friedrich Stadler, 53–65. Wien: Jugend und Volk. Falk, Jane. 1980. “The Conversational Duet.” Proceedings of the Sixth Annual Meeting of the Berkeley Linguistics Society: 507–14. Genette, Gérard. 1980. Narrative Discourse: An Essay in Method. Ithaca: Cornell University Press. Georgakopoulou, Alexandra. 2007. Small Stories: Interaction and Identities. Amsterdam: John Benjamins.

Performing Selves and Audience Design   275 Goffman, Erving. 1959. The Presentation of Self in Everyday Life. New York: Doubleday. ———. 1981. Forms of Talk. Philadelphia: University of Pennsylvania Press. Günthner, Susanne. 1999. “Polyphony and the ‘Layering of Voices’ in Reported ­Dialogues: An Analysis of the Use of Prosodic Devices in Everyday Reported Speech.” Journal of Pragmatics 31: 685–708. ———. 2005. “Narrative Reconstructions of Past Experiences: Adjustments and Modifications in the Process of Recontextualizing a Past Experience.” In ­Narrative Interaction, edited by Uta M. Quasthoff and Tabea Becker, 285–301. Amsterdam: John Benjamins. Hyvärinen, Matti. 2008. “Life as Narrative Revisited.” Partial Answers 6.2: 261–77. ———. 2012. “Against Narrativity Reconsidered.” In Disputable Core Concepts of Narrative Theory, edited by Göran Rossholm and Christer Johansson, 327–45. Bern: Peter Lang. Krämer, Nicole C., and Stephan Winter. 2008. “Impression Management 2.0: The Relationship of Self-Esteem, Extraversion, Self-Efficacy, and Self-Presentation within Social Networking Sites.” Journal of Media Psychology 20.3: 106–16. Kurkowska-Budzan, Marta, and Krzysztof Zamorski. 2009. “From the Editors.” In Oral History: The Challenges of Dialogue, edited by Marta Kurkowska-Budzan and Krzysztof Zamorski, xi-xviii. Amsterdam: John Benjamins. Labov, William. 1972. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press. Labov, William, and Joshua Waletzky. 1967. “Narrative Analysis: Oral Versions of Personal Experience.” In Essays on the Verbal and Visual Arts. Proceedings of the 1966 Annual Spring Meeting of the American Ethnological Society, edited by June Helm, 12–44. Seattle: University of Washington Press. Langellier, Kristin M., and Eric E. Peterson. 2004. Storytelling in Daily Life: ­Performing Narrative. Philadelphia: Temple University Press. Linde, Charlotte. 1993. Life Stories: The Creation of Coherence. Oxford: Oxford University Press. ———. 2009. Working the Past: Narrative and Institutional Memory. Oxford: Oxford University Press. Lucius-Hoene, Gabriele, and Arnulf Deppermann. 2002. Rekonstruktion narrativer Identität: Ein Arbeitsbuch zur Analyse narrativer Interviews. Opladen: Leske und Budrich. Marcus, Bernd, Franz Machilek, and Astrid Schütz. 2006. “Personality in ­Cyberspace: Personal Websites as Media for Personality Expressions and Impressions.” ­Journal of Personality and Social Psychology 90.6: 1014–31. McAdams, Dan P., Ruthellen Josselson, and Amia Lieblich. 2006. “Introduction.” In Identity and Story: Creating Self in Narrative, edited by Dan P. McAdams, Ruthellen Josselson, and Amia Lieblich, 3–11. Washington, DC: American ­ ­Psychological Association. McDonald, Mary Catherine. 2013. “Life as a Narrative: Re-Thinking Strawson’s Anti-Narrative Stance.” Philotheos 13: 219–37. Mildorf, Jarmila. 2006. “Sociolinguistic Implications of Narratology: Focalization and ‘Double Deixis’ in Conversational Storytelling.” In The Travelling Concept of Narrative, edited by Matti Hyvärinen, Anu Korhonen and Juri Mykkänen, 42–59. Helsinki: Helsinki Collegium for Advanced Studies.

276  Jarmila Mildorf ———. 2008. “Thought Presentation and Constructed Dialogue in Oral Stories: Limits and Possibilities of a Cross-Disciplinary Narratology.” Partial Answers 6.2: 279–300. ———. 2012. “Second-Person Narration in Literary and Conversational Storytelling.” Storyworlds 4: 75–98. ———. 2013. “Referential Frameworks and Focalization in a Craft Artist’s Life Story: A Socionarratological Perspective on Narrative Identity.” In Rethinking Narrative Identity: Persona and Perspective, edited by Claudia Holler and Martin Klepper, 103–16. Amsterdam: John Benjamins. Mishler, Elliot G. 1986. Research Interviewing: Context and Narrative. Cambridge, MA: Harvard University Press. Morson, Gary Saul. 1994. Narrative and Freedom: The Shadows of Time. New Haven: Yale University Press. Niederhoff, Burkhard. 2014. “Focalization.” In Handbook of Narratology, 2nd ed., edited by Peter Hühn, John Pier, Wolf Schmid, and Jörg Schönert, 197–205. ­Berlin: Walter de Gruyter. Ochs, Elinor, and Lisa Capps. 2001. Living Narrative: Creating Lives in Everyday Storytelling. Cambridge, MA: Harvard University Press. Page, Ruth. 2012. Stories and Social Media: Identities and Interaction. New York: Routledge. Pasupathi, Monisha. 2006. “Silk from Sows’ Ears: Collaborative Construction of Everyday Selves in Everyday Stories.” In Identity and Story: Creating Self in ­Narrative, edited by Dan P. McAdams, Ruthellen Josselson, and Amia Lieblich, 129–50. Washington, DC: American Psychological Association. Phelan, James. 2005. Living to Tell about It: A Rhetoric and Ethics of Character Narration. New York: Cornell University Press. Reichert, Ramón. 2013. “‘Biografiearbeit’ und ‘Selbstnarration’ in den Sozialen Medien des Web 2.0.” In Kultur – Wissen – Narration: Perspektiven ­transdisziplinärer Erzählforschung für die Kulturwissenschaften, edited by Alexandra Strohmaier, 511–35. Bielefeld: Transcript. Ricoeur, Paul. 1988. Time and Narrative, vol. 3, translated by Kathleen Blamey and David Pellauer. Chicago: The University of Chicago Press. Riessman, Catherine Kohler. 1990. Divorce Talk: Women and Men Make Sense of Personal Relationships. New Brunswick: Rutgers University Press. ———. 2008. Narrative Methods for the Human Sciences. Thousand Oaks: Sage. Ryan, Marie-Laure. 2005. “Narrative and Digitality: Learning to Think with the Medium.” In A Companion to Narrative Theory, edited by James Phelan and Peter J. Rabinowitz, 515–28. Malden, MA: Blackwell. Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson. 1974. “A Simplest ­Systematics for the Organization of Turn-Taking for Conversation.” Language 50.4: 696–735. Sanders, Robert E. 1987. Cognitive Foundations of Calculated Speech: Controlling Understandings in Conversation and Persuasion. Albany: SUNY Press. ———. 2005. “Validating ‘Observations’ in Discourse Studies: A Methodological Reason for Attention to Cognition.” In Conversation and Cognition, edited by Hedwig te Molder and Jonathan Potter, 57–78. Cambridge: Cambridge ­University Press. Sandino, Linda. 2010. “Artists-in-Progress: Narrative Identity of the Self as Another.” In Beyond Narrative Coherence, edited by Matti Hyvärinen, Lars-Christer

Performing Selves and Audience Design   277 Hydén, Marja Saarenheimo, and Maria Tamboukou, 87–102. Amsterdam: John Benjamins. Schegloff, Emanuel A. 1992. “Repair after Next Turn: The Last Structurally P ­ rovided Defense of Intersubjectivity in Conversation.” American Journal of Sociology 97.5: 1295–1345. Sklar, Howard. 2013. The Art of Sympathy: Forms of Ethical and Emotional ­Persuasion. Amsterdam: John Benjamins. Smith, Sidonie, and Julia Watson. 2010. Reading Autobiography: A Guide for ­Interpreting Life Narratives. 2nd ed. Minneapolis: University of Minnesota Press. Strawson, Galen. 2004. “Against Narrativity.” Ratio 17.4: 428–52. Tannen, Deborah. 1989. Talking Voices: Repetition, Dialogue, and Imagery in ­Conversational Discourse. Cambridge: Cambridge University Press. Thornborrow, Joanna, and Jennifer Coates, eds. 2005. The Sociolinguistics of ­Narrative. Amsterdam: John Benjamins. Vazire, Simine, and Samuel D. Gosling. 2004. “E-Perceptions: Personality ­Impressions Based on Personal Websites.” Journal of Personality and Social Psychology 87.1: 123–32.

15 Documenting Everyday Life Mind Representation in the Web Exhibition “A Finnish Winter Day” Mari Hatavara

The web exhibition “A Finnish Winter Day” is the result of a museum project that recorded everyday Finnish life on a single day in February 2011. The exhibition consists of photographs, narratives, and quotes that portray people from eleven cities and towns across Finland. The data were collected by a team of museum researchers, and the exhibition was produced by one researcher in conjunction with a web designer. The project’s aim was to record everyday life—to give future generations an opportunity to “look at what it was like today.”1 The same day is structured chronologically for each place: every portrayal begins with an introduction that details the people, their occupations, and their homes. The introduction is followed by a narrative report of the day’s events: morning, day, afternoon, and evening. Every page ends with quotations from the people portrayed under the s­ubtitle “thoughts about everyday life.” The narrative is in the form of third-person narration without a clear position of reporting. The tenses change between past and future, and deictic markers mostly follow the positioning of the “here” and “now” of the storyworld. The declared intention of the exhibition—to record and document “what it was like” on this day—touches upon a central issue in narrative studies: a narrative’s ability to convey the experience of living in a storyworld, the qualia of a mind represented (see Herman 2009, 73). Following the definition offered by David Herman (2009, 71–73; 2011, 5), I understand storyworlds as mental models where interpreters frame inferences about s­ ituations, characters, and occurrences based on discursive cues from narrative artefacts. With this in mind, I focus on the representative practices used to render the Finnish people who are the subjects of the exhibition. I examine the methods used to represent the minds of the informants and the kind of interpretative processes and strategies these methods enable and encourage in the readers and viewers of the exhibition. By investigating the means and modes of mind representation in a documentary setting, I use this exhibition as a test case. The ability to assume the point of view of another person in content and discourse is, perhaps, a fictional ability. The possible exceptionality of fictional minds has long been a subject of lively debate: from the linguistic structure of speech and thought representation to cognitive abilities and folkpsychological attributions, literary scholars have traced the anatomy of minds

Documenting Everyday Life  279 2

inside and outside of fiction. The classical “speech-category approach” (see Palmer 2005) claims that the ability to reveal the contents of another mind is unique to fiction. This approach designates certain d ­ iscursive possibilities to fiction only, especially those where a third-person narrator relates the inner thoughts or even the subconscious content of a character (see Cohn 1978, 5−6). In a parallel manner, embedded subjective points of view (including those not linguistically discernible) have been understood to not be possible in real-life narratives (Genette 1993, 66–69; cf. Nielsen 2013, 90). In particular, omniscient narration with the ability to relate several minds in the storyworld has been deemed distinctive to fiction (Culler 2004, 23–25). These distinctive features of fictional mind presentation are often studied through two opposing concepts: fiction and everyday storytelling situations. The object of my study falls between these two opposites because the object of representation is everyday social interaction, but the representation is an artificial one: it is an exhibition designed by a researcher based on interview materials collected by several teams of researchers. In her chapter, Jarmila Mildorf demonstrates the usefulness of narrative methods in exploring the points of convergence between conversational and literary storytelling. Her analysis exposes the multilayered communication at work in interview narratives where people construe their stories to accommodate several audiences. The web exhibition analysed in my chapter is somewhat similar to oral history interviews: researchers have interviewed people who have told about their lives knowing the documentary intention. My case, however, adds another layer to the already doubled dialogue (see Mildorf in this volume) between the informant and the researcher, namely the web exhibition written and designed from the interview material. The exhibition, which has a (primary) documentary intention, uses t­ echniques of mind representation that are regarded as fictional. The perspective of the perceived storyworld changes between the reporting researcher and the informant, and stylistic modes are used that blend the discourses of the researcher and the informant. For this reason, the classical distinction between the author and the narrator is valid here (Genette [1991] 1993, 69–78; Cohn 1999, 123–31). In my analysis, I consider the exhibition’s disparate parts—its photographs, narrative prose, and direct quotations of informants’ speech—to form one narrative where all elements contribute to the construction of the s­ toryworld and the characters found within. I also treat the exhibition as an artefact with an authorial intention behind the choices of selecting, framing, and arranging both pictorial and verbal material. Furthermore, I use narratological concepts to discuss the exhibition: I assume an authorial audience to whom the exhibition is directed, talk about the narrator when referring to the teller instance relating the events, and call the subjects of the story—the informants—characters. These choices indicate my aim to test tools created originally for the analysis of fictional narratives in the analysis of this documentary case. At this juncture, I will not dwell on questions of referentiality or truth value, but instead study fictionality in mind representation as is it utilized in

280  Mari Hatavara the exhibition.3 The attribution of an authorial intention, a narrator, and characters enables me to analyse the many layers of communication and intention at play, and to ask how characters’ experiences—their sensation of this day in their lives—are represented and framed by the narrator and the exhibition as a whole. Thus, I emphasize the embedded layers of agency that are often understood to be specifically characteristic to fictional narrative (Nelles 1997, 2). First, I investigate how focalization occasionally adopts the internal point of view of one of the characters; second, I analyse the discursive modes mixing the narrator’s and a character’s languages, and third, I examine how the choice of content and modes of representation frames the winter day portrayed. Finally, I discuss the results of my earlier enquires in conjunction with how the relation between the interpretative processes of fictional and other (non-fictional, documentary, everyday) representations are understood in the current theoretical discussion. In this chapter, my chief focus rests on how the exhibition represents the characters experiencing a Finnish winter day and how readers and viewers are invited to interpret and make sense of these people. My analytical practice follows the recent developments in the study of fictionality; thus, fictionality is not understood as a generic label, but as a rhetorical resource that is also capitalized in nonfictional contexts (Nielsen, Phelan, and Walsh 2015)4. A typical case of this kind of discursive blending is narrative journalism, which foregrounds the emotions of the subjects ­presented. In focusing on an arbitrary day in the everyday lives of the ­families, this exhibition is part of the emergent phenomenon of representing ordinariness, where people are invited and encouraged to share their experiences in different mediatized and institutional settings. Nancy Thumim (2012, 9–10), for example, has pointed out that the museum sector utilizes new digital techniques in negotiating the personal with the communal and in creating participatory practices and representative modes. This, one may add, taps into the “human interest” factor always present in fiction: ­fiction focuses on how human life is lived and understood (see Olsen 1987). However, in this exhibition, it is the researcher who tells and a different group—the ­informants—whose lives are portrayed. As Thumim (2012, 7–8) underlines, so-called self-representations in digital media are often representations of a set of people made by another set of people—in the case of this exhibition, it is the researchers collecting and organizing the material of the people portrayed. INTERNAL POINTS OF VIEW Gérard Genette ([1972] 1980; [1983] 1988; 1993) coined the term “focalization” to designate the person perceiving the story as opposed to the instance narrating it. More specifically, the theory of Genette’s term explains situations where a narrator uses limited access to the characters’ thoughts and information, for example, only observing the characters from the outside or being limited to what one character experiences. In a recent

Documenting Everyday Life  281 study, Henrik Skov Nielsen (2013, 76) pointed out that focalization as a possible restriction of knowledge actually concerns the relationship between the author and characters. Following his idea that focalization defines the relation between authors and characters, Nielsen (2013, 76) further argues that Genette’s typology on narrative situations offers different ways of mediating experientiality. Nielsen’s argument centres on the specificity of unnatural narratives—a ­subset of fiction—but he makes a comparison to what is deemed possible in real-life narratives. He maintains that unnatural narratives and real-life narratives essentially have two options available: heterodiegetic narration with external focalization or homodiegetic narration with internal focalization. In a nutshell, a person may either relate others from an external point of view or herself internally. These basic dichotomies prove to be complicated in my example, since the narrator’s position in relation to the storyworld is disputable. Even though the narrator (as a researcher living in the world documented) is inside the storyworld, and in that regard homodiegetic, she is not mentioned as part of the storyworld, which in this sense makes her heterodiegetic. This double occupation—inside the same world but outside of the events related—has interesting effects on the status of the exhibition. Discursively, as external to the storyworld related, the narrator is heterodiegetic and uses the third person to refer to the people portrayed. Using examples from both visual and verbal narration, I aim to illustrate several variations of—and exceptions to—the basic situations assumed possible in real-life narratives. In “A Finnish Winter Day,” instances of heterodiegetic narration with internal focalization are clear in cases where the narrator adopts someone else’s internal point of view. The “natural” combination of homodiegetic narration with internal focalization means someone reporting from her own point of view; in this case, the narrator, who occupies the world documented, visits the points of view of the informants. Here is an example of a morning in the city of Hämeenlinna, where a family with children live: Janne [the father] woke up to the sound of snow thomping down from the roof. Thaw had set in and loosened the snow on the tin roof. It was only six o’clock so there was still time to have a little nap. Eeva [the mother], however, had to get up, she had to leave for work in an hour. The first sentence immediately begins with the experiencing subject’s perception (a sound). This is followed by an explanation of the sound, which— since Janne was placed as the focalizer in the first sentence—may well be interpreted as Janne’s thoughts and rationale of the sound he heard. Next, both Janne’s and Eeva’s thoughts on the time and their plans for the day are described. In this example, temporal and spatial orientation is fixed to the experiencing characters, and focalization changes smoothly between Janne and Eeva. Even so, there is no clear distinction between the two experiencing

282  Mari Hatavara and thinking minds: Janne is most likely the focalizer in the third sentence and Eeva in the fourth, but the two sentences still logically follow into each other. The move from one mind to another while keeping the thematic orientation indicates the narrator’s selection of the parts of the minds to represent. The story from Hämeenlinna ends in the evening: “Janne lit the fire and soon it was time for the boys to come in; have an evening snack and get to bed. It was nice to go to sleep with the fireplace warming up the house.” This ending returns to the characters’ experience as the narrator sums up the feelings that produce the “niceness” of the setting. The narrative begins and ends with the characters’ point of view, but the narrator controls what is narrated. Textually, the examples come closest to thought report, which is the narrator’s mode of relating a character’s consciousness. Since the voice of the character is not clearly separate from that of the narrator, these examples do not fit a prototypical case of thought report as defined by Cohn (1978, 46–47; see Palmer 2005). They do, however, utilize the discursive freedoms allowed by this form of thought report—presenting the character’s consciousness as mental action (“woke up”), and summarizing inner development or reasoning (“had to,” “was nice”; Palmer 2005, 604). Examples of the characters’ points of view abound in visual focalization, as well. I have chosen one series of photographs to illustrate this (see photographs 1, 2, and 3). The example portrays Titta, an elderly woman in sheltered housing in the city of Lahti. At the beginning, the verbal narrator states that Titta’s options were limited: being childless, she had nobody to take care of her when she got old and developed mild disabilities, so sheltered housing was the logical choice. Photograph number one starts the morning sequence, and photographs two and three end the evening.

Figure 15.1  Photograph 1. Lahti City Museum, Picture Archive. Foto Tiina Rekola.

Documenting Everyday Life  283 The first photograph shows the character Titta seen from the outside. She is very small in the picture and visible only partially through the venetian blinds on the windows. According to the customary viewing (and reading) orientation, the viewer starts from the left of the picture and approaches

Figure 15.2  Photograph 2. Lahti City Museum, Picture Archive. Foto Tiina Rekola.

Figure 15.3  Photograph 3. Lahti City Museum, Picture Archive. Foto Tiina Rekola.

284  Mari Hatavara Titta, whose small size and distant positioning are visually enforced by the slats of the blinds aligning with the perspective of the picture. The next photograph gives an over-the-shoulder shot, where the point of view has moved indoors and the viewer both sees Titta and partially shares the view she would have if she were to look up from the paper she is reading. Titta is again very small and positioned low compared to the window and the balcony railing between her and the world outside. Whereas the viewer’s position in the first photograph was very alienated from Titta, the second photograph invites the viewer to partially share her (potential) view. The third photograph completes the series as it fully immerses the viewer in Titta’s perspective: a point-of-view shot gives the viewer the same landscape view Titta has through the venetian blinds. The visual angle is from a slightly higher position than Titta had in the previous picture. This enables the camera to show a kind of hypothetical focalization point: what Titta would see were she to stand up and look out. The balcony railing is still visible in the photograph as a bar between Titta’s apartment and the world outside. This series of photographs starts with a view from outside, moves into Titta’s sheltered housing flat, and ends with a (possible) view out. This series creates an opposition between the outside and inside worlds. The viewer first observes Titta as if she were an exotic creature behind the window and the horizontal slats. The viewer is then invited to share Titta’s view via an over-the-shoulder shot not only from behind the window and the venetian blind slats, but also from the balcony railing. Furthermore, the view is restricted by the window frames on both sides. These pictures suggest Titta is isolated in the assisted housing unit from the rest of the world on the other side of the window. The same impression of isolation is also provided by the verbal text: Titta is a 76-year-old woman living in sheltered accommodation for the elderly. She moved in about two years ago after being ill for a long time. When the flat she rented was put for sale, sheltered housing provided the single woman with no children with “a safety network” she lacked. Her close relatives either lived a long way away or were elderly themselves. At the end of the day the decision about moving into a sheltered housing complex was easy. Some adaptation was required when an independent life was replaced by a large community of different people living in the complex. The residents have their own flats with kitchens, and they are free to decorate their homes and lead an independent life. One can also choose to pass some of the everyday routines of the establishment. For example, sometimes Titta has her breakfast in her own room. She wants to avoid institutionalization and do as many things as possible by herself, even though her condition sometimes sets restrictions.

Documenting Everyday Life  285 In the past Titta had many cultural interests and travelled a lot, but now she finds her activities closer to home. Titta, who once had an international career, maintains her language skills for example by reading German books and by watching German police TV series. She is also interested in documentaries on historical issues and different cultures. The case in Lahti is the only one with a single person with no children (another case, from Turku, depicts a divorced father of two). This is mentioned as a problem, since it has resulted in Titta having no relatives to look after her in her old age. Clearly, the move to sheltered housing was a deci­ sion that required some serious deliberation, even if “at the end of the day the decision about moving into a sheltered housing complex was easy.” The move seems to have had a great influence on Titta’s life. The second paragraph reports that her former life has been replaced by the current situation, and the third paragraph sharply contrasts “the past” and “now.” The interior world of Titta’s present appears passive and confined, whereas the outside world belongs to her past and is active in nature. This further emphasizes the drastic nature of the change as a replacement, and the divide between her former life outside sheltered housing and her current one inside. Both visual focalization and the narrator’s reports on Titta’s thoughts suggest the same experience of isolation and the relinquishment of former activities. Visual focalization invites the viewer to gradually share Titta’s point of view, and the verbal discourse by the narrator uses deictic markers attached to Titta’s present time. This blend of the narrator’s and character’s viewpoints provokes the ambiguity always present in fictional presentation: the two points of view present offer the reader two options that may be mutually exclusive but still equally possible (cf. Cohn 2000, 309). Does the feeling of being isolated in a cage, cut off from the previous world, stem from Titta’s experience or from the verbal and visual narrator? Is Titta dissatisfied with her situation or is she as an elderly single person perceived as an anomaly from an external point of view? This ethically sensitive question may be probed with concepts that Seymour Chatman (1990, 143) introduced to examine the nuances of focalization. The first concept, “slant,” includes the attitudes and ideological implications of the narrator reporting, and the second concept, “filter,” covers the mental activity of the character inside of the storyworld with her experiences, feelings, memories, and perceptions. Visual framing in the sense of a camera angle is a choice made by the visual narrator and part of the narrator’s slant. Therefore, the practice of opposing the outside and inside views from different sides of the window suggests that the narrator at least shares if not creates the feeling of i­solation. Furthermore, the over-the-shoulder and the point-of-view shots of view do not align with Titta’s look. Therefore, they do not follow the character’s visual perception but suggest what she might see were she to look in the same direction as the camera. For this reason, the composition of the visual narration is dominated by the narrator’s choices, not by Titta’s perception of the world.

286  Mari Hatavara Verbally, the end of the first paragraph suggests the character’s filter: she is the one who has, before “the end of the day,” hesitated about the move. The second paragraph, however, includes no signposts for internal focalization, such as the sensations, feelings, or memories of the character. Moreover, the second sentence of the paragraph is clearly from the narrator’s point of view, since Titta is referred to as one of the group of residents. ­Overall, the introduction sums things up in a manner typical of a heterodiegetic narrator’s discourse. Formally, both visual and verbal narration suggest the dominance of narrator’s slant over the character’s filter. Nevertheless, it is possible that the narrator follows and adopts the points of view and feelings that Titta has provided herself. Besides the two minds—those of the narrator and the character—embedded in the text, the text presents a more obvious twofoldedness: there is a contradiction between the statement of the independent life being replaced and the statement that it is possible to lead an independent life in the flat. This contradiction offers another opportunity to interpret the opposition between life outside and life inside a sheltered home: the place offers the illusion of independence, but one restricted to the private flat. The “sheltered” aspect of the housing denotes the shield the house provides by surrounding these pockets of independence within the shared areas that are populated by both the residents and the personnel. Therefore, the strongest mental frame here may be an institutional one that penetrates the minds of both Titta and the narrator. Following this interpretation, both the narrator’s and Titta’s points of view are determined by her role as a senior citizen. Avoiding institutionalization, which is also mentioned in the text, is such a major issue that it becomes a dominant theme. MINDS AND THOUGHTS OF THE CHARACTERS Titta is approached from the outside before presenting or sharing her point of view. The move from an outside perspective to an inside view often applies discursively as well. In many instances in the exhibition, the mood and voice of the verbal narration start from the narrator’s position and move closer to a character. The morning at Raasepori, where a fisherman lives with his family, is described as follows: A dense darkness covered the Prästö island, but there was already light at the windows of Karl-Mikael and Camilla’s house. Karl-Mikael had got up early and was getting his things together for the day’s fishing trip. Camilla made some sandwiches for her husband. The boys had had their breakfasts and they would soon be off for school. This quotation starts with a hypothetical focalization from outside the house and swiftly changes the point of view indoors to observe the characters’

Documenting Everyday Life  287 actions. The last sentence could be interpreted to partly follow the thoughts of the boys as they are planning what they will do next. The mood certainly changes from an outside view to an inside view both in the sense of character portrayal and as a concrete move from outdoors to indoors. Additionally, parts of the discourse may be interpreted to follow the characters’ thoughts. The day in Helsinki, which also portrays a nuclear family, uses varying discursive modes when telling about the morning: Atro [the father] was up early, too. He wouldn’t have to hurry anywhere today. This was a period during which he could stay home and work on the manuscript for a play set for autumn premiere. There was one meeting scheduled for today, however, but it would take place at home. Still, the very first thing to do was to take Eppu the dog out for a walk. The deictic markers for time and place (“today” and “this”) follow the character’s positioning from the start, even though the narrator speaks here; this is one of the markers traditionally attached solely to fiction (see Herman 2011, 7). Furthermore, the quotation uses thought report or even free i­ndirect thought in the last two sentences, as Atro contemplates things he has to do. The interpretative decision between thought report (narrator’s verbalization) and free indirect thought (characters’ original verbalization modified by the narrator) is complicated here by the manner in which the exhibition is constructed. In fiction, there is no original verbalization by the character that would precede the modified version the narrator gives: the original verbalization is assumed by the reader. The situation is different when the narrative is based on real interviews. The way the final text reads may be the narrator’s (researcher’s) modification of an exact expression Atro used when the researchers documented his life (“There is one meeting scheduled for today, however, but it will take place at home. Still, the very first thing to do is to take Eppu out for a walk.”). The modifications from present to past tense and from future tense to conditional mood follow the pattern of free indirect discourse, as does leaving the deictic marker “today” unattached (see McHale 1978, 251). The informative addition of the words “the dog” are clearly from the narrator though, who wants to keep the reader up-todate. This possibility of free indirect discourse being based on an actual verbal report given by an informant to the researcher when conducting the interview is available in a documentary setting only, where the narrative has subtexts in the form of original interviews. The exhibition uses thought report quite often. The following examples are from Jyväskylä (a student couple with children) and Kuopio (Helena, a child minder). In Jyväskylä, Saara, the mother of the family, is visiting the university library: “Saara had to study efficiently, because she didn’t want to work on her studies at home in the evening.” The narrator here summarizes a character’s wishes, reasons, and intentions into mental action. In Kuopio,

288  Mari Hatavara Helena is described as content at the end of the day: “The busy day left her [Helena] feeling happy—the idea of retiring seems distant to Helena!” This extract first gives the narrator’s account on Helena’s overall feeling in the past tense and then moves temporally to use present tense.5 The expressivity of the use of the exclamation mark again offers two interpretative possibilities: it may be the narrator summing up Helena’s happy feelings in consonant psychonarration (see Cohn 1978) or a sign of Helena’s discursive agency coming close to free indirect thought (see McHale 1978). Based on the textual clues it is impossible to tell whether one or both—Helena and/ or the narrator—is excited about Helena not thinking about retiring in the near future. Whereas thought reports and free indirect thought representation are often used in the exhibition, the outer signs revealing the inner states of minds are rarely reported. Reports on characters’ facial expressions are only used on a few occasions. In this context, the social practice of inferring mental states and reasoning from facial expressions and other gestures, both inside and outside fiction, is relevant (see Zunshine 2006, 6–10; Herman 2011, 8, 14). In the section “Thoughts about Everyday Life,” where direct quotes are given, the narrator uses a reporting clause only a couple of times. Reporting clauses are used to give information on the character’s expressive tone or gesture. One example is from Lieto, where Lea and Simo, a retired couple, live. Simo always, well earlier sometimes, said that, that why don’t you ever talk all gooey to him and I said that Martti never has an evil word for me, ye-es—Lea [laughing about Martti the cat]. [Square brackets original – MH.] The rarity of the narrator’s reporting clauses is highlighted by their presence in square brackets. This suggests the reporting clause is not intended to be read as a part of the narrative but is additional information for the reader. Here it is important to keep in mind, again, that whereas a world in fiction only consists of information given or hinted in the representation, the d ­ ocumented reality is full of incidents and information, and the narrator chooses those to be presented. While all quotations chosen for the exhibition have been said in one or other tone of voice and accompanied by gestures, this is the one case where the narrator chooses to inform the reader of the laughter. The motivation for this exceptional use of a reporting clause is presumably a rhetorical one; the narrator wants to guide the reader’s interpretation of the situation. The interplay between what is said and the tone reported is crucial here. The content of what Lea says is complicated, since it includes the disnarrated, a report of what never happened: “Martti never has.” This disnarrated passage, originally a reply to the husband, strongly implies Simo—opposed to Martti—has sometimes had harsh words for Lea. After this interpretation, it is easy to assume the report of Lea’s laughter in the

Documenting Everyday Life  289 narrator’s reporting clause is expected to alleviate the unpleasant suggestion of spousal aggression the disnarrated offers. Since the reader is guided so obviously, the narrator’s intention is highlighted and the narrator’s rationale exposed. A similar kind of reference to negative feelings and attitudes can be found in the descriptive part of the story of the family of a dairy farmer from Oulu. The narrator describes earlier circumstances in which Tarja, the mother of the family, has moved in after marrying Henri, the son of the previous ­owners of the farm: Even Henri’s parents finally got used to having “a city girl as a daughter-in-law who messed everything up; put the cows outside and even changed their colours,” Tarja smiles as she talks about the change of generation. This reporting clause is embedded in the narrator’s descriptive passage on the family situation. A direct quote from Tarja is first given and then framed with the information of Tarja smiling. Tarja’s embedded words sum up a negative attitude attributed to Henri’s parents toward herself. The narrator invites the reader to interpret Tarja’s smile as alleviating her words and indicating that she—at least now, if not always—takes a humorous a­ ttitude toward the opinions of her parents-in-law. The use of a direct quotation from Tarja and framing it with her smile makes the negative attitudes seem like Tarja’s private joke. Once again, the narrator mitigates a negative ­feature of the world presented. Both instances of the narrator’s reporting clause are easily interpreted as an attempt to soften an implicit or explicit critical attitude stated in a character’s words. The narrator not only selects the phrases to be quoted and modifies the character’s thoughts and words in (free) indirect discourse, but also frames some of the quotations in a manner indicating their tone and thus changing the reader’s interpretation. This highlights the nature of self­ arrative, representations in digital media made by others. In documentary n it is the authorial narrator who decides how to present and frame what is quoted from a character (cf. also Nielsen 2013, 89). When the narrator is quite obvious in her rhetorical intentions, the reader’s interpretative interest is redirected toward the narrator’s opinions and attitudes, in addition to the mental states of the characters whose expressions are described. This highlights the authorial motivation behind the selection and framing of the material included. Both Lea’s laughter and Tarja’s smile may, inside of the storyworld, be interpreted as signs of discomfort in the social situation, where the negative attitudes of others are possibly present even as the interview is given. However, the fact that these gestures are included suggests a narrative intention to offer a more positive stance to the authorial audience. As I have argued elsewhere (Hatavara 2013, 166–67), the reader’s interpretative task differs between everyday storytelling situations and fictional

290  Mari Hatavara narrative: the former is direct person-to-person communication, whereas the latter is in a fixed textual form, framed by several embedded textual and intentional layers. In the case of the documentary web exhibition, the reader only has access to the text, but she may assume the reality the documentary refers to. This has at least two effects on mind representation. There is an overall authorial intention framing the narrative, which includes the narrator’s rhetorical intent. Mixed discursive forms like free indirect discourse may be presumed to derive from an original utterance of an informant, ­perhaps deductible from the given narrative. These different relations between the intentional layers, from the implied author to the characters, exemplify embedded intentions not directly discernible from each other—a phenomenon usually attached to fictional discourse. They reveal the need to postulate an overall authorial intention, which may indicate that the narrator is somewhat unreliable or at least skewed in her presentation of the world. FICTIONALITY AND THE EVERYDAY Fictionality as a local rhetorical resource has been examined in two recent articles: “Ten Theses about Fictionality,” by Henrik Skov Nielsen, James Phelan, and Richard Walsh (2015), and “Naturalizing and Unnaturalizing Reading Strategies: Focalization Revisited,” by Henrik Skov Nielsen (2013). In “Ten Theses about Fictionality,” the authors make the following points about fictionality: “[f]ictionality … [is] the form of the intentional use of invented stories and scenarios” (Nielsen et al. 2015, 62) and “fictionality attaches to the communicative act, not the object of representation” (Nielsen et al. 2015, 65). The first point applies to “A Finnish Winter Day” only in a very limited sense, since the subject matter of the stories is documented. However, the second point is very apt to the exhibition. The ways of relating the stories, and especially the ways of mind representation both in the sense of thought and consciousness representation and in the sense of adopting points of view, mark out the exhibition. Local fictionality in “A Finnish Winter Day” means using fictional techniques as parts of the communicative act. Fictionality functions as a rhetorical means to invite the reader to imagine events and scenarios but not to conceive of those events and scenarios as invented. Besides using internal focalization as an affordance rather than a restriction, the narrator uses several modes of thought and consciousness presentation. The exhibition functions with the principle of using techniques to enable access to characters’ minds mentally or textually in some instances—as a local opportunity—while still retaining the overall frame of documentarism. This applies to the use of a heterodiegetic (in the sense of not being mentioned) narrator together with focalization that varies between external and internal points of view while having discursive privileges over the characters’ thoughts and speech.

Documenting Everyday Life  291 If the web exhibition wants to assert something, namely, a view of the Finnish life style (“what it was like”)—and I am inclined to believe it does—while resorting to fictional discursive resources, the few rhetorical devices inviting us to infer a humorous tone in the reporting clauses are even more highlighted. Gregory Currie (2014, 317) points out that one way of expressing ideas without making a commitment to them via fiction is to put them to the mouth of a character. He then remarks that readers tend to believe what is said within a pretence if no other information is provided. Therefore, adding in the documentary setting, the narrator must reframe the negative implications of spousal abuse or intergenerational bullying that the reader may assume from the characters’ words. This is the more pressing in ­Currie’s (2014, 317–18) understanding of both assertive and fictional narratives being able to convey ideas (whether true or false). Even within the classical project of building a divide between fictional and nonfictional representational resources, Dorrit Cohn (1978, 7) remarks that despite distinguishing fiction from reality, the representation of inner minds builds the semblance of represented reality. The ability of mind and consciousness representation to build a semblance in reality—to make a storyworld plausible—may also be used in nonfictional contexts, such as museum exhibitions. The exhibition gains its force from the local fictionality created by the transgression of factual narration’s representational possibilities. This indicates the privileged positions of the rhetorical devices of fictionalization in the portrayal and understanding of others. Coming from the opposite end of the debate, in an attempt to create a unified model for mind representation both within and without fiction, David Herman (2011, 7–9) maintains that fiction has some unique language patterns. This claim may not be warranted in the light of this chapter and other studies that indicate the use of fictional devices in nonfictional ­contexts, even in firstand secondperson narratives (see Iversen 2013; Mildorf 2013; Mildorf in this volume). Herman (2011, 10) argues that “the procedures used to engage with the minds evoked in fictional narratives necessarily piggyback on those used to interpret minds encountered in other contexts (and vice versa)”. I agree with him but also think that the study of mind and consciousness representation is too geared toward the direction of assuming that spontaneous, everyday storytelling situations form the basis for all mind representation and attribution (cf. Fludernik 1996). What definitely needs more attention is how the traffic really flows both ways: not only do we use our everyday scripts to interpret fictional minds, we also utilize what we have learned from fictional minds when faced with real ones. Documentary representations like “The Finnish Winter Day” utilize representational modes from the whole range of fictionality and still maintain the factual frame with a careful selection of fictional resources. However, fictionalization is a mode that risks clarity of intentions; this may be seen in Titta’s case in particular. The question remains unanswered as to whether the abnormality of her life stems from her expressed emotion,

292  Mari Hatavara or whether it is constructed through the narrative or determined by a social and institutional script that dominates how the lives of the (single) elderly are understood. Embedded intentional frames (authorial, narratorial, and experiential) leave the reader with an interpretative task where decisive conclusions are impossible. Blending the minds of the documented and the documenters enables the portrayal of the informants’ subjectivity, but at the same time, it infiltrates the represented subjectivity with authorial intentions and compromises the authenticity of this represented subjectivity. CONCLUSION The techniques of representing minds of the characters in “The Finnish ­Winter Day” use several modes associated with fiction or—more recently— fictionality as rhetorics. The narrator uses varying focalization, occasionally adopting the characters’ point of view, displaying focalization rather as an opportunity than a restriction. The narrative relating one day in the informants’ lives may start directly from the internal point of view of a character (an informant) or move from an external (researcher’s) point of view to internal focalization. In a similar manner, the narrator’s report of the events uses varying discursive modes in both speech and consciousness representation. Furthermore, the design of the text directs the reader to benevolent interpretations in the few instances where ethically volatile material, such as spousal aggression, is mentioned. In direct quotations, embeddedness is textually present only when it is necessary to ensure the intended interpretation. The narrator uses reporting clauses portraying facial expressions or the tone of voice in situations where she wants to influence the reader’s interpretation of the content of a quotation. This highlights the power of the narrating instance over the informants narrated: the authorial narrator has access to the characters’ minds, quoting, relating, and summarizing their speech, thoughts, and feelings. This access to the characters’ minds places the experiencing subjects in ­ arrator the storyworld at the centre of attention. Even if—and just as—the n has the power to select what to tell about the winter day and how to tell it, the main interest of the exhibition rests in the human-interest factor— showing what it was like to live that Finnish winter day. Since the use of documentary and fictional modes fluctuate throughout the exhibition, the documented origin and authenticity of the experience comes under suspicion. Is the experience represented only a semblance of life, presented and organized to follow predetermined assumptions of the “niceness” of ­Finnish life, or can the reader trust that they have access to concrete examples of individual experience? In the end, the fictional resources utilized may make the character’s minds inaccessible to the reader and viewer of the exhibition because of the dominance of the authorial intention in narrative and textual design. “Niceness” as an overall frame may be suspected to dominate the

Documenting Everyday Life  293 portrayal, and the double (or triple) dialogue turns into an ideologically controlled representation. The minds accessed and presented by varying focalization and several discursive blends are quite markedly framed by the narrator’s point of view and the overall design of the text, which may make them inaccessible, modified, and hidden by the framing authorial intention. NOTES 1. The exhibition can be found at http://tako.nba.fi/suomalainentalvipaiva/en/#/ etusivu. The previous quotation is from the project presentation. In this chapter, quotes from the exhibition are labelled under the cities presented; each city is a web page of its own. 2. For a concise overview of the debate see Herman 2011, Introduction. For a discussion on the workings and functions of folk psychology in social interaction, see Hutto 2013. Matti Hyvärinen’s chapter in this book offers a detailed discussion on the transparency and accessibility of fictional (and actual) minds. 3. For more about historical versus fictional representation, see Hatavara 2014. 4. For a thorough discussion on fictionality vis-à-vis artistic and everyday discourse, see Grishakova 2008. 5. Or even future tense: in Finnish there is no formal difference between present and future tense.

REFERENCES Chatman, Seymour. 1990. Story and Discourse: Narrative Structure in Fiction and Film. Ithaca and London: Cornell University Press. Cohn, Dorrit. 1978. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton: Princeton University Press. ———. 1999. The Distinction of Fiction. Baltimore and London: The Johns ­Hopkins University Press. ———. 2000. “Discordant Narration.” Style 34.2: 307–16. Culler, Jonathan. 2004. “Omniscience.” Narrative 12.1: 22–34. Currie, Gregory. 2014. “Afterword: Fiction as Transcultural Entity.” In True Lies Worldwide: Fictionality in Global Contexts, edited by Anders Cullhed and Lena Rydholm, 311–24. Berlin and Boston: Walter de Gruyter. Fludernik, Monika. 1996. Towards a ‘Natural’ Narratology. London and New York: Routledge. Genette, Gérard. [1972] 1980. Narrative Discourse. Ithaca, NY: Cornell University Press. ———. [1983] 1988. Narrative Discourse Revisited. Ithaca, NY: Cornell University Press. ———. [1991] 1993. Fiction & Diction. Ithaca: Cornell University Press. Grishakova, Marina. 2008. “Literariness, Fictionality, and the Theory of Possible Worlds.” In Narrative, Fictionality, and Literariness. The Narrative Turn and the Study of Literary Fiction, edited by Lars-Åke Skalin, 57–76. Örebro University: Örebro Studies in Literary History and Criticism 7.

294  Mari Hatavara Hatavara, Mari. 2013. “Making Sense in Autobiography.” In The Travelling ­Concepts of Narrative, edited by Matti Hyvärinen, Mari Hatavara, and LarsChrister Hydén, 164–78. Studies in Narrative 18. Amsterdam: John Benjamins. ———. 2014. “Historical Fiction: Experiencing Past, Reflecting History.” In True Lies Worldwide: Fictionality in Global Contexts, edited by Anders Cullhed and Lena Rydholm, 241–60. Berlin and Boston: Walter de Gruyter. Herman, David. 2009. “Narrative Ways of Worldmaking.” In Narratologia: Narratology in the Age of Cross-Disciplinary Narrative Research, edited by ­ ­Sandra Heinen, 71–87. Berlin and Boston: Walter de Gruyter. ———. 2011. “Introduction.” In The Emergence of Mind: Representations of ­Consciousness in Narrative Discourse in English, edited by David Herman, 1–40. Lincoln and London: University of Nebraska Press. Hutto, Daniel. 2013. “Fictionalism about Folk Psychology.” The Monist 96:4: 582–604. Iversen, Stefan. 2013. “Broken or Unnatural? On the Distinction of Fiction in NonConventional First Person Narration.” In The Travelling Concepts of Narrative, edited by Matti Hyvärinen, Mari Hatavara, and Lars-Christer Hydén, 141–62. Studies in Narrative 18. Amsterdam: John Benjamins. McHale, Brian. 1978. “Free Indirect Discourse: A Survey of Recent Accounts.” ­Poetics and Theory of Literature; A Journal for Descriptive Poetics and Theory of Literature 3: 249–87. Mildorf, Jarmila. 2013. “‘Unnatural’ Narratives? The Case of Second-Person Narration.” In The Travelling Concepts of Narrative, edited by Matti Hyvärinen, Mari Hatavara and Lars-Christer Hydén, 179–200. Amsterdam: John Benjamins. Nelles, William. 1997. Frameworks: Narrative Levels and Embedded Narrative. New York: Peter Lang. Nielsen, Henrik Skov. 2013. “Naturalizing and Unnaturalizing Reading Strategies: Focalization Revisited.” In A Poetics of Unnatural Narrative, edited by Jan Alber, Henrik Skov Nielsen and Brian Richardson, 67–93. Columbus, OH: The Ohio State University Press. Nielsen, Henrik Skov, James Phelan, and Richard Walsh. 2015. “Ten Theses about Fictionality.” Narrative 23.1: 61–73. Olsen, Stein Haugom. 1987. The End of Literary Theory. Cambridge: Cambridge UP. Palmer, Alan. 2005. “Thought and Consciousness Representation (Literature).” In Routledge Encyclopedia of Narrative Theory, edited by David Herman, Manfred Jahn, and Marie-Laure Ryan, 602–7. London and New York: Routledge. Thumim, Nancy. 2012. Self-Representation and Digital Culture. Houndmills: ­Palgrave Macmillan. Zunshine, Lisa. 2006. Why We Read Fiction: Theory of Mind and the Novel. Columbus: Ohio State University Press.

Afterword A New Normal? Brian McHale

Here is a book of many moving parts all brought together under the roof of a single grand topic, which the editors characterize as “intermedial and interactive mind and world construction in different narrative environments”—or elsewhere, even more pithily, “how minds interact with and within worlds.” Gathered under that roof we find a range of disciplines and sub-disciplines, including “literary studies, social studies, language studies and interactive media studies,” and an even wider range of objects of analysis, “from ­literature to digital games and reality TV, from online sadomasochism to oral history databases, and from horror to hallucinations.” Given this book’s unity and diversity—unity of topic, diversity of materials and approaches—it ought be possible in an afterword such as this one to tease out a number of threads that weave in and out of the chapters, connecting disparate disciplines and objects of study. One could, for instance., trace the thread of the contending claims of cognitivist (or “epistemic”) vs. separatist (or “exceptionalist”) models of fictional minds and worlds. Or one could unravel the related thread of the feedback between fictional and real minds (“the conventions of mind ­representation in fictional and real-world environments interact and are mutually available,” the editors write, while editor Hatavara, writing in her own chapter, puts it this way: “not only do we use our everyday scripts to interpret fictional minds, we also utilize what we have learned from fictional minds when faced with real ones.”) One could track the affordances of different media and genres analyzed across the volume (what the editors call “potential affordances and, particularly, discrepancies between media”), or one could tease out the convergences and divergences among the different (sub-)disciplines represented here. These and many other lines of connection bind this volume together and might profitably be pursued from chapter to chapter. I invite you, dear reader, to pursue them for yourself, ably abetted by the volume’s editors, who have identified so many starting-points. I do not, however, propose to connect up these particular dots here in the afterword, except incidentally. Instead, I want to report something that I think I glimpse at the horizon of these essays, with potential implications for the field of narrative studies generally.

296  Brian McHale INSIDE OUT, OUTSIDE IN Theory-building depends partly on determining what counts as a prototypical object of a particular theoretical discourse. The more nearly prototypical the object, the better the theory ought to account for it, and the snugger the “fit” between theory and object. The further an object differs from the ­prototype, the more loosely or imperfectly we can expect the theory to apply, until finally we pass beyond the gravitational field of the theory itself to a range of objects on which it has no grip and about which it has nothing to say—its outside. If we ask, “What is the prototypical narrative?” contemporary narrative theory gives two not necessarily compatible answers. The tradition of ­literary narratology, both classical and post-classical, answers “the novel,” while the tradition of narrative theory in ethnography, sociolinguistics, and folklore studies—the tradition that runs from Labov (1972) through Ochs and Capps (2001) and down to Mildorf, its main representative in this ­volume—answers something like “everyday oral storytelling” or “conversational storytelling.” Narrative genres and practices that are the least like novels (in the one tradition) or the least like conversational storytelling (in the other) tend to be relegated to the fringes of narrative theory, classified as special, marginal, or mixed cases. Other contenders for prototypicality, such as cinematic narrative, graphic narrative, or life-writing, have had to struggle to get their claims heard in the arena of narrative theory. The power of Monika Fludernik’s (1996) natural narrative hypothesis, and the basis for its success, is to be found in the way it reconciles the two traditions in a single model. Conversational storytelling is indeed prototypical for natural narrative, but literary narrative—preeminently the novel—achieves prototypicality insofar as it observes the cognitive parameters of natural conversational narrative and striving to simulate naturalness. Since readers, too, observe these same cognitive parameters, they narrativize literary texts even (or especially) when they appear to diverge markedly from the template of conversational narrative. Only when narrativizing manifestly fails—for instance, in avant-garde or experimental literature— do readers finally abandon the attempt to conform texts to the model of natural narrative, relegating such texts to the “outside,” beyond the reach of narrative theory. One measure of the power of the natural narrative hypothesis is the strength and resourcefulness of the dissent it has generated, in the form of a countervailing unnatural narrative hypothesis, or unnatural narratology. Perhaps “dissent” is the wrong word, however. Heterogeneous though unnatural narratology is—there are weaker and stronger versions of it (see Alber et al. 2010)—its overall effect is, paradoxically, ultimately to confirm and reinforce the natural narrative hypothesis. This paradox is borne out by the essays in the present volume that investigate precisely the sort of texts that are typically deemed to threaten natural narrativity and narrativization— extreme or experimental literary texts. Rantanen, for instance, exhibits the

Afterword  297 novels and films of Duras’s India Cycle as evidence of “unnaratable” fictional minds that, refusing to be accessed, resist narrativization. Hyvärinen makes a parallel argument about paranoid and malevolent mind-reading in John Burnside’s anomalous novel, A Summer of Drowning. Kakko, in his chapter about hallucinatory narratives, is especially explicit on this point. Where narrativization of the kind that the natural narrative hypothesis presupposes can look like “a struggle to return rogue texts to the fold,” hallucinatory episodes, he says, are different, more like “friendly cat-andmouse games between the author and the reader”—not narrativization, but something else. According to the editors, “Kakko demonstrates how drug literature’s radical experiential otherness can challenge the canonized narratological notions of naturalization and narrativization.” Well, no, not exactly—not any more than Duras’s inaccessible privacies or Burnside’s deviant mindattributions challenge the naturalization of fictional minds. What they all do is demonstrate where the outermost limits of prototypical narrativity are to be located—and perhaps not even the outermost limits, but something short of that, where anomalous narratives can still be recognized precisely as anomalous, rather than being dismissed as something else, as non-­narratives. In short, such unnatural narratives serve as limit-cases, u ­ ltimately helping to reinforce the norm. Looking inward from the periphery, we can tell better than ever where the center lies. I wonder whether the horror fiction described here by Brümmer might actually be more extreme in some ways, and present more of a challenge than the experimentalism of Duras, Burnside, and the writers of drug ­experience. Certainly the horror genre is less canonical than the varieties of what is ultimately psychological realism in Duras and the others. But if Brümmer is right, then the “monstrous” narratives of horror fiction might have implications for the poetics of immersion generally, sufficiently radical as to constitute, not an outer limit of the natural-narrative prototype, but a new prototypical narrative altogether—a prototype in the periphery. I remain tentative in my claims for horror fiction, but less so for some of the other peripheral genres considered here that, like horror, seem to be undergoing promotion to the status of alternative models, challenging the prototypicality of the novel and conversational storytelling and normalizing “unnatural” narrative in various ways. I am thinking now of the chapters on role-playing digital games (Roine; Samuel et al.), sadomasochistic ­practices, both face-to-face and online (Harviainen), texting (Lyons), online trip reports (Kakko), online oral-history interviews (Mildorf ), documentary Web exhibitions (Hatavara), reality-tv shows (Mäkelä), and country and western songs (Palmer). Whether the authors of these chapters fully intend it or not, all of the genres and practices they analyze here seem to me to have the potential to constitute alternative narrative prototypes, with more or less radical knock-on consequences for the theorization of narrative. Change the prototype and you change the theory.

298  Brian McHale Lyons, for instance, following Ruth Page (2012) and others, analyzing “the episodic narrativity found in a variety of everyday stories told using electronic media (Facebook status updates, emails, tweets, etc.),” concludes that “the traditional understanding of what characterizes a prototypical narrative may not be fully representative of the current narrative practice.” In effect, what looks like “unnatural narrative” relative to prototypical narrative is normalized in texting. Roine, analyzing role-playing games, concurs, and even goes a step further. Not only do role-play games diverge from the norms of prototypical narrative, but they also function as though they constitute a new norm, a new center of gravity, drawing other media into their orbit. “The participatory nature of digital media,” she writes “has already profoundly affected the usage and reception of all media: they now appear to us as phenomena to be cut, pasted, reassembled, and distributed with ease.” Without saying so in so many words, Roine is performing a thoughtexperiment, asking, in effect, “What if role-playing games were treated as prototypical narratives? What would narrative theory (or media theory, etc.) look like then?” We could easily imagine others among these authors conducting similar thought-experiments, asking parallel questions. Harviainen: “What if S&M sessions were treated as prototypical narratives? What would narrative theory look like then?”—a profoundly disorienting prospect! Or Mäkelä: “What if reality-tv shows were treated as prototypical narrative?” The only one of these authors who might resist being characterized in this way, I suppose, is Palmer, who, rather than imagining folk and country ballads as constituting an alternative narrative prototype in the periphery, attempts explicitly to assimilate them into the novel. Their history, he says, runs parallel to the novel’s as far as the representation of consciousness is concerned, beginning in “behaviorism” and achieving over time increasing psychological depth. Nevertheless, Palmer’s own evidence partly contradicts his efforts at assimilation: ballads are, in at least one respect, distinctly a-prototypical. He discovers a narrative peculiarity of ballads, the attributional hole, as he calls it. While ballads, as sparse and succinct as they often are, nevertheless typically give motivations for secondary or follow-on actions, they often leave the key central act, the one that actually launches and drives the plot—murder, elopement, etc.—entirely unmotivated. What if that peculiarity of ballads were treated as in some sense prototypical? What sorts of knock-on consequences might that have for research into all kinds of narrative genres—for instance, biblical narrative (in the tradition of Sternberg [1985] and others), poetic narratives (see McHale 2009), or even the canonical novel itself? EITHER/OR, BOTH/AND Change the prototype and you change the theory: let’s say that this is so, and that what a number of the chapters in this volume do is propose, more or less

Afterword  299 deliberately, alternative prototypical narratives—not novels or conversational storytelling, but role-playing games, S&M sessions, trip reports, reality-tv shows, country and Western songs, and so on. How exactly would narrative theory change if new genres and practices such as these were promoted to prototypicality? I have already begun speculating about this in the pages you have just read. If the “monstrous” narratives of uncanonical horror literature that Brümmer describes were accepted as prototypical, then the poetics of immersivity would need to be overhauled. If the “attributional holes” that Palmer detects in ballads were regarded as prototypical instead of anomalous, then we might have to recalibrate all of our measures of “normal” narrative motivation. If what Mäkelä calls the “disturbingly ­manifold intentionality” of reality-tv shows of the Survivor type were regarded as prototypical, then literary narrative fictions might appear to be “unnatural” in ways that an unnatural narratologist could never have anticipated. And so on. Rather than try to pursue all of these intriguing thought-experiments, let me instead focus on just one. “What if role-playing games were treated as prototypical narratives?” I asked a moment ago. Well, what if they were? In the introduction, the editors of this volume rather firmly exclude the ­possibility of treating games as prototype narratives. The contentious debates in game studies, they write, though “sometimes adjacent to those of literary studies and similar disciplines,” have been “driven by the somewhat unique characteristics of games as media and performance.” In other words, what happens in game studies stays in game studies, and does not spill over into narrative more generally, because games are different and special—­ peripheral, not prototypical. Yes, but what if? According to Roine, role-playing games such as the Mass Effect Trilogy can be experienced as both storyworlds and game systems, both immersive and interactive. We engage with them in two r­ egisters simultaneously, or we freely “switch registers” (the terminology here is found in the chapter by Samuel et al.), toggling back and forth between realities—the storyworld of the game, the real world in which we play the game. “The fictional Mass Effect universe,” Roine writes, “can be perceived simultaneously as being [a] possibly existing world that is experienced from the inside—as if the events were happening to the player—and recognised as being blatantly constructed according to certain strategies and, perhaps, from familiar building blocks.” The editors claim that “Roine takes an intermediary stand between narratologists and ludologists” in the controversy that has so vexed game studies. Well, not exactly: she doesn’t so much take a stand between as to take both stands at once: both/and. It seems to me that the designers of Prom Week, in their report here on how its players relate to the game’s storyworld, come to a parallel conclusion when they talk about the “feeling of responsibility” that players develop, reflecting a certain empathy with characters but also empowering players to take action on the plane of the game: both/and again. Transfer this both/and stance from role-playing games to narrative in general, and the

300  Brian McHale consequences are potentially radical, as Roine herself seems to imply. “The enjoyment of fiction is seen too often as a ‘total’ experience,” she writes, and this is why the recognition of the structures or shapes is viewed as negating the immersion in the fictional reality …. [However] recognising structures and shapes is important because the player needs to be able to move from the general principles of the system—such as how quests are structured—to smaller details—such as how they will ­contribute to the development of the playable character. If we were to read narrative in general the way Roine reads role-playing games, in the key of both/and, then perhaps we would be challenged to think of all narratives as simultaneously worlds to be immersed in and systems to be contemplated from without, and to reflect on how each experience, immersion, and interaction inflected the other. That’s what might follow if role-playing games were regarded as prototypical narratives. Actually, Roine’s both/and stance toward immersivity and interactivity turns out not to be wholly unprecedented in classical and post-classical narrative theory. James Phelan (2007, 5–6), for instance, distinguishes among three sources of readerly interest in narrative: mimetic interest in the storyworld and its inhabitants, thematic interest in the ideological, philosophical, ethical, etc., issues raised by the narrative, and synthetic interest in the construction of the narrative and the aesthetic values that underwrite it. Some narratives are dominated by one or other of these three interests, but others feature two or all three of them in different combinations, and any given episode or passage may evoke two or three of these types of interest simultaneously or in quick succession. Obviously, Phelan’s theory would have no difficulty accommodating narratives of the kind described by Roine where immersivity (mimetic interest) and interactivity (synthetic interest) mingle and coexist. Not every theory would need to change, or not in every respect, if the prototype changed. However, Phelan’s theory of narrative could accommodate at least some of the implications of the new prototype, that isn’t necessarily the case with contemporary narrative theory generally, including some of the theories ­represented in the present volume. In particular, it seems to me that Andersson’s bold thesis about the incompatibility of “epistemic” (immersive) reading and “­aesthetic” (“whole-text”) reading would be seriously undermined if role-playing games, as Roine describes them, were accepted as prototypical. Andersson’s stance is aggressively either/or: either one can be transported to a storyworld and undertake to inform oneself about it, or one can grasp the structured text as an aesthetic whole, but not both; either one can immerse oneself in an alternate reality, or one can interact with a real-world ­artifact, but not both. The editors speak in terms of the competing claims of l­ifelikeness and composability—not terms that Andersson himself uses anywhere here, as far as I can see, but suggestive ones nevertheless. Perhaps Andersson is right that in the narratives that ­ arratives, presumably—one can he regards as prototypical—fictional literary n

Afterword  301 only entertain lifelikeness at the expense of composability, and vice-versa, as in a zero-sum game. But change the prototype to role-playing games, and wouldn’t the theory need to change accordingly? Ryan, too, the other big-picture theorist in this volume, might need to revisit her theory if role-playing games came to be regarded as prototypical narratives. Granted, Ryan doesn’t apply the categorical either/or logic that we find in Andersson. She never exactly suggests that the cognitive approach to “storyworld” and the ontological approach to “fictional world” are mutually exclusive or incompatible, though she certainly distinguishes between them sharply enough. But if she doesn’t share Andersson’s either/or logic, neither does she entertain here the possibility that storyworld and fictional world might co-exist somehow: both/and. Elsewhere in her chapter, in a similar move, she argues that story, world, and medium are competing and mutually exclusive objects of readerly interest in narrative. Intense focus on story detracts from world, as in jokes and tragedy; focus on world detracts from story, as in some fantasy and science fiction. As for the component of medium, “human attention is limited, and intense focus on the medium distracts people from the story and from the world.” The idea that one could focus on all three components, story, world, and medium, at the same time is utopian. There is much that is debatable about Ryan’s either/or logic here. Lutas, in his chapter, reminds us of Ryan’s conviction that self-reflexive paradoxical devices such as metalepsis block immersion in fictional worlds—in other words, either medium or storyworld is foregrounded, but not both. Not so, says Lutas, or anyway not always. Some of her cinematic examples seem to demonstrate an alternative both/and logic—both medium and storyworld—much as Roine’s role-playing games do. We might even be able to think of canonical literary counter-examples that contradict the logic of either medium or storyworld. How about Pynchon’s Gravity’s Rainbow, for instance, where readerly interest seems to be distributed pretty evenly across all three components all the time—complex science-fiction-style world-building, a ramifying thriller plot, and the densely textured medium of Pynchon’s language? Or how about Perec’s La vie mode d’emploi, where a mass of circumstantial “worldly” detail and proliferating stories simultaneously reveal and conceal, now-you-see-it-now-you-don’t, the procedural apparatus that generates the storyworld? All three components are there, occupying the foreground, all the time. These are, of course, exactly the sorts of extreme or experimental ­literary narratives that tend to get wheeled out whenever limit-cases are needed to mark the outermost orbit of narrativity—unnatural narratives, in short. Their “unnaturalness” converges with that of the new prototype narratives— role-playing games and reality-tv shows and S&M sessions and country-andWestern songs and all the rest—for which unnatural narrativity is normal. If unnaturalness is the “new normal,” as one might infer from these cases, then what might the consequences be for narrative theory as a discipline?

302  Brian McHale ONE, MANY One consequence might be that narrative theory, losing its anchorage in the “old” prototypes of the novel and conversational narrative, might become divergent and various, multiple narratologies instead of one—a separate narratology for each medium and intermedium, each genre and transgenre. This volume’s editors observe that in recent narrative theory “variation across media and genres has been downplayed in favour of searching for the cognitive universals of experiencing represented worlds.” This is no doubt the case, but perhaps the universalizing impetus has been exhausted for now, and the time has come to try something else. Editor Mäkelä, in her own chapter, notes that “narrative theorists of the cognitive paradigm … are keen to detect the same narrative universals at work in all media, especially in the ways that stories set up worlds and [populate] them with mental subjects.” However, her own point of departure, she writes, is “the assumption that different narrative genres and media have their own specific means of foregrounding mental action.” Since “mental action” is her topic, indeed one of the topics of this book as a whole, she leaves it at that, but one could go further and posit that “different narrative genres and media” have their own specific means of doing everything and hence require their own specific narratology. “This book,” its editors write, “explores the limits of both narrative objects and narrative studies.” Yes, no doubt it does but perhaps in ways the editors themselves don’t fully appreciate. Volumes such as this aspire to produce the effect of a unified discipline—unified, perhaps, only with difficulty and amidst controversy and contention, but nevertheless a discipline in which everyone is ultimately collaborating on the same great shared project. A volume like this one aspires to gather all the varieties of narrative in the world under the big tent of a single narrative theory: out of many, one. Do the developments I think I have glimpsed here—the loss or displacement of prototypical narrative, the proliferation of medium-specific narratologies— portend, rather, the dis-integration of our discipline? Out of one, many? And if it does, would that be a good thing or a bad one? REFERENCES Alber, Jan, Stefan Iversen, Henrik Skov Nielsen, and Brian Richardson. 2010. “Unnatural Narratives, Unnatural Narratology: Beyond Mimetic Models.” ­Narrative 18.2: 113–36. Fludernik, Monika. 1996. Towards a “Natural” Narratology. London: Routledge. Labov, William. 1972. “The Transformation of Experience in Narrative Syntax.” In Language in the Inner City: Studies in the Black English Vernacular, 354–96. Philadelphia: University of Pennsylvania Press. McHale, Brian. 2009. “Beginning to Think About Narrative in Poetry.” Narrative 17: 11–30.

Afterword  303 Ochs, Elinor, and Lisa Capps. 2001. Living Narrative: Creating Lives in Everyday Storytelling. Cambridge MA: Harvard University Press. Page, Ruth. 2012. Stories and Social Media: Identities and Interaction. London: Routledge. Phelan, James. 2007. Experiencing Fiction: Judgments, Progressions, and the ­Rhetorical Theory of Narrative. Columbus OH: Ohio State University Press. Sternberg, Meir. 1985. The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading. Bloomington: Indiana University Press.

This page intentionally left blank

List of Contributors

Greger Andersson, PhD, Professor of Comparative Literature, Örebro ­University, Sweden. Gero Brümmer, PhD, Postdoctoral Fellow, Heinrich-Heine-Universität ­Düsseldorf, Germany. J. Tuomas Harviainen, PhD, Postdoctoral Researcher, University of ­Tampere, Finland. Mari Hatavara, PhD, Professor of Finnish Literature, University of Tampere, Finland. Matti Hyvärinen, PhD, Professor of Sociology, University of Tampere, ­Finland. Tommi Kakko, PhD, Postdoctoral Researcher, University of Tampere, ­Finland. Dylan Lederle-Ensign, PhD Candidate in Computer Science, University of California Santa Cruz, USA. Liviu Lutas, PhD, Postdoctoral Research Fellow, Lund University, Sweden. Agnieszka Lyons, PhD, Postdoctoral Researcher, Birbeck, University of ­London, UK. Maria Mäkelä, PhD, Senior Lecturer in Comparative Literature / Postdoctoral Fellow, University of Tampere, Finland. Michael Mateas, PhD, Associate Professor of Computational Media, ­University of California Santa Cruz, USA. Frans Mäyrä, PhD, Professor of Information Studies and Interactive Media, University of Tampere, Finland. Josh McCoy, PhD, Assistant Professor of Computer Science, American ­University, USA. Brian McHale, PhD, Humanities Distinguished Professor, Ohio State ­University, USA.

306  List of Contributors Jarmila Mildorf, PhD, Senior Lecturer in English, University of Paderborn, Germany. Alan Palmer, PhD, Independent Scholar. Tytti Rantanen, MA, Doctoral Student, University of Tampere, Finland. Aaron Reed, PhD Candidate in Computer Science, MFA in Digital Arts and New Media, University of California Santa Cruz, USA. Hanna-Riikka Roine, Lic.Phil., Doctoral Student, University of Tampere, Finland. Marie-Laure Ryan, PhD, Independent Scholar. Ben Samuel, PhD Candidate in Computer Science, University of California Santa Cruz, USA. Mike Treanor, MFA & PhD, Assistant Professor of Computer Science, American University, USA. Noah Wardrip-Fruin, MFA & PhD, Associate Professor of Computational Media, University of California Santa Cruz, USA.

Index

Aarseth, Espen 75, 77 Abbott, H. P. 67, 147, 161, 164, 178 Abrams, M. H. 202n1 af Segerstad, Ylva Hård 123 agency 69, 73, 76, 84, 150, 164, 202n2, 209, 230, 280, 288 Alber, Jan 31, 44, 137–8, 296 alienity 173–4, 183–4 Alison, Laurence 106, 110 Alperson, Philip 108–9 alterity 173–4, 183–5 American ballad 205, 210, 214, 217–19 Argy, Stephanie 39 Atkinson, Gilbert 139n1 audience design 138, 260–1, 273 audience: authorial 261–2, 279, 289; implied 263, 266, 273 authenticity 113–14, 118, 151, 202n2, 247–8, 292 authoring 68, 70–4, 102 Bakhtin, Mikhail 117 Bakker, J I. 93 Baldry, Anthony 122, 125 Bamberg, Michael 122, 124 Banfield, Ann 16, 27n2 Baron, Naomi S. 123 Barthes, Roland 21 Bartle, Richard A. 70–1 Bataille, Georges 117–18 Baumeister, Roy F. 109 Bell, Alice 31, 44 Bell, Allan 138, 260–1 Berger, Peter L. 197, 203n6 Bessière, Irene 118 Bessière, Jean 46n1 biblical texts 52, 57–8 Bilandzic, Helena 129 Bogost, Ian 71, 96 Bordwell, David 20, 26n2, 163–4 Bortolussi, Marisa 55–6, 60–1

Bovbjerg, Dana H. 130 Brandhurst, Christoph 106, 110, 113 Breslow, Harris 139 British ballad 205, 207–8, 210–11, 215–19 Brix Jacobsen, Luise 51 Brock, Timothy C. 130 Brooks, Cleanth 26n1 Brooks, Peter 248 Brown, Ashley 114–15 Bruder, Gail A. 123, 130 Bruhn, Jørgen 31, 46 Bruner, Jerome 2–3, 109, 168, 225, 228, 231 Bruns, Gerald L. 196–9, 201 Budge, Gavin 191 Burleson, Donald R. 172 Burnside, John 223, 225, 238 Busselle, Rick 129 canonicity 168–72, 230–1 Capps, Lisa 258, 269, 296 Carroll, Noël 167, 171–3, 176–7 Carvalho, John 109 Cassell, Justine 135 causal attribution 214, 216 Cetina, Karin Knorr 203n6 character narration see first person narration character reactions 172, 174, 176–8 character: fictional 18, 70, 76, 83, 112, 116, 167, 177, 185, 206, 245; as actor 75; as avatar 75–6; playable character 67–8, 73–5, 77–81, 83 Chatman, Seymour 26n2, 285 Chion, Michel 162 Clandinin, D. Jean 2 classical narratology 46n2, 69, 122; see also postclassical narratology Coates, Jennifer 258, 271 Coates, Stephanie 122

308 Index cognitive approach 5, 12, 301 cognitive narratology 4, 123, 149, 151, 241, 253 Cohen, Jeffrey Jerome 171–2 Cohen, Norm 218 Cohn, Dorrit 2, 35–6, 51, 149, 163, 225, 279, 282, 285, 288, 291 Comme il Faut (CiF) 93, 94, 96–8, 100–2 conceptual 168, 177, 181, 183 confession 188–9; in reality TV 247–9 consciousness representation 241–2, 246–7, 249, 290–2 Constantine, Mary-Ann 208–9 context-sensitivity 169–70, 174 conventionality 238 Cortázar, Julio 22 Costikyan, Greg 69–70, 72 country music 205–20 Coupland, Nikolas 260 Cross, Patricia A. 112, 114 Crystal, David 137 Csikszentmihalyi, Mihaly 139n1 Culler, Jonathan 152, 156, 158, 196–8, 200–1, 279 Currie, Gregory 291 Cutmore, Tim R. H. 130 Dadds, Mark R. 130 Dagerman, Stig 52–4, 58 Dancer, Peter L. 106 Danielsson, Tage 37 deictic centre 125–6, 129–30, 132, 139 deictic shift 126, 129, 131–4, 136, 139 del Teso-Craviotto, Marisol 137 Deleuze, Gilles 108, 117, 162 Dennett, Daniel C. 246 Deppermann, Arnulf 257–8 De Quincey, Thomas 187–94, 197, 199–200, 202n2 description 168, 172–3, 177; see also evaluation desires, reasons, beliefs 225, 236–7 Dessen, Alan C. 137 dialogue 1, 42, 91, 150–1, 155, 208, 212, 238, 273; constructed 268–9; double 261, 279, 293; in game 72–3, 78, 80–3, 90, 98, 100–1 Dick, Philip K. 24 DiGiovanna, James 26 digital media 67, 70–2, 79, 83–4, 122, 258–9, 280, 289 direct speech 210–13, 218, 268, 272 disclosure functions 261–3, 273

discursive tools 122, 130, 135 Dixon, Peter 55–6, 60–1 documentary 278–80, 287, 289–92 Doležel, Lubomír 12, 16–17 Douglas, Mary 171 Doyle, Conan 56–7 Drachen, Anders 68–9 drug literature 187–8 Duchan, Judith F. 123, 130 Dumas, Alexandre 35, 42 Duranti, Alessandro 261 Duras, Marguerite 147–62 passim, 297 Eco, Umberto 18, 87 Elkanah, (Biblical) 56–7 Elleström, Lars 32–3 embodiment 1–4, 59, 227–8, 242, 252; in sadomasochism 112–13, 115; of narrators 16 Emmott, Catherine 53, 58–60 enactment 137 enactment imagery—description imagery 59 Ensslin, Astrid 124 Epstein, Seymour 139n1 Ermi, Laura 83 Eskelinen, Markku 4, 69 Etzersdorfer, Irene 260–1 evaluation 168–70, 172–5, 177; see also description event 23, 168–9, 171–2, 174; historical 205, 208, 210, 218; sequence of 123–7, 131–4, 154, 225 eventfulness 168–72, 174–5; see also event exceptionality thesis 149, 151, 225 experientiality 149, 151, 247, 253, 281 Expressive Processing 88, 94 expressivity 228, 241–49 Falk, Jane 271 Fauconnier, Gilles 123, 131, 134, 136, 139n2 Ferriar, John 202n2 fiction — and the “epistemic” and the “sepatist” approach to fiction 50–1, 53–6, 58, 61–2 fiction: experimental 25; horror 168, 171–2, 299; interactive 5; literary 50–2, 56, 61–2, 77, 242, 245–7, 249, 252–3; science 22, 24, 26, 67, 77, 84, 194, 301; visual 41 fictional mind 2–3, 149, 168, 206–7, 209, 225, 278, 291

Index  309 fictional narrative, narrative fiction 3, 5, 17, 34, 50, 52, 56, 59, 61–2, 128, 149, 163, 168–9, 206, 249, 252, 279–80, 291, 299 fictional worlds 3, 5–6, 23–4, 26, 35, 41, 46, 53, 110, 116–18, 170–1, 173–4; and immersion 30, 55, 177, 185; and metalepsis 41, 46; completeness/fullness of 17–18; in game 70, 72, 77, 79, 82, 84–5, 87–9, 92, 98, 102; ontology of 12–13, 16–17, 19–20, 169, 301; primary 22 fictionality 51, 278–80, 290–2 fictionalization 7, 291 fictive consciousness 148, 150, 152, 154–5, 159, 163; see also fictive minds fictive minds 147, 163; access to 149–52, 155–6, 164; interpreting 148–9, 152, 154, 157–8, 164; reading 148–52, 161; unreadable 147, 151, 161; unnarratable 147–8, 150–1, 159, 161, 164 Fielding, Raymond 38 first person narration 15, 133, 136, 261; see also homodiegetic narration Fludernik, Monika 30, 35–6, 38–9, 187, 198, 203n7, 253, 291, 296 focalization: external 211, 281; hypothetical 283, 286; internal 32, 281, 286, 290, 292; visual 247, 282, 285 Fokkelman, J. P. 56–8 folk psychology 56, 149, 151–2, 165n4, 225–8, 233, 293n2 forking path narrative 20–1, 249 Forster, E.M. 17–18, 25, 34 Frank, Katherine 114, 117 Franks, Anton 130 Frasca, Gonzalo 69 free indirect discourse 150, 211, 218, 287, 289–90 free indirect speech 228, 238 free indirect thought 287, 288 Frobenius, Maximiliane 124 Galbraith, Mary 130 Gallagher, Shaun 152, 228 game mechanics 77, 82 game studies 1, 4, 69, 103n1, 116, 299 gameplay 4, 67–83; passim 87–9, 91, 97–100, 103n1 gap-filling 55, 57–8, 61–2 Gazzaniga, Michael S. 107

Geisenhanslüke, Achim 172 Genette, Gérard 13, 15, 30, 33–7, 39–42, 44–5, 46n4, 155–6, 265, 279–81 Georgakopoulou, Alexandra 78, 83, 122–4, 258 Gerrig, Richard J. 169–70, 177 Gibbons, Alison 124 goals in games 68, 74–84 passim, 92, 94, 99, 103n2 Goffman, Erving 93, 113, 258, 262 Goodman, Nelson 3 Gosling, Samuel D. 259 Green, Melanie C. 130 Grishakova, Marina 293n4 Gubrium, Jaber F. 113, 202n5 Günthner, Susanne 258, 268 Halliday, MAK 122 hallucination: experience and 188, 199–201; language and 190–2, 195–201; linguistic turn and 196, 200; logos and 196, 199–200; Romanticism and 191; social constructionism and 197–8; the sublime and 190–1; veracity of 190, 192, 194 Hamlet 56–7 Hannah (Biblical) 56–7 Haralovich, Mary Beth 247 Harper, Richard 124 Hasan, R. 122 Hatavara, Mari 103n1, 241, 295 Heinke, Jörg 174 Heinze, Rüdiger 156 Herman, David 2–4, 12, 46n2, 50–1, 53, 56, 62n1, 107–8, 123, 127–9, 149, 161, 164, 165n4, 225, 238, 241, 253, 278, 286–8, 291, 293n2 heterodiegetic narration 34, 205–6, 211–14, 218–19, 281, 286, 290 Hewitt, Lynne E. 123, 129–30 Hibbert, Samuel 202n2 Hitchens, Michael 68–9 Hogan, Patrick Colm 169–70, 177–8 Holmes, Dylan 67 Holstein, James A. 113, 202n5, 248 homodiegetic narration 205–6, 211–13, 216, 218–19, 281 horror 171, 175, 179 horror literature 167–8, 170–2, 174, 185 Hühn, Peter 168–70 Hull, Thomas 172 Hutcheon, Linda 122

310 Index Hutcheon, Michael 122 Hutto, Daniel 223, 225–6, 228, 241, 253–4n1, 293n2 Hydén, Lars-Christer 150 Hyvärinen, Matti 33, 122, 150, 241, 273n2, 274n6, 293n2 Identification Principle 131, 135 identity and narrative 113, 257–60 Ijsselsteijn, Wijnand 140n3 imagination vs imagery 58–61 immanence 199–200 immersion 17, 52, 54, 58–60, 83, 167–70, 173–4, 176–8, 185, 297, 300; blocking of 30, 185, 301; conceptual 168, 177, 181, 183; emotional 168, 177, 182, fictional improvisation—and narratives 108–9, 118 India Cycle 147–50, 152–3, 155, 161 India Song 148, 159, 161–2, 164 intentional stance 241–2, 246; embedded 250 interactive narrative 71–2, 79, 84, 89, 98 interactivity 71, 258, 271 intermediality 31, 33 intermentality: distorted 241; embodied 242; materialized 231 interpretation 4, 12, 56–7, 61–62, 72, 100–1, 111–2, 115, 118, 129, 138, 196, 198, 225, 245, 248, 252, 260, 278, 280, 288–9, 292; mind attribution as 148–58 interview: narrative 257, 259–60; oral history 257, 260–1, 266, 273, 279 Ioana Vultur 53 Iversen, Stefan 137–8, 149, 152, 165n4, 225, 291 Jackson, Rosemary 107, 115, 117 James, Montague Rhodes 172–3, 175 Järvinen, Aki 4 Jay, Mike 187, 189 Jefferson, Gale 260 Jenkins, Henry 12, 24, 69 Jenks, Chris 107 Johansson, Christer 54 Johnstone, Keith 109 joint communicative space 127, 138 Josselson, Ruthellen 257 Jost, François 32 Joyce, James 26, 209 Juul, Jesper 4, 76, 103n1

Kafalenos, Emma 154 Kaprow, Allan 111 Kawin, Bruce F. 163 Keating, Patrick 247 Kim, John H. 75 King, Stephen 30, 167 Klastrup, Lisbeth 83, 89 Klevjer, Rune 4 Klimek, Sonja 30–1, 41, 44 Knuuttila, Sirkka 150–1 Koten, Jirǐ 12 Krämer, Nicole C. 259 Kress, Gunther 133 Krogh Hansen, Per 51, 62n1 Kurkowska-Budzan, Marta 273n5 Kuzmičová, Anežka 59–61 La femme du Gange 148, 153, 161 Labov, William 124, 133, 202n5, 259–60, 263, 273n4, 296 Landa, José Ángel García 150, 152 Lang, Sabine 29, 31, 36–7, 40–1, 46, 46n5 Langellier, Kristin 258 Le ravissement de Lol V. Stein 148, 150–1, 153–6, 158–61, 164 Le vice-consul 148, 152, 154–60, 162, 164 Leader, Darian 238n1 Lehrskov, Ulrik 116 Licoppe, Christian 123 Lieblich, Amia 257 life-likeness, narrative wholeness 54 Linde, Charlotte 258 linguistic resources 125–6 literary fiction 2, 50–1, 56, 61–2, 77, 242, 246–52 passim literary studies 1–2, 4, 29, 58 Lloyd, Annemaree 111 location: established in texting 123, 125–39 passim Looy, Jan Van 89 Lovecraft, H.P. 172, 176, 174 Lucius-Hoene, Gabriele 257–8 Luckmann, Thomas 197, 203n6 Ludlow, Fitz Hugh 189–92, 195–6 ludology 4, 69, 76 Machen, Arthur 168, 178–85 Machilek, Franz 259 Mains, Geoff 108–9 Mäkelä, Maria 122, 149, 151–2, 299 Malina, Debra 30 Marcus, Bernd 259

Index  311 Mass Effect -games 67–85 passim Matheson, Kim 112, 114 Mäyrä, Frans 70, 83 McAdams, Dan P. 257 McDonald, Mary Catherine 273n2 McHale, Brian 31, 253, 287–8 McJannet, Linda 137 McKenna, Terence 187–8, 193–201, 203n6 Mein, Georg 172 mental functioning 206, 253 mental life 216 mental model 3, 53, 59–61, 75–6, 128, 130, 278 mental space 131–138 mental state; performed or constructed 244–8; mental state attribution 205–6, 211, 213, 215, 219, 242 Meretoja, Hanna 165n2 metalepsis: ascending 41, 43; descending 41; diegetical 35; discursive 35; horizontal 41–4, 46n6; minimal 35–6; narrative 29–30, 34, 40, 45; ontological 35, 44; rhetorical 35–6, 39; vertical 40–3 Meyer-Minnemann, Klaus 29, 37, 41 Mildorf, Jarmila 115, 118, 124, 128, 138, 279, 291, 296 mind attribution 223–5, 232–3, 236, 242, 245, 250, 252, 254n1, 278, 291, 297; see also mental state attribution mind construction 1, 149, 161, 206, 209, 242 mind reading 51, 56–7, 148–9, 152, 161, 213, 223–8, 233, 236–8, 245, 253, 272, 297 mind: accessibility of 51, 149, 225, 238, 242; and genre 6, 225, 241–2, 248–9, 253, 254n1, 256; inaccessibility of 149–51, 225, 238, 292–3; materialized 241, 245–6; nexus between world and: 4, 241–2; performed 241, 244, 247–8; transparent 2, 51, 149, 163, 225 Mishler, Elliot G. 124, 260 Mitchell, David 18, 26 monster 171–3, 177 monstrosity 171, 174, 178, 184; see also monster Montola, Markus 116 Montoro, Rocío 124 Morris, Dave 72 Morson, Gary Saul 226, 269 Mortensen, Torill E. 113

Moser, Charles 106, 112 movement: in text messages 125, 127, 137 multiverse 22 Murray, Janet Horowitz 4, 71, 73–7, 81 Myers, David 107 narratability 25, 167–9, 202n5, 249–52 narratee 30, 41–43, 211–12, 218–19, 261–2 narrative distancing 167–8, 173, 175, 178, 185 narrative embedding 21 narrative level 11, 29, 31, 36, 41–2, 44 Narrative Practice Hypothesis 225 narrative situation 22, 218–9, 246, 281 narrative: factual 50; emergent 79–80; internet 260; literary 15, 123, 127–30, 138, 163, 273, 296, 299–301; natural 51, 238, 296–7; oral 123, 126–30, 138, 202n5, 265; unnatural 137, 281, 296–8, 301 narrativity: in Duras’s India Cycle 152, 162, 165n2; in sadomasochistic encounters 110; in text-messages 123–4, 138–9; in The Mass Effect Trilogy 67, 69, 76, 78–9; in traditional ballads 205; natural 296; prototypical 297; unnatural 301 narrativization 7, 160, 187–8, 198, 200–1, 248, 296–7; fallacy of 56 narrator functions 261, 273 naturalization 56, 152, 187, 196, 198, 200–1, 297 Nelles, William 280 Newmahr, Staci 106–8, 112–113, 116 Niederhoff, Burkhard 266 Nielsen, Henrik Skov 137–8, 155–7, 163–4, 165n1, 247, 279–81, 289–90 Nijnatten, Carolus van 230 nonfiction: nonfictional monsters 171; and fictionality 280, 291 Nordling, Niklas 106–7, 110 Norman, Donald 75 Nunberg, Geoffrey 131, 133, 135, 137 Oatley, Keith 129 Ochs, Elinor 258, 269, 296 Old Testament 56; see also biblical texts Olsen, Stein Haugom 280 ontological approach 12, 17, 301 Ortmann, David M. 113 Osborn, Joseph C. 100

312 Index Page, Ruth 68, 122, 124, 259 Palmer, Alan 2, 223–6, 253, 253n1, 279, 282 paradoxical narrative devices 29–31, 33, 34, 40, 42, 44–5 paranoia, paranoid psychology 151, 223, 228, 231, 236–8, 238n1 paraxis 107 Pasupathi, Monisha 258, 263, 268–9, 272 Patron, Sylvie 16 Pavel, Thomas 12 Pearson, Craig 102–3 performance: as player performance 4, 79–80, 83–4; discursive and bodily 245, 250; as social exchange 91, 117; narrative 30, 33, 78, 83; of ballads 208–9, 213; of mind attribution 250; of self 257–8, 272 Peterson, Eric E. 258 Petta, Paolo 139 Phelan, James 72, 77, 261–3, 269, 273, 280, 290, 300 photography 32, 175, 229, 278–9, 282–4 Pier, John 22, 33–6, 41–3 plot 17, 20, 24–25, 27n4, 27n6, 37, 39, 103n1, 178, 208, 249, 298; and genre 24, 167, 301; multiple 21; secret 77; unified 22 Porter, Gerald 208–9 positioning – in digital role-playing games 69, 74, 76, 78–9, 82, 84 possible worlds theory 12, 16, 22, 44, 89 possible worlds: and everyday world 128; and narrative fiction 17, 59; in game 6, 87–89, 94–103; in online environment 110–11, 114, 116–18; ontology of 16–17, 87 post-classical narratology 5, 46n2, 149, 157; see also classical narratology Prince, Gerald 50, 69, 169, 178 procedurality 71 proliferation: medial 12; narrative 12; ontological proliferation 12, 22; textual 12 qualia 224, 253, 278 Rabau, Sophie 46n6 Rajewsky, Irina O. 31 Rank,Stefan 139 recentering 89, 169–70, 173, 176–7 Redd, William H. 130

Reichert, Ramón 259 Richardson, Brian 33, 137–8, 164–5n1, 238 Ricoeur, Paul 56, 107, 257, 263 Riessman, Catherine Kohler 226, 260, 274n13 Riva, Giuseppe 140n3 Robbe-Grillet, Alain 151–2, 165n3 role-playing games (RPGs) 67–85 passim Rollings, Andrew 72 Ryan, Marie-Laure 4, 30, 35, 38, 44, 45, 50, 54, 59, 69, 71–2, 87–9, 96, 116, 169–70, 174, 237, 241, 249–50, 258 Ryle, Gilbert 2 Saarenheimo, Marja 150 Sacks, Harvey 260 sadomasochism 106–18 passim Sali, Serdar 98 Samuel, Ben 100 Sanders, Robert E. 256 Sandino, Linda 265 Sarraute, Nathalie 148, 152 Schaeffer, Jean-Marie 46n1, 53 Schegloff, Emanuel A. 260 Schiffrin, Deborah 124 Schmid, Wolf 168–9 Schober, Regina 31–2 Schütz, Astrid 259 scripting: in digital games 74–6, 79–81; sexual scripts 109–12, 114–15, 118 Searle, John R. 129, 206 Second Life 114 second-person perspective / third person perspective 226–7 Segal, Erwin M. 129 self: implied 115, 263, 266; performances of 258; selfpresentation 113, 258 self-reflexivity 4, 19, 30 sequentiality 123–5, 131, 139 Sicart, Miguel 70–1, 73–4, 80 Siegel, Carol 107 Sihvonen, Tanja 77 Sismondo, Sergio 203n6 situation model 128 Sixma, Tjarda 114–5 Skalin, Lars-Åke 50–2, 54, 56–7, 61, 62n1, 62n3 Sklar, Howard 273 small stories 122, 258–9 Smith, Barbara Herrnstein 11

Index  313 Smith, Sidonie 263 social mind (intermental mind); 224, 226, 238, 240–1, 253–4n1 social simulation 88–94 spatial mapping 128 speech-category approach 279 Sprott, Richard A. 108, 113 stage directions 15, 137 Stanislavski, Konstantin 109, 111 Stear, Nils-Hennes 107–8 Stephens, Alastair 103 Stephens, Christine 122 Sternberg, Meir 55, 298 Stevenson, Robert Louis 176 Stoker, Bram 176 story: afterstory 14; backstory 14, 68, 90, 93, 95; embedded 21–2, 258–9; pre-existing 51; story generation 88, 102 storytelling: conversational 256–8, 263, 265, 272–3, 296–7; everyday 262, 279, 289–91; storytelling situation 258, 263, 269, 279, 289, 291 Strauss, Neil 113 Strawson, Galen 165n2 structuralist poetics 197–8 Suoninen, Eero 230 Survivor (reality TV series) 240–53 syllepsis: horizontal 41; narrative 36–7, 39–40, 45; vertical 40 Tagg, Caroline 123 Tamboukou, Maria 150 Tannen, Deborah 268 Taylor, Charles 199 Taylor, Charles 199 tellability see narratability Tellegen, Auke 139n1 temporality 31, 35, 36, 124 text-messages 122–39 passim The Great God Pan 168, 178–80, 182, 184–85 Theory of Mind 93, 102, 206, 223, 225, 242, 245–6, 252, 253n1 Thibault, Paul J. 122, 125 third-person narration 15–16, 27n2, 150, 152, 155–6; in ballads 205–6, 211–13, 218–19; in text-messages 133, 136–7; see also heterodiegetic narration Thornborrow, Joanna 258 thought report (psycho-narration) 226, 238, 282, 287–9 Thumim, Nancy 280

Torner, Evan 109 traditional ballad 208–10 transgression: in metalepsis 40–6; in sadomasochism 109, 116, 118 transmediality 31–4 transportation 52, 58, 60–1, 130, 139n1, 169–70, 177; see also immersion Trosset, Michael W. 247 Turk, Horst 173, 183 universals: cognitive 3, 241–2, 245, 302; narrative 241, 302 universe 1, 3, 13 191, 250, 299; (meta) diegetic 30, 41–2; game 82, 84, 100 unreliable narrator 156, 237, 242, 290 vantage point 125, 128–9 Vazire, Simine 259 verbal imagery, speech-imagery, rehearsal imagery 59–60 Vilhjálmsson, Hannes 135 virtual reality / virtual world 23, 89, 107–8, 112, 114–118, 135 voice: narratorial 33–4; experiential 246; expressive 245; noncommunicative 247 voice-over 26n2, 34, 37–9, 162, 246–7 Vultur, Ioana 53 Wagner, Frank 30, 46n6 Waletzky, Joshua 124, 202n5, 263 Walsh, Richard 16, 51, 55, 62n3, 74, 76, 80, 165n5, 280, 290 Walton, Kendall L. 55, 57 Wardrip-Fruin, Noah 72, 83, 88, 94 Watson, Julia 263 Weinberg, Martin S. 106 Weinberg, Thomas S. 108 Werry, Christopher C. 137 Williams, Colin J. 106 Wilson, Christopher 253 Winter, Stephan 259 Wisker, Gina 172 Wistrand, Sten 56–7, 61 Wolf, Mark J.P. 24 Wolf, Werner 30–2, 34, 45, 176–7 Wolosky, Shira 164 World of Warcraft 114 world: complete 18; diegetic 36; embedded 22, 88; extradiegetic 43; game world 68–9, 73, 75–6, 79, 83–4, 89; incomplete 17; narrated 35, 36, 129; narrative 3, 5, 58, 169;

314 Index of representation 41; ontology of 3, 5, 12–13, 38, 41–4, 169–70; represented 3; textual 4, 11; virtual 3, 5, 89, 107, 114–16, 137, 164; see also fictional worlds and possible worlds worldmaking 3, 26, 53, 59–61 worldness 3, 17, 24–5, 82–3, 89

Yost, Megan R. 107 Zamorski, Krzysztof 273n5 Zubin, David A. 129 Zunshine, Lisa 223, 225, 241–2, 246, 252, 253n1, 288 Zurbriggen, Eileen L. 107 Zwaan, Rolf A. 128