How to Make Believe: The Fictional Truths of the Representational Arts [49, 1 ed.] 9783110441536, 9783110443875, 9783110435726

Citation preview

How to Make Believe

Narratologia

Contributions to Narrative Theory Edited by Fotis Jannidis, Matías Martínez, John Pier, Wolf Schmid (executive editor) Editorial Board Catherine Emmott, Monika Fludernik, José Ángel García Landa, Inke Gunia, Peter Hühn, Manfred Jahn, Markus Kuhn, Uri Margolin, Jan Christoph Meister, Ansgar Nünning, Marie-Laure Ryan, Jean-Marie Schaeffer, Michael Scheffel, Sabine Schlickers, Jörg Schönert

Volume 49

How to Make Believe The Fictional Truths of the Representational Arts Edited by J. Alexander Bareis and Lene Nordrum

ISBN 978-3-11-044153-6 e-ISBN (PDF) 978-3-11-044387-5 e-ISBN (EPUB) 978-3-11-043572-6 ISSN 1612-8427 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2015 Walter de Gruyter GmbH, Berlin/Boston Printing and binding: CPI books GmbH, Leck ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

Table of Contents J. Alexander Bareis and Lene Nordrum Introduction | 1

Section 1 – Theory Stein Haugom Olsen The Concept of Literary Realism | 15 Peter Lamarque Thought, Make-Believe and the Opacity of Narrative | 41 James R. Hamilton Narrative per se and Narratibility | 61

Section 2 – Literature and Film Tobias Klauk and Tilmann Köppe Distance in Fiction | 77 Sarah E. Worth Narration, Representation, Memoir, Truth, and Lies How We Diminish the Art of Narrative with Simple Truths | 95 Remigius Bunia Truth in Fiction | 113 Ira Newman Destabilizing Reality Postmodern Narrative and the Logic of Make-Believe | 139 Jukka Mikkonen The Place for External Considerations in Reading Literary Fiction | 155

VI | Table of Contents J. Alexander Bareis Fictional Truth, Principles of Generation and Interpretation Or: Why it is Ficionally True that Tony Soprano was Shot Dead | 165 Mario Slugan Deixis in Literary and Film Fiction Intra-ontological Reference and the Case of Controlling Fictional Narrators | 185 Liviu Lutas Metalepsis and Participation in Games of Make-Believe | 203 Sonja Klimek ‘I grieve’ as Make-Believe Generating Fictional Truth in Eighteenth-Century Lamentation Poetry (Günther, Haller, Young and Novalis) | 223

Section 3 – Theatre and Music Frank Zipfel Fictionality and Make-Believe in Drama, Theatre and Opera | 245 Matthew DeCoursey Making Meaning in the Theatre: Double Noesis | 269 Eran and Inbal Guter Impurely Musical Make-Believe | 283

Section 4 – Games E. M. Dadlez Make-Believe Wickedness vs. Wicked Making-Believe RPGs, Imagination and Moral Complicity | 309 Jason D’Cruz Agency and Volition in Make-Believe Worlds | 323

Table of Contents | VII

Chris Bateman Prop Perspective and the Aesthetics of Play | 335

Index | 353 Contact Addresses | 361

J. Alexander Bareis and Lene Nordrum

Introduction

In 1990, two fundamental books on the notion of make-believe were published: Kendall Walton’s Mimesis and Make-Believe and Gregory Currie’s The Nature of Fiction. Both books explain fictionality in analogy to children’s games of makebelieve, and in both books, the notion of imagination is central. Both works have greatly influenced research in the field of aesthetics in analytical philosophy, but perhaps more importantly, they have contributed to widening the notion of make-believe and its explanatory power beyond the domain of philosophy with the result that the make-believe explanation of fiction has become ‘Theory’. In the words of Jonathan Culler (2011, 3): “Works regarded as theory have effects beyond their original field.” Once the notion of make-believe started to spread into such areas as literary criticism and film studies, one could argue that it became ‘Theory’ with a capital ‘T’. Today, some twenty years later, make-believe theory is used in a wide array of different academic disciplines, and researchers from subjects as different as psychology and mathematics make use of the theoretical concept. The strongest impact in the last two decades, however, has been within the representational arts and its core fields of literature, film and visual arts where make-believe theory is on firm ground with numerous refinements and critical discussions to build on. Other areas are still in need of further investigation. The core idea behind make-believe theory is its analogy between the reception of works of art and the basic function of children’s game of make-believe. Two concepts serve key roles in this respect – imagination and fiction. Imagination is central to the understanding of how activities are connected to the appreciation of art, and fiction is central to the way scholars have divided the turf in various disciplines over a long time. To be more specific, once a game of make-believe is in process, participants will generate fictional truths according to the unwritten, but mutually accepted rules of the game. According to Walton, this game of make-believe works approximately the same way for all types of representational arts, e. g. novels, films, poems, pictures, plays, statues, dance and even music. Thus, works of representational art have in common that they are used as props in games of make-believe with the result that the ‘mechanics of generation’ of fictional truths are at work. Both Walton and Currie have given thorough accounts of how these mechanics of generation can be described, with rich examples from a variety of different fields within the representational arts, above all from literature and film. They have also emphasized that all the different categories of representa-

2 | J. Alexander Bareis and Lene Nordrum tional art generate fictional truths in their own special way. Thus, far from taking a position, as described and criticized by Peter Kivy (1993, 131), of philosophers who “hover above” their “subject matter like Gods from machines, bestowing theory upon a practice in sublime and sometimes even boastful ignorance of what takes place in the dirt of the mess of the workshop”, they have not been afraid of getting their hands dirty. In fact, in his presidential address from 2004 to the Society of Aesthetics and Art Criticism, Walton (2007, 156) responds to Kivy’s presidential address from 1992 by rhetorically asking: “But why must we choose between attending to particulars and developing grand theories?” We are convinced that such combination of top-down and bottom-up perspectives as made possible in make-believe theory provides fertile ground for further exploration of various similarities and differences between possible and specific instances of fiction as make-believe in the representational arts. Indeed, one of the starting points for the outline of the present volume was a wish that the academic practice in the aesthetics need not be an ‘either – or’ of theory or interpretation, or, as stated above, between the ‘big picture’ and the particular. Inspired by Walton and Currie, who have made their theoretical points by engaging deeply with different artworks resulting in genuine and exciting insight into the practice of a wide variety of cases within the representational arts, from pictures to novels and plays, we embrace the idea that since there is no inherent mutual exclusiveness of the general and the particular, there is no reason for mutual exclusiveness of theory and actual analyses of works of art, and hence of interpretation. Yet, it is certainly true that you cannot usually have your cake and eat it. Socalled grand theories come with a price, which is often paid by not being able to cover a large amount of very similar examples. Instead, similarities between different areas take precedence over thick analysis and depth. As Walton (2007, 157) puts it: “Grandeur consists in including lots of things under a single umbrella, highlighting what is in common among a wide array of instances.” In the long run, however, we believe that it will not suffice to deal with every possible detail of a limited number of examples, the hands need to get more dirty than that. The most obvious answer to this challenge, of course, is that grand theories need to be proven or falsified by collective efforts, which is the other main starting point behind this anthology. Our goal is to provide some further steps in the collective effort of testing, applying, proving and, whenever necessary, also falsifying the insights of aesthetic analytical make-believe theory for the representational arts. The question posed by the title of this volume – How to make believe?– is therefore meant as a starting point for a more thorough investiga-

Introduction | 3

tion into some of the vexing problems identified so far by a grand theory of representation in terms of make-believe. In slightly different terms: By asking ‘How?’ collectively, we wish to draw a wide range of different arts and different works of arts from different times of art history under the ‘umbrella’ of close inspection – but, again, not necessarily only for the reason of finding similarities in order to elaborate theory. Rather, we wish to point out important similarities across various art forms and ways of artistic expression, and in this process shed light on some of their differences as well. In this respect, we observe that aesthetic phenomena and puzzles represent the type of similarity of address to which the notion of make-believe has provided useful theoretical advances across different art forms – which is in line with Walton’s wide perspective in the use of make-believe theory. In his original work, Walton addressed works of literature, theatre, paintings, sculptures, photographs, films and music – and distinguished between verbal and depictive representations where the former comprise works such as novels and plays, and the latter paintings and photographs. Despite such wide attention, however, the framework of representational arts obviously includes even more artistic ways of expression than have received wide scholarly attention so far. To return to our observation at the outset, namely that make-believe theory is on relatively firm ground in the core areas of literature, film and visual arts, a third aim of this book is to expand the discussion to neighboring media. In so doing, the contributions of the book zoom in on an aspect which has been partly neglected so far, namely the role of narrative structures across different categories of representational arts. One of the reasons why this question is of value is that it has the potential of merging bottom-up and top-down perspectives. To be more specific, although narrative form is by no means essential for the notion of make-believe applied to representational art, it is central to most forms, and carful investigations of narrative structures across different art forms can therefore feed into general discussions of similarities as well as media-specific discussions from the perspective of one media. Crucially, since narratology is a well-established field, the discussions provided in this book can both contribute to depth of analysis through comparisons with similar examples addressed in the scholarly discussions within their respective art forms, and to generality through the lens of make-believe theory and similarities across art forms. The representations from a wide array of representational arts presented in the present volume all convey a form of narrative structure, albeit arguably somewhat dependent on one’s choice of underlying definition of narrative. In combining the discussions under the general umbrella of make-believe, we therefore hope to merge the grand perspective with the specific. The contributions have been divided into four sections. Section one, Theory, mainly deals

4 | J. Alexander Bareis and Lene Nordrum with theoretical issues of narration; section two, Literature and Film, includes a number of contributions focussing on different forms of literature and visual art forms, such as novels, memoirs, poetry, films and TV-series, but also photographs and pictures; section three, Theatre and Music, discusses performative art forms, including drama, theatre, opera and classical music; and section four, Games, deals with different forms of games, from virtual reality to role-playing games. The chapters in each section are presented below in their order of appearance. In the first section, Theory, we find contributions from Stein Haugom Olsen, Peter Lamarque and James Hamilton. STEIN HAUGOM OLSEN addresses the fundamental question of realism in literature. Discussing postmodernist views and criticizing them as monocriterial, Olsen investigates whether a criterion such as ‘objective/truthful representation of reality’ is a useful critical tool. He concludes that it leads to wrong assumptions about whole periods of literature, and therefore dismisses the concept as a valuable critical instrument. Instead, he suggests a “radically conventionalist view of realism” which approaches the concept of literary realism through realist works. Literary realism can be defined from the point of view of works in the realism period, Olsen claims, and this definition then opens up for a recognition of formal properties for the realist mode of writing that can be identified in works outside the period. Olsen thus sharpens the critical value of the concept literary realism, and at the same time pushes the epistemological problem of the relation of art to reality outside the scope of the concept. PETER LAMARQUE’S paper draws on his earlier work on Thought Theory, which addresses the question of how fictional works can elicit emotional responses to those works even if the reader knows that the work is fictional. Exploring the idea of narrative opacity, Lamarque argues that the cognitive benefits of literature need to be regarded through the lens of how a work’s content is presented, and how readers engage with the specific structuring of that content. In this discussion, Lamarque raises questions about the possible impact of the expectation on the part of the reader of what Lamarque refers to as literariness on the imaginings prescribed by fictional narratives. Lamarque’s paper thus offers an illuminating theoretical discussion of how Walton’s cross-medial theory with its focus on make-believe and imagining can be refined to explain nuanced responses to specific fiction genres. JAMES HAMILTON puts the question of narrative per se at the center of his investigation. For the sake of the argument, Hamilton assumes that there is no prior distinction between fictional and nonfictional narrative. If that is the case, he asks: What mental mechanisms, if any, would be involved in grasping and projecting narratives? Using a minimalist definition of narrative, Hamilton

Introduction | 5

seeks to explain narratives per se as bearers of basic aesthetic predicates, such as anticipation, and concludes that the so-called paradox of fiction, which is concerned with questions of emotion and suspense towards fictional representations, can be understood in a better way if examined from this minimalist perspective. In addition, the minimalist definition highlights transmedial aspects. In the second section of the book, Literature and Film, the chapters mainly discuss examples from literature and film, but even other forms of art are paid attention to, such as paintings, photographs and TV-series, sometimes also from an intermedial perspective. The section includes contributions by Tobias Klauk and Tilmann Köppe, Sarah E. Worth, Remigius Bunia, Ira Newman, Jukka Mikkonen, J. Alexander Bareis, Mario Slugan, Liviu Lutas and Sonja Klimek. Through philosophical stringency, TOBIAS KLAUK AND TILMANN KÖPPE take intitial steps towards a clear definition and explanation of the distinction between telling vs. showing in literary fiction. Their paper draws attention to the fact that although this distinction is often referred to in narratological research, it remains blurry beyond the general definition that telling describes the events in the literary work in a summary fashion, while showing gives a more detailed account of the going-ons of the storyworld. In their pursue of pinning down how the idea of ‘more detailed’ can be understood, Klauk and Köppe use the notion of prescription and distance and flesh out a theory of distance with the theoretical potential of explaining a narratologically salient, but hitherto slippery, distinction. SARAH E. WORTH’S paper tackles questions related to truth-depiction and genre-definition in the case of memoirs. From the perspective of Walton’s makebelieve theory, memoirs represent a particularly tricky case. To Walton, fictional worlds are dependent on fictional imaginings, Worth reminds us, but, interestingly, memoirs depend on a type of narration of events that essentially prompt very similar type of imaginings. Partly due to the shifty nature of personal memories, memoirs thus characteristically navigate the space between truth and fiction, narration and recount. According to Worth, a mistake is therefore made if memoirs are merely regarded primarily from the perspective of how well they fit facts. Rather, she concludes, memoirs represent a literary genre whose narratives should be regarded as primarily interpretive, and only secondarily as representations of truth. When memoirs are regarded in this way, cases in which memoirs are pulled of the shelves and harshly criticized because they do not line up with fact can be viewed in a different light. The distinction between fact and fiction is reflected in the contribution by REMIGIUS BUNIA as well. Bunia revisits the thorny question of how imaginations generated through fiction and nonfiction differ. He argues against a view where

6 | J. Alexander Bareis and Lene Nordrum this distinction can be regarded as emanating from linguistic form, in that language can be truly propositional, i. e., purely truth-generating, or truly aesthetogenic, i. e., purely perception-generating. Rather, Bunia develops the essentially cognitivist assumption that to distinguish between reality and fiction, we must look to mental processes rather than linguistic expression, and in so doing we will find that the distinction between fiction and nonfiction concerns the specific ways in which humans process their sensual experience to deal with reality. Bunia concludes that “the cultural and cognitive phenomenon called fiction resides in the plane spanned by the axis of liberty in communication and the axis of aesthetogenic expression”, meaning that the distinction between propositional and aesthetogenic is gradual. Intriguingly, Bunia’s notion of graduation also entails that nonfiction must be regarded in much the same way: the more a nonfiction text explores the nature of aesthetogenic expression, the more it develops a literary character. Bunia concludes that the concept of graduation outlines the contours of a particularly nuanced theoretical framework for the distinction between fact and fiction – a theoretical framework based on novel and, according to Bunia, more convenient, epistemological ground. IRA NEWMAN addresses the notion of appropriate vs. inappropriate imaginings in relation to the postmodern novel Jealousy, written by Alain RobbeGrillet. Newman particularly deals with a textual passage comprising no less than five inconsistencies, or logical contradictoriness, which makes it difficult to theoretically explain what imaginings the text might prescribe. According to Newman, however, Walton’s analysis of prescriptive imagining offers a way to understand such logical contradictoriness. That is, the notion prescriptive imagining includes the possibility that the imaginer can be directed to conduct an act of imagining that cannot be completed. Thus, the flexibility demonstrated in Walton’s account of prescriptive imagining, Newman argues, gives promise of a framework that can deal with postmodern texts such as Jealously – where realistic relationships of the kind we are used to finding between representational artwork and ‘the real word’ are undermined and rejected. Literature is the art form in focus also in the following article by JUKKA MIKKONEN. In his paper, Mikkonen discusses readers’ imaginative engagement with literary fiction particularly from the point of view of readers’ external considerations. Mikkonen defines imaginative engagement as involving reader responses both while reading and after reading a literary work, and proposes that a reader’s external perspective and imaginative engagement involve a range of different mental attitudes and activities that encompass several stages of literary interpretation. His definition of imaginative engagement broadens the concept in a way that enables Mikkonen to shed light on how readers’ external consid-

Introduction | 7

erations on literature may play into the work’s status as a work of fiction or a literary work, a perspective that is rarely addressed in the literature. A closer look at intermedial narrative structures is offered in the contribution by J. ALEXANDER BAREIS. He focuses on the generation of fictional truths in different forms of art. Based on well-known principles by Walton and Lewis, such as the Reality Principle and the Mutual Belief Principle, Bareis develops and exemplifies a new principle, the Principle of Media and Genre Convention, which becomes highly important in transmedial discussions. He illustrates this principle by a close reading of the last scene of the HBO-series The Sopranos, and highlights the complex relationship between interpretation and the generation of fictional truth. As Bareis shows, the relation between principles of generation and interpretation is comparable to issues discussed in the field of narratology, between narratological analysis and interpretation. Another paper with focus on narratives with different medial forms of representation is offered by MARIO SLUGAN. Slugan’s article engages in the discussion of the existence of fictional narrators, and addresses arguments presented by scholars such as Noël Carroll, Gregory Currie, Berys Gaut, Andrew Kania and George Wilson. Based on the deictic properties of verbs, Slugan argues for “the near-ubiquity” of controlling fictional narrators (CFN) in literature, but against the generalization of this argument to film. Slugan’s contribution is twofold: he convincingly adds new territory to the theorizing of fictional narrators, and he weaves together threads of arguments from different media, thus facilitating the quest for a full-fledged theory and understanding of fiction and fictionality across the representational arts. LIVIU LUTAS, next, explores metalepsis as a transmedial concept through examples from painting, photography, literature and film. Lutas’ primary point of departure is that narrative metalepsis is not restricted to a narrative voice, but rather deals with different levels of representation. According to this view, metalepsis is not restricted to literature, but becomes a full-fledged transmedial concept. The paper illustrates how one concept can shed light on both similarities and differences regarding the concept itself and the art forms in which it is argued to operate. Lutas thus adds both to the understanding of the art forms he addresses as well as to refining the definition of metalepsis as a theoretical construct. SONJA KLIMEK’S paper brings the volume back to literature, more specifically, to lyric poetry. She points out that the assumption that lyric poetry is necessarily fictional represents doctrinal knowledge in the English literature, but is challenged in the German literature even if the discussion on truth and fictionality in lyrics resonates at least back to Plato. Klimek’s point of departure is that there is a need to continue this discussion, and she specifically tackles the question of

8 | J. Alexander Bareis and Lene Nordrum how the principles for the generation of fictional truth are at work in lyric poetry. She underlines that to understand this process, a dual focus on production and reception is necessary. Using examples from eighteenth-century lamention poetry, Klimek concludes that since these poems typically lack clear signals of fictionality (such as genre names), contemporary readers cannot know whether the poem was meant to be fictional, and must therefore resort to constructing a game world in their own “game of make-believe” (Walton 1990), following, for example, the principles of minimal departure or the genre conventions of the time – if they have access to them. This means that distinguishing between truth and fiction for lyric poetry rests on a number of variables, including the historical context in which a poem was written and the genre conventions of the time of writing, and it follows that generalization across time and poems becomes difficult – if not impossible. The third part of the book, Theatre and Music, includes contributions from Frank Zipfel, Matthew DeCoursey and Eran and Inbal Guter. FRANK ZIPFEL’S works towards a multi-layered approach to how we can define transmedial fictionality in a theoretical account of fictionality applicable to drama, theatre and opera. The theoretical account rests on the tenet that fictionality designates phenomena that deal with fictional worlds. As the concept of fictionality is broadened beyond the literary narrative to stage performing art forms, Zipfel shows how it is essential to recognize that these fictional worlds evolve, and are dealt with, in symbiosis with the respective institutional practice. In essence, this means that he is able to illustrate how drama, theatre and opera are similar at the general level in functioning as fictional works, but different at the level of institutional practices in how the game of make-believe is dealt with and understood – both from the perspective of the stage performer and the audience. Similarly to Zipfel, MATTHEW DECOURSEY discusses questions regarding drama and play, but his focus is on performative actions. He specifically concentrates on theatre performance as an aesthetic object and suggests that the aesthetic experience of both actor and audience can be understood through the idea of double noesis, a concept taken from Husserl. Drawing on research from philosophy, psychology and neuroscience, DeCoursey gives a detailed account of how a process of double noesis is possible – with particular emphasis on the actor’s perspective. Central to DeCoursey’s account is the concept of emotion. Since actors are aware of the theatre performance as an aesthetic object, DeCoursey argues, they become capable of maintaining a double perception that juggles the perspectives of their stage character and actor persona – including the different emotions related to these perspectives. DeCoursey concludes that

Introduction | 9

this dual perception and the contingent double emotion are crucial in explaining the aesthetic force of theatre. The following paper broadens the realm of art forms under discussion even further, turning the attention to music. ERAN AND INBAL GUTER use Walton’s concept ornamentality to show how two types of listening strategies which has received considerable attention in Anglophone philosophy of music, the so-called narrative and formalist listening styles of instrumental music, can be considered as two ends of a continuum rather than as binary properties. Importantly, such a continuum allows for a type of double consciousness by both the listener and the performer that enables the merging of musical understanding of the narrative and the formalist type. Guter and Guter’s paper explains a richer set of ways in which individuals may engage with instrumental music, while simultaneously developing a Waltonian concept that has received relatively little attention (ornamentality). The paper is thus an excellent illustration of how different branches of philosophy dealing with separate types of art forms can crossfertilize in a way that adds explanatory power to perspectives forced into a gridlock in single theoretical branches. An entirely different form of representational content is by E. M. Dadlez, whose paper opens the fourth and final section of the book, Games, with contributions by E. M. Dadlez, Jason D’Cruz and Chris Bateman. E. M. DADLEZ discusses the philosophical question of the impact of fiction on issues related to ethics and morality from the perspective of role playing games (RPGs) and thus contributes to broadening the discussion of key philosophical issues across fictional representations in different art forms. More specifically, she draws attention to the question of whether RPG participants are more likely to become complicit with the morally suspect attitudes and perspectives of the character they represent than the readers and audiences of other fictional works are likely to adopt the moral stance of the works they observe. The reason for posing this question, according to Dadlez, is the valid observation that the act of role-playing in many cases at least appears to involve a first-person perspective. As Dadlez points out, “to role-play would be to make-believe believing and thinking and feeling as the character did”, and “so make participation in RPGs morally suspect in a vast number of cases”. In her article, however, Dadlez illustrates how role-playing scenarios are more complex than one might suspect, and that RPGs does not necessarily involve more imaginative immersion of a morally suspect nature than conventional fiction. In a similar way, JASON D’CRUZ brings core philosophical questions to a new context, namely Second Life, a multi-user virtual world where people engage with each other through avatars in an online environment. The question D’Cruz poses is whether avatar life can cause moral conflicts that are real. The question

10 | J. Alexander Bareis and Lene Nordrum is pursued through the theoretical lens of human agency: D’Cruz argues against the view that the operation of virtual-world avatars constitutes real agency in all cases. According to D’Cruz, there is an important difference between virtual worlds and the real world which is related to the property volition: Due to structural features of Second Life, people are not as likely to have as strong volitions about their avatars’ actions in the virtual environment as they have about their own real-world actions. D’Cruz’s paper features methodological novelty in using the discussion of the concepts agency and volition in relation to virtual worlds to shed light on the more conventional philosophical discussion of agency and volition in real-world contexts. This is yet another example of how discussions of concepts across different media can shed light on medium-internal discussions as well as refine theoretical concepts from a more general perspective. In the final chapter of the volume, CHRIS BATEMAN shifts focus to yet another new context of philosophical interest. Using Walton’s make-believe theory, Bateman explores the concept prop in the creation of fictional worlds in computer games. The paper presents three types of prop perspectives in computer games: the toy-view, the doll-view and the table-view. Bateman shows how these different views are essential in illustrating different ways in which gamers create fictional worlds and engage with the game. He also forwards the argument that since computer games generate rich fictional worlds, they cannot, at least not in an a priori sense, be excluded from the category of art – however such a category is defined. In keeping with the theme of the volume, Bateman’s paper thus adds to our understanding of the particular workings of computer games under the general umbrella of make-believe in the representational arts. * As a final note in this introduction, we would like to mention that the present anthology is a carefully compiled selection of papers that evolved in connection with a conference held at Lund University in March 2012. Keynote speakers at this conference were Peter Lamarque, Stein Haugom Olsen, Gregory Currie and Kendall Walton, to whom we would like to express our gratitude for their discussions of make-believe theory. Both the conference and this publication were financially supported by The Royal Swedish Academy of Letters, History and Antiquities and the Centre of Languages and Literature at Lund University, to which we are also grateful. Many thanks also go to two anonymous reviewers for their constructive criticism and support towards publication in the Narratologia series, as well as to the series editors for including this collection in the series. Special thanks go to Wilhelm Schernus for careful copy editing of the entire volume in the final stage of the editing process.

Introduction | 11

References Culler, Jonathan. Literary Theory. A Very Short Introduction. New York: Oxford University Press, 2011. Currie, Gregory. The Nature of Fiction. New York: Cambridge University Press, 1990. Kivy, Peter. “Differences.” Journal of Aesthetics and Art Criticism 51.2 (1993): 123–132. Walton, Kendall L. Mimesis as Make Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Walton, Kendall L. “Aesthetics – What? Why? and Wherefore?” Journal of Aesthetics and Art Criticism 65.2 (2007): 147–161.

| Section 1 – Theory

Stein Haugom Olsen

The Concept of Literary Realism 1 In The Rise of the Novel, Ian Watt (1957, 32) specifies a set of “narrative procedures” which, he maintains, constitute what he calls formal realism, “formal, because the term realism does not here refer to any special literary doctrine or purpose, but only to a set of narrative procedures”. Formal realism is characterized by the rejection of literary convention and traditional plots and instead an emphasis on creating the impression of fidelity to life and on inventing new plots. Secondly, there is the introduction of individual characters acting against a particular and particularised real life background rather than literary types acting against a background determined by the appropriate literary convention. This particularisation of character can also be seen in the rejection of type names and the adoption of real life names that do not place the character in a category. Thirdly, there is the emphasis on the importance of time, which manifests itself in the creation of a causal chain working itself out in the plot, in the creation of characters that develop through time, and in the attention given to the historical process. As a correlative to this emphasis on time, there is a particularized description of places. Formal realism is defined not only by a consistent time scheme, but also by a consistent geography and fully realised exteriors and interiors. Finally, it is characteristic of formal realism that it employs the language of concrete particularity and avoids ornate and poetic language (see Watt 1957, Ch. 1). Formal realism, as Watt defines it, is […] like the rules of evidence, only a convention; and there is no reason why the report on human life which is presented by it should be in fact any truer than those presented through the very different conventions of other literary genres. (Watt 1957, 32)

However, the fact that works employing the narrative procedures of formal realism cannot make any special claim to truth does not mean that realist works do not have a ‘positive relationship’ to reality which distinguishes them from other types of literary works. We must not forget, says Watt, […] that although formal realism is only a convention, it has, like all literary conventions, its own peculiar advantages. There are important differences in the degree to which different literary forms imitate reality; and the formal realism of the novel allows a more immediate imitation of individual experience set in its temporal and spatial environment

16 | Stein Haugom Olsen than do other literary forms. Consequently the novel’s conventions make much smaller demands on the audience than do most literary conventions; and this surely explains why the majority of readers in the last two hundred years have found in the novel the literary form which most closely satisfies their wishes for a close correspondence between life and art. (Watt 1957, 32–33)

Watt does not make the relational feature of imitation part of the definition of formal realism. It is an empirical fact that the realist mode of writing “allows a more immediate imitation of individual experience set in its temporal and spatial environment than do other literary forms” (see Watt 1957, 32–33). Watt’s concept of realism is descriptive. Labelling a work of literature realist in Watt’s sense does not identify it as possessing a certain kind of value. And Watt does not build into his concept of literary realism a reference to a historical period. It is not part of the definition of the concept that it should contrast with romanticism and modernism. As Watt defines it, formal realism can be a feature of literary works from any period. However, the point of introducing this concept of realism is that it can fruitfully be used to characterize certain important developments in literary history. Watt (1957, 32) points out that the narrative procedures which characterize formal realism “are so commonly found together in the novel, and so rarely in other literary genres, that they may be regarded as typical of the form itself”. Watt’s method when he defines a concept of literary realism is to specify the features of the realist mode of writing and only after that point to the empirical fact that this mode of writing satisfies readers’ wishes for a “close correspondence between life and art” to a higher degree than other modes of writing (Watt 1957, 32–33). In this, he departs from the traditional and dominant way of construing realism. From the very first application of the term realism to literary works in Mercure francais du XIXe siècle in 1826 and up to today, it has been assumed that it is a defining feature of such works that they somehow tell the truth about ‘reality’: Cette doctrine littéraire qui gagne tous les jours du terrain et qui conduirait à une fidèle imitation non pas des chefs-d’œuvre de l’art mais des originaux que nous offre la nature, pourrait fort bien s’appeler le réalisme: ce serait suivant quelques apparences, la littérature dominante du XIXe siècle, la litterature du vrai. (Borgerhoff 1938, 839)1 [This literary doctrine which gains ground every day and leads to faithful imitation not of the masterworks of art but of the originals offered by nature could very well be called ‘re-

|| 1 I quote this passage from Borgerhoff (1938, 839). Borgerhoff in his turn quotes it from an article by A. David-Sauvageot (1899), On Romantisme. Translation, René Wellek (1963, 227).

The Concept of Literary Realism | 17 alism’. According to some indications it will be the literature of the nineteenth century, the literature of the true]. (Translation, R. Wellek 1963, 227)

For those authors and critics who adhered to the ‘doctrine of realism’, such as the practitioners of the nineteenth-century realist novels, it was a basic assumption of that they were telling the truth about social reality. “French Society”, says Balzac in the preface to The Human Comedy, […] was to be the historian, I had only to be the secretary. I would draw up the inventory of the vices and virtues, collect the main effects of the passions, portray characters, select the principal events of Society, and compose types of combining the traits of several homogenous characters; and I might thus succeed in writing that history forgotten by so many historians, the history of manners and morals. [...] But if I was to deserve the praises which any artist must aspire to, I must needs study the causes or central cause in that immense assembly of events. (Balzac 1972, 143)

Realism as a literary doctrine builds into the concept itself the notions of truth and reference, and critics of all colours have adopted ‘truthful/objective representation of reality’ as a central element in the concept of literary realism.2 The concept of realism it is argued, has both a ‘real life’ and a literary application and the conceptual core of both uses is the same. “What unites the two areas of meaning – the real-life and the literary meanings [of the term ‘realism] –”, says J. P. Stern, […] is their representational quality, is the fact that they both designate a ‘standing-forsomething’, and a process of selection: they designate not a content but a condition, or at least an outlook, and a form. They refer to a way of thinking (in the one case) and a way of writing (in the other), each of which is positively related to the real world. (Stern 1973, 42)

The concept of realism as the ‘truthful/objective representation of reality’ also acquires an evaluative element. The ‘truthful/objective representation of reality’ is a positive literary achievement. In Lukács (see Lukács 1963 [1957], 56), the positive evaluation embodied in this concept is taken to its limit in statements like “The great realist writer is alone able to grasp and portray trends and phenomena truthfully in their historical development”. And the view that realist works have a special kind of value because they give a ‘truthful/objective representation of reality’ is combined in Lukács, and in a number of other influential critics, with the view that the concept of realism has a special application to a certain period in literary history because in this period realism reaches its high|| 2 Wellek’s (1963, 240–241, 242, 253) formulation is that realism is “the objective representation of contemporary social reality”.

18 | Stein Haugom Olsen est level of achievement. “[T]hat willed tendency of art to approximate reality, which critics call realism”, says Harry Levin, in the first chapter of The Gates of Horn, […] has nowhere been more clearly recognized or more expertly cultivated than in modern France. A great tradition, reaching down from the French Revolution into our own day, has been sustained with conspicuous mastery by at least one novelist in each of five interlinked generations: notably by Stendhal, Balzac, Flaubert, Zola, and Proust. (Levin 1963, 3–4)

The tradition is ‘great’ because it is ‘realist’.

2 If one builds into the concept of literary realism an epistemological claim and makes ‘truthful/objective representation of reality’ a criterion for a work to be realist, then certain consequences follow. In the 1970s, realist literature was attacked by post-modernist critics (to use a convenient label) because, it was argued, the ontological and epistemological assumptions on which literary realism rested were mistaken. Realist literature was premised on a realist epistemology which assumed that the reality which realist literary works describe exists independently of being described in language, that this reality is ultimately knowable and can therefore be truly and exhaustively described. Postmodernist critics thought they had arguments for rejecting these assumptions. But if these assumptions could be rejected, then the claim made for realist literature that it provided an objective representation of an independently existing and knowable reality was not merely mistaken but meaningless. Once one had gained this insight, one would no longer find realist literary works worth reading. Indeed, only ‘naive’ readers could then endure reading realistic novels: “The best-seller list still figures numerous realistic novels, but sophisticated readers find no substance in them, and few critics claim any lasting greatness for them” (Hume 1984, 39). Moreover, if the purpose of realist fiction could not be to give a true-to-life representation of social reality, then it had to have another purpose. This was found in its ideological function: By imposing an apparent order on experience where there is none, realist literary works ‘naturalized’ and made the reader accept as meaningful the social order, and in this way prevented the reader from disrupting this social order by questioning it. Realism was seen as “complicit” in “preserv[ing] various consciousnesses from doubt”

The Concept of Literary Realism | 19

Lyotard 1984 [1979], 84). The function of realist works of art, whether novels, photographs, or movies was held to be […] to stabilize the referent, to arrange it according to a point of view which endows it with a recognizable meaning, to reproduce the syntax and vocabulary which enable the addressee to decipher images and sequences quickly, and so to arrive easily at the consciousness of his own identity as well as the approval which he thereby receives from others – since such structures of images and sequences constitute a communication code among all of them. This is the way the effects of reality, or if one prefers, the fantasies of realism, multiply. (Lyotard 1984 [1979], 84)

Realism, in the vocabulary of the post-modernist critic, retained its evaluative dimension but the evaluation it carried was negative. This hostile view of literary realism left post-modernist critics with a somewhat delicate problem: What to do with the established tradition of realist literature, a tradition which included all the great European nineteenth-century novels, French, English, and Russian. If they rejected these novels as fit only for ‘naïve’ readers, they would put themselves in a position not unlike that of Tolstoy (1930 [1896]) in What Is Art when he rejected the whole tradition of Western literature because of its morally depraved character. Tolstoy did not, of course, gain acceptance for his evaluations, and all he achieved was to discredit the criteria he applied. So, post-modernist critics tried to solve this problem by showing that the great nineteenth-century novels were not after all realist. Fairly typical is the way in which J. Hillis Miller (1974, 455–73) deals with Middlemarch in his article Narrative and Fiction where he argues that far from being a coherent representation of reality, a history as it purports to be, Middlemarch is in fact a sustained effort to ‘demythologize’ the whole notion of history: The effort of demythologizing in Middlemarch, then, can be defined as a dismantling of various versions of the metaphysical system on which the traditional idea of history depends. In spite of its recourse to the conventional locus standi of defining itself as a displaced form of history, the novel, so to speak, pulls the rug out from under itself and deprives itself of that solid ground without which, if Henry James is right, it is ‘nowhere’. Her fiction deprives itself of its ground in history by demonstrating that ground to be a fiction too, a figure, a myth, a lie, like Dorothea’s interpretation of Casaubon or Bulstrode’s reading of his religious destiny. (Miller 1974, 467)

And Middlemarch does not impose a false unity and order on experience and it has no final meaning: Against the notion of a work of art which is an organic unity and against the notion that a human life gradually reveals its destined meaning, George Eliot opposes the concepts of a text made of differences and of human lives which have no unitary meaning, for whom

20 | Stein Haugom Olsen ‘every limit is a beginning as well as an ending’ (III, ‘Finale’, 455). (...) And in place of the concept of elaborate organic form, centered form, form organized around certain absolute generalizable themes, George Eliot presents a view of artistic form as inorganic, acentered, and discontinuous. (Miller 1974, 468)

However, this move, reading realist novels in a non-realist way, simply pushes the problem on to another level: For the post-modernist critic then has to explain why readers in the last two hundred years have read these novels as realist works of literature. The explanation is simple enough. Readers in the last two hundred years have been blinded by their ideological blinkers. It takes a special kind of reader to recognize that what looks like a realist representation of reality, is really something quite different. It is only “for those who have eyes to see it” that Middlemarch is an example of a work of fiction which not only exposes the metaphysical system of history but also proposes an alternative consonant with those of Nietzsche and Benjamin in the citations with which I began this paper. Like Benjamin, George Eliot rejects historicism with its ideas of progress and of a homogeneous time within which that progress unfolds. Like Benjamin she proposes a view of the writing of history as an act of repetition in which the present takes possession of the past and liberates it for a present purpose, thereby exploding the continuum of history. (Miller 1974, 468)

Well, it is not quite enough to have eyes to see here. One also has to have read Nietzsche and Benjamin. It is a notable feature of the post-modernist approach to realism that it is monocriterial: Any kind of literature that ‘represents reality’ is realist. No other criteria are employed. This cuts the concept of literary realism off from its historical context and it does not give any role to the formal features of the nineteenth century realist novels in a characterization of realism. It opens up for a multiple rhetorical use of the words realist and realism through which it can be insinuated that a work of art is a kind of deception practiced on naïve readers: Realism is a culturally relative concept, of course, and many avant-garde movements have successfully introduced formal changes in the name of increased verisimilitude. But the term is useful in distinguishing between those forms which tend to efface their own textuality, their existence and discourse, and those which explicitly draw their attention to it. Realism offers itself as transparent. The rejection of the concept of a literary form which reflects the world, however, has led some post-Saussurean critical theorists to use the phrase ‘classic realism’ to designate literature which creates an effect or illusion of reality. This is not just another gratuitous piece of jargon. ‘Classic realism’ makes it possible to unite categories which have been divided by the empiricist assumption that the text reflects the world. (Belsey [1980, 51])

The Concept of Literary Realism | 21

But if one needs to invent a new term to refer to such works as have been called simply realist, then the question does arise whether the concept of realism itself is not losing its usefulness as a critical instrument. Indeed, this kind of criticism in effect denies the existence of realist literature. If no literary work can be realist in the sense that it ‘truthfully/objectively represents reality’, then the concept of realism becomes a concept with a null extension and only an ideological function.

3 There are various ways to respond to ‘The rejection of the concept of a literary form which reflects the world’. An immediate and direct response was simply to reject the rejection and insist that literary realism is a form of truth-telling. Realist works of literature are interesting and valuable because they do tell the truth about human existence: War and Peace is interesting not only as an account of a particular campaign and of the manners and morals of a certain group of people in a certain place and time. It is great because it has much to say about war and peace at other places and other times. The reality described in a great fiction stands metonymically for a larger reality, or for a whole, infinite class of realities. (Tallis 1988, 208)

The critics who chose this response had two main lines of argument. One line of argument was empirical and the empirical attack had two prongs. It was pointed out that the post-modernist argument against realism was normative in character. The post-modernist argument was that because the assumptions on which realism rested were false, it should no longer be possible to write realist novels or to tolerate reading such novels. According to this view, as summarized somewhat parodically by Tallis, realism […] has been evicted from the Literature shelves and squats ingloriously (but lucratively) in paper-back bookstands on railway stations and at airports. It has, with a few striking exceptions, lost its literary function; and if it has any political function, it is to support the status quo, to collude with one version of reality and pass it off as if it were reality itself. The serious reader must look elsewhere, outside of realism, for instruments to sharpen his perception of the world and to heighten his awareness of the significance of things; even, paradoxically, for fictions with which to explore reality itself. (Tallis 1988, 1)

However, declaring that it should no longer be possible for serious authors to write realist novels or for serious readers to tolerate reading them, did not make it so. For the truth is that

22 | Stein Haugom Olsen […] realistic fiction is very much alive and kicking, though realism seems to have conceded much of the experimental high ground. Realistic novels of a very high standard, as well as the rubbish, continue to be written and read. And realism continues to be appreciated – if the awarding of major literary prizes can be counted as a criterion. The great majority of novels short-listed for the Booker Prize are realistic. (Tallis 1988, 1–2) So the facts are not entirely in favour of those who, like Michael Boyd, claim that the modern novel defines itself in terms of its rejection of the conventions of formal realism. Nevertheless, there are many critics who wish that the modern novel would behave in this way, believing that those who continue to write in the realist or naturalist traditions are ‘like headless chickens unaware of the decapitating axe’.3

The second prong of the empirical attack was to focus on the actual situation with regard to readers and critics. The post-modernist argument disenfranchised most contemporary readers of literature and all readers of literature that did not live in the contemporary world. Post-modernist critics were not the first to engage in this high-risk sport. Virginia Woolf was particularly good at it with her contempt for the ‘middlebrows’, among whom she also placed those who practiced criticism from within the academy and who wrote books about Shakespeare, not out of love of his works but in order to make a living (see Woolf 1966 [1942], 196–203). Disenfranchising readers is a high-risk sport because one leaves oneself open to the obvious challenge to produce the empirical evidence on which this disenfranchi-sement is based. For Virginia Woolf and her likeminded artist colleagues “The awkward question was, how could the superiority of the intellectual or artist be demonstrated, and in what, precisely, did it consist?” (Carey 1992, 72). Virginia Woolf produced no evidence why those who wrote books about Shakespeare in order to make a living could not write good books about Shakespeare, except citing what she thought was their bad taste in furniture, music, painting etc. Her attack on the middlebrows were a kind of visceral social cum intellectual snobbery that is well captured by E. M. Forster in his rendering in Howards End of the relationship between the Schlegels and Leonard Bast. Post-modernist critics, it was pointed out, faced exactly the same question, but with a twist. For among the readers disenfranchised by the post-modernist critical argument, one would find the majority of distinguished twentieth- and twenty-first-century critics who have written on the realist tradition and produced interpretations of the works in that tradition. To call critics like Harry Levin, George Levine, and George Becker (see Levin 1963; Becker 1963 and Levine 1981, 2008) ‘naïve’ and ‘unsophisticated’, one would have to stretch the || 3 Tallis (1988, 1–2). The references are to Boyd (1983, 9) and Scholes (1967, 6).

The Concept of Literary Realism | 23

meaning of these terms beyond the limit of their usefulness. And one would be hard put to explain what it is that makes these critics more naïve and more unsophisticated than e. g. J. Hillis Miller, Paul de Man, Catherine Belsey, or Terry Eagleton. Indeed, the defenders of realism argued, it is difficult to believe that even ordinary (in the sense of non-professional) readers are naïve and unsophisticated to the extent that post-modernist critics seem to assume. “Surely”, says Pam Morris (2003, 42) commenting on Hillis Miller’s (1971) article “The Fiction of Realism: Sketches by Boz, Oliver Twist, and Cruikshank’s Illustrations”, “it is immensely condescending to assume that most of Boz’s nineteenthcentury readers would have naïvely confused his appeal to imaginative conjecture with an hour by hour factual account of some actual man’s last night alive”. The condescension to and contempt for ‘the ordinary reader’ was a pervasive feature of post-modernist critical theory. “Lest in speaking of ‘contempt’ for the reader I be accused of exaggeration”, says Raymond Tallis (1988, 55), “consider this passage from Critical Practice”: The success with which the Sherlock Holmes stories achieve an illusion of reality is repeatedly demonstrated. […] According to The Times in December 1967, letters to Sherlock Holmes were then still commonly addressed to 221B Baker Street, many of them asking for the detective’s help. (Belsey 1980, 113)

And he comments: The appreciative reader of realistic novels, it seems, is in grave danger of confusing fiction and life at a very elementary level indeed. He or she is like the child who writes to Father Christmas or the fully paid up member of the Munch Bunch Club who expects to be visited by Rosy Raspberry or Pete Pepper. No wonder critics have to be alert, watchful and wakeful among the sleeping, if readers not only imbibe the implicit ideology of the novels they read but imagine that what they read is factually true as well as fictionally plausible! (Tallis 1988, 55)

The second argument deployed by the defenders of realism against the postmodernist position was theoretical. Again this argument had two prongs. There was the focus on the non-sequiturs, contradictions, and logical fallacies in the post-modernist espousal of ontological and epistemological scepticism. This aspect of the argument was aimed at establishing that there were no general epistemological reasons for rejecting literary realism as a mode of writing whose purpose was a true and accurate representation of an independently existing reality, be it physical or social. This argument was essentially negative. It did not offer any positive grounds for accepting that successful literary realism did in fact provide ‘an accurate representation of contemporary social reality’.

24 | Stein Haugom Olsen The other aspect of the theoretical argument engaged directly with the analyses offered by post-modernist literary critics of nineteenth-century realist novels. The response by defenders of realism to the postmodernist interpretation of realist novels as texts that (for “those who had eyes to see” [Miller 1974, 468]) focused attention on their own fictionality, was to point out that these interpretations were not in the end based on the texts analysed but in a theory that dictated the outcome of the interpretation even before it had begun. Commenting on Hillis Miller’s (1971) discussion of Sketches by Boz, Gerald Graff has this to say: It may be true, of course, that Dickens’s devices of characterization and imagery do indeed signal to the reader the literary and fictive quality of the text. But nothing in this common convention whereby fictional works concede their lack of factual truth justifies the conclusion that such works call attention to their inability to refer to external reality. It is only by confusing the pseudo-factual or pseudo-informational level of fictional reference with the larger thematic level that one can infer that literary works necessarily call their referential claims into question. Like de Man, Miller has resorted to the strategy, discussed earlier in this chapter, of using the fact that literary action does not present itself as literally true as an excuse to sever literature from referential claims, or to see it as invariably commenting on the futility of such claims. For all its heavy documentation, then, Miller’s reading of the Sketches rests not on textual evidence from the work itself but on a theory that tells him in advance what this evidence must be evidence for. (Graff 1979, 176)

Again this argument was essentially negative. It established that interpretations of nineteenth-century realist novels as texts calling attention to their own conventionality and through that to their inability to refer to anything beyond themselves have serious methodological weaknesses. However, there was no further argument to establish that such novels do actually represent correctly contemporary social reality and that this is what makes these novels worth reading when the representation is successful. Defenders of realism seemed to take for granted that once they had removed the grounds for accepting the ontological and epistemological scepticism on which the post-modernist critical position was based, then it became obvious that successful realist works of literature were read because of the insights they offered into ordinary social reality: If one insists on seeing all novels as congeries of semiotic systems intricately functioning in a pure state of self-referentiality, one loses the fine edge of responsiveness to the urgent human predicaments that novels seek to present to us – through the most inventive variety of artifice, whether disguised or manifest – lives that might seem like our lives, minds like our minds, desires like our own desires. That has been what most novelists quite

The Concept of Literary Realism | 25 clearly have tried to accomplish in their writing, and that is what still makes the reading of novels for most people, intellectuals included, one of the perennially absorbing activities of modern culture. (Alter 1984, 21)

In other words, it is taken for granted that if one recognizes the futility of “seeing all novels as congeries of semiotic systems intricately functioning in a pure state of self-referentiality”, then one will automatically accept that the novel presents human predicaments – “lives that might seem like our lives, minds like our minds, desires like our own desires” (Alter 1984, 21).

4 One reason why the critics who took up the ‘defence of realism’ may not have felt the need for a positive argument to support the view that realist works, in so far as they are realist, offer ‘an accurate representation of contemporary social reality’ was that they had behind them a solid critical tradition. When critics like Lukács, Becker, Levin, and Levine use the concept of realism they insist explicitly that realist works of art are truthful/objective representations of reality. And, like these critics, the defenders of realism were able to rely on a widely shared experience between readers of realist literary works “over the last two hundred years” that it is this type of work that, in Watt’s (1957, 32–33) words, “most closely satisfies their wishes for a close correspondence between life and art”. However, even if one rejects the epistemological and ontological scepticism of post-modernist critics, this concept of realism proves problematic once one puts pressure on it by making evaluative claims about realist literature. This is glaringly obvious in the works of Georg Lukács who uses the criterion truthful/objective representations of reality to condemn whole schools of writing. Not only does he attack naturalist writers like Zola, but he dismisses literary modernism as essentially decadent and vicious. Modernism “leads not only to the destruction of traditional literary forms; it leads to the destruction of literature as such” (Lukács 1963 [1957], 45). “Modernism means not the enrichment, but the negation of art” (Lukács 1963 [1957], 46). Employing a concept of realism which has an inbuilt reference to truthful/objective representation, leads to the same kind of problems for the critics who ‘defend’ realism as it did for poststructuralist critics, since they are forced to evaluate realist works of art more highly than works belonging to other schools or traditions. It may be argued that this problem arises for critics like Lukács only because of their concept of reality. With a wider or a different concept of reality, the problem would not arise. This may be true, but to open the concept of reality

26 | Stein Haugom Olsen would not make the criterion ‘truthful/objective representations of reality’ more useful as a critical instrument. When Virginia Woolf attacks the British realist writers, the Great Edwardians, John Galsworthy, H. G. Wells, and, in particular, Arnold Bennett, she rejects realism because their work does not represent reality: I asked them – they are my elders and betters – How shall I begin to describe this woman’s character? And they said: ‘Begin by saying that her father kept a shop in Harrogate. Ascertain the rent. Ascertain the wages of shop assistants in the year. Discover what her mother died of. Describe cancer. Describe calico. Describe –.’ But I cried Stop! Stop! And I regret to say that I threw that ugly, clumsy, that incongruous tool out of the window, for I knew that if I began describing the cancer and the calico, my Mrs. Brown, that vision to which I cling though I know no way of imparting it to you, would have been dulled and tarnished and vanished forever. That is what I mean by saying that the Edwardian tools are the wrong ones for us to use. They have laid enormous stress upon the fabric of things. They have given us a house in the hope that we may be able to deduce the human beings who live there. To give them their due, they have made that house much better worth living in. But if you hold that novels are in the first place about people, and only in the second about the houses they live in, that is the wrong way to set about it. (Woolf 1966, 332)

Virginia Woolf rejects the conventions of nineteenth-century realism, the description of the fabric of things, the focus on the outward fact and the emphasis on circumstantiality of detail, and in her practice she goes on to develop new techniques, those of the interior monologue and stream of consciousness, the techniques of modernism. Her argument for developing new techniques and rejecting the techniques and conventions of the realists is that they are inadequate for rendering human character. Virginia Woolf, the modernist, makes use of the same criterion as Balzac, the realist, but she uses it to criticize the way in which the realists write about reality. The criterion is an epistemological one: The techniques of realist writers are inadequate if one want give a true picture of human character. “Whether we call it life or spirit, truth or reality”, she (Woolf 1966, 105) says in Modern Fiction, “this, the essential thing, has moved off, or on, and refuses to be contained any longer in such ill-fitting vestments as we provide.” Her argument is that if one confronts realist novels with ‘reality’, then they are untrue. One can falsify realist descriptions by looking and seeing if they correspond to reality: Look within and life, it seems, is very far from being ‘like this’. Examine for a moment an ordinary mind on an ordinary day. The mind receives a myriad impressions – trivial, fantastic, evanescent, or engraved with the sharpness of steel. (Woolf 1966, 106)

The Concept of Literary Realism | 27

So, if one opens up the concept of reality as it occurs in the criterion ‘providing a truthful/objective representation of some aspect of reality’, this criterion cannot be used to distinguish between the modes of writing that literary critics have been identifying as realism and modernism. The concept of realism has then lost its usefulness as a critical concept. The attack on literary realism by post-modernist critics was part and parcel of a wider attack on humanism. An unanalysed concept of realism with its apparent conceptual link to reference and truth provided an easy target for critics who wanted to undermine humanist values. And the response to postmodernist critics was not really a defence of realism, but, in Gerald Graff’s (1979, 193) words, “a realist defence of humanism”. And in this struggle the concept of realism was itself a casualty. When the concept of realism is put under pressure, as it was by Lukács and post-modernist critics, it breaks, and it breaks at a very important point. A concept of realism which licences the conclusion that modernism leads to the destruction of literature, that only naïve readers can find any satisfaction in realist literary works, or that realist novels are not really realist, or that, contrary to all empirical evidence, realism is dead, is simply useless as a critical tool. The effect of the debate between post-modernist critics and the defenders of realism was to derail any fruitful discussion of realism for thirty years. “The number of critical books on realism from the past couple of decades”, says Rachel Bowlby (2010, xiv) in the Foreword to a recent collection of articles that attempt to get the discussion of realism back into critical focus, “can be counted on the fingers of one hand.” However, the collection that Bowlby is introducing does not represent a new departure. In Bowlby’s introduction, one recognizes many of the issues that were raised by the defenders of realism discussed in this section: [W]hen realism does get mentioned it is usually in the form of a passing, knee-jerk dismissal of it as something self-evidently without interest, not to say a bit dumb. Realism normally comes stuck with one of a set menu of regular adjectival accompaniments, and whether it’s gritty, or vulgar, or kitchen-sink, or photographic, the standard formulations reinforce the way it is seen as itself formulaic, something we already know about and need have no interest in exploring: it is predictable and simple, and serves only as the foil (or the cling-film) for showing up the more exotic or more complex courses that are always to be preferred to it. Nowhere is this clearer than in the regular scorn for realism’s crudely ‘linear’ narratives, its naively ‘omniscient’ narrators, and – worst crime of all – its facile assumptions of linguistic ‘transparency’, all of these being qualities that are quite untransparent and unanalyzed in their own meaning but essentially damning in their aim. (Bowlby 2010, xiv–xv)

28 | Stein Haugom Olsen Large parts of Bowlby’s introduction are dominated by a continued polemics against the post-modernist rejection of realism, and none of the articles in the collection manages to cut themselves off from the conceptual framework that dominated the polemics between post-structuralist critics and the defenders of realism.

5 The concept of literary realism, then, has a central place in literary criticism as well as in literary history. However, as we have seen, employing a concept of literary realism that has ‘objective/truthful representation of reality’ as a central element leads critics to adopt counterintuitive views about the nature of realist literary works, to confusion as to whether there actually are any such works, and licenses unacceptable and downright ridiculous conclusions about the relative value of the whole periods of literary works. Such a concept of literary realism simply does not function as a useful critical tool. It would seem, then, that the only way to establish the concept of literary realism as a fruitful and interesting critical tool is to reject ‘objective/truthful representation of reality’ as a criterion for a literary work being realist. If one rejects this criterion, then one is left with two possible interpretations of the concept. Without committing oneself to the ontological and epistemological scepticism that is part and parcel of the post-modernist package, it is possible to adopt what one may call a radically conventionalist view of realism. On such a view the concept of literary realism can be exhaustively characterized in terms of a set of techniques, conventions, subjects etc., and in spite of its name, it carries no reference to reality, social or physical. The reason for calling this view radical conventionalism would be that it is possible to contrast it with the merely conventionalist view that though literary realism can be defined in terms of a set of techniques, conventions, subjects etc., the mode of writing so defined has nevertheless a positive relationship to reality. The problem with radical conventionalism, which cuts the connection between the concept of literary realism and reality and sets the claim of ‘a positive relationship to reality’ at nought, is that it has to dismiss as an illusion what seems to be the widely shared experience between readers of realist literary works, “over the last two hundred years”, i. e. that it is this type of work that, in Watt’s (1957, 32–33) words, “most closely satisfies their wishes for a close correspondence between life and art”. It also has to dismiss the strongly held view of many of the best critics who have been writing on realism that realist works do

The Concept of Literary Realism | 29

indeed have a positive relationship to reality. Moreover, it makes the concept of literary realism much less interesting as a critical instrument and it would make it difficult to explain why critics find it illuminating to identify realist elements in works that are clearly not realist or intended to be realist. It would seem, then, that the only account that would yield a fruitful and interesting concept of literary realism is the conventionalist account, which nevertheless accepts ‘a positive relationship to reality’ as a criterion for a literary work being realist. The problem for such an account would be to provide an interpretation of the criterion of ‘positive relationship to reality’ without falling back on the notion of ‘truthful/objective representation’. A possible departure for such an interpretation can be found in formulations like the following: Realism in literature connotes a way of depicting, describing a situation in a faithful, accurate, life-like manner; or richly, abundantly, colourfully; or again mechanically, photographically, imitatively. (Stern 1973, 40)

The following passage from Arnold Bennett’s Anna of the Five Towns seems to illustrate such a narrative procedure: In the slip-house begins the long manipulation which transforms raw porous, pliable clay into the moulded, decorated, and glazed vessel. The large whitewashed place was occupied by ungainly machines and receptacles through which the four sorts of clay used in the common ‘body’ – ball clay, China clay, flint clay, and stone clay – were compelled to pass before they became a white putty-like mixture for shaping by human hands. The blunger crushed the clay, the sifter extracted the iron from it by means of a magnet, the press expelled the water, and the pug-mill expelled the air. From the last reluctant mouth slowly emerged a solid stream nearly a foot in diameter, like a huge white snake. Already the clay had acquired the uniformity characteristic of a manufactured product. (Bennett 1990 [1902], 117–118)

The passage is taken from a chapter called On the Bank. The chapter presents to the reader a complete description of a pottery factory with all the production processes that takes place in the factory. This passage is a description of the ‘slip-house’ and the reader has no reason to doubt that it is a correct description of how a slip-house in such a factory would look. The description is not, however, a description of a real slip-house. It is ‘life-like’ but not a description of any particular entity in reality. The criterion that Stern is building into the concept of realism in the above formulation appears in the discussion of realism as verisimilitude. Though Stern does not use the term itself, it has figured in the discussion of realism almost as prominently as the phrase ‘objective/truthful representation of reality’. “What is

30 | Stein Haugom Olsen realism as understood by the theoretician of art?”, asks Jakobson in an early article that first introduced the view that realism is merely a convention: It is an artistic trend which aims at conveying reality as closely as possible and strives for maximum verisimilitude. We call realistic those works which we feel accurately depict life by displaying verisimilitude. (Jakobson 1987 [1921], 20)

However, to use the concept of verisimilitude to explain the concept of realism is to commit the fallacy of obscurum per obscurius. The concept of ‘verisimilitude’ is as problematic as the concept of literary realism and the history of literary criticism since Jakobson’s article was published does not provide any discussion of the concept of verisimilitude that goes beyond the definition provided by the Oxford English Dictionary (1989) “the appearance of being true or real; likeness or resemblance to truth, reality, or fact; probability.”4 The closest to a discussion of the concept that can be found in literary criticism occurs in A. D. Nuttall’s A New Mimesis (Nuttall 1983, 60) where he argues against Jakobson’s relativist view of realism and points out that verisimilitude in all its uses “retains as its clear but open meaning, ‘approximating to what is really the case’”.5 The problem he fails to address is what it is to ‘approximate’ to ‘what is really the case’.

6 Though there is no analysis of the concept of verisimilitude in literary criticism or literary theory, one does find such an analysis as part of the discussion in philosophy of science about truthlikeness that started with Popper’s attempt in Conjectures and Refutations to give an account of […] the idea of a degree of better (or worse) correspondence to truth or of greater (or less) likeness or similarity to truth; or to use a term already mentioned above (in contradistinction to probability) the idea of (degrees of) verisimilitude. (Popper 1989 [1963], 233)

|| 4 verisimilitude, Oxford English Dictionary (1989), http://www.oed.com/view/Entry/-222523 (13 August 2012). As late as 2003, this is as far as Pam Morris (2003) in her introductory book, Realism takes the discussion of the concept. In Beaumont’s recent collection of articles, A Concise Companion to Realism (2010), the term verisimilitude occurs only six times and is never explained. 5 Nuttall (1983, 60). See also Harry Levin (1963, 3), who talks about “that willed tendency of art to approximate reality”.

The Concept of Literary Realism | 31

Popper introduced the notion of verisimilitude in order to be able to make sense of scientific progress. Popper accepted Hume’s criticism of induction and took it to imply that no interesting theory about the world can be verified, nor can one show that one theory is more probable than another. While no theory can be verified, a theory can, according to Popper, be falsified and it is this possibility of falsifying a theory which makes a theory scientific. However, falsification is a blunt instrument and it does not provide any basis for distinguishing between theories along a scale of falsity. In order to make sense of the idea of cognitive progress and to be able to account for what appeared to be scientific progress, a basis for such a scale had to be provided. If one could make sense of the idea that some theories, though they were false, were nevertheless truer than other theories, then this provided a basis for a meaningful conception of scientific progress. What was needed was a concept of graded truthlikeness, and this Popper called verisimilitude. A very simple example will give an idea of how the notion of truthlikeness could be made meaningful. If the slice of reality is, say, the weather conditions in a particular location which can be exhaustively described along three dimensions, humidity (humid/dry), temperature (hot/cold), wind (windy/still), then one would have eight possible descriptions of the weather conditions. Humidity

Dry

Dry

Dry

Dry

Humid

Humid

Humid

Humid

Temperature

Hot

Hot

Cold

Cold

Hot

Hot

Cold

Cold

Wind

Windy

Still

Windy

Still

Windy

Still

Windy

Still

Table 1: Weather Conditions

If the weather at any one time was dry, hot, and windy the description of it as humid, cold, and still would be further from the truth and thus less ‘truthlike’ than a description which characterized the weather as humid, hot, and still. And this description would be less truthlike than a description of the weather as dry, hot, and still. Popper’s formal definition of the notion of verisimilitude has over the years been challenged, developed, and refined.6 However, this notion of verisimilitude is aimed at providing a quantitative measure of the truthlikeness of scientific theories so as to provide a basis for deciding between competing theories about the same slice of reality. This slice of reality can be given a description || 6 For a recent, easily accessible account of the discussion of this notion of verisimilitude, see Graham Oddie (2009). For a more detailed account see Sjoerd D. Zwart (2001). See also Ilkka Niiniluoto (1998).

32 | Stein Haugom Olsen which at its limits represents the whole truth about it (‘The weather at place p at time t is hot, dry, and windy’ when the weather is hot, dry, and windy) and it is only in a relation to this ideal that the notion of truthlikeness makes sense. There is no slice of reality to which Arnold Bennett’s description of the sliphouse in Anna of the Five Towns refers and which could be given a total true description to which Bennett’s description could approach and of which there could be alternative descriptions which were more or less truthlike than Bennett’s description. There is also the fundamental problem that if one interprets verisimilitude as truthlikeness in the above sense, then using the concept of verisimilitude as a criterion for a work being realist, brings the same problems as using the criterion of ‘objective/truthful representation of reality’. Verisimilitude would then simply mean ‘approximating to an objective/truthful representation of reality’. This interpretation of verisimilitude does not take us beyond Nuttall’s (1983, 60) formulation that verisimilitude in all its uses “retains as its clear but open meaning, ‘approximating to what is really the case’”.

7 Consider again the claim made by Balzac in the preface to La comedie humaine [French Society], says Balzac, […] was to be the historian, I had only to be the secretary. I would draw up the inventory of the vices and virtues, collect the main effects of the passions, portray characters, select the principal events of Society, and compose types of combining the traits of several homogenous characters; and I might thus succeed in writing that history forgotten by so many historians, the history of manners and morals. [...] But if I was to deserve the praises which any artist must aspire to, I must needs study the causes or central cause in that immense assembly of events. (Balzac 1972, 143)

An alternative to interpreting Balzac here as making a truth-claim would be to interpret him as making a claim about probability. This would be a claim that his novels are to be read as an account of a range of types of fact (a range-offacts-account), morals and manners, which might occur, though it may not as a matter of fact have occurred, in the world outside the work. He may be construed as seeing his task as subtly different from that of the historian who writes about actual, particular political and social events and facts. Balzac, it may be argued, does not here express referential intentions, but an intention to produce an account that is life-like, authentic, that has a high degree of verisimilitude,

The Concept of Literary Realism | 33

because it presents a scene that has a high degree of probability. The positive relationship between the realist literary work and external reality could then be interpreted as being the same as that between a hypothesis and the facts about which it hypothesises. The explication of the concept of verisimilitude in terms of probability has the advantage that it permits an explanation of how the ‘positive relationship’ to the world can dictate a mode of writing. A probable account has certain features which, if they do not constitute, at least contribute to, its probability. It can be argued that probability is enhanced by such features as Ian Watt (1957) lists as characteristic of formal realism: Introduction of individual characters acting against a particular and particularised real life background, the rejection of type names and the adoption of real life names that do not place the character in a category, the creation of a causal chain working itself out in the plot, and the creation of characters that develop through time, the use of the language of concrete particularity, and the avoidance of ornate and poetic language. Descriptions presenting circumstantial detail mutually support each other and in this way enhance probability. They also present conditions under which certain types of actions and events take place in the real world, and thus they enhance the probability of the description of such actions and events in literary works. When Balzac (1972, 143) says that he “must needs study the causes or central cause in that immense assembly of events”, he can naturally be interpreted as saying that he wants to give action- and event-descriptions that are probable in the light of a set of fact-representations which he provides as a backdrop. Indeed, the strategy of defending the probability of a representation of actions and events in a literary work by pointing to the background of conditions for these actions and events which the work also represents, is a familiar one.7 And it is this feature of the realist work that provides the basis for the slogan ‘letting the facts speak for themselves’ and for the view of the realist novel as a ‘laboratory’ for ‘testing’ human reactions to find their true causes. However, the explication of the concept of verisimilitude in terms of probability still leaves the notions of truth and reference at the heart of the concept of literary realism. Probability is a relational term and the relation of probability is that between the realist literary work and social reality. To make probability part of the definition of the concept of literary realism would still leave the concept open to the same abuses as were opened up by using ‘objective/truthful representation of reality’ as a defining feature of a realist literary work. || 7 For a discussion of the concept of probability that such an argument takes for granted, see Newsom (1988, 85; 92–95).

34 | Stein Haugom Olsen Moreover, the application of the concept of probability to fictional descriptions leads to some additional problems. Probability, at least in its modern sense,8 is credibility based on evidence, and is a matter of degree. The stronger the evidence, the more probable it is that the description is a true or correct description. However, to ask how probable it is that Arnold Bennett’s description of a slip-house quoted in section V is true or correct is distinctly odd. There are two ways to deal with this oddity. One is simply to accept it. One would then have a fictional description which, being fictional, per definition is not a description of a real slip-house. The answer to the question about the degree of probability of this description would have to be that there is zero probability of the description being correct or true. Yet one would intuitively say that this description has a high degree of verisimilitude. Rather than accept the oddity and assign this description zero probability, one might, following Robert Newsom, argue that it is logically absurd to ask about evidence for a made up event-description: Probability statements about the characters or plots in imaginative literature logically belong to a different space than do statements about the real world, and there is something fundamentally strange – or absurd – about discussions of the probability of Mr. Pickwick, say. Why should there ever be a question about the ‘probability’ of people and events that are known certainly never to have existed or occurred? (Newsom 1988, 96)

The logical strangeness arises because to claim probability for literary works involves a category mistake in Ryle’s sense. Literary works as literary works are neither probable nor improbable just as they are neither true nor false. Both these lines of argument lead to the conclusion that the notion of verisimilitude when employed to characterize a literary work cannot be elucidated through the concept of probability. It is interesting to note that realism as an aesthetic doctrine declined in France at the end of the last century just as the new social sciences came into their own. It has been argued that realism declined because it became obvious that the social sciences performed so much better the descriptive and explanatory tasks which the realist authors tended to see themselves as performing.9 If one accepts Newsom’s point that it is logically absurd to ask about the evidence for a literary description, then this decline could be interpreted as resulting || 8 The theory that there is a distinctive ‘modern’ concept of probability is presented in Hacking (1975). However, theorists writing about the application of the concept of probability to literary works have argued that the ‘new’ concept is essentially an extension of the classical concept as found in Aristotle (see Lane Patey [1984] and Newsom [1988]). 9 For a summary of this argument, see Hemmings (1974, 361).

The Concept of Literary Realism | 35

from disillusionment on the part of realist writers and their audience as they discovered their category mistake.

8 The conception of verisimilitude as truthlikeness and verisimilitude as probability are conceptions that tie verisimilitude directly to an objectively given reality. These conceptions represent construals of verisimilitude as related in a positive way to truth, as an “[approximation] to what is really the case” (Nuttall 1983, 60). They reflect one main element in the non-technical concept of verisimilitude.10 However, the non-technical concept of verisimilitude can also be construed as having no such positive relationship to truth. Under this construal, a description or narrative has verisimilitude when it has “the mere appearance or show of being true or in accordance with fact; an apparent truth” The Oxford English Dictionary 1989). Verisimilitude as an ‘appearance of what is really the case’ is radically different from verisimilitude as an ‘approximation to what is really the case’ in that the former introduces a subjective element that is absent from the latter: An appearance is always an appearance to someone and whether a description or narrative has the appearance to someone of what is really the case, will depend upon his conception of reality. It is this concept that is utilized by radical conventionalists when they argue that literary realism is a matter of convention only. For if one defines literary realism using this concept of ‘verisimilitude’, then literary realism is made relative to the individual conception of reality. And this conception, it is argued by radical conventionalists, is culturally determined. Authors and critics have come to think that realist works have a positive relationship to reality because they have mistakenly (naïvely) failed to recognize that their conception of reality is culturally constructed. Radical conventionalists conclude from this that realism is a mode of writing defined by a set of conventions without any positive relationship to an independently identifiable ‘reality’. It is possible, however, to conclude differently, i. e. that the concept of verisimilitude by itself is inadequate to define literary realism and that it has to be supplemented with a specification of the concep|| 10 I take the non-technical concept of verisimilitude to be that which one finds in dictionaries. This is the way in which The Oxford English Dictionary (1989) defines ‘verisimilitude, n.’: 1. The fact or quality of being verisimilar; the appearance of being true or real; likeness or resemblance to truth, reality, or fact; probability. 2. A statement, etc., which has the mere appearance or show of being true or in accordance with fact; an apparent truth.

36 | Stein Haugom Olsen tion of reality that makes a description possessing ‘an appearance of what is really the case’ a realist description. That is, only descriptions embodying a certain conception of reality are realist descriptions. Or, to turn this around, the form of a realist mode of description is dictated by a certain conception of reality. The problem of defining literary realism then becomes the problem of how to specify the relevant conception of reality, how to justify the claim that this is the relevant conception. The constraints on this conception of reality are twofold. The conception of reality embodied in a realist work must be such as to secure the positive relationship to reality that is the justification for calling this mode of writing ‘realist’. That is, the concept of reality embodied in a realist work of literature may be culturally constructed, but central to this construction must be procedures for comparing and adjudicating between different conceptions of reality with regard to how closely they ‘approximate to what is really the case’. Some conceptions of reality will ‘approximate’ more closely ‘to what is really the case’ than others and it is this relationship of ‘approximation’ that secures the positive relationship that a realist work of literature has to reality. There are a number of advantages that result from the move of making verisimilitude a relationship between a literary work and a conception of reality rather than between the literary work and reality itself. A conception of reality is always aspectual. There is no conception of reality as such. If ‘appearance of being true or real’ is used as a criterion of something being a realist work of literature, then this appearance is only of a certain aspect of reality and to define realism is to specify one particular aspect or a range of aspects embodied in the concept of reality that is relevant for the definition of literary realism. The fact that a concept of reality is aspectual enables us to deal with the sort of criticism that Virginia Woolf directs against Bennett, Galsworthy, and Wells in the name of a ‘different’ reality. The concept of reality embodied in realist literary works simply does not represent that aspect of reality. Realist literary works present ‘social reality’ and not ‘mental reality’. Secondly, the burden of the epistemological claim is moved from the concept of literary realism and placed on the conception of reality which the realist work embodies. The advantage of this is that the attack on literary realism by post-structuralist critics can now be seen for what it was: An attack on the possibility of distinguishing between true and false conceptions of reality and not an attack on literary realism. And it explains Gerald Graff’s (1979, 193) diagnosis that the defence of realism was really “a realist defence of humanism.” Equally important is the possibility this move opens up for interpreting ‘approximating to what is really the case’ not in terms of truth or falsity, but in terms of a set of standards defining what it is to approximate to what is really the case. Radical

The Concept of Literary Realism | 37

ontological and epistemological scepticism of the kind represented by poststructuralist critics does not affect such standards. The second constraint on the concept of literary realism has its basis in its two related uses. It is used as a period concept as well as a concept designating a ‘perennial’ mode of writing. “[T]he period concept of realism”, says Wellek (1963, 225), has to be construed “as a regulative concept, a system of norms dominating a specific time, whose rise and eventual decline it would be possible to trace and which we can set clearly apart from the norms of the periods that precede and follow it.” The concept of reality embodied in realist description must enable the critic to develop a characterization of literary realism as a mode of writing which can be distinguished from other modes of writing (romantic, fantastic, classical etc.), a mode of writing that is typical of the period of literary realism and which sets the period of realism clearly apart from the periods that precede and follow it. As was noted above, the criterion of verisimilitude or ‘the appearance of being true or real’ is not sufficient to distinguish literary realism from literary modernism, if the aspect of the conception of reality on which the similarity in appearance rests, is not specified. The system of norms ‘dominating a specific time’ will specify what this aspect is and thus also specify the content of the concept of literary realism.

9 The aim of the present argument has not been to produce yet another definition of literary realism, but to provide some constraints which must be imposed on the concept of literary realism if it is to be a useful critical instrument in literary studies. The main problem in discussions of literary realism has been the assumption that realist literary works must be understood as making an epistemological claim and that making good on this claim is what makes realist literary works important. The present argument has removed the burden of the epistemological claim from the concept of literary realism and placed it on the conception of reality which the realist work embodies. This gives the critic a concept of literary realism that is defined in relation to a period, but that enables him to specify a set of formal properties for the realist mode of writing that can then be identified in works outside that period. In this way, this concept of literary realism secures a close connection between the period concept and the aspect concept of literary realism. In doing this, the present argument aligns itself with the view that the period of realism is the period when the realist mode of

38 | Stein Haugom Olsen writing became dominant, but that elements of this mode can also be identified in literary works of all other periods. And the view is rejected that there is another ‘problem of realism’ which is “the whole fundamental epistemological problem of the relation of art to reality” (Wellek 1963, 224). For this latter problem cannot be approached by help of the concept of realism.

References Alter, Robert. “Mimesis and the Motive for Fiction.” Motives for Fiction. Robert Alter. Cambridge, Mass.: Harvard University Press, 1984, 3–21. Balzac, Honoré de. “Preface to The Human Comedy.” The Nineteenth Century Novel. Critical Essays and Documents. Ed. Arnold Kettle. Translation, Petra Morrison. London: Heinemann, 1972, 140–153. Becker, George J. Realism in Modern Literature. New York: Frederick Ungar, 1980. Beaumont, Matthew (Ed.). A Concise Companion to Realism. Oxford: Blackwell, 2010. Belsey, Catherine. Critical Practice. London: Methuen, 1980. Bennett, Arnold. Anna of the Five Towns. Harmondsworth: Penguin, 1990 [1902]. Borgerhoff, Elbert, “Réalisme and Kindred Words: Their Use as Terms of Literary Criticism in the First Half of the Nineteenth Century.” PMLA 53.3 (1938): 837–843. Bowlby, Rachel. “Foreword.” A Concise Companion to Realism. Ed. Matthew Beaumont. Oxford: Blackwell, 2010, xiv–xv. Boyd, Michael. The Reflexive Novel: Fiction as Critique. Toronto: Lewisburg Bucknell University Press, 1983. Carey, John. The Intellectual and the Masses: Pride and Prejudice among the Literary Intelligentsia, 1880–1939. London: Faber and Faber, 1992. David-Sauvageot, A. “Romantisme.” Histoire de la langue et de la littérature française des origines à 1900, Vol 8. Ed. Louis Petit de Julleville. Paris: Armand Colin, 1899. Graff, Gerald. Literature against Itself: Literary Ideas in Modern Society. Chicago: University of Chicago Press, 1979. Hacking, Ian. The Emergence of Probability. A Philosophical Study of Early Ideas about Probability, Induction, and Statistical Inference. Cambridge: Cambridge University Press, 1975. Hume, Kathryn. Fantasy and Mimesis. Responses to Reality in Western Literature. London: Methuen, 1984. Hemmings, Frederick Willliam John (Ed.). The Age of Realism. Harmondsworth: Penguin, 1974. Jakobson, Roman. “On Realism in Art.” Language in Literature. Ed. Krystyna Pomorska and Stephen Rudy. Translation, Karol Magassey. Cambridge, Mass.: Belknap Press, 1987 [1921], 19–27. Levin, Harry. The Gates of Horn: A Study of Five French Realists. New York: Oxford University Press, 1963. Levine, George. The Realistic Imagination: English Fiction from Frankenstein to Lady Chatterley. Chicago: University of Chicago Press, 1981. Levine, George. Realism, Ethics and Secularism. Essays on Victorian Literature and Science. Cambridge: Cambridge University Press, 2008.

The Concept of Literary Realism | 39 Lukács, Georg. The Meaning of Contemporary Realism. Translation, John and Necke Mander. London: The Merlin Press, 1963 [1957]. Lyotard, Jean-François. The Postmodern Condition: A Report on Knowledge. Translation, Geoff Bennington and Brian Massumi. Manchester: Manchester University Press, 1984 [1979]. Miller, J. Hillis. “The Fiction of Realism: Sketches by Boz, Oliver Twist, and Cruikshank’s Illustrations.” Dickens Centennial Essays. Eds. Ada Nisbet and Blake Nevious. Berkeley: University of California Press, 1971, 85–153. Miller, J. Hillis. “Narrative and History.” ELH 41.3 (1974): 455–473. Morris, Pam. Realism. London: Routledge, 2003. Newsom, Robert. A Likely Story. Probability and Play in Fiction. New Brunswick: Rutgers University Press, 1988. Niiniluoto, Ilkka. “Verisimilitude: The Third Period.” British Journal for the Philosophy of Science 49.1 (1998): 1–29. Nuttall, Anthony David. A New Mimesis. Shakespeare and the Representation of Reality. London: Methuen, 1983. Oddie, Graham. “Truthlikeness.” The Stanford Encyclopedia of Philosophy (Winter 2012 Edition). Ed. Edward N. Zalta. Stanford: The Metaphysics Research Lab, Center for the Study of Language and Information, Stanford University, 2009. http://plato.stanford.edu/archives/win2012/entries/davidson/. (12 August 2013). Oxford English Dictionary, 2nd ed. http://www.oed.com/viewdictionaryentry/Entry/222523. Oxford: Oxford University Press, 1989 (9 January 2011). Patey, Douglas Lane. Probability and Literary Form: Philosophic Theory and Literary Practice in the Augustan Age. Cambridge: Cambridge University Press, 1984. Popper, Karl R. Conjectures and Refutations: The Growth of Scientific Knowledge, 5th ed. London: Routledge, 1989 [1963]. Scholes, Robert. The Fabulators. New York: Oxford University Press, 1967. Stern, Joseph Peter. On Realism. London: Routledge & Kegan Paul, 1973. Tallis, Raymond. In Defence of Realism. London: Eward Arnold, 1988. Tolstoy, Leo. What Is Art? and Essays on Art. Translation, Aylmer Maude. Oxford: Oxford University Press, 1930 [1896]. Watt, Ian. The Rise of the Novel: Studies in Defoe, Richardson and Fielding. London: Chatto and Windus, 1957. Wellek, René. “The Concept of Realism in Literary Scholarship.” Concepts of Criticism. Ed. Stephen G. Nichols Jr. New Haven: Yale University Press, 1963, 222–255. Woolf, Virginia. “Middlebrow.” Collected Essays. Vol. II. London: The Hogarth Press, 1966 [1942], 196–203. Woolf, Virginia. “Mr. Bennett and Mrs. Brown.” Collected Essays. Vol. I. London: The Hogarth Press, 1966 [1924]: 319–337. Woolf, Virginia. “Modern Fiction.” Collected Essays, Vol. II. London: The Hogarth Press, 1966 [1919], 103–110. Zwart, Sjoerd D. Refined Verisimilitude. Dordrecht: Kluwer Academic Publishers, 2001.

Peter Lamarque

Thought, Make-Believe and the Opacity of Narrative 1 I am going to explore the idea of the opacity of narrative by weaving together threads from earlier work of mine, notably so-called Thought Theory and the perennial issue of literature and truth. I shall not say much about the problem of emotional responses to fiction but will draw on other aspects of Thought Theory. As a response to the paradox of fiction, Thought Theory is simply the idea that thought-clusters formed in engaging with works of fiction can cause emotional responses to those works regardless of knowledge that the works are fictional. If we are terrified watching the movie of the green slime, then according to Thought Theory what frightens us is the thought of the green slime as brought to mind by images in the movie; we are frightened by the thought, but we are not, in normal cases, frightened of the thought (see Lamarque 1981).1 In this present context, I am more concerned with how the thought-clusters themselves are formed. On the question of literature and truth, I have defended the view, with Stein Haugom Olsen, that the appraisal of a work’s truth, in any literal sense of the term, is not an indispensable focus for what gives interest to literary works appreciated as literature (see Lamarque and Olsen 1994). Again rather than defending that view, I want to look at aspects of it that relate both to Thought Theory and to the opacity of narrative. Let me briefly situate these current ideas with the work of Kendall Walton on fiction. Although Walton and I disagree on the paradox of fiction, I am broadly sympathetic to his general approach and its two core ideas: that works of fiction prescribe imaginings and that readers of fiction project themselves into fictional worlds by playing games of make-believe with themselves and the sentences of fiction as props. But I have come to think that while this framework might provide a generic account of fiction per se, it needs careful refinement if it is to illuminate literary fiction. More, I think, needs to be said about the content of what is imagined and the processes by which the imagining takes place when works of literature are concerned. What interests me is how the constraints of || 1 Noël Carroll labeled this account “Thought Theory” in Carroll (1990, 79–80).

42 | Peter Lamarque genre, literary form, narrative voice, point of view, irony, allusion, and other such modes, which are commonplace in the literary sphere, operate on imagining when we read literature as literature and not merely as fiction. I hope to show through a series of examples that the model offered by Walton or indeed Gregory Currie, with its focus on make-believe and imagining, is not adequate (at least as the whole story) when dealing with narratives that demand these more nuanced levels of response. Let me first introduce the idea of the opacity of narratives.

2 In our book Truth, Fiction, and Literature, Olsen and I (1994) remarked on a certain kind of referential opacity in reports about fictional narrative content. One example we used focused on the name ‘Canterbury’ in Geoffrey Chaucer’s Canterbury Tales. Here is a relevant passage from the Tales: And specially, from every shires ende Of Engelond, to Caunturbury they wende, The hooly blisful martir for to seke, That hem hath holpen whan that they were seeke. Bifil that in that seson, on a day, In Southwerk at the Tabard as I lay Redy to wenden on my pilgrymage To Caunterbury with ful devout corage, At nyght were come into that hostelrye Wel nyne and twenty in a compaignye Of sondry folk, by aventure yfalle In felaweshipe, and pilgrimes were they alle, That toward Caunterbury wolden ryde. (Geoffrey Chaucer, The Canterbury Tales, Prologue, lines 15–27 [Chaucer 1957, 17])

The modern city of Canterbury contains both a cathedral and a university, the former medieval in origin, the latter founded in the 1960s. The cathedral that exists today in Canterbury is the very one in which Thomas Becket was murdered (in 1170) and that became the site for later pilgrimages. While it would be entirely accurate to report that Chaucer’s pilgrims were travelling to a city with a cathedral, it would seem at least misleading to report that the pilgrims went to a city with a university (or even one that would later have a university). Having a university is a modern aspect of Canterbury that has no place in Chaucer’s narrative. No doubt one might be able to formulate an appropriate de re report

Thought, Make-Believe and the Opacity of Narrative | 43

about the contents of the Canterbury Tales to the effect that it is true of Canterbury itself, under any description, that it was the destination of the pilgrims. But it does not seem right, without suitable qualification, simply to substitute the definite description ‘the city in Kent with a university’ for the name ‘Canterbury’ in the report ‘Chaucer’s pilgrims went to Canterbury’. The sentence ‘Chaucer’s pilgrims went to the city in Kent with a university’ seems, again without qualification, to be false, or certainly misleading. But what interested me and Olsen was less the preservation of truth in the substitution (salva veritate) so much as the preservation of the narrative (which we, in our own coinage, called salva fictione [Lamarque and Olsen 1994, 126]). One inference that we drew from this was that the name ‘Canterbury’ in The Canterbury Tales is not used in a purely transparent manner. Although the name has a referential function in the Tales its use does not licence just any substitution of co-designative names (or singular descriptions) if the narrative is to be preserved. A further consequence of this – going beyond the merely semantic – would be to say that only certain aspects of the city of Canterbury (roughly speaking those aspects contemporaneous with the Tales themselves) are relevant in the picture readers are invited to form of the narrative content. To bring to mind aspects of the modern city of Canterbury – like its having a university – is at best a distraction in reflecting imaginatively on the goings on of Chaucer’s pilgrims. Similarly, although the expression (in the quoted lines) “The hooly blisful martir” clearly refers to St Thomas Becket, it would be a move beyond Chaucer’s narrative to substitute the definite descriptions “the saint whose shrine at Canterbury was destroyed by King Henry VIII” or “the saint whose assassination was the subject of a play by T. S. Eliot”, although these descriptions might figure in critical commentaries on Chaucer’s narrative. What the example shows is that the kind of opacity Olsen and I had in mind is apparent in three related but distinct occurrences: in certain kinds of narrative itself; in reports of narrative content; and in propositional attitudes (such as thinking, imagining, believing) taken towards narrative content. As far as narrative itself is concerned, it should be noted that the relevant species of opacity is not a feature of narrative per se, but only of certain kinds of narrative, notably (but not exclusively) those of fiction or literature. Only where the mode of narration is salient – in other words where the form in which a story is told matters in the appraisal of the narrative – will co-referential substitutions be blocked. Those narratives primarily concerned with imparting information – from homely conversational narratives to those of history or biography – will normally use proper names or other referential devices transparently. Of course the expression ‘mode of narration’ recalls the Fregean ‘mode of presentation’ and there will be cases even in information-imparting narratives where mode of narration

44 | Peter Lamarque is significant. Simple Fregean cases establish this; a narrative that states ‘Hesperus is Phosphorus’ is unlikely to licence substitution to ‘Hesperus is Hesperus’ where information-content is lost. We will see later how important mode of narration is to literary narratives. The opacity of narrative reports follows largely from the opacity of narrative. Where it matters to a narrative what form the narration takes, it will likewise matter to a report about the narrative that it treats its references opaquely (and indeed, as we shall see, preserves sense in other ways as well). But the general constraints on narrative reports are less those deriving from features of narrative, rather those deriving from features of speech and the reporting of speech. The principles of oratio obliqua and oratio recta apply in parallel fashion to the reporting of narrative. As we have seen, it is often entirely permissible, in the right context, to report narratives under de re or transparent formulations. And it might be appropriate in critical commentaries to invoke supplementary information about the subjects of narratives to enhance understanding or appreciation. Of most interest for the present enquiry is the effect of narrative opacity on the psychological attitudes that bear on narrative content. What is it to reflect on, hold in mind, imagine, or even make believe the content of a fictional narrative? To what extent are those mental states themselves subject to a corresponding opacity, limited, one might say, to de dicto rather than de re formulations?

3 Thus we are brought back to Thought Theory and indeed to the further issue of fiction and cognition. As mentioned, I shall not be defending Thought Theory here but developing certain elements of it. Five core aspects serve to characterize the theory. The first is the distinction between thought and belief. Bringing a thought to mind, reflecting on it as a thought, is not the same as believing a proposition to be true. One can entertain a thought without being disposed to assert its content as a world-directed truth. Furthermore, a second point, thoughts in the appropriate sense might encompass merely images or ideas and thus need not take the form of fully articulated propositions: that a is F or that some x is G. As such they need not even be candidates for truth. A third element is that of a thought-cluster. Here, the idea is that it is not likely to be isolated thoughts that cause significant emotional reactions. On its own, the thought, say, that Othello killed his innocent wife Desdemona in a fit

Thought, Make-Believe and the Opacity of Narrative | 45

of jealousy – or even less so the generalised thought of a jealous husband killing his innocent wife – is unlikely to elicit strong emotion. The intense and growing emotion in our response to Shakespeare’s play arises out of the precise delineation of this event, and events leading up to it, within a complex dramatic structure. It is the development of thought-clusters round the characters and round the fictive events – and the precise way those thought-clusters are constituted or formulated – that makes these episodes vivid and moving. A fourth element from Thought Theory seeks to account for the specificity or directedness of the elicited emotion. Given that the characters are fictional, what determines that it is specifically Othello that we fear and Desdemona that we pity in responding to the play rather than other individuals who might be in a similar predicament? The suggestion in my original paper is that again we need to appeal to thought-clusters, but more precisely that we need to ground the directedness of the emotions in both a causal and a content-based connection between the thoughts brought to mind and the descriptions and images in the work itself. Even if we learn of Othello and Desdemona from a secondary source without having read or viewed Shakespeare’s play, there must be a causal chain (perhaps of a Kripkean kind) traceable in principle from our thoughts to the play itself in order to ground our emotional responses (that is, to show the responses to be directed to those particular characters). But a causal link is not sufficient; there must also be an intentional content to the emotions, a way of characterising the objects of the emotions (what they are directed at), that connects them to the intentional content or meaning of the play (for example, the characters might come to mind under the original descriptions of them). The final aspect of Thought Theory was merely hinted at in my original presentation of the theory: that there are likely to be some kinds of isomorphism between the structure of a literary work – or at least parts of a work – and the structure of thoughts that the work elicits and which generate the emotional responses. It is that final element that is the starting point for what I want to propose now. Using the notion of narrative opacity, I shall argue that the cognitive benefits of literature are best explained through exploring the distinctive ways in which a work’s content is presented and the ways readers engage with the particularities of that content. To understand how the great works of literature can, speaking metaphorically, reshape our minds or change our inner landscape in a positive manner, we can appeal to the very elements of Thought Theory just outlined. It is not just that we come to think new thoughts when we engage with literary works – that is obvious – but what is of more interest is how we acquire those thoughts and how they can re-order our minds in subtle ways. How we

46 | Peter Lamarque think becomes as salient as what we think, just as how the content is presented can help determine just what that content is.

4 So, let us see how such a view might be supported. It seems an anodyne truth that a reader’s engaging with a fictional narrative involves, at least partially, the reader bringing to mind thoughts that relate to the propositional content of the narrative. Chapter 3 of F. Scott Fitzgerald’s novel The Great Gatsby begins with this sentence: “There was music from my neighbour’s house through the summer nights” (Fitzgerald 1967, 45). A reader would have little difficulty entertaining the thought of – indeed imagining – music coming from a neighbour’s house through summer nights. That is an entirely general thought with little specificity of detail and involving no singular reference. But of course the quoted sentence itself is grounded in a context and it does contain singular reference in the form of a definite description, “my neighbour’s house”. The possessive pronoun my needs to be negotiated. In fact it is a reference to the narrator, Nick Carraway, who has already introduced himself into the narrative. Although the sentence was written by the author Scott Fitzgerald the I in the narrative does not refer to him but to this fictional character. A reader ignorant of the fact that this is a novel with a fictional narrator could bring the wrong thought to mind, perhaps a singular thought with Scott Fitzgerald as its object. An inattentive reader might also conjure quite the wrong image from the phrase “my neighbour’s house”, perhaps supposing this to be a humdrum suburban neighbourhood; in fact, of course, the reference is to the house belonging to Jay Gatsby and there had been a passing description of it in an earlier part of the novel: The [house] on my right was a colossal affair by any standard – it was a factual imitation of some Hôtel de Ville in Normandy, with a tower on one side, spanking new under a thin beard of raw ivy, and a marble swimming pool, and more than forty acres of lawn and garden. (Fitzgerald 1967, 11)

This description somehow needs to inform the imagining when a reader brings the relevant thought to mind about “my neighbour’s house”. Finally, there is the reference to music. A reader who supposes this to be, let us say, loud rock music from a noisy neighbour would again be making an over-hasty and incautious inference. We are told a page or so later what kind of music to imagine:

Thought, Make-Believe and the Opacity of Narrative | 47 By seven o’clock the orchestra has arrived, no thin five-piece affair, but a whole pitful of oboes and trombones and saxophones and viols and cornets and piccolos, and low and high drums. […] playing yellow cocktail music. (Fitzgerald 1967, 46)

The simple example tells us much about what is involved in ‘bringing a thought to mind’ in reading a literary narrative. The sentence under discussion, “There was music from my neighbour’s house through the summer nights” seems quite unexceptional, yet the surface conceals the complexity in responding to literary fiction. Merely appealing to entertaining thoughts, imagining, or makingbelieve does not itself do justice to those complexities. A great deal of scenesetting – the fictional mode, narrative voice, reference, implicit and explicit, textual connectedness – must be in place for the appropriate thought and imagining to be grounded: ‘appropriate’ in the sense of being adequately informed by the relevant surrounding context. The presence of metaphor or imagery can make matters even more complicated. Consider, for example, the very next sentence at the beginning of Chapter 3: “In his blue gardens men and girls came and went like moths among the whisperings and the champagne and the stars.” Why “blue gardens”? (Artificial, not natural?) Why “moths”? (Attracted by the light? Short-lived? Creatures of the night?) Is it enough merely to say that readers make believe that men and girls came and went like moths among the whisperings and the champagne and the stars? The literal propositional content seems inadequate on its own to capture the thought that the literary context invites. Before we pursue these additional complexities, a word is needed about the relation, mentioned earlier, between thought and belief. A reader can entertain a thought without believing it to be true, that is, without a disposition to assert it. This is sometimes associated with the imagination and it might be considered the basic response to a fictional narrative. However, belief formation is an important element even in the fictional context: Beliefs are formed about what is true-within-the-fiction and without such beliefs, readers would not be able to grasp the content of the narrative. In the description of Gatsby’s parties, the narrator relates this: “[O]n Mondays eight servants, including an extra gardener, toiled all day with mops and scrubbing-brushes and hammers and gardenshears, repairing the ravages of the night before” (Fitzgerald 1967, 45). The reader will not only entertain this thought, but will take it to be true of the fictional world, or, more cautiously, at least something the narrator wants the reader to take as true. Furthermore, other truths are clearly implied, or facts that readers are invited to infer: Gatsby has at least eight servants and more than one gardener, his weekend guests leave a trail of mess and damage, the clearing up is a lengthy process (“toiled all day”), etc. However, not every thought enter-

48 | Peter Lamarque tained – even thoughts of a propositional form – leads to a corresponding belief. Readers form hypotheses or suppositions about fictional events and characters, holding in mind thoughts that they are not necessarily disposed to assert as truths: Readers need to judge whether and where the narrator is reliable. The ambivalence Nick Carroway shows towards Gatsby and other characters suggests that readers are not always getting an unbiased, or straightforwardly truthful, account. The most important consideration, though, which brings us back to narrative opacity, is that the fictional content represented in our propositional attitudes, either as beliefs or thoughts, needs to match to some degree of accuracy the perspectival nature, as I shall put it, of the narrative content itself: content given to us not transparently, but from a point of view, that is, essentially, not merely contingently, through the perspective of the descriptions constituting the content. This opacity, which goes beyond the mere referential opacity previously discussed in the Canterbury example, constrains the fittingness of the relevant beliefs or thoughts. Should there be a marked discrepancy between the mode in which the content is represented in the minds of the reader and its mode of representation in the narrative text itself, then dangers of misrepresentation or misunderstanding arise even if the characterisation in other contexts – for example, as part of a narrative report – might count as a reasonable paraphrase of the text. What is it for narrative to be opaque, or for its content to be presented perspectivally? In the context of poetry, the thought that form and content are somehow intimately connected is orthodoxy in the literary community, even if it can be formulated in different ways.2 But the lesson for narrative must be carefully applied. Consider the first stanza of Thomas Hardy’s poem The Darkling Thrush (Hardy 1994 [1900], 134): I leant upon a coppice gate When Frost was spectre-grey, And Winter’s dregs made desolate The weakening eye of day. The tangled bine-stems scored the sky Like strings of broken lyres, And all mankind that haunted nigh Had sought their household fires.

|| 2 For a classic statement, see Bradley (1926); for a recent reformulation (and defence of Bradley), see Lamarque (2009).

Thought, Make-Believe and the Opacity of Narrative | 49

The lines are presented as part of a small but intense narrative: The poem represents a speaker out on a cold and desolate winter evening who hears a thrush’s song (“An aged thrush, frail, gaunt and small, / In blast-beruffled plume, / Had chosen thus to fling his soul / Upon the growing gloom”) and infers “Some blessed Hope, whereof he knew / And I was unaware.” Terry Eagleton (2007, 121) has noted that this first stanza “highly compressed yet utterly lucid, is without an ounce of surplus fat” and that it is “remarkable for its tight interweaving of abstract allegory and keenly observed naturalistic detail”. The mood of wintry melancholy is reinforced in the succession of negative images: “[d]regs”, “desolate”, “weakening”, “tangled”, “scored” (i. e. cut or scratched), “broken”. The ghost-like quality of the scene is given in “spectre-grey” and “haunted”. For a reader to form an appropriate thought (and image) of the setting and narrative content, it seems that these epithets must play an essential, not merely contingent, role. It is not as if other ways of capturing the scene would be just as effective, for the scene itself derives its very identity (including its mood) through these exact lines. Form-content identity in poetry is sometimes explained in terms of unparaphrasability, but that is inadequate. Hardy’s lines could be paraphrased with varying degrees of precision for different purposes. In a school lesson, for example, it might be entirely legitimate (and helpful) to summarize in different words what is going on in the stanza. What seems clear, though, even truistic, is that the summary, or paraphrase, is not equivalent to the poem; it is not a matter of indifference if a reader reads the poem or the paraphrase. Part of the reason for this is that if one is reading the poem as a poem, then one should be receptive to the overall experience that the poem affords and that experience is partially determined by the very words and structures themselves. There is always more to a poem than just a core meaning that could be expressed in other ways. In this respect poems are different from other modes of discourse – including conversation and even philosophy – where alternative expressions can be just as effective in imparting an idea or message. When Cleanth Brooks coined the phrase Heresy of Paraphrase in relation to poetry, he was not saying that paraphrase of poetry is impossible or pointless – he acknowledges it might be useful in certain contexts – but that it should never be deemed substitutable for the poem itself. A poem never just makes a statement where what is stated is indifferent to how it is stated. It might be thought that form-content indivisibility is a peculiarity of poetry where fine-grained attention to the language used is integral to the kind of experience that poetry offers, and for which it is valued. But narrative opacity also shows how form helps determine content in prose narrative. In a literary novel, it is not a matter of indifference how scenes are depicted, as if the very same scenes might have been described in any other manner. The point, as with poet-

50 | Peter Lamarque ry, is not about the possibility or otherwise of paraphrase. Paraphrase of scenes in novels is entirely possible and occurs in ‘plot summaries’, but once again few would hold that a plot summary is as good as, or equivalent to, the original narrative form. Note that this is true for anything that might be classed as a literary narrative. There are other kinds of narrative – perhaps stories for children made up by parents or someone recounting the day’s events – that are not so bound to precise modes of narration. It is sometimes thought that the mark of the unparaphrasability of poetry is that no substitutions can be made – even synonyms for synonyms – while retaining the identity and integrity of the whole. But, it is said, such a strict rule does not hold for narrative prose. Synonyms in many cases could be substituted without any loss of work identity. However, the status of this claim is obscure. Perhaps all the vital meaning in a literary narrative could indeed be preserved through such (small, intermittent) substitutions and that might show that form is not so tightly tied to content in prose cases. But the idea that this might somehow licence rewording of literary works is absurd at many levels. A Jane Austen or Charles Dickens novel, let’s say, gets its identity, and through that its literary value, in the precise words – give or take minor issues of textual corruption, etc. – written by the author. Wilful changes, even preserving sense, would be unacceptable and undermine work identity. Even more significantly, it could never be said with confidence that any given synonym substitution is ‘harmless’ to a literary narrative. To read a literary work seriously is always to attend to, or be on the lookout for, linguistic nuance. A small example. Here, a critic notices the recurrence of the word peep in Dickens’s Bleak House: The word ‘peep’ echoes through the book, neatly conveying the partial and fragmented view of things which people usually achieve. Esther, for example, after her arrival at Bleak House, watches the dawn prospect from her window change ‘at every new peep’ and then decides to ‘take a peep at the garden’. Mrs Snagsby, curious about Tulkinghorn’s business with her husband, […] ‘peeps at them through the window-blind’; and, shortly afterwards, at Nemo’s death-bed, Miss Flite ‘peeps and trembles just within the door’. It is obviously no coincidence that the youngest member of the Jellyby family, a relentlessly curious and peripatetic child, should be named ‘Peepy’. The term ‘peep’ suggests a childish inquisitiveness, or a timid and old-maidish activity. But in Bleak House curiosity can quickly take less innocuous forms. (Ousby 1977, 978)

A single word in so long a novel might seem of marginal significance – and thus easily substitutable – but the lesson from narrative opacity is that there is a standing assumption that form of narration counts in the characterisation of content. The word peep, in this example, is not accidental, it has a function and

Thought, Make-Believe and the Opacity of Narrative | 51

salience in the narrative: It connects scenes, it holds nuances (‘childish inquisitiveness’), it contributes to an atmosphere (‘a partial and fragmented view of things’). Part of the pleasure of reading Dickens is savouring his use of language; the distinctive features of his characterization and scene depiction are determined by their precise linguistic delineation. In that sense form and content are indivisible. Consider another seemingly minor passage, also from Bleak House: My Lady Dedlock (who is childless), looking out in the early twilight from her boudoir at a keeper’s lodge, and seeing the light of a fire upon the latticed panes, and smoke rising from the chimney, and a child chased by a woman, running out into the rain to meet the shining figure of a wrapped-up man coming through the gate, has been put quite out of temper. (Dickens 1977, Ch. 2, 11)

Here are the critic Ian Ousby’s observations about this passage: When the reader encounters that passage for the first time the parenthetical phrase serves adequately to account for her [i. e. Lady Dedlock’s] apparent detachment and her concealed chagrin. Childless herself, she can only be a silent and distant spectator of such domestic scenes. But when the passage is read in the light of the novel’s denouement it takes on a more precise reference to her fate. The spectacle of the child ‘chased by a woman’ recalls Lady Dedlock’s separation from her illegitimate daughter, Esther Summerson. One further notes in chapter 18 that the lodge is actually the scene of Lady Dedlock’s first interview with Esther. (Ousby 1977, 976–977)

What the critic is noticing is a resonance in the passage with other parts of the novel. The description bears significance beyond its surface meaning. It does more than relate facts in the fictional world and this suggests that it is wrong to view narrative prose (in literature) as merely offering a window transparently onto a world, as if all that matters is that facts about that world be placed before us however described or as if the world itself somehow exists independently of the descriptions that constitute it. Perhaps David Lewis’s account of a fictional world, as a world where the fiction is ‘told as known fact’, might have these consequences. According to Thought Theory, the strength of an emotional reaction to fiction is likely to increase as thought-clusters take shape. But more needs to be said about the formation of these clusters; it would be wrong to give the impression that the matter is largely passive, an accumulation of thoughts and images through an essentially linear, and causal, progression. Needless to say, the process is more complex. To consolidate the core thesis that works of literature can serve in some sense to reshape our minds, we must focus on the role of distinctively literary modes of reading in the forming of thought-clusters.

52 | Peter Lamarque At the heart of narrative opacity is the idea that a reader’s attention to textual nuances, implicit evaluations, symbolic resonance, humour, irony, tone, allusions, or figurative meanings in the textual content will help give precise shape to the thoughts that the content brings to mind. Similarly the very ordering of the material, the manner in which information is imparted in a narrative, and the setting up and fulfilment of expectations can structure the reader’s perspectives. Let us return briefly to The Great Gatsby. The character of Gatsby himself develops slowly in the novel. Crucial information about his true background (his criminal activities, his modest upbringing, his obsession with Daisy) takes time to emerge. Early in the novel, he is a mysterious, aloof figure attending, but not participating in, the lavish parties at his house. People turned up uninvited: “Once there they were introduced by somebody who knew Gatsby, and after that they conducted themselves according to the rules of behaviour associated with an amusement park” (Fitzgerald 1967, 47). They gossiped about him: ‘He’s a bootlegger’, said the young ladies, moving somewhere between his cocktails and his flowers. ‘One time he killed a man who had found out he was the nephew to Von Hindenburg and second cousin to the devil. Reach me a rose, honey, and pour me a last drop into that there crystal glass’. (Fitzgerald 1967, 67)

The narrator seems torn in his attitude to the fabulous opulence of Gatsby’s world, and indeed to Gatsby himself, an ambivalence displayed in his narration. At times the hollowness of it strikes home, for example, in reflecting on the atmosphere round Gatsby’s house when the guests leave: A wafer of a moon was shining over Gatsby’s house, making the night fine as before, and surviving the laughter and the sound of his still glowing garden. A sudden emptiness seemed to flow now from the windows and the great doors, endowing with complete isolation the figure of the host, who stood on the porch, his hand up in a formal gesture of farewell. (Fitzgerald 1967, 62)

But as Gatsby’s doomed relationship with Daisy emerges, a more tragic and thus sympathetic figure develops, shaping the reader’s own attitudes and again drawing on those of the narrator. At the fateful reunion of Gatsby and Daisy, Gatsby’s vulnerability and the hopelessness of his quest are captured in this somewhat wistful description: As I went over to say goodbye I saw that the expression of bewilderment had come back into Gatsby’s face, as though a faint doubt had occurred to him as to the quality of his present happiness. Almost five years! There must have been moments even that afternoon when Daisy tumbled short of his dreams – not through her own fault, but because of the

Thought, Make-Believe and the Opacity of Narrative | 53 colossal vitality of his illusion. It had gone beyond her, beyond everything. He had thrown himself into it with a creative passion, adding to it all the time, decking it out with every bright feather that drifted his way. No amount of fire or freshness can challenge what a man can store up in his ghostly heart. (Fitzgerald 1967, 102–103)

In the final meeting between the narrator and Gatsby, the narrator’s ambivalence is out in the open: ‘They’re a rotten crowd’, I shouted across the lawn. ‘You’re worth the whole damn bunch put together.’ I’ve always been glad I said that. It was the only compliment I ever gave him, because I disapproved of him from beginning to end. First he nodded politely, and then his face broke into that radiant and understanding smile, as if we’d been in ecstatic cahoots on that fact all the time. His gorgeous pink rag of a suit made a bright spot of colour against the white steps, and I thought of the night when I first came to his ancestral home, three months before. The lawn and drive had been crowded with the faces of those who guessed at his corruption – and he had stood on those steps, concealing his incorruptible dream, as he waved them goodbye. (Fitzgerald 1967, 160)

Here, then, is a character whose very being is given, as I put it, perspectivally, that is ‘under-a-description’, ‘from a point of view’. The narrator’s shifting attitude towards Gatsby determines the reader’s own shifts in thought and imagining. The thought-clusters that develop in a reading of the novel are not just a succession of propositions held in mind. They are shaped, manipulated, and given structure through the linguistic resources of the narrative prose. This is narrative opacity. Far more is going on in the reader than just an amassing of facts about a fictional world. That world is constituted both by the descriptions given and the images evoked. It is also shaped in other ways as well, by the strong undercurrent of wider themes that inform the thought-clusters and provide literary significance: themes about the American dream, illusion, wealth and happiness, new money and old, style, status, cynicism, hope and loss. Recurrent symbols – the eyes of Doctor T. J. Eckleburg, the green light at the end of Daisy’s dock – give further structure to the developing thoughts. Gatsby and his world lodge in the mind not as mere character and plot, but as a far more complex constellation of ideas and images.

5 So far, I have concentrated on thoughts directly related to character and incident in fiction, thoughts, we might say, at a first-order level. I have sought to emphasise that belief acquisition even at this level is perspectival. We come to

54 | Peter Lamarque believe what is the case in the narrative world by developing thoughts that are shaped in our minds by how the fictional content is presented. The considerable potentialities here can be overlooked if too simple a model of responding to fiction is followed. But there is another level of thought-development in response to fiction that is even more complex and is sometimes thought to be at the heart of the cognitive yield of works of literature. This concerns thoughts that arise out of, or through reflection on, the first-order thoughts. The particularities of a work of fiction can coalesce into a second-order overview where broader thoughts of a thematic kind relating to the whole work take shape. We are still in the realm of thoughts and thought-clusters, but they seem to be of a higher order and for many people are associated with just the kinds of insights or truths commonly attributed to literature as its principal cognitive payoff. For a final example, let us consider George Meredith’s novel The Egoist (1979 [1879]). The plot itself is fairly convoluted, but the central theme, indicated in the title, is clear enough. At the core of the novel are two main protagonists, Sir Willoughby Patterne (“The Egoist”) and the beautiful and resourceful Clara Middleton, to whom he is engaged to be married. As it becomes evident to Clara just what demands Sir Willoughby will make of her in the marriage, drawing her into his egoistic world and allowing her no independence of thought, she grows more and more concerned and seeks to escape from the engagement. He refuses this, and constantly misinterprets her motives. Eventually, she tries to run away, even at the cost of losing everything, but gets no further than the railway station. Meanwhile, Clara becomes acquainted with Laetitia Dale, who had once herself been courted by Sir Willoughby and seemed not to have Clara’s concerns about his ‘egoism’, indeed was still in love with him. Clara contrives to reignite this relationship, which she believes would suit both parties as well as releasing herself. The novel ends with Clara finally managing to get out of her engagement with Sir Willoughby, who has been persuaded to pursue Laetitia instead, even though the latter is more open-eyed now about his egoism. A double marriage concludes the novel, between Sir Willoughby and Laetitia (after laying down some conditions of her own) and Clara and Vernon Whitford, a cousin of Sir Willoughby and a literary scholar who had been living at Patterne Hall. Meredith called the novel a ‘comedy in narrative’ and the ‘happy’ ending suggests just such. But the themes of egoism and in particular the plight of women in Victorian society are powerfully drawn and seriously meant. The critic Robert M. Adams writes this of Clara Middleton illuminating both the character and the core theme: She appeals to us not on the basis of legal or ethical technicalities, but as a human being caught in an inhuman system of commodity relationships. By her father, by Willoughby,

Thought, Make-Believe and the Opacity of Narrative | 55 by “right-thinking society” as a whole, she is valued for her purity, her docility, her serviceability to men. Meredith’s intuition of the suffocating web that can be woven about a young woman by playing on these “virtues” of hers is a tremendous imaginative achievement. What makes her struggle so exasperating is that breaking the net would be so easy for her, if only she would forfeit her fragile reputation. How she guides herself surely, instinctively, without calculation or crudity, through a labyrinth of blinding dilemmas to safety is the heart of Meredith’s book. (Adams 1979, viii)

Certain images run through the novel helping to illustrate Clara’s predicament and her relation to Sir Willoughby: Clara as an objet d’art is one of the central images. She will ornament Sir Willoughby’s house and his life. She will be static, at home; Sir Willoughby pictures her awaiting his return from masculine pursuits; it is the classic Victorian male image of the wife. Closely linked with this is the image of possession, of enslavement, which recurs varied and insistent throughout. […] A second group of images sharpens the focus. These are images of hunting, prey and sacrifice. Meredith describes Clara running like a hare pursued by hounds. She is also “a victim decked for the sacrifice”. This particular phrase returns us to the ornament image for she goes on to describe herself as “the garlanded heifer you see on Greek vases”. (Calder 1979, 476–477)

These images have a literary function in connecting scenes, elaborating the character of Clara, and of course giving vitality to the more abstract themes of egoism and the oppression of women. In fact the idea of egoism is more complicated, and nuanced, than might be supposed when simply tied to the selfishness of Sir Willoughby: The egoist we understand to be Sir Willoughby, yet Clara describes herself as an egoist, and it is her ‘egoism’ that prompts her rebellion, her awareness of herself as an independent, individual woman. This idea of egoism, which runs through practically everything that Meredith wrote, is double-edged. Egoism is on the one hand the core of selfish, limited, self-aggrandizing behaviour; on the other it is the spur to what Meredith calls insurrection. A sense of oneself is vital. An ignorance of self in others is disastrous. Sir Willoughby and Clara demonstrate the two facets of the same idea. (Calder 1979, 474)

Meredith subscribed to an evolutionary conception of humans, whereby modern behaviour often recalls or reverts to our primitive ancestry. The theory is presented explicitly in the novel: The Egoist is our fountain-head, primeval man; the primitive is born again, the elemental reconstituted. Born again into new conditions, the primitive may be highly polished of men, and forfeit nothing save the roughness of his original nature […] he has become the civilized Egoist; primitive still, as sure as man has teeth, but developed in his manner of using them. (Meredith 1979, 324–325)

56 | Peter Lamarque There is a close connection between the themes of The Egoist and John Stuart Mill’s essay On the Subjection of Women (1869), which Meredith had read and admired. Indeed it has been claimed that “what Meredith has done in The Egoist is to dramatize the ideas of Mill” (Hill 1979, 523). Here, some connections are spelt out: Mill describes the legal dependence of women as “the primitive state of slavery” which has not lost “the taint of its brutal origin”. Women are in the situation of bond-servants, educated to accept their captive state. “Men do not want solely the obedience of women, they want their sentiments”, Mill writes, expressing the thesis of The Egoist precisely. “How many are the forms and gradations of animalism and selfishness, often under an outward varnish of civilization and even cultivation!” he exclaims in a voice that could as well be Meredith’s. “There is nothing which men so easily learn as this self-worship,” he is observing a little later, and to all that Mill has to say about the effect upon men and women alike, and upon the harmony of their relations in the state of marriage, which the inferiority imposed on women is bound to produce, Meredith is in fullest accord. (Hill 1979, 523–524)

The novel, then, offers themes of a serious kind that bind together the particulars of character and plot and give them higher-order significance. It is sometimes supposed that it is at this thematic level that we enter the realm of the truth-telling contribution that novels can afford. We learn from novels, it is sometimes said, because the novels offer truths that thematic interpretations uncover. The Egoist has literary value on this view at least partially because it reinforces Mill’s arguments about the subjection of women, offering insights both into that debate and into an analysis of egoism as a manifestation of primitive human behaviour. No doubt there are contexts where Meredith’s novel can count as a powerful contribution to analyses of the role of women in Victorian society and to philosophical and socio-psychological debates about egoism. But it should not be thought that the primary role of literary interpretation is to extract social, moral or political truths that have general application to a wider field of enquiry. Another role seems far more closely tied to the literary enterprise – it relates to the kind of perspectivalism described earlier. Here, interpretation seeks to heighten our experience of the fictional world: in the terms of Thought Theory, to help develop and bind together thought-clusters. This is where the earlier metaphor of a novel reshaping our minds or changing our inner landscape again comes into play. Of course, no adequate response to The Egoist could fail to reflect on the themes identified. But the themes can be seen as providing another perspective through which the thoughts and images arising from the novel’s particulars take shape. Rather than asking whether Meredith is right to try to explain the selfish behaviour of Victorian men as a throwback to primitive ancestral dispo-

Thought, Make-Believe and the Opacity of Narrative | 57

sitions, it does better justice to his novel to ask instead how effective the concept of the primitive is in organising and making sense of the material presented in the novel itself. We might get a far deeper and more lasting ‘insight’ into these themes by calling them to mind through the particulars (the characters, the dialogue, the scenes) than by seeking to formulate generalized truths suitable for extra-literary debate. This is where the ‘vision’ of the novel lies. Its power is in the particulars, not the generalities.

6 Let me end by drawing some of the threads together. I have argued that the content of literary fictional narrative is infused with a kind of opacity. The content is given and thus constituted, as we might say, ‘under a description’. This is true not just for the fictional characters and fictional incidents described in the narrative, which acquire their nature and very existence from the modes of their presentation, but also for any real-world setting which itself is presented under a perspective. The opacity applies not just to proper names and their referential function. What I have called perspectivalism runs deep in narrative representation; tone, humour, irony, connotation, allusion, narrative voice, and other aspects of representation, colour all narrative that aspires to literary status: Or perhaps more accurately, one should say that readers come to literary works with an expectation that narrative perspective of this kind is salient, that the modes of representation are significant. I have shown how the relevant kind of opacity occurs in narratives themselves, in reports of narrative, and in thoughts about narrative content. All this complicates, although probably remains consistent with, the idea that fictional narratives prescribe imaginings. But it focuses attention on the constraints under which that imagining takes place. I have concentrated on the thoughts and thought-clusters that readers form in responding to works of literature under the expectation of what might be called literariness. My emphasis has been on the way that narrative can shape the thoughts of those attending to it, not merely by imparting beliefs about narrative worlds, but by informing the thoughts with the unique perspective through which the worlds are given. Although I have not said much about it, I believe that such perspective-imbued thoughts, arising from attention to the modes in which narrative represents its content, can go some way towards explaining how and why we respond emotionally as well as cognitively to narrative. This is where I hold that Thought Theory can contribute both to the problem of emotional response and to the problem of literature and knowledge.

58 | Peter Lamarque Attending to fictional particulars under the perspective of authorial and textual control, subject always to genre or cultural conventions, can shape the mind by inducing and guiding thoughts and thought processes. The building up of thought-clusters, of the kind that according to Thought Theory elicit emotional responses, can reconfigure our minds, usually, although not inevitably, in positive ways. Novels, we say, leave their mark on us. The great ones stay with us forever. This is what makes the experience of reading a novel valuable and valuable for its own sake. What we go on to do with the thoughts thus accumulated is another matter altogether. They might affect our subsequent actions and attitudes, they might re-order our conception of ourselves. They might have no practical effect whatsoever. But any effects they do have will be contingent and largely dependent on local psychological dispositions of individual readers. This is why one should be careful in proposing too ambitious a truthrevealing capacity for works of fiction over and above the capacity to present particularities perspectivally and literally in thought-provoking ways.3

References Adams, Robert M. “Preface.” George Meredith: The Egoist. Ed. Robert M. Adams. New York: W. W. Norton & Co., Ltd, 1979, vii–ix. Bradley, Andrew Cecil. Oxford Lectures on Poetry 2nd ed. London: Macmillan, 1926. Calder, Jenni. “The Insurrection of Women.” George Meredith: The Egoist. Ed. Robert M. Adams. New York: W W Norton & Co, Ltd, 1979, 472–480. Carroll, Noël. The Philosophy of Horror: or Paradoxes of the Heart. London: Routledge, 1990. Chaucer, Geoffrey. The Works of Geoffrey Chaucer 2nd ed. Ed. Fred N. Robinson. Boston: Houghton Mifflin Co., 1957. Dickens, Charles. Bleak House. Eds. George Ford and Sylvère Monod. New York: W. W. Norton & Co., Ltd, 1977. Eagleton, Terry. How To Read a Poem. Malden, MA: Blackwell, 2007. Fitzgerald, F. Scott. The Great Gatsby. Harmondsworth: Penguin, 1967. Hardy, Thomas. The Collected Poems of Thomas Hardy. Ware: Wordsworth Editions Ltd, 1994 [1900]. Hill, Charles J. “Theme and Image in The Egoist.” George Meredith: The Egoist. Ed. Robert M. Adams. New York: W W Norton & Co, Ltd, 1979, 518–524. Lamarque, Peter. “How Can We Fear and Pity Fictions?” British Journal of Aesthetics 21 (1981): 291–304. Lamarque, Peter. “The Elusiveness of Poetic Meaning.” Ratio 22:4 (2009): 398–420.

|| 3 There is some overlap between this paper and another paper of mine, see Lamarque (2012). And parts of both are reproduced in Lamarque (2014).

Thought, Make-Believe and the Opacity of Narrative | 59 Lamarque, Peter. “Thought Theory and Literary Cognition.” Understanding Fiction: Knowledge and Meaning in Literature. Eds. Jürgen Daiber, Eva-Maria Konrad, Thomas Petraschka, and Hans Rott. Münster: Mentis: 2012, 67–80. Lamarque, Peter. The Opacity of Narrative. London: Rowman & Littlefield International, 2014. Lamarque, Peter, and Stein Haugom Olsen. Truth, Fiction, and Literature: A Philosophical Perspective. Oxford: Clarendon Press, 1994. Meredith, George. The Egoist. Ed. Robert M. Adams. New York: W. W. Norton & Co, Ltd, 1979. Ousby, Ian. “The Broken Glass: Vision and Comprehension in Bleak House.” Charles Dickens: Bleak House. Eds. George Ford and Sylvère Monod. New York: W. W. Norton & Co., Ltd, 1977, 974–984.

James R. Hamilton

Narrative per se and Narratability This paper is exploratory. The approaches I take in it are novel, I believe; but I also am only guardedly optimistic about its conclusions.1 I began thinking about the issues discussed in this paper because I wanted to know what predicates, usually associated with the contents of narrative theatrical performances, might also be carried by performances themselves. As will become evident, this is not a question about how much diegetic material is to be found in the contents of theatrical performances. Nor is it to ask how that diegetic material works in performances. It is, instead, to ask if what narratologists call the medium of performance bears some of these same features. My initial guess was that a moment in performance should be capable of being suspenseful, for example, independently of any suspenseful moment occurring in the content of the performance. After all, if what we sometimes feel just before the diver leaves the board can be called suspense, then suspense could also be something which can be induced, and perhaps is induced routinely, by the physical and vocal actions of performers. The importance of that fact is that it can become an aesthetic or critical issue of some importance in particular cases whether, say, performance suspense lines up with, precedes, trails, or is otherwise juxtaposed to suspenseful moments within a performance’s content.2 I confess that I have not found everything for which I was looking in this paper. I do not, for example, believe I have yet found general support for my initial guess. But I do believe I have made serious headway.

|| 1 This paper began life as the basis of a presentation to the conference How to Make Believe. The Fictional Truths of the Representational Arts held at the University of Lund, Sweden, in March 2012. Since then, it has been substantially revised for the European Society for Aesthetics, held in Braga and Guimaraes, Portugal, June 2012, and again for the Conference in Honor of Peter Goldie, held at the University of Manchester, September 2012. This version differs as well from each of those. 2 This fact alone should show us why explorations of the capacities of different narrative media, such as those undertaken by Ryan (2005), for example, are important to philosophers working in aesthetics.

62 | James R. Hamilton

1 Introduction I begin by noting a distinction between two ways we can ask the question I wish to engage with. The first goes like this: What is the difference between the narration of a sequence of events and a description of the same sequence? You may recognize this first question as a classical question of the discipline of narratology. However, as a number of narratologists have observed, narratology was initially grounded in, and focused upon, the study of the novel and of written fictional narratives more generally. Only gradually did it broaden out to include nonfictional narratives. Still, until quite recently, it has largely remained focused upon written narratives, whether fictional or nonfictional. In the last three decades, however, challenges to this way of thinking have been prominently put forward by such figures as Monika Fludernik (1996; 2008), MarieLaure Ryan (2005), Brian Richardson (1987, 1988, 2011), and Ansgar Nünning and Roy Sommer (2008, 2011). The origins of this new way of thinking go further back, to at least Claude Bremond (1964), who claimed: [Story] is independent of the techniques that bear it along. It may be transposed from one to another medium without losing its essential properties: the subject of a story may serve as argument for a ballet, that of a novel can be transposed to state or screen, one can recount in words a film to someone who has not seen it. These are words we read, images we see, gestures we decipher, but through them it is a story that we follow, and it could be the same story. (Bremond 1964, 4)3

While this is true, Ryan (2005, 1–3) claims it is too simple. For some media, she contends, are better at narrative than others, in terms of having more and greater capacities for employment of various particular narrative techniques. And this pair of facts – that story is not tied inherently to any particular medium but that different media possess inherent, distinctive, and sometimes limiting sets of narrative techniques – should lead us to think that a more nuanced account of ‘narrative’ and ‘narrativity’ than has usually been given is necessary.

|| 3 The original article reads: “La structure de celle-ci est indépendante des techniques qui la prennent en charge. Elle se laisse transposer de l’une à l’autre sans rien perdre de ses propriétés essent ielles: le sujet d’un conte peut servir d’argument pour un ballet, celui d’un roman peut être porté à la scène ou à l’écran, on peut raconter un film à ceux qui ne l’ont pas vu. Ce sont des mots qu’on lit, ce sont des images qu’on voit, ce sont des gestes qu’oji déchiffre, mais à travers eux, c’est une histoire qu’on suit ; et ce peut être la même histoire.” The translation I use here is from Ryan (2005).

Narrative per se and Narratability | 63

Accordingly, and in concert with a number of other narratologists, Ryan pursues an account of narrative and narrativity that is both inherently mediumindependent but also sensitive to differences among media, and that narrativity is a matter of degree rather than categorical. I do not take a position within the disputes among narratologists about how the questions they pursue should be answered. I only note that, like the metaphysically inclined philosophers, they are asking questions about what makes any representation a narrative.4 I will return briefly to the issue about differences in media in the last section of this paper. For now I wish only to note that I aim to avoid the metaphysical question in any of its forms. Instead, I want to ask something somewhat different, namely: What is the minimum necessary that must be presented to someone for her to be positioned to tell a story, as a way of demonstrating she has understood what was presented to her? The difference is that the first question is metaphysical, aiming at discovering what makes a representation into a story, while the second, the one I will address in this paper, is epistemological, aiming at discovering when someone is justified in narrating the events presented in the prompt. Another way of putting the latter question is this: What sort of evidence must someone have in order to make a certain kind of judgment, namely, that whatever it is that someone has encountered is appropriately ‘narratable’? (see Goldie 2003). I use this term to mark a normative category. But there are at least two ways of thinking of ‘narratability’ as a normative category. Gerald Prince (2008, 24– 26) takes one of those ways when he holds that a sequence of events is narratable just when an agent understands it to be “worthy of being narrated”. This category of narratability is – for lack of a better term – morally normative. The contrast class for Prince is that which is ‘narrativizable’, that is, that which can be set forth as a narrative. But, as he is quick to point out, pretty much everything can be narrativized. That might, of course, be true. And, in any case, I shall not dispute it here. The other way of marking the category, and the path I take, is purely epistemic. I will agree for the sake of argument that pretty much any sequence of events could be represented and expressed by an agent in the form of a narrative. This view seems plausible to a number of people. But it does not seem nearly as plausible to anyone to hold that an agent is always epistemically justified in narrativizing all bits of one’s experience. One way of putting this is to say || 4 This is true even of Fludernik’s (1996) view, with which my own view about theatrical spectating, developed elsewhere, is consonant largely because it focuses on the mental apparatus involved in grasping a narrative. The details are significantly different, however.

64 | James R. Hamilton that it is possible to narrate a sequence of events – for example, perhaps, one’s own life – that may not in fact comprise a narrative on any plausible metaphysical account of what a narrative is. I do not wish to settle any substantive issues in making that observation; I only want to stress the possibility. And, henceforth, when I write about a sequence being ‘narratable’ I shall only mean ‘epistemically justifiably narratable’. For the question I want to address is about the conditions that make a sequence of events justifiably narratable, epistemically. For sake of brevity, however, I will usually drop the modifiers. In what follows, I also do not abandon the more traditional metaphysical questions altogether. Indeed, some of the same kinds of issues I wish to address are raised in the attempts at answering the metaphysical question in its various forms. However, it should also be obvious that, because my focus is on the epistemic question, I am under no compulsion to enter the debate between those who, in taking up the minimalist strategy, think of themselves as providing minimal criteria for the existence of a story and those philosophers and narratologists, like Gregory Currie, Marie-Laure Ryan and Monika Fludernik, who think of narrativity as a graded property, as something that comes in degrees and is specifiable by reference first to features of ‘richer’ or ‘successful’ narratives (Currie and Juriedini 2004; Currie 2010; Fludernik 1996; Ryan 2005).5 In sum, my use of the minimalist strategy is not aimed at determining when we actually have a narrative; it is rather to determine when a spectator in the theater, for example, is justified in telling a story, rather than, or in addition to, describing the events or images she has witnessed. But I confess the picture I am about to draw looks initially a lot like minimalist metaphysics and philosophy of mind. Indeed, I openly help myself to some of that material because it is where the discussion helpful to answering my question has taken place. I only ask for patience at this point.

2 Narrative per se Consider some variations on this famous pair of sentences: (1) The king died and then the queen died. (2) The king died and then the queen died of grief.

|| 5 Goldie’s (2011) review of Currie (2010) also seems to endorse this view.

Narrative per se and Narratability | 65

According to E. M. Forster (1927, 130), the first is a story, the second a plot. The first answers the question, ‘and then?’ while the second answers the question ‘why?’6 David Velleman (2003, 3) declares that he uses ‘story’ to mean what Forster calls ‘plot’. On Noël Carroll’s (2001) influential view (canvassed below), only the second is a story. And I shall use (2) as a model in what follows. I do not thereby endorse either Velleman’s or Carroll’s theories about minimal narratives. My point, instead, is to simply get to a minimum that all would agree to. And (2) offers a sufficient model. Taking up the indisputably smaller (if not the indisputably smallest) will do for present purposes. Peter Lamarque (2004, 394) claims that “narratives can be identified from formal features alone of individual sentences or sentence strings and no implications about reference, truth, subject matter or discursive ends, can be drawn from such formal identification”. There is surely something right about this. For some sentence or string of sentences to be a narrative, all that seems required is that (A) at least two events or states must be depicted, (B) the relation between the events or states must be depicted as temporal, and (C) there must be some loose ‘non-logical’ connection depicted between them (see Lamarque 2004, 393). Lamarque’s (C) is, however, too loose, at least as it stands. Some similarity constraint has to be imposed on the depiction of the events so as to avoid the result that the depiction of just any two events whatever, no matter how distanced from each other historically, thematically, causally, or emotionally could count as a narrative depiction. I think Lamarque wants to rule this out by specifying there be some sort of ‘non-logical connection’. But both being-depictedside-by-side and being-depicted-one-after-the-other are instances of ‘nonlogical connections’. Neither, however, takes us beyond the satisfaction of (A) and (B). So we need something stronger. Let us call what Lamarque offers us (C-LOOSE). How should we tighten up Lamarque’s (C-LOOSE) so that the ‘nonlogical’ connection between the events/states is still ‘loose’ in some sense while we also remain ‘minimalists’? Consider the following candidates for a ‘similarity constraint’ – a condition that marks some causal or quasi-causal similarity between antecedent and succeeding event(s):

|| 6 Narratologists will find the formulations here, most of which are taken from philosophical accounts of these issues, quite familiar because they trace what are now fairly standard positions in debates within their own discipline.

66 | James R. Hamilton –

–

–

–

(C-INUS): the connection between the events/states must be an insufficient but necessary part of an unnecessary but sufficient condition; (Carroll 2001, 123–124); as Carroll makes clear, the INUS conception of causal connection was first articulated in Mackie (1965); (C-EMOTION): the connection between the events/states must be such that the pair of events/states sets up and resolves or “completes” an “emotional cadence” in the audience for the story (Velleman 2003, 6); or (C-AGENT): the events/states are the intentional doings or non-doings of one or more ‘agents’, understood broadly as “planning mechanisms”. (Hamilton 2007, 65–68) (C-CONTROL): the “cumulative end point [of the temporally ordered chain of events/states] is in a position to control what came before it”. (Kreiswirth 2000, 299)

Setting these forth disjunctively might suggest they are mutually exclusive alternatives. But do we have to choose among these? I do not think we need to do so as a general matter, so long as we think of them as different, but equally plausible, ways of specifying the similarity constraints needed to make a narrative in a particular case. This is another way to take the disjunctive presentation. Accordingly, which particular constraints are chosen will depend upon the context and point of the narrative and which possibilities are to be regarded as properly ignored, given the context and purpose of the narrative. I model this thought on David Lewis’s (1996) contextualism about knowledge. It is also supported by Currie’s contextualist approach to the question when to determine whether one telling is narrative. It is, he argues convincingly, a matter of what you are comparing that particular case of telling to (see Currie 2010, 35–36). As a general matter, then, we can say a depiction is a narrative if it satisfies (A) and (B) and (C-LOOSE), so long as (C-LOOSE) is understood as a placeholder for one of the foregoing formulations (or some other good alternative) that specifies the similarity constraint at work in a particular case. A final note on this point: The list of candidates is not intended to be exhaustive. Nor am I committed to the distinctness of each of these. I suspect that (C-CONTROL) is not significantly different from (C-EMOTION); and there also might be other overlapping cases. It will suffice if the formula for specifying the similarity constraint yields a fair and approximate understanding of narrative per se.

Narrative per se and Narratability | 67

3 Narratives per se as bearers of basic aesthetic predicates So, consider again the model of a minimal narrative: (2) The king died and then the queen died of grief. Now consider these variations: (4) The king died and then the queen died of laughter. (5) The king died and then the queen died of boredom. (6) The king died and then, after a very very very long time, the queen died of laughter. (7) The king died, the king died, the king died, the king died, and then the queen died of boredom. (8) The king died after the queen had died of grief. (9) The king died violently soon after the queen had been poisoned. (10) The king died, and soon the queen shall die of grief. (11) “The king died, and I shall soon die of grief.” (12) The queen declared she had died long before the king had died. (13) ‘Did the King die?’ the Queen asked. ‘Then I shall die of grief.’ Each of these satisfies the minimal criteria. Each also allows us to consider features Gérard Genette (1990) suggests can be present in any narrative: order of presentation, speed of presentation (acceleration, deceleration, ellipsis, pauses), frequency (iterative narratives), mode (direct access to the mental life of characters: indices, internal monologues, free indirect discourse), and voice (author-narrator relationship). Consider now some syntactical features of these minimal stories. Most introduce a prepositional phrase, e. g. ‘of boredom’, modifying the second event. One introduces an adverbial modifier to the verb, indicating the manner in which the first of the events occurred. Consider as well some semantic features of these stories. Some involve forecast. Some – e. g. (6), possibly (7), and (12) – involve metaphorical uses of terms referring to events. Some, e. g. (9), introduce possibilities for causal explanation that must be – and usually are – supplied from outside what is depicted. Some, e. g. (13), require explicit recognition of a narrator. One (11) introduces a first-person utterance in which the author and the narrator appear to be the same individual.

68 | James R. Hamilton In addition to these individually specifiable features, plausibly some of these stories can be paired in terms of whether they recount the same story events. In setting the pairs beside each other, one might be prompted to think that some drive home the emotional point of the same story events more effectively than do their counterparts. Those that strike us as employing metaphors may employ the same metaphor as well as recount the same event. But our interest in those tellings may be different. One might leave out something of interest that we would like to have filled in. One might be more mundane than the others. And one, if it is a first person narrative, might be taken to be more compelling than the others, perhaps for precisely that reason. Without enriching the budget very much, if at all, one can see how to go on, staying with the indisputably smaller cases and yet developing story variants that have different tones, express different moods, demand different qualitative assessments. Among those could easily be stories that are dull, rambling, labored, episodic, disjointed, elegant, or graceful. Listing them out together, moreover, brings out the fact that they could also bear relational predicates to one another, predicates such as brash, tacky, edgy, innovative. At least at the level of the particular analysis of the effects of these stories, I do not take a position on which effects are caused by which of these stories. Indeed, I think there are empirical questions about some of those effects, and I do not have any data to help sort those out. But notice that the features I have just adduced as attaching to some narratives just because they are narratively structured in particular ways are classic instances of aesthetic features. Therefore, I think we have no reason to simply rule it out a priori that the same story events cannot be presented in different ways each carrying different aesthetic interests and different foci of aesthetic interest. Some who have certain metaphysical scruples will be seriously worried about whether we should speak of ‘aesthetic features’ or ‘aesthetic properties’ at all. Certain Nominalist worries, for example, have to do with whether there are properties, but rather only predicates. Other ontological worries have to do with whether there are distinctively aesthetic properties/predicates or whether, instead, they only supervene on genuine base-level physical properties/predicates. None of those worries are germane to the present discussion. So, with apologies to the metaphysicians among the readers, I ignore those issues altogether.

Narrative per se and Narratability | 69

4 Richer predicates How far can we take the generation of interesting predicates onto minimal narratives? I should think we would want to be able to show that certain special predicates can apply to narratives per se before we were ready to pronounce that they are capable of bearing sufficient aesthetic interest to be the basis of a ‘literature’. If we thought otherwise, we might have to concede that there are distinctively literary predicates that are the center of our interests in literature and that they are not derived, nor even derivable, from narrative or narrativity, per se. I am not ready to make that concession, at least not just yet. However, I see the point. One might reasonably think that, in addition to the basic aesthetic features I have defended as attaching to narratives per se, at least three further literary predicates need to be discussed, namely suspenseful, ironic, and tragic. A literature that did not allow these predicates to be applicable to at least core instances of narratives would be – by our lights at least – a much diminished example of the kind. But, as I stated earlier, it is not part of my project to defend a minimalist metaphysics of narration. Instead I now turn to another feature of the stories adduced so far that gives room to the discussion I do wish to encourage.

5 An explanation I think that any account of narrative that allows narratives to make room for, indeed to prompt, anticipatory readiness also provides both authors and their audiences opportunities for narrative play and pleasure in that play. I further believe that minimal narratives can do this. By being structured in any of the ways listed above, narratives per se allow for certain possibilities of presentation that make them interesting. The possibilities are first that what is narratable can be presented with a different temporal ordering than the ordering of (or at least retrievable from) the events presen-ted (Carroll 2001, 126), and second that what is narratable may induce recognition of what Carroll (2001, 125, 127) calls retrospective significance – the significance of one depicted event is often evident only upon the depiction of another. Both explicitly fall out from Carroll’s (C-INUS), but they are available from any of the other ways of filling in the connection indicated in Lamarque’s (C-LOOSE). These same two possibilities are also available on another and a different kind of account, a ‘scalar account’ of ‘narrativity’, found in work by Gregory

70 | James R. Hamilton Currie and Jon Juriedini (2004). This account allows that “we know [any given depicted event] happens in the story for a purpose, though it may not be clear yet what the purpose is, or that this purpose connect this event with much else the story contains” (see Currie and Juriedini 2004, 419). And this fact enables the account to include the same features about narratives per se that we have just adduced, namely the distinction between narrative and ‘real’ time and the possibility of retrospective signi-ficance. Consider the following stories from the previous list: (6) The king died and then, after a very very very long time, the queen died of laughter. (7) The king died, the king died, the king died, the king died, and then the queen died of boredom. (8) The king died after the queen had died of grief. (9) The king died violently soon after the queen had been poisoned. These stories are sufficient for triggering at least some level of suspense. How is that? Partly it is to do with the waiting occasioned in (6) and (7). But mostly it is to do with the facts that in all the cases (a) recognition of the event depicted first does not tell you what will happen next nor what has just happened before, and (b) recognition that something has gone before or will come after is likely to be triggered by the phrases that follow the depiction of the first event. Recognition of the phenomena first depicted is sufficient for inducing uncertainty regarding what comes next, even when the uncertainty itself goes unrecognized. And, on a wide (or ‘eliminativist’) notion of suspense, this is sufficient for its production (see Uidhir 2011). a sense of uncertainty here is automatic, quick, and occurs below the level of awareness. Two things are worth discussing briefly at this point. The first is whether the unconscious triggering of anticipation can plausibly be taken as a sign of comprehension; and the second is whether, as I have construed them, there really is any longer a distinction to be drawn between anticipation and suspense. As regards the first, the question is provoked by the claim that recognition of one set of phenomena, that which is first depicted, is sufficient for inducing uncertainty regarding what comes next, even when the uncertainty itself goes unrecognized. We may worry that any mental process having an unconscious component cannot be taken to be a sign of comprehension because comprehension must be a process of which we are aware. The worry points out that we need a way to make explicit how one agent, A, tells when another agent, B, comprehends, based only upon observing behavior of which even B seems to be unaware. But that is ready to hand (see Hamilton

Narrative per se and Narratability | 71

2007, 75–78). The test, so to speak, is simply for A to wait and see what B subsequently does. Were B to react in a way that is consistent with comprehension, B would eventually demonstrate that by further behavior – some of which may be a correct verbal recounting of the narrative – or B would not do so. And that is how we tell; we wait and see. To those who worry about how B tells whether she has comprehended, I only point out that this involves a standard question in traditional epistemology, whether one must know that one knows in order to know at all. And I do not think so. But that discussion would take us much further afield than I have space for in this essay. As regards the second, the concern would seem to be that, given how I have worked out a story about what propels a reader or spectator to describe what she encounters when utilizing a narrative structure (whether she is aware of that structure or not), there would no longer appear to be a distinction to be drawn between anticipation, which underlies grasp of any narrative, and suspense, which would seem to be a feature only of some narratives. But that seems counterintuitive to some people. I do not think it so counterintuitive. However, I do see the problem. And I have no answer to it, at least not one that I regard as satisfactory. But, for reasons I will now adduce, I think the position that occasions the problem has real benefits independently of this problem and, so, should be adopted at least provisionally.

6 Three nice payoffs If we take suspense to be a collection of anticipatory states and we emphasize the element that sub-personal awareness plays in setting up these anticipatory states, we are able to clear up several key questions concerning whether uncertainty is necessary for suspense, whether knowledge of a narrative’s outcome makes uncertainty impossible, and whether we feel suspense on repeated exposures to the same narrative. These are the three items that have been thought to generate the so-called paradox of suspense (see Varroll 1996). For example, the view makes room for the thought that while uncertainty may be necessary for suspense, the awareness one is uncertain need not be. In this way, it also allows for a compati-bilist approach to the apparent paradox. I do not pursue this thought in any detail here. I am content to point out that Mag Uidhir (2011) has shown how this more relaxed analysis of suspense, by not taking it to be a distinct mental state but rather something more like a collection of anticipatory states, succeeds precisely in dissolving the paradox in a way that should satisfy

72 | James R. Hamilton those who initially thought the paradox was a genuine and troubling feature of narratives. Second, the fact that one may anticipate an outcome without knowing ahead of time that it is an outcome will be important to first and perhaps many initial encounters with a narrative. Upon repeated encounters, as one might expect, one’s knowledge of the outcome will ordinarily kick in and interfere with anticipatory setups. However, the fact that our initiating states are so frequently induced sub-doxastically means they can also be manipulated in much the same way, and even blocked. So, for example, Richard Gerrig (1989) reports on experiments in which suspense is maintained because a reader’s certain knowledge of outcome is actually undermined by a kind of conceptual priming effect generated by bits of information that reader does not realize are defeaters of what she knows. Gerrig does not use the term ‘conceptual priming effect’ in his essay; but the experiments he describes can be re-described in these terms without loss of content. And, as a second example, Yun-Gyung Cheong and R. Michael Young (2008) show that by determining what story elements to tell and which to hide, they are able to manipulate the level of suspense created by computer generated narratives by affecting readers’ plan-based reasoning, much of which takes place below the level of consciousness. A third payoff of this analysis, even though it has been offered as an explanation of explicitly verbal narratives, is that the same kinds of anticipatory states that underwrite or trigger the reactions of readers also underwrite the reactions of theatrical spectators. Notice that the analysis has focused upon the reactions of readers rather than the intentions of authors. As such, it is consistent with at least this much of the phenomena Goldie (2003, 310) has explored by employing the notion of narratability: Insofar as a thinker can take on and think through a sequence of events or images present to mind and “is able to grasp coherence and meaningfulness”, then the object of the thinker’s reflection is “narratable”.7 Goldie (2003, 310) also holds that, in the case of an individual reflecting on his own life, that the sequence is narratable “does not imply the existence of a narrator as such, or of an audience”. This is part of what I have been looking for in this paper.

|| 7 Goldie (2003, 310) adds that she must also be able “to evaluate and respond emotionally to what happened”. He does so because he is concerned with stories we tell about our lives that we use to understand ourselves and guide our decisions. A much weaker notion of narratability is to be found if one leaves it, as I have in the text, with an agent’s capacity to grasp sense and meaningfulness in a sequence. And, I believe, this is a desirable feature of my account. (See Hohwy 2007 and Gallagher 2000).

Narrative per se and Narratability | 73

7 Concluding remarks The first two are important results of the analysis I have provided, but as my discussion of the third payoff suggests, its importance here is that in that same examples we can also see, at least partially, how to answer the questions that are the ultimate focus of this paper. What I think we can now do is answer this: What is the minimum necessary that must be presented to someone for her to be positioned to tell a story? What triggers her ability to tell a story, to narrate, is similar to the sort of thing that enables her later to explain her reaction to the diver, the performer dropping a penny, or to the image of a character that a performer displays or the way she displays her character’s verbal behavior. Consider just the case in which a spectator grasps that her recognition of one event depicted first does not tell her what will happen next, nor what has just happened before. As such, even assuming she knows nothing more than that something else will happen, she is enabled by her own recognition of the event depicted in its context to grasp its coherence with other depicted events and its sense in relation to them.8 Thereby she is able to tell a story. The enabling really is enough for that.

References Carroll, Noël. “The Paradox of Suspense.” Suspense: Conceptualizations, Theoretical Analyses, and Empirical Explorations. Eds. Peter Vorderer, Hans Jürgen Wulf and Mike Friederichson. London, New York: Routledge, 1996, 71–91. Carroll, Noël. “On the Narrative Connection.” Beyond Aesthetics: Philosophical Essays Cambridge: Cambridge University Press, 2001, 118–132. Cheong, Yun-Gyung and R. Michael Young. “Narrative Generation for Suspense: Modeling and Evaluation.” Interactive Storytelling. First Joint International Conference on Interactive Digital Storytelling, ICIDS 2008. Erfurt, Germany, November 2008, Proceedings. Eds. Ulrike Spierling and Nicolas Szilas. Berlin, Heidelberg: Springer, 2008, 144–155. Currie, Gregory. Narratives & Narrators. A Philosophy of Stories. Oxford: Oxford University Press, 2010. Currie, Gregory and Jon Juriedini. “Narrative and Coherence.” Mind & Language 19.4 (2004): 409–427. Fludernik, Monika. Towards a “Natural” Narratology. London, New York: Routledge, 1996. Fludernik, Monika. “Narrative and Drama.” Theorizing Narrativity. Eds. John Pier and José Ángel García Landa. Berlin, New York: De Gruyter, 2008, 355–384. || 8 This will be so, even if she is not necessarily prepared to invest it with emotional import, as Goldie requires.

74 | James R. Hamilton Forster, Edward Morgan. Aspects of the Novel. New York: Harcourt and Brace, 1927. Gallagher, Shaun. “Philosophical Conceptions of the Self: Implications for Cognitive Science.” Trends in Cognitive Science 4.1 (2000): 14–21. Genette, Gérard. “Fictional Narrative, Factual Narrative.” Poetics Today 11.4 (1990): 755–774. Gerrig, Richard J. “Suspense in the absence of uncertainty.” Journal of Memory and Language 28.6 (1989): 633–648. Goldie, Peter. “One’s Remembered Past: Narrative Thinking, Emotion, and the External Perspective.” Philosophical Papers 32.3 (2003): 301–319. Goldie, Peter. “Book Review of Gregory Currie’s Narratives and Narrators: A Philosophy of Stories.” British Journal of Aesthetics 51.3 (2011): 335–338. Hamilton, James R. The Art of Theater. Malden, Oxford, Carlton: Wiley-Blackwell, 2007. Hohwy, Jakob. “The Sense of Self in the Phenomenology of Agency and Perception.” Pysche 13.1 (2007): 1–20. Kreiswirth, Martin. “Merely Telling Stories? Narrative and Knowledge in the Human Sciences.” Poetics Today 21.2 (2000): 293–318. Lamarque, Peter. “On Not Expecting Too Much from Narrative.” Mind and Language 19.4 (2004): 393–408. Lewis, David. “Elusive Knowledge.” Australasian Journal of Philosophy 74.4 (1996): 549–567. Uidhir, Christy Mag. “An Eliminativist Theory of Suspense.” Philosophy & Literature 35.1 (2011): 120–133. Nünning, Ansgar and Roy Sommer. “Diegetic and Mimetic Narrativity: Some further Steps towards a Narratology of Drama” Theorizing Narrativity. Eds. John Pier and José Ángel García Landa. Berlin, New York: De Gruyter, 2008, 331–354. Nünning, Ansgar and Roy Sommer. “The Performative Power of Narrative in Drama: on the forms and function of dramatic storytelling in Shakespeare’s plays.” Current Trends in Narratology. Ed. Greta Olson. Berlin, New York: De Gruyter, 2011, 200–231. Prince, Gerald. “Narratives, Narrativeness, Narrativity, and Narratability.” Theorizing Narrativity. Eds. John Pier and José Ángel García Landa. Berlin, New York: De Gruyter, 2008, 19–28. Richardson, Brian. “Time is Out of Joint: Narrative Models and the Temporality of the Drama.” Poetics Today 8.2 (1987): 299–309. Richardson, Brian. “Point of View in Drama: Diegetic Monologue, Unreliable Narrators, and the Author’s Voice on Stage.” Comparative Drama 22.3 (1988): 193–214. Richardson, Brian. “Endings in Drama and Performance.” Current Trends in Narratology. Ed. Greta Olson. Berlin, New York: De Gruyter, 2011, 181–199. Ryan, Marie-Laure. “On the Theoretical Foundations of Transmedial Narratology.” Narratology beyond Literary Criticism: Mediality, Disciplinarity. Ed. Jan Christoph Meister in collaboration with Tom Kindt and Wilhelm Schernus. Berlin, New York: De Gruyter, 2005, 1–23. Velleman, J. David “Narrative Explanation.” The Philosophical Review 112 (2003): 1–25.

| Section 2 – Literature and Film

Tobias Klauk and Tilmann Köppe

Distance in Fiction

Many twentieth-century writers of fiction were interested in a particular distinction between two different ways of presenting events in a story: The author may either tell what is going on in the storyworld, or she may show it. While telling amounts to a summarizing description of the events, showing involves a more detailed account of what is going on in the storyworld. In particular, showing is supposed to involve a rendering of an observer’s sense experiences of the scene. Rather than merely learning that something has happened, the reader gets an impression of what is to be seen, heard, smelled etc. on the scene. Showing, that is to say, puts the reader in the shoes of a participant in the event. In his study The English Novel, Ford Madox Ford explains: If I say ‘The wicked Mr. Blank shot nice Blanche’s dear cat!’ that is telling. If I say: ‘Blank raised his rifle and aimed it at the quivering, black-burdened topmost bough of the cherrytree. After the report a spattered bunch of scarlet and black quivering dropped from branch to branch to pancake itself on the orchard grass!’ that is rather bad rendering, but still rendering. (Ford 1983, 122)1

Of his pair of example sentences, the first one is considerably shorter and gives a summarizing description of what happened together with a rather sweeping evaluation of the participants, while the second one supplies a more detailed account of the incident itself. Here, we are given some striking details, not only of Blank’s actions, but also of how the cat in the tree looks from beneath the tree, how the shot sounds, and how the cat tumbles to the ground. Similar examples of a juxtaposition of telling and showing abound in popular literature on how to write fiction. Ford’s claim that the novelist “has to render and not to tell” (Ford 1983, 122) does not seem to have lost much of its force for authors of fiction today. In contrast, current narratology, thanks to its prevalent self-understanding as the systematic study of narrative rather than a guide to successful story writing, is not concerned with recommending showing or telling; the distinction itself, however, has found its way into many introductory textbooks to the field. To give but one further example, Rimmon-Kenan (2002, 109) in her Narrative Fiction, contrasts “John was angry with his wife” (telling)

|| 1 Ford speaks of rendering rather than showing, but from his example, it becomes clear that it is the distinction between telling and showing that he is up to. A look into the history of the subject and dozens of references can be found in Friedman (1955).

78 | Tobias Klauk and Tilmann Köppe with “John looked at his wife, his eyebrows pursed, his lips contracted, his fists clenched. Then he got up, banged the door and left the house” (showing). So far, we have given but a vague exposition of the showing vs. telling distinction, and in doing so, we heavily relied on examples. This is the approach in the bulk of current narratological treatments of the phenomenon, i. e. the showing vs. telling distinction is mentioned, but it is hardly ever studied in detail, let alone explained.2 In this article, we shall try to take some first steps in this direction. We start by introducing some basics of current narratological accounts of the distinction. In doing so, we shall substantiate our claim that a fresh start is needed and lay bare some of the conditions of adequacy a successful theory of the phenomenon should meet. After this, we shall outline the contours of an alternative approach to the showing vs. telling distinction. Before we start, some preliminary remarks are in order: Firstly, a quick look at current narratological textbooks reveals that the showing vs. telling distinction has been taken up under many different labels. We have already encountered Ford Madox Ford’s terms telling vs. rendering; others speak of scene vs. summary, dramatic mode vs. narrative mode, mimesis vs. diegesis, or narratives with small distance (showing) vs. narratives with large distance (telling).3 We have decided to stick with Genette’s terms (small distance, i. e. showing, vs. large distance, i. e. telling) for the simple reason that Genette’s treatment of the issue is hugely influential and his terms seem to be widespread in narratology by now. Moreover, these fresh terms might prove useful for the purpose of distinguishing among the many different interpretations theoreticians have come up with in the history of the telling vs. showing distinction. But apart from that, of course, it does not matter much what terms one uses once one has made it clear what they designate. This will be our main task in what follows. Secondly, and more importantly, there are some things called distance in narratology which have nothing to do with the distinction at hand. Thus, the term distance has been taken to cover “the similarities and differences between any two agents involved in narrative communication” (Phelan 2005, 119), such as author, narrator, character, and audience (see also Eder 2008, Ch. 12). This is not the kind of distance we are interested in here. Thirdly, and still more importantly, before one sets out to offer an explication of a term, one needs to have a rough idea about the very phenomenon it is

|| 2 Notable recent exceptions are Linhares-Dias (2006); Mellmann (2010); Johansson (2012). Their approaches differ considerably from our approach, and our dealing with them has to await another occasion. 3 For references, see e. g. Rabinowitz (2005); Chatman (1978); Stanzel (2008, 190–194).

Distance in Fiction | 79

supposed to cover. A quick look into narratological literature reveals that it is not clear whether the telling vs. showing distinction captures a single phenomenon or several different ones. Although narratologists seem to agree on the classification of core examples, they do not agree about what makes the respective passages of texts examples in the first place.4 The phenomena that are taken to lie at the heart of the telling vs. showing distinction include: (1) the relations of a fictional narrator to what is told, including his or her spatial, temporal and epistemic position as well as his or her motives or ‘self-awareness’; hence a narrator who is close to the events, eager to report and hence not ‘self-aware’ establishes the showing mode, while a more distant, epistemically privileged narrator who is able to reflect on his business establishes the telling mode (see Toolan 2001, 134); (2) the very ‘perceptibility’ or ‘presence’ of a fictional narrator, i. e. the question whether, or how strong, the passage of text invokes the impression that one is told about the events (see Chatman 1978, 32; Stanzel 2008, 192); (3) the question whether the passage of text explicitly mentions characters’ traits, themes, meanings, morals or evaluations (telling), or leaves them implicit, i. e. to be detected by the reader (showing) (see Friedman 1955, 1169–1170; Rimmon-Kenan 2002, 108; see also Bal 2009, 75–76); (4) grammatical or stylistic features, such as different modes of speech representation (direct speech, indirect speech, free indirect speech, etc.), marked commentary or evaluations, or combinations of these, which influence the ‘speed’ of the narration, or which are taken to be textual clues for (1), (2) or (3). Many narratological treatments indeed present combinations of these four phenomena;5 Genette, whose views will be considered in some detail below, is an apt example of this. We believe, however, that while it is possible to inquire into any of these distinctions and to investigate passages of texts focusing on any of these aspects, the label narrative distance should be reserved for a phenomenon || 4 In other words: They seem to agree on the extension of the terms, but they do not agree on the intensions of the terms. For the following survey, compare also Klauk and Köppe (2014) and (2014a). 5 In his article Showing vs. Telling in The Routledge Encyclopedia of Narrative Theory, Peter Rabinowitz (2005) claims that there are further distinctions “[b]uried beneath this apparently straightforward duality”, such as objectivity vs. partiality or the direction of the implied reader’s attention towards either the story or the storyteller. We shall not attempt at a resurrection of these further distinctions.

80 | Tobias Klauk and Tilmann Köppe that is not yet to be found on the list. What is central to a narrative with small distance is that the reader gets an impression of what is to be seen, heard, smelled etc. on the scene. Showing, that is to say, somehow puts the reader in the shoes of a participant in the event. The telling vs. showing distinction, then, centrally involves reader-responses. It is, at bottom, a reception-phenomenon. This understanding of the phenomenon is not revisionary. There are many who seem to have understood Genette and his predecessors in precisely this way. Thus, for instance, the German narratologists Matías Martínez and Michael Scheffel (1999, 47) explain that from passages of small distance, the reader gets the “[illusion of the immediate presence of the narrated (our translation)]”6, or the “[impression of an immediate presence (our translation)]” of the narrated (Martínez and Scheffel 1999, 50).7 Similarly, Rimmon-Kenan (2002, 109) claims that, in the showing mode, “narrative texts create the illusion of mimesis”. According to these authors, then, small narrative distance is primarily about reader-responses. We believe that this understanding of the distinction can be traced back to antiquity (under the heading mimesis vs. diegesis), but will not argue the point here. The reader-response approach to narrative distance has the advantage that it does not render the distinction superfluous. In contrast, the accounts listed above run the risk of trying to establish a distinction that has in fact already been made, albeit in different terms.8 We do believe that the reader-response approach to narrative distance even identifies some kind of common denominator to at least some of the current theories of narrative distance. Again, we will not argue the point here.9 As any account of narrative distance should be able to dodge typical problems in earlier accounts, we now discuss such typical problems. We concentrate on Genette (1980), and leave it to the reader to judge if and to what degree other

|| 6 German original: “[D]ie Illusion einer unmittelbaren Nähe zum erzählten Geschehen.” Very similarly, Franz K. Stanzel (2008, 192) claims that ‘mimesis’, i. e. small distance, is connected to the intention [to evoke the illusion that the narrated is to be perceived ‘in actu’ (our translation)] by the reader. German original: “[D]ie Illusion zu erwecken, das Erzählte sei gleichsam ‘in actu’ wahrzunehmen.” 7 German original: “Der Eindruck einer unmittelbaren Präsenz.” 8 Genette (1980, 166) claims that concerning what he calls the narrative of events, small distance “is simply a product of features that do not belong to it in its own right, and so we have no reason to linger over it.” On his distinction between ‘narrative of events’ and ‘narrative of words’, see our comments below. See also Rimmon-Kenan (2002, 109). 9 But see Klauk and Köppe (2014).

Distance in Fiction | 81

accounts suffer from the same problems. In any case, analyzing Genette’s approach will offer us important insights for our own theory. Genette (1980, 162) explains that “the narrative can furnish the reader with more or fewer details, and in a more or less direct way, and can thus seem (to adopt a common and convenient spatial metaphor, which is not to be taken literally) to keep at a greater or lesser distance from what it tells”. Genette explicitly claims to mean by ‘distance’ James’ (and Lubbock’s) telling vs. showing distinction, which he takes to be the same as Plato’s distinction between pure narrative (or diegesis) and mimesis (see Genette 1980, 162–163).10 His interpretation of Plato’s distinction leads Genette (1980, 163–164) to distinguish between narrative of words and narrative of events, the idea being that the showing mode actually can only be established in the former case. Readers, so the argument goes, can be directly presented with the spoken words of a character, but they cannot be directly presented with any other (non-verbal) event of the story. The rationale for this distinction seems to be that a character’s speech or thought might be given in its exact wording through direct speech and interior monologue, while this (quasi-) quotation is not available once events other than utterances or thoughts are being narrated. When asking how mimesis is created, Genette therefore gives different answers for the two cases. For narrative of words, he distinguishes direct speech from free indirect discourse and narrativised speech (see Genette 1980, 171–173). Direct speech gives us a character’s words or thoughts in the showing mode, while every other narrative of words is mediated and, therefore, in the telling mode. For narrative of events, on the other hand, this way of understanding the distinction is not available. Instead, Genette here refers to the two characteristics he finds in Plato’s conception, which he understands as the relative presence or absence of a narrator, and the quantity of information given (see Genette 1980, 166). Now, there are several problems with Genette’s account. Firstly, and most importantly, Genette at no point distinguishes clearly between a description of the phenomenon and the narrative means employed to achieve the distance effect. For example, the two narrative means important for narrative of events

|| 10 We are here concerned with Genette’s take on the distinction in his Narrative Discourse (1980). In his Narrative Discourse Revisited (Genette 1988), he repeats most of his earlier analyses with the clarification that the mimesis vs. diegesis distinction is equivalent to the narrative vs. dialogue distinction (Genette 1988, 45), which means, as a consequence, that the reader does not enter the picture. Genette’s historical claims are dubious at best, but we cannot argue the point here.

82 | Tobias Klauk and Tilmann Köppe also feature in the initial description of the phenomenon given in the quote above. Secondly, the distinction between narrative of words and narrative of events is problematic. Even if a narrative quotes a character’s speech, readers are of course not directly presented with the character’s utterance. What readers are actually presented with are printed words on a page, not the speech of a person (including a particular tone of voice, intonation, talking speed, pauses, and circumstances of the utterance, to name but a few indispensable properties of actual utterances). By quoting a character’s utterance, the author can invite you to imagine that these are the very sentences the character actually used in his or her utterance. The words on the page, then, do not literally show readers anything, irrespective of the words being about a speech act or any other kind of event. Also, there are clear examples in which the words a character speaks are not the words viewers are supposed to imagine. We are not supposed to imagine Othello speaking in the beautiful verse only a brilliant poet could produce (see Walton 1990, 174–183). Thirdly, quantity of information and presence (or absence) of a narrator are neither the only narrative means to achieve effects of distance, nor can they alone characterize the phenomenon. For one, the sheer amount of narrative information can hardly be decisive. Imagine a very detailed description of developments in the foreign trade volume: Would that cause readers to have the impression that they are directly presented with anything? We do not think so. A high quantity of information is not sufficient for small distance. But neither is it necessary: It is surely possible to evoke the effect by providing striking details of an event rather than by giving lengthy descriptions. The beginning of Helmut Krausser’s short story Spielgeld is a case in point: Berlin, Hotel Interconti, Schachturnier, dritter Tag, dritte Runde, August, Spätnachmittag. Im größten Saal des glitzernden Bronzentempels ticken 300 Uhren. (Krausser 1996, 27) [Berlin, hotel Interconti, chess tournament, third day, third round, August, late afternoon. In the largest hall of the sparkling bronze temple 300 clocks are ticking. (Our translation)]

Not only for someone acquainted with a setting such as this, Krausser’s brief description might count as an instance of showing due to the careful selection of information (rather than its sheer mass). Thus, what really is at stake is, appar-

Distance in Fiction | 83

ently, not the amount of information, but a certain type of information such that, as Lubbock puts it, the scene is placed before us (see Lubbock 1954, 65).11 Genette’s second criterion, the narrator’s presence, is also treacherous. Consider the example sentences by Rimmon-Kenan (2002, 109), quoted above: “John was angry with his wife” vs. “John looked at his wife, his eyebrows pursed, his lips contracted, his fists clenched. Then he got up, banged the door and left the house.” The two passages do not differ with respect to the presence or absence of a narrator, but can still be said (and have been said) to show the difference between telling and showing. The same is true of Ford Madox Ford’s example quoted above. Also, it is quite easy to find narratives with small distance which feature a clearly discernible fictional narrator. An example is Peter Weiß’ Der Schatten des Körpers des Kutschers, which opens thus: Durch die halboffene Tür sehe ich den lehmigen, aufgestampften Weg und die morschen Bretter um den Schweinekofen. Der Rüssel des Schweines schnuppert in der breiten Fuge wenn er nicht schnaufend und grunzend im Schlamm wühlt. Außerdem sehe ich noch ein Stück Hauswand, mit zersprungenem, teilweise abgebröckeltem gelblichem Putz, ein paar Pfähle, mit Querstangen für die Wäscheleinen, und dahinter, bis zum Horizont, feuchte, schwarze Ackererde. (Weiß 1964, 7) [Through the half opened door I see the clayey, stamped path, and the rotten planks, and the hog house. The pig snout sniffs the wide gap, if it doesn’t dig into the mud, snuffling and grunting. Also, I see part of the house wall, with cracked, partly flaking finery, a few posts, with crossbars for the clothes lines, and behind them, to the horizon, moist, black soil. (Our translation)]

As these examples illustrate, the presence or absence of a narrator simply does not characterize the phenomenon we are interested in. Hence, passages of text can have small distance (respective be in the showing mode) regardless of the presence or absence of a narrator. But we must be careful here, for things are more complicated. Why do so many different accounts of the showing vs. telling distinction emphasize the role of a fictional narrator? Answering this question allows us to introduce another important distinction. The very phrase ‘telling mode’ could be taken to point to someone who || 11 Maybe this is what Hermann Hesse had in mind when he said about Goethe’s prose that it seems to contain a wealth of descriptive detail in very few remarks: “Der Leser erinnert sich an Szenen, an Personen, Begegnungen, Gespräche, und wo immer er nachschlägt und nachprüft, findet er, was in seinem Gedächtnis breit und detailliert dastand, präzise und sparsamst ausgedrückt” (Hesse 1975, 178). [The reader remembers scenes, persons, encounters, conversations, and wherever he looks and checks, he finds expressed precisely and most economically what stood wide and detailed in his memory” (our translation)].

84 | Tobias Klauk and Tilmann Köppe tells you what is going on in the fiction. Hence, one may think that the distinction captures whether readers either imagine experiencing something or being told about something. Unfortunately, this attempt at a straightforward definition easily leads to confusion and it seems to be based on one as well. We have already seen that a text with a narrator is not the opposite of a text that asks readers to imagine experiencing something. Rather, talk of ‘either experiencing something or being told about something’ mixes up two distinctions: The first one distinguishes texts with a narrator from texts without (or with a hidden or background-narrator). Let us call this distinction telling vs. not telling. The second one distinguishes texts that give readers the impression of perceiving something from those that do not. Let us call this distinction small distance vs. large distance. These two distinctions give rise to four different cases (small distance with telling, small distance without telling, large distance with telling and large distance without telling).12 It is easy to find examples for all four cases. In what follows, we are only interested in the distinction between small distance and large distance. Note that this is no longer Genette’s distinction. His distinction between small and large distance is, if we read him correctly, a prominent case of the mixing up of the distinction we are interested in with the telling vs. not telling distinction. Moreover, rather than pursuing the question of the narrative means that might establish small distance, we shall confine ourselves to taking a closer look at what we take to be the distance-phenomenon itself. Before we do so, let us pause here briefly and summarize the main distinctions we have drawn so far. Firstly, we have distinguished between distinct phenomena that are labeled telling vs. showing in current narratology. Secondly, we have singled out one phenomenon that we take to be basic, namely the distinction between different impressions a reader may get upon reading the text. (We have called this reader-response approach to narrative distance.) Thirdly, we have argued that one needs to distinguish between the distancephenomenon itself (namely a particular impression on the side of the reader) on the one hand, and the narrative means that are prone to establish it on the other. Fourthly, we have detected another sub-division within the reader-response

|| 12 Note that in texts with telling there are two levels which can have small or large distance, namely the act of telling and the told. Thus, to return to the example from Peter Weiß, it is possible to imagine experiencing the narrator telling you about what he sees, and it is possible to imagine seeing what he tells you he sees. It is of course also possible to have small distance on one level and large distance on the other.

Distance in Fiction | 85

approach, namely the distinction between telling vs. not telling and small distance vs. large distance.13 With these distinctions in mind, we can now proceed with our analysis. As we have noticed, narrative distance is predicated of texts and, at the same time, we are dealing with a reception phenomenon. The easiest way to take account of both features of narrative distance consists in viewing it as a responsedependent property of texts. To say of a passage of text that it has small distance, then, amounts to saying that it provokes a particular response on the part of the reader.14 This brief formulation, of course, needs a lot of additional information. In particular, one needs to specify when (under what conditions), how (by what means and in what way) and in whom (depending on what further requirements on the side of the reader, such as particular sensibilities or a particular stock of knowledge) a response of the respective kind (which kind?) is provoked. But all beginnings are difficult, so rather than presenting a full-blown theory, we shall take some first steps in what we consider to be a fruitful direction. As noted by Genette and others, passages of small distance do not literally make us ‘see’ the narrated, nor do such passages make us literally perceive it in any other way. All we see are printed words on a page. Now, narratology is of course well aware of this fact. Passages of small distance are therefore often said to give us but the illusion of the presence of the narrated. However, since the term ‘illusion’ has, at least to our ears, a ring of misperception or false perception to it, we shall avoid it. Rather than making mistakes of some kind, in cases of small distance readers make-believe perceptually accessible aspects of the fictional world.15 || 13 It may also be worth mentioning here that we have found reasons to drop a distinction Genette and many of his followers in current narratology consider useful, namely the distinction between narrative of events and narrative of words, in the context of explaining narrative distance. This is not to say, of course, that the distinction does not prove useful in other contexts. 14 See LeBar (2005), for an introduction of response-dependence. 15 In at least some cases readers could be said to make believe perceiving, or imaginatively perceive, the narrated. According to Walton, all imagining might contain a (minimal) de se element (see Walton 1990, 28–29). Often, but not always, this de se element amounts to imagining “from the inside”: “a form of self-imagining characteristically described as imagining doing or experiencing something (or being a certain way), as opposed to imagining merely that one does or experiences something or possesses a certain property” (Walton 1990, 29). Imagining perceiving something, then, is a subtype of imagining experiencing something. There is some debate in the philosophy of mind and psychology on whether visual imagining (or imagining visual aspects of something) simply amounts to imagining seeing something (see Currie 1995).

86 | Tobias Klauk and Tilmann Köppe The concepts of make-believe, or imagination, are the cornerstone of many recent theories of fiction. Most notably, in his Mimesis as Make-believe, Kendall Walton (1990) maintains that works of fiction are props in games of makebelieve. Readers of fiction are invited to make believe what the works are about. Works of fiction, in other words, have the primary function of prescribing imaginings.16 The many details of Walton’s theory are not pertinent to our present purpose. However, we believe that his theory provides the framework for a theory of narrative distance. To begin with, the make-believe approach to fiction offers a systematic account of the relation between texts and reader responses. The connection is conceived of as a rule-governed and hence normative one (see Walton 1990, 39–40, passim). Readers attend to works of fiction adequately by imagining what the works prescribe them to imagine, or else inadequately by failing to do so. This very fact is captured by be notion of prescribed imaginings. To give but one rather trivial example, Dickens’s novel Great Expectations prescribes imagining that the protagonist’s name is Pip, and if we imagine that his name is Henry, we have surely done something wrong. Readers imagine all kinds of things when reading narrative fiction, of course. Usually, prescribed imaginings will form a subclass of these imaginings; they are the foundation from which we may proceed to embellish fictional worlds as we please. They tie our imaginings to the text, as it were, and they do so in a way that is not merely causal. The notion of a prescription is especially important in our context, for it allows us to be more specific concerning the link between narrative texts and reader responses. In general terms, the idea is this: Narrative texts which exhibit the distance phenomenon prescribe imaginings that have to do with perception. In what follows, we shall try to flesh out some of the details of a theory of distance that is based on this idea. Firstly, it only makes sense to ask whether a passage of text has the property of prescribing perceptual imaginings under the condition that the respective passage of text is about experientially accessible things at all. Passages of text with the potential for some degree of distance must be about sufficiently concrete matters that can be experienced through sensory perception. Consider the following passage:

|| We shall sidestep these questions here. In any case, it seems that they have not been addressed in previous accounts of narrative distance. 16 Walton speaks of authorized games of make-believe and mandated, or invited, or prescribed, imaginings.

Distance in Fiction | 87 A great variety of expressions of ordinary language can be translated into modal logic. Not only do possibly and necessarily fall into this group, but all the other expressions we have already encountered such as must, might, and have (in some cases). (Girle 2009, 4)

When asked whether this passage displays small or large distance, one might want to answer that it displays large distance, since it does not “create the illusion of mimesis” (Rimmon-Kenan 2002, 109). But to give this answer would only mean to promote conceptual unclarity. The passage is not about experientially graspable things in the first place, and therefore the question of how vivid, complete or good such things are presented to the reader does not even arise. It is not a passage to which the concept applies. For passages of text that feature fictional characters, a related question arises. Is it possible for readers to imagine perceiving the thoughts of a character? Literary texts regularly render the thoughts and feelings of characters. In cases of internally focalized texts, veridical perception and introspective description of the character’s impressions can be difficult to distinguish.17 Take as an example the opening passage of Paul Bowles’ The Sheltering Sky: He awoke, opened his eyes. The room meant very little to him; he was too deeply immersed in the non-being from which he had just come. If he had not the energy to ascertain his position in time and space, he also lacked the desire. He was somewhere, he had come back through vast regions from nowhere; there was the certitude of an infinite sadness at the core of his consciousness, but the sadness was reassuring because it alone was familiar. He needed no further consolation. In utter comfort, utter relaxation he lays absolutely still for a while, and then sank into one of the light, momentary sleeps that occur after a long profound one. Suddenly he opened his eyes again and looked at the watch on his wrist. It was purely a reflex action, for when he saw the time he was only confused. He sat up, gazed around the tawdry room, put his hand to his forehead, and sighing deeply, fell back onto the bed. But now he was awake; in another few seconds he knew where he was, he knew that the time was late afternoon, and that he had been sleeping since lunch. (Bowles 1969, 3)

The passage gives us the thoughts, feelings and perceptions of the protagonist. It further does so in an introspective way, from Port’s point of view. Should we count such cases of introspection in fiction as cases to which the concept of distance applies? On the one hand, readers never have such a strong feeling of a directly graspable reality as when a text directly gives the thoughts and feelings

|| 17 In terms of Genettean narratology: It can be difficult to decide whether we are dealing with an internally focalized passage of text that describes things from the point of view of a character, or rather with a zero-focalized passage of text that describes the point of view of the character, including what she perceives, feels, etc.

88 | Tobias Klauk and Tilmann Köppe of fictional characters. That is a strong reason for applying the concept of distance to such cases. On the other hand, thoughts and feelings are not perceptually accessible.18 Since central cases to which the notion of distance applies have to do with perception, this is a strong reason not to count introspective passages as having small or large distance. In this paper, we stick to central cases and therefore our proposals below do not apply to cases of introspection. If one wanted to include cases of introspection, our proposal could be adapted accordingly.19 Secondly, a further peculiarity of the distance phenomenon is that distance seems to be a gradual phenomenon. We can easily take Rimmon-Kenan’s (2002, 109) example sentences (quoted above) and add a passage (2) such that its distance lies between the two original passages (1) and (3): (1) “John looked at his wife, his eyebrows pursed, his lips contracted, his fists clenched. Then he got up, banged the door and left the house.” (2) “John tried to remain calm, but he could not do it with his wife directly in front of him. Equally unable to confront his wife directly, all he could do was to rush out of the house immediately.” (3) “John was angry with his wife.” The easiest way to accommodate the gradual nature of distance is to restrict talk about distance to comparisons between at least two passages. A passage of text should not be described as possessing small distance simpliciter, but as showing small distance in comparison to a second passage or group of passages. This yields the intuitively plausible result that the very same passage can have a relatively large distance in the context of (and in comparison to) one text while it has a relatively small distance in the context of another text. However, one surely cannot repeat the step demonstrated above ad libitum. We can find a || 18 Strictly speaking, that is not true, since thoughts and feelings are very well perceptually accessible. We often see that somebody feels sad, and we can tell, on a regular basis, what people think just by watching them. The Rimmon-Kenan example is a case in point. However, these are not the cases that are at stake here. Our question at this point is if passages like interior monologue, in which thoughts and feelings (instead of being perceived) are explicitly stated in a way that is meant to approximate how they occur in the character’s mind, should count as having small distance. 19 An account that included introspective passages would have to decide if such passages can differ in distance among themselves. It is not clear to us what the answer to that question should be. Depending on the answer, the definitions discussed below could be easily augmented (by speaking of ‘perceptually or by introspection’), or would have to include introspection as a kind of baseline of minimal distance.

Distance in Fiction | 89

fourth, maybe a fifth passage of text about John’s anger which ranks somewhere between the three passages – but then it gets exceedingly difficult to find further refinements of the scale. This shows that although distance is a gradual phenomenon, one should not hope for any number of fine-grained distinctions between passages of texts. It is time to combine the considerations above with our thoughts on fiction, which amount to something like the following preliminary definition of distance: (D1) A passage of text T of a narration N has smaller distance than a passage of text U of a narration M if, and only if, T prescribes more constraints on readers’ imaginings about perceptually accessible aspects of the fictional world than U. (D1) catches the gradual nature of distance by comparing passages of texts. By relying on the notion of a prescription, it is not only compatible with the best available theories of fiction, but captures both the intuition that distance is a property of texts and that distance has to do with reader responses that depend on textual features in a rule-governed way. Moreover, (D1) has certain advantages over earlier accounts of distance. Most importantly, it describes the concept of distance by staying neutral on the question of which narrative means can be used to produce small or large distance. This neutrality (as to the narrative means) is also one reason why (D1) demands constraints on readers’ imaginings instead of, say, claiming that the passage of text has smaller distance if it describes a scene in greater detail. The latter is hardly correct, as we have seen from the Krausser-example above. There is another reason why (D1) speaks of constraints, rather than, say, simply claiming that a passage of text has smaller distance if it prescribes more, or more striking, imaginings about perceptually accessible aspects of the fictional world than its respective contrasting passage. Presumably, any fictional narrative invites us to imagine what it is about in a vivid manner. There is nothing wrong with vividly imagining what Rimmon-Kenan’s (2002, 109) examplesentence “John was angry with his wife” is about. What is characteristic of small distance is that we are dealing with a more specific prescription. Passages of small distance thus add some particular constraints to the general invitation to vividly imagine what the narrative is about. Thus, the quoted example sentences featuring small distance (“John looked at his wife, his eyebrows pursed, his lips contracted, his fists clenched. Then he got up, banged the door and left the house” [Rimmon-Kenan 2002, 109]) constrain what we are to imagine about the look of John’s face and the sound of the door (this pertains to the contents of the

90 | Tobias Klauk and Tilmann Köppe imagination); indeed, we may even be invited to imagine seeing John’s face and hearing the door (this pertains to the manner of the imagination).20 Thus, the function of small distance may be compared to that of a road sign that sets up a speed limit: The road sign, as it were, takes for granted that people drive on the road; it specifies a particular aspect of the way in which this is to be done. Similarly, (D1) acknowledges that, generally, people tend to vividly imagine what a narrative is about; but this is not to deny that, sometimes, there is small distance on top of this. In spite of these assets, however, (D1) is not the last word on the matter. One problem is that (D1) cannot account for cases in which the difference in distance is not produced by a difference between two passages of text but by the context of those passages. Imagine a novel in which Angela meets a guy in chapter one, whom she then meets again in chapter three. The relevant passage from chapter three reads: ‘And there was Peter again in his handsomeness!’ The relevant passage from chapter one reads: ‘During the party, she noticed an attractive man. Later she found out that his name was Peter.’ Now imagine a second novel (or a revised version of the first). Chapter three stays exactly the same, but chapter one has changed significantly. The relevant passage now reads: ‘During the party, she noticed a guy whose delicate laughter and sparkling eyes immediately drew her towards him. With light footsteps he moved elegantly through the crowd. Later she found out that his name was Peter.’ If we compare the identical sentences from the third chapters across the two books, the second has smaller distance. The reason, however, does not reside in the passages themselves but in their context. Thus, (D1) needs a clause which clarifies that we are talking about passages of text in the context of a text. (D2) does the trick:

|| 20 Compare again note 15 above.

Distance in Fiction | 91

(D2) A passage of text T in the context of a narration N has smaller distance than a passage of text U in the context of a narration M if, and only if, T, in the context of N, prescribes more constraints on readers’ imaginings about perceptually accessible aspects of the fictional world than U in the context of M. Unfortunately, (D2) is still subject to counterexamples. Imagine two descriptions of a person’s nose, one of which goes into minute detail covering ten or more pages. Although such a passage constraints readers’ imaginings more than any short description of a nose, it seems that the distance of such a passage will be larger. Whether one is willing to accept this conclusion might depend on the exact nature of the two texts, but the details of the counterexample are less important than the mechanism behind it. Among other things, (D2) tries to map the intuition that distance has to do with imaginatively simulating ordinary perception. And while in most cases such imaginings simulate perception better if they are more precise, in certain cases the opposite happens – more detail leads to imaginings that are further removed from perception. We therefore need a further clause to immunize (D2) against such ‘alienation’ counterexamples: (D3) A passage of text T in the context of a narration N has smaller distance than a passage of text U in the context of a narration M if, and only if, either T prescribes readers to imagine events as in a normal perceptual situation while U does not, or T in the context of N prescribes more constraints on readers’ imaginings about perceptually accessible aspects of the fictional world than U in the context of M (given that T and U both prescribe readers to imagine events as in a normal perceptual situation). (D3) has the benefit of being immune to ‘alienation’ counterexamples such as the nose which is described for ten pages. In order to see the new feature of (D3) at work, compare the following two passages: ‘The horse stood on the meadow. Small points of light danced happily, went blurry until they were in focus. They had different colors; some were rather red or blue than white.’ ‘Peter looked through the telescope. Small points of light danced happily, went blurry until they were in focus. They had different colors; some were rather red or blue than white.’

92 | Tobias Klauk and Tilmann Köppe We take it that the second quote has smaller distance than the first, for we can easily make sense of the sentences as describing what Peter sees (through the telescope, that is). What is strange about the first quote is that things seem to differ from usual perceptual situations (indeed, it is hard to make sense of what we are to imagine at all). The second quote, in contrast, introduces the dancing lights within the context of ordinary perception (i. e. this is what one may see when looking through a telescope), and it is this difference that makes for the difference in distance of the two passages. One can keep on finding counterexamples and improvements to the already rather complex (D3) and its successors, but we would like to stress again that it is not our aim to arrive at the perfect definition of distance. Rather, (D1) to (D3) are presented here in order to highlight important features of the concept. We therefore proceed no further at this point. We close by taking a step back and answering a general concern one might have with our account. We explained above how our account draws heavily on Walton’s theory of fiction. Talk of prescribed imaginings has its place primarily with fictional texts – or so it might seem.21 And therefore our account seems to apply only to fictional texts. This, however, would not be a desirable constraint. The concept of distance applies to nonfictional texts just as well as to fictional texts, as long as perceptually accessible things are described by the text. Furthermore, it is not convincing to understand the fictional cases as the core cases, as the following example shows: Davis Creek is only a small trickle during most of the year and sometimes not even that. Originating at the foot of a high rock battlement known as Fiftymile Point, the stream flows just four miles across the pink sandstone slabs of southern Utah before surrendering its modest waters to Lake Powell, the giant reservoir that stretches one hundred ninety miles above Glen Canyon Dam. (Krakauer 1996, 88)

The distance of this passage can be compared to the distance of other passages of Krakauer’s (nonfiction) book. So how does our account deal with nonfictional texts? An obvious attempt to expand the considerations of the previous sections to nonfiction would be to substitute talk of prescribed imaginings by talk of prescribed beliefs. Unfortunately, this does not suffice. If we replace ‘readers’ imaginings’ by ‘readers’ beliefs’ in (D3), the result is utter nonsense. It is not prescribed beliefs that are restrained by the quote from Krakauer. In both the || 21 Note that Walton employs a very inclusive notion of fiction; see his 1990, 70–105, passim.

Distance in Fiction | 93

fictional and the nonfictional cases, it is imaginings rather than beliefs that are at the core of the distance-phenomenon. Surprisingly, then, we see no reason why (D3) cannot apply straightforwardly to nonfiction. In fact, we would argue that D3 applies to all texts that describe perceivable aspects of the fictional or actual world. Such texts prescribe imaginings concerning those aspects regardless of whether they are fictional or nonfictional. And importantly, they can be compared in regard to how much they restrain prescribed imaginings of perceptually accessible things.22

References Bal, Mieke. Narratology. Introduction to the Theory of Narrative. 3rd ed. Toronto/Buffalo/ London: University of Toronto Press, 2009. Bowles, Paul. The Sheltering Sky. London: Penguin, 1969 [1949]. Chatman, Seymour. Story and Discourse. Narrative Structure in Fiction and Film. Ithaca/London: Cornell UP, 1978. Currie, Gregory: “Visual Imagery as the Simulation of Vision.” Mind & Language 10 (1995): 25– 44. Eder, Jens. Die Figur im Film. Grundlagen der Figurenanalyse. Marburg: Schüren, 2008. Ford, Madox Ford. The English Novel. Manchester: Carcanet Press, 1983 [1930]. Friedman, Norman. “Point of View in Fiction. The Development of a Critical Concept.” PMLA 70 (1955): 1160–1184. Girle, Rod. Modal Logics and Philosophy. 2nd ed. Ithaca: McGill-Queen’s UP, 2009. Hesse, Hermann. “Wilhelm Meisters Lehrjahre” [1911]. Eine Literaturgeschichte in Rezensionen und Aufsätzen. Ed. Volker Michels. Frankfurt/Main: Suhrkamp, 1975, 159–183. Johansson, Christer. “Telling and Showing: A Semiotic Perspective.” Disputable Core Concepts of Narrative Theory. Eds. Christer Johansson and Göran Rossholm. Bern: Peter Lang, 2012, 147–182. Klauk, Tobias, and Tilmann Köppe. “On the Very Idea of the Telling vs. Showing Distinction.” Journal of Literary Semantics 43 (2014): 25–42. Klauk, Tobias, and Tilmann Köppe. “Telling vs. Showing.” Handbook of Narratology. 2nd ed. Ed. by Peter Hühn et al. Berlin/Boston: De Gruyter, 2014, 846–853. [2014a] Krakauer, Jon. Into the Wild. New York: Villard, 1996. Krausser, Helmut. “Spielgeld.” Das Liebesleben des Giacomo Müller. Erzählungen. Reinbek bei Hamburg: Rowohlt, 1996 [1989]: 27–41. LeBar, Mark. “Three Dogmas of Response-Dependence.” Philosophical Studies 123 (2005): 175–211. Linhares-Dias, Rui. How to Show Things with Words. A Study on Logic, Language and Literature. Berlin/New York: de Gruyter, 2006. || 22 Work on this article has been supported by the Courant Forschungszentrum Text-strukturen at Goettingen University.

94 | Tobias Klauk and Tilmann Köppe Lubbock, Percy. The Craft of Fiction. London: J. Cape, 1954 [1922]. Martínez, Matías, and Michael Scheffel. Einführung in die Erzähltheorie. München: Beck, 1999. Mellmann, Katja: “Voice and Perception. An Evolutionary Approach to the Basic Functions of Narrative.” Toward a Cognitive Theory of Narrative Acts. Ed. Frederick L. Aldama. Austin: University of Texas Press, 2010, 119–141. Phelan, James. “Distance.” The Routledge Encyclopedia of Narrative Theory. Eds. David Herman, Manfred Jahn and Marie-Laure Ryan. London/New York: Routledge, 2005, 119. Rabinowitz, Peter J. “Showing vs. Telling.” The Routledge Encyclopedia of Narrative Theory. Eds. David Herman, Manfred Jahn and Marie-Laure Ryan. London/New York: Routledge, 2005, 530–531. Rimmon-Kenan, Shlomith. Narrative Fiction. London/New York: Routledge, 2002 [1983]. Stanzel, Franz K. Theorie des Erzählens. 8th ed. Göttingen: Vandenhoeck & Ruprecht, 2008. Toolan, Michael. Narrative: A Critical Linguistic Introduction. London: Routledge, 2001 [1988]. Walton, Kendall L. Mimesis as Make-Believe. On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Weiß, Peter. Der Schatten des Körpers des Kutschers. Frankfurt/Main, 1964.

Sarah E. Worth

Narration, Representation, Memoir, Truth, and Lies How We Diminish the Art of Narrative with Simple Truths Narrative can only loosely be said to be representational in the strictest sense of the word. Writers of narrative,1 both fiction and nonfiction, must manipulate time, causation, emphasis, and perspective of events in order to craft an interesting story – one that is compelling, appealing, and interesting. Events are not the same as stories; stories are always interpreted, condensed, and told through a certain perspective that transform events into something that can be consumed and made sense of by readers or listeners. Events must be crafted into stories by narrators; thus my resistance to allow them to be, in the strictest sense, purely representational. Memoir incorporates aspects of narration, aims at what might be called personal historical accuracy, but also seeks to develop character and plot. At the same time, memoirs are supposed to accurately represent a life. Presumably, memoirs are supposed to be true. However, I question whether truth is a quality that we can attribute to memoir at all, but if it is, in what qualified sense. The question of truth attribution is especially relevant in cases of fabricated memoir, where some or all aspects of the account have been made up or exaggerated. When cases of fabricated memoir come to public attention, they are often criticized widely and harshly, but notably on the basis of truth failure rather than on literary aspects. Moreover, I suggest that memoir, taken as a literary genre, should be considered as something more significant than just records of incidentally true events. What counts within the genre of memoir might begin with true events, but those events are crafted into narratives, and in the process of turning events into stories, something altogether different is created. Condemning a memoir merely because it does not line up exactly with that which can be documented, indicates standards of criticism that focus on reference status (and reference failure) rather than the literary aspects that I take to be characteristic of the genre.

|| 1 Although narrative covers a wide array of kinds of storytelling, I am focused here on the various forms of literary narration including fiction, literary nonfiction, memoir, and autobiography.

96 | Sarah E. Worth In this paper, I will look at the genre of memoir as a particular form of narrating events and life stories in order to address the question of how memoir might be said to be meaningfully true. I will look specifically at examples of memoir which have been found to be fabricated in some way so as to tease out the particular issues with truth and falsity, as well as the kind of expectations the reading public has with the genre. I will frame my discussion of truth and memoir with the following three suggestions; first, I take it that memoir is a genre that deserves careful consideration by philosophers. Its special combination of narration and appeal to truth should give it a space of its own for philosophical inquiry. Memoir is a distinctive genre that cannot be aligned either with fiction or nonfiction despite the fact that it shares qualities with both of those literary genres. It is neither simply nonfiction nor fiction, truth nor lie. Its features lie in a vague literary successtype distinction that should have to do more directly with the quality of the prose and the effectiveness of the story. Secondly, I take the example of memoir, and fabricated memoir in particular, to raise specific philosophical problems for theorists who approach the ways in which we engage with narrative. In general, philosophers talk about the ways in which narrative prose influences our imagination, and how those imaginings in turn influence our beliefs. Memoir and fabricated memoir force us to address the ways in which these narratives can be called representational and to question the ways in which we ascribe truth. Simplistic or reductionistic versions of the correspondence theory of truth are inadequate to characterize this distinctive genre. Thirdly, I believe that narration is an interpreter between what we take to be truth and its relationship to events. This leads me to suggest an alternative paradigm with which to describe the kind of success a memoir might achieve. How one characterizes narration and which notion of truth one ascribes to, will dictate the ways in which either can be used and applied meaningfully. For instance, Hayden White (1987) has argued that narrative is always a construction that alters the events represented, or the so-called ‘pure facts’. He (White 1987, 47) suggests that there is a “transition from the level of fact or events in the discourse to that narrative [and] this transition is effected by a displacement of the facts onto the ground of literary figurations or, what amounts to the same thing, the projection onto the facts of the plot structure”. Seen in this way, the fact that narrative elements have been applied to ‘pure facts’ or unnarrated events will result in a narrative which must always be an imaginative construct which differs significantly from ‘what happened’. Louis Mink (1978, 145) also defends this constructionist view of narrative and says that no narrative can “defend its claim to truth”. White and Mink both subscribe to a form of positivism which suggests that we cannot have any pure perceptions of given facts. We

Narration, Representation, Memoir, Truth, and Lies | 97

approach the world with concepts, categories, and various folk psychological explanations of why the world is the way it is. This leads them to question the epistemic legitimacy of all narratives. This seems to me a problem, however, in that they reject narrativity despite the fact that it exhibits a characteristic that is found in all kinds of knowledge (scientific, social scientific, historical, etc.) and is likely found in all forms of explanation. On this account there can be no truth beyond pure facts, and pure facts are impossible for us to perceive. I cannot support this version of positivism as my account of narrativity simply because of the impossibility of any means of truth being attainable. I take it that we do not experience the past as having an implicit narrative structure (if we did, all historians would need to do is retrieve the narrative) but rather we describe the past, as well as our own experiences, imposing the structure of narrative on those experiences. Thus, I do not take the extreme position that all narrative is fiction and/or metaphor that White might be characterized as having, but rather I acknowledge the fact that narrative is a construction of the sort which can be said to be accurate or faithful or have verisimilitude, but it is not something that can be said to have correspondence truth attributed to it in the way that individual sentences or utterances can be said to be true. Whereas individual propositions can have correspondence truth, it is a category mistake to say that correspondence truth can be meaningfully applied to narratives as a whole. Historical narratives are largely constructed explanations of events in the past as they are understood by the teller. Narratives must always be constrained by limited perspective, the choice of what is left out and what to include, and the way in which one chooses to construct the elements of the story among others. It is for these reasons that I do not find memoir to be purely representational, nor do I find it to be necessarily placed in the realm of fiction as White believes. This is important as I will go on to discuss how the criticisms of fabricated memoir seem also to be misplaced because of the kind of construction I take it to be.

1 Issues with and examples of fabricated memoirs Memoir is a literary genre that has been socially and historically constructed. It is not a strict genre category with clear boundaries or an essential character, but, as with other literary genres, there are some literary conventions that remain constant throughout the narratives that we classify as memoir. Historian of memoir Ben Yagoda (2009, 1) defines memoir as a narrative “understood by

98 | Sarah E. Worth its author, its publisher, and its readers to be a factual account of the author’s life”. Memoir is often differentiated from autobiography in that memoir only covers a portion of a life, whereas autobiography usually covers an entire life. The exact designation of these genres is anything but static, and the functional designations of the two genres have reversed over the centuries. The memoirs I am concerned with are primarily contemporary ones which make claim to accurate representations of life events, but have failed in some significant way. I will consider theories of fiction, which are distinct from the literary genre of fiction, in order to help explain how it is we approach narrative texts designated in a particular way. I take the theoretical approaches to fictionality to be related to my discussion of memoir in non-incidental ways. Memoir has some reliable, documentable reference to actual events. It subscribes, perhaps, to a narrow correspondence theory of truth. Many people think truth is merely that which can be documented or otherwise supported by testimony – merely that which can be said to have ‘really happened’. What is not documentable or verifiable is the insight gained by one who experiences a certain event. This insight is often one of the most rewarding aspects of a memoir for its readers. One’s perspective is not something one can prove with documentation. Without insight, memoirs would merely be lists of places, people, and events. These chronological lists are hardly gripping stories, devoid of the plot devices woven in for coherency and intrigue. Assuming that one could say that a memoir can be true, however, assumes that we can write about the experiences of our lives with some real veracity; that there might be some comprehensive standard of truth to which these particular written works aspire in order to be classified as memoir. In the last several years, an ever-increasing number of memoirists have been accused of fabrication in one form or another; Yagoda (2009, 246) claims that the last four decades will likely be remembered as the ‘golden age’ of memoir fraud. As the review process to ensure factual accuracy is not nearly as stringent as with traditional journalism, the lies, exaggerations, and fabrications created by memoirists often are found only after publication. Yagoda says that [t]he higher the stakes – if the books’ ‘facts’ are in support of charged political issues (as with slave narratives or Holocaust memoirs), if it makes unlikely or melodramatic representations […], if it hurls slurs at some recognizable individuals, or if it starts really selling […] – the quicker the faker will be outed. (Yagoda 2009, 246)

James Frey is one such notorious fabricator, largely because his book sold so well, in addition to the fact that he notoriously fooled American talk show hostess Oprah Winfrey. Frey published his memoir, A Million Little Pieces, about his

Narration, Representation, Memoir, Truth, and Lies | 99

struggle with drug addiction. As it turns out, Frey had originally tried to market the book to publishers as fiction but it was rejected numerous times because the story was too unbelievable (The Smoking Gun). When he said it was all true, Random House published it. The book ended up being the biggest selling nonfiction book of 2003 (King, 2006). Oprah put it on her ‘Book Club’ list, and the book sold even better. The fact checking website The Smoking Gun found several lies and exaggerations in the work and Frey was called out as a fake. In a public apology Frey issued in The New York Times he says that I didn’t initially think of what I was writing as nonfiction or fiction, memoir or autobiography. I wanted to use my experiences to tell my story about addiction and alcoholism, about recovery, about family and friends and faith and love, about redemption and hope. (Frey, 2006)

In subsequent interviews, Frey (The Smoking Gun) claims that what he wrote is a book; it was a story, not something that needed to be categorized with such scrutiny. Even later he (Mozes) says that he has trouble making sense of the distinction between fiction and nonfiction at all. James Frey, Binjamin Wilkomirski, Margaret Seltzer/Jones, Greg Mortensen,2 Rigoberta Menchu, J. T. LeRoy, Herman Rosenblat, and Misha Defonseca, among others, all wrote memoirs that were marketed and sold as true accounts. All of their books sold well and many were given high literary praise. They wrote accounts of drug addiction, surviving genocide and the holocaust, and planting the seeds of education in an undereducated country. All were stories of hope and inspiration, and all of the stories were about people who overcame what were thought to be insurmountable odds. However, all of the accounts also had factual inaccuracies or were found to be made up whole cloth, which forced their works to be questioned, and for some, to have their works pulled from the booksellers’ shelves entirely. Binjamin Wilkomirski, for example, fabricated a Holocaust memoir, not having been in the war at all. His memoir received multiple awards, including the National Jewish Book Award, the Prix Memoire de la Shoah, and the Jewish Quarterly literary prize. He was called upon to testify as a witness to the horrors of the Holocaust, he was a paid guest speaker at several child survivor groups of the Holocaust, and his stories were recorded as part of the oral histories in the Holocaust museum in Washington DC. As one critic (Maechler 2001, 281) said || 2 Greg Mortensen did not pen the book Three Cups of Tea (Relin and Mortensen, 2006) which is about him, rather David Oliver Relin wrote what is in the style of a memoir but Relin is really a biographer.

100 | Sarah E. Worth about his entirely fabricated narrative, “once the professed interrelationship between the first person narrator, the death camp story he narrates, and historical reality are proved palpably false, what was a masterpiece becomes kitsch”. The work was transformed from a heartbreaking, reliable, first person testimony to a vacuous lie representational of nothing that had in fact happened. The same words went from being inspirational to insulting to those who had actually lived through, suffered, and died. This particular kind of fabrication gives credence to Holocaust denial and so is exceptionally damaging, in ways other fabrications might not be. Wilkomirski was exposed as a fraud and all of his literary awards were revoked (Eskin 2002, 236). The books were pulled from the shelves and are no longer available.3 Historians, however, have insisted that Wilkomirski’s account is “consistent with the way young children’s memories for traumatic events are sensed, stored, and related; consistent with the struggle to make sense of a chaotic, horrific world during the Holocaust and in the aftermath” (Eskin 2002, 144). One historian at the Holocaust museum in DC said that even though the account is not one that really happened to the author, it is an account that is well-told and is representative of many children’s experiences in the concentration camps. She believes that in 50 years there will not be very many memoirs of the Holocaust left in circulation, but believes this should be one of the ones available because of its verisimilitude. She does not care that the experiences described did not, in fact, happen to the author, but says that the main thing is that “the story is well-told” (Eskin 2002, 219). This case is particularly difficult for some because of the topic, but it also highlights the change in attitude from one state to another (high literary praise to Kitch) while the text remains the same. With all of the bad press in each of these cases, critics rebuked authors over the simple veracity of merely what could be documented. Based on the number of law suits,4 books pulled from the shelves,5 scandals, awards revoked, and damning news articles, it seems that readers were less interested in the narrative elements of quality prose and the insight that comes with self-reflection than they were in the documented and documentable facts. Popular but simplistic notions of truth have become increasingly influential as many of these authors have been trudged through legal courts as well as the || 3 Although his memoir, Fragments, is no longer being sold, the entirety of the book is republished along with Stefan Maechler’s book, The Wilkomirski Affair: A Study in Biographical Truth. 4 James Frey had a class action law suit for $ 2.35 million filed against him, which he lost, and was required to refund $ 13.95 to readers who bought the book thinking it was true (Litte 2007). 5 Margaret Jones’ (real name Margaret Seltzer) Love and Consequences was taken out of publication as well as Wilkomirski’s and Rosenblat’s Holocaust memoirs.

Narration, Representation, Memoir, Truth, and Lies | 101

court of public opinion. I believe that the simplistic truth standards the public applies to memoir negatively influence the license that narrative should have to do exactly what it needs to do – narrate, describe, and incite imaginings. Memoirists in particular have been given a standard which in many ways is impossible to attain since now the only events that can be said to be documented or documentable can be said to be true. This is evidenced by the kinds of vicious public dismissals fabricated memoirs get. This does not excuse intentional lying or fabrication, but it seems to me that there should be some room within the genre of memoir for elaboration and insight that might not be as widely acceptable in some other more literal accounts of history, journalism, or even autobiography. The examples of fabricated memoir run the gamut in terms of the severity of lies, self-delusion or denial on the part of the author and the stakes of the topic. Margaret Jones made up her memoir Love and Consequences whole cloth. James Frey exaggerated. Herman Rosenblat, a real concentration camp survivor, fabricated a Holocaust love story to make something inspirational come out of the devastation of the Holocaust (he also fooled Oprah Winfrey with this book) (Rosenblat, 2009). Rigoberta Menchu made herself into a composite character. All of the atrocities she described as part of the Guatemalan Civil War happened (Menchu, 1984). They just did not all happen to her. Menchu received the Nobel Peace Prize in 1992 for her work promoting rights for indigenous Guatemalans. Conservative writer David Horowitz (Grandin 2010, 3) has actively tried to discredit Menchu calling her a “Marxist terrorist” and has suggested that her book and rights work constitutes “one of the greatest hoaxes of the 20th century”. All of these authors were discredited in ways that go well beyond just what they wrote in individual texts. It is as if a scientist invented data and might rightfully never be able to get another job. But the standard seems too high to me, the derision too much. My argument is that because memoir is not simply, directly, or literally representational, because memoir is inherently narrated, the standards necessarily need to be shifted away from mere event recitation. That which might be slavishly representational could be understood in a correspondence-type truth paradigm, but the constructed nature of narration, I argue, should necessarily shift the standards of a correspondence truth to a more value-laden notion of truth. Perhaps truth is not the appropriate value term to give memoir at all since expression, insight, and conveyed emotion are what we tend to value as readers of the genre. In other words, the kind of narrative elements found in memoir should be interpreted as primarily interpretive and only secondarily as being imitative or representational of actual events.

102 | Sarah E. Worth

2 Memoir and its relation to truth What is the difference between a true or accurate memoir and one with a few (or a lot of) mistakes or lies? How should a memoir be read which we know is entirely fabricated? If memory is largely unreliable, how might that influence the correspondence reliability of memoir? As I take it, memoir is closely related to the kinds of everyday stories that we tell about ourselves. These stories have a significant influence on personal identity construction, which in turn has an impact on the ways in which authors of memoir write about themselves.6 Those who talk about themselves as victims, often see themselves as victims. Those who talk about themselves as heroes tend to see themselves as heroes. When one writes formally for an audience, as with a memoir, it matters which one of these filters is used to convey the events as the result is skewed from the correspondence truth of what happened. Further, insights gleaned are dependent upon the facts that get narrated. If the events did not happen, then the subsequent insight must be called into question. As with fiction and nonfiction, however, textual engagement and intrigue are incited by plot devices as much as anything else. Creating well-developed characters, understanding motivations, and building expectations are essential to keep the readers engaged in any kind of narrative. It is for these reasons that I suggest that the ‘truth’ of the events should be evaluated only secondarily to effective prose. The imaginative force of a successful narrative should be dependent upon the ways in which events, causal relationships, and characters are portrayed either well or badly.7 I want to suggest that it should not be a matter of primary importance whether a narrative is true or false, fiction or nonfiction, documented fact or embellished fancy.

|| 6 I take my view of personal identity of selfhood construction from a number of theorists who have argued to this relation more fully elsewhere, namely Daniel Dennett, and Jerome Bruner. Dennett (1992) argues that the self is a “fictional construction” that he refers to as the “center of narrative gravity”. This self is a conglomeration of the stories we tell about ourselves. Bruner argues at length that “we organize our experience and our memory of human happenings mainly in the form of narrative – stories, excuses, myths, reason for doing and not doing, and so on”. Narratives, he claims “are a version of reality whose acceptability is governed by convention and ‘narrative necessity’ rather than by empirical verification and logical requiredness, although ironically we have no compunction about calling stories true or false” (Bruner 1991, 4–5). 7 See Worth (2008) where I argue that ‘well-told’ stories are more compelling in many ways, than ‘badly-told’ stories, primarily because of our tendency to learn to organize information from story form. Well-told stories are better organized in terms of cause and effect, and suspense, for instance, than badly-told stories.

Narration, Representation, Memoir, Truth, and Lies | 103

But it does matter since the ways in which works are judged, understood, and critiqued are dependent upon what readers believe to be true about a story. What is taken as true and what is taken as false in advance of reading matters deeply to the ways in which we make meaning of a narrative. True stories are ones in which the obstacles one overcomes are constrained by reality in ways in which fictional or fabricated obstacles are not. The problem with this is that there is no clear line in narrative form between truth and lies. Narration itself blurs the line between any easy distinction we might make between truth and falsehood. I will concede that the events of a memoir might be scrutinized to a certain degree by the ‘truth’ of what did or did not happen, although I take it to be a category mistake to apply the notion of ‘truth’ to a memoir as a whole. The narrative form, despite its ‘true’ representational content, disallows it to be considered entirely true in this sense. Another interesting case of memoir to consider is one in which a confused sense of self lends itself to an inaccurate memoir – not because of intentional lies, but because of genuine self-delusion. This seems to be the defence of Misha Defonseca’s memoir, Misha: A Memoire of the Holocaust Years. She describes how her parents were taken from her Belgian home in the early part of WWII. Her father was a war criminal who gave up state secrets. That much is true. She concocted a story as a child, however, that her parents were sent to Auschwitz and she ran away from the care of her grandfather to find them. Along the way, she was adopted by a pack of wolves who took her in as their cub. The acceptance of the usually vicious animals was demonstrated in stark contrast to the inhumanity of the Nazis. As she repeated the story over and over through her adult years the story became very real to her. Eventually, after years of living in the States, her friends convinced her to publish this moving memoir. After initial praise for fearlessness and great humanity, fact checkers provided the publishers with evidence that her father was taken by the government as a war collaborator, but Misha was not a child of the Holocaust. She (Sheilds 2008) later explained that “the book is a story, it’s my story. It’s not the true reality, but it is my reality. There are times when I find it difficult to differentiate between reality and my inner world”. This story, about her searching for her (incidentally, non-Jewish) parents, was likely a coping mechanism that she used to help her deal with a tragedy that happened to her in her youth. She told the story repeatedly and it became more and more real to her. In this case, then, it is not as clear that she was intentionally trying to deceive, but instead she was trying to create an identity that was more palatable than the one circumstances provided her. Several of the examples of fabricated memoir seem to come out in ways that where authors did not set out to intentionally deceive their audience, but rather

104 | Sarah E. Worth their own mistaken views on their lives get translated into a story that does not lend itself to ‘truth’ in any standard account. When the stories we tell about ourselves are put into a public form like memoir, we are certainly subject to the scrutiny of criticism and fact checking, but as readers we might also take into consideration the construction that goes into selfhood well before we map ourselves into a memoir.

3 The possibility of transparency in memoir If it is possible that the self is what memoir aims at getting an accurate view of, then the notion of transparency, as discussed in issues in photography, can help to frame some of the issues with truth and memoir. Kendall Walton (1984, 250) explains that one of the things that appeals to us about photographs is their realism, but what is distinct about photographs is that they are always of something which actually exists. Photographs are an ideal of representation. Portraits – paintings – can be life-like, but their relationship to their subject is different from a photograph’s relationship to its subject. Walton (1984, 251) claims that photography is like an aid that can help us with our vision. Photographs can help us “see into the past”. His notion of photographic transparency allows us to literally see our dead relatives, to see the events of the past, and to see ourselves as younger than we are now. Paintings, portraits specifically, are not transparent in this way, Walton says, but always opaque. This becomes relevant as he develops a theory of perception and suggests that “to perceive things is to be in contact with them in a certain way” (Walton 1984, 269). He (Walton 1984, 270) says that “a mechanical connection with something, like that of photography, counts as contact, whereas a humanly mediated one, like that of painting, does not”. Ultimately, Walton (1984, 269) argues that there is a certain kind of “contact” that is achieved through photography that is different than with any other kind of perception. He (Walton 1984, 270) says that photography produces this kind of contact with its subject in a way that portraiture can never achieve. This contact, he advocates, also has to do with a sense of first person experience or direct access. In storytelling, this first person testimony is incredibly valuable to the kinds of things that we believe about other people’s stories. Although Walton’s argument is specifically about photographs and visual perception, I want to suggest that his account might serve as a useful metaphor to inform some of my claims about memoir. I take it that the ‘perception’ of memoir is something parallel to our ideal perception of what photography can

Narration, Representation, Memoir, Truth, and Lies | 105

offer us; transparency, accuracy, and realism. Walton suggests that the transparency we have with photographs is akin to the first person experience of seeing an object. What we learn about the object in a photograph (or a memoir) might be wrong, but there is something important to be said for the direct perception. The ideal of memoir might be like photography in that it is, ideally transparent – it presents a view of the author’s experience that can be accurate, realistic, and maybe even said to be generally true. As readers of the genre of memoir, I take it that we want something akin to the kind of photographic transparency that Walton talks about. It seems to me that there is an expectation of something akin to photographic transparency that readers have when reading memoirs. I take this to be the case because of the kind of criticism lodged at the fabricated memoirs. Fraudulent memoirs offer us the illusion of transparency when in fact the realism, accuracy, and ultimately the truth of the experience have been falsified. Readers who believe they are getting a transparent view into another’s experiences, and later find out it was faked, feel justifiably duped. The contact the reader felt with the author is thrown into question, and no matter how good the plot and the prose, readers do not know how to process what they initially took to be truth. Presumably, readers need to reassess and adopt a different stance towards the work, but in reality this seems to be somewhat difficult to do. This shift is particularly difficult because it is not the case that I found something out to be false that I initially thought was true. Rather, I found out something was false that I was intentionally misled about.

4 The truth about what happened One of the important aspects of memoir, highlighted by the cases of fabrication, is that the author presents a particular state of affairs as having happened, as having historical accuracy, when such events did not happen or happened differently than portrayed. Sometimes there is more at stake with this issue than others. Which events are important enough to have historical accuracy? The exact number of days someone spent in jail?8 The exact injustices done to one during a time of war?9 Perhaps events in which many people were affected || 8 James Frey said in A Million Little Pieces that he spent 87 days in jail, when it was documented as only 3 hours. 9 See the example of Rigoberta Menchu above who, in a first person narrative, described what happened to her directly when in reality the events she described are a compilation of things that happened to her, her family members, and others she heard about second hand.

106 | Sarah E. Worth would be more important than individual perspectives like the Holocaust – or events that have documentation would be more important to get right. The issue of historical accuracy can be divided into two further elements; memory and history. I will deal with each of them in turn. Memory is not something that we have easy, clear access to. Psychologists often argue against what might be called the “videotape recorder view of memory” (Einstein and McDaniel 2004, 34). This view claims that we have a detailed record of our experiences that are faithfully recorded in the hinterlands of our minds. Although we might need to go to great lengths to retrieve some of the memories, like through hypnosis, the memories are always in there, they are retrievable, and they are accurate. Psychologists argue that this simply is not true and this folk understanding of memory is completely misguided. Instead, the consensus from the psychological literature is that memories are anything but clear or literal representations of events. Many memories are irretrievable and the ones we do have are often inaccurate. Memories are interpreted and reinterpreted through time, through our retelling of events, the occurrence of new events, the need for consistency in our own beliefs, and of our own basic need for narrative construction (Lynn and Payne 1997, 55). Further, in terms of simple descriptive memory, there is a distinction to be made between semantic and episodic memory. Semantic memories are small, individual memories that can be verified and agreed upon and even validated, such as the name of my elementary school, the date of my husband’s birthday, and the number of inches in a foot. Episodic memories are series of events that are strung together, at least in a chronological order, if not also in a causal sequence. These are the kinds of memories that are very susceptible to revision of all sorts.10 That is, they are revised by subsequent experience, other memories, others’ renditions of the events, coherence or consistency making, and even our own conscious or

|| 10 Beginning in the 1980s there were a large number of people who claimed to have recovered repressed memories from their childhoods of some sort of childhood trauma, most often of sexual abuse or their involvement in satanic rituals. These claims have been proven almost completely unreliable. Recovered memories are most often made available through hypnosis and suggestion. When people are hypnotized they lose their inhibitions and are more open to suggestion and they can ‘remember’ things that didn’t happen. Physical evidence is rarely found or is no longer available in many of the cases. Forensic scientist Terence Campbell (1995) claims that “the accumulated data related to long-term memory indicate that people rarely remember much that occurred before the age of 4, and moreover, memories from between the ages of 4 and 8 tend to be quite unreliable” as well.

Narration, Representation, Memoir, Truth, and Lies | 107

unconscious revision to make the events or ourselves more likable or appealing.11 The insights from psychology are important in terms of the argument that although we long for the videotape ideal, and we may even insist that we have it at times, this kind of transparency of memory is simply not available. Semantic memories are easily fact-checked and allow for falsification, but it seems as if the episodic memories are the ones that are memoir-worthy; they are the sequences, the connected feelings, and eventually the insights. It is these episodic memories that build the backbone of a memoir. Think about the kind of controversies that ensued from the following titles: The Da Vinci Code (Dan Brown), Satanic Verses (Salman Rushdie), and even Harry Potter (J. K. Rowling). All of these titles were marketed and sold as fiction, but were highly criticized because the perception was that what they were presenting was true. Of course, all fiction can refer to real facts and include true propositions. In these cases, however, the represented content was taken as threatening in a way fiction standardly need not be. The controversies are not really about what may or may not be ‘true in the fiction’ as Walton might say, or more accurately, as Walton might say ‘made fictional in the fiction’ (see Walton 1990, 41). The controversies arose over the supposed truth or accuracy of the relationship between the text and the real world. What was at stake in these examples is that what is made fictional had a bit too much overlap with what is true in the real world. In the case of The Da Vinci Code, for example, since most of the historical, art historical, religious, and geographical aspects were based on facts, critics were quick to debunk any aspect they could (especially the religious conspiracy aspect of the story). This one fictional narrative spawned dozens of books attempting to discredit it. The genre of fiction allows for true propositions, historical accuracies, real people and events etc., but what makes it fictional is the way in which we take it. For Walton, what makes something fictional is something whose function is to be used as a prop in a game of make believe (Walton 1990, 72). For Gregory Currie, perhaps, what makes something fictional has more to do with the intentions of the author, or how it is used as a fictional utterance (Currie 1990, 18). Whichever theory of fiction we use, often, fiction writers are criticized if they get this kind of contextual information wrong.12 Fiction allows for both true and fictional propositions, but memoir is

|| 11 See Galen Strawson (2004), for instance, who outlines the differences between episodic and semantic memories, but who also argues against the notion of a narrative constructed self. 12 See Christopher Bartel (2012, 213–221). Here Bartel discusses the criticism of historical period fiction that gets historical detail wrong, intentionally or unintentionally.

108 | Sarah E. Worth criticized for any falsities. For the most part, readers of fiction know how to take narratives that are marketed as fiction. I am not sure the same is true for readers of memoir. Walton distinguishes between the real world and fictional worlds as though they are ontologically separate – ontologically separate in our minds at least. Fictional worlds (not the genre of fiction) are held apart by our attitude towards them as being make-believe, but it is this attitude that is one of the defining features of the way Walton describes what it is to be fictional. I want to focus on the element of narrative that is shared between fiction and nonfiction rather than on the ontological distinction of fictional worlds and so-called real worlds. With the notion of worlds I hope to convey the coherence not just of fictional intentional objects but the conceptual maps with which we imagine that which is fictional. With this shift in focus, what we have is, rather than the traditional split between fiction and nonfiction, a focus on narration and that which is not narrated; non-narrated experiences or events are merely those which have not been edited by, or crafted into story. Non-narrated events might include chronological lists, memories, or descriptions, but they do not include plot or causation. I want to argue that narration must necessarily be prior to any notion of fiction or nonfiction. What narrated worlds share are the kinds of imaginings that the narration prompts. The props in Walton’s make-believe account prompt certain kinds of imaginings, but we have similar kinds of imaginings with narrated real (true) events as well. Walton’s account of the way in which we use fictional props can help me to make some important distinctions. His theory focuses mainly on the representational arts, and the make-believe is largely about the kind of imaginings, beliefs, and emotional responses that result from our engagement with fiction. When we engage with a fiction, it incites certain kinds of imaginings, but the imaginings are not constrained by merely what is said to be “true in the fictional world” (Walton, 1990, 36). As Walton explains, beliefs, unlike imaginings, are correct or incorrect. Beliefs aim at truth. What is true and only what is true is to be believed. We are not free to believe as we please. We are free to imagine as we please. So it may seem. Imaginings are constrained also; some are proper, appropriate in certain contexts, and others not (see Walton 1990, 39). There is a parallel to be drawn between truth and belief, and fictionality and imaginings. What is fictional promotes certain kinds of imaginings and what is true promotes certain kinds of beliefs. Walton (1990, 41) says “what is true is to be believed” and what is fictional is to be imagined. He warns, however, that

Narration, Representation, Memoir, Truth, and Lies | 109 there is a persistent temptation to go one step further and to think of fictionality as a species of truth. The temptation is both reflected in and nourished by the fact that what is fictional is colloquially described as true in a fictional world [...] [but] what we call truth in a fictional world is not a kind of truth. (Walton 1990, 41)

It is rather a kind of internal consistency that gives fictional worlds the ontological distinction that we want to grant them. Anna Karenina’s world is complete because we can imagine it, in the same way, using the same mental imagery and cognitive mechanisms that we use to imagine our own worlds, and other real worlds that are described to us. So the fact that we have certain kinds of imaginings seems to be the linchpin between Walton’s account and mine. The worlds he describes are dependent upon fictional imaginings. The worlds I am interested in exploring are dependent upon the narration of events which prompt the same kinds of imaginings. What makes memoir so unique on this question is this intersection between narration and personal history – between story and truth. Walton’s use of props in his theory of fiction is enormously important. Props in games of make-believe are the signposts or rivets that fasten fictional worlds together. They are one of the things that can give fictional worlds internal coherence. As Walton (1990, 41) explains, we tend to think of “fictional worlds as remote corners of the universe where unicorns really do roam, where a war is actually fought over how eggs are broken, where it is true that a bear hides in a thicket a few feet from Eric”. But with my focus on personal narratives or memoirs, the purely fictional world is not a different imagined land looking for internal coherency, but rather a narrativized world where the props are the true events which populate that world and whose narrative is the glue which gives the account coherence. So whereas Walton’s make-believe theory relies in large part on so-called fictional worlds, and so-called real worlds being implicitly distinct, I want to suggest that with the kind of worlds narrative provides us, that distinction is misleading. So which is it – are the facts more important or the story – are the beliefs more important or the imaginings? It strikes me that aestheticians have come across a similar problem and have dealt with it at length in the case of forged paintings. The same painting which was once praised as a Vermeer is now criticized as a Van Meegeren. But I do not think it is the same kind of example. These memoirs are not forged, per se, they are just invented, fabricated. They are fictional when we thought they were fact. What differentiates fact from fiction is a different kind of distinction than who the author or painter was. What is at issue in the case of memoir is much more foundational – it is the standard of truth that we use to judge narratives. It seems to me that we prioritize a corre-

110 | Sarah E. Worth spondence theory of truth over a more comprehensive version of a coherence theory or perhaps over a different kind of valuation of something like verisimilitude. When we are talking about not just the mere representation of events, but life writing, correspondence should not be a dominating value. A coherence theory makes much more sense when writing about putting together the meaningful events in one’s life enough to create a comprehensive story out of them. I take it that the genre itself should have a greater emphasis on story and coherence than merely what can be documented. This takes me back to my initial claim that narration can only loosely be said to be representational. If representation is any sort of clean or accurate or true copy of the way the world really is or what events truly took place, then narration almost immediately excludes itself based on the necessarily perspectival, time-edited, imploded personal history, and the inherently self-focused descriptions memoir creates. To reduce our criticism of memoir to merely whether or not it is a true account of events that took place diminishes the power of what life writing can and should be. In the end, the argument I support is one that suggests that fictionality is not the main cause of our imaginings, but it is the coherence of narration itself that prompts all kinds of imaginings and those imaginings cannot be diminished by the subjection of overly-simplistic truth categories.

References Bartel, Christopher. “The Puzzle of Historical Criticism.” Journal of Aesthetics and Art Criticism 70.2 (2012): 213–221. Bruner, Jerome. “The Narrative Construction of Reality.” Critical Inquiry 18.1 (1991): 1–21. Campbell, Terence. “Repressed Memories and Statutes of Limitations: Examining the Data and Weighing the Consequences.” American Journal of Forensic Psychiatry 16.2 (1995): 25–51. Currie, Gregory. The Nature of Fiction. New York: Cambridge University Press, 1990. Dennett, Daniel. “The Self as a Center of Narrative Gravity.” Self and Consciousness: Multiple Perspectives. Eds. Frank S. Kessel, Pamela M. Cole, Dale L. Johnson and Milton D. Hakel. Hillsdale, NJ: Erlbaum Press, 1992, 103–115. Defonseca, Misha. Misha: A Memoire of the Holocaust Years. Bluebell: Mt. Ivy Press, 1997. Einstein, Gil, and Mark McDaniel. Memory Fitness: A Guide for Successful Aging. New Haven: Yale University Press, 2004. Eskin, Blake. A Life in Pieces: The Making and Unmaking of Binjamin Wilkomirski. New York: WW Norton & Company, 2002. Frey, James. A Million Little Pieces. New York: Anchor Books, 2003. Frey, James. “A Note to the Reader.” http://www.nytimes.com/2006/02/01/books/01cndftex.html?_r=0 The New York Times, 1 February 2006 (24 January 2013).

Narration, Representation, Memoir, Truth, and Lies | 111 The Smoking Gun. “A Million Little Lies Exposing James Frey’s Fiction Addiction” http://www.thesmokinggun.com/documents/celebrity/million-little-lies. 4 January 2006 (January 16, 2013). Grandin, Greg. “It Was Heaven That They Burned.” The Nation. (27 September 2010). Jones, Margaret. Love and Consequences: A Memoir of Hope and Survival. New York: Riverhead Books, 2008. King, Larry. “Interview with James Frey.” http://transcripts.cnn.com/TRANSCRIPTS/0601/11/lkl.01.html CNN Transcripts. 11 January 2006 (16 January 2013). Litte, Jane. “Random House to Settle James Frey Class Action Suit.” http://dearauthor.com/features/industry-news/random-house-to-settle-james-freyclass-action-suit/ Dear Author, 7 November 2007 (1 February 2014). Lynn, S. J., and D. G. Payne. “Memory as the Theater of the Past: The Psychology of False Memories.” Current Directions in Psychological Science 6 (1997): 55. Maechler, Stefan. The Wilkomirski Affair: A Study in Biographical Truth. Translation, John E. Woods. Including the entire text of Fragments. New York: Schocken Books, 2001. Menchu, Rigoberta. I, Rigoberta Menchu: An Indian Woman in Guatemala. Brooklyn: Verso, 1984. Mink, Louis. “Narrative Form as a Cognitive Instrument.” The Writing of History. Eds. Robert Canary and Henry Kozicki. Madison: University of Wisconsin Press, 1978, 129–140. Mozes, Suzanne. “James Frey’s Fiction Factory.” http://nymag.com/arts/books/features/69474/ New York Books, November 2010 (1 February 2013). Relin, David Oliver, and Greg Mortensen. Three Cups of Tea: One Man's Mission to Promote Peace . . . One School at a Time. New York: Penguin, 2006. Rosenblat, Herman. Angel at the Fence. New York: Berkley Books, 2009. Sheilds, Rachel. “Adopted by Wolves? Bestselling Memoir was a Pack of Lies.” http://www.independent.co.uk/news/world/europe/adopted-by-wolves-bestsellingmemoir-was-a-pack-of-lies-790000.html. The Independent World, 2008. (16 January 2013). Strawson, Galen “Against Narrativity.” Ratio (2004): 428–452. Walton, Kendall. “Transparent Pictures: On the Nature of Photographic Realism.” Critical Inquiry 11 (1984): 246–277. Walton, Kendall. Mimesis as Make Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. White, Hayden. The Content of the Form: Narrative Discourse and Historical Representation. Baltimore: Johns Hopkins University Press, 1987. Wilkomirski, Binjamin. Fragments: Memories of a Wartime Childhood. Translation, Carol Brown Janeway. New York: Schocken Books, 1996. Worth, Sarah. “Storytelling and Narrative Knowing.” The Journal of Aesthetic Education 42.3 (2008): 42–55. Yagoda, Ben. Memoir: A History. New York: Riverhead Books, 2009.

Remigius Bunia

Truth in Fiction 1 Introduction In the framework of Kendall L. Walton’s theory, a newspaper article is, in the typical case, nonfiction, while a novel is fiction. However, both are made of words, and both can be understood as prescriptions to imagine. But in fact, Walton (1990, 58) defines fictionality, and nothing but fictionality, as “prescription to imagine”. Even though a factual proposition might prompt a reader to imagine something, this is not what he thinks a proposition usually does. When someone says that planet Earth pivots around the sun, this is not, from his point of view, to be considered as an invitation to imagine cosmic objects moving through space. How does he draw the line between fictional prescriptions to imagine and nonfictional propositions that might provoke some imagination despite their sheer propositionality? The answer is that he implicitly assumes two realms of language. On the one hand, we have a pure propositional mode of expression; on the other hand, we have a more imaginary mode of expression that prompts us to create mental simulations of perceptions, images, and ideas. I am going to detail this distinction between the two realms in great length in the following article. Neither realm is linked to either fiction or propositions (whether or not we understand fiction and propositions along Walton’s lines of reasoning). This distinction will nonetheless prove helpful in analyzing both Walton’s concept of fiction and other theories. In fact, it will even contribute to a better understanding of language and communication. Thus, my purpose is not to expand the existing theories of language and fiction, but to introduce some new concepts that might lead us towards new insights. In the first realm, we find propositions that clearly relate to reality. Here, we assume that word-based1 language can convey these propositions. Linguistic expressions and thoughts are considered equivalent – at least this equivalence is deemed possible and should be strived for (plain style, precise definitions, etc.). In Descartes’s famous wording, ‘clear’ and ‘distinct’ thoughts can be ex|| 1 I use the adjectives word-based, linguistic, and verbal without any nuance of differentiation and try to avoid the expression ‘verbal’ since it can be associated with the grammatical category of verbs, which plays an important role in my essay.

114 | Remigius Bunia pressed in a precise language. Let me call this type of description between thinking and speaking the propositional realm. In the second realm, we encounter different forms of expression: pictures, sounds, letters on paper, and (if I may mention Walton’s famous example) the stump of a tree. These expressions incite us to imagine perceptions. However, we are not prompted to interpret these imaginations. They are supposed to behave very much like immediate experience: When we scrutinize the sight of a tree (as to find out what kind of tree it is), we can do so with a painted tree or, equally, a rich verbal description of a tree. (I will speak of immediate reality or immediate experience when I refer to a situation characterized by the essential absence of symbols.)2 We face things that take place in our minds and whose relationship with reality remains ambiguous. The impression itself is definite – even if, of course, a description can present us with a deliberately equivocal representation of an impression (as William Turner and Claude Monet do in many of their paintings). Let me call this ‘territory’ the aesthetogenic realm (this neologism3 implying that ‘perceptions are generated’). My distinction applies to the mental processes a particular representation prompts. One should not be tempted to feel reminded of Nelson Goodman’s (1997 [1976]) notions of density and sample, since the distinction between propositional and aesthetogenic realms does not deal with the quality of the representation. My inquiry rather adopts a cognitivist perspective. Although I try to challenge most of the analytical tradition and include some of the poststructuralist critique, my argument will be ‘naturalistic’. As most cognitivist theories of today, my view is based on little biological evidence. The point of the cognitive approach is not ‘biologism’, but to consider phenomena from an operational perspective. Thus, I do not ask how things are, but how they are processed; and this implies that my inquiry is about the way humans process their sensual experience in order to deal with reality. With his description of fiction, Walton opened up a new perspective on representations that redirects our attention towards mental processes. When he speaks of games, he has stopped discussing the relationship between reality and word-based descriptions. Instead, he focuses on the processes of dealing with certain real entities that prompt us to imagine. In this sense, he adds an operational ingredient to the static analytic descriptions that prevail in most || 2 Of course, all experience involving symbols is in some way equally immediate, since it can involve a human being in the same way intellectually or emotionally as symbolic communication. The point in question here is that any experience that we would call direct is defined by an absence, not by a presence. 3 As Peirce (1905, 163–165) noted, it is better to invent a neologism than to risk confusion.

Truth in Fiction | 115

theories on fiction. I will focus on these operational aspects within his theory and link them to some insights from cognitive sciences. Certainly, the critique that has been repeated some dozens of times aims at a particular confusion within Walton’s model. While he distinguishes between the newspaper article and the novel, he does not fundamentally differentiate between a newspaper photograph and an ‘artful’ photograph. Put differently, he does not distinguish between a photograph serving to represent the truth and a photograph serving a significantly different purpose, namely being fiction. Many scholars have tried to re-establish the difference between fiction and nonfiction photographs without giving up the fundamental insights given by Walton’s model (e. g. Bareis 2008). My objective, however, will be different from this. While I fully admit a difference between a faithful and a delusional photograph (as Walton too would), I doubt that a representation can be fully propositional. Additionally, I posit that the distinction between propositional and aesthetogenic is gradual. If my theoretical framework turns out to be advantageous, then we face the question of discerning the difference between ‘fact and fiction’ once more on a new and more convenient epistemological basis.

2 Knowledge, memory, and the human body I will introduce some expressions that are useful for my argument. The human body uses its perceptions and its memory to change its memory and to act. Among the possible perceptions, we have acoustic and optic impressions of word-based language. Whenever impressions change the memory, I will speak of information (with reference to the impression) and of learning (with reference to the changes of memory). Whenever changes of memory occur due to internal processes without new impressions, I will speak of reasoning. What is ‘stored’ in our memory can be called knowledge. My notion of knowledge is very neutral and essentially equivalent to that of memory; I do not assume that someone who has a specific knowledge must know of this knowledge and can give any account of it; she need not ‘believe’ it, and my knowledge need not be accurate. Further, I will not differentiate between dispositional (‘knowing how’) and propositional knowledge (‘knowing that’), since any propositional knowledge is nothing but an operational knowledge. To know that something is the case only becomes manifest in operations that rely on this knowledge. I will expand on this when I discuss the cognitive processes involving knowledge. No knowledge has an operation-free state. In stating this, I do not discard this philosophically

116 | Remigius Bunia popular distinction en passant, but the insufficiency of this distinction is a core matter for my argument. Despite this, we can distinguish two fundamental types of knowledge and information, respectively, namely: viable or illusive. I define viable as practically relevant. (I take this term from Stephan Packard’s unpublished work.) Viable knowledge and information guide us in our daily practical decisions (whether consciously or not). Illusive knowledge and information, however, are used only when the situation requires the reference to this piece of knowledge – and not for a practical decision. An illusive piece of information or an illusive piece of knowledge exist for their own sake; they have no specific purpose in coping with the world. Let me give some examples. Of course, illusive information can be emotionally important or entertaining. When a widower remembers his dead wife’s traits, this knowledge is illusive (unless he shares these memories with others); but it might be vital for his emotional well-being. Let’s take the Sherlock Holmes stories as a second example: I may know that Watson works with Sherlock Holmes. Usually, this is illusive knowledge. It is useful only in the context of reading Arthur Conan Doyle’s stories or discussing these stories with friends (for purposes of entertainment). However, for a literary critic, knowledge about Watson and Holmes can be relevant: It is practical knowledge for writing articles about them. Nevertheless, it remains illusive as this knowledge can only be activated with an explicit reference to the story. Quite differently, I may find valuable information about the architecture in Paris in a novel by Victor Hugo. Then, I gain viable information. It is important to see that my notion of knowledge is based on its operationality only. It does not yield a concept of ‘factuality’. The knowledge is adequate to reality if it helps us cope with reality, not if the knowledge is ‘correct’ in any sense of the word. However, viable knowledge can consist of good descriptions of reality. Thus, I do not doubt that we can have good and correct ideas of how the world actually works. Physics, in particular, gives a very precise understanding of the elementary qualities and behaviors of matter. This understanding is viable even in a very practical sense since this knowledge allows us to erect tall buildings, to predict certain future events, and so on. In my terminology, the terms knowledge and memory label the same phenomenon, seen either from the phenomenological or the biological point of view. I will thus relate my concept of knowledge to recent neurological insights. Since the brain is complex, simple dichotomies are of course misleading. Nonetheless, at least some basic functional distinctions can be based on analyses of both lesions and anatomical structures. On the one hand, I will discriminate

Truth in Fiction | 117

between idiosyncratic and generic knowledge. On the other hand, I will distinguish two operational spaces: the image space and the dispositional space. 1. Some will know the idiosyncratic knowledge by the more widespread name of episodic memory. This knowledge is the ability to recognize singular events and entities. These have a specific position in time and space, inalienable features, and the memory is useful only with regard to the particularities of the respective events and entities. Your parent’s face, the look of your apartment, your first love: all these specific memories add up to your idiosyncratic knowledge. I prefer the notion of idiosyncratic instead of episodic, because less ephemeral memories also belong here: Abraham Lincoln’s face, the first lines of Dante’s Commedia, and the consequences of the Vienna Congress are idiosyncratic knowledge, but far from any episodic contextualization. Generic knowledge, however, pertains to patterns that are encountered in many instances. I know how trees look like, how coffee ought to be brewed, and how houses are to be dealt with (take the door, not the window, when you wish to get inside). 2. Let us turn to the difference between two operational spaces, the imaginative and the dispositional. The image space constructs copies of the perceptional data being received. When a complex being is seeing a tree, the cortex creates a copy of the particular impression it finds on the retina. So does the image space with all other senses. But the image space does more than that. It compares the impression with earlier impressions and thus recognizes the sources of the impression. In the image space, the current, the recalled, and the constructed perceptions pop up and are subject to further mental processing. The dispositional space, however, responds to the current bodily state (including its perceptions and thoughts) and engages the body (including its brain) in some action. Actions in this sense need not be carried out; they can remain in the state of an impulse. And actions can target the mind itself: One of the most important actions is the retrieval of generic and idiosyncratic memories, the recognition of the things and persons being around as such and such. As long as both spaces perform their tasks without interacting with each other, their range of effect is limited: The dispositional space assesses the reactions deemed necessary to cope with present discomforts and joys, and the image space provides the dispositional space with the data. In higher animals, among them humans, the dispositional space starts acting on the image space, which in turn keeps informing the dispositional space. This feedback loop allows the dispositional space to draw its conclusions not only from the body’s monitoring system, but to make the mind monitor the products of its own thinking processes. This is why Antonio Damasio (2012) describes the interlock between the image space and the dispositional space as the interface where ‘consciousness

118 | Remigius Bunia comes to mind’, as he playfully puts it. When the dispositional space can feed the image space with contents (from either memory or deliberate construction), what we perceive inside our mind is no longer restricted to body reports about its exterior, but also about our interior. It is important to note that we only sense the products of the image space as contents of our consciousness. In contrast, the operations within the dispositional space remain inaccessible to consciousness (Damasio 2012). Therefore, we cannot say how we can recognize specific objects; we just know that a tree that we are seeing is a tree, but we cannot tell what we are doing in correctly identifying it. As another consequence, philosophical tradition tends to overlook the powerful mental performances taking place without our consciousness knowing; and tradition has thus wrongfully equated consciousness and mind. We tend to take the highly structured contents of our consciousness for the products of our exceptional abilities, while in fact both the process of structuring and the re-creation of structured impressions in the image space should fill us with awe. Francis Bacon’s and Claude Lévi-Strauss’s complaints about the human tendency towards abstraction face a biological explanation here. My distinction between propositional and aesthetogenic communication directly refers to the difference between dispositional and imaginative operations. On the one hand, a propositional communication activates the abilities to recognize things, persons, and processes; and these abilities are housed in the dispositional space. This is all that is behind elementary symbolic signs: One learns to deal with the symbol for an experience in much the same way as with the immediate experience. Whether I hear people shout ‘fire’ and the alarm bell or I see the flames in my room myself, I should run out of the building as fast as I can. Damasio even presumes that language use is dispositional; this is an intuitively sound hypothesis, given the fact that I cannot explain how I know what the word ‘tree’ means better than how I identify a tree in the streets. I just do. On the other hand, an aesthetogenic communication asks the recipient to turn on their image space and to populate it with impressions. I can be asked to imagine the details of an object that is known to me; but I can also be made construct an object I have never seen before.4 As defined above, the aesthetogenic and the propositional pole encompass a continuum. This is to say that the dispositional space and the image space can contribute varying amounts to the experience evoked by communication. To || 4 This is why Walton is wrong when he assumes that imagination works by “moving existing furniture” only. Humans can also construct new furniture through aesthetogenic representations!

Truth in Fiction | 119

imagine a unicorn (for the first time in a lifetime), a minute departure from the propositional pole is sufficient. One just has to plug a generic horn into the head of a generic white horse, and the imaginative work is done. However, to imagine the landscapes and the beings described in J. R. R. Tolkien’s The Lord of the Rings requires much more activity. I will speak of imagery (Damasio) when the operations in the image space create settings that substantially take off the ground of dispositionally managed experience. Inversely, imagination serves as the term for filling the image space with remembrances. When I spoke of either memory or construction delivering content to the image space, I implicitly referred to this distinction: The dispositional space can make the image space create experiences that come from the past (imagination) or out of the void (imagery). Of course, humans do not restrict their use of imagery to art, but they employ it in all kinds of planning (what shall I do after finishing my meal tonight? will I earn enough money to buy the car next year? what, by the way, can we humans do to save the planet’s climate?).5 Imagery is where both rational construal of future events and aesthetic creations take place, where they may meet, and where they more often than not contaminate and enrich each other. Again, the distinction between imagination and imagery is not a question of all or nothing. Only seldom does imagination produce exact reproductions of the lived experience of the past (this is why humans are mediocre witnesses), on the one hand. And on the other hand, at least some generic knowledge is necessary for imagery. This becomes manifest when we analyze how aesthetogenic communication works. Here, we can observe how the mind prefers to be instructed about images. In a personal communication, Walton expressed some doubts about the distinction between aesthetogenic and propositional, thus shedding much light on the intricate dependency of aesthetogenic communication on generic knowledge. He asked whether it would be an aesthetogenic way of describing a sunset if I laid a grid over my visual impression and, row by row, column by column, told someone the RGB color for every pixel aloud. (This is how computer software encodes so-called bitmaps: A picture is a list of numbers representing colors.) Walton considered this a counterexample, because such an utterance probably would not produce any immediate visual experience within the recipient. Rather, it would leave the indulgent recipient with a time-

|| 5 Interestingly, we know from experimental studies that the paratextual warning ‘fiction’ activates the brain region that is also active when someone makes plans for the future. The inverse paratextual warning, ‘fact’, rather activates the regions related to instant bodily action (see Altmann et al. 2012).

120 | Remigius Bunia consuming absurd task. This thought experiment hints at the complex nature of human cognition. Although the retina and the cortex indeed do nothing but measure colors pixelwise and compose them until they form a mental image, the acoustic communication of rasterized visual information annoys and exhausts the mind. The explanation for this difficulty is simple. The image space as defined above can only deal with existing forms, mostly stored as generic knowledge, and the number of forms must be limited to a rather small number. These forms can be accessed again though dispositional knowledge – – – – –

in the visual (shapes, geometrical relationships, contours of known objects, colors, reflection properties, and so on); in the acoustic (pitch, intensity, echo, distance, position, and so on); in the haptic (warmth, smoothness, softness, humidity, itch, burn, irritation, and so on); in the olfactory (an extremely long list of individual incomparable impressions such as thyme, tomato, and gasoline); and in taste (salty, sweet, bitter, sour, and umami).

A linguistic aesthetogenic description will use these elements and also complex generic forms (such as persons, things, processes) in order to prompt the imagery of a new experience. When I ask you to imagine the odor of cumin, my request can remain propositional; when I ask you to combine cumin with cherry mentally, the task is aesthetogenic; and it only works if you can recall the scents mentally (a capacity seldom trained in our culture) and if you can deduce the impression from imagining it (a capacity even more rarely trained here, alas). If this example strikes you as exotic, think of musicians who can imagine the acoustic impression of particular instruments playing particular notes; many trained musicians can experience a tune or a whole symphony by reading the score. In our visual culture, we train our optic capacities, and those who live in Europe nowadays, are not born blind and have not suffered any severe neurological impairment can easily imagine the picture of a cherry lying on a violin. A successful communication will combine only as many elements as can be processed. It is easy to imagine a green dog, a yellow cow, and a blue mouse standing in front of the Arc de Triomphe of Paris; but it is difficult to imagine thousands of such colorful animals in rows and columns. If we were able to imagine a picture with so many elements, we could construct an experience from the enumeration of pixels too. Essentially, both techniques ask for the same capacity.

Truth in Fiction | 121

3 Perception, language, and human communication Since in this theoretical framework, knowledge is a disposition to act, it cannot be described in ‘static’ sentences. This is why I have great difficulty in describing knowledge by means of natural language. I can say: ‘I know that apples grow on trees.’ But it remains largely unclear to what extent my ability to produce such a sentence relates to the agricultural insights conveyed through this sentence. More specifically, I put forward that it is wrong to equate human cognition and word-based descriptions; there is no evidence whatsoever that all higher human capacities can be resolved into language. When we scan the Western philosophical tradition in search of what has been counted as the cardinal higher abilities, we find, firstly, symbolic reasoning, secondly, spatial and, thirdly, quantitative reasoning. By consequence, tradition has tended to emphasize the importance of conscious efforts, above all of word-based reasoning. This is why, until today, we face a widespread biased view on our own cognitive capacities and limitations. However, we now know that the ‘lower’ abilities that operate without involving consciousness have much impact on our actions and reasoning. Not only does the continuous analysis of perceptual data prove very complex, but more astonishingly, psychologists now convince us that intuition is a ‘higher’ ability of assessment, operating without any notice from consciousness. Intuition serves to make fast and heuristic judgments; and some psychologists – for instance Daniel Kahneman (2011) – insist that conscious logical reasoning more often than not contributes to the fast intuitive decisions even in complex situations that would apparently benefit from logic (such as inferences from statistical data). Note the famous finding that lengthy intensive conscious comparisons between economic options lead to an inferior quality of decision compared to intuitive fast thinking. Unaware of our astonishing faculties of unconscious reasoning, the classical philosophical assumption, inherited from antiquity, is that language is specific for human cognition and that it can express virtually everything the human mind deals with. This is why the proposition has been given so much credit in the tradition. And it is true that a proposition can express a state of the human mind in an extraordinarily compact way. What it contains is knowledge. But if verbal descriptions turn out to be far away from mental representations, if they are nothing but an incitation to produce mental simulations of perceptions, then there is no reason to prioritize them over other forms of communication. While the tradition assumes that language always encodes a static and stable

122 | Remigius Bunia interpretation of past perceptions, I suggest that language in many cases encodes a perception requiring a new interpretation. This claim needs more clarification. One might interpose that human communication can develop forms of expression that evade the need of new interpretations. We have biological terminology, mathematical formulation of physical laws, juridical abstraction, and so on. Yet even within these fields of high clarity, there are degrees of interpretability: A mathematical formula allows for less variance than statutory laws; probably only in mathematics, we can attain the complete absence of ambiguity. And then again, we obviously have the diametrical opposite pole, that of poetic language. On this side of the spectrum, language can turn into a means of expression that triggers interpretation, but gives few hints at limits. In fiction, verbal description and reality usually diverge to a great extent (I will have to return to this point as to give a more precise phrasing than this). In the following sections, I will reinterpret the initial notions of propositional and aesthetogenic ‘realms’ as modes of communication. The scientific language will be the propositional extreme within this range, and the poetic language will be its aesthetogenic opposite. In order to deploy this analysis of communicative modes, I adopt Ronald W. Langacker’s (2008) fundamental assumption that the structure of ordinary language reflects the regularities of human processing of perceptions. Grammar is not made to represent reality, but to represent our impressions of reality. In accordance with Langacker’s theory, neuropsychology shows that humans structure their perceptions by handling them as things, persons or verbs (that is, processes, which include movements); some humans use their skills to identify things also when recognizing humans (prosopagnosia), therefore relying on coarse optical information such as glasses or singular haircuts instead of facial features which the neurotypical would recognize (see Humphreys and Riddoch 2005, 214–215; Shiffrar 2005, 239). It is thus natural to posit that knowledge is about persons, things, and verbs, and that any piece of knowledge can be described by presenting the persons, things, and verbs at stake. Langacker then demonstrates that language remodels the ways we structure our perceptions. For example, the syntax in a particular language can reflect the way humans differentiate between figure and ground. Although not all languages rely on a system with predicates and subjects, most languages do, and all languages have structures for objects, humans, and processes. These grammatical structures thus echo the way humans manage their perceptions.6 || 6 Aristotle’s original idea of proposition, expounded in his Hermeneutics, relies on the combinations of noun and verb. According to his theory, this combination can be either true or false.

Truth in Fiction | 123

Two ways of dealing with perceptions have to be distinguished. They too are poles on a continuum. One the one hand, perceptions can be savored; on the other hand, they can be compacted. I call a perception savored if it is being relished in its details and riches without a comprehensive interpretation of the impressions. Typical examples are: examining a masterpiece painting or enjoying a complex meal. A perception, however, that is being subsumed under general categories (which one has in mind) will be termed compacted. When a general category is at hand, it is usually connected to a specific word or languagelike expression. With respect to the perception of objects, Langacker (2008) speaks of ‘thingification’: Whenever our cognition isolates objects within our perception, these objects become things in our thinking and in our language. Compacted experiences are thus most easily resolved into a linguistic description. They are at the very core of what language can easily convey. It is straightforward to link these notions to the distinction between propositional and aesthetogenic modes of communication. The more a communication aims at compacted perceptions, the more it approaches the propositional end of the spectrum. Accordingly, the more it tries to convey savored perceptions, the more it draws near the aesthetogenic pole. Both communications and memories tend towards the propositional pole since humans have a strong penchant for compacting their experiences. The reason for this inclination is obvious: It is much easier to store many compacted experiences in memory than to remember many details that defy categorization. As Bacon stated in the seventeenth century, the human minds tend to produce abstractions (see Bacon 1990 [1620]; Lévi-Strauss 2009 [1962]; Bunia 2011). Similarly, the psychologists Richard E. Nisbett and Eugene Borgida (1975, 939) conclude that humans tend to infer the general from the particular and refuse to take the general into account when dealing with the particular. All in all, the human mind absolutely prefers abstractions over the complexities of reality. Communications and memories are thus closer to the propositional pole than to the aesthetogenic. However, they do not fully reach it. In ordinary lan|| In the seventeenth century, the idea of proposition became more abstract; the predicate was no longer the process (expressed by a verb in the grammatical sense of the word) attributed to an agent, but it became a list of attributes that were attached to an object. This is when the verb ‘to be’ became to be viewed as the essential way of coupling the subject and the predicate. But it thus lost the link to the way humans process information; here, Aristotle had the better intuition. My critique of propositions can be reduced to the statement that all phenomena in reality cannot be brought into the logical structure of a proposition, nor can human higher cognition be described by ‘translating’ the contents of human reasoning into any kind of linguistic expression.

124 | Remigius Bunia guage, even the most precise words and memories rely on some perceptional associations. This is because cognition needs a perceptional basis to identify an instance of compacted experience; and these traits of recognition usually accompany the memory of the experience. This is why everyday ‘propositions’ that aim at precision nevertheless have noticeable tendencies towards vagueness, and this is also why everyday ‘propositions’ without any need of precision show a great amount of vagueness. In scientific language, by contrast, precision becomes possible because the communication evades the traps of both human perception and human ordinary language. This is why quantification plays a certain role. In chemistry, for instance, we have precise ideas of substances, but we cannot decide which substance we have by means of our senses alone. Science made it possible to escape the fallacies of our senses, and this is why scientific communication can indeed reach the propositional pole. However, the precision attained remains restricted to the particular field it has been developed for. It is impossible to describe my feelings in a fight with a friend by presenting a mathematical or a chemical formula. In all social interactions and in all everyday contexts, a precise language not only does not exist, it would be of little use, since all communication benefits from a certain amount of vagueness. Scientific communication may have established means beyond everyday language that allow scientists to achieve precision, but these means are restricted to the fields the communication is about. Of course, we face different intensities of precision within the sciences with physics doubtlessly achieving the highest degree of linguistic exactitude. The example of physics, which makes use of mathematics in order to attain precision, shows that scientific communication can to some extent rely on modes of communication that transcend language (for instance, when the whiteboard becomes an indispensable tool to put down an equation that is virtually impossible to express by words). Nevertheless, scientists can talk to each other about scientific topics, and they do so quite successfully in many cases. My notion of communication includes both the word-based and the wordtranscending communicative means. Let me resume the argument: Scientific communication is limited to specific scientific purposes. This seeming restraint is the necessary counterpart of the benefit. Only if an expression is bound to a specific use, can it be precise. This is why science was so diffuse until the empirical revolution; and this is why Bacon was so skeptical about theory and language in general. The elasticity of natural language is characteristic of everyday communication. Strongly propositional thinking implements a severe control of the usage of all expressions involved. This is what restricts scientific communication to

Truth in Fiction | 125

such a narrow range of applications. In ordinary language, however, the control is very loose. It is possible to describe new experiences with old words. Yes, the propositional realm invites everyone to compact their experience. But the very feature of being able to express almost everything with great readiness can only subsist at the price of a tendency to redraw the borders of ‘correct’ usage. In ordinary language, the ‘meaning’ of words quivers and changes, both within the same speaker during their life and between different speakers of the same language. It does so because a speaker can and will adjust the usage of their language to every new situation.7 But it is impossible to comment on these shifts of usage in a meta-language, since such a meta-language does not exist yet, and it would spoil the efficiency of natural language. The propositional pole of expression relies on strict control of usage, which would waste cognitive resources and diminish the readiness to adapt to new situations. In fact, ordinary language always and without exception has a tendency to generate a slightly aesthetogenic drive in every utterance. As mentioned above, grammar is modeled after the way sensual perception is processed. Even if one encounters an apparently propositional mode of expression, and even if one employs seemingly clear-cut ‘concepts’, the combination of words suggests an arrangement of ideas that conforms to the structures of perception. What is more, the products of compactification, the nouns and the verbs, aggregate a broad range of possible perceptional experiences; they thus trigger broad association. (The degree to which this happens can of course be controlled by the choice of words. ‘Cat’ is more specific than ‘animal’.) An aesthetogenic ingredient is essential to all linguistic and nonlinguistic modes of everyday communication. It warrants the required syntactical and conceptual suppleness. Nouns and verbs tend to entail imaginations, and this is why a nonscientific proposition can never be precise.8

|| 7 In scientific language, such a shift is only possible with an explicit change of definitions. One cannot just use the expression ‘mass’ to mean something new without announcing the semantic modification. Thus, one will not encounter common ground issues in the sciences in the same way as one incessantly does in the humanities. Of course, someone might be confused by a surprising use of the Einstein summation convention, but this is not an essential problem of terminology, but a well-known and readily explicable choice of communicative means. 8 Let me avoid another misunderstanding. I do not doubt that imagination is essential for scientific reasoning. Quite on the contrary, in all sciences and in mathematics, imagination is fundamental. However, one needs the precision to condense and to communicate the fruits of these adventurous imaginations, and here vagueness does not contribute to the promulgation of new ideas. And again, it is true that in the process of creating totally new ideas, scientific

126 | Remigius Bunia Yet, I am still speaking of utterances staying close to the propositional pole. Certainly not all experiences are compacted as much as possible. Were this assumption true, it would be unthinkable to encounter an experience that seems ineffable – ‘beyond expression’. Some experiences call for a reproduction that can be savored. Although language tends to convey compacted information, it can also make a recipient actually relive an experience. In the latter case, we usually speak of representation. Monika Fludernik’s (1996, 49) idea of a presentation of experientiality concurs with this notion of representation. Not only language, but also pictures and music can produce savored experiences. I will call any unit of communication (a text, for instance) that can be savored representation. Since abuse is always possible and everything can be savored, it is more scrupulous to say: A representation consists of utterances and other pieces of communication that tend towards the aesthetogenic pole of expression. The propositional realm of cognition is shaped by the capacity to compact experiences; the aesthetogenic, however, is characterized by the capacity to arrange and to detail experiences. The idea of a proposition thus highlights the cognitive ability to isolate certain objects and certain actions, to structure our sensual experience by dissecting it into things and verbs. The idea of imagination, however, stresses the cognitive ability to process the full width of sensual details that come with an experience.

4 Representations vs. immediate experience A perception can trigger very different actions, depending on our tastes, our education, and our cognitive abilities. The perception itself can trigger different kinds of response. Among these reactions, there may be such that bear consequences on my future behavior and such that I do not need to take into account. However, the difference between the perception of a representation and the perception of immediate reality is astonishingly small with regard to possible responses. I will substantiate this claim in the present section, even if I will not use the result until the middle of the next section. A thought experiment will clarify the contention. The sight of an apple falling onto the ground can make me

|| communication uses a lot of aesthetogenic devices in order to help others participate in the development of new views.

Truth in Fiction | 127

– – – – – – –

measure the time interval of the dropping, think about the nature of gravitation, think of eating, ponder upon the vanity of all existence, form the words ‘the apple is falling down’, ignore and forget the impression altogether, or use the apple in a game of make-believe.

The last option is the most interesting one. Think of a simple game. Imagine the apples being bombs that target my friends and me while we are running around a tree. Whenever an apple drops and hits the ground (or even one of us) in close distance, the victim has to play dead. Of course, this is quite a silly game. But when we are done with this game, there is no reason why I cannot study the phenomenon of gravitation afterwards – or even within the game if I have the time and the leisure, being the first one to be hit by an apple, forced to lay down and be still. My claim is that perceptions – whether they be verbal or nonverbal – can prompt the same actions. To read ‘an apple is falling down’ can make me visualize the scene. Again, the perception can make me: – – – – – –

think of eating, think about the nature of gravitation, ponder upon the vanity of all existence, form the words ‘the apple is falling down’, ignore and forget the impression altogether, or use the sentence in a game of make-believe.

As far as these points are concerned, there is no apparent difference between seeing an apple and reading about the apple. This similarity is the reason for the widespread tendency to state a functional equivalence between perceiving immediate reality and perceiving a description of reality. The most important similarity is the fact that the sentence can incite me to start operations of a wide range. Yet there are important differences. The most important difference is that I can study the nature of gravitation only by observing apples (that is, by observing nonverbal reality), while reading and thinking cannot give me any valuable hint. Again, this was Descartes’ and Bacon’s complaint about the former tradition. One might interject that today we have great books on gravitation that ground themselves on observations; and yes, it is not a foolish idea at all to read them if one feels the sudden urge to learn something about this fundamental

128 | Remigius Bunia physical interaction. Yet, until today students in the natural sciences are trained to verify many pieces of elementary scientific knowledge in their lab courses. It is important to uphold the Baconian spirit: Thinking and reading tempt the human mind to abandon physical reality and to turn to the deceptive products of human abstractions. Only when I intend to learn something about language and human cognition should I look at words; but if I am interested in nature, I must scrutinize nature.9 The key to understanding the similarities and differences between representations and immediate reality is the notion of information. Even Gregory Currie (1990, 95), who is far from being prone to using ‘positivist’ expressions with regard to fiction, calls the process by which visual media create impressions “presenting us with information”. In fact, all experience can convey information. Information is – to quote Gregory Bateson’s (2000 [1971], 315) ingenuous description – an experience that makes a difference with regard to processing the world. So the question is what counts as information, and what kind of information fiction usually presents us with. But in almost all situations in one’s life, one has to tell apart viable pieces of information from illusive ones. In order to succeed in doing so, one has to use one’s existing viable knowledge acquired during childhood and beyond. In this sense, the analysis of experience relies on former experience – and of course on a long personal history of failures and successes, punishments and rewards. One also learns where to direct one’s attention: which pieces of a sentence may have multifaceted meaning, and which events and things around deserve an examination and an interpretation that surpass the usual degree of contemplation. I will now turn to the effects fictional and factual information exercise on recipients.

5 Influence exercised by fictional and factual representations Can we learn anything from the description of viable information when we ask if fictional representations can contain knowledge or have any influence on us? In order to approach an answer, let us admit that some people read newspaper

|| 9 There are many more important differences. Let me only mention the media in which representations appear. Verbal media are in fact a radical case, as they show a strong inclination towards the propositional pole. Paintings, however, in particular abstract art, tend towards the aesthetogenic pole. This is why literature is the most eccentric case.

Truth in Fiction | 129

articles and novels for the exactly same purposes. They relish both, either for ephemeral entertainment or for gaining knowledge, or for a combination of both. Both trigger the processing of ideas or simulated perceptions. Both invite us to analyze these ideas and perceptions and to distil viable information from them. However, I consider fiction to be more remote from describing reality than a newspaper article. How can I explain my cautious attitude towards fiction? Let me – despite some discord concerning the details – agree with Peter Lamarque and Stein Haugom Olsen (1994) in stating that the identification of fiction depends on a given context and that this identification changes my attitude towards the representation in question. Whenever we realize (from whatever signals) that we are dealing with fiction, we no longer treat the fictional perception as ‘trustworthy’. Equally important, we would not take fictional accounts for ‘mendacious’ either. They may be false, but if they are, they do not deceive. This is different with newspaper articles. If they are false, deceit is probable (the alternative option of explanation being journalistic negligence). It is clear that both newspaper articles and fictional stories can be illusive. Doubtlessly, newspaper articles can also be viable. But can fictional stories be viable in any way? My answer will be yes. And this is precisely the point where caution is required. To claim that fiction conveys some pieces of useful information quite often triggers a response full of disapproval. Such resistance occurs both with those who attribute an empathic value to literature and with those who analytically keep fiction apart from all other modes of communication. First of all, it certainly depends on the context (including genre, recent developments in the arts, and so on) whether a recipient expects a representation to be faithful with regard to the things and verbs themselves, or faithful to the regularities behind them. Education, knowledge, and intelligence can direct the discrimination between viable and illusive pieces of information within fiction. We are taught not to believe in what we recognize when the label ‘fiction’ is invoked (as the default behavior); at the same time we tend to believe nonetheless because disbelief requires effort (see Gerrig 1993 [1988]). Of course, it is simple not to believe in the existence of Alice’s Wonderland, since the physical ‘laws’ in Wonderland differ so much from real-world physics. But it requires much effort not to believe that real-world Romans dressed the way we see Romans do in the movies. This conflict between trained disbelief and natural confidence can be described in a quite simple way. On the one hand, the label ‘fiction’ makes us very cautious with regard to entities and events in the foreground, and less so with the background. The more shaped the piece of information is, the less trustworthy it is. To eliminate the metaphor of ‘shape’ and to put it in the terminology

130 | Remigius Bunia developed above: All compact elements within a representation receive much doubt. The rule is simple: The more compact an element is (individual human persons, pieces of furniture), the less a reader will be inclined to rely on its ‘fidelity’. Fidelity here only means to decide whether any information about the element is viable. One is trained to shun this assumption. On the other hand, a reader is not expected to cast much doubt on descriptions that can be savored. Any experience that is evoked aesthetogenically (which entails that there cannot be any ‘elements’) will simply be accepted. This is why the word ‘fiction’ has seldom been the name of a category to classify paintings; although we have always had paintings that depict entities exhibiting more or less similarity to reality, that is, paintings that bear much resemblance to other representational techniques in that they possess various possible degrees of realism. The point is not that we are not trained, or not aware of possible inaccuracies, since precision in paintings is by necessity aesthetogenic in nature. The problem is that the aesthetogenic mode of communication asks us to simulate the experience, to compact it ourselves and for ourselves only for economy’s sake, and to interpret it. Paintings do not convey very compact pieces of information. In the case of factual representations (such as news reports) the situation is the reverse. It is exactly the pieces of information close to the propositional pole that count as viable. The more aesthetogenic a news report becomes (the more ‘vivid’ perceptional details it describes, for instance), the less it is considered a report. Occasionally, such texts are even considered ‘literary’. If one accepts the notion of literary journalism at all, it surely means that the words and pictures try to express such experiences that are not readily compacted. In all cases, techniques that accomplish an aesthetogenic mode of communication tend to be identified as ‘poetic’. It is noteworthy that poetic language can emerge in very different ways: as an uncommon application of grammar or lexicon, as exuberant expression, as diligence for detail, as beautiful style, as original, as perfectly prototypical language, or many more. In all these cases, I suggest that the language operates within the aesthetogenic realm. This means that language can also evoke the simulation of an experience that spontaneously defies being compacted. Fiction – and again in particular literary fiction – can therefore convey viable information. Of course, the variety of possible uses is enormous. Let us take the indelicate but instructive example of pornography. It is instructive because, contrary to superior forms of fiction, it regularly does not resist being compacted; quite the reverse, the content of pornographic representations can, unless they aim to promote less-known sexual practices, be perfectly captured in the propositional realm without any loss of nuance. Yet, pornographic representa-

Truth in Fiction | 131

tions heavily rely on aesthetogenic techniques; stating simple facts in a propositional style would not have the desired effect. The viability of pornography amounts to effecting sexual arousal, regularly used to facilitate masturbation. The information a pornographic representation provides a consumer with is thus a difference making a difference with regard to the physical state of the body. Let me give another example that displays dissimilarities only on the surface: With regard to basic bodily and mental effects, banal commonplace fiction – stereotype love stories, stereotype thrillers, or stereotype crime stories, for instance – does not differ much from pornography. All these forms of fiction produce excitement and emotions, even though the specific kind of agitation depends on the reader’s dispositions and on the genre. For instance, commonplace fiction allows recipients to use its aesthetogenic qualities to dream themselves away to an entertaining or soothing experience that does not rely on any specific quality of the fictional work. If reluctant to acknowledge the right to relief of this kind, one might call such a use of fiction escapism. An interior escape may be quite a different bodily and mental effect than sexual relief, but from a structural point of view this is only a difference in goals, not in means.10 In all these cases of low-key fiction, the aesthetogenic realm remains predictable. The patterns these works of fiction are made of do not tend to challenge a reader’s intellectual capacities. Advanced fictional works of art, however, achieve a high degree of aesthetogenic novelty and singularity. In these cases, it is less likely to detect any kind of viable information hidden within the fictional work. The result obtained in the previous section of this article says that the perception of a representation and the perception of immediate reality do not differ much in the responses they can trigger. (Again, this may vary between individuals and between contexts, and it surely depends on many factors.) This result is important for the understanding of advanced fiction insofar as rich fiction produces an experience that asks for an extensive interpretation. This interpretation can aim at very different fields of human experiences; it can be mainly sensual, or political, or purely self-contained, or anything else. Following the rich representation through the aesthetogenic realm means that – no matter if the work of fiction intermittently contains compacted propositional segments – it never invites anyone to draw specific (that is, compact) conclusions. However,

|| 10 From an aesthetic point of view, all these forms of fiction do not deserve any attention, of course. Although I draw a sharp distinction between high-key and low-key representations, my approach here is not value-based but ‘structural’.

132 | Remigius Bunia it can open up a whole range of perspectives – often even such that are discordant or dissonant or logically inconsistent among each other. There is much to learn from rich experiences. They present us with settings that may match our current frame of (compacting) thought. But they can also stage a setting that defies our habitual and acquired modes of structuring the world. In particular, in the field of the social, a rich representation can show us places and configurations that question and contradict the reader’s current beliefs and opinions (which are viable knowledge that dictates their everyday behavior). A rich representation might even employ techniques that provide access to a person’s point of view and point of experience – this is Franz K. Stanzel’s (1985 [1979]) personal narrative situation or, in this respect equivalently, Gérard Genette’s (2000 [1972]) internal focalization, combined with the exposition of mental states as described by Dorrit Cohn (1978) in Transparent Minds. In this case, a representation achieves to activate a systematic sympathy with a represented person that can lead to reconsideration of one’s former attitudes towards persons with a similar disposition to this person. For example, a representation can show the situation of the poor, the gay, or the immigrated by departing from a general depiction of ‘their’ situation and, more movingly, by concentrating on a single individual’s fate and feelings among ‘them’.11 None of these experiences could be presented in the propositional realm, simply because this realm requests, prior to the consumption of an act of communication, the mental presence of the very categories into which one aims to embed the situation; we can only compact experiences if we already know how to compact them. In the propositional realm, however, the categories must be present (in the mind or in the language12) beforehand; this follows from the very definition of the propositional realm. A proposition such as ‘the poor live in dreadful circumstances’ does not have much emotional and intellectual effect since it fully relies on existing schemes of structuring. This is why factual de-

|| 11 Representations can present readers with rich details of experiences they have had before. But there is no necessity to this. First, many representations are banal and not rich at all. Second, a particular recipient might be totally unresponsive with regard to representations and never be moved by any one. Third, a particular reader might be well acquainted with the situation in question herself and thus not learn anything; in this case, she might even have the impression that the representation is inaccurate. Rich representations that change some people are thus not at all a regular case. 12 Whether I speak of mind or language depends on the perspective, not on a difference with regard to the phenomenon. In my point of view, a semantic phenomenon resides both in minds and in communicative rules, and one can single out either perspective for analytical purposes only.

Truth in Fiction | 133

scriptions that aim at moving people and changing their attitudes towards society and the world employ techniques in the aesthetogenic realm. Otherwise, they would not have much impact on the hearers. In this sense, it is neither false nor unfair to accuse them of ‘fictionalizing’ and of ‘rhetorizing’ facts. Of course, this accusation comes from those who demand that everyone stay as near as possible to the propositional pole of communication. I am referring to the anti-rhetorical tradition starting in the seventeenth century, the Logique de Port-Royal for instance, with its more recent heirs in analytical philosophy. The anti-rhetoric critique has also attacked those factual representations that include ‘poetic features’ in order to become persuasive. As sketched above, this article primarily deals with the opposite phenomenon: fictional representations that produce viable knowledge, that is, representations that are labeled as ‘not trustworthy’ by social institutions, but that accomplish to change the way we think and act anyway. A fictional representation can array complex and manifold experiences that force us to reshape our schemes simply because we have none that fit while reading. Of course, this only applies to advanced works of fictional art, but this is also why these works are considered ‘intellectually demanding’ and why they are considered advanced. This effect is thus not at all broad but rather restricted to a small group of readers who devote their time to challenging texts and who are willing to interpret the experience. Only then are they forced by the representation to reconsider their attitudes. A more cursory consumption of works of fiction will regularly leave only banal traces, such as (totally inaccurate) knowledge about the clothing customs in Ancient Rome. This is maybe why the famous and popular theory-of-mind thesis about the function of fiction as put forward by Lisa Zunshine (2006) still lacks empirical proof and faces classical critique from philosophers (namely Currie 1998). Of course, both an intellectually unresponsive and quick consumption of a rich fictional representation and an exhaustive reading of porn or shallow love-stories will not lead to any improvement in mindreading. This is also why low-key representations do not receive any attention from academic hermeneutics. They can only be analyzed on the grounds of cultural, anthropological, or sociological assessment. Quite pertinently, they are considered to exhibit existing patterns of compacting experience. As a consequence, such a reading of porn and trivial love-stories can reveal mental schemes prevalent in society. My theory helps to argue that such deductions, widespread in the humanities, are, in methodological terms, perfectly sound. The knowledge acquired in processing the breadth of aesthetogenically generated experiences is by necessity viable. This is because the schemes of structuring immediate perceptional experiences cannot willfully turn off our ability to see what we are seeing, to hear what we are hearing, to taste what we

134 | Remigius Bunia are tasting, and to feel what we are feeling. When one has learnt to identify a tree in the park, one will be able to recognize it in a painting;13 and inversely, when one has acquired the knowledge of a blind person’s way of life through a fictional text, one will imagine a real blind person’s life along these patterns when one meets them. German courts have realized the consequences of this human disposition. The lawsuits that in the recent past effectively prohibit the distribution of novels depicting real people in their private life never base their bans on propositional elements but on the aesthetogenic passages that suggest similarities between the real person and the fictional figure and, and this second point is crucial, that actually show their intimate life (Bunia 2005). It does not matter at all if the ‘propositional content’ is truthful or altered. In fact, the courts state that it is equally wrong if the aspects are significantly altered since the alterations do not prevent any reader to imagine the real person’s private life as depicted in the fictional story. Since humans tend to infer the general from the particular, fictional representations can easily inform about general patterns by showing particular examples. 14 Perhaps this is what Aristotle means in his Poetics (1451a.36–1451b.7) when he states that fiction is more truthful than historiography since it shows the general. Today, it is possible to put it a bit more precisely: Since humans tend to deduce general patterns from individual examples, fiction can demonstrate these patterns and make recipients learn the general facts, even though the character names in the fictional text do not necessarily refer to real persons.

|| 13 In some cases, one also has to learn to decode the specifics of the respective medial transposition. Unfortunately, this aspect has been overstressed due to Ernst Gombrich’s (2002 [1960]) ingenious analysis, which postulates the opposite. As for photography and drawing, the transposition is so easy that even many animals are capable of performing it. This is used, for example, to prevent small birds from flying against transparent windows by pasting paper in the shape of big dangerous birds on the glass. In a way, one can be told the truth in a sense Currie develops. He draws the distinction between two meanings of truth: “Perhaps A la recherche du temps perdu expresses important truths about love, time, and memory. In that case there is much truth in Proust’s novels, and in this sense the truth in Proust’s fiction is genuine truth. This sense of ‘truth in fiction’, which appears prominently in literary evaluations, is not the subject of this chapter. That […] it is true in The Turn of the Screw that there are ghosts: These claims exemplify the kind of truth in fiction that will concern us here” (Currie 1990, 53). It is the other ‘kind of truth’ that I am concerned with at this moment. Au lieu of using the overladen and incendiary word ‘truth’, I stick to the notion of viability, which does not imply that mental states correspond to reality, but denotes that they pertain to reality.

Truth in Fiction | 135

I (Bunia 2007, 150–162) have, in an earlier work, proposed the notion of reality co-principle to designate the possibility to make such deductions, to infer viable knowledge from fictional representations. It is a co-principle to the reality principle as defined by Walton (1990, 145) and others. The latter states that we construct fictional representations, using the real world as blueprint: Our knowledge of reality equips our mental simulations with default qualities of all entities and processes. This is exactly what happens in the propositional realm. There, we rely on an existing and learnt ability to structure reality. The reality co-principle, however, expresses the fact that fictional accounts can give us viable information. It banks on the specific nature of the aesthetogenic realm where an experience generated by a rich representation can make us reconfigure the way we deal with the world. The co-principle thus says that everything evoked by a rich representation and vaguely fitting our contemporaneous viable knowledge can be merged with our viable knowledge – deliberately or inadvertently. Put differently, the co-principle states that (again: whether intentionally or involuntarily) every fictional description may be tested if it can operate as viable knowledge in other contexts. It is important to clarify a tricky point. The reality co-principle does not exclude the possibility that someone also tests some plain statement in a work of fiction for its truthfulness. Of course, if we read in a novel about a specific battle Napoléon fought, we may check if the novel reports historical events accurately. A fictional representation may of course have elements that tend towards the propositional pole; and even then, the reality co-principle remains in force. But, and this is why I extensively expand on the gradual distinction between propositional and aesthetogenic modes of communication, it is easier to consciously doubt the accuracy of information stemming from the propositional realm, while it requires almost impossible efforts to attain the same caution in the aesthetogenic realm. In both cases, some effort is needed. Only the propositional realm provides us with the simple but powerful grammatical technique of negation. This enables a recipient to question the accuracy of a particular element. The aesthetogenic realm, however, knows shades, forms, and continuous qualities only. It is difficult to doubt anything specific, since the aesthetogenic realm does not feature anything one could pinpoint. For instance, it does not make any sense to negate a specific shade of red in a painting or the expression of an unusual form of pleasure in a poem. A recipient who is reluctant to accept the representation can fail or even refuse to go along with the way it works, but cannot negate a specific aesthegenically created experience. Only propositions can be rebutted, but experiences one has lived through cannot. I purport that the reality co-principle is a real-world fact, a fact about human cognition. It can therefore be tested in empirical research, and it should be.

136 | Remigius Bunia I am modestly suggesting some lines along which such research can be undertaken. Until then I am quite confident that despite a lack of empirical evidence my analysis corresponds to reality to a much greater extent than the normative and traditional clear-cut separation of fiction and reality. At least, I see the (non-scientific) evidence in large-scale cultural developments and in smallscale personal experience. Again, this is not to say that the line between fiction and nonfiction is blurry in practice. Of course, we have a whole range of cultural techniques (Olson’s and Lamarque’s institutions) that govern both the identification of fiction as such, and the socially acceptable uses of fiction. What is more, we have a good grip on some intrinsic features of fiction that strike us even when we find a representation devoid of all contexts suggesting fiction. The liminal cases – those which make us waver – such as poetic journalism or documentary fiction may be more instructive; still they are, in direct comparison, statistically negligible. In practice, the difference between fiction and nonfiction is both vital and unambiguous.

6 Fiction reconsidered Since my analysis departs from Walton’s insights about representations in general and not so much about fiction, my results relate to the aesthetogenic realm of human communication rather than to fiction in particular. The aesthetogenic realm may indeed be typical of fiction (although the degrees to which works of fiction approach the aesthetogenic pole vary enormously); yet it is present almost everywhere for flexibility’s sake, and to a more manifest degree, it is there in every factual narrative – whether in the news or in nonpublic accounts. Finally, I am now going to address the distinctiveness of fiction compared with other forms of communication. My assertion is simple. Fiction is a practice that grants communication more liberty than is ceded by other cultural practices. I am thus refuting the conclusions offered by those – widespread – theoretical attempts that seek to detect the specific rules that guide the use of fiction. Contrary to these, I suggest that a work of fiction itself lays out which uses may be possible. Walton describes fiction as a game that is played with rules imposed by a specific prop, and he implies that in many respects the rules are neither conclusive nor complete. Not only are we not told what to do with the work of fiction, but – what is worse – the rules are defined by the things we are playing with.

Truth in Fiction | 137

In how far a work of fiction defines the rules is a matter of degree. Certainly, the more common works of fiction adhere to habitual usages, the more simple theories of readers’ responses will be valid. However, the interesting cases are those works of fiction that demand an individual interpretation. They are not at all rare; the bookshelves are full of them. Niklas Luhmann goes as far as to stating that every work of art – whether or not fictional – sets up its own rules of usage. He speaks of an artwork’s ‘self-programming’ (Luhmann 2000 [1995], 202–210). It follows that fiction draws on two resources: the cultural invention of art with its immense liberty and its self-programming on the one hand, and the powers of massively aesthetogenic modes of communication on the other. Unfortunately, both resources do not simply yield a ‘definition’ of fiction; the problem is that low-key fictional representations such as porn make only sparse use of both resources. Even so, my conclusion is that the cultural and cognitive phenomenon called fiction resides in the plane spanned by the axis of liberty in communication and the axis of aesthetogenic expression. Inversely, anything that is localizable in this plane will readily be labeled as fiction. After all, every work of fiction can be an invitation to fathom the possibilities of its use, to seek viable information, or to enjoy it without any specific purpose. But fiction never promises anything. You are always warned not to trust any of the information you receive, to always be prepared that any fictional representation can mislead you in many ways – even without you knowing. And yet, stay hopeful that you will find some most valuable piece of knowledge in a work of fiction.

References Altmann, Ulrike, Bohrn, Isabel C., Lubrich, Oliver, Menninghaus, Winfried, Jacobs, Arthur M. “Fact versus Fiction: How Paratextual Information Shapes our Reading Processes.” Social cognitive and affective neuroscience 09 (2012). doi: 10.1093/scan/nss098. Aristotle. Poetics. Ed. and translation, Stephen Halliwell. Cambridge, Mass.: Harvard University Press, 1995. Bacon, Francis. Neues Organon / Novum Organum. 2 vol. Ed. Wolfgang Krohn. Hamburg: Meiner, 1990 [1620]. Bareis, J. Alexander. Fiktionales Erzählen: Zur Theorie der literarischen Fiktion als MakeBelieve. Göteborg: Acta Universitatis Gothoburgensis, 2008. Bateson, Gregory. “The Cybernetics of ‘Self:’ A Theory of Alcoholism.” Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution and Epistemology. Chicago: University of Chicago Press, 2000 [1971], 309–337. Bunia, Remigius. “Fingierte Kunst: Der Fall Esra und die Schranken der Kunstfreiheit.” Internationales Archiv für Sozialgeschichte der Literatur 32.2 (2005): 1–21.

138 | Remigius Bunia Bunia, Remigius: Faltungen: Fiktion, Erzählen, Medien. Berlin: Erich Schmidt, 2007. Bunia, Remigius. “Empirie: Zu Francis Bacon, zur Geschichte der mathematisierten Physik und der statistischen Methode sowie zur Hermeneutik als empirischer Wissenschaft.” Scientia Poetica 15 (2011): 29–66. Cohn, Dorrit. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction. Princeton: Princeton University Press, 1978. Currie, Gregory. The Nature of Fiction. Cambridge: Cambridge University Press, 1990. Currie, Gregory: “Realism of character and the value of fiction.” Aesthetics and Ethics: Essays at the Intersection. Ed. Jerrold Levinson. Cambridge: Cambridge University Press, 1998, 161–181. Damasio, Antonio. Self Comes to Mind: Constructing the Conscious Brain. London: Vintage, 2012 [2010]. Fludernik, Monika. Towards a “Natural” Narratology. London: Routledge, 1996. Genette, Gérard. “Discours du récit.” Figures III. Paris: Seuil, 2000 [1972], 65–282. Gerrig, Richard. Experiencing Narrative Worlds: On the Psychological Activities of Reading. New Haven / London: Westview, 1993 [1988]. Gombrich, E[rnst] H[ans]. Art and Illusion: A Study in the Psychology of Pictorial Representation. London: Phaidon, 2002 [1960]. Goodman, Nelson. Languages of Art: An Approach to a Theory of Symbols. 2nd ed. Indianapolis: Hackett, 1997 [1976]. Humphreys, Glyn W., and M. Jane Riddoch. “The Neuropsychology of Visual Objects and Space Perception.” Blackwell Handbook of Sensation & Perception, Ed. E. Bruce Goldstein. Malden: Blackwell, 2005, 204–236. Kahneman, Daniel. Thinking, Fast and Slow. London: Allen Lane, 2011. Lamarque, Peter, and Stein Haugom Olsen. Truth, Fiction, and Literature: A Philosophical Perspektive. Oxford: Clarendon, 1994. Langacker, Ronald W. Cognitive Grammar: A Basic Introduction. New York: Oxford University Press, 2008. Lévi-Strauss, Claude. La Pensée sauvage. [Paris]: Plon, 2009 [1962]. Luhmann, Niklas. Art as a Social System. Translation, Eva M. Knodt. Stanford: Stanford University Press, 2000 [1995]. Nisbett, Richard E., and Eugene Borgida. “Attribution and the Psychology of Prediction.” Journal of Personality and Social Psychology 32.5 (1975): 932–943. Peirce, Charles Sanders. “What Pragmatism Is.” The Monist 15.2 (1905): 161–181. Schiffrar, Maggie. “Movement and Event Perception.” Blackwell Handbook of Sensation & Perception. Ed. E. Bruce Goldstein. Malden: Blackwell, 2005, 237–271. Stanzel, Franz K. Theorie des Erzählens. 3rd ed. Göttingen: Vandenhoeck & Ruprecht, 1985 [1979]. Walton, Kendall L. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Zunshine, Lisa. Why we Read Fiction: Theory of Mind and the Novel. Columbus: Ohio State University Press, 2006.

Ira Newman

Destabilizing Reality Postmodern Narrative and the Logic of Make-Believe

1 Introduction: free and directed imagining In Kendall Walton’s Mimesis as Make-Believe (1990) imagining is transferred from an unstructured environment, where mental agents either initiate acts of imagining or passively experience random imaginative thoughts, to a context regulated by rules, which set out (or prescribe) the imaginings the mental agent is supposed to assume (see Walton 1990, 39). It is this shift to a regulated environment that is the key to understanding the representational arts, where imagining is understood to function under restrictions that, through long established social institutions, are built into the very painting, the novel, the play or the film the audience encounters. And so it is not one’s free imagining that is enlisted when confronting a painting, but a normatively laden act of make-believe, where the painting lays down directions to an appreciator about what is to be imagined if the appreciator is to interact appropriately with the painting. For example, if one sees a particular reddish-shaped paint fleck on a twodimensional canvas, one is supposed to imagine seeing a building’s red roof; or if one reads a particular set of descriptive sentences in a novella, one is expected to imagine a man on his deathbed regretting a life wasted on trivial pursuits. It is through complying with these directions that the audience is able to incorporate normative standards into its imaginative responses; these criteria establish what is legitimate and what is not legitimate in the audience’s cognitive and psychological operations. It is in this way that imagining as a response to the representational arts becomes analogous to truth seeking in the acquisition of knowledge (see Walton 1990, 41). Just as truth embodies a normative standard for what is to be believed (and is independent of the subjectively held beliefs of the inquirer), so the imaginative rule embodies a normative standard for what is to be imagined (and is likewise independent of the subjective inclinations of the imaginer). The success or failure of the imaginer in responding to representational art is a function simply of whether the imaginer has successfully complied with the specific rules to imagine that are embedded in the particular representational artwork.

140 | Ira Newman If we apply this notion of prescriptive (or directed) imagining to a specific literary work, we can say that audiences encountering the words of the play Hamlet may have all sorts of imaginings occurring as they experience the work. Nevertheless, there would be some imaginings that fall into the category of appropriate imaginings (insofar as they were mandated or prescribed by the words of the text) and some imaginings that clearly fall outside that category. Thus, if one were suitably encountering the words in Hamlet, it would be appropriate to imagine that the ghost of Hamlet’s father appeared to Hamlet; but it would not be appropriate in the same situation to imagine that Hamlet killed Claudius immediately after the ghost of Hamlet’s father appeared to him.1 While this analysis may seem unproblematic as an account of traditional representational literary forms (which have held sway for many centuries in Western literature), there are some surprising – yet illuminating – implications when this theory is applied to some of the experimental literary works appearing in the last half century, in particular the group of literary works classified as postmodern narrative. And that is the subject of my interest in this paper, where I turn attention to the novel (or – some might say – anti-novel) Jealousy, written by Alain Robbe-Grillet (1965 [1957]). Specifically, I turn my attention to one section of that puzzling work and attempt to show that Walton’s theory has some illuminating approaches for dealing with the unusual textual passage that com|| 1 See Walton (1990, 60). What term should one use to characterize this duly imagined proposition? ‘True’ is not an accurate term, since the affirmation is not being applied to the actual world. ‘False’ is an irrelevant term, since although correctly characterizing the proposition’s relation to the actual world, the actual world once again is not the object of interest (i. e., we are not interested in whether it is true in the actual world that a ghost appears to a prince). What we are interested in is whether the proposition has been mandated, by the representation, to be imagined; to characterize success here Walton employs the term fictional. So the term ‘fictionality’, for Walton’s system, serves a purpose it does not ordinarily serve, since built into its meaning is not only a descriptive meaning (telling that some state of affairs is fictional), but a clear normative dimension as a term used in assessing levels of success or excellence (such as we find in terms having a dual descriptive and normative meaning, as in ‘right’, ‘good’, ‘just’, ‘healthy’, ‘honest’ and ‘fair’). And so, those propositional imaginings that have been directed by the marks or words of the representational work are now labeled fictional, as in ‘It is fictional that the ghost of Hamlet’s father appeared to him’: meaning it would be appropriate to imagine that the ghost of Hamlet’s father appeared to Hamlet, upon suitably encountering the words in the play Hamlet. Conversely, those propositional imaginings that have not been directed through the marks or words of the representational work (whether explicitly or by implication) are denied the label fictional, as in ‘It is not fictional that Hamlet killed Claudius immediately after the ghost of Hamlet’s father appeared to him’: meaning it would be inappropriate to imagine that Hamlet killed Claudius immediately […], upon suitably encountering the words of the play Hamlet (See Walton 1990, 60).

Destabilizing Reality | 141

prises that section. My speculation is that Walton’s remarks can present promising frameworks for dealing with other postmodern works as well.

2 A paradoxical postmodern text So here is a curious section from Robbe-Grillet’s novel, Jealousy. The passage presents a series of sentences appearing to describe the plot of a novel: (M) The main character of the book is a customs official. This character is not an official but a high-ranking employee of an old commercial company. This company’s business is going badly, rapidly turning shady. This company’s business is going extremely well. The chief character - one learns - is dishonest. He is honest, he is trying to re-establish a situation compromised by his predecessor, who died in an automobile accident. But he had no predecessor, for the company was only recently formed; and it was not an accident. Besides, it happens to be a ship (a big white ship) and not a car at all. (Robbe-Grillet 1965 [1957], 137)2

Now there are no less than five pairs of inconsistent sentences here. This means that the states of affairs that have to obtain for both propositions in each respective pair to be simultaneously true would be, logically speaking, impossible: For example, in the first pair it is not possible for the main character’s being a customs official and the main character’s not being a customs official both to obtain.3 So, if logical contradictoriness is at the heart of this passage, what is it (to invoke Walton’s language) that the words of the text are prescribing (or directing) an imaginer to imagine? There seems to be a puzzle here. Walton (1990, 64–66) does make some remarks that directly bear on this problem by acknowledging that prescriptions to imagine can result in inconsistent propositions. He is careful, however, to distinguish between two kinds of prescriptive situations leading to this result. As it turns out, this distinction will at first create even more questions, but in the end it will lay the foundation for a resolution of the puzzle – or at least that is what I will argue. || 2 Curiously enough, this excerpt from the novel appears as an epigraph to an important section in Mimesis as Make-Believe (Walton 1990, 63). I say ‘curiously enough’ because I had scoured Jealousy (which I had read years before) for just such a passage, without realizing that Walton, too, had zeroed in on exactly the same one. 3 A proposition is what is proposed or stated in a declarative sentence and is capable of truth or falsity. The same proposition can be expressed by two different sentences if the sentences are synonymous in meaning or intertranslatable. I owe the technical terms involving propositions, states of affairs and possibility to Plantinga ([1979, 426–429]; [1974, Ch. 4]).

142 | Ira Newman

(a) In the first kind of situation the prescriptions are made separately, as if in distinct moments: We are directed first to imagine p and then subsequently, we are directed to imagine not-p. There is no overt connection between the two. (b) In the second kind of situation, however, the prescriptions are conjoined, and we are directed to imagine p and its denial, not-p, as a conjunction (p & not-p). This changes the meaning considerably since it requires us to imagine both p and not-p as true at the same time. Let us consider each situation in turn and see how useful it may be in formulating a prescriptive understanding of section M.

2.1 The separate prescriptions approach The virtue in what we can call the separate-prescriptions approach is, as Walton (1990, 182–183) indicates, that it avoids a “sense of paradox” by “mut[ing] the clash” between inconsistent propositions: It does so precisely “by disallowing the fictionality” (or the mandate to imagine) “their conjunction”.4 If the text mandates that we imagine p and then subsequently mandates that we imagine not-p, these two independent mandates do not entail a (third) mandate to imagine something entirely different, namely the conjunction of p and not-p. If this third mandate were the case, we would have a genuinely paradoxical situation, since (according to truth-functional logic) to assert that p & not-p is true entails || 4 As a preliminary move, we could search other sections of the entire text to see if an interpretation is available that does in fact comply with the separate-prescriptions approach, yet in an entirely innocuous manner logically speaking. Searching broadly through a text – in order to understand what a text is prescribing that we imagine at any particular representational point – is essential, since the individual sentences are not just atomic entities, but exhibit a holistic relationship among the various sentences comprising the text (See Walton 1990, 145–146). Walton (1990, 174) himself urges that “fictional truths [...] may be mutually dependent”, thereby requiring us to “go back and forth among provisionally acceptable fictional truths” until a “convincing combination” is found. Yet in this particular case, this process leads nowhere, since while there is in fact an earlier section in the novel (Robbe-Grillet 1965 [1957], 74–75) where the two central characters (A. and Franck) are discussing a novel they are reading, and suggesting alternative (and conflicting) “probable outcome[s]” – which indeed is what M itself can be understood to be setting out, it is not clear that M is in fact dealing with the novel they are talking about. So that pathway leads to a blind alley.

Destabilizing Reality | 143

that both of its components (namely p and its denial, not-p) would also be claimed to be true, simultaneously. So if the separate prescriptions approach can apply to section M, so much the better. There is a way textual section M could model this logical format. If we adopt the standard distinction between the story (or the events the narrative is talking about) and the discourse (or the way the narrative is talking about these events), we could conceive the discourse to be presenting a single narrative voice, which happens to be in the process of formulating a story.5 It does so by simply trying one formulation of a propositional statement in the story, then its contradictory, then another propositional statement, then its contradictory, and so on – but without any final resolution in its overall process of composition. If this is plausible, we could reformulate the propositional mandate in this way: Imaginers are being directed, by the words of Jealousy, to imagine that there is an unidentified narrator (or narrative voice) formulating a sentence s, then a sentence not-s, a sentence t, then a sentence not-t, etc. This will remove any hint of logical incoherence. The problem with this attempt to reduce a potential paradox to a coherent naturalistic view of the narrator is that it is not borne out by the text. From what we can tell in a close reading of the text, as well as from the work of some careful interpretive critics, there is no indication of a narrator undergoing a creative compositional process. In fact, there is no indication of a narrator at all; or even a disembodied (and fantastic) narrative voice (which we might attribute to some of Samuel Beckett’s works); or even – following a still weaker interpretive model – a group of separate narrative voices, each with its own alternative in the respective contradictory pair.6 || 5 For the discourse-story distinction see Chatman (1978). 6 Richardson (2006, 104) objects to reaching for “naturalistic assumptions to comprehend” texts of this kind, for which there is “no mechanism offered (such as different narrators with different memories and agendas) to explain away the often outrageous contradictions”. Richardson (2006, 87) calls this maneuver of sequential contradiction denarrative and although it is used by some authors (e. g., Nabokov) in a limited way, where “the stability of the represented world is not seriously challenged”, in Beckett and Robbe-Grillet “the denarration is global and undermines the world it purports to depict” (Richardson 2006, 91). Abbott (2008, 169) offers a similar objection to naturalizing a Robbe-Grillet text containing a string of contradictory sentences: In this case it is the novel, In the Labyrinth, and the naturalizing attempt is to suggest it is the expression of the deranged mind of a narrator protagonist. Another reason to deny such interpretive maneuvers is that section M appears quite surprisingly toward the end of the text, with similar sets of contradictory sentences appearing only sporadically through the course of

144 | Ira Newman Suppose we resist this view by proposing that it makes no sense to conceive a narrative without a narrator: If there is a narration, there is the presumption of a narrator (however far in the background) who speaks, writes or thinks the words of the narration. But it would be incorrect to offer this objection. Under the prescriptions model, narrative texts (without narrators) can confront the audience, with no accompanying difficulties about what the audience is expected to do. As Walton (1990, 365) suggests, the text’s basic prescription is that the audience imagine the propositional content expressed by the sentences of the text; but following this direction does not entail that the audience is also directed to imagine a narrator. This disconnect between narrative and narrator is particularly evident when an alleged speaker, writer or thinker of the text’s words is so thin that we have no idea of ‘what sort of person’ he or she is; in Walton’s (1990, 365) view there would be no point in cases of this sort in “insisting that there are narrators”. Things can change dramatically, however, depending on the textual elements that become perspicuous. When the text makes it evident that events are described in tones of contempt or irony, for example, it might be useful to imagine a person (that is, a narrator) who is expressing these attitudes toward the events narrated (see Walton 1990, 366). In this case there is a pragmatic value in postulating a narrator, in order to achieve a sensitive critical interpretation of these aspects of the text. Yet that is what is lacking in Robbe-Grillet’s textual section M. There simply are insufficient textual indications – even on the most generous interpretation – of an attitude, feeling or viewpoint that is expressed toward the events described. So there are no pragmatic reasons to presume a narrator behind the text, one who might save the text from logical incoherence through uttering the contradictory sentences as a set of separate and logically unconjoined propositions. For these reasons, the separate-prescriptions approach appears to fail if we want to remain faithful to the text and its most persuasive critical interpretations.

|| the text’s progress. In fact, for the vast body of the text, there is no indication of revision at all, but only a surefooted description of what is occurring through the largely visual scenes presented. If there is a question about comprehending the text, it lies mainly in the iterative, circular and non-progressing narrative development, which is a pervasive formal feature of the entire text. Readers are trying to figure out not only what is taking place in the alleged story, but just as significantly, what the point is in being told the things that are conveyed to them about rows of banana trees, squashed centipedes and settings at the patio table.

Destabilizing Reality | 145

2.2 The conjunctive approach If we cannot find a solution in the separate-prescriptions approach (a), it looks as if we are forced to consider the conjunctive approach (b), that is, p & not-p. This is the approach we should be trying to avoid at all cost, since the logical clash between the propositions now confronts us with the full power of its demands on the reader. We can illustrate this by forming a compound proposition drawn from one pair of propositions included in section M: ‘This company’s business is going badly’ (which we can name p); ‘This company’s business is going extremely well’ (which we can name not-p).

Let us furthermore call the new compound sentence we are forming, C (consisting, as it does, of the conjunction, p & not-p). So now we have: C: ‘This company’s business is going badly and this company’s business is going extremely well’.

We can easily see that C is a contradiction: For to affirm p is true (in the first of the two conjuncts) prohibits affirming p is not the case (in the second conjunct).7 Moreover, logically speaking, we can say that the contradiction is not only false, but that it is necessarily false: which means that the state of affairs to which C corresponds (the company’s business both going badly and extremely well) not only does not obtain (which expresses C’s falseness), but is impossible to obtain (which expresses C’s necessary falseness). There simply is no possible world in which this state of affairs could obtain. But if this is so, we are as enmeshed as ever in the tangles of the puzzle we started out with; for what is sentence C prescribing the imaginer to imagine (to invoke Walton’s language again) if the conjunction is so mangled in its proposi-

|| 7 Truth-functional logic holds that if a conjunction of two propositions (p & q) is true, both of its constituent propositions must also be true. So if p is considered true and if not-p is also considered true (which we are presuming as the truth status for all the sentences of M), their conjunction (p & not-p) should also be considered true. But since p & not-p is a contradiction (for if p were true, its negation not-p cannot also be true), p & not-p can never turn out to be true, since one of its conjuncts will always turn out false. So p & not-p is necessarily false, in terms of its logic.

146 | Ira Newman tional content that a “sense of paradox” (Walton 1990, 183) will have been projected into the ‘world’ (or propositional cluster) the representation generates?8 There is an available line of argument that can be pursued here. As Walton (1990, 64) points out, fictional worlds are not possible worlds. Possible worlds technically must not include both members of an inconsistent pair of propositions, such as the propositional constituents of C (dealing with the company’s business going both badly and extremely well); only one of these propositional constituents may be included. But exemption from such strict demands is one of the (desirable) freedoms granted to imaginers of fictional worlds. And we are quite familiar with countless instances where imaginers of fictional worlds can be prescribed to imagine logically or empirically impossible states of affairs: These include, humans transformed into frogs or giant bugs (as Gregor Samsa experienced in Kafka’s Metamorphosis), or people going back in time (as Scrooge did in A Christmas Carol).9 Moreover, as Walton notes, somehow audiences are not bothered by these problems in inconsistency or impossibility, which “they accept without batting an eye” (Walton [1990, 66]). Assuming these reports of audience behavior are true, two questions about their implications emerge. First, if it is the case that audiences are not bothered by inconsistency or impossibility in these sorts of imagining, what is the explanation for this odd result? What qualifications are audiences incorporating in their psychological and interpretive responses that allow them to be modally so carefree? Second, how do these familiar cases of imagining impossible state of affairs (and our understanding of how audiences deal with them) relate to the Robbe-Grillet text, M?

3 Visualizability Let us start with visualizablity and ask whether visualizability could provide a clue to why, for example, we can imagine Scrooge traveling back in time and sitting next to a younger version of himself, left alone at a boarding school at

|| 8 Walton (1990, 62–63) proposes, at one point, a conceptually lean sense of ‘fictional world’, referring only to the “class or cluster of fictional truths belonging to it”. But he does go on to point out how in some cases we might think of fictional world as referring to a ‘place’, as in “it is fictional that there are unicorns someplace else – at a ‘fictional place’ which we might think of as a ‘fictional world’”. 9 See Walton (1990, 64; 66).

Destabilizing Reality | 147

Christmas?10 Or why we can imagine Gregor Samsa (in The Metamorphosis) waking up one morning to discover he has been morphed into a giant bug overnight, while still retaining all the memories and emotions of Gregor the man? 11 We surely seem able to visualize these states of affairs – as impossible as they may be, in that we imagine ourselves seeing these states, or imagine how these states might visually appear to a perceiver (who is not necessarily ourselves). Undoubtedly this is in no small way due, in each case, to the writer’s literary powers to evoke a highly visualizable scene for us to imagine seeing.12 And yet one could argue that visualizability is a red herring if we think we are going to learn anything useful from the phenomenon about why imaginers accept without batting an eye the violation of empirical and ontological laws. One reason for its questionable utility is because it is not clear how to describe what it is that is being visualized. This is important because if visualizing were to amount to genuinely imagining a particular impossible state of affairs, one would have to be visualizing (in the Metamorphosis case) a bug-human hybrid in some way. But suppose I visualize (or imagine seeing) a bug squirming in a human bed and imagine hearing a poignant query ‘what has happened to me?’ Is it so clear what it is that I am visualizing? Am I indeed visualizing a bughuman hybrid, as the standard I have proposed requires? Or is it something else instead, something that does not fill the bill? For instance, am I visualizing merely a dreamer’s dreamed image of a bug (as in an Henri Rousseau painting – || 10 “They went, the Ghost and Scrooge, across the hall, to a door at the back of the house. It opened before them, and disclosed a long, bare, melancholy room, made barer still by lines of plain deal forms and desks. At one of these a lonely boy was reading near a feeble fire; and Scrooge sat down upon a form, and wept to see his poor forgotten self as he had used to be” (Dickens 1986 [1843], 26). 11 “When Gregor Samsa woke up one morning from unsettling dreams, he found himself changed in his bed into a monstrous vermin. He was lying on his back as hard as armor plate, and when he lifted his head a little, he saw his vaulted brown belly, sectioned by arch-shaped ribs, to whose dome the cover, about to slide off completely, could barely cling. His many legs, pitifully thin, compared with the size of the rest of him, were waving helplessly before his eyes. ‘What’s happened to me?’ he thought”(Kafka 1972 [1915], 3). 12 I am not claiming the text is prescribing that I visualize the states of affairs to which the sentences point. The text prescribes merely that I imagine the states of affairs in question. But since frequently imagining is causally connected with acts of visualizing (my act of imagining may easily trigger a mental state of visualization), we tend to think of successful visualizing as a sufficient indicator (or test) of successful imagining. Thus, if I can visualize (or imagine seeing) Gregor Samsa morphed into a bug, I infer I can imagine that state of affairs. Since beliefs about the alleged causal connection between imagining and visualizing appear so deeply embedded in our epistemological frameworks, it is worth inquiring to determine if the test is a valid one.

148 | Ira Newman on which Walton (1990, 170–171) has commented)? Or am I visualizing simply a bug, and along with that visualization, an oral (and quite unvisualizable) monologue by a man, Gregor Samsa? Without further guidance any one of these three possibilities is a plausible characterization of what I am visualizing; yet the last two are insufficient for establishing any conceptual connection between visualizing and imagining an impossible state of affairs. The Scrooge case is no different. If I visualize two individual humans occupying two separate locations in the same room – one having the visual appearance of a younger Scrooge, the other the visual appearance of the mature Scrooge, is it clear that I am visualizing the impossible state of Scrooge viewing a younger version of himself? Or is it that I am visualizing two different and independent individuals whom I think of as two temporal slices of the same person? This last question has expressed the problem – but also inadvertently suggested the solution – in dealing with the visualizability test. What my previous questions suggested was that there is an inherent indeterminacy in visualization, because by itself visualization cannot lead to a clear determination of the identity of what it is that I imagine myself seeing (that is, the intentional object, or visual content, of my mental act of visualizing). It is only when an interpretation, or a way of taking these images, is applied to the process, that what is visualized acquires a clear identity. We can visualize Gregor Samsa morphing into a bug only because we think of the visual images we are having as images of Gregor Samsa morphing into a bug – and not as images of one of the other candidates I have offered, which a correct reading of Metamorphosis (and its prescriptions to imagine) would disallow.13 What this means is that it was not the visualization itself that led the way to the imagining, but the interpretation that presented a guide for how to ‘read’ the images. In essence, it was the reading that counted and not the imagined seeing; or, more charitably, if the visualization were to count for something, it could not do so without the determining effect of the meaning that, once grasped by the imaginer, indicated how the images should be taken. If the projecting of meanings into the intentional objects of imaginative acts is the critical feature, and not visualizability (which is merely a contingent feature, as well as a dead end for theorizing), then meaning projection may be the key to understanding how the imaginer confronts the logical and empirical inconsistencies, and impossible worlds, in the cases at hand.

|| 13 I owe this point to Tidman (1994, 300–301).

Destabilizing Reality | 149

4 Work meaning One of Walton’s approaches to representational works is to suggest the global process through which prescriptions to imagine are generated. Prescriptions are not typically generated in a linear manner through following one atomic unit and then the next in the textual sequence, but as a result of an extensive ‘feedback” system, which derives “from one’s overall assessment of the work’ (Walton 1990, 184, 174, 145–146). The resources for achieving this are distributed widely throughout a representational work and when brought to bear on local elements, they frame the array of meanings that an audience can apply to the work. These resources include: conventions of the medium (Walton 1990, 165– 166), the genre and tradition to which the representational work belongs (Walton 1990, 184, 186), the work’s overall themes and morals (Walton 1990, 184), the artist’s presumed intentions (Walton 1990, 183), even knowledge of the outside world (Walton 1990, 184, 186, 171). All influence how imaginers are to determine what the work is prescribing the imaginer to imagine. It is this overall feature – which I shall call work meaning – that can address our original question of why imaginers can accept without batting an eye the violation of empirical and ontological laws. Take the Scrooge case. Instead of viewing Scrooge’s adventure in time travel as just an odd deviation from logical or empirical norms, the work (in the form of the story line of A Christmas Carol) suggests certain meanings based on what the story line is about: It is a tale of redemption deriving from a review of a person’s past history. This theme ties into certain larger wishes – fantasies even – that we harbor (as human beings) to review and correct the past in an effort to set things right: It is part of our distinctive moral and psychological makeup. An enticing imaginative way to express a wish of this sort is achieved through converting a review of the past, which we perform by recounting events as we recall them in memory, into a literal re-view of the past, performed through a quasi ‘journey’ to the past where we can actually view events once again. For if such a ‘visit’ could be accomplished, hindsight (along with our knowledge of how things did in fact turn out) could enable us to scrutinize past events from a more enlightened perspective than the limited one we were forced to accept in our first (and only) experience of them. In this way, travel to the past, which from a literal standpoint defies all that is logically or naturally possible, acquires a quasi-allegorical dimension through expressing or signifying our longings to revisit now logically unreacha-

150 | Ira Newman ble (yet haunting) events.14 The impression of paradox, which the representation of time travel should be expected to elicit from the audience, becomes deflected by an appeal to the meaning time travel has for those who have been taught to think about it along certain iconographic or symbolic lines.15 But deflection is an essential feature here, because the paradoxes connected with time travel and splitting the self still remain (see Walton 1990, 183). What really matters is the relevance of these paradoxes, however; and when the power of meaning, as expressed by the work itself, transforms the impossibilityladen prescriptions into vehicles for symbolism, interest in the particular paradox is overridden by a consideration of its meaning.16

5 Imaginative failure What can we say now about section M in Robbe-Grillet’s text? I find two important statements in Mimesis as Make-Believe that can finally suggest a frame|| 14 Ultimately it is these fantasies, in symbolically infusing their moral and psychological meanings into the representations of time travel, that enable audiences to imagine the content of what would ordinarily be viewed as a metaphysically incoherent set of events. So (in Walton’s terminology) what the audience is directed to imagine is not simply a state in which Scrooge sees a younger version of himself occupying a spatial position different from the one he occupies at their fantastic meeting point in the past, but an allegory-laden state in which these events unfold. 15 I say “taught to think about” these paradoxes, in a wide open sense of ‘taught’. This teaching process could include many implicit forms of teaching, through cultural and linguistic conditioning, say, where symbolic associations can become as easily a part of a person’s ways of ‘seeing the world’ as the concepts (e. g., causality, mind, justice) that post-Kantian philosophers indicate become constitutive of a person’s conceptual framework for grasping the world. Empirical studies into the cognitive and cultural development of these symbolic associations may illuminate this process; but equally, critical studies of these symbolic associations (conducted in literary and artistic iconography, as well as in psychoanalytic and linguistic studies) could contribute greatly to our understanding. The latter process may have affinities with the model Robert Brandom (2000, 56) (following Wilfrid Sellars) projects onto the socratic method, as it is employed in philosophical analysis: It “make[s] explicit the rules we have adopted for thought and action” – in this case, the rules or practices for symbolic thought and action. 16 Using the example of a dreamer experiencing an ontologically incoherent dream (where the same person appears on different occasions as both Joan’s father and Joan’s boss), Walton (1990, 177) admits the paradox is still there, yet the “true character” of the dream – “what is important about it” and “an understanding of it” – will only be pointlessly interfered with “by dwelling on” the anomalies. That is to say, it is the symbolism of the dream that counts and not its logical deviations.

Destabilizing Reality | 151

work for dealing with this troublesome textual section, as well as promising theoretical guidelines for dealing with other postmodern narrative texts. A) Walton distinguishes two sorts of paradox-laden representational works. First, he acknowledges those (like the aforementioned Scrooge and Metamorphosis cases) whose underlying paradoxes are overridden by what I have been calling work-meaning. But there is also a second group of paradox-laden representational works that Walton (1990, 183) directs us to: those whose underlying paradoxes are not at all softened by work-meaning. On the contrary: In the latter category, representational works “highlight their [logical] anomalies, and appreciators relish them” (Walton 1990, 176). These are works such as Hogarth’s False Perspective (reprinted in Walton’s text [1990, 65] and commented on by him), where (to invoke Walton’s terminology) the representation (due to the optical illusions generated by perspectivally drawn figures) appears to prescribe imagining “that a woman is and is not within reach of a man whose pipe she is lighting” (Walton 1990, 147). Escher drawings also come to mind (see Walton 1990, 176). In these cases the questions the works raise about their paradoxical underpinnings are legitimate ones and central to appreciating them, since that is their whole ‘point’ (Walton 1990, 176). Their paradoxes are not softened by an embracing canopy of meaning; instead they “wear [their] paradoxes on their sleeve” (Walton 1990, 183). So here is one way to grasp Robbe-Grillet’s section M and sentence C (“This company’s business is going badly and this company’s business is going extremely well.”) The search for a way to deflect attention from the logical waywardness of this necessarily false proposition (p & not-p) or from the impossible state of affairs to which it corresponds is not only a hopeless task, but (if we follow Walton) one of the very points the text is generating for the audience of imaginers.17 B) Now we approach a more stark conclusion. Walton’s approach reminds us what the basic formula in his system is. The concept of mimesis (under Walton’s view) derives not from a correspondence framework, where a set of propositions is conceived to correspond to (or mirror) an antecedently structured world: That would (arguably) be one way to understand truth-seeking in the actual world, and it would also be a traditional way to understand mimesis in || 17 With the depictive representations of Hogarth and Escher things might not be so bleak, since these representations are pictorial, and have a richly detailed visual dimension – which may generate a more full-bodied imaginative response by the imaginer. One can only speculate if there had been a counterpart to such pictorial renderings, in the form of a similarly clever visual rendition of the content of the Robbe-Grillet text, would that have led to a less desiccated imaginative response?

152 | Ira Newman the representational arts. Under that view a representational artwork models or depicts some general aspects of human life (such as types of human action or kinds of discourse – whether in the form of personal speech, narrative voice or written history). While this view may successfully characterize most realistic forms of representational art, it fails in the case of postmodern works, especially the problematic text M. In the latter, it is precisely the realistic relationships connecting representational artwork and human action, mind and speech that are undermined and rejected. And it is for these situations that Walton’s view – which conceives mimesis as a prescriptive rather than a modeling relationship – offers a more flexible critical instrument for reconstructing the imaginative encounter with the literary artwork. For if mimesis derives from a prescriptive conceptual environment, where the marks or words in the representation prescribe that the imaginer should imagine something, there are no limits on what the content of the imagining must be: It could be anything, whether realistic or not. Moreover – and to the specific point on which we are focused here – prescribing or directing that we imagine something does not even entail that the imagining is realizable. In Walton’s view (1990, 64) “there can be prescriptions to imagine a contradiction even if doing so is not possible”.18 In the end it is this dark interpretation of M that wins the day. For there appears no good critical analysis that allows an interpretation of the contradictions of M that will alleviate the paradox by finding a work-meaning solution. There is no moral or human dimension of meaning to rescue the bewildered intellect from its relentless confrontation with logical contradiction. And so there is nothing left to do but to focus on those contradictions, which curiously, in the hands of a stylist such as Robbe-Grillet, can become a positive literary experience through leading the audience to attend to what it would be like to have a fictional world shorn of every vestige of logical, moral or any other structure of meaning.19 The representation directs the imaginer to conduct an act of

|| 18 Are we not leaping into another paradox here? Can one prescribe what one knows is not capable of being executed? If a so-called prescriber knows it is impossible to carry out the prescription, shouldn’t this person be characterized as having failed to offer a prescription at all? Clearly there are conceptual issues at work here. But a correct analysis might be that it is in the trying that one meets the criterion of doing something, however the trying is carried out, and however successful one is in completing the doing in its most optimal form. 19 Stylistically, it begins with the words, which describe objects and events with just enough meaning to hint that these do have narrative histories or functions within human life, but with never enough depth to suggest what those histories or functions are within the particular situation presented. As a result, objects and events are described with a rigorously maintained

Destabilizing Reality | 153

imagining that he cannot complete – possibly cannot even begin. Like a John Cage silence it is the space where instrumentally produced sound does not appear. Indeed that is the way M works as you read it. However mysterious the remaining parts of the literary work Jealousy is (with its repetitions, themes and variations, dull and dry descriptions of a banana plantation and other objects in the physical environment), when M occurs, it shuts down everything, and in Walton’s language we become aware of a prescription to imagine that not only cannot be accomplished, but cannot be visualized. In Robbe-Grillet we look into the space of a fictional null world, where whatever hint of a story or character or metaphor or image is presented to us winds up merely a tease that evolves into nothing.

6 Concluding remark To conclude, let me say that it is this sense of nothing that is at the heart of the title of my paper. As a result of interacting with Robbe-Grillet’s postmodern narrative text, audiences undergo a destabilized sense of the traditional storytelling form against which this postmodern text seems quite forthrightly juxtaposed. Coherent story lines, psychologically integrated characters and narrators, even unified and recognizable narrative voices – all standard features in traditional storytelling – are thrown into doubt, if not abandoned outright, in this kind of literary work. It is understandable why the audience is left in a paralyzed interpretive state, with the difficult project of finding effective strategies for managing these new narrative realities. Walton’s analysis of prescriptive imagining, along with the possibility of breakdown in these very prescriptions to fulfill their mandates to imagine, opens a critical framework that promises a way to deal with such elusive and resistant texts.

References Abbott, H. Porter. The Cambridge Introduction to Narrative. 2nd ed. Cambridge: Cambridge University Press, 2008. || focus on their outermost appearances only, limiting the reader’s attention merely to the surfaces of these things – whether the surfaces are understood in terms of spatial, temporal or even narrative categories. See the illuminating analysis by Roland Barthes (1965).

154 | Ira Newman Barthes, Roland. “Objective Literature: Alain Robbe-Grillet.” Translation, Richard Howard. Two Novels by Robbe-Grillet: Jealousy and In the Labyrinth, by Alain Robbe-Grillet, 11–25. New York: Grove Press, 1965. Originally published as “Littérature objective: Alain RobbeGrillet” (Critique 86–87 [juillet-août 1954]: 581–591). Brandom, Robert B. Articulating Reasons: An Introduction to Inferentialism. Cambridge, Mass.: Harvard University Press, 2000. Chatman, Seymour. Story and Disourse: Narrative Structure in Fiction and Film. Ithaca, N.Y.: Cornell University Press, 1978. Dickens, Charles. A Christmas Carol. 1843. Reprint, New York: Bantam Books, 1986. Kafka, Franz. The Metamorphosis. Ed. and translation, Stanley Corngold. New York: Bantam Books, 1972. Originally published as Die Verwandlung (Leipzig: Kurt Wolff Verlag, 1915). Plantinga, Alvin. “God, Possible Worlds and the Problem of Evil.” Philosophy as It Is, edited by Ted Honderich and Myles Burnyeat, London: Penguin Books, 1979, 417–446. Originally published in God, Freedom and Evil, by Alvin Plantings (New York: Harper & Row, 1974), 29–55. Plantinga, Alvin. The Nature of Necessity. Oxford: Clarendon Press, 1974. Richardson, Brian. Unnatural Voices: Extreme Narration in Modern and Contemporary Fiction. Columbus: Ohio State University Press, 2006. Robbe-Grillet, Alain. Jealousy. Translation, Richard Howard, in Two Novels by Robbe-Grillet: Jealousy and In the Labyrinth. New York: Grove Press, 1965. Originally published as La Jalousie (Paris: Les Editions Minuit, 1957). Tidman, Paul. “Conceivability as a Test for Possibility.” American Philosophical Quarterly 31.4 (1994): 207–309. Walton, Kendall L. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990.

Jukka Mikkonen

The Place for External Considerations in Reading Literary Fiction 1 Introduction The theories of fiction formulated by Kendall L. Walton, Gregory Currie and Peter Lamarque together with Stein Haugom Olsen define the audience’s appropriate attitude toward a fictional work in terms of make-believe. When reading fictional literature, these philosophers propose, readers are basically to make believe the propositional content of the work and its truth and reference. Nevertheless, imaginative engagement with a work of literary fiction commonly involves, and sometimes arguably necessitates, readers reflecting on the fictional content in different ways in relation to reality. The philosophers mentioned maintain that there are two different ways to approach and speak of the content of a work of fiction, such as true propositions or apparent authorial assertions in the work; Walton and Currie make a distinction between imagination (make-believe) and belief/disbelief, whereas Lamarque and Olsen speak of the audience’s internal and external perspectives to the fictional content. In this brief essay, I shall discuss readers’ imaginative engagement with literary fiction and the place for external considerations in it.1 I shall propose, first, that the external perspective to fictional content may be seen to involve different mental attitudes and take place in different stages of literary interpretation, and second, that the external considerations need not be insensitive to the work’s status as a work of fiction or a literary work. By imaginative engage|| 1 The interest in the role of external considerations in literary interpretation is often related to the question of the cognitive value of literature, a matter that also underlies this essay. Of course, one may argue for the cognitive value of literature without referring to the external perspective which readers may take on the fictional content, or end products such as knowledge. For example, Walton and Currie both find that engaging in a game of make-believe may be, to some extent, cognitively valuable, as it may enhance the imaginer’s understanding of herself and others, see e. g. Walton (1990, 35, 272) and Currie (1995, 55–56; 2005, 219). In turn, Noël Carroll (1998, 141–145) argues that the act of comprehending the narrative, which requires the reader to mobilize the knowledge she possess, has cognitive gains as it enlarges the reader’s understanding of things by building and reorganizing conceptual connections. Carroll thus maintains that the cognitive gains of literature are an integral product of imaginative engagement with the works.

156 | Jukka Mikkonen ment, I mean a variety of responses which take place not only when one is reading the work or following the story, but also after the work is finished, and which all include reflecting on, or thinking of, the story content, or entertaining its sense as (literally) unasserted. I thus use the term in a broader sense than vivid, pretence-like imagination, which I think is a substantial, although partly genre-relative, form of imaginative engagement with fiction.

2 Fiction and reflection Imaginative engagement with a work of fiction characteristically involves readers reflecting on the content of the work, that is, sentences, passages or utterances in the story, in relation to reality, or drawing on their beliefs about the actual world. Arguably, different literary genres emphasize different kinds of imagining, and different modes of reading set different interests and significance for so-called worldly considerations, that is, for reflecting how the content of the work figures outside the fictional realm. The elementary need for the drawing on actual world beliefs in imaginative engagement with fiction relates to implied fictional truths and imaginative supplementation, in which readers have to fill in the gaps in the narrative, partly with their knowledge about reality, in order to make the story coherent. This activity is however to a large degree spontaneous and unnoticeable and does not need the involvement of worldly considerations in the sense of reflecting on the fictional setting in the light of the actual world or, say, comparing the two. The reader simply applies her actual beliefs, such as her basic beliefs concerning human behaviour, to the story. On a more conscious level, a reader may feel the need to speculate or reason a character’s motives and actions – for example, in the light of her folk psychological knowledge – in order to comprehend the character and the story events. Acknowledging the need for this kind of external consideration, which often comes in the form of a problem, is not unusual in literary interpretation. Walton, for instance, thinks that if one is ascertaining a work’s implicit fictional truths, such as the reason for a character’s antisocial behaviour, even by making a library visit and consulting medical studies on neurological diseases (extrapolation based on the Reality Principle) or historical sources concerning mutual religious beliefs in the author’s society (extrapolation based on the Mutual Belief Principle), one is still participating in a game of make-believe (see Walton

The Place for External Considerations in Reading Literary Fiction | 157

1990, 159–160.2 What is controversial is the role of external considerations in literary appreciation, as in evaluating the significance of a literary theme or the way it is being developed in the work. As the act of evaluating a work as a literary work is generally distinguished from the act of imaginatively engaging with the work (fiction), I shall discuss these questions, imaginative engagement and appreciation, separately. There are also different kinds of communication which authors may aim to perform in or through their works.3 By their fictional representations, authors may, for instance, make or imply assertions or perhaps advance hypotheses, provide insights or encourage understanding, or intend to draw the audience’s attention to states of affairs in the actual world. This sort of authorial communication is perhaps best seen on an arbitrary scale ranging from explicit assertions that are intended to be believed, and are characteristic for didactic fictions, to invitations to, for example, philosophical, moral, social or political contemplation which may serve assertoric or thematic purposes, that is, which may be intended to change readers’ beliefs or strengthen their response to the fictional setting by means of their beliefs about reality. Respectively, readers may have different kinds of cognitive intentions or expectations. They may “read for life” (Gibson 2004, 110), with a motive for gaining understanding of reality or themselves. For example, readers who read Dostoyevsky or Tolstoy partly in order to gain insights about human nature and real life, arguably read the works properly as works of fiction, assuming that they take them to be literally non-assertive. Whether such cognitive expectations have a place in literary interpretation is however a matter of dispute. While one may argue that a reader’s cognitive expectations may be part of the literary response,4 one has to admit that from a literary point of view, a reader whose

|| 2 Walton however argues that in this case, an extrapolation based on the Reality Principle, in which the character is seen to suffer from an inherited neurological disorder, makes the reader’s imaginative engagement with the work smoother and more vivid than an extrapolation based on the Mutual Belief Principle, in which the character is seen to be possessed by the devil. Following the Mutual Belief Principle, the imaginer in Walton’s example has to leave out her actual religious convictions and scientific beliefs, for instance, whereas following the Reality Principle, the imaginer can “transpose more of what is true (and what she knows to be true) about herself into the world of her game” (Walton 1990, 160). 3 See Trivedi (2001, 193–194) for different kinds of communication which literary works may attend to. 4 For example, Noël Carroll (2007, 32, 36–37) has argued that approaching a realist novel in terms of the social and psychological insight the work is meant to deliver is part of the literary response to the work.

158 | Jukka Mikkonen only interest in a work of literature is in obtaining worldly truths (or insights) is not properly attending to the work. The make-believe theories of fiction maintain that to engage with a fictional story is basically to make believe, or imagine or pretend, that the story told is true and to set aside questions of truth and reality. In these theories, readers’ external considerations of fictional matters are, in turn, commonly discussed in terms of readers’ belief or disbelief in fictional propositions. For instance, Walton and Currie argue not only that a reader may simultaneously make believe and believe a fictional proposition, but also that an author may intend an utterance in her work, or perhaps the whole content of her work, to be both madebelieve and believed (see e. g. Walton 1990, 71, 78, 94, and Currie 1990, 48–49). Walton (1978, 21) proposes that the “dual standpoint” which readers characteristically take on fictive content, their being inside and outside the fictional world, is “one of the most fundamental and important features of the human institution of fiction”.5 However, Walton and Currie say relatively little of the role of external considerations in literary interpretation, whereas Lamarque and Olsen’s discuss this topic extensively. In Truth, Fiction, and Literature (1994), Lamarque and Olsen maintain that readers may take a dual perspective on the fictional content of a work. In their view (Lamarque and Olsen 1994, 143–144), certain kinds of description in fiction invite both internal and external reflection. As one of their examples, they use the narrator’s generalization about “shallow natures” in George Eliot’s Middlemarch. Lamarque and Olsen (1994, 147) assert that when reflected from the internal perspective, the generalization enhances the reader’s understanding of a fictional character, Rosamund Vincy, whom the generalization concerns, while reflecting the generalization from the external perspective “involves thinking of ‘shallow natures’” beyond the fictional setting. Although Lamarque and Olsen (1994, 148) say that these perspectives “nicely interact”, they take the external perspective to have a narrow and limited role in the literary-fictive response. Lamarque and Olsen (1994, 148) give the external perspective this limited role because they find that generalizations and other explicit or implicit propositions in fiction are “almost never to be taken at face value, merely as general || 5 Walton (1990, 49–50) discusses these two perspectives mainly in terms of participation and observation, that is, attending to fictional propositions and noting the ways by which their fictionality is generated. Moreover, he (Walton 1990, 289) suggests that one’s observing of one’s own participation in a game of make-believe may have cognitive value as it “may make it easier to see connections between possible or actual fictional experiences and actual or possible real-life ones.”

The Place for External Considerations in Reading Literary Fiction | 159

assertions about the world issued directly by the author”. As they (Lamarque and Olsen 1994, 148) see it, such a response would be “to abandon the cognitive distance of the fictive stance”. As for literary interpretation, the interest which generalizations have from the external perspective is, for Lamarque and Olsen (1994, 148), “conditioned by their role under the internal perspective”. In Lamarque and Olsen’s view (1994, 147–148) to reflect a general proposition in a work of fiction from the external perspective is to consider it for its truth, or truth-conditions, which serves to enhance fictional characterization. From the above, it can be concluded that the distinction between the internal and the external perspective, which may be used for various purposes, as Currie (2010, Ch. 4) has shown in his recent work on narratives, is a great help in clarifying the speak of our encounters with fictional works and objects. However, I think that the external perspective is much more complicated and has a greater role in general reading public’s literary experience than is made clear in Lamarque and Olsen’s account.

3 External considerations in literary interpretation The internal and the external perspectives on the fictional content of a work arguably both admit both different degrees and various modes, depending on the type of the work. When it comes to the internal perspective, de se imagining, visual imagining, and affective response, are certainly characteristic of, but none of them perhaps necessary for, the perspective. The imaginative engagement with a highly dramatic and emotional narrative surely differs from that with a work which contains large amounts of, say, explicit philosophical speculation, both as for the interaction between the perspectives, and for the kinds and intensity of imagining.6 In addition, metafictional novels often exploit and ridicule the reader’s expectations derived from her encounter with vivid and coherent realistic works, whereas the Epic Theatre aimed to get rid of escapistic make-believe and keep the spectator constantly aware of the fictionality of the play she is watching.7

|| 6 See Roger Scruton’s article Fearing Fictions (2010). 7 These effects have been acknowledged and studied in make-believe theories. Walton, for one, has approached them with his distinction between participation and observation and his notion of reflexive representation.

160 | Jukka Mikkonen Conversely, external considerations come with different propositional attitudes and may be seen to involve different mental processes from supposing a proposition via determining its truth-conditions to ultimately assessing its truth-value. Further, the reader’s act of externally reflecting on the fictional content, even considering it for its truth, may be seen to admit degrees with regard to her engagement with the story (the internal perspective). There are authors’ explicit assertions in fictions, especially in those of the didactic sort, which the author clearly intends the reader to believe and which could be assessed outright without much dangering the cognitive distance of the fictive stance. For instance, the assertion which Voltaire conveys by presenting Pangloss’s views in Candide, namely, that Leibniz’s metaphysics is ridiculous, may surely be assessed as an assertion by the author (conveyed by Pangloss’s speeches and his misfortune) while engaging with the fictional story. In more ‘literary’ cases, in turn, there seems to be less legitimate place for external considerations, at least for those in which the fictional content would be formulated as assertions of the author or the work and assessed for their truth. This means that as readers pay attention to the ways by which, say, philosophical views are being dramatically developed in the work, their focus of imagining is in the internal perspective and their external considerations primarily feed the fictional charaterization, as Lamarque and Olsen propose. Peter Kivy has suggested that while external considerations of general views expressed in fictional literature might not have place in the very act of following the story, the considerations may still be considered part of the encounter with the work. Kivy remarks that there is a distinctive temporal dimension in literary experience, as large novels, for instance, are typically read in parts. According to him (Kivy 1997, 23), external considerations take place in the ‘gaps’ and ‘afterlife’ of the literary experience, that is, the reading of the work. Kivy (1997, 131–134) maintains that when the reader is contemplating an issue – a hypothesis – raised by the work after reading it, she is still enjoying the work. Kivy’s proposal attempts to preserve the unity and coherence of the fictive stance and yet allow a place for substantial external considerations in literary experience. But what would make a reader render a literary work as a hypothesis and genuinely consider it, or to look for support for it or attempt to verify it, which Kivy thinks is a typical response to fictional content? In addition to particular cognitive expectations, there is the widespread conception in literary culture that great works of art often offer insights about reality and human nature and the curiosity and fascination with the epistemic uncertainty caused by the literary-fictive mode of speaking. In fictional literature there is ultimately no general way to achieve certitude of the meaning of the assumed thematic claims a work makes or whether the work also makes these claims of the world. For

The Place for External Considerations in Reading Literary Fiction | 161

example, Dostoyevsky’s Idiot may be seen to be built upon an implicit statement or hypothesis that ‘A perfectly sincere, noble, and unselfish, Christ-like person would be exploited and laughed at’, but whether the work claims so (or otherwise) and whether it claims so (or otherwise) of the world is something a reader cannot be sure of. This uncertainty invites readers to think twice, to thoroughly reflect on the fictional content. Second, there is what I shall call, for lack of a better term, implication. Akin the internal perspective requiring filling in the story, the external perspective on thematic statements and other fictional content can hardly escape demands of exploring their real-world implications and consequences; arguably, the aptness, deepness and perceptivity of a fictional representation is assessed partly by reflecting how the representation figures in external application.8 (Literary-characters-cum-metaphors from Don Quixote to Bartleby alone illustrate how extensive cultural results external considerations may have.) But here one has to be cautious. If we want to retain the notions of authorized response to a work of fiction and imaginings permitted by the work, allowances to free play of imagination may lead to the danger of blurring the distinction between fiction and personal fantasy. The legitimacy of external considerations in literary interpretation is often discussed in terms of the focus and direction of imagination. One’s reflecting on the fictional world in the light of the real world is typically considered part of the literary response (recall Walton’s example of ascertaining implied fictional truths), whereas one’s reflecting on the real world in the light of the fictional world is deemed improper. Lamarque and Olsen make a distinction between the use and appreciation of a work and argue that thinking of the thematic statement a work makes or implies in relation to reality is to use the work for extraliterary purposes, namely, illuminating one’s understanding of reality, not the work. Appreciation, next, which Lamarque and Olsen consider the core of criticism, and thus a model for all literary interpretation, is about identifying and aesthetically enjoying the theme of the work and the way it is being developed (see Lamarque and Olsen 1994, 256, 408–409; Lamarque 2009, 134). However, it is questionable whether there is a core of criticism that is constituted of analysis and appreciation of themes. Thematic criticism itself is a rather contested approach in contemporary criticism, which raises an objection to Lamarque and Olsen’s description of criticism.9 Although appreciation derived || 8 As Jerrold Levinson (1997, 967) perceptively remarks, “[e]valuations of works as naive, puerile, or shallow, or as penetrating, insightful, or wise, implicate assessments of the cognitive adequacy of those works in a way that is hard to get around”. 9 Lamarque (2009, 134–135) argues that psychoanalytic and Marxist approaches to literature, for instance, do not approach literary works as literature and “cannot be seen as continuous

162 | Jukka Mikkonen from the abstract core of criticism characterizes a substantial feature of literary interpretation, it does not sufficiently describe literary experience tout court, as it does not cover other typical values which the general reading public looks for in literature. Finally, a focal problem in the discussion on literary interpretation in the philosophy of literature has been that it has to a great extent ignored the reader and considered her a passive agent whose task is merely to reconstruct the meaning of the work. While ‘sophisticated’ and ‘unsophisticated’ are apt labels to differentiate, say, readers’ conceptions of reference in literature, they are rather broad labels if used to describe the diverse interests which general readers have in literature (as literature). Anders Pettersson (2008), for one, has remarked that empirical studies of literary reception on actual readers’ motives for reading novels show that the general reading public primarily looks for entertainment and learning (that is, gaining knowledge of the world and understanding of themselves and of other people). If we are interested in what sorts of literary values there are, and what readers commonly look for in literature, we should better acknowledge the broader practice of literature and different modes of reading. Although it is clear that the philosophy of literature is distinct from literary sociology and empirical studies of literature, it is also clear that a priori-debates on readers’ typical expectations never close.10

References Carroll, Noël. “The Wheel of Virtue: Art, Literature, and Moral Knowledge.” Journal of Aesthetics and Art Criticism 60.1 (2002): 3–23. Carroll, Noël. “Literary Realism, Recognition, and the Communication of Knowledge.” A Sense of the World: Essays on Fiction, Narrative, and Knowledge. Eds. John Gibson, Wolfgang Huemer and Luca Pocci. New York: Routledge, 2007, 24–42. Currie, Gregory. The Nature of Fiction. Cambridge: Cambridge University Press, 1990. Gibson, John. “Reading for Life.” The Literary Wittgenstein. Eds. John Gibson and Wolfgang Huemer. London: Routledge, 2004, 109–124. || with a tradition of reading that rests on pleasures of a more universal kind associated with literary art.” 10 For an example of conflicting views of readers’ typical exceptions in reading literature, see John (1998, 331–332) and Lamarque (2009, 254). For remarks about the concept of the reader in the debate on the cognitive value of literature, see Kivy (1997, 20–21) and Carroll (2002, 8). – I want to thank the Alfred Kordelin Foundation for funding my post doctoral study, which this article is a part of.

The Place for External Considerations in Reading Literary Fiction | 163 John, Eileen. “Reading Fiction and Conceptual Knowledge: Philosophical Thought in Literary Context.” Journal of Aesthetics and Art Criticism 56.4 (1998): 331–348. Kivy, Peter. “On the Banality of Literary Truths.” Philosophic Exchange 28 (1997): 17–27. Lamarque, Peter. The Philosophy of Literature. Malden: Blackwell, 2009. Lamarque, Peter and Stein Haugom Olsen. Truth, Fiction, and Literature. A Philosophical Perspective. Oxford: Clarendon Press, 1994. Levinson, Jerrold. Review of Truth, Fiction, and Literature: A Philosophical Perspective, by Peter Lamarque and Stein Haugom Olsen. Philosophy and Phenomenological Research 57.4 (1997): 964–968. Pettersson, Anders. “Three Problematic Aspects of Analytical Aesthetics.” The Nordic Journal of Aesthetics 35 (2008): 60–74. Scruton, Roger. “Fearing Fictions.” A Companion to the Philosophy of Literature. Eds. Garry L. Hagberg and Walter Jost. Malden: Wiley-Blackwell, 2010. Trivedi, Saam. “An Epistemic Dilemma for Actual Intentionalism.” British Journal of Aesthetics 41.2 (2001): 192–206. Walton, Kendall L. “How Remote are Fictional Worlds from the Real World?” Journal of Aesthetics and Art Criticism 37.1 (1978): 11–23. Walton, Kendall L. , Make-Believe. On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990.

J. Alexander Bareis

Fictional Truth, Principles of Generation, and Interpretation Or: Why it is Fictionally True that Tony Soprano was Shot Dead

1 Introduction The aim of the following discussion is to clarify some of the principles by which fictional truths in representational arts are generated, and which consequences the generation of fictional truths has in relation to the interpretation of representational works of art. The clarification builds on the original work by David Lewis (1978, 1983) in his seminal article on truth in fiction, and on Kendall Walton’s concept of make-believe (Walton 1990). Walton’s description of the principles at work within what he calls the mechanics of generation is based on Lewis’ account, but also goes beyond it on several points. In the past few years, several suggestions for principles of generation have been put forward, and as new principles have been proposed, the range and validity of the established principles have been questioned. Yet, the relation between the mechanics of generation, between what is true in a work of fiction and how to interpret a work of fiction, has not been a topic of intense research. In the following, I first clarify how I use the term fictional truth in this article. Next, I give an overview of the existing variations of principles as put forward by different scholars, followed by a discussion of different types of fictional truths, in particular the notion of implied and direct fictional truths. In a next step, I discuss the different principles of generation that have been suggested besides the two ‘original’ principles by Lewis and Walton, the so-called Reality Principle and the Principle of Mutual Belief, and also add suggestions for new ones myself. I then shift focus towards a discussion of the relation between the principles of generation of fictional truths and the methodological issue of interpretation. My discussion is exemplified by a close reading of the final scene of the TV-series The Sopranos. The paper ends with a discussion of the relation between the generation of fictional truth and the notion of interpretation.

166 | J. Alexander Bareis

2 Different types of fictional truth What counts as a fictional truth? According to Walton, fictional truth has nothing to do with the notion of truth itself: What we call truth in a fictional world is not a kind of truth. The phrase ‘In the world of the Unicorn Tapestries,’ preceding ‘a unicorn was captured’, does not indicate in what manner or where or in what realm it is true that a unicorn was captured, or anything of the sort. This is not true, period. (Walton 1990, 41–42)

Instead, Walton talks about something being fictionally true, a concept which is captured by the phrase ‘in the world of the work’. In Walton’s (1990, 35) words: “To call a proposition fictional amounts to saying only that it is ‘true in some fictional world or other’.” It is fictionally true in Kafka’s Metamorphosis that Gregor transforms over night into a ‘verminlike’ shape. The phrase “in some fictional world or other” quoted above (Walton 1990, 35) is important for the continuing discussion. In Walton’s approach, there can be different types of fictional worlds, in particular worlds that are created in a game of make-believe with a representational work of art. This is related to the level of generalization of fictional truths. Saying that it is fictionally true in Kafka’s Metamorphosis that Gregor turns into a kind of vermin is a proposition that probably holds true in most fictional worlds played or associated with the Metamorphosis, but not necessarily in all fictional Metamorphosis-worlds. The fictional truth that Gregor was transformed into a giant cockroach, however, is a fictional truth that may not be true in all games with the Metamorphosis as a prop.1 Walton has developed the notion of game world and work world for this and similar differentiations, and this distinction is important in the discussion of the interconnection of fictional truth and interpretation in the last section of the article. A problem with Walton’s distinction between work world vs. game world is that it is rather idealistic, and not absolute. To illustrate: What is fictional in the || 1 I adopt the terminology by Walton in this paper. This is Walton’s (1990, 69) own definition of his key terms: “Representations, I have said, are things possessing the social function of serving as props in games of make-believe, although they also prompt imaginings and are sometimes objects of them as well. A prop is something which, by virtue of the conditional principles of generation, are fictional, and the fact that a given proposition is fictional is a fictional truth. Fictional worlds are associated with collections of fictional truths; what is fictional is fictional in a given world – the world of a game of make-believe, for example, or that of a representational work of art.” A prop can be basically everything – a novel, a movie, but also a hobbyhorse or a cloud formation.

Fictional Truths, Principles of Generation, and Interpretation | 167

world of the work is what is fictional in any game world with the prop in question. Here is how Walton describes this, with the painting Un dimanche aprèsmidi à l’Île de la Grande Jatte (1884–1886) by Georges-Pierre Seurat as example: This points to the conclusion that what is fictional in La Grande Jatte is what is (or would be) fictional in any game in which it is the function of the painting to serve as a prop, and whose fictionality in such games is generated by the painting alone. (Walton 1990, 60)

A fictional truth, therefore, has to be true in relation to a fictional world. However, whether it is a truth in relation to the world of the work or to the world of a single game of make-believe, or both, remains unsaid. Thus, if we want to understand the concept of fictional truth entirely, we have to dig deeper, and we must differentiate between different types of fictional truths. Walton makes a difference between direct fictional truths and indirect fictional truths. That Gregor is turned into a kind of vermin is a direct fictional truth, since the text explicitly and directly states this: “When Gregor Samsa woke up one morning from unsettling dreams, he found himself changed in his bed into a monstrous vermin” (Kafka 1996, 1).2 In comparison, it is not a direct fictional truth that Gregor was transformed into, say, a bug or a beetle, or cockroach, even though this actually might be the fictional truth that many readers of Kafka’s short story generate. For example, in Vladimir Nabokov’s reading of the story it certainly was true that Gregor has wings (or at least wing cases) like a beetle, and that he definitely was not transformed into a cockroach. It is true in the world of the game that Nabokov played with the story that Gregor was transformed into a beetle, since Nabokov generated these implied fictional truths. Here is the passage in Nabokov’s Lectures on Literature: Now what exactly is the “vermin” into which poor Gregor, the seedy commercial traveller, is so suddenly transformed? It obviously belongs to the branch of the “jointed leggers” (Arthropoda), to which insects, and spiders, and centipedes, and crustaceans belong. […] Commentators say cockroach, which of course does not make sense. A cockroach is an insect that is flat in shape with large legs, and Gregor is anything but flat: he is convex on both sides, belly and back, and his legs are small. He approaches a cockroach in only one || 2 I chose to quote the Norton Critical Edition and the translation by Stanley Corngold from 1972, which was reprinted in the Norton edition from 1996, edited by Corngold. The translation problem of the first sentence of Kafka’s Verwandlung is by now legendary. For a critical discussion, see Corngold (in Kafka 1996, 87–88), who himself translates ‘monstrous vermin’, and the discussion by Hartmut Binder in the same edition (185–190), who quotes Kafka’s letter to his publisher in which he explicitly forbids an illustration of the insect. Malcolm Paisley, the curator of the manuscripts at The Bodleian Library in Oxford and editor of the critical edition published by F. Fischer, translates ‘monstrous insect’.

168 | J. Alexander Bareis respect: his coloration is brown. That is all. Apart from this he has a tremendous convex belly divided into segments and a hard rounded back suggestive of wing cases. In beetles this cases conceal flimsy little wings that can be expanded and then may carry the beetle for miles and miles in a blundering flight. Curiously enough, Gregor the beetle never found out that he had wings under the hard covering of his back. (Bowers 1980, 258–259)

The discussion above nicely illustrates a number of important issues at hand.3 More explicitly, it exemplifies how Nabokov uses the so-called Reality Principle (discussed in the next section) to generate an implied fictional truth. Nabokov infers from the text that Gregor has turned into a beetle, since the characteristics of beetles in reality and the description in the fictional text seem to coincide. Since the text does not state explicitly that it is a fictional truth that Gregor is transformed into a beetle, it is not part of what Walton calls the work world, or at least it seems not to be.4 It is a fictional truth that is obviously, or at least likely to be, true in the personal, individual game of make-believe that Vladimir Nabokov has played with the story. It is therefore a truth of Nabokov’s game world.

|| 3 Stacie Friend has used the very same passage, but for entirely different purposes in her article from 2011. I was a temporary member of the Aesthetics discussion group at the Department of Philosophy, University of Michigan, where she discussed the topic in 2003. I have afterwards used the passage myself: in my thesis (Bareis 2008), as well as in an article on the principle of genre convention (Bareis 2009). 4 The German original version states that Gregor was transformed into “einem ungeheueren Ungeziefer,” literally ‘a monstrous vermin’ (Kafka 1995, 5). A number of translators used ‘gigantic’ instead of ‘monstrous’, in particular Willa and Edwin Muir, whose translation is the one used by Nabokov for his lecture on Kafka. Nabokov corrected this very phrase in his teaching copy of the Metamorphosis, as shown in a facsimile print in the Bower’s edition of his Lectures on Literature (Bowers, 250), but he only corrected the ‘gigantic insect’ into ‘monstrous insect’, with the ‘insect’ remaining as a flawed translation. Yet, in his manuscript cited above, he correctly uses ‘vermin’. However, the only valid attribution of a fictional truth in an authorized game of make-believe should be based on the original version of a text, and the original does not state that Gregor was transformed into a beetle, but into ‘a [sic] vermin’. The question of what counts as a work to begin with obviously becomes vital here, and from a strict viewpoint of edition philology, translations should probably be regarded as works in their own right. This, of course, has far-reaching consequences for questions of intentionality, and hence interpretation.

Fictional Truths, Principles of Generation, and Interpretation | 169

3 Principles of generation Before Walton in Mimesis as Make-Believe singled out two principles of generation based on Lewis’ account, the so-called Reality Principle and the Mutual Belief Principle, other scholars have suggested principles as well: For example, Marie-Laure Ryan made the case for the Principle of Minimal Departure (Ryan 1980), which is also based on Lewis’ initial ideas and can be said to be ‘something in between’ the Reality Principle and the Mutual Belief Principle. For Ryan, reality, or any mutually-believed version of it, is the point of departure from which one should only deviate if the work in questions explicitly tells one so. One could say that Ryan’s solution incorporates Walton’s distinction between direct and indirect fictional truths discussed above. However, the direction is reversed: For Walton, make-believe is the primary point of departure, and the generation of fictional truth is the primary principle at work. Simply put, Walton starts out from a principle of fiction – ‘just accept anything, otherwise you can’t play’ – while Ryan starts out from reality – ‘you might have to imagine things that are not as in reality, but try to keep it as real as possible’. Kai Mikkonen (2011, 116) has described Ryan’s principle as “an economy in reasoning concerning the with[held] information in fiction, related to a multiple-world model frame”. In his discussion of the principle, he points to a number of difficulties regarding its applicability, in particular the applicability to metafictional texts. Mikkonen is in good company pointing out potential problems with the principles. Difficulties with, and counterexamples to, the Reality Principle and Mutual Belief Principle are easy to come by. For example, Richard Woodward (2011, 163) has recently pointed out that both “principles are seriously inadequate when measured against the complexities of fictional generation, but both principles have their place”. It seems they should be considered to be rules of thumb, a possible and meaningful way to reason about truth in fiction, but they can never take the place of interpretation. As Paisley Livingston (2005, 191) has put it, “the problem of fictional truth should not be dissolved in some holistic account of work-interpretation”. If the generation of fictional truths and interpretation remain separate issues, as Livingston suggests, the problem of how exactly to generate fictional truth, and by virtue of which principles, remains, and the question of which fictional truth to generate remains vital in discussions about different interpretations. The question of which principle of generation to use can also be a question of which method of interpretation one employs, and which kind of interpretation one pursues. Therefore, I agree with Livingston that there is a difference worth maintaining between the generation of fictional truth and the question of interpretation, but I will argue at a later

170 | J. Alexander Bareis point in this article that the two issues are intertwined in a way that makes it necessary to look at both in order to be able to make a substantial contribution regarding any of the parts. Thus, the two original principles have received a great deal of criticism, and this has led to the introduction of a number of additional principles. Some of these are variations of the discussions by Lewis and Walton, in particular the so-called Principle of Coherence, suggested by Mikael Pettersson (2008), and the Principle of Genre Convention which I have suggested earlier, particularly in relation to unreliable narration (Bareis 2013). In addition to Genre Convention, however, I aim to show that there is a need for a Principle of Media Convention, since the medium to which a prop belongs plays a major role in the generation of fictional truths. However, the difference between medium and genre is far from obvious, and I will therefore always include both terms. Additionally, I will suggest yet another principle, which I will call the Principle of Fiction. I will start the discussion of the above-mentioned principles with the Principle of Coherence as my point of departure. Yet, as the evolving discussion will show, there is a need to discuss the principles in connection to each other, rather than isolated, since there is a strong tendency of overlap and even competition between the principles. The Principle of Coherence proposed by Pettersson is based on the discussion of a particular example in Walton’s Mimesis as Make-Believe, where Walton suggests that it is an implied fictional truth that Mrs. Verloc in Joseph Conrad’s The Secret Agent has committed suicide. In the novel, the actual suicide is not narrated directly, yet a newspaper article about a “Suicide of Lady Passenger from a cross-Channel Boat” (Conrad, cited by Walton 1990, 141), narrated at a later point in the story, makes it an indirect fictional truth that Mrs. Verloc has in fact committed suicide.5 As Pettersson (2008, 241–242) shows, not only coherence is part of the process of generating fictional truths, but also relevance and knowledge about genre. It makes for a coherent story if the reader infers the fictional truth that the ‘Lady Passenger from a cross-Channel Boat’ who committed suicide is Mrs. Verloc, and not a random person, since that would not make any sense at all. Also, the above-mentioned Principle of Fiction, which can be paraphrased as ‘things are narrated for a reason’, seems partly at play here: In many cases, most parts of a work of representational art are where they are for a reason. This is also hinted at by Pettersson (2008), in his foregrounding of rele|| 5 To be more precise, the narrator describes how Comrade Ossipon reacts towards the newspaper item, which implies that Comrade Ossipon understands that it was Mrs. Verloc who committed suicide, without actually stating the fact.

Fictional Truths, Principles of Generation, and Interpretation | 171

vance. The Principle of Fiction in relation to relevance can be paraphrased as a version of Chekhov’s Gun: Remove everything that has no relevance to the story. If you say in the first chapter that there is a rifle hanging on the wall, in the second or third chapter it absolutely must go off. If it’s not going to be fired, it shouldn’t be hanging there. (Bill 1987, 79)

So, fictional particulars should be there for a reason. Obviously, this does not mean that every detail of every fictional narrative has to be highly charged with meaning in the story, but entirely pointless fictional truths are hard to justify, and might be judged as an aesthetic flaw of the work. In relation to the abovementioned novel The Secret Agent, the question is: Why should Conrad’s narrator spoil the reader’s time by narrating about a random person committing suicide? Especially since the fictional truth that it was Mrs. Verloc makes for a coherent story. So, in this particular case, one could argue that it is fictionally true, owing to the Principle of Coherence, that Mrs. Verloc is the woman on the boat who has committed suicide, and that the Principle of Fiction, with its economist connotations in the sense of Chekhov’s Gun, also hints at this. Therefore, in the case of Mrs. Verloc’s suicide, two different principles cooperate. The Principle of Fiction covers yet another aspect in the mechanics of generation of implied truths and in the acceptance of direct fictional truths, and that aspect is akin to what Walton (1990, 175) has called the notion of ‘silly questions’: “Why do all thirteen of the diners in Leonardo’s Last Supper line up in a row on the same side of the table?” The answer is of course that Leonardo wanted to show the faces of the disciples, not their backs. It spoils the game of make-believe to ask these kinds of questions – and that is why Walton calls them silly. The notion of fiction as make-believe means that you have to play along, to suspend your disbelief and not to suspend your charity when engaging with representational works of art: [T]here is a principle of charity at work. The generation of fictional truths is sometimes blocked (if not merely deemphasized) just, or primarily, because they make trouble – because they would render the fictional world uncomfortably paradoxical. But usually more than charity is involved. Recognition of a question’s silliness, and so of one reason for disallowing or deemphasizing fictional truths that give rise to it, may depend on awareness of various demands to which the artist must respond. A decision to disallow anomalous fictional truths is especially plausible when it is evident that there are other reasons for the presence in the work of the features that appear to generate them – when, for instance, they are needed to make the fictional world accessible to the audience, or to enhance appreciator’s games of make-believe. (Walton 1990, 183)

172 | J. Alexander Bareis In the case of the Last Supper, making the world and salient features of it accessible to the viewer is of course the reason why Leonardo decided to group the disciples the way he did – despite the somewhat awkward positioning from a realist point of view. As we can see, the mechanics of generation should not be considered as highly stable or unambiguous. ‘Silly questions’ may block the generation of fictional truths according to an established principle of generation. Different principles may rule each other out, or might compete with each other, or, as shown above, it is sometimes the case that several principles work together. For example, questions of genre and coherence often coincide when we generate implied fictional truths. A telling example is the case of unreliable narration, which demands knowledge of the genre or the narrative technique of unreliable narration in order to be detected (see Bareis 2013). If a recipient is not familiar with the notion of unreliable narration and its specific signs, it is much more difficult – indeed, it may even be impossible – to decipher the hints in the narration that point at this implication.6 This is the reason why certain narratives have not been regarded as examples of unreliable narration while their creators certainly had the intention of creating an unreliably narrated narrative.7 This is also the point where the Principle of Genre Convention and the Principle of Media Convention meet. Let me exemplify this with the famous last scene in the last episode of the by now classical HBO-series The Sopranos.

4 The fictional truths of The Sopranos The Sopranos is the work of David Chase. Particularly interesting about The Sopranos is that the producer can be regarded as the equivalent to an author of literary fiction, as far as the work as a whole can be seen as an intentional work of art. As pointed out by Salman Rushdie (quoted in Thorpe 2011),8 in some TVseries the producer can “have control in the way that you never have in the cinema. The Sopranos was David Chase”. Intentionality, hypothetical or real, is

|| 6 There are of course cases of partially mimetic unreliable narration which can be recognised even by readers without prior knowledge of the genre convention. 7 I have discussed a number of examples in Bareis (2013), in particular for the novel Der fernste Ort (The Farthest Place) by Daniel Kehlmann. 8 See also Polan (2009, 81) on the season charts Chase was using when assigning the writing of episodes to other writers.

Fictional Truths, Principles of Generation, and Interpretation | 173

therefore of relevance for the interpretation in a similar fashion as in literary fiction. In the final episode of the seventh and final season of The Sopranos, after eight years (between 1999 and 2007) and 85 hour-long prior episodes, the screen simply turns black simultaneously as the sound turns mute for as long as 10 seconds, before the final credits start to roll. Many viewers suspected that their TV-set had broken down, or that their cable had failed. One of the executives of HBO, who saw the episode before it was aired, “assumed that Chase dared to conceal the ending from him and was furious” (Davis 2008, 10). But obviously, the ten seconds of complete silence and blackness were intended exactly this way by the creator and producer of the show, David Chase. The ending left many viewers in despair: In the days immediately following June 10, the date on which the last episode of HBO’s hit show The Sopranos was aired, the word ‘Sopranos’ itself was the most frequent search term on Internet search engines. And in the first few minutes after the episode finished, the HBO Web site crashed and remained down for several hours due to the volume of people trying to log into it. (Polan 2009, 1)

What people most probably tried to find out was of course how to make sense of the ending – they were searching for an answer to their question ‘what happened?’ If we put this into the terminology of the make-believe theory, the question is: What, fictionally, is true about the ending of The Sopranos? At this point, it is important to emphasize that it is not a question of interpretation yet, but a question of establishing ‘fictional facts’. Yet, for many participants in the discussion, it seems to be rather a question of interpretation. Therefore, this example should suit us perfectly in order to clarify how the generation of implied fictional truth is related to interpretation. Obviously, the final episode of The Sopranos left many viewers in a state of wondering, and this includes even ‘professional interpreters’ such as critics and researchers. In her 2009 book The Sopranos, for instance, Dana Polan does not give the reader a definite answer to the question what fictionally happens during the last ten seconds of the show. Instead Polan (2009, 119), in a chapter titled Against Interpretation, points out that “its meanings and morality are frequently indeterminate”. Even though the quotation above is not about the ending in particular, Polan’s (2009, 59) discussion of the ending is rather vague, actually arguing that the last scene does deliver closure, since “there is no need to show what might happen next, since in a sense we know already what life had been like, is like, and will be like for each character: Tony thus will always be seeking quiet moments with his family, even as he knows that menace could come from anywhere and at any time.” This, to my mind, is an attempt at inter-

174 | J. Alexander Bareis pretation, not an attempt to answer the question of what is implicitly fictionally true. And, so I will argue, there is both a need and a final answer to what fictionally has happened. My own reading of what is fictionally true in the last episode of The Sopranos, on the contrary, comes to a rather firm conclusion: It is a fictional truth of the work world (or at least should be a fictional truth that is true in any game world with the work) that Tony was shot dead and that the last ten seconds of blackness and silence are what Tony sees and hears – that is, nothing, because of his death. The reason for this claim lies in the application of the Principle of Media and Genre Convention. The key to the conclusion lies in the filmic convention of point-of-viewshots, with which you need to be familiar. Without such knowledge it is extremely hard to generate the fictional truth that the last ten seconds of the final Sopranos episode, the ten seconds of silent black screen, is what Tony sees and hears at the very moment of his death. But this, I argue, is the best possible interpretation of this scene. The black screen and the silence is what Tony ‘sees’ and ‘hears’ after he was hit by a bullet, presumably directly in the head. In order to strengthen my argument, I will show that the Principle of Media/Genre Convention can help to explain why the fictional truth that Tony is shot is a fictional truth that can be generated in any game of make-believe with the Sopranos as a prop, as long as you are aware of the principle at work. In filmic narratives, the so-called point-of-view (POV) shot is often prepared by a close-up or gaze shot of the person whose thoughts or vision we are about to see. The POV shot is an established filmic way of expressing what a character sees, what he or she thinks, feels or dreams. If there is no shot before, the reverse shot afterwards usually clarifies what actually went on in the scene. This convention is used in The Sopranos throughout the whole series, and has been a standard convention in film and TV for many decades. What is problematic about the usage of a POV shot in the final scene is the fact that there is no reverse shot, which is usually also part of the convention. But, and this helps to explain the usage in the last scene, there is a significant scene in which the POV technique is used with a reverse shot that prepares for the final scene. The scene in question occurs in episode 14 of the last season, titled Stage 5 aired on 15 April 2007 nearly two months before the final scene, which was aired on 10 June 2007. In this scene, there is a shoot-out at a restaurant. Steve van Zandt’s character Silvio is having dinner with his girlfriend and another couple, the mafia member Gerardo ‘Gerry’ Torciano, and his girlfriend. The two couples are seated opposite of their partners. Silvio sits next to Gerry’s girlfriend, and opposite of his own girlfriend, who is seated to the right of Gerry. In a typical POV shot over the shoulder of Silvio’s girlfriend, we see what Silvio’s

Fictional Truths, Principles of Generation, and Interpretation | 175

girlfriend sees: She is looking at Silvio opposite her, who is talking about a wine they once drank. Gerry, to her left, is not visible from this perspective. Suddenly, both the sound and speed of the audio- and video track change dramatically. It starts with a tinnitus-like sound on the audio track, which makes it impossible to hear what Silvio says. The speed of the video track slows down to slow motion. Stains of blood slowly start to cover Silvio’s face. At that point, the only thing the viewer can hear is the tinnitus-like sound, muting everything else. In a following neutral reverse shot, the audience can then see the whole chain of events, all four characters around the table, and an assassin shooting Gerry. The viewers can see how an assassin comes to the table, pulls up a gun, and fires shots at the person next to Silvio’s girlfriend, the mafia member Gerry Torciano. This means that the viewer sees the same couple of seconds, the actual shooting, from two perspectives: first, from Silvio’s girlfriend’s point of view, who is looking at Silvio and sees bloodstains on his face, and second, from a neutral point of view, the so-called reversed shot. The viewer can see how Gerry falls on the floor with multiple shot wounds, and is shot several more times by the assassin, as he lies on the ground. At the same time, Silvio and the two girlfriends are hurrying away from the scene. During the reverse shot, the speed of the audio and video track is back to normal – the viewers can hear the shots at the point in time when they are fired. This leaves us with the conclusion that the tinnitus-like sound and the blood stains on Silvio’s face are what Silvio’s girlfriend hears and sees during the shooting: Her hearing went ‘blind’ because of the shots fired right next to her, while she could already see the bloodstains flying across the table into Silvio’s face. We have to generate the (more or less) implied fictional truth that the POV shot of Silvio’s girlfriend established what she hears and sees at that very moment of the shooting. It is fictionally true that she saw the blood on Silvio’s face without hearing a shot, since her hearing was ‘deafened’ by the shot fired in close proximity to her. What this scene establishes, then, is on the one hand the filmic technique of POV shots, and on the other hand a specific relation between hearing and seeing shots fired. Both aspects will turn out to be decisive for the understanding of the much-discussed final scene. There is more to be considered. In the same episode as discussed above, Silvio talks to Tony about what happened at the restaurant, which establishes the relationship of hearing and seeing shots more firmly. Silvio, whom the viewer could see when the shots were fired through the point of view of his girlfriend, explains to Tony that he didn’t even hear the shots. Further, in the previous episode 13, aired on 8 April 2007, the ‘You-never-hear-the-shot theory’ is established for the first time. Tony and his brother-in-law, Bobby Baccalieri, are talking about how it probably feels to be shot, assuming that one would not

176 | J. Alexander Bareis even hear the shots, since the speed of a bullet is faster than the speed of the sound of the shot fired. From the above, we see that there is good reason to generate the following fictional truth about what happened in the final scene of The Sopranos: The ten seconds of black silent screen are a POV shot of what Tony ‘hears’ and ‘sees’ after he has been shot. Which is nothing, as he was killed instantly by a shot in the head. What the final episode does not show is the equally conventional reverse shot, which would show Tony shot in the head at the table of the diner, with his family around him. It is also clear that the reason why so many viewers were upset is the missing reverse shot in the final episode, which would have shown what actually, fictionally speaking, had happened from a neutral point of view. The concepts of direct and indirect fictional truths neatly explain the root of this frustration: The reverse shot would have been equivalent to a direct fictional truth of what happened. Without it, all there was left to do for the viewers was to generate indirect fictional truths, and making choices about which principle of generation to use. To be more specific, according to the Principle of Genre and/or Media Convention, it is fictionally true that the last ten seconds of the final episode of the Sopranos depict what Tony sees and hears after he has been shot: nothingness. It is fictionally true, since POV shots always generate the fictional truth by media-specific convention that what you see and hear is what the character in question sees and hears. This was how the POV shot worked in the scene from episode 14 when Gerry was murdered, which was first shown by means of a POV shot of Silvio’s girlfriend. Together with primary fictional truths, especially the ‘You-never-hear-the-shot theory’ established in direct conversation among fictional characters, it is fictionally true of the last scene in the Sopranos that Tony was shot in the head. The ‘you-never-hear-the-shot theory’ is an issue for the generation of the fictional truth using the Principle of Coherence and/or the Principle of Fiction as in Chekov’s gun, which I will comment on in more detail in the following section: All the foreshadowing by means of conversations and instances where a shot was not heard, all these fictional truths, only make sense in relation to the final scene. Otherwise, their existence in the series appears to be redundant. What becomes evident is that the Principle of Genre and/or Media Convention can help to establish fictional truths. But the example also shows that other principles either cooperate or sometimes actually suggest dissimilar fictional truths. In the case of the final scene, my reading relies heavily on similar scenes from earlier episodes, which brings the Principle of Coherence into play – at least to some extent. In other words, it makes more sense and is more coherent to generate fictional truths in a similar way when similar narrative and filmic

Fictional Truths, Principles of Generation, and Interpretation | 177

techniques are used at an earlier stage. On the other hand, the answer to the question whether one does not in fact hear the shot that kills one is, for obvious reasons, impossible to answer. Since sound travels faster than bullets usually do, and keeping in mind that it is probably rather difficult to hit an exact spot in the brain where a shot will kill you instantly, it seems more realistic to assume that one actually hears the shot before one dies.9 However, it is hardly meaningful to argue that the ‘You-never-hear-the-shot theory’ is a mutually-believed fact of reality. Instead, it strikes me as a fictional truth established through the work itself – it is a fictional truth of the world of The Sopranos that you never hear the shot that kills you, and it is mutually believed in the fictional world by the characters in that world, as established through a number of prior reliable direct fictional truths.

5 Fictional truths, summarized At this point, I would like to summarize the different principles that have been discussed so far: the Reality Principle, the Mutual Belief Principle, the Principle of Minimal Departure, the Principle of Coherence, the Principle of Genre and/or Media Convention, and the Principle of Fiction. So far, the Principle of Fiction has not gained any thorough attention. In my understanding, the Principle of Fiction works in two ways – from a perspective within fiction, and from a perspective as fiction. A typical example of the Principle of Fiction would be that we accept certain unnatural or illogical circumstances and even infer those as fictional truths within the work, even if the Reality Principle and the MutualBelief Principle would advise otherwise. For example, we accept that the next inhabitant of Liliput is only three inches high, and we play along and infer fictional truths accordingly. We also accept that James Bond always finds an empty parking spot right in front of every building in every metropolitan city in the world, and we do not generate the fictional truth that James Bond is extremely lucky because of this. To some extent, this principle overlaps with the Principle of Genre and/or Media Convention – it is part of the genre convention that Othello speaks in verse, yet it is not fictionally true that he is a poet because of this. In other respects, this reasoning along the lines of the Principle of Fiction coincides with genre convention in a different way: We know that a character cannot be the murderer in a piece of Agatha-Christie-like crime story if we also || 9 There are plenty of theories about this to be found on the internet. Yet for obvious reasons, it is extremely hard to come by solid or even experimental evidence on the topic.

178 | J. Alexander Bareis know that the relevant revelation takes place in the middle of the book. So, different principles sometimes help appreciators to find out what is not fictionally true. Such an intentional strategy can be said to be the reason behind the generation of fictional truth with another genre convention, the narrative technique of unreliable narration. This narrative device can be found both in high-brow literary prose (e. g. Vladimir Nabokov’s Lolita) and in mainstream Hollywood cinema (e. g. A Beautiful Mind), and is by now regarded by narratologists to be the result of an (at least hypothetically) intentional strategy by an author (for an overview, see Bareis 2013). Sometimes, what we see in a movie or a TV-series is not true in the fictional world. It is not true in the fictional world of A Beautiful Mind that there is a little girl called Marcee who is friends with John Nash. Thus, in literature, sometimes what we read and consider to be a primary fictional truth and hence to be believed is not one. In cases of unreliable narration, direct fictional truths can be said to be reversed. In the classic story by Ambrose Bierce, An Occurrence at Owl Creek Bridge, we are first supposed to make believe that the protagonist succeeds in running away from the bridge where he was about to be hanged. When we reach the end of the story, we are supposed to make believe that he is not. What is said to be true in the beginning of the story will be reversed by implied fictional truths at a later point, and the narrator will turn out to be unreliable. The narrative technique of unreliable narration plays with the very tension between prima facie direct or primary fictional truths and implied fictional truths that are generated because of viewers’ and readers’ knowledge about the genre convention of ‘twist movies’ and unreliably narrated literary texts, or, as in the case of A Beautiful Mind, the usage of so-called fallible filters. If you do not know the genre, which entails the possibility of being unreliable for aesthetic purposes, then the narrative device will not work, since the necessary implied fictional truths will not be generated by the recipient. The authorial intention of creating a piece of unreliable narration will not be recognized. In cases of unreliable narration, it is implicitly fictionally true that it is fictionally untrue what you are prescribed to imagine in the first place. This underlines that the generation of fictional truths with the help of principles of generation is the result of what Walton has called the mechanics of generation, but that these mechanics do not work mechanically. Instead, the generation of fictional truth and the usage of principles of generation are the meeting point of fine-grained handywork of artists on the one hand, and the efforts of participants in games of make-believe, deciphering a work of art, on the other hand. This meeting point is formed by an utterly intricate relationship, which touches the core of aesthetic appreciation – that is, the question of interpretation.

Fictional Truths, Principles of Generation, and Interpretation | 179

6 Fictional truth and interpretation The interrelations of different principles of generation are fuzzy. There is the possibility of principles ruling each other out, of competing with each other, of overlapping with each other, of partially enhancing each other, and even of leaving the recipient with an unsolvable paradox. In unreliable narration, (often) directly generated fictional truths turn out to be untrue, and it is necessary to generate the (often) implied fictional truth that directly generated fictional truths are fictionally untrue. For a literary scholar, this result is highly unsatisfactory, yet hardly surprising. The history of literary scholarship proves this in manifold ways – for example, a quarrel over divergent readings of literary texts can be understood as the clash between competing usages of the different principles. The core of this problem seems to be located in the relationship between what Walton calls the mechanics of generation and interpretation. If we look at the mechanics of generation from the perspective of interpretation, we have to consider both obvious differences between the two concepts, but also a relationship of necessary collaboration. As the discussion of the last episode of the Sopranos has shown, fictional truths generated according to the Principle of Media and/or Genre Convention are essential to generate the fictional truth that Tony was shot dead, which has to be considered a core insight for any interpretation of the final episode of the series. Yet, there are numerous examples of fictional truths generated with this and other principles that play absolutely no role for an informed interpretation of the episode. Furthermore, interpretation is a wide concept, used in different ways, and for different purposes. From this perspective, one could argue that the principles form a set of tools, and even some kind of basic methodology, to find out what fictional truth there might be in a given work. Interpretation would then be the process of deciding which fictional truths we believe to be most adequate. I will call this the methodology thesis. In the theory offered by Walton, the closest we come to explaining this difference is the differentiation between work world and game world, but this will not suffice. Walton’s differentiation was designed to single out typically private ways of reasoning about fictional worlds – it is, for example, a fictional truth of my private game of make-believe, of the game world, that I am scared of the green slime in the movie. Typically, all fictional truths that we generate about ourselves are not part of the work world. The world of the work, that is the fictional truths that are fictionally true in any game played with a work, remains an ideal notion rather than a fruitful methodological tool.

180 | J. Alexander Bareis With regard to the question of fictionality, Walton’s framework is essentially non-intentional. For Walton, unintentionally produced objects can be props in games of make-believe. This does not imply, however, that the intention of an object’s maker is of no interest whatsoever. It might be neither a necessary nor sufficient condition for a game of make-believe to take place, which is Walton’s position in the question, and the main interest of his theory. Thus, the theory of make-believe is not a theory of interpretation, but a theory of fiction. If we want to connect the mechanics of generation of fictional truth with the question of interpretation, we cannot avoid the question of intention. The notion of intention is highly important for connecting the generation of fictional truth and interpretation. As I have tried to show, different principles seem to be the appropriate choice when generating certain fictional truths with a prop. The same prop, or, to be precise, any detail of it, might be used to generate fictional truths that are widely or slightly different, sometimes even mutually exclusive. This is also why Livingston (2005, 191) has warned against the tendency to overemphasize the role of implied fictional truths generated according to established principles within the mechanics of generation, as cited earlier: “the problem of fictional truth should not be dissolved in some holistic account of workinterpretation.” Even so, and this is a point I would like to emphasize, we have to make clear what role the generation of implied fictional truths according to the established principles play in relation to an informed interpretation of a work by a literary scholar. Are the ‘mechanics of generation’ and the usage of different principles merely the interplay as described above, or do they compose a way to establish a methodology of interpretation of works of fiction? Livingston proposes an answer to this question in his suggestion to look at the notion of appropriateness: Elucidating the problem of fictional truth in terms of appropriateness of make-believe in response to a work’s content has the advantage of foregrounding the issue’s normative character (in a broad, axiological, and not exclusively moral sense). In other words, the proponent of some analysis of fictional truth should be able to say why someone ought or ought not to allow that it is appropriate, in response to some work of fiction, to make believe some proposition. And ‘appropriateness’ has the advantage of expressing a suitably open and permissive position, that is, one which embraces both epistemic goals and other sorts of anticipated pay-offs or values. (Livingston 2015, 196)

In the passage quoted above, Livingston explicitly uses the term ‘analysis of fictional truth’, and not ‘interpretation’. Yet, his description correlates with the currently widely-accepted view on interpretation as a practice involving some extent of normativity, and judgment, as expressed by Gregory Currie (2003, 291): “Interpretation requires some degree of thought rather than the operation

Fictional Truths, Principles of Generation, and Interpretation | 181

of merely subpersonal level processes, as with understanding literal meaning. It requires judgment applied to the object of interpretation […].” What Currie calls ‘subpersonal level processes’ such as understanding literal meaning could to some extent be attributed to the generation of direct fictional truths. Understanding the literal meaning of the first sentence of Kafka’s The Trial and generating the direct fictional truths associated with it are closely related: “Someone must have been telling lies about Joseph K., for without having done anything wrong he was arrested one fine morning” (Kafka 1988, 9). In understanding the literal meaning of this sentence, one probably comes very close to the direct fictional truths prescribed to imagine by this sentence: In my opinion, those primary fictional truths are that I am told that there is a person named Jose[f] K., and that K.’s opinion is that he had done nothing wrong, from which he concludes that someone must have been telling lies about him.10 This, again, is where the gap is located between the generation of fictional truths and interpretation. Livingston ends his interrogation with the following heuristic: Intentional heuristic: in determining what is true in a given fiction, the reader, viewer, spectator, etc. should adopt the thoughts and background premisses that the author(s) of that fiction settled on in creating the work, yet only in so far as these assumptions mesh sufficiently with the features of the utterance (including the text, audio-visual dislay, or other expressive means). (Livingston 2005, 207)

For the question of fictional truth, this heuristic might suffice. It does allow for a necessary integration of intention into the process of generating appropriate fictional truths. However, it does not solve the question of the relation between fictional truths in a work of fiction and the interpretation of a work of fiction. Within narratology, and in particular within the German literature on narratology, there has been a vital discussion about the distinction between the terms ‘Beschreiben’ [describe] and ‘Verstehen’ [understand], and the concepts of analysis and interpretation as respective practices.11 Without going into any || 10 Again, the question of translation becomes important. The Muirs’ translation above is slightly flawed, in particular the ‘fine morning’ is problematic. The narrator of The Trial is a seemingly highly objective one, and refrains most of the time from evaluative statements like this. It is not a truth of the work world of Kafka’s Der Process in its original German version that the morning K. woke up was a ‘fine’ one. Also, the German ‘verleumdet’ in the original could be translated with ‘slandered’, and there is no acute need of anglicizing Josef’s name. Yet, the Muirs’ translation probably is the most spread and by now canonical one. 11 In particular, Tom Kindt and Hans-Harald Müller (2003a, 2003b) have discussed this distinction, as well as Oliver Jahraus (1999). For a recent overview, see Bareis (2014).

182 | J. Alexander Bareis detail in this debate, it is rather obvious that the said distinction eventually boils down to the same question that was at the focus of this article – the relation of fictional truth and interpretation. As is the case for the debate in narratology, a final and decisive answer to the question of the exact relation between the generation of fictional truths and interpretation is probably not possible. Rather, a gradual scale of fictional truths becoming (parts of) interpretations seems more adequate. Also, interpretation itself can, and certainly should be, divided in different types, with different usages, as suggested by Göran Hermerén and others.12 If interpretation is not understood as a single, fixed term, but as process and/or product of different activities for different purposes, with different methodologies and on different levels, the relation of the ‘mechanics of generation of fictional truths’ and relevant meanings of interpretation might be easier to explore. However, and this is an emphasis I so far have not seen among theorists of interpretation within analytical philosophy, the notion of fiction has to be taken into consideration more prominently. As I have aimed to show above, the generation of fictional truths and the intricate interplay of manifold principles at work when generating them strongly point to this. When theorizing about the interpretation of works of fiction, a theory of fiction explaining the generation of fictional truths should be the point of departure. If we do that, then hopefully Tony Soprano will have the chance to fictionally rest in peace.

References Bareis, J. Alexander. Fiktionales Erzählen. Zur Theorie der literarischen Fiktion als MakeBelieve. Göteborg: Acta Universitatis Gothoburgensis, 2008. Bareis, J. Alexander. “Was ist wahr in der Fiktion? Zum Prinzip der Genrekonvention und die Unzuverlässigkeit des Erzählers in Patrick Süskinds Die Geschichte von Herrn Sommer.” Scientia Poetica 13 (2009): 230–254. Bareis, J. Alexander. “Ethics, the Diachronization of Narratology, and the Margins of Unreliable Narration.” Narrative Ethics. Eds. Jakob Lothe and Jeremy Hawthorn. Amsterdam, New York: Rodopi, 2013, 41–55. Bareis, J. Alexander. “Fiktion, Analyse und Interpretation. Der fiktionstheoretische ‘blinde Fleck’ der Erzähl- und Interpretationstheorie.” Vor der Theorie. Immersion, Materialität, Intensität. Ed. Mario Grizelj, Oliver Jahraus and Tanja Prokic. Würzburg: Königshausen & Neumann, 2014, 307–322. Bill, Valentine T. Chekhov: The Silent Voice of Freedom. New York: Philosophical Library, 1987. || 12 See in particular Göran Hermerén (1983), and the anthology by Staffan Carlshamre and Anders Pettersson (2003).

Fictional Truths, Principles of Generation, and Interpretation | 183 Bowers, Fredson (Ed.). Vladimir Nabokov. Lectures on Literature. London: Weidenfeld and Nicolson, 1980. Carlshamre, Staffan and Anders Pettersson. Types of Interpretation in the Aesthetic Disciplines. Montréal: McGill-Queen’s University Press, 2003. Currie, Gregory: “Interpretation in Arts.” The Oxford Handbook of Aesthetics. Ed. Jerrold Levinson. Oxford, New York: Oxford University Press, 2003, 291–306. Davis, J. Madison. “It’s Over When the Soprano Eats Onion Rings.” World Literature Today (January–February 2008): 9–11. Friend, Stacie. “The Great Beetle Debate: a Study in Imagining with Names.” Philosophical Studies 153 (2011): 183–211. Hermerén, Göran. “Interpretation: Types and Criteria.” Grazer Philosophische Studien 19 (1983): 131–161. Jahraus, Oliver. “Die Unhintergehbarkeit der Interpretation im Rahmen literaturwissenschaftlicher Theoriebildung.” Interpretation, Beobachtung, Kommunikation. Avancierte Literatur und Kunst im Rahmen von Konstruktivismus, Dekonstruktion und Systemtheorie. Eds. Oliver Jahraus and Bernd Scheffer. Tübingen: Niemeyer, 1999, 214–291. Kafka, Franz. The Trial. Translated by Edwin and Willa Muir. London: Penguin, 1988. Kafka, Franz. The Metamorphosis. Translation, Backgrounds and Contexts, Criticism. Translated and ed. Stanley Corngold. New York, London: W. W. Norton & Co., 1996. Kafka, Franz. Die Verwandlung. Stuttgart: Reclam, 1995. Kindt, Tom, and Hans-Harald Müller. “Narrative Theory and/or/as Theory of Interpretation.” What Is Narratology? Questions and Answers Regarding the Status of a Theory. Eds. Tom Kindt and Hans-Harald Müller. Berlin, New York: De Gruyter, 2003a, 205–219. Kindt, Tom, and Hans-Harald Müller. “Wieviel Interpretationen enthalten Beschreibungen? Überlegungen zu einer umstrittenen Unterscheidung am Beispiel der Narratologie.” Regeln der Bedeutung. Zur Theorie der Bedeutung literarischer Texte. Ed. Fotis Jannidis, Gerhard Lauer, Matías Marínez, and Simone Winko. Berlin, New York: De Gruyter, 2003b, 286–304. Lewis, David. “Truth in Fiction.” American Philosophical Quarterly 15 (1978): 37–46. Lewis, David. “Postscripts to ‘Truth in Fiction’.” Philosophical Papers I. New York, Oxford: Oxford University Press, 276–280. Livingston, Paisley. Art and Intention. A Philosophical Study. Oxford: Clarendon Press, 2005. Mikkonen, Kai. “‘There is no such thing as pure fiction’: Impossible Worlds and the Principle of Minimal Departure Reconsidered.” Journal of Literary Semantics 40 (2011): 111–131. Pettersson, Mikael. “What’s the Story? On the Issue of Truth in Fiction.” Fact and Fiction in Narrative. An Interdisciplinary Approach. Ed. Lars-Åke Skalin. Örebro: Örebro University Library Press, 2005, 227–250. Polan, Dana. The Sopranos. Durham, London: Duke University Press, 2009. Ryan, Marie-Laure. “Fiction, Non-Factuals, and the Principle of Minimal Departure.” Poetics 9 (1980): 403–422. The Sopranos. Producer David Chase. HBO, 1999–2007. Thorpe, Vanessa. “Salman Rushdie says TV dramas comparable to novels,” The Observer (12 June 2011) . Walton, Kendall L. Mimesis as Make-Believe. On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Woodward, Richard. “Truth in Fiction.” Philosophy Compass 6 (2011): 158–167.

Mario Slugan

Deixis in Literary and Film Fiction Intra-Ontological Reference and the Case of Controlling Fictional Narrators

1 Introduction Noël Carroll (2006), Gregory Currie (2010), Berys Gaut (2004, 2010) and Andrew Kania (2005) have garnered much attention recently claiming that no fictional narrators exist but explicit ones such as Ishmael from Herman Melville’s Moby Dick. George Wilson (2007, 2011) responded. Elsewhere I have argued in more detail against Carroll (2006) and Kania (2005), but for reasons different than Wilson (2007); (Slugan 2010). In fact, in the same place I have argued that Wilson (2007) also gets it wrong. In this introduction, I will briefly run through this territory again in order to articulate my own argument for the near-ubiquity of controlling fictional narrators (henceforth CFNs) in literary fiction based on deictic properties of verbs.1 In the second section, I will address objections to my argument stemming from Gaut (2004, 2010) and Currie (2010) by demonstrating that their various appeals to indeterminacy do not suspend deictic properties. The focus on the second-person narratives in the third section will allow me to flesh out another important property of deixis in fiction – intra-ontological reference – and to, at the same time, clear up some confusion about the effect usually referred to as ontological destabilization. The discussion will serve as a bridge to the final section where I will argue against the identification of CFNs in film based on metaphorical attribution of deictical properties to film. I will conclude with the criticism of Wilson’s latest attempt to ubiquitously attribute CFNs to film. Wilson (2007) replied to Carroll (2006) and Kania (2005) concerning their proposal that we need not make believe fictional declarative sentences as assertions, and that in turn we need not imagine a narrative agency making those assertions. Rather, according to Carroll and Kania, fictional declarative sentences may be understood merely as containers for propositional content. Thus, sentences like ‘Katie loves Hubbell’ need only express the proposition ‘that || 1 Controlling narrators are narrators whose narrating coincides with the whole of the text and not merely part of it (Currie 1995, 265–270).

186 | Mario Slugan Katie loves Hubbell’. Even if that were the case, Wilson argues, there are sentences in literature, such as conditionals and interrogatives which often employ illocutionary force over and beyond that of pure declaratives. Wilson (2007) asks us to consider the following: “Katie loves Hubbell. Many people thought so. But was it true?” and concludes that the extra illocutionary force found in the question is a sufficient marker of some internal minimal narrating agency. Currie (2010) and Gaut (2010) replied that both utterances following the initial declarative ‘Katie loves Hubbell’ can be thought of as issued from the author and not the fictional narrator. Finding no definite answer to these objections, Wilson (2011, 121) concluded that “these considerations rest finally on claims about the phenomenology of our imaginative engagement with novels and kindred works of literary fiction”. I have proposed that Wilson’s inability to resolve this issue hinges on his failure to see that he, unlike Kania (and others it would seem), models this imaginative engagement on oral narration similar to fictional monodrama and not on the transcribed version of such oral narration (see Slugan 2010, 27–28). However, although both accounts of oral narration fit Chatman’s definition of the text as “any communication that temporally controls its reception by the audience” (Chatman 1990, 7), the proper way to engage in a novel would be to focus on the published text and not on the author-text complex. Once such an understanding of the text is secured, we can directly address the view that declaratives may be understood merely as propositional content and then proceed to an argument stronger than Wilson’s. As previously noted, Carroll et al. claimed that a sentence like ‘Katie loves Hubbell’ is not a fictional assertion but merely a container of propositional content. Yet, propositional content in analytic philosophy is usually discussed in the form ‘X does Y’ or ‘X is Y’. The present simple in these sentences does not describe an action occurring at the moment of speaking, but expresses a fact, a state of affairs or a generalization. These sentences are easily understood without any recourse to temporality. Literary narratives on the other hand, regularly employ verb tenses to express, however imprecisely, the time of a particular event or state of affairs. In the case of ‘Katie loves Hubbell’ the verb in present tense simple does not merely express a state of affairs or generalization as it does in the sentence ‘a mother loves her child’. It says more. It says that with respect to the present Katie is in love with Hubbell. We cannot simply translate sentences which use verbs narratively into propositional content of the form ‘at one point in time X does/is Y’ without losing relevant information. The propositional content must retain a reference to the time frame expressed by the verb tense. But then, how can we fully make believe fictional propositional content P such as ‘X was Y’ without recourse to a present temporal position at which X

Deixis in Literary and Film Fiction | 187

might no longer be Y? This present temporal position, as Émile Benveniste (1971) elaborates, can only be understood as the moment of speaking about the event contained in P. Thus, even if we made-believe that narrative sentences merely contain propositional content, presuming the chain of reasoning is followed through, we should also make believe a controlling fictional narrator no different than the one established by make-believing sentences as fictional assertions. The argument – the linguistic version of the ontological-gap argument2 (henceforth LOGA) – may be formalized as follows: (1) ‘Present’ can only be understood by positing a speaker/writer cotemporaneous with it (Benveniste 1971, 227); (2) The temporal dimension of the narratively used verbs can only be understood with reference to the present tense i. e. the ‘present’ (Benveniste 1971, 226–227);3 (3) (1) and (2) entail that the temporal dimension of the narratively used verbs can only be understood by positing a speaker/writer contemporaneous with the ‘present’ which they refer to; (4) Temporal deictic terms such as ‘now’, ‘yesterday’, ‘tomorrow’ can only be understood by positing a speaker contemporaneous with the ‘present’ which they refer to; (5) (3) and (4) entail that narratively used verbs behave like temporal deictic terms as far as the understanding of their temporal dimension is concerned; (6) In literary narratives, deictic terms used fictionally refer to the agent of fictional narration (the fictional speaker) and not to the agent of actual writing (the actual writer) (see Genette 1980, 214); (7) (5) and (6) entail that any text of literary fiction in which: a) There is no explicit CFN, b) there is at least a single narrative usage of a tensed verb, and c) the usage of which cannot be ascribed to any particular character, has an implicit CFN – i. e. the agent contemporaneous with the present referred to in the tensed verb.

|| 2 Levinson (1996) has produced an (epistemological version of the) ontological-gap argument criticized successfully by Kania (2005) and Wilson (2007). 3 Although disagreeing about the exact details, linguists and narratologists such as Monika Fludernik (1996, 188–189) appear to allow for exceptions to this proposition (e. g. epic preterite and epic present). However, because narratives with effaced narrators written exclusively in one of these tenses make up for only a small subclass of literary fiction, I will not pursue this issue further.

188 | Mario Slugan If this argument is sound, then we can assign CFNs to a huge subclass of literary fiction.4 Non-experimental subclasses which must be excluded include novels written exclusively in the form of direct speech and epistolary novels.5 Experimental texts avoiding the narrative use of tensed verbs should be dealt with on case-to-case basis.

2 Possible objections to LOGA One way to counter this argument is of course to deny that one or some of its propositions hold. Proposition (2) has been attacked on at least one occasion. Currie (2010, 78) invokes the fact that a majority of fictional tales are recounted in the past tense and asks: “How are we to understand this relation if not by postulating an agent who provides testimony of what is past?” He opts for an alternative, however, by claiming that in fictional utterances such as: ‘He sat on the floor’, a breakdown of representational correspondence takes place. According to Currie, a fictional character such as Othello provides an example of such breakdown: Othello is not represented as possessing poetic virtuosity within the fictional world despite the fact that he is represented as uttering exactly those words that the performer acting Othello is uttering.6 Similarly, the argument goes, we should not think that just because every nonfictional narrative report not based on pre-cognition is bound to use the past tense in order to represent the pastness of what took place prior to the act of uttering, the fictional utterance has the same representational function as the nonfictional one. Currie believes that the standard function of the past tense in fictional utterances is only to eliminate explicit markers of their own fictional status by mirroring standard nonfictional practices. What Currie fails to notice is that even if the breakdown of representational correspondence does occur in fictional utterances employing past tense, other || 4 The argument makes no commitments about the ontological status of fictional characters. 5 Rimmon-Kenan (2003) believes that even epistolary novels have a fictional ‘editor’. However, such claims are susceptible to the same arguments Kania (2005) makes against Levinson (1996). The position might be called the material version of the ontological gap argument for somebody fictionally needed to compile all of the letters. It fails because it is fictionally indeterminate whether compilation took place or not. 6 Using the phrase ‘fictional world’ does not mean I support explanations of fiction in terms of possible worlds (e. g. Thomas Pavel 1986; Lubomir Doležel 1998). In fact, my account is neutral as far as this matter is concerned. I turn to the ‘world’ metaphor to emphasize the spatiotemporal coordinates around which deixis revolves.

Deixis in Literary and Film Fiction | 189

non-representative, i. e. analytic properties of such utterances remain operational. Consider utterances such as: ‘He is walking down the street and runs into an acquaintance of his’. Regardless of whether these utterances are fictional or not they can easily be understood to represent the pastness of past events within the proper context of narrative recounting, such as for example: ‘Let me tell you what happened to Joe yesterday’. Thus, other tenses may represent pastness in appropriate contexts. In other words, the breakdown of representational correspondence between time frames and tenses employed is no extraordinary phenomenon. However, there is another important analytic property here, one usually hinted at in the accounts of the experience of hearing or reading a story told in such a way. Such accounts usually say something like: The event is made more immediate to the audience. The narratological basis of this effect is the shift to the present tense, which allows for the contemporaneous positioning of the narrating agency with the past event being described. And it is exactly this bond between the narratively used tense and the narrating agency I refer to as the analytic property Currie misses. This is best understood in relation to Benveniste’s (1971, 226–227) claims that the present is nothing but “the coincidence of the event described with the instance of discourse that describes it” and that “the time at which one is” is nothing but “the time at which one is speaking”. Thus, there is a discrepancy between the representational function of a particular tense on the one hand, and the analytic relation between the narrating agency and the event recounted (established by the temporal dimension of that tense) on the other. Currie misses this because he focuses exclusively on utterances employing past tense where these two are conflated. Another possible objection to my argument boils down to an appeal to imaginative indeterminacy. According to Gaut (2010, 211) imaginative indeterminacy presents ways to ward off asking silly questions – i. e. questions to which there are no answers in the fictional world, and that are best answered with reference to the aesthetic practices and intended effects of the work in question. One type of indeterminacy is invoked in order to allow for contradictions within the fictional world once they are made explicit. For instance, we need not imagine how ‘Marcel’ – the narrator of Marcel Proust’s In Search of Lost Time – knows the subjective states of characters he has never spoken to in order to fictionally take him at his word when he recounts those states.7 This is entirely in line with Gaut’s (2004, 205) own formulation of the Reality Heuristic (henceforth RH): “We should attempt, other things being equal, to render them [i. e. || 7 I use ‘Marcel’ following Genette but there is more to be said about fictional ‘Marcel’ as opposed to actual Marcel in the next section.

190 | Mario Slugan fictional worlds] as like the real word as we can.” Because ‘Marcel’ is an explicit narrator with the property of omniscience clearly imputed to him, ‘Marcel’s’ omniscience overrides the heuristic which would have us think of ‘Marcel’ as only being able to recount what he witnessed. In other words, ‘Marcel’s’ omniscience is precisely that which contravenes the ‘other things being equal’ qualification. However, if no explicit narrator were present, the RH ought to take charge. This would mean in turn that postulating implicit narrators, as I do via deixis, would lead us to ask silly questions about the omniscience of such a narrator. Gaut’s proposal is to avoid posing silly questions whenever possible, which means we should not postulate anything non-explicit. The other type of indeterminacy suspends the chain of reasoning by implication. This gives another slant to Currie’s discussion of the breakdown of representational correspondence. For example, imagining that Othello speaks in beautiful verse does not entail that we must also imagine how a mere soldier could have acquired such talent; nor should we look for internal reasons why all of the characters in Leonardo da Vinci’s Last Supper find themselves on one side of the table. That the characters in Shakespearian tragedies are masters of verse is a convention of the genre and representational and compositional clarity explain, in no small part, why the characters of Last Supper are all on the same side of the table. On this variant, the skeptic could simply refuse to follow the chain of reasoning, arguing that we need not make believe a present temporal position implied in the temporally used verbs, and in turn, that we need not make believe an agency occupying that position. All of these objections, Currie’s included, implicitly introduce what I believe to be another type of indeterminacy only hinted at by Gaut. I propose to call it medium indeterminacy. All appear to state, in one way or another, that just because the language is employed fictionally and not actually, we are free to ignore the analytic entailments which would hold for its actual employment and which would in turn disqualify the entailment that there is a narrative agency behind the temporally used verbs. In the second type of indeterminacy, asking how Othello can be in such a superb command of verse is silly because the story is not concerned with providing an answer. And asking about the character placement in Last Supper again proves otiose from the perspective of what is represented. However, asking about the existence of narrators from a linguistic perspective cannot be a silly question in this way, for the response pertains neither to the level of fictional diegesis nor to the level of aesthetic practices and intended effects. It pertains to the properties of the medium of language. In other words, it is not some chain of implications internal to the story or the story-world that is followed, but one pertaining to the analytic properties of language. Put differently again, if we were free to ignore the chain of reasoning in

Deixis in Literary and Film Fiction | 191

LOGA we would also be free to dismiss semantic and syntactic properties that ‘loves’ in the fictional statement ‘Katie loves Hubbell’ possesses. We could also claim that it is not fictional that Othello speaks in verse. But all of this is clearly fictionally false. This discussion also provides an answer to the objection stemming from the first type of indeterminacy. It should be the case that the RH applies not only to the fictional world but to the medium as well. Thus, when there are no explicit narrators, we can, and should, track the chain of implications which the language medium affords us. Perhaps Gaut could respond to this by saying that avoiding silly questions should take precedence over the application of the RH in its language medium aspect. But here the choice to follow one chain of analytic entailments in order to argue against Wilson’s take on ‘omniscient’ narrators, as Gaut does, and not another (e. g. the one pertaining to the linguistic version of the ontological argument) seems rather arbitrary. The initial articulation of LOGA revolved around the Genettian assumption (6) that deixis in literary fiction is intra-ontological, i. e. that deictic reference cannot cross ontological boundaries between fiction and actuality. In the next section, I aim to address the question whether this assumption holds more generally and, if not, how it affects LOGA. In order to do so, I turn to fictional second-person narratives in literature and the ontological status of the ‘you’ address therein. Then, in the last section, I examine the supposed analogue to ‘you’ address in fiction film and the role it plays in the arguments for the existence of CFNs in film.

3 Second-person addresses in literary fiction Since Bruce Morrissette’s seminal paper Narrative ‘You’ in Contemporary Literature from 1965, scholarship on second-person fictional narratives has proliferated steadily. Various typologies have been proposed and many papers have been written to explain the rhetorical effects of such stories.8 Of course, the textual presence of ‘you’ address is neither a necessary nor a sufficient condition for the text to count as a second-person narrative. Usually, the defining trait is that the protagonist needs to be addressed, i. e. that the narratee is the protagonist.9 Yet || 8 For a recent overview of the scholarship on the subject see Rolf Reitan (2011). Of particular importance is the special volume of the journal Style on second-person fiction (Style 1994.3). 9 Monika Fludernik (1994a) points out that the use of second-person pronouns is not even necessary – made-up pronouns will do.

192 | Mario Slugan even if the ‘you’ address is neither necessary nor sufficient for second-person narrative, insights from this field of study will be very helpful for our own investigation. Two issues need to be discussed in this respect: The ontological status of ‘you’ addresses and the instances of second-person fiction in which ‘you’ addresses function as self-addresses. There is general agreement among second-person narrative scholars that, although addressing fictional characters on most occasions, ‘you’ may address actual readers on others. An oft-cited example is the opening sentence of Italo Calvino’s If on a Winter’s Night a Traveler: “You are about to begin reading Italo Calvino’s new novel, If on a winter’s night a traveler” (Calvino, 1981). Consider proposition (6) again, reformulated to include both the addresser and the addressee: (6’) In literary narratives deictic terms used fictionally refer to: 1) The agent of fictional narration (the fictional speaker) and not to the agent of actual writing (the actual writer), and 2) the fictional addressee and not the actual reader. On first inspection Calvino’s sentence suggests that proposition (6’) should allow for second-person addresses as exceptions. Indeed, at least one theoretician explicitly states the possibility of the inter-ontological breach by distinguishing between the character-address and the actual reader-address. David Herman (1994, 381) juxtaposes the “fictionalized address”, “which entails address to and/or by the members of some fictional world and thus constitutes ‘horizontal’ address […]”, to the “‘actualized address’ or ‘apostrophe’, which […] entails address that exceeds the frame (or ontological threshold) of a fiction to reach the audience, thus constituting ‘vertical’ address”. Such analysis is flawed, however. Genette’s claim that deixis must be intraontological remains intact. The most important thing to remember is that the address to the reader is fictional (or fictionally true or it is true in fiction).10 This does not mean that the strange effect usually described with (imprecise) phrases such as ‘the reader is pulled into the fictional word’ or ‘the boundary between fiction and reality is destabilized’ does not take place. It does mean, however, that although the actual reader is prescribed by the text to makebelieve herself as addressed and as partaking in the act of fictional reading, strictly speaking, the ‘you’ address in question does not cross ontological borders. The further possibility of confusion stems from the fact that the actual || 10 My account is neutral with regard to the construal of fictional truth.

Deixis in Literary and Film Fiction | 193

reader is actually reading. Moreover, one can say that the novel’s opening line may be understood as producing an actual address. This is true, but in that case, the actual address directed to the actual reader is no other than the actual author. In both accounts the intra-ontological reference of deixis remains intact. There is also widespread agreement that the narratee-character in secondperson narratives may at the same time be the fictional narrator, if the text allows for the ‘you’ addresses to be construed as self-addresses. An oft-cited example of this is Edna O’Brien’s novel A Pagan Place. Paired with the possibility to fictionally address the actual reader, this creates an interesting situation in which the actual reader is no other than the fictional narrator. Although I am not familiar with such an instance of second-person narrative, it is not difficult to imagine what one would look like.11 Consider a possible opening sentence of a fictional novel: ‘You, yes you, you who are reading these lines at the very moment you are writing them down’. In this case there seems to be no way around the conclusion that it is fictionally true that ‘you’ – the actual reader – is at the same time the fictional narrator. And if this is the case, then (6’), although true as far as the claim about intra-ontological reference of deixis, needs to be rearticulated. We should say the following: (6’’) In literary narratives deictic terms used fictionally refer to: 1) The fictional speaker and 2) the fictional addressee (both of whom may be an actually existing entity). Wilson (2011) would add that there is nothing curious about this because, according to the description of his preferred phenomenological engagement with fiction, it is not rare that actual entities act as fictional narrators. In Robert Coover’s The Public Burning, to give just one example, actual Nixon fictionally narrates the story. In other words, Wilson is saying something like q: It is fictionally true that actual Nixon narrates the events of June 1953 recounted on the pages of Coover’s book. Others such as Ljubomir Doložel and Gérard Genette would presumably claim something more like r: It is fictionally true that a fictional entity bearing the name ‘Nixon’ (which is supposed to invoke actual Nixon) is the one who narrates.12 According to LOGA, the deixis employed in the text of the book refers to an entity with fictional spatiotemporal coordinates. By itself, LOGA cannot decide between whether actual Nixon or a fictional entity by the || 11 I disagree with Fludernik’s (1994b) brief account of Margaret Atwood’s Happy Endings. 12 Doložel would do so, since he subscribes to fictional worlds as possible ones (see Doložel 1998), and Genette does so in order to avoid naïve practices in literary criticism.

194 | Mario Slugan name of ‘Nixon’ is fictionally narrating. The matter seems to come down to the question of the preferred phenomenological engagement with fiction. Yet at least one conclusion Wilson draws from this is worth addressing at greater length. It does not follow that Wilson is warranted in identifying actual authors as fictional narrators in the cases of effaced narrators or narratives in which the narrator remains nameless. Wilson finds it plausible that the actual Graham Greene is the fictional narrator of The Heart of the Matter and that the actual Jane Austen is the fictional narrator of Emma. There is certainly nothing in the text that explicitly precludes us from saying this, but following Wilson’s line of reasoning neither is there anything in Macbeth to preclude us from saying that Lady Macbeth had exactly six children. However, such assertions are indeterminate at best. Moreover, on many occasions they are most probably fictionally false.13 Consider transparent science fiction narratives such as Philip K. Dick’s Do Androids Dream of Electric Sheep. Here it would be implausible to make believe that Dick is fictionally narrating Deckard’s story for it would be at odds with the story’s future setting. Wouldn’t it be easier to assume that somebody fictional, somebody ‘naturally’ belonging to his world is recounting Deckard’s story? To make believe Dick as the fictional narrator puts unnecessary strain on the fictional world, which would be served better by a nameless fictional narrator of Dick’s world. Think of this as an application of the RH. In short, the plausibility of Wilson’s description of phenomenological engagement appears to rest on picking out examples which are both ‘realistic’ and ‘contemporary’ to the time of actual writing. This is why, on some occasions, it does not seem unnatural to pick out the author as the fictional narrator of narratives with effaced narrators since the fictional world is very close to the author’s. I would advise, however, to refrain from such identifications if we cannot say (with reasonable certainty) that they are fictionally true. Let us return to the case of the actual reader as the fictional narrator. I think it possible that, regardless of the way I (or someone else) worded a fictional sentence with the intention to fictionally address the actual reader, an interpretation might be devised which claims that ‘you’ does not do the job it was intended to do. On this account, it would only be fictionally true that ‘you’ addresses a fictional narratee and not the actual reader. In some sense, this would be enough to disqualify the claim that an actual person can also be a fictional narrator. However, it could be objected that such an argument would be highly interpretative and at odds with the standard descriptions of phenomenological engagement with ‘you’ engendered by novels like Calvino’s. If we agree that || 13 Here I am close to agreeing with Currie (1990) that fictional truths are a matter of degree.

Deixis in Literary and Film Fiction | 195

‘you’ in Calvino’s and my example fictionally addresses the actual reader, it would appear that actual ‘you’ fictionally acting as the narrator is no different from actual Nixon fictionally narrating, and that fictional ‘you’ doing the same is no different from ‘Nixon’ recounting. We should be more precise about this, however, especially because of the gnawing impression of ontological fuzziness which seems to afflict the examples of ‘you’ addresses somewhat differently than the instances concerning Nixon or ‘Nixon’ as the narrator. The difference is of course that in these texts, proper names denote, but do not address.14 It does not matter that there is a conflation between the protagonist and the narrator in my example. They are merely roles of one and the same entity with spatiotemporal coordinates within the fictional world. From the perspective of there being only one such entity, actual ‘you’ is completely like Nixon and fictional ‘you’ is completely like ‘Nixon’.15 It does seem to be more difficult for the actual reader to accept actual ‘you’ as the fictional narrator than it would be for her to accept actual Nixon as the narrator. This is usually accounted for in terms of the discrepancies between the narrator’s properties and those of the actual reader. But if this were the only issue, it would be no easier for actual Nixon to accept actual Nixon as the fictional character. I suggest that there is more to the ontological fuzziness than what we might call proneness for fictional engagement with ourselves. The cause of this gnawing impression resides, at least partially, in the double address ‘you’ instantiates even in narratives in which ‘you’ is taken to fictionally address actual ‘you’. Simply put, even in my example there remains a lingering doubt that it is not only the actual ‘you’ who is fictionally addressed, but the fictional ‘you’ as well, a ‘you’ belonging to the fictional world more ‘naturally’ than the actual reader, not unlike the anonymous narrator in Dick’s novel. Again, this concerns neither the extent to which the actual reader is fictionally engaged with herself as the protagonist, nor whether ‘you’ hovers somewhere between horizontal and vertical address, as is the case in what Herman calls double deixis. It concerns uncertainty whether we as actual readers are the only ones being fictionally addressed. With ‘Nixon’ and Nixon we opt for one or the other according to the preferences of our phenomenological engagement with fiction and run with it. In the case of actual ‘you’ addresses, we make believe ourselves doing things, but cannot help but make believe some|| 14 ‘Denote’ should be taken with reservation here: It is not meant to open the can of worms known as ‘fictional reference’ but only to distinguish the general function of denotation and the one of address. Also, ordinary use of proper names to address people is not important here. 15 The case of ‘Marcel’ or Marcel doesn’t even differ as far as the separation of the roles is concerned.

196 | Mario Slugan body else as well, who, being ‘naturally’ fictional, remains a separate addressee. We might call this Janus-faced deixis.16

4 Analogues to second-person addresses in fiction film Turning to film, enunciation theorists such as Mark Nash (1976) and André Gaudreault and François Jost (1999) have traced the equivalents of ‘you’ addresses to the character’s look at the camera. Although they themselves are unclear whether such addresses cross fictional boundaries, as in the case of their literary counterparts, there is no compelling reason to think that they do.17 Of more importance for this discussion is that, according to scholars applying linguistic theories to film, such addresses are informative of CFNs. In other words, a look at the camera is a marker of enunciation which “trace[s] the presence of something beyond the diegesis, a ‘grand image-maker’” (Gaudreault and Jost 1999, 48). On this account, ‘you’ addresses (together with other shots including exaggerated foregrounds or low angle point of views) establish the presence of a CFN because they are discursive in Benveniste’s sense of expressing subjectivity. The argument seems to say something like s: If, according to Benveniste, every instance of ‘you’ discourse entails ‘I’, and if the camera-look is equivalent to ‘you’ address, then there must be ‘I’ uttering this address and this ‘I’ is no other than the filmic narrator/enunciator/grand image-maker, i. e. the CFN. Although the application of enunciation theories has suffered well-argued criticism by David Bordwell (1985, 21–26 and Nöel Carroll (1988, 150–160), it mysteriously persists (usually by not engaging with this criticism at all).18 I will not rehash Bordwell et al.’s arguments but merely point to another way of dismissing the enunciator theorists’ claims as they pertain to the question of the filmic narrator. I will also ignore the ontological contradictions present in their

|| 16 I would like to thank Alexander Bareis and Lene Nordrum for suggesting this name. 17 Inadvertent looks at the camera might best be dealt with by bracketing them off from what we are to make-believe and just treat them as actual looks at the camera, glitches in the game of make-believe. 18 Enunciation theorists I discuss are no exceptions.

Deixis in Literary and Film Fiction | 197

understanding of the CFN and assume it is only fictionally true that the CFN narrates.19 When Christian Metz (1991, 769) tries to find a common denominator for all the filmic devices evoking subjectivity, he effectively explains why they are not discursive: “All figures of enunciation consist in metadiscursive folds of cinematic instances piled on top of each other.” The use of the term ‘metadiscursive’ should not fool us here. As can be seen from his famous statement that the filmic shot resembles enunciation more than it does a word (e. g. picture of X ought to be understood as saying ‘Here is X’), Metz (1986) assumes filmic narrative is already discursive, i. e. that it regularly employs filmic versions of linguistic deixis. Because these figures are both discursive and reflexive, they must also be metadiscursive. What Metz actually says is that these figures are metanarrative in Linda Hutcheon’s sense of the word. A metanarrative text “provides, within itself, a commentary on its own status as fiction and as language, and also on its own processes of production and reception” (Hutcheon 1980, xii). This, in turn, is a subclass of Viktor Shklovsky’s, or later, Bertolt Brecht’s well-known concepts of estrangement and alienation effect, respectively. The point is that metanarrative moments are often metadiscursive only in oral and especially in written narrative, because they relate to the narrator’s subjectivity in one way or another. The most famous example is of course the CFN in Laurence Sterne’s Tristram Shandy who fictionally regularly addresses the audience and comments on the process of writing. However, metanarrative moments need not produce fictional truths about the existence of a narrator. If a character who is not a narrator addresses an actual reader by saying: ‘Dear sir, you should not think of me as a mere character’ – arguably an analogue to a character’s look at the camera – this statement generates no fictional truths about the narrator. The only ‘I’ ‘you’ entails is the character uttering the sentence. And this can be informative of the CFN only if the character is the CFN. By way of analogy, the only ‘I’ the camera-look entails is the character doing the looking. The enunciator theorist might reply that we have taken things too literally. We should not be so close-minded, but should understand the look as emanating not only from the character but as a function of the CFN. Following Francesco Casetti’s (1988, 45–50) account of cameralooks, we should say something like: ‘This [the look] is for you; it comes from the cinema, that is, from me [the CFN].’ However, such accounts are purely || 19 See for instance Gaudreault (2009, 87, 96) on monstroator and Metz’s conflicting views on great-image maker’s spatiotemporal position (Metz 1986, 40) and embodiment (Metz 1991, 759).

198 | Mario Slugan metaphoric and have little or no regard for what is fictionally true. For a metanarrative moment in a filmic text to be discursive, it would have to fictionally tell us something qualitatively about the narrator, but from the examples listed above it clearly does not. Simply inserting deixis into one’s linguistic descriptions of film similar to Casetti does not do the trick. There is general agreement about the near absence of CFNs in film among analytic philosophers. The reason is that, generally speaking, images lack the structural abilities of language to automatically generate fictional truths about the existence of agents fictionally responsible for their production. The lone voice in favour of CFNs remains Wilson with his Fictional Showing Hypothesis, which he presents as a necessary condition for the existence of minimal narrating agency. As it stands, I believe it has been adequately criticized.20 Rather than pursuing old arguments, I will construct a new and more fundamental reproach to Wilson’s understanding of CFNs in film. In his latest work, Wilson (2011, 55) notes that the Fictional Showing Hypothesis rests on the Imagined Seeing Thesis. Therefore, the Fictional Showing Hypothesis is invalid if the Imagined Seeing Thesis is. Wilson convincingly dismisses Carroll’s (2006) criticism of the Imagined Seeing Thesis, demonstrates that Currie’s (1995) concept of perceptual imagining is in relevant aspects no different than his own thesis, and elaborates the thesis’ conceptual coherence in a detailed manner. My intention is not so much to deny the relevance of Wilson’s thesis for the epistemology of film, but to demonstrate its inapplicability to the discussion of CFNs.21 The reason for this stems from the confusion between game-worlds and work-worlds. Wilson claims that it is fictional in the work (or work-fictional) that spectators are seeing cinematic images and in turn being presented with them.22 He bases this claim on Kendall Walton’s understanding of the work-world (see Walton 1990, 57–61). More specifically, for soand-so to be work-fictional means that it is the work’s standard function to prompt appreciators to make believe so-and-so:

|| 20 It has been shown that by subscribing to complex fictional entities such as ‘naturally iconic images’, Fictional Showing Hypothesis fails to meet the RH and that it invokes the material version of the ontological gap argument. See Gaut (2004), Carroll (2006) and Slugan (2010). For a critique of Slavoj Žižek’s understanding of enunciation see Slugan (2013, 739–741). 21 The argument is produced for the Mediated Version of Imagined Seeing Thesis Wilson endorses but works for all of its versions. 22 Wilson illuminative account of seeing so-and-so in terms of having an impression of its being as if seeing so-and-so is orthogonal to my argument.

Deixis in Literary and Film Fiction | 199 [I]t is a standard function of a cinematic work of fiction to prompt viewers to imagine―to make believe―themselves being shown the narrative events and circumstances of successive shots. Moreover […] it is fictional in the movie […] Since ‘fictional showing’ is putatively what the movie’s images are meant to achieve, and ‘imagined seeing’ is putatively what movie viewers do in response to those images, it is often easier to formulate certain points in terms of one thesis rather than the other. But, to repeat, the two theses are utterly interdependent, although, of the two, the Imagined Seeing Thesis is probably the more fundamental. (Wilson 2011, 55)

Walton has more to say about the work’s standard function. He claims something is work-fictional as long as it holds for any game the work prompts (Walton 1990, 60). Otherwise it is game-fictional. To use Walton’s example, although it is both work- and game-fictional that there is a couple strolling in Georges Seurat’s La Grande Jatte, it is only game-fictional that Richard is seeing the couple strolling. This depends on the fact that it is Richard who is appropriately playing an authorized game of make-believe with the painting as a prop. If Mary were to do the same, then it would be game-fictional that she is seeing the couple. Now, Wilson cannot exchange Richard and Mary for a variable X and say that it is work-fictional that X is seeing the couple, for this is clearly false. An alternative might be to say something like: Although there is no work-fictional character that sees, there is subject-less work-fictional ‘seeing’. But this would not work either. For when we are seeing a couple strolling in La Grande Jatte, we are imagining seeing the couple but we are not imagining some subject-less ‘seeing’ on top of that. Thus, the Imagined Seeing Thesis and, in turn the Fictional Showing Hypothesis, are game- but not work-fictional.23 As such, they cannot tell us anything about CFNs because their (non-)existence is workfictional.24 This, of course, does not mean that film as a medium necessarily precludes the existence of CFNs.25 Films such as Alexander Sokurov’s Russian Ark consisting exclusively of point of view shots accompanied by intradiegetic sounds only illustrate a case analogous to literary autodiegetic narrators recounting their stories in the present tense. Mocumentaries such as Rob Reiner’s This is Spinal Tap or recent horror films such as Oren Peli’s Paranormal Activity make it work|| 23 The following demonstrates at least a possible inadvertent admission of this: “They [spectators] imagine, falsely but quite legitimately, about the shots of the film, that those have been transparently derived from certain visible constituents of the fictional world that the film creates” (Wilson 2011, 90, italics in the original). That there is confusion here is supported by direct contradiction with the in-text citation above. 24 Note that LOGA does not depend on any abstract imagined reading or hearing. 25 For more details see Slugan (2014).

200 | Mario Slugan fictional that there is a fictional recording of fictional events identical to the actual recording we see.26 If we modified these films slightly, we could even construct examples of second-person narratives in which it is fictional that the actual spectator is the narrator. For instance, having characters in mocumentaries address the actual spectator and making it fictional that she is holding the camera and filming would amount to this. We can leave the working out of these details to actual filmmakers. Regardless of their implementation, however, we can be certain that any addresses to the audience would remain intraontological.

References Benveniste, Émile. Problems in General Linguistics. Translation, M. E. Meek. Coral Gables, Fla: University of Miami Press, 1971. Bordwell, David. Narration in the Fiction Film. Madison: University of Wisconsin Press, 1985. Calvino, Italo. If on a Winter’s Night a Traveler. Translation, William Weaver. New York: Harcourt Brace, 1981. Casetti, Francesco. Inside the Gaze: The Fiction Film and Its Spectator. Translation, Neill Andrew. Bloomington, IN: Indiana University Press, 1988. Carroll, Noël. “‘Introduction’ to ‘Part IV: Film Narrative/Narration’.” Philosophy of Film and Motion Pictures. Eds. Noël Carroll and Jinhee Choi. Malden, MA: Blackwell, 2006, 175– 184. Chatman, Seymour. Coming to Terms: The Rhetoric of Narrative in Fiction and Film. Ithaca, NY: Cornell University Press, 1990. Currie, Gregory. The Nature of Fiction. Cambridge: Cambridge University Press, 1990. Currie, Gregory. Image and Mind: Film, Philosophy and Cognitive Science. Cambridge: Cambridge University Press, 1995. Currie, Gregory. Narratives & Narrators: A Philosophy of Stories. Oxford: Oxford University Press, 2010. Doležel, Lubomir. “Possible Worlds of Fiction and History.” New Literary History 29.4 (1998): 785–809. Fludernik, Monika. “Introduction: Second-Person Narrative and Related Issues.” Style 28.3 (1994a): 281–311. Fludernik, Monika. “Second-Person Narrative: A Bibliography.” Style 28.3 (1994b): 525–539. Fludernik, Monika. Towards a Natural Narratology. London, New York: Routledge, 1996. Gaudreault, André and François Jost. “Enunciation and Narration.” A Companion to Film Theory. Eds. Toby Miller and Robert Stam. Oxford: Blackwell, 1999. 45–64. Gaudreault, André. From Plato to Lumière: Narration and Monstration in Literature and Cinema. Translation, Timothy Barnard. Toronto: University of Toronto Press, 2009. || 26 I would suggest that the exceptionality and ‘strangeness’ of such examples is a good reason not to think we imagine anything like fictionally transparent images when watching movies.

Deixis in Literary and Film Fiction | 201 Gaut, Berys. “The Philosophy of the Movies: Cinematic Narration.” The Blackwell Guide to Aesthetics. Ed. Peter Kivy. Malden, MA: Blackwell Publishing, 2004. 230–53. Gaut, Berys. A Philosophy of Cinematic Art. Cambridge: Cambridge University Press, 2010. Genette, Gérard. Narrative Discourse: An Essay in Method. Translation, Jane E. Lewin. Ithaca, NY: Cornell University Press, 1980. Herman, David. “Textual ‘You’ and double deixis in Edna O’Brien’s A Pagan Place.” Style 28.3 (1994): 378–410. Hutcheon, Linda. Narcissistic Narrative: The Metafictional Paradox. Waterloo, Ont: Wilifred Laurier University Press, 1980. Kania, Andrew. “Against the Ubiquity of Fictional Narrators.” Journal of Aesthetics and Art Criticism 63.1 (2005): 47–54. Levinson, Jerold. “Film Music and Narrative Agency.” Eds. David Bordwell and Nöel Carroll. Post-Theory: Reconstructing Film Studies. Madison, WI: University of Wisconsin Press, 1996, 248–82. Metz, Christian. “Problems of Denotation in the Fiction Film.” Ed. Philip Rosen. Narrative, Apparatus, Ideology. New York: Columbia University Press, 1986, 35–65. Metz, Christian. “The Impersonal Enunciation, or the Site of Film (In the Margin of Recent Works on Enunciation in Cinema).” New Literary History 22.3 (1991): 747–772. Morrissette, Bruce. “Narrative ‘You’ in Contemporary Literature.” Comparative Literature Studies 2 (1965): 1–24. Nash, Mark. “Vampyr and the Fantastic.” Screen 17.3 (1976): 29–67. Pavel, Thomas. Fictional Worlds. Cambridge, MA: Harvard University Press, 1986. Reitan, Rolf. “Theorizing Second-Person Narratives: A Backwater Project?” Strange Voices in Narrative Fiction. Ed. Per Krogh Hansen. Berlin, New York: De Gruyter, 2011, 147–174. Rimmon-Kenan, Shlomith. Narrative Fiction. London, New York: Routledge, 2003. Slugan, Mario. “An Asymmetry of Implicit Fictional Narrators in Literature and Film.” Postgraduate Journal of Aesthetics 7.2 (2010): 26–37. Slugan, Mario. “The Rhetorics of Interpretation and Žižek’s Approach to Film. ” Slavic Review 72.5 (2013): 739–749. Slugan, Mario. “Some Thoughts on Controlling Fictional Narrators in Fiction Film.” American Society for Aesthetics Graduate E-Journal 6.2 (2014): 1–7. Walton, Kendall. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Wilson, George. “Elusive Narrators in Literature and Film.” Philosophical Studies 135.1 (2007): 73–88. Wilson, George. Seeing Fictions in Film: The Epistemology of Movies. Oxford: Oxford University Press, 2011.

Liviu Lutas

Metalepsis and Participation in Games of Make-Believe Narrative metalepsis is a concept stemming from narratology. That is why its use is sometimes restricted to the field of literature. Indeed, because of its close connection to the voice of a narrator – defined by Gérard Genette (1980 [1972], 235) as the transgression of the boundary separating the narrator from the diegetic universe – narrative metalepsis apparently requires the linguistic frame of narrative discourse. However, with the acceptance of narratology in other disciplines and media, the narratological concept of metalepsis has come closer and closer to what Werner Wolf (2005) calls a transmedial device – that is, a device whose use is not restricted to one single medium. This transmediality does not mean that the device is used in exactly the same manner in all media. Rather, as Wolf (2005, 104) points out, the transfer of the concept of metalepsis from literature to other forms of art or media makes it possible to “highlight […] formal, functional and historical similarities” between these. One might add that a comparison of metalepsis as it appears in different media could be used for refining the original definition of the term itself. It must be said that Genette hinted from the beginning at the transmediality of the concept of metalepsis. Not only did he (Genette 1980, 235–237) explicitly mention other media in relation to metalepsis – theatre, paintings and cinema – but he stressed its representational aspect when he distinguished between “two worlds, the world in which one tells, the world of which one tells” (in original, Genette 1972, 245). Eleven years after the first definition, Genette (1988, 88) emphasized the importance of representation for the metaleptical transgressions even more strongly, defining them as “deliberate transgression of the threshold of embedding”.1 As Jean-Marie Schaeffer (2005, 325 and 327) pointed out, this redefinition permitted the transfer of the concept to the field of representation in general. It is, as we shall see, thanks to this close relation to representation that metalepsis can be analysed according do Kendall Walton’s theory of make-believe (1990, 4), in which representation is central to the point of being interchangeable with the concept of fiction (see even Ryan 2001, 106). What appears as a constant of metalepsis throughout its different medial incarnations is its reflexive dimension. Indeed, since it points to the existence of two levels of representation, it seems beyond doubt that metalepsis possesses || 1 In original: “Transgression délibérée du seuil d’enchâssement” (Genette 1983, 58).

204 | Liviu Lutas the capacity of signaling the constructed nature of the story.2 The question is if this reflexive capacity automatically entails an anti-illusionist effect on the recipient. This is for instance the conclusion of Debra Malina (2002, 138), who defines metalepsis as a “breach in narrative structure that undermines the narrative’s illusions”. In this paper, I try to challenge theories according to which narrative metalepses automatically block the reader’s participation and immersion in a work of fiction. The paper starts with a theoretical introduction, in which I discuss the concepts which will be important in the other parts of the paper. In these other parts, I analyse some examples from three different media in order to challenge the theoretical concepts I present first.

1 Make-believe and narratology – A theoretical prolegomenon I will start by presenting my understanding of the concept which is central to this paper, and indeed to the rest of this volume: Make-believe. I will follow Kendall Walton’s definition of the concept: For me, make-believe is, as it is for Walton, the attitude of a recipient – not only a reader – towards a work of fiction. We, as recipients, are intended to make believe, in other words, to pretend, that the world represented in the fiction work is true. Admittedly, the concept of make-believe is much more complex than this short definition. Still, a short definition like this one is useful when trying to analyse the relation between the world of the writer and that of the reader – in other words, the world of the representation – on the one hand, and the fictional world – the diegesis, in the terminology of Gérard Genette, or the represented world – on the other. Obviously, as I already mentioned in the introduction, the make-believe theory is important when analysing the narrative device which is at the core of my presentation: the narrative metalepsis. Indeed, according to Gérard Genette’s definition from 1972, “metalepsis is the transition from one narrative level to another” (Genette 1980, 234; in original 1972, 243). Narrative levels are the level of the teller and the level of the story itself; as stated in the second quotation from Genette (1980, 236): Metalepsis is the transgression of “a bound|| 2 See Wagner (2002, 239); in original: “Les métalepses, dans leur diversité, ont toutes en commun le fait de signaler l’essence construite du récit, c’est-à-dire le procès de la textualisation.”

Metalepsis and Participation in Games of Make-Believe | 205

ary 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” (original emphasis).3 In a more recent translation of the passage into English – “the world where narration takes place and the world which is narrated” (Kukkonen, 2011, 4) – the representational dimension appears more clearly, something that is also highlighed in Sonja Klimek’s (2009, 169) use of the terms: “the level of representation (or the world ‘où l’on raconte’) and the level of what is represented (‘celui que l’on raconte’).” The common point of the concepts of metalepsis and make-believe is thus transparent. As it seems, both metalepsis and make-believe have to do with the relationship between the world of representation and the represented world. This can explain why, even though the definitions above do not seem to allow that, the use of metalepsis has lately been extended outside the field of literature, even to non-verbal media. This takes me to another important aspect of the premises of this paper: the transmediality of the concept of make-believe. I agree with Kendall Walton’s view that all make-believe is in essence fiction, and all fiction is make-believe. Walton makes his point by comparing works of fiction from different media with children’s games of make-believe: In order to understand paintings, plays, films, and novels, we must first look at dolls, hobbyhorses, toy trucks, and teddy bears. The activities in which representational works of art are embedded and which give them their point are best seen as continuous with children’s games of make-believe. Indeed, I advocate regarding these activities as games of make-believe themselves, and I shall argue that representational works function as props in such games. (Walton 1990, 11)

The transmediality of the concept of fiction is underscored several times by Walton. For instance, he (Walton 1990, 75) argues that “not all fiction is linguistic. Any adequate theory of fiction must be able to accommodate pictorial fictions, for instance, as well as literary ones”. Admittedly, the alleged synonymy between fiction, make-believe and representation as well as the transmediality of concepts like fiction or metalepsis could be debated, but the main aim of this paper is not to take part in such a debate. I will use examples from different media – painting, literature and film

|| 3 In original: “Tous ces jeux manifestent par l’intensité de leurs effets l’importance de la limite qu’ils [les auteurs] s’ingénient à franchir au mépris de la vraisemblance, et qui est précisément la narration (ou la représentation) elle-même; frontière mouvante mais sacrée entre deux mondes: celui où l’on raconte, celui que l’on raconte” (Genette 1972, 245).

206 | Liviu Lutas – in order to illustrate my points, and I will regard fiction as a mode of representation. One of the most important aspects when studying the theory of makebelieve in relation to narrative metalepsis is the relation between participation – the recipient’s “experience of being ‘caught up in a story,’ emotionally involved in the world of a novel or play or painting” (Walton 1990, 6) – and appreciation – a more reflexive and distanced attitude towards the work of fiction, the “aesthetician’s” attitude (Walton 1990, 6). Kendall Walton dedicated a whole chapter, chapter 6, or a total of 80 pages in his book Mimesis as Make-Believe, to the problem of participation (Walton 1990, 209–289). According to Walton (1990, 213), “viewers of paintings and films, spectators of plays, readers of novels […] participate in the games in which these works are props much as children participate in games”. In other words, “[c]hildren are almost invariably characters in their games of make-believe” (Walton 1990, 209). This shows us the importance of participation in the activities of reception of representational works of art. The concept of participation is very close to what Marie-Laure Ryan calls immersion. This is obvious when looking at Ryan’s (2001, 14) definition of immersion in relation to reading: “In the phenomenology of reading, immersion is the experience through which a fictional world acquires the presence of an autonomous, language independent reality populated with human beings.” The other possible reaction to works of art, appreciation, is characterized by a more distanced, or reflexive, attitude. Appreciation usually involves participation, but this is not always the case. According to Walton (1990, 275), especially the participation of the kind “being caught up in a story” is discouraged for instance by works “declaring or displaying their fictionality”. In Walton’s own words: One obvious way in which works sometimes discourage participation is by prominently declaring or displaying their fictionality, betraying their own pretense […] We are not likely to imagine very vividly that we are being told truths about actual events, when we read a story, if (and while) it forces on our attention the fact that it is just a story, that it is intended as a prop for games of make-believe rather than a report of actual events, that what it says is not true. (Walton 1990, 275)

So how exactly can fictionality be prominently declared or displayed? There can be several ways to achieve that, of course, to be found for instance in postmodern fiction, especially of the self-reflexive kind. Indeed, as noticed by MarieLaure Ryan, postmodern ‘high literature’ has become almost contemptuous of the so-called immersivity of popular literature, inherited from the nineteenthcentury novel. Today’s ‘high literature’, according to Ryan (2001, 160), “dares to

Metalepsis and Participation in Games of Make-Believe | 207

use such modes [that is modes encouraging participation and immersion], only under the protection of scare quotes and ironic distance”. Or, as Jean Ricardou stated already in 1967 about the French nouveau roman, a novel today is not the writing of an adventure any longer, but the adventure of writing.4 Postmodern writers thus prefer to exhibit the fictitiousness of their creations and to concentrate on the writing process. Most scholars tend to take the link between self-reflexivity and high art for granted, in the same way as they accept the link between immersion and popular culture today. Such dichotomies lead to different paradigms of values. Film theorist Robert Stam (1992, 129) has for instance noted how “the mainstream of film criticism, both academic and journalistic, subtended by the assumption that truly serious films must be verisimilar representations concerning plausible characters set in believable social contexts”. In the eyes of such critics, Stam (1992, 129) continues, “[r]eflexive films, in this sense, form an object of paranoia [since they] deprive the cinematic game of its illusion”. But what appears clearly is that Stam himself, in spite of his pretense to bring together reflexivity and illusion, represents the other paradigm of criticism, having a rather negative view of so-called illusionistic films. This happens when he criticizes Stanley Kubrick for not having succeeded to transfer Nabokov’s reflexive style in the adaptation of Lolita to the silver screen: The mainstream fiction film’s relative impermeability to reflexivity also explains the partial failure of Stanley Kubricks adaptation of Lolita (1962). While the novel constantly flaunts its own status as linguistic artefact, the film is largely cast in the illusionistic mould, presenting rounded characters in plausible settings through a self-effacing style. (Stam 1992, 160)

Surprisingly enough, Stam had earlier highlighted that mainstream productions such as television commercials often make use of reflexivity, but here he (Stam 1992, 16) discards this tendency by stating that it is “calculated to mystify rather than disenchant”. In other words, according to Stam, when used in popular fiction, reflexivity has the same effect as immersion: The exception only confirms the rule. Some scholars seem to nurture a certain negative attitude towards participation and immersion. Immersion, claims Valentine Cunningham, is viewed with great suspicion by many literary theorists, who rather prefer a detached, reflexive form of reading. According to these scholars, Cunningham (2002, 61) || 4 In Ricardou’s (1967, 111) own words in French: “Le récit n’est plus l’écriture d’une aventure, mais l’aventure d’une écriture.”

208 | Liviu Lutas points out, “[t]he magic that texts want to work on you is highly suspect. Their music is the dangerous Song of the Sirens. These delusive spells must be broken”. Marie-Laure Ryan (2001, 10) points towards one reason why theorists disregard immersion: “The major objection against immersion is the alleged incompatibility of the experience with the exercise of critical faculties.” In other words “immersion promotes a passive attitude in the reader, similar to the entrapment of tourists in the enclosed virtual realities of theme parks or vacation resorts” (Ryan 2001, 11), a phenomenon that Ryan (2001, 10) describes as a “semiotic blindness”. The low-esteem in which participation is held is also emphasized by Jay David Bolter (2001, 155): “Losing oneself in the fictional world is the goal of the naive reader or one who reads as entertainment. It is particularly a feature of genre fiction, such as romance or science fiction.” Certainly, from an ideological point of view, one reason for the negative attitude towards the immersive reception of works of art is that losing the critical faculties opens up for the risks of indoctrination. This is alluded to in the following quotation from Stephen Greenblatt (2001, 82), whose term ‘absorbing’, besides its obvious closeness to immersion and participation, can be seen as referring to a possible instrument for conveying an ideology: “The story, however pleasurable or absorbing, is shoring up ideological propositions or confirming the legitimacy of institutional arrangements.” As I mentioned above, certain scholars find that immersion can be stopped, or at least made difficult, by the use of certain techniques which exhibit the fictitiousness of the world in a work of art. Let us put this in Marie-Laure Ryan’s (2010, 14) words: “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.” Among such techniques, there is undoubtedly the narrative metalepsis. Indeed, as I already alluded to in the introduction, a great number of literary theorists have emphasized the so-called anti-illusionist effect of this narrative device. According to Frank Wagner or Debra Malina, whom I quoted above, it is the essence of all metalepses to break the mimetic illusion. Other scholars, such as John Pier or Jean-Marie Schaeffer (2005, 12–13), consider that most metalepses are anti-illusionistic, but admit that some metalepses could have the opposite effect. This is also the opinion of Sonja Klimek (2009), Werner Wolf (2005) and Monika Fludernik (2005, 77), who claim that some metalepses can actually reinforce realistic effects. As a matter of fact, there are scholars who would even argue that metalepses epitomize all ‘fictional immersion’, since they put on stage one of its principal characteristics: the duality of the mental state during the act of immersion. Especially Jean-Marie Schaeffer (2005, 325) and Jean Bessière (2005) have highlighted this paradoxical aspect of metalepsis, mentioned by other scholars in relation to all reflexive techniques. For instance,

Metalepsis and Participation in Games of Make-Believe | 209

film theorist Richard Allen (1995, 4) claims that this duality of the mental state is what characterizes all film experience – indeed all media experience. He (Allen 1995, 3) coins the term ‘projective illusions’ in order to be able to account for the duality inherent in the fact that “the film spectator knows it is only a film and actively participates in the experience of illusion that the cinema affords”. A similar duality characterizes all experience of fiction according to Peter Lamarque (1996, 14), who coins the dichotomy ‘internal perspective’ – corresponding to “imaginative involvement” – versus ‘external perspective’ – or “awareness of artifice”. “They need not be incompatible”, Lamarque (1996, 14) argues, “for there is no reason why they cannot be held simultaneously”. Against this, Marie-Laure Ryan (2001, 284) claims that such duality cannot occur simultaneously: “Literary texts can thus be either self-reflexive or immersive, or they can alternate between these two stances through a game of in and out […] but they cannot offer both experiences at the same time.” If I were to position myself in relation to the opinions above, I would claim that narrative metalepsis can have an anti-illusionist and thus anti-immersive effect. But I do not think that narrative metalepsis can be tied so simply to antiillusionist effects only. Indeed, self-reflexive devices in general should not automatically be linked to an anti-immersive effect. However, I do not follow Jean Marc Limoges’s (2009) or Sonja Klimek’s (2009) arguments when they make lists over which types of metalepses create illusionist effects, and which create anti-illusionist effects. I rather think that it is the individual context in which the metalepsis appears that creates one effect or the other. The same type of metalepsis could create an illusionist effect in one work and an anti-illusionist effect in another, as indeed the same type of metalepsis could be ‘funny’ (“bouffonne” in Genette’s terminology (1972, 244)) in one case, and ‘eerie’ (“fantastique” according to Genette (1972, 244)) in another. That is why the recipient – and of course the critic – must consider every individual occurrence of metalepsis separately and go through an interpretation process that would contribute to solving these problems. After a long period of time during which metalepsis was usually automatically linked to anti-illusionist effects and reflexive works of art, more recent work has brought to the surface a number of (empirical) examples that might show that metalepsis possibly allow readers’ immersion in fictional worlds. I will continue in the same spirit, and present some other such examples from three different media.

210 | Liviu Lutas

2 Metalepsis in paintings – trompe l’œil or clin d’œil I will start by analysing metalepsis and its effect in relation to make-believe in some paintings. This might seem surprising, since the original definition of metalepsis is connected to literature. Still, painting gives us one of the best examples of illusionistic representations: the so-called technique of the trompe l’œil. The trompe l’œil, which could be translated as ‘deceiving the eye’, is a kind of painting which gives the viewer the illusion that the depicted objects or persons are real. According to Richard Allen (1995, 82), “[i]n a trompe l’œil painting I see the painting as if it were an object and not a pictorial representation at all; the trompe l’œil thus entails a loss of medium awareness”. This leads Allen to the conclusion that this technique is only based on illusion, consequently losing what is required for it to be a representation or an artifact; It would not even correspond to the Waltonian make-believe concept, since the recipient actually believes that what he or she is seeing is the empiric reality. However, this view of the trompe l’œil is not accepted by all scholars. The French theorist Daniel Bergez insists that the complexity of this technique cannot be fully appreciated until the beholder becomes aware of the illusion he or she is subjected to. The deceived eye, Bergez (2009 [2004], 43) concludes, [enjoys simultaneously its illusion, the recognition of the illusion, the admiration of the tour de force of the artist, and the rivalry between art and reality (my translation; my emphasis)].5 In other words, since it brings together illusion and knowledge about the illusion, trompe l’œil is for Bergez the epitome of appreciation, especially according to Walton's earlier definition of the concept.6

|| 5 In original: “’l’œil’ ’trompé’ jouit simultanément de son illusion, de la reconnaissance de celle-ci, de l’admiration pour le tour de force de l’artiste, et de la séduction d’une rivalité entre l’art et la réalité” (Bergez, 2009 [2004], 43). 6 It is interesting to notice that in an article published 12 years before Mimesis as MakeBelieve, Walton (1978, 21) defined the concept of appreciation somewhat differently, allowing the simultaneity of the inside and outside perspective: “We, as it were, see Tom Sawyer both from inside his world and from outside of it. And we do so simultaneously […] The dual standpoint which appreciators take is […] one of the most fundamental and important features of the human institution of fiction” (Original emphasis).

Metalepsis and Participation in Games of Make-Believe | 211

Fig 1: Freskenzyklus in der Camera degli Sposi im Palazzo Ducale in Mantua, Szene: Gewölbefresko, Detail, Andrea Mantegna [Cycle of frescoes in the Camera degli Sposi in Palazzo Ducale in Mantua; scene: vault fresco, detail, Andrea Mantegna]. Photo: The Yorck Project: 10.000 Meisterwerke der Malerei. DVD-ROM, 2002, ISBN 3936122202. DIRECTMEDIA Publishing GmbH

It seems to me rather difficult to decide whether illusion and awareness about illusion occur simultaneously or not in the case of trompe l’œil. What is sure, however, is that there is a certain genre of trompe l’œil paintings in which the use of metaleptic breaches of the frame could be seen as a way to reveal the illusion. I am thinking of paintings, often frescoes, from which characters or objects seem to step out into the real world thanks to painted frames that are being crossed. This genre saw the day of light quite soon after the discovery of the central perspective in painting, and reached a very high degree of artistry and perfection already in works like Mantegna’s frescoes, painted between 1465 and 1474 in Camera degli Sposi in Mantua (see Schneider Adams 2001, 262–267). Especially the di sotto in sù ceiling of this room (see Fig. 1 above) is a marvel of illusionism which gives the beholder the impression that the ceiling opens into the sky, and thus towards God. The illusion is created for instance by using foreshortened angels scattered around a balustrade. But the balustrade also works as a frame, and the fact that three of the angels are on the wrong side of it gives the beholder the impression

212 | Liviu Lutas that they are about to fall down into our world. This is a case of metaleptic transgression of the frontier between the represented world and the beholder’s world. The question is if this metalepsis works as a so called clin d’œil, that is, a blink of the eye to the beholder in order to point out to him or her that this is just a painting. If this were the case, the illusion of the trompe l’œil would be broken. But does this really happen? In an analysis of a similar painting from the baroque period, Giovan Battista Gaulli’s ceiling fresco from Chiesa del Gesú in Rome (1674–1679), Sonja Klimek argues convincingly against the interpretation of such metaleptic stepping out of the frames as undermining aesthetic illusion. “Rather than create a sense of fictionality”, Klimek (2009, 180) says, “the aim of the trompe l’œil is to suggest to the observer that the gate to heaven is wide open and that the world of the observer has opened up to the world of God”. Even though one could argue for a possible change of perspective, and suggest that the world of God has opened up towards our world, it would be difficult to contradict Klimek’s conclusion that metalepsis in these cases rather contributes to the illusion than breaks it. Another example of trompe l’œil which could be illuminating in this context is the way in which the German photographer Thomas Struth represents people watching museum paintings in a series of 17 photographs presented in Hamburg 1993–1994.7 As noted by Laura M. Sager Eidt (2008, 34–35), “many of Struth’s photographs blur the boundaries between photograph and painting [...] by stressing the symmetry and correspondences between the paintings and the museum visitors”. The last photograph in the exhibition catalog, Galleria dell’Accademia I, Venedig [Venice] 1992 (see Fig. 2 below), is especially interesting in the context of this chapter, since the painting itself is a trompe l’œil. Indeed, Veronese’s banquet scene Feast in the House of Levi from 1573 achieves “not only a perfect illusion of depth, but also three dimensionality, to which the pillars and the stairs on both sides contribute” (Sager Eidt 2008, 36).

|| 7 See the exhibition catalog, i. e. Belting (1998).

Metalepsis and Participation in Games of Make-Believe | 213

Fig. 2: Galeria dell’Accademia I, Venice 1992, Thomas Struth, 1992, Photo: © Schirmer/Mosel, Munich and Tate, London 2013.

The photographer exploits the trompe l’œil on a new level of illusionism, giving the viewer of the photograph the impression that the world of the painting and the world of the museum visitors are the same. The camera perspective makes the people in the painting have the same size as the museum visitors, something that confuses the viewer of the photograph as to where the different persons belong. Thus, viewers could get the impression that some museum visitors have entered the world of the painting, while some painted characters have left the painting and stepped out into reality. Consequently, they might get the impression that the real world of the museum is an extension of the painting. Or is it the other way around? Is the painting an extension of the real world? What is certain is that the metaleptic technique of Struth’s photographs could have an illusionistic effect on some viewers, quite the opposite of the undermining of the mimetic effect which Debra Malina or Frank Wagner finds characteristic of all metalepses.

214 | Liviu Lutas

3 Metalepsis in literature – an entrance into the world of fiction The above-discussed technique of the trompe l’œil can finds its equivalent in literature in the works belonging to nineteenth-century realism. Indeed, this kind of literature is often considered illusionistic. Thus, it is the short story Gambara, published in 1836 by Honoré de Balzac, that Émile Benveniste (1966, 241) quotes in order to illustrate the perfect illusion of a story seeming to narrate itself: [Nobody speaks here ; the events seem to narrate themselves (my translation)].8 But as in the case of the trompe l’œil, the immersion into these illusionistic narratives occurs through a metaleptic device. As Ryan (2001, 125) notices, “[t]o dramatize the description, Balzac often resorts to the device of figuratively pulling the reader into the scene through a second-person address”. Admittedly, the metaleptical effect in this Balzacean technique is rather weak, of a rhetorical kind described by Monika Fludernik.9 Below, I will analyse two literary examples which in some ways can be considered as more elaborated versions of Balzac’s rhetorical metalepses. I have previously analysed these two works (Lutas, 2011), but from another perspective: as examples of metalepses in detective fiction. The first example is a short story written by Stephen King in 1993: Umney’s Last Case. There is a very salient case of metalepsis in this story: The author, a certain Samuel D. Landry, enters the world of a detective story he has written and forces the main character, who is actually his own creation, Clyde Umney, to switch places with him. The way in which the author enters the diegesis is explained in the story, but in quite an unconvincing way. Landy claims that he found Umney’s weak point and started to tear apart the whole diegetic world in order to make a place for himself in it. What seems to have happened is rather that Landry was literally absorbed by his own story. An argument for such an interpretation is that what Landry seeks in the novel is a form of escape from his miserable real life, like readers of so-called escapist literature, but in the literal sense of the word escape: “Here’s a world”, Landry says, “where I’ll never get any older, a year where all the clocks are stopped at just about eighteen months before World War II, where the newspapers always cost three cents, where I can eat all the eggs and red meat I want || 8 In original: “Personne ne parle ici; les événements semblent se raconter eux-mêmes” (Benveniste 1966, 241). 9 Fludernik’s (2005, 79–81) term is métalepse rhétorique.

Metalepsis and Participation in Games of Make-Believe | 215

and never have to worry about my cholesterol level. […] In this world, beloved sons never die of Aids and beloved wives never take overdoses of sleeping pills” (King 1993, 534). The example could be read as the physical realisation of the concept of participation according to Kendall Walton. One interesting issue, which will be discussed in relation with the film examples below as well, is that it is the author, not the reader, who is absorbed. But this is of minor importance for the point I am making; what matters is that through this switch of places between the author and the character he created, there occurs a violation of the boundary between the real world, that is, the world of the author, and the diegetic world, a world whose existence is ontologically different from the author’s reality. The other interesting issue is that both the author and the character are aware of their ontological statuses even after the switch. Oblivion of the medium and of the fictionality of the work seems thus not to be necessary in order to become immersed into the world of the book. On the contrary, Landry, the author, has a talk with his character in the diegetic world before he expels him from there into the harsh reality. The second example is the novel The Eyre Affair by Jasper Fforde. In this novel, those who enter the diegetic worlds are the readers, not the author, as happened in Umney’s Last Case. It is a non-typical detective story, where the main crime that has to be solved is the literal kidnapping of the character of Jane Eyre from the original manuscript of Charlotte Brontë’s novel. The detective who has to solve the crime, a 36-year-old woman by the name of Thursday Next, is chosen because of her abilities as a child to literally glide into the diegesis of the novel Jane Eyre. As I (Lutas 2011) have argued previously, metalepses of the kind that appear in The Eyre Affair can be interpreted as an allegory of the process of reading. Indeed, the way in which these intrusions occur are clear arguments in such a direction. One way of entering fictional worlds in The Eyre Affair is for instance by use of a special technical device called the prose portal. Another way is by getting literally absorbed by the novel when reading very attentively, as does a Japanese woman whom Thursday Next asks for advice in order to refresh her memory of how she did this as a child: “How do you manage it?” asks Thursday. “I just can – she answered simply – I think hard, speak the lines and, well, here I am” (Fforde 2001, 325). This case too, similarly to Landry in Umney’s Last Case, can be considered as the realisation of the literal meaning of the figure of

216 | Liviu Lutas speech ‘being absorbed by the story’.10 The Japanese woman has developed this capacity to literally enter the text, succeeding to control it. Something similar happened to Thursday as well, when she was a child, but in a way that she could not control. Besides, it was an event that she even starts doubting if it really happened, writing it off “as the product of an overactive imagination” (Fforde 2001, 69). Both literary examples above show rather clearly that narrative metalepsis does not necessarily block immersion. On the contrary, metalepsis here works as the literal realisation of the dream that many authors or the readers have: to enter the world of the book. The film examples in section four below are more complicated from this point of view, and consequently more ambiguous.

4 Metalepsis in film – breaking the illusion In this section, I will analyse five scenes from two films by the Greek director Giorgios Lanthimos: one scene from Alps and four from Dogtooth. In both these films, make-believe is not only an issue concerning the reception of the film by willing film viewers, but is thematized and dramatized in the films. Indeed, games of make-believe are the main subject of both of these films, which both try to diagnose how illusion works in real life. The games of make-believe are used by consenting players in the first film, as a means to achieve a cathartic reaction, but are imposed on some persons in the second film, as a means to create a total illusion. The differences between the two films can cast new light on the concept of make-believe, especially when it comes to the participation aspect. A narrative metalepsis occurs only in the fifth scene, but its effect in relation to make-believe is especially interesting to analyse. The first scene comes from Lanthimos’s film from 2011, Alps. In this scene, a couple hires a young girl who is supposed to act their daughter, who died in an accident some time before. Everything is set up as if it were a film, or rather a theatre play. The girl enters the house, goes to her room, and lets in her boyfriend in secret, without her parents knowing, as it were. The two lovers are surprised by the parents who rush in and see their daughter kissing her boy-

|| 10 It is interesting to note that Tzvetan Todorov had mentioned the realisation of the literal meaning of a figurative expression as typical of the genre of fantastic; “le surnaturel naît souvent de ce qu’on prend le sens figuré à la letter”, he claimed, concluding further that the fantastic “réalise alors le sens propre d'une expression figure” (Todorov 1970, 83–84).

Metalepsis and Participation in Games of Make-Believe | 217

friend – who is actually the dead girl’s real boyfriend. The parents react by getting really angry, and throwing out the young man. We as films viewers are first presented with the scene as a part of the cinematic diegesis, and do not become aware of the fact that it is a set-up situation until we suddenly see that the parents are standing in the dark, in the hallway, looking at the young couple already before they surprise them kissing, that is, when they are not supposed to be there. Our own mimetic illusion is thus broken when we realize that we have been deceived about the nature of the meeting taking place on the screen. However, this does not block our immersion. We are rather recentred, to use Ryan’s (2001) term, in a new coherent situation, trying to understand what the play inside the film is about (but even Wolf 2005, 103). But our situation as viewers is not completely the same as the parents’ situation. Admittedly, the parents watch the scene too, as we do, but under different premises. First of all, they staged the scene themselves, in an attempt at selfdelusion that would give them comfort in the process of grief. Secondly, they have the physical possibility to intervene in the scene and change its course, something that they actually do when they find the girl’s acting unsatisfying. Thus, it appears clearly that their illusion is less complete than ours, even though their goal is total immersion in a world where their daughter is not dead. The fact that we, as recipients, see the parents watching the scene is in some ways comparable to the photographs taken by Thomas Struth analysed above, where people watching paintings are depicted. The comparison is relevant since there is an essential difference between the two cases. If the public in Struth’s photographs, as we saw above, is put at the same level as the painting in a way reminiscent of a trompe l’œil, it is quite clear for us that the public, that is the parents, watching the scene in Alps is not at the same level as the diegetic world they try to create. Or rather, one should say, it is the staged scene that never really acquires a diegetic-level ontologically separated from the reality of the parents. As Lanthimos (La Porta 2011) himself stated in an interview, “Alps is about a person who tries to enter a fabricated world”. As we have seen above, however, this does not really succeed. Admittedly, the parents get some comfort in their grieving process by staging the scene, but they are never completely deluded, and the scene never acquires the effect of a trompe l’œil for them. The other example, Lanthimos’ earlier film Dogtooth, can be seen as the other side of the coin: As Lanthimos (La Porta 2011) expresses it himself, “it is the story of a person who tries to escape a fictitious world”. What is interesting is to see if this succeeds better than in Alps. I will analyse four short scenes from the film Dog-

218 | Liviu Lutas tooth. The first three are necessary as background to the analysis of the metalepsis occurring in the fourth scene. The film is about a couple who isolate their three children inside a country estate, telling them that what is outside the fence is dangerous, and not reality. The first scene shows how this is done. When one of the two sisters throws her brother’s toy airplane over the fence, the father opens the gate and takes the car to the place where the airplane had landed, in order to pick it up while still sitting in the car, and drive back without stepping on the ground outside the fence – as if that were an extremely dangerous thing to do. What counts from the point of view of immersion is that the isolation is not complete here either. As in Alps, the reality outside the fence exists, is integrated in the world inside the fence, and could be accessed if the children only started to question the rules made up by their parents. From the point of view of the parents, the situation is quite similar to Alps too. Here as well, it seems difficult to be a complete part of the fictional world, since the reality outside is not really expelled. The second and the third scenes from Dogtooth show how difficult it is for the parents, in other words for the creators of the fiction, to become a part of their created world. As in Alps, they never forget the artificiality of the world, since they are those who must construct it. In scene two, one of the girls sees an airplane in the sky, but her reaction shows that she is a victim of her parents’ indoctrination. Indeed, she thinks that the plane is a small toy of the same kind as her brother’s, and she says that she wants it to fall, so she can pick it up. The mother answers her that whoever deserves it will get it. The children interpret this as it were a competition, so in the third scene, when the parents throw a toy airplane on the lawn exactly when a real airplane is passing above them, making the children believe that the real airplane had fallen in their garden, they all start running to get it. Certainly, the airplane might be seen as a metaphor for freedom, but in the context of this film, it shows how illusion can work when creating a fictitious world. If we are to see the children as the recipients of the fiction created by their parents, it appears clearly that the airplane is the prop needed to make them believe that the sky is just a part of their small world. Except that it is not a game of make-believe here. The children actually believe that their world is like that. But the cost is rather high: They are prisoners in the small space delimited by the fence. Belief thus seems necessary in order to accomplish complete immersion. The last scene, the most ambiguous of the four, may be interpreted in such way that maybe even the parents have succeeded to be a part of the created world. However, another interpretation could be quite the opposite: that one of the children starts understanding that there is an exit out of the illusion. In this scene, the whole family is sitting on a sofa in the living-room watching a video

Metalepsis and Participation in Games of Make-Believe | 219

film of themselves having a picnic on the lawn. The first strange thing that happens is that when one of the sisters on the sofa is shown by the camera moving her lips as if she were talking, the sounds she produces are those from the video film they are watching. Admittedly, there need not be anything metaleptic about this, since the girl might only mimic the words which she already knows had been uttered. The following clip, however, reinforces the metaleptic effect. Indeed, the camera turns from the sofa to the TV-set, and shows us the family sitting on the lawn. But when one of the sisters in the TV scene moves her lips as if speaking, the words are incongruous with the situation. They seem to come rather from the people sitting on the sofa watching, both because of what is said and because of how it is said. It is as if a metaleptic breach has occurred between the world of the TV viewers, their reality, and the world beyond the TV screen. If we consider this scene a metalepsis, one consequence could be that the world created by the parents inside the fence really is ontologically isolated. The TV-set becomes a boundary, or an interface, that isolates the estate much better than the fence does. Thus, even the parents would have succeeded in becoming a part of the world of their fiction. The metalepsis would in that case be similar to the metalepsis in Umney’s Last Case, that is, a door into the fiction, and an escape from reality. The parents’ make-believe would in such a case be transformed into belief. However, from the children’s point of view, the metalepsis could be seen as a signal that there actually exists a possibility to break the boundary, and to leave the world where they are ontologically imprisoned. This interpretation seems more adequate when considering the film’s structure. Indeed, the elder of the two sisters starts questioning her state, and eventually makes an attempt to escape. The ambiguity of the film’s ending does not allow the viewer to conclude whether this attempt succeeds or not, but that is of minor importance. The fact that the girl dares cross the boundary, at the risk of losing her life according to her episteme, shows that the illusion has started to break. In such a case, the metalepsis would rather be a door out from the fiction, quite the opposite of the other interpretation.

5 Conclusion The fourth scene of the film Dogtooth is a good example of how difficult it is to give a final answer as to the effect of metalepsis when it comes to participation in games of make-believe. Narrative metalepsis is obviously a complex device,

220 | Liviu Lutas with ambiguous effects, which could both hinder and enhance participation. An important point is that self-reflexivity, which can be seen as characteristic of metalepsis, does not necessarily block the experience of being caught up in the story, as Kendall Walton has claimed. This conclusion is valid for the other cases I studied in this essay too. I have shown for instance that metalepsis can contribute to the illusionistic effect of some trompe l’œil paintings, and that it can permit an author or a character to enter a work fiction in literature in a way that can be seen as the realization of the proper sense of the metaphor ‘absorbing the reader’. The fact that the examples studied above were taken from three different media shows that the concept of metalepsis should not be restricted to literature and to the narrator’s voice. As Jean-Marie Schaeffer has claimed, metalepsis rather plays with the levels of representation. This is why the concept of metalepsis is compatible with Kendall Walton’s theory of make-believe, a theory in which the representational aspect is of capital importance. Considering narrative metalepsis in different media is obviously beneficial for the concept of metalepsis itself, but maybe even more beneficial for the understanding of some artifacts which are not usually analysed with the help of narratological tools.

References Allen, Richard. 1995. Projecting Illusion. Film Spectatorship and the Impression of Reality. Cambridge: Cambridge University Press, 1995. Belting, Hans. “Photography and Painting: Thomas Struth’s Museum Photographs.” Museum Photographs. Ed. Hans Belting. München: Schirmer/Mosel, 1998, 5–28. Benveniste, Émile. Problèmes de linguistique générale. Paris, Gallimard, coll. “Tel,” 1966. Bergez, Daniel. Littérature et peinture. Paris: Armand Colin, 2009 [2004]. Bessière, Jean. “Récit de fiction, transition discursive, présentation actuelle du passé, ou que le récit de fiction et toujours métaleptique.” Métalepses. Entorses au pacte de la représentation. Eds. John Pier and Jean-Marie Schaeffer. Paris: Éd. de l’EHESS, 2005, 279–294. Bolter, Jay David. Writing Space: Computers, Hypertext, and the Remediation of Print. Mahwah: Lawrence Erlbaum Associates, 2001. Cunningham, Valentine. Reading after Theory. Oxford: Blackwell Publishers, 2002. Fforde, Jasper. The Eyre Affair. London: Hodder and Stoughton, 2001. Fludernik, Monika. “Changement de scène et mode métaleptique.” Métalepses. Entorses au pacte de la représentation. Eds. John Pier and Jean-Marie Schaeffer. Paris: Éd. de l’EHESS, 2005, 73–94. Genette, Gérard. Figures III. Paris, Éditions du Seuil, coll. “ Poétique”, 1972. (Translated into English as: Narrative Discourse: An Essay in Method. Ithaca: Cornell UP, 1980). Genette, Gérard. Narrative Discourse: An Essay in Method. Ithaca: Cornell UP, 1980.

Metalepsis and Participation in Games of Make-Believe | 221 Genette, Gérard. Nouveau discours du récit. Paris, Éditions du Seuil, coll. “Poétique”, 1983. (Translated into English as: Narrative Discourse Revisited. Ithaca: Cornell UP, 1988). Genette, Gérard. Narrative Discourse Revisited. Ithaca: Cornell UP, 1988 Greenblatt, Stephen and Catherine Gallagher. Practicing New Historicism. Chicago: University of Chicago Press, 2001. King, Stephen. “Umney’s Last Case.” Nightmares and Dreamscapes. London: Hodder and Stoughton, 1993, 501–540. Klimek, Sonja. “Metalepsis and Its (Anti-)Illusionist Effects in the Arts, Media and Role-Playing Games.” Metareference across Media. Ed. Werner Wolf. Amsterdam/New York: Rodopi, 2009, 169–187. Kukkonen, Karin. “Metalepsis in Popular Culture: An Introduction.” Metalepsis in Popular Culture. Eds. Karin Kukkonen and Sonja Klimek. Berlin/New York: De Gruyter, 2011, 1–21. Lamarque, Peter. Fictional Points of View. Ithaca and London: Cornell University Press, 1996. La Porta, Domenico. “There is No New Greek Cinema.” Cineuropa 3 September 2011. http://www.cineuropa.org/it.aspx?t=interview&l=en&did=209034 (22 June 2015). Limoges, Jean-Marc. “The Gradable Effects of Self-Reflexivity: On Aesthetic Illusion in Cinema.” Metareference across Media. Ed. Werner Wolf. Amsterdam/New York: Rodopi, 2009, 391– 407. Lutas, Liviu. “Narrative Metalepsis in Detective Fiction.” Metalepsis in Popular Culture. Eds. Karin Kukkonen and Sonja Klimek. Berlin/New York: De Gruyter, 2011, 41–64. Malina, Debra. Breaking the Frame: Metalepsis and the Construction of the Subject. Ohio: Ohio State University Press, 2002. Pier, John and Jean-Marie Schaeffer. “La métalepse aujourd’hui: Introduction.” Métalepses. Entorses au pacte de la représentation. Eds. John Pier and Jean-Marie Schaeffer. Paris: Éd. de l’EHESS, 2005, 7–15. Ricardou, Jean. Problèmes du nouveau roman. Paris: Seuil, 1967. Ryan, Marie-Laure. Narrative as Virtual Reality. Immersion and Interactivity in Literature and Electronic Media. Maryland: John Hopkins University Press, 2001. Ryan, Marie-Laure. “Fiction, Cognition, and Non-Verbal Media.” Intermediality and Storytelling. Eds. Marina Grishakova and Marie-Laure Ryan. Berlin/New York: De Gruyter, 2010: 8–26. Sager Eidt, Laura M. Writing and Filming the Painting. Ekphrasis in Literature and Film. Amsterdam, New York: Rodopi, 2008. Schaeffer, Jean-Marie. “Métalepse et immersion fictionnelle.” Métalepses. Entorses au pacte de la représentation. Eds. John Pier and Jean-Marie Schaeffer. Paris: Éd. de l’EHESS, 2005, 323–334. Schneider Adams, Laurie. Italian Renaissance Art. Boulder: Westview Press, 2001. Stam, Robert. Reflexivity in Film and Literature: From Don Quixote to Jean-Luc Godard. New York: Columbia University Press, 1992 [1985]. Todorov, Tzvetan. Introduction à la littérature fantastique. Paris: Seuil, 1970. Wagner, Frank. “Glissements et déphasages, Notes sur la métalepse narrative.” Poétique 130. avril, Paris: Éditions du Seuil, 2002, 235–253. Walton, Kendall. “How Remote are Fictional Worlds from the Real World?” Journal of Aesthetics and Art Criticism 37 (1978): 11–23. Walton, Kendall. Mimesis as Make-Believe. On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990.

222 | Liviu Lutas Wolf, Werner. “Metalepsis as a Transgeneric and Transmedial Phenomenon: A Case Study of the Possibilities of ‘Exporting’ Narratological Concepts.” Narratology Beyond Literary Criticism. Ed. Jan Christoph Meister. Berlin/New York: De Gruyter, 2005, 83–107.

Sonja Klimek

‘I grieve’ as Make-Believe Generating Fictional Truth in Eighteenth-Century Lamentation Poetry (Günther, Haller, Young and Novalis) Poetry […] is a strictly solitary affair; solipsist even. I ought not to talk to you about it. But… well… (Pascal Mercier: Lea)

1 Problems in modern lyricology During the last few decades, research in fictionality and the generation of fictional truth has mainly centred on literature, especially fictional narrative texts, like novels or novellas (see Zipfel 2011, 162). The present interdisciplinary volume attempts to broaden this focus and to study how these principles work in other art forms and media. Within this framework, my paper treats the question of how the principles of generation of fictional truth are at work in texts for which we often cannot be sure, in advance, whether they are fictional or not – that is, in lyric poetry. A long-lasting criticism of the study of lyric poetry is that it is undertheorized and dominated by intuitive interpretation (see e. g. Müller-Zettelmann 2000, 1). In comparison with the prom-inent progress of terminological clarity and methodological approaches of contemporary narratology, for instance, this judgement seems altogether justified. Nonetheless, during the last three decades, great efforts have been made to fight the lack of theoretical reflections in lyricology.1 Highly differentiated studies have been published in German. Interestingly, they do not seem to have been much noticed in the Anglophone world, where still today, scholars largely adhere to the view that all lyric poetry must be, by definition, fictional (see Zymner 2009, 10), a conviction that most likely goes back to Beardsley’s (1981) paper on “Fiction as Representation”.2 In newer || 1 For the term itself and a short review of some of the most important newer works in this discipline, see Zymner (2009, 7). 2 From a speech act theoretical point of view, Beardsley suggests to dissociate author and speaker of a poem and thus to interpret the poem as being placed by its author in a ‘makebelieve context’ (see Zipfel 2001, 301).

224 | Sonja Klimek times, Müller-Zettelmann (2002, 142) assumes that the communication system of a poem always contains an ‘intratextual persona’ that is necessarily a ‘fictional product’.3 The present paper assumes that the basic understanding of lyric poetry as being always fictional is too strict. As Zymner (2009, 14–15) has argued, there is no obvious reason why a poet should not be able to use words that refer to some circumstances in extra-textual reality or to refer to himself by the linguistic sign ‘I’ in a poem – just because the speech he produces is structured in verse. To assume that this is impossible by definition in lyric poetry would not only prescribe a certain kind of reading of each and every poem, but also lead to implicit exclusions of many texts from the genre of the lyric, such as for example some parts of occasional poetry and the whole spectrum of committed poetry: Obviously, activist poets or pious chant-writers as well as any lover presenting an original poem to his or her beloved would probably affirm that their lyric production is nonfictional. As Lamping (2000, 127) states, for ‘referential political lyric poetry’, there is even something like an ‘obligation to truth’ for the poet. In fact, the discussion of whether the first-person speaker of a poem is to be regarded as the author herself or himself or as a purely fictive entity can (with regard to some texts that would be called lyric poetry today) be traced back to antiquity. Plato’s claim that dithyrambic poetry is the direct speech of the poet – in contrast to the mimetic representation of speeches from different personae in drama (where the poet talks in disguise) and to the mixture of both manners in epic texts (see Lamping 2000, 117–119) – laid the foundation for an understanding of poetry as the authentic expression of a real person.4 During the Middle Ages, antique Greek poetics were mostly lost in European contexts, so there were completely different concepts of fictionality and alterity at work in the semi-oral and performative cultures of literature and poetry.5 But beginning with the Italian Renaissance, texts we would nowadays call lyric poetry were revaluated, both in production and in theory, and became the third genre of literature, next to, and of equal value to, Epics and Dramatics.6 Baroque poetry, || 3 Müller-Zettelmann (2002, 143–144 and 148–149) understands subjectivity in lyric poetry as the result of a special way of using language in order to generate “the illusion of a (speaking, writing or thinking) I-origo”, a model that interestingly includes the “possibility of a complete lack of attributes suggesting subjectivity or authenticity”. 4 Of course, the texts Plato had in mind are not identical with the great variety of poems we nowadays summarise under the term of lyric poetry (see Lamping 2000, 119). 5 For more details, see the various papers in Bleumer and Emmelius, eds. (2011). 6 This trias first occurred in Giovanni Giorgio Trissino’s Poetica (1529/1563) and was renewed by Charles Batteux in Les beaux-arts réduits à un même principe (1746) (see Burdorf 1997, 3, and

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 225

with recourse to antique forms and poetics, followed the rules of imitatio and aemulatio; poetic speech acts were highly standardised and organised according to the art of rhetoric. In German poetry, for example, Plato’s idea of the Lyric as the authentic speech of a poet was not readopted until the era of the so-called Erlebnisdichtung and Genieästhetik of Herder, Klopstock and Goethe, who, in their turn, were influenced by English discourse on the ‘original genius’. This discourse was established in the Conjectures on Original Composition (1759) by the English poet and theoretician Edward Young and transmitted through the comments of the German translation of Batteux’s poetics by Johann Adolf Schlegel who – in opposition to Batteux – stated that lyric poetry was always the presentation of authentic feelings (see Zipfel 2011, 163). Since then, and for more than an entire century, lyric poetry in England as well as in Germany was understood to be the authentic and ingenious expression of a gifted individual, a ‘poet’ in the most emphatic sense of the word. This dictum of the authenticity of poetry was destroyed in connection with the rise of modernism by authors like Charles Baudelaire in France and Stefan George in Germany, who wrote highly complex texts that foregrounded their artificiality. Nonetheless, in the Germanophone aesthetics of the nineteenth through at least the middle of the twentieth century, the dictum of the aesthetics of the original genius remained common-sense: Theoreticians (e. g., Hegel, Dilthey, Staiger and Hamburger7) believed that the Lyric was – in contrast to the Epic and the Dramatic – a nonfictional genre (see Zipfel 2011, 64, and Lamping 2000, 102 and 114–117). Today, most Germanophone lyricologists would claim that lyric poetry can be fictional as well as nonfictional. For example, in his well-known compendium to the study of lyric poetry Burdorf (1997, 169–170) states that “the lyric is able to create alternative worlds, independently from its degree of fictionality”. For him, lyric poetry, in contrast to the Dramatic and the Epic, is the only genre in which fictionality is not a central category. Here, Burdorf refers to Anderegg (1985, 123–124) who first uttered that the question of fictionality should not stand at the centre of lyric theory because for him, lyric poetry seems to be “in a strange way indifferent to the category of fictionality”. To this, Zipfel (2001, 34) disagrees, specifying that the ‘indifference’ of lyric texts to questions of fictionality means that poems as presentations of facts with a particular linguistic form || Burdorf 2000, 502–503) for whom lyric poetry was “not imitation of actions, but of feelings” and therefore fictional (see Zipfel 2011, 163). 7 Käte Hamburger (1977) presumes that lyric poetry can never be fictional because for her, the speaker of a poem is always identical with the poet and that he is always referring to a ‘real’ experience, be it only an imagined inner one. This is contrary to Beardsley’s position who later postulated that each speaker of each poem must be different from the author (see above).

226 | Sonja Klimek – as any other presentation of facts – can be either fictional or nonfictional.8 Agreeing with Zipfel’s position,9 Zymner (2009, 20) concludes that lyric poetry can be either fictional or not, that is, can either “talk about fictive things or about non-fictive ones”. The present paper is a plea to carefully raise the question of fictionality for each single poem, with regard to the historical context it stems from, and to the genre conventions of the time it was written. Notably, this perspective involves a focus on the production as well as on the critical reception of the text: What happens when readers use a specific piece of lyric poetry as a prop in their own game of make-believe (Walton 1990, 69) to generate their personal fictional truths of a game world (Walton 2011, 465), and what happens if they read the same text as nonfictional? In the following, I will study these questions with regard to some poems that stem from a period of time relatively shortly before and after the baroque convention of rhetoricity and aemulatio was replaced by the later convention of the primacy of authentic expression of an ingenious poet’s true feelings, that is, from the eighteenth century. Concerning their topic, I have chosen poems that belong to the lyrical sub-genre of lamentation poetry on the assumption that the exceptional psychological circumstances of a ‘person’ or ‘persona’ in mourning makes it improbable to modern readers to think that the feelings expressed in the text have been simulated. As soon as the reader of a lyric text finds an ‘I’, saying something like ‘I grieve’, he or she can identify the ‘I’ with the author, or decide to read the ‘I’ as an invented character. || 8 German Original (Zipfel 2001, 34): “Die ‘Indifferenz’ lyrischer Texte gegenüber der Fiktion bedeutet vielmehr, daß Gedichte als Sachverhaltsdarstellungen, die durch eine besondere sprachliche Form ausgezeichnet sind, wie alle anderen Sachverhaltsdarstellungen entweder fiktional oder nicht-fiktional sein können.” 9 See Zipfel (2011, 163): “Daraus folgt, dass einerseits fiktionale Texte in Verse gegliedert sein können – oder auch nicht; und dass andererseits das in Lyrik Dargestellte fiktiv und/oder die Darstellungsweise in Gedichten fiktional sein kann – oder auch nicht.” [From this it follows that, on the one hand, fictional texts can be written in verse – or not; on the other hand, the things depicted by lyric poetry can be fictive and/or the mode of representation in poems can be fictional – or not. (My translation)] And Zipfel (2011, 165–166): “Während bei der Produktion und Rezeption von narrativen Texten die Zuordnungen zur institutionellen Praxis Fiktion oder Nicht-Fiktion eine wichtige Rolle spielt und in der Regel schon durch den Paratext eindeutig festgelegt wird […], bleiben viele lyrische Gedichte im Hinblick auf die Frage nach ihrer Fiktionalität intentional unterbestimmt, sodass dem Rezipienten in dieser Hinsicht ein größerer Interpretationsspielraum eingeräumt wird.” [Whilst the production and reception of narrative texts, the attribution to the institutional practices of fiction or nonfiction plays an important role and usually is clearly determined by the paratext […], concerning lyric poetry, many poems stay intentionally under-determined with regard to their fictionality, so that the recipient has a larger range for his interpretation (my translation).]

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 227

2 Fictionality and the generation of fictional truth in lyric poetry: attempts at categorisation Based on the differentiation between the Rede-Status [State of the speech] and the Rede-Funktion [Function of language] Lamping (2000, 103–110) has developed a scheme in which he establishes four forms of fictionality of a poem: At the core, there is semantic fictionality, which means that a text presents fictional propositions (i. e., talks about fictive things), and pragmatic fictionality, which means that the situation of the speech is fabled and the speech act therefore is fictional. Because these two modes of fictionality exist independently from each other, their combination leads Lamping (2000, 108) to a list of four possible categories: Lyrical poems that are “fictional, but not fictive”, “fictive, but not fictional”, “fictive as well as fictional” and those which are “neither fictive nor fictional”. As Lamping admits, the problem with these terms is that they deviate from the common use of language: At present scholars widely agree that a text that exists in our reality is either fictional or not fictional, but never ‘fictive’. ‘Fictive’ is an attribute of things, while ‘fictional’ is an attribute of “texts and stories and so on” (Gabriel 1991, 36; see Lamping 2000, 14). Zymner presents a similar categorisation of fictionality in lyric poetry. Like Lamping, Zymner (2009, 11–12) uses the distinction between semantics and pragmatics of fictionality to distinguish between ‘fictionality’ and ‘factuality’, and to combine this differentiation with the two criteria ‘authenticity of the lyric subject (as the author or the fictive persona?)’ and ‘authenticity of the proposition (fictional or factual speech?)’. This leads him to the following four categories: 1.

Autorfiktionale Lyrik [author-fictional lyric poetry] (the author himself speaks, he talks about invented or fabled things); 2. Personafiktionale Lyrik [persona-fictional lyric poetry] (the author invents a fictive ‘persona’ who then speaks about invented things); 3. Autorfaktuale Lyrik [author-factual lyric poetry] (the author speaks as himself and tells about true things or facts from reality); 4. Personafaktuale Lyrik [persona-factual lyric poetry] (the author invents a ‘persona’ and lets her talk about true things or facts from reality). One problem I see in applying these four categories to the study of lyric poetry is that in some cases, the reader cannot be sure whether the things told are fictive or not, which makes it difficult to distinguish whether the poem is ‘fictional’ or

228 | Sonja Klimek ‘factual’. Moreover, the term ‘factual’ is highly contested for a literary text: There are alternatives to fictionality and factuality, as, for instance, lies and fallacies. For this reason, both Zipfel and I use Lamping’s ‘nonfictional’ instead of ‘factual’, as Zymner does. Another problem for the application of Zymner’s four categories to concrete text analysis is that for many poems, the decision whether to understand them as speech produced by the author or by a persona is already part of their interpretation. In their explication of the German term Lyrisches Ich [‘Lyrical Self’ or ‘Lyrical I’, that is, the persona or voice of the poem], Fricke and Stocker (2000, 509–511) argue that many pieces of lyric poetry (leaving aside role poems and ballads) are characterised by a kind of ‘drifting fictionality’ that results in a state between fictionality and nonfictionality, with the Lyrical I as an “Instanz, […] deren personale Identität und fragliche Fiktionalität […] gleichsam in der Schwebe bleiben” [entity, whose personal identity and moot fictionality remain quasi in abeyance (My translation)]. Thus, the ‘Lyrical I’ explicitly includes the option for the reader to interpret this ‘I’ in the one or in the other way, as the author or as a persona.10 Ryan’s (1980, 43) notion of the so-called principle of minimal departure seems helpful when analysing the reception of lyric poetry: “This principle states that whenever we interpret a message concerning an alternate world, we reconstrue this world as being the closest possible to the reality we know.” Ryan (1980, 43) continues that, in what she calls “natural discourse”, “the referents of the pronouns ‘I’ and ‘you’ are reconstrued as retaining the personality of the actual speaker as fully as possible, but in fiction they [the pronouns ‘I’ and ‘you’] are immune to the principle of minimal departure”. Applied to lyric poetry, this means that as long as a poem does not contain any explicit signal of fictivity, concrete readers can understand the text as nonfictional, spreading the application of Ryan’s principal of minimal departure also to the pronoun ‘I’ in the text, thus identifying the speaker of the poem with the author – or they can decide to read it as a fictional text with a fictive speaker.

|| 10 Of course, there are genres where we can be absolutely sure that the ‘I’ in the text is not the author, for example in role poems with a female speaker and a male author.

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 229

3 Eighteenth-century lamentation poetry 3.1 The Epicedium – Johann Christian Günther’s late-baroque occasional poetry In the baroque epicedium, the poetic ‘lamentation of the dead’ is not a free invention, but usually refers to a concrete situation. Therefore, the epicedium belongs to the category of Occasional Poetry. The death of a real person is bemoaned in a rhetorically highly regularised manner. The historical context of these texts is clear: Someone has died and another person, a poet, writes an epicedium, a lamentation poem for the funeral, either because he is in mourning himself, or because a friend or a relation of the dead pays him for this service. In both cases, the poet keeps a copy of the text for the later collection of his own poetic works. When we take, for example, the poems of Johann Christian Günther, a famous late baroque German poet at the beginning of the eighteenth century, we find an abundance of ordinary panegyric lamentation poems such as, for instance, the Latin epicedium about the widow of the German emperor Leopold I, Eleonora Magdalena, through which Günther (1998, 172) supposedly wished to introduce himself to the Court in Vienna (see Bölhoff 1998, 1014). Panegyric lamentation poems are a genre known to all readers of baroque collections of poetry. We can assume that the genre convention of the panegyric lamentation poem clearly signalised that the ‘I’ (or the ‘we’) hints to the author himself (or to a collective of which the author is part), even if his words are not the personal expression of his own feelings, but the rhetorically highly regularised enumeration of standardised topoi which are expected to be addressed at the funeral of a potentate. So would it be just to apply Zymner’s third category of ‘author-factual lyric’? In Günther’s lamentation poetry, we also often find subtitles like Jm Nahmen eines andern [In the name of someone else (my translation)], as, for example, in the poem about a noble widow called Anna Helena von Zedlitz.11 This subtitle serves as a genre-signal. It alerts the reader to apply the genre convention of remittance lamentation poetry to the text, which means that the || 11 The title that sounds typically baroque in its full length reads (in German): “Die auch in Jhrer Asche / Als ein Muster Des preißwürdigen Frauenzimmers verehrte / Frau Anna Helena, verw von Zedlitz etc. / Welche / A 1713. den 27. Jan in dem HErrn entschlieff. / Jm Nahmen eines andern” (Günther 1998, 127)

230 | Sonja Klimek text that follows this subtitle does not give the poet’s (that is, Günther’s) opinions. Günther, as a baroque poet laureate, does not deplore the passing of the lady of Zedlitz because he is himself in mourning, but because he is apt and well-educated enough to pronounce the supposed grief of the deceased’s relatives or friends in a suitable, dignified and even supra-individual or universal way, and therefore the poet is being paid.12 Thus, the genre convention generates a ‘truth’ that is not necessarily fictional, but ‘supposedly true’, despite the author speaking not in his own name, but in the name of another person for whom the pronounced feelings may be true. Such conventional remittance work is neither personal nor purely fictional. It is one of the examples to which it is difficult to apply one of Lamping’s and Zymner’s four categories: We cannot use the terms ‘author-fictional/nonfictional’ because the subtitle clearly states that the ‘I’ of the text is not to be identified with the real author Günther. Nor can we apply one of the categories of ‘persona-fictional’ or ‘persona-nonfictional’ because the ‘I’ is not a fictive ‘persona’, but a real person, that is, the person who ordered the poem. And it is also questionable whether the act of speaking is fictional or nonfictional at all because the real person who is supposed to be in mourning does not utter the propositions of the text, but only receives them as the poet has put them. In this tricky case, we make more progress if we leave aside Zymner’s and Lamping’s system and move to another attempt at explaining fictionality in lyric poetry. Walton draws an analogy between political speechwriters and poets: Speechwriters don’t use the words they inscribe (not insofar as they are simply speechwriters); they mention them. They produce a text for someone else. The speechwriter doesn’t assert the declarative sentence he comes up with; his client asserts them when she delivers the speech. (Walton 2011, 460–461)

Indeed, the analogy to baroque remittance lamentation poetry ‘in the name of another’ is very strong: “The client […] may appreciate and admire the speechwriter’s skill and pay him well for his services” (Walton 2011, 461). However, Walton is not concerned with baroque remittance works, but with poetry in general. He draws his own conclusion: A poet’s words might strike me as just the right way of expressing a thought I thought I had. Or, on reading a poem I might decide then and there to assert it to myself. Alternatively, I may think of the words as clarifying my thoughts, as well as providing a means of expressing them. So I will say that poets sometimes serve not exactly as speechwriters, but as thoughtwriters. (By ‘thoughts’ I include not just ‘intellectual’ ideas, but any feel|| 12 See in general Segebrecht (1977).

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 231 ings, emotions, sentiments, attitudes, etc., that might be expressed by means of words.) (Walton 2011, 462)

So Walton introduces an alternative to Zymner and Lamping’s binary distinction between ‘author’ and ‘persona’ and to the theoretical concept of an ‘abstract speaker’ (see Burdorf 1997, 195) or an ‘implied lyric narrator’ (see Hühn and Schönert 2002, 295–296). For Walton (2011, 462), the poet “need not expect readers to imagine a fictional narrator using the words of the poem, or to suppose that she herself, the author, meant them ‘seriously’. The poet might expect the reader merely to recognize her invitation to use the words himself, to recognize her role as a thoughtwriter.” I am not sure whether this is the case for all lyric poetry,13 but for explicit remittance works such as Günther’s poem on the death of Anna Helena von Zedlitz, Walton’s description fits quite well. Baroque remittance lamentation poetry is neither fictional nor nonfictional, we cannot clearly decide whether it prescribes imaginings or not, the ‘I’ or the speaker is neither the author nor a persona. To understand how the genre convention of this type of poems worked in their times and how baroque readers used these texts in their own ‘games of make-believe’, the best approach seems to be Walton’s notion of thoughtwriting. Finally, there is another problem with the current reception of baroque lamentation poetry: Readers of today might not know about the ancient genre convention. The ‘mutual belief principle’ presumes that readers know the conventions of a genre at the time an older text was written. So if readers today do not know the baroque rules of the genre epicedium, they might read Günther’s lamentation poems – according to the later genre conventions of what in German is called Erlebnisdichtung [that is, the poetry of experience] – as authentic statements of the author himself. This is not to be disqualified as the ‘mistake’ of an uneducated, naïve reader. It should rather be described as a phenomenon occurring in the reception of baroque texts nowadays when readers use old texts as props in their own games of make-believe. People who have read a bit more of early eighteenth-century poetry may know that Günther also wrote remittance works where the deceased is the speaker of the text. This sounds odd today, but it fits perfectly with the habits of poetic epitaphs in the seventeenth and eighteenth century. Under Günther’s || 13 Walton (2011, 466) himself does not claim this to be true for all lyric poetry, but he lists several subgenres where the notion of ‘thoughtwriting’ fits quite well, as for “political and religious songs […], prayers, chants, mantras”. Poems do not even have to be meant as thoughtwriting by their authors, but in the concrete acts of reception, they have the “propensity to function as thoughtwriting” (Walton 2011, 473).

232 | Sonja Klimek various works, we find a poem called Den seeligen Hintritt / Der / F Agnetha Philippina Rüdigern / […] den 28. Febr A 1715. / J C G [On the blessed passing of Mistress Agnetha Philippina Rüdigern / on the 28th February 1715 (my translation)] (Günther 1998, 144). Here, the author gives his own name (at least the initials) in the headline of the poem. Nevertheless, the ‘I’ of the text is Agnetha herself, speaking in the following nine stanzas about her life, her family, her death and her hope for the future life in God, finishing in a consolation for her family and the typical Christian memento mori: Jhr folgt mir endlich alle; / Genung, ich geh voran” [You’re all going to follow me one day. I just go ahead (my translation)] (Günther 1998, 146). So can we talk about this poem as an example of ‘persona-factual’ lyric poetry, just because the speech act itself is clearly invented? Is it even ‘persona-fictional’? Agnetha is after all not an invented character, but a real person. The information she gives about her life and death are correct. Even if, once again, the four categories are not easy to apply, the example of this text shows that roleplayings and unnatural speech situations such as the talking of a dead person are not unfamiliar to baroque lamentation poetry. For some examples of Günther’s late-baroque lamentation poetry, the question of fictionality cannot simply be answered with yes or no. In the collected poems of the same author, we also find lamentation poems which are devoted to friends of Günther himself, such as the sonnet Den höchst-schmertzlichen Fall / eines lieb-gewesenen Schul-Freundes / beweinte Dessen betrübtester Bruder [The most grievous fall / of a once beloved schoolmate / bewept by his most aggrieved brother (my translation)] (Günther 1998, 139), where the characteristic ‘In the name of another’ is missing. The text could represent a typical baroque style exercise on a well-known story, as the names ‘Jonathan’ and ‘David’ (as the speaker calls himself) hint to the biblical friendship (see 1 Sam. 18–20) and David’s lamentation Carmen for his dead friend (see 2 Sam. 1). But readers of today know from the annotations in the modern edition that the poet Günther really had a schoolmate with the similar name ‘Johann’ who had died at that time (see Bölhoff 1998, 990–991). So this text can indeed be read as an example of what Zymner names ‘author-factual’ (or at least ‘author-nonfictional’) poetry. In this case, the biblical allusion would serve as a device to highlight how deep the author’s friendship to the deceased was and, at the same time, to present the author in the role of David, the ancient poet-king. This literary re-shaping of actual life by recurring to the archetypical narrative of David bemoaning his best friend probably did not hinder the impression of authenticity. Readers today can, as Bauer (2000, 120) puts it with regard to a similar device in Paul Fleming’s early modern poetry, see through this role-play and look for an individual, afflicted ‘I’ that has not yet any other means of expression at his dispos-

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 233

al but the topoi of literary tradition and the norms of antique rhetoric. – Nonetheless, without the extra biographical information, readers would not know if the text was authentic or not and therefore could also read it as an example of purely fictional role poetry.

3.2 Albrecht von Haller: personal grief between baroque convention and ingenious expression of the self Some years after Günther, the famous Swiss polymath Albrecht von Haller writes a suite of lamentation poems in which he describes the death of his young wife and deplores his loss. In the first edition of these texts, we find a poem entitled Über Marianens anscheinende Besserung, Den 16. Octobr. 1736 [On Mariane’s apparent recovery, 16 October 1736 (my translation)]. The title already includes the date of the supposed improvement of the illness, suggesting a kind of diary style and thus claiming authenticity. Nonetheless, later editions of the same poem give the 30 of October (see Boschung 1994, 22). This change of date might be motivated by the author’s wish to heighten the abruptness of the change described by the following poem, because his wife died 31 October. To think her feeling better 30 October and to have her die one day later is more dramatic than to have two weeks between these events. So, even if the name of Haller’s wife and the date suggest authenticity, the later change of dates hints to a certain kind of stylisation. The Trauer-Ode. Beim Absterben seiner geliebten Mariane gebornen Wyß. Novembr. 1736 [Mourning ode, [written on the occasion of] the death of his beloved Mariane, born Wyß, in November 1736 (my translation)] (Haller 1768, 161– 166) claims to be written shortly after the death of the poet’s wife. While the title still mentions the speaker of the poem in the third person (seiner [his]), the stanzas themselves are written in first person narration: A ‘Lyrical I’ deplores the sudden loss of his wife in a foreign country. When reading about Haller’s personal fate in his diary (see Boschung 1994, 25), it is difficult not to apply Ryan’s principle of minimal departure, interpreting the text as the purely authentic expressions of a mourning widower who goes outside into nature to start his mourning in a forest. Zymner (2009, 12) does so when choosing the text as an example of his third category (‘author-factual’ lyric poetry). In general, research on Haller has done so for a long time, emphasising the unusualness of Haller’s poem, as the poet claims to ignore the rules of baroque lamentation poetry by expressing his own personal grief so directly. The ‘Lyrical I’ explicitly distances himself from the conventions of baroque lamentation poetry:

234 | Sonja Klimek […] Nicht Reden, die der Witz gebieret, Nicht Dichter-Klagen fang ich an; Nur Seufzer, die ein Herz verlieret, Wann es sein Leid nicht fassen kann. Ja, meine Seele will ich schildern, Von Lieb und Traurigkeit verwirrt, […]. (Haller 1768, 161)14

Indeed, these phrases seem to prepare the Erlebnisdichtung [the poetry of experience] of Klopstock and the young Goethe that was created about forty years later, as Siegrist (1967, 32–33) states. But is the speaker of the poem completely trustworthy? Are his words really ‘sighs of a heart’, not ‘speeches generated by cleverness’ or typical ‘complaints of a [professional] poet’? – Readers that have been accustomed with Erlebnisdichtung might believe so. Nevertheless, Haller lets his speaker use Dichter-Klagen [complaints of a poet (My translation)] to express his grief: The idea to go out into an allegoric landscape and there to think of the unreachable beloved is an established topos of literature, which can be found in Dante’s Divina Commedia and Petrarch’s Canzoniere (see Braungart 2011, 135) and hence is not Haller’s original idea. Of course, such a stylisation does not mean that Haller does not really deplore the loss of his wife, but it shows how he falls back on well-established stereotypes when referring to his own indi-vidual feelings. Still today, lamentation poetry is usually read as being nonfictional. Readers tend to identify the ‘I’ of the text with the real text-producer. However, the example of Günther’s remittance work ‘In the name of another’ has shown that this is not correct for all baroque texts of this genre. Genre conventions of the epicedium allowed a speech that cannot be classified as being fictional or nonfictional, but can be understood with Walton’s term thoughtwriting. But if a text shows no signs of a fictional speech act or a role poem, today’s readers might apply what Ryan (1980, 43) calls the principle of minimal departure, reconstruing the ‘I’ of the text as more or less ‘the personality of the poet’. Nevertheless, as the example of Haller has shown, even in cases where the story of the poetic ‘I’ shows many striking parallels to the life of the real poet, the content of the text might have been standardised anyway in literary topoi. Readers might not expect a poet to play a game of make-believe when saying ‘I grieve’; but the genre convention of baroque lamentation includes this option. || 14 [No speeches, generated by cleverness, / Nor complaints of a [professional] poet will I utter; / Only sighs that come straight from a heart / When it cannot comprise its grief. / Yes, I will depict my soul / As it is perplexed by love and sadness […]. (My translation)]

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 235

3.3 Fictional texts being read as nonfictional ones: ‘The principle of minimal departure’ and the reception of Edward Young’s Night Thoughts and Novalis’ Hymnen an die Nacht About mid-eighteenth century, the baroque genre convention of lamentation poetry must have changed, at about the time when Edward Young, the abovementioned English priest and poet, published his long Christian poem The Complaint, or Night Thoughts on Life, Death and Immortality in about 10 000 blank verses, starting in 1742 with The COMPLAINT. NIGHT the FIRST to NIGHT the FOURTH. The eminent success of his work brought him to publish successively further Nights, until NIGHT the NINTH and LAST. The CONSOLATION (1745). Right from the publication of his first Nights, Young wanted his obviously fictional text to be read in a nonfictional, autobiographical context, assuming that the authenticity of grief, the immediate impression of his own familiar losses, would enhance the critical reception of his poem. Therefore, in the preface, Young generated what in the framework of Walton would be called a primary fictional truth by giving the fictional text a nonfictional coat: “AS the Occasion of this Poem was Real, not Fictitious; […] [T]he Facts mentioned did naturally pour these moral Reflections on the Thought of the Writer” (Young 1989 [1742], 35). This ostentation of factuality lead the readers of Young’s time to apply Ryan’s (1980, 43) aforementioned principle of minimal departure by reconstruing the pronoun ‘I’ ‘as retaining’ the personality of ‘the actual speaker’, that is, as Young himself. They filled the blanks of the text world in their own personal game worlds “as much like the real one as the core of primary fictional truths permits” (Walton 1990, 144–145),15 even if by its form, the Night Thoughts do not compose a traditional epicedium, but ‘moral Reflections’ (Young 1989 [1742], 35) with narrative passages in verse, with characters called Narcissa, Philander and Lucia. Readers of the bilingual, English-German translation with extensive commentary by the translator Johann Arnold Ebert from 1760 even got to know the ‘exact occasion’ of this poem, which is claimed to be the real death of || 15 This is what Walton (1990, 140–145) calls the reality principle; ‘Primary fictional truths’ are generated “directly” by the propositions of a fictional text, whereas ‘implied fictional truths’ are generated “indirectly”, that is, those fictional truths have to be inferred by the reader in their personal game of make-believe, since they are not spelled out explicitly. This process is guided by several principles. Beside the Reality principle, Walton suggests a Mutual-belief principle. For a discussion of those and other principles, see the contribution by Bareis in this volume.

236 | Sonja Klimek Young’s step-daughter in 1736 and her husband in 1740, the year when Young’s own wife died as well. Ebert even identified the fictive characters of the poem with persons from reality: According to him, ‘Philander’ and ‘Narcissa’ are identical to Young’s step-daughter and her husband, while ‘Lucia’ would be the equivalent of Young’s wife .16 Probably, it was Young’s own preface, in combination with the translator’s commentaries that forced the autobiographical reading, thus making Young one of the most important poets for the German movement of the Erlebnisdichtung [the poetry of experience], in the second half of the eighteenth century (see Perels 2006, 65–71). Then, at the end of the century, readers began to reject the highly artificial construction of the Night Thoughts and the mass of theological, philosophical, literary and scientific thought that was transported within the verses because they perceived it to be inconsistent with the proclaimed authenticity and the immediate event-character of Young’s poetry. Obviously, something had changed in the way people expected ‘good’ poetry to be like. Thus the clergyman’s popularity in Germany fell again (see Perels 2006, 65). Readers of the late eighteenth century no longer appreciated the genre convention of baroque lamentation poetry that included the option of ‘I grieve’ as make-believe, of expressing personal thoughts and feelings through rhetorical norms and set phrases or through fancied situations like the voice of the dead talking (see above, Günther’s poem to Agnetha Philippina Rüdigern where there is a true occasion of grief beneath the unnatural speech situation). They expected a poem to be nonfictional and genuine, an authentic expression of the author without any literary staging. For readers of the late eighteenth century with their specific genre expectations, Young became something of an impostor. His Night Thoughts fell out of favour. Nevertheless, we still find a trace of reception in the works of the German romanticist Novalis. Under the date of 23 April 1797, during the time of mourning for his dead fiancée Sophie von Kühn, about three weeks before the quasimystic experience he claims to have had, which is called the ‘Sophie-Event’ (see Novalis 1975, 35–36) at her grave,17 Novalis (1975, 30) states in his diary:

|| 16 See Young (1786): Ebert’s commentary on page 192, third night. 17 In short phrases, the diary describes a strangely joyous personal experience in which death and time seem to count no more: “Dort war ich unbeschreiblich freudig – aufblitzende Enthusiasmus Momente – Das Grab blies ich wie Staub, vor mir hin – Jahrhunderte waren wie Momente” [There, I was joyful beyond words – flashing moments of enthusiasm – I blew [her] grave away from me as if it were dust – centuries were like moments (my translation)]. Novalis declares to have had the feeling of his dead fiancée being close to him “ihre Nähe war fühlbar –

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 237

“Abends in Youngs Nachtgedanken geblättert” [In the evening, I browsed in Young’s Night Thoughts (my translation)]. This short statement gave rise to a lot of research in order to identify relations between Young’s long Christian blankverse poem and Novalis’ prose-verse mixture text Hymnen an die Nacht (Hymns to the Night, written between 1797 and 1800, first print 1800), a text that was supposed to be the result of working through his own grief (see Kamla 1945; Perels 2006, 72–73). Especially the third hymn was supposed to be highly autobiographical due to some semantic and syntactic similarities to the description of the ‘Sophie-Event’ in Novalis’ diary.18 As well as for Young, research on Novalis has long shown that the autobiographical reading of the Hymnen an die Nacht is quite inadequate, that is, that the text shows more signs of a highly artificial, constructed situation than of an autobiographical, quasi diarist immediate confession. 19 Nevertheless, readers of the early nineteenth century tended to read the texts of Novalis as authentic expressions of personal grief (see Perels 2006, 65), using them as props in their own games of make-believe to create a game world where the unfortunate poet deplores his own loss.

4 Generating fictional truth in readers’ game worlds The study of some examples of eighteenth-century lamentation poems has shown that for those examples of lyric poetry, the question of fictionality is not a question to which the answer is simply yes or no. At the beginning of the early modern period, there must have been something like the ‘institution’ of fictionality, even for lyric poetry, as the paratexts of Günther’s, Haller’s and Young’s || ich glaubte sie solle immer vortreten.” He felt as if she could step out of her grave every now and then. 18 The third hymn describes a kind of ‘unio-mystica’ experience in which the beloved dead is identified with God. The theme of the grave that turns into dust returns (“Zur Staubwolke wurde der Hügel”), as well as the idea that time counts no more (“Jahrtausende zogen abwärts in die Ferne, wie Ungewitter”) (Novalis 1978, 135). 19 Dick (1967, 162–170), for example, has shown that the description of Novalis’ ‘unio-mystica’ with Sophie at his grave is far from a spontaneous revelation but can, in many details, be traced back to some letters the poet had written in the weeks after Sophie’s death, before that special evening at the graveyard. For a general review of the older positions and a new presentation of the highly artificial intertextual elements in the Hymnen an die Nacht, see Perels (2006, 63–67).

238 | Sonja Klimek poems have shown. However, the rules of this ‘institution’ seem to have changed during the eighteenth century. The genre expectations for lamentation poetry changed, baroque conventions were lost. In a retrospect from today, our categories of fictionality or factuality may prove to be unsuitable.20 Even differentiated categorisations such as those of Lamping (2000) and Zymner (2009) are not always applicable, as the case of Günther’s remittance works has shown. Despite some special cases where a title or a subtitle clearly indicates that the ‘I’ of the text does not refer to the poet himself, the recipient often cannot know whether the text was intended as fictional or nonfictional by the author. In texts like Günther’s epicedium to the lady of Zedlitz, the phrase Jm Namen eines andern” [In the name of an Other] functions – according to the genre convention of baroque lamentation poetry – as an inquit-formula. To ignore such an inquitformula would lead the recipient to apply the principal of minimal departure to construct his personal game world in a way obviously not meant by the author, identifying him with the speaker of the text. This is always the case when ignoring inquit-formulas and other signals of narrative fictionality. Let us take, for example, the little motto from Pascal Mercier that opens the present paper. It is torn from Lea, a fictional text, as indicated by the genre Novelle [novella] in the subtitle. In its original length, the quotation reads: “‘Poetry’, hatte sie gesagt, ‘is a strictly solitary affair; solipsist even. I ought not to talk to you about it. But… well…’” (Mercier 2009 [2007], 232). The quotation marks and the paratext (Novelle) clearly state that the ‘I’ does not refer to the real author, but to a fictive persona. In other words, it is not the professor of philosophy Peter Bieri (publishing his fictions under the pseudonym ‘Pascal Mercier’) who asserts the proposition with this idea of poetry being something written exclusively for solitary lecture and reflection,21 but an English-speaking nurse within Mercier’s novella that is written in German. Early eighteenth-century occasional poems were not exclusively destined to solitary lecture, but also to be read out to a public on a certain occasion, as, for instance, a funeral. In the early modern perception, a poem could at the same time be stereotyped and nonetheless (in some way) be ‘true’. In addition to this, in the case of lyric poetry, there are often no inquit-formulas or other clear signals of fictional|| 20 For this problem, see Köppe (2014), who is merely concerned with prose fiction. 21 This point of view that lyric poetry were texts expressing subjectivity and being made for silent and solitary lecture (“stille Lesen von irgendwie Subjektivität ausdrückenden Gedichten”) has dominated since around 1800 (see Zymner 2009, 194). So the texts analysed in the present paper stem from the century when, at its beginning, this opinion did not yet exist and, during its course, just started to rise.

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 239

ity. So in contrast to narrative texts with genre names as signals of fictionality, concerning lyric poetry the recipient will, in most cases, never know whether a poem was meant to be ‘fictional’ or something else. The ‘Lyrical I’ often rests in a state of ‘drifting fictionality’ (see Fricke and Stocker 2000) which allows the reader to reconstrue it as a fictive persona or as a mask through which the poet can speak himself. If there are no obvious signals of fictionality, a reader of today can only decide how to construct the game world in his own ‘game of make-believe’ (Walton 1990), following, for example, the principles of minimal departure or – if he knows them – of genre conventions. Many elements in these personal game worlds of text reception are constructed in addition to the elements of the work world and therefore exist relatively autonomously from whatever might have been the case in the game world which the author imagined during his act of text production.

References Anderegg, Johannes. Sprache und Verwandlung. Zur literarischen Ästhetik. Göttingen: Vandenhoeck & Ruprecht, 1985. Bauer, Barbara. “Naturverständnis und Subjektkonstitution aus der Perspektive der frühneuzeitlichen Rhetorik und Poetik.“ Künste und Natur in Diskursen der Frühen Neuzeit. Ed. Hartmut Laufhütte. Wiesbaden: Harrassowitz, 2000, 69–131. Beardsley, Monroe C. “Fiction as Representation.” Synthese 46 (1981): 291–313. Bleumer, Hartmut and Caroline Emmelius (eds.). Lyrische Narrationen – narrative Lyrik. Gattungsinterferenzen in der mittelalterlichen Literatur. Berlin: de Gruyter, 2011. Bölhoff, Reiner. “Stellenkommentar.” Günther, Johann Christian. Werke. Ed. Reiner Bölhoff. Frankfurt a.M.: Deutscher Klassiker Verlag, 1998, 952–1548. Boschung, Urs. Albrecht von Haller in Göttingen 1736–1753: Briefe und Selbstzeugnisse. Bern: Huber, 1994. Braungart, Georg. “Naturlyrik.” Handbuch Lyrik. Ed. Dieter Lamping. Stuttgart and Weimar: Metzler, 2011, 132–139. Burdorf, Dieter. Einführung in die Gedichtanalyse. 2nd ed., reworked and actualised. Stuttgart and Weimar: Metzler, 1997. Burdorf, Dieter. Lyriktheorie. Reallexikon der deutschen Literaturwissenschaft. Vol. 2. Ed. Harald Fricke et al. Berlin and New York: de Gruyter, 2000, 502–505. Dick, Manfred. Die Entwicklung des Gedankens der Poesie in den Fragmenten des Novalis. Bonn: Bouvier, 1967. Fricke, Harald and Peter Stocker. “Lyrisches Ich.” Reallexikon der deutschen Literaturwissenschaft. Vol. 2. Ed. Harald Fricke, Klaus Grubmüller, Jan-Dirk Müller, Friedrich Vollhardt, Klaus Weimar. Berlin and New York: de Gruyter, 2000, 509–511. Gabriel, Gottfried. Zwischen Logik und Literatur. Erkenntnisformen von Dichtung, Philosophie und Wissenschaft. Stuttgart: Metzler, 1991.

240 | Sonja Klimek Günther, Johann Christian. Werke. Ed. Reiner Bölhoff. Frankfurt a.M.: Deutscher Klassiker Verlag, 1998. Haller, Albrecht von. Versuch Schweizerischer Gedichte. Zehnte, rechtmäßige, vermehrte und veränderte Auflage. Göttingen 1768, 161–166. Hamburger, Käte. Die Logik der Dichtung. 3rd ed., reworked. Stuttgart: Klett/Cotta, [1958] 1977. Hühn, Peter and Jörg Schönert: “Zur narratologischen Analyse von Lyrik.” Poetica 34 (2002): 287–305. Kamla, Henry. Novalis’ Hymnen an die Nacht. Zur Deutung und Datierung. Copenhagen: Ejnar Munksgaard, 1945. Köppe, Tilmann. “Fiktionalität in der Neuzeit.” Fiktionalität. Ein interdisziplinäres Handbuch. Eds. Tobias Klauk and Tilmann Köppe. Berlin, New York: de Gruyter, 2014, 419–439. Lamping, Dieter. Das lyrische Gedicht. Definitionen zu Theorie und Geschichte der Gattung [1989]. 3rd ed. Göttingen: Vandenhoeck & Ruprecht, 2000. Mercier, Pascal. Lea. Eine Novelle. 3rd ed. München: btb, 2009. Müller-Zettelmann, Eva. Lyrik und Metalyrik. Theorie einer Gattung und ihrer Selbstbespiegelung anhand von Beispielen aus der englisch- und deutschsprachigen Dichtkunst. Heidelberg: Winter, 2000. Müller-Zettelmann, Eva. “Lyrik und Narratologie.” Erzähltheorie transgenerisch, intermedial, interdisziplinär. Eds. Vera and Ansgar Nünning. Trier: Wissenschaftlicher Verlag Trier, 2002, 129–153. Novalis. “Hymnen an die Nacht.” Schriften. Die Werke Friedrich von Hardenbergs. Ed. Paul Kluckhohn and Richard Samuel. Vol. 1: Das dichterische Werk. 2nd ed. Stuttgart: Kohlhammer, 1978, 131–156. Novalis. Schriften. Die Werke Friedrich von Hardenbergs. Ed. Paul Kluckhohn and Richard Samuel. Vol. 4: Tagebücher, Briefwechsel, Zeitgenössische Zeugnisse. 2nd ed. Darmstadt: Wissenschaftliche Buchgesellschaft, 1975. Perels, Christoph. “Von Young zu Novalis. Hardenbergs ‘Hymnen an die Nacht’ und die ‘Night Thoughts’ in Deutschland.” Zwischen Aufklärung und Romantik. Neue Perspektiven der Forschung. Ed. Konrad Feilchenfeldt, Ursula Hudson, York-Gothart Mix and Nicholas Saul. Würzburg: Königshausen & Neumann, 2006, 61–86. Ryan, Mary-Laure. “Fiction, Non-factuals, and the Principle of Minimal Departure.” Poetics 9 (1980): 403–422. Siegrist, Christoph. Albrecht von Haller. Stuttgart: Metzler, 1967. Walton, Kendall L. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Walton, Kendall L. “Thoughtwriting – in Poetry and Music.” New Literary History 42 (2011): 455–476. Young, Edward. Night Thoughts. Ed. Stephen Cornford. Cambridge and New York: Cambridge University Press, 1989. Young, Edward. Dr. Eduard Youngs Klagen, oder Nachtgedanken über Leben, Tod und Unsterblichkeit. In neun Nächten. […]. Aus dem Englischen ins Deutsche übersetzt, durchgehends mit kritischen und erläuternden Anmerkungen begleitet, und […] herausgegeben, von J[ohann] A[rnold] Ebert. http://www.textkritik.de/young/index.htm#fn2. Braunschweig: Ludolph Schröders Erben, 1786 (11 March 2013). Zipfel, Frank. Fiktion, Fiktivität, Fiktionalität. Analysen zur Fiktion in der Literatur und zum Fiktionsbegriff in der Literaturwissenschaft. Berlin: Erich Schmidt, 2001.

Generating Fictional Truth in Eighteenth-Century Lamentation Poetry | 241 Zipfel, Frank. “Lyrik und Fiktion.“ Handbuch Lyrik. Theorie, Analyse, Geschichte. Ed. Dieter Lamping. Stuttgart and Weimar: Metzler, 2011, 162–166. Zymner, Rüdiger. Lyrik. Umriss und Begriff. Paderborn: Mentis, 2009.

| Section 3 – Theatre and Music

Frank Zipfel

Fictionality and Make-Believe in Drama, Theatre and Opera For a long time, concepts of fictionality1 have mostly been discussed with regard to literary narratives. Even today, the chief amount of the scholarly work concerned with theoretical issues of fictionality is based on, and limited to, considerations referring to narrative literature, where narrative is mostly understood in the narrow sense of a story being told by a narrator or a narrative instance.2 It is certainly to the credit of Kendall Walton’s influential Mimesis as Make-Believe that the discussion about fictionality has been opened beyond the boundaries of literary narration, and a need for elaborating on a transmedial concept of fictionality has been put on the agenda. In fact, a transmedial concept of fictionality can be useful in grasping relevant features of works of art which are not literary narratives (in a narrow sense). Thus, fictionality cannot only be a relevant category for investigations into novels or short stories, but also for the explanation of drama and poetry as genres and for the analysis of particular dramas and lyrical poems. Moreover, fictionality is a concept that may help us to grasp specific aspects of non-literary art forms like film, theatre, opera and dance. I agree with Walton insofar as I share his opinion that we can profit form a transgeneric and transmedial concept of fictionality. However, I will not follow Walton’s overall approach because I am not very comfortable with his assumptions about the relationship between representations and fiction, i. e. with his somewhat counterintuitive claim that all representations are fictions.3 My own approach to a transmedial understanding of fictionality starts from what I have called elsewhere a multilayered concept of fictionality (Zipfel 2014),4 a concept which I will briefly elaborate in the first section of my paper. The

|| 1 In this paper, I will use the term fictionality rather than fiction. The notion of fiction in English is very much associated with literary narration, whereas fictionality can be used to designate works of art in different art forms or media. 2 I rely on the distinction between a narrow and a wide concept of narration. For narration in a narrow sense, the mediation by a narrator is essential, which means that narration in a narrow sense designates the result of the telling of a story by a narrator. Narration in a wide sense simply means story presentation. In this understanding, all kinds of story presentation (in films, games and other media), with or without a specific narrator, qualify as narration (see. e. g. Ryan [2008]; Jannidis 2003). 3 See New (1999, Ch. 3); Zipfel (2001, 23–24). 4 For a similar approach, see also Ryan (2010).

246 | Frank Zipfel other three sections will deal with fictionality of drama as a literary genre and of theatre and opera as art forms. As a matter of fact, since Walton opened up the path for a transmedial understanding of fictionality, investigations regarding fictionality as a relevant feature of art works have been extended to various fields, especially to the realm of the feature film. Comparatively little has been said, however, about the specific fictionality of drama, probably because it has either been taken for granted or is supposed to be explicable in terms of narrative fiction. Also, the specific fictionality of theatre and opera performances has rarely been discussed. The goal of this paper is to meet this desideratum and to develop some preliminary and introductory thoughts about how we can conceive of the fictionality of drama as a literary genre and of the fictionality of theatre and opera as specific art forms. I contend that such an investigation is bound, on the one hand, to heighten our awareness of transmedial similarities between differing genres, between differing art forms and between art works realised in different media; yet, on the other hand, it will also allow us to lay out the media-specific differences between these genres and art forms.

1 A multilayered approach to a transmedial concept of fictionality Attempts to define the concept of fictionality have been elaborated across theoretical frames, e. g. in terms of textual semantics, in terms of mode of appreciation, in terms of authors’ intentions and also in terms of institutional practice (see e. g. Zipfel 2001; Gorman 2005). All these attempts have given us fruitful, but mostly partial, explanations of how to understand fictionality as an artistic phenomenon. In my opinion, a fully-fledged theory of fictionality should be able to explain the complex artistic and social interactions that are involved in the production of, and in the response to, so-called fictional works. Therefore, I favour a multilayered approach to the concept of fictionality. I call this approach ‘multilayered’ because it encompasses three different elements that are situated on different analytical levels. The three components of this multilayered approach are: 1) The fictional world, 2) the game of make-believe and 3) institutional practice. The different levels addressed by the different components are obviously the referential or semantic level for the fictional world, the author/reader or communication level for the game of make-believe and the institutional or pragmatic level for institutional practice. It is beyond the scope of this paper to give a detailed account of this multilayered concept of fictionality; I therefore have to limit myself to some very general considerations.

Fictionality and Make-Believe in Drama, Theatre and Opera | 247

I argue that fictionality is a concept that designates phenomena that in some way or another deal with fictional worlds, that the way these fictional worlds are dealt with involves a game of make-believe and that this game is embedded in an institutional practice. A slightly more elaborate account would run as follows: A fictional work, in general, is a work produced according to the conventions of the institutional practice of fictionality. Such a work has as a rule been composed and published by the author with the specific intention that an audience uses the work as a prop in a game of make-believe and that audiences will actually adopt the so-called fictional stance towards the narration because they recognize this intention. To function as a prop in a game of makebelieve means that a work mandates imaginings by virtue of conditional principles of generation which are part of the conventions of the institutional practice. Propositions whose imaginings are mandated by, or can be inferred from, the work in question are called fictional truths, and fictional worlds can be understood as a set of fictional truths.5 Conceived in this way, fictionality can be seen as a transmedial concept because the outlined multilayered approach to fictionality does not focus on a specific art form or medium. Out of the three elements of the multilayered concept of fictionality, I will not elaborate on the notions of the fictional world and institutional practice. I understand the notion of the fictional world very much like Walton: As a set of fictional truths which can legitimately be inferred from the work in question.6 As for the concept of institutional practice, I rely on the explanations of Lamarque and Olsen (Lamarque and Olsen 1994, Ch. 2 and 10).7 What I will briefly elaborate on is the concept of the make-believe game, because the differing concepts of make-believe are the basis for my investigations into the fictionality of drama, theatre and opera. A great many of the differing theories about fictionality converge on the assumption that readers or audiences are in some way or other invited to take a specific fictional stance towards fictional works. This fictional stance is more often than not described in terms of make-believe or concepts closely related to it. However, the question about what make-believe means with regard to fictionality has been vividly debated during the last 20 years. The discussion about the so-called paradox of fiction is perhaps one of the best examples of the dispute around fictional stance and make-believe (see Zipfel 2012). While it is be|| 5 It is quite obvious that this concept is heavily indebted to the thoughts of Kendall Walton and Gregory Currie (for the concept of make-believe and different understandings of it), as well as to Peter Lamarque and Stein Haugom Olsen for their institutional approach to fictionality. 6 For a more detailed account, see Zipfel (2011). 7 For a discussion of possible amendments to this concept, see Alward (2010).

248 | Frank Zipfel yond the scope of this paper even to give a brief account of this debate, I would like to sum up some of the most prominent contributions by singling out three different recurring interpretations of fictional stance and make-believe respectively, even if I risk undue oversimplification of the matter. To a certain extent, the three differing interpretations of make-believe can be paralleled with three different concepts of imagination: Imagining de dicto, imagining de re and imagining de se (see Alward 2006).8 First, in his seminal Mimesis as Make-Believe, Kendall Walton introduces the concept of make-believe by means of an analogy to children’s play. Just as children in a game of cowboys and Indians believe for the duration of the game that some of them are cowboys while others are Indians, or that a wooden stick is a rifle with which an enemy can be shot, the audience of fictional works engage in a similar game of make-believe concerning what they are reading, hearing or seeing for the duration of their engagement with the work. Fictional works can thus be conceived as props in games of make-believe. Moreover, they are props in games of make-believe by virtue of conditional principles of generation (see also Bareis 2014). They mandate imaginings, and imaginings for Walton involve imagining about oneself or imagining de se. Second, Gregory Currie has proposed a slightly different interpretation of make-believe in his early writings about fictionality.9 In Currie’s opinion, our make-believe regarding fictional narrations is “not merely that the events described in the text occurred, but that we are being told about those events by someone with knowledge of them” (Currie 1990, 73). Thus, Currie opens the path for a narratological understanding of make-believe. Understanding makebelieve as imagining one is being told a story in the form of a series of known facts is more or less tantamount to the traditional narratological assumption that fictional texts encompass a fictional narrator who is different from the real author. Moreover, this concept of make-believe encompasses imagining de dicto and de re, but not imagining de se. Of late, however, this understanding of make-believe has been repeatedly confronted with criticism from scholars who deny the assumption that all fictional art works have narrators and also from those who advocate that even in fictional literature there are narratorless narrations.10 The question of whether it is correct to assume a fictional narrator for every kind of fictional narration or whether there are narratorless narrations is || 8 About the different meanings of imagination, See also Stevenson (2003). 9 I specifically refer to Currie’s book The Nature of Fiction (1990). Currie has changed his view in his later work, especially in Narratives and Narrators (2010). 10 See e. g. Gaut (2004); Kania (2007); Thomson-Jones (2007); Walsh (2007, Ch. 4); Currie (2010); Köppe and Stühring (2011); Köppe and Stühring (2015).

Fictionality and Make-Believe in Drama, Theatre and Opera | 249

indeed an intriguing one – and there do not seem to be any clear-cut answers to it. Quite possibly, the answer depends on how the concept of narrator is understood, i. e. whether one adheres to a narrow definition of narrator as a personalized teller or to a wider definition of narrator as some kind of story-presentation device.11 A third conception of make-believe and the fictional stance focuses solely on imagining de dicto. This view of make-believe and fiction-specific imagination is prominent in the work of Peter Lamarque, Noël Carroll and Roger Scruton, among others. They opt for an understanding of imagining that Carroll has called “suppositional imagination” (Carroll 1997, 184) and which can be explained as “the capacity, which all rational beings exhibit to some degree, to entertain thoughts without affirming or asserting them” (Scruton 2010, 96). In these theories, to imagine is understood as a form of mental activity which can be described as being akin to holding something before the mind, as “entertaining the proposition that p unasserted” (Scruton 1974, 88) or “to contemplate the supposition” or as “entertaining a thought non assertively” (Carroll 1990, 88), or simply as “to entertain a thought” (Davies 2009, 272).12 I will not try to argue in favour of any of the three different understandings of the fictional stance and make-believe outlined. In my opinion, each of them has its advantages and disadvantages. All three understandings grasp important parts of the intuition they seek to explain, i. e. why fictional works do not have to be labelled as lies or falsehoods even if they do not present real events, and how it is that fictional works can be, and are, taken seriously in some way or other. 13

2 The fictionality of drama Traditionally, literature is divided up into three major genres: Narrative, drama and lyrical poetry. In so far as the distinction between drama14 and narration is || 11 See e. g. Wilson (2007); Bareis (2008, Ch. 3.3); Thon (2014); Zipfel (2015). 12 See also Gibson (2007, Ch. 5). 13 For a more detailed discussion of the differing conceptions of the fictional stance see Zipfel (2013). 14 In this paper, I use drama and dramatic texts to denote the specific literary genre constituted by literary texts that are meant to be represented in a theatrical form, or, to put it more plainly, that are “written to be performed” (Weidle 2009, 225). I will not use drama in the broad sense that is usually attached to it, as when Carroll maintains that “drama is a dual-tracked or two-tiered art form” and distinguishes between “the art of composing play texts [...] and the art

250 | Frank Zipfel concerned, this division is based upon a narrow definition of narration. It distinguishes drama from narration by claiming that dramatic texts lack narratorial mediation and that in drama the story is presented in an unmediated way: “The novel tells, the drama shows” (Nünning and Sommer 2011, 204).15 This assumption, however, can be questioned since in the printed text of a play, the lines of the characters are presented – so to speak – via the mediation of what is usually called stage directions or didascalia. The length and importance of these stage directions can vary considerably from one play to another, ranging from mere indications about which character utters which lines, to rather extensive descriptions of stage settings, characters’ appearances, clothing or non-verbal actions. It can be argued that when we read a dramatic text, we play a game of make-believe in which we perceive these stage directions as some kind of mediator’s or even narrator’s voice. Moreover, Korthals maintains that stage directions can be seen as an example of extradiegetic narration (see Korthals 2003, 111). This understanding of dramatic texts as a kind of narrative in a narrow sense is rather pertinent when we look at some modernistic or naturalistic dramas, such as the plays of August Strindberg, George Bernard Shaw, Sean O’Casey or Luigi Pirandello. In these plays, the stage directions are not confined to the indications about which lines belong to which character, but can be quite extensive and are often expressed in a patently narrative mode (see Schenk-Haupt 2007, 35–36). When we thus consider dramatic texts as narratives, more or less everything that can be said about narrative fiction is also true about drama. With such a view, there is no difference at all between the fictionality of drama and the fictionality of narrative texts. The problem, however, with regarding drama as a narration in a narrow sense is that we have to ignore one specific and most important feature of dramatic texts, namely that they are generally meant to be represented in some kind of theatrical performance. [S]uch works are typically written with the express aim that they be incorporated into performances; and even when they are not written with this aim, the choice to write a work in || of performing them” (Carroll 2006, 106). Like Fortier, I reserve drama for the first of the two, and use theatre for the second (See Fortier 2002, 10). I also do not see the advantage of calling the printed dramatic text ontologically defective as Bennett does, which means that he is bound to create a new ontological category, the work, which encompasses the printed drama and all representations of it but is neither of them: “Neither the text alone, nor any performance or group of performances alone, is the work itself; rather, the work comes into existence repeatedly by being interpreted in performance” (Bennett 1985, 4). 15 See also Segre (1980, 40), who claims that “the mediation of the I-writer” which characterises narration “has been eliminated” in drama.

Fictionality and Make-Believe in Drama, Theatre and Opera | 251 dramatic form rather than, say, a work of narrative prose generates a warranted association between the literary work and the tradition of theatrical performance. (Irving 2009, 45)16

For this reason, I find Searle’s proposal that “the illocutionary force of the text of a play is like the illocutionary force of a recipe to bake a cake” (Searle 1975, 329) rather convincing.17 In Searle’s view, the text of a play “is a set of instructions for how to do something, namely, how to perform a play.” In a similar vein, Noël Carroll regards play texts as comparable to blueprints (Carroll 2006, 109). So, dramatic texts can be described as instructions for a theatrical production or, to put it more mildly, as invitations to stage a play. To conceptualize plays in this way does not mean, however, that play texts are not regarded as artworks in their own right and only as recipes, or “semiporous formulas to be filled with execution,” as Carroll seems to imply (Caroll 2006, 108–109). It simply suggests that we see those texts as artworks that are conceived in the form of instructions for a theatrical production. When viewed this way, a text of a play is not a narration in the narrow sense, but still qualifies as narration in the wider sense. Moreover, in this view, the question about the fictionality of drama does not arise in the way in which it is usually discussed with regard to fictional narrations in the narrow sense, since plays would be seen as fictional narrations (in the wider sense) that present a story in the form of an instruction or invitation. The whole problem about who assumes the responsibility for the assertions made in the text and for the overall assertive status of the text, a problem which has been one of the major topics of the discussion about fictionality in narratives (in a narrow sense),18 vanishes because the overall speech act of the text is not an assertive one. The question that arises, however, is whether a play considered as an instruction for a theatrical performance can be regarded as fictional and, if so, in what sense. Of course, dramatic texts are fictional in the sense that they present fictional stories and in the sense that fictional worlds can be inferred; in other words, the feature that fictional narration (in the narrow sense) and dramatic texts have in common is the fact that fictional worlds are presented. The interesting question, however, is how the conception of plays as instructions for a

|| 16 See also Segre’s assertion about the relationship between text and representation: “[...] it is obvious that the text has been composed primarily in view of such representation” (Segre [1980], 39). 17 See also Schenk-Haupt (2007, 34), who contends that “[s]tage directions […] carry the illocutionary force of directives”. 18 See e. g. Searle (1975); Currie (1990); Cohn, (1990); Cohn (1999); Zipfel (2001).

252 | Frank Zipfel theatrical performance influences and alters the game of make-believe readers play with dramatic texts. After all, one could argue that plays are instructions by a real person (author) for real people (director, actor).19 Indeed, interpretations of the fictional stance entailing a fictional narrator meet some difficulties when we talk about the fictionality of a play. For instance, the view that our make-believe regarding fictional narrations is that we are being told about those events by someone with knowledge of them does not apply for instructions or invitations. One could even argue that the concept of fictionality only applies to texts whose dominant speech acts have a word-toworld direction of fit (i. e. are assertions, descriptions, explanations etc.) and not to texts whose dominant speech acts present a world-to-word direction of fit (i. e. requests, instructions, invitations etc.). But to take this view would probably mean oversimplifying the problem (see Searle 1979, 4; Zipfel 2001, 311–312). Another way of putting the question is: What kind of imagining is entailed in the make-believe game with dramatic texts? Readers of a play certainly engage in imagining de dicto, but what about imagining de re, let alone imagining de se? For example, the literary critic Monika Fludernik describes the way we read a play as follows: In reading a play, we imaginatively ‘stage’ it in our minds. Although this is a metaphorical way of expressing it, this reading process is different from the reading of fiction because – owing to the explicit staging information in the stage directions – it involves more visualisation than does novel reading. (Fludernik 2008, 363)

Now there are several aspects of this passage I do not agree with. It seems obvious to me that reading a play does not necessarily involve more visualisation than reading a novel; in most cases it will probably be just the opposite. Moreover, adopting the fictional stance or engaging in a game of make-believe in my opinion does not necessarily involve visualisation or any other kind of pretended sensory perception. The passage raises an interesting question, however, namely whether plays are to be read in the same way as novels with regard to their fictionality i. e. with regard to what readers are invited to imagine. One could argue that when we read Shakespeare’s tragedy of star-crossed lovers, we are not only invited to imagine that Romeo meets Juliet, but also to imagine that Romeo’s meeting Juliet is enacted on a stage (or is to be performed in a theatrical production). Imaginings mandated by dramatic texts then differ || 19 See Issacharoff’s assertion: “Didascalia are addressed by real person (the author) to real people (director and actors)” (Issacharoff 1987, 88). See also Schenk-Haupt’s claim that stage directions are factual (Schenk-Haupt [2007, 35]).

Fictionality and Make-Believe in Drama, Theatre and Opera | 253

from those mandated by narrative texts in the following way: With narratives we are invited to imagine the depicted events and what we are told about those events, whereas with dramatic texts, we are invited to imagine the enactment of the depicted events on a stage. When we read plays, we do not simply imagine fictional worlds, but fictional worlds as enacted worlds. Or to put it differently: Imagining de dicto would be the same in both cases: Readers imagine the fictional world as depicted. Imagining de re, however, would differ: With novels, readers are invited to make believe that they are told a story, and with plays, readers are invited to imagine that the story is being enacted. After what has been said, there are two different ways of reading a play with regard to its fictionality. In more specific terms, one can argue that the way a play is perceived depends upon the goal a reader pursues while she is reading. A reader who reads the play because she is interested in the story is likely to apprehend the play in a similar way as she does with novels and thus engage in a similar game of make-believe. The director or actor who reads the play in order to stage it, in contrast, will regard the text as instructions and thereby engage in a drama-specific game of make-believe with regard to what the text invites him or her to imagine. Still, the game of make-believe entailed by dramas is not the same as the one entailed by narrations: Dramatic texts mandate the reader to imagine the story as enacted, whereas narrative texts mandate us to imagine the story as told. This also means that the principles for generation of fictional truth are not the same and that the fictional worlds entailed by a narration and by a drama that both present the same story might not be identical. It may sound odd that the presentation of a story in a play should give rise to a different fictional world than the one conveyed by the narrative presentation of exactly the same story. Maybe this is an instance where the difference between the concepts of story and discourse, which is usually considered to be a rather clear-cut one, begins to blur. Be this as it may, I would argue that a comprehensive reading of a play entails perceiving the text as one that is meant to be enacted. Readers are invited to imagine a fictional world, but they are also invited to reflect on the theatrical enactment of that fictional world, and, furthermore, to consider the impact of the instructive mode of the text on what is to be considered as fictional truth.

254 | Frank Zipfel

3 The fictionality of theatre20 It has often been said that theatre as an art form is invested with a double fictionality, dramatic fiction on the one hand, and theatrical fiction on the other (cf. Birkenhauer 2005, 107). The concept of dramatic fiction points to the fictionality of the written play on which I have elaborated in the previous section. The concept of theatrical fiction can be understood as expressing the fact that the representation of a story in a theatrical form produces a specific fictional performance world. For a better understanding of the meaning of the expression theatrical fiction, it will be necessary to take a closer look at the particular conditions, possibilities and conventions of theatre as a specific form of art. The minimal definition of theatre proposed by Eric Bentley in his Life of Drama, i. e. “A impersonate B while C looks on” (Bentley 1964, 50)21, expresses the two major defining features of theatre: (1) The co-presence of (part of) the producing agents22 and the audience while the work of art is presented, and (2) the fact that theatre is based on an activity that in general terms can be called role-playing or play-acting. And when David Osipovich’s answers the question: “What is the sine qua non of theatrical performance?” in his insightful article dating from 2006 (Osipovich 2006, 465), what he does is to unpack and specify the aspects already mentioned in Bentley’s short definition. Osipovich’s lists three necessary conditions for theatrical performance: 1) At least one performer and at least one observer in the same room and at the same time, 2) a pretense on the part of the performer that the interaction between performer and observer is somehow other than it actually is, and 3) an awareness on the part of the observer that the pretense is occurring. (Osipovich 2006, 465)

There have, of course, been many other attempts at defining the basic features of the cultural practice of theatre, e. g.. in terms of semiotics, codes and subcodes (see e. g. Elam 2002), in terms of iconicity, deflection of reference and similarity on the material level (see Rozik 1992, 2008, 2009), in terms of time shift and duality of time structure (see Limon 2009), and in terms of display behaviour, attending and understanding (see Hamilton 2007). I will not try to discuss these different approaches here, but I would argue that in some way or other acting and enactment play a crucial role in all of them, as they do in Bent|| 20 For this section see also the third section of Zipfel (forthcoming). 21 See also Fischer-Lichte (1983, 16). 22 Not all the producing agents involved in a theatre production are present on stage (e. g. director, stage and costume designer, make-up artists etc.).

Fictionality and Make-Believe in Drama, Theatre and Opera | 255

ley’s and Osipovich’s approach. Consequently, in the following, I will elaborate on some aspects of Osipovich’s definition, which presents the advantage that its central element, i. e. pretence, can easily be related to concepts of fictionality. Osipovich uses the concept of pretence in order to describe what actors do on stage. Actually, Osipovich’s contribution is part of a widespread discussion among theorists of theatrical performance (such as David Saltz, Noël Carroll and James R. Hamilton) about whether actors on stage pretend to perform illocutionary acts and more generally about whether or not acting is related to some kind of pretence – a discussion that originates from Searle’s influential article on the logic of fictional narrative discourse in which pretended speech acts play a central role. Unfortunately, there are quite differing understandings of what pretending means and of what pretence entails. Thus, the answer to the question of whether and how acting on stage involves pretending mostly depends on how the concept of pretence is used. In fact, on the one hand, it can be argued that pretence is essential to what actors do on stage, i. e. that actors pretend to perform not only speech acts, but also other actions. However, on the other hand, it can be argued that it is misleading to say that an actor who plays Othello pretends to murder somebody, because what he actually does is to perform the action that in the respective game of the production counts as murdering. This action can consist of mimicking strangling or stabbing somebody and is thus a kind of pretence, but it can also consist of a clapping of hands if that is the sign for murder agreed upon in the given production. Again, it is beyond the scope of this paper to enter this debate.23 For our context, it suffices to note that the intuition behind the talk of pretence in relation to play-acting does catch something specific about theatre performances and that this something can be related to what Walton sees as the core of fictionality, i. e. make-believe. Actually, David Saltz maintains that “most of what Walton has to say about the audience’s interaction with representational artworks is directly applicable to the actor’s own imaginative engagement within the theatre event. To a large extent, actors usurp the role that Walton carves for audiences of artworks” (Saltz 2006, 212). Undoubtedly, some insight can be gained in relating Walton’s concept of the fictional stance to play acting, but in my opinion, Saltz’s statement turns the actual picture on its head. Therefore, let us have a closer look at the similarities and dissimilarities between the fictional stance and children’s games on the one hand, and between children’s games and play acting on the other hand. || 23 See Saltz (1991); Hamilton (2007, 2009); Meskin (2009) and especially Alward (2009) for a detailed account of the discussion.

256 | Frank Zipfel I start with a comparison between the fictional stance and children’s games. Walton explains the fictional stance in terms of a game of make-believe. The explanation is illuminating insofar as it highlights the fact that the fictional stance like children’s games entails a rule-governed practice and the engagement with a game-world which is not the actual world. However, there are at least two major differences between children’s games and the fictional stance: (1) Children are mostly the inventors of their make-believe games whereas audiences of fictional art works are not, and (2) children’s engagement in their games is an active one, insofar as they physically impersonate the characters of their games, whereas the engagement of the audience in a fictional world is a passive one insofar as it is purely mental and belongs to the realm of imagination (see Sainsbury 2010, 15). It seems to me that it is Walton’s metaphorical explanation of the fictional stance that is responsible for the fact that we see a similarity between the audience’s engagement with fictional works and playacting. In other words, it is only because the fictional stance has been explained in terms of children’s games, which, like play-acting, encompass make-believe, that such a resemblance appears. Now let us turn to the similarities and dissimilarities of children’s games and acting on stage. The two phenomena are obviously similar in many ways, but there are also some major differences, as Hamilton has pointed out. Children invent their games and characters as they play along, actors typically enact scripted stories; children usually impersonate character-types (an Indian, a cowboy, a gangster or more specifically, a chief, a sheriff and a crime boss), actors usually impersonate individuals (even characters like Harpagon in Molière’s L’avare are individuals although they are meant to represent a type); children play only for themselves, whereas “there is no non-audience practice of theatre” (Hamilton 2007, 51).24 Thus, one can say that actors engage in a game of make-believe that is usually based on pre-existing scripts and as a rule entails enactment of individuals for an audience. In my opinion, this kind of make-believe is an important and essential element of the social practice and art form of theatre as we know it, even if there are nonrepresentational examples of theatre.25 From what has been said about the similarities and differences between children’s games and the fictional stance on the one hand, and children’s games and acting on the other, it becomes evident that when the concept of || 24 See also Hamilton (2009, 100–101). 25 See the discussion between Meskin (2009) and Hamilton (2009) about acting as display behaviour.

Fictionality and Make-Believe in Drama, Theatre and Opera | 257

make-believe is applied to actors and their enactment, the term means something slightly different from what Walton describes as the fictional stance. For instance, in contrast to the make-believe that explains the fictional stance, the make-believe of actors cannot be described simply as imagining. While imagining may be an important component of acting, actors also have to perform real actions, both linguistic actions, such as uttering the lines that the displayed character is supposed to utter, and non-linguistic actions, i. e., more or less elaborate physical actions.26 Thus, features like technique, embodiment and presence are important components of make-believe in acting – and there is of course a lot more to say about these features of acting and enactment, but that lies beyond the scope of this paper.27 For a fully-fledged explanation of theatrical fiction, it is necessary to turn from the realm of production to the realm of the audience. It seems legitimate to say that a theatre audience is invited to make believe what is represented on stage. Moreover, the different ways of understanding make-believe as constitutive of the fictional stance that has been discussed earlier can easily be applied to theatre audiences. The difference between theatre and literary narrative lies of course in the fact that with texts, audiences engage imaginatively with what they read, whereas in theatre, audiences engage imaginatively with what is enacted and thus presented to them visually as well as acoustically. Through this act of imagination, the audience engages with the performed story or with the production world of a performance.28 When one speaks of the production world, one wants to highlight the fact that although different productions of the same play show the fictional world of the written drama, every single production presents its own production world.29 The production world is characterized by the numerous choices that have to be made in order to stage a play, i. e. casting, costumes, make-up, setting and lighting etc. (see Wolterstorff 1980, 253–254). Moreover, one can distinguish

|| 26 See Meskin (2009, 55–56); Fischer-Lichte (2010, Ch. 3). 27 See e. g. Crease and Lutterbie (2006); Jaeger (2006); Erickson (2006); Rozik (2002). 28 For a more detailed account, see Zipfel (2014). For a differing approach, see Hamilton (2007, Ch. 4). 29 Hamilton advocates the view that different theatre productions are not to be conceived as presentations of one and the same text because every theatre production/performance is an art work in its own right (see Hamilton [2007, Ch. 2]). I do not want to go into the arguments of the discussion this thesis has provoked, but rather follow Carroll (2009), D. Davies (2009), Irving (2009) and Thom (2009), who in one way or other argue that we do not need to discard the idea that a theatrical performance is of a certain written play in order to consider theatre as an independent art form.

258 | Frank Zipfel between the production world and the performance worlds of a production. As theatre performances are typically live enactments of a story, there will be (more or less important) differences between the performance worlds of different performances. The relationship between performance worlds and the production world can be conceived as a token-type relationship. Although the uniqueness of theatre performances is an important feature of theatre as an art form, it cannot be discussed here. What is more interesting when we consider the make-believe of the audience, is that, in Walton’s terms, one can say: Theatre performances/productions mandate imaginings; thereby a set of fictional truths is entailed and this set constitutes the fictional performance/production world. Further, what counts as fictional truth in a production world depends upon the principles of generation that are specific to the art form of theatre. These principles can be described by what Hamilton has developed in what he calls his feature-salience model (see Hamilton 2007, Ch. 6). This model tries to explain how spectators converge on the characteristics of what is going on in a performance, and can be understood as a way of explaining the particular way in which fictional truths and fictional worlds are generated in theatre. The problem related to the questions of how to establish what features of a production are relevant and of how these featured are to be interpreted in order to describe the fictional production world can be attributed to the fact that in theatre productions “iconic elements are combined with non-iconic ones that have no reflection in the fictional world” (Rozik 2002, 116). Thus, establishing what counts as a salient feature in a specific production depends upon conventions and rules that are located on different levels of abstraction: The overall (historically variable) conventions of theatre performances, the specific conventions that the theatrical style of the production entails and the even more specific rules that a particular production establishes. There are numerous striking examples of the dependency of fictional truth on theatrical conventions. In the following, I can only hint at a small number, taken from the works of Rozik and Hamilton. Specific techniques of representation which are applied in order to overcome aural or visual limitations of the audience in the theatre, e. g. speaking in a loud voice, strong make-up, or wearing cothurnus in ancient Greek theatre, differ considerably from everyday life behaviour. However, these differences are usually not taken into account when it comes to establishing what is part of the

Fictionality and Make-Believe in Drama, Theatre and Opera | 259

fictional world (cf. Rozik 1992, 11).30 Moreover, “actors may speak in alexandrines, but this does not entail that in the fictional world the characters are supposed to speak in this metre” (Rozik 2002, 116), and […] a landscape painted on a cyclorama does not entail that this is so in the fictional world; an actor may speak to himself in soliloquy, but this does not entail that the character speaks to himself in the fictional world; an actor may use a pink leotard in order to enact nakedness, but this does not mean that the character is not actually naked [...].” (Rozik 2002, 116)

So in order to determine what is a salient feature of a production and what is fictionally true in the production world, audiences must have some knowledge about the conventions of theatre as an art form and about specific styles of theatrical performance and they must recognize whether any particular rules are established by the production itself (cf. Hamilton 2007, Pt. III). To summarize, one can say that the fictionality of theatre can be described by the fact that the active make-believe of the actors induces a passive makebelieve by which the audience imagines a fictional world which is generated in the audience’s interpretation of what is presented in the theatrical performance according to the conventions of the theatrical production.31

4 The fictionality of opera Similar to the way the word drama can refer to a written play or a staged production of a play, the word opera can have two meanings: (1) A work of art that consists of the compilation of (literary) text and a musical score and which is meant to be represented in a theatrical form; and (2) theatrical productions of (1). In the second understanding, opera can be seen as “the staging of an encounter between music and theatre” (Campana 2012, 203) or simply as “staged sung drama” (Williams 2006, 1) or more precisely as “a multivalent art form [that] combines dramatic and literary traditions with vocal and instrumental music and the visual and plastic arts to tell a story” (Zeiss 2012, 179) or even as

|| 30 See also Elam (2002, 163–165) on syntactic orderliness, informational intensity, illocutionary purity and floor-apportionment control. 31 The distinction between active and passive here refers to the distinction between physical and mental activity. The make-believe of actors is active insofar as it encompasses physical activities, while the make-believe of the audience is passive insofar as it encompasses ‘only’ mental activities.

260 | Frank Zipfel “the combination of the dramatic + the narrative + the thematic + the verbal + the visual + the auditory” (Hutcheon and Hutcheon 1996, 5).32 Throughout the history of opera and opera criticism there have been major debates about which of the two most important elements of opera, i. e. text and music, has, or should have, precedence. While I cannot recount this sophisticated discussion about the aesthetics of opera here,33 it is interesting to note that it has often been argued that opera is much closer to the literary form of the novel than to that of drama. It is certainly true that there are quite a few arguments that can support the assumption that opera displays an affinity with the novel. One such similarity concerns the representation of time.34 In opera, time can be brought to a standstill in a similar way as in narrative texts, and in contrast to spoken theatre, quite a few arias are not actually sung monologues. In theatrical monologues, fictional time moves on while the character displays her thoughts and feelings. These thoughts and feelings actually often undergo important changes during the monologue. Arias, in contrast, often represent only a momentary emotional state which does not undergo any change during the aria. More often than not, an aria simply repeats and thereby amplifies the same emotion over and over again, for example in the very typical death moment arias. Also, in the operatic ensemble, time is usually brought to a stand-still as an ensemble is more often than not the representation of the momentary emotional state of several characters at the same time. Another similarity between opera and the novel concerns narrative presentation. In the aftermath of Wagner’s influential theory on Musiktheater, it has been argued that opera displays a narrator or narrative instance. Halliwell (1999), for example, claims that the orchestra in an opera corresponds to the narrator in a novel, and that this orchestra-narrator can take on the different roles of the literary narrator: “It is omniscient in that it is privy to all the thoughts of the characters it ‘creates’. It can, however, deliberately limit its scope and therefore function both as an authorial narrator and as a figural narrator (thus abandoning the possibility of omniscience), or somewhere on the continuum between the poles” (Halliwell 1999, 150). The orchestral accompa|| 32 Kerman’s definition “Opera is a type of drama whose integral existence is determined from point to point and in the whole by musical articulation” (Kerman 1988, 10) seems to encompass both meanings. See also Kivy’s view of opera as “drama-made-music” (Kivy 1988, Ch. 11) and Durante (1997). 33 See e. g. Kerman (1988, Ch. 1); Kivy (1988, Ch. 9); Scheit (1995); Hutcheon and Hutcheon (1996, 6–8); Zipfel (2007, 145–146); Pratt (2009); Morris (2012). 34 See e. g. Dahlhaus (1989); Williams (2006, 10); Zeiss (2012, 182–184).

Fictionality and Make-Believe in Drama, Theatre and Opera | 261

niment can thus take on the role of an authorial commentator and even the role of the organizer of the action, or it can recede into figural narration when for example the “orchestra closely identifies with one of the characters on stage by, for example, doubling the vocal line or the repetition of thematic material identified with that character” (Halliwell 1999, 151). On these grounds, Halliwell (1999, 135) argues that “opera is more musical novel than musical drama” (see Zeiss 2012, 189). The parallels that can be drawn between particular features of opera and particular features of the novel are certainly illuminating. Nonetheless, one can question the thesis that the orchestral score functions as a kind of a narrator on different grounds. There is of course the fact that it may be difficult to establish precisely what exactly the orchestra-narrator is saying as music is not a codified semantic system in the way language is.35 But even if we leave aside the problem of interpretation, the fact remains that text and music are “profoundly incompatible systems with different purposes and different means of achieving those purposes” (Corse 1987, 13), and that there are many different ways of combining literary and musical ideas (see Roberts Finlay 1986). Thus, text and music may not always be intended to add up to a homogenous work of art, and the role of the orchestra may not always be legitimately interpreted as some kind of presentation of or commentary on the story. Moreover, the conceptualization of the orchestra as narrator may hold for opera when regarded as musico-literary text (opera1), but when we look at opera as performed on stage (opera2), all the theatrical elements have to be considered in view of the presentation and narration of the story. The equivalent to a literary narrator would thus not be identical with the orchestra, but it would be the result of the complex interaction between all the elements that an opera performance is comprised of, i. e. the staging elements, especially singers as actors, costumes, setting, lighting etc., and the musical elements, especially the orchestral music and the voices of the singers. Therefore, explaining opera2 as a musical novel is unsatisfying because the parallel reaches both too far and not far enough: The analogy between opera and literary narrative is overstressed and, more importantly, the theatrical components of an opera performance cannot be explained by means of this analogy. Opera2 is after all a theatrical art form (see Williams 2012), and thus, it can be described by means of a slightly altered version of the minimal definition of

|| 35 Or as Kivy puts it: “Music is a self-sufficient pattern of sound with ‘syntactic’ but not ‘semantic’ properties” (Kivy 1988, 57).

262 | Frank Zipfel theatre performance mentioned in section 2: A impersonates B by means of singing with orchestral accompaniment while C looks on. The question that interests us in the present context is whether there is a specific kind of make-believe related to opera performances. At an abstract level, the investigation about what is considered as fictional truth in the performance world of an opera production can be led in the same way as for theatrical performances; that is, by applying Hamilton’s feature salience model.36 On a practical level, however, the principles of generation will differ considerably because the conventions that establish whether a specific feature is to be considered as salient are different. All theater is built on performing conventions, some more distant than others from a direct simulacrum of life-as-we-live-it. [...] Whatever ineffable truths about life an opera production may convey, in its onstage action it is considerably farther from life-as-we-liveit, considerably less ‘realistic’ and more ‘theatrical’ in its conventions than most other forms of drama. (Littlejohn [1992, 33])

The most obvious, but to a certain extent also the most odd, conventional feature of opera as opposed to spoken theatre is of course the orchestral accompaniment and the singing. It has to be noted though, that the musical component of an opera performance or production is not part of the performance /production world. It is not fictionally true in the performance world that the actions are accompanied by orchestral music or that the characters sing, unless it is part of the story that characters actually present songs, such as in the second act of Tannhäuser. Rather, the audience is invited to imagine that the characters in the performance world act, speak and sometimes utter their private thoughts with no music involved in all this, while they are hearing the orchestra play and the singers sing. Moreover, the way the orchestra executes the score and the specific voices of the singers are part of the performance world because the differences in timbre, tempo or accent that occur from one production to another inevitably influence the way in which actions and characters of the fictional world are perceived. So opera as an art form is based on a paradoxical || 36 In some way it is true that the text of an opera “prescribes the actuality of the performance (pitch, volume, pace, rhythm, duration) much more exactly than the merely verbal text of drama” (Bennett 1985, 9). Nonetheless I do not agree with Bennett’s conclusion that a “performance of an opera is thus more a realization of the text than an interpretation”, or with his claim that “whereas the various performances of a drama, as interpretations, are felt to diverge from the basic understood text, in opera, by comparison, we have the sense of various performances converging toward the sort of ideal realization that is not ordinarily inferable from a verbal text alone” (Bennett 1985, 9).

Fictionality and Make-Believe in Drama, Theatre and Opera | 263

set of conventions which entail that in their make-believe, audiences are to ignore the fact that there is music, but nevertheless acknowledge the way in which the musical elements are executed. This paradoxical situation seems to be responsible for the fact that the conventions of opera are less realistic than the ones of most of the forms of spoken theatre. This has a certain bearing on different features of the make-believe that entail the performance world. One example: The fact that the soprano singing the title role in Verdi’s La Traviata is a rather corpulent prima donna would not necessarily be regarded as an indication that it is a fictional truth about the character of Violetta that she is obese. When, however, the character of Marguerite in the play Dumas himself extracted from his novel La Dame au Camélia is played by a very corpulent actress it would be legitimate to ask whether it is fictionally true in the performance world that Marguerite suffers from some kind of eating disorder.

5 Concluding remarks What I present here is only a first draft of an account of fictionality in drama, theatre and opera. A lot more would have to be said in order to flesh out the specificities of fictionality in these particular art forms. It can be noted, however, that the fictionality of all three art forms involves make-believe, fictional worlds and institutional practices. The common feature of the three art forms lies in the fact that it is part of the respective institutional practices that the works are to be used as props in games of make-believe, i. e. that they are regarded as mandating specific imaginings. What imaginings specific works mandate depends in part upon their particular features, but also, and crucially so, on the conventions of the institutional practices that govern the art form in question. Moreover, there is make-believe involved on the production side in the art forms that are based on live performances (theatre and opera). It has to be noted however, that the concept of make-believe involved in the theatrical practice of acting and singing/acting is slightly different from the one that applies to audiences. The differences in the understanding of make-believe in the particular elaborations on fictionality follow from the differences in the institutional practices that underlie the different art forms. Nonetheless, the depicted similarities and dissimilarities in the fictionality of drama, theatre and opera show that the multilayered concept of fictionality enables us to explain and discuss the fictionality of differing art forms. This concept has the dual benefit of both highlighting what these art forms have in common and how they differ.

264 | Frank Zipfel More specifically, it shows how they function as fictional works, but it also grasps the differences between the differing art forms with regard to their fictionality.

References Alward, Peter. “Leave Me out of It. De Re, but Not De Se. Imaginative Engagement with Fiction.” The Journal of Aesthetics and Art Criticism 64.4 (2006): 451–459. Alward, Peter. “Onstage Illocution.” The Journal of Aesthetics and Art Criticism 67 (2009): 321– 331. Alward, Peter. “Word-Sculpture, Speech Acts and Fictionality.” The Journal of Aesthetics and Art Criticism 68.4 (2010): 389–399. Bareis, J. Alexander. Fiktionales Erzählen. Zur Theorie der Literarischen Fiktion als MakeBelieve. Göteborg: Acta Universitatis, 2008. Bareis, J. Alexander. “Fiktionen als Make-Believe.” Fiktionalität Ein interdisziplinäres Handbuch. Eds. Tobias Klauk and Tilmann Köppe. Berlin/Boston: De Gruyter 2014, 50–67. Bennett, Benjamin K. “Cinema, Theater and Opera: Modern Drama as Ceremony.” Modern Drama 28.1 (1985): 1–21. Bentley, Eric. The Life of Drama. New York: Atheneum, 1964. Birkenhauer, Theresia. “Fiktion.” Metzler Lexikon Theatertheorie. Eds. Erika Fischer-Lichte, Doris Kolesch and Matthias Warstat. Stuttgart: Metzler 2005, 107–109. Campana, Alessandra. “Genre and Poetics.” The Cambridge Companion to Opera Studies. Ed. Nicholas Till. Cambridge: Cambridge University Press, 2012, 202–224. Carroll, Noël. The Philosophy of Horror or Paradoxes of the Heart. New York: Routledge, 1990. Carroll, Noël. “Fiction, Non-Fiction, and the Film of Presumptive Assertion. A Conceptual Analysis.” Film Theory and Philosophy. Eds. Richard Allen and Murray Smith. Oxford: Clarendon, 1997, 173–202. Carroll, Noël. “Philosophy and Drama. Performance, Interpretation, and Intentionality.” Staging Philosophy. Intersections of Theater, Performance and Philosophy. Eds. David Krasner and David Z. Saltz. Ann Arbor: The University of Michigan Press, 2006, 104–121. Carroll, Noël. “Basic Theatrical Understanding. Considerations for James Hamilton.” The Journal of Aesthetic Education 43.3 (2009): 15–22. Cohn, Dorrit. “Signposts of Fictionality. A Narratological Perspective.” Poetics Today 11 (1990): 775–804. Cohn, Dorrit. The Distinction of Fiction. Baltimore: Johns Hopkins University Press, 1999. Corse, Sandra. Opera and the Uses of Language. Mozart, Verdi, Britten. Rutherford, N.J.: Fairleigh Dickinson University Press, 1987. Crease, Robert L. and John Lutterbie. “Technique.” Staging Philosophy. Intersections of Theater, Performance and Philosophy. Eds. David Krasner and David Z. Saltz. Ann Arbor: The University of Michigan Press 2006, 160–179. Currie, Gregory. The Nature of Fiction. Cambridge: Cambridge University Press, 1990. Currie, Gregory. Narratives and Narrators. A Philosophy of Stories. Oxford: Oxford University Press, 2010.

Fictionality and Make-Believe in Drama, Theatre and Opera | 265 Dahlhaus, Carl. “Zeitstrukturen in der Oper.” Vom Musikdrama zur Literaturoper. Aufsätze zur neueren Operngeschichte. München: Piper 1989, 27–40. Davies, David. “Rehearsal and Hamilton’s ‘Ingredients Model’ of Theatrical Performance.” The Journal of Aesthetic Education 43.3 (2009): 23–36. Davies, Stephen. “Responding Emotionally to Fictions.” The Journal of Aesthetics and Art Criticism 67 (2009): 269–284. Elam, Keir. The Semiotics of Theatre and Drama. 2nd ed. London: Routledge, 2002. Erickson, Jon. “Presence.” Staging Philosophy. Intersections of Theater, Performance and Philosophy. Eds. David Krasner and David Z. Saltz. Ann Arbor: The University of Michigan Press, 2006, 142–159. Fischer-Lichte, Erika. Semiotik des Theaters 1. Das System theatralischer Zeichen. Tübingen: Narr, 1983. Fischer-Lichte, Erika: Theaterwissenschaft. Eine Einführung in die Grundlagen des Faches. Tübingen/Basel: A. Fracke, 2010. Fludernik, Monika. “Narrative and Drama.” Theorizing Narrativity. Eds. John Pier and José Angel García Landa. Berlin: Walter de Gruyter, 2008, 355–383. Fortier, Mark. Theory/Theatre. An introduction. 2nd ed. London: Routledge, 2002. Gaut, Berys. “The Philosophy of the Movies: Cinematic Narration.” The Blackwell Guide to Aesthetics. Ed. Peter Kivy. Malden MA: Blackwell, 2004, 230–253. Gibson, John. Fiction and the Weave of Life. Oxford: Oxford University Press, 2007. Gorman, David. “Theories of Fiction.” Routledge Encyclopedia of Narrative Theory. Eds. David Herman, Manfred Jahn, and Marie-Laure Ryan. London: Routledge, 2005: 163–167. Halliwell, Michael. “Narrative Elements in Opera.” Word and Music Studies. Defining the Field. Eds. Walter Bernhart, Steven Paul Scher and Werner Wolf. Amsterdam: Rodopi, 1999, 135– 153. Hamilton, James R. The Art of Theater. Malden, MA: Blackwell, 2007. Hamilton, James R. “Replies to Criticism.” The Journal of Aesthetic Education 43.3 (2009): 80– 106. Hutcheon, Linda and Michael Hutcheon. Opera: Desire, Disease, and Death. Lincoln: University of Nebraska Press, 1996. Irving, Sherri. “Theatrical Perfomance and the Works Performed.” The Journal of Aesthetic Education 43.3 (2009): 37–50. Issacharoff, Michael. “How Playscripts Refer. Some Preliminary Considerations.” On Referring in Literature. Eds. Anna Whiteside and Michael Issacharoff. Bloomington: Indiana University Press, 1987 84–94. Jaeger, Suzanne M. “Embodiment and Presence. The Ontology of Presence Reconsidered.” Staging Philosophy. Intersections of Theater, Performance and Philosophy. Eds. David Krasner and David Z. Saltz. Ann Arbor: The University of Michigan Press, 2006, 122–141. Jannidis, Fotis. “Narratology and the Narrative.” What is Narratology? Eds. Tom Kindt and Hans-Harald Müller. Berlin: Walter de Gruyter, 2003, 35–54. Kania, Andrew. “Against them, Too: A Reply to Alward.” The Journal of Aesthetics and Art Criticism 65.4 (2007): 404–408. Kerman, Joseph. Opera as Drama. New and Revised Edition. Berkeley: University of California Press, 1988. Kivy, Peter. Osmin’s Rage. Philosophical Reflections on Opera, Drama, and Text. Princeton: Princeton University Press, 1988.

266 | Frank Zipfel Korthals, Holger. Zwischen Drama und Erzählung. Ein Beitrag zur Theorie geschehensdarstellender Literatur. Berlin: Erich Schmidt, 2003. Köppe, Tilmann and Jan Stühring. “Against Pan-narrator Theories.” Journal of Literary Semantics 40.1 (2011): 59–80. Köppe, Tilmann and Jan Stühring. “Against Pragmatic Arguments for Pan-Narrator Theories: The Case of Hawthorne’s Rappaccini’s Daughter.” Author and Narrator: Transdisciplinary Contributions to a Narratological Debate. Eds. Dorothee Birke and Tilmann Köppe. Berlin/Boston: De Gruyter 2015, 13–44. Lamarque, Peter, and Stein Haugom Olsen. Truth, Fiction, and Literature. A Philosophical Perspective. Oxford: Clarendon, 1994. Limon, Jerzy. “Theatre’s Fitfh Dimension. Time and Fictionality.” Poetica 41 (2009): 33–54. Litteljohn, David. The Ultimate Art. Essays Around and About Opera. Berkeley: University of California Press, 1992. Meskin, Aaron. “Scrutinizing the Art of Theater.” The Journal of Aesthetic Education 43 (2009): 51–66. Morris, Christopher. “‘Too Much Music’: The Media of Opera.” The Cambridge Companion to Opera Studies. Ed. Nicholas Till. Cambridge: Cambridge University Press, 2012, 95–116. New, Christopher. Philosophy of Literature. An Introduction. London: Routledge, 1999. Nünning, Ansgar and Roy Sommer. “Diegetic and Mimetic Narrativity: Some Further Steps towards a Narratology of Drama.” Theorizing Narrativity. Eds. John Pier and José García Landa. Berlin: Walter de Gruyter, 2008, 331–354. Osipovich, David. “What is a Theatrical Performance?” The Journal of Aesthetics and Art Criticism 64.4 (2006): 461–470. Pratt, Scott L. “Opera as Experience.” The Journal of Aesthetic Education 43.4 (2009): 74–87. Roberts Finlay, Carolyn. “Structural Pragmatics. A Semiotic Approach to the Opera Text.” Yearbook of Comparative and General Literature 35 (1986): 25–39. Rozik, Eli. “Theatrical Conventions: A Semiotic Approach.” Semiotica 89.1 (1992): 1–12. Rozik, Eli. “Acting. The Quintessence of Theatricality.” SubStance 31.2/3 (2002): 110–124. Rozik, Eli. “The Homogeneous Nature of the Theatre Medium.” Semiotica 168.1 (2008): 169– 190. Rozik, Eli. Fictional Thinking. A Poetics and Rhetoric of Fictional Creativity in Theatre. Brighton: Sussex Academic Press, 2009. Ryan, Marie-Laure. “Transfictionality across Media.” Theorizing Narrativity. Eds. John Pier and José García Landa. Berlin: Walter de Gruyter, 2008, 385–417. Sainsbury, R. M. Fiction and Fictionalism. London, New York: Routledge 2010. Saltz, David Z. “How to Do Things on Stage.” The Journal of Aesthetics and Art Criticism 49.1 (1991): 31–45. Saltz, David Z. “Infiction and Outfiction. The Role of Fiction in Theatrical Performance.” Eds. David Krasnern and David Z. Saltz. Staging Philosophy. Intersections of Theater, Perfomance and Philosophy. Ann Arbor: Michigan University Press, 2006, 203–218. Scheit, Gerhard. “Die Oper als Gesamtkunstwerk.” Literatur Intermedial: Musik – Malerei – Photographie – Film. Ed. Peter V. Zima. Darmstadt: Wissenschaftliche Buchgesellschaft 1995, 93–125. Schenk-Haupt, Stefan. “Narrativity in Dramatic Writing. Towards a General Theory of Genres.” Anglistik 18.1 (2007): 25–42. Scruton, Roger. Art and Imagination. A Study in Philosophy of Mind. London: Methuen, 1974.

Fictionality and Make-Believe in Drama, Theatre and Opera | 267 Scruton, Roger. “Feeling Fictions.” A Companion to the Philosophy of Literature. Eds. Garry L. Hagberg and Walter Jost: Malden/MA: Wiley-Blackwell, 2010, 93–105. Searle, John R. “The Logical Status of Fictional Discourse.” New Literary History 6.2 (1975): 319–332. Searle, John R. Expression and Meaning. Studies in the Theory of Speech Acts. Cambridge: Cambridge University Press, 1979. Segre, Cesare. “A Contribution to the Semiotics of Theatre.” Poetics Today 1.3 (1980): 39–48. Stevenson, Leslie. “Twelve Conceptions of Imagination.” British Journal of Aesthetics 43.3 (2003): 238–259. Thom, Paul. “Works, Pieces, and Objects Performed.” The Journal of Aesthetic Education 43.3 (2009): 67–79. Thomson-Jones, Katherine. “The Literary Origins of the Cinematic Narrator.” British Journal of Aesthetics 47.1 (2007): 76–94. Thon, Jan-Noël. “Toward a Transmedial Narratology. On Narrators in Contemporary Graphic Novels, Feature Films, and Computer Games.” Beyond Classical Narration. Unnatural and Transmedial Challenges. Eds. Jan Alber and Per Krogh Hansen. Berlin/Boston: de Gruyter 2014, 25–56. Walsh, Richard. The Rhetoric of Fictionality. Narrative Theory and the Idea of Fiction. Columbus: Ohio State University Press, 2007. Walton, Kendall L. Mimesis as Make-Believe. On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Weidle, Roland. “Organizing the Perspectives: Focalization and the Superordinate Narrative System in Drama and Theater.” Point of view, Perspektive, and Focalization. Modeling Mediation in Narrative. Eds. Peter Hühn, Wolf Schmid and Jörg Schönert. Berlin: De Gruyter, 2009, 221–242. Williams, Bernhard. On Opera. New Haven: Yale University Press, 2006. Williams, Simon. “Opera and Modes of Theatrical Production.” The Cambridge Companion to Opera Studies. Ed. Nicholas Till. Cambridge: Cambridge University Press, 2012, 139–158. Wilson, Geoge M. “Elusive Narrators in Literature and Film.” Philosophical Studies 135.1 (2007): 73–88. Wolterstorff, Nicholas. Works and Worlds of Art. Oxford: Clarendon, 1980. Zeiss, Laurel E. “The dramaturgy of opera.” The Cambridge Companion to Opera Studies. Ed. Nicholas Till. Cambridge: Cambridge University Press, 2012, 179–201. Zipfel, Frank. Fiktion, Fiktivität, Fiktionalität. Analysen zur Fiktion in der Literatur und zum Fiktionsbegriff in der Literaturwissenschaft. Berlin: Erich Schmidt, 2001. Zipfel, Frank. “Erzählter Gesang: Literarische Opernreflexionen.” Literarische Medienreflexionen. Künste und Medien im Fokus Moderner und Postmoderner Literatur. Eds. Sandra Poppe and Sascha Seiler. Berlin: Erich Schmidt, 2007, 137–154. Zipfel, Frank. “Fictional Truth and Unreliable Narration.” Journal of Literary Theory 5.1 (2011): 109–130. Zipfel, Frank. “Emotion und Fiktion. Zur Relevanz des Fiktions-Paradoxes für eine Theorie der Emotionalisierung in Literatur und Film.” Emotionen in Literatur und Film. Ed. Sandra Poppe. Würzburg: Könighaus & Neumann, 2012, 127–153. Zipfel, Frank. “Imagination, fiktive Welten und fiktionale Wahrheit. Zu Theorien fiktionsspezifischer Rezeption von literarischen Texten.“ Fiktion, Wahrheit, Interpretation. Philologische und philosophische Perspektiven. Eds. Eva-Maria Konrad, Thomas Petraschka, Jürgen Daiber and Hans Rott. Münster: Mentis 2013, 38–64.

268 | Frank Zipfel Zipfel, Frank. “Fiction across Media: Towards a Transmedial Concept of Fictionality.” Storyworlds across Media. Toward a Media-Conscious Narratology. Eds. Marie-Laure Ryan and Jan Thon. Lincoln/London: University of Nebraska Press 2014: 103–124. Zipfel, Frank. “Narratorless Narration? Some Reflections on the Arguments For and Against the Ubiquity of Narrators in Fictional Narration.” Author and Narrator: Transdisciplinary Contributions to a Narratological Debate. Eds. Dorothee Birke and Tilmann Köppe. Berlin/Boston: De Gruyter 2015, 15–80. Zipfel, Frank. “Ein institutionelles Konzept der Fiktion – aus einer transmedialen Perspektive. Überlegungen zur Fiktionalität von literarischer Erzählung und theatraler Darstellung” (forthcoming).

Matthew DeCoursey

Making Meaning in the Theatre: Double Noesis The experience of an actor in a play is aesthetic in nature. In one way, this statement seems too obvious to need demonstration: What is an actor involved with if not the aesthetic? In another way, it is not obvious at all. From Kant onwards, aesthetics has been discussed in terms of the experience of an audience, not in terms of an artist of whatever kind. Some, following Wolfgang Iser’s work on literature (1978), have made a sharp distinction between the artistic as the domain of the artist and the aesthetic as the domain of the audience (see Jackson 2007, 37). Yet theatre brings out the interaction of aesthetic appreciation and artistic endeavour. In theatre, there is generally a person whose job it is to appreciate the artistic merits of actors’ performance and make changes based on what she sees: that is, the director. One should not, however, think of directors as always dictators, moving actors around like chessmen: Both aesthetic appreciation and artistic effort often take place in the social space between the director and the actor. That social space may be larger, of course, including other actors, the designer, and so on. This paper is primarily concerned with the aesthetic experience of the actor – but we shall be aware throughout that this aesthetic experience is built in a process of rehearsal, in interaction with a director, a directorial team, or an entire collective. Let us begin with the experience of the actor at the moment of performance. What I mean is this: You are an actor. You are onstage. You are standing in the right place to make the impact that the director and the cast want, as developed in rehearsal. You are listening to a speech by another actor. When the speech is over, you must leave a pause of just the right length and deliver your own lines, in the awareness of what the play is about, and how your lines will contribute to the overall impact of the play. The next actor needs to interrupt you. You have a few extra words ready in case the next actor is late on his cue. You are aware of where each actor is standing, how this tableau relates to the story as a whole. You know where you will exit, at what point, and how you should walk when your cue comes. Clearly, I am writing here about a play that is performed from a text, and I will assume a text with a clear, classic dramatic structure: complication-crisis-dénouement. It is rehearsed, rather than improvised, and rehearsed in a meticulous way: All pauses are timed, things have been tried, decisions made about how to perform. This discussion will assume a good production, in which all actors are aware of, and invested in, artistic decisions as to the way they will play their parts. This is not the only form of theatre, but it is an espe-

270 | Matthew DeCoursey cially important kind, because it is so meaningful to so many people in the world. In this article, I will show that a phenomenological view of time and of the production of meaning can provide a sound basis for characterizing the experience of the actor as aesthetic. This paper can be thought of as a commentary on a distinction made by Kendall Walton in his 1978 paper Fearing Fictions. In that paper, he distinguished between really fearing and make-believedly fearing. He suggested that this distinction satisfactorily accounts for the coexistence of contradictory feelings in spectators at a play or a film. In this paper, I will construe this distinction in terms of the creation of meaning by consciousness, and in the end suggest that two distinct egos exist for a time in the mind of the aesthetic experiencer. In the process, I will also seek to place the experience of makebelieve in the context of theatre into the framework of phenomenological aesthetics by connecting the activity of make-believe with the special experience of time set out by Mikel Dufrenne (1966). This paper will postulate the existence of an aesthetic object. That is, theatrical performance creates a ‘performance text’, distinct at once from the written text and from the private responses of each theatregoer or actor. The plot of a play constitutes an autonomous aspect of a larger aesthetic object. James Hamilton (2007, Ch. 5) has argued recently in The Art of Theater that it is possible to identify a “basic theatrical understanding”. He argues, very plausibly, that it is possible to identify certain aspects of a play as salient: That is, one must understand these aspects in the same way as others in order to be able to say that one has experienced the same play. He calls his approach feature-salience analysis, remarking, Features of a performer are salient to a spectator for an act or set of facts just when the learner-spectator, under a suitable common knowledge requirement, can notice those features as regularities in the behavior of the performer and when the learner-spectator concludes (1) that some pattern – and hence some set of facts – obtains, (2) that whenever those features appear in the same context then the same set of facts obtains, and (3) that every other learner-spectator will conclude both (1) and (2). (Hamilton 2007, 95)

Hamilton’s stress on the social nature of the theatrical experience is valuable, though in the fashion of an analytic philosopher, he supposes that it is necessary for spectators to reason out what others will think, whereas the experience of theatre does not seem so involved. Hamilton does not approach the question of the aesthetic object in his book, but it seems apparent that even a basic theatrical understanding must be aesthetic in nature. In a play with a classic dramatic structure, directors often use the structure to give shape to the experience at once of the audience and of the

Making Meaning in the Theatre: Double Noesis | 271

actor. That dramatic structure is one of the key elements of theatrical aesthetics.1 In Macbeth, for example, the whole structure of the play turns around the murder of Duncan: Every scene in the earlier part of the play looks toward the future murder of Duncan (“When you durst do it, then you were a man…”, 1.7.49), and the rest of the play plays out the consequences of the murder (“I am in blood / Stepp’d in so far that, should I wade no more, / Returning were as tedious as go o’er”, 3.4.135–137). Surely no one would say that a spectator had understood the play who had not realized that Duncan’s murder is first planned, then executed, then consequences are played out. In phenomenological aesthetics, the process of prediction and fulfillment has a very deep role, as David Lewin showed with respect to music in a famous article (1986). If we translate Hamilton’s basic theatrical understanding into the language of the phenomenological aesthetician Mikel Dufrenne (1966), we find that ‘basic understanding’ allows access to expressed time, which has narrative structure, as opposed to the represented time which mimics the time of the transcendent world, the time of the clock. Dufrenne takes Macbeth as an example: [D]uration is a function of the way in which the represented characters live time. Because Macbeth is fascinated, then corrupted, by crime, his will hurls him toward his own destruction; the fall of a soul into the snare of fatality is thus a movement uniformly accelerated. It is Macbeth’s time – that of the evil project which he forms and of which he is captive, caught in its trap like the consciousness of a dreamer – which guides the theatrical time. (Dufrenne 1966, 140)

For Dufrenne, it is in the nature of the aesthetic object to provide a movement from ‘represented time’ to ‘expressed time’. ‘Basic theatrical understanding’, then, is aesthetic in nature. Dufrenne’s notion of expressed time has received support from an unexpected quarter. The psychologist Daniel Kahneman has found that people evaluate experiences not in the manner of represented time but in the manner of expressed time. In an experiment, he asked subjects to put their hand into painfully cold water for sixty seconds. Then, using the other hand, they had a second experience: ninety seconds of pain, with the first sixty identical, and last || 1 This assertion works especially well with John Dewey’s aesthetics. His insistence that the aesthetic is in an experience, a bounded experience with a shape, can easily refer to narrative structure (Dewey 1934, 35). Other views of aesthetics, such as those of Kant, stress the form of the artistic work, the interaction of parts (Kant 2000, 109). In theatre, there are a lot of parts to connect, but it seems clear from much theatrical practice that dramatic structure is often a central element.

272 | Matthew DeCoursey thirty less painful. Asked which of the two they wished to repeat, 80% of the subjects chose the second option. They knew the pain had lasted longer, but still preferred a story with a happier ending (Kahneman 2011 381–384). From this and other evidence, Kahneman draws two rules about remembered human experience: (1) Peak-end rule: The global retrospective rating was well predicted by the average of the level of pain reported at the worst moment of the experience and its end. (2) Duration neglect: The duration of the procedure had no effect whatsoever on the ratings of total pain. (Kahneman 2011, 380)

Dufrenne views expressed time as marking literary aesthetics; Kahneman sees it as what constructs the memory of an episode. They have in common an assertion that time as measured by the clock is trivial in their two different contexts. Kahneman observes something in the way people experience and remember. Dufrenne shows how expressed time is part of the basic material for the construction of literary art – including plays. Kahneman postulates ‘two selves’, one which experiences from moment to moment, and which he thinks must experience in represented time. The remembering self looks back on experience in expressed time. He complicates he picture, however, by pointing out that when people take vacations, they choose experiences that they think will be ‘memorable’, and that holidaymakers make choices based on the remembering self rather than the experiencing self (Kahneman 2011, 386–390). People take photos during their holidays, because they are looking to remember. Kahneman does not conclude, though it seems reasonable, that his ‘remembering self’ is also part of the experiencing self on the moment. Further, when performing in, or watching, a play, people seem to experience expressed time on the moment, not only in retrospect. That means that the two ‘selves’ are not as distinct as Kahneman suggests. These observations give us some further knowledge of the aesthetic object. It is made up of expressed, that is, structured, time, and has relatively little to do with represented time, or time by the clock. But is the object so described perceptible to both the actor and the audience equally? One might think not. The actor sees not only the events on the stage, but backstage too. The actor will probably miss parts of the play while changing costumes, or may have to go to the green room because the stage manager does not want the whole cast in the wings during performance. The experience of the actor may be broken up with small irrelevancies. Her husband has sent a text message to a silenced phone. She pauses to study for an exam during a long break between appearances onstage.

Making Meaning in the Theatre: Double Noesis | 273

Despite the differences in direct perception, the actor may experience the aesthetic object no less than does the spectator. Where an actor understands the need to give a pause a certain length, it is because of a dramatic moment that is made so by its place in the plot, as well as by the immediate arrangement of actors, speed of speaking, and the many other circumstances made by small decisions in a theatrical production. A well-prepared actor is aware of the location of this pause in the play as a whole, its significance in terms of the lines around it, and what actors are looking where, and standing how. The actor remembers everything that has already happened at this point in the play and looks forward to what will happen in the rest of the play. This location in the time of the plot, in expressed time, creates the context for a small artistic decision: How long must this pause be? The actor, in order to perform well, must be intensely, intuitively aware of the aesthetic object in all its significance. The aesthetic object has been thought through consciously in rehearsal, but by performance time has become an intuitive reality to the actor. Actors have been part of the decision-making, and usually cannot perform well unless they know. Also, unforeseen events on the stage may create the need for a new aesthetic decision on the fly, and well-prepared actors can make such decisions successfully. It is true that the actor may not be present for the whole performance, may sit in the green room until called by the stage manager. This is immaterial for an actor who has been present through rehearsals and is familiar with the play as a whole. The actor looks back over a structure made familiar in the course of rehearsals. To watch and listen again is at most to be refreshed in memory. The notion of expressed time refers to Husserl’s assertion that human beings do not experience time as it is measured by the clock. Husserl, using music to make his point, asserted that real human beings incorporate memories of the past (retention) and anticipations of the future (protention) as well as immediate experience of the present moment, and that the phenomenal experience of the present moment is never independent of retention and protention (see Smith 2007, 210–217). Paul Ricœur (1980), following Husserl, made a parallel point about literature, arguing that for human beings, time is created by narrative. If Ricœur is right, this is something that is always true in some way, but it is when we perform in theatre that we are most aware of the fact. Performing in a play constitutes, then, a special experience of time, an experience in which we are very fully aware of what precedes each moment and what will follow it. Ricœur, following St. Augustine, calls the effect that this has on consciousness distentio animi, a stretching of the soul. By this he means that the perception of passing time requires the activity of the mind, and Augustine and Ricœur equally find this activity most fully in singing a song, because at every moment, one remembers the earlier part of the song and looks forward to an end already known. The

274 | Matthew DeCoursey experience of performance, then, stretches the soul, and we may suggest that it does so more for the performers than for the audience, as they are more actively engaged in the creation of the soul-stretching experience. Theatre also brings with it a doubleness of perception, similar to distentio animi in that both require an effort of the will. An actor must necessarily be constantly aware of the real-world nature of what surrounds him or her. Actors are not allowed to walk off the edge of stages, or to leave the stage by the wrong exit. Actors are trained never to forget that there is an audience watching them. That is why they turn their backs to the audience only on purpose, never by accident. They are aware of the audience at every moment of their presence onstage. At the same time, the significance imposed upon real-world objects and events must have reality for them so that they can respond appropriately. Perhaps in a given production, the daggers of the murder are not really daggers, but dowels painted black. ‘Macbeth’ and ‘Lady Macbeth’ must treat them as daggers in so far as they are parts of the fiction, but must remember that they are dowels in so far as they are objects in the real world. Wooden dowels may be much lighter than steel daggers, but the actors must focus, and treat them as if they had the weight of steel. Dowels are the same at the two ends, but actors must focus and recognize that in the story, the two ends are different. The blood dripping from them is of great importance to the play, but in many productions, there is no real blood on the daggers, even if they are shaped as daggers. The actors must create the image of the blood which is invisible by treating the daggers as if they were bloody. The same applies to people: In real life, ‘Macbeth’ and ‘Lady Macbeth’ may have a relationship based on laughing and wit, but that relationship should not intrude on this performance – unless its intrusion fits into the aesthetic object, modifying it in a positive way. Hamilton calls this double focus, but appears uncomfortable with the notion that the two kinds of focus appear at the same time for spectators. He says that spectators find “their attention going back and forth between” the awareness of the actor as a real-life individual and the actor as a character (Hamilton 2007,103). He sees it, then, as similar to the duck-rabbit picture discussed by Wittgenstein: We can see one at a time but not both together.2 When we look at actors, however, the simultaneous doubleness of focus is inevitable. One cannot forget either the real nature or the fictional nature of the dowels. The actor also need not forget that other actors are people she knows. This awareness does not

|| 2 See Wittgenstein (1953 [1997], Part 2, Par. 11).

Making Meaning in the Theatre: Double Noesis | 275

prevent the actor from playing her part, from seeing other actors also as their characters.3 Recent research by Daniel Kahneman and other associated with him casts light on this double perception. He separates System 1, or intuitive perception, from System 2, or fully conscious, connected thought. He takes the words ‘bananas’ and ‘vomit’ and sets them side by side (Kahneman 2011, 50). He asserts, and is able to show from research, that there is an attenuated impact on us, as if we had experienced vomiting a banana. In the scene from Macbeth, the idea of a dagger and the physical presence of a dagger are brought together in the perception at once of the actors and of the audience. The word ‘blood’ has already occurred many times in the play, often placed in the context of murder. This includes the blood on the imaginary dagger in Act 2, Scene 1. The blood is present in our minds, present to our emotions, when we look at the dowel. This experience, however, is not attenuated. It is heightened. Kahneman speaks of the impact of two words placed together, and of the power of ‘priming’ on people’s interpretations and perceptions. This perception goes beyond what Kahneman discusses, to something powerful, perhaps capable of changing lives. If ‘banana’ and ‘vomit’ together produce an emotional response, the mere production of emotional response is not the result of artistry, or perhaps we might say, is only the result of a very low level of artistry, only what is required to bring these two words together. In Macbeth, the transformation of the dowels creates a very strong effect. We might, in simple terms, say that the technique of the actors creates the heightening. The actors must avoid basic mistakes: Mumbling, turning backs to the audience, allowing faces to go blank between lines. Beyond that, actors must appear to feel the emotions of the character and allow the emotions of the character to affect the audience. Emotion seems to be somehow involved in the phenomenon of double noesis. We have already used Husserl’s ideas in characterizing the actor’s experience of time. We can better characterize the actor’s experience of objects on the stage with reference to Husserl’s notion of noesis combined with Kendall Walton’s understanding of make-believe. Husserl’s noesis serves to characterize our way of making sense of things in our environment. Noesis is the attribution of meaning to an object as perceived (see Smith 2007, 258–260). The perception of the object in the consciousness is the noema. The process of noesis is the result of operative or act intentionality. || 3 Julia Walker (2006, 36–39) characterizes this experience as simultaneous, using a nineteenth-century account of a theatrical performance together with her own experience.

276 | Matthew DeCoursey In operative intentionality, we perceive involuntarily; in act intentionality, we turn our attention toward an object, and as a result, we constitute it somehow. In normal daily life, we try to constitute objects to match their real nature: When we drive, for example, we see an object, and we want to constitute it as solid (threatening the car and occupants) or as insubstantial in real-world terms. If a cardboard box blows onto the road in front of my car, I want to know if I can drive right over it or not. A perceived object, for Husserl, has a horizon: It has potential for use (see Smith 2007, 300–304). If I see a chair, it is important to know if it is sittable, or if it will collapse under my weight. Husserl’s noesis has an important advantage over Hamilton’s understanding of double focus: It is nonverbal and instantaneous. For Hamilton spectators must be constantly reasoning about what other spectators are likely to think about what they see. Kahneman’s research is again relevant here, because he sees System 1 as operating automatically, without any need to ask our permission (2011, 50–58). He points out that if we see a word written in a language we understand, we cannot avoid understanding it. Further, we cannot avoid connecting the word with our network of associations, as the reader of ‘banana’ and ‘vomit’ must, of necessity, connect the two. In this example, the theatre company builds that network of associations in rehearsal. The intuitive awareness of dramatic structure and symbolic connections, by performance time, has that same inevitability for the actor. In performance, there is little need to think through the connections in any conscious way. The actor must necessarily constitute the stage, the backstage, the other actors, the stagehands and the audience in a natural manner. Even if an actor is playing a ghost, another actor cannot walk through him. A character may fictionally be whispering, but the people in the back row still need to hear her. The actor must also constitute other actors, props and so on in conformity with the fiction. There are therefore two separate processes of noesis going on at the same time. The constitution of other actors, props and so on is a process of makebelieve, developed by Kendall Walton in his book Mimesis as Make-Believe (1990). He suggested that make-believe often involves a process of rule-making. In one of his examples, two boys playing in the woods decide that all tree stumps are bears. They run away from bears or hunt them, as the case may be. The preparation of a play is considerably a process of creating such rules, the first of which is that the actors are not themselves, but it is fictional, in Walton’s terminology, that they are someone else. Such rules may make something insubstantial, perhaps made up of light, into a solid object. They may make two dowels into daggers. The fictionalized objects then have horizons that differ from the horizons of real objects. A dowel cannot in the real world be a murder

Making Meaning in the Theatre: Double Noesis | 277

weapon, unless one is very ingenious, whereas the deadliness of the fictional object is crucial to its significance. These daggers appear in Macbeth, Act 2 Scene 2, at a moment of great dramatic significance to the play. Lady Macbeth appears and has a soliloquy on her anxiety over the success of the murder, in the course of which she reveals to us that she could not bring herself to kill Duncan by her own hand. Macbeth appears carrying the daggers, now covered with Duncan’s blood. Lady Macbeth is in such a state that she fails to see them for 33 lines. She then insists that he should return to lay them beside the grooms. He refuses, and she returns the daggers herself. The daggers are murder weapons, the instrument of the most significant event of the play. Their bloodiness is the proof of Duncan’s death as well as the symbol of Macbeth’s guilt, and ultimately also Lady Macbeth’s. She famously tries to wash that same blood off her hands in the last act. Yet in many productions, the blood is not made visible, even when, as in Act 2 Scene 2, the blood is meant to have physical reality. When I directed the play myself, I did not make the blood visible. In our imaginary production, they are yet further from their natural appearance, for they are only dowels. Our actors have a problem. They must remain aware that the dowels are dowels. At the same time, they must by their treatment of these objects make them emotionally central to the scene they are performing. That is to say, they must make them emotionally central to the scene in the experience of the audience. A very frequent way of resolving this problem is through the emotions of the actors themselves. If they regard the daggers with (for example) horror, the audience will regard them the same way. One aspect of rehearsal in many productions is the development of emotional response of this kind. An actor may be asked to remember relevant incidents in his or her own life, be interviewed in character about the daggers, or work out improvisations involving other imaginary objects. The actor’s double focus on an object appears crucial to the aesthetic experience of the actor in the case of this play. The actor must see it as it is, but give it emotional significance for him- or herself. There is a great variety of emotional significances that may be given to the bloody daggers, but they must be given significance. The actor must in imagination see the invisible blood and feel the significance of the blood to the dramatic situation and consequently to the play. The focus of the two actors’ eyes on the daggers, the expression in their faces, their bodies and their voices, will characterize the importance of the daggers to the audience. In this case, the daggers are so important to the action of the play that this moment of focus determines important things about the play as a whole.

278 | Matthew DeCoursey Must the actor feel the emotion on the spot, while performing? Most programs for training actors and for rehearsing plays seem to assume so. The name of Stanislavsky is most associated with emotional memory, with the use of imagination to kindle emotion in the actor that will then be used to create performance (see Carnicke 2000). This would include, presumably, the transformation of the dowels into daggers . Other programs, such as Joan Littlewood’s, work outside-in, meaning that actors do physical exercises that create emotion in them, which, in its physical manifestation developed this way, then becomes part of the play (see Barker 2000). This is not the place to resolve this question, but the view that the actor must feel the emotion in performance is widely held. The imaginative focus that makes the dowels into daggers not only gives a double noesis to the object; it makes it possible for the actors to have double consciousness and double emotion. The actor Simon Callow explains the experience: When you’ve defined, by analysis or intuition, those parts of yourself that coincide with the character, you let them rip, while consigning the parts which are irrelevant to temporary oblivion. The sense of expanding into areas of yourself which have been dormant or repressed is exhilarating, even a little frightening: ‘Do I have this in me?’ Parts of yourself that you’ve never dared to reveal, as an ordinary inhibited citizen, burst arrogantly forth. It may be your wit, or your romanticism, your sexuality or your homicidal tendencies. Unless it’s your own lust, longing, or craving, the audience will only be intellectually aroused: the thing will have been referred to, but not experienced. And you need to enjoy it. (Callow 2003, 165–166)

This double consciousness is useful to them in doing their job as actors. The character Macbeth, in this scene, may well be hysterical, close to losing control. As a director, I would not want my actor to be hysterical, and would like him, as an actor, to remain calm. I would want the emotion of the character to be real to him, to be felt and not just thought, but I would still want the actor to remain calm and in control. This appears difficult, but it happens every day in theatres. The character is in torment, but I would like my actor to enjoy the moment, to feel the control and learn confidence from this moment, and generally speaking, actors do. It appears that there is in the consciousness of actors a double source of meaning, something like a double transcendent ego in Husserl’s terms. The two are obviously not independent of each other, and they must be ultimately one, but there is also a clear division: The normal ego persists, seeing the dowels as dowels; and the fictional ego also exists, seeing the dowels as daggers. The fictional ego exists within the fiction, in the midst of the emotions of the charac-

Making Meaning in the Theatre: Double Noesis | 279

ter. It is not at all rare for the two to have contrary emotions. To take another example from Walton, think of a father playing with his son (1990, 242). He imitates a monster and advances on the child, who runs away screaming but smiling. It is fictional that he is a monster, and the son responds fictionally to the monster with fear; but it is also true that he is the boy’s father, and the child responds truly to the excitement of a game. The child in this example constitutes his father in two ways at the same time: As a monster and as his father. For Husserl, the constitution of the object is a logical process that does not involve emotion, and emotional response plays no role in the constitution of the object. That position appears to be untenable in this instance. The child willingly constitutes his father as a monster because of an emotion, or perhaps because of a potential emotion: the desire to feel the joyful excitement of non-threatening fear. The difference between the child and the actor is in adult discipline. The child has no need to keep the two separate, focus on one more than another. The actor must keep the pleasure of acting hidden for the sake of the audience and of other members of the creative team. The actor must, in a disciplined fashion, show the character only, and seek to present the aesthetic object as determined by the whole creative team or by the director. This discipline also applies to the emotional intensity given to the construction of the aesthetic object. The child feels the fictional fear, but can withdraw from it if it is too much for him. The actor must feel what the role requires without withdrawing from it (unless of course it is part of the artistic plan that all actors must withdraw from feeling). The daggers here can function as a synecdoche for the aesthetic object as a whole. In themselves, they are only black dowels; in the context of the play, they become daggers – so with every object in the play and with every actor when seen from the outside or from the inside. In themselves, they are one thing; in the play they are something different. Actors see both simultaneously. In mimetic theatre, the aesthetic experience of the audience often appears as an extension of the aesthetic experience of actors. When directors urge beginning actors to ‘focus’ on the emotions of their character, this is in order that the audience should find the emotions of the play real, feel with the characters or feel about the apparently real emotions of the characters. The action of the play acquires emotional reality because the art of the actors succeeds in providing significance for the action of the play. The audience ends up caring what happens. As Paul Woodruff (2008, 22) writes, “[t]heater is the art of finding human action worth watching, and it mostly does this by finding human characters worth caring about”.

280 | Matthew DeCoursey One of the most salient features of theatrical noesis is that people intend to create the meaning, and go to a lot of trouble to create it right. Husserl, as I mentioned briefly earlier, suggests that there are two kinds of intentionality, operative and act. Operative intentionality contains no intentionality in the commonsense meaning of the term. ‘Intentionality’ here is the aboutness of consciousness: what the consciousness is directed toward. ‘Act’ intentionality does involve intentionality in the commonsense meaning of the term: one turns one’s attention toward something. I think the problem of the double ego I mentioned earlier can be resolved if we simply hold that the construction of the aesthetic object is always marked by act intentionality. At the rehearsal phase, it is an act of will to see the dowel as a dagger. After rehearsal, in performance, the perception of the dowel as a dagger is considerably naturalized, made intuitive. I think the problem of the double ego I mentioned earlier is resolved if we simply hold that the construction of the dowel as dagger is always marked by act intentionality. At the same time, this act-intentional, learned intuitive awareness does not suppress or replace operative intentionality. We normally remain aware of where we really are. One must in practice realize that the daggers are really dowels. By extension, actors often retain a full awareness that Lady Macbeth is in fact their friend Sophie, and so on. Similarly, when the child runs away screaming but smiling from his father, he has learned to constitute the father as a monster intentionally, in the commonsense meaning, as an act of will. His awareness that the father is really his father remains in operative intentionality. As we have seen, the child and the actor both may have two opposing emotions at the crisis moment: one, within the fiction, negative, and one, outside the fiction, positive. I would suggest that emotion is integrally involved in the act of artificial noesis. The child puts emotion into his intentional act of constituting his father as a monster. He is not disciplined about it, and he does not rehearse, so the boundaries between the two are unstable. You can generally tell from looking at him that the fear is fake. So with bad actors and beginning actors not sufficiently rehearsed: you can tell that the emotion is fake. Everybody also knows that you have to be careful with small children. If you make the monster image too vivid, the child will start to believe it and genuinely become terrified. That part is different with beginning actors. I have never seen a beginning actor respond that way. The problem is always the other way, making the emotion real enough to affect the audience. In an older person, it seems the operative intentionality is strong and hard to overcome. Husserl thought that noesis was a conceptual process, and that emotion was restricted to response to the noema once formed. This material, together with the experience of rehearsal, suggests that emotion is central to the process of

Making Meaning in the Theatre: Double Noesis | 281

artificial noesis. The dowels only become daggers by dint of emotion, only become vivid and real because there is an emotion. If emotion is central to the creation of meaning here, then the problem of empathy is also central. How does the emotional response of the audience relate to the actual emotions of the actor onstage? Can an actor coldly manipulate the signals of emotion so as to make the audience feel things that she does not feel herself? Current research in neuroscience offers some hope of answering this question. The discovery of mirror neurons (see Rizzolatti and Craighero 2004) has created a concerted effort to describe the interrelations of voluntary and involuntary systems in the exercise of empathy.4 As we come to understand more about the operation of empathy in neural terms, it should become possible to define the interrelation of empathy with meaning more clearly. This paper has suggested that the aesthetic experience of both actor and audience can be understood through the idea of double noesis. We have seen that certain notions from the psychological research of Daniel Kahneman support this statement, and that it is credible to think of emotion as central to artificial noesis in the theatre. Kahneman’s System 1 is involved with immediate, short-term awareness of a system of association and with emotion. The creation of a production may be seen as building such a system of association, which then becomes intuitive to actors on the stage. Through their awareness of the aesthetic object constructed in rehearsal, actors achieve double perception of all that is around them in the course of acting. Through that double perception, they may experience double emotion. That double perception and the consequent double emotion are important parts of the aesthetic force of theatre. There remains a question as to the relation of the actor’s feeling to the feeling of the audience. Is it necessary to the audience’s response that the actor should really feel, as Callow asserts? A wide range of research is relevant here, including some research on the brain.

References Barker, Clive. “Joan Littlewood.” Twentieth Century Actor Training. Ed. Alison Hodge. Abingdon, UK: Routledge, 2000, 113–128. Callow, Simon. Being an Actor. New York: Picador, 2003. Carnicke, Sharon Marie. “Stanislavsky’s System: Pathways for the Actor.” Twentieth Century Actor Training. Ed. Alison Hodge. London: Routledge, 2000, 11–36.

|| 4 See Spunt and Lieberman (2013); den Ouden et al. (2005); Iacoboni et al. (2005).

282 | Matthew DeCoursey den Ouden, Hanneke E. M., Uta Frith, Chris Frith and Sarah-Jayne Blakemore. “Thinking about Intentions.” NeuroImage 28 (2005): 787–796. Dewey, John. Art as Experience. New York: Perigee, 1934. Dufrenne, Mikel. “The World of the Aesthetic Object.” The Phenomenology of Aesthetic Experience. Translation, Edward S. Casey. Evanston: Northwestern University Press. Rpt. In The Continental Aesthetics Reader. Ed. Clive Cazeaux. Abingdon, UK: Routledge, 1966. 129– 150. Hamilton, James. The Art of Theater. Malden, Mass.: Blackwell, 2007. Iacoboni, Marco, Istvan Molnar-Szakacs, Vittorio Gallese, Giovanni Buccino, John C Mazziotta and Giacomo Rizzolatti. “Grasping the Intentions of Others with One’s Own Mirror Neuron System.” PLoS Biology 3.3 (2005): e79. doi:10.1371/journal.pbio.0030079. Iser, Wolfgang. The Act of Reading: A Theory of Aesthetic Response. Baltimore: Johns Hopkins University Press, 1978. Jackson, Anthony. Theatre, Education and the Making of Meanings: Art or Instrument? Manchester: Manchester University Press, 2007. Kahneman, Daniel. Thinking, Fast and Slow. London: Penguin, 2011. Kant, Immanuel. Critique of the Power of Judgment. Translation, Paul Guyer and Eric Matthews. Cambridge: Cambridge University Press, 2000. Lewin, David. “Music Theory, Phenomenology, and Modes of Perception.” Music Perception 3 (1986): 327–392. Ricœur, Paul. “Narrative Time.” Critical Inquiry 7.1 (1980): 169–190. Rizzolatti, Giacomo and Laila Craighero. “The Mirror Neuron System.” Annual Review of Neuroscience 27 (2004): 169–192. Smith, David Woodruff. Husserl. Abingdon, UK: Routledge, 2007. Spunt, Robert P., and Matthew Lieberman. “The Busy Social Brain: Evidence for Automaticity and Control in the Neural Systems Supporting Social Cognition and Action Understanding.” Psychological Science 24.1 (2013): 80–86. Walker, Julia. “The Text/Performance Split across the Analytic/Continental Divide.” Staging Philosophy: Intersections of Theatre, Performance and Philosophy. Ed. David Krasner and David Z. Saltz. Ann Arbor: University of Michigan Press, 2006, 19–40. Walton, Kendall L. “Fearing Fictions.” The Journal of Philosophy 75.1. (1978): 5–27. Walton, Kendall L. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Wittgenstein, Ludwig. Philosophical Investigations. Translation, Gertrude Elizabeth Margaret Anscombe. Oxford: Blackwell, 1953/1997. Woodruff, Paul. The Necessity of Theater: The Art of Watching and Being Watched. New York: Oxford University Press, 2008.

Eran Guter and Inbal Guter

Impurely Musical Make-Believe The mode of narration appropriate to absolute music, that is, “pure instrumental music without text, title, programme, dramatic setting, or any other extraordinary music apparatus” (Kivy 2009, 157), has been a matter of controversy in contemporary Anglophone philosophy of music for some three decades. Musicologist Anthony Newcomb put the issue succinctly: A large component of most music lies in its power […] to delight with its patterns in sound. (Hanslick likened music to the arabesque and the kaleidoscope.) But in some music these patterns seem to force upon some of us recognition of meaning connected to other aspects of our life – of a representational and expressive power. (Newcomb 1997, 132)

The crux of the problem is the conviction, which germinated in Eduard Hanslick’s seminal treatise On the Musically Beautiful (Hanslick 1986 [1885]), that pure instrumental music lacks representational, narrative, semantic, or other extra-musical content, hence whatever meaning such music has consists in what Richard Wollheim (1987, 45) called “the much maligned property of the decorative”. Yet arguably, such extra-musical content, and accordingly, the power of music to be ‘about’ such content in some sense, seem to play a prominent role in our understanding and appreciation of pure instrumental music and seem vital to the value that it holds for us. Without extra-musical content we are left with pure repetitive form. The problem concerning the mode of narration appropriate to absolute music patently relates to the unique unfolding of such music qua ‘the fine art of repetition’, to use the catchy title of a book by Peter Kivy (1993a). It is not at all clear how we might explain why and how pure repetitive patterns of sound relate to our lives, or whether we can do so at all. In this paper we probe this problem in the somewhat unexpected context of Kendall Walton’s theory of ornamentality. In his seminal book, Mimesis as Make-Believe (1990), and in a number of other papers (1986, 1988, 1994, 1999, 2011, 2012), Walton carefully integrated the art of music into his theory of the representational arts, arguing that not only bona fide instances of programme music, but also works of so-called absolute music are thoroughly capable of generating fictional truths by engaging the listeners with props in introspectively imaginative games of make-believe. Hence our discussion in this paper is bound to read Walton against the grain. Yet one of the philosophically attractive features of Walton’s theory of make-believe is the way that it is couched in terms of imaginative transactions with artifacts, placing homo ludens in the limelight. Walton gives unusual precedence to the experiencing subject – what

284 | Eran Guter and Inbal Guter would she identify as props, and how would she use them in appropriate games of make-believe? In our discussion we will tap into this essentially Waltonian intuition in order to show that a possible remedy for the problem concerning the mode of narration appropriate to absolute music can actually be found in his fruitful ideas on ornamentality. For an elegant initial setup for our discussion, we turn to a famous passage from E. M. Forster’s novel Howards End, which Peter Kivy wittily invoked in his book Music Alone (Kivy 1990). In this passage, four characters are listening to Beethoven’s fifth symphony, “the most sublime noise that has ever penetrated into the ear of man” (Forster 1963 [1910], 31). Two of the listeners in that scene – Helen Schlegel and her younger brother Tibby – are of particular importance for Kivy as well as for our present concerns, since they exemplify seemingly incompatible types, or modes, of listening to music. Helen – “who can see heroes and shipwrecks in the music’s flood” (Forster 1963 [1910], 31) – dwells on imaginative introspection; Tibby – “who is profoundly versed in counterpoint, and holds the full score open on his knee” (Forster 1963 [1910], 31) – savors tönendbewegte Formen. For Kivy, an eminent advocate of formalism, this sharp contrast between Helen-type listening and Tibby-type listening ultimately means that Helen cannot really be a pure listener ‘of the work’. Rather, Helen-type listening is a case of attending an amalgamation of the music per se with the listener’s own imaginative, introspective, often unruly contribution. Since, according to Kivy (1990, 91), “our enjoyment of music is of cognitively perceived musical sound”, only Tibby-type listening is justifiable as a genuine experience of pure instrumental music. To be sure, postulating the mutual exclusivity of these two types of listening is unwarranted. Indeed, under the pressure of criticism, Kivy eventually withdrew his original claim concerning the unique authenticity of Tibby-type listening, admitting that neither his metaphysical nor his consequentialist strategies in Music Alone for establishing the claim that ‘pure’ music requires ‘pure’ listening are viable (Kivy 1993b). However, we may still regard the Tibby-Helen distinction as opening up, and charting, a continuous rich spectrum of congruent modes of listening. Thus, our investigation is driven by the following initial questions: (1) Can Walton’s theory of make-believe accommodate such a continuum? (2) How may such accommodation intervene in the current philosophical debate concerning the problem of absolute music?

Impurely Musical Make-Believe | 285

1 Formalism and narrativism Let us begin by situating our discussion vis-à-vis the current debate on the subject. In Antithetical Arts: On the Ancient Quarrel between Literature and Music (2009), Peter Kivy portrayed the current philosophical divide concerning the problem of absolute music as a watershed in aesthetic thought between two opposing traditions about how to understand music: formalism and narrativism. According to Kivy, whereas formalists insist that musical meaning inheres solely in the formal features of ‘music alone’, narrativists are actually literary interpreters of absolute music. They employ a literary analogy in order to attribute meaning to pure instrumental music, perhaps even to render absolute music as akin to literary art. Like Helen, they see heroes and shipwrecks in the music’s flood: disembodied actions of indeterminate agents. And Forster’s original choice of words, employing the verb to see rather than to hear, is actually philosophically important here, because (Kivy argues) this is precisely the problem: We don’t really hear heroes and shipwrecks in the music’s flood, but rather, quite figuratively, we behold such things in imagination or perceive them mentally, against the music’s flood. The most popular form of present-day narrativism is the persona theory. The main idea of this theory was first voiced by the musicologist Edward T. Cone (1974, 94), who claimed that “any instrumental composition, like the instrumental component of a song, can be interpreted as the symbolic utterance of a virtual persona”. Variants of the persona theory were subsequently developed and defended independently by Jerrold Levinson (1990, 306–335) and by Jenefer Robinson (2005). Such theories commonly postulate some sort of indefinite persona in the music, which is supposed to serve as a hinge for our deeply felt interest in pure instrumental music. The idea is that just as narrators in literary works express beliefs, attitudes, intentions, and emotions by means of the words of the text, musical personae express emotions (feelings, attitudes) by means of the sounds (or ‘gestures’) of the music. So listeners’ experiences, like those of readers, involve something like recognizing and responding to another person, one who experiences and expresses the emotion in question. As Kivy pointed out, the musical persona performs two functions for the narrativist: [T]o give absolute music a fictional content that is supposed to account for its artistic substance and interest, at least in part; and to explain how absolute music is capable […] of arousing what I have been calling the “garden-variety” emotions […]. (Kivy 2009, 101)

From the formalist’s point of view, the persona theorist wishes to take an easy way out of the problem concerning pure instrumental music (Kivy 2002, 113–

286 | Eran Guter and Inbal Guter 119, 2009, 101–117). Persona theory may seem to fare better in explaining why we consider pure instrumental music valuable, or at least it seems to be able to do so more naturally, but only on pain of allowing the uncontrollable overpopulation of the canon of pure instrumental music with shadowy musical personae. The problem is that we cannot individuate such musical personae from one another or re-identify any of them over time in any clear sense. The analogy with the literary model runs out of steam when we come to realize that characters in literary fiction are interesting because they are concrete individuals with relatively rich personal histories who are situated in relatively detailed circumstances; they are not abstract entities, or ‘empty suits’, as Kivy (2002) calls them. In the last analysis, musical personae may possess very few fictional resources to hold our interest or to move us emotionally. Moreover, construing musical personae as hinges for our emotional involvement with pure instrumental music may involve a number of misconceptions and ambiguities. For instance, Walton (2011, 459) made the general point that “on the persona hypothesis, expression turns out to be a species of representation: Music represents itself as someone’s expression of feelings, as a story represents itself as someone’s reports of a series of events”. This eventually muddies some important distinctions between stories and pure instrumental music. But even on their own terms, persona theories seem to falter. Kivy (2009, 107–108) pointed out that in Levinson’s account we always feel the same emotion that the musical persona is supposed to be expressing. Characters in literary fiction do not restrict our responses in this way. This might be a result of misconstruing the concept of empathy as mirroring in the sense of taking on the emotion of the persona (see Levinson 1990, 320–321), rather than in the sense of reacting to his or her emotion with a different emotion. Robinson’s (2005) “new Romantic theory of expression in the arts” follows in the footsteps of Robin George Collingwood (1964 [1938]) by maintaining that readers and listeners take on, experience for themselves, or “recreate in imagination” the emotion they observe in the narrator or persona of an expressive work. However, such a conception might be open mutatis mutandis to the charge, which has already been effectively leveled against Collingwood’s original theory, that a work of art can have expressive qualities without there necessarily being a prior act of expression by an (implied) artist (see Tormey 1971). Stephen Davies (2003) made the important point that the listener projects the purported persona onto the work after the emotive content has already been recognized; hence the musical persona is hardly an essential precondition of expressive hearing. The arguable demise of the persona theory notwithstanding, Kivy’s own enhanced formalism seems to fare no better in explaining why, and how, the empty pleasure of pure instrumental music could be valuable in itself. While the

Impurely Musical Make-Believe | 287

persona theorist may have taken an easy way out, the formalist seems adamant to push things to a state of gridlock. Kivy invariably insists that we cannot explain the appeal of absolute music by appealing to its narrative content, because if it has narrative content then it isn’t absolute music qua an art of purely abstract but perhaps expressive sound (2009, 119–120). In Antithetical Arts, Kivy attempted to account for the value that pure instrumental music holds for us by observing a kinship between the kind of ‘morally uplifting experience’ which such music affords, and the inexplicable notion of mystical experience pertaining to aesthetic emotion which Clive Bell (1958 [1914]) promulgated a century ago: “[W]hat strikes me about Bell’s descriptions of his ‘aesthetic emotion’ are how closely they resemble the kind of ecstatic experience of absolute music I have been trying to describe […]” (Kivy 2009, 249). In fact, Kivy seems to have aligned himself even more closely with another Bloomsbury aesthetician, Roger Fry (1920, 199), who maintained that “those who experience [aesthetic emotion] feel it to have a peculiar quality of ‘reality’ which makes it a matter of infinite importance in their lives”. Fry did not shy from the ultimate conclusion: “Any attempt I might make to explain this would probably land me in the depths of mysticism.” Following a certain reading of Bell by Nicholas Wolterstorff, Kivy appeals to the religious overtones and allusions in Bell’s language in order to bathe the empty pleasure of pure instrumental music in profundity. There is no doubt that the members of the Bloomsbury group saw a very close and deep connection between ethics and aesthetics. Yet this connection derives primarily from the conceptual construal of the idea of significant form. Jaakko Hintikka (1995) pointed out that the aesthetics of Bell and Fry was suffused in the epistemology of George Edward Moore, which was predominant in Bloomsbury. The idea that significant forms are the basic objects of aesthetic experience was invariably connected with the idea that objects of acquaintance are the basic data of epistemological experience. Hintikka’s point is that significant form is actually analogous to a sense-datum as an object of immediate experience. It is an objective entity; it is not an object (form) that merely appears beautiful, but is rather what we experience as beautiful. However, such aesthetic objects cannot be identified with perceptual objects (patterns of color on canvas) or with the perceptual sense-data whose being consists in exhibiting those patterns. To do so would be to commit the naturalistic fallacy. “Hence”, Hintikka (1995, 23) concludes, “an object of an aesthetic experience must according to Fry or Bell have intrinsic properties other than perceptual or geometrical ones. It must intrinsically be a bearer of certain values. This is what the Bloomsbury critics express by saying that a significant form must have moral and spiritual values.”

288 | Eran Guter and Inbal Guter Thus, there was nothing religiously mystical about the kind of object to which the Bloomsbury aestheticians attributed the aesthetic emotion. The moral and spiritual values possessed by a significant form are intrinsic properties of what is seen or otherwise perceived in a work of art. The lesson to be learned from this brief historical excursion is that Kivy cannot seriously take recourse to the religious or otherwise mystical overtones of Bell’s (or Fry’s) idea of the aesthetic emotion without also admitting the much-contested conceptual presuppositions of early-twentieth-century British realism, which served as the ultimate philosophical justification for splicing aesthetics and ethics in Bloomsbury. We remain skeptical whether Kivy would be willing to mount a defense of Moore’s epistemology for this particular purpose. Our survey of the current debate remains incomplete, but it is now full enough to observe the contours of the debate. As things stand, the debate on both sides of the fence seems to have backtracked unwittingly into problematic philosophical presuppositions of yore, at least in the cases of Robinson and Kivy. Nonetheless, each side seems to capture something important about our experience of pure instrumental music. Champions of Helen-type listening are keen to retain the insight so beautifully expressed by Ludwig Wittgenstein (1980, 73): “For me this musical phrase is gesture. It insinuates itself into my life. I adopt it as my own.” But, as Stephen Davies (2003, 168) observed, while the imaginary narrative of the progress of the musical persona is indeed possible, it follows, and only contingently, the grasping of the music’s nature, the recognition of its unity, integrity, symmetry, and so forth, upon which the narrative projection is cast. So Helen cannot do without Tibby. Champions of Tibbytype listening are keen to retain the equally important insight that music actually delights us with its patterns in sound. Yet Tibby might not attain profundity without Helen. To quote again from Wittgenstein (1980, 70): “Understanding music is a manifestation of the life of humankind.”

2 The Tibby-Helen continuum We shall now proceed in a direction which might appear counterproductive at first, especially against the backdrop of our discussion in the previous section. We shall focus on understanding pure instrumental music in terms of musical ornamentality. As noted before, such an understanding of pure instrumental music has been a hallmark of formalism at least since Hanslick, and presumably even as early as Kant (1952 [1790], 72), who suggested ranking “fantasias (without a

Impurely Musical Make-Believe | 289

theme), and indeed, all music that is not set to words” in the same class with “designs à la grecque, foliage for framework or on wallpapers”. Hanslick famously argued in On the Musically Beautiful that a certain kind of ornamentation in the visual arts may afford a serviceable image of how music is able to produce beautiful forms without specific feelings as its content. “[L]et us think,” he wrote (Hanslick 1986 [1885], 29), “of an arabesque not dead and static, but coming into being in continuous self-formation before our eyes”. Also: “Music is a kind of kaleidoscope […] Music produces beautiful forms and colours in ever more elaborate diversity, gently overflowing, sharply contrasted, always coherent and yet always new, self-contained and self-fulfilled” (Hanslick 1986 [1885], 29). He qualified this analogy (inconsistently) by admitting that “by contrast with arabesque, music is actually a picture, but one whose subject we cannot grasp in words and subsume under concepts” (Hanslick 1986 [1885], 30). Still, he seriously begged the question by rendering the distinction between visual ornamentality, which is “no more than a mechanically ingenious plaything”, and musical ornamentality, which is an exalted human achievement, in terms of the latter’s being a “direct emanation of an artistic spirit” (Hanslick 1986 [1885], 29). Kivy (1993a, 348) seized upon this conception, dubbing it “the wallpaper model”, and boldly suggested that we bite the bullet: We should retain the consistency of Kant’s original insight, which the comparison with wallpaper means to uncompromisingly make plain, that “music, indeed, possesses no content. It is pure, empty decoration: arabesque”. Elsewhere (1993a, 360–373) he went even further by arguing that pure instrumental music is not an art in the sense of that term most likely to be applied to it in the present day, that is, as a fine art. However, by focusing here on understanding pure instrumental music in terms of musical ornamentality, we do not wish to endorse or pursue the formalist agenda. To anticipate, Walton’s theory affords a much richer conceptual framework for the discussion of musical ornamentality. At this point, we would like to argue that the idea of musical ornamentality enables the opening up of a continuous rich spectrum between what we earlier dubbed Tibby-type listening, that is, the kind of musical understanding which typifies the formalist, and Helen-type listening, that is, the kind of musical understanding which typifies the narrativist. In order to do this, we first need to bracket the historical issue concerning the status of music as a fine art, and to dissociate the discussion of musical ornamentality from the formalist agenda. Kivy’s historical line of argument (1993a, 360–373) is that the notion of a fine art was secured in the eighteenth century upon the formation of the modern system of fine arts (see Kristeller 1990), and that its application to music rested

290 | Eran Guter and Inbal Guter on the idea of an essentially representational practice. Subsequent philosophy of art was bound to struggle with the implications of this theory-laden inclusion of pure instrumental music in the modern system of the arts. In reading history in this way, Kivy’s strategy is clear: to prompt a re-evaluation of prevailing narrativist proclivities in contemporary Anglophone philosophy of music. There are two good reasons for bracketing (for the purpose of our present discussion) the historical issue concerning the establishment of music’s status as fine art and its ensuing prestige. The first reason is that these are matters on which the account of musical ornamentality per se can remain agnostic. The second reason is that the problem, raised by Kivy may be quite overstated, as Philip Alperson suggested, if there is another way of understanding the notion of the fine arts (as arts of beauty rather than arts of imitative play), which could accommodate musical ornamentality or even both musical representation and musical ornamentality at the same time. There is yet another reason for dissociating the discussion of musical ornamentality from the formalist agenda: The splicing of these two distinct philosophical strata in the first place was the result of an unnecessary fixation on the visual analogue of the arabesque, a too-literal application of Kant’s rather pedestrian example. Kivy’s own meditation on a Persian carpet is quite telling in this regard (1993a, 349–358). The example is cleverly calculated to show precisely what the enhanced formalist needs to say: that music is a multidimensional (i. e., polyphonic), quasi-syntactic, and ultimately deeply expressive and deeply moving pattern. Still, the arabesque model is neither the only nor even the primary non-musical analogue for musical ornamentality; and more importantly, in some instances of musical ornamentality this model is plainly not the right kind. According to Alperson (1992, 220), “we can find decorative elements within each of the major musical categories of Western instrumental music of melody, harmony and rhythm”. Such instances of musical ornamentality run the gamut from local devices such as trills, turns, grace notes, tremolos, and flourishes, through various means of embellishment of melodic passages and of rhythm, to phrasing accents and timbral manipulations, and ultimately to larger compositional practices. “In fact”, Alperson (1992, 220) submits, “whole musical styles – and thus the history of musical taste and appreciation – may be understood, if not identified, in large part by reference to decorative practice”. It is important to observe that most of the examples in this vast array of musical phenomena actually fall outside of the arabesque model. The gist of the arabesque model is the idea of sequentiality, as shown in Kivy’s use of the Persian carpet example to flesh out his account of repetition in classical form. However, not all instances of musical ornamentality are explicable, neatly or

Impurely Musical Make-Believe | 291

even at all, in terms of sequentiality. For example, short trills, turns, grace notes and their ilk are all local phenomena which have no bearing on musical structure. They decorate in the most straightforward sense, and more often than not, they can be removed in performance without significant musical loss. The cadenza, which is ornamental with respect to the work as a whole, is patently external to the formal plan of the piece. This is most clear where the cadenza has been composed by someone other than the original composer of the concerto, or is improvised by the performer. More complex counterexamples can be found in certain calculated uses of such ornamental devices as accelerando and ritardando to undermine sequentiality by way of surprise. For example, in the second movement of Beethoven’s Piano Sonata No. 31 in A-flat major, op. 110, the restatement of the opening theme in measures 96–111 is varied with ritard and a tempo markings in its second appearance. One cannot predict this change in tempo since both the antecedent and the consequent phrases return completely and identically, hence they are readily referred to the original statement of the material. Thus, the embellishment of the tempo in the second repetition turns out to be an unexpected interruption, which in effect weakens the sense of sequentiality altogether. Other instances of musical ornamentality are profoundly a matter of musical practice, products of cultural and social conditions. Consider, for example, the basso continuo in the musical practice of the seventeenth and eighteenth centuries. This is a bass line with appended accidentals and numbers (‘figures’) that provides the harmonic outline in many of the musical compositions of the Baroque era. Execution of this characteristic notation assumes the proficiency of the player in conveying the ‘coded information’ by realization of the symbols. However, although the symbols may designate particular forms of harmony, they do not prescribe a fixed distribution of the voices. The role of the accompanist is to “fit to each piece a correct performance of its harmony in the proper volume and with a suitable distribution of tones” (Bach [1753/1787] 1949, 174). The practice of basso continuo also encourages the performer to reach beyond the score in the sense of refining the accompaniment with embellishments and varied imitations which are based on the motivic material. The accompanist needs to practice discretion in striking a good measure of variability against the principal part while avoiding tedious repetition. Even more importantly, he needs to practice discretion in identifying the uniqueness of a certain piece and designing its appropriate accompaniment for the particular occasion of its performance. For instance, he needs to take into consideration the circumstances of the performance, its location and audience, and the size of the ensemble. Consequently, the execution of basso continuo may vary in

292 | Eran Guter and Inbal Guter each performance; it is dependent on the judgment and skill of the accompanist. In sum, to account for the ornamental practice of the basso continuo, we must acknowledge the crucial role played by the accompanist in the compositional process, finalizing and determining a musical composition in accord with its contemporary performance practice. In this case, compositional decisions taken by the performer need not follow a pure compositional method, but rather bring together practice and method. The arabesque model of musical ornamentality falls short of explaining such quintessential cases of musical ornamentality. The rich variety of musical ornamentality is hardly surprising considering the enormous scope of non-musical ornamentality, which encompasses all kinds of objects and practices, from tapestries to the sleek contours of race cars, from glass-blowing to bodily modification, from interior decoration to social decorum. It is important to observe that in many such cases, the ornamental aspect of such objects or practices is perfectly compatible with full-fledged mimesis. If we consider such objects and practices as ornamental, it is not because they each have a certain form, but because they impinge on our way of living in a certain form. So to look for the point of such ornamental objects and practices is to ask how they mesh with our life. One is not likely to take this to be the point of musical ornamentality if one takes the arabesque to be one’s paradigmatic example. Having cleared up these issues, we may now return to the musicologist’s intuition: [I]n music as in the other arts (verbal, filmic, literary, painterly) aspects of agency are not continuously displayed, nor are aspects of narration. Both are only intermittently operative. Even the most “expressive” music […] at times simply swirls or dreams or chugs along in its decorative function. But in this it is essentially no different from painting and literature. It may differ from them only in the balance of these functions. (Newcomb 1997, 133; emphasis in original)

In effect, Anthony Newcomb introduces here what we would dub the TibbyHelen continuum, that is, the idea that musical ornamentality (sans formalism) and narrativism not only are not mutually exclusive, they are in fact mutually permeable, continuous with each other, and complementary. The balance of these functions may differ across the musical repertoire, across different styles and performance practices of pure instrumental music, and even across various renditions of a given musical work, invoking a need to take into account the audiences and traditions to which they appeal as well as what Alperson (2008) recently called ‘the instrumentality of music’ – a richly layered understanding of the art of music in terms of culturally and historically

Impurely Musical Make-Believe | 293

situated practices of human beings. Thus, we are inclined to accept Alperson’s suggestion that we may want to think about instrumental music as encompassing a continuum, from music which approaches the limit, on the one hand, of a pure art of decoration, to music which approaches the limit of full-blown imitation on the other. (Alperson 1992, 227)

Kivy (2009, 131–139) raised a couple of general objections to Newcomb’s original suggestion to consider such a continuum. To complete our move in this section of our essay, we shall now answer these objections. Kivy’s first objection is that the very idea of such a continuum patently begs the question about the narrative content of pure instrumental music: “[U]nless one simply assumes that absolute music has agents and narrative content […] one cannot argue about the degree to which it has them” (Kivy 2009, 134; emphasis in original). However, if the continuum concerns modes of listening (broadly construed as modes of human engagement with music or musical phenomena), and if we can admit a posteriori that narrative content is possible, as Stephen Davies (2003) suggested, then Kivy (1993b, 23) has already conceded that he “cannot, in a non-question-begging way, rule out ‘impure’ listening, tout court, as being listening not ‘of the work’”. In other words, Helen is here to stay. Kivy’s second objection derives its logical force from the Sorites paradox, as Kivy moves to establish that the very idea of a continuum in this case is a non sequitur. If the continuum is a matter of how much narrative content it possesses, then, according to Kivy (2009, 134), “that is perfectly consistent with its having, as the formalist insists, none at all, just as the bald man possesses zero degree of hair”. However, the appeal to the Sorites paradox is actually unwarranted here given the inherent vagueness of the concepts under discussion. These concepts are irrevocably derived from ordinary language. This is clearly the case even with regard to the foundational concepts of musical formalism itself. Heinrich Schenker (1954 [1906], 6), for example, construed the very concept of repetition by means of an analogy with the procreative urge of living creatures. In the final analysis, Kivy’s second objection is a red herring. Thus we pre-empt Kivy’s objections to the Tibby-Helen continuum.

294 | Eran Guter and Inbal Guter

3 Musical ornamentality as impurely musical make-believe We are now set to address the first question we posed in our introduction to this essay – can Walton’s theory of make-believe accommodate the Tibby-Helen continuum? – and also to answer it in the affirmative. The kind of musical make-believe that we wish to explore here is impure in two different senses. It is patently impure in the formalist’s sense, as Kivy (1990, 42–67) had argued. This need not disturb our present discussion. It is also impure in a particularly Waltonian sense, because it impinges on Walton’s characterization of music (in Mimesis as Make-Believe and elsewhere) as an art of imaginative introspection. This is where we propose to read Walton’s theory against the grain. We believe that Walton’s characterization of music can be resisted on independent philosophical as well as musicological grounds, but we shall leave that for another occasion. Our point is that his theory of make-believe nonetheless allows the kind of move that we propose here, and as we shall see, Walton himself actually approached such ideas at one point, although he never developed this any further. In Mimesis as Make-Believe, Walton (1990, 274–289) offered an intriguing account of ornamentality in terms of psychologically inhibited games of makebelieve. This account has received surprisingly little attention in the literature, and to the best of our knowledge, it has never been applied to music. According to Walton, ornamental works belong to the category of representations that positively discourage participation, especially the psychological participation that would constitute the experience of being caught up in a story. Works are ornamental insofar as they inhibit or interrupt certain imaginings. Such representations deliberately distance the appreciator from the fictional world or even hinder the imagining of what is fictional, effectively undercutting appreciators’ roles as reflexive props. In such cases, Walton contends, our stance is more akin to that of an onlooker than a participant in games of make-believe, although that we “observe” is not someone’s actually playing a game but rather the kind of game that might be played. We step back and examine the prop, contemplating the games it might inspire and the role it would have in them. We may marvel at a work’s suitability for use in games of certain sorts; we may be fascinated by the combination of fictional truths it generates (this amounts to an interest in the work world); we may admire the artist’s skill and ingenuity in devising ways of generating fictional truths; we may delight in the devices by which participation is inhibited. (Walton 1990, 275)

Impurely Musical Make-Believe | 295

Walton usefully explicated his notion of ornamentality as psychologically inhibited games of make-believe by suggesting that ornamentality consists in fictional worlds in which other fictional worlds are embedded. Embedding a fictional world within another one puts it at a certain emotional ‘distance’ from us. According to Walton (1990, 287), the first-order world, that is, the world of the framing game, “is frequently very sparse, consisting of scarcely more than the work itself together with, by implication, its artist and his creative activity, and inviting minimal participation”. Our participation is in this first-order game of make-believe: In participating in it we may imagine that there is another game which we could participate in […] But imagined participation is not actual participation, and imagined participation, let alone imagining merely that there is a game to participate in, does not constitute involvement in a fictional world. We stand apart from the internal fictional world and observe it through its frame. (Walton 1990, 284)

Walton’s theoretical structuring of psychologically inhibited games of makebelieve as consisting of two game-worlds, one embedded in or framed by another, makes his theory of ornamentality particularly proximate to the Tibby-Helen continuum. The rudiment of a continuum or a sliding scale is already in place. According to Walton, the inhibitions wherein ornamentality subsists are usually partial and the interruptions are temporary, and different ornamental elements may differ in their capacity to inhibit participation. Also, according to Walton (1990, 283), ornamental works remain genuine representations: “Their ornamentality merely alters what they are representations of.” The two game-worlds may interrupt or interfere with each other, and they generally interpenetrate. In some cases, their juxtaposition might even be the point of the piece. On the whole, Walton maintains that ornamentality coexists with full-fledged mimesis; it is an aesthetic phenomenon which is much more widespread in art and in life than we tend to acknowledge. We can neatly map the two poles of the Tibby-Helen continuum onto Walton’s dual-game formation. Let us be reminded that E. M. Forster’s delightful characters, Tibby and Helen, serve here as idealized placeholders for different modes of human engagement with music or musical phenomena. The kind of game in which Tibby, “who is profoundly versed in counterpoint, and holds the full score open on his knee” (Forster 1963 [1910], 31), participates while listening to Beethovens fifth symphony exemplifies Walton’s first-order game-world which includes the work (the symphony) and, by implication, the artists (the composer and the performers) and their creative activities. In the limiting case, the point of Tibby’s game may very well be merely to delight in the devices by which participation is inhibited. This may indeed correspond to Kivy’s descrip-

296 | Eran Guter and Inbal Guter tion of the ‘empty pleasure’ of absolute music. But, as Walton pointed out, the point of such games ordinarily extend to appreciating how suitable the work is for use in games of certain sorts and for generating fictional truths as we appreciate the compositional ingenuity and the performing skills in devising ways for generating such fictional truths. So, beyond the limiting case, Tibby’s game is all about reaching out to the second-order world, to the embedded game-world, Helen’s game, in which it is fictionally true that she see heroes and shipwrecks in the music’s flood. The Tibby-Helen continuum stretches perpendicularly from the framing game of make-believe into the game-worlds embedded therein. It is important to note why Helen’s use of Beethoven’s symphony is not illicit. She sees heroes and shipwrecks in the music’s flood, which means that she has not forgotten about the primary, framing world. She does not mistake the internal, second-order game-world for the primary, framing world, as though it were the world of the work. Helen-type listening does not amount to using the work as a prop in a way in which it is only fictional that it is to be used (see Walton 1990, 285). Only in the limiting case would Helen-type listening indeed be listening “not of the work” in this sense. This dialectic of temptation and resistance fleshes out the idea of artistic recalcitrance pertaining to the musical material and its promise of artistic achievement. Certain musical works tend to inhibit more; hence they require some effort to achieve vivid imagining, which might be essential for playing the piece with musical understanding. Other works tend to inhibit less; hence they require some effort to achieve distance, which might be essential for a musically balanced performance. The very idea of recalcitrance in this context underscores aspects of what Philip Alperson (2008, 46) calls “the instrumentality of music”. Alperson (2008, 46) argues for a rich ontology of musical instruments which accommodates their understanding as musically, conceptually, and culturally situated; the musical instrument “as an amalgam of material object, the performer’s body, and bodily dispositions as habituated by the developments of various musically related skills”. Accordingly, Alperson contends, the performance of musical works is a kind of musical practice in which the object of aesthetic appreciation is legitimately regarded as the work-in-performance. That is, there is a kind of double consciousness of the performance as a performance of the work as a musical entity and of the performer’s achievement in performing the work. (Alperson 2008, 47)

We would like to suggest that the artistic recalcitrance afforded by the TibbyHelen continuum (as seen from the vantage point of Walton’s theory of ornamentality) exemplifies precisely this sort of experience of ‘twofoldness’.

Impurely Musical Make-Believe | 297

One type of example of the artistic challenge involving what we may dub Tibby-to-Helen recalcitrance is found whenever we encounter what music critics often call cerebral or analytic renditions of otherwise highly expressive works, for example Tchaikovsky’s Pathétique symphony. Familiarity with the tonal idiom, format, and rhetoric of such works may render them significantly more apt to be experienced ‘from within’, in Walton’s sense (2011). Yet a musician may wish to do more than offer listeners means of expressing feelings or emotions. A conductor might want us to experience the first movement of the Pathétique symphony not just as an expression of conflicting emotions, ours or someone else’s (a musical persona, perhaps), but also as an ingenious symphonic structure. He may show ingenious artistic skill precisely in inhibiting or interrupting the listeners’ ability to lose themselves in any second-order fictional world beyond their initial appropriation of the musical sounds. He might hold back a little the tempo of the third movement, a defiant march, in order to let the inner voices and overlapping textures shine through. Such a finely balanced performance would unequivocally show an artistic interest in reflexivity, in the means by which participation is promoted and the kinds of participation one might engage in. For the price of lacking a certain kind of passion in Tchaikovsky, one would gain a fuller understanding of how the music works by being geared toward concrete aspects of the actual performance. Walton contends (1990, 288–289) that this is in general one of the great cognitive values of ornamentality. The converse category of Tibby-to-Helen recalcitrance is even more illuminating. In such cases, if the music is to work at all, reflexivity must give way to, and in effect coexist with, vivid imagining. A relatively simple example is the long trill. Among musical embellishments, the long trill has a special function: the prolongation of a long note. Keyboard instruments are not capable of sustaining long notes or modifying their dynamic level once the action of striking the key has been completed. In his definitive treatise on keyboard techniques, Essay on the True Art of Playing Keyboard Instruments (first part published in 1753; second part in 1787), Carl Philipp Emanuel Bach (1949 [1753/1787], 150) argued that the “empty space” may result in a “consequent monotony due to a lack of sonority”. If the score calls for a long note in either the upper or the lower voice, then the long trill is commonly regarded (even where it is not actually specified in the score) as a suitable way for the performer to compensate for the acoustic limitations of the instrument so that “the listener will believe that he is hearing only the original note” (Bach 1949 [1753/1787], 150). This example shows the prevalence and importance of the idea of enhancement for any complete theoretical understanding of musical ornamentality. The idea of enhancement, which was urged upon Kendall Walton by Patrick

298 | Eran Guter and Inbal Guter Maynard (Maynard 1991; Walton 1991, 416–420), pertains to cases in which an ornament induces imaginings which direct and vivify one’s perception of the thing ornamented. In the case of the long trill, the thing ornamented is the ‘empty space’, which the long note prefigures. Thus the imaginings, which the long trill induces, facilitate the listener’s grasp of the unfolding of the phrase’s structure. Importantly, the idea of enhancement emphasizes the roles actual things – their physical features as well as their functions – play in make-believe, and other ways in which fiction and reality are intertwined. No less important is the idea that enhancement may coexist with reflexivity. That is, focusing attention on the physical properties of an object apart from their role in generating fictional truths may still vivify imaginings concerning that object. It is noteworthy that Walton (1991, 417) has actually acknowledged that “enhancement of the kind Maynard has in mind is especially important in music, in the ornamentation or embellishment or elaboration of melodies or motives. There is much to be done in the way of explaining what the enhancement of an ornamented melody consists in, and what role imaginings have in the process”. To the best of our knowledge, Walton never developed this idea, or any of its theoretical implications, in his published writings. Another important result of Maynard’s suggestion is the extension of Walton’s notion of ornamentality to include ornamenting or decorating in the transitive sense (the decoration of actual things in our environment). We can comment on this only very briefly here. In such cases, Walton (1991, 417) agrees, “one’s attention is likely to be drawn to the thing decorated and to the relation of the decoration to it, and this might distract one from full participation in a game with the decorating representation”. This extension significantly opens up the discussion of musical ornamentality to accommodate important questions concerning how music meshes with our social reality. For instance, Ernst Gombrich pointed out that music may follow the laws of social decorum in ways analogous to cases of symbolism in visual decoration. Gombrich wrote: The minuet, the hymn, the pastoral, the aria, are all associated with moods which the composer can accept of modify much as the decorator can treat the music room, the boudoir, or the garden house by accepting or flouting conventions. (Gombrich 1979, 301)

These are aspects of the musical work-in-performance which call for further exploration in the framework of Walton’s theory of ornamentality. The case of the long trill clearly shows that enhancement in music, even in the simplest case, is indispensable, as opposed to the crude example of the sort of automobile decorations that induce us to imagine that a car is fast (color

Impurely Musical Make-Believe | 299

stripes, trim, or simply the styling of the body). In the latter case, if we remove the embellishment, it would be more difficult, yet still quite possible, to imagine that this car is fast. The embellishment is merely ‘stuck on’ what it decorates. In the case of the long trill, if we remove it, then it will be quite impossible to imagine that the long note persists. An aural gap will remain in the middle of the music. The performance will simply fail. The indispensability of embellishments in music received an elaborate theoretical defense across the board in C. P. E. Bach’s Essay on the True Art of Playing Keyboard Instruments. Bach (1949 [1753/1787], 79) suggested considering the many uses of embellishments for enhancement (in the sense explored here): “They connect and enliven tones and impart stress and accent; they make music pleasing and awaken our close attention. Expression is heightened by them; let a piece be sad, joyful, or otherwise, and they will lend a fitting assistance.” He (Bach 1949 [1753/1787], 79) argued that “embellishments provide opportunities for fine performance as well as much of its subject matter. They improve mediocre compositions. Without them the best melody is empty and ineffective, the clearest content clouded”. Importantly, there is a natural emphasis in Bach’s argument on the idea that the indispensability of embellishments is not merely a matter of the formal properties of the music, but rather enables us to appreciate music as work-in-performance, in Alperson’s (2008) sense. Another, strikingly different example of enhancement is found in Arnold Schoenberg’s Piano Piece op. 11, no. 1 (see Fig. 1). One of the most distinct characteristics of common-practice tonality is its ability to convey the sense of directed motion toward a goal. This goal is achieved through a ‘cadence’, a closing formula which contributes to the affirmation of tonality. Schoenberg’s ‘emancipation of the dissonance’ in such works as his Piano Piece op. 11, no. 1 yields an abundance of unresolved dissonances, which obscure the tonal center. As Schoenberg (1975, 134) himself admitted, when tonality is thus abandoned, “the necessity for fortifying the key may be questioned”. Schoenberg argued that in such cases the recalling of tonal centers would not compensate for the absence of directed motion. Yet he also argued that unity and coherence cannot be achieved solely by means of tonality: [T]here are other laws that music obeys, apart from these [i. e., the laws of sound at the origin of tonality] and the laws that result from the combination of time and sound: namely, those governing the working of our minds. This latter forces us to find a particular kind of layout for those elements that make for cohesion. (Schoenberg 1975, 259)

300 | Eran Guter and Inbal Guter

Fig. 1: Arnold Schönberg, 3 Klavierstücke für Klavier, op. 11/1. © Copyright 1995 by Wiener Urtext Edition, Musikverlag Ges.m.b.H. & Co., K.G., Wien/UT 50195

Schoenberg (1995, 249) approached the cadence in its role as “restoration of the state of rest”. He maintained that when tonal means are unavailable, this state of rest can be restored by other means of articulation. In op. 11, no. 1, Schoenberg replaced the harmonic cadence by ritardando and langsamer markings that served as means of punctuation so that beginning and ending could be clearly delimited. In effect, the embellishment of tempo carries the musical rhetoric of the piece. The performer must vividly imagine closure for the music to work. Schoenberg’s atonal idiom in this piece palpably makes enhancement musically indispensable. If we remove the embellishment of tempo, that is, if we fail to make the gesture in performance, or if we refuse to imagine that this is closure, then the music is bound to fail. We maintain that this sort of enhancement by means of gesture is vital to many other contemporary musical idioms and underlie a mode of listening which is conducive to the appreciation of such music, one which unsympathetic listeners often refuse to engage in. Another example is needed in order to show a distinguishing case: the advent of Tibby-to-Helen recalcitrance which does not involve enhancement. Consider Frédéric Chopin’s piano etude op. 10, no. 6 (see Fig. 2). This piece calls our immediate attention to its highly chromatic harmony. It displays a four-part

Impurely Musical Make-Believe | 301

polyphonic texture with chromatic sixteenth-note figuration occurring within the inner voices. The intricateness of the polyphonic structure, together with the presence of decorative semi-tonal movement, leads to increased tension among all four layers. This creates a ‘centrifugal force’ which drives the large-scale tonal plan of the piece toward distant tonal relationships (see Finlow 1992, 72).

Fig. 2: Frederic Chopin, 12 Etudes, op. 10/6. © Copyright 1973, 2005 by Wiener Urtext Edition, Musikverlag Ges.m.b.H. & Co., K. G., Wien/UT 50205

The extended function of the chromatic figuration ultimately re-defines the performer’s considerations. Normally, performing an etude requires special attention to a particular technical challenge. A successful performance of the etude amounts to the attainment of a pre-assigned goal in the performer’s learning process. However, in this particular etude, the ornamental figuration is not merely a subtle accompaniment which only requires advanced playing technique; rather, the figuration also significantly intensifies the tonal motion, the

302 | Eran Guter and Inbal Guter continuity, and the overall tension of the piece. Without it, only a bare harmonic skeleton is left, marked by reduced tension and motion. Without it, the ground for remote tonal relationships in the large-scale tonal plan weakens significantly. In this piece, the particular artistic challenge for the performer is to move beyond the normal focus in this genre on technique, and hence to reconsider the expressive range of his playing. This artistic challenge pertains to Tibby-toHelen recalcitrance in the sense that expressive playing in this piece presupposes reflexivity: acute awareness of the broad effect of the chromatic figuration on harmonic motion, tonal tension, and the large-scale tonal plan – the areas which, in the genre of the etude, patently inhibit psychological participation. Such reflexivity determines an expressive form of playing which enables the performer to clearly distinguish the different layers from one another while emphasizing the tense interaction among them as well.

4 Musical ornamentality and thoughtwriting From the vantage point of Walton’s theory of onrnamentality as psychologically inhibited games of make-believe, the gist of musical ornamentality is reflexivity. Ornamentality curtails imaginative immersion in a work-world as it deflects the appreciator back to the world of a framing game, which consists of the work itself together with the performer and his creative activity, and by implication also the composer (if he is not also the performer) and his creative activity. We suggested that the reflexivity pertaining to musical ornamentality involves a kind of double consciousness (or twofoldness) of the musical performance as a performance of the work as musical entity and of the performer’s achievement in performing the work. We also suggested that deflecting back to the framing world-game – that is, becoming interested in the means by which imaginative participation is promoted and the kinds of participation one might engage in – involves the appreciation of significant connections with our social environment and with our lives. Importantly, such double consciousness pertains not only to the impurely musical make-believe of the performer but also to that of the (non-performing) listener. As Walton (1987, 78) suggested long ago, the listener’s appreciation involves some sort of empathy with the act of making the sounds. That is, appreciation by performing is primary even when we refer to the experience of a (non-performing) listener. We can now turn to the second question, which we posed in the introduction to our essay – how may this conception of musical ornamentality intervene

Impurely Musical Make-Believe | 303

in the current formalist-narrativist debate concerning the problem of absolute music? In the previous sections of our essay, we carefully traced and explicated the situation. Paying one last tribute to our loyal literary pathfinders, Tibby and Helen, we can put matters succinctly: Tibby and Helen can be seen as belonging with each other in the sense that Helen’s introspectively imaginative gameworld is framed (and psychologically inhibited) by Tibby’s reflexive yet imaginatively enhancing game-world; musical ornamentality consists in the mutual, two-way artistic recalcitrance which traverses this dual-world formation. This dynamic theoretical construal of musical ornamentality steers clear of both the Scylla of formalism and the Charybdis of the persona theory. It is time to place the last piece in our theoretical puzzle. In a recent paper entitled Thoughtwriting – In Poetry and Music, Walton (2011) advanced a different literary model for understanding musical experience. Walton is now suggesting that music and music-making can frequently be understood in terms of what he dubs thoughtwriting, that is, as “texts” which are composed “for others to use in expressing their thoughts (feelings, attitudes)” (Walton 2011, 455). The paradigmatic example for thoughtwriting, Walton (2011, 468–469) argues, is lyric poetry, which, in contrast to novels, for instance, is especially likely to express thoughts or ideas in a manner that the reader will find particularly apt for his expression. Thus he means to deflect not only the idea of hypothetical narrators and musical personae, whose prevalence and importance in music, he believes, “have been seriously exaggerated” (Walton 2011, 459) but also reduce the prominence of novels or stories (hence the prominence of the notion of “storytelling”) in the explanation of readings (silent ones included) as performances (see Kivy 2006). “Poetry and music”, Walton (2011, 473) contends, “are strikingly similar in their propensity to function as thoughtwriting”. Thoughtwriting in music requires that we follow along, thinking the music as we listen. “In ‘performing’ music in these ways”, he (Walton 2011, 473) writes, “listeners are likely to be using the sounds (the sound types, anyway, or tokens of the types they hear) to express their own feelings or emotions, or feelings or emotions they try on, ones they experience at least in imagination”. Yet Walton does not offer any conception for a mechanism of appropriation for music. If music is a perceptual art (in contrast to literature [Walton 1990, 334]), then the application of Walton’s new literary model to music calls for an elaboration of such a mechanism. The very idea of musical performance as some sort of ‘text’ which is presented for others to use for their own expression requires further explication. We maintain that a suitable conception for a mechanism of appropriation for Walton’s new idea of thoughtwriting in music is actually ready at hand in our present suggestion to explain musical ornamental-

304 | Eran Guter and Inbal Guter ity as psychologically inhibited games of make-believe. In order to clearly distinguish Walton’s idea of appropriation in music from cases of mere arousal of thoughts (feelings, attitudes) by means of music, Walton’s notion of appropriation in music requires the sort of reflexivity which we expounded in the previous section of this paper. To appropriate musical sound, to be able to utilize it, would necessitate some reflexive know-how. Of course, this seems quite trivial when we think of thoughtwriting in the case of a listener who happens to be also the actual performer of the music. Our suggestion complements and completes Walton’s recent idea of thoughtwriting in music in the following ways. It underscores the full import of music as a performing art by locating the germ of thoughtwriting in the double consciousness of the performance as a performance of the work as a musical entity and of the performer’s achievement in performing the work. This significantly expands and enriches the scope of what we may regard as the ‘text’ offered for appropriation in musical performances. The sound offered for appropriation is imbued with understanding of the circumstances of its introduction into the listener’s life. That is, the impression the musical sound makes on us is connected with “the whole range of our language games”, as Wittgenstein (1980, 51–52) put it. This is particularly true when complex phenomena such as tunes, rhythmic patterns, and musical gestures are presented for possible use by the listener or performer, as the examples from Schoenberg and Chopin in the previous section show. It also accounts for differences in thoughtwriting across different types of music, and across different types of listeners. Ultimately, such a contextually rich conception of thoughtwriting undercuts the force of the standard formalist charge of begging the question about the purported meaning of pure instrumental music. Finally, our suggestion pulls together Walton’s insight in Mimesis as MakeBelieve that the psychological distance afforded by ornamentality “makes for less direct but more significant connections with our lives” (1990, 288) and his recent idea of experiencing something “from within”, of making an expression our own, thinking the music that we hear assertively ourselves (2011). Conceived in terms of musical ornamentality, as we suggest, the appropriation of musical sound may yield thoughts that connect us to the world in multifarious ways. In contradistinction to formalism and to persona-theory narrativism, the ultimate point of impurely musical make-believe may be to enable us to appreciate how music meshes with our lives.

Impurely Musical Make-Believe | 305

References Alperson, Philip. “The Arts of Music.” The Journal of Aesthetics and Art Criticism 50.3 (1992): 217–230. Alperson, Philip. “The Instrumentality of Music.” The Journal of Aesthetics and Art Criticism 66.1 (2008): 37–51. Bach, Carl Philipp Emanuel. Essay on the True Art of Playing Keyboard Instruments. Tr. William J. Mitchell. New York: W. W. Norton, 1949 [1753/1787]. Bell, Clive. Art. New York: Capricorn, 1958 [1914]. Collingwood, Robin George. The Principles of Art. Oxford: Clarendon, 1964 [1938]. Cone, Edward T. The Composer’s Voice. Berkeley: University of California Press, 1974. Davies, Stephen. “Contra the Hypothetical Persona in Music” [1997]. Repr. in Themes in the Philosophy of Music. Oxford: Oxford University Press, 2003, 152–168. Finlow, Simon. “The Twenty-Seven Etudes and Their Antecedents.” The Cambridge Companion to Chopin. Ed. Jim Samson. Cambridge: Cambridge University Press, 1992, 50–77. Forster, E[dward]. M[organ]. Howards End. Harmondsworth: Penguin, 1963 [1910]. Fry, Roger. Vision and Design. London: Chatto & Windus, 1920. Gombrich, E[rnst]. H[ans]. The Sense of Order: A Study in the Psychology of Decorative Art. Ithaca, NY: Cornell University Press, 1979. Hanslick, Eduard. On the Musically Beautiful: A Contribution towards the Revision of the Aesthetics of Music. Tr. Geoffrey Payzant. Indianapolis: Hackett, 1986 [1885]. Hintikka, Jaakko. “The Longest Philosophical Journey: The Quest of Realism as a Common Theme in Bloomsbury.” The British Tradition in 20th Century Philosophy. Proceedings of the 17th International Wittgenstein Symposium. Eds. Jaakko Hintikka and Klaus Puhl. Vienna: Hölder-Pichler-Tempsky, 1995, 11–26. Kant, Immanuel. The Critique of Judgement. Tr. J. C. Meredith. Oxford: Oxford University Press, 1952 [1790]. Kivy, Peter. Music Alone: Philosophical Reflections on the Purely Musical Experience. Ithaca, NY: Cornell University Press, 1990. Kivy, Peter. The Fine Art of Repetition: Essays in the Philosophy of Music. Cambridge: Cambridge University Press, 1993a. Kivy, Peter. “Listenings: A Response to Alperson, Davies, and Howard.” The Journal of Aesthetic Education 27 (1993b): 22–30. Kivy, Peter. Introduction to a Philosophy of Music. Oxford: Oxford University Press, 2002. Kivy, Peter. The Performance of Reading: An Essay in the Philosophy of Literature. Malden, MA: Blackwell, 2006. Kivy, Peter. Antithetical Arts: On the Ancient Quarrel between Literature and Music. New York: Oxford University Press, 2009. Kristeller, Paul Oskar. “The Modern System of the Arts” [1951–1952]. Repr. in Renaissance Thought and the Arts: Collected Essays, exp. ed. Princeton, NJ: Princeton University Press, 1990, 163–227. Levinson, Jerrold. Music, Art, and Metaphysics. Ithaca, NY: Cornell University Press, 1990. Maynard, Patrick. “Real Imaginings.” Philosophy and Phenomenological Research 51.2 (1991): 389–394.

306 | Eran Guter and Inbal Guter Newcomb, Anthony. “Action and Agency in Mahler’s Ninth Symphony, Second Movement.” Music and Meaning. Ed. Jenefer Robinson. Ithaca, NY: Cornell University Press, 1997, 131– 153. Robinson, Jenefer. Deeper than Reason: Emotion and Its Role in Literature, Music, and Art. Oxford: Oxford University Press, 2005. Schenker, Heinrich. Harmony. Tr. Elisabeth Mann Borgese. Chicago: University of Chicago Press, 1954 [1906]. Schoenberg, Arnold. Style and Idea. Tr. Leo Black. Ed. Leonard Stein. Berkeley: University of California Press, 1975. Schoenberg, Arnold. The Musical Idea and the Logic, Technique, and Art of Its Presentation. Ed. and tr. Patricia Carpenter and Severine Neff. New York: Columbia University Press, 1995. Tormey, Alan. The Concept of Expression: A Study in Philosophical Psychology and Aesthetics. Princeton, NJ: Princeton University Press, 1971. Walton, Kendall L. “What Is Abstract about the Art of Music?” The Journal of Aesthetics and Art Criticism 46 (1986): 351–364. Walton, Kendall L. “Style and the Products and Processes of Art.” The Concept of Style, rev. ed. Ed. Berel Lang. Ithaca, NY: Cornell University Press, 1987, 72–103. Walton, Kendall L. “The Presentation and Portrayal of Sound Patterns.” Human Agency: Language, Duty and Value. Eds. Jonathan Dancy, Julius Matthew Emil Moravcsik, and Christopher Charles Whiston Taylor. Palo Alto, CA: Stanford University Press, 1988, 237–257. Walton, Kendall L. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, MA: Harvard University Press, 1990. Walton, Kendall L. “Reply to Reviewers.” Philosophy and Phenomenological Research 51.2 (1991): 413–431. Walton, Kendall L. “Listening with Imagination: Is Music Representational?” The Journal of Aesthetics and Art Criticism 52 (1994): 47–61. Walton, Kendall L. “Projectivism, Empathy and Musical Tension.” Philosophical Topics 26 (1999): 407–440. Walton, Kendall L. “Thoughtwriting – In Poetry and Music.” New Literary History 42 (2011): 455–476. Walton, Kendall L. “Two Kinds of Physicality in Electronic and Traditional Music.” Bodily Expression in Electronic Music: Perspectives on Reclaiming Performativity. Eds. Deniz Peters, Gerhard Eckel, and Andreas Dorschel. New York: Routledge, 2012, 114–221. Wittgenstein, Ludwig. Culture and Value. Tr. Peter Winch. Ed. Georg Henrik von Wright. Chicago: University of Chicago Press, 1980. Wollheim, Richard. Painting as an Art. Princeton, NJ: Princeton University Press, 1987.

| Section 4 – Games

E. M. Dadlez

Make-Believe Wickedness vs. Wicked Making-Believe RPGs, Imagination and Moral Complicity A lot has been said in the philosophical literature about the extent to which fictions can make us complicit in the perspectives they endorse or project. Even the most controversial allegations of complicity, however, invariably distinguish between the mere depiction and the outright imaginative adoption of morally problematic perspectives (for instance, a perspective from the vantage point of which genocide is an attractive course of action). Simply to show that there are likely to be people who harbor immoral perspectives on the world, and to show further how it is they come to see things that way (and perhaps why some circumstances make those perspectives difficult to resist) is not yet to endorse them. Even those who claim that merely to imagine the rightness of genocidal perspectives makes the imaginer complicit in them will nonetheless exonerate readers or viewers who merely imagine some person’s approval of genocide. It is, after all, possible to imagine people’s problematic attitudes without on that account imagining genocide approvingly. To imagine genocide approvingly or to imagine it right is held to be problematic, however. Since one cannot imagine what one cannot conceive, an ability to imagine the rightness of genocide is thought to tell us something rather disturbing about the imaginer’s conception of rightness. A distinction corresponding to that outlined above can be drawn between the content of one’s imagining and the manner in which one imagines that content. So, as is evident, such grounds for moral criticism of fiction are pretty much restricted to cases in which there is an outright endorsement of morally suspect propositions – that is, cases in which it is true in the fictional world that some ascertainably noxious course of action is correct or admirable. While this enables a criticism of such things as violent pornography and white supremacist fiction, it leaves the majority of fiction unscathed as regards moral criticism. The glaring potential exception here, however, is that of role-playing games (or RPGs), participation in which seems to mandate the imaginative adoption of troubling character attitudes and so condemn the majority of RPGs to charges of immorality. That is, if we undertake an analysis of role-playing in terms of Kendall Walton’s account of make-believe (1990), an account to which RPGs seem fortuitously adaptable, it appears to involve the adoption of character attitudes

310 | E. M. Dadlez from a first-person perspective. One would not merely make believe or imagine that the character took a pro-genocide stance. One would instead make believe or imagine approving of genocide, in accordance with one’s assigned role. To imagine being a character would, using this line of reasoning, necessitate adopting character beliefs and attitudes in imagination. To role-play would be to make believe believing and thinking and feeling as the character did. That would be equivalent to imaginatively entering into some work’s suspect endorsement and so make participation in RPGs morally suspect in a vast number of cases. In what follows, I will outline some historical contexts in which questions about complicity have arisen and then move on to the applicability of such intuitions to gaming. I will try to give a clear idea of what role-playing, especially live action role playing, involves. Role-playing scenarios are more complex than one might suspect, and it is worthwhile to focus on specific examples in order to test the applicability of various claims. I will offer a few examples from my own experience. Finally, I will return to a selection of contemporary theoretical perspectives on the complicity issue in order to explore their implications for the material canvassed, and for RPGs in general. Historically, the point of interest in the philosophical literature has been less focused on complicity than on our imaginatively resisting it – resisting, that is, points of view that we regard as morally troubling. Does this make complicity the default assumption, with resistance taking center stage as an intriguing deviation from the norm? Whether or not that is the case, such considerations go back to David Hume’s Of the Standard of Taste, in which he proposes that we will resist imaginative immersion in morally suspect perspectives: [W]here the ideas of morality and decency alter from one age to another, and where vicious manners are described, without being marked with the proper characters of blame and disapprobation; this must be allowed to disfigure the poem, and to be a real deformity. I cannot, nor is it proper I should, enter into such sentiments; and however I may excuse the poet, on account of the manners of his age, I never can relish the composition […]. We are not interested in the fortunes and sentiments of […] rough heroes: We are displeased to find the limits of vice and virtue so much confounded: And whatever indulgence we may give to the writer on account of his prejudices, we cannot prevail on ourselves to enter into his sentiments, or bear an affection to characters, which we plainly discover to be blameable […]. The case is not the same with moral principles, as with speculative opinions of any kind. These are in continual flux and revolution [...]. Whatever speculative errors may be found in the polite writings of any age or country, they detract but little from the value of those compositions. There needs but a certain turn of thought or imagination to make us enter into all the opinions, which then prevailed, and relish the sentiments or conclusions derived from them. But a very violent effort is requisite to change our judgment of man-

Make-Believe Wickedness vs. Wicked Making-Believe | 311 ners, and excite sentiments of approbation or blame, love or hatred, different from those to which the mind from long custom has been familiarized. And where a man is confident of the rectitude of that moral standard, by which he judges, he is justly jealous of it, and will not pervert the sentiments of his heart for a moment, in complaisance to any writer whatsoever. (Hume 1987b, 246–247)

Tamar Gendler (2000) investigates this phenomenon, and proposes the possibility of imaginative investment that is free of moral or epistemic complicity. Kendall Walton focuses on the Humean distinction between moral principle and speculative opinion, wondering why the endorsement of questionable moral stances is so imaginatively off-putting when imagining ‘speculative errors’ doesn’t throw us for a loop. “There is science fiction”, he points out, “why not morality fiction?” (Walton 1994, 37). He suggests that imaginative resistance might involve an individual’s inability to comprehend how the relation of dependence between moral properties and those natural properties upon which she believes they supervene could differ from what she believes it to be (1994, 46). A number of accounts have been offered along such lines, the most straightforward of which simply claim that it is impossible to imagine what one cannot in the first instance conceive. So imagining the rightness of what would ordinarily be regarded as a wrong act would be impossible unless one believed it possible for acts of that kind to be permissible (though perhaps only in limited circumstances, such as those occurring in the fiction). Since one cannot imagine what one cannot conceive, one’s ability to imagine the rightness of the act in question attests to some belief about the possibility of its being right. This is the way in which claims of moral complicity might be grounded. And RPGs, at least on the face of it, seem to offer far more scope for moral complicity in dubious perspectives than run-of-the-mill fiction. This is because RPGs quite frequently seem to require imaginative immersion in the particular world views of depraved or evil characters whereas imaginative engagement in fiction typically does not. Let us explore the nature of RPGs. Often regarded as a form of interactive and collaborative story-telling, RPGs involve the player in the collaborative construction of a fictional world. Roleplaying games are not, in other words, an advanced form of Monopoly. Players take on character roles in a pre-established fictional setting, narrating or acting out character actions and decisions. What is fictionally or make-believedly the case is established initially by the pronouncements of the Game Master (GM), who plays a role very loosely analogous to that of the omniscient narrator of a work of fiction. The GM will continually update the players on what events transpire in the world of the game. What is make-believedly true is also determined

312 | E. M. Dadlez by the pronouncements of the players themselves concerning their characters’ attempted or intended actions, the success or failure of which is determined by a roll of the die in concert with the pre-established parameters concerning every character’s skill set, capacities, and resources (as adjudicated by the GM). If one’s character is physically fit and possessed of formidable martial arts skills, for instance, attempting to grapple with a fictional enemy is likely to prove successful. With no martial skills in one’s repertoire, hiding is likely to be more conducive to one’s survival. It should be remembered that it is only within a player’s power to make an attempt to act. Whether it succeeds or not is a matter of luck and one’s arsenal of talents and documented resources. A player’s words – at least those spoken from a perspective within the world of the game – make it fictional that the character utters those very things. A character could be a compulsive liar, of course, so it is simply true that the utterance occurred in the world of the game, not necessarily that it correctly describes fictional states of affairs. A player’s third-person rather than first-person pronouncements about his or her character’s mental state, beliefs, and history – information that is typically supplied to the GM at the beginning of the game and sometimes to other players – is true in the world of that game, at least once it has passed muster in whatever review process is required for that game. Personal character histories are sometimes subject to a consistency check (e. g. against the character’s established skill set) and to organizational verification and approval in the case of RPGs with a wide (sometimes international) participant pool. An example of this will be given later. For the moment, however, it is enough to point out that an approved list of character beliefs, preferences, and history also provides us with a set of propositions taken to be true in the world of the game. Depending on the RPG, a player may devote a considerable amount of time and energy to developing a character’s backstory and to inventing a set of character quirks and susceptibilities, provided none of these are at odds with the character’s established set of skills and capacities. Even more energy may be devoted to getting a convoluted character history and bloodline approved by the RPG parent organization. In the case of larger live-action role-playing (LARP) organizations like the Camarilla, the world of the game encompasses thousands of players (perhaps tens of thousands, if non-dues-paying participants are counted) the personal histories of whose characters are vetted for consistency within the world of the larger game. The Camarilla was the “sanctioned massive multiplayer live-action experience for White Wolf’s World of Darkness”, according to their Wiki,1 a club that is now coming out as a self-sustaining non-profit organi|| 1 White Wolf. Camarilla Wiki. http://wiki.white-wolf.com/camwiki/index.php?title=About_-

Make-Believe Wickedness vs. Wicked Making-Believe | 313

zation. In LARP, a player’s actions or, more accurately, some subcategory thereof, make it fictional that the character behaves in that way. It is also frequently the case that some abbreviated gesture makes it fictionally the case that some specified power (or superpower, as the case may be) has been exerted in the world of the game. Whether it has the effect the player intends is once again determined by the ubiquitous ten-sided die. These are established conventions to which all players quickly grow accustomed. All of this seems relatively straightforward, allowing us to regard the RPG as a kind of group project, in which participants are co-authors of a fictional world – collaborating architects, each of whom contributes to the construction of the world of the game by generating make-believe truths. The complicity issue arises not because of this group authorship, but because of the first-person vantage point that role-players adopt. If the role-player makes-believe and thereby imagines being the character, and that character has been, as the result of the player’s own invention, endowed with certain traits and beliefs and preferences, does not the player imagine having those beliefs and desires and preferences? Notoriously, RPGs are infested with unapologetically wicked and nefarious characters. Indeed, many RPGs are devoted to such characters almost exclusively. Take RPGs in which all of the characters are vampires, for instance. The world of the game is entirely devoted to the activities of beings who regard the human race as a food source. Filled with contempt for the insipid pleasures of the Twilight series of books and films, members of the Camarilla scoff at the metaphorically vegan proclivities of Twilight’s heroes. Camarilla players obsess about a character’s clan and covenant and bloodline, about attributes and skills and blood potency. They obsess even more about clan politics. And there are clear moral conflicts in the world of the game, especially regarding the power plays of various characters. Very few of these, however, will involve moral qualms about using any stray human being as the equivalent of a protein shake in a handy easy-to-open container. Developing a character for the Camarilla involves developing a character history consistent with the (initially limited) skills and attributes one has selected at the outset of the game, and which increase in scope and degree with every game played and sometimes with every feat achieved. Establishing one’s character as the inheritor of a powerful bloodline, for instance, enables one to acquire, via experience points, additional skills and abilities available only to the members of those bloodlines, something that can prove extremely advantageous in the world of the game. Because a character’s possession of some par|| the_cam (27 June 2012).

314 | E. M. Dadlez ticular bloodline presumes historical connections to other characters in the fictional world, a complex approval process is required. One must either get a player whose character already possesses the bloodline to go along by agreeing to make it fictionally the case that she or he sired one’s character (a procedure that could place one’s own character under their control) or one must invent a story about the bloodline’s acquisition that remains consistent with states of affairs in the world’s work by eliminating inconsistencies. In the case of my own character, a nineteenth-century Polish vampire named Haecceity (Hex, for nonphilosophers), I eliminated inconsistencies by making her a patricide. Time did not improve Hex’s character. In latter days, she became an arms smuggler, invariably supplying military aid and equipment to the underdog, subsisting on Janjaweed and Fritos. It was true in the world of the work that Hex was a meddler in political affairs, something that was a general annoyance to the vampire hierarchy in that it threatened to expose the existence of vampires to their food supply. But it was also true in that world that she had very few qualms about exploiting or killing individual people. If she preferred to kill those of whose politics she disapproved, self-interest would still always quell squeamishness in a pinch. To make believe that one is this kind of a person is to make believe that one is both ruthless and exploitative, and to further imagine that these things are an exhilarating expression of the imposition of one’s will upon the world. The first person imaginative stance is clearly problematic, but there remain questions about the extent to which such a stance is actually taken up. My principal enjoyment of groups such as the Camarilla has always involved the invention of characters and histories, as delineated above. The creative challenges, the formulation of justifications, the weaving together of disparate attributes in order to create a colorful persona – all of these things are engrossing. The games themselves, in my estimation, are very often not. Admittedly, a couple of colleagues and I joined up mainly in order to facilitate doing a little philosophy in relation to the RPG question. We also signed on in order to enjoy ourselves, of course, but the LARP aspect of things proved a profound disappointment at least half the time. In my experience, one spends far too much of one’s time standing about interminably, waiting for the possible instantiation of one’s prospective actions to be ascertained by rolling ten-sided dice or pulling cards. Ironically, mortal combat has probably the most potential to become truly soporific, as the prospects for the success of each and every move of each and every player (and each and every runner away) is calculated, in a kind of glacial slow-motion number-fest. Any prospect of imaginative immersion is dead as the proverbial doornail. Surprisingly, more imaginative involvement can attend those moments when LARP reverts to tabletop strategies

Make-Believe Wickedness vs. Wicked Making-Believe | 315

and the impossibility of physically miming the proposed actions leads to a series of narrative descriptions of what one’s character is doing. These can be more emotionally engrossing than attempts at physical replication. It is only fair to say that I am reliably informed that live-action roleplay in venues inaccessible to me proceeds in an infinitely more engaging manner and has largely dispensed with the deceleration of the game on account of necessary calculations.2 Jeepform LARPing,3 for instance, is designed to minimize interruptions and to standardize GM directives (complex games will involve more than one GM) in such a way as not to inhibit or curtail the imaginative immersion of participants. Game plots differ quite radically from the action/adventure format that I have described hitherto, as do varieties of GM involvement in orchestrating the game. Some suggestions for GMs in these contexts frankly make their roles seem much more like those of omniscient narrators in works of fiction. The GM may be encouraged to announce startling events in startling way: Four persons are walking through a dark forest. Suddenly, the game master cries out: ‘What was that?!’ Depending on the game, the GM may also inform the player what his or her character is thinking or feeling, or what his or her character may do. The variations on roleplay are complex, and this is not the place to speculate at any length about the effects this has on player immersion except to acknowledge that it is probably very much enhanced. The events in many Jeepform games occur in real time and the focus seems to be on character interaction and on communication. It is very clear, moreover, that some games can prove intentionally disturbing or claustrophobic, even psychologically grueling: The Journey, by Fredrik Axelzon A desolate barren winter landscape. Metal wheels against rusty tracks and a perpetual pendulum. Hungry and exhausted. Strangers and shadows. Carrying shards of the past. Will they ever reach their destination? Is it only a dream? Hope is waning, fatigued. Only the name remains. The turning point. The journey – a scenario that takes the players for a dark and quiet ride through a frozen landscape in the shadows of the apocalypse of civilization. Focus is placed on depicting the fatalistic mood, barren dialogues and a constant tone of silence. The scenario

|| 2 Many thanks to Mark Silcox, who is much better informed about the state of gameplay in the world at large than I. Any mistakes in this paper are entirely my own fault. 3 Vi åker jeep – the home of Jeepform. http://jeepen.org/games/ (27 June 2012).

316 | E. M. Dadlez is scene-based and influenced by Cormac McCarthy, jeep format and classic Swedish freeform of 2010. Players: Duration: Capsule definition: Premier:

4 Approx 4 hours Barren landcape, horror, silence, despair Fastaval 2010 (winner for Audience Award)4

Clearly, there is an enormous potential in roleplay for imaginative immersion, whether or not that potential is always realized. And despite the difficulties and interruptions that afflict most games launched for the benefit of somewhat less dedicated participants, it cannot be denied that there are moments in both LARP and tabletop RPGs of the action/adventure variety during which emotional and imaginative engagement amount to more than the simple pretense that was my own roleplay staple and fallback (and that I assume is typical of the experience of other players of the action/adventure genre of games as well). When someone shoots one’s character in the head at the behest of power figures in the vampire hierarchy for reasons which are clearly indefensible, one will feel nearly as miffed as one’s character is supposed to be (the character is miffed rather than dead because she is bullet-resistant in the world of the game). My own experience of live action roleplaying (circa 2011) bears this out. The following replicates our dialogue to the best of my recollection: Nimrod Vampire Flunky [unexpectedly whips out pistol in form of finger and shoots Hex in the head. Speaks apologetically]: Bang. Nothing personal. Just following orders. Hex [outraged by turn of events both within and outside of the world of the game]: A gun? Really? Are you kidding me? GM: Three points worth of damage [Hex’s brains are leaking nastily in the world of the game]. Hex: I stab the expletive. Nimrod: Hey! GM: With what? You’re not armed. Hex: I stab him with my ballpoint! I stab him in the eye! Nimrod [defensively]: It wasn’t my idea! Hex: What, your nimrod flunky status is just an autonomy-sucking black hole that exonerates you of any responsibility for your actions?

There is a reason why drinking is not permitted during Camarilla games. Emotions can run high at least in part because character lives are frequently at risk. || 4 Fredrik Axelzon, The Journey, Vi åker jeep – the home of Jeepform. .

Make-Believe Wickedness vs. Wicked Making-Believe | 317

That sort of play gives rise to both actual and fictional outrage and distress. The elimination of a character to the development of which one has devoted time and energy is more than a mere annoyance. It heralds not only the inability to use the persona on which one has lavished so much attention, but the need to start from scratch in developing a new one. And starting from scratch generally involves a kind of demotion, the loss of all experience points and powers. No more ability to read minds (or compel cooperation or teleport to safety). One is back at ground zero, powerless relative to the majority of one’s fellows. It places one at a considerable disadvantage. Injuries in the world of the game also restrict or inhibit one’s ability to employ one’s talents and powers, and so again represent a disadvantage, one that a player may (unsurprisingly) feel was arbitrarily or unfairly imposed. To borrow from Aristotle, where events in the world of the game do not follow one another of necessity but instead take on a suspiciously deus ex machina flavor that favors some characters over others, players may become irritated with actual people, such as the GM, rather than, or in addition to, fictional ones. In the example above, my annoyance had both fictional and actual objects. Make-believedly, I was annoyed with the flunky who shot me and with his misguided vampire overlords. Actually, I was annoyed with all of the players who thought this would be a good way to move the plot forward. Clearly, one kind of reaction can bleed over into the other. Imagined distress over damage to a character can quickly become irritation with the GM for permitting incapacitating events to unfold. Annoyance over an illconsidered player decision can underwrite imagined outrage on behalf of one’s character over another character’s action. I am inclined to depart from Walton’s account of our emotional response to fiction in that I believe our affective response to fictional events usually involves real emotion. That is, I believe genuine emotions can take fictional objects and Walton does not, maintaining that such affective reactions are only quasi-emotional. I would say the annoyance was real in both cases. Whichever account one prefers, however, it seems clear that we can posit a strong connection between active imagination and affective response and go on to link this combination of affect and cognitive activity to the kind of complicity some philosophers target. I treat affective response here as a kind of watershed that enables us to distinguish between make-believe that is mere pretense and make believe that involves what Richard Moran would call active imagination (1994). Kendall Walton’s account of make-believe in his groundbreaking paper Fearing Fictions from 1978 was initially misinterpreted in some quarters in such a way as to ally making believe with pretense rather than imagination. But merely pretending to take up an attitude does not necessitate experiencing the affective response that would typically accompany that attitude’s possession. As pretense, making-

318 | E. M. Dadlez believe that one is in danger need not afflict one with any of the symptoms of fear, for instance. One can pretend to be in danger and to be afraid without trembling, without a rapid pulse, without shallow breathing, whether we think of these as the symptoms of terror or of quasi-terror. If making-believe is thought of as vivid imagining, however, an affective response seems a typical concomitant. Actively imagining that one is in immediate peril (or that some third party is) – the kind of imagining that occurs when one is entirely engrossed in a film or novel or play – seems much more likely to produce an affective reaction. The concern about RPGs would focus on a role-player’s taking up the perspective of an immoral character. Merely imagining that there is a person with certain despicable traits is not morally problematic in any necessary way. To invent a character, to make it fictional in a variety of ways that the character makes certain decisions and performs certain actions, even to make it fictional that the character feels a certain way about those decisions and actions, does not require the imaginative adoption of character attitudes. So, on the one hand, we can have some hypothetical role-player who pretends to be nefarious, and who both imagines and makes-believe that her character is triumphantly empowered by destructive and exploitative actions. This is that kind of activity in which authors of many kinds of fiction engage – they imagine characters and character attitudes without necessarily imaginatively inhabiting them. A role-player who merely pretends makes inferences about her character’s probable reactions to events based on the history she has devised and the set of skills and attributes with which her character is endowed. If engaged in LARP, she will attempt to exhibit behaviors that reflect the attitudes she assumes the character would have. On the other hand, we can have a role-player who imagines being nefarious. And what this would amount to, I think, is imagining those exploitative and destructive actions triumphantly, with exhilaration and pleasure, as proofs of personal achievement. It is always possible, of course, to regard one’s character’s actions as a kind of strategic triumph of gaming acumen on any number of grounds: that of maintaining character authenticity, say, or of finessing a larger-than-usual quantity of experience points. There, the object of one’s pleasure is real and is believed to be: one’s own skill or inventiveness. However, what I have in mind here are affective responses like triumph or pleasure which are directed toward distinctly fictional targets – for instance, the abuse of a living person within the world of the game. Proponents of the view that fictions can make us complicit in the perspectives they endorse would almost undoubtedly consider the adoption of a triumphant and approving attitude toward imaginary events to be the equivalent of entering into an endorsement – in this case of despicable behavior. If one imagines being a despicable person, at least of the

Make-Believe Wickedness vs. Wicked Making-Believe | 319

kind I have tried to describe, then one imagines believing and feeling as that person would, entertaining the thought of various nefarious actions with pride and approval (or quasi-pride and quasi-approval, if one opts for Walton’s account). To return to the approach originally canvassed, it would be maintained that the ability to imagine the abuse triumphantly and approvingly suggested something about one’s core beliefs – one would have to believe it possible for abuse to be good or desirable if one was capable of imagining it approvingly or imagining it as right, since one cannot imagine what one cannot conceive. A shadow is cast here over the conception of desirability or rightness and, indeed, over what can be taken to constitute a triumph. And this is not the only line of argument enabling us to cast moral aspersions on our hypothetical role-player. Berys Gaut proposes that: [A]ttitudes [as taken up toward fictitious objects, in the context specified above] are partly directed toward kinds, not just individuals. When the rape fantasist imagines his fictional women, he is imagining them as women, that is, as beings of a kind that also has instances in the real world; and that he imagines them as women is, of course, essential to his imaginative project. Thus, by virtue of adopting such an attitude toward his imagined women, he implicitly adopts that attitude toward their real-life counterparts. (Gaut 1998, 187–188)

This dovetails in several respects with a point made earlier by Bijoy Boruah in his (unjustly undervalued) bid to lay the paradox of fiction to rest. Boruah ties our affective responses to fiction to evaluative beliefs about universals. Thus, fear for a beleaguered character in a fiction would be in part grounded in the belief that to be placed as the character is placed would be dangerous (see Boruah 1988, 108–117). Triumph and pleasure in effecting some character’s annihilation would, in the same way, be tied to some belief that to exert control over another in such a way can amount to a personal achievement. This is not an existentially committed belief, granted, but it is a belief that clearly applies to the world as well as the work, and it is a belief liable to ethical assessment. The properties on the instantiation of which the emotion depends (danger for fear, personal achievement for triumph, underserved suffering for pity) involve conceptual and epistemic commitments on our part – in regard to what is believed to be dangerous, say, or respecting what can count as an achievement or what can constitute desert. Many of these have obvious moral overtones, and would be directly implicated in responses to fiction of the kind under consideration. Aaron Smuts’ work on taking pleasure in the pain of fictional others may impli-

320 | E. M. Dadlez cate RPGs as well.5 Smuts proposes that it is bad to delight in the suffering of fictional characters, or at least in their undeserved suffering. The claim here is that fiction may encourage responses that it is morally bad to have. RPGs encourage just such reactions to the extent that players take up first person perspectives of the kind canvassed above, and might therefore be criticized for these reasons. Having reviewed the preceding line of argument, it becomes important to distinguish between an indictment of RPGs in general, and an indictment of the way certain players approach the game. Once that distinction is made, it is evident that the outlook for RPGs may not be so grim. No special moral indictment would attach to RPGs of any stripe that would not also attach to any work of fiction with a villain, except in cases where some immoral perspective were taken to be true in the world of the game as a whole. That is, an RPG in which it was true that genocide was admirable and right could be criticized on the same grounds as white supremacist works like The Turner Diaries. Yet it would be difficult to claim, even in the case of organizations like the Camarilla, that overarching and false moral assumptions (e. g. assumptions that exploitativeness was invariably laudable or that destructiveness ranked as an achievement) were true in the world of each and every game. The proliferation of villainous types in RPGs, on the other hand, introduces the possibility of a player’s adopting problematic perspectives. But this is more or less comparable to the case of any fiction featuring a villain. Consider that it is always possible, even in the case of a fiction which extols every virtue, to empathize with the conceited villain and entertain his villainy with approval or smugness in imagination. The hypothetical reader has missed the point of the book, we might argue, or has demonstrated something quite unattractive about himself, but we certainly cannot condemn the book just on account of making such a thing possible. So while there may well be a case against imaginative immersion in villainous viewpoints, this would not amount to a wholesale case against RPGs. Also, it is unclear how often imaginative immersion in morally problematic roles actually occurs. It is, as has been indicated, quite possible to imagine a villainous viewpoint in the third person, to imagine that there is one, without imagining villainously. One can roleplay without imaginatively adopting the character’s moral stance from a first-person perspective. Furthermore, action/ adventure RPGs, the very games that feature the most floridly noxious villains, are not particularly conducive to thoroughgoing and prolonged imaginative || 5 Aaron Smuts, Pleasurably Regarding the Pain of Fictional Others, draft manuscript, 2012. I am grateful to Aaron for sharing his work.

Make-Believe Wickedness vs. Wicked Making-Believe | 321

immersion simply because of the necessary and virtually constant interruptions caused by dice-rolling, card pulling, character-sheet-checking, and out-ofcharacter colloquy with the GM and other players. We move in and out of imaginative immersion when we read novels and watch films as well, but there are more, and more continual, obstacles to immersion in the case of these types of RPGs. This raises an entirely new set of questions that will have to be explored at length elsewhere but that are worth mentioning here. It is a little disturbing to find that features that make a game less imaginatively and artistically engaging are the very ones that prevent the kind of problematic moral complicity being investigated.6 The fact is, that games are almost always better and more exciting (by way of artistry and interest and their capacity to engross) when character perspectives are thoroughly explored and exploited. One of the most effective ways of doing that is by means of the adoption of the alien perspective in imagination. Moreover, the adoption of the problematic moral perspective can contribute to the enhancement of art and enjoyment and thereby serves a good rather than a problematic end. So any argument regarding complicity would eventually have to weigh the costs in possible habituation against the benefits of creating a better overall narrative. Good art, or at least good games, might involve moral risk. At present, however, it appears that any case against RPGs is weak. The very fact that we transition between fictional and extra-fictional perspectives muddles the objects of our affective attitudes. It can be quite unclear whether a player’s triumph concerns her strategy as a player or her choice of action in the world of the work. Affective responses to in-game and extra-game events become tangled in such a way that it is difficult to distinguish among them. And even fictional objects of affective reactions might be difficult to determine. One might be proud of having succeeded at some endeavor without on that account being overjoyed about the way one went about it. Haecceity might be chuffed about ridding the world of a dangerous warlord without on that account being thrilled that she used up a couple of human beings in the process. The upshot is simply that RPGs, like more conventional fiction, might sometimes endorse morally problematic perspectives or might simply describe such perspectives. Only the former provides a possible reason for criticizing the fiction itself. Entering into a morally suspect perspective in imagination would put the onus on the imaginer, not the work.

|| 6 Thanks to Mark Silcox for raising this troubling point.

322 | E. M. Dadlez

References Axelzon, Fredrik. The Journey. Vi åker jeep – the home of Jeepform. http://jeepen.org/games/thejourney/ (n.d.) (27 June 2012) Boruah, Bijoy H. Fiction and Emotion: A Study in Aesthetics and the Philosophy of Mind. Oxford: Clarendon Press, 1988. Gaut, Berys. “The Ethical Criticism of Art.” Aesthetics and Ethics: Essays at the Intersection. Ed. Jerrold Levinson. Cambridge: Cambridge University Press, 1998, 182–203. Gendler, Tamar. “The Puzzle of Imaginative Resistance.” Journal of Philosophy 97.2 (2000): 55–81. Hume, David. Treatise of Human Nature. Ed. Lewis Amherst Selby-Bigge. Oxford: Clarendon Press, 1978. Hume, David. “Of the Standard of Taste.” Essays: Moral, Political and Literary. Ed. Eugene F. Miller. Indianapolis: Liberty Classics, 1987b, 226–249. Moran, Richard. “The Expression of Feeling in Imagination.” Philosophical Review 103 (1994): 75–106. Smuts, Aaron. “Pleasurably Regarding the Pain of Fictional Others,” draft manuscript, 2012. Walton, Kendall L. “Fearing Fictions.” Journal of Philosophy 75 (1978): 5–27. Walton, Kendall L. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Walton, Kendall L. “Morals in Fiction and Fictional Morality (I).” Proceedings of the Aristotelian Society Supplement 68 (1994): 27–50. White Wolf. Camarilla Wiki. http://wiki.white-wolf.com/camwiki/index.php?title=About_the_cam (n.d.) (27 June 2012).

Jason D’Cruz

Agency and Volition in Make-Believe Worlds Second Life, an online multi-user environment created by Linden Labs in 2003, resists categorization either as game that an individual may play, or as site where an individual may reside. In 2014 there were about 1 million active users /inhabitants.1 Second Life players, called residents, interact with each other through avatars, graphical representations of their online alter-egos. Residents converse, learn languages, form relationships, run businesses, publish magazines, exhibit art, perform live music, engage in scientific collaboration, and participate in religious organizations. The population of Second Life also includes non-player characters (NPCs), avatars that are animated by computer programs rather than by people. A prominent feature of Second Life, one that makes it particularly philosophically interesting, is the lack of a goal, implicit or explicit, that is stipulated by the game design. In Second Life, you are not given a quest. Residents are not tasked to save a princess, unravel a mystery, or build an empire. Rather, they must set their own goals and decide for themselves what is worth pursuing. It only makes sense to talk of winning or losing Second Life to the extent that it makes sense to make those same kinds of judgments of real life. Success is relativized to particular goals, and those goals are determined by individual players. Common goals include finding a soul mate, learning a skill, creating a community, and striking it rich. (Second Life has its own currency, the Linden dollar, used to buy and sell virtual goods and services). This absence of a gamefurnished goal makes the virtual environment compelling and immersive for some, and utterly mystifying for others. Playing Second Life requires residents to initiate actions that are consistent with what Bernard Williams (Williams and Smart 1973) called their ground projects, thereby inviting players to exercise their capacity for agency in the virtual world. Some will find this invitation tantalizing, an occasion to transcend corporeal limitations and to live a literal second life. Others will find the invitation absurd, seeing actions in the virtual world as without significance. This investigation into human agency in virtual worlds begins with a short case study of marital infidelity on Second Life. I examine considerations adduced by J. David Velleman in favor of the view that Second Life residents exercise real agency while animating their virtual bodies. I then argue that im|| 1 Current user metrics for Second Life. http://www.lindenlab.com/releases/infographic-10years-of-second-life (13 May 2015).

324 | Jason D’Cruz portant disanalogies between Second Life and real life should give us pause in attributing full-blooded personhood, and therefore full-blooded agency, to Second Life residents. In particular, I argue that residents are less likely to have volitions, desires about the content of their will, in the sense articulated by the philosopher Harry Frankfurt (1971). Ultimately, the degree to which residents have volitions is a function of the degree to which they care about the lives of their alter-egos. This is a contingent matter that will vary greatly among individual players. However, I argue that structural features of Second Life make it less likely that individuals care about their Second Life avatars in the same way that they care about their real lives. Finally, I point out ways that the analysis of agency in Second Life illuminates aspects of agency in the real world.

1 Velleman on virtual agency A widely-discussed Wall Street Journal article chronicled the life of Second Life player Ric Hoogestraat, a 53-year-old man with a long ponytail, graying sideburns, a salt-and-pepper handlebar mustache, and a large paunch. His look is described as that of a “cross between a techie and the Grateful Dead fan that he is” (Alter 2007). During different periods of his life, Hoogestraat worked as an elementary school teacher, a call center operator, a computer graphics instructor, and a vendor for herbs and essential oils at Renaissance fairs. But Mr. Hoogestraat also has another persona. For an average of six hours a day, and fourteen hours at a stretch on weekends, he inhabits an avatar named Dutch Hoorenbeek in Second Life. His life as ‘Dutch’ is rather different from his life as ‘Ric’: [Dutch] looks like a younger, physically enhanced version of [Ric]: a biker with a long black ponytail, strong jaw and thick handlebar mustache. In the virtual world, he’s a successful entrepreneur with a net worth of about $ 1.5 million in the site’s currency, the linden, which can be earned or purchased through Second Life’s Web site at a rate of about 250 lindens per U.S. dollar. He owns a mall, a private beach club, a dance club and a strip club. He has 25 employees, online persons known as avatars who are operated by other players, including a security guard, a mall concierge, a manager and assistant manager, and the “exotic dancers” at his club. He designs bikinis and lingerie, and sells them through his chain store, Red Headed Lovers. (Alter 2007)

Ric Hoogestraat’s parallel existence has proven rather distressing for his long suffering (real) wife, Sue Hoogestraat, who pays household bills, cooks, does laundry, takes care of their three dogs, and empties ashtrays around the house. Even worse from her perspective is the fact that in Second Life, Ric is married to

Agency and Volition in Make-Believe Worlds | 325

a tall, wiry redhead named Tenaj Jackalope, an avatar played by a 38-year-old Canadian woman whom he has never met in person. According to the article, ‘Dutch’ and ‘Tenaj’ share a remarkable degree of intimacy and trust; Ric and Sue, not so much. The case raises a number of fascinating psychological, sociological, and ethical questions. There is also a very practical question to be answered: Is Mr. Hoogestraat cheating on his wife? The answer to this question depends on whether we understand Mr. Hoogestraat to have really done anything unfaithful. In a paper entitled Bodies, Selves (2008), J. David Velleman describes the manner in which virtual agency is exercised in multi-user role-playing games. Velleman notes that a player of Second Life will typically attribute to herself the various actions that his Second Life avatar performs in the virtual world. She will say things such as ‘I wielded the knife’ or ‘I was late for class’ or ‘I got married to Cindy last year’. Velleman argues that we have good reason to take these avowals literally. He maintains that although the player’s actions in the virtual world are merely fictional, the player is really performing them. Velleman contends that when an individual engages in virtual play, that individual (really) has a (fictional) body. Although the body is fictional, the player’s relation to the body is real in the respects that are most significant for physical actions. The avatar is directly controlled by the individual’s intentions, which are motivated by the individual’s beliefs and desires. This makes animating an avatar on Second Life both strange and familiar, since “even if you never play video games, you already have an avatar by default; your default is your body” (Velleman 2008, 415). Velleman’s characterization of virtual agency is in sharp contrast to the way that we ordinarily interpret the action of actors in theater and film, or the action of children in the context of pretend games. When an actor reports ‘After I got off the subway I stabbed Tony Soprano’ or ‘I commit suicide in Act V,’ it would be very odd to interpret what he says as literally true. Similarly, in pretend games, it would be very odd to interpret literally a child’s utterance of ‘I killed the pirate’. So why should matters be different for virtual play? Velleman articulates two central disanalogies between the phenomenology of make-believe in pretend games and acting, and the phenomenology of virtual play. Unlike the imaginary world of children’s fantasies, the virtual world of Second Life is both determinate and recalcitrant.2

|| 2 Velleman understands his account of virtual play as a challenge to Kendall Walton’s account of fiction and make-believe (Walton 1990). In this paper I do not comment on whether the

326 | Jason D’Cruz Velleman points out that virtual worlds, in contrast to worlds of pretend play, have a high degree of determinateness in relation to the knowledge of the participants (2008, 408). In games or pretend play, players are on an equal footing with authors of fiction: The facts of the fictional world are limited to the contents of their fictionalizing. By contrast, in a virtual world, each player must explore the world in order to learn what it is like. Moreover, since their beliefs about the virtual world are constantly being reality-tested, there is always the chance that they might be getting things wrong. A related feature of the virtual play typical of games like Second Life is that players cannot make stipulative additions to the virtual truths in the game (Velleman 2007, 407). For this reason, the virtual world manifests a kind of recalcitrance that is characteristic of real life. This recalcitrance is part of what explains the players’ intense psychological engagement for extended durations. They must deploy strategic reasoning and struggle to achieve their goals, just as they do in real life. Although a child playing pretend pirates may imagine that his hypothesis turns out to be false, it will only turn out false if he so decides. As a result, participants in games of make-believe cannot be frustrated or disappointed in a way that it is possible for inhabitants of Second Life. Unlike a pretend game of pirates where a child can simply stipulate, ‘I found the treasure! It was in the ship’s hull!’ and thereby make it true in the fiction, in virtual games like Second Life, the player must actually search for and discover the treasure. The treasure may very well be difficult to find, and there is an open possibility that the player, despite his best efforts, will not find it. A player’s hypothesis that the treasure is located in, say, the ship’s hull, may turn out to be false. Velleman (2008, 412) argues that this aspect of virtual play accounts for its distinctive phenomenology: “[P]layers who send their avatars into unknown regions of the virtual word are genuinely curious about what they will find; they do not attribute a fictional curiosity to their avatars to account for their fictional explorations.” Replying to Velleman, Robson and Meskin (2012) advert to Kendall Walton’s (1990) classic example of prop-based pretence to point out that children who play a game of make-believe game where stumps count for bears “may be genuinely curious as to what is true in this fictional world; there are so many places where stumps could (genuinely) be uncovered after all and as such so many places where we could (fictionally) stumble upon a hidden bear”. Robson and Meskin are correct that it is possible to genuinely strive for knowledge and that || putative distinctiveness of virtual play provides reason to think that Walton’s view cannot accommodate virtual play. For an answer in the negative, see Robson and Meskin (2012).

Agency and Volition in Make-Believe Worlds | 327

it is possible to get things wrong in standard games of make-believe, and that Velleman overstates the disanalogy between Second Life and prop-based games of pretence. A child who is playing a game in which ‘tree stumps’ stand for ‘bears’ may be in error about a number of things that are true in the fiction. For example, he may fail to notice that there is a bear ‘hiding’ on the other side of the creek, and his lack of knowledge may have adverse consequences in the game. Be that as it may, Velleman is right that such games are in general much more fluid and amenable to stipulative change than are virtual worlds. A child may simply decide that the bear across the creek is a ‘friendly bear’, or that the skinny tree stumps across the way do not stand in for bears after all; rather, they stand in for deer. (New rule: Thick stumps for bears, skinny stumps for deer). With a few quick and easy stipulative revisions, the imagined world can become markedly less recalcitrant. Granted, in games requiring interpersonal cooperation, this fluidity will depend partly on the amenability of one’s playmates. And one’s playmates may not be amenable. Again, the difference between virtual play and games of make-believe is not as cut and dried as Vellemen makes it out to be. But I think that Velleman is right to point out that virtual worlds like Second Life tend to be noticeably more determinate and recalcitrant than children’s games of make-believe, and this is part of what explains the psychological engagement they produce. Robson and Meskin also point out that paradigms of canonical fictions can be just as recalcitrant, determinate, and thereby just as emotionally gripping as games like Second Life. Again; they are right about this. However, Second Life is nonetheless distinctive in that it combines the recalcitrance and determinateness of a, say, realistic novel with the participatory aspects of games of makebelieve. Finally, Robson and Meskin maintain that Walton’s account of fictiondirected ‘quasi-emotions’ (Walton 1990, 245) is just as applicable to the virtual case as it is to canonical fictions: One of us has, today alone, encountered giant spiders, killer robots and murderous ogres in virtual environments but has no need for the extensive post-traumatic counselling that encountering any of these things in the real world would likely necessitate. Indeed all of Walton’s arguments that our fear in watching the green slime approaching is not genuine (our desire to repeat such experiences, the fact we don’t get up and leave the cinema, our continued belief that the object of our fear does not exist etc.) seem readily applicable, with slight adaptations, in the virtual monster case. (Robson and Meskin 2013, 12)

Robson and Meskin are probably right about virtual monsters, but I think that cases like that of Ric Hoogestraat should give us pause. Were Tenaj to abandon

328 | Jason D’Cruz or publically humiliate Hoorenbeek in Second Life, it seems unlikely that Hoogestraat would be keen to repeat the experience. Velleman (2008, 412) notes that people in virtual relationships “describe themselves as being in love, not as authoring a fictional romance”. He (Velleman 2008, 412) avers that, “They do no experience themselves as artists inventing characters; they experience themselves as the characters, behaving in character, under the impetus of their of thoughts and feelings.” Robson and Meskin counter that the widely reported accounts of love on Second Life are in fact relationships between real people, not between fictional entities, or as Velleman suggests, between chimerical creatures, compounded of fictional bodies and real minds. Robson and Meskin appeal to Walton’s view that one can genuinely perform certain acts by pretending to perform others (for example, one may genuinely express affection by pretending to blow a kiss). They contend that a person may (genuinely) profess love for another person by pretending to have their avatar profess love for the other person’s avatar. They draw the analogy of two (real) people communicating their love to each other via sock puppets. I suspect that this model of love on Second Life, although strange, does not capture the deep strangeness of Second Life romantic relationships. The qualities that one chooses to present in the guise of a Second Life avatar can often be very far from the person one really is; indeed, that is often the point. As a result, the ‘person’ one falls in love with may not exist at all outside of the game. Moreover, one may have markedly different preferences with regard to romantic partners in real life than the person one falls in love as. One may object on behalf of Robson and Meskin that meticulously curated self-presentations and even outright falsifications are not uncommon in many real relationships, especially in the age of online dating between strangers. Moreover, it is not uncommon for people to report falling in love with a fantasy or idealization, or for people to report being self-deceived about what they ‘really want’. However, the capacity for, and more importantly, the acceptability of, outright fictionalizing about the self in Second Life makes it odd to characterize Second Life romance as romance between two (real) people. Tenaj did not fall in love with a balding, overweight, unemployed, married man. Rather, Tenaj fell for an unattached, virile, risk-taking and dynamic entrepreneur. As Velleman points out, in virtual worlds the actual players are usually not known to one another, and there is no way to look behind the avatar to the person who ani-

Agency and Volition in Make-Believe Worlds | 329

mates it (Velleman [2008, 410] dubs this feature “opacity”).3 As a result, to fall in love on Second Life is neither to author a fictional romance nor is it to fall in love with a real person. The capacity for frictionless and endlessly repeatable self-invention is an aspect of a fundamental disanalogy between agency in the real world and agency in virtual world that I explore in the rest of the paper. This disanalogy does not rule out the metaphysical possibility of real agency in virtual worlds; rather, it explains why structural aspects of virtual worlds like Second Life tend to undermine the enabling conditions of agency.

2 Virtual agents as wantons In his classic paper Freedom of the Will and the Concept of a Person published 1971, Harry Frankfurt contends that the essential mark of personhood is to be found not in the capacity for reason but in the structure of the will. It is in virtue of the structure of will that a person is “a type of entity for whom the freedom of its will may be a problem” (1971, 14). In Frankfurt’s terminology, persons have both first-order and second-order desires. First-order desires are directed toward bringing about some state of affairs through action. Second-order desires are desires about what motives and desires to have. These kinds of desires require a capacity for reflective self-evaluation that only persons have: Besides wanting and choosing and being moved to do this or that, men may also want to have (or not to have) certain desires and motives. They are capable of wanting to be different, in their preferences and purposes, from what they are. Many animals appear to have the capacity for what I shall call “first-order desires” or “desires of the first order”, which are simply desires to do or not to do one thing or another. No animal other than man, however, appears to have the capacity for reflective self-evaluation that is manifested in the formation of second-order desires. (Frankfurt 1971, 12)

Any given agent will have a great number of first-order desires that compete for supremacy. Frankfurt defines an agent’s will as the first-order desire that actually motivates her to act; it is the effective desire. A person has a volition when she wants a certain desire to be her will. Volitions are a proper subset of secondorder desires. To illustrate the case of a second-order desire that is not a volition, Frankfurt offers the example of doctor who, for the sake of getting a better || 3 Robson and Meskin (2012, 17–18) point out that Velleman exaggerates the degree of opacity in Second Life.

330 | Jason D’Cruz understanding of what it feels like to be an addict, desires to desire a drug. The doctor here merely wants to feel the pull of the desire without being motivated by it, without that desire becoming his will. Volitions, on the other hand, are desires that are directed toward an agent’s will. For Frankfurt (1971, 17) a wanton is someone who is indifferent to her will. A wanton has first-order desires, but no second-order volitions. The class of wantons includes nonhuman animals as well as very young children. These beings may have desires and the capacity to act on them, but they do not have the capacity to care about their will. Adult human beings can act more or less wantonly to extent that they have or lack volitions of the second order. A wanton may have a highly developed capacity to deliberate about how to do what she wants to do; she may be excellent at engaging in instrumental reasoning. However, she does not broach the question of what her will is to be. Wantonness comes in degrees. Even individuals who are characteristically thoughtful and reflective will at times forget themselves.4 And even those who are habitually heedless and unmindful will sometimes be displeased or disgusted with their own desires. The degree to which one is a wanton is proportional to the extent that one lacks volitions, desires about which desires will constitute one’s will. I contend that we have good reason to understand the residents of Second Life as markedly more wanton in Frankfurt’s sense, because structural features of Second Life diminish the pressure to form volitions. As a result, although we may understand them as agents in a ‘thin’ sense, we should hesitate to attribute ‘full-blooded’ agency, the agency of persons, to them. To begin, consider the following structural features of Second Life (SL) that are disanalogous to that of real life (RL). Some of these features may seem too obvious to mention, but attending to them carefully is essential to understanding the structure of the will of SL residents, and therefore the nature of their personhood and agency. (1) In SL, you escape your past. (Typically, no one in SL knows you from RL). (2) In SL, you escape your future. (You can always log off when things get sticky without the irrevocable implications of a RL suicide).

|| 4 The tendency to forget oneself at times need not always rule out agency. Indeed, such selfforgetting may be indicative of a wholeheartedness and spontaneity that is constitutive of agency par excellence. See Velleman (2008b) and D’Cruz (2011). However, it is the systematic and pervasive absence of reflection on motives ultimately undermines agency.

Agency and Volition in Make-Believe Worlds | 331

(3) If your body or your persona in SL no longer pleases you, you can abandon your SL avatar and animate a new avatar. (4) If you ‘die’ in SL, you still have your real life to live. If you die in real life, you can no longer animate your SL avatar. (5) While you inhabit SL, you must also inhabit RL. But not vice-versa. (6) The seriousness of the consequences of your actions in SL are mitigated by the fact that 1–5 are true for other residents of SL as well. (7) The reasonable expectations of other players are modulated by a mutual understanding of 1–6. Velleman is right that the characteristics of recalcitrance and determinateness significantly increase the psychological immersiveness of games like Second Life. But as immersive as Second Life can be, the above-mentioned structural features simultaneously render life in Second Life relatively ‘low stakes’. The felt imperative to endorse or to disavow the desires that one has as a Second Life resident is markedly less urgent than the analogous felt imperative in real life. The felt need to ‘identify’ with one’s desires is greatly diminished when the consequences of one’s actions are so much less weighty. In real life, the question, ‘What does this desire say about me?’ can be both troubling and urgent. Individuals will often be concerned with the states of affairs they desire even if they are unlikely to bring them about, especially if those desires are in conflict with deeply held values. (Consider the paedophile consumed with self-hatred. He may hate not only the thought of acting on his desires, but also the desires themselves.) In Second Life, the question, ‘What does this desire say about me?’ can easily be deflected to, ‘What does this desire say about the alter-ego I am creating through this enactment?’ Players of Second Life can without difficulty move between the perspective of an agent who is performing an action to that of an author who is creating a character. Such a shift in perspective from agent to author will alter the applicable norms of responsibility. In general, there is nothing morally troubling about creating of fiction with an evil character; being an agent of evil, or even having desires that one conceives of as evil, is a different matter altogether.5 None of this is to insinuate that actions in Second Life never have real consequences. One can, for instance, break another person’s heart on Second Life. Moreover, it is sometimes the case that the actions of Second Life players are || 5 It is not important for my argument whether or not desires themselves are an appropriate object of moral assessment. All that matters for my argument is that agents often feel them to be so.

332 | Jason D’Cruz expressive of desires that players have in their real lives. But such desires can easily be written off by players of the game as mere ‘desire-like imaginings’. The connection between ‘desire’ and ‘desire-like imagining’ is highly complex and uncertain even for theoreticians,6 and therefore fertile ground for obfuscation on the part of ordinary players. This obfuscation may at times be strategic, a subtle way of evading troubling questions about the ultimate source in one’s psyche of one’s fiction-bound conations. Because of structural features (6) and (7) listed above, it is not hard for players to think that fellow players cannot or will not be injured by one’s actions in the virtual world. This may sometimes be an all too convenient rationalization. (Imagine how Ric Hoogestraat might feel to discover that ‘Tenaj’ is carrying on a covert affair with another Second Life resident). Nonetheless, the perspective of ‘it’s only a fiction’ is readily available. The extent to which taking up this perspective is justified will in part depend upon the extent to which the game inhabitants with whom one is interacting are emotionally invested in their virtual lives. But it is the availability of this perspective rather than its justification that distinguishes Second Life from real life, and which ultimately diminishes the sense of agency in Second Life.7 While these structural features cast doubt on Second Life as a locus of real agency, they simultaneously shed light on why Second Life is often conceived as a locus of emancipation. In Second Life, individuals are able to ‘try on’ identities without fear of judgment or recrimination. The anthropologist Tom Boellstroff describes some of complex relations between Second Life sexuality and realworld desire: Since sexual orientation was not typically seen to be embodied online (or offline) in the way that gender, race, age, and disability were, the sexual orientation of Residents was often unclear. For some, Second Life provided a virtual closet where they could live out same-gender desires that they were unwilling or unable to enact in the actual world. For instance, I encountered many cases of bisexual men and women who were heterosexually married, had chosen to be monogamous with their opposite gender spouse in the actual world, but with their spouse’s blessing pursued same gender sex and even relationships on Second Life. For others, Second Life could serve as a venue to grow comfortable with gay and lesbian identity before coming out in the actual world. (Boellstroff 2008, 165)

|| 6 See, for example, Currie (2002) and Nichols (2004). 7 In fact, there are instances when individuals will take up something analogous to this perspective in real life. Such a state, if enduring and pervasive, is indicative of a disorder of psychological dissociation, manifested either as depersonalization (a sense of unreality in one’s self), or derealization (a sense of unreality in one’s surroundings). See Simeon (2004).

Agency and Volition in Make-Believe Worlds | 333

When an individual’s desires and values conflict, the incongruous desires often occasion guilt, shame, and self-loathing. Typically, the individual seeks to extinguish them, or at least prefers that they wither away. In some instances, however, the opportunity to explore and to experiment with desires that one does not presently endorse leads one to view the objects of those desires as desire-worthy, or at least no longer inconsistent with one’s values. In ordinary life, there is a strong pressure to achieve congruence between felt motivation and reflective valuation. At times this pressure can be stifling. But on the whole we want the things that we desire to overlap with the things that we judge to be desire-worthy: We want to be wholehearted. Such pressure is key to personhood and to agency. The structural features of Second Life that relieve this pressure account both for its appeal, and for the fact that Second Life residents are neither fully persons nor agents. The singularity of real life, the fact that there is no ‘take-two’, makes urgent the imperative to desire and to pursue what is by our lights valuable. Second Life lacks in this urgency. In real life, we are sometimes given second chances, opportunities to make amends for past sins, solecisms, and indiscretions. However, we are never given the opportunity literally to erase the past and start over, free of regret, remorse, or reckoning. It is this feature of life that raises the stakes and makes meaningful agency possible. As mentioned early, the considerations adduced in this essay do not constitute an argument against the metaphysical possibility of virtual agency. Indeed, in rare cases, individuals will become so engrossed in Second Life that they care more about what happens in the virtual world than what happens in real life. In such cases, the prospect of Second Life suicide may feel like a real existential threat, and betrayals may reach across metaphysical boundaries of virtual and real. Such may have been the case with Hoogestraat/Hoorenbeek. If it were, then we would be wise not to ignore the question of whether he was also unfaithful to Tenaj, his wife in Second Life.

References Alter, Alexander. “Is this man cheating on his wife?” Wall Street Journal Aug 10 (2007). Boellstorff, Tom. Coming of Age in Second Life: An Anthropologist Explores the Virtually Human. Princeton: Princeton University Press, 2008. Currie, Gregory. “Desire in Imagination.” Conceivability and Possibility. Eds. Tamar S. Gendler and John Hawthorne. Oxford: Oxford University Press, 2002, 201–222. Current user metrics for Second Life. http://secondlife.com/xmlhttp/secondlife.php (n.d.) (19 February 2010).

334 | Jason D’Cruz D’Cruz, Jason. “Volatile Reasons.” Australasian Journal of Philosophy 91.1 (2013): 31–40. Frankfurt, Harry. “Freedom of the Will and the Concept of a Person.” The Journal of Philosophy 68.1 (1971): 5–20. Nichols, Shaun. “Review of Recreative Minds.” Mind 113 (2004): 329–334. Robson Jon and Aaron Meskin.“Videogames and the first-person.” Mimesis: Metaphysics, Cognition, Pragmatics. Eds. Gregory Currie, Petr Kotátko and Martin Pokorny. London: College Publications, 2012. Second Life. http://secondlife.com/?lang=en-US (n.d.) (20 January 2014). Simeon, Daphne. “Depersonalisation Disorder: A Contemporary Overview.” CNS Drugs 18. 6 (2004): 343–354. Velleman, J. David. “Bodies, Selves.” American Imago 65.3 (2008a): 405–426. Velleman, J. David. “The Way of the Wanton.” Practical Identity and Narrative Agency. Ed. Kim Atkins and Catriona MacKenzie. New York: Routledge, 2008b. 169–192. Walton, Kendall L. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Williams, Bernard and J. J. C. Smart. Utilitarianism: For and Against. Cambridge: Cambridge University Press, 1973.

Chris Bateman

Prop Perspective and the Aesthetics of Play 1 Games as art? When a child or an adult play, they interact with props that are not, in themselves, usually considered to be art, even the case of artistically-crafted artifacts. Toys and games, no matter how beautifully or artfully constructed, are generally assumed to lie outside of the sphere of art per se, and are generally viewed merely as leisure activities. Yet the experience of play is intrinsically aesthetic, as Niedenthal (2009, 4) correctly observes: “Whether or not we believe games to be works of art, it is undeniable that games can give rise to an aesthetic experience, as currently understood.” This opens the door to considering whether games and conventional artworks have something important in common. The issue as to what is or is not to be considered art is a matter of serious dispute: As Walton (2007, 148) notes, this question is “troubled and seriously contested”, and may in fact be essentially insoluble. Because the world of aesthetics does not agree as to what constitutes art, attempting to get other domains to be accepted for consideration is an uphill challenge, to say the least. Getting games seriously considered as potential artworks is all but impossible – the crass nature of the content of most commercial videogames seems to speak loudly against the medium. Nonetheless, the poor quality of most movies, television shows, plays and novels – as remarked by author Salmon Rushdie amongst others (see Weitz 2011) – has not prevented these media from producing at least some works that are maintained as worthy of cultural esteem and suited to the label of ‘art’: Perhaps the exclusion of games from the category of art reflects the natural reluctance to let the kids eat at the grown-ups table. The year 2011 saw the publication of two very different books on the relationship between games and aesthetics: Kirkpatrick’s Aesthetic Theory and the Video Game and my own Imaginary Games (Bateman 2011a). There had, however, already been various stirrings towards game aesthetics, or an aesthetics of play, as neatly summarized by Niedenthal (2009) two years earlier. Nonetheless, theoretical resources for the aesthetic appreciation of games as artefacts (or potentially as artworks) had been conspicuous by their absence prior to this year. Although arguing for the case that games qualify as artworks is beyond

336 | Chris Bateman the scope of what I am discussing here,1 at the very least we can say that these new approaches to games aesthetics help narrow the perceived gap between the medium of digital games and other media more traditionally presumed to provide the basis for culturally esteemed artworks. Kirkpatrick’s method deploys Rancière’s resurrection of a Kantian aesthetics of Form to apply classical aesthetic ideals to digital games, and denounces attempts to relate play to fiction. This line of argument is rooted in what has been termed ludology (see Frasca 1999 and Juul 2001), which takes an exceptionalist stance towards games in general, and digital games in particular, viewing them as requiring special techniques and approaches and unsuitable for study within existing paradigms, such as narratology. Narratologists have unsurprisingly been hostile to this attempt to erect a fence to keep them out of games: Simons (2007), for example, accuses ludologists of mounting arguments that are “ideologically motivated rather than theoretically grounded, and don’t hold up against closer scrutiny”. Kirkpatrick inherits the limitations of ludology, yet still offers much of value to the understanding of the aesthetic nature inherent to the play of digital games. He is particularly cogent when exploring the movements of the hands of players as they engage with fast-paced digital games, which he likens to dance. Additionally, he offers a staunch defense of the relevance of aesthetics to games: The experience gamers have with video games is aesthetic, even if most of them would not identify it as such, and the fact that it is aesthetic is probably the most important thing about it. Primarily, it is what games feel like to players that matters, both in the sense of explaining why players play and of accounting for the importance of video games in contemporary culture and cultural theory. (Kirkpatrick 2011, 32–33)

My own approach is radically different, taking the path Kirkpatrick (2011, 50) expressly disavows and derisively terms “rapprochement with fictionality”. Imaginary Games (Bateman 2011a) systematically adapts Walton’s make-believe theory of representation (Walton 1990) to games of all kinds, including digital games. I distinguish between two aspects of Walton’s theory, what has been termed pretence theory, which concerns distinctions between truth and fiction, and the core of Walton’s method which I term prop theory. The essence of prop theory is that certain objects serve as props in games of make-believe that prescribe various imaginings that collectively constitute a fictional world in relation to that specific prop. || 1 Although, see Bateman (2011b) for an attempt at mounting this argument.

Prop Perspective and the Aesthetics of Play | 337

Thus, there is a fictional world associated with the book Little Dorrit (Dickens 1857) that we participate with via a game of make-believe akin to that of a child who uses a toy horse or gun as a prop in their informal games. What distinguishes the two imaginary games is the extent that the prop is intrusive upon the resulting game (see Walton 1991, 45): Dickens’ text leaves little room for the reader to improvise, prescribing specific imaginings about the Marshalsea debtors’ prison, the actions of William Dorrit and his children and so forth. Precisely this intrusiveness, Walton suggests, is characteristic of great artworks – they offer intense emotional engagement as a result of their capacity to guide the game of make-believe that is played with them. If, as Walton suggests, our participation with representational art is best understood in terms of analogies with children’s games of make-believe then all (representational) art was always already intimately connected with games and play. On this reading, it makes little sense to attempt to cordon off digital games from consideration as potential artworks, as some critics attempt to do (see e. g. Ebert 2010). If we take Walton’s prop theory seriously, we must allow that any object that is sufficiently intrusive on make-believe has an aesthetic element that could be used to defend the claim that games are contiguous with other culturally-valued art forms such as cinema and theatre. This does not necessarily make all games art – but it does mean that digital games cannot be excluded a priori from consideration as potential artworks.

2 Fiction vs. function The most generous interpretation of the scope of Walton’s prop theory must be somewhat agnostic about the ontological implications of the term prop, allowing it to be applied in numerous different ways. A sentence in a written play is a prop, as is the entirety of the manuscript; the foil in the hand of an actor playing Hamlet is a prop, as is the actor themselves, and indeed the entire performance of the play. These observations build upon Walton’s discussions of prompters and objects of imagining, which demonstrate the decomposability of the prop concept (1990, 21–27). What props have in common is that they prescribe specific imaginings, and what they prescribe, we imagine, depends upon which props are taken to be in effect. For example, although the audience is always present in the case of the performances of stage plays, that audience is not usually taken to be part of the play: The prop or props in effect concern only what happens on stage (although there are notable exceptions, such as British pantomime play, when the audience directly participates with the fictional world of the play by,

338 | Chris Bateman for instance, yelling ‘look behind you!’ to warn a hero of a ‘hidden’ villain, and the hero dutifully responds to this interjection because it is part of the form of the pantomime that they do). This ontological agnosticism is a strength for prop theory when it engages with digital games, since the various situations allow for multiple interpretations depending upon which props are considered. On the one hand, for most digital games, a visual display (often a television or a monitor) serves as a central prop in that the images rendered onto a screen are one of the chief sources of the game’s prescriptions to imagine. On the other, the objects we perceive on this screen can also be taken to be props. Playing Ms. Pac-man (Namco 1982), we can distinguish a yellow circular blob as the titular character. This sprite – as 2D graphical objects are termed – is the prop that prescribes we imagine where Ms. Pac-man is in the fictional world of the game, while the sixteen red and two blue pixels that depict her bow and the four red pixels that depict her lipstick are the props that prescribe we imagine she is female. (Alternatively, the word ‘Ms.’ in the title could also be used as a prop to prescribe the same imagining). The prop theory approach places the player’s perceptions and experiences at the heart of the aesthetic understanding of play, and runs radically contrary to some positions in the game studies community that would seek to give priority to something far more concrete than the fiction – say, the nature of the program code that produces the game’s content. Kirkpatrick (2011, 63), for instance, decries Juul’s acceptance of fictional worlds in connection with digital games suggesting there is an “exaggeratedly spatial concept of games” at work: Juul’s notion that the game is a fictional world continues his earlier reification of the game program as a space, that is, he continues to treat it as if it existed prior to the player’s actions with it, or as if the three-dimensional space on the screen really has three dimensions even if no one looks at it. In fact, games only provide us with spatial sensations (the feelings arising from perceptions of space) when we play them in time. (Kirkpatrick 2011, 63)

Following Walton, Juul’s commitment to thinking in terms of fictional worlds (2005) no longer seems strange – even though Kirkpatrick (2011, 23) is correct to suggest that digital games can only provide ‘spatial sensations’. But of course, the same is equally true of paintings, for which Kirkpatrick is strangely happy to talk of a fictional world, and must surely also be true of novels. Perhaps movies and television shows can be claimed to actually depict space, as per Walton’s transparency thesis whereby we ‘see through’ photographic images (see Walton 2008). Either way, if our concern is the player’s experience, their imaginative engagement in a fictional world is arguably the key point of interest, no matter

Prop Perspective and the Aesthetics of Play | 339

how that world came to be depicted (although this may also be of great interest, but perhaps not of expressly aesthetic interest). One of the reasons proponents of ludology occasionally find themselves needing to distance themselves from fiction (despite Juul’s atypical commitment to the fictional world as a means of understanding digital games) is its fundamental thesis that the elements of the game that are of primary importance are functional, not representational. There is considerable merit to this perspective: In understanding Chess, it will be misleading to focus on the sense in which the Rook represents a castle – particular since most castles do not move at all, let alone unlimited distances in straight lines. Its function as a piece in the game of Chess is radically more vital than its vestigial representationality. It is considerably less clear that this is so in the case of, say, Grand Theft Auto: Vice City (Rockstar North 2002). Kirkpatrick, as the ludologists’ representative in the field of aesthetics, suggests that players of Vice City are pulled into a different way of interpreting the fictional world of the game by its demand that they take action: For example, the arrows in Grand Theft Auto: Vice City focus the player’s attention on those characters as elements that have to be ‘played’ (usually killed). There is a corresponding reduction or degradation of the rest of the presented scene, which now becomes just a backdrop when previously it was foregrounded and replete with coherent fictional significance. The switch of focus is a move into playing the game as an object, or cluster of objects, with priorities now set by action. It is a move away from concern with the fictional world as a setting. (Kirkpatrick 2011, 69)

From my own player studies, I would acknowledge that there are players – particularly those who prefer what I have elsewhere termed the systems aesthetic (Bateman 2014) – for whom this stripping away of the representational trappings is the natural form of play. This deeply pragmatic approach to games is more interested in exposing the underlying systems and exploiting them to attain victory than an imaginative experience of a fictional world, and can be connected to what Baron-Cohen (2003, 97–107) calls ‘a systemizing brain’. It would, however, be premature to believe that all players have this utilitarian bias towards the fictional worlds of their games. The players of certain games often bring their own aesthetic values into the digital games they play, and can be far more deeply involved with the fiction than might otherwise be assumed. Outside of digital games, for example, players of certain diceless tabletop role-playing games provide clear evidence that fiction may be judged above function: In the play of these verbal storytelling games, which resemble improvisional theatre with actors but no audience, the

340 | Chris Bateman content of the fiction is all that player experience can possibly be engaged with, since the functional mechanics have been sheared away almost entirely. Even in respect of Grand Theft Auto, concluding that the fiction is a mere backdrop misunderstands most players’ relationship with the game. One of the most publicized aspects of the design of the 3D rendered games in this franchise, including Vice City, has been the fact that players can pay a prostitute to have sex with them, then kill the prostitute and get their money back. This has been a source of great hilarity among many gamers, and a source of significant moral outrage amongst critics of digital games. Examined from the perspective of the functional aspects of the game, the prostitute is a deeply minor element – a means of recovering lost health, but by far the most inefficient one offered. If this were all that mattered, the fictional prostitutes would have been safe from harm. Instead, players murdered these fictional prostitutes because it amused them to do so, and part of the humor was precisely because players got their money back at the end, which has fictional significance but very little functional significance (particularly since the money involved is very small, and certainly not large enough to constitute motivation). It is only because the fictional worlds of Grand Theft Auto engender a playfully amoral atmosphere that this activity is a source of humor – different depictions, differences in tone, could easily have made this disturbing. As a point of comparison, could we imagine how grisly a Chuck Jones cartoon would be if the cat-and-mouse violence were animated with the gruesome realism of a horror movie? The interpretation of fiction is extremely sensitive to stylistic issues, and this is as true for digital games as it is for other media. Although not an entirely fair comparison, Manhunt (Rockstar North 2003), which the developer released one year after Vice City, featured extremely brutal depictions of violence and sold poorly despite heavy investment in marketing. Manhunt sold just over a million units, its sequel less than half of this. Conversely, Vice City sold over sixteen million units and its sequel more than twenty million.2 The players of Grand Theft Auto games enjoy them precisely because they offer a playground world in which to work off frustrations, one in which moral judgment is suspended via humor and cartoonish depiction, in direct contrast to Manhunt’s grit and gore. More than one player I have interviewed has suggested they particularly preferred to play a Grand Theft Auto game after a particularly irritating day at work. The fiction is vital to this catharsis, just as it is in movies and novels: The physical humor of The Three Stooges would not be

|| 2 See VG charts www.vgchartz.com (23 May 2012).

Prop Perspective and the Aesthetics of Play | 341

effective if we were not appropriately prepared to perceive their angry violence as comedic (see Scheff 1979, 127). Despite this, it would be wrong to think that digital games can be understood solely in terms of their fictional content, as Sicart has explicitly argued in connection with Manhunt (2009, 52). Assessing the ethical implications of the play of digital games often does involve the kind of detached utilitarian appreciation of the content noted by Kirkpatrick, but it would be just as wrong to conclude that the functional reading has priority over the fiction as it would to imply the contrary. I have argued that for digital games, and indeed many other kinds of games, it is precisely in the relationship between function and fiction that the aesthetic experience of play can be located (see Sicart 2009, 52).

3 Prop perspective 3.1 Toy-view One particularly common form of digital game in circulation today is known as a first person shooter. The genre’s most commercially successful manifestations, such as the Modern Warfare franchise (Infinity Ward 2007 onwards), can sell in impressive numbers – over twenty five million units for the third installment, for instance. The form is known as a ‘first person’ shooter because the action is depicted as if through the eyes of the protagonist, such that all that can be seen is a gun being held in a disembodied hand. Yet it is rather odd that we should think of this perspective as ‘first person’, since the original use of this phrase is to refer to prose text in which the narrative is presented by the author using ‘I’ and ‘my’. In the case of a first person shooter such as Modern Warfare, the best comparison to literature would be second person – ‘You’ hold the gun in your hand, ‘you’ shoot the ‘bad guys’. The shortcomings of taking terms from verbal grammar and applying them to digital games has been remarked upon by Thon, who notes that such distinctions as ‘first person’ are “inappropriate in its reference to grammatical categories that cannot be applied to audiovisual presentation of space in such a straightforward manner” (Thon 2009, 282). Thon expands upon Neitzel’s adaptation of Mitry’s The Aesthetics and Psychology of the Cinema to propose subjective, semi-subjective and objective points of view as more appropriate terms (see Neitzel 2002 and Mitry 1998). I support this general approach, and even more so Thon’s cleaned up versions of Neitzel’s point of action in digital games, something conceptually distinct from other media for which the participant by

342 | Chris Bateman definition lacks an ability to exert significant influence on the fictional world. However, by coming at issues of perspective from Walton’s prop theory, a set of metaphors for describing prop perspective becomes available that is potentially illuminating for the aesthetics of play. In the case of the first person shooter, and any other game rendered in Thon and Neitzel’s subjective point of view, the fictional world is specifically depicted in such a way as to show an object being held by hands that prescribe the player imagines it is they who are holding this object. Most frequently, the object in question is a weapon of some kind, usually a gun (although knives, swords and bows occasionally appear). In terms of Walton’s parallel with children’s games of make-believe, the kinds of props that most relate to the situation in subjective perspectives are toys e. g. toy guns and toy swords. Just as the child holds the toy weapon in their hands and imagines themselves wielding it in battle against imaginary foes – or against other children also wielding such toy weapons – players of first person shooters or other similarly presented games imagine themselves wielding the weapon shown on screen and using it to fight their imaginary foes – or other players who are also looking at a screen prescribing they imagine they are holding such a weapon. For this reason, I propose to term this perspective toy-view. The purpose of the metaphor that underlies toy-view is to emphasise the comparison between playing with toy objects and playing with games that render digital imagery of equivalent objects. The first person shooter is not so much about people as it is about guns, and such games offer dozens of firearms to play with. Modern Warfare 3 (Infinity Ward/Sledgehammer 2011), for instance, provides some 49 different guns and an additional 14 grenades and similar devices, not to mention a riot shield, a knife, and a slew of modifications such as silencers and laser targeting scopes. Borderlands (Gearbox 2009) uses procedural generation of firearms to boast 17,750,000 guns (see Robinson 2009). These are games about guns, and the elements of their design proceed naturally from the decision to depict a gun in such a way as to prescribe that players imagine they are holding those guns in their hands. The comparison between toy guns and shooting games goes further, since many digital games actually place a toy gun in the hands of the player as a control device. Games such as Duck Hunt (Nintendo R&D1 1984) pioneered this for home consoles with the NES Zapper peripheral, although in the arcade shooting games had been around since the late 1890s (see Mangels 1952, 194). These early shooting gallery games, however, could not be considered to deploy toy-view in the sense being used here, which implies the digital rendering of a fictional world rather than a shooting gallery prop, per se, not to mention that most early shooting galleries used actual firearms, typically .22 caliber rifles (see Mangels

Prop Perspective and the Aesthetics of Play | 343

1952, 194). However, games such as the aforementioned Duck Hunt or the popular arcade cabinet scrolling shooter Operation Wolf (Taito 1987), could be considered toy-view even though a physical toy gun prop is held in the player’s hands, since the depictions on screen are intended to harmonize with the gun as an interface device. Toy-view is not constrained to refer solely to shooting games, however, as there are other kinds of toys available in digital games. Driving games, for instance, often offer a subjective point of view in which the dashboard of a motor vehicle serves as a prop that prescribes players imagine they are driving, for example, a truck in 18 Wheeler: American Pro Trucker (Sega AM-2 2000). The trinkets depicted on the dash in this game move in response to the player’s actions to further the sense of connection between the player and the fictional world, what has been termed immersion (see Murray 1998). Cameras are another kind of prop that benefits greatly from toy-view, and although digital games based on photography are rare, they have enjoyed some commercial success with titles such as Pokémon Snap (HAL Laboratory/Pax Sofnica 1999). An unusual case is Project Zero (Tecmo 2001), known in the US as Fatal Frame, in which the player must fight for survival against ghosts using a magical camera that is deployed in toy-view – photographing the spirits harms them and eventually traps them, rendering them harmless. The use of the subjective perspective evokes an extremely claustrophobic atmosphere reminiscent of the use of the equivalent to toy-view in recent movies such as The Blair Witch Project (Myrick and Sánchez 1998) or Cloverfield (Reeves 2008), which use the pretence of a video camera held by one of the central characters to justify the use of subjective point of view. Despite the occasional use of a perspective equivalent to toy-view in film, this is not the usual way movies approach their camera work. Indeed, an earlier example of subjective perspective in film, Dark Passage (Daves 1947), helps show up the limitation of the form. Whereas toy-view in digital games makes sense when the activity focuses around the actions of specific toys – guns or cameras for their respective forms of shooting, for instance – Dark Passage’s portrayal of Humphrey Bogart’s hands picking up books and so forth feels deeply artificial and distances the audience from the action on screen. The horror movie examples trade on the engendered claustrophobia – but usually our interests in films lie elsewhere, namely on the emotional experiences of the character. This requires a different perspective.

344 | Chris Bateman

3.2 Doll-view In a pivotal scene of Kurasawa’s epic Seven Samurai (1954), we see Toshiro Mifune’s character rescuing a young child from a burning mill. Once safely in the river, he breaks down in tears realizing that the child is in the same situation he was in when he was a youngster. Mifune’s character is posing as a samurai, but is actually a farmer who has acquired a samurai birth certificate by some unknown means. The other samurai know he is a fake, but tolerate him, and it is only when he is forced to confront his past via the child that his tough exterior becomes cracked and he breaks down in tears. The camera centers upon the weeping Mifune, the child in his arms and the burning mill in the background, flames reflected in the water. It is a powerful and moving scene that simply could not have been replicated in a subjective point of view precisely because our interest in movies are the actors and actresses, and their portrayal of their various emotional states. Indeed, sometimes a reaction shot of the face of a performer is more effective than depicting what is actually occurring. In digital games, such emotional scenes rarely occur, and when they do it tends to be within a short animated movie sequence embedded into the flow of the game. In this respect, thinking of digital games as ‘mixed media’ may not be misleading. Nonetheless, just as the camera places people at the centre of almost all of the most critical shots, many digital games place the animated character whom the player controls in the centre of the screen, where the player is able to enjoy the animations of this imaginary person defending against attack, clambering over obstacles or simply moving around the world. The animated model is the player’s focus, and for this reason I propose calling this semisubjective or third person perspective in games doll-view. The animated character model in games is often called the avatar, although as I argue in Imaginary Games, this is highly misleading since what is meant by ‘avatar’ is the player’s link to the fictional world of the game (the source of the point of action, in Neitzel’s phrase) and in some games, such as text adventures, there is no such depiction, yet there is still clearly an avatar, or else the player could not act in the world. Instead, I suggest calling the depiction of the avatar the doll, or perhaps for clarity, the avatar-doll. When the avatar-doll is the focus of attention, the game is operating in doll-view. Toy-view makes sense when the player’s interactions with the fictional environment are simple – shooting or driving, for instance. When they have more detailed interactions, it becomes necessary to see the avatar-doll in the context of the world itself. For example, Super Mario 64 (Nintendo EAD 1996) pioneered the adaptation of what is termed a platform game from its traditional 2D repre-

Prop Perspective and the Aesthetics of Play | 345

sentation into a three dimensional fictional depiction of a world. The titular character, Mario, interacts with this world with a variety of animations relating to his agility – jumping, climbing, swimming, crawling and so forth. These kind of interactions are very difficult to conduct in toy-view, as Mirror’s Edge (DICE 2007) amply demonstrated. The 2D platform games that Super Mario 64 built upon can also be considered doll-view, and the key point is that the position of the doll with respect to the environment is crucial to the play of these games, thus necessitating a semi-subjective perspective. The view is still partially subjective as Thon (2009, 284) notes: “Although the spatial position of the avatar is not the same as that of the camera, the camera’s position is always linked to the avatar.” Dollview places the player’s fictional representative squarely into the world they belong to, making it easier to orient with respect to it – or with respect to enemies. Fighting games such as the classic one-on-one karate game The Way of the Exploding Fist (Barnett 1985) use doll-view because the relationship of bodily limbs on one doll to another is central to the play of the game, and this would be impossible to judge in toy-view. Just as toy-view in a first person shooter is analogous to the child’s toy gun, the doll-view in most games that use this perspective is analogous to a child’s doll or action figure. The term ‘action figure’ is illuminating, since boys have traditionally not been allowed to play with ‘dolls’ as a result of gender stereotyping that is endemic to the construction of these kinds of toys (see Attfield 1996, 80–89). Nonetheless, ‘doll’ is surely the most appropriate generic term to use in these contexts, and the play with the avatar-dolls of digital games is deeply parallel to children’s play with dolls of other kinds. An alternative term, model-view, could be applied to cases where the depiction of the avatar is not humanoid – for instance in a racing game that uses a semi-subjective viewpoint. In fact, almost all contemporary digital racing games offer a choice between toy-view and model-view, since different players prefer different perspectives. In my own player studies, I have noticed players preferring to race in toy-view insist that it is ‘more immersive’ while those who prefer to race in model-view complain about limited peripheral vision (‘racing in first person is like having to drive with toilet rolls strapped to your eyes’). Although the primary benefits of doll-view lie in showing the relationship between the doll and the world, or between the doll and other dolls, it would be wrong to think that games using this kind of prop perspective avoid using guns. Even though targeting of firearms is more natural to control in toy-view, a large number of doll-view gun games (‘third person shooters’) are available. Indeed, the Grand Theft Auto games are examples of games with a heavy element of gun-play, but a preference for doll-view. This is actually not entirely surprising

346 | Chris Bateman when the diversity of play activities within such games are considered: Driving, flying, shooting and (more recently) swimming and parachuting are all being offered within the context of one fictional world. Doll-view (including modelview) is able to cope with any conceivable situation and is better suited to such playground worlds, although this advantage comes at the price of increased development costs as a result of the far larger suite of animations required for the dolls that are depicted. Although doll-view is intended as a metaphor for appreciating 2D and 3D digital games, like toy-view’s precursor shooting galleries, doll-view has its precedents in earlier forms of play. In particular, players of tabletop roleplaying games of the kind pioneered by Dungeons & Dragons (Gygax and Arneson 1974) often use lead miniatures to represent the players’ characters in spatial relationships with the monsters they fight. These figurines may not be animated, but they still serve the same essential role as an avatar-doll in a digital game, and the comparison with children’s dolls in their corresponding games of make-believe is still apposite.

3.3 Table-view Tabletop role-playing games descended from wargames of the kind published by Avalon Hill in the 1950s, 60s and 70s. But in these games, there was no specific link between players and characters – this was precisely the breakthrough of Gary Gygax and Dave Arneson in conceiving of the kind of play inherent to a game like Dungeons & Dragons and its descendents. (As I describe in Imaginary Games (Bateman 2011a), this game was innovative in multiple different ways and has had far more influence on the history of digital games than is usually recognized.) In the wargame, the fictional world of the game is depicted primarily by a map that is laid out on a table, sometimes a very large table. I advocate calling the corresponding prop perspective table-view to highlight this ancestral connection to tabletop wargames. Table-view corresponds broadly to Thon and Neitzel’s objective perspective. As Thon (2009, 284) notes, this point of view “offers the possibility to observe a large game space without being constrained by the spatial perspective of an avatar or comparable entity”. However, table-view can equally be applied to less directly representational games, including abstract games whereby the notion of a table survives solely in the sense of there being a two dimensional space of play. Chess and Go are examples of conventional games that could be

Prop Perspective and the Aesthetics of Play | 347

claimed to represent table-view, as can certain puzzle games such as the popular match-3 genre pioneered by Bejeweled (PopCap 2001). The parallel with tabletop wargames helps draw out the point that tableview need not by represented in two dimensions since lead miniatures of tanks and so forth are typical to this form of play. Similarly, most strategy games, including what are termed real-time strategy games (despite very little strategic play) can be considered table-view games. Whether the games in consideration are isometric grids, such as in Sid Meier’s Civilization (Microprose 1991) or fully rendered battlegrounds, such as in the Total War franchise (Creative Assembly 2000 onwards), the player messes around with their little toy soldiers just as if they were models on a table – although with the added entertainment value of detailed computer animations. The most common use of table-view in digital games, however, is as a map. This still fits Thon’s (2009, 284) criteria for objective perspective (“observe a large game space without being constrained”) and is even more explicitly linked with the idea of a table, since laying out a large map on a table is a classic use of maps. Indeed, when maps appear in movies they are often positioned upon table surfaces, as in the World War II film Sink the Bismarck! (Gilbert 1960), in which the sequences depicting battles between warships in the North Atlantic are narratively connected by scenes in the British admiralty’s War Room where the positions of naval vessels, and their possible movements, are tracked upon a large ocean map. Within digital games, the use of the map can be central to the play of the game. For instance, in the Castlevania series – typified by the fourteenth installment, Castlevania: Symphony of the Night (Konami 1997) – the individual environments the player encounters in the fictional world tend to be simple corridors of one kind or another. It is only via the map screen that players link these up into an overall space – specifically Dracula’s castle. The map prescribes that players imagine that they are exploring this castle, and without this prop the entire game feels dull and pointless. Table-view can thus have a complementary role to toy-view and doll-view in providing additional prescriptions to imagine that radically affect players’ sense of participation in a tangible (although imaginary) fictional world.

4 Art and the aesthetics of play Examining digital and other games by considering prop perspective helps to draw attention to the extent that the games of make-believe associated with

348 | Chris Bateman these kinds of play are intrusive in Walton’s sense – they create rich fictional worlds, and as such cannot be excluded from consideration as candidates for the cultural esteem associated with ‘art’. However, prop perspective also draws attention to the ways in which digital games, and indeed most tabletop games (with the possible exception of tabletop role-playing games), do not engage their participants in the ways we normally associate with works of art. Seven Samurai and Little Dorrit are masterly artworks that draw deeply against our empathy for human life; Grand Theft Auto and Modern Warfare are entertaining playgrounds in which death and carnage engage and amuse. An aesthetics of play is necessary to appreciate what games are good at doing, which is often very different to those things that are conventionally considered ‘art’. There are many different approaches to an aesthetics of play, as Kirkpatrick, Niedenthal and myself have elucidated in radically different ways – and these are needed to help draw out the aesthetic experience that occurs in games of all kinds. This is something that happens in the confluence between their functional elements – what Juul (2005) summarizes as rules – and the fictional elements, which can be gainfully explored using Walton’s prop theory. However, the rise of abstract art in the twentieth century – as represented by artists such as Matisse, Mondrian and Kandinsky – demonstrates that aesthetic and cultural value can occur in cases beyond what had been previously presumed. These artists, it might be claimed, made use of something akin to table-view to break out of an artistic vision of painting previously constrained to doll-view. In the same way, digital games at the start of the twentieth century are breaking free of the utilitarian vision of games as popular entertainment, a space in which gun-play and the addictive pursuit of goals dominates, and carving out an aesthetic niche conditioned by, yet transcending, its predecessors. Developers such as Tale of Tales, Thatgamecompany and thechineseroom, and creative individuals such as Mory Buckman, Ferry Halim, Rod Humble, Ed Key, Deirdra Kiai, Jordan Magnuson and others too numerous to mention are experimenting in the space between function and fiction that digital play offers, with results that are worthy of critical appraisal. As I suggested earlier, the art world is still reluctant to let the kids eat at the grown-ups table. However, I would already suggest that in terms of their contributions to their respective media The Graveyard (Tale of Tales 2008), Journey (thatgamecompany 2012) and Proteus (Ed Key and David Kanaga 2012) are at least on a par with early Bergman movies such as The Seventh Seal (1957), which admittedly has aged badly when compared to later Bergman films such as Persona (1966). Yet digital games are still a young medium – Bergman and Kurasawa had nearly a century of cinematic experience to draw against; today’s ‘artgame’ creators have mere decades of prior material to refer to. Still, from a

Prop Perspective and the Aesthetics of Play | 349

personal perspective I can no longer doubt – as I once did – that games can be great artworks, for some have already moved me deeply. The kids table has suddenly gained infinitely greater aesthetic interest.

References Attfield, Judy. “Barbie and Action Man: Adult Toys for Girls and Boys.” The Gendered Object. Ed. Pat Kirkham. Manchester, UK: Manchester University Press, 1996, 80–89. Barnett, Gregg. The Way of the Exploding Fist [Home Computer Game]. Melbourne: Beam Software/Melbourne House, 1985. Baron-Cohen, Simon. The Essential Difference: Men, Women and the Extreme Male Brain. London: Penguin/Basic Books, 2003. Bateman, Chris. Imaginary Games. Winchester: Zero Books, 2011a. Bateman, Chris. “Prop Theory for Game Aesthetics.” 3rd Annual Conference on Videogame Cultures and the Future of Interactive Entertainment. http://www.interdisciplinary.net/wp-content/uploads/2011/06/batemanvpaper.pdf. Oxford, UK: InterDisciplinary.net, 2011a (21 January 2014). Bateman, Chris (2014). “Implicit Game Aesthetics.” Games and Culture. http://gac.sagepub.com/content/early/2014/12/11/1555412014560607.abstract (accessed 10 March 2015), article published online before print. Bergman, Ingmar. Det Sjunde Inseglet (The Seventh Seal) [Motion Picture]. Stockholm: AB Svensk Filmindustri, 1957. Bergman, Ingmar. Persona [Motion Picture]. Stockholm: AB Svensk Filmindustri, 1966. Creative Assembly. Shogun: Total War [PC Game]. Redwood City, CA: Electronic Arts, 2000. Daves, Delmer. Dark Passage [Motion Picture]. Burbank, CA: Warner Bros, 1947. DICE (EA Digital Illusions CE). Mirror’s Edge [Console Game]. Redwood City, CA: Electronic Arts, 2007. Dickens, Charles. Little Dorrit. London: Bradbury and Evans, 1857. Ebert, Roger. “Video Games Can Never Be Art.” http://www.rogerebert.com/rogersjournal/2010/04. Roger Ebert’s Journal, 16 April 2010 (14 January 2014). Frasca, Gonzalo. “Ludology Meets Narratology: Similitude and Differences between (Video)games and Narrative.” Parnasso#3 (1999). http://www.ludology.org/articles/ludology.htm. Ludology.org: video game theory (23 May 2012). Gearbox. Borderlands [Console Game]. New York, NY/Novato, CA: 2K Games, 2009. Gilbert, Lewis. Sink the Bismarck! [Motion Picture]. Los Angeles, CA: 20th Century Fox, 1960. Gygax, Gary and Dave Arneson. Dungeons & Dragons [Boxed Role-playing Game]. Lake Geneva, WA: Tactical Studies Rules, Inc, 1974. HAL Laboratory/Pax Sofnica. Pokémon Snap [Console Game]. Kyoto: Nintendo, 1999. Infinity Ward. Call of Duty 4: Modern Warfare [Console Game]. Santa Monica, CA: Activision, 2007. Infinity Ward/Sledgehammer Games. Call of Duty: Modern Warfare 3 [Console Game]. Santa Monica, CA: Activision-Blizzard, 2011.

350 | Chris Bateman Juul, Jesper. “Games telling stories? – A brief note on games and narratives.” http://www.gamestudies.org/0101/juul-gts/. Game Studies 1.1 (2001) (23 May 2012). Juul, Jesper. Half-Real: Video Games between Real Rules and Fictional Worlds. Cambridge, MA, MIT Press, 2005. Key, Ed and David Kanaga. Proteus [PC Game]. Privately published, 2012. Kirkpatrick, Graeme. Aesthetic Theory and the Video Game. Manchester: Manchester University Press, 2011. Konami. Castlevania: Symphony of the Night [Console Game]. Tokyo: Konami, 1997. Kurosawa, Akira. Schinin no samurai (Seven Samurai) [Motion Picture]. Tokyo: Toho, 1954. Mangels, William J. The Outdoor Amusement Industry: From Earliest Times to Present. New York, NY: Vantage Press, 1952. Microprose. Sid Meier's Civilization [PC Game]. Austin, TX: Microprose, 1991. Mitry, Jean. The Aesthetics and Psychology of the Cinema. London: Athlone Press, 2008. Murray, Janet H. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. Cambridge, MA: MIT Press, 1998. Myrick, Daniel and Eduardo Sánchez. The Blair Witch Project [Motion Picture]. Santa Monica, CA: Artisan Entertainment, 1998. Namco. Ms. Pac-man [Arcade Game]. Tokyo: Namco, 1982. Neitzel, Britta. “Point of View and Point of Action: A Perspective on Perspective in Computer Games.” Paper presented at the Internationale Conference ‘The Challenge of Computer Games’, University of Łodź, Poland, September 2002 [manuscript]. Niedenthal, Simon. “What We Talk About When We Talk About Game Aesthetics.” http://dspace.mah.se/dspace/bitstream/handle/2043/13326/Niedenthal_What%20we %20talk%20about.pdf;jsessionid=5CF254F2FCD73CA299380ABD44539DD8?sequence=2 Proceedings of DiGRA 2009: Breaking New Ground: Innovation in Games, Play, Practice and Theory. Eds. Barry Atkins, Helen Kennedy and Tanya Krzywinska. London: Brunell University, 2009 (21 January 2014). Nintendo EAD. Super Mario 64 [Console Game]. Kyoto: Nintendo, 1996. Nintendo R&D1. Duck Hunt [Console Game]. Kyoto: Nintendo, 1984. PopCap Games. Bejeweled [Browser Game]. Seattle, WA: PopCap Games, 2001. Reeves, Matt. Cloverfield [Motion Picture]. Hollywood, CA: Paramount Pictures, 2008. Robinson, Andy. “Gearbox interview: Randy Pitchford on Borderlands’ 17 million guns.” http://www.computerandvideogames.com/220328/interviews/gearboxinterview/?page=2#top_banner. Computer and Video Games 28 July 2009 (24 May 2012). Rockstar North. Grand Theft Auto: Vice City [Console Game]. New York, NY: Take Two Interactive, 2002. Rockstar North. Manhunt [Console Game]. New York, NY: Take Two Interactive, 2003. Scheff, Thomas J. Catharsis in Healing, Ritual, and Drama. London: University of California Press, 1979. Sega AM-2. 18 Wheeler: American Pro Trucker [Arcade Game]. Tokyo: Sega, 2000. Sicart, Miguel. The Ethics of Computer Games. Cambridge, MA: MIT Press, 2009. Simons, Jan. “Narrative, Games, and Theory.” http://gamestudies.org/0701/articles/simons Game Studies 7.1 (2007) (23 May 2012). Taito. Operation Wolf [Arcade Game]. Tokyo: Taito, 1987. Tale of Tales. The Graveyard [PC Game]. Kirkland, WA: Valve Corporation, 2008. Tecmo. Project Zero (AKA Fatal Frame) [Console Game]. Tokyo: Tecmo, 2001. Thatgamecompany. Journey [Console Game]. Tokyo: Sony Computer Entertainment, 2012.

Prop Perspective and the Aesthetics of Play | 351 Thon, Jan-Noël. “Perspective in Contemporary Computer Games.” Point of View, Perspective and Focalization: Modelling Mediation in Narration. Eds. Peter Hühn, Wolf Schmid and Jörg Schönert. Berlin: de Gruyter, 2009. 279–299. Vgchartz. http://www.vgchartz.com (n.d.) (23 May 2012). Walton, Kendall L. Mimesis as Make-believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990. Walton, Kendall L. “Reply to Reviewers.” Philosophy and Phenomenological Research 51.2 (1991): 423–427. Walton, Kendall L. “Aesthetics – What? Why? and Wherefore?” The Journal of Aesthetics and Art Criticism 65.2 (2007):147–161. Walton, Kendall L. “Transparent Pictures: On the Nature of Photographic Realism.” Marvelous Images: On Values and the Arts. Kendall L., Walton, Oxford University Press, 2008, 79– 116. Weitz, Gidi. “Interview / Salman Rushdie is not afraid.” http://www.haaretz.com/weekend/magazine/interview-salman-rushdie-is-not-afraid1.389961. Haaretz Wednesday 23 May 2011 (23 May 2012).

Index Abbott, H. Porter 143 Adams, Robert M. 54–55 Allen, Richard 209–210 Alperson, Philip 290, 292– 293, 296, 299 Alter, Alexander 324 Alter, Robert 25 Altman, Ulrike 119 Alward, Peter 247–248, 255 Anderegg, Johannes 225 Aristotle 34, 122–123, 134, 317 Arneson, Dave 346 Atwood, Margaret 193 Augustine 273 Austen, Jane 40, 194 Axelzon, Fredrik 315 Bach, Carl Philipp Emanuel 291, 297, 299 Bacon, Francis 118, 123–124, 127 Bal, Mieke 79 Balzac, Honoré de 17–18, 26, 32–33, 214 Baudelaire, Charles 225 Bareis, J. Alexander 5, 7, 115, 168, 170, 172, 178, 181, 196, 235, 248–249 Baron-Cohen, Simon 339 Bartel, Christopher 107 Barthes, Roland 153 Bateson, Gregory 128 Batteux, Charles 224–225

Bauer, Barbara 232 Beardsley, Monroe C. 223, 225 Becker, George 22, 25 Beckett, Samuel 143 Beethoven, Ludwig van 284, 291, 295–296 Bell, Clive 287–288 Belsey, Catherine 20, 23 Bennett, Arnold 26, 29, 32, 34, 36 Bennett, Benjamin K. 250, 262 Bentley, Eric 254 Benveniste, Émile 187, 189, 196, 214 Bergez, Daniel 210 Bergman, Ingmar 348 Bessière, Jean 208 Bierce, Ambrose 178 Bieri, Peter 238 Bill, Valentine T. 171 Binder, Hartmut 167 Bolter, David 208 Bölhoff, Reiner 229, 232 Bordwell, David 196 Borgida, Eugene 123 Boruah, Bijoy 319 Boschung, Urs 233 Bowers, Fredson 168 Bowlby, Rachel 27–28 Bowles, Paul 87 Bradley, Andrew Cecil 48 Brandom, Robert 150

354 | Index

Brecht, Bertolt 197 Bremond, Claude 62 Brontë, Charlotte 215 Brown, Dan 107 Bruner, Jerome 107 Buckman, Mory 348 Bunia, Remigius 5–6, 123, 134–135 Burdorf, Dieter 224–225, 231 Cage, John 153 Calder, Jenni 55 Callow, Simon 278, 281 Calvino, Italo 192, 194–195 Campana, Alessandra 259 Campbell, Terence 106 Carroll, Noël 7, 41, 65–66, 69, 155, 157, 162, 185– 186, 196, 198, 249–251, 255, 257 Casetti, Francesco 197–198 Chase, David 172–173 Chatman, Seymour 78–79, 143, 186 Chaucer, Geoffrey 42–43 Cheong, Yun-Gyung 72 Chopin, Frédéric 300–301, 304 Collingwood, George 286 Cohn, Dorrit 132, 251 Conrad, Joseph 170–171 Coover, Robert 193 Corngold, Stanley 167 Corse, Sandra 261 Culler, Jonathan 1 Cunningham, Valentine 207

Currie, Gregory 1–2, 7, 10, 42, 64, 66, 70, 85, 107, 128, 133–134, 155, 158– 159, 180–181, 186, 188– 190, 194, 198, 247–248, 251, 332 Damasio, Antonio117–119 Dante Aligheri 117, 234 Davies, David 209, 257 Davies, Stephen 286, 288, 293 Davis, J. Madison 173 Defonseca, Misha 99, 103 De Man, Paul 23–24 Dennet, Danie 102l Descartes, René 113, 127 Dewey, John 271 Dick, Manfred 237 Dick, Philip K. 194–195 Dickens, Charles 50–51, 86, 147, 337 Dilthey, Wilhelm 225 Doležel, Lubomir 188 Dostoyevsky, Fjodor 157, 161 Doyle, Arthur Conan 116 Dufrenne, Mikel 270–272 Eagleton, Terry 23, 49 Ebert, Johann Arnold 235– 236 Ebert, Roger 337 Eder, Jens 78 Einstein, Gil 106 Elam, Keir 254, 259 Eliot, George 19, 158 Eliot, T. S. 43

Index | 355

Escher, M. C. 151 Eskin, Blake 100 Fforde, Jasper 215–216 Fitzgerald, F. Scott 46–47, 52–53 Fleming, Paul 232 Fludernik, Monika 62–64, 126, 187, 191, 193, 208, 214, 252 Frankfurt, Harry 329–330 Friedman, Norman 77, 79 Friend, Stacie 168 Frey, James98–101, 105 Fricke, Harald 228, 239 Fry, Roger 287–288 Ford, Ford Madox 77–78, 83 Forster, E. M. 22, 65, 284– 285, 295 Gabriel, Gottfried 227 Galsworthy, John 26, 36 Gaudreault, André 196–197 Gaulli, Giovan Battista 212 Gaut, Berrys 7, 185–186, 189–191, 198, 248, 319 Gendler, Tamar 311 Genette, Gérard 67, 78–81, 83–85, 132, 187, 189, 192–193, 203–205, 209 George, Stefan 225 Gerrig, Richard J. 72, 129 Gibson, John 157, 249 Girle, Rod 87 Goethe, Johann Wolfgang 83, 225, 234

Goldie, Peter 61, 63–64, 72– 73 Gombrich, Ernst Hans 134, 298 Goodman, Nelson 114 Gorman, David 246 Graff, Gerald 24, 27, 36 Green, Graham 194 Greenblatt, Stephen 208 Günther, Johann Christian 223, 229–234, 236–238 Gygax, Gary 346 Hacking, Ian 34 Halim, Ferry 348 Haller, Abrecht von 223, 233– 234, 237 Halliwell, Michael 260–261 Hamburger, Käte 225 Hamilton, James R. 4, 66, 70, 245, 255–259, 262, 270– 271, 274, 276 Hanslick, Eduard 283, 288– 289 Hardy, Thomas 48–49 Hegel, Georg Wilhelm Friedrich 225 Hemmings, Frederick William John 34 Herder, Johann Gottfried 225 Herman, David 192, 195 Hermerén, Göran 182 Hesse, Hermann 83 Hintikka, Jaakko 287 Hogarth, William 151 Horowitz, David 101 Hugo, Victor 116

356 | Index

Hühn, Peter 231 Humble, Rod 348 Hume, David 31, 310–311 Hume, Kathryn 18 Husserl, Edmund 8, 273, 275– 276, 278–280 Hutcheon, Linda 197, 260 Iser, Wolfgang 269 Jahraus, Oliver 181 Jakobson, Roman 30 Jannidis, Fotis 245 Johansson, Christer 78 John, Eileen 162 Jones, Chuck 340 Jones, Margaret (= Margaret Steltzer) 99–100 Jost, François 196 Juriedini, Jon 64, 70 Juul, Jesper 2336, 338–339, 348 Kafka, Franz 146–147, 166– 168, 181 Kahneman, Daniel 121, 271– 272, 275–276, 281 Kanaga, David 348 Kandinsky, Wassily 348 Kania, Andrew 7, 185–188, 248 Kant, Immanuel 269, 271, 288–290, 336 Kehlmann, Daniel 172 Kerman, Joseph 260 Key, Ed 348 Kiai, Deirdra 348 Kindt, Tom 181

King, Stephen 214 Kirkpatrick, Graeme 335– 336, 338–339, 341, 348 Kivy, Peter 2, 160, 162, 260– 261, 283–290, 293–295, 303 Klauk, Tobias 5, 79–80 Klimek, Sonja 5–8, 205, 208– 209, 212 Klopstock, Friedrich Gottlieb 225, 234 Köppe, Tilmann 5, 79–80, 238, 248, 250 Korthals, Holger 250 Krakauer, Jon 92 Krausser, Helmut 82 Kreiswirth, Martin 66 Kubrick, Stanley 207 Kühn, Sophie von 236 Kukkonen, Karin 205 Lamarque, Peter 4, 10, 41, 43, 48, 58, 65, 69, 129, 136, 155, 158–162, 209, 247, 249 Lamping, Dieter 224–225, 227–228, 230–231, 238 Langacker, Ronald W. 122– 123 Lanthimos, Giorgios 216–217 LeBar, Mark 85 LeRoy, J. T. 99 Levin, Harry 18, 22, 25, 30 Levine, George 22, 25 Levinson, Jerrold 161, 187– 188, 285–286 Lévi-Strauss, Claude 118, 123

Index | 357

Lewin, David 271 Lewis, David 7, 51, 66, 165, 169–170 Limoges, Marc 209 Lincoln, Abraham 117 Linhares–Dias, Rui 78 Littlejohn, David 262 Littlewood, Joan 278 Lubbock, Percy 81, 83 Luhmann, Niklas 137 Lukács, Georg 17, 25, 27 Lutas, Liviu 5, 7, 214–215 Lyotard, Jean-François 19

Mondrian, Piet 348 Monet, Claude 114 Moore, George Edward 287– 288 Moran, Richard 317 Morris, Christopher 260 Morris, Pam 23, 30 Mortensen, Greg 99 Muir, Edwin 168, 181 Muir, Willa 168, 181 Müller, Hans-Harald 181 Müller-Zettelmann, Eva 223– 224

Maechler, Stefan 99–100 Magnuson, Jordan 348 Malina, Debra 204, 208, 213 Mantegna, Andrea 211 Martínez, Matías 80 Matisse, Henri 348 Maynard, Patrick 298 McDaniel, Mark 106 Meredith, George 54–56 Mellmann, Katja 78 Melville, Herman 185 Menchu, Rigoberta 99, 101, 105 Mercier, Pascal (= Peter Bieri) 223, 238 Meskin, Aaron 255–257, 326–329 Metz, Christian 197 Mill, John Stuart 56 Miller, J. Hillis 19–20, 23–24 Mikkonen, Kai 5–6, 169 Mink, Louis 96 Mitry, Jean 341

Nabokov, Vladimir 143, 167– 168, 178, 207 Nash, Mark 178, 196 Neitzel, Britta 341–342, 344, 346 Newcomb, Anthony 283, 292–293 Newman, Ira 5–6 Newsom, Robert 33–34 Niedenthal, Simon 335, 348 Niiniluoto, Ilkka 31 Nisbett, Richard E. 123 Nordrum, Lene 196 Novalis 223, 235–237 Nuttal, A. D. 30, 32, 35 Nünning, Ansgar 62, 250 Olsen, Stein Haugom 4, 10, 41–43, 129, 155, 158–161, 247 O’Brien, Edna 193 O’Casey, Sean 250 Oddie, Graham 31

358 | Index

Osipovich, David 254–255 Ousby, Ian 50–51 Packard, Stephan 116 Paisley, Malcolm 167, 169 Pavel, Thomas 188 Peirce, Charlses Sanders 114 Peli, Oren 199 Petrarch, Francesco 234 Pettersson, Anders 162 Pettersson, Mikael 170, 182 Pier, John 208 Pirandello, Luigi 250 Phelan, James 78 Plantinga, Alvin 141 Plato 7, 81, 224–225 Prince, Gerald 63 Proust, Marcel 18, 134, 189 Polan, Dana 172–173 Popper, Karl R. 30–31 Rabinowitz, Peter J. 78–79 Rancière, Jacques 336 Reiner, Rob 199 Reitan, Rolf 191 Relin, David Oliver 99 Ricardou, Jean 207 Richardson, Brian 62, 143 Ricœur, Paul 273 Robbe–Grillet, Alain 6, 140– 144, 146, 150–153 Robinson, Jenefer 285–286, 288, 342 Robson, Jon 326–329 Rowling, J. K. 107 Rozik, Eli 254, 257–259

Rushdie, Salman 107, 172, 335 Rimmon-Kenan, Shlomith 77, 79–80, 83, 87–89, 188 Rosenblat, Herman 99–101 Rousseau, Henri 147 Ryan, Marie-Laure 61–62, 64, 169, 203, 206, 208–209, 214, 217, 228, 233–235, 245 Ryle, Gilbert 34 Sager Eidt, Laura M. 212 Sainsbury, R. M. 256 Saltz, David Z. 255 Schaeffer, Jean-Marie 203, 208, 220 Scheffel, Michael 80 Schenker, Heinrich 293 Schenk-Haupt, Stefan 250– 252 Schernus, Wilhelm 10 Schlegel, Johann Adolf 225 Schneider Adams, Laurie 211 Schoenberg, Arnold 299–300, 304 Schönert, Jörg 231 Scruton, Roger 159, 249 Searle, John R. 251–252, 255 Segre, Cesare 250–251 Seltzer, Margaret 99–100 Seurat, Georges-Pierre 167, 199 Shakespeare, William 22, 45, 252 Shaw, George Bernard 250 Shklovsky, Viktor 197

Index | 359

Siegrist, Christoph 234 Silcox, Mark 315, 321 Simons, Jan 336 Slugan, Mario 5, 7, 186, 198– 199 Smuts, Aaron 319–320 Sokurov, Alexander 199 Sommer, Roy 62, 250 Staiger, Emil 225 Stam, Robert 207 Stanzel, Franz K. 78–80, 132 Stern, Joseph Peter 17, 29 Sterne, Laurence 197 Stocker, Peter 228, 239 Strawson, Galen 107 Strindberg, August 250 Struth, Thomas 212–213, 217 Tchaikovsky, Pyotr Ilyich 297 Tallis, Raymond 21–23, 341– 342, 345–347 Thon, Jan–Noël 249 Tidman, Paul 148 Todorov, Tzvetan 216 Tolkien, J. R. R. 119 Tolstoy, Leo 19, 157 Toolan, Michael 79 Trissino, Giovanni Giorgio 224 Turner, William 114 Uidhir, Mag 70–71 Van Meegeren, Hans 109 Van Zandt, Steven 174 Velleman, J. David 65–66, 323–331

Verdi, Guiseppe 263 Vermeer, Johannes 109 Voltaire 160 Wagner, Frank 204, 208, 213 Wagner, Richard 260 Walker, Julia 275 Watt, Ian 15–16, 25, 28, 33 Weidle, Roland 249 Weiß, Peter 83–84 Wellek, René 16–17, 37–38 Wells, H. G. 26, 36 White, Hayden 96–97 Wilkomirski, Binjamin 99– 100 Williams, Bernard 259–260, 323 Williams, Simon 212 Wilson, George M. 7, 185– 187, 191, 193–194, 198– 199, 249 Winfrey, Oprah 98, 101 Wittgenstein, Ludwig 274, 288, 304 Wolf, Werner 203, 208, 217 Wollheim, Richard 283 Wolterstorff, Nicholas 257, 287 Woodruff, Paul 279 Woodward, Richard 169 Woolf, Virginia 22, 26, 36 Yagoda, Ben 97–98 Young, Edward 223, 225, 235–237 Young, R. Michael 72

360 | Index

Zeiss, Laurel E. 259–261 Zipfel, Frank 8, 223, 225– 226, 228, 245–247, 249, 251–252, 254, 257, 260 Žižek, Slavoj 198 Zola, Émile 18, 25

Zwart, Sjoerd D. 31 Zunshine, Lisa 123 Zymner, Rüdiger 223–224, 226–233, 238

Contact Addresses J. Alexander Bareis

[email protected]

Chris Bateman

[email protected]

Remigius Bunia

[email protected]

E. M. Dadlez

[email protected]

Jason D’Cruz

[email protected]

Matthew DeCoursey

[email protected]

Eran Guter

[email protected]

Inbal Guter

[email protected]

James Hamilton

[email protected]

Tobias Klauk

[email protected]

Sonja Klimek

[email protected]

Tilmann Köppe

[email protected]

Peter Lamarque

[email protected]

Liviu Lutas

[email protected]

Jukka Mikkonen

[email protected]

Ira Newman

[email protected]

Lene Nordrum

[email protected]

Stein Haugom Olsen

[email protected]

362 | Contact Addresses Mario Slugan

[email protected]

Sarah Worth

[email protected]

Frank Zipfel

[email protected]