Storification of the Self

Transcription

1 Faculty of Arts and Philosophy Storification of the Self Narrative Unreliability in the Metafictional First-Person Novel Supervisor: Prof. Dr. Jürgen Pieters Dissertation submitted in partial fulfilment of the requirements for the degree of Master in de Vergelijkende Moderne Letterkunde by Hanne Carlé August, 2009

2 I m telling you stories. Trust me. Jeanette Winterson 2

3 Acknowledgements First and foremost, I wish to thank my supervisor Prof. Jürgen Pieters for giving me the opportunity to write this dissertation, for his help in delineating the subject and for the time invested in me. Without his guidance and encouragement, this work would not have been what it is now. In addition, I would like to thank Prof. Vervaeck for his time and for providing me with a peripeteic insight when it was desperately needed. I wish to express my sincere gratitude to my family for allowing and encouraging me to pursue this study. I would like to thank my father for the genuine interest and the moral assistance, my mother for the inexhaustible support and for my daily dose of vitamins, my sister Jole for her enthusiasm and for putting up with me, my aunt Greet for believing in me and my grandparents who, I m sure, have lost quite some sleep worrying over this paper. I would also like to express my warm-hearted gratitude to my friends for their support and listening ear. More specifically, I would like to thank Kim for being there and for letting me miss her birthday party when I needed the time to write; Bob for the positive attitude and for being himself; Tinne for the liters of coffee and the diversion; Milica for her sharing generosity; An for the encouraging phone calls; René for the witty messages; and Karen for her spicy postmodern humour that always lifts my spirits. 3

4 Table of contents Introduction p. 6 I. Theoretical survey 1. Theory and definition of the concept of narrative (un)reliability p The Founding Father: Wayne C. Booth p Problematizing the implied author p The respective roles of authorial agency, textual clues and reader response in the interpretation process p Towards a working definition? p Typological distinctions p Facts and ideology p Unreliable Narrators vs. Fallible Characters p Reading Narrative Unreliability p Textual clues to recognize narrative unreliability p The scope of unreliable narration p Estranging and bonding unreliability: the effects of (un)reliable narration on the authorial audience p Theory and definition of narrative self-consciousness p Definition of metafiction p The birth of literary self-awareness p The metafictional paradox p A typology of self-conscious narration p Reading metafiction: freedom and responsibilities for the reader p A double ethical implication p Unreliable narrators in metafictional novels p The narrator: subjective, unreliable and self-conscious p Locating self-conscious and unreliable narration p The ethical implications of unreliable narration in metafiction p Quadruple distancing p L effet-valeur according to Jouve p. 43 4

5 (Dramatic) irony p. 46 II. Dave Eggers, A Heartbreaking Work of Staggering Genius 1. Introduction p The author p The novel p The plot p Analysis p Paratext p Self-conscious parodying of the memoir: title, epigraphs and preface p A helping hand: table of contents and acknowledgements p The memoir p The narrator as flawed human being p Overt self-consciousness p Conclusion p. 65 III. Philip Roth, Operation Shylock. A Confession 1. Introduction p The author p The novel p Analysis p Paratext p Intratext p Narrator(s) in the twilight zone p Narrative self-consciousness p. 73 IV. Conclusion p.76 Bibliography 5

6 Introduction When we read a story, we surrender ourselves to the one who tells it. However, what happens if this narrator is not telling us the truth, the whole truth and nothing but the truth? What if the person depicting our fictional world is deforming the image, whether deliberately or not? In such cases, it is our job as skilled and attentive readers to detect the error on the narrator s part, and recompose the fictional truth to our best ability. We will be guided down the yellow brick road by the author, who will often give us plenty of subtle or less covert hints in order to doubt our companion s trustworthiness. The term narrative unreliability was coined by Wayne C. Booth in his widely acclaimed work The Rhetoric of Fiction (1961), and in spite of recent criticism, his views continue to serve as the foundation for several studies on the topic today. Though unreliable narration might strike us as a feature of predominantly contemporary, (post)modern literature, the phenomenon dates back to the novel s infancy. Once we start doubting the character that guides us through the fictional universe, we become aware of this world s artificiality. The reflecting upon fiction as a construct in explicit or implicit terms is called metafiction, and like unreliable narration, the phenomenon is older than we might think at first. Already in the opening lines of the Don Quijote we notice traces of this proto-postmodernism when the narrator discloses to us that he encountered difficulties when writing his book (Currie 1995: 5). In this dissertation, I will investigate unreliable narration in contemporary metafictional novels. First, I will outline the basic theoretical foundation of both my central concepts, providing the reader with a definition, a typology and an overview of the different academic stances on the subject. Secondly, I will for both techniques investigate the specific consequences for the reader. I will then continue the theoretical survey by trying to determine a common ground between the two narratological phenomena: where are they located, i.e. in what part are they authorial constructions or interpretations by the reader? How can we recognize an unreliable narrator or metafictional passages? How do both narratological phenomena manipulate readers and push them towards a certain interpretation? Finally, I would like to examine whether metafiction and unreliable narration affect each other when used simultaneously in a text, and more specifically whether the use of both techniques has an influence on the effect produced in the reader s mind. For this purpose, I will put theory into practice in the second part of this work and analyse two postmodern novels: Dave Eggers A 6

7 Heartbreaking Work of Staggering Genius and Philip Roth s Operation Shylock. A Confession. 7

8 I. Theoretical Survey 1. Theory and definition of the concept of narrative (un)reliability Once upon a time there was an unreliable narrator, meaning it was not a reliable one. To determine the nature of the not-so-reliable fabulator, we will start by looking at his positive counterpart. 1 Kathleen Wall is one of the first theorists to reflect upon this opposition. In her 1994 work on Kazuo Ishiguro s The Remains of the Day, she defines the reliable narrator as the rational, self-present subject of humanism, who occupies a world in which language is a transparent medium that is capable of reflecting a real world (1994: 20). Likewise, Ansgar Nünning argues in Reconceptualizing Unreliable Narration that [t]he notion of unreliability presupposes that an objective view of the world, of others, and of oneself can be attained (2005: 41). As has been demonstrated throughout history, humans are keen to believe they are capable of providing veracious accounts of events. Think for example of eighteenth-century humanist optimism and nineteenth-century realism. The contemporary reader however, infused by a postmodern disbelief in this possibility, will immediately remark that no such entirely objective narrator exists. I would therefore prefer to define the reliable narrator as a character that reflects a coherent rather than a real world, and does not trigger any textual discrepancies within the fictional universe. In the following part of this dissertation, I will try to provide the reader with a survey of the different views on narrative unreliability and main problems standing in the way of a consensus on an unambiguous definition. I will begin with Booth s more text-centred theory, elucidate the notion of the implied author, and briefly mention some of Booth s like-minded colleagues. The vision of these scholars was strongly contested by cognitive theorists, in particular by Ansgar Nünning, who proposes a more reader-oriented approach. Subsequent to the cognitive-constructivist perspective, I will discuss the rhetorical approach as forwarded by James Phelan, who campaigns the recursive relations between author, text and reader as responsible for narrative unreliability. Finally, I will try to establish a working definition, which I will use as an instrument for analysis in the second part of this work. 1 As both novels in my analysis feature male unreliable narrators, I have chosen to use strictly male or pural pronouns for the sake of simplicity. 8

9 The Founding Father: Wayne C. Booth I have called a narrator reliable when he speaks for or acts in accordance with the norms of the work (which is to say the implied author s norms), unreliable when he does not. (Booth 1961: ) Booth defines the unreliable narrator in his 1961 work The Rhetoric of Fiction as a narrator whose values and/or perceptions differ from those of the implied author, and thus distinguishes itself from its reliable counterpart through a certain distance between himself and this so-called implied author. By introducing this important narratological construct, enfant terrible Booth clearly positions himself against the anti-authorial theories of the in those days dominant New Criticism. He believes the implied author is a second self the flesh-and-blood author creates while writing, and which is the source of the norms, beliefs and final purpose of the text. He asserts that this doppelgänger of the author is an entity that the reader can know, can reconstruct based on the given text. However, he also warns readers that the continuity between the flesh-and-blood person and the implied author can not be interpreted as a total identification: the same real author can create various implied authors in as many texts. According to Booth, the implied author can communicate with readers on two levels: in the first place through the narrator s direct telling of the story, yet at the same time the writer can add a second layer of communication that takes place between him/her and the readers behind the narrator s back and that is signposted by textual indicators. Consequently, when faced with a work of fiction attentive readers can at a certain point detect a distance between the literal meaning of the narrator s words or actions and a second, underlying meaning they are invited to infer. The utterances of the narrator can also be contradictory to what readers believe to be the norms and values of the text as a whole. In such cases, any claim of the narrator s part will not be taken for granted, yet be scrutinized: once the unreliability is detected, readers truth-sensitive antennas are extended, as it were. Booth believes it is possible to draw up an exhaustive list of the textual signals that cause readers to consider a narrator as unreliable. According to the scholar, the unreliability can either be regarding facts or values. Moreover, he claims that a narrator either is or isn t reliable and remains so throughout the entire text. However, as Vera Nünning demonstrates in Unreliable Narration and the Historical Variability of Values and Norms, The Vicar of Wakefield as a test case of a Cultural-historical Narratology, a text and its narrator can be interpreted as reliable in one era 9

10 and as unreliable in another. It seems thus, that however important Booth s coining of the term and theoretical foundations were, he was indeed wide of the mark in these last assumptions. In 1978, Seymour Chatman develops the communication model ; following in Booth s footsteps (he will later make several adjustments to this model, due to criticism by cognitive and rhetorical schools). According to Chatman, a double communication takes place within every fictional text: R author I author (Narrator) (Narratee) I reader R reader Chatman defines the implied author as the patterns of the text or the codes and conventions of the text. 2 This symmetrical model is also used by Monika Fludernik in her important narratological work in The Fictions of Language and the Languages of Fiction. Let us concentrate on the narratological levels of interaction between characters within the fiction, narrator and narratee, implied author and implied reader and finally the real author and his or her public. The communication on each of these levels is of a very different kind from the interaction on the others, and the status of sender and recipient, too, needs to be redefined for each level. (1993: 442) In these text-immanent models, narrative unreliability is a result of the interplay between the discursive levels in a text, and more specifically of the discrepancy between the narrator s point of view and the general meaning or values of the text. 1.2 Problematizing the implied author A vigorous opponent of the text-oriented approach Booth and Chatman promote, is Ansgar Nünning. In his work Reconceptualizing the Theory, History and Generic Scope of Unreliable Narration, he claims that both the concept of the unreliable narrator itself, and more specifically the axiom of the implied author are terminologically imprecise and theoretically inadequate. 2 Chatman, Story and Discourse, Quoted from Phelan 2005: 40 Note: Chatman does nuance the concept of the implied author, which he believes should be renamed inferred author to incorporate the reader s role in the process. 10

11 The postulation of essentialized and anthropomorphized entities designated unreliable narrator and implied author ignores both the complexity of the phenomena involved and the dynamics of literary communication and the reading process, standing in the way of a systematic exploration of the cognitive processes which result in the projection of unreliable narrators in the first place. (2005: 30) According to Nünning, Booth regards unreliable narration as a text-immanent issue, thus neglecting the important role of the reader in the reception and interpreting process. As results from the above citation, the cognitive scholar has specifically strong objections against the theoretical construct of the implied author, which, he argues, is a repository for all difficulties narratology encounters concerning the relation between the author and the reader. [C]ognitive narratologists have argued that instead of postulating an unreliable narrator, whose reliability or unreliability is gauged against the norms of an anthropomorphicized entity designated the implied author, it would be more sensible to conceptualize the relevant phenomena in the context of frame theory as a projection by the reader who tries to resolve ambiguities and textual inconsistencies by attributing them to the narrator s unreliability. (2005: 32) Likewise, Mieke Bal observes that the implied author is a remainder category, a kind of passepartout that serves to clear away all the problematic remainders of a theory (1981: 209). Nünning claims that even in Booth s own definition of the implied author, the phenomenon is not forwarded as a purely textual feature, but rather as a mental image that the reader constructs of the author based on the text. Moreover, the scholar draws our attention to the fact that placing the responsibility for the text in the hands of an anthropomorphicized agent/creator reintroduces the intentional fallacy Wimsatt & Beardsley tried to heed us for: i.e. that the author creates a text with one, ideal interpretation. Susan Lanser believes that the law of parsimony is to be held high and the concept therefore should be abandoned altogether, as [i]t not only adds another narrating subject to the heap but it fails to resolve what it sets out to bridge: the author-narrator relationship. 3 Because Booth bases his theory of unreliable narration on this contested notion of the implied author, it no longer seems sustainable. What does Nünning propose instead? His definition of unreliable narration is as follows: Unreliable narrators are those whose perspective is in contradiction to the value and norm system of the whole text or to that of the reader. The phenomenon of unreliable narration can be seen as the result of discrepant awareness and dramatic irony. The general effect of what is called unreliable narration consists of redirecting the reader s attention from the level of the story to the 3 Susan Lanser, The Narrative Act: Point of View in Prose Fiction, Princeton University Press, 1981: 49f. Quoted from Nünning 2005: 36 11

12 speaker and of foregrounding peculiarities of the narrator s psychology. (2005: My emphasis) Nünning believes that the concept of the unreliable narrator is a naturalization strategy of the part of the reader in order to accord for ambiguities within the text, and proposes a cognitive reconceptualization of the narrative phenomenon. According to the cognitivist, the reader will try to make sense of any textual discrepancies, and does this by attributing them to the narrator s flawlike character. In a similar vein of thought, Tamar Yacobi admits that the technique of unreliability is a cognitive construction of the reader in the meaning-making process, is an inference that explains and eliminates tensions, incongruities, contradictions and other infelicities the work may show by attributing them to a source of transmission (1981: 119). Yacobi calls unreliable narration one of the many reconciling and integrating measures readers have at their disposal in order to account for contradictions in the text (1981: 114). Surely, it is true that the historical variability of the reception of unreliable narration, as demonstrated by Vera Nünning, endorses her namesake Ansgar s view that unreliability is not a textual feature, but is dependent on reader reception and thus a reading strategy. In contrast to Booth, Nünning contends that it is not merely the distance between the world view of the narrator and that of the implied author that causes the reader to perceive of the narrator as unreliable, but also that between the narrator and the world view and standards of normalcy of the reader. A positive side to the cognitivist approach is that the vague and abstract notion of the implied author is left out. However, Greta Olson argues in her article Reconsidering Unreliability: Fallible and Untrustworthy Narrators that even though Booth s model of unreliable narration is text-oriented and Nünning s is reader-oriented, the latter is too aggressive in his attack. Olson argues that the two models are structurally similar, and that Nünning throws out Booth s theoretical baby with the bathwater. Both models have a tripartite structure that consists of (1) a reader who recognizes a dichotomy between (2) the personalized narrator s perceptions and expressions and (3) those of the implied author (or the textual signals). (2003: 93) Simplified, Booth locates the authority to judge a narrator as unreliable with the implied author, whereas according to Nünning that authority lies with the reader. Nünning follows a more constructivist stream of thought, allowing for an infinite number of different interpretations. 12

13 A second positive aspect of Nünning s reader-reception-oriented approach to unreliable narration is that it takes into account the conceptual frameworks readers bring into the text and on which they base themselves when constructing meaning. Determining whether a narrator is unreliable is not just an innocent descriptive statement but a subjectively tinged value-judgment or projection governed by the normative presuppositions and moral convictions of the critic, which [...] recuperate[s] textual inconsistencies by relating them to accepted cultural models. (2005: 40) Besides the intratextual signals Booth points to, Nünning also draws on the reader s extratextual frames of reference, claiming that all readers have knowledge, norms and moral standards set up in their minds prior to the reading of the text. These frames of reference range from general world-knowledge and moral, cultural, social or linguistic values, to specifically literary norms, like e.g. knowledge of genre conventions, pre-existing texts and generic stereotypes. The readers world view and norms will certainly have an effect on whether or not they deem a narrator unreliable. Nünning provides an example to illustrate this: a normal person will read Nabokov s Lolita as going against the opinion of normal moral standards; a pederast is likely to have a different perspective on the text. In this light, the interpretation of a narrator as (un)reliable is not caused by a structural feature of the text, but is governed by pragmatics, and as such, Nünning calls for an interactive model of the reading progress in which cognitivist theories, alongside pragmatics and frame theory, can function as an instrument for studying the reception of the unreliable narrator. One of Nünning s supporters is Monika Fludernik, who considers unreliable narration as one of several possible naturalization theories readers can use to fit a text into their own world view. However useful and necessary the recognition of the important role the reader has in the construction of the figure of the narrator, there is a downside to this approach. In the complex postmodern age we live in today, how can we establish objective norms the unreliable narrator deviates from? Can we speak of common sense and normal moral standards in our diversified world? The problem is that an objective, generally agreed-upon standard by which to judge a narrator simply cannot be drawn up, and the danger of overrelativization renders an objective study of the phenomenon extremely difficult. 13

14 1.3 The respective roles of authorial agency, textual clues and reader response in the interpretation process The cognitivist theories are criticized by the so-called rhetorical approaches for overstating the role of the reader at the expense of the author s agency and to a lesser extent of the textual signals of unreliability. They rightly point out the fact that, like the reader, the author is also defined by frames of reference which have an impact on the text. James Phelan argues in his work Living to tell about it that the writing of a text is a rhetorical act in which authors want to communicate something to their audience, and do so by creating a narrator who tells something to a narratee. The interpretation of a text depends on the interaction between author, text and reader. Phelan asserts that the rhetorical approach locates meaning in a feedback loop among authorial agency, textual phenomena (including intertextual relations), and reader response. In other words, for the purposes of interpreting narratives, the conception assumes that texts are designed by authors in order to affect readers in particular ways, that those designs are conveyed through the language, techniques, structures, forms, and dialogic relations of texts as well as the genres and conventions readers use to understand them, and that reader responses are a function, guide, and test of how designs are created through textual and intertextual phenomena. (2005: 18) The difficulty of the authors task is to render the message while restricting theirselves to the narrator s view. To denominate the creative agent behind the text Phelan maintains the contested notion of the implied author, which he redefines as a streamlined version of the real author, an actual or purported subset of the real author s capacities, traits, attitudes, beliefs, values and other properties that play an active role in the construction of a particular text (2005: 45). In contrast to Booth, he does not consider the implied author a structural feature of the text, but rather as the agent responsible for bringing the text into existence and infuses the text with his or her values. Symmetrically, he adds the notion of the implied or authorial audience, a hypothetical pool of ideal readers for whom the implied author constructs the text. The most important change Phelan makes with respect to previous theories is that he moves the implied author outside the text, but keeps the authorial audience inside the text. As he puts it, the implied author creates textual phenomena for a hypothetical audience, and it is the individual reader s task to become part of that authorial audience by making the right inferences about the author s intention. This implies that in every text a double communication takes place: two speakers, one explicit and one implicit, impart a message to their respective audiences, the narratee and the authorial 14

15 audience. The most important difference between authorial and narrative audience is that the first positions itself towards the text with the tacit knowledge that the characters and events are synthetic constructs, while the second operates within the discourse. Character narration is an art of indirection: an author communicates to her [sic] audience by means of the character narrator s communication to a narratee. The art consists in the author s ability to make the single text function effectively for its two audiences (the narrator s and the author s, or to use the technical terms, the narratee and the authorial audience) and its two purposes (author s and character narrator s) while also combining in one figure (the I ) the roles of both narrator and character. (2005: 1) The use of an unreliable narrator is one of the rhetorical devices an author can employ in order to establish a multilayered communication and achieve a desired effect on the reader. At a certain moment the attentive reader will notice textual inconsistencies, and deduce that the narrator s point of view does not coincide (entirely) with that of the implied author. As Tamar Yacobi puts it, the reader recognizes the character s interference with the facts or their significance (1981: 118). It is this moment of anagnorisis that will trigger readers to consider the narrator unreliable. Where Nünning holds readers alone responsible for the interpretation of a narrator as unreliable, Phelan sees the author as someone who deliberately designs textual discrepancies as signals of unreliability which allow the implied author and the authorial audience to set up a communication behind the narrator s back (this secret communication is another of Booth s notions Phelan preserves). The effect of the unreliable narrator s story thus depends on the implied authors ability to communicate an underlying message to the authorial audience, while restricting themselves to the vision and voice of the narrator as this character addresses the narratee. The tension between the respective viewpoints of implied author and character narrator create the unreliability in the narration, which the reader can infer from textual signals. The feeling that may come out of this is that the narrator is the puppet and readers the kids in the audience trying to warn Little Red Riding Hood that the wolf is right behind her. However, once a certain age the illusion experienced as a child wears away and the puppeteer becomes ever more visible. Likewise, the detection of the manipulating hand of the implied author in a text requires a certain level of interpretative maturity. Phelan stresses that it is important that readers notice the different communicative levels interacting within discourse. The narrator tells her story to her narratee for her purposes, while the author communicates to her audience for her own purposes both that story and the narrator s telling of it. (2005: 18) 15

16 Phelan makes an interesting distinction between these two purposes, which he both calls telling functions. The telling functions that follow the narrator-narratee track, he labels narrator functions; those that run along the narrator-authorial audience track, when the ignorant narrator unintentionally discloses underlying information to the reader, Phelan calls disclosure functions. The rhetorical scholar claims that the latter will ultimately trump the first, because the communication between the implied author and authorial audience is the final goal of the text. Disclosure functions are foregrounded when the author s indirect communication with the reader interferes with the narrator s direct communication to the narratee, as is the case with unreliable narration. 1.4 Towards a working definition? In my perspective, all three of the approaches mentioned above have a point to make, and I would argue that a synthesis of them is probably the clue that will lead us closer to discover the working of the complex phenomenon of unreliable narration. I have opted to retain the contested notion of the implied author, because as Tamar Yacobi argues, the mechanism of narrative unreliability implies a manipulative agent that designed and uses the flawlike character narrator for a specific purpose. The relations between implied author and reader are by definition functional and hence located within the framework of an act of communication. Therefore, when the reader infers an unreliable narrator (or any other fallible observer) who unconsciously reveals his eccentricities and distortions... even this informative aspect of narration forms a part of the intentionality underlying the overall act of communication. To construct an hypothesis as to the unreliability of the narrator is then necessarily to assume the existence of an implied (and by definition reliable) author who manipulates his creature for his own purposes. (Yacobi 1981: 123) Moreover, it will result from my analysis that the concept of an authoritative figure will turn out to be a useful instrument for detecting the values the text as a whole tries to convey. Vincent Jouve claims in La Poétique des Valeurs that the implied author is the figure that allows us to evaluate the character of the narrator on a higher ontological level: it is une instance textuelle qui se distingue du [narrateur] puisqu elle permet d évaluer [le narrateur] (2001: 91). Where I agree with Ansgar Nünning s rejection of Booth s definition of the implied author as a purely structural feature of the text, Phelan s or Yacobi s for that matter reconceptualization of the notion does appear to be useful. In the rest of this dissertation, I will interpret the concept of the implied author as a creative agent that is responsible for the text, and addresses a hypothetical ideal audience. 16

17 The unreliable narrator then, is a character that grants us access to the fictional world, and that is created by the implied author out of language for a specific communicative purpose. The structure of this narrative strategy is, like Booth proposed, tripartite: the reader judges the narrator s (un)reliability based on textual signals, intercepts a non-literal, underlying meaning of the text which express the text s general intention or values and norms. The respective degree of importance attributed to each of these components can vary from text to text. Determining narrative unreliability is not just a gratuitous value-judgement of the reader, but a decision based on a broad range of textual signals. The reader will import conceptual frameworks and knowledge of standards of normalcy into the text, and likewise take those of the (implied) author into account. After this judgement, the reader can predict whether the narrator is likely to be unreliable again and assume a reading strategy. 1.5 Typological distinctions Facts and ideology Several theorists have attempted to make distinctions between various sorts of unreliable narration. The most important division seems to be based on whether we are dealing with a factual of an ethical type of unreliability: Booth claims a narrator can be unreliable about either facts or values, Nünning feels a narrator can be either misrepresenting or misinterpreting, and Dorrit Cohn distinguishes between a factual kind of unreliability and ideological kind that is attributed to a narrator who is biased or confused, which she labels discordant narration. Fludernik takes this typology to the next level, as she adds a third axis. She discovers three modes of narrative unreliability: factual inaccuracy, lack of objectivity and ideological unreliability. In a more elaborate typology, Phelan believes a narrator can be unreliable along three axes: 1. the axis of facts and events: in this case the narrator is guilty of unreliable reporting; 2. the axis of understanding and perception: this would be unreliable reading on the narrator s part; and 3. the axis of ethics and values: here he speaks of unreliable regarding. The rhetoricist makes a further subdivision between six types of unreliable narration which can occur along one or several of the axes: misreporting, misreading, misregarding; underreporting, underreading and underregarding. Misreporting occurs when narrators give us an erroneous account of things, often because of a lack of knowledge or mistaken values on their part. 17

18 Misreading is due to the narrators lack of knowledge and understanding. Misregarding involves unreliability along the axis of ethics: the narrator makes a bad judgement call. Narrators that are underreporting do not tell us everything they know. We speak of underreading when a narrator makes another bad judgement call due to a lack of knowledge. Finally, underregarding occurs when narrators cannot grasp the scope of the event and judge it insufficiently. An important remark Phelan adds to this typology is that narrators do not have to be (and often cannot be) fitted neatly into one of these categories. A narrator can be unreliable in different ways at different points in his or her narrative (2005: 52). According to Phelan, a further distinction between unreliable narrators must be made based on the effect the narration has on the authorial audience, and I will elaborate on this topic when discussing the reception of narrative unreliability Unreliable Narrators vs. Fallible Characters Greta Olson provides us with an important observation concerning narrative unreliability in her article Reconsidering Unreliability: Fallible and Untrustworthy Narrators. She claims it is necessary to make a differentiation between downright untrustworthy narrators and characters that are fallible, i.e. who are not in the position to provide readers with a veracious or authoritative account of events, because they are mistaken in their judgements or biased. Unreliable and untrustworthy suggest that the narrator deviates from the general normative standards implicit in the text. For this reason the narrator cannot be trusted on a personal level. By contrast, inconscience and fallible imply that the narrator makes mistakes about how she perceives herself or her fictional world. The first terms concern the narrator s qualities as a person and the second her ability to perceive and report accurately. 4 (2003: 96. My emphasis) Olson claims that a narrator s fallibility is situationally motivated or that external circumstances cause the narrator s mistaken judgement and that this kind of unreliability is excusable. However, untrustworthy narrators are dispositionally unreliable; any utterance of this type of character will be scrutinized by the reader. Like Phelan, Olson argues that different types of unreliability require a different reading attitude and that a decision on whether a narrator is fallible or untrustworthy allows the reader to predict whether the narrator is likely to always misreport or is prevented by circumstances from telling the tale straight. 4 Note: Olson adopts the term inconscience from Booth s work, which he defines as the mistaken belief of the narrator in his/her own qualities. 18

19 A final notion I would like to elaborate on is sincerity. A fallible character can simultaneously be sincere and unreliable as it does not provide the reader with the required information. Lionel Trilling has defined sincerity in his essay Sincerity and Authenticity as the congruence between avowal and actual feeling. 5 Liesbeth Korthals Altes asserts: [s]incerity has been defined as the adequation of speech or writing to one s intention and character, to one s inner self and one s deeds. Although it is a necessary element of truth-telling, it does not guarantee a truth claim as reliability would but rather the speaker s commitment to his/her saying and said and to the values and meanings implied. (2008: 109) Altes maintains that in speech act and communication theory, sincerity is considered as one of the basic conditions for meaningful human communication, as it entails the projection of the image of a trustworthy speaker, suggesting a basis for reciprocity (2008: 114). In spite of her belief in the necessity of moral genuineness, Altes remarks that in contemporary postdeconstructionism culture, irony and unreliable narration may appear the more sincere mode of expression, as they highlight the risks of communication and the hermeneutic work that are in fact the norm. 5 Lionel Trilling, Sincerity and Authenticity, Harvard University Press, 1972: 2. Quoted from Korthals Altes 2008:

20 2. Reading Narrative Unreliability After having briefly described the main theoretical currents underlying the notion of narrative unreliability, I will now scrutinize the causes and effects of the literary phenomenon. First, I will look at the textual signals that can guide a reader to conclude on such unreliability; several of the authors mentioned above have attempted to draw up a (complete) list of these indicators. Secondly, I will investigate which effect this conclusion has on the reading experience and the meaning the reader infers from the text. 2.1 Textual clues to recognize narrative unreliability Why do readers consider a character narrator untrustworthy? There are several intratextual signals that trigger such an interpretation. Ansgar Nünning claims that [u]nreliable narrators tend to be marked by a number of textual inconsistencies. These may range from internal contradictions within their discourse over discrepancies between their utterances and actions, [...] to those inconsistencies that result from multiperspectivval accounts of the same event. (2005: 44) Kathleen Wall talks about verbal tics or verbal habits of the narrator. As I mentioned above, Booth claims in his The Rhetoric of Fiction that narrative unreliability is a textual phenomenon, which can consequently be subjected to a systematic investigation. In his more recent work A Rhetoric of Irony, the author puts theory into practise and provides us with a list of textual indications of a narrator s untrustworthiness. These include: paratextual elements; direct warnings that the narrator should not be confused with the author; obvious grammatical, stylistic, or historical mistakes on the part of the narrator; conflicts between fictional facts; discrepancies between the values asserted in the work and those of the author in other contexts (1975: 47-86). Nünning provides us with a more exhaustive list. He discovers the following textual indications of narrative unreliability: the narrator s explicit contradictions and other discrepancies in the narrative discourse; discrepancies between the narrator s utterances and actions; divergences between the narrator s description of himself and other character s descriptions of him; contradictions between the narrator s explicit comments on other characters and his implicit characterization of himself or the narrator s involuntary exposure of himself; contradictions between the narrator s account of events and his explanations and 20

21 interpretations of the same, as well as contradictions between the story and discourse; other characters corrective verbal remarks or body signals; multiperspectival arrangements of events and contrasts between various versions of the same events; an accumulation of remarks relating to the self as well as linguistic signals denoting expressiveness and subjectivity, which indicate a high degree of emotional involvement; an accumulation of direct addresses to the reader and conscious attempts to direct the reader s sympathy; syntactic signals denoting the narrator s high level of emotional involvement, including exclamations, ellipses, unmotivated repetitions etc; idiosyncratic verbal habits like stylistic peculiarities/ violation of linguistic norms; explicit, self-referential, metanarrative discussions of the narrator s believability; an admitted lack of reliability, memory gaps, and comments on cognitive limitations; a confessed or situation-related prejudice; paratextual signals such as titles, subtitles, and prefaces. 6 I want to stress that Nünning feels that explicit self-referential metanarrative discussions can be an indication of unreliable narration, a technique that both the authors I study in my analysis have clearly grasped and make extensive use of in their novels. Phelan elaborates on the interesting notion of redundant telling, which he defines as a narrator s apparently unmotivated report of information to a narratee that the narratee already possesses (2005: 11). He attributes this redundancy to the author s need to communicate information to the audience, for example when the audience knows less than the narratee, and claims that redundant telling can equal necessary disclosure. Vincent Jouve argues in his work Poétique des Valeurs that the technique of redundant telling is an important instrument of which narrators dispose to indirectly communicate values to the reader. Jouve claims narrators fulfil three functions through which they can steer readers perception of the text: 1. la fonction idéologique, i.e. when the narrator explicitly formulates a judgement towards characters or events; 2. la fonction de régie: the narrator organises the text according to preference. This function includes the use of redundant telling or the juxtaposition of conflicting viewpoints in order to emphasize and/or express a stance towards an event or character; and 3. la fonction modalisante (the narrator assumes the role of porte-parole of the text). 6 Nünning, Unreliable Narration zue Einführung: Grunzüge einer kognitiv-narratologischen Theorie und Analyse unglaubwürdigen Erzählens, in Unreliable Narration: Studien zur Theorie und Praxis unglaubwürdigen Erzählens in der englischsprachigen Erzählliteratur, edited by Ansgar Nünning, Trier, WVT, Freely adapted from Olson, My italics. 21

22 Certains personnages «ont toujours raison» leurs commentaires (prévisions, analyses, jugements) sont toujours confirmés par les évènements. Un tel personnage fonctionne comme interprète véridique, voire comme porte-parole des valeurs de l œuvre. Une fois qu un tel personnage est constitué, tous ses commentaires tendront à fonctionner comme des commentaires «autorisés». 7 The employment of an unreliable narrator thus points to a high degree of ideological manipulation by the implied author. 2.2 The scope of unreliable narration Both novels I will analyse in the second part of my dissertation feature homodiegetic firstperson narrators. The distinction between homodiegetic and heterodiegetic narration was introduced by the French theorist Gérard Genette, who asserts that a homodiegetic narrator is a character that encounters itself at the same ontological level as other characters, that can talk to them and partakes in the events of the story. A heterodiegetic narrator on the other hand, is a narrator that is situated on a higher ontological level than other characters, and cannot interfere with the story world (think of the objective omniscient narrator in realist novels). Some theorists, like Bruno Zerweck, argue that there exists a package deal between narrative unreliability and homodiegetic narration, because there can be no interpretation of unreliability unless there is an anthropomorphic character that the unreliability can be ascribed to. Zerweck argues that [t]he concept of narrative unreliability is inapplicable if a narrative is transmitted in an impersonal mode or if a text is extremely metafictional. 8 Moreover, heterodiegetic narration seems to have an authoritative aura caused by its nonparticipation in the related events, as existence outside the fictive world [...] naturally links up with reliability (i.e. objectivity resulting from distance and lack of involvement) as well as with omniscience (the range of knowledge which exceeds that inherent in the human condition) (Yacobi 1981: 120). However, more recently Manfred Jahn, Dorrit Cohn and Tamar Yacobi have among others explored the extensive possibilities of heterodiegetic narration. Greta Olson argues in Reconsidering Unreliability: Fallible and Untrustworthy Narrators that she 7 Susan Suleiman, Le roman à thèse, op. cit. p Quoted from Jouve, Poétque des Valeurs, 2001: Bruno Zerweck, Historicising Unreliable Narration: Unreliability and Cultural Discourse in Narrative Fiction, in Style 35, 1, 2001: Quoted from Gunther Martens, Revising and Extending the Scope of the Rhetorical Approach to Unreliable Narration, in Narrative Unreliability in the twentieth-century first-person novel. De Gruyter, 2008: 77 22

23 would agree with Cohn and Yacobi that the same inferences about character, trustworthiness, and personality can be applied to narrators who do not take part in their stories. The degree to which these disembodied voices appear to be part of full-fleshed characters determines the reader s perception of whether they can be thought of as fallible and trustworthy. The less personalized the narrative voice is, I would argue, the more inappropriate it is to infer unreliability. (2003: 106) It does seem to be the case that homodiegetic narrators are more likely to be fallible or untrustworthy, as they do not have the bird s-eye view heterodiegetic narrators have over the narrative world and are thus more liable to erroneous judgements and interpretations. However, character narrators who tell stories they do not take part in can still deliberately deceive or give readers incomplete accounts of events. 2.3 Estranging and bonding unreliability: the effects of (un)reliable narration on the authorial audience Action triggers reaction: once readers have made up their minds whether or not to trust the narrator, this mindset will have consequences for the further reading strategy. The effect of the detection of unreliable narration is twofold: on the one hand the phenomenon refocuses the reader s attention on the narrator s mental processes 9. Ansgar Nünning claims that [t]he general effect of what is called unreliable narration consists of redirecting the reader s attention from the level of the story to the speaker and of foregrounding peculiarities of the narrator s psychology. (2005: 38-39) On the other hand, the change in the readers perception of the narrator s character will cause a change in their emotional response to both the narrating figure and the text as a whole. As the cognitive-constructionist approach to unreliable narration has postulated, a certain narrator can be considered either reliable or unreliable according to the readers respective frames of reference. The variation in interpretation of the narrator can cause a potentially enormous variation in the experience of the narrative world. Ansgar Nünning claims that [d]ecisions which readers make about a narrator s unreliability tend to determine many aspects of the represented world that readers (re-)construct (2005: 67). Once readers consider a narrator to be untrustworthy, they will naturally be suspicious of this character s other utterances. In this respect, Phelan makes an interesting distinction between types of unreliable narrators based on their respective effect on the authorial audience. 9 Wall 1994: 23. Quoted from Nünning 2005: 39 23

24 More specifically, I want to distinguish between estranging unreliability, by which I mean unreliable narration that underlines or increases the distance between the narrator and the authorial audience, and bonding unreliability, by which I mean unreliable narration that reduces the distance between the narrator and the authorial audience. (2008: 9) When confronted with the first type of unreliable narration, the authorial audience understands that the narrator s perspective and the implied author s do not coincide. Readers will consequently be on their guard for the rest of the reading process and every utterance by the narrator will be scrutinized, to the extent that identification with the narrator is made entirely impossible. Reading the second type of unreliability, the authorial audience paradoxically recognizes that the viewpoints of the implied author and the authorial audience may not coincide, yet at the same time understands that in this case the implied author endorses some of the unreliable communication. The reader is aware that the unreliable narrator is the instrument that the implied author uses for a specific purpose, and will consequently play along with the character narrator. Bonding unreliability thus turns the reader in to the accomplice of the implied author. 24

25 3. Theory and definition of narrative self-consciousness Desocupado lector: sin juramento me podrás creer que quisiera que este libro, como hijo del entendimiento, fuera el más hermoso, el más gallardo y más discreto que pudiera imaginarse; pero no he podido yo contravenir al orden de naturaleza, que en ella cada cosa engendra su semejante. Miguel de Cervantes Saavedra, Don Quijote De La Mancha, prologue In the two following parts of this dissertation I will take a closer look at my second central concept: that of contemporary metafiction. In the sixties the postmodern novel was born: writers eschewed the possibility of meaning-making in their work and started analysing both the construction methods of the novel and the arbitrariness of language. Authors of the period turn to playfulness, parody, pastiche and puns as a way of maintaining identity within the postmodern chaos. Interestingly, the use of these self-reflective devices makes metafictional writers at once creators and critics of their own work. Scorned by some critics as the last convulsion of an extinguishing novel form, jubilantly received by others as a fresh wind in a vibrant -if fossilized- story tradition, self-conscious fiction or metafiction has the effect of an earthquake in the literary field. Linda Hutcheon argues in her work Narcissistic Narrative that art has always been illusion, and that it has often, if not always, been aware of that status. She claims self-consciousness is a broad cultural phenomenon, not limited by art form or even by period (1980: 17). Indeed, the erosion of literary foundations is discernible in fiction from the very beginning: early examples of novelistic self-awareness include Cervantes Don Quijote which is not only the first realistic novel, but also the first self-reflective one in that it is aware of and parodies both its own status as a written text (cfr. supra) and the literary conventions of its time ; the phenomenon of the mise en abyme; Shakespeare s plays within plays; frame narratives like Boccaccio s Decamerone; and perhaps above all Sterne s parodic novel The Life and Opinions of Tristram Shandy, Gentleman. Likewise, we can think of early self-referentiality in painting, like the painter in Velasquez Las Meniñas and Jan Van Eyck s literal selfmirroring in the portrait of the Arnolfini-couple. In the following section of this chapter, I will more or less maintain the same structure as in the above discussion of unreliability, starting with the definition of and conceptual problems 25