Subjective Perspectives in Ian McEwan's Narrations

1. Introduction

The quality of [McEwan's] prose is so high that it is easy to sustain interest throughout many readings. It is terse, dry and evocative of powerful emotions. His grammar is meticulous, his words precise and his style direct. When writing about his work, it is almost impossible to précis anything without losing most of its impact, and one can seldom do better than quote paragraphs wholesale. (Byrnes, 2002: 1)

Ever since McEwan's first publications, his work has received considerable attention from critics and scholars. Malcolm claims that he is "certainly one of the most noteworthy of contemporary authors." (Malcolm, ix)^[1] The flood of reviews that welcome every one of his new publications and the number of publications about his work seem to acknowledge that most critics and scholars share this opinion.

Thus, it is not surprising that McEwan has been awarded with a number of prizes for his work. He received, for example, the Sommerset Maugham award for First Love, Last Rites (1976), the Evening Standard award for his screenplay "The Ploughman's Lunch" (1983), the Whitebread Prize for Fiction (1987) and the Prix Fémina Etranger (1993) for The Child in Time, and the Germany's Shakespeare Prize in 1999. Enduring Love was shortlisted for the Whitebread Prize in 1997, as were The Comfort of Strangers and Black Dogs for the Booker Prize for Fiction. In 1998, McEwan finally won the Booker Prize for Amsterdam. Atonement received the WH Smith Literary Award (2002), the National Book Critics' Circle Fiction Award (2003), the Los Angeles Times Prize for Fiction (2003), and the Santiago Prize for the European Novel (2004). For his novel Saturday, he won the James Tait Black Memorial Prize in 2006.

Ian McEwan's family background does not suggest such a fantastic literary career. He was born on 21st June 1948 in Aldershot, Hampshire, as the son of a Scottish sergeant major of the British Army. He spent his childhood on British military bases in England, Singapore, and Libya. During his schooldays at Woolverstone Hall School (1960-66), McEwan developed a keen interest in English Literature. Nevertheless, he took a year off after school, worked for the refuse disposal service and made a trip to Greece before he studied English Literature at the University of Sussex, Brighton, where he received a B.A. degree in 1970. Subsequently, he took a creative writing course taught by Malcolm Bradbury and Angus Wilson at the University of East Anglia, Norwich, where he received his M.A. in English Literature in 1971. During this time, McEwan wrote a number of short stories, several of which were included in his two volumes of short stories First Love, Last Rites and In between the Sheets. In 1972, McEwan made trip to Amsterdam and Afghanistan in his own bus, and after his return, he worked as a teacher for English as a second language in Norwich. Since 1974, he has been living by his writing. In 1982, he married Penny Allen (journalist, astrologer and alternative practitioner), but the marriage was divorced in 1997. He has got two sons from this marriage. In 1997, he married Annalena McAfee, a literary journalist. In 1989, he was awarded an honorary Litt.D. by the University of Sussex, Brighton.

McEwan has written his first short stories in 1970 and managed to publish a view in magazines. However, after publishing his first collection of short stories (First Love, Last Rites [1975]) his career took off. Since 1975, McEwan has published two volumes of short stories (FLLR, and In between the sheets [1978]) and nine novels – The Cement Garden (1978), The Comfort of Strangers (1981), The Child in Time (1987), The Innocent (1989), Black Dogs (1992), Enduring Love (1997), Amsterdam (1998), Atonement (2001), and Saturday (2006). McEwan has also ventured into other realms. He has written a collection of stories for children (The Daydreamer [1995]), three screenplays (The Imitation Game & Other Plays [1980], The Ploughman's Lunch [1985], and Soursweet [1988]) and an oratorio (Or Shall We Die? [1983]). McEwan's work has continuously attracted the attention of critics and reviewers, and McEwan has been praised to be one of the leading representatives of the young generation.^[2]

1.1 Thematic Focus

Despite the extraordinary praise of McEwan's work, it has been discussed most controversially. The fact that he often engages taboo subjects like masturbation, incest, regression, child abuse, dismemberment, sadism-masochism etc. earned McEwan the reputation of an author who writes to shock his audience.^[3]

Although this literature of shock is especially – even exceedingly – prevalent in his first novel and his short story collections, McEwan's desire to startle the reader is evident throughout his canon (even his two most recent novels, The Innocent and Black Dogs, contain scenes overwhelming in their vicious verisimilitude). (Slay, 1996: 7)

McEwan himself appears to be surprised about these attributions and objects them.^[4] Slay also modifies his statement quoted above and asserts that McEwan is certainly not a "writer solely of shock horror" (Sl, 1996: 7), and Nünning affirms that McEwan's prose is not straining after effect.^[5] However, the dark atmosphere in his narrations^[6] is often described as being repugnant:^[7] "Although some [critics] find his unsavoury and often appalling subject matter repugnant, even repulsive, most contend that its shock value is reflective of the world itself." (Slay, 1996: 5) Like many other critics, Slay asserts that McEwan uses the shock value of his narrations to "warn […] the people of a bewildered age." (Ibid.)

While the reasons for McEwan's intense employment of repulsive topics are discussed controversially, it seems to be undisputable that the narrations are extremely shocking. However, it has hardly ever been considered why the narrations cause such unease to the reader. It appears that critics and scholars tacitly agree that the shocking effect of the narrations is caused by McEwan's choice of repugnant subjects. The topics, however, cannot be the only reason for the fact that readers feel uncomfortable with McEwan's narrations. If this were the case, the extreme emotional response to McEwan's narrations would appear to be entirely disproportionate in view of topics of crime and horror stories or thrillers that can be viewed as being just as macabre or perverse. However shocking the topics, crime stories and thrillers do not seem to shock the readers as much as McEwan's narrations do. Although readers might condemn outraging behaviour of characters in crime stories, the topics nevertheless seem to cause only a tingly thrill, not revulsion and repugnance.

In my opinion, the shock value of McEwan's narrations is mainly caused by his particular way to present these topics. In keeping with this, I consider the form of McEwan's narrations to be as important as their content. Scholars usually assume that formal aspects of McEwan's narrations can be neglected, and mainly concentrate on the interpretation of the content of McEwan's narrations. Nünning and Oeser, for instance, claim that McEwan does not experiment formally in his narrations,^[8] and furthermore Malcolm states that "metafictional concerns, that is, self-reflexive concerns with fiction's possibilities and problems" (Malcolm, 2002: 10) are only considered on the level of topic:

He [McEwan] is certainly frequently concerned with literature itself as a topic, and the difficulties, possibilities, and complexities of giving an account of things and of telling stories in general. Such concerns, however, in his later fiction at least, remain on the level of topic (that is, they are discussed as subject matter) without penetrating the technical level of the text itself (that is, they are not embodied on the level of narration, language, narrative organization, or genre, among others). (Ibid.)

Although McEwan's narrations do not primarily appear to experiment with the form, it can be discovered that McEwan seems to experiment with the employment of perspectives,^[9] subjective perspectives in particular. In fact, it can be argued that he taps the full potential of the employment of subjective perspectives in his narrations as the reader is confronted with the subjectivity of perspectives on all levels of textual communication. In my opinion, McEwan's most outstanding accomplishment is his ability of depicting subjective perspectives in all consequence. The absence of morality in many of McEwan's narrations, for example, which is usually regarded as an underlying topic, can also be seen as a result of depicting consistently a specific subjective perspective.

In the context of subjective perspectives, the aspect of unreliability in McEwan's narrations has to be considered. Often, McEwan's I-narrators are regarded as "unreliable narrators".^[10] However, it is striking that even in narrations without an "unreliable narrator" the reader seems to be nevertheless confronted with unreliability. In my opinion, the aspect of unreliability cannot be tied up with the concept of an "unreliable narrator", but it is instead linked to the subjectivity of perspectives. The aspect of unreliability seems to constitute a main feature of subjectivity, and thus, unreliability in McEwan's narrations appears to be a result of depicting coherently a specific subjective perspective. Therefore, unreliability will be examined in some detail in this paper.

The choice of topics, the absence of moral censor, the remarkable isolation of characters and the general unreliability in McEwan's narrations are all aspects that can be linked to McEwan's consistent depiction of extreme subjective perspectives. The consequent employment of subjective perspectives in addition with the absence of any objectivity or reliable 'truth' in general, which confronts the reader^[11] with his own subjectivity, is, in my opinion, mainly responsible for the shocking effect of McEwan's narrations. Thus, the formal aspect of coherent depiction of subjective perspectives appears to be just as important as the aspect of content with regard to McEwan's narrations. Therefore, this paper will examine the subjective perspectives in McEwan's narrations in all detail.

The nature of the subjective perspective has not been examined at large in narratology, although parts of this issue have been discussed thoroughly e.g. "unreliable narration", "perspectives" or the "subjective novel", etc. An applicable structure for analysing the subjective perspective on all levels of textual communication is inexistent so far. However, such a structure is required for analysing the subjective perspectives in McEwan's narrations and therefore it will be developed in the frame of this thesis. This new developed structure could also prove to be valuable in textual analysis in general.

1.2 Outline

Before analysing the subjective perspectives in McEwan's narrations in chapter 6, the theoretical basis will be provided in the preceding chapters. In chapter 5, an applicable structure for analysing subjective perspectives on all levels of textual communication will be presented. The findings of considerations in the chapters 2 – 4 will provide the basis for this structure.

In chapter 2, the relevant narratological context will be introduced. For example, diverse models of textual communication will be discussed, and narratological termini will be considered. In the third and fourth chapter, the context of subjectivity and of unreliability will be examined respectively. The aspect of unreliability will be located in the system of textual communication that has been developed in chapter 2. In this context, the perspectives on the textual level will be introduced.

Based on the considerations in the chapters 2 – 4, an applicable structure for analysing subjective perspectives on all levels of textual communication will be developed in chapter 5. In the examination of subjective perspectives, the different levels of textual communication (level of fabula, mediation level, textual level) will be considered. As the intra- and the interpersonal aspects of the levels of fabula and of mediation are similar for an examination of the subjective perspectives, they will be treated in a single chapter of story-internal perspectives (5.1). However, the perspectives on the mediation level are nevertheless superior to the perspectives on the level of the fabula, and some aspects are only relevant for the perspectives of the mediation level. These will be treated in 5.2. In 5.3, the perspectives on the textual level will be further clarified.

Having developed an applicable structure for analysing the subjective perspectives in narration in the fifth chapter, this structure will be applied to McEwan's narrations in the following chapter in all detail. In 6.1, the intrapersonal dimension of perspectives in McEwan's narrations will be analysed. For this purpose, the main aspects of the intrapersonal dimension will be summarised in groups (6.1.1). Chapter 6.1.2 will focus on the effect that these aspects have on the protagonists or respectively on the reader. Namely, the aspects of morality and of the ordinariness of actions will be discussed. In 6.2, the structure of interpersonal relation of perspectives which has been developed in 5.1.2 will be applied one-to-one to McEwan's narrations. Thus, the issues of 'understanding of other perspectives' (6.2.1), 'misunderstanding other perspectives' (6.2.2) and 'no understanding of other perspectives' (6.2.3) will be examined. While the chapters 6.1 and 6.2 mainly describe the subjective perspectives in McEwan's narrations, the focus of chapter 6.3 will be on aspects that concern the evaluation process of subjective perspectives. Hereby, it will be analysed in how far the perspectives on the mediation level evaluate the perspectives of the fabula (6.3.1) and in how far the perspectives on the textual level evaluate the perspectives on the mediation level (6.3.2). In 6.3.3, the evaluation process of the reader concerning the perspectives on the textual level will be scrutinised. In this context, it will be focussed on the reliability of the perspectives on the textual level. The seventh chapter will summarise the results of this paper.

The chapters 2 – 4 deal respectively with a variety of topics that are not divided into single sections in order to maintain a fluid reading in a continuous text. However, in order to enable the reader to associate the text with the respective topic, keywords are highlighted. In the chapters 5 and 6.3, the highlighted catchwords facilitate the reader to follow the structure of the text, which is quite complex. In 6.1 and 6.2, sub-topics are captioned.

The terminology that is respectively relevant for a specific chapter will be italicised in that chapter. Terminology that is not in the focus of a chapter and that has been defined before will not be italicised anymore in order to enhance readability. Termini that are frequently used will be abbreviated according to the list in the appendix.

2. Narratological Context and Terminology

In this chapter the narratological context is expounded and the terminology is established which subsequently will be used. This determination of key terminology is fundamental for the development of a system to analyse subjective perspectives in narration (chapter 5) that will be applied to McEwan's narrations in chapter 6. For that reason, a model of textual communication will be developed that emphasises the distinction of the levels of textual communication. Further­more, the terms "perspective" and "point of view" are differentiated. Moreover, the categories of the narrating authority and intradiegetic narrating authorities will be derived from Bal's concept of focalization and placed in the textual communication system. Finally, the notion of shifting intradiegetic narrating authorities will be examined using A. and V. Nünning's concept for structuring multi-perspectival narrations in a paradigmatic and a syntagmatic way.

The distinguishing difference between narrative texts and dramatic texts is the existence of a narrator in narrative texts. This narrator is not the actual author but is invented by the author.^[12] Chatman's diagram^[13] illustrates the flow of textual communication:

Real author à Implied author à (Narrator) à (Narratee) à Implied Reader à Real Reader

In his model, the real author and the real reader are outside the narrative transaction, while the 'implied author' and the 'implied reader'^[14] are text-immanent. However, they are not part of the story. Wolgang Weiß^[15] summarises various models of textual communication. The advantage of his illustration is that it shows the different interlocking levels of textual communication:

illustration not visible in this excerpt

The white frame shows the ontological reality of author and reader; in grey, the level of the discourse is shown, which embeds the narrator and the fictive addressee, and in pink, the level of the story is illustrated, which embeds the protagonists of the narration. Autor' and Leser' are both constructions that represent those aspects of the text which are relevant concerning its reception, e.g. in children's books, the child as the recipient is implied in the text itself.^[16]

For this thesis, a model has been developed that contains aspects of both models, but mainly emphasises the distinction of the different levels of textual communication:

illustration not visible in this excerpt

= fabula = mediation level = textual level

The differentiated terminology of the different levels is crucial for the analysis instrument that will be developed in chapter 5.^[17] Here the story includes, beside the fabula (Bal, 1985), the mediation-level^[18] as well, since the narrators, focalisers and narratees appear to be part of the story though on a different level.^[19] This assumption is congruent to Mieke Bal's presupposed textual communication model (1985; see next paragraph). The mediation-level itself excludes the fabula. The characters act out the story – they are the centre of the story in the fabula itself (orange). The narrator of the fabula is outside the central story on a mediation-level (yellow). Although he still belongs to the story, he operates on another level of this interlocked system as long as he does not interfere with the fabula but only narrates it or comments on it.^[20] In this context, a narratee is only existent, if the narrator explicitly addresses the narratee; otherwise, the narratee is directly substituted by the 'implied reader'. In addition to the narrator, a focaliser is another narrating authority who operates on the mediation-level although the narrated views, feelings and ideas belong to the corresponding protagonist on the fabula. This will be set out in detail later in this chapter. The 'implied author' and the 'implied reader' are both on a text-internal level, which is outside the story (yellow + orange). It is called here the textual level^[21] (grey). Although 'implied author' and 'implied reader' are text-immanent, they do not 'interfere'^[22] with the story. This model of textual communication creates the context for the assignment of subjective perspectives in narration.

Before approaching the different manifestations of subjective perspectives in narration (chapter 5), it seems advisable to have a look at the term "perspective". The term is derived from the Latin noun Perspectiva, which basically means the "science of sight".^[23] Its rising epistemological significance in the past centuries has developed the epistemological dualism of subject and object, which finally culminates in the theory of perspectivism, an approach in which all cognition depends on the recognizing subject (cf. Nünning, 2000: 9). Nünning's preference for the term "perspective" originates in its slightly shifted meaning in the last century:

Der im 20. Jahrhundert zugrunde gelegte Begriff von Perspektivität ist Ausdruck eines intellektuellen Pluralismus, der die Relativität aller Standpunkte betont; er situiert den Begriff 'Perspektive' innerhalb des wahrnehmenden Individuums: 'Perspektive' meint demnach nicht die Beziehung eines Menschen zur Welt, sondern bezeichnet die Voraussetzungen, auf deren Basis ein Individuum Beschreibungen von Welt erzeugt. (Nünning, 2000:10)

The term "point of view", which often is used in the same sense as "perspective" in narratology, is strongly connected to Stanzel's Theorie des Erzählens (1979). Despite the appreciation of Stanzel's theory in general, the undifferentiated use of the term point of view for the categories of his Erzählsituationen has frequently been criticised.^[24] In comparison, Genette distinguishes between the narrator und the focaliser: the narrator "speaks" and the focaliser "sees".^[25]

Mieke Bal has examined focalisation in more detail in her essay "Focalization" (1985)^[26]. She defines it as following:

Focalization is the relationship between the 'vision', the agent that sees, and that which is seen. […] Consequently, focalization belongs in the story, the layer in between the linguistic text and the fabula. (1996: 118)

Bal's meaning of focalizor, which is derived from her definition of focalization, is congruent to the narrating authority^[27], which includes both the narrator and the focaliser. Furthermore, Bal obviously situates the term in a textual communication system when she localises it "in the story […] between the linguistic text and the fabula" (ibid.). She refers to a level of textual communication which is congruent to the mediation-level.

In this context, Bal develops the concept of character-bound focalization according to which she combines I-narrator and focaliser in one category of intradiegetic narrating authorities.^[28] She points out that character-bound focalization"will have a technical advantage over the other characters. The reader watches with the character's eyes and will, in principle, be inclined to accept the vision presented by that character" (118-119). On the other hand, it brings about "bias and limitation" (119) and is therefore 'subjective'. A third-person narrator (which she calls an external focalizor), conversely, may appear 'objective' because his bias is not apparent: "The [external] focalizor's bias is, then, not absent, since there is no such a thing as 'objectivity', but it is unclear." (120) So even external third-person-narrators are finally biased, limited and subjective, though the subjectivity is not as obvious as in intradiegetic narrating authorities.

The phenomena of the subjective perspective of an intradiegetic narrating authority (Bal's character-bound focalizors) being broken by a multitude of subjective perspectives deserves special attention. This is the case when the narrating authority shifts from one focaliser or I-narrator to another.^[29] Ansgar and Vera Nünning have studied and defined this phenomenon in detail.^[30] In order to examine the structure of multi-perspectival narrations, A. and V. Nünning consider the single perspectives and their relation to each other.^[31]

The reconstruction of the single perspectives is called the paradigmatic dimension and the relation between the single perspectives is called the syntagmatic dimension (2000: 52). The paradigmatic dimension is concerned with a variety of parameters, which are:

- "der quantitative Umfang des Perspektivenangebots"
- "die Breite des Spektrums der Figurenperspektiven"
- "Grad an Konkretisierung der Figurenperspektiven
- "Maß an Ausgestaltung der Erzählerperspektive"
- "Grad an Repräsentativität"
- "Grad an Zuverlässigkeit oder Glaubwürdigkeit"
- "Grad an Autorität" (Nünning, 2000: 52ff)

The syntagmatic dimension is concerned with assignment and correlation of the perspectives which can be contradictory or correspondent:

- "hierarchische Relationierung der Perspektiven"
- "quantitative Relationierung der Perspektiven"
- "syntagmatische Relationierung der Perspektiven"
- "temporale Relationierung der Perspektiven"
- "lokale Relationierung der Perspektiven"
- "informationsmäßige Relationierung der Perspektiven"
- "normative Relationierung der Perspektiven"
- "inhaltliche Relationierung der Perspektiven" (Nünning, 2000: 55ff)

These categories refer to multi-perspectival narrations and certainly cannot be directly applied to mono-perspectival narrations. However, the idea of examining the structure of subjective perspectives in terms of a syntagmatic and a paradigmatic dimension appear likely to be helpful. They form the context for the narratological structure of the story-internal perspectives, which will be discussed in detail in chapter 5.1.

Summary

In this thesis, a model of textual communication is developed which emphasises the distinction of the different levels of textual communication in order to place the related perspectives on those levels:

a) level of the fabula – perspectives of characters
b) mediation-level – perspectives of narrating authorities and narratee
c) textual level – textual perspective and perspective of the 'implied reader'
d) level of ontological reality – perspectives of the author and the reader

The story includes fabula and mediation-level but it is not an 'extra' level of this interlocked system. It is mentioned in the diagram in order to provide the context of the terms story-internal or story-external perspectives.

The term perspective is preferred to the term point of view, which is strongly connected to Stanzel's theory. Genette's distinction of narrator and focaliser invites an exploration of the concept of focalisation.

In this context, Bal's combination of focaliser (in Genette's sense) and narrator into one category, which is subsequently called narrating authority, is useful. She places this in a textual communication system between fabula and linguistic text, which is congruent to the mediation-level. Furthermore, she combines I-narrator and focaliser (in Genette's sense) into a category of what she calls 'character-bound focalization', which subsequently will be called intradiegetic narrating authorities, and describes their limited and biased perspective, their subjectivity.

In the context of subjectivity, Bal's concept of shifting intradiegetic narrating authorities which might feign objectivity is of special interest. In order to define and classify this phenomenon of multi-perspectival narrations, A. and V. Nünning have explored the structure of those narrations with the

help of a syntagmatic and a paradigmatic dimension. The concept of exploring structures of perspectives in a syntagmatic and a paradigmatic way promises to be useful although this concept will have to be expanded to cover mono-perspectival narrations as well.

3. Subjectivity in Narration

As the question of subjectivity is a fundamental aspect of this thesis, it will be dealt with in some detail in this chapter. First of all, the term "subjectivity" will be defined. After that, the subject-object-opposition and the relation between 'I' and 'reality' will be examined. It will be shown how the dissimilar conceptions of reality lead to differing approaches towards dealing with reality. In accordance with the uneven conception of reality, the difference between "subjective" and "objective novel" will be examined. Moreover the consequences of the development of a consciousness for subjectivity will be examined with regard to the protagonists and the author of a novel. Finally, the "subjective perspective" will be described in the context of the textual communication system as it is the crucial focus of this analysis.

The term "subjectivity" has various meanings. For example, the encyclopaedia Deutsches Fremdwörterlexikon defines the term "Subjektivität" as follows:

fachspr. in der Philosophie und Ästhetik in der Bed. 'Eigenschaft, subjektiv zu sein; subjektives Wesen, Individualität, Innerlichkeit', auch allgemeiner ver­wendet für 'rein persönliche Auffassung, subjektive Einstellung; Eigenart, Eigentümlich­keit' und leicht abwertend für 'Einseitigkeit, Voreingenommenheit, Unsachlich­keit'. (1978: 546)^[32]

The definitions show that subjectivity implies not only the aspect of an individual point of view but also that subjectivity is seen in opposition to objectivity which allows the more negative meaning. While subjectivity stands for restriction or for having no real foundation (see above), objectivity seems to embody some universal knowledge. This ' subject-object-opposi­tion ' (see Gessner-Utsch, 1994: 19ff) has been widely discussed. In her work, Bettina Gessner-Utsch has described fundamental considerations on this subject-object-dualism which can be applied to the subjective perspective as well.^[33] The decisive question con­cerns the relation between 'subject' and 'world', between 'I' and 'reality'. Gessner-Utsch has dis­tinguished two distinct reality-conceptions: a) the idea of an 'objective reality' and b) the conviction of the existence of innumerable subjective reality-models^[34]. The difference between these models of 'objective' and 'subjective' reality lies in the kind of relation between reality and subject:

Die Formulierung 'objektive Wirklichkeit' besagt, dass dabei an eine Wirklichkeit gedacht wird, die in ihrer Existenz und Beschaffenheit als unabhängig von jeglichem Subjekt an­gesehen wird, eine Wirklichkeit also, mit der die Gesamtheit der Subjekte in gleicher Weise konfrontiert ist. Ganz am anderem Ende einer Skala von möglichen Wirklichkeits­konzeptionen [...] steht dieser Auffassung, die natürlich per definitionem von der Wirk­lichkeit spricht, die Vorstellung einer Pluralität subjektiver und gleichermaßen gültiger Wirklichkeitsentwürfe gegenüber, deren Gestaltung den einzelnen Individuen obliegt. (5)

A. J. Ayer phrases the two opposite concepts as "the world as it is in itself" and "the world as we conceive it" (1973: 12). The ' radical constructivism ' concept goes even further and claims that all reality is constructed.^[35]

This thesis supports a moderate approach. Gessner-Utsch presents a model in which not everything is constructed, but the constructed reality is based on some 'raw material':

Solche Ansätze [Schulz, 1979] stellen einen Mittelweg zwischen rein realistischen Positionen auf der einen und rein idealistischen auf der anderen Seite dar, indem sie unter Beibehaltung der Vorstellung einer jeglichen Wahrnehmungs- und Erkenntnisakt als "raw material" [Ayer, 1973] zugrundeliegenden 'objektiven Wirklichkeit', die vom Subjekt aber im Erkenntnisprozeß erst strukturiert und damit 'realisiert' wird, der wirklichkeits-gestaltenden Leistung von Subjektivität Raum lassen. (23)

Considering modern or post-modern literature subjectivity and the subjective perspective in narrations is not only a stylistic device but in a world of individualism, it has to generally be taken into account. Gessner-Utsch claims that the central issue of the 'subjective novel' is not the concreteness of 'the reality' but the examination and the analysis of the perceived world: "[…] statt 'der Wirklichkeit' stellt er [der Bewusstsein(strom)roman]^[36] die Auseinandersetzung des Bewusstseins mit Wirklichkeit in den Vordergrund." (1)

The opposition of subjective and objective reality-models is expressed in the different approaches 'reality' is dealt with. Gessner-Utsch explains that, under the conviction of the existence of an 'objective reality', the role of the subject is minimised to recognise this reality, while under the conviction of the existence of 'innumerable subjective reality-models', the subject has to constitute reality itself. (5) She explains this difference in more detail:

... wenn von 'der Wirklichkeit' die Rede ist, meint man einen umfassenden und allgemeingültigen Rahmen, in dem sich alle Menschen und Ereignisse situieren lassen. Zu diesem Rahmen gehören u.a. auch festgelegte Werte und Normen, die die Orientierung für das Leben des Einzelnen vorgeben. Geht man dagegen davon aus, dass im Grunde jeder Mensch seine eigene Wirklichkeit entwirft, dann muss – bzw. kann – auch die Sinngebung, die Zielvorstellung, die die Orientierung liefert, vom Individuum selbst entwickelt werden. (5)

In a novel with the idea of an 'objective reality', the protagonist is not supposed to create a purposeful reality but only to recognise a given meaning. When he has managed to find his place in the given reality, the process to assimilate this reality ceases. So the process of examining reality in an 'objective novel' is finite.^[37] In a 'subjective novel', the process of examining reality is infinite because the process of creating reality never ceases. Hence, 'the' reality is never fixed but constantly in motion.

Gessner-Utsch embeds the development of the 'subjective novel' into the historical situation of a crumbling picture of an 'objective reality'.^[38] Nevertheless, this process is not described as excluding. Instead, objective and subjective reality-models overlap (9). While Gessner-Utsch is mainly concerned with the characters, Fielitz examines especially the subjectivity of the author:

Basierte der realistische Roman des 19. Jahrhunderts noch auf der Annahme, ein Autor könne ein getreues Abbild der Realität in seinem Roman geben, so setzte sich im 20. Jahrhundert immer stärker die Annahme durch, dass die Welt nicht objektiv beschreibbar sei, da der Autor selbst Teil dieser sei und diese entsprechend nur 'gefiltert' durch seine eigenen Sinne und Erfahrungen erfassen und abbilden könne. (2001: 55)

The question arises as to what aspects are the characteristics of a subjective narration. Gessner-Utsch's approach of defining the 'subjective novel' ("subjektiver Roman") concentrates on the topic of the novel: "Roman, in dem in erster Linie Subjektivität thematisiert wird" (3). Jerome Hamilton Buckleys calls a novel subjective when the author "display[s] an innermost self to a putative audience" (1974: VII) with which he refers to the autobiography genre. For Stanzel, all I-narrators are biased^[39] by which he means that they are subjective. Nevertheless, the presence of an I-narrator is not a definite indicator for a 'subjective novel' as Gessner-Utsch explains with the example of Robinson Crusoe, which – in spite of the subjective perspective of the narrator – can more be seen as an 'objective novel' because of its objective reality-model (7ff). However, Stanzel's idea that perspectivity and subjectivity are related^[40] cannot be denied. In fact, as soon as a perspective is to be found, it cannot be objective – it is a subjective perspective. Hence, I will not concentrate on the 'subjective novel' but will focus in the following on the various manifestations of the 'subjective perspective' in narrations which is more suitable in regard to McEwan's narrations.

On the foundation of these considerations, the subjective perspective can be examined now in some detail. A subjective perspective might be found in every level of a narration. In the fabula, the characters quite naturally are subjective and take up subjective perspectives. The subjective perspective, though, is only apparently perceivable when the protagonists either narrate or focalise them­selves. If they do not narrate or focalise themselves, they still have a subjective point of view but it is not directly perceivable – it is filtered by another authority (usually the narrating authority) with its own subjective interpretation of the events. A narrator who does not participate in the fabula is also subjective, although the extent of the subjective restriction surely varies from narrator to narrator.^[41] The narrator is still on the story-level. Even outside the story-level, there are subjective perspectives to be found. This means that there may exist a restricted and exclusive perspective through which the narration is to be entered – or valued (perspectives of the 'implied author' and 'implied reader', see next chapter). Even though the reader might have the freedom to correspond to this perspective or to disagree with it, it might still be understood. Usually this perspective with its subjec­tive values and norms is taken for granted unless a form of unreliable narration is suspected.

Summary

The definitions of the term "subjectivity" show that subjectivity is always part of a subject-object-dualism, whose central question is the relation between 'I' and 'reality'. In this context, Gessner-Utsch distinguishes two distinct conceptions of reality: the idea of an 'objective reality' in contrast to the conviction of the existence of 'innumerable subjective reality-models'. Ayer calls the opposites "the world as it is in itself" and "the world as we conceive it". The concept of the radical constructivism even claims that all 'reality' is constructed and that there is no 'objective reality' at all. Nevertheless, in this thesis a moderate approach is supported in which the constructed subjective reality is based on some 'raw material'.

The opposite reality-models manifest different approaches to deal with reality: under the presumption of the existence of an 'objective reality', the subject has to recognise this reality while under the presumption of the existence of 'innumerable subjective reality-models' the subject has to construct his own subjective reality by interpreting the perceived world. Hence, the process of examining reality in what Gessner-Utsch calls a 'subjective novel' is infinite and the process of creating reality never ceases, while it is finite in an 'objective novel' and finishes as soon as the protagonist has managed to recognise a given meaning.

In a historical situation in which a consciousness for individuality and subjectivity is being developed, the picture of an 'objective reality' crumbles. As a consequence, the task of the protagonists in novels changed from 'discovering reality' into 'constructing reality'. Also, the objectivity of the author was increasingly questioned during the 20th century, although the subjectivity of the author is peripheral in this context. Concerning the subjectivity of perspectives, the existence of an intradiegetic

narrating authority is a first obvious indicator for a subjective perspective. Additionally, perspective and subjectivity are closely linked and as soon as a perspective is to be found, it cannot be objective – but is instead a subjective perspective.

Placing the concept of the subjective perspective in the textual communication system, it is apparent that subjective perspectives do exist on each level: the perspectives of the characters in the fabula, the narrating authorities/narratees on the mediation-level and the perspectives on the textual level are all more or less subjective. The examination of McEwan's narrations in chapter 6 will show that his work is exemplary for the intensive employment of subjective perspectives on all levels.

4. Unreliability in Narration

Whenever dealing with subjective perspectives, sooner or later the question of reliability of those perspectives arises because of their bias and their limited vision. For this reason, in this chapter the concept of unreliability in narrations will be explored in some detail. At first, the term unreliable narrator will be defined and examined. Then, the premises for judgments of unreliable narrations in the context of the subjective perspective will be explained. Furthermore, actual and normative unreliability of narrators and focalisers will be distinguished and located in the textual communication system. In this context, the concepts of a textual perspective (t.p.) and of a perspective of the' implied reader' (p.i.r.) (or the recipient's role [r.r.]) are introduced and an explanation will be given of how these perspectives might prove to be unreliable. Here the point of reference is clarified as well as the problems that arise from the use of systems of values-and-norms as point of reference. The introduced terminology will be fundamental in the analysis of subjective perspectives in McEwan's narrations.

First of all, the context to the terms unreliability or unreliable narration is to be examined. Since Booth created the term unreliable narrator, it has become a fixed term in the discussion of narratology. Booth defines it as follows:

For lack of better terms, I have called a narrator reliable when he speaks or acts in accordance with the norms of the work (which is to say, the 'implied author's' norms), unreliable when he does not. (1961: 158f)

Although this definition has found entrance into most encyclopaedias, Booth himself and others have regarded it as being vague and indistinct, especially since the term of the 'implied author' – or his norms – are not clearly defined. Stanzel criticises in particular that Booth uses the term unreliable narrator for I-narrators and focalisers equally without any differentiation. (Stanzel, 1989: 202) In Stanzel's opinion, only an I-narrator is able to be unreliable – not a focaliser. Certainly Stanzel is right in claiming that a figure that is not aware of being in the process of narrating cannot be unreliable on purpose. Nevertheless, excluding the concept of unreliability entirely for focalisation seems to be going too far since the given perspective of a focaliser can be unreliable for the reader as well.

Since Booth's exposition, many narratologers have discussed the subject of unreliable narration.^[42] It is described in particular detail in Ansgar Nünning's anthology Unreliable Narration – Studien zur Theorie und Praxis unglaubwürdigen Erzählens in der englischsprachigen Erzählliteratur (1998) . However, in this thesis, which is based on the theory of subjectivity, the issue of unreliability will be treated from a slightly different point of view.^[43] Here, the premise for any judgment of a perspective of a figure is the theory of a moderate constructivism (see chapter 3), which implies that all subjective perspectives are to be regarded as being of equal value and that there is no 'objective reality' (only some 'raw material') – thus no objective values and no norms. Therefore, it is not possible to judge any subjective reality-model to be generally out of the norm. In keeping with this, it can be said that all narrators, focalisers and recipients are equally as subjective because of their limited perspective. In a world of multiple subjectivity, the term unreliable narrator fits only for an I-narrator who purposefully tries to deceive the narratee or who modifies the text-internal truth on purpose. All other characters are merely 'subjective'.

Also Stanzel and Fludernik assume that only an I-narrator can be an unreliable narrator. (Cf. Stanzel, 1989: 202; Fludernik, 1996: 213.) It seems that all other forms of unreliability – normative unreliability of a narrator^[44] and actual or normative unreliability of a focaliser – are to be found on a story-external level. It is not possible to assign them directly to a narrator or focaliser who only describes his subjective perspective. These forms of unreliability have to be assigned to the perspective of the text – the textual perspective (Textperspektive; Hof, 1984: 49) – or to the perspective of the 'implied reader'. This hypothesis will be explained in the following passages.

When the story-level is left, it can be ascertained that the same story with the same narrators, values and norms, occurrences and details, can still be viewed from various perspectives. The 'implied author' – to stick to the vague but nevertheless well-know term initially – takes up one of these various perspectives. Each perspective (each 'implied author') is, of course, subjective with its own subjective reality-model, own values and norms, own criteria of selection and subjective limitations. Nevertheless, these aspects can hardly be extracted from the text. (Cf. Stanzel, 1989: 201; Rimmon-Kenan, 1983: 101.) Furthermore, it appears that the 'implied author' is regarded as a kind of shadow of the real author whose intentions or 'perspective' are represented. Obviously, this is not the case. Thus the concept of the 'implied author' remains most questionable. (Cf. Nünning, 2000: 8f.) As Nünning claims that every per­spective is in­stalled in a subject (Nünning, 2000: 10), it appears that there can not be such a perspective without the 'implied author'. Nevertheless, it is conceivable that the real author 'installs' a perspective, like setting up a pair of binoculars. Through these binoculars, the individual reader has to look in order to be able to perceive the story at all: it is his entrance to the story. One lens of these 'binoculars' prescribes a perspective, its field of vision, the distance and the angle onto the story. However, it does not determine the individual interpretation nor does it inform about the real author's intention of the story. This perspective does not depend on the existence of an 'implied author' whose existence is disputed anyway. Moreover, the term 'im­plied author' often implies a real person with an objective perspective, which is certainly not the case. As this perspective is story-ex­ternal but nevertheless text-immanent, I prefer Hof's term Textperspektive – textual perspec­tive – which already implies subjectivity and a selection out of a multitude of subjectivities.

Apart from the textual perspective, the author creates another perspective which the recipient is supposed to enter – the perspective of the 'implied reader' ("impliziter Leser")^[45] or the recipient's role (Leserrolle; Hof, 1984: 74).^[46] Thus the author creates a perspective for the reader which im­plies a preset attitude towards the protagonists or narrators of the narration – this implied at­titude constitutes the second lens of the pair of binoculars. It is the reader's only emotional entrance to the narration. This will be illustrated by the following example of the intended dissociation from a 'mad monologist'.^[47] The recipient's role determines – with the help of various textual signals^[48] – that the reader has to dissociate himself from this protagonist even though he is forced to take in the events in the atmosphere of an intimate conversation. So the textual perspective suggests an atmosphere of intimacy while the recipient's role demands dissociation. The resulting conflict is quite intended. But how does the reader take up his role? And who or what is unreliable?

Staying on the story-internal levels, it is worth noting that the 'mad monologist' – inside his subjective reality-model – does not necessarily lie or deceive on purpose. Often, he is even extremely precise in the description of the events. If he leaves out some details which he thinks unworthy of telling or which he has suppressed, he is – viewed from his subjective perspective – still 'honest' and most reliable.

This form of 'mad monologist' is – inside the story – reliable concerning the actual reliability. Even the normative reliability is intact inside the world of the story – actions, judgments, values and norms of the narrator are authentic, logical and consistent – taking the subjective world of the narrator as point of reference.^[49] Only when the reader leaves the text does he compare the world that is offered in the story, its values, norms and premises with his own world, values, etc.^[50] It is here that the discrepancy develops – not between the values of the 'implied author' and the narrator (cf. Nünning, 1998: 17). And this discrepancy can only de­velop once the reality-model of the story-world actually collides with the reality-model of the reader. Since the respective values and norms of the single reader, his reality-model, are unknown, the textual perspective and the recipient's role orientate themselves by the tacit consent of valid values and norms of a given culture in a specific historical situation – Busch calls it a 'compromisable real­ity-model'.^[51] A reader who holds the same values and norms as the 'mad monologist' will not mistrust him, he will not dissociate himself from him and he will not take up the recipient's role. Only the premise of knowledge and acceptance of the compromisable reality-model in the recipient's role creates the ambiguity of the text, the dramatic irony (Nünning, 1998:17), the possibility of 'reading-between-the-lines' (Chatman, 1978: 233) or the unreliability. This compromisable reality-model (c.r.m.) is outside the textual world.^[52] It is the 'point of reference outside the closed world of the text'^[53] which is – according to Hof – necessary in order to have a yardstick for measuring or judging reliability or unreliability. Thus the reader mistrusts the narrator only after leaving the textual world, e.g. when he discovers that the textual perspective is set too close to a character or a narrating authority that is regarded in the compromisable reality-model as someone rather distanced. Inside the textual world, it is not the narrator who is unreliable; instead, it is either the given textual perspective, like in the above case of a 'mad monologist', or the given recipient's role that is not reliable. This will be explained in the following.

The recipient's role implies a basic attitude towards the narrating authorities and the characters of the fabula. Usually this involves the knowledge of the compromisable reality-model and the acceptance of the c.r.m. in the recipient's role. This attitude is usually based on the distance between the c.r.m. in the r.r. and the reality-model of the narrating authority or character and it is installed at the beginning^[54] with the help of textual signals. If the distance appears to be rather small, attitudes like identification or sympathy are supported, while an obvious wide distance rather provokes distrust, dissociation, or possibly even pity. The reader has to take up this given role because he enters the story without any prior knowledge. If he discovers in the course of the narration that the given perspective (the attitude towards the protagonists that is expected and adopted) collides with his own (or rather with the c.r.m. in the r.r.), he distrusts the narrator as a consequence. But again, it is not the narrator who is unreliable, instead – in this case – it is the recipient's role. The reader experiences it as a breach of trust that he could not rely on the given perspective which he took on in good faith (compare Coleridge's "willing suspension of disbelief"). Hof calls this distrust concerning the reliability of the narrator a 'game of the author with the reader's expectations' (Hof, 1984: 46).

It appears that a narration is unreliable, when either the textual perspective is unreliable – the reader discovers that the given perspective on the events (textual perspective) does not correspond with the one he would choose for himself – or the recipient's role is unreliable – the reader discovers that the required attitude towards the characters and narrating authorities does not correspond with his own. In addition, the incongruity concerning the distance of these two perspectives towards the narrating authority can provoke doubts and irritations concerning the reliability of the whole text.

Hof and Nünning also assume that the reliability of a narrator is not a text-immanent aspect (Hof, 1984: 54; Nünning, 1998: 17), but that it depends on the recipient.

Im Falle eines unreliable narrator resultiert dramatische Ironie aus einer Diskrepanz zwischen den Wertvorstellungen und Absichten des Erzählers und den Normen und dem Wissensstand des realen (nicht eines impliziten) Lesers. (Nünning, 1998: 17)

Nevertheless, Nünning does not come to the obvious conclusion that unreliability is developed on the story-external level as well. Instead, he refers in the same context again to the lack of reliability of the narrator which the reader can see through with the help of textual signals (1998: 17). In conclusion, he comes to the following definition: "Als unreliable narrators sind solche Erzählinstanzen zu bezeichnen, deren Perspektive im Widerspruch zum Werte- und Normensystem des Gesamttextes stehen." (Ibid.)

Comparable to the problems with Booth's definition, the question arises how to extract the system of values and norms from the text. In fact, it is just as impossible to extract the system of values and norms of the 'implied author' (as the assumed subject of the textual perspective) from the text. However, its orientation towards a tacit consent of values and norms of a given culture, the compromisable reality-model, makes this system far more reliable, even if this consent constantly changes. The change of the system also explains the change of the interpretation of a text in the course of time (cf. Hof, 1984: 54). If values and norms, philosophies of the world, ideas of normality, etc. of a cultural area change, it is possible that the distance between the system of values and norms of a putative outsider in a narration and the one of the c.r.m. can change. If the distance decreases, a text might suddenly show aspects of unintentional comic effect or it might lose all its shock value. All dramatic irony that is based on this distance can only be maintained, if the changes of this compromisable reality-model do not affect the distance between the reality-models of the text-internal authorities and 'implied recipient'. However, this still does not ensure that every real reader will actually experience this discrepancy. This would only be the case if in fact all readers had equal norms and values or reality-models.

In consequence, some theoretical approaches appear to be rather questionable when they compare the system of values and norms of a narrating authority to 'the' general norm or to 'the' common values and relate them to 'the objective truth' without taking the processional character and the subjectivity of values, norms and reality-models into consideration.

Überblickt man die Forschung zum Thema unreliable narration, so ist unübersehbar, dass sie auf einer Reihe von fragwürdigen konzeptuellen Prämissen basiert, die in der Regel implizit, unreflektiert und uneingestanden bleiben. Die zentrale epistemologische Prämisse, die dem Konzept zugrunde liegt, besteht in dem Glauben an die Möglichkeit, dass ein Subjekt dazu fähig ist, etwas objektiv zu erkennen und ein Geschehen wahrheitsgetreu wiederzugeben. Außerdem liegt diesem Konzept die Vorstellung zugrunde, dass es nur eine wahre und vollständige Darstellung eines Geschehens – "an authoritative version of events" (Wall, 1994: 37) – geben kann und dass diese verbindliche Version im Prinzip rekonstruierbar und erzählbar ist. (Nünning, 1998: 20f)

Nünning criticises especially the fact that the assumptions of most critics are tacitly based on a concept of normality based on common sense which is taken for granted.^[55] He argues that there is no generally accepted standard of normality which could function as a yardstick for unbiased and inter-subjective comprehensible judgments.^[56] Nevertheless, the ambiguity in the factual internal truth and the reader's inner dissociation of the values and norms of the narrating authority in unreliable narration still work quite well in general despite the infinite variety of subjective perceptions. In fact, it can be assumed that the reader – despite his own system of values and norms – has knowledge of a general

compromisable reality-model which is presupposed in the perspective of the 'implied reader' and that he accepts it – at least in his recipient's role.

Summary

In a world of multiple subjectivity, only a narrator that is actually unreliable can be called an unreliable narrator. All other forms of unreliability (actual and normative unreliability of a focaliser and normative unreliability of a narrator) merely indicate the subjective perspective of the narrating authority.

On the story-external but still text-immanent textual level, the real reader finds a pair of perspectives that are his only entrance to the story: the textual perspective and the recipient's role (or p.i.r .). The textual perspective establishes structural aspects that determine the factual distance to the story; while the recipient's role decides on the emotional distance to the story. It presupposes an attitude towards characters and narrators that is based on the distance between the reality-model of the narrating authority and the compromisable reality-model presupposed in the recipient's role. These perspectives – taken up in good faith by the real reader at first – create certain expectations concerning, for example, how to evaluate characters or the reliability of the narrating authority. When the reliability of the presupposed expectations are questioned with the help of textual signals, it is usually experienced as the unreliability of the narrator, but it seems that in fact the unreliability lies instead in the perspectives on the textual level.

The point of reference here is the compromisable reality-model that orientates itself by a tacit consent of values and norms of a given culture in a given historical situation. It is presupposed in the recipient's role. A great distance between this consent and the system of values and norms of a character or narrator creates the dramatic irony that determines normative unreliability. However, this consent is constantly in flux throughout the decades of time and differs from culture to culture. Only when the changes do not influence the distance between the compromisable reality-model and the reality-model of the narrating authority/character will the dramatic irony be maintained. Hence, this compromisable reality-model does not correspond to remarks about 'the general norm'.

These explanations about unreliability will be a fundamental aspect in the analysis in chapter 6, since a main feature of McEwan's characters seems to be their outstanding subjective perspective which continually calls their reliability into question.

5. Subjective Perspectives in Narration

In order to provide an instrument to analyse the subjective perspectives in McEwan's narrations, in this chapter a structure to analyse the subjective perspectives in narrations will be developed. The structure will be based on the explanations and terminology of the past three chapters.

[...]

---

^[1] References that refer to a whole sentence (or passage) will be given after the last sentence. However, if the sources merely refer to a phrase or to single words, they will be given directly afterwards.

^[2] See e.g. Slay (1996), Malcolm (2002), Nünning (1989).

^[3] "McEwan […] immediately acquired notoriety as a 'shocking sensationalist' […]." (Lewis, 1991: 622) See also e.g. Slay (1996), Malcolm (2002), and Blackwood (1978).

^[4] See interviews with e.g. Begley (2002), Hamilton (1978), and Fortin/Louvel/Ménégaldo (1995).

^[5] "Da sexuelle Themen ohne Romantik oder Erotik, auch ohne jede pornographische Effekthascherei behandelt werden, ist die Einschätzung M.s als Sensationsautor, der schockieren will, unberechtigt." (Nünning, 1989: 1954)

^[6] "Die Atmosphäre in der Welt seiner Erzählungen ist […] düster und melancholisch." (Ibid., 1954)

^[7] "McEwan has always been taken seriously by critics and scholars, although they have not always liked his books." (Malcolm, 2002: 4) "McEwan's imagination seemed as though it could usefully swabbed out with Dettol. Vomit, excrement, mucus, slime, globs of squashed frog, rancid food and green mould encrusted his pages. Characters were equally putrid. Sickening smells wafted from their various bodily zones. Their behaviour was noxious." (Kemp, 1992: 6)

^[8] "Seine Werke sind zwar formal nicht zur experimentellen Literatur zu zählen, erschließen aber neue Themenbereiche." (Nünning, 1989: 1953) "Von der Skizze "Hin und Her" […] einmal abgesehen, stellt McEwan keine Formexperimente an, sondern ist einer realistischen Stilhaltung verpflichtet." (Oeser, 1994: 3)

^[9] See e.g. Kermode: "[McEwan has] a keen technical interest in […] the point of view." (Kermode, 2001: 1)

^[10] See e.g. Nünning: "Viele seiner Ich-Erzähler sind von einer fixen Idee besessen und erinnern an den Erzähltypus des 'verrückten Monologisten', der in der literarischen Moderne […] einen festen Platz hat." (1989: 1953) See also Malcolm, i.a..

^[11] In the following, the male form will always be used for both genders.

^[12] Compare Chatman: "That it is essential not to confuse author and narrator has become a commonplace of literary theory." (1978: 147)

^[13] Ibid., 151.

^[14] As the concept of the 'implied author' and the 'implied reader' is controversial the term will be set in inverted commas throughout the text.

^[15] Weiß, 1979: 132.

^[16] See Fielitz, 2001: 33f.

^[17] Concerning the different levels, the terminology is not standardised. According to a diagram in A Glossary of Contemporary Literary Theory (2000: "Story and Plot") which is based on Fludernik's narrative levels (1993), it becomes apparent that even the term "story" itself is not uniform for Chatman, Stanzel or Rimmon-Kinnan. Furthermore, the levels of textual communication are not exactly congruent to the narrative levels where differing levels are defined. So Fludernik's terminology can not be adopted here.

^[18] Here Chatman and Genette, e.g. use the term "discourse" and "discours", but respectively imply a different meaning with those terms. In this context, a term was looked for that defines exclusively the level of textual communication on which the narrator/focaliser and the narratee operate. Because of the lack of such a term it is called here "mediation-level" in accordance to Stanzel's definition "mediation by teller or reflector" (taken from Fludernik, 1993: 62).

^[19] It is debatable whether an omnipotent extradiegetic narrator is part of the story. But the extradiegetic narrator will usually be directly substituted by the 'implied author' as he is experienced as such, and then he is outside the story. Nevertheless, limited third-person-narrators and certainly all I-narrators and focalisers are experienced as being a part of the story.

^[20] Only an acting I-narrator who is narrating while he is acting is inherent in the fabula. All other narrators do not interfere but only reflect upon the fabula – which is in the past. Even an I-narrator gives only "its vision of a fabula in which it participated earlier as an actor" (Bal, 1985: 126), so he reflects upon his earlier self.

^[21] In the Weiß illustration, the level of the 'implied author' and 'implied reader' is not classified. Also, Fludernik's survey of narrative levels does not explicitly take this level into account. It seems to be implied somewhere between "the text on page" and "narration as enunciation" (1993: 62).

^[22] Although the 'implied author' appears to be responsible for the story, story and 'implied author' are strictly separated and also perceived as such, while an I-narrator or a focaliser is an invented part of the story and may even be interwoven with the fabula.

^[23] "Perspectiva […] war ein mittellateinisches Substantiv, das die Wissenschaft von der Sehkraft, d.h. die physikalischen und physiologischen Gesetzmäßigkeiten des Sehens, und die Lehre vom sichtbaren Licht bezeichnete. Die spätere Bedeutungsvielfalt ist v.a. darauf zurückzuführen, dass visuelle Wahrnehmung als ein Medium der Erkenntnis betrachtet wurde. Die traditionelle Korrelation von visuell-optischen und kognitiven Aspekten […] stellt ebenso eine Grundkonstante des Perspektivenbegriffs dar wie dessen enge Verbindung mit einem erkenntnistheoretischen Dualismus [the "split between subject and object", Guillén, 1971: 292f]." (A. Nünning, 2000: 7f).

^[24] See e.g. Nünning: "Zur Kritik an diesem Begriff [Erzählperspektive] und an point of view -Modellen vgl. Lintvlet (1981), Basic (1983), Cohan (1986), Kablitz (1988) und A. Nünning (1990)." (Nünning, 2000: 11) See also Fielitz, 2001: 43.

^[25] Genette (1980). See Fielitz, 2001: 43f.

^[26] In: Susan Onega, José Angel García Landa (eds.), (1996: 115-128). All quotations out of this work within this chapter will be indicated by the number of the page in brackets only.

^[27] In order not to mix Bal's and Nünning's terminology, which would necessarily lead to confusion about the term "focaliser" the expression narrating authority shall subsequently describe Bal's category of "focalizor". Though an external focaliser is theoretically possible it is usually experienced and classified as an extradiegetic narrator. Subsequently the term focaliser here will refer to an internal or intradiegetic focaliser exclusively. Compare Nünning: "Da extradiegetische Erzähler in der Regel in Personalunion zugleich als Fokalisierungsinstanz fungieren, fallen die Typen 1a [extradiegetisch multiperspektivisch erzählte Texte] und 2a [extradiegetisch multiperspektivisch fokalisierte Texte] in der Praxis meist zusammen." (Nünning, 2000: 44).

^[28] For the reason behind the change in terminology, see above.

^[29] "Character-bound focalization (CF) can vary, can shift from one character to another. In such cases, we may be given a good picture of the origins of a conflict. We are shown how differently the various characters view the same facts. This technique can result in neutrality towards all the characters. Nevertheless, there usually is never a doubt in our minds which character should receive most attention and sympathy. On the grounds of distribution, for instance the fact that a character focalizes the first and/or the last chapter, we label it the hero(ine) of the book." (Bal, 119)

^[30] See: Multiperspektivisches Erzählen. Zur Theorie und Geschichte der Perspektivenstruktur im englischen Roman des 18. bis 20. Jahrhunderts. (Nünning, 2000)

^[31] "Die Analyse der Perspektivenstruktur eines narrativen Textes besteht demzu­folge aus zwei Schritten: der Rekonstruktion der Einzelperspektiven und ihrer Relationierung zueinander." (Ibid., 51)

^[32] The Oxford Advanced Learner's Dictionary of Current English (1988) defines 'subjective' as follows: "1 (of ideas, feelings, etc) existing in the mind, not produced by things outside the mind; not objective […] 2 (of art and artists, writing, etc) giving the personal or individual point of view or feeling (opp to realistic art, writing, etc). 3 (gram) of the subject. […]" (861)

^[33] All quotations from this work within this chapter will be indicated by the number of the page in brackets only.

^[34] "zahlloser subjektiver Wirklichkeitsentwürfe" (GU, 4)

^[35] Schmidt explains the theory as follows: "Wahrnehmung und Erkennen bilden [...] nicht eine objektive Wirklichkeit ab, sondern sie errechnen bzw. konstruieren etwas, das wir erkennend als Wirklichkeit akzeptieren und dementsprechend wir uns verhalten und handeln." (1985: 121).

^[36] Gessner-Utsch regards the 'stream-of-conscious novel' as one kind of subjective novels.

^[37] "In einem Roman dieses Typs [objektiver Roman] ist dem Protagonisten/der Protagonistin also nicht die Gestal­tung einer eigenen sinnvollen Wirklichkeit, sondern die Aufdeckung eines von vorneherein gegebenen Sinnes aufgegeben. [...] weil dem Protagonisten/der Protagonistin die Einpassung in eine vorhandene Wirklich­keit aufgegeben ist, stellt sich seine/ihre Subjektivität als Resultat eines endlichen Prozesses dar. Wenn die 'objektive Wirklichkeit' aufgedeckt und der Platz in ihr eingenommen ist, ist 'die Auseinandersetzung des Ichs mit der Wirklichkeit' abgeschlossen [...]." (GU, 8)

^[38] "M.E. wurzelt dieser Romantyp [der 'subjektive Roman'] in einer geschichtlichen Situation, in der die Überzeugung vom Vorhandensein einer 'objektiven Wirklichkeit' [...] brüchig wird und sich somit der Bedarf nach einem Roman ergibt, der den Veränderungen in der Wirklichkeitskonzeption Rechnung trägt, indem er dem Protagonisten/der Protagonistin nicht mehr die – nunmehr unerfüllbar gewordene – Aufgabe des Wirklichkeitsentdeckers, sondern – zumindest teilweise – die des Wirklichkeitserstellers zuschreibt." (GU, 9)

^[39] "Etwas verallgemeinernd und vereinfachend könnte man sagen, dass alle Ich-Erzähler per definitionem parteiliche und somit mehr oder weniger unverlässliche Erzähler sind." (Stanzel, 1989: 200)

^[40] "Perspektivierung in diesem Sinne bedeutet somit Subjektivierung." (Stanzel, 1989: 166)

^[41] The extent of the restriction probably determines the reader's perception of the narrator as belonging to the story or not. If he is perceived as a story-external figure, he is directly substituted by the textual perspective.

^[42] See e. g. Iser (1972), Chatman (1978), Stanzel (1989), Lanser (1981), Rimmon-Kenan (1983), Hof (1984), Nünning (1993a/1998), Fludernik (1996/1998), Busch (1998), i.a..

^[43] The focus of this paper lays not so much on single analytical aspects of unreliable narration but on its embedding in the context of the analysis of subjective perspectives. Moreover, Nünning uses the 'values and norms of the text' ("Werte- und Normensystem des Gesamttextes"; 1998: 17) as point of reference. This concept cannot be used here. (The reasons are explained in detail later in this chapter.)

^[44] Nünning distinguishes actual and normative unreliability ("faktische und normative Unglaubwürdigkeit"; 1998: 12). Actual unreliabiliy regards the text-internal facts, the substance of the presented facts. Normative unreliability regards the values and norms of the narrator or focaliser ("moralisch-ethischen Wertmaßstäbe und Urteile des Erzählers"; ibid.).

^[45] See Iser: "Der implizite Leser meint den im Text vorgezeichneten Aktcharakter des Lesens und nicht eine Typologie möglicher Leser." (1972: 8f)

^[46] Nünning claims that Iser's attempt to use the 'implied reader' as point of reference is as useless as using the 'implied author' as point of reference. (1998: 14) Still, the concept of a recipient's role or of the perspective of an 'implied reader' is, in my opinion, quite useful in this context. Although both terms define the same perspective, it is helpful to hold on to both terms, since they emphasise different aspects of the same perspective respectively.

^[47] In a philosophy of multiple subjective realities, the term 'verrückter Monologist' is untenable, because there are no objective ideas of 'normality' and 'craziness'. Therefore this expression is always set in inverted commas. However, it is used in its known sense to save the reader sidetracking examples and explanations.

^[48] In this thesis, the term textual signals is always used according to Gaby Allrath's exposition "'But why will you say that I am mad?' Textuelle Signale für die Ermittlung von unreliable narration" (Allrath, 1998: 59-79) in order to omit sidetracking examples. However, it should be mentioned that her statements about textual signals cannot be integrated without difficulties to this thesis. Continually, Allrath claims that the mentioned aspects lead the reader to recognise the subjectivity of the narrator, which seems to expose his unreliability. She does not take into account that every single perspective is finally subjective. Under the presumption of the concept of 'moderate constructivism', the so-called objectivity is nothing else but a subjective perspective that claims for itself to be a standard. Does subjectivity automatically signalise unreliability? Although every character is subjective in the end, we do not mistrust every character just because of his subjectivity. Possibly the extent of subjectivity is decisive – but is it possible to measure subjectivity? It seems that it is not the existence of subjectivity that is decisive but – as will be shown – only the distance between the examined subjective perspective and the compromisable reality-model in the ontological reality of the reader. However, the textual signals as pointed out by Allrath nevertheless sow doubts concerning the reliability of the narrator and may indicate the unreliability of perspectives despite the universality of subjectivity.

^[49] Even Nünning considers it possible to dispute the unreliability of some narrators: "Die von ihnen erzählten Geschichten mögen keine 'objektiven' Darstellungen der entsprechenden Ereignisse sein, aber sie schildern denkbar zuverlässig die Illusionen, Selbsttäuschungen und Harmonisierungsversuche der Erzähler." (Nünning, 1998: 19) Here Nünning seems to indirectly assume that the existence of a strongly limited subjective perspective is already sufficient to prove some kind of unreliability.

^[50] Nünning rightly criticises that the scientific literature usually only names 'values and norms' in the text-external frames of references (Culler, 1975). His list is much more precise: "- allgemeines Weltwissen, - das jeweilige historische Wirklichkeitsmodell [...], - explizite oder implizite Persönlichkeitstheorien sowie gesellschaftlich anerkannte Vorstellungen von psychologischer Normalität oder Kohärenz, - moralische und ethische Maßstäbe, die in ihrer Gesamtheit das in einer Gesellschaft vorherrschende Werte- und Normensystem konstituieren, - das individuelle Werte- und Normensystem, die Perspektive bzw. das Voraussetzungssystem des Rezipienten." (Nü, 1998: 29f) In the scope of this essay, it will not be possible to always list all these aspects so I will usually use the terms "values and norms" or "reality-model", which shall imply all other aspects.

^[51] See Busch: "D.h. verschiedene Weltsichten sind grundsätzlich gleichberechtigt, die Anerkennung eines kompromissfähigen Wirklichkeitsmodells ist zur gegenseitigen Verständigung jedoch notwendig." (1998: 45)

^[52] This consent is not fixed, but it is a convention which underlies a constant change and which, furthermore, depends on the cultural tradition. (Cf. Hof, 1984: 54.) Therefore, an author can never be sure, if the intended irony or ambiguity will be attained. Only characters that lie so far outside the 'norm' that the possibility of identification is to a large degree excluded will succeed in reaching most of the readers with the intended effect.

^[53] "Bezugspunkt außerhalb der geschlossenen Welt des Textes" (Hof, 1984: 24).

^[54] Cf. Allrath: "[…] dabei [kommt] den am Anfang des Erzähltextes vergebenen Informationen sowie den metatextuellen Vorinformationen in den Nebentexten eine zentrale Stellung zu, da diese die Basis für die Zuord­nung von Referenzrahmen und damit für das (zumindest vorläufige) Verständnis des Textes bilden." (1998: 75)

^[55] "Anstatt jedoch das damit angesprochene Problem der jeweils relevanten Bezugsrahmen explizit zu erörtern, gehen die meisten Kritiker stillschweigend von einem als selbstverständlich vorausgesetzten Normalitätsbegriff aus, der im gesunden Menschenverstand gründet." (Nünning, 1998: 21)

^[56] "Problematisch sind sie [solche normativen Kategorien] vielmehr auch deshalb, weil es keinen allgemein akzeptierten Standard von 'Normalität' gibt, der als Maßstab für unvoreingenommene und intersubjektiv nachvollziehbare Urteile fungieren könnte." (Ibid., 22)