Unreliable narration in Kazuo Ishiguro's "The Remains of the Day". A theoretical approach

Table of contents

1. Introduction

2. Unreliable Narrator and Narrative Situation in General Terms

3. Detecting Unreliable Narration in The Remains of the Day

4. Restated thesis statement / Summary of main points

Bibliography

1. Introduction

In this paper I am going to explore the phenomenon Unreliable Narration and I am going to show that the narrator in Kazuo Ishiguro’s novel The Remains of the Day, which was originally published in 1989, can be read as an example of this phenomenon. This topic seems to be particularly relevant in view of the following: Ever since Wayne C. Booth has introduced the concept of the unreliable narrator in 1961, it became an integral part of literary studies (cf. Zerweck 151). Accordingly, Ansgar F. Nnning considers the aforementioned concept “[…] to be among the basic, important, and indispensable categories of textual analysis” (Nnning 83; Nnning 89). Besides, Kazuo Ishiguro’s novel received the Booker Prize for Fiction¹ which contributed to the book’s significant influence on British literature and its history.

My paper is subdivided into two main parts. These are “Unreliable Narrator and Narrative Situation (in General Terms)” and “Detecting Unreliable Narration in The Remains of the Day.” Firstly, I am going to define the crucial term unreliable narrator. Then, I am going to briefly present / outline two long accepted theoretical approaches. Namely, Frany Stanzel­’s concept of Narrative Situation and Grard Genette’s Structuralist Theory. I do this in order to familiarize the reader with the terms narrative situation, voice and focalization. Indeed, being aware of these narratological concepts is of vital importance for the reader. That is, it enables the latter to better keep track of how to analyze the Narrative Structure in The Remains of the Day properly. The next step is applying these concepts to, the concrete example of, the narrator in The Remains of the Day. Secondly, I will list and explain several respective clues which backup my thesis of Unreliable Narration. More precisley, I will point out passages of the here discussed novel that specifically meet the criteria of inconsistency, incoherence and lack of correspondence. The approach to my subject will be a theory and analytical method based, close reading of The Remains of the Day. I am going to argue that, since the narrator of The Remains of the Day meets the criteria of inconsistency, incoherence and lack of correspondence, he should be considered unreliable.

2. Unreliable Narrator and Narrative Situation in General Terms

2.1 Definition of unreliable narration / What is an unreliable narrator?

The following paragraph aims to provide a preferably accurate definition of Unreliable Narration. My proposal is inspired by the cognitive-narratological approach of the German scholar Ansgar Nnning and is based on Michael Meyer’s strategical account of criteria for unreliable narration. More precisely, my suggested definition is, in fact, a synthesis of these two, latter models. By pointing out the terminological, theoretical and methodical problems that arise from/ stem from/derive from the standard definition(s) and theories of U.N., I will justify my choice of how to deal with the respective subject. That is, I am going to make clear that my approach does, in no way, exhibit those blatant shortcomings, and is thus particularly adequate for the issue of this paper.

“There is indeed a peculiar discrepancy between the importance generally attributed to the question of reliability in narrative and the unresolved [terminological, theoretical and methodical] issues surrounding the concept” (Nnning 89). The esteemed Tamar Yacobi did phrase this discrepancy in a concise manner: “There can be little doubt about the importance of the problem of reliability in narrative and in literature as a whole […]. [But] the problem is (predictably) as complex and (unfortunately) as ill-defined as it is important” (Yacobi 113). According to A.Nnning, the abovementioned difficulties are due to Wayne C. Booth’s rather vague explication of Unreliable Narration (cf. Nnning 3). The latter did define the phenomenon as follows: “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’s² norms), unreliable when he does not” (Booth 158-159). Additionally, Booth distinguishes reliable from unreliable narrators by considering “the degree and kind of distance” (155) that exists between the mediator³ and the implied author. Though Booth freely admits to the inadequacy of his proposed terms⁴, almost all narratologists adopted his terminology by providing virtually verbatim versions. Accordingly, the defintions provided in glossaries of literary terms and narratological studies agree on the following: A narrators reliability is to be hold suspect if his moral norms are at odds with those of the implied author. In this context, G. Price’s explication may be of assistance to illustrate the communis opinio on the subject U.N.:

A narrator whose norms and behavior are not in accordance with the implied author’s norms; a narrator whose values (tastes, judgements, moral sense) diverge from those of the implied author’s; a narrator the reliability of whose account is undermined by various features of that account (Prince 101).

However, the concencus regarding Unreliable Narration does not conceal the fact that conventional definitions and theories are surrounded by a number of terminological, theoretical and methodical issues. First, the term unreliable narrator does lack precision. That is because most definitions leave unclear what is exactly meant by unreliability. Therefore, it is controversial whether the shortcomings that such a mediator exhibits are rather moral or epistemological. Although Booth response to this unresolved matter is anything but definite, his explanation indicates that he conceives unreliability primarily as a moral shortcoming. More precisely, from Booth’s view, the ethical deficit comprises the discrepancy between the implicit author’s norms and those expressed by the mediator. Consequently, both Booth’s notion of reliability and that of those who/of the ones who draw on it, are to be branded as fraud(ulent labeling). After all, what seems to be at stake is not the narration’s veracity, namely its factual credibility, but rather norms and values (cf. Nnning 11).

Second, not only does a clear lack of terminological precision prevail, instead there are also a number of serious, theoretical problems. For example, the key issue, in this respect, is that the prevailing definitions provide only one guideline for deciding whether a narrator is to be judged unreliable. That is, the terminologically ill-defined and theoretically controversial concept of the implied author (cf. Nnning 13,16). Despite these obvious deficiencies, almost all theorists who deal with unreliability stick to this concept with stubborn insistence: “[…] most theorists […] who have written on the unreliable narrator take the implied author both for granted and as the only standard according to which unreliability can be determined” (Nnning 86). The tenacity with which narratologists hold on to the implicit author, as the only yardstick for ascertaining a mediator’s unreliability, proves the subsequent assumption true. Namely, the concept serves as a catchall for all the unresolved aspects of narrative theory. The latter presumption is confirmed by the following utterance: “[…] [T]he implied author is a remainder category, a kind of passepartout that serves to clear away all the problematic remainders of a theory” (Bal 209). Nevertheless, the introduction of the i.a. has certainly not managed to resolve the issues of defining unreliable mediation. Accordingly, the fact that the crucial question, namely how the implied writer’s norms are to be established at all⁵, is unsettled, emphasizes the susequent. That is, referring to the implicit author doesn’t help to untangle the concept of unreliable mediation (cf. Nnnng 14-15).

Besides, the notion of the implicit writer is also from a theoretical perspective quite problematic since/as it results in the fallacy that it is (,and thus the untrustworthy⁶ mediator likewise,) a solely textual phenomenon. However, many explanations⁷ define the implied author as a reader-created product which is based on the whole text. Moreover, the myth that the unreliable mediator and its base resp. refernce figure (i.e., the implied writer) are purely textual phenomena is proven wrong by the following empirical fact. That is, professional literary critics usually differ significantly in their unreliability assesments, which concern one and the same text (cf. Nnning 16). Therefore, if […] Chatman […] writes that ‘we might better speak of the inferred than of the implied author,’ he implicitily concedes that one is dealing with something that has to be worked out by the reader. […] [H]owever, the notion of an inferred author is quite different from an implied author[.]. [T]he latter is a creation by the author, which the reader may or may not gauge in practice, whereas an inferred author is a creation by the reader, which may or may not correspond to the implied author projected by the flesh-and-blood person who wrote the text […] (Nnning 92).

Indeed, this reveals the difficulty which results from linking the implicit writer to the unreliable mediator. More precisely, narratology deprives itself of the possibility of developing more differentiated and adequate models for explaining the functioning, semantic content and pragmatic effects of unreliability (cf. Nnning 16). That is why, I take my cue from the cognitive narratologist A.Nnning who “argue[s] for the abandonment of the implied author and for a radical reconceptualization of the unreliable narrator” (Nnning 89-90; cf. 87;17).

Third, the established theories of the phenomenon are also methodologically unsatisfactory as they either leave open how the impression of flawed reliability is actually created, within the reading process, or they offer merely highly metaphorical explanations of it. Instead of addressing the textual signals/clues that do evoke this feeling, the majority of narratologists employ vague descriptions or tropes resp. verbal images. Thus, S. Chatman settles for the opaque allusion that recipients do question the mediator’s veracity by ‘reading out between the lines’ : “We conclude by ‘reading out,’ between the lines, that the events and existents could not have been ‘like that,’ and so we hold the narrator suspect” (Chatman 233). Not only does Chatman’s repeated use of inverted comas unveil the absence of concrete criteria for determining a narrator’s untrustworthiness, instead the impression of questionable methodology is also intensified by the metaphorical language. The second prevailing metaphor which theorists apply in order to explain how the reader detects/apprehends a narrator’s unreliability is that something is happening ‘behind the narrator’s back’ (cf. Nnning 90; 15). More specifically, the implied author sends, so to speak, a secret, ironic message to its equivalent (i.e. the implied reader). It is obvious that this image also doesn’t provide much of an enlightment as to how untrustworthiness may in fact be discerned. That is because, on the one hand, the structural phenomena (i.e., implied author and implied reader) are anthropomorphized without contributing to a better understanding of the narrative mediator’s specific characteristics. On the other hand, this metaphor, likewise the other, boils down to (nothing other than) elevating the individual’s intuition to the solely clue for a mediator’s unreliability. And that, without even answering the question as to how the communication behind the narrator’s back may in fact be performed. All in all, do such metaphors, though vivid, not manage to conceal the fact that they provide anything but a clearly structural method for analyzing narrational unreliability. That is, “[…] they do not tell the critic anything about the range of [nameable] textual and contextual clues to a narrator’s unreliability” (Nnning 90).

The matter of shedding light on how unreliable mediation ‘functions’ requires, instead of relying on vague metaphorical descriptions, rather a systematic survey of the textual signals that are responsible for undermining a narrator’s credibility⁸. A. Nnning itemizes the following textual signals⁹ for establishing an untrustworthy narrator:

- Explicit contradictions of the mediator as well as other internal discrepancies within the narrational discourse¹⁰.
- Discrepancies between the narrator’s actions and utterances.
- Other conflicts bewteen story and discourse.
- Discrepancies between the narrator’s rendering of events and his / her explanations, evaluations, and interpretations of them¹¹.
- “pragmatic indications […] such as frequent occurrences of speaker-[centered] and addressee-[resp. reader-] oriented expressions” (Nnning 97).
- The narrator’s deliberate attempts of guiding a reader’s literary reception.
- The storyteller more or less self-consciously admits his own unreliability. This can be done, for example, “by alluding to a [..] faulty memory or limited knowledge” (Nnning 98).

Such textual signals can encourage recipients to call into question the mediator’s credibility. This is especially the case if certain signs do pile up and various indicators occur together. Nevertheless, it is important to note that textual discrepancies (i.e., textual signals) alone are not sufficient for clearing up the question whether resp. to what extent a narrational authority’s credence appears doubtful: “[…] the identification of an unreliable narrator depends upon both textual and extratextual information located in the reader’s mind […]” (Nnning 99). Thus, within the frame of a cognitive theory, textual features are merely granted a signaling effect. This is because, eventually, it depends upon the individual recipient how textual inconsistencies or discrepancies are (re)solved- within the reception process. Besides, the option of eliminating any significant discrepancies by attributing them to a narrator’s lack of reliability resp. credibility, is only one out of several possibilities (cf. Nnning 28-29).

Apart from these textual signals, the matter of extratextual reference to models of reality or society’s notion of reality, plays a decisive role in the reader’s skeptical attitude towards a narrator’s trustworthiness. However, instead of making sweeping statements, it is rather appropriate to distinguish the various external frames of reference by resorting to the models of frame theory¹². According to Nnning, the contextual frameworks can be divided into two major groups. The, for this paper relevant, first group includes those frames of reference that refer to the ‘real’ world (i.e., real world frames). That is, they refer to the empirical reality resp. to the socially prevailing concept of reality. These frameworks have the following in common: They are either based on the text’s reference to reality or on the assumption that the text (i.e., fictional model of reality) is (at least) compatible with the dominant view of the world at the time of writing ¹³ . Accordingly, deviations¹⁴ from what is commonly-received as reality may indicate that the mediator is unreliable.

In order to analyze the phenomenon in a more differentiated and comprehensible manner, one should distinguish (, just as with the textual signals,) between those different contextual frames. That is why, the most important extratextual frameworks are mentioned below:

- General world-knowledge (i.e., knowledge about the text’s historical background).
- Conventional notions of psychic normality resp. normal psychological behavior
- Moral and ethical standards that a given society holds to be constitutive of its norms and values

While referring to moral norms is of crucial importance for the question of untrustworthy mediation, the other contextual frameworks remain widely ignored by the research literature¹⁵. However, for an adequate and methodically comprehensible theory, it is essential that all categories, which are used as a yardstick for gauging a mediator’s unreliability, are mentioned explicitly. After all, the abovementioned textual signals frequently gain their meaning by merely being linked to the recipient’s contextual frames. This is true for many cases, such as for contradictive textual information. By drawing on historical background knowledge or moral standards, the latter may be interpreted as clues to a storyteller’s lack of credibility (cf. Nnning 29-30). Consequently, the bottom line is that unreliable mediation is not to be regarded as a purely textual phenomenon. Instead, it rather depends upon both textual- and extratextual information, which are, thus, to be considered equally.

A. Nnning’s systematic account of clues to u.n. does coincide with Michael Meyer’s strategical account of criteria for u.n. . Since the latter complements Nnning’s approach by the more tangible criteria of inconsistency, incohrence and lack of correspondence, it makes sense to synthesize these approaches. Accordingly, Nnnings textual signals are to be subsumed under the criteria of inconsistency and incoherence¹⁶. The contextual clues (,which are identified by drawing on the abovementioned frames of reference,) are to be subsumed under the criterion lack of correspondence. Thus, I define the said phenomenon as follows: A narrator should be considered unreliable if he meets the criteria of inconsistency, incoherence (i.e., Nnning’s textual signals) and lack of correspondence (i.e., Nnning’s contextual clues).

[...]

¹ “[…] a literary prize awarded each year for the best original novel written in the English language and published in the United Kingdom. The winner of the Booker Prize is generally assured international renown and success […]” (https://en.m.wikipedia.org/wiki/Booker_Prize,).

² “A term coined by Wayne C. Booth in The Rhetoric of Fiction (1961) to designate that source of a work’s design and meaning which is inferred by readers from the text, and imagined as a personality standing behind the work. As an imaginary entity, it is to be distinguished clearly from the real author […], [and] […] is also to be distinguished from the narrator […] (https://www.oxfordreference.com/view/10.1093/oi/authority.20110803095959435,).

³ For reasons of stylistic variation, the term ‘mediator’ is used interchangeably with ‘narrator.’

⁴ This assertion is revealed by the subsequent statement: “Our terminology for this kind of distance in narrators is almost hopelessly inadequate” (Booth 158).

⁵ In virtue if this unsettled issue, one can entirely resp. unreservedly agree with S. Rimmon-Kenan’s rsum “[…] that the values (or ‘norms’) of the implied author are notoriously difficult to arrive at” (Rimmon-Kenan 101).

⁶ For reasons of stylistic variation, the term ‘untrustworthy’ is used interchangeably with ‘unreliable.’

⁷ See, for instance, the explanation provided by S. Rimmon-Kenan who determines the implied author ”[…] as a construct inferred and assembled by the reader from all the components of the text” (87).

⁸ Accordingly, several recent scholarly works on the problem do emphasize that the identification of an unreliable narrative agent does not depend solely on the individual recipient’s intuition at all. Furthermore, these studies demonstrate that there is, indeed, a whole range of definable “ ‘Kriterien für die Ermittlung der Unglaubwürdigkeit eines Erzählers‘ […], ‘die es erlauben, uns unabhängig von individuellen Intuitionen über die Wahrheit oder Falschheit von Äußerungen zu verständigen‘ “ (Nünning 16).

⁹ Nünning differentiates systematically between contextual and textual clues resp. signals. He defines the latter as follows: (textual signals) are all formal and thematic features of a literary piece that suggest, without drawing on extratextual aspects, to question a narrational agent’s credibility (cf. Nnning 27).

¹⁰ Literary scholars distinguish between two levels of a narrative. That is, the “discourse (Diskurs, Erz hlweise), the form of how something is told by whom to whom, and story , the content of what is told” (Meyer 66).

¹¹ “In such cases a narrator’s commentary ‘is at odds with the evidence presented in the scene he comments upon.’ The critic can establish such a difference by analyzing those utterances in which the narrator’s subjective bias is particularly apparent and comparing the world-view these imply with the story itself” (Nnning 96).

¹² “Cognitive narratology is based on the premises of a frame theory originally derived from research on artificial intelligence[…]. In the context of frame theory, the reading process can be conceptualized as the construction and projection of a system of hypotheses and schemata- or frames- with the help of which the potential meaning of textual signals is worked out by the reader” (Zerweck 153).

¹³ Highlighted in italics to indicate that these content-related refinements are made by drawing on M. Meyer’s criterion lack of correspondence (cf. Meyer 81).

¹⁴ - in shape of descriptions, comments or interpretations-

¹⁵ “Kathleen Wall is the only critic to date who has at least briefly discussed the importance of contextual information for recognizing an unreliable narrator: ‘Part of the way in which we arrive at suspicions that the narrator is unreliable’, Wall argues, is through ‘using what we know about human psychology and history to evaluate the probable accuracy of […] a narrator’s assertions’“ (Nnning 99).

¹⁶ Except for the last three textual signals. These are addressed separately in 3.1 .