Abstract
As one of the master stylists of our time, John Banville has honed his own unique style of writing. The typical Banville novel is a first-person confessional narrative of an aging male character troubled by his painful memories of failure and loss. In a struggle to cope with their traumatic life experiences, Banville’s protagonists attempt to find answers to haunting existential questions and rediscover their identities in the face of emotional fragmentation. This sense of dislocation and displacement thus emerges as a major theme of Banville’s fiction and his works generally revolve around the internal conflicts of a ‘divided self’. The purpose of this study is to investigate how the language of the novels reflects the inner split of the characters and what linguistic mechanisms Banville exploits to create the ‘divided self’ effect. This article examines a particular linguistic structure used as a pervasive narrative feature: sensory modality. I will more specifically explore sensory modality patterns with co-referential subject and object pronouns (referred here as ‘special effects’) analyzing them in the light of Systemic Functional Grammar as mental transitivity processes and will demonstrate how they constitute a powerful stylistic tool for constructing the image of the divided personality and for conveying self-disunity in retrospective novels.
Keywords
1. Introduction
The ‘divided self’ in Banville’s novels is generally acknowledged to occupy a central place in the writer’s prose but since the main approaches to Banville appear to be grounded in Postmodern theory, psychoanalysis and philosophy, the problem of the ‘divided self’ has so far been treated from one of these theoretical frameworks and has hardly been studied from a stylistic and narratological standpoint. While previous investigations of the topic such as O’Connell’s Winnicottian reading of Banville (2013), D’hoker’s exploration of different ‘visions of alterity’ (2004), or Ghassemi’s research of Banville through the prism of Lacan (2016) have shown many aspects of splitting, their approach to the ‘divided self’ is rather based on an intuitive interpretation and not on linguistic evidence. Yet, a linguistic exploration of a literary work is extremely beneficial as it opens new perspectives for interpretation and contributes to our understanding of a text. For this reason, I will attempt to analyze how the theme of splitting is encoded in the language of the novels and, more particularly, in a set of consistently-applied sensory modality patterns.
As narratologists have shown, a first-person text is basically divided between the consciousness of the narrating self and the experiencing self. Briefly defined, the narrating self is the mature present self who narrates the story and the experiencing self refers to the past self whose experiential point of view can be adopted in Free Indirect Style at certain points in the narrative. Thus, while third-person narration presupposes the existence of character and narrator as two separate entities, the I in a first-person novel fulfils both roles by default. This fact has been duly acknowledged in narrative literature and these two functions of the first-person have received various designations: narrating I – experiencing I (Spitzer, 1922); narrating self – narrated self (Schmid, 2008); teller – reflector (Stanzel, 1981); narrator – focalizer (Rimmon-Kenan, 1983); speaker – experiencer (Fleischmann, 1990), representing consciousness – represented consciousness (Chafe, 1994); remembering self – remembered self (Adamson, 2001), monologist – hero (Kakavoulia, 1992).
From a cognitive-stylistic perspective, these two selves are often explored in terms of the different properties that co-referential pronouns may denote. Emmott (2002), applying cognitive studies of selfhood and conceptual metaphor in a more text-oriented framework, explores this split of the I in narrative texts and utilizes the term enactor to separate between these two conceptual entities. Narrative enactors present different versions of a character in space and time which superimpose on one another in a story-world creating a complex and multi-layered image of a protagonist. To separate between these spatio-temporal manifestations of a given protagonist Emmott proposes the distinction present enactor for the narrator-I and past enactor for the character-I. An example of my choice illustrating this split can be seen in the following extract from a retrospective novel:
Even though the referring expressions in these sentences agree in grammatical person and number, they denote different spatio-temporal manifestations of the protagonist – a current self recollecting an image of his/her past self. The remembering self, underlined with a single line, is exemplified by the two ‘I’-s in subject position above, whereas the remembered self, marked by double underlining, is represented by the reflexive ‘myself’ and the object pronouns ‘me’. Therefore, the grammatically same pronoun denotes different narrative enactors. Although the self-referential I depicts one and the same individual, it splits it into two chronologically distinct selves. Every I (I, myself, my, me) has its own temporal dimension and mental realization. The I in first position acting as subject is clearly the mental representation of the adult self narrating the story, whereas the I (myself, me) acting as object refers to the mental representation of his/her young or childhood self. Emmott (2002) calls this overlapping of character-in-present and character-in-flashback the ‘split self’ phenomenon.
She also proposes a categorization of how such splits are realized in the act of narration or by what means the split self is represented in narrative. According to her, the different versions of the self can be put together by means of narrative juxtaposition, double/multiple voice narration or even the ‘special effects’ technique (Emmott, 2002: 170–174). The first two types are most commonly applied to memory and flashback episodes which are the most simple and accessible form of juxtaposing and splitting the self into present and past enactors, while the third technique is the most sophisticated of the three in that it literally makes duplicates out of the characters and allows them to lead a double existence in the same physical location by overstepping the boundaries of natural laws. Such ‘special effects’ Emmott recognizes as being characteristic predominantly of the genres of science fiction and fantasy in which imagination is fully unleashed and time travelling made possible and, secondarily, of imaginary sub-worlds. Emmott illustrates this technique with examples from J. K. Rowling’s Harry Potter and the Prisoner of Azkaban, where Harry and Hermione use a Time-Turner magical device to go back 3 hours in time and observe their earlier selves: This is three hours ago, and we are walking down to Hagrid’s,’ said Hermione. ‘ ‘We need to keep out of sight of Hagrid’s front door, or
In terms of grammar, the ‘special effects’ technique is in fact a type of sensory modality, as defined by Palmer (2001). Palmer’s classification of modality differentiates most basically between propositional and event modality. Propositional modality is further divided into epistemic (related to the speaker’s judgement) and evidential (related to the speaker’s evidence), with epistemic subdivided into speculative, deductive and assumptive and evidential – into sensory and reported modality. Event modality falls into two groups: deontic and dynamic. According to Palmer’s typology, sensory modality is concerned with evidence that comes from the speaker’s senses and covers propositions acquired by visual, auditory and any other perception. Some of the examples that he gives clearly contain visual and auditory markers like the ones presented above – I hear/heard someone running along and I see it coming (Palmer, 2001: 45). Emmott’s ‘special effects’ technique, in which the speaker acts as both subject and object of the proposition, can be viewed then as a private case of sensory modality. And since my study deals specifically with this type of sensory modality which has co-referential subject and object pronouns, I will also adopt Emmott’s term ‘special effects’, as it more precisely describes the particular linguistic phenomenon that I am researching.
In her study, Emmott focuses primarily on fantasy and science fiction as the main source of ‘special effects’ and basically sees this technique as serving various plot functions in time-travel stories. Hence, the double self in them is part of the unreality of the genre. But this duplicating of characters can be a powerful resource in retrospective novels as well and may have significantly larger implications for a narrative. For this reason, I would like to explore other important manifestations of the ‘special effects’ technique, this time in the genre of retrospective fiction, by paying closer attention to its use in terms of narrative voice and perspective. An important point that needs to be taken into account in this respect is the fact that the oppositional terms narrating self – experiencing self and present enactor – past enactor introduced above are not exactly equivalent and it shouldn’t be assumed that there is one-to-one correspondence between them. Emmott adopts a more mechanical separation between present and past enactor, which is mainly a temporal distinction, while the division between narrating and experiencing self in narrative terms is much more subtle. Therefore, the past enactor does not always correspond to the experiencing self as past experience in first-person narration can be represented from the point of view of the narrating self.
Apart from that, the ‘special effects’ examples that Emmott identifies are all framed in the direct speech of the characters and despite the doubling, they stem only from the point of view of the speaking character. What I am going to do is research instances when this technique is weaved straight into the narrative texture and explore in more depth the psychological significance of such instances of doubling when not restricted to genres that are unrealistic by definition. I will aim to prove that this is a technique which has got profound implications for the construction of subjectivity in memory novels. Moreover, I will delve deeper into the semantics of the constructions that create this special effect and will analyze more broadly their function in narrative texts and the stylistic meanings that they convey.
2. ‘Special effects’ in Banville’s fiction
2.1 I see myself…
In this section I will examine in detail how John Banville employs the above-defined ‘special effects’ technique in his first-person retrospective novels to implicitly reinforce the drama of self-division that all his characters go through. I argue that the co-referential pronouns I and myself do not merely divide a protagonist into current self and past self, juxtaposing the adult and the child versions of the protagonist, but can also signal a hero’s displaced sense of self and emotional disconnection from himself and the world. The first type of ‘special effects’ technique that I have discovered in Banville is duplication in the field of vision, as demonstrated in this passage from The Sea: 1. In fact, I do not recall under what circumstances exactly I managed eventually to get inside the Cedars.
Here Max, the narrator, recalls the first time he got inside the Graces’ house, the Cedars. As can be seen, he does not give a coherent narrative account located in linear time and evoking a precise memory, but rather presents random episodic visions of himself. He does not remember the exact circumstances around entering the house, what he recalls is being right inside ‘the sanctum’. So, in addition to the enactor split between character-in-present (I) and character-in-flashback (myself), we can add an extra layer of division, this time between observation and experience. The observational stance of the I-narrator is signalled twice – once by the very verb and once by the pronominal expressions. Firstly and more obviously, the verb ‘see’ directly reveals the visual nature of the protagonist’s memory. He sees the events from his past as if they are unfolding in front of his eyes, with him being a mere spectator. Secondly, the use of the pronoun ‘myself’ can further reinforce the position of the speaker as an eyewitness.
Proof of the latter can be found in Lyons (1982) who sees the addition of ‘myself’ in memory reports as a semantic contrast between the subjective experiencing self and the objective observing self and illustrates this with the following examples: 1. I remember switching off the light. 2. I remember myself switching off the light (Lyons, 1982: 107).
According to Lyons, the difference between these two sentences lies in the subjective agency conditioned by the deictic use of I. The first case introduces an experiential memory and gives a subjective perspective of the event. In the second case we have a descriptive account of an event observed objectively. In Lyons’s own words: (1) exemplifies the normal subjective mode of reference to what is necessarily a personal and incommunicable experience, whereas (2) is not only closer in grammatical structure to (3) I remember you switching off the light, but can also be interpreted, like (3), as reporting the memory of something observed, rather than experienced (Lyons, 1982: 107).
Therefore, if the first sentence foregrounds the subjective experience of the speaker, the addition of myself in the second sentence generates a more distanced perspective on the self. Of course, Lyons’s distinction between subjective experiencing self and objective observing self shouldn’t be blindly equated to the concepts of experiencing self and narrating self in traditional Narratology. Lyons’s argument holds true about language in general but it doesn’t take into account pure narration and free indirect style in fiction. If we map his statements onto the study of narrator’s report and free indirect style, they won’t correspond fully because both statements can be used in both modes depending on the context of the narrative. In other words, the use of the pronoun myself will not always automatically evoke the perspective of the narrating self as this matter is defined by other linguistic features as well. What it seems to achieve, rather, is a pictorial presentation of the memory compared to the more factual presentation presupposed by the first statement.
Similarly, in the afore-mentioned example from The Sea the narrator clearly distances himself from the child’s experience and does not present it as something that he has gone through but rather as something that he is observing at the moment of speaking. So instead of the internal perspective of the experiencing self, we are given the external perspective of the narrator-observer. The psychological difference between the two selves is additionally reinforced by the phrase ‘with the twins watching me go’. Obviously, the character-in-flashback cannot have an external view of what happens behind his back. Rather the protagonist Max watches his past life from outside and has a panoramic vision of himself, i.e. his boy self, turning away from the green gate and the twins watching after him.
But the sensory modality construction ‘I see myself’ seems extremely unusual from a functional and a grammatical point of view. According to Halliday’s Systemic Functional Grammar (Halliday and Matthiessen, 2014), which discriminates between six different transitivity processes, this is a mental clause containing the verb of mental process ‘see’ and the pronoun I is both subject and object or both senser and phenomenon in this case. Though Halliday says that the phenomenon is not restricted and can be a thing, a fact or an act, the option of having the same senser and phenomenon linked syntactically with the verb of perception ‘see’ is an ontological oddity. Since seeing always happens from inside, it doesn’t usually presuppose seeing yourself as an object in the outside world.
Yet, the semantic pattern ‘I see myself’ is a particularly prominent feature in all Banville novels and is the predominant mode of introducing memory episodes by the protagonists. Such reminiscences are typically narrator-mediated because they happen in the narrator’s consciousness which evokes here and now what happened there and then. The narrator’s position in his immediate speaker-now is affirmed by the use of the present tense in the initial introductory phrase ‘I see myself’, as in the following memory episode from The Untouchable: 2. I see
A more profound analysis of this passage, however, brings forth a paradox. We can easily notice that the construction used after the phrasal unit ‘I see myself’ is that of the present participle, which is a non-finite verb form (e.g. sitting, looking, falling, hearing, wandering, crooning). In Hallidayan terms, the phenomenon in mental clauses is not always an object but can also denote an act, macrophenomenon, or a fact, metaphenomenon. While an object is linguistically represented by a noun phrase (e.g. He saw
And this exactly is the paradox in this passage. On the one hand, the introductory expression ‘I see myself’ clearly marks the action as observed, with the verb of perception ‘see’ and the use of ‘myself’ as an object in the outer world. On the other hand, the non-finite verb form in the macrophenomenal clause gives a subjective nature to the experience. In the above example the protagonist reports a scene from his childhood in which he sits in the bay window. Beginning with ‘I see myself as a little boy’, the narrator distinctly distances himself from the events claiming his observational stance only to continue in a highly subjective mode immersed in the consciousness of the child rendering what he saw and heard right there on the scene. The experience of the young self is even more strongly subjectivized with the additional use of participial forms within the macrophenomenal clauses themselves because the further embedded participials the rain falling, poor Freddie wandering and crooning syntactically related to the first participial forms sitting, looking and hearing again signal the perspective of the experiencing self.
What also needs to be considered is the use of tense within this semantic pattern. The previous two passages from Banville both introduce the memory episode in the present tense, I see myself, but in other extracts the pattern is employed in the past tense, I saw myself. The phrase in present tense indubitably positions the narrator in his immediate now splitting the co-referential pronouns into adult self and young self, with the adult self making a current report of how he sees himself as a child. But when does the act of seeing happen in a past tense account when the narrator declares ‘I saw myself’? Consider the following example: 3. They were beyond me, they could not touch me – could they see me, even, or was I now outside the spectrum of their vision? And yet how avidly I observed them, in hunger and wonderment. They surged around me at a sort of stumble, dull-eyed and confused, like refugees.
Here the protagonist Freddie recalls how strange he felt among people, almost as a different species. In this scene he describes an occasion when he moved through a crowd in the street. The first sentence is filtered through the mind of the experiencing past self and renders his free indirect thought (could they see me? was I outside the spectrum of their vision? now). Further on we notice that mid-paragraph the narrator takes an observing position – I saw myself bobbing head in place of the more expected I bobbed head. Instead of continuing the past-tense memory account by just enumerating the actions, e.g. They surged around me at a sort of stumble, dull-eyed and confused, like refugees.
If we return to the memory process in Banville, we cannot but notice that memory in his works has a strong visual quality. The act of remembering comes close to recording spontaneous visions that have opened before the protagonists’ eyes. But this immediately puts under question the authenticity of the events – as the actions may have happened, so they may well not have happened in reality and might be simply imagined – ‘seen’ – at the moment of narration. More evidence in support of this hypothesis can be found in other examples which employ the same technique and use the same sensory modality construction ‘I see/saw myself’ but unarguably present imaginary scenes: 4. Usually these imaginary, clean killings took place at night, and involved sentries. 5. The thought occurs and at once there I am, in shirt-sleeves and concertina trousers,
Both passages demonstrate narrated imagination. In the first example the narrator Viktor describes how he and Nick used to imagine being sent on a secret mission and how he wondered what it would be like to kill a person conjuring up hypothetical scenes of murder in his mind. In second place we have the narrator Max from The Sea who narrates his spontaneous thoughts in the immediate present. Upon noticing that the grass needs cutting, he suddenly conjures up being behind the mower. But if memory and imagination are narrated in the same way, then memory is strongly equated to imagination. In this line of thought, the memory report loses its truthfulness and becomes highly questionable.
As we have seen so far, the use of a construction like ‘I see myself doing something’ in retrospective narration creates a peculiar doubling of the protagonist into both senser and phenomenon. The narrator’s observational stance is strongly countered by the vivid and lifelike visions that he presents. The vividness of these images can be attributed to the use of the present participial forms following the verb of perception ‘see’. The use of a present participle (e.g. I saw him
An alternative theory proposed by Kirsner and Thompson (1976) frames the distinction between a present participial clause and an infinitival clause in terms of ‘boundedness’ in time summarized by Halliday as follows: ‘The difference between them is a temporal one: the participial clause represents the process as unbounded in time, while the infinitival one represents it as bounded in time’ (cf. Halliday and Matthiessen, 2014: 252; Kirsner and Thompson, 1976). Therefore, if a retrospective text presents the narrative events as unbounded in time, it further questions their authenticity as they are perceived through the senses as still unfolding.
A recent study on the semantic difference between the do-form and the doing-form in perception verb complements (Ono, 2004) sheds new light on the topic approaching the subject from a pragmatic and cognitive perspective. Briefly, this hypothesis postulates that: a. The do-form (‘pure perception’) represents the perceiver’s pure perception of an event with which the perceiver does not experience psychological concern. b. The doing-form (‘cognitive perception’) represents the perceiver’s instantaneous cognition of an event with which the perceiver experiences psychological concern (Ono, 2004: 424).
In other words, the bare infinitive complement reflects raw perception before it is mentally cognized. The perceiver’s neutral seeing of events is equated to a video camera passively recording a scene. Pure perception is also regarded by Ono as ‘objective perception’ or ‘physical perception’. At the other pole, the present participle complement indicates that the speaker is psychologically interested and emotionally engaged in the object of perception and that the perceiver’s consciousness is involved in it. Cognitive perception is also ‘subjective perception’ or ‘psychological perception’. Though there might be some logic in Ono’s proposal, his contrast between subjective and objective, once again, shouldn’t be accepted as coinciding with the narratological notions of experiencing self and narrating self because all represented perception in studies of narrative is subjective. Therefore, Ono’s opposition does not correspond to the distinction between ‘subjective’ and ‘objective’ in narrative studies. If his semantic distinction has to be mapped onto the narratological mode of represented perception, it could be interpreted as depicting degrees of perception or types of perception, with ‘pure perception’ reflecting a weaker form of perception which would be more passive and unengaging, while ‘cognitive perception’ would indicate a stronger, more active and more engaging type of perception but they would both be equally subjective in the narratological sense of the word.
In this sense, introducing memories through the sensory modality pattern ‘I see myself doing something’ has the effect of presenting the past events as detached from the speaker. The memory episodes are presented not as remembered (‘I remember myself’/‘I recall myself’) but as seen (‘I see myself’) with the phrase ‘I see myself’ maintaining the observational position of the narrator who stays outside the past events and merely observes. But the use of the present participle in the macrophenomenon of this mental clause imparts great vividness to the scenes and suggests the subjective participation of the entity designated as phenomenon because the action is presented as incomplete, unbounded in time and provoking a psychological reaction (i.e. a stronger degree of perception according to my redefinition of Ono’s hypothesis). Thus, the narrative becomes split between the outside perspective of the speaker and the inside perspective of the speaker’s image taking part in the visualized scene, which indeed creates a ‘special effect’, though different from the one postulated by Emmott (2002). This has even further narratological consequences in that it challenges the strict binary opposition between narrating I and experiencing I. As we have seen, these two perspectives become fused in a special way when this sensory modality pattern is applied.
Linguistically speaking, the passages from Banville present memories on the borderline between observed and experienced and create a more sophisticated and stylistically-nuanced ‘special effect’ with the protagonists constantly wavering between observation and experience, objective report and subjective account and an external and internal perspective of the events. This ‘perceptual retrospection’, as I dare term it, which in terms of language does not distinguish between fact and illusion, further breaks up the events between remembered and imagined, i.e. between memory and imagination. What is more, a past-tense rendering of the construction presupposes an actor-spectator duplication within the experiencing self itself. Therefore, the ‘special effect’ of visual duplication in memory narratives does more than to split a character into current self and past self and is a highly complex device that conceals multiple layers of meaning.
2.2 I hear myself…
As the above section showed, the semantic pattern ‘I see/saw myself’ in Banville’s novels is one powerful construction that splits the self into two entities. However, Banville does not restrict his presentation of the ‘divided self’ to the visual domain but extends it to other areas of perception, as well. Thus, in some passages the narrative events are filtered not through visual perception but through auditory perception, or the protagonist reflects experiences through his sense of hearing utilizing a construction like ‘I heard myself’. Unlike the phrase ‘I see/saw myself’, which is unrealistic by definition because we never have a vision of ourselves from outside, unless reflected in a mirror, ‘I heard myself’ is not that impossible. Since seeing always happens from inside, we can only see objects that are positioned externally. Hearing ourselves, however, is feasible because we can hear both our own voices and noises and those of others. Yet, the construction ‘I heard myself’ when used in narrative texts can still imply personal displacement. The three examples I have selected from Banville all illustrate disconnection from the self by foregrounding the perception of one’s own speech rather than simply representing the speech itself: 6. There was that gasp of gratitude we could not restrain when someone of the ones in authority over us chose to relent in the prosecution of some trivial order of the day. 7. What a memory I have, to retain so many things and so clearly; I must be imagining them. I stood with my hands plunged in the pockets of my overcoat and squinted about. Bleak thoughts in a dying season. Then, to my considerable surprise, 8. A weeping woman is a terrible spectacle.
In all these examples the narrator alienates himself from the actions performed by expressly relating them to his sense of hearing – I heard myself doing it instead of I did it, I heard myself asking instead of I asked or I heard myself saying Polly’s name instead of I said Polly’s name. By explicitly thematizing the auditory perceptions of his actions rather than simply presenting the actions themselves, the narrator diverts the focus from the content of the utterance, i.e. what I said or asked, to the perception of one’s own voice while speaking. The stylistic effect of this shift is that it creates the impression of a voice disconnected from the self.
In a recent study exploring the grammatical signs of the divided self (Haiman, 1998) Haiman claims that ‘the perfect participant […] looks outward and is visually unaware of himself or herself’ (Haiman, 1998: 69) referring to Langacker’s notion of total participation on a speaker’s part (1985). According to Langacker, a speaker’s total nonrepresentation, which also includes the avoidance of the first-person pronoun, is the strongest marker of subjectivity. Drawing on the idea that any overly specified self-reference can be interpreted as a grammatical signal of alienation, Haiman identifies five linguistic signs of the divided self in first-person report which he arranges in increasing order from the weakest to the strongest: self-address; self-reference; self-conscious emotions; affectation and distinct representations of the self by reflexive pronouns. He pays greatest attention to the last type which he also names ‘mind-body splits’ and illustrates with the following examples:
1 a) I expect — to win. 2 b) I got — up.
a) I expect myself to win. b) I got myself up.
The reflexive pronoun myself in the second set of sentences indicates a separate entity and constitutes an extreme form of alienation between subject and object. It should also be noted that according to Haiman’s classification this is the strongest sign of division of all five in his list. Haiman himself comments that: The representation of reflexivity by a separate (nonclitic) reflexive pronoun in sentences like (5b) and (6b) [herein designated as (1b) and (2b)] originally signalled the recognition of not one but two participants and thus implied some kind of detachment from the self (Haiman, 1998: 72).
The detachment is especially underlined with introverted verbs. Haiman makes use of the distinction introverted-extroverted verbs, with the first denoting an action typically performed on oneself, i.e. reflexivity is incorporated in it (e.g. eat), and the latter denoting an action typically performed on others (e.g. feed).
Along these lines, the co-referential pronouns I and myself linked with an extrovertly-oriented verb like hear indubitably create the impression of two participants. And this totally undermines the status of the individual as a single unified whole who can act of his own volition. By saying ‘I heard myself saying something’ instead of ‘I said something’ the speaker makes the strong suggestion that the voice does not emanate from his individuality but rather acts of its own will. The speaker does not act as a volitional agent and isn’t in control of his own voice. What he commits is a kind of unintentional, involuntary action, which deprives him of any agency and presents a picture of him as a subject whose actions and perceptions are detached from each other. This sensory modality pattern even implies stronger agency in the act of hearing than in the act of speaking, which turns the hero into a mere perceiver of speech that is occurring without him being responsible for it. Thus, the voice disconnected from the body and the speaker’s lack of control over it represent a different type of inner splitting that characterizes Banville’s fiction. The dissociation of hearing and speaking creates the image of a personality that is poorly integrated and that fails to construct a coherent view of the self.
2.3 I feel myself…
An even stronger self-discontinuity than the one introduced by phrases like ‘I see/saw myself’ or ‘I heard myself’ can be found in the narration of memory through a linguistic expression like ‘I feel/felt myself’. The reason for this lies in the semantics of the verb ‘feel’ which, despite still being a verb of perception associated with the sense of touch, in its other meaning denoting psychological states is essentially different from the verbs ‘see’ and ‘hear’. While ‘see’ and ‘hear’ are primarily used to describe our perception of the outside world, ‘feel’ represents the perception of our inner feelings. And though in rare cases ‘see’ and ‘hear’ justify reflexivity, as, for example, hearing your own voice or seeing yourself in a mirror or photograph, the object that follows them is generally different from the perceiving subject. But even in those rare cases the perception of the self comes from outside. Therefore, it can be said that ‘the object of feel is not the objective outward world as in the case of see and hear, but the subjective inner sense arising in the perceiver’ (Ono, 2004: 419). But it should also be borne in mind that the verb ‘feel’ can take two kinds of objects: a physical one – when you feel something happening in the outer reality (e.g. I felt the earth tremble/trembling.), and a mental one – when you feel a certain emotion rising in your inner world (e.g. I felt my interest rise/rising.) (Ono, 2004: 419). According to the taxonomy of free indirect style proposed by Sotirova (2013) and Rundquist (2017) the first use of the verb feel, when it is oriented to the external reality, is normally part of narrated perception, whereas the second use of feel, directed towards the internal world, belongs to the mode of psycho-narration.
So, apart from seeing and hearing themselves and dwelling simultaneously outside and inside, Banville’s protagonists go one step further in their self-disintegration by dividing themselves even within the inner realm of feeling. The construction I feel myself is either employed in the present tense and linked to the consciousness of the narrating self or in the past tense to convey the internal state of the experiencing I. An example from The Book of Evidence illustrates the use of this pattern from the point of view of the narrating I who speaks from his current immediate now: 9. There is so much to be said I do not know where to begin.
Here, the construction I feel myself is used in the physical sense of the word feel to describe the speaker’s bodily sensation of movement. Obviously, this excerpt presents the perception of the narrating I. But it should also be noted that the physical movement of staggering does not take place in reality but rather unfolds in the imagination of the narrator who likens his frustration of having too much to say to a metaphorical action of staggering backwards under the weight of this burden. Yet, he does not only say I stagger backwards but expresses his feeling with the more dissociative I feel myself staggering backwards as if the process of feeling and the process of staggering have as a performer two different agents, thus implying involuntary kinesthetic activity. Moreover, even though the narration takes place at the discourse level of the narrating I speaking from the present moment, there is still a sense of the narrator being subdivided into two entities: a reflective narrator giving an account and a narrator experiencing the act of narration.
In other cases, this construction refers to a mental awareness of a certain emotion. In Banville’s novels it is mainly utilized in the past tense to portray the internal state of the experiencing past I: 10. The wind was pounding at the house now, rattling the windows and setting the rafters creaking and making the stove shoot out spurts of smoke through tiny gaps in the door and along the rim of the red-hot lid. Sitting there,
In this passage the protagonist Oliver recalls a visit to his sister Olive. Interestingly, the first sentence, represented in free indirect style, portrays the subjective viewpoint of the experiencing I signalled by the use of the progressive forms (was pounding, rattling, setting, making), the past perfect tense (had always been) as well as the temporal deictic adverb now. In the second sentence, however, the focalizer distances himself spatially by positioning himself there (also reiterated in the third sentence had always been there) and additionally detaching himself from the events by adding one more agent (I felt myself being absorbed instead of I felt absorbed). The reflexive myself, as Haiman points out, is indicative of a separate being, which standing as a complement to the verb of inner sensation ‘feel’, is a strong marker of a deep internal split. As if the participant of the first process I feels how the participant of the second process myself is absorbed into the rhythm of the room and this split happens entirely in the realm of the experiencing I.
In short, the construction I feel/felt myself doing something is another tool that helps to intensify the protagonists’ feeling of disconnection with their own selves but the lack of internal unity that shows up in these passages draws a different picture of the ‘divided self’ in Banville’s novels. Compared to the visual and auditory splits, this division works on a deeper, more internalized level. This time the perception is not related to the outer world but is directed towards the heroes’ most inner experience. Consequently, the heroes’ integrity is more deeply affected and their self-disintegration more firmly exhibited as they cannot hold things together in a single, unified and solid whole.
3. Conclusion
This article has investigated the use of a specific type of sensory modality which utilizes pronouns of the same person as subject and object and which is called by Emmott ‘special effects’ technique. While Emmott (2002) has to be credited for identifying this linguistic structure, it has to be noted that her examples are limited to a specific genre and not very realistic due to the fantastical context they appear in, which inevitably results in limitations in the analysis itself. In my research, I have examined this modality construction in a more detailed way by viewing it in the context of retrospective fiction and in relation to memory. What I have found out is that although the ‘special effects’ technique is a narrowly-defined sensory modality pattern, it has a far-reaching significance and conveys complex meanings. My study makes use of a larger set of verbs of perception. I have presented a wide range of examples to give a broader view of how the ‘special effects’ technique works in different contexts and how this creates multiple layers of stylistic meaning. I have also explored in greater detail the function of these sensory modality patterns when used in the present and the past tense, the effect of the present participle forms in them as well as the variety of semantic nuances associated with each pattern.
The fact that this linguistic pattern is not an isolated case but a highly consistent pattern inevitably turns it into a stylistic device. Thus, co-referential pronouns linked with verbs of perception can intensify a protagonist’s internal split in retrospective novels by dividing him into both subject and object of the perception. The persistent use of this narrative pattern results in the splitting of the narrative report between observation and experience, perception and participation and, in some cases, between memory and imagination. I have argued that this type of self-expression reflects a splintered self-image and foregrounds the unfathomable nature of the remembered self – a self that remains beyond comprehension. The lack of self-continuity and identity integration implicit in these sensory modality markers is therefore a strong linguistic sign of the protagonists’ discordant identity and divided sense of self.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Antonia Stoyanova’s research is funded by a Vice-Chancellor's Scholarship for Research Excellence (EU) from the University of Nottingham.
