The concept of focalization is central to discussions of fictional consciousness in prose narrative, but as we shall see, its application to the same in comics is problematic without some medium-specific adjustment. In the section to follow, I outline Genette’s theory of focalization and discuss conceptual modifications imported from film scholarship to address the difficulties created by the constantly shifting focalization in visual narrative. Included in the discussion is Branigan’s assertion that several different narrations can run simultaneously in images, and Borkent’s exploration of the concept of embodiment from cognitive linguistics demonstrates how visual modality allows the reader to construct composite viewpoints.
In a 2008 article which takes for its subject the presentation of minds, Mikkonen concludes that frameworks for the exploration of mind construction current in literary scholarship require some reassessment if they are to be usefully and productively applied to comics. The concept of focalization as it exists in research devoted to prose narratives is one such concept which requires a medium-specific reevaluation if it is to be a helpful tool in the study of fictional consciousness in texts which comprise a visual as well as a verbal track. Gérard Genette’s theory of focalization continues to provide the basis for investigation of fictional consciousness in spite of the many subsequent revisions it has undergone since the publication of Narrative Discourses in 1972 (1980 in English translation), and Ann Miller devotes a chapter of her 2007 book Reading Bande Dessinée to the application of Genette’s theory to French-language comic strips, with some modifications adopted from film scholarship.
Miller begins by restating that focalization falls under Genette’s ‘mood’, a category which reflects the linguistic concept of modality in that it ‘enables information to be affirmed with greater or lesser degrees of certitude or subjective investment’ (Miller, 2007: 105). Focalization, according to Genette’s model, is displayed in the processes of selection and restriction by which narrative information is conveyed. Genette separates the focalizer (who sees?) from the narrator (who speaks?), and he divides focalization into three types depending on the level of restriction to information, where zero focalization is an entirely unrestricted omniscient point of view, internal focalization presents the view of a particular character, and external focalization is limited to strictly behaviourist accounts with no access to mental processes.
In zero focalization, the narrator has knowledge of the character’s thoughts. As we have seen, in the comics medium thoughts can be rendered by way of various devices such as thought balloons, dreams, visions and fantasy sequences. In figure 18 below, Iris is seen contemplating her forthcoming mastectomy and how it will affect her fantasy persona; in figure 19, Iris’ dream features a metaphorical pre-echo of her imminent cancer diagnosis in the form of a huge boulder which is heading her way.
Figure 18 TIW p. 215
Figure 19 TIW p. 11
In the case of internal focalization, the information rendered is dependent on and coloured by the orientation and perspective of one of the characters. What is narrated is likely incomplete and possibly unreliable, but in representing that particular character’s experience of being in the story world, the narration is endowed with its own epistemological veracity. In terms of purely physical viewpoint, internal focalization can take the form of the restricted field of vision of a particular character; in comics, this means that the panel shows the reader only that which the character can see at that moment. An example from TIW (see figure 20 below) shows Iris in hospital following her mastectomy. The image shows her hands and the rest of her body under the bedcovers, and the reader is positioned in a way which makes it clear that the image represents Iris’ field of vision. Mikkonen notes how it is possible to reveal this subjective viewpoint by the inclusion of body parts which belong to the character doing the seeing at the edges of the image (2017: 161).
Figure 20 TIW p. 262-263
Conversely, characters are only seen from the outside in cases of external focalization. This category is exemplified in behaviourist fiction, in which no access is granted to the inner life of the character apart from that which the reader can glean through observing the character’s actions and speech. Figure 21 below shows Iris after having made an appointment with the doctor. Iris herself says nothing, but the reader can deduct from the worried expression created by her frown and lowered brows that Iris is anxious about the outcome of the examination. The constant tapping of Iris’ pencil demonstrates her mental distraction, and the sequence of the panel triad will lead the reader to assume that Iris has been tapping her pencil continuously for an hour; the clock in the first panel reveals the time to be 3pm and Iris’ appointment is at 4pm. Pictorial runes exaggerate the effect of the tapping pencil: the pencil’s movement is shown through repeated versions of its form alongside motion lines, and the volume of the tapping sound is increased by the echoes which shadow the sound effect.
Figure 21 TIW p. 30
Genette discusses five categories in all: order, duration, frequency, mood and voice. The first three categories are largely concerned with narrative time; as we have seen, the mood category deals with narrative perspective, and the final category of voice ‘addresses the question of who tells the story, and what traces of the narrator’s presence may be discerned in the text’ (Miller, 2007: 105). Heterodiegetic narrators are external to the storyworld and can utilise all three types of focalization; by contrast, homodiegetic narrators exist within the storyworld and cannot logically use zero focalization (with, as always, the exception of experimental texts).
Although Genette’s work remains a popular starting-point for further enquiry, it has been subject to numerous revisions and refinements by narratologists and interested parties from other disciplines. Genette himself reformulated ‘who sees?’ as ‘who perceives?’ in his 1988 revision of Narrative Discourse; Mikkonen notes that this was an attempt to capture not only what is seen, but to represent the ‘affective, perceptive, and conceptual centre orienting the narrative’ (Mikkonen, 2017: 151). In other words, this was a revision which represents a movement away from the purely visual aspect of focalization implied by earlier formulations. Mieke Bal, one of Genette’s earliest revisionists, preferred to jettison external focalization altogether, and redefined this category in terms of the relation between the seer and the seen. Bal’s modifications to Genette’s framework have been well-documented, and others have debated what should be understood and encompassed by the term focalization: Shlomith Rimmon-Kenan (2002) argues in favour of the inclusion of mental processes and ideological orientation alongside the purely perceptual facet of focalization, a position that is opposed by first-generation narratologist Seymour Chatman (1978). Achim Hescher (2016) would discard focalization entirely in the case of comics, but this position is certainly opposed by Mikkonen (2017) and Baetens and Frey (2015), who argue that while existing narratological theories are obviously unsuited in their present form for wholesale transferral to the study of comics, they contain much that is useful nevertheless, provided medium-specific adjustments and additions are implemented.
By way of adapting Genette’s work to render it more suitable for application to comics, Miller supplements Genette’s framework with the work André Gaudreault and Francis Jost, who argue that Genette’s three categories of focalization refer to knowledge, whereas ‘ocularization’ represents the visual perspective of a character, thus separating what a character knows from what they see. Furthermore, Gaudreault and Jost identify a subcategory of ocularized shots which ‘bear traces of the subjectivity of a character through deformation’ (Miller, 2007: 106), as can be seen in figure 22 below. As Iris succumbs to the anaesthetic, the sides of the frame gradually encroach across the panel until an entirely black panel indicates that Iris is now unconscious. The numbers of the surgeon’s countdown also become increasingly wobbly and malformed as the surgeon’s voice become fainter and more indistinguishable for Iris.
Figure 22 TIW p. 242-243
Subjective traces ‘may be further extended through images which represent purely mental processes such as dreams’ (Miller, 2007: 106). Gaudreault and Jost argue that the status of these images is made clear by modalization operators such as a foregrounded change in the panel’s shape to indicate a dream sequence, but if these operators are not present, the reader has to distinguish between what is real in the storyworld and what is real to the character: internal focalization presents character perception as fact. Iris’ second dream in chapter 8 features a deliberately misleading opening sequence which shows Iris waking on the morning of her scheduled mastectomy. The dream-status of the following episode is uncertain for perhaps twenty panels. Iris’ encounter with the paper dolls in chapter 5 is even more puzzling: Iris is intoxicated, of course, and this sequence could be written off as a drunken nightmare if it were not for the continued presence of the paper doll bearing the personal ad which leads Iris into communication with Henry. The doll can be seen pinned to the wall above Iris’ computer from this point onwards, and it is Iris’ action of discarding and burying the doll in chapter 16 which marks the end of her fantasy-romance. The doll’s sudden appearance in an unfamiliar location at this crucial juncture in the narrative imbues it with metonymic significance: the doll is a stand-in for Henry and the fantasy world Iris formerly cherished. Nevertheless, the question of the doll’s physical existence in the storyworld remains unclear.
It would be useful at this point to summarise that which has been discussed so far. Graphic storytelling allows for zero, internal and external focalization in heterodiegetic narration. Panels that are externally focalized naturally do not include subjective imagery or a point-of-view shot which would suggest the filtering consciousness of a character. Characters are depicted from an external viewpoint and thought balloons do not feature in externally focalized images because this device allows access to a character’s mental activity. At the other extreme, zero focalization implies omniscience and access is given to the characters’ inner life, although it is rare that all characters are treated equally in this regard (Miller, 2007: 110). It is more common that one or two characters will be selected for such privileged access. In TIW the reader has access to Iris’ thoughts and feelings, and very occasionally those of Maud, although in the case of the latter access is usually restricted to a thought balloon containing a question mark to indicate Maud’s confusion or her frustrated attempts to understand Iris’ behaviour.
Internal focalization provides access to one particular character’s mind. However, ‘it is rare that images are restricted to the ocular viewpoint of that character’ (Miller, 2007: 109), although examples can be found. In figure 23 below, the reader sees Maud from Iris’ perspective, indicated by Maud’s gaze and her placement in the frame, and the tails of Iris’ speech balloons which point to the speaker at a position somewhere outside of the panel’s frame. Film theorist Edward Branigan, whose work I shall turn to shortly, contends that the point-of-view shot is composed of two shots (Branigan, 1984: 103). Following Branigan’s analysis, Iris’ presence at the window is established as a point in space in the panel on page 296. In the next panel, her glance is directed towards an ‘off-camera’ object, in this case, Maud. The ‘camera’ is located from Iris’ position in the panels on pages 298 and 299 and focuses on Maud-as-object. Maud’s thought balloon on page 299 indicates her puzzlement that Iris could have forgotten her hospital appointment. (In fact, the presence of this thought balloon in an ocularized image means that the reader is presented with two inner lives simultaneously: Iris’ field of vision and Maud’s mental confusion.)
Figure 23 TIW p. 298-299
Ocularized views such as the example discussed above are possible in localised instances, but while it is a relatively simple matter to sustain one particular fictional consciousness in prose texts, it is impractical, difficult, and arguably undesirable to do so in comics (Mikkonen, 2008: 316). Focalization in comics is constantly shifting and as a result of this perpetual movement, Miller states that graphic storytelling ‘tends to exhibit a certain permeability between inner and outer worlds’ (2007: 119). Film scholars have noted a similar phenomenon with respect to their own medium, and Branigan cites the example of Robert Montgomery’s 1946 film Lady in the Lake, which sustains the point of view of the investigating detective, Philip Marlowe, throughout its duration (Branigan, 1992: 142-160). Montgomery as Marlowe is visible onscreen only in reflections or when his hands come into the frame, and the camera renders as accurately as possible this character’s ocularized view. However, cinema audiences quickly wearied of the experiment and found the viewing experience to be frustrating. A far more successful experiment with first-person consciousness can be found in Channel 4’s sitcom Peep Show, which features ocularized shots simultaneously with voice-overs, so the viewer sees what the character can see and hears what the character is thinking. This inner access is restricted entirely to the two main characters, Jeremy and Mark, played by Robert Webb and David Mitchell respectively. The popularity of this show attests to its success, unlike the now infamous failure of Montgomery’s film. Peep Show’s appeal rests on the fact that internal focalization is often deployed to set the visual and verbal tracks against each other for comic effect, and the shifting viewpoint between the two leads allows the audience to witness Jeremy and Mark from external as well as internal positions.
The work of film scholar Edward Branigan has already been briefly touched upon, but I wish to consider in more detail now his argument that several different narrations can operate simultaneously. Branigan defines the character as an agent who provides information by simply living in their world and talking to other characters who inhabit the same world, and focalization depends upon that character experiencing something. Branigan draws a distinction between looking/listening, which is intersubjective and can be reported by a narrator, and seeing/hearing, which is a personal experience and can only be recounted through means of internal or external focalization.
Character experiences can be rendered internally through point of view shots and dream sequences, or externally through close-ups and eyeline matches. As well as seeing and hearing, Branigan notes that focalization extends to ‘thinking, remembering, interpreting, wondering, fearing, believing, desiring, understanding, feeling guilt’ (Branigan, 1992: 101).
On the question of different narrations running concurrently, Branigan quotes fellow film scholar Stephen Heath, who argues that there is no real dichotomy between a subjective point of view shot and an objective non-point of view shot. The latter can be the basis over which the former runs, and an external shot can be overlaid with consciousness. This line of reasoning has enormous consequences for the interpretation of visual narrative in that it removes the onus to positively identify a single focalizing source for each panel, and it becomes possible to describe an image in terms of numerous levels of focalization. As previously stated, visual narratives feature numerous transitions from panel to panel and cannot sustain the depiction of a single consciousness without contrivance. Heath’s argument suggests there is ground for claiming that fictional consciousness in visual narrative could be sustained over extended passages if it is transposed over a basis of externally rendered images.
To exemplify this point, the image from TIW below (figure 24) presents an apparently objective narratorial third-person perspective because Iris is not present and there is nothing to suggest that this is an ocularized image based on her field of vision. Nevertheless, traces of Iris’ subjectivity are present and her mental preoccupations are clearly reflected in the symmetry of the image which is neatly divided into two. Her computer takes centre-stage, with Henry’s kiss forming the focal point of the entire panel. The paper doll seen to the left of the computer represents the fantasy world of Iris’ online romance, but the letter from the hospital which is pinned above the computer on the right-hand side provides a stark reminder of the reality of Iris’ cancer and her imminent mastectomy. The panel following (figure 25) shows continuity of spatial arrangement in that the computer screen is still visible, thus indicating to the reader that the location is identical to the previous panel but the ‘camera’ has moved closer to the letter so that its text is fully legible. This refocusing of the reader’s attention on Iris’ real situation is a narratorial move to underline the escapist nature of her fantasy.
Figure 24 TIW p. 140
Figure 25 TIW p. 141
To demonstrate composite viewpoints over a more extended sequence, I refer below to pages 131-139 in TIW, which depicts a messaging conversation between Iris (balletgirl42) and Henry (sailorbuoy-39). The reader is aware that Iris has misrepresented herself to Henry, and she is writing as balletgirl42, a persona she has invented as a distraction from her diagnosis. What is unapparent to the reader on a first reading, however, is that Henry’s persona, sailorbuoy-39, is also a fiction: he too, is not what he claims to be. The voices of the two characters – or at least, their personae – appear onscreen in the form of messages, and the messages are depicted as panels within the wider surrounding panel. As such, the frame of each message operates in the same way as a speech balloon. The frame of the panel on page 132 (figure 26) takes the form of the computer screen and water drips from the bottom, leaking out of the frame’s dimensions. The frame’s contents have also switched, and now include an image of the lighthouse Henry is describing.
Figure 26 TIW p. 132
Horstkotte and Pedri would likely label this a shift in visual vocabulary (2011), and this shift marks the movement into Iris’ consciousness and her fantasy of Henry. (Of course, it is also true that these images represent Henry’s own fantasy version of himself.) The reader reads Henry’s words and witnesses Iris’ fantasy simultaneously (figure 27).
Figure 27 TIW p. 135-136
On another narrative level, the reader knows that Iris has been diagnosed with cancer and is shortly to undergo a mastectomy; this narrative level is still present throughout Iris and Henry’s conversation, and resurfaces in the image of the letter from the hospital.
In a 2017 article published in the journal Cognitive Linguistics, Mike Borkent considers how viewpoints are constructed from ‘multiple modalities’ (2017: 539) and he brings more recent cognitive theories to the study of focalization. He argues that theories of ‘embodiment, domains, mental simulation, and mental space blending’ (2017: 539) can provide a framework for the analysis of multimodal viewpoint construction in comics. Borkent considers carefully the communicative and expressive possibilities of the comics medium and notes that the range of signs through which comics relay their narratives (referred to previously in the discussion of Peirce’s classifications) have their own ‘viewpoint prompting features and conventions’ (2017: 541). Images communicate through iconicity and on the surface seem to do little more than point to the storyworld existence of an object through the depiction of something resembling that object, but even in this case, Borkent maintains that the ‘variable formal features’ of such images ‘can significantly reconstrue content’ (2017: 540). Iconic signs in comics interact with other signs in turn to construct viewpoints that are multimodal and multilayered.
Borkent uses the cognitive linguistic concept of embodiment to explain how comics readers ‘activate lived experiences – from basic sensorimotor schemas up to cultural knowledge – to flesh out and interpret limited communicative cues’ (2017: 542). This is to say that readers’ gap-filling activities extend far beyond mentally completing the spaces, or gutters, between panels. Real-world knowledge is drawn upon to complete the information given in the panel. Borkent notes that the panel itself ‘acts as a window of attention’ (544; cf. Cohn 2013, Bridgeman, 2005, and Jahn, 1996) presenting a view into the storyworld that is not unfiltered: the composition of the panel and its style of presentation can be mobilised to influence the reader’s inference processes. Such processes might include completing a half-shown image, filling in a blank background, and importing ‘cues from prior panels to contextualise the action’ (Borkent, 2017: 544), all of which is reminiscent of Catherine Emmott’s work on contextual frame theory which focuses on how readers build mental frames of reference in relation to characters, settings, and so on, to create, maintain and update a storyworld (1997). An artist does not have to produce endless replications of one panel because the reader will assume that information not shown is still present; furthermore, this kind of ellipsis enables the foregrounding of communicative signs which contain salient narrative information. In addition, readers interpret signs with the aid of domain networks. The concept of ‘domains’ originates from research in semantics, and refers to pockets of stored information which people activate whenever that particular domain is encountered. In the example from TIW discussed in this section, the domain is ‘online dating’, activated by the images of the personal ad on pages 125 and 126 (see figure 28 below). The reader is able to make inferences connected with this domain, concerning roles, values, interactions and ideology. Given that Iris and Henry are both less than honest about their true identities, the sub-domain of catfishing is also activated for those readers who are aware of such things.
Figure 28 TIW p. 125-126