Speech and Thought Representation
Just as readers ascribe mental activity to a character by watching that character’s actions and facial expressions, readers also make assessments based on a character’s utterances both in the form of speech and thought.
In their immensely popular and seminal work Style in Fiction (2007), Geoffrey Leech and Mick Short set out a taxonomy of the ways in which speech and thought can be represented in prose fiction. While not all of their categories can be accommodated by the comics medium, it is worthwhile to investigate the points of cross-over and divergence. The speech categories are as follows (Leech and Short, 2007: 268):
If Direct Speech (DS) represents the words of the character, then any move to the left on this continuum indicates an increased narratorial presence in the utterance, and movements to the right of DS demonstrate a stripping-away of the narrator; utterances in Free Direct Speech (FDS) therefore, are in prose fiction devoid of any narratorial markings such as quotation marks, and the utterance is presented as if produced entirely of the character’s own volition. It is arguable that Indirect Speech (IS) can only occur in graphic narratives when one character is recounting the words of another. Both IS and FIS (Free Indirect Speech) require the presence of a narrator, and this is a highly contentious area in comics studies.
I have provided examples below from TIW to exemplify two of Leech and Short’s categories of speech representation. The first, and most obvious, is the use of the speech balloon to represent DS. In figure 6 below, Iris is in conversation with Granma Suggs. The reading path in the Western tradition travels across the image from left to right and top to bottom, and the conversation between Suggs and Iris follows this path. Suggs makes five utterances and Iris two in this single image, but the direction of the tails of the speech balloons and the spatial layout makes it clear who says what when. Of note here is Ball’s idiosyncratic use of speech balloons to encapsulate sound effects (‘knock knock’), although in this instance it is entirely possible that Dr Magic does actually utter these words.
Figure 6 TIW p. 160
The example in figure 7 below shows how NRSA (Narrative Report of a Speech Act) can be employed in this medium. The utterances of those wearing the ‘I kicked the ass of cancer’ T-shirts cannot be heard by Iris, a fact which is reinforced by the verbal track (her question ‘What?’). The speech balloons do not contain recognisable lexical items, but indecipherable symbols. This device, used several times and in various circumstances throughout TIW, constitutes an NRSA because while the reader is aware of the utterance and the context in which it has been made, the actual words used remain a mystery. In this particular example, it is clear that Iris cannot hear the words spoken because she is distanced from those speaking; however, when those on the platform issue a collective ‘Bye!’, as indicated by the speech balloon with more than one tail, Iris can just about hear this word, demonstrated by the fractured appearance of the lettering.
Figure 7 TIW p. 502
The image in figure 8 provides an interesting example of an utterance that is a mixture of DS and NRSA: Iris’ stricken explanation is offered in between sobs, inside a segmented speech balloon. It is possible that the words written here present Iris’ utterance in its entirety, but it is equally possible that her conversation is truncated into a series of subject headings to convey to the reader the substance of a far longer utterance as Iris tells Maud of her diagnosis and her fears. Iris’ words are represented as NRSA but with elements of DS: the reader knows an utterance was made and is also cognisant of some of the words used, from which the substance of a much lengthier speech can be guessed at.
Figure 8 TIW p. 86
The comics medium in general offers only limited space for verbal utterances and Ball has found an economical solution here. Lengthy utterances encroach on the image inside the frame and risk tipping the balance too far in favour of the verbal track. This latter consideration explains in part the relative scarcity of thought balloons in comparison with utterances in DS; this observation refers to TIW and other comics in general, although exceptions can, of course, always be found. For example, Guy DeLisle tells a fictionalised account of the story of Christophe André’s kidnap and escape in Hostage, in which utterances in captions represent the thoughts of a man who cannot communicate with others because he is a prisoner and because he cannot speak either Chechen or Russian. The eponymous hostage is left alone for long stretches of the narrative. He talks out loud to himself on occasion, but by far the majority of his utterances are rendered in a form of Direct Thought (DT): words that appear in a caption that is a coloured differently from the rest of the text (see figure 9). This form of caption-box in comics is conventionally reserved for the narrator’s voice, but in this particular story, the captions communicate both narrative information and the protagonist’s thoughts. The reader is thus informed of narrative developments, and in the same manner is apprised of the hostage’s anxiety through his mental cogitations. Without the latter, DeLisle’s text would lack both tension and narrativity.
Figure 9 Hostage by Guy DeLisle, p. 286-287
Having touched on the mechanisms for representing thought, I return briefly to Leech and Short in order to reproduce their categorisat¬ions of thought presentation as shown above (2007: 276). Leech and Short point out that in prose fiction, Indirect Thought (IT) is the norm for thought presentation, as opposed to DT being the norm for speech presentation. The reason is simple: the rendering of a character’s thoughts verbatim is perceived as highly artificial in prose fiction (277), and it is more usual for a character’s thoughts to be filtered in some way through the narrator. Many comics arguably lack a narrator entirely, and the option of IT in comics is therefore problematic, to say the least. Writers and artists do, however, make use of DT, but its use is necessarily localised. If used too often, DT very quickly appears contrived, and, as is the case with lengthy spoken utterances, risks overloading the verbal track at the expense of the visual. If the verbal track is allowed to significantly outweigh the visual, there comes a point where the identity of the text as a comic comes into question and the visual track is reduced to mere illustrations of the verbal. Even here, however, exceptions can be found: Posy Simmonds makes free and abundant use of passages of prose text in conjunction with stretches of narrative that is more recognisable as comics in format. (Simmonds, 2001, 2009, 2018). The panel in figure 10 below shows an example of DT in TIW, one of only a handful of similar examples in the book.
Figure 10 TIW p. 17
Other thought balloons do appear from time to time, often containing nothing more than a question mark to indicate confusion on the part of a character. What has not yet been touched upon is the use of silence in comics. Baetens and Frey note that ‘the number of graphic novels including large wordless sections and sequences is steadily increasing’ (2015: 152). Some texts, such as Shaun Tan’s enduringly popular The Arrival, are entirely wordless (Tan, 2006), and this silence ‘invite[s] the reader to gain understanding through observation and deduction, and to decode the narrator’s (or the protagonist’s) intentions, to let symbols and icons “talk”, [and] to deliver information on the implicit level’ (Adler, 2011: 2278). Indeed, in TIW, approximately 40% of over 700 panels have no dialogue (although a percentage of this figure includes panels which feature a sound effect). This figure represents a high proportion of panels which exclude the verbal track, forcing the reader to engage with the visual, and to follow Iris’ story through what can be deduced from her facial expressions, postures, gestures and actions. As Adler points out, ‘[s]ilence functions…not only as a simple absence of speech…but also as a vehicle of a large variety of emotions and mental states connected to the protagonists’ (2011: 2278). Figure 11 shows Iris, exhibiting hair loss from chemotherapy, in the hole at The Helping Hand. Polly, a resident at the clinic, visits Iris and advises her to ‘think about all that sadness and tomorrow, leave it down here’ (TIW, 481). The three wordless images in figure 11 are powerfully eloquent of Iris’ silent despair.
Figure 11 TIW p. 482-483
Speech and thought balloons are capable of rendering far more than just the lexical items of a verbal utterance: the visual track can be deployed to communicate tone, volume, and other extra-linguistic information besides. Baetens and Frey mention ‘grammatextuality’, a concept coined by French theoretician Jean-Gérard Lapacherie, which refers to the visual form of the words in comics (2015: 153). The concept includes ‘the form of the lettering, the configuration of the words in the speech balloons and the insertions of these balloons in the panels, the presence of letters and other written symbols within the fictional world, [and] the presence of…onomatopoeias’ (154). An example from TIW is shown in figure 12 below. Iris has been told that her right breast is to be removed following the discovery of two cancerous tumours. The appearance of the hand-lettered word ‘mastectomy’ which hangs over Iris’ head communicates to the reader Iris’ misgivings about the operation. This is emphasised in the visual track by Iris’ wide-eyed expression and the absence of the lower half of her face as a visual pre-echo of the removal of her breast.
Figure 12 TIW p. 64
In another example from the same moment in the narrative, Iris’ understanding of her situation is rendered as shown in Figure 13. Iris is asked if she understands the information she has just been given. This panel gives us the response ‘yes’ and ‘I’ve got breast cancer’. The word ‘yes’ appears to come from the tree which features as part of the text’s many pictorial metaphors, but the words ‘I’ve got breast cancer’ form the tree’s roots, thus rendering visually Iris’ understanding that the cancer is rooted within her and is growing inside her breast; the breast itself is represented as the hill upon which the tree (the nipple) is standing.
Figure 13 TIW p. 60
Miodrag notes that ‘bold format, underlining, capitals, and shifts in text size visualize the modulations of tone that are natural to speech’ (2013: 71). The convention of using capital letters for the equivalent of shouting has passed into more modern forms of communication such as email and text messages: the larger the font and the heavier the type, the louder the volume. Miodrag is careful to note however that ‘[s]pacing and typography…are vital elements in comics’ visual arsenal, but do not operate as signs standing in for an identifiable signified’ (78). Her main concern, as we have seen, is to dismiss arguments that insist on the parity of word and image, but she concedes that ‘the speech balloon remains a convincing proponent of comics-as-language (or, at any rate, as symbol system) insofar as it reads as a conventionalized sign, recognized as a visual signifier of sound whose border and script affect how we interpret particular instances of its use’ (101). Miodrag notes that the speech balloon is exclusive to comics and that readers recognise ‘by learned convention their relationship to the diegesis: not visible in the world-of-the-work as they are to the reader, these forms represent diegetic material nonetheless, visualizing for the reader what is audible for characters’ (100-101). Achim Hescher also mentions this quality of speech balloons: ‘[b]alloon speech is verbal (as speech or writing) and pictorial (in its outline or shape), and diegetic as speech; the balloons, in contrast, are non-diegetic (for the characters see no speech balloons floating around them)’ (2016: 149).