Let’s examine an interesting narratological problem in a typical, if not unspectacular mainstream comicbook. Werewolf by Night #23 (Nov. 1974) concerns the eponymous anti-hero running afoul a deranged, disfigured movie actor who murders his former co-workers in revenge for a on-set stunt that went horribly wrong. From pages three to six of the comic, we see Buck Cowan, the Werewolf’s human comrade, explain how the actor became a serial killer in an extended flashback sequence.
Specifically, let’s look at the fourth panel of the page and spell out its larger implications for mainstream comicbook narration more generally. The image depicts the actor swinging past a statue and over a fire pit and in the upper-right corner Buck narrates in a caption box stating, “But his arrogance evaporated midway through his swing…as he failed to clear a spike on the dragon-statue —”
At least since Wayne Booth, we have know that narrative is all a matter of rhetoric; even if it seems as if were simply seeing or reading narrative details there is always an act of narration or “telling” occurring. In the Werewolf panel we are presented with a double act of telling, encompassing both the disembodied author / artist as well as the embodied “narrator” / Buck as a textual construct designed to order, authenticate and respond to story information, or in narratological terms fabula. And if we assume that Buck’s verbal narration across the page is continuous and in more-or-less real time, then the image within the panel can only be attributed to this outside narrative agency. The simple reason is this: the image contains details that aren’t in Buck’s exposition. Specifically, Buck’s verbal account depends on the narrative detail of the “dragon-statue”; however, the dragon-statue, with its relationship to the film set and the danger in its adjacent pits, is only described through the visual rendering delivered in this and the preceding panels.
This disjuncture of word and image could be understood as contradiction with respect to what David Bordwell and others have called the communicativeness of plot, that is relative rate of disclosure or revelation with respect to story information. In this case, Buck’s version of the story is far less communicative than the images that accompany it. How then are we to understand this extra information that is clearly not part of Buck’s verbal story yet absolutely essential to make sense of his captions? One could assume that they were implied by Buck’s story, that they were delivered to the Werewolf in the gutters between the panels. Or we could even assume that the images are the synthesized thoughts and mental images of either Buck as he remembers or the Werewolf as he listens to Buck’s recounting. However, none of these narrative solutions are really any more than needlessly baroque inventions imposed onto the text to paper over the fundamental division between verbal and visual semiotic resources in comicbook narration.
Much of strangeness of this panel is easy to miss in the general course of reading, largely because of the form’s reliance on codes of sequencing and plotting derived from other media, particularly film. A rather old-fashioned use of the flashback in film (and television) has a diegetic narrator begin recounting or remembering the past followed by the story of the past delivered in a completely omniscient fashion, that is incorporating visual and audio information that the ostensible narrator could have known, or at least couldn’t have seen from that particular angle, or heard with that particular fidelity, etc. This is not insinuate that many film flashbacks are “wrong” only that they have conventionally established flexibility with respect to narration norms. I suspect that the panel in Werewolf was composed with the same conventions in mind, but the suspended presence of the internal narrator, through the spoken captions, complicates this simple translation.
On the one hand, all this suggest nothing more than the common sense conclusion that the image and the word will always be at odds in the narration of plot. But conversely we can use this panel to suggest the centrality of multimodality in comicbook composition. Multimodality, as described by Gunther Kress and others, describes the fact that semiotic material may be crafted out of information in any number of modes — a story may be told in a poem, a novel, a picture, etc. — and that we live in a world in which communication and expression more frequently than not incorporates multiple modes to convey meaning. Of course, comic art has always been multimodal, using mixtures of verbal and visual resources to do its work. The panel in Werewolf demonstrates a narrative constructed in a way that it can only be decoded through multimodality; we would never understand exactly how the actor died without the images and we would have never understood why without Buck’s discussion.
Yet this begs a deeper question: if comicbook narration depends on a combination of verbal and visual resources, then how is one to decide what should be rendered by which? Either technique could be used to capture the raw material of story information. Of course, there seems to be certain proclivities, or perhaps habits; the visual seems suited for the description of figures and environment and the verbal seems better suited for characterization and exposition. However these functional roles certainly aren’t steadfast, nor should they be. We need only look to the representations of sound effects, particularly in the hyper-baroque, panel-filling variants typical of the better work of Walt Simonson and Howard Chaykin, to observe a figure that complicates this too-easy verbal / visual dichotomy; they are both verbal and visual at once. But more deeply, we should get used to questioning the traditional functions of the visual and the verbal in comicbooks, as neither is indefinitely attached to a single role. Dave Sim’s work on Cerebus, which alternates pages of only verbal text, dense illustrative drawings, with more traditional comicbook paneling and sequencing (particularly in Jaka’s Story), seems exemplary in its ability to disrupt these functional tendencies.
Tendencies in what semiotic mode is used for what purpose could also be used to think about comicbook categorization. In other words, I suspect that there are great similarities in the strategies and practices to capture certain elements in image and others in word that could be traced through and across texts, genres and schools. For example, I have discussed in other works the production norms accompanying Classical Heroic Form (CHF), roughly understood as mainstream superhero comics. More than just similarities of theme and subject manner, these works share visual strategies and accompanying work practices that, for example, highlight visual dominance and the search for the most “active” pose for figures within panels. In other words, traditional makers of CHF made coordinated decisions to tell stories with images of exaggerated poses and, secondarily, with the dialogue that accompanies them. They also made clear decisions (clear in the process and clear in the result) about which resource should be used for which function (i.e. images do this…words do this). This is not how all comicbook narration is done or should be done, it is one coordinated, historical solution, or one set of historical poetics.
Moving forward it might be interesting to attempt to categorize comicbooks more by resource and function, which could perhaps put comics unlike in subject matter or “intention” into deeper conversation with one another. Further, it would be likewise fascinating to construct a series of historical studies using this model to consider other historical narrational systems (a la Bordwell), in addition to how, why and when they developed.