Inferential Revision in Comics Page Interpretation
A Hermeneutic Approach to Renegotiating Panel Comprehension

Stephan Packard (Köln)

 

What can structuralist or hermeneutic concepts add to the detailed explanations of comics comprehension offered by linguistics or cognitive studies? I argue that comics page interpretation should be extended beyond the most immediate cognitive aspects of comics page comprehension, allowing for logically later stages in that process during which a consideration of the different areas of meaning creation will be more flexible. Such an approach will be especially useful to explain artistically prompted processes of inferential revision on the distinct levels of syntactic cohesion, semantic coherence, and storyworld consistency. This usefulness rests on the ultimately non-binding manner in which expectations of comics’ structures are shaped by heautonomic rules, among others, which I will conceive as auto-poetic principles of organization that contextualize comics as an intermedium between more strictly structured art forms, as principles of an art form in several genres, and as rules introduced by and applicable only to individual pieces of poetry or art. Taking my examples from the original run of Amazing Spider-Man, I depart from one expertly handled surprising twist in a panel to discuss different versions of the comics/language analogy and the shaping of audience expectations afforded by different conceptions of rules for a visual language. Using a complex page to illustrate how inferential revision develops, I go on to point out one exemplary overlap with theories of panel segmentation, where some ideas about a panel structure divided into several domains can serve as mediating concepts between individual artistic heautonomies and a more generalized visual language. This directs attention to artistic devices, not only in cases of surprising revelations, but when they invite inferential revision in panel sequence comprehension in general.

A ›Bad Grammar‹ Panel: Surprising Compositions

In Amazing Spider-Man 47 from 1967, the following panel (fig. 1)¹ stands out by its simple play with expectations:

Supervillain Kraven the Hunter thinks he is about to punch superhero Spider-Man. But the hero is in a different place than Kraven expects. That brief moment in the storyworld is underscored by the unusual structure of the panel. Its main motion is dominated by Kraven’s elongated body shape and the accompanying motion lines that indicate movement from right to left, against the reading direction. Meanwhile, Spider-Man is tucked away in the top-right corner, observing Kraven’s futile movement and, in fact, even presenting it to an imagined audience through his openly gesturing hands, while commenting on Kraven’s failure. This gesture exploits an ambiguity between the finger gestures that Spider-Man routinely uses to shoot web fluid, and a more general pose of presenting and pointing. Web shooting would fit the context of the combat scene and the genre, but a contemplative demonstration fits the surprising calmness of Spider-Man’s dialogue. What is more, we do not see any webs. The slower, demonstrative gesture contrasts with the dynamic body pose. This fits the complex temporality of the panel, as Spider-Man appears as the foregrounded scene’s presenter only after that foreground has likely been taken in by the observer-readers.²

That Spider-Man is a commentator rather than a participant becomes clear only through an almost immediate revision of the expectations built up by the ongoing scene. John Romita’s ingenious artwork is complemented nicely by Stan Lee’s words, as both characters underline the turn against expectations in their dialogue. Kraven clearly expresses anticipation: »Once I get my hands on you« he begins, the ›once‹ outlining that hope, and he continues in the future tense: »I’ll –«, only to interrupt himself in a deflated grunt. Spider-Man then frames the interrupted sentence as an analogy to the aborted combative movement, admonishing Kraven for ending »a sentence with an expletive« and going on to describe such »bad grammar« as »unforgivable«. Taken together with his speech style and calm pose, his comment moves Kraven’s words as well as the villain’s physical movement to the level of object language, framed and evaluated by the hero’s metalingual reflection.

How far may we take Peter’s metaphorical merging of the tactically failed action with a syntactically infelicitous statement? Perhaps all three interruptions – from panel structure through combat movement to dialogue – come together: As the surprising discursive panel structure underlines one character’s failed action in the story, while another character compares that failure to a syntactical mishap, does the panel structure itself mirror a mistreatment or flaunting of discursive expectations?

It is noteworthy how each of the elements assembled to create this small surprise for the observer-readers is taken from established conventions of the art form, genre, or issue. In the immediate artistic vicinity of this panel, the villain does usually come from the right when confronting the hero, which poses the antagonistic movement against the direction of the panel sequence. A quick survey of the same issue confirms this: Of the 38 panels that present Kraven in a clear binary panel, i.e., a panel with a distinguishable left and right space, he is placed on the right-hand side 22 times. Of the remaining 16 panels, 14 lack any appearance of the protagonist. 11 of these are comprised of arrangements at the beginning of new panel sequences, which set up Kraven as the main actor in a new scene, in three directly continuing runs of 4 (1967, 7–8), 4 (11–12) and 3 (19) panels each, before the perspective turns to move him back to the right side in every case. In addition to these 38 binary compositions, three more complicated panel sequences with more than two dominant spaces per panel (13–14, 16, and 19) underscore the rule by having Spider-Man belatedly enter a sequence already begun with Kraven as the main actor, whereupon the spatial configuration turns as they engage one another: See the first of these instances as one case in point (fig. 2), where Kraven’s plane of action, continued from the first panel, opens up to the space above in the second panel, where the perspective shifts and Spider-Man appears vertically from above. Spider-Man eventually re-establishes himself on the left side of the fourth panel, with circular motion lines and Spider-Man’s rounded and dynamic pose emphasizing the turnaround. Once again, Kraven punches empty space, though in a less surprising fashion. One might argue that the ›bad grammar‹ panel that follows in a different fight two pages later encapsulates quite the same turnaround, but compresses it into just one image, thus creating a greater surprise by the more sudden resolution, while also drawing attention to this flaunting of expectations through the dialogue.

Thus, as a general rule, whenever Spider-Man appears in a clearly binary panel together with Kraven, Spider-Man is drawn in the left half of the panel while Kraven is on the right; and should the panel be divided in more complex ways, the sequences resolve by returning to the same binary arrangement. The only two exceptions, in which binary panels are arranged differently, are the two moments of surprise in Spider-Man’s and Kraven’s combat: As Kraven suddenly uses his special weapon in this issue (17–18), he gains the upper hand by dominating the action from the left, winning that fight. Conversely, when the misplacement of the hero is under the latter’s control, as in the ›bad grammar‹ panel, the antagonist is immediately defeated and has to flee, notably by flipping around in a complicated somersault that takes the direction of his movements altogether out of the concluding sequence (fig. 3).

So we expect Kraven to come from the right, but are surprised that his movement is presented as misdirected in the ›bad grammar‹ panel. Similarly, finding Spider-Man crouching in a corner with demonstratively spread out fingers is an equally expected motif for the arachnoid protagonist; but we do not expect such sovereign immobility from him at this point, as opposed to moments before or after a fight, such as when he enters the parallel scene from above (in fig. 2). A third well-established device put to specific use in this panel’s surprise is the time differential between the elaborate dialogue and the immediacy of the fast-paced action sequence’s movements. This is, of course, a consequence of the ›Marvel Method‹ employed in producing this comic, which had Romita draw out whole sequences without a detailed script and Lee adding his famously verbose dialogue and captions later (cf., e.g. Groth). This discursive structure affords both the narrator and the characters much greater liberty to comment upon an event than the mere diegesis would render plausible. Lee often uses this opportunity to comment on the visually conveyed information through the narrator’s or various characters’ voices in attitudes that would also make sense from an extradiegetic position. In one such instance in this issue, he has Spider-Man react to a perhaps too-early encounter between hero and antagonist, which hence must remain without further consequence for the story, by suggesting: »[…] let’s skip the preliminaries and get to the last act!« (fig. 2). That Spider-Man can comment on these events, almost as if he were a member of some audience or a reader-observer, underlines his superiority. (This device is completely erased in moments of actual threat to the character.)

In the ›bad grammar‹ panel (fig. 1), Spider-Man uses this sovereign attitude to turn the antagonist’s speech into his direct object of observation, indirectly mocking the dynamic movement whose disruption it parallels. The panel effectively has three narrative speeds: the punch is slowed down by the time requirement implied by Kraven’s words and, as his punch and speech are interrupted, the discourse slows down again for Spider-Man’s lengthier and metalingual comments. This takes us out of the fight scene for good, which ends in the next panel with Kraven’s retreat, somersaulting away alone against an empty panel foil.

Notably, we can neither explain the surprise offered in this panel as a complete break from nor as a complete fulfillment of the rules and the expectations that they shape. Obviously, the relationships and connections between these rules, expectations, narrative strategies, interpretations, inferences, and revisions require further elaboration. In the instance discussed so far, we have to assume that we are dealing with two different implementations of those rules, leading us to distinguish between two somewhat distinct moments in comprehension: First, we are led to expect one thing, but then, at a logically (not necessarily chronologically) later moment, we are prompted to revise our previous inferences. Since the extent of this segment is confined to just one panel, it is more limited than in typical cases of narrative unreliability (cf. Booth; Phelan; Shen; Packard 2013). While calling the instance narratively unreliable might still be technically accurate, exploring that concept in this context would add only a little to our understanding of the specific devices in play in this panel structure, taking us from the terminological traditions of grammar or a visual linguistics to those of rhetorics. Of the two techniques in question, one we have explored so far and the other adopted from unreliable narration, the latter is at home in storytelling in general, while the former is taken from comics’ detailed aesthetic devices. Nevertheless, they certainly overlap in at least two ways: the observer-readers are expected to come not to one, but to two different interpretations, and they are to consider both interpretations to have been created in accordance with some rules, as opposed to a text merely contradicting itself directly and without justification. While the readers' response to a felicitous instance of unreliable narration must accept that they have been deceived fairly as they reach a surprising twist, the observer-readers of the ›bad grammar‹ panel might not even have reflected upon their shift in interpretations, if the dialogue had not pointed it out. But even without this broad hint, they would still make the same distinct inferences – Kraven’s punch is directed at Spider-Man, yet Kraven’s punch is not in the direction of Spider-Man, one very briefly after the other.

While all of the elements that are used to build towards the surprise in this panel are taken from established conventions, it is their combination that breaks expectations. We expect Kraven to move from right to left, but do not anticipate that space to be suddenly empty. We expect Spider-Man to crouch, but not in the middle of a fight. We expect the dialogue to stretch the temporality of moments in combat through commentary, but in this case the commentary repeats and describes the previous two visual surprises by echoing them in lingually parallel constructions. If one were to think of the rules of panel structures as a grammar, by close analogy or loose metaphor, one might be hard-pressed to decide whether the panel is ungrammatical, or whether it uses the assumed grammar of comics in an especially striking and perhaps unusual way. Clearly, the panel is coherent, and the depicted events in the world are consistent, at least with regard to genre rules, which allow both for overblown violence and for lengthy intermittent dialogue from the violent actors. Even narrative expectations are ultimately fulfilled rather than violated, as the comparison with the longer scene (fig. 2) demonstrates: we might very well expect Spider-Man to evade Kraven’s blows, and yet it might just take a bit more effort to realize that he does exactly that in this compressed panel. If the surprise thus concerns the process of textual or media comprehension, does it make sense to say it is prepared for on a level of syntactical cohesion rather than semantic coherence or diegetic consistency?

What are the Rules? Inferential Revision and Heautonomy in Comics

One way to expand on that question is to reconsider what kinds of rules build expectations for comics comprehension in the first place, and to ask which – if any – of them may be understood in analogy to verbal language. The treatment of cohesion and coherence in comics, at least when summarized under those terms, has recently been dominated by linguistic and cognitive approaches. The concepts emerging from the associated disciplines may often seem to overlap with an inheritance – or baggage – that comics studies has carried over from those decades starting in the 1970s in which approaches in emergent comics studies were mostly adopted from Russian formalism, literary criticism, the history of art, as well as visual and cultural studies, and thus moved closer to linguistic work, even as they eschewed a specialized linguistic vocabulary for a more generalized semiotic set of concepts.³ Some accounts undertake reconciliation, sometimes even attempting to build bridges from McCloud’s very broad concept of ›closure‹ through psychological gestalt theory and narratological concepts of implied readership to the lucid explicitness of a detailed multimodal discourse analysis or the strong empirical foundations of experimental cognitive studies.⁴ Others, including some of the most prolific authors in each field, tend to strictly separate hermeneutic traditions of free interpretation from empirically grounded insights into media comprehension.

The same question does come up in these debates: To what extent can the exploration of comics’ coherence and cohesion be understood as the detailing of an – if multimodal – language? But while the development of multimodal discourse analysis for comics studies (Bateman and Wildfeuer 2015 and beyond) as well as the emergence of a differentiated concept of visual language in cognitive studies (Cohn 2013) have spelled out specific, if partially differing, relationships between verbal language and the structures that supposedly underlie comics, hermeneutic traditions still seem to be using both the language metaphor and accompanying general semiotic vocabulary in a less clarified way.⁵

Let us consider this spectrum of conceptual approaches by looking at the case of the ›bad grammar‹ panel once again. McCloud’s idea of ›closure‹ (McCloud, chapter 3) is itself always already conceptually transformative. It is first introduced as the process by which we complement a limited visual depiction by expanding partially represented spaces, objects, and protagonists: For a boy depicted in one panel as walking along a street, we will assume that the street continues beyond the confines of that segment cut off by the panel borders. For a body depicted from the head down to the waist, we will assume legs that are not shown. McCloud imagines this process to be motivated by the limits of our visual perception: We will only see part of a street we are walking along, yet assume that the remainder of the world persists (60), and we will be able to envision the whole of a Pepsi bottle from seeing only one part of its surface on a supermarket shelf (63). In these parts of his implicit cognitive theory, McCloud is basically following the foundational concepts as well as examples of psychological gestalt theory. In what should perhaps be more clearly marked out as a second step beyond completing each gestalt based on the salient parts of a whole, this idea of ›closure‹ that is in play when »a mere shape or outline is enough«, is then taken to encompass the »constant, even overpowering« process by which the projected sequence of movie images or the cathode-ray tube television’s »single point of light, racing across the screen« dissolves into complete, moving images in the observers’ perception (McCloud, 64).

One important difference that might be momentarily lost in this series of ideas is that the single, incomplete view of the street, bottle, or body allows us to re-examine the visual evidence as first presented, once we realize what our inferences have added (which is exactly how McCloud employs those same pictures as he demonstrates the idea). In other words, it allows for direct inferential revision. The ›bad grammar‹ panel does not force us to accept a surprising twist in a sequence of moving images on a screen, which we might then revise only based on our memory or by halting a video and re-examining what we saw by interrupting the most obvious order of viewing (fig. 1). Instead, Spider-Man’s demonstrative gesture and verbal commentary directly invite us to refocus our attention on the left two-thirds of the panel and reconsider Kraven’s powerfully directed punch as altogether misdirected, his dynamic motion as aborted rather than dominating. As McCloud moves on to a third step, now summarizing the ›closure‹ across two or more panels that brings together individual depictions to one semantically coherent whole (which he illustrates and perhaps equates with a supposedly syntactic movement that brings panels together cohesively across gutters), he emphasizes the same distinction: »The closure of electronic media is continuous, largely involuntary and virtually imperceptible. But closure in comics is far from continuous and anything but involuntary!« (68) It is here that a conception originally comprised of the building blocks of immediate perception and cognition extends to a theory of implied readership, which could easily connect to the relevant theories in literary criticism (cf. Iser). For a larger sequence in Spider-Man (fig. 2), we might think of Spider-Man’s introduction to the sequence as coinciding with his first full appearance in panel 2. But we might equally assume a different reading experience which follows Kraven’s movement across the top row of panels 1 to 3, and which is then prompted only by Spider-Man’s more prominent depiction in the top left of panel 3 to discover his less prominent appearance at the top of panel 2. If we instead follow the dialogue, we find that his intervention is signaled already in panel 1, as we first read his words coming in via a speech bubble from off-panel. Empirical studies and cognitive theories on the actual order of gaze movements and inferential readings can elucidate these processes much further. However, beyond and in addition to their results, it will remain true that these three possible discursive orders – Spider-Man introduced by dialogue in panel 1, Spider-Man introduced as a small depiction in panel 2, and Spider-Man introduced as a large depiction in panel 3 – are all arguably salient for the process of re-examination. The whole set-up, and especially the inquisitive construction of the dialogue (»anyone else would have waited« – »That voice! I’d know it anywhere!« – »next thing we know« etc.), invites the observer-readers to engage in just such a re-examination.

While such inferential revision is an important aspect of textual comprehension (cf. McNamara and Magliano; Kendeou and O’Brien; Kendeou 2014) as well as the comprehension of narrative in general (cf. Poynor et al.; Rapp and Kendeou), especially as we deal with narrative unreliability or other devices that negotiate ambiguities (cf. Dutke and von Hecker), there is evidence that it takes on additional importance when ostentatiously multimodal media products engage us in bringing together information across auditory and visual domains (Evans et al.), and similarly across single panels, panel sequences, and image and text combinations in comics (Cohn and Magliano; Cohn 2014; Cohn and Kutas 2015; 2017). In addition to fast revision processes in which »both the evaluation of mismatches and the revision of no longer relevant information can occur at an inferential processing level« (Pérez et al., 1106), artistic effects will often develop their full impact as continued discursive discussion ensues well beyond the immediate moment of primary textual comprehension. As reader-observers build arguments in self-reflexion or in discussion with one another, previously revised, suppressed, or obsolete information might resurface, as elements that were misleading or had to be reinterpreted are reconsidered. In the examples from Amazing Spider-Man 47 and many others, we find cues for such revision in the three-part temporal structure and the anaphoric pointers in demonstrative gestures and reverse movements, coaxing us towards just such a recognition. Arguments that evaluate the plausibility of such revisions imply different logical orders for when each inference is reached. Those assumptions need not be congruent with (but must always remain interested in) direct neurological or cognitive activity. Instead, the sequence of interpretations that they conceive belongs to the logical progression of the on-going reconstruction of plausible semioses. It is here that concepts of hermeneutics, rhetorics, and poetology provide useful descriptive as well as some normative concepts.⁶

Such an introspective and individual, or collective and public, re-negotiation of interpretation is, of course, possible for almost all kinds of perception. But what makes it especially relevant for comics – among several other fields of media studies – is the fact that artistic products are less reliably bound to the validity of assumed and even demonstrably employed cognitive rules than are most uses of verbal, spoken or written language. The established cases of exceptions in lingual communication belong to verbal art: poetic and rhetorical language use routinely invites us to re-examine the interpretation of syntax as well as semantics in the light of later textual elements. As Thierry Groensteen has famously argued, comics studies is on a fundamentally different footing compared to traditional linguistics because the regularities by which panels are arranged into larger units are always already artistic (Groensteen, 21–23; see Packard 2018, 55, for a previous discussion in CLOSURE). This sets comics apart from verbal art, where the formalist tradition suggests that poetry is as grammatical as standard language, only more so, with the ›poetic function‹ being tantamount to a dominance of the ›grammatical function‹ (Jakobson, 25). In this view, what binds the elements of language together is equally what binds the elements of a poetic text together even more closely. But Groensteen believes that there is no standard, non-exceptional rule for associating sequences of panels with one another, so that any case of ›iconic solidarity‹ consists of an artistic intervention. Instead, all cohesion among comic panels has to be considered as ultimately heautonomous, with sequential art creating rules for its own use within the confines of individual pieces of art, genres, or the art form as a whole. Beyond the artistic level on which Groensteen focuses, such heautonomy may also be argued to hold for comics as a whole, if we take the claims by Dick Higgins and others seriously, who include the art form among those ›intermedium‹ expressions for which any clear rule of interpretation is missing, so that the audience is encouraged, entitled, and forced to create their own rules in interaction with the semiotic affordances of the material. However, researchers should remain attentive to any poetic rules or rhetorical devices introduced below the level of the art form or its various genres, i.e., as devices whose semantic aura is created by the repeated or emphasized use in even just one work of art. Thus, heautonomies take hold on at least three different levels, for comics altogether among all intermedia, for comics as a specific art form as well as for its various genres, and within the confines of each artwork.

But does any of this hold true in light of recent research? Given the experimental evidence in favor of a universal grammar of visual languages to be re-differentiated in different local and historically bound traditions, one might be tempted to just reject Groensteen’s premise altogether. Re-arranged panels from a Peanuts strip, for instance, appear more or less acceptable to readers in new sequences even independently from the semantic coherence of recurring referents and the consistency of the narrative conveyed, as Cohn et al. (2012) demonstrated early on. This clearly suggests a category of grammatical acceptability that mirrors that famous example from Chomsky’s concept: »Colorless green ideas sleep furiously« (15), in which a sentence is recognized as syntactically correct even though it is semantically incomprehensible. In other words, malformed comics panel sequences can violate rules of cohesion and rules of coherence separately from one another. Therefore, if the structures underlying those effects can be unified as a langage beyond their special traditions or in langues and even ›dialects‹, such as those subsumed under a specific Japanese visual language (Cohn and Ehly), does that not allow for some version of a universal grammar for said visual language? Perhaps the addition of poetic rules within each art work remains imaginable, but how can we still argue that comics are an intermedium for which no definite rules of interpretation hold? How can we argue that iconic solidarity has no general structure on which to build special artistic effects?

The crux of my argument is that these two positions do not necessarily contradict one another. The fact that a language structure underlies all iconic solidarity neither entails that there is one standard solution or degré zéro to serve as a foil for every artistic device, nor that discursively explicable revision cannot engage with the cues that allowed the original inference to take place in ways that re-examine the various choices made by the artists and writers involved in shaping their communication. For the example discussed above, the fact that the ›Marvel Method‹ implies a two- or three-tiered temporal structure neither implies that we can uniquely reformulate a panel using that structure in one unequivocal ›standard‹ shape in which it would have appeared in some kind of normalized visual lingual code, the way we might imagine a verse turned to prose or a sequence of alliterations and rhymes removed to leave a more prosaic word choice; nor that this three-tiered structure has no further hermeneutic implications. As Saussure famously argues (45), the fact that onomatopoeia cannot be overlooked in many lexicalized terms such as ›cuckoo‹ and ›shatter‹ neither entails that lexicalization makes no further difference – ›cuckoo‹ is not the same as ›Kuckuck‹ –, nor does lexicalization conversely eradicate the onomatopoetic effect – whoever says ›cuckoo‹ cannot help but invite comparisons to the bird’s call they just mimicked. And even where we find a more specific rule, such as the placement of the antagonist in binary panels with the protagonist in early Amazing Spider-Man comics, the existence of the rule is neither sufficient to force the shape of a panel, nor does the violation take the panel outside the realm of properly shaped graphic sequences. Does the panel engage in flaunting a grammatical rule to convey its surprising meaning? Crucially, that question has no clear answer, because there is no culturally accepted point of reference for the explication of the rule thus violated that would allow the violation and hence its meaning to be interpreted as unequivocal. Visual language structures have to be sufficient to allow for the comprehension of a sequence, but that does not leave us without ambiguity as to when standards, rules, or even necessities of that visual language have been disregarded. Perhaps such more strictly binding standards might be possible even for visual language, and perhaps some dialects of verbal language also lack such standards; the difference, then, might well be entirely cultural, in a restrictive sense of that word. But such is the cultural situation in which mainstream comics – including Amazing Spider-Man – currently exist. They make good use of the opportunities this lenience affords them.  
 
Counting Alligators: Inferential Revision in Panel Domains

This might become even more obvious once we move away from the previous example, which might arguably be confined to the lowest level of a heautonomy, i.e., that of the special self-referential rules in individual pieces of art, which might not be subjected to any kind of normalization according to a more or less universal visual language. Consider instead this page (fig. 4), depicting the early stage of another confrontation between Spider-Man and a supervillain. It is taken from Amazing Spider-Man 6, written by Stan Lee, but realized by the original artist Steve Ditko (15). Here it is the Lizard that threatens Spider-Man’s life. As before, Spider-Man is put in a position to observe and comment upon the movements of his antagonist; but contrary to before, this does not interrupt the flow of the antagonist’s movement, nor does a surprising effect dominate the nevertheless ingenious work with panel structures. Instead, Spider-Man’s position is well justified, as he is carefully approaching the Lizard’s hideaway in the swamp. The surprising development begins with Spider-Man losing control in panel 4, just after the mid-point of the page, slipping from the wall of the ruin he has just scaled, only to be discovered by the Lizard and his alligators.

The referential development on this page creates a nicely symmetrical effect and culminates in an ostentatiously binary panel, with Spider-Man on the left side and the Lizard on the right side of a middle axis, emphasized by the edge of the wall that insufficiently hides Spider-Man. Leading up to this iconic scene, two mirrored triangles assign both characters specific positions in the overall layout: Spider-Man dominates panel 1 on the left and panel 4 on the right before being taken up in the left half of the larger panel 5, whereas the Lizard dominates the complementary panels 2 on the right and 3 on the left before appearing in the right half of panel 5 on an (almost) equal footing. That symmetry is broken by the very first appearance of the Lizard in the background of panel 1, as well as the smaller figure of Spider-Man in the far background of panel 3. In panel 1, the Lizard is introduced as the object of Spider-Man’s and the observer-reader’s gaze, while the narrator’s comment aligns them by telling us that Spider-Man reverts to his alter ego as photographer Peter Parker to evaluate the page as pictures: »The scene is so amazing, so fraught with drama, that Spider-Man takes a few fast pictures of it […]!« As Spider-Man’s civilian identity, Peter Parker, works as a photographer for the Daily Bugle, here his position is likened to one who aesthetically contemplates rather than engages in the scene presented. Equally, the Lizard is framed as an object to be studied and commented upon, much as Kraven was in the ›bad grammar‹ panel. But this time, the commentary just reinforces that the pictorial sequence is well-formed, dramatic, and poignant, rather than criticizing the villain’s words or motions.

Depending on whether an interpretation begins with the narrator’s caption and the dialogue, or first focuses on the dominant elements in the pictorial sequence, the Lizard’s appearance in panel 1 might either be obvious from the start or need to be re-discovered once the second panel brings his shape into dominance. The dialogue in the second panel fittingly comments upon just such a reversal: »We shall be the first!«, the Lizard exclaims from the position of his appearance in the second place, announcing an evolutionary reversal for himself and his alligators. Following this declaration, the perspective continues to move around his body in subsequent panels, moving him to the left side in panel 3 before Spider-Man’s reintroduction from the left in panel 4 flips the scene around again. Note that, diegetically, the Lizard must have turned around himself, as the first panel has Spider-Man looking at his back, whereas we look at the Lizard’s back in the third panel when he starts moving away from us and towards Spider-Man’s supposedly unchanged position. Even then, it is difficult to reconcile Spider-Man’s hanging on to a wall on his left in panels 1 and 4 with the point from which he seems to be observing the Lizard in panel 3; or to precisely locate the lower ground he has fallen onto in panel 5, given the Lizard’s scaling of steps up to that space in panel 3. Clearly, the consistency of information about the storyworld does not rest on any strictly upheld topography. Instead, the spatial information given has a coherent discursive function supported both by the general rules of cohesive sequential chains of reference and the artful symmetrical layout, placing Spider-Man and the Lizard in operatively different parts of each panel as the sequence evolves from the Lizard’s introduction as a framed object of observation, to his claim of dominance and primacy, through to Spider-Man’s now symbolically fraught fall to a position lower than the reptiles, from which the fight will ensue on the next page.

That discrepancy is possible because the Lizard’s space, while diegetically clearly positioned in this ruin within this swamp, is constructed quite differently in the aesthetics of the panels. The diegetic placement is already ambiguous, as the ruin marks the center of the Lizard’s realm, but his character is constructed around a theme of nature disrupting culture, such that it becomes Spider-Man’s place to observe the Lizard from the ruin of threatened human architecture, whereas the Lizard moves around outside of it. In contrast to this, the space that houses the Lizard within the panel aesthetics is well defined, not by the walls or even the swamp but by the alligators that surround him.

Their unusual treatment on this page is best explained from a point of view of panel segmentation into several domains. Going back to Krafft’s strongly hermeneutic but structurally linguistic theory of a comics grammar and many further efforts towards a structural analysis of panel sequences, we find a number of theories that distinguish several domains into which the elements in each comics panel may be subsumed (cf. Packard 2023 for a detailed recent model). The domains mediate between semantic expectations of coherence and distributive functions for sequential cohesion, as they are motivated both by the semantic content relevant to each panel and by recurrences and referential shifts across panel sequences. One such domain traditionally encompasses what Krafft refers to as ›Raumzeichen‹, spatial signifiers that define a middle ground between the depiction of salient actors as – usually strongly cartoonized – signifiers of actors or ›Handlungszeichen‹ and the indeterminate background of shapeless or merely geometrically and colorfully decorated panel foils or flat areas inside panel frames. As Krafft was perhaps first to notice, among the two recurring structural differences between actors’ and spatial signs is, first, that the former will usually have a definite, unbroken circumference, such that the panel outline is less likely to cut their bodies apart and that, whenever they do so anyway, the closure of McCloud’s first order, i.e., the imagination of the missing parts, is easily accomplished and unequivocal due to the well-established gestalt of their body shapes. Secondly and by the same measure, actors’ signs are countable: we know that there is one Spider-Man and one Lizard, two agents Dupond and Dupont in Tintin, three nephews for Donald in the Duck comics – and, in any case, the consistency of a scene requires that observer-readers may keep track of the pertinent actors. In contrast to this, spatial signifiers may depict a forest with shifting and changing amounts of trees, a meadow with uncountable blades and bushels of grass, and even, as in the case under consideration, a building with some unclear borders, walls, windows, stairs, and levels. Krafft goes on to describe the cohesion of a sequence by pointing out the necessity to repeat actors’ signs throughout a panel sequence, and the possibility to signal the beginning of a new sequence by setting up new actors to characterize new scenes. Spatial signs, on the other hand, are reducible up to the point of complete omission, constituting the oft-discussed dropped backgrounds in comics panel sequences (cf. e.g. Edlin and Reiss). Conversely, new elements can appear in these spaces merely as a change in depiction or focus, without signaling their sudden appearance in the storyworld. An additional tree, blade of grass, or even wall or window is acceptable as new information about unchanged objects. Spatial domains in panels are productive in this sense (cf. Packard 2006, chapter 5), as they bring forth new elements in the progression of a panel sequence. In both senses, spatial domains in comic panel sequences are depicted by ›open signifiers‹ (Krafft; Packard 2015).

Semantically, we would expect the alligators to be depicted as actors’ signs. Spider-Man clearly thinks of them as independent agents when he considers their potential threat: »[…] huge alligators! But they seem to be obedient to him!« The Lizard addresses them as his »pets« in panel 1, includes them in the first-person plural in panel 2, exhorts them to follow him in panel 3, and orders them to fight in panel 5. However, even these lingual representations allow us to dismiss an exact count of the animals: the Lizard is accompanied by several alligators, but how many there are exactly remains unclear. The pictorial realization treats these beasts differently than it does the actors’ signs for Spider-Man and the Lizard. Even though the circumference of their bodies and the distinct coloring set the alligators apart, their exact number remains unclear. On the following page (16, fig. 4), it is also their exact position that becomes impossible to determine, as their tails hit the walls of the ruin from what, in panel 4, seems to be not just in front of and behind, but also underneath and above Spider-Man’s position. Even on the previous page, there is a tendency to have the alligators’ bodies cut off by panel borders.

That this treatment of the alligators is possible is easy to explain in terms of Krafft’s constructive rules for panel sequences. One might equally argue that a concept of relevance automatically reduces the amount of alligators from any specific number to ›several‹ (cf. Forceville).⁷ However, the salience of this implementation of the animal depictions in the panel and sequence structure becomes clearer when we consider how the association between the panel elements and panel domains is subject to inferential revision in the course of this page. What might appear to be an objectified image to be gazed upon in panel 1 and would most likely be expected to appear as actors’ signs in the progression of the scene, is actually employed as a flexible denotation of a space that surrounds the Lizard, travels with him, and ultimately tears down the more readily recognizable spatial arrangement of the architectural ruin. The alligators are to the Lizard what the ruin is to Spider-Man. Thus, the number of actors is reduced to the binary logic of protagonist and antagonist, and the space of the binary panel at the end of this page (fig. 3) that sets up their fight actually contains two, not one, semantically loaded spaces: the house from which Spider-Man falls, and the realm of alligators with which the Lizard ascends.

Hands and Faces in a Semiotic Anatomy of Comics

If we want to continue the hermeneutic bent of this approach, some validation might come from the title of each episode. The issue in which Kraven the Hunter surprisingly fails to connect his hand to Spider-Man is titled, »In the Hands of the Hunter!« The earlier issue in which Spider-Man fails to merely observe and is instead directly confronted with the Lizard is titled, »Face-to-face with… the Lizard!« But the validation of such a hermeneutic approach should not be based on such far too easily found (and cherry-picked) clues. The point of a close hermeneutic analysis of the cohesion and coherence across panel sequences must not be to disregard visual language in favor of an unbound freedom of free association. Its usefulness, however, may lie in detailing the engagement with the rules of visual language as the inferential revision of each syntactic cue on the page may play a role for the re-interpretation of a comic’s meaning in the constant process of reconsidering and debating the indications that rely upon, but also play with the grammar of comics to allow for their poetry and rhetorics, summarized in an equally defined and productive semiotics of comics pages’ structures and elements.

 

_______________________________________________________

Bibliography

• Bateman, John A., and Janina Wildfeuer: Linguistically-Oriented Comics Research in Germany. In: The Visual Narrative Reader. Ed. Neil Cohn. London: Bloomsbury, 2016, p. 19–65.
• Bateman, John A., and Janina Wildfeuer: Defining Units of Analysis for the Systematic Analysis of Comics. A Discourse-Based Approach. In: Studies in Comics, 5.2 (2015), p. 373–403.
• Booth, Wayne C.: The Rhetoric of Fiction. Chicago: University of Chicago Press, 1961.
• Chomsky, Noam: Syntactic Structures. The Hague and Paris: Mouton, 1957.
• Cohn, Neil, Martin Paczynski, Ray Jackendoff, Phillip J. Holcomb and Gina R. Kuperberg: (Pea)nuts and Bolts of Visual Narrative. Structure and Meaning in Sequential Image Comprehension. In: Cognitive Psychology 65.1 (2012), p. 1–38.
• Cohn, Neil: The Visual Language of Comics. Introduction to the Structure and Cognition of Sequential Images. London: Bloomsbury, 2013.
• Cohn, Neil: You’re a Good Structure, Charlie Brown. The Distribution of Narrative Categories in Comic Strips. In: Cognitive Science, 38.7 (2014), p. 1317–1359.
• Cohn, Neil, and Marta Kutas: Getting a Cue before Getting a Clue. Event-related Potentials to Inference in Visual Narrative Comprehension. In: Neuropsychologia, 77 (2015), p. 267– 278.
• Cohn, Neil, and Sean Ehly: The Vocabulary of Manga. Visual Morphology in Dialects in Japanese Visual Language. In: Journal of Pragmatics, 92 (2016), p. 17–29.
• Cohn, Neil, and Marta Kutas: What's Your Neural Function, Visual Narrative Conjunction? Grammar, Meaning, and Fluency in Sequential Image Processing. In: Cognitive Research: Principles and Implications, 2.27 (2017), p. 1–13.
• Cohn, Neil, and Joseph P. Magliano: Visual Narrative Research. An Emerging Field in Cognitive Science. In: Topics in Cognitive Science, 12.1 (2019), p. 197–223.
• Dutke, Stephan, and Ulrich von Hecker: Comprehending Ambiguous Texts. A High Reading Span Helps to Constrain the Situation Model. In: Journal of Cognitive Psychology, 23.2 (2011), p. 227–242.
• Eckhoff-Heindl, Nina: Comics begreifen. Berlin: Reimer, 2023.
• Edlin, Lauren, and Joshua Reiss: Measuring Inter-subjective Agreement on Units and Attributions in Comics with Annotation Experiments. In: Zeitschrift für Semiotik (2023), special issue, ed. Stephan Packard and Janina Wildfeuer, forthcoming.
• Evans, Julia L., Craig Selinger and Seth D. Pollack: P300 as a Measure of Processing Capacity in Auditory and Visual Domains in Specific Language Impairment. In: Brain Research, 1389 (2011), p. 93–102.
• Forceville, Charles: Visual and Multimodal Communication. Applying the Relevance Principle. New York: Oxford University Press, 2020.
• Frahm, Ole: Die Sprache des Comics. Hamburg: Philo Fine Arts, 2010.
• Groensteen, Thierry: Système de la bande dessinée. Paris: Presses Universitaires de France, 1999.
• Groth, Gary: Editorial. In: The Comics Journal, 75 (1982), p. 4.
• Higgins, Dick: Intermedia. In: Something Else Newsletter, 1.1 (1966), p. 1–3.
• Iser, Wolfgang: The Implied Reader. Patterns of Communication in Prose Fiction from Bunyan to Beckett. Baltimore: Johns Hopkins University Press, 1974.
• Jakobson, Roman: Linguistics and Poetics. In: Selected Writings. III: Poetry of Grammar and Grammar of Poetry. Berlin: deGruyter, 2010, p. 18–51.
• Kendeou, Panayiota: Validation and Comprehension. An Integrated Overview. In: Discourse Processes, 51.1.2 (2014), p. 189–200.
• Kendeou, Panayiota and Edward J. O’Brien: The Knowledge Revision Components (KReC) framework. Processes and Mechanisms. In: Processing Inaccurate Information. Theoretical and Applied Perspectives from Cognitive Science and the Educational Sciences. Ed. David Rapp and Jason Braasch. Cambridge: MIT, 2014, p. 353–378.
• Krafft, Ullrich: Comics lesen. Untersuchungen zur Textualität von Comics. Stuttgart: Klett-Cotta, 1978.
• Lee, Stan (W), Steve Ditko (A): Spider-Man: Face-to-face with… the Lizard! (Amazing Spider-Man 6). New York: Marvel, 1963.
• Lee, Stan (W), John Romita (A): Spider-Man: In the Hands of the Hunter! (Amazing Spider-Man 47). New York: Marvel, 1967.
• McCloud, Scott: Understanding Comics. The Invisible Art. New York: Harper Collins, 1994.
• McNamara, Danielle, and Joseph P. Magliano: Toward a Comprehensive Model of Comprehension. In: Psychology of Learning and Motivation. Advances in Research and Theory, 51 (2009), p. 297–384.
• Packard, Stephan: Superior Unreliability. Thoughts on Narrators in Comics on the Occasion of Spider-Man 2012/13. In: Comics Forum. Ed. Ian Hague.
  <https://comicsforum.org/2013/08/12/superior-unreliability-thoughts-on-narrators-in-comics-on-the-occasion-of-spider-man-201213-by-stephan-packard>. 12 Aug, 2013. Accessed 1 Feb, 2023.
• Packard, Stephan: Anatomie des Comics. Psychosemiotische Medienanalyse. Göttingen: Wallstein, 2006.
• Packard, Stephan: Closing the Open Signification. Forms of Transmedial Storyworlds and Chronotopoi in Comics. In: Storyworlds, 7.2 (2015): Transmedial Worlds in Convergent Media Culture, ed. Jan-Noël Thon, p. 55–74.
• Packard, Stephan: ›37-mal die Welt gerettet‹. Eine kleine Typologie der Selbstreferenz und Reflexivität sowie ihrer Verwechslung im Comic, auch anhand Chris Ware und Tom King. In: Closure. Kieler e-Journal für Comicforschung 4.5 (2018), p. 47–61. <http://www.closure.uni-kiel.de/closure4.5/packard>. 10.05.2018.
• Packard, Stephan: Cohesion in Panel Graphs. A Psychosemiotic Approach. In: Zeitschrift für Semiotik (2023), special issue, ed. Stephan Packard and Janina Wildfeuer, forthcoming.
• Pérez, Ana, Kate Cain, María C. Castellanos and Teresa Bajo: Inferential Revision in Narrative Texts. An ERP Study. In: Memory and Cognition, 43 (2015), p. 1105–1135.
• Phelan, James: Living to Tell about It. Ithaca: Cornell University Press, 2005.
• Poynor, David V., and Robin K. Morris: Inferred Goals in Narratives. Evidence from Self-Paced Reading, Recall, and Eye Movements. In: Journal of Experimental Psychology. Learning, Memory, and Cognition, 29.1 (2003), p. 3–9.
• Rapp, David N., and Panayiota Kendeou: Revising What Readers Know. Updating Text Representations During Narrative Comprehension. In: Memory and Cognition, 35 (2017), p. 2019–2032.
• Saussure, Ferdinand de: Cours de linguistique générale. Ed. Charles Bally and Albert Sechehaye. Paris: Payot, 1916.
• Shen, Dan: Unreliability. In: The Living Handbook of Narratology. Ed. Peter Hühn et al. <https://www-archiv.fdm.uni-hamburg.de/lhn/node/11.html>. 31 Dec, 2013. Accessed 1 Feb, 2023.

Table of Figures

• Fig. 1: Amazing Spider-Man 47 (1967, 15).
• Fig. 2: Amazing Spider-Man 47 (1967, 13).
• Fig. 3: Amazing Spider-Man 47 (1967, 15).
• Fig. 4: Amazing Spider-Man 6 (1963, 15).
• Fig. 5: Amazing Spider-Man 6 (1963, 16).

 

• 1]   This discussion expands and corrects a very brief engagement with this panel in my earlier attempt at a psychosemiotic theory of comics (Packard 2006, 112–113).
• 2]   Here and in the following, I conceive of comics’ audiences as observer-readers in line with Nina Eckhoff-Heindl’s concept of the ›lesend-betrachtenden‹.
• 3]  Cf. the excellent overview, at least for German comics studies, in Bateman and Wildfeuer (2016).
• 4]   One instance is found in the Call for Papers for ComFor’s 2021 annual conference.
• 5]   For instance, the – ingenious – treatise by Frahm, purporting to describe the language of comics‹  – Sprache des Comics – demonstrates and then subverts, but still continuously relies on this trope.
• 6]  Together with the linguistic conceptualizations, they might be summarized in a general pragmaticist semiotics; cf. Packard 2023, forthcoming.
• 7]   My thanks to Charles Forceville for pointing this out at ComFor’s annual conference in 2021.