Online Magazine of the Visual Narrative - ISSN 1780-678X

Home

Editorial Board

Policy

Issue 1. Cognitive Narratology

Narration in Comics

Author: Pascal Lefèvre
Published: August 2000

Abstract (E): Cognitive narratology states that the reader arrives with schemata, cognitive frames for a meaningful organisation of various interrelated concepts, based on previous experiences. However, the reader is also cued to perform a specific activity by the artwork. Reading is a dynamic and continuous process. Models, the basis of the reader's schemata, assumptions, inferences and hypotheses, are set up by both extrinsic and intrinsic norms. Comics readers use similar constructive procedures as in constructing reality from real-life perceptions, and narration should be understood as the organisation of a set of cues used in constructing a story. The close reading of a Flemish 1947-comic strip shows that even such a simple one-page gag can implement various strategies and formal organisations besides pure narration. There are many aspects and elements without any strict narrative function in comics. The whole of a narrative is far more than the sum of its events.

Keywords: narration, Vandersteen, schemata

1. Introduction

During the last 20 years a small but growing number of narratologists (Bordwell, 1985; Reid, 1992; Fludernik, 1996; Herman, 1997) have stressed the role of the reader and the way a 'text' engages this reader in semantic exchanges. In the same way I shall focus on the way comics cue the reader to a specific reading through their form and how readers are able to activate relevant categories of knowledge, with or without explicit textual clues to guide them (Herman, 1997). Certain concepts from cognitive psychology might help us with the description of a reader's understanding of a graphic narrative. In his interesting book Moving Pictures, Torben Grodal (1997) explained the close links between narrative and cognitive-emotional activation. He argues that mental models and image-schemata can provide more general descriptions of the cognitive and emotional role of the human body than psychoanalysis or psychosemiotics can (Grodal, 1997: 282). In this study I will try to show some of the assumptions, inferences, hypotheses, schemata the reader needs to activate a comic and how a comic can cue the reader to perform a specific activity. From this perspective comics don't differ that much from other media as film, novel, theatre... Besides, I shall also briefly analyse how comics can deviate from the proposed definition of narration, where narration begins and where it ends.

My definition only describes the most essential characteristics of narration. For me, the core definition of a narrative is: 'A formal system that the reader interprets as a interesting representation of a series of logically and chronologically related events, caused or experienced by actors.' It is a formal system because it is a set of elements that depend on and affect one another. Form is the overall system of relations the reader perceives in a comic. The reader has access to the formal system of the comic through his visual sense. In comics (or other narratives) meanings are constructed by the reader's interpretation of the formal system (drawings and texts). His unifying or framing act of perception and interpretation activates the comic. The events themselves are not included in the narrative, only the way they are represented (drawn). If the reader does not perceive the representation as interesting, he would not continue his interpretation and reading activity. An event can we define as something which happens or develops over a period of time. Usually the narrative includes beginning and end of this development, but they are in any case implicitly contained. Previous events are seen as causes for later events; what happens first has consequences for what happens later (this implies a chronological structure). And actors can be defined (Bal, 1997: 5) as agents who perform actions (causing or experiencing an event). Actors are not necessarily human, they can be animals, robots, supernatural creatures, extraterrestrials...

This definition of a narrative excludes pure constatations, descriptions, ideas, images, reasoning, arguments, catalogues, etc. However, these could become narrative if they included the necessary elements. Still, a comic (or another form of narration) does not consist solely of narration (Groensteen, 1988). Narrative comics sometimes feature sequences where the narration is of secondary importance. Such sequences then display other, more salient characteristics, like categorical or poetic ones. Non-narrative elements can be incorporated in the comic and they can be of some importance for the narrative.

I will test my definition in close reading of a specific page from "De Familie Snoek" (The Pike Family), a Flemish one-page gag comic from 1947 by Willy Vandersteen, who later became the best-selling comics artist in the Low Countries.

I thank Tom Pijnenborg for his help on the English translation and Jan Baetens for his valuable comments on the first draft of this paper, which was first presented at ICAF, September 1999.

2. Reading a comic

Firstly, I will disregard the interpretation of separate panels, because I want to focus on the sequential aspect of comics, and more specifically their narrational qualities. However, I certainly will not neglect the fact that, since several events can be represented in one panel, a single image can be narrative according to my definition. In that case the sole panel consists of several virtual panels or frames as in some photos (a.o. Marey) or in paintings (a.o. "Nude Descending a Staircase No. 2." by Duchamps). But also in the domain of the comics there are countless examples wherein a single panel seems to contain a representation of a cause and an effect, of several distinct positions of a moving thing or character. But the reader has always to divide mentally what is given as one. The readers adds mentally frames inside the panel.

The Family Snoek, "With all modern conveniences"

Secondly, I reject any notion of a strict separation between form and content; in my opinion form is more than an empty and coded container for content (Lefèvre, 1999). Although verbal description can never grasp exactly what one sees in an image, I will, however, for the sake of convenience, assume that one can give an approximate verbal description.

An image can show actions or objects, and the reader usually has no difficulties distinguishing between both. The reader also accepts the notion of the frame. He assumes that the fictive world does not end at the border of a panel. He also accepts the intern 'hors champ'; for instance, when the stairs leading to the basement are not depicted a third time, the reader does not interpret this as if the stairs disappeared, that the world of this comic is unstable and volatile.

In fact, the reader does not come unbiased or innocent. He arrives with schemata, which are cognitive frameworks for the meaningful organisation of various interrelated concepts based on previous experiences (Sternberg, 1996: 508). A schema is a configuration of knowledge about objects and events, including general information. The schema expresses typical information, not the unique features of a specific thing, situation or event (Haberlandt, 1994: 147). A special kind of schemas are scripts. A script is a structured representation describing a stereotyped sequence of events in a particular context. It is a structure for a schema involving a common understanding about the characteristic actors, objects and sequence of actions in a stereotypical situation (Sternberg, 1996: 508). Although the exact workings of these schemata are not completely understood, I believe there is some truth in the concept. More and more narratologists are applying this cognitive concept to their studies of narration. Monika Fludernik makes an important distinction between story schema (the fictional world) and discourse schema (telling). She does not entirely reject the story vs discourse distinction, but relegates it to those parameters which depend on a realistic cognization of both the story-world and the narrational act (1966: 336). At the same time, we can use also Bordwell's distinction (1985: 150-153) between external and internal schemata. Paraphrasing Bordwell, we could say that the reader applies the external schemata to the comic, matching the expectations appropriate to the norms with their fulfillment within the comic. Greater or lesser deviations from these norms stand out as prominent. But at the same time, the viewer is alert for any norms set up by the comic itself. The intrinsic norms may coincide with or deviate from the conventions of the extrinsic set. Finally, the reader may encounter foregrounded elements the moment the comic diverges to some degree from intrinsic norms. "In a sort of feedback process, these deviations may then be compared with pertinent extrinsic norms." (Bordwell, 1985: 153) Throughout this process, both intrinsic and extrinsic norms set up models, which form the basis of spectators' schemata, assumptions, inferences, and hypotheses.

It seems reasonable that comics readers use similar - or even identical - constructive procedures as they do when they construct reality from their real-life perceptions (Grodal, 1997:29). For example, a reader has no difficulties discerning the important actors from the background (décor and supernumerary figures). It is quite similar to everyday life where one for example, does not expect a house to become a character.

There are, of course, striking differences with real life. First of all, the pictures of a comic consist only of static lines and colors on a two dimensional plane. The artificial nature of the images strikes the viewer, but usually doesn't hinder an easy interpretation. The various projection and denotation systems (Willats, 1997) express different kinds of truths. The artificial nature of the images makes it possible that fiction is not bound by the same physical laws as our reality is. Deviations (e.g. the powers of Superman) are not problematic for the reader, because he knows it is make-believe. There are devices to make the weirdest things acceptable in fiction. It is the degree of coherence of the fictive world (e.g. its own set of physical laws) which is crucial for the acceptance by the reader. Or as Fludernik (1996: 316) confirms: "Realism is always an effect of interpretation, but this effect can be thwarted by radical inconsistencies in the represented fictional world." In comics some inconsistencies, for example in the representation of space, are not even perceived or considered as troubling by the reader.

Intrinsic norms delivered by the comic can construct models which form the basis of readers' schemata, assumptions, inferences and hypotheses (Bordwell, 1985: 153). The world of a comic needs not to be static, it can change over time. However, if changes are too abrupt, readers will get confused and loose interest in the story. In a traditional comic the fictive world can be incoherent as long as the narrative concept is respected. In our example of "De Familie Snoek" the bathtub is not drawn exactly the same. The second time it is depicted, tiles are up to the edge, while in the other cases the tub has a border. It is a detail, which most readers won't even notice. Even when a reader sees this change in the diegetic world, it would not necessarily undermine his believe in a stable diegetic world. Maybe he looks for some external reason as the author's lack of carefulness for such details.

The reader has to accept that the arrangement of the panels on the page is not random, but directed, and that the panels are interconnected. They form a sequence of successive situations.

According to cognitive psychologists our interaction with the environment and with other people consists of sequences of events: "Usually we have no difficulty understanding and participating in events, and we know the likely consequences of actions. We are also familiar with the objects and the roles of participants in typical events (...)" (Haberlandt, 1994: 144-145). If we are already familiar with sequences of events in our daily life, we probably should have no problems understanding representations of such sequences in comics. The only problem could be the way a sequence is represented: some representations ask more imagination from the viewer. And there are also other differences. While sequences seem to be continuous in daily life, this is not possible in comics. Comics are by necessity an elliptic medium. Unlike films or stage plays, comics can only represent fragments of events (and objects). A comic has to suggest a whole sequence of events by representing only some single significant actions. In our case study, we see the artist did not show all phases of simple events like Leonard stopping his work in the cellar, climbing the stairs and walking in the hall. Still, the reader can easily interpret the whole on the basis of its similarities to known sequences in daily life.

A comic showing all successive phases in detail could be quite boring. The traditional reader prefers the story to move forward, so most actions are limited to the essential phases. Showing Leonard returning to his work in the cellar would be redundant. The reader understands Leonard returned, even though this action is not depicted. Several events are systematically left out: actions like going down to the cellar, climbing the stairs and opening the bathroom door. These are implied in the events shown. Although there is no real evidence that the 'missing' events occured in the diegetic world, most readers just assume on the basis of what is shown that they happened. (Experimental comics can play with such assumptions.) Furthermore, the story itself is but a fragment, while daily life continues until death. Because of the resolution situated in the last panel, there is no internal need to prolong the story, even though Leonard Snoek will certainly reappear in the next episode.

There are a lot of basic conditions (shemata) needed to read a comic. A crucial shema to understand a comic is the elliptic and fragmented nature of the medium. The history of comics is partly a history of the refinement of elliptic and fragmented storytelling.

An extrinsic norm crucial to comics is the interpretation of a figure reappearing in several panels as one and the same figure shown at different moments in time (usually in chronological order). Still, an artist can subvert this convention: think of the typical games with disguises, masks and doubles in detectives, or McCay's experiments, Fred's multiplication of a single character.

Chronological order enables causal inference, the conclusion that something (an event) causes something else (Sternberg, 1996: 498). The reader expects cause-effect relations between successive events. Usually it is assumed that the event represented in the second panel happens after the event represented in the first one. In the first panel we see a character at work, complaining. In the second panel the same character is shown walking to the right, probably in a different room. He certainly stopped working. The caption ("I'm fed up. A nice hot bath is what I need!") refers to his work and indicates it is done. Although we can not be absolutely certain that this panel follows the first one chronologically (the second panel could depict an earlier event) or that the location is still the same (the house), we would expect deviations from these conventions to be clearly indicated. We assume the comic respects the extrinsic norms of traditional narration. Of course, there is a lot of trust involved in our acceptance of conventions, and artists can play with such expectations. In Vaughn James' The Cage there are, however, too few indications of time to be certain of any chronological order and we can only guess at possible sequences. In that case the traditional scripts - structured representation describing a stereotyped sequence of events in a particular context - are of no use. There is no apparent actor in The Cage, only buildings and objects in various situations. Since we can not reassemble the chronological structure, causal inference is nearly impossible. The comic urges us to adapt to a new situation and question our reading habits. It forces us to look for parallels and rimes (Vidal, 1986: 190-193).

3. Case-study "De Familie Snoek": narrative understanding

Relying on basic skills and expectations, the average reader should not have any serious problems in understanding what is going on. The images and the text give us all the information needed. They show the most important actions in their chronological order and with causal relations. Leonard returns to work each time from sheer necessity, because there always is somebody using the bathroom. His return to work, reinforcing the mounting tension, is almost the only possibility. He is too dirty to sit on a sofa, and the gag would fail if he went to the kitchen and relax with a glass of beer. Finally, Leonard is forced to find another solution (boiling the water in a kettle), because the gas runs out

Since we know the different characters from previous stories, we can better understand the underlying relations and we recognize the woman introduced in panel 9 as Leonard's wife. From the successive frustrations we understand how the tension is rising and why Leonard explodes emotionally in the penultimate panel. It is not as much the objective problems (occupied bath, no gas), but the growing determination - even obsession - of the character and the multiplying obstacles which cue the humor.

The last panel only makes sense considering Leonard's strong determination to take a bath with hot water. At the same time the use of the small sink as a bathtub is ridiculous. However, any other solution would not have been funny; e.g. Leonard boiling water in a kettle to fill the bathtub, or going out for gas. Moreover those solutions would have taken more than one panel (or a lot of explaining text). Using the sink is the humorous solution, demonstrating once more Leonard's determination to have it his way.

Some of the schemata which help the reader understand this comic:

- The publication format and the caricatural style suggest a humorous content.

- The schema of a family, and more specific the schema of a middle class family in Belgium at the end of the forties. With this background one can understand why Leonard's wife complains and why he cares so much about what she says. It would not make sense in another social context, where a man, in his own house, giving in to his brother or his son (who is even poking fun at him), would be unthinkable. Like his American equivalents of the family comic strips, Leonard is a henpecked husband who can rarely cope with the situation. The use of this formula is important and stresses the fact that those old roles had not altogether disappeared in 1947. Familiarity with living conditions in Belgium in the forties would allow even better understanding. Most families had no bathtub and still washed themselves at the sink. In 1947 only 56% of the families had access to the public water supply, and only 7% had a separate bathroom or shower (Matthijs, 1988: 104). Such facilities generally belonged to the higher classes. Also the use of an electric cooker is quite advanced for its Belgian context and time. Although the Snoek family appears typical middle-class, they are in some aspects quite progressive and modern. Working in his cellar, Snoek ressembles the working class, but his house seems to belong in the (higher) middle class.

- The schema (script) of taking a bath, including a sequence of actions: heating water (if there is no warm water), undressing, going in the water, rubbing with water and soap, getting out of the water, drying the body with a towel. The fact that, in the final panel, Leonard only partly conforms to the schema (script) of taking a bath is funny. For instance, he does not actually wash himself, but just sits in the sink, reinforcing the fact that, in spite of his attempts and creative mind, it is no more than a ridiculous substitute, especially since he wants to act like it is a bathtub. His position is humiliating; naked, neither covered by the tub, water, nor any clothes. His expression of anger clearly shows how he feels about it, and even the kettle is hissing loudly at this pitiable scene.

4. Nonnarrative aspects

Trying to understand a comic the reader can concentrate also on nonnarrative aspects as the formal qualities of the comic. I am aware that disparate readers may perceive quite disparate semantic and syntactic elements in the same comic, since an unbiased and value-free approach is impossible. Signs can generate more than one reading, because they are determined by a large number of factors. Therefore they are called polysemic. As Nick Lacey (1998: 90)shows, readers are at liberty to pick any reading of a text, as long as it can be justified by that text.

Non-linear relations between the two halves of the page:

In our western culture readers are trained to read each page from left to right, top to bottom. The arrangement of the panels on the page and the numbering of each panel (in the upper left corner) are based on this orientation. However, we do not simply read an ordered succession of isolated panels. Like a story, which is more than a simple chain of consecutive events, the page as a whole (a typical unit of a comic's conventional format) invites us to contemplate non-linear relations between the panels. Aspects or parts of any panel can be linked to similar aspects or parts of other panels (Baetens & Lefèvre, 1993: 72; Groensteen, 1999: 173-174).

The formal organisation of the page is intriguing. It is divided in four tiers (strips) of equal dimensions, and the arrangement of the panels is identical in both upper tiers:

- The first panel shows Leonard working in the cellar. This panel has the same position in both strips, as well as the same composition, point of view and framing, and the character is depicted in a similar position at the same location.

- The second panel shows Leonard going to the bathroom. Again, this panel has the same position, point of view and framing in both strips, and the character is depicted in a similar "location" (on his way to the bathroom) and moving in the same direction.

- The third panel shows Leonard in the bathroom, confronted with the occupied bath, and again the panel has the same position, point of view and framing in both strips, and the characters are placed on the same spot, in the same location. (There is some "noise" as well; the length of the panels is not exactly the same and the bathtub is drawn a little bit differently the second time around).

The organisation of the bottom half of the page is somewhat different, but still partly refers to the upper half:

- The same 'orthogonal projection system' (Willats, 1997: 43) is used: all objects are depicted perpendicular to the scene, never oblique or canted, the screen-tone haloes are the same, ...

- The first panel of the third strip resembles that of the upper strips, with the same position, point of view and framing. The character is even depicted in the same location (although the stairs are not shown), but with a change of direction: his head is still turned to the left side of the page, but his body is moving in the opposite direction.

- The second panel of the third strip refers to the last panel of the upper strips, with the same point of view and framing, and showing the same location. Only the character's position has changed, from the side to the center and upside down (flipped, literally and figuratively).

- The last strip with only two big panels contrasts to the other three strips with three panels. Moreover, for the first time the reader could get the impression of a split-panel (eg. the skirting board seems to continue), though there is a suggestion of three gutters in between.

However, the last panels of the page (the last one of the third strip and the entire fourth) are quite new and only refer to the previous ones in that the last two feature the same character:

Halfway down the page there is a remarkable break in the structure introduced in the first two strips. On the narrative level this formal break coincides with the end of the repetitive part and with Leonard's decision not to return to the cellar (obvious from the way he throws his hammer to the ground). This break is stressed further in switching from three to two panels in the final strip.

There are also non-linear relations between background-elements. In panels 3, 6, 8 and 11, the sides of the tub in the bathroom and the wall above the sink in the kitchen are decorated with similar tiles. In the first three panels they occupy the lower left corner of the panel, in the last one they are placed in the upper right corner (above the ground), and they are fewer in number. While the other characters (or at least their heads) are placed above (or behind) the tiled surfaces, Leonard is always positioned beside or in front of them, even in his tub substitute. He can not hide: as opposed to the others, his whole body is visible almost all the time. The similar tiles also act as a formal motivation of the substitution of the kettle (on a small electric stove) for the gas-heater, and of the small kitchen sink for the bathtub. Futhermore, while the tub has two taps (hot and cold water), the sink only has a single tap (cold water only), stressing the contrast between hot and cold, and the reduction of the tub.

5. References

BAETENS, Jan & LEFEVRE, Pascal. 1993. Pour une lecture moderne de la bande dessinée. Bruxelles: Centre Belge de la Bande Dessinée.

BAL, Mieke. 1997. Narratology Introduction to the Theory of Narrative. Second Edition, University of Toronto Press.

BORDWELL, David. 1985. Narration in the Fiction Film. London: Methuen.

BORDWELL, David, & THOMPSON, Kristin. 1986. Film Art, An Introduction, Second Edition. New York: Alfred A. Knopf.

FLUDERNIK, Monika. 1996. Towards a 'Natural' Narratology. London: Routledge.

GRODAL, Torben. 1997. Moving Pictures, A New Theory of Film Genres, Feelings, and Cognition. Oxford: Clarendon Press.

GROENSTEEN, Thierry. 1988. "La narration comme supplément." In: Groensteen, Thierry (ed.). 1988. Bande Dessinée, Récit et Modernité. Paris: Futuropolis & CNBDI. pp. 45-69.

GROENSTEEN, Thierry. 1999. Système de la bande dessinée. Paris: PUF.

HABERLANDT, Karl. 1994. Cognitive Psychology. Boston: Allyn and Bacon.

HERMAN, David. 1997. "Scripts, Sequences, and Stories: Elements of a Postclassical Narratology." In: PMLA, vol. 112, number 5, october 1997, p. 1046-1059.

LACEY, Nick. 1998. Image and Representation, Key Concepts in Media Studies. New York: St. Martin's Press.

LEFEVRE, Pascal. 1999. "Recovering Sensuality in Comic Theory". In: International Journal of Comic Art. Vol. 1, No. 1, Spring/Summer 1999.

MATTHIJS, Koen. 1988. Belgoscope, De Belgen, de Vlamingen en de Walen. Wie ze zijn. Waar ze wonen en hoe ze leven. Tielt: Lannoo/De Financieel Ekonomische Tijd.

REID, Ian. 1992. Narrative exchanges. London: Routledge.

STERNBERG, Robert J. 1996. Cognitive Psychology. Forth Worth: Harcourt Brace College Publishers.

VIDAL, Jean-Pierre. 1982. "La capture de l'imaginaire." In: VAUGHN-JAMES, Martin. 1982. La Cage. Paris: Impressions Nouvelles, p. 185-199.

WILLATS, John. 1997. Art and Representation. New Principles in the Analysis of Pictures. Princeton, New Jersey: Princeton University Press

Comics mentioned

MCCAY, Winsor. 1903. "Saved". In: Life. November 5, 1903.

VANDERSTEEN, Willy. 1947. De familie Snoek. In: De Nieuwe Standaard. April 19th , 1947.

VAUGHN-JAMES, Martin. 1975. The Cage. Toronto: The Coach House Press.

This site is optimized for Netscape 6 and higher

site design: Sara Roegiers @ Maerlantcentrum