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



Home archive




Issue 16. House / Text / Museum

The construction of space in comics

Author: Pascal lefèvre
Published: February 2006

Abstract (E): This paper on diegetic space is not only based on parts of my PhD Willy Vandersteen’s Suske en Wiske in the dailies (1945-1971), A Theoretical Framework for the Formal Analysis of Comics (originally written in Dutch in 2003), but also on some earlier publications in Dutch and French such as Het hors champ in de strip (Lefèvre, 1989), Pour une lecture moderne de la bande dessinée (Baetens & Lefèvre, 1993), Architecture dans le neuvième art (Lefèvre, 1996).

Abstract (F): Le présent article est basé à la fois sur notre thèse de doctorat: Willy Vandersteen’s Suske en Wiske in the dailies (1945-1971), A Theoretical Framework for the Formal Analysis of Comics (originellement écrit en néerlandais) et sur quelques articles déjà parus en néerlandais et en francais, tels Het hors champ in de strip (Lefèvre, 1989), Pour une lecture moderne de la bande dessinée (Baetens & Lefèvre, 1993), Architecture dans le neuvième art (Lefèvre, 1996). Il propose quelques outils d'analyse pour l'analyse de l'espace fictionnel ou diégétique en bande dessinée.

keywords: space, diegesis, panel, comics, sequence


Among other activities the reader constructs during his reading of a comic the diegetic space of a story. The diegetic space is the fictive space in which the characters live and act. This space can be constructed by the reader in various ways: both by elements that appear inside the frame of a panel, and by elements which remain unseen (which are called in French 'hors champ'). This non-visualised space does not only refer to the virtual supposed space outside the frame (in French called 'hors cadre') of a certain panel, but also to the supposed 'hidden' space within the borders of the panel itself (in French called 'hors champ interne'): for instance figures can overlap one another and hide parts for the eye of viewer. While some elements may not be visualized, they can be suggested by direct and indirect means: an element can directly indicate its presence outside the visualized space (e.g. a shadow or a balloon inside the frame that indicate the presence of someone outside the frame), or indirectly unexpected elements can pop up in a later panel: e.g. while the first panel shows us a close-up of a person, the second panel by enlarging the frame can show that this close-up is just of photograph on a wall and not an acting character in this scene (for example in the opening sequence of City of Glass the first telephone we see, turns out to be a drawing on a telephone directory). The artist has thus a powerful tool, namely framing at his hands: by limit the scope for the viewer and therefore the available information, he can cause a reader to make wrong inferences. Leaving elements outside the frame is regularly used in cliffhangers: an unseen but threatening monster is usually more effective than a represented one.



The construction of space in separate panels


The construction of space is a dynamic process, not only for each individual panel, but also for complete sequences. Several cues can help or obstruct the reader in this process.

In the first place every flat image has to deal with its fundamental two-dimensional aspect: the picture can try to deny the flatness by suggesting an illusionary depth or, on the contrary, can accentuate this flatness (like Trondheims's Bleu , 2003). In most comics two-dimensional composition represents a three-dimensional space in which the action occurs. In the course of history visual artists have developed several means to suggest a voluminous space on a flat surface.. Spatial relations between figures or objects in a picture can be described by projection systems. Willats (1997) following Booker (1963) defines projection systems in terms of primary and secondary geometry. Primary geometry is viewer-centered and describes pictures in terms of projection rays: "The geometry of projection of lines or rays from objects in the scene and their intersection with the picture plane to form an image or picture." (Willats, 1997:369). Most technical drawings can be described by primary geometry, but other formal projection systems as the reversed perspective can not be described by primary geometry. In those cases an object-centered system is needed, like secondary geometry, which Willats (1997:369) like Booker (1963) defines as: "The two-dimensional geometry of the picture surface, obtained without recourse to the idea of projection." From the renaissance till the end of the 19 th century linear and aerial perspective was the most used projection system in European art, but in other periods and other places other projection systems were used: for example the reversed perspective in Byzantine and Russian icons (Willats, 1997:12) or the 45° oblique in Asians paintings and drawings. Each method has its possibilities and limitations, so the choice of a certain projection system has many consequences. While linear perspective offers only one possible view on an object, object-centered projection systems can offer various views on the same object (e.g. cubist effects) or respect the relative distances (e.g. 45° oblique). The intrinsic qualities of the object to represent can play a role in the choice of the projection system (Palmer, 1999:370). Objects that appear on a flat surface can never show the complete reality of that three-dimensional object. The flat and unmoving image can only use monocular cues to suggest depth: interposition or overlapping, convergence, relative size, density gradient. Not all depth cues were everywhere and in all times used (for an historical overview see Solso, 1994:192).

While various depth cues can lead to the same conclusion, sometimes they can contradict each other, which can cause tension (Arnheim, 1971:126). [1]

The construction of space by the reader has several goals: a certain space is necessary to situate the action. Therefore most artists use stereotypical icons (like the Statue of Liberty for New York or the pyramids for Egypt) because such well known buildings or monuments can be easily recognized by the readers. Moreover most buildings can already by their form indicate which function they have (e.g. farms, airplanes, houses.). Space can also suggest other meanings: the way a person has decorated or organized his room can suggest what can of personality he or she has (orderly or messy, classic or modern, etc.). Furthermore space can express a certain mood or be a symbol for an underlying concept or a scene or even a complete story. When Jacques de Loustal draws a building he wants foremost in his subtle coloring to catch the atmosphere; he is not interested in spectacular architecture but in unsightly or banal buildings (de Loustal, 1990). The rigid, monumental forms of the architecture of Urbicande (in Schuitens and Peeters' La Fièvre d'Urbicande , 1984) suggest an authoritarian system, that suppresses its inhabitants as insignificant parts. No wonder that the authors found their inspiration in Stalinist and fascistic buildings and in futuristic architectural projects of early 20 th century. Schuiten and Peeters use architecture in metaphorical way in their Dark Cities -series.



Construction of space in a sequence


In general the reader expects that the diegetic space of a comic is sufficiently coherent: he expects - in analogy with daily life - a consistent space, because he tries on the basis of cues (given in the panels) to form a global image of the complete space. Some authors such as the French François Bourgeon dedicated great attention to the (re)construction of a diegetic space: for his story about the Middle Ages Le Dernier Chant des Malaterre (1989), he even built scale models and drew plans of his locations (Thiebaut, 1992:57-68).

The reader knows the cues to construct a space: he recognizes in the linear perspective depth cues, he is conscious of the unseen but virtual space outside the panel borders, to link the fragments together the reader is looking for overlaps. Without the necessary overlaps, the readers can only believe that the various fragments belong to the same and consistent space. By and large, readers will not check every diegetic space in all its details for its degree of contingency: he knows that the diegetic world is not completely the same as his daily reality and he accepts the existence of fictive worlds with their own rules and principles (e.g. imaginary worlds as cities on other planets). Some contradictions of the diegetic space remain unnoticed, usually the suggestion that the various fragments belong together is sufficient for the reader. Scores of comics suggest a coherent diegetic space without giving sufficient proof. Seldom in a sequence are all the corners of one room shown or a global view of the space is presented. The reader's expectation of a consistent diegetic space is often wrong. Berthomé (1990:44) argues that the décor changes according the needs of the moment and he gives the example of the Asterix village (the same houses occupy different locations in various stories of the series). One has to make a difference between changes that do not affect the illusion of a consistent diegetic space and those changes that weaken this illusion or belief. Readers accept that not in each panel every detail of the décor is repeated: the décor might disappear temporally from the reader's view to accentuate the actions of the characters. By and large, readers do accept these codes, they are not surprised that elements disappear and reappear. Except, this temporally disappearances of the décor, there may be also more unexpected changes in the diegetic space, that affect the represented space intrinsically. Even in series like The Dark Cities ( Cités Obscures) an attentive reader may notice such inconsistencies: for instance in La Fièvre d'Urbicande (1983) changes the big world map (on the wall of Robricks office), in Brüsel (1991) is the same window of the Wappendorf's house drawn with different numbers of panes (p. 51 et 54), also Wappendorfs, invention the 'solenoïde' (p. 52 et 53) does not carry always the same number of rings.

Such inconsistencies do not have to surprise us, because unlike in cinema there is no camera that registers a material décor, in comics every panel has to be composed again on the blank page. Characters and décor can only exist in comics if they are represented in some way or another. Even if panels seem to offer the same view on a certain space, everything which is not drawn again will be absent. Small changes will no obstruct the reading, because they are not considered as radical inconsistencies of the represented world (Baetens & Lefèvre, 1993:31-32). Usually these are details of lesser importance and their visibility may depend both on the comic and the reader: if a comic pretends to be a realistic depiction of our world, the reader will expect a sufficient degree of consistency. In a humorous drawn comic the reader will accept more voluntary inconsistencies in the representation of the diegetic space.

In some comics the changes are a little more visible, because they are more important details, but even then they mostly remain noticed by the average reader. Not only in more experimental comic strips as Krazy Kat , but also in mainstream comics (e.g. the village of Astérix, the house of Suske en Wiske , the Moulinsart castle of Haddock, or the office of Mortimer) the décor can be volatile. The characters themselves never do notice the bizarre changes of their space: for them their environment seems to be stable and consistent - but the attentive reader knows otherwise. If a very attentive reader notices such inconsistencies, he may both become frustrated that his realistic expectations were fooled and delighted because he found some 'mistakes'.



Extradiegetic space


Except that diegetic space in every comic has also an extradiegetic space, namely the space outside the fictive world of the comic. The extradiegetic space is the material space that surrounds the individual panels: not only the whites between the panels, but also the real space in which the reader is located. Of course characters are not expected to be aware of this space, only in self-referential exceptions characters deal with that aspect: the frame of a panel crumbles upon Little Nemo, Philémon climbs by means of the panel frames from one panel into another. In Martin Vaughn-James' The Cage (1975) the white is used as a kind of mat placed over the drawings (Baetens & Lefèvre, 1993:35-36) Furthermore the extradiëgetic space can be integrated by means of a character seemingly looking the reader straight in the eyes and addressing the reader in his speech balloon; but the extradiegetic space can never be directly represented in the comic - except when for instance a real mirror would be pasted on the page.





Arnheim, Rudolf. 1971 (1954). Art and Visual Perception, a Psychology of the Creative Eye . Berkeley: University of California Press.

Baetens, Jan, & Lefèvre, Pascal. 1993. Pour une Lecture Moderne de la Bande Dessinée . Bruxelles: Centre Belge de la Bande Dessinée.

Berthomé, Jean-Pierre. 1990. 'Décors, décors' in: CinemAction: cinema et bande dessinée . Paris: Corlet - Télérama, été 1990, p. 41-47.

de Loustal, Jacques. 1990. Carnet de voyages . Paris: Futuropolis.

Lefèvre, Pascal. 1988. 'Het hors champ in de strip' in: Andere sinema . N° 6; 1988. p. 41-42.

Lefèvre Pascal. 1996. Architectuur in de negende kunst/Architecture dans le neuvième art . Arnhem: NBM-Amstelland Bouw BV.

Lefèvre, Pascal. 2003. Willy Vandersteens Suske en Wiske in de krant (1945-1971). Een theoretisch kader voor een vormelijke analyse van strips. Leuven: Doctoraat Sociale Wetenschappen, Katholieke Universiteit Leuven.

Palmer, E. Stephen. 1999. Vision Science. Photons to Phenomenology . Cambridge, Mass.: MIT Press.

Solso, Robert L. 1994. Cognition and the visual arts . Cambridge, Massachusetts: MIT Press.

Thiebaut, Michel. 1992. Dans le sillage des sirènes. Autour des Compagnons du Crépuscule de François Bourgeon . Tournai: Casterman.

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





[1]An artist can deviate from the normal proportions between the figures or objects: in the Middle Ages the most important figures like Christ or Maria were painted a lot bigger than the other figures. Also in comics disproportions are often used: for instance to make a character more visible in car his body and head can be enlarged in comparison to the car.


Pascal Lefèvre teaches the history and theory of comics at the Sint-Lukas school for fine arts (Brussels) and had published widely on the subject.



Maerlant Center Institute for Cultural Studies

This site is optimized for Netscape 6 and higher

site design: Sara Roegiers @ Maerlantcentrum