Stories in Pictures (and Non-Pictorial Objects) –
A Narratological and Cognitive Psychological Approach
Narratological studies have quite frequently focused upon linguistic structures, considered to be paradigmatic cases of narrativity, whereas pictorial signs (such as icons and symbols) or indices have received comparably much less attention.
In this paper, however, I intend to outline some basic and regularly occurring narrative aspects of ‘pictures’ (in a wide sense), i.e. painting, reliefs, sculptures, and the like. Furthermore, I shall assume (influenced by approaches from cognitive psychology, e.g. the work of Roger Schank et al.) that cognition to a considerable extent consists of the storage and retrieval of action scripts or schemata (i.e. narrative structures) which may occur on various levels of abstraction. These schemas incorporate generalized knowledge about event sequences as well as scene schemas including inventory and spatial-relation information.
Through previous experiences we acquire a large amount of such culturally based event and scene stereotypes (along with idiosyncratic variations), either due to our previously acquired, direct familiarity with instances of events, or due to our acquaintance with written, oral, and of course pictorial descriptions of them (e.g. religious or mythological tales). They include settings, sub-goals, and actions in attempting to reach specific goals.
The production and comprehension of pictorial signs, as I finally will claim, is frequently based upon the existence and activation of such mentally stored action and scene schemas on part of the beholders. Actually, even things in general, whether artificial or natural objects, may be capable of expressing or triggering such narrative structures, thus to some extent of “telling us stories”. Thus in this paper, I shall present and discuss some examples of pictures (and non-pictorial objects) where narrative structures become activated and, indeed, their recognisability and comprehensibility as such seems to presuppose these structures.
Quite frequently, narration has been associated with verbal discourses, whether in written or oral form, where, briefly put, events or situations are represented in a time sequence. Accordingly, theoretical discussions concerning narrativity have usually focused upon literature and drama, though also on cinema films and television. However, the ability of static pictures to represent actions and to narrate stories seems to have received much less attention in art theory contexts. Among art historians, on the other hand, the narrative aspects of visual art have of course constituted a prevalent focus of interest, though chiefly from a descriptive, interpretative, and historical point of view. Still, attempts to elucidate any deeper psychological and philosophical aspects involved in visual narrativity have usually occurred on a rather superficial level, consisting of scattered remarks, intuitively based hypotheses, or the like. Any continuous and systematic treatment of narrative and temporal imagery, compared to the vast amount of discussions concerning the rendering of space and perspective, seems to be largely absent.1
This relative lack of theoretical interest is of course somewhat surprising; visual narratives are undoubtedly occurring in most historical and cultural contexts. With regard to Western art, we may find examples of pictorial "story-telling" at least as early as in ancient Egypt, Greece, and Rome as well as in the Middle Ages, the Renaissance, and so forth (and numerous examples from the Orient and Asia could unquestionably be mentioned).
Still, any deeper theoretical reflections on this matter occur often only as scattered remarks even among art historians - which for obvious reasons is unfortunate. Erwin Panofsky, one of the most influential art historians with outspoken theoretical concerns, may be credited for having elaborated the so-called iconographical or iconological methods. According to Panofsky, a fruitful investigation of works of art should be striving for an analysis of their meaning-aspects (in contradistinction to their formal aspects). These aspects occur on several levels.2 First, we have a pre-iconographic level - the depiction of human beings, animals, natural or artificial objects, etc. The identification of gestures, expressive qualities, and simple actions would also belong to this level. A second interpretative level - the iconographical analysis - consists of identifying the subject matter or the theme of the artwork. An iconographical interpretation would demand an identification of the depicted agents as certain persons (for example, the Virgin Mary or Heracles) or maybe personifications with certain attributes and would, if necessary, contain some reference to relevant myths or tales (i.e. complex action sequences).3 So far, so good.4 The exact nature, however, of such narratives (in contrast to e.g. the rendering of space and perspective), i.e. the various means used by the artist in order to convey them and the presuppositions needed on part of the beholder in order to understand them, is analyzed disappointingly scarce. It should be pointed out that Panofsky by no means seems to be an exception in that respect; indeed, among art historians, as well as aestheticians, problems of narrativity in pictorial art seem hardly to have received any continuous and thorough attention compared to those issues mentioned earlier.
To some extent, of course, this neglect seems to be understandable. Usual conceptions of pictorial representation appear to be irreconcilable with the common sense idea of narration as being temporal and sequential, or, put in another way, as a “temporal program” explicitly manifested by a work. Paintings seem to present themselves as holistic and almost immediately graspable, but verbal narratives as linear, requiring a temporally successive perceptual process. Now, as, for instance, the narratologist Gerald Prince has proposed, a minimal requirement for something to be a narrative consists of “the representation of at least two real or fictive events or situations in a time sequence neither of which presupposes or entails the other.”5 Thus, not all modes of discourse may be properly regarded as narratives. For example, arguments are usually considered to be (e.g. deductive or inductive) forms of persuasion, relying on logic, though not necessarily strict syllogisms, in order to prove the validity of an idea, or point of view. Expositions can be described as acts of expounding, setting forth, explaining, or conveying information (e.g. about a narrative’s plot, characters, setting, and theme), while descriptions (whether indefinite or definite) render the properties of things, verbally or visually (re-) present persons, places, events, or actions, as well as nonvisible or abstract states of affairs. And explanations, such as deductive-nomological or teleological ones, can, briefly put, be defined as descriptive statements attempting to clarify the causes, contextual circumstances, and consequences of certain facts. None of these discursive modes have an internal time sequence seemingly required by narrative structures; they seem to be static or atemporal. Still, narratives may very well make use of, incorporate or overlap with arguments, expositions, descriptions, or explanations.
Although Prince admits of many different manifestations and of varying degrees of narrativity (some narratives may be ‘more narrative’ than others)6, it seems, however, that he adheres to a rather essentialist definition of the concept, where the necessary, perhaps even sufficient, characteristic consists of the “event-sequence” criterion. Thus a sentence such as “The water boiled then World War II started” would, according to Prince, qualify as a minimal narrative.7 However, to call such an extremely reduced event sequence a narrative seems to be rather counterintuitive.8 As, for example, Noël Carroll has argued, such an example should rather be counted as a mere chronicle, where the crucial “narrative connection” is missing. Such a connection does not necessarily consist of strict causal entailments, rather, “[in] most narratives, the earlier events in a sequence of events underdetermine later events.”9 Inspired by J. L. Mackie’s discussion of so-called INUS conditions, Carroll argues, though, that a narrative connection occurs when there is “an insufficient but necessary part of a condition that itself is unnecessary but sufficient for an effect event”.10 An example of such INUS conditions would be the sentence “The thief enters the bank to rob it, but subsequently, as he exits, he is apprehended by the police”. Although the robbing of the bank is causally relevant, it does not causally determine the arrest. Apart from INUS conditions, a narrative connection seems, according to Carroll, also to require perspicuously ordered temporal relations between the occurring events, it concerns the career of at least one unified subject (rather than just adding up disparate or disconnected subjects), and it is structured in a globally forward-looking manner (rather than being orientated ‘backwards’).11 Other narratologists, such as Monika Fludernik, have attempted to delineate narratives from other forms of discourse by also stressing the representation of human or at least anthropomorphic protagonists (such as speaking animals) performing goal-directed actions, being anchored in particular (existential) time-space settings.12
Be this as it may, most discussions concerning narratives have, as already mentioned, focused upon verbal and literary narratives, whereas other kinds, such as pictorial ones, have been treated quite casually. At the first glance, such representations are indeed most favorably manifested by "genuine" temporal arts, such as poetry, drama, literature in general, motion pictures, and the like which “by nature” have a sequential structure. Pictures, on the other hand, are “by nature” static, only capable of representing timeless situations or single, momentary instants (thus the notion "static picture" in itself would appear to be tautological).
Accounts on these lines have been put forward by, for example, Lord Shaftesbury, James Harris, and Gotthold Ephraim Lessing. Harris, in his Discourse on Music, Painting and Poetry (1744), distinguished between media such as music, which is concerned with motion and sound, and painting, which renders shapes and colours. Pictures, according to Harris, can “of necessity [only represent] a punctum temporis or instant”. Interestingly, though, he also admits that “in a Story well known the Spectator’s Memory will supply the previous and the subsequent… [This] cannot be done where such knowledge is wanting.” Indeed, he doubts whether the rendering of a historical situation in a painting even would be intelligible “supposing history to have been silent and to have given no additional information.”13
A more well-known and much debated account was put forward by Lessing, who in his Laokoon: oder über die Grenzen der Malerei from 1766 attempted to characterize the distinctive features of painting vs. poetry qua signs, claiming that the representation (or "imitation") of actions does primarily (and best) occur in poetry.
"Objects which exist side by side...are called bodies. Consequently bodies with their visible properties are the peculiar subjects of painting. Objects which succeed each other...are actions. Consequently actions are the peculiar subjects of poetry."14
The claims put forward by Lessing are thus that painting essentially is an art of space concerned with the rendering of bodies, while poetry is an art of time, the latter being privileged in narrating actions, that is, the succession of events in time. But poetry cannot render actions without being “joined to certain agents. In so far as those agents are bodies or regarded as such, poetry describes also bodies, but only indirectly through actions”.15 And “bodies…exist not only in space, but also in time… [P]ainting can imitate actions also, but only as they are suggested through forms.” That is, painting is capable of indicating actions, though only indirectly through suggestion, namely by preferably choosing the most pregnant, arrested movement in an imagined action sequence.
"Painting, in its coexistent imitations, can use but a single moment of an action, and must therefore choose the most pregnant one, the one most suggestive of what has gone before and what is to follow."16
The representation/perception of actions in painting is thus not impossible per se, but it demands more effort, it is less ‘convenient’ compared to poetry. Actually, the difference between painting and poetry is rather a matter of degree than a matter of kind: poetry represents actions directly, painting only indirectly. Moreover, Lessing may very well be criticized for committing a naturalistic fallacy: from a factual description of the two genres as primarily spatial or temporal, being their ‘essential nature’, he comes to the normative conclusion that these genres ought to be restricted to those ‘natural’, functional characteristics. 17 But is there any reason why we should adhere to such a rigorous normative position, just by referring to various degrees of ‘convenience’ or ‘ease’?
Still, it seems quite possible that narratology's primary concern with ‘temporally extended’ arts such as literature, movies, and so on could has been influenced by similar essentialist lines of thought. We may ask, however, whether and to which extent conceptions of pictorial representation as basically static and non-temporal actually are tenable. Of course, it has frequently been admitted that the perception of pictures in itself is a temporal, successive process. As, for instance, Etienne Souriau, in his essay "Time in the Plastic Arts" (from 1949), has argued, the view that pictorial works are seen "in its entirety in a single instant...is clearly false"; rather, viewing a picture, as other visual works of art, involves "a period of contemplation wherein successive reactions take place."18 This is not only the case when it comes to three-dimensional objects, such as inspecting a sculpture or walking through a Gothic cathedral, but also two-dimensional paintings demand similar efforts. We should note, however, that according to Souriau, the fictive time inherent in a pictorial representation "radiates...around the prerogative moment represented... [being] a structural center from which the mind moves backward to the past and forward to the future", and thus in this respect his view bears a close similarity to Lessing's.19 Ernst Gombrich, to mention another example, has likewise maintained that "...[t)he reading of a picture...happens in time, in fact it needs a very long time...We do it, it seems, more or less as we read a page, by scanning it with our eyes...We build it up in times and hold the bits and pieces we scan in readiness till they fall into place as an imaginable object or event, and it is this totality we perceive and check against the picture in front of us."20
A number of experiments on eye movements and picture perception suggest indeed that the perception of pictorial representations involves something like a temporally extended scanning activity and feature analysis. The Russian psychologist Alfred Yarbus, one of the pioneers in this field of research, studied the saccadic movements of beholders' eyes when encountering different kinds of visual stimuli, such as photographs or paintings.21 Eye movements do not occur arbitrarily, but may rather be described as a systematic scanning process where the beholder fixes his attention on one feature at a time for a very brief period (about 300 msec.), and then moves on to focus on another feature. Now, it does not seem to be especially controversial to admit of temporal processes involved in the perception of pictures, thereby repudiating that any instantaneous understanding of either medium or message is possible. Still, it may be argued that this kind of temporality rather is dependent on the viewer's activities than on the object itself which by nature is static and temporally "frozen", so to speak.22 However, such a narrow and essentialist view on the concept of pictorial representation might be put into question by pointing to a number of counter-examples. What about stage design or scene painting, for example? In numerous cases such pictures have by no means been static, but made use of moveable parts such as representations of clouds, waves, not to mention various light effects such as strokes of lightning - thus creating a changeable pictorial scene (and I am here not even taking moveable subjects, i.e. actors, into consideration). We may also think about stained glass windows in Gothic cathedrals, changing due to the varying intensity of light filtered through them, or fountains or sculptural installations making use of water effects. In the 20th century, there are even further examples of non-static pictures or at least borderline cases, such as mobiles or op art-paintings (for example Bridget Riley’s "Crest", 1964).
Within contemporary aesthetics, it has widely been claimed that any attempts to define concepts such as "art" in essentialist terms by referring to necessary and jointly sufficient conditions (whether perceptual, functional or procedural) which members of this category are supposed to possess are doomed to failure. Rather, we should think about this category as being like a family whose members resemble each other in some, but not in all or in commonly shared respects. These complicated networks of similarities constituting the class of artworks are, borrowing a Wittgensteinian term, called family resemblance. This line of reasoning is of course quite familiar to those who are acquainted with contemporary aesthetics, most notably analytic aesthetics. Moreover, numerous cognitive psychologists have, following Eleanor Rosch's pioneer work, attempted to investigate (by means of quite strict experimental procedures) the nature and acquirement of categories in general, most notably that of taxonomic categories.23 The results obtained from these experiments support, according to Rosch, the assumption that categories, psychologically speaking, usually do not have clear-cut boundaries, but rather possess a graded structure.24 This means that there are certain category members which are experienced as cognitive reference points (or the clearest cases of category membership), while other members gradually deviate from them, although they still belong to the category in question. Put in another way, categories are formed around their most representative instances, which have something like a prototypical character.
When it comes to narratives, we may likewise conceive of them as constituting a category with fuzzy boundaries; also in this case it seems rather problematic to insist on a too rigid and essentialist view concerning this category's nature.25 Narratives may be intertwined, it seems, with descriptions, expositions, arguments, and explanations; meaning-bearers may be more or less narrative; and narratives may be manifested in various genres, as those mentioned before. But if we admit of the existence of temporal and narrative aspects in pictorial representations, the question still remains in which way(s) clear-cut (still) pictures, reliefs, or sculptures possess such features, and whether or in which respects some pictures might be regarded as ‘more narrative’ than others. Thus, let us take a closer look at some ways in which pictures seem to have a relatively straightforward narrative function (with temporal ingredients).26
First, we have numerous historical examples where static, monoscenic, and quite distinct pictures are linked in narrative series having a fixed reading order, frequently horizontal or vertical. Modern instances of this kind of pictorial narration can be found in strip cartoons, but do actually occur as early as in antiquity and the Middle Ages (e.g. scenes from the life of St. Ambrose on the back of the altar in S. Ambrigio, Milano, c. 850; the scenes from the Old and New Testaments on the bronze doors of the Hildesheim Cathedral, c. 1015; Giotto's Passion scenes in his frescoes in the Arena Chapel, c. 1306; Gaudenzio Ferrari's Passion scenes in S M delle Grazie, Varallo, c. 1513; William Hogarth's series of moralizing engravings in the 18th century; cf. illustr.1).
Illustration 1: Bronze doors from St. Mary’s Cathedral, Hildesheim (c.1015), showing the Creation of Adam and Eve and the Expulsion from the Garden of Eden (left), birth and Passion of Christ (right); vertical reading orders: left top-down; right down-top.
Second, and relatively often discussed by art historians, there are single pictures showing different events and persons in the same pictorial space. In these cases, sometimes called "continuous narratives" or cases of "simultaneous succession", various phases in an event series are represented simultaneously.27 Such forms of pictorial narration can also be found throughout history (e.g. the epic-documentary representation on the column of Trajan of the emperor's war against the Dacians, c. 101 - 106 AD.; Massacio's fresco "Tribute Money", showing St. Peter three times in the same pictorial space, c. 1427; Fra Filippo Lippi's depiction of the Banquet of Herod in the Cathedral of Prato, 1460s; Bernardino Luini's Crucifixion in S. M. degli Angeli, Lugano, c. 1530).28 Illustration 2 shows an especially interesting example, “The Legend of the Relics of St. John the Baptist” by Geertgen tot Sint Jans, c. 1484. In the background we may see the separate burials of the head and the body after the decapitation of the Baptist, assumingly having occurred in the first half of the
1st century AD. (illus.2a). In the foreground the opening of the tomb and the burning of the limbs on the orders of the Roman Emperor Julian the Apostate is rendered, about 362 A D. (illus. 2b, c); in the center the rediscovery of the rescued remains in the 13th century is shown (illus. 2d, e). However, the last scene also includes a group portrait of the Knights of the Order of St. John Convent in Haarlem, thus relocating the scene in the late 15th century (when the relics were given to the Order by the Turkish sultan, being the concrete reason for commissioning this painting). The implied time span in this pictorial narrative is thus remarkably extended, stretching over a period of more than 1.000 years.
Illustration 2: “The Legend of the Relics of St. John the Baptist”, Geertgen tot Sint Jans,
Illustration 2b. Illustration 2c.
Illustration 2d. Illustration 2e.
Of course, numerous examples may be found where these two forms of pictorial narration are intertwined, for example in Lorenzo Ghiberti's reliefs on the Baptistery doors - the so-called Porta del Paradiso - in Florence, 1424 - 52, showing ten separate, though narratively linked, scenes from the Old Testament. These scenes constitute a narrative series consisting of distinct pictures, beginning with Adam and Eve, then showing Kain and Abel, Noah, Abraham and Isaak, Joseph, Moses, Josua, David, and Salomon. However, almost all of these reliefs are polyscenic or continuous narratives. In the picture showing Adam and Eve, for example,, we can distinguish between various scenes in the same pictorial space: (i) the creation of Adam; (ii) the creation of Eve; (iii) the temptation and fall of Adam and Eve, and (iv) the expulsion from Paradise.
Now, in our present context I would like to focus upon a third kind of pictorial narration, namely in static pictures which seem to have a less straightforward narrative function, that is, where a frozen scene in a tacit action sequence has been chosen, from which what has preceded and will follow has to be inferred by the beholder. Thus Lessing's idea of the "pregnant moment" would very well fit into this category; only an arrested moment is directly represented, though implying a wider, temporally extended action sequence. The hellenistic sculpture group "Laocoön and his Two Sons" (1st century AD., illus. 3), which Lessing himself discussed at length, may be mentioned as an example where the depicted scene is referring to a series of mythological events. The pregnant ‘moment’ rendered here consists of the death struggle between the priest Laocoön and his two sons with two snakes, sent by
Illustration 3: “Laocoön and his Two Sons”, 1st century A.D.?
Athena as punishment for his attempt to warn the Trojans from taking a wooden horse, having Greek warriors hidden inside, into the city. Thus a beholder acquainted with the relevant narrative background might very well see this sculpture as a significant or crucial moment within a narrative sequence, stretching backwards in time as well as into the future, i.e. where Laocoön and his sons die, and the Trojans become defeated by the Greek.
Mythological, religious, political, or otherwise widespread narratives like this have of course frequently been pictorially rendered, that is, by visualizing more or less significant segments of implied narrative structures. However, we may also think about static pictorial scenes which either simply refer to more common (or even everyday) action schemes or which are narratively, so to speak, more or less indeterminate. In the next section, I shall discuss, by taking research within cognitive psychology into account, how and in which ways pictures may have narrative implications and give rise to the emergence of narrative mental representations among beholders.