22.3 Jason Gendler

Julien Jacques Simon

December 3, 2013

Exposition in Narrative Cinema

¶ 1 Leave a comment on paragraph 1 0 Jason Gendler
Univ. of California, Los Angeles

¶ 2 Leave a comment on paragraph 2 0 This paper explores film’s recourse to conveying exposition. It will address four topics: how to differentiate expositional from non-expositional narrative material; the extent to which the literary means of defining exposition are applicable to other media such as film; ways in which exposition might be conveyed through imagery, as opposed to written or spoken text, and how one can identify exposition in stories in which the plot (or syuzhet) rearranges the order of the story (or fabula), such as in stories with flashbacks, or framing or embedded narratives.

¶ 3 Leave a comment on paragraph 3 0 By “exposition,” I am referring to Russian formalist Boris Tomashevsky’s use of term to indicate a narration’s presentation of fabula events that antedate the first scene of the syuzhet (i.e., things that happened before the start of the plot).[1] It serves an important contextual function in narration, providing, in Meir Sternberg’s words, “the general and specific antecedents indispensable to the understanding of what happens” in the story world. However, identifying exposition solely in temporal terms creates problems for stories whose syuzhets rearrange the order of the fabula, such as ab fine narratives like Sunset Blvd, where the syuzhet begins near the end of the fabula and then flashes back to an earlier point. Were exposition defined solely in temporal terms, most of the narrative of films like Sunset Blvd. could qualify as exposition, simply because much of the narrative consists of fabula material that antedates the start of the syuzhet. Such a definition seems a gross overstatement, one that renders the term “exposition” so broad as to be nearly meaningless. In order to preserve the use value of the term “exposition” in such films, it is necessary to distinguish expositional from non-expositional material in other terms aside from temporal order. Meir Sternberg has done just that, devising nuanced and flexible means of differentiating expositional material from non-expositional material in literature, which, as we shall see, are also useful for thinking about exposition in film.

¶ 4 Leave a comment on paragraph 4 0 Defining Exposition
Quantitatively, Sternberg identifies exposition through the intrinsic norm a given narration establishes for the pace of the communication of information. Usually, most scenes unfold at the pace of “real” time, which Sternberg calls the “scenic norm.” Literary exposition, however, is often indicated when the pace is increased, and scenes (located in the beginning of the fabula) become summarized.[2] When exposition is presented at the very start of the syuzhet, the exposition ends when the narration first transitions from summary to the scenic norm, a point which Sternberg refers to as the first “discriminated occasion.”[3] Quantitatively, then, expositional material is summarized fabula material antedating the first discriminated occasion in the syuzhet.

¶ 5 Leave a comment on paragraph 5 0 In ab fine narratives, then, rather than identifying as exposition all fabula material that occurs before the first discriminated occasion, instead only that material which is summarized should be considered expositional. Thus (ignoring for the moment the issue of what might count as summary in film), the majority of Sunset Blvd. should not be considered exposition because quantitatively, much of it is at the pace of the “scenic norm” rather than summary. However, by itself, the quantitative indicator is insufficient for distinguishing exposition in all cases, both because sometimes it is difficult to differentiate summary from the “scenic norm,” and because the quantitative indicator is still rather broad. Therefore, Sternberg also identifies three qualitative indicators for distinguishing expositional from non-expositional material.

¶ 6 Leave a comment on paragraph 6 0 These are specificity, concreteness, and actional dynamics. Specificity refers to how much detail a passage contains. The less specific the passage, the less detailed it is, and the more expositional it becomes.[4] Concreteness refers to the degree to which the events described are habitual or reoccurring. The more habitual or reoccurring the events or actions of a passage, the less concrete it is, and the more expositional it becomes.[5] Actional dynamics refers to the degree to which a passage develops and advances the action of the narrative. Expositional passages do not actively advance the narrative, but portray “a state of affairs that is essentially static or stable.”[6] In other words, non-expositional material is specific, concrete, and dynamic: it contains material details of actions and events; these actions and events are also singular and nonrecurring, and are also, “essentially dynamic or developmental, introducing into the once-stable state of affairs the first disturbing, destabilizing element… which causally leads, by necessary or probable sequence, to the next states of the action.”[7] Lacking these qualitative indicators, narrative material shades more toward exposition: it summarizes, describes recurring events, and describes stable states of affairs. When combined with the quantitative indicator, these qualitative indicators sufficiently narrow the range of possible expositional material to the point where they provide a thorough means of distinguishing expositional from non-expositional material in most literary cases.

¶ 7 Leave a comment on paragraph 7 0 In general, Sternberg’s indicators are matters of degree rather than absolutes, and context is crucial, because expositional material need not inhabit all of these indicators simultaneously. Even though the various indicators are “usually concomitant and interdependent,” the context of the narrative material will determine their relative use value.[8] They merely indicate that certain material is probably exposition, rather than functioning as hard and fast rules. This gives them some flexibility, which is beneficial both because they make his theory more robust, and because it also saves them for use in other media.

¶ 8 Leave a comment on paragraph 8 0 Exposition in Film
Sternberg’s indicators work very well in film when exposition is conveyed through voiceovers or titles, which also rely on spoken and written language, and which are thus subject to variations in pace, specificity, and concreteness. For example, the voiceover that begins Casablanca is purely expositional, and conforms to all of Sternberg’s indicators. It states, “With the coming of the Second World War, many eyes in imprisoned Europe turned hopefully, or desperately, toward the freedom of the Americas. Lisbon became the great embarkation point. But, not everybody could get to Lisbon directly, so a tortuous, roundabout refugee trail sprang up: Paris to Marseille; across the Mediterranean to Oran; then by train, or auto, or foot across the rim of Africa to Casablanca in French Morocco. Here, the fortunate ones through money, or influence, or luck might obtain exit visas and scurry to Lisbon, and from Lisbon to the new world. But the others wait in Casablanca, and wait, and wait, and wait.”

¶ 9 Leave a comment on paragraph 9 0 Quantitatively, the pace is clearly heightened: the voiceover describes actions and events that take days or weeks to complete, but which takes the voiceover only a minute to describe. Qualitatively, the voiceover lacks specificity: there are no details that describe life on the refugee trail, or anyone’s particular journey over it. Instead, it reads as a summary of what refugees do. It also lacks concreteness: all of the actions it describes refer to the reoccurring actions many people continually undertake during this period. Finally, the voiceover also lacks actional dynamics: it describes the stable state of the refugee trail at the onset of World War II, which is the background against which actionally dynamic events will take place.

¶ 10 Leave a comment on paragraph 10 0 Given that Sternberg’s indicators are meant to describe literary texts, it is not terribly surprising that they accurately describe filmic exposition when it is conveyed through language. This is especially the case for specificity: Exposition conveyed through dialogue, voiceover, or titles in film will almost always lack specificity in comparison to the imagery, because no matter how detailed the exposition, it is unlikely to equal the innumerable details with which images are naturally imbued (even sparse images).[9] Therefore, his indicators remain quite useful for describing exposition in (classically narrated) films, which typically deliver exposition through dialogue, titles, or voiceover.

¶ 11 Leave a comment on paragraph 11 0 Visual Exposition
However, a film’s image track is not idle during moments of auditory exposition. When exposition is conveyed through conversation between characters, often we might say that the audio is expositional but the image is non-expositional. Voiceovers are another matter, however, because they raise more directly the possibility of visual exposition. In Casablanca, the voiceover is accompanied by imagery, all of which should also be considered expositional, because even though the images may pass for specific and concrete, they serve as illustrations of the non-concrete things the voiceover describes, thereby ascribing to them the same non-concreteness as the voiceover. However, this example tells us nothing about the possibility of visual exposition when there is no voiceover to provide the right contexts. In order to preserve the possibility that exposition can be conveyed through imagery in film – more than as mere illustrations of auditory exposition – it is necessary to find visual equivalents for Sternberg’s literary indicators, and to expand the ways in which exposition can be indicated through context.

¶ 12 Leave a comment on paragraph 12 0 Even though most film imagery is shot at a ratio of 24 frames per second (a standard that indicates the “real time” of “discriminated occasions,”) it’s possible to use the quantitative indicator for visual exposition in film. Like literary narratives, film narratives also manipulate the pace of scenes.[10] Film can use ellipsis to hasten the pace by omitting parts of the fabula. Ellipsis most resembles literary summary when many are strung together in quick succession, as in montage sequences. A montage sequence can “summarize” fabula information by making the imagery representative of larger swathes of the story. For example, midway through Casablanca, there is a flashback to Rick and Ilsa’s Parisian romance, the beginning of which is conveyed through a montage sequence summarizing their romantic history (which occurs before the start of Casablanca’s first “discriminated occasion”).[11] With the ellipses of montage sequences, the imagery is imbued with the increased pace that the quantitative indicator is meant to identify.

¶ 13 Leave a comment on paragraph 13 0 Films can also accord with the quantitative indicator through what David Bordwell calls “compression,” where the pace is condensed, presenting, as Bordwell writes, “a series of actions in such a way that no missing time can be detected.”[12] In Play Time (1967), for example, the scene set at the Royal Garden restaurant takes place from evening to dawn and contains no noticeable ellipses, but lasts for only 45 minutes of screen time.[13]

¶ 14 Leave a comment on paragraph 14 1 Compression is also useful for indicating summary because it can occur even in scenes paced at the “scenic norm.” This possibility is most evident when objects or still imagery provide “compressed” accounts of other events. In the beginning of Rear Window (1954), for instance, the camera lingers over various objects in Jeff’s apartment, soliciting inferences from viewers about previous events. The combination of Jeff’s cast, the shattered camera, the crash photo, as well as the camera movement’s deliberate emphasis of each of these objects, solicits an inference about an anteceding event: Jeff broke his leg and camera in the process of photographing the crash. The narration is paced at the “scenic norm,” yet the information it reveals about anteceding events is “compressed,” because the “duration” of these events last only as long as it takes for a viewer to infer the cause of Jeff’s injury. Essentially, the imagery of this scene functions identically to that of exposition conveyed through language: in both cases, the narration solicits inferences from viewers about events that antedate the first discriminated occasion, compressing the duration of the events into the time it takes to view the imagery or listen to the dialogue, even though the scene is paced at the “scenic norm.”

¶ 15 Leave a comment on paragraph 15 0 Moreover, the crash photo and ruined camera also conform to Sternberg’s qualitative indicators. These objects are unspecific relative to the degree of detail available in the rest of the scene, offering few details beyond the most basic causal implications. We infer that Jeff broke his leg taking a photo of a crash on a racetrack, but we don’t know what the crash sounded like, how Jeff reacted to it, or the decisions that led him to be injured in the first place, all specific details that presumably would have been more readily available had the injury been presented as a scene in the film.[14] Additionally, other objects in Jeff’s apartment provide viewers with non-concrete information. Individually, these objects and photos show concrete events – they consist of particular actions that take place at a particular time – however, much like in a montage sequence, taken together, they provide non-concrete, expositional information about Jeff’s habitual or reoccurring professional activities – he appears to be a photographer, one whose assignments involve risk of danger. Most of these objects in Jeff’s apartment are also actionally static. The broken camera and racetrack crash photo might be considered exceptions, since they explain the cause of Jeff’s broken leg, and thus introduce a “destabilizing element” into a previously stable state of affairs. However, considering that Jeff’s broken leg is also a stable state of affairs that sets up the conditions for the rest of the entire narrative, one might consider the actional dynamism of the photo and camera to be severely mitigated.

¶ 16 Leave a comment on paragraph 16 1 Exposition through Flashback: Explanatory and Illustrative Flashbacks
As the Rear Window example shows, when the exposition is a part of a scene taking place in the present, Sternberg’s indicators are indeed useful for indicating visual exposition. However, there still remains the tricky issue of flashbacks. With the exception of flashbacks that employ montage sequences, as in the example from Casablanca, flashbacks to narrative material that antedate the first discriminated occasion would seem to pose a problem for the usefulness of Sternberg’s indicators in film, since such flashbacks often convey exposition, but demonstrate none of Sternberg’s indicators: they are often paced at the “scenic norm”; they appear specific and concrete because they consist of particular actions or events that take place at particular times, and they are often actionally dynamic, revealing information that destabilizes states of affairs. Such flashbacks would seem to resemble any other scene in a film, as well as the flashbacks of ab fine narratives like Sunset Blvd., where the flashback takes up most of the duration of the syuzhet.

¶ 17 Leave a comment on paragraph 17 0 In fact, the latter half of the Casablanca flashback conforms to this description: in the flashback’s final two scenes, Rick, Ilsa, and Sam close down La Belle Aurore café and plan their escape from Paris, and then Rick discovers Ilsa has abruptly and inexplicably ended their romance and left him to flee on his own. These scenes are not really a part of the montage sequence that comprises the first half of the flashback because they are not summaries of lengthier processes or events, nor are they representative of greater swathes of fabula time, but instead they reveal the specific, concrete, and actionally dynamic circumstances of Rick and Ilsa’s parting. Yet these scenes are just as expositional as the first half of the flashback: despite not conforming to Sternberg’s indicators, they still fulfill exposition’s basic function, providing the antecedents necessary to fully understand the story’s action. How, then, can we account for such flashbacks within the parameters outlined thus far, and in a way that differentiates them from the flashbacks comprising the majority of ab fine narratives?

¶ 18 Leave a comment on paragraph 18 0 One answer is to consider the contexts in which the flashbacks appear. In the latter half of the Casablanca flashback, the crucial contextual factor is that the flashback fills in a specific gap in knowledge. Such flashbacks might be called explanatory flashbacks, and they are expositional because they provide a previously missing cause for events or character behavior already narrated by the syuzhet. The second half of the Casablanca flashback is expositional because it provides the specific cause for Rick’s complex reaction to Ilsa’s appearance at his café in Casablanca: they were lovers, and she abandoned him. The ab fine flashback, on the other hand, rather than providing specific information to fill in a gap in knowledge, instead does the work of establishing the various figures and relationships of the story in the first place. Double Indemnity, for example, solicits many specific questions about the situation that opens the syuzhet (why is Walter Neff driving so frantically? Why is he going to his office at night? Why does he appear to be unsettled? etc.). However, the subsequent flashback doesn’t provide immediate answers to these questions. Instead, it provides much broader information: where and when the action is taking place; Walter’s background; how he met Phyllis Dietrichson, his immediate attraction to her, and so on.

¶ 19 Leave a comment on paragraph 19 0 As these examples suggest, the difference is a matter of the kinds of questions the flashback information provides answers to. The ab fine flashback establishes the story world and the major figures and conditions of the narrative to come, whereas the explanatory flashback fills in a specific gap in the story. This difference explains why the explanatory flashback is expositional but the ab fine flashback is not (despite neither according with Sternberg’s qualitative or quantitative indicators): the former fulfills exposition’s basic purpose – assisting viewers with more fully understanding the non-expositional story events – while the latter establishes the basic components of the entire story. It’s a difference between providing a small piece of information that can be plugged into our larger knowledge of the narrative, and providing the basis for the entire narrative in the first place.

¶ 20 Leave a comment on paragraph 20 0 Aside from explanatory flashbacks, flashbacks can also be made expositional if the context makes the flashback non-concrete. Sternberg accounts for this possibility when he writes that it is possible for stories to delve into scenes in the past that “may pass for concrete,” but which are “purely expositional” because of the “wider narrative framework in which they are set.” These passages are “concrete in themselves,” but they become “deconcretized” by ultimately conveying non-concrete information, such as illuminating “an habitual state of affairs” or reinforcing “an engrained trait.”[15] We might call such flashbacks illustrative flashbacks: seemingly concretized episodes are made expositional because they serve as illustrations of habitual states of affairs or engrained traits.

¶ 21 Leave a comment on paragraph 21 0 For example, in Zindagi Na Milegi Dobara (2011), Arjun is on a vacation celebrating his friend’s engagement, but can’t enjoy himself because of his demanding job, which requires him to frequently interrupt their vacation for work. Midway through the film, a flashback reveals the dissolution of his most recent romantic relationship. In the flashback, Arjun fights with his (now ex-) girlfriend because he canceled the vacation they planned for her birthday when it conflicted with a sudden work opportunity. What would otherwise appear to be a specific and concrete scene can be classified as exposition because it is “deconcretized” by its context: the flashback illustrates that Arjun’s workaholic behavior is a trait engrained so deeply that he inadvertently lets it destroy a serious romantic relationship.

¶ 22 Leave a comment on paragraph 22 2 In conclusion, Sternberg’s qualitative and quantitative indicators are useful for identifying exposition in film because it often accords with them, especially when narrated through language. Such exposition is often unspecific relative to the scene in which it appears, non-concrete, and actionally static. When exposition is narrated visually, his indicators remain useful because they have filmic equivalents such as ellipsis and compression, and because film has available various stylistic recourses for conveying specificity and concreteness. Even when a film employs flashbacks, there are contextual means through which expositional and non-expositional flashbacks can be differentiated, based on whether or not the flashback is explanatory of a specific gap in knowledge, and whether or not it is illustrative of engrained traits or habitual or reoccurring activities.

¶ 23 Leave a comment on paragraph 23 0 Notes
[1]. Boris Tomashevsky, “Thematics,” Russian Formalist Criticism: Four Essays. Trans. Lee T. Lemon and Marion J. Reis (Lincoln: University of Nebraska Press, 1965), 72. See also See Gerald Prince, Dictionary of Narratology, Revised Edition (Lincoln: University of Nebraska Press, 2003), 28. This definition is opposed to the less precise use of the term that can be found in various screenwriting manuals and even in some narratalogical works, such as Gustav Freytag’s use of the term, where it is often synonymous with “introduction.” See Gustav Freytag, Freytag’s Technique of the Drama: An Exposition of Dramatic Composition and Art. Trans. Elias J. MacEwan (Chicago: S. C. Griggs & Company, 1895). For a discussion of Freytag’s insufficiency, see Meir Sternberg, Expositional Modes and Temporal Ordering in Fiction (Bloomington: Indiana University Press, 1978), 5-7.
[2]. For more on duration in literature, see Gérard Genette, Narrative Discourse: An Essay in Method, trans. Jane E. Lewin (Ithica: Cornell University Press, 1980), 86-112.
[3]. Sternberg, 19-21.
[4]. Ibid., 24.
[5]. Ibid.
[6]. Ibid., 26.
[7]. Ibid.
[8]. Ibid.
[9]. At least, this is the case in classical narration. Some experimental or animated narratives might be so visually simple as to be less detailed than whatever specificity might be contained in written or spoken text.
[10]. David Bordwell, Narration in the Fiction Film (Madison: University of Wisconsin Press, 1985), 81-82.
[11]. Casablanca’s first discriminated occasion is most likely the point when a Vichy officer reads out a notice regarding the murder of two German couriers, and calling for the arrest of suspicious characters.
[12]. Bordwell, 82.
[13]. Ibid.
[14]. The availability of such details would of course depend on the way in which this hypothetical scene would be shot, but it likely would include some degree of increased detail, relative to the way in which the event is presented in the finished film. Moreover, even if those details were unavailable in a hypothetical injury scene, the scene would still be more specific than the way in which it is presented in the actual film, simply by virtue of the finished film conveying it through stationary objects and still imagery rather than moving imagery. Other stylistic choices could also potentially influence imagery’s specificity, obfuscating details through devices like distant framing, obstructive staging, low-key lighting, fast-paced editing, distorted sound design, or blurred focus.
[15]. Ibid., 30.