11. Historical-Materialist Narration
The Soviet Example
n its widest scope, leftist political cinema has no pertinence as a mode of narration. Political fiction films can
appeal to classical narrational norms (e.g., the work of Costa-Gavras) or to conventions of the art-cinema mode (e.g., Man of Marble, 1976). But within left-wing filmmaking we can discern one clear-cut narrational tradition. Although this tradition has influenced both classical and art-cinema norms, it possesses a distinct set of narrational strategies and tactics. These originate in the Soviet "historical-materialist" cinema of the period 1925-1933.1 will take twenty-two films as prime instances of this mode: Strike (1925), Potemkin (1925), The Devil's Wheel (1926), Mother (1926), Moscow in October (1927), The End of St. Petersburg (1927), October (1928), Zvenigora (1928), Lace (1928), Storm over Asia (1928), Arsenal
(1929) , The Ghost That Never Returns (1929), The New Babylon (1929), Fragments of an Empire (1929), Old and New (1929), Goluboi Express (1929), Earth
(1930) , Mountains of Gold (1931), Ivan (1932), A Simple Case (1932), Twenty-six Commissars (1933), and Deserter (1933). (Certainly By the Law (1926), Bed and Sofa (1927), Alone (1931), and others might be added to the list, but the above seem to me the least disputable cases.) After considering the Soviet variant, I will sketch out how the mode changed in later years.
HISTORICAL-MATERIALIST NARRATION
235
Narration as Rhetoric
Like much Soviet art of the 1920s, the historical-materialist film has a strong rhetorical cast. It uses narrational principles and devices opposed to Hollywood norms for purposes that are frankly didactic and persuasive. Within Soviet culture generally, artists and political workers debated how aesthetic practices could be translated into utilitarian ones. One position, exemplified by the extreme left wing of Constructivism, called for an end to "art," a hopelessly bourgeois category. But on the whole, both artists and politicians wanted to maintain "the aesthetic" as a distinct (if subordinate) space. Some, like Kuleshov, saw their work as part of a long-range process of basic research; pursued in the scientific spirit, their experiments could eventually reveal the laws of socialist art. Other creators made art obedient to "social command." Here the artwork was endowed with immediate utility as "agitprop." Patriotic music, the mass spectacles celebrating the October Revolution, and much of Mayakovsky's poetry are examples. No matter how practical the end, the social-command view clung to a conception of the distinctly aesthetic. "Art," wrote Lunacharsky and Sla-vinsky in 1920, "is a powerful means of infecting those around us with ideas, feelings, and moods. Agitation and propaganda acquire particular acuity and effectiveness when they are clothed in the attractive and mighty forms of art."1 Thus, the instrumental aim provided—at least for a time—an acceptable framework for experiment.
In Soviet cinema, the double demand of poetic and rhetoric shapes basic narrational strategies. There is the tendency to treat the syuzhet as both a narrative and an argument. Soviet cinema is explicitly tendentious, like the roman a these; the fabula world stands for a set of abstract propositions whose validity the film at once presupposes and reasserts. Strike offers a very clear instance. Not only is this the story of a single strike, it is a discourse on all the Russian strikes that occurred before 1917. The exact locale and time are unspecified; instead, the film is broken into six parts explicitly labeled as typical stages: seething in the factory; "Immediate Cause of the Strike"; "The Factory Stands Idle";
"The Strike Is Prolonged"; "Engineering a Massacre"; "Liquidation." The film concludes:
Extreme close-up: Eyes stare out at us.
Expository title: "And the strikes in Lena, Talka, Zlatovst, Yaroslavl, Tsaritsyn, and Kostroma left bleeding, unforgettable scars on the body of the proletariat."
Extreme close-up: Eyes stare out at us.
Expository title: "Proletarians, remember!" The film's argument works by appeal to example; the narrative cause and effect demonstrate the necessity for the working class to struggle against capital. While later films did not utilize the nakedly argumentative structure of Strike, they did rely on the presupposition that the narrative should constitute an exemplary case for Marxist-Leninist doctrine.2 Furthermore, Strike's example is a historical one; the fabula is based on fact. Other Soviet films take up this referential impulse, creating a "realistic" motivation for the fabula events.
The most obvious result of "rhetoricizing" the fabula world is the changed conception of character. Narrative causality is construed as supraindividual, deriving from social forces described by Bolshevik doctrine. Characters thus get defined chiefly through their class position, job, social actions, and political views. Characters also lose the uniqueness sought to some degree by classical narration and to a great degree by art-cinema narration; they become prototypes of whole classes, milieux, or historical epochs. Diego's existential crisis in La guerre est finie would be unthinkable in Soviet historical-materialist cinema. As M. N. Pokrovsky put it, "We Marxists do not see personality as the maker of history, for to us personality is only the instrument with which history works."3 The single character may count for little, as seen in some films' attempt to make a group of peasants or workers into a "mass hero." Such an approach to character had already been evident in Soviet revolutionary literature and theater of the 1918-1929 era.4
True, the Soviet cinema recognized degrees of individuation: the anonymous agents of Moscow in October, Eisen-stein's physically vivid but generally apsychological characters like the sailor Vakulinchuk, the more detailed delinea-
236
HISTORICAL MODES OF NARRATION
tion of individual behavior in Pudovkin, and the intensely subjective characterization in Room's films. Nonetheless, psychological singularity remains quite rare. Sometimes, as in October, the more psychologically motivated the character (e.g., Kerensky, with his Napoleonic lust for power), the surer the character is to be denigrated as a bourgeois.
Character types find their roles within specific generic motivations. There is the genre of "studies of revolution," either in historical or contemporary settings. Here the film tells a story of successful struggles (Potemkin, October, The End of St. Petersburg, Moscow in October, Zvenigora) or currendy emergent ones (Storm over Asia, Mountains of Gold, The Ghost That Never Returns, Goluboi Express, Twenty-six Commissars). The revolution film may also pay tribute to heroic failures (Strike, Mother, Arsenal, The New Babylon). A second genre portrays contemporary problems in Soviet life, usually involving remnants of capitalist or feudal behavior (Fragments of an Empire, Lace, The Devil's Wheel). There is also a genre that matches the literary formula of the "production" novel: a dam must be built (Ivan), or the countryside must be collectivized (Old and New, Earth). Some films combine genres: A Simple Case (historical revolution and problems of contemporary life) or Deserter (emergent revolution plus production goals). All these genres evidently give the film an opportunity to create a fabula that will make each character emblematic of forces within a politically defined situation.
One task of tendentious narrative art is to create conflicts that both prove the thesis and furnish narrative interest. In these films, the viewer is likely to know, or quickly guess, the underlying argument to be presented and the referential basis of the fabula world. (There can be no doubt that the October Revolution will succeed.) Most of our interest thus falls upon the question of how history takes the course it does.
In a general sense, the Soviet historical-materialist film answers this by adhering to the two schematic patterns which Susan R. Suleiman identifies in the roman a these. There is what she calls the "structure of confrontation," in which a psychologically unchanging hero represents a
group in his struggle against adversaries.5 Such is Marfa in Old and New, or the Chinese coolie in Goluboi Express. This structure provides a fairly traditional curve of dramatic conflict. There is also the "structure of apprenticeship" in which the typical individual moves from ignorance to knowledge and from passivity to action.6 The specific shape which this dramatic development takes in Soviet literature of the period has been summarized by Katerina Clark. She points out that the Socialist Realist narrative often centers on a character who moves from a spontaneous, instinctive form of activity to a disciplined, correct awareness of political ends and means.7 Mother, as both novel and film, is the canonic instance. The mother acts spontaneously but incorrectly, and her positive qualities are offset by the danger she poses to the revolution. By accepting the tutelage of her son and the Party, she is able to become a martyr to conscious revolutionary activity. The result of this pattern is that potentially affirmative characters are shown initially in a rather bad light: they may be naive (Mother, the sailor in The Devil's Wheel, Filiminov in Fragments of an Empire) or worse— cowardly (Renn in Deserter), lascivious (Pavel in A Simple Case), rowdy (the delinquents in Lace), treacherous (the peasant in The End of St. Petersburg) or greedy (the peasant in Mountains of Gold). The cause-and-effect chain then works to convert the characters) to disciplined socialist activity. The drama—and the spectator's hypotheses—come to be based on how and when the apprentice's conversion will take place.
To some extent, the didactic aim of the Soviet cinema created a storehouse of topoi, or argumentative commonplaces, which the filmmaker could use to structure the syuzhet. But these were not so narrow that they stifled experimentation. The narrative-plus-argument pattern was open to poetic exploitation in many ways. The use of character prototypes—the sturdy worker, the activist woman, the bureaucrat, the bourgeois "man out of time"—allowed stylistic embroidering. "The figure of a cinematic character," declared Pudovkin, "is the sum of all the shots in which he appears."8 It was up to the director not to give the character individuality but to use film form to make the type vivid.
HISTORICAL-MATERIALIST NARRATION
237
Pudovkin could draw on the techniques of poster art and contemporary fiction, Eisenstein on theater and caricature, Dovzhenko upon cartoon art and Ukrainian folklore. Commonplace rhetorical points could be sharpened by stylistic devices. The opening sequence of Arsenal powerfully demonstrates how, given the topos "The czar's war destroys the Russian peasantry," a film's shot-to-shot relations could still be made highly unpredictable. Similarly, in Old and New, Marfa's decision to organize a collective is presented so that her misery in the fields ("Enough!") is alternated with her oratory before her friends; impossible to say where one scene [eaves off and the other begins. Rhetorical demands provided generic and realistic motivation for an experimentation with the medium akin to that in Soviet avant-garde art generally. Thus Old and New's localized breakdown of classical order and duration is motivated by the whole film's juxtaposition of past and present. In Russian Formalist terms, the rhetorical aim enabled the films to "defamiliarize" classical norms of space and time.
Once the filnvuses poetic procedures for rhetorical ends, the narrational process becomes quite overt. The narration comes forward as a didactic guide to proper construction of the fabula.
There is an especially clear index of this. In the classical Hollywood cinema of the silent era, the narration almost invariably employed many more dialogue titles than expository ones—usually four to twelve times as many. In some films of the late 1920s, there are no expository titles at all. The reason is obvious: an expository tide creates a self-conscious narration that is only occasionally desirable in the classical film. But the Soviet films I am considering here have a much higher proportion of expository titles. In most of these films, dialogue titles outweigh expository ones by a ratio of only four to one, and some of the films actually contain more expository than dialogue titles. In later years, the Soviets' use of nondiegetic or "contrapuntal" sound montage had a comparably overt effect.
Overt narration is also signaled through nonlinguistic means. Some cinematographic techniques—the dynamic camera angle that creates many diagonals; the abnormally
high or low horizon line; slow and fast motion; the extreme close-up that picks out a detail; the 28-mm lens that distorts space; vignetting and soft focus—were quickly identified with the Soviet cinema, but despite their often cliched employment, we must see them as striving to suggest a narrational presence behind the framing or filming of an event. It is here that Pudovkin's concept of an "ideal observer" has some relevance. Critics were quick to spot and personify this camera eye; one wrote of Potemkin: "It is like some grotesque record of a gargantuan news photographer with a genius for timing and composition."9
"Realistic" though such films as Potemkin and The End of St. Petersburg were often felt to be, the staging of the action tends to create highly self-conscious narration. The set may present a perspectivally inconsistent space, as in the warden's office in The Ghost That Never Returns or in the cafe in The New Babylon. Lighting may also be manipulated, as when in Storm over Asia the cut-in close-ups of the fox fur are lit in ways completely unfaithful to the overall illumination of the Mongol home. Figures are often placed against neutral background, either realistically motivated ones (a peasant or worker fiercely silhouetted against a cloudless sky) or more stylized ones, as in the initial attack on the woman on the Odessa Steps (fig. 11.1) or the abstract cut-ins from The End of St. Petersburg which we examined earlier (fig. 7.50-7.55). The figures will often be placed in unnaturally static poses as well. While Dovzhenko made the most systematic use of this, we find the device in other films as well: in The Ghost that Never Returns, characters freeze in place during an attempted suicide; in Twenty-six Commissars, a crowd listens to a speech while standing in abnormally fixed postures. In contrast, the figure behavior may be what was called at the time "grotesque" or "eccentric"— stylized figure movement that makes the scene difficult to construe as a real event. Strike's dwarfs and clownish bums are usually cited here, but we could add the petty thieves in The Devil's Wheel, Kerensky and company in October, the priest in Earth, and the prison warden in The Ghost That Never Returns.
What gives the narrational presence away completely is
238
HISTORICAL MODES OF NARRATION
11.1. Potemkin
the propensity for frontality of body, face, and eye in these films. We have seen how the classical film favors a modified frontality of figure placement; our sight lines are marked out, but the characters seldom face or look directly toward us. The Soviet cinema tends to stage the action much more frontally. Furthermore, the characters frequently look out at the camera. Sometimes this is motivated as another character's point of view, but not nearly as often as it would be in Hollywood. And at some point, frontality becomes an unabashedly direct address to the camera. Again and again characters turn "to us" without the slightest realistic motivation. The end of Strike, with its staring eyes, is probably not the best example, since such concluding confrontations form a minor convention of classical epilogues too.10 But when, in the middle of a scene, a soldier gazes out and asks us, "What am I fighting for?" (End of St. Petersburg), or when a character confides in us (Zvenigora), or mugs and winks at us (Lace), or asks whether it is all right to kill the enemy (Arsenal), or turns to us for help during a fistfight (Twenty-six Commissars), we must acknowledge that narration is not simply relaying some autonomously existent
profilmic event. Now the narration overtly includes the profilmic event, has already constituted it for the sake of specific effects. Ideas of montage within the shot, montage "before filming," and montage "within the actor's performance," so current in the late 1920s, testify to filmmakers' notion that narration should include self-conscious manipulation of the profilmic event, the material that normally pretends to go unmanipulated. This narration is not only omniscient; it announces itself as omnipotent.
What brings together film techniques like intertitles, cinematography, and mise-en-scene is the key concept of Soviet film theory and practice: editing, usually called montage. As conceived in Soviet artistic practice during the 1920s, montage in any art implies the presence of a creative subject actively choosing how effects are to be produced. Summarizing the views of many practitioners, Felicie Pastorello writes aptly: "Montage is an act (and not a look), an act of interpreting reality. Like the engineer and the scholar, the artist constructs his object, he does not reproduce reality."11 In objecting that Soviet montage "did not give us the event; it alluded to it," Bazin was putting his finger on exactly this refusal to treat film technique as a neutral transmitter.12 The didactic and poetic aspects of Soviet cinema meet in a technique which insists, both quantitatively and qualitatively, upon the constant and overt presence of narration.
It comes as no news that Soviet montage films rely upon editing, but some comparative figures may spruce up the obvious. The Soviet films I am considering contain between 600 and 2,000 shots, whereas their Hollywood counterparts of the years 1917-1928 typically contain between 500 and 1,000. (I am counting intertitles as shots.) Hollywood canonized the average shot length as five to six seconds, yielding a common figure of 500-800 shots per hour. The Soviet films, however, average two to four seconds per shot and contain between 900 and 1,500 shots per hour. This means that only the very fastest cut Hollywood films of the teens (such as Wild and Woolly) approach the Soviet standard, while the fastest-cut Hollywood films of the 1920s fall at the slower end of the Soviet scale. And nowhere in Hollywood filmmaking of any period can one find editing as quickly paced as in
HISTORICAL-MATERIALIST NARRATION
239
the most rapid montage films: an average shot length of under two seconds in Potemkin, Deserter, Goluboi Express, and A Simple Case.
The reliance on cutting has qualitative consequences. In the Hollywood film, especially after the coming of sound, a few sequences will be fairly heavily edited while others will contain longer takes. By rejecting such a "crossbred" de-coupage, the Soviet films provide a ubiquitous and constant level of rhetorical intervention. This cinema goes beyond those narrational asides which we found in the art cinema; these films do not offer a reality (objective, subjective) inflected by occasional interpolated "commentary"; these films are signed and addressed through and through, the diegetic world built from the ground up according to rhetorical demands.
Thus there are always more cuts than needed for lucid cueing of fabula construction. Even the simplest gesture may be broken into several shots. Crosscutting endlessly juxtaposes actions in different locales. By virtue of what the Soviets called "concentration" cuts, a simple transition from long shot to medium shot gets splintered into several shots. Jump cuts break up a single camera position. Montage also operates on intertitles: in October, the narration chops speeches into brief phrases. The relendess presence of montage in these films aims to keep the spectator from construing any action as simply an unmediated piece of the fabula world. Whereas Bazin worried that cutting changed the profilmic event from something real into something imaginary, the Soviet filmmakers believed that not cutting would change the syuzhet from a rhetorical construct into something (falsely) descriptive.
Montage makes the narration self-conscious in yet another way: through rhetorical tropes. The Soviet films furnish an anthology of both "tropes of thought" and "tropes of speech." The former are buried or ellided formal arguments, such as the schematic argument-from-example that under-girds the Soviet film and the tendency of the narration to argue by analogy (as when crosscutting links two social agents and makes us infer a shared motive or political view: bourgeoisie/police, proletarian/peasant). Tropes of speech,
or figures of adornment, can be mimicked by editing too. These films teem with rhetorical questions, metaphors or similes (the bull and the strikers in Strike), synecdoches (a general's medals substituting for the general in The End of St. Petersburg), personifications (the squirming concertina in Arsenal), understatements, hyperbole, antitheses, and many other classical figures. October uses paranomasia, or punning, when the narration presents Kerensky's political rise as a climb up an apparently endless flight of stairs; the play is based on the Russian word lestnitsa (stairs), as used in the phrase ierarkhicheskaia lestnitsa, or "table of military ranks." In the same film, the intercutting of Kerensky with a Napoleon statue cites the simile Lenin used in a 1917 Pravda article, "In Search of a Napoleon," while the montage of statues and artillery probably is meant to revive Lenin's synecdoche "With icons against cannons."13 The prominence of stylistic organization in these films cannot be read as sheer artistic motivation; the didactic ends often make film style operate as compositionallý justified ornamentation.
All these techniques invest the narration with a high and consistent degree of overtness in all the respects we have been considering since Chapter 4.
Degree and depth of knowledge. The narration of these Soviet films is omniscient. The conventional knowledge-ability afforded by crosscutting is particularly visible in these works because the crosscutting is not only that of a last-minute rescue: crosscutting is constantly drawing marked comparisons. Firing cannons áre likened to popping champagne corks (Goluboi Express). While a boy is borne to his grave, his lover is at home, in an ecstasy of despair (Earth). More unusually, the syuzhet will "flash back" without the motivation of character memory, as when at the close of Old and New the narration gives us glimpses of earlier scenes of Marfa's struggle. The narration may also overtly anticipate what will happen later in the film. The most striking example comes from the opening of Storm over Asia, where shots of landscapes are interrupted by near-subliminal flashes of the saber that the protagonist will wield in the last scene. The narration likewise has no need to justify spatial manip-
240
HISTORICAL MODES OF NARRATION
ulations by character knowledge: we can cut to any locale. In Potemkin, as the marines prepare to fire, the narration cuts away to the bugle, the imperial crest, and other objects which yield ironic juxtapositions. In The End of St. Petersburg, the narrator can situate the political activity in relation to lyrical landscapes. In The Ghost That Never Returns, when the police agent fires his pistol, the narration prolongs suspense by holding on such details as drifts of sand and a hat rolling in the wind. In Lace, a quarrel is interrupted by cutaway shots of a poster on the wall.
Communicativeness. The narration's authority rests in part on its refusal to withhold what the mode defines as crucial fabula information. Such information includes the story's historical context, political arguments, and character background. The film's fabula action consists either of the struggle of a protagonist to achieve a goal or of the growth of a spontaneous protagonist to socialist discipline and awareness. It is this linearity that the narration respects. The syuzhet does not equivocate about characters' motives or behavior. The exposition is concentrated and preliminary, furnishing relevant and valid information about the characters' pasts; there will never be what Sternberg calls "anticipatory caution," let alone a "rise and fall of first impressions." The narration, in fact, takes the opportunity to be "overcommunicative" by using many devices that ensure redundancy: conformity of character to type, of type to situation, or of situation to historical-political presuppositions. In Ivan, a street loudspeaker will often reiterate the narrational information already supplied by other means. The celebrated overlapping editing of Soviet practice displays not only the narration's authority (ability to restage the profilmic event, to "remount" it in editing) but also the narration's urge to insist on certain gestures. Scenes like that of the woman running through the doors in Ivan and the cream separator test in Old and New resemble traditional oratorical amplifications of set topics (grief, success).
Self-consciousness. We have already seen the extent to which camera position and lens length, frontality of figures, static poses, to-camera address, and the constant use of montage all create the sense of a self-conscious address to
the audience. The expository title can focus this effect. The narration can interject maxims (a quotation from Lenin in Potemkin), slogans ("All power to the Soviets!" in October), and rebuttals (in Goluboi Express, a reactionary cries, "Stop the train!" and an expository title shoots back: "But can you stop a revolution?"). The narration will also usurp the characters' own voices. In many Soviet films, information that could easily be given in dialogue titles will be supplied by expository titles, as in the beginning oiThe End of St. Petersburg, when the peasant family must send some members to work in the city. In one episode of Twenty-six Commissars, the narration becomes a witness's testimony to the action. And some titles could plausibly come from the fabula world but, because they are not signaled as quotations, instead suggest that the words are routed through the narration. Moscow in October intercalates an orator and expository titles, while in Arsenal, we cannot locate a speaker for such lines as "Where is father?" Nothing could be stronger evidence for this tendency than the insistence on retaining exhortatory expository titles after the arrival of lip-synchronized sound. In the remarkable Mountains of Gold, expository titles repeat what we have already heard a character say, and they even argue with a speaking character! Unlike their contemporaries in Europe, who envisioned the titleless film as the goal of a "pure" experimental cinema, the Soviet filmmakers saw the linguistic resources of the expository title as an instrument for rhetorical narration.
Attitudinal properties. The very constitution of genres and the didacticism of the narration in this mode make the narration openly and unequivocally judgmental, often satirically and ironically so. Judgments can be carried by interti-tles, especially in the exposition: how many Soviet films begin by rendering an oppressive state of affairs in the images and then interjecting ironic titles ("All is calm etc.)? The narration throws its voice to cheer for the opposition or quotes characters to mocking effect (the figure known to classical rhetoric as "transplacement"). In Goluboi Express, decadent bourgeois proclaim, "Ah, Europe, culture, civilization"; later the narration intercuts the same phrases with statues, policemen, and troops. In October, the
HISTORICAL-MATERIALIST NARRATION
24I
2. Mountains of Gold
Bolsheviks arrested during the July Days are called traitors and spies; later, when Kerensky releases them to defend Petrograd, the narration sarcastically recalls the epithets. The ne plus ultra of this process may be seen in the intercutting of batdefield and stock exchange in The End of St. Petersburg, in which the same phrases ("Forward!" "The deal is over!" "Both parties are satisfied!") apply with brutal irony to both milieux. Once this "tone of voice" has been established, the images can reinforce it by typage (grotesque costumes and demeanor of the bourgeois types, valorization of protagonists), camera work (the low angle as connoting power or solitude), lens length (the wide-angle lens for distortion and caricature; see fig. 11.2), and music (e.g., comic music to parody the opposition); The specific rhetorical tropes already mentioned will often, of course, work to judgmental effect as well.
Predictable Fabula, Unpredictable Narration
By treating the syuzhet as an argument by example, and by gathering a powerful rhetorical thrust, the Soviet historical-
materialist cinema created a distinct organization of narration, with effects on cinematic style already discussed. Another result was an idiosyncratic approach to the spectator, one that is neither as "totalitarian" as liberal-humanist critics often assume nor as radical as some recent theorists of textuality have claimed. The films' mixture of didactic and poetic structures calls for viewing procedures which deviate from classical norms yet remain unified by protocols specific to this mode.
Broadly speaking, the viewer brings to these films a few highly probable schemata. Already-known stories, drawn from history, myth, and contemporary life, furnish a fairly limited range of options for the overall cause-effect chain. Knowledge of the different genres, especially when the film treats a historical subject, further limits what can plausibly happen. The viewer also possesses a sense of how the mode creates character and signals salient conflicts. And the ending is likely to be known, at least in general outline. In syuzhet terms, the narration further strives to eliminate any ambiguity at the level of causality (motives, goals, preconditions) or at the level of the rhetorical point made. Most narrational difficulties presented by these films cannot be explained under the rubrics of realism or subjectivity; the problems are clearly marked as proceeding from the self-conscious narration. On the whole there is little room for the gamelike equivocations and the interpretive subtlety valorized by art-cinema norms.
These films therefore sacrifice many resources of other narrational modes. There is relatively little curiosity about how events came to be as they are; macrosocial historical causes are often taken for granted. Suspense is limited to questions of how the inevitable will occur or, in the case of characters who are not "public" personages, whether the character will survive, move to correct consciousness, and so forth. The syuzhet may assume that because the historical event or rhetorical point is already known, not all of the links need to be shown. In Deserter, the process of converting the German worker Renn from a traitor to a good proletarian is completely skipped over; the narration simply assumes that a stay in the Soviet Union suffices to bring him around. The
242
HISTORICAL MODES OF NARRATION
end of Potemkin neglects to mention that the rebelling sailors were eventually captured, but the viewer is supposed to understand that whatever the outcome of this episode, the entire 1905 revolution was a harbinger of 1917. Moreover, if there are political disputes within Soviet communism about the case considered, it is often wiser for the filmmaker to omit explanation than to risk being criticized. Vance Kepley has shown that many elliptical moments in Dovzhenko's films result from skirting sensitive issues.14 We shall later see how The New Babylon tries to avoid disputes about why the Paris Commune failed. Again, the omnipotent narration works as a reliable guide: any "permanent" breaks in the causal chain signal not a lack of communicativeness but a tacit appeal to the audience's referential schemata.
The historical-materialist film compensates for its limited narrative schemata by unusually innovative spatial and temporal construction. If the story outline is often predictable, stylistic processes often are not. At the barest perceptual level, narration will jolt the spectator. Consider the opening of Twenty-six Commissars:
1. Long shot: Oil field
2. Tide: "Baku"
3. Explosion
4. Title: "1918"
5. Explosion
6. Explosion
This is our introduction to the revolutionary brigade. Strike begins with abstract shots of the factory, including silhouettes and an upside-down, reverse-motion reflection of the factory in a puddle. The narration of Deserter establishes the river docks in a lyrical tranquillity before startling us with shots of chains dropped from ships—shots that intersperse black frames with bursts of imagery and thus create an almost annoying flicker. The conventionality of the large-scale narrative articulations promotes a moment-by-moment "microattention" to the unfolding syuzhet. Like the orator embroidering a commonplace, the narration takes for granted that we understand that part of World War I was fought around Baku, that Strike will be about a workers' walkout, that Deserter is set in a dockyard. The task is to
make these givens vivid, or as the Soviet directors were fond of saying, perceptible.
What renders these stylistic processes more unpredictable than the procedures of classical narration? Most obviously, the Soviet films I am considering define themselves against many spatial and temporal norms of classical Hollywood narrative. All the procedures of titling, cinematography, editing, and mise-en-scene I have already mentioned constitute an alternative stylistic paradigm. Eyelines will not necessarily cut neatiy together; characters will not necessarily ignore the audience; framing will not necessarily be symmetrical or centered. Similarly, principles of spatial and temporal continuity, of tight linkage of cause and effect, and so forth do not hold in this mode. As in the art cinema, style becomes more prominent here because of its deviation from the classical norm.
To the extent, however, that the Soviet devices function within a paradigm, the viewer can apply schemata based on this extrinsic norm to make sense of the films. But this process is more difficult than in the classical mode because of the great emphasis the Soviets placed upon deviating from extrinsic norms. Again as in the art cinema, variations often proceed from authorial differences: Dovzhenko is more likely to use slow motion than Eisenstein is, Room is more apt to match shots "classically" than are his contemporaries. Still, nothing in Strike prepares us for the alternating of two successive scenes in Old and New; nothing in Mother anticipates the montage of black frames in Deserter. It is not just that the filmmakers developed; the search for ever more "perceptible" effects pushed them to try new devices in every film. In general, narration became more elliptical, images became briefer, gaps became greater, fabula events underwent more expansion and amplification. Virtually any device—soft focus, slow or fast motion, upside-down camera positions, single-source lighting, handheld camera movement—could create a film's distinctive intrinsic norm. It would be up to the viewer to make sense of the unpredictable procedure by slotting it into accustomed syuzhet functions and patterns. We have already seen this at work in our examples of spatial discontinuity in Earth and The End of St.
HISTORICAL-MATERIALIST NARRATION
243
11.3. Earth
Petersburg in Chapter 7. Because each film strives to attain great stylistic prominence—-the intrinsic norms marking significant differences within the "Soviet style" itself—the v !i wer must use the extrinsically normalized principles as guidelines. The task, as in art-cinema narration, is to grasp each film's unique reworking of the paradigm. This is done by calling on procedural schemata that urge: when in doubt, construct a fabula event as perceptually forceful and politically significant.
Faced with the shocks of this jarring style, the spectator can, at least up to a point, deal with it cognitively. The 11111 lortant strategies are those of "filling in" and "linking and distinguishing." Such activities form a part of the viewer's task in any narrational mode, but, with Soviet montage cinema, they play a major role at the level of temporal and spatial construction.
The very idea of montage demands that we fill in gaps. As Shklovsky put it in describing intellectual montage, the editing works "through its non-coincident components— its aureoles."15 Every shot change offers the filmmaker a chance to create a break in time and space. Classical editing usually avoids perceptible gaps at this level; at most they are suppressed or temporary. Soviet montage flaunts its spa-tiotemporal gaps and will not always plug them. The Soviet tendency to minimize or omit establishing shots asks the spectator to fill in the overall milieu. For similar reasons, the So\iet directors never canonized the over-the-shoulder reverse shot; instead of this extra cue that the classical style provides we are often presented with no clear information about characters' distances or angles of interaction. Thus when the cutting pattern violates the 180-degree rule of Hollywood practice, the viewer must construct a set of hypotheses about character position. Entire sequences (e.g., Pax el's trial in Mother) or whole films (Earth) can make sense on the basis of comparatively few cues for characters' placement.
From the elimination of the establishing shot the Soviet directors drew two conclusions, one quite radical. First, you need not find or create an entire profilmic event: partial views can create a locale that need never have existed in
front of the lens. The spectator will infer a unified space based on assumptions about real spaces and about the sort of space that films usually present. The more radical discovery was that viewers could be asked to unify spaces in physically impossible ways. Supplied with strong spatial cues, such as character eyelines or earlines, the spectator will infer an "abstract" space that could not exist empirically. In Twenty-six Commissars, the Bolshevik prisoners are massacred in the desert. A wounded man staggers to the top of a hill and shouts: "Be calm, comrades!" There is a cut to the oil fields of Baku, many miles away. Suddenly workers in the fields freeze in place, as if hearing his cry. There follows a series of shots in which a striker at Baku "watches" the execution of the commissars. And after the massacre, the workers stand in silent homage before a spectacle they could not possibly see or hear. Comparably "abstract" spaces can be found in many Soviet films; as we shall see, The New Babylon relies on them to a considerable degree.
The spectator must fill in temporal gaps too. Here is a passage from Earth:
1. Medium shot: In his house the father bellows (fig. n-3)-
244
HISTORICAL MOOES OF NARRATION
ii.4. Earth j 1.5. Earth
11.6. Earth
11.7. Earth
11.8. 11. g.
Earth Earth
11.10. Earth
11.11. Earth
2
3
4
5'
6
7
8,
9
10.
11.
12. 11.7),
13-14.
"Ivan!"
Long shot: Against sky, he calls, rightward (fig. 11.4). "Stephen!"
Medium shot: He calls, rightward (fig. 11.5). "Grigori!"
Medium close-up: He calls, leftward (fig. 11.6).
"Have—"
"you killed—"
"my—"
"Vassily?"
Long shot, as (3): The father looks straight out (fig.
Extreme long shot: Empty landscape (fig. 11.8). Medium shot: Over father's shoulders, two men together (fig. 11.9), Track back with father as he strides to the camera, revealing a third man in the background (fig.
11.10) .
15. Medium shot: The father walks up to Khoma (fig.
11.11) .
The narration has created a spatial gap—the abrupt transition from the house to the outdoors in shots 1-3—and some temporal ones. If the father shouted "Ivan!" in the house, we must assume that he consumed time in getting out to the hillside. Yet the rhythmic alternation of tide and image suggests that perhaps "Ivan!" was shouted outside too. This yields an ambiguity about the frequency of the fabula event. Later, after the father has hollered and apparently gotten no response (shots 12-13), another cut takes us immediately to a group of three men (shot 14)—presumably those he summoned by name. Without warning, the cut has skipped over the fabula duration required for the group to assemble. But when the father turns and walks away, shot 15 reveals that a fourth man is present—Khoma, the youth who did kill Vassily. His arrival has been withheld for the sake of surprise. Dovzhenko's style is unusually oblique, but his reliance on ellipses is only an extension of a general Soviet tendency to ask the spectator to see any cut as embodying a possible break in fabula time.
Because these Soviet films suggest that we fill in missing pieces of space and time, the spectator must tolerate a degree
HISTORICAL-MATERIALIST NARRATION
245
11.12. Potemkin
11.13. Potemkin
f of cognitive strain. At the start of a sequence, we may be "^uncertain about exactly what is happening; the narration ^has plunged us abruptly into a stream of details. We must < patiently trust that the narration will eventually clarify or fjustify what seems unsettled. Early in Old and New we see f rnen sawing timbers while families look on; only gradually do we grasp (thanks chiefly to an intertitle) that brothers are dividing their property by sawing the family house in two. In '-Arsenal, the fight for possession of the locomotive is inter-;rupted by a series of very close shots of a woman turning to the camera and leaping up; cut back to the locomotive; only after this do we get a shot that establishes the woman at the , telegraph key in a railroad office. It is as if the narration, rushing to give us the emotional core of the situation, later takes the opportunity to flesh out time, place, and causality. In Chapter 7, we have already seen how sequences in Earth • and End of St. Petersburg create "open" spatial relations that only eventually get closed: father and son quarreling (back to back or not?), troops firing (on the Bolsheviks or on the general?). In sum, the stability of broad causal schemata in these films allows the narration to create a process of hypothesis testing in a film's moment-by-moment unfolding. Film style works to retard the likeliest meaning, and the spectator adopts a wait-and-see strategy.
Occasionally we wait and never see. Some spatiotemporal gaps we can never close at any denotative level. At the end of the Odessa Steps sequence in Potemkin, the baby carriage jitters down the steps, intercut with shots of the staring woman with the pince-nez. Then:
1. Medium shot: The carriage begins to flip over (fig. 11.12).
2. Medium shot: A swordsman starts to swing his saber (fig. 11.13).
3. Close-up: He slashes downward (fig. 11.14).
4. Close-up jump cut: He slashes (fig. 11.15).
5. Close-up jump cut: He draws back and starts to slash again, shouting (fig. 11.16).
6. Close-up: Blood runs from a woman's eye, and her pince-nez is shattered (fig. 11.17).
We can, I think, construct the fabula action in several
ways. (A) The soldier has slashed at the woman with the pince-nez. Reasons: shots 2-6 can be construed as a group, making shot 6 a reaction shot; the frontality of the soldier's attack (perhaps a subjective point of view) is congruent with that of the woman's orientation. (B) The soldier has slashed
246
HISTORICAL MODES OF NARRATION
11.14. Potemkin
11.15. Potemkin
at the baby in the carriage. Reasons: shots 1-5 hang together; the cossack is observed from a low angle, befitting his assault on the carriage; the woman has earlier been seen some way up the steps; the woman's wound is not plausible as coming from a saber. (C) The baby carriage overturning,
11.16. Potemkin
11.17. Potemkin
the cossack slashing, and the wounding of the woman are unconnected events, crosscut. Reason: all the inadequate and incompatible cues present in (A) and (B). (D) The cossack slashes at both the carriage and the woman: an "impossible" profilmic event. Rather than decide on a single
HISTORICAL-MATERIALIST NARRATION
247
x8. Potemkin
11.19. Potemkin
1*
1
construction, we should recognize that exactly this mixing oi cues, this shaking of scenic components loose from a iinivocal fabula world, enables the narration to create an "open" space from which can be selected maximally forceful images of brutality—with five of the six addressed directly to the viewer. The spatial gaps become permanent ones, creating vivid rhetorical effects.
The act of filling in must then include our willingness to accept, in the name of perceptibility, very great violations of 1 onventional or internally consistent space and time. What 1 lse can explain the spectator's assimilation of shots in which the film strip is flipped side-to-side or upside down? The fabula event may be presented not as ambiguous but as contradictory: an officer sits in inconsistent positions from shot to shot (Storm over Asia); a coolie is slapped once, but in a different way from shot to shot (Goluboi Express); a worker assaults his boss in two locales at once (The End of St. Petersburg); a priest raps his cross in one palm, then—or rather also—in the other (Potemkin; see figs. 11.18-11.19). The Soviet directors assumed that if syuzhet material cannot be unified at the denotative level, the spectator will look for ways to unify it connotatively. Thus ideologically defined
argumentative schemata and the explicit and constant presence of a narrator allow the viewer to place incompatible presentations within a larger affective dynamic.
Besides filling in gaps, the spectator must link and distinguish elements. One consequence of Soviet film's stress upon "perceptibility" is that we are expected to fine tune our sensitivity to the representation of space and time. Similarities and dissimilarities among images weigh more in this mode than in the classical narrative. Soviet directors are fond of calling on short-term memory in order to permute images in palpable ways, as Dovzhenko does in the Earth segment (figs. 11.3-11.11) or as Boris Barnet does in Moscow in October by varying the same shot (figs. 11.20-11.22). In Potemkin, the narration frequently cuts from one character to another as each executes a similar gesture (making a fist, running a machine); denotatively we must pick different individuals out of a smooth passage of movement (even while connotatively we must see them as linked in the performance of similar actions). By using editing to achieve temporal dilation, these films rely on the viewer's ability to construct one movement out of several overlapping representations onscreen. And some films, in particular Pudov-
248
HISTORICAL MODES OF NARRATION
11.20. Moscow in October
11.21. Moscow in October
kin's, utilize devices which lie on the very threshold of perceptual discrimination, such as sporadic black frames, single-frame montage, and barely discernible jump cuts.
The spectator's ability to draw likenesses and contrasts can work closely with the rhetorical aims of the mode. Storm
11.22. Moscow in October
over Asia features a celebrated sequence in which the British commander and his wife prepare to visit the Buddhist mmple. The narration crosscuts the couple's preparation— shaving, washing, dressing—with functionaries scurrying around cleaning the temple. More than temporal simultaneity is evoked here. The narration draws analogies between objects in each line of action: the temple feather duster is likened to the wife's powder puff, the priest's collar to her necklace. Expository titles remark ironically, "There are ceremonies / and rites / among all races." Since the immediate causal function of the scene is minimal, the fact that it is given extensive treatment invites the viewer to dwell on its rhetorical implications. The spectator must take the visual similarities between the British and the Buddhists as cues to a conceptual likeness; the intertitles reinforce the link. The rhetorical effect is double: to satirize fastidious upper-class hygiene, as solemn and self-righteous as a religious ritual; and to mock the church as a thing of this world, as vain as the decadent imperialists. Like many crosscutting episodes in Soviet film, this sequence asks the viewer to liken "unlike" things. Conceptual parallelism replaces causal logic as the basis of the syuzhet. Ultimately, however, these argumenta-
HISTORICAL-MATERIALIST NARRATION
249
tive connotations "feed back" into the causal nexus, since the similarity between imperial and Buddhist authorities anticipates the complicity of rulers to be exhibited during the colonialists' visit to the temple.
The locus classicus of this abstract tendency is the notorious "intellectual montage" of Soviet cinema, of which the Storm over Asia sequence could count as a fair example. But note that the narration can achieve high-level "intellectual"—that is, rhetorical—judgments in two ways. The possibility that entranced Eisenstein was what Metz has called the nondiegetic insert—one or more images that possess no denotative reality in the fabula world. The slaughter of the bull intercut with the massacre of the workers (Strike) is a pure case, as are the Kerensky/peacock comparisons and the "God and Country" sequence in October. More common Soviet practice, however, was the rhetorical combination of images taken from the diegetic world. The sequence from Storm over Asia is an instance: both the commander's household and the temple exist on the same level of fabula "reality." In fact, this second possibility proves to be the richer one, since it allows the narration to present images initially designed to denote fabula information and then to recall them for more connotative purposes. Eisenstein had a penchant for repeating identical shots in very different situations across the film. In October, images hailing the February revolution—for instance, troops with upraised rifles—are repeated during the October Revolution. After the Potem-kin's crew has pitched Smirnov overboard, the narration inserts an image of the maggoty meat that had precipitated the mutiny. Such a shot becomes what one theorist of the time calls a "refrain."16 At the start of Arsenal, a worker is shown in a very disjointed series of images; much later in the film, as urgings to strike sweep through the arsenal, the same worker suddenly looks up. The use of the refrain multiplies the potential functions of each montage fragment, making the film a collection of intrareferential bits frozen in a mosaic, a total "spatial" order.
It may seem odd that I have said so little about what is for most viewers the salient quality of Soviet montage: the speed of the cutting. All of the films I have picked out contain
passages of rapid editing, and some present shots only one frame long. Often this technique is motivated by violent action or by tense emotional confrontations; the rapidly cut battle scene or police attack is a convention of these works. Just as often, though, accelerated rhythmic editing functions as the narration's instrument. Fast cutting not only embodies causal climaxes but creates rhetorical ones. Any rapidly cut sequence becomes ipso facto significant (not least because fast cutting tends, paradoxically, to stretch out the syuzhet duration devoted to an episode). For the spectator, rapid editing is the most self-conscious effort of the rhetorical narration to control the pace of hypothesis formation. We have repeatedly seen that any rapid flow of fabula information, via editing or other means, compels the spectator to make simple, all-or-nothing choices about story construction. Under the pressure of time—certainly long before half a second—we must give up trying to predict the next image and simply accept what we are given. Soviet fast cutting takes care to combine and repeat shots or actions that we have already seen, so that we can gather a total impression from repeated bursts. Far from being passive subjects inundated by the film's spray of imagery, we continue to apply rhetorical and narrative schemata; we continue to fill in, to liken, to discriminate; but we do so at a suprashot level, unifying the sequence from the top down by using prototypes like "battle," "strike rally," "police attack," or whatever—all the while registering the sheer perceptual force of the style.17
The New Babylon
The film work of Grigori Kozintsev and Leonid Trauberg grew out of their experimental theater group, Factory of the Eccentric Actor ("Feks" for short). These young men were initially interested in achieving grotesque effects by manipulating the profilmic event. Feks's The Cloak (1926) transposes verbal grotesquerie (Gogol's skaz style) into visual terms through setting, costume, and acting. The stylization of the profilmic event serves to emphasize narrational in-