History and Ontological Realism

The preceding chapter referenced Soviet writer and theorist of socialist realism Maxim Gorky’s admiration for the romanticism at the heart of myths and the need, as he saw it, to bring that romanticism into realist representation in Communist art—a formula for Communist aesthetics that would become a core part of socialist realism in general and Mao’s “combination of revolutionary realism with revolutionary romanticism” in particular. In that light, although the romanticist element helped to imbue the films with mythical import and affective force, the challenge of the realist side of the formula was to make romantic myths convincing by grounding them in apparent reality, using all the tools of cinematic realism—whether ontological, perceptual, fictional, or social.8

The Chinese revolutionary films discussed in both this and the preceding chapter drew heavily upon founding myths of the PRC, including, in the case of war films, the Communist-led victories over the Japanese occupation of 1937–45 and the battle against the Nationalists or Guomindang during the civil war both before and after the Japanese occupation. Calling these founding myths, of course, does not imply that they never happened in actual history; rather, the repeated retelling of those stories in the most inspiring form possible, both to the generations that lived through the events and to the new generations that would grow up under Communist rule, was deemed essential for building communal identity and legitimizing the CCP’s leadership and the PRC state itself.

The narration of these myths in the form of feature-length fiction film was an exercise in creating the sort of “prosthetic memories” of historical events at which cinema had long excelled. Alison Landsberg defines prosthetic memory as “a new form of public cultural memory” under modernity that “emerges at the interface between a person and a historical narrative about the past, at an experiential site such as a movie theater or museum.”9 Through an experience like a public film screening, a “person sutures himself or herself into a larger history” in a way through which he or she “does not simply apprehend a historical narrative but takes on a more personal, deeply felt memory of a past event through which he or she did not live.”10 Because of the sort of personal, affective investment involved, “prosthetic memory has the ability to shape that person’s subjectivity and politics.”11 Landsberg theorizes that with modern technologies like photography and cinema, “it becomes possible to have a mediated memory that one nevertheless experiences as real or genuine.”12

Cinematic realism facilitates the credibility of a fiction film’s vision of history even though audiences understand that they are watching a fictionalized representation of past historical events rather than a direct documentation of those events. The ontological realism of the film image—the fact that photographic media directly record whatever real things are before the camera—plays a key, but somewhat devious, role in achieving the sense of credibility of the historical representation. Philip Rosen’s analysis of this kind of cinematic representation of history and the way it is bolstered by the ontological realism of film is premised on the fact that even a fiction film can be considered a documentary of its actors’ performances, directly recording those as well as any objects or settings (even if they are actually film props or studio sets) in front of the camera. A historical film thus offers “an image that represents not one but two past times”—the time of the historical events it references but also the time of the film’s making.13 Strictly speaking, only the latter representation possesses the ontological or “indexical” realism of the photographic image—the direct existential link between the people and objects being filmed and the resulting image. However, in connecting the fictionalized representation of history to the actual historical events referenced, the film in a sense transfers this ontological realism and thereby lends an “indexical credibility” to the fictionalized representation of history.14 Such films thus enact “a conversion from one time to another, from the past time of the camera’s operation to the temporal setting of a different past, that of the narrative,” and the persuasiveness of the fictionalized historical narrative is increased by the fact that “as photographic cinema there is a certain documentary force at work in these images.”15 The sense of “referential pastness” of the historical film is further increased by the careful use of historically convincing props, costumes, buildings, hairstyles, and countless other features that provide what Roland Barthes called the “reality effect” attained through the use of incidental details in the modern realist novel, in that such details serve “to convince the reader [or viewer] of a direct link between the signifier and referent by means of an impression of precise accuracy.”16 Here Rosen (via Barthes) borrows the terms of structural linguistics, in which a sign is broken down into a signifier and a signified, both of which are distinguished from an actual referent in the material world. For Rosen, the historical film on the screen is a signifier, with the signified being the fictionalized story acted out before the camera at a previous time—a story, however, that claims to have a referential relationship to a historical real before the time of performance. It is precisely through the ontological realism of the medium that a fictionalized historical film gains documentary force that transfers from the signified fictional performance to the historical referent itself. That is, to be a convincing representation of history rather than just the diegeticized fiction, the historical film must elide the signified (the performance) and seem to make a direct link from signifier (the film viewed) to referent (historical reality). The connection is precarious, because “the diegeticized profilmic, as historical representation, inevitably falls short of the kind of documentary authority claimed by the image as an indexical trace of the profilmic.”17 That is, the recording of the performance has an actual ontological realism—what Rosen calls indexical credibility or what Bill Nichols calls the indexical “stickiness” of a documentary film image—that the claim to historical representation lacks.18 However, Rosen suggests, “there is . . . a kind of passage or play between document and diegesis, but with a kind of residue, as if the diegeticized film attempts to retain something of the factual convincingness of the document.”19 The ontological realism of the image subtly but powerfully lends credibility to the film’s claim of historical representation, even though it only directly shows us a fictional performance.

To put it briefly in terms of the 1961 version of The Red Detachment of Women, discussed in chapter 4, audiences at the time of its release, even if they knew that there really was a women’s detachment of Communist revolutionaries on tropical Hainan Island in south China during the Chinese civil war, would also have known that they were watching a performance of fictionalized characters; historical films need not claim that their main characters are based on real historical figures if they can only convince the audience that this is what the real people were like. No original audience of Red Detachment of Women would have naively thought they were seeing history directly; they knew they were watching a fiction film. However, if the characters “came to life” through the performance, and most of the incidental details seemed historically accurate, the actual ontological realism tying the film to the performance could transfer and lend a “residue” of heightened realism to the claim of historical representation. It is not that the spectators suffer from the delusion that they really are directly seeing history but that the imprinted trace of the past in the moving image gives it, as Bazin says, “the irrational power to bear away our faith”—which, as Tom Gunning points out, makes Bazin’s argument about the power of ontological realism far from just a description of a logical process of using photographs as signs.20 In this way, cinema’s ontological realism or indexical stickiness helps to enable films to imprint as current generations’ prosthetic memories of past historical events even though the viewers do not naively and literally read the moving image as an actual indexical imprint from the historical events referenced.

Needless to say, this transfer of indexical stickiness or ontological realism from the fictional performance (Rosen’s “diegeticized film”) to the historical referent is in a basic sense a contrivance. Because the historical film’s ontological realism is in reality an index of the film’s performance rather than of its historical referent, such a film arguably has less of an actual indexical tie to the real historical event than would, say, a newspaper illustration by an early “newspainter” sent along as a correspondent to a war zone in the early days of modern print media or a print journalist’s written account of an observed historical event.21 Although paintings, drawings, lithographs, and newspaper stories lack the ontological realism of the photographic image, they are indexical insofar as the artist or writer uses firsthand, on-the-scene impressions to create the image or narrative. Nonetheless, the persuasiveness of film’s ontological realism provides a unique fuel for the credibility of cinema as historical representation.

Of course, such a crafty imbuing of indexical credibility to a historical fiction film is vulnerable to a sudden loss of persuasiveness, because the representation of history in fact lacks the “documentary authority” that it possesses as a document of a performance. The “passage” or transfer of such credibility to historical representation can suddenly break down for any number of reasons—from a character too obviously having a hairstyle from the time of filming rather than the historical time represented to a changed context of reception, whether following a cultural or ideological shift or just a different perspective afforded by multiple viewings of the same film. In any of those cases, one can suddenly start to view the performance as a performance rather than as a convincing historical representation.

Autonomization and Derealization

The sense of realism cultivated by Chinese revolutionary historical films is in continual tension with the films’ conventions, whether they be the generic conventions of the war film, the conventions of Chinese socialist realism, or the conventions of ballet or Beijing opera in the case of the yangbanxi films. Conventions serve to orient the viewer and tell the story but also potentially call attention to themselves in a way that undercuts the realism of the historical representation. The tradition of Chinese revolutionary cinema not only gradually became less realist and more romanticist but in fact increasingly displayed and eventually culminated in a very distinctively Chinese version of what David Bordwell has called “socialist formalism.” Bordwell has used the term to refer to Soviet socialist realist filmmakers who drew upon their more avant-garde 1920s predecessors to create moments of formalism even during the Stalinist period as well as later Soviet filmmakers who conducted similar explorations in parallel with the renewed interest in formalist montage during the French New Wave.22 In the case of China, I use socialist formalism to refer first to the more stylistically formalist moments in Chinese revolutionary films of the Seventeen Years—the appearances of flamboyant montages, extreme close-ups, superimpositions, dramatic off-screen gazes, and so on that tend to happen at “epiphanic” moments as discussed in the preceding chapter—and second in particular to the yangbanxi films, which are among the most formalist instances of mass cinema in the history of the medium, comparable, say, to Hollywood musicals of the first decade or so of the sound era, such as The Wizard of Oz (Victor Fleming, 1939).

But the phenomenon discussed here is not only a matter of whether cinematic style is more realist or more formalist; it is also a question of the developmental dynamics of cinematic conventions or stereotypes. Returning to Jörg Schweinitz’s theory of film stereotypes discussed in earlier chapters, we can identify a gradual process of what he calls derealization. Character types within a genre, for example, following a process of increased conventionalization and autonomization, may eventually be “experienced not so much as direct representations of reality but as roles played out in an imaginary world.”23 Character stereotypes in this way become less “vehicles of knowledge” about reality or history than “personified agents of a repeatable experience of pleasure found in a ritualized, self-similar game constantly offered anew by the individual films of a genre.”24 In other words, there is an intrinsic tendency toward derealization or denaturalization that occurs with the repetition of genre stereotypes.

Taking the example of the villainous landlord from the preceding chapter, immediate audience recognition of this character type was facilitated by its repetition across many films as well as literature and drama of the revolutionary period. Krista Van Fleit has shown how Chinese Communist critical debates about the issue of “typicality” from the 1930s to the 1950s sometimes centered on the landlord as type.25 The stereotyping of the landlord in cinema was further bolstered by the phenomenon of “villain stardom,” in which the same actors were enlisted to play similar roles in different films, such as Chen Qiang’s 陈强 repeat performances, a decade or so apart, as evil landlord in The White-Haired Girl and the original Red Detachment of Women.26 (In the latter, the landlord’s name, Nan Batian, means literally “tyrant of the south,” as if the audience needed more evidence of his stereotypical villainy.) The tendency toward derealization or consumption of the landlord stereotype as stereotype would have reached its peak with the yangbanxi films, including the remakes of Red Detachment of Women first as a revolutionary model ballet film in 1970 and again as a revolutionary model opera film in 1972. In a study of gender in the model operas, Rosemary Roberts has argued that counterrevolutionary villains in those works are systematically emasculated and feminized as part of an overall gendered system of representation. She notes that in both the model ballet and model opera versions of Red Detachment of Women, the villainous landlord and his cronies have longish hair in a style more like the female characters in the same films than like the heroic male characters, and they wear the kinds of traditional silk gowns that had been rejected as effeminate in twentieth-century China.27 The emasculation of Nan Batian in these films is also expressed by the shortened stature of the submissive crouches he tends to assume in both films when in the presence of the films’ hero, Hong Changqing (who replaces the slave-turned-revolutionary Wu Qionghua as the main protagonist in the Red Detachment yangbanxi films, in contrast to the original film, which had focused on her as heroine). These instances of systematic gendered caricature would have increased the sense of derealization of the landlord character type, who would have appeared less as a realistic representation and more as simply a genre stereotype—a reading reinforced by the shift from the more classical fictional realism of the original film to the opera aesthetics of the final version, because Chinese opera in general relies on repeated and relatively rigid character role types that are immediately identifiable through signifiers of make-up, costuming, and performance.

The derealization that results from the repetition and autonomization of character types occurs as well with other types of conventions, such as repeated narrative scenarios. For example, the preceding chapter discussed the repeated scenes of execution and martyrdom in Mao-era revolutionary films. The idea of a revolutionary hero or heroine going bravely to his or her death in sacrifice to the revolution had maintained a powerful hold on the imaginations of Chinese youths going back to the late Qing dynasty, when idolized heroes included Russian nihilist revolutionary Sofya Perovskaya, who had been executed by hanging in 1881 at age twenty-seven, as well as the young rebels against Manchu rule Zou Rong 邹容 and Qiu Jin 秋瑾, who became celebrated revolutionary martyrs at ages twenty and thirty-one, respectively, during the final decade of the dynasty.28 In Chinese cinema, scenes of revolutionaries being apprehended and appearing gladly to accept execution go back at least to Daybreak in 1933. At the end of that film, the heroine, Lingling, who has transitioned from country girl to factory worker to prostitute and, finally, to Republican revolutionary, primps her hair and straightens her clothes before being led to the firing squad—a conventional detail still evident, for example, in the 1971 popular revolutionary model opera film The Red Lantern (Hong deng ji 红灯记; Cheng Yin 成荫).29 By the time of the latter, several other conventions had become standard in scenes where martyrs are executed—scenes repeated across dozens of Chinese revolutionary films, including The Song of Youth and all three versions of Red Detachment of Women, in addition to many more.30 Aside from the ritual primping of hair and straightening of clothes in preparation for a dignified death, such conventions included the playing of the Communist “Internationale” on the nondiegetic soundtrack during scenes of execution, the martyrs shouting “Long live the Communist Party!” and “Long live Chairman Mao!” just before dying, and often a cutaway to an extreme long shot of a sublime natural landscape immediately after the death (which almost always remains off-screen). As with other revolutionary film conventions, execution scenes attained their most rigorously formalized expression in the yangbanxi films, specifically The Red Lantern opera film and the Red Detachment of Women ballet and opera films.

Such repetitions of revolutionary film conventions have important consequences in terms of their impact on the viewer. First, audience affective experience will shift significantly from fiction emotions to artifact emotions. Carl Plantinga uses the former term to label emotions that “take as their object some element of the film’s fictional world” and the latter for emotions that “take as their object the film as a constructed artifact.”31 Fiction emotions are stirred by the process generally called “identification” with fictional characters (though Plantinga himself avoids that term because of its psychoanalytic connotations). In the original Red Detachment of Women, when Changqing is burned at the stake while his would-be lover Qionghua watches from afar, the spectator presumably feels intense admiration for his bravery as well as strong sympathy for the agonizing loss she is suffering. Artifact emotions, in contrast, might include annoyance at the predictability of the screenplay (Changqing shouts “Long live the Chinese Communist Party!” as his dying words), admiration for the outstanding ballet dancing skills of the performer playing Changqing in the model ballet film version of the story, or appreciation of how the cinematographer’s mobile framing adds dynamism and variation in shot scale to the lengthy shots of the character dancing and singing in the opera film version during the minutes before his execution. Such emotions are stirred not by immersion in the story world but by the film as a constructed and performed work of art. It stands to reason that as genre conventions, through abstraction and repetition, become derealized and recognized more consciously by audiences as conventions, audience affective experience will shift at least somewhat from fiction emotions to artifact emotions, appreciating not so much the fictional realism of the films’ worlds as the skill or artistic invention with which expected conventions are repeated or varied. It also seems likely that the yangbanxi films would have brought such a shift to a culmination, particularly given that Chinese opera has traditionally been viewed in a way that emphasizes artifact emotions; when a seasoned Beijing opera viewer shouts hao 好! (bravo!) while viewing a performance of a canonized opera scene, the object of approval is less the fictional event or character than the artistic excellence of a particular singing performance, acrobatic sequence, or standardized pose (liangxiang 亮相) struck by a particular performer.

Aside from shifting the spectator’s affective experience from fiction emotions to artifact emotions, the endless repetition of revolutionary film stereotypes also would have altered the films’ perceived relationship to history, in that film scenes increasingly would be read as referring to other film scenes employing the same conventions rather than to the prior historical reality that is referenced. As Mikhail Iampolski put it in his study of intertextuality in cinema, a seeming quotation of one film within another “violates the link between sign and objective reality (the mimetic link), orienting the sign toward another text rather than a thing.”32 In other words, such intertextual citations work at cross-purposes with the effect of ontological or indexical realism cited earlier in our discussion of Rosen’s ideas of film and history: whereas Rosen saw the historical film as effecting a direct link from signifier (the film) to referent (history), short-circuiting the signified (the recorded fictional performance), Iampolski implies that intertextual repetition calls attention to the signified (the fictional performance) because of its relation to other fiction films, thereby lessening the effect of realism (“the link between sign and objective reality”). In Rosen’s terms, a yangbanxi film of the Cultural Revolution in particular would have been experienced not as an accurate representation of history but as “a document of performed virtuosity” owing to its “foregrounding of artifice.”33 History is converted into conventionalized spectacle in a way that loosens the film from its referential claims to historical reality.

In this context, we can add to the trilogy of Red Detachment of Women films another series of remakes to consider: as mentioned in the preceding chapter, the 1955 black-and-white revolutionary war film Guerrillas on the Plain was remade twice in 1974, as a color film in the same classical mode as the original and also as a revolutionary model opera film under the slightly different title Fighting on the Plain.34 This set of examples allows us to look not only at how classical-style fiction films of the Seventeen Years were turned into formalized yangbanxi spectacle films of the Cultural Revolution, as in the case of Red Detachment, but also, in the next section, at how classical-style films began to be remade in the final years of the Cultural Revolution, with small but telling differences from those before the Cultural Revolution.

The 1955 Guerrillas on the Plain focuses on Communist guerrilla commander Li Xiangyang as he leads guerrilla actions against occupying Japanese forces in Hebei province in 1943, in coordination with the Communist Eighth Route Army. The fictional Li Xiangyang was said to be based on a real-life war hero, guerrilla fighter and deputy regimental commander Zhen Fengshan 甄风山, so the film makes a relatively direct claim to historical referentiality. In addition, like the original Red Detachment of Women, it is an engaging and skillfully made war movie with charismatic actors in the main roles, particularly Guo Zhenqing 郭振清 as Li Xiangyang, but also Fang Hua 方化 as the main villain, Japanese captain Matsui. The opening sequence puts us realistically on location in a river gorge, and in general, the film aims for the same sort of combination of social and fictional realism that a contemporaneous Hollywood war film would have displayed. It also similarly concludes with a major battle victory within an ongoing war (just as does Sands of Iwo Jima, the 1949 Hollywood war film used as a point of comparison in a different context in chapter 4).

The 1974 revolutionary model opera film version makes some significant changes to the story—introducing a heroic elderly woman as one of the main characters, eliminating some scenes, adding others—but the biggest changes are simply the vast differences in aesthetic between classical-style fiction films and revolutionary model operas, which combine a Cultural Revolution Maoist worldview with techniques drawn from Beijing opera. Model opera films, such as Fighting on the Plain and the opera version of Red Detachment of Women, adopted many characteristics of traditional Chinese theater that defy the norms of the mimetic fictional realism found in mainstream classical-style cinema, including the other two film versions of Guerrillas on the Plain. These include elements of both mise-en-scène and narration.

In terms of mise-en-scène, Chinese opera sets are traditionally more spare and artificial than those of Western theater. While not nearly as minimal as those of traditional opera, the Cultural Revolution model dramas had sets and props that, when set to film, defied the supposed naturalism of the cinematic medium with blatantly artificial objects, from architecture to vegetation and natural scenery. Human figures were similarly stylized. Rather than individualized protagonists and antagonists with realistic idiosyncrasies and psychological depth, characters would conform to a limited number of role types that determined their behavior. The stereotyped roles of traditional Chinese opera (the young scholar, the beauty, the warrior, and so on) were replaced with those of revolutionary mythology (the virtuous worker, peasant, or soldier; the self-sacrificing Communist hero; the villainous landlord or Japanese occupier). Though not as extreme as in traditional operas, characters in the model opera films nonetheless generally have makeup that exceeds lifelike verisimilitude and costumes that appear highly schematized rather than realistic. This includes color codification of costume and makeup for different character types, such as wan-faced villains wearing black or green and ruby-cheeked heroes wearing red or white.

Narrative elements are similarly stylized in Beijing opera. Actions are expressed as much through symbolic signs (semiosis) as through verisimilar representations (mimesis), most obviously in the use of liangxiang (striking a pose) to signify important narrative moments and emotional states. Chinese opera forms use representations that are often “suppositional” (jiading 假定) or “subjunctive” (xuni 虚拟) rather than mimetic, meaning that elements of the story world are supposed in the mode of “as if” rather than directly represented.35 For example, in the first revolutionary model opera film, Taking Tiger Mountain by Strategy (Zhiqu Weihu Shan 智取威虎山; Xie Tieli 谢铁骊, 1970), when the hero, Yang Zirong, has to fight off a tiger while riding his horse through the forest, neither the horse nor the tiger is seen; instead, the horse riding is represented through the traditional tangma 趟马 performance technique, in which the actor represents horse riding by going through a series of poses and dancing motions while holding a stylized riding crop, while the tiger is indicated only by an audible roar on the soundtrack and the character’s gestures of tracking and shooting something unseen by the audience. Both horse and tiger are cued to be supposed by the viewer but not directly shown. Performances in general rely on gestural language and pantomime. Emotions are not so much realistically performed as codified through standardized gestures.36 Speech is stylized rather than imitative of speech in regular life, and the frequent arias, of course, break even more sharply from any semblance of mimesis of reality (just as the Hollywood musical was among the most formalist of popular film genres and thus the most apt to break with classical Hollywood norms of fictional realism).

The yangbanxi and the films made from them were consumed in ways marked by repetition, attractions, and narrative distension, all of which pushes spectator affect toward artifact emotions rather than fiction emotions. Particularly by the time the yangbanxi films were distributed, PRC audiences already were familiar with the narratives through stage performances, books, and other product tie-ins,37 and in any case, the films were viewed repeatedly owing to the relative lack of other options in movie exhibition. Consequently, as with traditional opera, any given spectator would already be familiar with the narrative from previous repetition and thus not processing plot revelations so much as enjoying the staging of a spectacle depicting an already well-known story. For that reason, as a cinematic mode, yangbanxi, like their precedents in traditional theater, are much more focused on presenting attractions than on efficiently conveying cause-and-effect chains of narrative events.38 They frequently linger extensively on each narrative moment far beyond the length necessary to transmit it as plot information, instead offering prolonged visual spectacle and commentary relating to each event, particularly through arias and dance movements. The 125-minute-long Fighting on the Plain model opera film thus extends 42 percent longer than the 1955 narrative film version and 25 percent longer than the 1974 narrative film version.

A nearly nine-minute scene that begins one-third of the way into Fighting on the Plain demonstrates the conventions described as well as the artistic quality and excitement as spectacle that the yangbanxi films were capable of achieving. The hero—whose name in the opera version has been changed from Li Xiangyang to Zhao Yonggang (赵勇刚; literally “surpassingly brave and strong”)—has disguised himself as a coachman for a horse wagon, with several of his guerrilla comrades pretending to be coolies, all arriving undercover at a Japanese-held area ostensibly to help transport grain. In keeping with the suppositional aesthetic of Chinese opera, the actual horses are not seen, though they are heard in the form of off-screen neighing and hooves clopping on the soundtrack. While the mise-en-scène is elaborate, it is still obvious that the action takes place on an artificial studio set, with painted backgrounds to represent the distant landscape and many objects that are obvious as props rather than real-world things, including a prominent tower that is featured in the action to come. In cinematographic reinforcements of character stereotypes, shots of Zhao Yonggang tend to be framed at a slightly low angle so that we look up to him, while shots of the Japanese enemy and their Chinese collaborators tend toward high-angle framing so that we look down on them. Zhao makes his entrance carrying an elaborate horse whip and singing a brief aria, the first of two times he breaks into song in the scene before his cover is blown. His makeup gives his brightly illuminated cheeks a rosy hue, whereas the sallow, scowling face of the Japanese soldier in charge tends to be in shadow. Even before the real action starts, Li Guang 李光, the opera performer playing Zhao, frequently spins around or breaks into other dancelike moves punctuated by liangxiang poses accompanied by synchronized Beijing opera cymbals.

After Zhao is revealed to be a Communist guerrilla rather than a simple cart driver, the scene erupts into an exhilarating staging of highly acrobatic hand-to-hand combat between the Japanese soldiers and Zhao and his men. This segment of the scene lasts nearly a minute and a half, averaging just over two and a half seconds per shot, though the highlight is a fifteen-second continuous shot in which a handheld camera puts us right in the midst of the action, bobbing and weaving as the blows are struck and the bodies fly in all directions. By the end of the fight, Zhao alone has dispatched twenty-five enemy soldiers, using punches; kicks; throws; the occasional improvised weapon, such as a sack of grain; and, finally, a Japanese saber he seizes from the enemy to dispatch his last few opponents. When the fight is over, he concludes with a victorious solo liangxiang pose, after which his men quickly reassemble for a group liangxiang pose that ends the scene in tableau vivant fashion (Figure 25). The fight segment of the scene is remarkably similar to the choreography of contemporaneous martial arts and kung fu films from Hong Kong studios like Golden Harvest and the Shaw Brothers, and there could be no better evidence for combining opera films and martial arts cinema into one “shadow opera mode” of Chinese filmmaking, as proposed by Chris Berry and Mary Farquhar.39 The entire scene, including the lead-up to the fight, has sixty-eight shots, nearly half of which (46 percent) feature mobile framing of all types: pans, tilts, tracks, crane shots—even two paired zooms on the saber (one in, one out) when Zhao seizes it.40 Considering also the mostly smooth editing continuity of matches-on-action—but with the occasional more expressive, slightly discontinuous sorts of cuts characteristic of Hong Kong action sequences—the scene lays to rest any notion that the yangbanxi films are merely stage documentaries rather than highly cinematic works of art. On the contrary, despite their rejection of classical-style cinematic realism, their basis in theater, and their adoption of many Chinese opera conventions, the yangbanxi films in general are carefully and creatively engaged in the most distinctively cinematic of formal elements, including editing and camera movement.41

Formalism and Ideology

Still, however cinematic they would become in their film versions, all the yangbanxi began as stage performances. Aesthetic factors that lean away from realism toward stylization, abstraction, and theatricality no doubt were partly what made yangbanxi adaptations the focus of filmmaking during 1970–73, when film production resumed after ceasing entirely during the early stage of the Cultural Revolution throughout 1967–69. The yangbanxi films’ rejection of certain common understandings of realism was an intentional part of their politically inflected offering as prescriptive “models” for new cultural production appropriate to the Cultural Revolution. Although Mao’s wife, Jiang Qing—the former film actress widely credited with sponsoring, steering, and indeed sometimes micromanaging the production of the yangbanxi and their film adaptations—claimed that even model theater should aim to “reflect . . . real life,”42 she also explicitly attacked and ruled out modes of critical social realism that some artists and critics had advocated in periods of relative liberalization of cultural policy during the Seventeen Years, even though the advocates of those theories already had long since been purged.43 Such notions of critical realism were considered to be opposed to the sanctioned formula of combining revolutionary realism with revolutionary romanticism, and the ongoing criticism of them was deemed an essential counterpart to the promotion of the yangbanxi as exemplary revolutionary works of art.44 Their prescriptive quality is suggested by the very naming of the works and their heroes as yangban 样板 (“template,” “prototype,” or “model”). Jiang Qing herself acknowledged that “the art of Peking opera is based on exaggeration” and that yangbanxi should aim “to model the image of the contemporary revolutionary hero on our opera stage.”45 As a result of such intentional idealization, as Pang Laikwan notes, “yangbanxi heroes were deemed the ultimate heroes—far greater than any ordinary person could aspire to become.”46 The Fighting on the Plain fight scene just described exemplifies the point: the hero’s defeat of twenty-five opponents in just a minute and a half of hand-to-hand combat is more in keeping with a martial arts or superhero movie than with a historical war film.

Such extreme larger-than-life exaggeration is readily apparent if we compare the scenes of the hero Hong Changqing’s execution in The Red Detachment of Women as transformed from the original 1961 film into the later Cultural Revolution model ballet and opera film versions. In the 1961 film, as described in the last chapter, the moment when Changqing is burned at the stake as the local villagers helplessly watch (and his implied potential romantic partner, Wu Qionghua, likewise looks on from afar) is romanticized and melodramatized. The final close-up of his face as he defiantly shouts “Long live the Chinese Communist Party!” just before flames rise to cover him no doubt constitutes an “epiphanic moment” of high heroism. Still, even in this moment, there are plenty of markers of a verisimilar form of realism: on-location extreme long shots mark the scene as plausibly exterior (even if the closer shots were likely done on a studio set), and his tight binding to the tree on which he is immolated, with his arms tied behind him, show him as believably tied down and immobilized by the enemy; his ripped, rumpled, and dirty clothing, along with the dirt and profuse sweat evident on his face, all realistically show the dire situation he is in (Figure 26a). In contrast, the model ballet and model opera versions, both of which combine into one scene his refusal to sign a confession and his resulting execution, transform the believable setting in a tropical village to a distinctively theatrical, stagelike location, where, through gestures, dance, and poses, the landlord villain offers Changqing a paper on which to write his confession and Changqing rejects his cowering captor in heroic fashion. In the subsequent execution, no attempt is made to make the fire look realistic, and Changqing dances under his own power to his own execution bonfire, where, unbound, he finally strikes a defiant—even downright victorious—liangxiang pose that suggests immortality even as the obviously fake flames surge around his body (Figure 26b).

Recalling the discussion in chapter 1 of the transition in acting from a “histrionic” stage style to a “verisimilar” film style during the early silent film era, as described by Roberta Pearson, the yangbanxi films obviously swing very strongly back to the “histrionic” pole.47 Pearson notes that in the histrionic mode of stage melodrama and early cinema, meaning often was communicated through discrete gestures or poses that acted as semantic units rather than inseparable parts of a continuous flow of performance: “Actors deliberately struck attitudes, holding each gesture and abstracting it from the flow of motion until the audience had ‘read it.’”48 Similarly, the model opera films’ liangxiang poses, in which the actors periodically come to a momentary rest, are an essential part of the performance and a principal means by which the performance is made legible and enjoyable to the audience. These ritualized, frozen postures punctuate passages of movement, dialogue, or song and dance and thus become both syntactic and semantic units in the narrative. As in the histrionic stage style of Western acting analyzed by Pearson, they strive not for resemblance to real human behavior (mimesis) but rather for a codification of emotion (semiosis) that the audience can read through their familiarity with dramatic conventions, while also enjoying the rhythm of the performance, with its periodic pauses to mark the ends of particular segments of narrative.

Returning to the Guerrillas on the Plain series, aside from the 1955 original and the 1974 model opera remake already discussed, there remains the odd fact that, in the same year the opera film was released, the original also was remade in its more classical fiction-film style, but this time in color. This was not an isolated phenomenon. When non-yangbanxi feature filmmaking resumed in 1973 for the first time since the beginning of the Cultural Revolution, in the absence of officially sanctioned original screenplays, one practice was to reshoot previously approved black-and-white films of the Seventeen Years as color films. Besides Guerrillas on the Plain, other films that were remade in the period 1973–75 included Conquering the Flood (Zhan hong tu 战洪图; Su Li 苏里 and Yuan Naichen 袁乃晨, 1966), The Younger Generation (Nianqing de yidai 年轻的一代; Zhao Ming 赵明, 1965), Pine Ridge (Qingsong ling 青松岭; Liu Guoquan 刘国权, 1965), Reconnaissance across the Yangtze River (Du jiang zhencha ji 渡江侦察记; Tang Xiaodan 汤晓丹, 1954), and Fighting North and South (aka From Victory to Victory; Nanzhengbeizhan 南征北战; Cheng Yin and Tang Xiaodan, 1952).

When a historical war film is remade, the temporality of its representation becomes even more complicated. Recalling Rosen’s point that a historical fiction film represents two different past times—the time of the documented performance of the fiction and the time of the historical events referenced—these Cultural Revolution remakes add a third representation, that of the earlier film that is being imitated, with which Chinese filmgoers of the mid-1970s would have been familiar before seeing the remake. Such an added level of fictional signification would have further problematized the transfer of the “indexical stickiness” of the documented performance’s ontological realism to lend credence to the historical representation because not one but two documented fictional performances mediate the relation of film to actual history. Consequently, while the 1974 model opera film remake of Guerrillas on the Plain would have been unlikely to be taken as a realistic historical representation owing to the extremity of the opera form’s stylization and theatricality, the more conventionally realistic remake also would have any claimed tie to history weakened because of being a re-presentation of a prior representation of history.

Beginning in 1966, Xing Ye 邢野, cowriter of the script for the original 1955 Guerrillas on the Plain, was enlisted by Jiang Qing to join a team that began the process of adapting the screenplays for what would eventually become the model opera version of the story as well as the 1974 color remake. His account of that process sheds light on the issues of realism and the representation of history as the Mao era advanced from the Seventeen Years to the Cultural Revolution.49 He reports that the writers had access to many actual stories from the life of Zhen Fengshan—the real-life war veteran on whom the fictional protagonist was based, who was still alive during the first stage of the revision process and talked with the writers. In Xing’s and other writers’ views, those stories could have enriched the screenplay: “In terms of the complexity of history and the richness and uniqueness of a figure’s personality, these incidents would have enabled the character to be written much more brilliantly.”50 Xing reports that one writer’s participation in the project was ended after he tried to add a new plot incident based on Zhen’s actual war experiences. The guerrilla leader had been involved in an unusual incident in which a Japanese captain had kidnapped Zhen’s wife in an attempt to force his surrender, so Zhen in turn had kidnapped the captain’s daughter-in-law and forced a hostage exchange to get his wife back. (The ploy worked, but he was disciplined by the army for pulling such a stunt.) When Jiang Qing saw the incident added to a draft script during the revision process, she angrily rejected it as defiling the reputations of the CCP and the Eighth Route Army, ordered it cut, and fired the writer from the project.51 Not simply an interesting anecdote, this incident—and the broader rejection of adding real biographical and historical details that would make the story more rich and complex and the character more rounded—illuminates just how prescriptive Chinese revolutionary realism had become; as Pang Laikwan has put it, “the regime found fiction more reliable than reality.”52

Xing laments that if they had been allowed to incorporate such details into the film, they may even have approximated something like the accomplishments of the classic Soviet film Chapaev (Georgi Vasilyev and Sergei Vasilyev, 1934) in depicting its eponymous war hero as “a man with a character that was so complicated, so full of historical depth and breadth.”53 The comparison is telling, given that Chapaev has long been regarded as embodying the type of complexity and ambiguity that would soon disappear from most Stalinist cinema as the dogmas of Soviet socialist realism took firmer hold of cultural production. Calling it “the masterpiece of Soviet films with historic heroes,” André Bazin celebrated “with what intelligence the failings of Chapayev, even in his manifestly most heroic acts, are suggested without diminishing him at all psychologically.”54 In contrast, later Soviet historical films about Stalin turn the Soviet leader into such a superhero that he becomes “History incarnated,” making it “impossible to define him through his character, his psychology, or his personality,” which Bazin compares unfavorably with the commander in Chapaev.55 Xing Ye’s memoir of the rewriting of Guerrillas on the Plain makes clear that the flattening and idealization of the hero as a character and the altering of historical reality resulted both from ideological strictures on the arts and from the desire to please audiences. For example, noting that the wiping out of an entire large Japanese military unit and the killing of its commander never happened in Hebei in 1943, Xing explains that “our thinking was still in pursuit of so-called ‘combining revolutionary realism and revolutionary romanticism’” and that these fictional exaggerations would be more “entertaining to the masses.”56

Such distortions of history had existed in the original film as well. In fact, in terms of the intertextual relations between the three films, whereas the model opera film version took many liberties with the story and characters of its predecessor, the more classical color remake of the same year is remarkably faithful to the original performance of almost two decades earlier (if not to real historical events). Most of the script remains the same, and often, cinematically, it looks like a shot-by-shot imitation, with even the less dramatic shots often mimicking the original in the details of their mise-en-scène, framing, and even the nonverbal gestures and expressions of the actors (Figure 27). And although the main hero is played by a different actor in the remake to retain his youth, in another instance of “villain stardom,” the Japanese commander Matsui is once again performed with great expressiveness by Fang Hua, who had played him in the original (and had been typecast in Japanese enemy roles going back to his earliest films in the late 1940s).57

In the context of such surprising sameness, the differences between the 1955 film and the 1974 color remake become all the more notable, revealing the changed political circumstances and the heightened prescriptiveness of culture and the arts in the wake of the Great Leap Forward and the Cultural Revolution. The intervening Sino-Soviet split, for example, requires that a reference to the “great Soviet Army” fighting Adolf Hitler be changed to “all the people in the world will defeat Hitler this year.” At the end of the film’s climactic battle, the Japanese commander, who had been given the dignity of at least attempting to commit ritual suicide (seppuku) in the original film, now is simply trying to run away at the moment he is caught (and, soon after, shot dead). But the most notable differences in the largely identical scripts are those that best fit what Chris Berry calls the “pedagogical” and “didactic” tendencies of Mao-era cinema, as discussed in the preceding chapter.58 Whereas Mao Zedong is never mentioned in the original 1955 feature, the 1974 remake adds lines including the following:

• With the help of the Party, Chairman Mao, and the Masses, we can fulfill any mission!
• The hearts of the common people are all with the Party. Stick to the strategy of protracted warfare introduced by Chairman Mao, and the [Japanese] devils won’t live for long.
• As Chairman Mao says, “the strategy of risking life is shortsighted in military operations. We need to look forward and plan wisely.”
• We are the people and the army led by the Communist Party and Chairman Mao! You can muster all the divisions of your country, and we will still wipe out the invaders without mercy.

The remake also adds two brief musical sequences, in which montages of the good characters fighting and then (in the second such sequence that ends the movie) celebrating victory are accompanied by a nondiegetic choral soundtrack with rousing lines like “Chairman Mao’s strategy is mighty!” and “Chairman Mao leads us bravely forward!”

Particularly given how closely the film’s script in general hews to its 1955 precedent, when the remake is viewed after the original, the now-obligatory lines of homage to Mao and the party feel awkwardly inserted, as if they are brief rhetorical rituals that interrupt the momentum of the film’s dialogue and plot. Their addition to the screenplay reflects a shift in ideological rhetoric in the course of the Cultural Revolution. Many have noted that the early stage of 1966–68—marked by the chaos of local uprisings, often violent struggle sessions within schools and other communities, Red Guard factional warfare, and mass migrations of young people to Beijing to be reviewed by Mao himself in Tiananmen Square—was quite different from the much longer, later stage from 1969 to 1976. As Pang puts it, “the Cultural Revolution could be separated roughly into two periods: the first three years corresponded to a period of semi-anarchy, after which the state largely resumed its functions.”59 In terms of art and culture, Pang continues, the earlier stage’s “drive was to establish a new culture, while the second part aimed to solidify the status quo.”60 By 1974 the almost liturgical recitations of praise to Mao and the party in the Guerrillas on the Plain remake likely struck audiences at the time less as the sort of intensely passionate youthful commitment to Mao and revolution that the early Red Guards experienced and more as the voice of the state enforcing the expected ideological orthodoxy.

Indeed, in his book on the cultural aspects of the Cultural Revolution, Paul Clark makes several observations that suggest that the formalist and formulaic nature of Cultural Revolution culture, notably the yangbanxi, gradually lessened their effectiveness as propaganda. According to Clark, Cultural Revolution formulas like the “three prominences,”61 partly through sheer repetition, resulted in a kind of banality in which, for example, the endless copying of heroic figures’ images “rendered them into kitsch icons” that were “empty of real substance,” akin to the omnipresence of Coca-Cola or Nike advertising logos in the visual culture of capitalist societies.62 Similarly, according to Clark, the model operas in particular “failed because of their very success in filling China’s stages and screens.”63 While the cliché of “800 million people watching eight shows” (the initial “eight model works”) is an overstatement of the paucity of artistic offerings during the Cultural Revolution,64 it was still the case that, as Clark puts it, the model operas “could not be avoided during the Cultural Revolution,” to the point that although they may have been “enjoyed . . . on a first or even fifth viewing,” they eventually became tiresome.65 Clark notes that by 1974, Chinese audiences welcomed any new feature films not based on yangbanxi, including the color remakes of older films, just to have a break from the repetitive hyping of the same yangbanxi heroes—indicating that, “by the mid-1970s, mass cynicism regarding the rhetorical hyperbole surrounding the model theatrical works was widespread.”66 As a rare Westerner who lived in China as the Cultural Revolution drew to a close in 1974–75, Clark remembers that there was a notable rise in irony by the end of the period: “A certain tone in singing an aria from one of the model operas, a certain flick of the head in exaggerated parody of one of the central heroes, a clever rewording of a well-known verse could provide an outlet for a largely unspoken but shared sense of the ridiculous.”67

Performative Shift and Transcendent Master

All these observations on the repetitive and ritualized nature of political rhetoric in art by the time the two Guerrillas on the Plain remakes were released suggest that the transition from the initial stage of the Cultural Revolution to its later period just before Mao died may have reproduced in 1970s China what Alexei Yurchak has dubbed a “performative shift” in ideological rhetoric in the Soviet Union after the death of Stalin. Drawing on the distinction in linguistics between constative statements, which claim to accurately describe an objective reality, and performative utterances, which are not really evaluable as true or false but instead are meant to perform an action by their very utterance, Yurchak argues that after the death of Stalin, “the performative dimension of ritualized and speech acts rises in importance (it is important to participate in the reproduction of these acts at the level of form), while the constative dimension of these acts becomes open-ended, indeterminate, or simply irrelevant.”68 Or, as Yurchak puts it, “it became increasingly more important to participate in the reproduction of the form of these ritualized acts of authoritative discourse than to engage with their constative meanings.”69 In other words, owing to what Yurchak calls the “hegemony of form,” the ideological discourse, while losing none of its omnipresence as form, became oddly detached from reality, in that Soviet citizens started to experience it precisely as performative rather than as constative, as keeping up an appearance rather than being real.

This distinction corresponds closely with the development of the historical war film in Mao-era China from the Seventeen Years to the Cultural Revolution, including, in the case of the latter, both the yangbanxi films and the remakes of earlier classical-style fiction films. The loss over time of those films’ credibility in claiming to represent actual historical reality and their tendency to instead become generically stereotypical, artistically formalist, and ideologically formulaic exercises is not just analogous to but part of a performative shift in revolutionary rhetoric. Moreover, the broader performative shift in Chinese Communist ideological discourse illustrated by Chinese socialist formalism in cinema may ultimately have contributed to the same sort of instability of ideological meaning that Yurchak finds in Soviet rhetoric following the performative shift after Stalin’s death, leading to possibilities for “increasingly frozen” and “replicated” ideological forms to suddenly develop new, unintended meanings or to lend themselves to the kinds of cynicism and irony Clark saw in China by the end of the Cultural Revolution.

The potential for this instability, explains Yurchak, grows out of what he dubs “Lefort’s paradox,” referring to French philosopher and anti-Stalinist socialist Claude Lefort, and in particular the latter’s ambitious 1974 essay “Outline of the Genesis of Ideology in Modern Societies.”70 Lefort explains ideology in modern societies in terms of a basic contradiction or aporia that arises when human communities become aware of their own historicity or contingency, a phenomenon that is considered to be constitutive of the modern condition insofar as global modernities—including Communist movements in China and other non-Western societies—are inflected by certain modes of Enlightenment thought that helped to launch the modern age. In particular, modern forms of social organization and government are based on a conception that human societies are self-constituting. Forms of government are not consecrated by God or other supernatural origins but are created by human beings themselves as the primary agents of their own history. Whereas premodern civilizations conceived of their social organization as being determined from elsewhere—the divine right of kings, the mandate of heaven, and so on—a modern society realizes that its structure is created immanently through its own historical processes, so that humans in theory also can change, abolish, or recreate their social forms and institutions at will according to what they think will lead to happiness and prosperity. The freedom this grants, however, is necessarily accompanied by a lack of any firm grounding—given the instability and contingency of the historical real itself—a tenuousness that must then be dissimulated if a stable political and social order is to be instituted. Any hegemonic political power must at least partially disavow its own contingent origins and instead develop a system of representation that consecrates its system of rule and makes it seem anchored to something permanent. In a modern society, “the external character of the ‘elsewhere’” comes not from “religious or mythical knowledge” but from “the transcendence of ideas”—ostensibly eternal values, such as those of the Enlightenment itself.71 This, for Lefort, is the very definition of ideology in modern societies. Certain ideas, “realized as pure transcendence” and ostensibly immutable, function to conceal the historicity and contingency of the social by claiming a certainty and stability beyond the historical real, an ability to grasp the social totality from some vantage point outside of it; however, because modernity itself teaches us that both these ideas and the social system they serve originated precisely in the social and in history, they are necessarily “haunted by tautology,” by the contradiction between the actual contingency and immanence of their creation and their claim to lie outside that contingency and connect to something transcendent that helps to establish the stability of the state and the social order.72 In the Chinese party state, what presented themselves as transcendent in order to solve that tautology were first the reified party, embodiment of the masses, in which the heroes in this and the preceding chapter found immortality, and second, especially during the Cultural Revolution, Mao himself.

It is part of the complexity of the Cultural Revolution that Mao was both the revolutionary advocating for shattering the stability of ossifying structures of society and culture and also the transcendent figure offering ideological stability through the turbulence. Mao’s launching of the Cultural Revolution was rather unique in that he called in effect for a rebellion against the very state and party structures that he had helped to build. The danger that Mao sensed was what Wang Hui summarizes as the “statification” of the revolutionary party, in which its function as an insurgent historical political movement gives way to its integration into the framework of the state.73 As Lefort has put it in the case of the USSR, “bureaucracy became a class in the Soviet Union at the very moment when, paradoxically, it integrated itself entirely into the state.”74 For Wang, the Cultural Revolution was an attempt to forestall that tendency, and its failure meant precisely the further “statification” of the party and depoliticization of social life that occurred afterward in the reform era. But the process of statification arguably resumed not only after 1976 but after the early, more anarchic phase of the Cultural Revolution was over in the late 1960s. For Lefort (writing about Stalinism), with the transformation of a revolutionary party into a party state, “the claim to circumscribe within the boundaries of the organization access to the truth of revolutionary doctrine” is one of the aspects of party ideology that becomes totalitarianism once the party has been consolidated into a state structure.75

In his study of the Soviet Union, Yurchak notes in particular Lefort’s suggestion that totalitarian modern societies can conceal the paradox of their own historicity through the figure of a “master” who is seen as standing outside both ideological discourse and history and having external knowledge of objective truths in a similar way to which an imagined religious or mythical “elsewhere” provided legitimacy to premodern social orders.76 This, of course, was the ideological function of Stalin during his rule of the USSR—an idea that echoes Bazin’s argument that Stalin was represented in Soviet cinema as the “God of History,” who cannot be allowed to “fall into human contingency.”77 In revolutionary China, the imaginary external position that rendered ideological discourse stable was of course embodied by the figure of Mao and all his metonyms, such as the scriptural “little red book” of his quotations. This had not been true in the early years of the PRC, when the party explicitly rejected “personality cults”; however, with the launching of the Great Leap Forward in 1958, as documented by Maurice Meisner’s study “The Cult of Mao Tse-tung,” Mao began “appearing on the historical scene as supreme leader and utopian prophet,” assuming a “unique and transcendent role,” a “cult of the individual” that Mao himself argued was justified as long as “the individual concerned represents the truth. If he does, then he should be revered.”78 In the following decade, with the Cultural Revolution, “the cult of Mao was to manifest itself in forms far more extreme than Stalin’s ‘personality cult’ ever had assumed,” with explicitly religious connotations, such as aligning Mao with heaven, the sun, and thereby “the forces of the cosmos”; calling “Mao Zedong thought” a “magic weapon” in a manner reminiscent of Daoist mysticism; and labeling his foes with terms drawn from the popular Buddhist supernatural imagination of monsters and demons.79 Meisner notes that after the Cultural Revolution waned in fervor in 1968 and Mao himself moderated its radicalism, “the cult of the Chairman, ironically, grew ever more extravagant,” but now appeared less like a “spontaneous mass revolutionary movement” and more like “the performance of the established rituals of an orthodox church”—an observation in keeping with Yurchak’s idea of a “performative shift” and my descriptions of the increasing formalization of film style and conventionalization of cinematic narrative (including obligatory homages to Mao in film scripts).80 As Meisner concludes, the Mao cult became “one of the most extreme examples in modern history of the alienation of the social power of the people into fetishized political authority.”81

The socialist realist gaze into off-screen ideological space by the heroes in revolutionary films reproduces the structure of Mao’s supposed transcendental perspective and can be read as a veritable illustration of Lefort’s paradox. Ideological discourse must claim to represent an “objective truth” outside of it, but the very external nature of this truth renders it unrepresentable within the discourse—which is thus forced to resort to an array of rhetorical sleights of hand. On one hand, the unrepresentable Truth is suggested as an off-screen sublime presence that, if not available to the direct gaze of the audience, is at least secondarily accessed by the relay of the socialist realist gaze of the heroic character, who, it is suggested, has gained access to it by becoming a true Communist. On the other hand, the Truth can yet be symbolized by the substitute image of a loaded ideological symbol, such as a flag, a simple signifier that nonetheless seems to transcendently anchor the entire symbolic field. In their access to ideological truth, fictional Communist protagonists thus served as Mao surrogates, representing the party’s access to the Truth through their own heroism. This imaginary link from the revolutionary romanticist hero to the party as a whole and ultimately to Mao himself became obvious, for example, in the case of the Red Lantern model opera film, in which the martyred hero Li Yuhe was played by lead actor Qian Haoliang 钱浩梁, who was said to have been picked in part because of a physical resemblance to Mao himself. So strong was the imaginary link between fictional heroes and the Great Helmsman that Qian was richly rewarded by Jiang Qing and eventually became deputy minister of culture toward the end of the Cultural Revolution (only to be persecuted soon thereafter, following the fall of the Gang of Four).

Coming to a Standstill

In yangbanxi cinema, with the entire film embracing socialist formalism, the overdetermination of every detail by ideological rather than realistic motivations goes hand in hand with an unmooring of form from reality, potentially destabilizing that very ideology in the manner that Yurchak describes with the performative shift. The stylized performance becomes such pure spectacle that the ideological content, precisely by being so rigorously formalized, threatens to become a mere surface itself, a superficial appearance that in fact signifies nothing. For example, the briefly frozen liangxiang poses of traditional drama, as adapted to yangbanxi, offer an obvious opportunity to turn the socialist realist gaze into something rigorously standardized as a discrete unit of meaning, in the same way that the histrionic poses of nineteenth-century Western melodrama were abstracted from the flow of the performance to be read as semantic units by the audience, as described by Pearson. Certainly many of the liangxiang struck by yangbanxi heroes appear to be gazes toward some off-screen horizon that could indicate the superior ideological vision of the socialist romanticist hero. However, in the sheer repetitiveness of the poses as well as the artificiality of the sets, the sense that the hero is in fact looking at something can be lost. Instead, particularly on repeated viewings, such looks could have struck audiences as being purely formulaic, possibly even vacant. In this sense, the adaptation of body language, particularly that of the eyes, of liangxiang poses from traditional opera to revolutionary yangbanxi may have unintentionally clashed with the precedent set by the socialist realist gazes in films of the mimetic revolutionary realist tradition preceding the Cultural Revolution. Whereas the latter may more plausibly have convinced the viewer that the Communist hero actually did see the Truth of correct ideology, the yangbanxi hero may too plainly have been looking at nothing (Figure 28).

Mao himself embodied the fundamental contradiction between the desire to continue and further radicalize the revolution and the practical need to maintain a stable nation-state. With his call for a “continuous revolution,”82 Mao clearly sided in theory with the idea that the dialectics of revolution must continue within a Communist state even to the extent of threatening that state itself if a contradiction develops between the state bureaucracy and the revolutionary masses. In the course of the late 1950s and early 1960s, what Mao had initially identified as “contradictions among the people” in socialist society developed into a coherent theory of continuous revolution that found contradiction to be ubiquitous and permanent even in a society in socialist transformation, classes to be permanent features of society, and revolution in the form of class struggle consequently to require renewal indefinitely.83 These views were based in part on the dialectical materialist conviction that reality develops through contradiction. Without contradiction, there can be no evolution or progress; reality and history itself would be in stasis.

Chapter 3 cited Paul Pickowicz’s argument that while melodrama served to introduce “basic Marxist notions of class struggle, capitalism, and imperialism” in leftist films of the Republican era, those “ideas were swallowed up by the melodramatic genre and reduced to stereotypes and caricatures” in a narrative mode based on “repetition and ritual.”84 In that chapter, I argued that realism and melodrama are more intertwined than simply opposed to each other, and in chapter 4, I argued further that the match between melodrama and Communist art is hardly a surprising one, because melodrama’s goal of “wresting the true from the real” is of a kind with the Maoist priority of going beyond any surface-level realism to capture the inner truths of society and history as determined by the dialectical “laws” of historical materialism. Here, however, I will argue that the fundamental disjunction that occurs when melodramatic aesthetics are used to articulate Marxist dialectical theory—the disjunction identified by Pickowicz—finds a quintessential expression in the yangbanxi films.

In fact, the harnessing of melodrama by Marxism is not just something that was added by twentieth-century Communists. In a remarkable essay from 1948, Wylie Sypher argued that the writings of Marx and Engels themselves suffered from a mismatch between their dialectical philosophy and their sometimes melodramatic narration. Philosophically, they saw a world of complex, immanent dialectical processes without transcendent truths or final solutions; aesthetically, however, they narrated their ideas as melodrama, with its moral Manichaeism and ultimate revelations and closures.85 Capital, for example, juxtaposes polarities like capital and labor as mythical oppositions, with the injustices committed against the masses set to be righted in a final, dramatic—even apocalyptic—revolutionary gesture. In a more recent study that cites Sypher, Elisabeth R. Anker reads The Communist Manifesto as a melodrama.86 Jane Gaines also extended Sypher’s original critique, arguing that it is the theory of contradiction, particularly that between capital and labor, that, as she put it, “produces the ‘melos’ in Capital,” and that “the labor/capital dichotomy suggests characters acting out a scenario.”87 Gaines further asserts that melodrama has long been a preferred artistic mode of revolutionary periods because of its ability to quickly dichotomize conflicts and identify enemies: “revolutionary melodrama can be depended upon to narrate intolerable historical conditions in such a way that audiences wish to see wrongs ‘righted,’ are even moved to act upon their reaffirmed convictions, to act against tyranny and for ‘the people.’”88 Drawing also on Sergei Eisenstein’s critique of Charles Dickens and D. W. Griffith, Gaines nonetheless suggests that the dualism of melodrama is incompatible with dialectical theory, in which contradictions do not just face off and perhaps get decided by transcendent intervention but rather ceaselessly develop and transform through an immanent dynamic.

With these critiques of the relationship between Marxism and melodrama in mind, I would like to focus on one particular formal quirk of the 1970s model opera films, one repeated so often as to be considered a hallmark of their style. I am speaking of the tendency for the films to end in a frozen tableau vivant, which, among other things, serves as a veritable illustration of the “three prominences,” as it inevitably features the leading hero at the center, with other heroes and positive characters arrayed around him or her, all striking dramatic poses of, if not final triumph (the struggle will continue, after all), then more precisely the absolute certainty of final triumph. Nine of the ten revolutionary model opera films released from 1970 to 1975 employ tableau vivant endings, in which the actors stand motionless in a pictorial formation even as the shot goes on, sometimes combined with mobile framing during the frozen pose.89 Tableaux vivants have a history in Western culture of the eighteenth and nineteenth centuries, having become both a parlor game among the Western European aristocracy, in which “living pictures” would reproduce the compositions of famous paintings, for example, and a technique used occasionally in European drama of the time, often with a pedagogical or didactic message.90 More relevant in the case of the model opera films is the fact that the freezing of characters in liangxiang poses is part of the aesthetics of traditional Chinese opera, often as punctuation at the end of songs, acrobatic routines, scenes, or plays. But whatever their dramaturgical origins, the concluding tableaux of the yangbanxi films drive home the message of the certainty of Communism’s final victory through the cessation of all further movement among the actors, as if history itself has come to a stop in the full realization of Mao Zedong thought. The ending of Fighting on the Plain is typical of the model opera films. Having dispatched all the Japanese soldiers and finally shot the main villain, Kameta, when he refused to surrender, all the Communist soldiers and their civilian supporters gather in a typical frontal formation, with Zhao Yonggang in the center, and together chant the drama’s last words: “Long live Chairman Mao! Long live the Chinese Communist Party!” The script’s final instructions are as follows: “Together, excitedly, strike a majestic and heroic liangxiang pose.”91 The camera zooms in on the pose, and the film ends.

It is with this in mind that I would like to return to Wylie Sypher’s essay “Aesthetic of Revolution: The Marxist Melodrama” of over seventy years ago. Sypher’s central argument concerns what he calls the “fallacy of melodrama,” which he equates with the “fallacy of revolution.” To summarize, whereas dialectical philosophy conceives of a reality of immanent contradictions driving ongoing processes of transformation and increasing complexity, melodrama instead depicts a world of stark polarities in which clear ethical choices lead to dramatic action converging upon a climax: “The tensions must concentrate toward a last overwhelming tableau, a final stasis beyond which one must not think.”92 In its melodramatic depiction, “the culmination of the class struggle is the revolutionary stasis. . . . History is no longer a process; it is a gesture.”93 Staged this way, according to Sypher, “the revolution is a mythical act; the negation of negation is final justice.”94

If this were true, the dialectic would come to a standstill, the essence of revolutionary truth having been realized. But the actual philosophy of dialectical materialism is in line with Sartre’s “existence precedes essence” dictum: human constructs have no abstract, absolute truth but emerge from and change with the material conditions of history and a physical reality always in flux. The implication of Lefort’s paradox is that, with the hardening of ideological orthodoxy that especially tends to accompany the institutionalization of political and economic power structures, the assumption increasingly is reversed: essence precedes existence; reality must be reducible to the now seemingly immutable categories and abstractions of the ruling ideology.

The theatrical stasis described by Sypher could hardly find a more literal representation than the frozen tableaux vivants that conclude most of the yangbanxi films, and it is in their conclusion with, to borrow Sypher’s words, “a stasis so conclusive that it appears arbitrary” that we find a fundamental disjuncture between yangbanxi aesthetics and Marxist philosophy—and indeed between Maoist philosophy and Maoist political rule. Though Mao had clearly committed to the theory of continuous revolution, in reality, after the Cultural Revolution’s early phases, what people saw depicted in cultural forms like the yangbanxi was on the contrary what Sypher called the revolutionary stasis, in which “the dialectic has been transvalued to a symbolic act, an apocalypse” expressed in a “theatrical stasis.”95 In her essay on the “prescriptive dramatic theory” of the Cultural Revolution, Ellen Judd, too, critiques the yangbanxi as in the end being more about resolution than about conflict and thus being strangely vulnerable to the “charge of portraying the end of class struggle—that being the main type of contradiction and conflict presented,” which goes against Mao’s very idea of continuous revolution.96 If we recall the Shanghai left-wing films of the 1930s discussed in chapter 2, here we have something like the opposite extreme in revolutionary cinema: whereas the 1930s films lacked plot resolution or even any positive configuration of the possibility of revolution, instead just leaving openings through which such a vision might appear, the 1970s yangbanxi films end with reassuring and seemingly complete resolutions through spectacular frozen displays of the revolution in a state of seemingly final victory.

Overdetermination and Excess

Cultural Revolution ideology and culture are widely supposed to have been characterized by the repression of human sexuality, yet it has become commonplace to point out the irony that the yangbanxi and the films made from them were very much able to inspire erotic fantasies. Paul Clark notes that poster stills of the heroines from the model ballets The Red Detachment of Women and The White-Haired Girl—whether leaping with splayed legs or chained in a dungeon—adorned the walls above peasant brick beds and student bunks alike throughout China during the Cultural Revolution.97 Wang Shuo 王朔, the infamous “hooligan literature” author of the late 1980s and after, as well as the irreverent film actor and director Jiang Wen 姜文 have both recalled that their youthful sexual feelings were aroused by The Red Detachment of Women.98 In the 2006 Dutch documentary Yang Ban Xi: The 8 Model Works (Yan Ting Yuen), a middle-aged artist also remembers the Red Detachment ballet as his youthful favorite because the tight, skimpy uniforms of the dancing women soldiers exposed much of their legs: “Our first sexual feelings were aroused by yangbanxi. At last we’d discovered something real in the revolution.”

Why do these anecdotes attract our interest, aside from being mildly amusing? Of course, we can interpret these men’s childhood reactions as evidence of “resistance” against totalitarian control, minor acts of heroism in the struggle of the individual mind to be free even while subject to propaganda. However, framing teenage boys’ usual interests as evidence of political dissidence perhaps caters too easily to Western Cold War ideological fantasies of Mao-era China. We could more constructively view these responses as evidence for Ban Wang’s contention, in his landmark study of sublimation in Mao-era ideology and culture discussed in the last chapter, that

despite its puritanical surface, Communist culture is sexually charged in its own way. High-handed as it is, Communist culture does not—it cannot, in fact—erase sexuality out of existence. Rather, it meets sexuality halfway, caters to it, and assimilates it into its structure. Communist culture was attractive to some extent precisely because it incorporated sexuality.99

As Rosemary Roberts puts it specifically in regard to the presentation of women’s bodies in The Red Detachment of Women, “the sexualisation of female costume in fact functions to heighten the appeal of the class message encoded into female characters.”100 At the same time, as I suggested in chapter 4, revolutionary films aiming to sublimate the libidinal energy of certain of their elements—whether physically attractive performers or the film conventions of love stories—necessarily also made it possible for spectators simply to enjoy those elements in their unsublimated forms. The athletic bodies in motion in the Red Detachment of Women model ballet film are no exception—and despite the expression of specifically male heterosexual desires in the examples mentioned earlier, no doubt we can be sure that the yangbanxi films inspired erotic fantasies across genders and sexualities. Chris Berry has pointed out that the dance that pairs the heroine, Qinghua, with her Communist Party savior, Changqing, functions as a pas de deux even if its narrative content is about political “guidance and inspiration” rather than “individual romantic passion.”101 I would add that a dance Qinghua performs as a fight with an enemy soldier much later in the film functions as well as a pas de deux, with a hint of a more deviant eroticism of gender role reversal, as Qinghua physically dominates her opponent in an intimate way through dance before finally dispatching him, eroticizing the strong women figures who were characteristic of Chinese revolutionary representation.

To conclude this discussion of Cultural Revolution cinema, however, I would like to emphasize a broader implication of these anecdotes of the consumption of yangbanxi films as erotic spectacle, namely, the necessary presence of cinematic excess even in forms of cinema that are said to be the most rigorously constructed according to ideological principles that inform every detail. Kristin Thompson defines cinematic excess as anything in a film that lacks or exceeds narrative function or motivation: “at that point where motivation fails, excess begins.”102 Classical Hollywood-based film narration draws our attention away from excess with its “struggle by the unifying structures to ‘contain’ the diverse elements that make up its whole system,” keeping the focus on the story being told.103 The yangbanxi films of the Cultural Revolution would seem, on one hand, to offer an extreme example of such “unifying structures,” given what Barbara Mittler calls their “absolute monolithic quality (一元化), of perfect univalence” that, “at least in theory, allowed for one and only one interpretation.”104 Such a view is reinforced by anecdotes about, for example, Jiang Qing forcing a reshoot of a model opera film because she was not satisfied with the shade of red on the heroine’s scarf.105 And yet, those young viewers’ focus on patches of exposed skin on the screen show that any shot of any film, no matter how carefully and redundantly determined by political messaging, nonetheless offers countless details that could lend themselves to what Paul Willeman and Christian Keathley have labeled “cinephiliac moments” in a film, which capture our interest in a way irreducible to intended messaging and yet lead to a powerful response in the spectator.106 As John Mullarkey puts it, whether intended or not, any film offers “unseen aesthetic adventures of matter both on the surface and beneath the subject of figuration.”107 The possibility of such adventures—such cinephiliac moments—in audiences’ experiences of the yangbanxi films testifies to the material richness of the film image as well as the imagination of the spectator. While we usually think of the yangbanxi as a form of cinema in which every detail is most strictly motivated by politics and ideology, they could be vehicles of rampant cinematic excess. Particularly when we consider how many times the films and the plays on which they were based were viewed by the typical PRC citizen, we can speculate that for many, they would have become almost pure examples of excess. After the political message had long since become ignored cliché, the spectacle nonetheless could retain interest; the number of notes the brigade party secretary might melismatically hit in a single syllable during an aria would hold much more attraction than the commune crop yield goals about which she might be singing. The strictest subjugation of art by politics thus paradoxically can result in art for art’s sake.

It was with that conviction that the article that was my first, tentative effort to outline some of the arguments developed in this and the preceding chapter concluded with the speculation that, despite their deserved reputation as an ultimate example of art being subjected to politics, “the lasting legacy of the yangbanxi [films] may well continue, somewhat ironically, to be the triumph of their formal innovation over their political agenda.”108 Since then, several major studies by other scholars have reinforced this conviction. Mittler’s A Continuous Revolution: Making Sense of Cultural Revolution Culture includes several quotations from artists and audiences who lived through the Cultural Revolution confirming that, although everybody knew that the yangbanxi were intended as political propaganda, it was the artistic quality that actors, artists, and musicians achieved under Jiang Qing’s high standards that made them widely enjoyable.109 A guiding theme of Pang Laikwan’s The Art of Cloning: Creative Production during China’s Cultural Revolution is that a certain autonomy of the aesthetic and the sensorial accompanied Cultural Revolution art even at its most politicized. Chris Berry’s essay “Red Poetics: The Films of the Chinese Cultural Revolution Revolutionary Model Operas” similarly concludes that it is a paradoxical autonomy of form that lends the yangbanxi an ongoing aesthetic appeal even in the post-Mao era.110 In a comprehensive study of the stage model operas, Xing Fan concludes that the exacting pursuit of aesthetic beauty characterized them in all their aspects, from acting to music to costume and set design.111 It makes for a curious dialectical reversal—that a prescriptive mode of realism ostensibly steeped in political utilitarianism would result in a triumph of artistic form over content. However, if my argument is correct, such a development in the arts corresponds and contributes to the overall performative shift in ideological discourse after the radical grassroots political struggles of the early Cultural Revolution subsided.

At the same time, although I have discussed them as the culmination of Chinese socialist formalism, from a post-Mao historical perspective, one real value of the yangbanxi films may be precisely their realism—but in a very particular sense. Recall Rosen’s distinction between the two past times represented by any historical film—the time of the performance before the camera, captured directly through cinema’s ontological realism, and the time of the historical events referenced, a much more indirect signification. Quite likely, even audiences at the time appreciated the yangbanxi films as direct recordings of extraordinary opera and ballet performances rather than as representations of historical reality. In that sense, they enjoyed the film medium’s realist capacity to record precisely the moment of performance. Today, still, we value them as records of those performances as well as broader indices of an extraordinary moment of revolutionary culture saturated in performativity, displaying with a kind of terrible beauty how the revolutionary impulse not only to struggle but to transcend can stall and ultimately be thwarted by its own aporias.