Postsocialism, Post–Socialist Realism, and the Modernization Discourse

The term postsocialism, though rarely used in China itself, has been employed frequently by scholars in the West to describe the condition of China after Mao, particularly after the 1980s. Arif Dirlik first proposed the term in 1989 to refer not to a capitalist China that had discarded socialism but rather to China still in a state of “actually existing socialism” (as opposed to an idealized form) that nonetheless sought to redefine the relation between Chinese socialism and capitalist modernity.7 Within a few years, the articulation of Chinese socialism to global capitalism became so tight that a cynic might well say that the emerging “socialism with Chinese characteristics” was merely a face-saving euphemism for capitalism itself, owing to the ongoing economic boom and the accompanying marketization, if not outright privatization, of more and more sectors of the economy, including the film industry and other cultural institutions. A number of scholarly books on contemporary China have featured the term postsocialism, just as many works on Eastern Europe and the former Soviet Union have used postsocialist, post-Communist, or post-Soviet to describe those societies since 1989.8

Specifically in the realm of Chinese cinema studies, the term postsocialism was used as early as 1994 in an essay published by Paul Pickowicz and another coauthored by Chris Berry and Mary Farquhar, both of which focused on Black Cannon Incident (Hei pao shijian 黑炮事件; Huang Jianxin黄建新, 1985), with Berry and Farquhar also applying the term to Yellow Earth (Huang tudi 黄土地; Chen Kaige, 1984).9 Berry would employ the label in discussing a set of films from even earlier in his book Postsocialist Cinema in Post-Mao China, which covers films made in the late 1970s to early 1980s.10 Yingjin Zhang later parsed the concept of postsocialism as a label for historical periodization, a structure of feeling, a set of aesthetic practices, and a regime of political economy, all as reflected in cinema of the “urban generation” that emerged in the 1990s.11 Berry emphasizes that “postsocialism is a disaggregated phenomenon that emerges gradually in fits and starts in different parts of the socio-cultural formation in different ways and at different times.”12 I agree with this flexible use of the term, even though elsewhere I have argued that a generalized condition of “postsocialist modernity” did not really predominate until the 1990s.13

I first used the specific phrase “postsocialist realism” in a discussion of the early films of Jia Zhangke 贾樟柯, in which I argued that those films are post–socialist realist in two senses: first, they are a realism of the postsocialist social and economic condition, and second, they self-consciously supplant and oppose the cinematic aesthetics of socialist realism and its successor, state-sanctioned “main melody” or “leitmotif” (zhuxuanlü 主旋律) cinema, as well as commercialized entertainment cinema.14 For the purposes of this chapter, I hyphenate post-socialist to highlight this second sense, more clearly retaining the implied “socialist realism” against which post–socialist realism began to differentiate itself soon after the Cultural Revolution. Post–socialist realism invokes the legacy of Maoism—sometimes directly through nostalgic or parodic citation, but more generally by developing as its dialectical other in form. The decades-long phenomenon of post–socialist realism began with critical discussions in 1979 and in films of the early Deng Xiaoping 邓小平 reform era, and it eventually blended into more contemporary practices, such as independent digital video (DV) filmmaking, which is discussed in both this and the next chapter.

In their landmark 1979 essay “On the Modernization of Film Language,” filmmaker Zhang Nuanxin 张暖忻 and critic Li Tuo 李陀 in many ways anticipated the course of development of post–socialist realism in Chinese cinema over multiple decades and “generations.”15 As Yingjin Zhang has noted, although the essay was first known as a manifesto for the Fourth Generation filmmakers—most of whom began making films immediately following the Cultural Revolution, although they had been trained earlier—many of the ideas in the essay did not come fully to fruition—or result in films that gained widespread recognition, particularly internationally—until the films of the Fifth and Sixth Generation Chinese filmmakers (many of whom had studied with instructors from the Fourth Generation).16 Indeed, it is remarkable how well the essay both encapsulates the significance of its historical moment—just months after Deng Xiaoping had consolidated power and launched the reform era—and portends developments in Chinese cinema that would play out over the decades to come. The authors not only addressed and deepened points that recently had been made by others—most notably a critique of Chinese film as having become overly dramaturgical—but also began what would be a decades-long engagement with inter- and postwar Western film theory and practice, in particular, the methods advocated by French critic and theorist André Bazin and the “global neorealism” indebted to the Italian Neorealist movement that he championed.17

The polemic against the overdramatization of Chinese film had been launched earlier in the same year, when Bai Jingsheng 白景晟 urged Chinese cinema to “throw away the crutch of theater.”18 The essence of drama, according to Bai, was conflict (chongtu 冲突), and Chinese filmmakers were all too apt to use dramaturgical methods of plot construction. Film, he argued, does not have to focus so exclusively on fictional characters and their conflicts but can express things through natural landscapes or objects as well as through techniques distinctive to film. In their essay that soon followed, Zhang Nuanxin and Li Tuo reiterated Bai’s criticism of overdramatization, but they placed it in the context of a teleological narrative of world cinema history, the urgent overall message of which was that Chinese cinema had fallen behind and must “catch up with the development of world cinematic art.”19 Even in that larger frame, drama becomes cinema’s foil, with the progress of film art measured by how much it distances itself from drama. Early film is said to have been dependent on theater but to have gradually established its own cinematic aesthetic by the end of the silent era. The coming of sound prompted a temporary retrogression, in which dramatic principles again came to dominate film, but movements like Italian Neorealism and the French New Wave helped to steer cinema back to its own path and made significant advances in film aesthetics. The overarching history told by Zhang and Li is summed up by their claim that “a trend in the modern development of world film art is that the narrative mode of film language (or one could say film’s structural method) has increasingly shed its theatrical influence and instead in every respect advanced toward becoming more cinematic.”20

Zhang and Li’s teleological narrative of film history suggested that cinema had a universal history from which China had quite simply been absent. Of course, the idea that China must catch up with the world was not so subversive for the essay’s historical moment, playing as it did into the overall ideology of the early Deng era; indeed, Zhang and Li explicitly cite the objective of “meeting the wider aesthetic needs of the people for cinematic art that can realize the Four Modernizations in the New Era.”21 Such appeals to the ideology of progress implicit in the official slogan of modernization may have been in part a rhetorical strategy in that, if filmmakers and film critics wanted to see more attention paid to artistic form and less to political content, ironically, their arguments might have been most effectively framed within the political master narrative of the Four Modernizations. Similarly, the rejection of “dramatic conflict” at this moment was in part a coded, tentative way of challenging the socialist realist orthodoxy that class struggle should be at the center both of art’s function and its content. In lamenting the excessive influence of drama, with its overreliance on “conflict,” these critics were obliquely critiquing not only dramaturgical film aesthetics but, more broadly, the instrumentalization of film by revolutionary politics. In the rhetoric of these essays, “drama” stands in for the aesthetics of socialist realism in general, so that the call for film to evolve from the dramatic to the properly cinematic is not simply about the intermedial relationship of film to theater but more fundamentally about a new ideology of depoliticization, calling for an increased autonomy of art from politics in the reform era. Zhang and Li repeatedly make the obligatory gesture of blaming the Gang of Four for the stalled development of cinema in China, and shortly after their first mention of the Gang of Four, they somewhat daringly go further: “What needs to be pointed out is that the trend that, in artistic creation, one can speak only of politics rather than of art, only of content rather than of form, only of the artist’s worldview rather than of artistic skill, did not completely begin with the rampage of the Gang of Four.”22 They go on to urge a more sophisticated view of the Marxist theory of the relation of form to content and, implicitly, the relation of art to political economy, or superstructure to base.

Zhang and Li explicitly praised French and Italian politically inflected films that “do not rely on the concept of dramatic conflicts to develop the plot” but instead “employ a ‘realist’ [jishi] method from beginning to end, imbuing the films with a very strong documentary style, making it seem as if what we see is not a fabricated story but rather is like an actual record of a real political incident.”23 The methods of cinematic realism thus are explicitly tied to a realistic representation of history, in implicit contrast to the highly formalized revolutionary myths that came to dominate filmmaking in the Mao era. The essay also credits André Bazin with leading a group of critics and filmmakers in challenging Eisenstein’s montage theory and instead pursuing the “long-take theory.”24 Indeed, the essay launched what would become a decades-long fascination with Bazinian aesthetics among many Chinese critics and filmmakers—in terms of both the promulgation of Bazin’s film theory and the championing of Italian Neorealism as a model. As Cecile Lagesse puts it, Bazin’s “approach to cinematic realism would allow filmmakers and critics to rethink cinema’s relationship to reality after the Cultural Revolution and to free cinema from its former role as an ideological conduit.”25 Less than a year after the landmark essay on the modernization of film language, Li Tuo followed with a more specific essay titled “An Attention-Worthy School of Cinematic Aesthetics—On the Long-Take Theory,” coauthored with Zhou Chuanji 周传基.26 Lagesse documents a number of other significant articles that would follow over the years, and even those only scratch the surface of Bazin’s penetration into the Chinese discourse on cinema and the remarkably consistent importance of his ideas for various filmmakers from the Fourth Generation through the Sixth Generation.27 Jia Zhangke, a film theory concentrator in the Beijing Film Academy (BFA) literature department during the mid-1990s who would go on eventually to become the most renowned Sixth Generation filmmaker, virtually paraphrases Bazin in many of his interviews, even if he rarely mentions him by name. Bazin’s influence continued into the new century; the leading film journal Contemporary Cinema (Dangdai dianying 当代电影), for example, published a special section on Bazin in 2008. A full-text search for Bazin in the Chinese Academic Journals Database turns up references to him in 601 articles published from 1979 to 1989, 799 articles published from 1990 to 2000, and an astounding 2,407 articles published between 2001 and 2011.28 I will give some detailed examples of how Bazinian aesthetics manifested in post–socialist realist Chinese films later in this chapter, but first we will consider how these films subverted the genre conventions—and thereby the historical claims and political ideology—of Mao-era socialist realism.

Genre Subversion

Many films of the post-socialist era adopted genre stereotypes of the Mao era, but (to paraphrase Schweinitz) rather than adhering to the conventionalized worlds of their generic precedents, they instead reflexively conduct a critical discourse on them.29 Their undermining of prescriptive realism relies partly on an unmasking gesture that claims a truer sense of realism: “In its systematically organized references to the mythological world of the genre, the film seems to wish to express: Look, this is the abject reality hidden behind by the mythical world of appearances.”30 I have written previously about one example, Black Cannon Incident, which cued and then systematically subverted the genre conventions of the Mao-era spy thriller.31 In that film, a black cannon Chinese chess piece comes to embody the entire mythical world of antiforeign suspicion that the counterespionage genre upheld, which eventually is revealed to be an ideological fantasy empty of any connection to reality in the context of China during the Four Modernizations.

In the films examined in chapters 4 and 5 of this book, the main legitimating story of the People’s Republic of China (PRC) told by cinema was the master narrative of Communist victories in the civil war against the Nationalists as well as the war of resistance against occupying Japan, so first we will focus on post–socialist realist films dealing with war and the military. The undermining of the myth of Communist military prowess began in particular with some early films of the so-called Fifth Generation—the first generation of filmmakers to enter (in 1978) and later graduate (1982) from the Beijing Film Academy, the nation’s primary film school.32

One of the earliest feature films made by the Fifth Generation, The One and the Eight (Yi ge he ba ge 一个和八个; Zhang Junzhao 张军钊, 1983), follows a group of Eighth Route Army soldiers as well as nine Chinese prisoners they are escorting as they all retreat from the advancing enemy during the early stages of the Japanese occupation. Controversial and only released later after more editing, the stunningly shot film is the first for which Zhang Yimou worked as the primary cinematographer. It begins with an authoritative-sounding voice-over narration setting the historical stage with rhetoric about the Communist Party uniting the people in patriotic resistance to the Japanese, thus cueing the genre expectations for anti-Japanese war films set by countless films of the Mao era featuring heroic resistance led by the Communist Party. The genre is quickly defamiliarized, however, as the retreating troops are unsure even about whether their Chinese prisoners are good guys or bad guys, and often mutual mistrust as well as the barren wasteland through which they retreat—featured in extreme long shots that make the human figures seem tiny and vulnerable—appear to be as much of a threat as the Japanese enemy. All the main characters were played by actors who looked like the desperate and in some cases outlaw characters they were supposed to play. In his insider memoir, Ni Zhen tells of how the actors shaved their heads and suntanned their bodies in preparation, to the point that they scared some locals while passing through train stations to reach the shooting location. Such a lifelike depiction of bandits, writes Ni Zhen, contrasted starkly with the “caricatured villains” of the earlier Chinese war movies, “who were as stereotyped as Chinese opera masks.”33 Other markers of authenticity—like on-location shooting in the rugged and sweltering Hebei desert and elaborately weathered uniforms worn by the soldiers—gave the film a very different look from the studio sets and crisp costumes of the earlier films.34 In the end, rather than sharing much DNA with patriotic war films like Guerrillas on the Plain, The One and the Eight turns out to be more in the lineage of Western philosophical antiwar films like All Quiet on the Western Front (Lewis Milestone, 1930), The Grand Illusion (Jean Renoir, 1937), and A Walk in the Sun (Lewis Milestone, 1945), with undercurrents of absurdism and existential bleakness and an art film mode of narration. In a typical bit of dialogue, the question “We’ve been running for a day—are the [Japanese] devils chasing us, or are we chasing them?” is answered with “It’s war: you chase me, and I chase you.” Instead of inevitable triumph under Communist leadership, the film conveys war as a redundant cycle of pointless exhaustion.

Later, Zhang Yimou would serve again as chief cinematographer for The Big Parade (Da yuebing 大阅兵), made by a BFA classmate, Chen Kaige. Although also a military film, in this case, the setting is contemporary rather than historical, and there is no enemy with which to contend, just preparations for the upcoming National Day military parade in Tiananmen Square. Often read as an ambiguous allegory about individualism versus collectivism,35 The Big Parade features voice-over narration by multiple characters with quite different voices and viewpoints and a drill sergeant who expresses doubt about whether he even understands his own soldiers. The film’s release was delayed until 1986 partly because the Guangxi Film Studio, the military, and the Ministry of Culture Film Bureau insisted that it should end with the titular big parade on National Day—whereas Chen had planned to end it with footage of an empty parade ground.36 Even in its compromised final cut, the film is striking for its depiction of a People’s Liberation Army without any requisite heroism (though with humanity) and for its attention to the vast differences in personality and motivation among the soldiers. Paul Clark observes that the film’s “profound ambivalence” regarding military service matched the filmmakers’ own ambivalent feelings about China, and the lack of any obvious patriotism itself marks the great distance of the emerging post–socialist realism from the univocal nationalism of Mao-era war films.37

Whereas The One and the Eight and The Big Parade undermine the myth of Communist military victory largely through their ponderous art film style and relative lack of heroic action, a much later post–socialist realist film, Jiang Wen’s 姜文 Devils on the Doorstep (Guizi laile 鬼子来了; 2000), represents a more thorough dismantling of revolutionary-era war film conventions through black comedy and parody. The film’s Chinese title, “the devils are coming,” immediately cues the expectations of the historical master narrative of resistance against Japanese aggression. Using the common slur against the Japanese occupiers, that exact line of dialogue had appeared in battle scenes of multiple classic Mao-era war films.38 As Gary Xu has detailed, the director’s own memory of watching such films while growing up molded his imagination of the war, and it is those mediated prosthetic memories that he playfully explores in Devils on the Doorstep, beginning with the fact that it is filmed almost entirely in black and white, in keeping with most of its Mao-era precedents.39 As Devils on the Doorstep begins, the setup appears to confirm such genre expectations. A Communist guerrilla fighter knocks on the door of a local peasant’s hut to seek assistance for the underground resistance in a Japanese-occupied area. A similar scene had appeared near the beginning of Railway Guerrillas, for example, in which guerrilla fighters knock at a village hut, a voice from inside asks, “Who is it?” and the guerrillas reply, “Me. Hurry and open the door!” In Devils on the Doorstep, the peasant inside the hut is the hapless protagonist, Ma Dasan, and the situation is immediately made comical by the fact that Dasan is having sex with his girlfriend inside when the knock occurs. While his girlfriend scurries to hide, Dasan has the same brief exchange with his unseen visitor not once but three times: “Who is it?” “Me.” In Railway Guerrillas, the villager opens the door, and a warm reunion ensues, the conventions of socialist realism dictating that the Communist cadre must be handsome, courageous, and in solidarity with the peasants on whose behalf he fights to liberate the country, while the peasants whose help is enlisted would in turn be inspired to revolutionary consciousness by the brave cadre. But in Devils on the Doorstep, the lone Communist character is hardly seen, except as a handgun and then a sword extended into the frame from off-screen—weapons that are used to threaten the fearful peasant Dasan into compliance with the guerrilla’s demands: to keep watch over a kidnapped Japanese soldier and his translator until the Communists return. Neither the unseen guerrilla nor the scared and confused Ma Dasan appears to be particularly admirable, and the entire film is in fact devoid of heroism. The Chinese people under occupation are depicted hardly more favorably than either their brutal Japanese occupiers or the dim-witted American soldiers who appear at the end of the film, after the Japanese have at last been defeated. Even the casting of the film mocks its Mao-era generic precedents. Attentive viewers would have recognized the actor playing the would-be executioner “One-Stroke Liu,” an ostensibly legendary swordsman who turns out to be merely a crazy old man, as the eighty-plus-year-old Chen Qiang 陈强, the “villain star” who had portrayed evil landlords in the Mao-era classics The White-Haired Girl and the original Red Detachment of Women—his menacing performances in those films now being reduced to ridiculous parody. Stylistically, much of Devils on the Doorstep employs aspects of neorealism, with mostly handheld cameras shooting often in claustrophobic interior spaces lit by a single light source, plunging much of the frame into darkness, and many minor characters being played by nonprofessional actors, some of whom also served as crew members.

Still other classics of post–socialist realism perform similar generic subversions, not of war films but of the party-promoted master narrative of China’s liberation from poverty and feudal oppression and its subsequent economic development under Communism. The later stage of the Cultural Revolution, for example, had featured not only revised versions of war films like Guerrillas on the Plain but also films like The Fiery Years (Huohong de niandai 火红的年代; Fu Chaowu 傅超武, Sun Yongping 孙永平, and Yu Zhongying 俞仲英, 1973), about improved methods of steel production, and Hongyu (红雨; Cui Wei 崔嵬, 1975), about expanding rural health care through training “barefoot doctors.” Both films had included obligatory villains—class enemies who tried to sabotage the progress in socialist construction—so that the “drama” of class conflict still was depicted, but mainly they convey an image of ever-improving national wealth and well-being through socialist development. These types of narratives also came under the critique of the post–socialist realist films beginning in the 1980s.

One example is Yellow Earth, the definitive early Fifth Generation tour de force featuring the team of Chen Kaige as director and Zhang Yimou, soon to become a globally famous director himself, as cinematographer. The film is set on the remote Loess (or “yellow earth”) Plateau along the Yellow River in north China during the civil war period. Gu Qing, a handsome visiting cadre from the Communist base area in Shaanxi, is idolized by Cuiqiao, a young woman oppressed by the feudal patriarchal traditions of her poverty-stricken region. For the Chinese spectator of the immediate post-Mao period, this would quickly bring to mind the structure of apprenticeship found in such red classics as The Song of Youth and The Red Detachment of Women, in which a spunky young woman with instinctive revolutionary tendencies is mentored by an attractive, young, male Communist Party member. In those cases, as we saw in chapter 4, a romantic bond between heroine and cadre is suggested through the visual conventions of Hollywood romance, even if it is absent in explicit form in the script. In Yellow Earth, however, Cuiqiao is a girl of fourteen, complicating the audience’s perception of any implicit romance, even though clearly something like a teenage crush is happening (Figure 29). To make matters worse, the cadre utterly fails to save her from the awful fate bestowed on her by local traditions. Cuiqiao is married off against her will to a much older man, and she eventually dies trying to flee across the Yellow River to search for the Communist base after Gu Qing neglected to return to her village before her forced marriage to give her a chance to fulfill her dream of joining the revolution. The thwarting of the norms of Chinese socialist realism, both artistic and ideological, was quite conscious and intentional. As Chen Kaige himself has said, “from the perspective of art, and even politics, we had a very strong desire to reject the entire age that had come before us. The films made before and during the Cultural Revolution were simply outrageous. . . . It was a very conscious thrust pushing us to negate everything that had come before.”40

The brilliance of Yellow Earth—aside from the stark beauty of Zhang Yimou’s cinematography—is the way in which the crux of the story, including its deep subversiveness, is told in the image track while being largely absent from the film script itself. The film is an excellent illustration of John Mullarkey’s point (already mentioned in chapter 4) that “the art of cinema involves a fabulation of images, a narrative of images that may well (indeed, will) ‘thwart’ the textual narrative, interrupting it, contradicting it.”41 The filmmakers take full advantage of this potential of cinema and use it to undermine the familiar character stereotypes and narrative scenarios of Chinese revolutionary cinema and its master narrative of liberation and progress under Communist revolution and rule. In fact, if one is not familiar with the conventions of Mao-era revolutionary cinema, it is impossible to grasp Yellow Earth’s subtle subversiveness—as is evident from a clueless review published in the New York Times when the film was screened belatedly in New York City in 1986 (the same reviewer would be equally befuddled by The Big Parade, interpreting both essentially as Communist propaganda).42 Evgeny Dobrenko has observed that “the word yields to censorship more easily than the visual image,” giving the example of Eisenstein’s Ivan the Terrible, which visualized a screenplay approved by Stalin “in such a way that the ideological task of the film was completely subverted.”43 Much the same was the case with Yellow Earth, through the details of its images and even its pacing.

Sometimes the slightest of hesitations in performances or editing rhythms can undercut the anchoring status of the socialist realist hero. One example of the image track’s subtle sabotage of socialist realist conventions is its defamiliarization of the “socialist realist gaze,” which Stephanie Donald first defined in a discussion of Yellow Earth. Her close reading of one moment in the film—precisely when Gu Qing is making his eventually broken promise to Cuiqiao—shows how that customary authoritative look toward an off-screen horizon, indicating the sure ideological vision of the Communist hero, is devastatingly hollowed out through a combination of performance cues and editing.44 To that example I would add a minor moment early in the film, when Gu Qing has just arrived at the village and is invited to join a wedding feast in progress. The locals, played by nonactors who lived where the on-location filming took place, appear very believable as hardworking peasants suffering from extreme scarcity. There is a peculiar moment when Gu is led to sit at a table with some local men, who are told he is a soldier from the Communist base area in Yan’an who is visiting the farming village to collect folk songs (following the Maoist cultural practice of drawing on local folk arts and repurposing their forms with revolutionary content). The men all nod and just say “Oh” together in acknowledgment, staring at him as a strange outsider. When Gu invites them to sit and begins to sit down himself, all the men remain standing. Gu catches himself and straightens back up to a stand, whereupon the peasants all sit down, out of sync with him. The film then cuts to a medium close-up of Gu looking around at the men in puzzlement, as if he does not understand how to relate to them. It is a small moment in which the images speak much louder than the words. At formal meal gatherings in China, it is customary for the highest-ranking person to sit first, so the momentary confusion about the timing of sitting down indicates a lack of shared sociality between the visiting soldier and the locals, who first wait while he begins to sit, then go ahead and sit down themselves while he remains standing. As was the case with Devils on the Doorstep, the mutual estrangement between soldier and peasant undermines the message—sent in countless ways in previous revolutionary films—of Mao’s ideal of “the soldiers and people as close as fish and water” (junmin yu-shuiqing 軍民魚水情). The rift here is established clearly at the beginning of the film but through images rather than dialogue (which is remarkably sparse in the film); as Jerome Silbergeld notes in his illuminating reading of Yellow Earth, with such a bare-bones script, “unspoken gesture, even reticence itself . . . become all the more poignant and revealing.”45 The inexpressible, unbridgeable gaps in understanding between the revolutionary party members and the people for whom they claim to speak call into question the cadre’s very mission: to connect with the rural masses by collecting and repurposing folk song melodies.

I will add two more examples of the genre subterfuge in the film. I have noted elsewhere that one of the central messages of Mao-era cinema can be summed up as “Communists have more fun!”46 In historical films of the 1950s and 1960s, regions controlled by the Nationalists are usually depicted as literally darkened areas in which the common people weep, their oppressors frown or sneer, and nobody seems truly happy. In contrast, the Communist base areas are invariably depicted as bathed in bright light and filled with lighthearted, unalienated laborers at work—people laughing and playfully splashing each other as they wash clothes in Red Detachment of Women, for example. Often the people and soldiers in the Communist-held territories break into joyful and energetic song or dance. The message that “Communists have more fun” is severely problematized (possibly even lampooned) by the famous waist-drum dance sequence in Yellow Earth. The film cuts temporarily to the Communist base in Yan’an, from which the Communist cadre had come to the girl’s village and to which he had returned after falsely promising that he would come back to fetch her before the date of her forced marriage. True to generic precedent, the waist-drum dance sequence depicts vigorous Communist soldiers performing an energetic, joyous dance in the bright sunlight, enjoying the blessings of living in a liberated area. However, the entire sequence is undermined by the shot that immediately precedes it—a subtly horrifying shot of Cuiqiao’s bridal veil being raised by the dark, grizzled old hand of the husband who essentially is about to rape her. The joy of the liberated area is put directly in the context of the fate of the helpless girl for whom Communism held only a false promise, undermining the dance’s vivacious joy and energy.

A related key moment occurs at the very end of the film, when Gu Qing at last returns to the village, too late to rescue Cuiqiao but perhaps still in time to help her silent little brother, Hanhan, who also had grown attached to Gu Qing during his earlier visit. At the end, Hanhan sees Gu approach over a hill on the horizon in extreme long shot and struggles to run to him through a throng of local men engaged in a ritual rain prayer to help their drought-ravaged local farming community—a rite symbolic of the feudal squalor and superstition of the preliberation village. Repeated cuts from shots of Hanhan running through the crowd to Gu Qing arriving on the horizon, in the exact same shot repeated three times, end with a final shot of the suddenly now empty horizon. As the soundtrack plays a song that ends with the line “It’s the Communists who save the people,” the last shot of the film shows only the barren yellow earth with a now empty horizon, the Communist savior having apparently disappeared. At this point, the film has become more of an abstract allegory than a realistic story, but the image’s undermining of the message in the script could not be more clear.

The leading critic Li Tuo, coauthor of the “Modernization of Film Language” essay, compared watching Yellow Earth to answering the door at a party in expectation of welcoming an old friend, only to be confronted by a complete stranger, resulting in an awkward silence and even possible hostility.47 One of the film’s most controversial aspects was its “depiction of the poverty of the land and its people.”48 This would not have been as awkward for viewers—the film being set in the 1930s, depicting people who are supposed to have been suffering from the deprivation of preliberation feudalism—except that the filmmakers used mostly local farmers to play roles other than the four main characters. This neorealist technique provoked the unsettling realization that local farmers of the early 1980s, more than three decades after liberation, continued to live under great hardship, thus subverting the narrative of progress under Communism. That is, the documentary technique of using local nonprofessionals as extras in the film severely disrupts the historical master narrative of socialist realism, insofar as it places the visible poverty of peasants in the Communist present, not just in the feudalist past, thus undermining the entire master narrative of the Mao era as bringing liberation from oppression and hardship. The implication is that the party thus far had failed the people, just as Gu Qing had failed Cuiqiao.

Of course, the critiques of the Mao era found in Fourth and Fifth Generation films of the 1980s arguably helped to legitimate the policies of the new reform-minded Communist leadership. However, later waves of post–socialist realism would extend the critique of the myth of progress to the reform era itself. While the generation that emerged in the 1980s could be seen as subtly undermining the legitimacy of the previous party leadership while implicitly endorsing the new order, by the 1990s, a successive generation of post–socialist realist filmmakers would challenge the myth of progress during the post-Mao era as well. The early films of the pioneering independent directors Zhang Yuan 张元 and Wu Wenguang 吴文光, for example, often focused on disillusioned youths who had lost the idealism of the 1980s, while many slightly later independent films, particularly those following in the footsteps of Jia Zhangke’s early films, such as Xiao Wu (小武; 1997), reinforced the critique of the reform-era economy by featuring its losers—pickpockets, prostitutes, migrant laborers, coal miners, small-time hustlers—again, often played by nonprofessional actors. The latter trend continues well into the new millennium, including in the form of low-budget, “underground” fictional films and documentaries shot on digital video. In their very choice of human subjects and experiences to highlight, such films both call into question the myth of historical progress and firmly reject socialist realism’s tropes of heroism. Jia himself expresses an utter lack of interest in creating heroes and notes that his film-school understanding of how to build drama enabled him precisely to undercut it. Even in scenes with plenty of potential tension and conflict, instead of following the “rules of drama,” Jia says he would ask himself, “What would the logic of realism dictate for a scene like this?”49 In this and many other ways, Jia and similar post–socialist realist directors echo the “Modernization of Film Language” essay’s disparagement of drama and follow in the footsteps of the Italian Neorealist filmmakers admired by Bazin, who praised “the pride of place they all give to the representation of reality at the expense of dramatic structures.”50 We now turn to the specific neorealist techniques that provoke such an impression of reality rather than artificial drama.

Stylistic Neorealism

Italian Neorealism was a reaction against both Hollywood’s dream factory and the prewar Italian “white telephone” films that largely imitated Hollywood.51 Chinese post–socialist realist films similarly were a reaction against socialist realism and the Hollywood-based classical style on which socialist realist as well as reform-era mainstream Chinese entertainment films drew. Here I will consider several methods that they emphasized instead, including on-location shooting, the use of nonprofessional actors, an emphasis on contingency, and the interrelated deployment of detail, ambiguity, and elliptical editing, before moving on to a lengthier discussion of one of their stylistic mainstays: long-take/long-shot cinematography.

We already have seen how Yellow Earth used on-location shooting and nonprofessional actors drawn from local residents to lend a powerful sense of realism to its depiction of place and social setting. Not just the exterior landscape scenes but even interior scenes, such as those set in Cuiqiao’s family’s cramped and dimly lit hovel, give a strong feeling of authenticity to the rural poverty depicted. Insofar as any on-location shot serves as a documentary image of a real place, even in a fiction film, on-location shooting is generally taken as better than studio sets in building cinematic realism. The rigorous dedication to on-location shooting for Yellow Earth contrasts, for example, with the socialist realist Plains films discussed in chapter 5, including the original 1955 Guerrillas on the Plain, the 1974 remake, and the 1974 opera film Fighting on the Plain. While the model opera film quite obviously was shot entirely on artificial sets, the two more “classical” versions also often alternated—even within the same scene—between on-location exterior longer shots that established the setting and closer shots of characters talking that almost certainly were shot at a different time on a studio sound stage. Viewers would likely have taken little conscious note of the discrepancy owing to careful continuity editing that guided their engagement with narrative developments rather than details of background setting, but the cuts between on-location and studio shots, combined with the use of more editing rather than longer takes (see later), nonetheless would have lessened those films’ impression of realism. The strong preference for on-location shooting in the post-Mao era resulted in the practice of small teams of collaborators—the director, cinematographer, production designer, and sound designer, for example—extensively scouting the locations where the early Fifth Generation films would be shot. This was the case for Yellow Earth, The One and the Eight, and an even earlier Fifth Generation film to be discussed shortly, The Red Elephant (Hong xiang 红象; Zhang Jianye 张建亚, Xie Xiaojing 谢小晶, and Tian Zhuangzhuang 田壮壮, 1982).52 During shooting, the use of natural lighting even for many interior scenes resulted in images that could be unusually dark, sometimes to the dismay of studio executives reviewing the rushes.53

Later post–socialist realist filmmakers, such as Jia Zhangke, took on-location shooting and, especially, nonprofessional actors to an even stronger level of commitment. His groundbreaking Xiao Wu ends with a title that informs the viewer, “The entire film is interpreted by non-professional actors,” and in several exterior scenes, unwitting “extras” look directly at the camera, so that he sacrifices a bit of the fictional realism that calls for the camera to be an invisible observer in favor of the documentary realism gained from shooting on location with crowds of real people on city streets.54 Such a reliance on documentary techniques was common among China’s first generation of independent filmmakers (a group that mostly overlaps with the Sixth Generation label, at least at the beginning of their careers) who emerged in the 1990s and in many ways continued and intensified neorealist trends begun by their predecessors in the 1980s but also inched closer to documentary-style realism. China’s pioneering independent fiction film director Zhang Yuan used a raw, verité documentary style in his films that shared a close affinity with the videos of then emerging independent documentary filmmaker Wu Wenguang and his peers. In fact, Zhang’s first feature film, Mama (妈妈; 1990), about a mother’s struggles to raise her disabled child, intercut the fictional story with actual documentary interviews with mothers of disabled children in Beijing. The same filmmaker later made documentaries of his own as well as a string of underground films focusing on marginalized figures, such as rock musicians and closeted gay lovers. Soon a handful of other film directors—including Jia Zhangke—were making low-budget, independent films in a similar style that emphasized the values of documentary or “on-the-spot” realism (jishizhuyi 纪实主义) imbued with a strong sense of live shooting on location (xianchang 现场).55 Still later, independent DV filmmakers continued the trend well into the new millennium, with directors like Liu Jiayin 刘伽茵 and Song Fang 宋方 shooting fiction films that feel like documentaries because they are shot in the filmmakers’ actual family homes and feature their own family members playing versions of themselves and generally appearing unscripted (although in fact they are acting out prewritten scenarios).

One result of the post–socialist realist preference for on-location shooting and nonprofessional actors is an opening up to, and valuing of, contingency entering the frame. In his classic work of realist film theory, Siegfried Kracauer asserted that film “gravitates toward unstaged reality” because of the ability of photographic media to capture “actual physical existence” beyond human intentions.56 Whereas prescriptive realism, in its effort to depict a world conforming to an officially sanctioned ideal, would seek to minimize unscripted, contingent elements that may distract from the intended film message, the neorealist countermovement in cinema welcomed the unpredictability that comes with shooting either in nature or in the streets with nonprofessional actors. The small accidents that might result represent unforeseen intrusions of the real into the fictional world, which, no matter how minor or subtle they may be, contribute to the sense of realism in the film through their very lack of human intentionality. For a small example, we can turn to the very first feature film directed by the Fifth Generation, the children’s film The Red Elephant, about three ethnic minority children in southern Yunnan who go on an adventure together in the tropical rainforest, trying to track a legendary red elephant. When, early in their journey, the children come to a river that is too big to swim across safely, they turn a floating log into a makeshift ferry, with one boy push-poling the log into the middle of the river while the other boy and girl, along with the girl’s pet German shepherd, sit or squat on the log. The first shot of the whole procession comes as a series of minor revelations for the viewer, beginning with just the surface of the river, then the log entering the frame from off-screen right, mostly submerged, but with the older boy wielding the pole standing on it, so that he almost seems to be standing on the water itself. As the log floats through the frame, the other boy, the girl’s dog, and the girl with her basket enter the frame in succession. About sixteen seconds into the twenty-six-second shot, the camera begins panning left to follow the figures on the improvised ferry. At the exact moment that the camera starts moving, one of the dog’s paws slips briefly off the log into the river, almost as if the camera movement was momentarily destabilizing to the precarious vessel. Of course, the dog’s brief slip and its timing were entirely coincidental, but the chance moment indicates a poetics of openness in the film, in which the children, the various animals (elephants, monkeys, birds, and even a leopard, in addition to the dog), and the jungle itself have some agency in shaping the film’s impressions due to their unpredictability (much like the wave in the Lumière brothers’ 1895 actuality Boat Leaving the Port, as discussed in chapter 1). A small detail like a paw slip during a leisurely, entertaining long take can feel like a cinematic event in itself to a viewer attuned to the revelations of small details or “cinephiliac moments.”57 As Mullarkey points out, both animals and child actors in a film potentially interrupt the intended performance of a fiction by introducing the “randomness of the Real,” and the same can be said of on-location shooting and nonprofessional acting.58 This openness to contingency is one way that many post–socialist realist films tend toward apophatic realism (see the introduction), in that they implicitly acknowledge that cinema confronts a reality that includes much that is unknown and unanticipated.

Bazin wrote that “neo-realist filmmakers . . . endow their films with a sense of the ambiguity of reality.”59 Techniques for doing so include not just on-location shooting, unpredictable humans or animals in the mise-en-scène, long shot distances, deep staging, and long takes (to be discussed shortly) but also narratives that are themselves discontinuous, inconclusive, or open ended. In this specific sense, in addition to other reasons (their critical realism, for example), many post–socialist realist films echo the masterpieces of the Left-Wing Film Movement during the golden age of Shanghai cinema in the 1930s (chapter 2).60 Several films of the late 1970s and early 1980s, for example, were noted for having the kinds of ambiguous endings that would have been unthinkable in socialist realist cinema. The banned film Bitter Love (Ku lian 苦恋, aka Taiyang he ren 太阳和人; Peng Ning 彭宁, 1980) infamously ended with an overhead shot of the body of its dying protagonist graphically tracing a question mark in the snow in which he lies. The next year, Narrow Street (Xiao jie 小街; Yang Yanjin 杨延晋, 1981)—so metacinematic and unconventional that Chris Berry has called it “self-conscious and possibly Godardian”—offered three different possible plot endings and left its audience to weigh them.61 A little later, Yellow Earth ended with the previously mentioned ambiguous images of the now-empty horizon, leaving viewers suspended in doubt. The trend of ambiguity and inconclusiveness only grew stronger when the independent film movement of the 1990s and beyond took over the post–socialist realist tendency. For example, in Jia Zhangke’s coming-of-age epic Platform (Zhantai 站台; 2000), the ending seemingly skips years ahead in time and shows a couple we thought had long since broken up seemingly now with an infant sharing a domestic space. With no dialogue, we are left to interpret the scene despite the gaps in our knowledge.

Another key feature of post–socialist realism, mentioned from the earliest critical writings of the reform era, was the importance of incidental detail. In their seminal essay on the modernization of film language, Zhang and Li approvingly noted that “in Italian Neorealist cinema, life is generally depicted through ordinary details,” in contrast to films that follow “theatrical rules,” with their “complicated twists, outlandish plots, and exaggerated, violent conflicts.”62 In a later critical essay published in 1984, Yang Ni 杨妮 continues and deepens the critique of overly dramatic film and, like Zhang and Li, points to Italian Neorealism as the model for making films less dramatic.63 For her, this does not mean that a film should lack a plot or conflict but that the plot should be relatively loose and possibly have multiple strands and that any conflicts should be realistic rather than overly dramatized. Film narratives, Yang argued, should not necessarily have “dramatic plots in which closely related events, cause-and-effect relationships, and step-by-step developments lead to a climax” but rather should have “parallel plots,” “few or no dramatic climaxes created out of intensified contradictions,” and—most important—narrative lines that are frequently interrupted by minor details.64 Also like Zhang and Li, Yang invokes Bazin as she argues for film as a fundamentally documentary medium that depends most crucially on the indexicality of photography: “Between the recorded image on film and the object there should be a relationship like that of fingerprint to finger.”65 In keeping with the idea of film as what Kracauer called “the redemption of physical reality”—a phrase quoted in Yang’s essay—the natural development of the art, according to Yang, involves “discarding theatricality and absorbing documentary devices.”66

The minor detail not only functions as a marker of realism in the Bazinian or Italian Neorealist sense; it also participates directly in the process of the subversion of socialist realism. Take, for example, one instance of the irruption of detail in a moment from the climax of Devils on the Doorstep. The antihero protagonist Ma Dasan has gone on a vengeful rampage in which he tried to hack to death Japanese prisoners of war after their surrender to Chinese troops. He is sentenced to death, and the Chinese commander assigns the task of execution to none other than the now-imprisoned Japanese soldier whom Ma Dasan previously had been hiding at the behest of the anonymous Communist guerrilla. In terms of its revolutionary film precedents, this would be the most dramatic moment in the film, when the main protagonist is to be executed by the Japanese enemy, but of course, it is undermined by the facts that the war is over and the protagonist did nothing particularly heroic to lead to his martyrdom. Moreover, at the very moment when the Japanese prisoner is poised to behead the kneeling Ma Dasan with his samurai sword, an ant suddenly appears on Ma’s neck just where the sword is to cleave his flesh. Distracted, in a moment of bizarre dark humor, the executioner stops what he is doing to flick the ant off the condemned man’s skin. Given the conventions of heroism in countless Chinese revolutionary films, the interruption of this crucial dramatic moment by a mere ant crystallizes the intervention of post–socialist realism as a whole, serving simultaneously as the sort of true-to-life detail celebrated in Bazin’s writings on Italian Neorealism and also as itself a disruption of the codes of socialist realism, in which executions have a very specific form of dramatized representation that viewers would have internalized (as chapters 4 and 5 discussed).

In her discussion of Mao-era socialist realist fiction writer Li Zhun, Krista Van Fleit notes that while details are “used frequently to give texts a feeling of authenticity,” they nonetheless had to be “closely monitored,” in keeping with the official (prescriptive) doctrine of realism.67 As Georg Lukács argued in championing nineteenth-century realism over naturalism as a model for socialist literature, the point of a detail in a work of art is not simply that it “corresponds photographically to life” in a way that might be “purely accidental, arbitrary and subjective”; instead, the detail must be chosen or imagined in a way that makes it “a necessary aspect of the accurate reflection of the total process of objective reality.”68 Details must not be autonomous but must be part of the “correct reflection of the totality” as envisioned by a prescriptive notion of realism.69 The alternative “danger of details becoming important in themselves” risks “a disintegration of the composition into disconnected and autonomous details.”70 Post–socialist realism is less certain about the totality to be represented and instead embraces the risks of autonomous details. As Bazin writes in one of his classic essays on Italian Neorealism, a study of Vittorio De Sica’s Umberto D. (1952), “the narrative unit is not the episode, the event, the sudden turn of events, or the character of its protagonists; it is the succession of concrete instants of life, no one of which can be said to be more important than another, for their ontological equality destroys drama at its very basis.”71 In other words, to the extent that a film narrative becomes a series of closely observed moments, each equal in its accumulation of details, the film serves the antidramaturgical agenda of post–socialist realism. Of course, this is a matter of degree, as these films still maintain narrative structures. On first viewing, Xiao Wu seems to be a series of random events in the life of a small-time thief whose life is spiraling downward, but, as Michael Berry has shown, the seemingly random series of events is in fact organized into three discrete sections of the film, centering sequentially on his relationships with his former best friend, his fleeting girlfriend, and his estranged family.72 More generally, despite his focus on the small details of lives at the margins of society, Jia Zhangke arguably engages in a form of critical realism that Lukács would recognize for its implicit critique of how neoliberal capitalism has played out in China. Some later post-socialist filmmakers of the DV generation (who will be considered at more length in the next chapter) who pursue an almost “mumblecore” style of mundane realism,73 such as Li Hongqi 李红旗 and Liu Jiayin, take the idea of a film that is a succession of concrete details and ordinary events of everyday life, all of seemingly equal importance, well beyond even what Sixth Generation filmmakers like Jia Zhangke had done.

If the detail is a realist technique that defies cinematic convention by leaving in what would normally be taken out as irrelevant, its seeming opposite—the ellipsis—achieves its realist effect by cutting out what would conventionally be kept in as highly relevant. Of course, virtually all films have a variety of “minor” ellipses that simply skip ahead to the next significant event in the plot’s causal chain.74 In such cases, for reasons of economy, story content is left out owing to its insignificance. It is the elision of significant events that helps to make a film’s neorealist style. Gilberto Perez describes such ellipses as central to both cinematic modernism and realism—the latter “because reality always exceeds its representation, so any representation necessarily leaves out much of reality.”75 As chapter 3 already noted, Bazin praised ellipses for being realistic insofar as in our real lives as well, “we do not know everything that happens to others.”76 Apophatic realism, consequently, is verisimilar in its own way, insofar as it depicts a reality that is always partly unknowable. This is part of the radicality of a film like Yellow Earth in comparison to the certainties of socialist realism. Its director explained that “the quintessence of our style can be summed up in a single word: ‘concealment.’”77 Even the apparent death by drowning of the film’s young protagonist, Cuiqiao, is indicated so indirectly that some viewers take as an open question whether she escaped to join the revolution or died trying. The ambiguity of the final scene of Platform, mentioned earlier, also is due to the ellipsis that precedes it, seemingly skipping over crucial story events, including a rekindling of a romance and a subsequent marriage and childbirth. Michael Berry has described how, in the process of editing Platform down from its much longer original version, Jia intentionally chose to leave key narrative events out of the final cut, thus forcing the audience to deduce their nature in a way that challenges the usual conventions of cinematic narration but more closely mirrors our actual experience of change.78

The Long Take

In her “Formal Look at Realism,” Kristin Thompson observes that in film history, realism conventionally presents itself as “departures from the prevailing classical norms.”79 In the case of early post–socialist realism, of course, the classical norm is not Hollywood itself but rather the form of classical-style cinema taken by mainstream Mao-era popular cinema. In the post-Mao era, the long take became one of the techniques that post–socialist realist films would foreground as a break with Mao-era cinema’s prescriptive mode of classical fictional realism. Zhang and Li’s 1979 “Modernization” essay praised the “long-take theory” of the Bazinian school as having raised a challenge to “montage theory” in the West decades earlier, and it suggested that, at the least, long-take techniques should be combined with montage to allow for more artistic possibilities.80 Bazin had advocated long takes—combined with longer shot distances, deep focus, and deep-field staging, putting more information from the fictional world on the screen in each shot—as intrinsic to a realist style, owing to their preservation of objects and events in their actual duration within their broader surroundings. In this view, rather than simply showing an event, montages of closer shots construct the event through analytical editing, cutting it up into parts so that the editing determines what parts of the event are important to pay attention to at each moment as it proceeds, thus leading the spectator to particular meanings. In contrast, long-take/long-shot cinema was said to preserve the continuity of space and time, the ambiguity of the image, and the freedom of the spectator to choose what details are most meaningful.81 For post–socialist realist critics and filmmakers, the fact that montage was tied to Soviet film theory only reinforced its association with conflict-based revolutionary cinema, and a long-take aesthetic—combined with longer shot distances—would seem consistent with the desire to wrest more autonomy for art from politics, presenting viewers with an objective “reality” without rigorously guiding their reaction to it through a finer carving of time and space. As Thompson notes, such ostensibly realist techniques often have their strongest effect when a new realist trend is beginning, “with the effect of realism diminishing as more extreme uses of the same approach appear in other films. . . . The repetition of the same realistic traits gradually makes their conventional nature apparent, and the early films that seemed strikingly realistic appear mannered upon re-viewing.”82 It is not surprising, then, that while long-take cinematography was used selectively in the 1980s to great effect, it only became a rigorously pursued distinctive style with later filmmakers of the 1990s and the first decade of the twenty-first century, and eventually it would appear as its own type of formalism when applied in a “parametric” fashion—meaning the narration of the story becomes secondary to its stylistic patterning—in extreme forms of “slow cinema” in China.

One way to measure historical changes in the aesthetics of shot length is simply to calculate the average shot length (ASL) of films from different eras. The ASL of classical Hollywood films between 1930 and 1960 ranged from eight to eleven seconds, but it decreased to five to eight seconds by the 1970s and has become even shorter since, to the point that many Hollywood films today have ASLs as short as two seconds or less.83 As we might expect, Mao-era revolutionary cinema in the classical mode had ASLs similar to Hollywood at the time. Taking some of the films examined most closely in chapters 4 and 5 as examples, the original 1961 Red Detachment of Women had an ASL of 8.7 seconds, while the 1955 and 1974 versions of Guerrillas on the Plain had ASLs of 8.3 and 7.7 seconds, respectively. Dong Cunrui, the 1955 revolutionary film mentioned earlier as context for Devils on the Doorstep’s post-Mao parody, also comes in at 8.3 seconds. Interestingly, the yangbanxi films of the Cultural Revolution had much longer ASLs than classical-style fiction films of the time. The 1974 model opera film Fighting on the Plain (discussed in the preceding chapter), for example, averaged nearly sixteen seconds per shot, with a couple of shots—both during arias—lasting well over two minutes. This suggests a potential contradiction between the post-Mao film critics’ condemnation of an excessively dramaturgical cinema, on one hand, and their preference for long takes, on the other, because the best example of long-take cinema in 1970s China appears to have been precisely the model opera films.

Yangbanxi films aside, in the immediate aftermath of the Cultural Revolution, just after Zhang and Li’s “Modernization” essay appeared, not much change in shot length was yet evident. Evening Rain, the 1980 film mentioned at the start of this chapter, had an ASL of 7.1 seconds, for example. Soon, however, shot lengths began to expand. Narrow Street had an ASL of 9.5 seconds. The mid-1980s films of the Fourth Generation filmmaker Wu Tianming 吴天明—River without Buoys (Meiyou hangbiao de heliu 没有航标的河流; 1983), Life (Rensheng 人生; 1984), and Old Well (Lao jing 老井; 1987)—had ASLs of 9, 12.1, and 11 seconds, respectively, on average about 30 percent longer than the films of the Mao era mentioned previously. Another prominent Fourth Generation film, Wild Mountains (Ye shan 野山; Yan Xueshu 颜学恕, 1986), had an even longer ASL of 12.7 seconds. As for the early Fifth Generation films of the same period, The One and the Eight had an ASL of 11.6 seconds, and Tian Zhuangzhuang’s The Horse Thief (Dao ma zei 盗马贼; 1986) had an ASL of 11.8; on the other hand, Yellow Earth averaged less than 10 seconds per shot, so there is no apparent tendency toward longer shot lengths among the early films of the Fifth Generation in comparison to the contemporaneous films of their Fourth Generation elders—further evidence that the “generation” labels may be of less use in predicting film style than the particular historical moment in question.84 In any case, these statistics show a common preference in the mid-1980s for a significantly longer-take style than in either Mao-era or contemporary Hollywood cinema.

In fact, some of the more experimental films of the early 1980s—including Evening Rain, Narrow Street, and Life—would have had longer ASLs had it not been for the tendency to indulge in the occasional very rapid montage sequence, often to represent memories or fantasies of the protagonists, which itself came off as the kind of subjective modernist (as opposed to socialist realist) technique also recommended by critics like Bai Jingsheng, Zhang Nuanxin, and Li Tuo. This hints at the possible flattening effect of ASL as a measure of long-take aesthetics, because an average says nothing about the effects of shots on either extreme away from the average. Median shot length (MSL)—though much more tedious to calculate—would be a slightly better measure of the viewer’s experience through most of the film, but that, too, fails to capture the unique aesthetics of either a brief, impressionistic rapid-fire montage sequence or a slow and contemplative long take. The latter’s aesthetic of lingering, even when inserted into a film with otherwise normal shot lengths, exposes many more details of the shot’s content and form while also almost inevitably increasing the ambiguity of the shot’s meaning, simply because the viewer has time to reflect in multiple ways on the shot before it ends. As Bazin theorized, “the very nature of editing’s analysis of reality confers a sole meaning upon the dramatic event”; long shots and long takes with depth of field build “ambiguity into the structure of the image, if not as a necessity,” then “at least as a possibility.”85

For example, about two-thirds of the way through The Red Elephant—the early Fifth Generation children’s film set in the rainforest of southern Yunnan—on the morning after the children’s second night in the jungle, a spectacular forty-seven-second tracking shot stands out precisely because it is more than six times longer than the film’s ASL of 7.4 seconds. Beginning as a relatively tightly composed long shot of the three kids sprawled asleep on the thick trunks and branches of a giant tree, the shot then tracks far, far back on a slight slope through the jungle canopy above, until the children’s figures are just tiny details that, had the shot not started with them, would not even have been noticed amid the grandeur of the enormous tree in which they rest and the wider rainforest that surrounds them. The effect would have been much more easily approximated with a zoom, but the bravura track imparts a corporeal sense of movement that makes the shot more thrilling, and the length of the take gives the viewer time to shift attention from the children at the center of the tale to the natural ecosystem they have entered, which now seems to engulf the viewer as well. At moments like this, the film hints at the potential of its makers at the beginning of their careers, including assistant cinematographer Zhang Yimou.

For the new film aesthetics of the 1980s, among filmmakers, there was perhaps no more central figure than Wu Tianming, director of several films, producer of some of the most celebrated early films of the Fifth Generation filmmakers, and, perhaps most important, head of the Xi’an Film Studio during the key years of 1983–89, when he provided a protective environment for talented film artists to take chances with experimental or “exploratory” films (tansuo pian 探索片), defending them when necessary from higher-level cultural bureaucrats. Wu arguably presided over a new high point of Chinese socialist art, when artists were encouraged to innovate without needing to prioritize an official political or ideological agenda, yet continued government ownership of the film studios still provided them with insulation against market forces and the profit motive, with all the different sorts of restrictions and censorship those entail.

As with The Red Elephant, the key films Wu himself directed in the 1980s were notable for their selective use of the long take. In River without Buoys, one extraordinary 114-second shot (more than a dozen times longer than the film’s average), of a drinking session that will not end well, uses multiple camera movements to alternate its shot distance between extreme close-up, close-up, medium shots, and long shots, all while using tracks and pans in combination to navigate the surface area of the ramshackle river raft that is the setting for most of the film. Old Well features a 158-second shot (more than fourteen times the film’s average) at one of the film’s most dramatic moments. The protagonist, Sun Wangquan (played by Zhang Yimou, also the film’s co-cinematographer), and his former girlfriend, Zhao Qiaoying, have been trapped underground by a a disastrous collapse while digging a well. For more than a minute and a half, the dimly lit shot has no dialogue, just Sun painstakingly dragging Zhao from where they both had been searching for a companion buried in the accident to a higher place where they could lean against a wall. The lingering take, with no soundtrack other than the occasional gasp of the characters and the sound of rocks shuffled by their movement, exemplifies how what Bazin (following Henri Bergson) called the “cinema of duration”86 can become a cinema of endurance for a spectator expecting a shot to be mainly a delivery vehicle for narrative information, as the shot forces us to endure the painstaking silent crawl across the rocks with the characters. Such a shot would have been unthinkable in the context of Mao-era socialist realism, which both moves the plot along much more efficiently and emphasizes clear acts of heroism or villainy, not brute struggles for existence in the face of random fate. Finally, in the shot’s second half, we find Sun and Zhao tearfully collapsing into each other’s arms, preparing to confess their ongoing love for each other (Sun is now married to someone else) before they presumably will die together underground (though they do not). These examples show how, in the 1980s, both the Fourth and Fifth Generation filmmakers employed a similar style that used somewhat longer average takes than previous Chinese cinema, punctuated by occasional extreme long takes. At the same time, they show that a long take can have very different effects, depending on all the other elements that affect the viewer’s sense of pacing, including mobile framing, dialogue, character movement, and plot developments. The long take in River without Buoys is a virtuoso display of cinematographic technique, but it does not feel the least bit “slow,” because the camera moves five times to provide very different framings, the shot distance changes accordingly, and character dialogue proceeds throughout the shot, so that narrative information is passed along both by the image and by what is revealed by the characters’ speech. The long take in Old Well, in contrast, particularly in its first minute and a half, when not a word of dialogue is spoken, feels slow to the viewer.

Despite the increase in shot length in Fourth and Fifth Generation post–socialist realism in the 1980s, it really was not until the Sixth Generation in the 1990s and early 2000s that an extreme long-take aesthetic became more of a rule throughout some films, no doubt owing not only to the continuing relevance of Bazin in Chinese cinematic discourse but also to the influence of such globally recognized long-take auteurs as Taiwan’s Hou Hsiao-hsien 侯孝贤 and Tsai Ming-liang 蔡明亮, as well as others, such as the Iranian filmmaker Abbas Kiarostami and the Hungarian Béla Tarr. Jia Zhangke’s first feature, Xiao Wu, had an ASL of over three times the average of the 1980s films mentioned earlier, and several of his later works would extend it much further to well over one minute in films such as Platform, Unknown Pleasures (Ren xiaoyao 任逍遥; 2002), and Still Life (Sanxia haoren 三峡好人; 2006). In interviews, Jia makes clear that his preference for the style has a strongly Bazinian, liberal humanist basis: “In my long shots and long takes, my goal is to respect the viewer’s agency, and even to give my films a sense of democracy. I want audiences to be able to freely choose how they want to interact with what’s on screen.”87

At the same time, because Jia’s extra-long takes often linger on characters and spaces without offering obvious new narrative information, they have much more of a tendency to distend time than was the case in the Fourth and Fifth Generation films of the 1980s, and this very slowness represents a particular view on the postsocialist condition in China. In a reading of Xiao Wu—which would become almost a template for later independent filmmakers exploring the drudgeries of life for marginalized figures in contemporary Chinese society—Chris Berry has shown how Jia’s long takes offer a particular experience of postsocialist time, as lived by the losers rather than the winners of reform-era China’s pseudo-capitalist modernization project. As a low-class migrant from the countryside who now survives as a small-time thief in a county-level city (Jia’s own hometown of Fenyang), the protagonist Xiao Wu experiences neither the progressive revolutionary time of earlier socialist realism nor the equally progressive postsocialist time of rapid economic modernization but rather seems to stand still as an “onlooker” while time passes him by.88 This insight into the implications of the long take in post–socialist realist cinema applies to a great many films made in the two plus decades since Xiao Wu set a new standard for Chinese neorealist independent cinema. Indeed, a very similar argument might be made about time and the long take in the more recent and even more devastatingly bleak masterpiece of Chinese slow-cinema realism An Elephant Sitting Still (Da xiang xidi er zuo 大象席地而坐; Hu Bo 胡波, 2018), which is nearly four hours long, with an ASL of more than two and a half minutes.

In 2016 Moira Weigel observed that “over the past two decades, . . . ‘art’ cinema and ‘slow’ cinema have become increasingly synonymous.”89 The “slow wave” that “has swept the festival circuit” has featured films with narratives that “are nondramatic or nonexistent,” scripts that “are minimal and repetitive, with little dialogue,” and extra-long takes in which often neither the camera nor the figures in the frame move much. The slow art cinema phenomenon has been global, but its most notable practitioners have tended to come from Eastern Europe, Iran, and East and Southeast Asia, including Taiwan and Thailand in addition to mainland China and South Korea. In China itself, within the broader post–socialist realist trend, slow cinema has overlapped with the trend of independent cinema focused on subaltern subjects, and it has embraced DV technology, not least because DV allows for unlimited shot lengths (whereas traditional cinema was at least limited by the length of a film reel, generally about eleven minutes). Independent filmmaker Liu Jiayin, with her Oxhide (Niupi 牛皮; 2005) and Oxhide II (Niupi er 牛皮贰; 2009), exemplifies how the emerging digital generation has pushed long-take slow cinema to new extremes. Oxhide II, for example, has no overarching narrative beyond the simple event of a family (Liu and her own parents, playing themselves) making and consuming a meal of Chinese dumplings together. Lasting over two hours, the film has only nine shots, each one photographed at a forty-five-degree angle from the previous one, so that the camera gradually circles the table around which the family chats, cooks, and eats. The rigorous formal structure, combined with the extraordinary ASL of nearly fifteen minutes per shot, means that while the film has many of the standard markers of neorealism—on-location shooting, nonprofessional actors playing ordinary people, and so on—it also falls under the mode of “style-based” or “parametric” narration, in which a stylistic “system” (the camera revolves 360 degrees around the table clockwise at forty-five-degree increments through extraordinarily long takes) is as important to the experience of the film as its plot.90 This example shows that, when taken to an extreme and repeated over many films, the long-take style begins to appear less as a type of realism and more as its own kind of formalism, if the latter is defined as form or technique calling attention to itself rather than to the narrative content or “reality” the film conveys.

Prescriptive Capitalist Realism and Post–Socialist Realism

Insofar as there remains a nonmainstream neorealist alternative to mainstream cinema in China, it has increasingly morphed from a realist style defined against socialist realism into simply a realism of the postsocialist condition that continues to employ techniques of Bazinian and documentary realism. Such a style extends aspects of the preceding post–socialist realism, and it will continue to be explored in the next chapter in the context of the digital revolution in filmmaking. With Liu Jiayin’s Oxhide films, for example, what lingers as a backdrop for the film’s style or story is no longer the legacy of the Mao era so much as the legacy of postsocialism itself. While the stylistic explorations go beyond the then-established conventions of neorealism (to the extent of reaching a parametric formalism in Oxhide II, as just described), the only real drama is the struggle of Liu’s father to support the family as a skilled craftsman (doing leather work—hence the films’ titles), trying to keep a small shop afloat in a competitive economy of global markets and rising rents. The films’ context is contemporary capitalism, with the socialist legacy fading far into the background.

With the massive social and economic transformations that only accelerated in China after its formal entry into the World Trade Organization in 2001, as post–socialist realism became postsocialist realism (a realist depiction of the postsocialist condition rather than a self-conscious challenge to the bygone conventions of socialist realism), its other—the types of prescriptive cinemas against which it defined itself—also transformed. Initially, that had been Mao-era socialist realism, which in the reform era was reinvented in the more or less subtly propagandistic “main melody” or “leitmotif” films of the current era—government-sponsored big-budget films that continue to promote patriotism by retelling the foundational myths of the PRC, the CCP, and modern China in general.91 Insofar as those films reinforce heroic legitimating narratives of the party and state, they constitute an updated version of the prescriptive realism of the Mao era, and low-budget neorealist independent films obviously remain an oppositional alternative cinema. However, other forms of prescriptive film storytelling also are implicitly opposed by the ongoing neorealist trend.

Some of those will be discussed in the next chapter, but here it is particularly noteworthy that the overidealized and artificial cinematic world that was once constructed by socialist realism gave way in part to a new, prescriptive capitalist realism. This new form of prescriptive realism legitimates not just the party or government but the broader mode of production of post-Mao China, namely, the articulation of neoliberal capitalism to Chinese state socialism, leading to an economic system that requires ordinary people to adapt themselves to the global capitalist economy.

In recent years, the phrase “capitalist realism” has been used to describe the sense in the post-Soviet age that there is no alternative to capitalism in the world today.92 Although that idea certainly is relevant to China today, I refer more specifically to the concept of capitalist realism as proposed by journalism and sociology scholar Michael Schudson, who, in a direct play-off of “socialist realism,” used the term to refer to the capacity of advertising in capitalist societies to simplify, typify, and abstract everyday life to “its own plane of reality” that “celebrates” and “promotes” the values of capitalism in the same way that socialist realism does those of Communism.93 Just like socialist realism, capitalist realism “does not claim to picture reality as it is but reality as it should be—life and lives worth emulating.”94 Xiaobing Tang has used this concept of capitalist realism to analyze the visions presented by consumer advertising in post–New Era China, arguing that “both socialist realism and capitalist realism are utopian art forms designed to alleviate the anxiety of everyday life and to help their respective viewers cope with the secular condition of modernity.”95

Go Lala Go (Du Lala shengzhi ji 杜拉拉升职记; Xu Jinglei 徐静蕾, 2010) exemplifies capitalist realism as a form of prescriptive cinema. The film was based on a popular novel and both stars and was directed by one of China’s most famous celebrities at the time, Xu Jinglei. It tells the story of how a young Chinese woman joins an international corporation and finds career success by gradually mastering the politics and methods of survival and advancement in such a competitive environment. She essentially is represented as a model worker under capitalism, an example for audiences to emulate in the same way that they were supposed to emulate the heroes of socialist realist films in the Mao era.96 The slickness, commercial sex appeal, rapid pacing, and high production values of Go Lala Go could not contrast more with the relentlessly down-to-earth realist aesthetic and plodding pace of Oxhide II, released during the same year, and the struggles of Liu Jiayin’s family to survive under current economic conditions—grasped only through the conversation as they make their dumplings—contrast more directly with Du Lala’s fantasy world of corporate success than with the utopian visions of socialist realism.

This chapter began by saying that the question of where Chinese cinema should go after the Cultural Revolution was answered in theory—in the “Modernization of Film Language” article, in particular—before it was worked out in practice. While various techniques were advocated for overcoming the formalization of film language under Mao-era cinema, many of which were explored in detail in the course of the chapter, the fact that they were laid out first in a theoretical essay should alert us to the realization that they, too, were potentially prescriptive, in that their repetition could lead to the same sort of formalization and even ossification that we saw in the case of Mao-era revolutionary cinema.

Echoing the points about the vicissitudes of realism by Roman Jakobson and Kristin Thompson cited earlier, John Mullarkey has observed that “the one sure thing about each new realism is that it affects us precisely when new, only in order to become passé subsequently, an obvious style refracted through time to emerge as an imitable artefact.”97 As reform-era China progressed from the shadow of the Cultural Revolution to the center of the twenty-first-century capitalist world economy, post–socialist realist aesthetics inevitably transformed from a progressive tendency seeking to counter and replace Mao-era socialist realism into an arguably “conservative tendency” (following Jakobson’s dichotomy)98 that has sustained itself as a niche style favored by the international art film and festival circuit. If the Fifth Generation’s early “exploratory” films, such as The One and the Eight, Yellow Earth, Black Cannon Incident, The Horse Thief, and The Big Parade, had reopened questions of identity and cultural roots, their follow-ups in the next decade—such as Raise the Red Lantern (Da hong denglong gaogao gua 大红灯笼高高挂; Zhang Yimou, 1991) and Farewell My Concubine (Bawang bie ji 霸王别姬; Chen Kaige, 1993)—began to feel like ponderous national allegories calculated to win their makers prizes at international film festivals as dissident Chinese artists. Similarly, gritty urban Sixth Generation films like Xiao Wu felt entirely fresh upon release, but, following the growing worldwide fame of Jia Zhangke, it seemed that many of the cinematic techniques that produced such a strong impression of realism in Chinese independent cinema of the 1990s (on-location shooting; use of nonprofessional actors; long-shot and long-take cinematography; and a focus on the daily lives of marginal characters, such as prostitutes, thieves, and migrant workers) by the very early 2000s had become conventions reproduced for consumption by the global festival audiences that had made such films successful commodities in the first place. However accomplished such independent films as The Orphan of Anyang (Anyang de gu’er 安阳的孤儿; Wang Chao 王超, 2001) and Blind Shaft (Mang jing 盲井; Li Yang 李杨, 2003) might have been, one could not help but notice their similarity in look and theme to earlier Chinese art house hits, particularly Xiao Wu, and many other films tried to achieve the same effect without the skill displayed by those. The gritty realist Chinese indie film had become a product designed for a particular elite consumer. By the end of the 2000s, Shelley Kracier noted that although Jia Zhangke himself had struck out in “new directions,” “Jia-ist cinema, through its profound effect on most younger independent Chinese directors, seems lately more restrictive than liberating in its influence. Film language in ‘mainstream’ indie Chinese films (both docs and features) seems to have temporarily congealed into something like formulaic liturgies: fetishization of the long take, the distant camera, the objective tone, the unedited minutiae of daily life.”99

Italian Neorealism is generally thought to have played itself out by the early 1950s, due to the dynamic described earlier, in which techniques that seemed fresh, provocative, and revelatory became, upon repetition, conventionalized. Nonetheless, the style—which had drawn on other precedents in world cinema from 1930s French Poetic Realism all the way back to Robert J. Flaherty’s Nanook of the North (1922)—traveled globally and is still evident in trends around the world today. The editors of the volume Global Neorealism “contend that this film style relinquished its exclusive Italian nationality soon after World War II” and “acquired many nationalities and became a citizen of the world.”100 The anthology Opening Bazin similarly shows how the French critic of the mid-twentieth century had a global legacy that stretched up to and beyond the end of the century. The Chinese post-Mao neorealist trend that I call post–socialist realism may, too, have become conventionalized in many critics’ eyes, just like its Italian precedent, but the ongoing attraction of many of its techniques, perhaps deployed in increasingly radical or inventive ways to avoid seeming formulaic, suggests that their power goes beyond mere novelty. Insofar as they invoke a reality that is as rich, complex, ambiguous, and often boring but occasionally exhilarating as everyday life, they retain their attraction and power for some Chinese filmmakers seeking to resist the prescriptive strictures of both officially sanctioned mythologies and commercial entertainment conventions.