WP 7: Distance

Approaching Reality: Epistemic Distance, Political Crises and Temporal Imaginations in the Sino-French Dialogue on Cinema Ontology

Victor Fan

In recent years, an increasing number of film scholars have become interested in setting up a dialogue between Bazinian ontology and a similar line of discussions among the Shanghai filmmakers and critics of the 1920s. Such a desire to open up a comparative space in cinema ontology is motivated by a growing interest among film scholars in those theoretical traditions “outside” the Euro-American “canon,” and by a hope to find inspirations “elsewhere” to reconfigure the established notion of cinema ontology in response to the digital image. Yet, how we conduct a dialogue and to what end we do so are precarious problems. It is because the idea of searching for a conceptual alternative in a discourse outside Europe and America is often based on the assumption that “China” and the “West” are fundamentally different, and in the end, instead of putting two bodies of knowledge into conversation, scholars sometimes inadvertently reproduce an imagined epistemic distance by too eagerly asserting their differences. In addition, by holding Bazinian ontology as the golden standard, it is often tempting to use the Chinese film theoretical writings to “improve” Bazin or vice versa, instead of locating possible blind spots that are shared by these two seemingly distantiated discussions—aporias that might in fact help us work through the “digital” problem by addressing certain theoretical limits within the historical imagination in the first half of the 20th century.

Attempts to make Bazin talk to the Shanghai film critics of the 1920s came to the fore in the mid-1980s. Between 1985 and 1986, two film scholars from Beijing, Chen Xihe (b. 1949) and Zhong Dafeng (b. 1955), set out to find an alternative concept of ontology specific to the history of Chinese cinema. “Does China have its own system of film theory?” asked Chen. “In comparison with the state of theoretical discourses in the West, is the state of theoretical discourses in China complete? If incomplete, in what sense is it incomplete? Can we follow another specific logic to understand the state of theoretical discourses in China?”1 In an attempt to answer these questions, Chen and Zhong argued that the Chinese thinking on cinema ontology was not based on the image as reality, but on the relationship between drama, representation and life.2 At the time, this series of questions sounded exciting: It would be too good to be true if we could identify a notion of cinema ontology alternative to the one built upon Bazin’s writings.3 Yet, the desire to find a theoretical alternative in the name of opening a comparative space is based on an ethnological assumption that “China” is posited outside the rest of the “World,” as though these two cultural discourses, no matter how closely they converse with each other, are fundamentally discreet. Hence, despite the fact that Chen and Zhong managed to construct a whole system of ontological thinking out of the theoretical writings from the 1920s, what their discussion does is simply maintain the imaginary distance between China and the World.4

By focusing so much on building an epistemic distance between China and the imagined West, Chen, Zhong and a whole generation of film theorists who worked along this line missed a potentially more interesting difference between Bazinian ontology and the early film thinking among the Shanghai critics: the latter’s insistence on a delicate distance between the cinematographic image and reality. In many film theoretical writings in the 1920s, critics have the tendency to use a peculiar term to describe cinematographic reality: bizhen—often translated as “lifelike,” but more properly understood as “approaching reality.” This ambiguously defined term is useful for us not only because it hints at the fact that the Shanghai film critics of the early 20th century were, after all, equally interested in the question of reality as Bazin and his French predecessors (that is, the whole generation of photogénie writers before him). But it is also interesting because the distance between the image and reality that such a line of thought adamantly maintains stands to challenge the confidence of Bazin—and to different degrees, the philosophers to whom he was indebted, including Henri Bergson (1859-1941), Jean-Paul Sartre (1905-1980) and Maurice Merleau-Ponty (1908-1961)—in believing that the perceiving subject (in our case, the spectator) apprehends the image-consciousness in its totality and immediacy.5 What I shall demonstrate is that the term “approaching reality” suggests a certain gap between a spectator’s affective state and the image-consciousness it apprehends, a gap that may contain within itself multiple potentialities in the way the spectator is put into a relationship with time.

How we should approach this distance and the epistemic difference it implies is a serious problem. On the one hand, it is tempting to read our new discovery with the Bazinian paradigm and try to prove that above all, these historical theories could be reconfigured in accordance with the critical language to which we are accustomed. On the other hand, it is equally tempting to simply use the Shanghai film theory as a tool to improve Bazin. What I want to do here is to take up a methodology proposed by Thomas Elsaesser: seeing each historical attempt to propose a theory of cinema ontology as the symptom of a crisis—be that cinematic, sociopolitical or philosophical. In so doing, I can map that crisis onto our current crisis of thinking cinematographic reality in relation to the digital image. What I suggest here—as Elsaesser did—is not that the digital image has actively transformed the ontology of the moving image, but that it inspires us to think “retrospectively” and “retroactively” that the way we used to define cinema was in itself limited by certain historical imaginations. Perhaps more importantly, by acknowledging that it is possible to doubt the line of film theoretical thinking that led towards the intellectual environment that nurtures Bazin’s film thinking—i.e. French phenomenology and the photogénie movement—and to locate the blind spot that both schools of ontology share, a new comparative space may emerge that can offer us further potential to rethink what cinema can be, that is, the cinema that “has not yet been invented.”6 In the intervention that follows, I argue that the very term “approaching reality” can offer us an insight into this gap between cinema spectators’ affective states and the image-consciousness they apprehend. It also reveals the historical limitations in our understanding of time beyond those terms as distance and boundary, beginning and end, progression and regression—temporal potentialities that are currently opened up by the digital image.

The Myth of Chinese Ontology: A Symptom of Post-Revolutionary Crisis

The effort of identifying a line of thinking that can be known as “Chinese ontology” emerged out of a sociopolitical crisis in the mid-1980s. Chen and Zhong belonged to the first generation of graduate students being trained after the Cultural Revolution (1966-76) in the Beijing Film Academy’s (BFA) Department of Literature (equivalent to critical studies). Studying film theory and criticism in the late 1970s and early 1980s was an adventure. On the one hand, volumes of film magazines and entertainment newspapers from the 1920s and 1930s that were banned—if not partially destroyed—during the Cultural Revolution were once again made available to researchers. On the other hand, as Cecile Lagesse points out, discussions on Bazinian theory by film theorists and makers resurfaced in academic journals between 1979 and 1980, with the publication of “Tantan mengtaiqi de fazhan” (“Discussion on the Evolution of Montage,” 1979), by the BFA professor Bai Jingsheng, and “Yige zhide zhongshi de dianying meixue xuepai—guanyu ‘chang jingtou lilun,’” (“A School of Thought on Film Aesthetics Worth Paying Attention To: On the ‘Long Take Theory’”) by the screenwriter Li Tuo.7 Yet, as Li and his partner Zhang Nuanxin (director, 1940-1995) suggest, film scholars and makers at the time were not only excited by the possibility of learning about what had happened in the capitalist sphere outside, but were also eager to be inspired by Euro-American and Japanese film theories to come up with technical innovations that could be used to represent and negotiate the sexual desires, social ambitions, economic anxieties and identity crises experienced by the post-revolutionary generation.8

Nonetheless, the blossoming interest in Euro-American theory and Bazinian ontology also met with some constraints. One set of constraints was methodological. In the 1980s, the fascination with Bazin was coupled with a revival of interest in pre-1966 socialist theories, namely, Soviet montage and the Frankfurt School of film criticism. As Lagesse argues, this complicates the discursive contours of discussions on film theory in two ways. First, Li and Zhang were influenced by Brian Henderson’s article “Two Types of Film Theory” (1975), which had the effect of persuading Chinese readers to see Soviet montage and Bazinian ontology as two dichotomized schools of thought based on film forms.9 As a result, Bazinian ontology was reduced to a formal fascination with the long take and a preoccupation with the French bourgeoisie of the 1940s. Because of this impression, Bazin’s line of thought met a certain resistance from the scholastic circle, as the slowly changing official cultural policy—with its own internal inconsistencies and political contestations—was still hanging onto the protocol of safeguarding the “shehui zhuyi jingshen wenhua” [socialist spiritual civilization].10 This did not mean that Bazin was banned in the academic discourse; it only meant that his works were to be studied and critiqued within an epistemic space informed by the “socialist spiritual civilization.”

As Lagesse points out, one such solution was found in the translation of Siegfried Kracauer’s Theory of Film: The Redemption of Physical Reality (1960) by Shao Mujun.11 In the translation, Shao uses the word “xieshi zhuyi,” or “the principle of describing or representing reality,” instead of “xianshi zhuyi” [the principle of reality] to translate Kracauer’s term “realism.” Lagesse, following the view of Hao Jian, argues that “this term might help erect a bridge between Bazin’s realism and an earlier Chinese social realism—translated as shehui xieshizhuyi [the principle of describing reality in society]—that characterized most of Shanghai film production from the 1920s to the 1940s.”12 The “principle of describing reality in society” was not in use between the 1920s and 1940s; it was retroactively labeled upon Shanghai cinema during this period by the film historians of the 1950s, and was publicized—if not coined—by Mao Zedong in his Yan’an Talk in 1942.13 It was then widely discussed and disseminated during the Seventeen Years Period (1949-66) by screenwriter and theorist Xia Yan (1900-1995).14 Shao’s gesture of connecting Kracauer’s notion of realism (and as a result of that, Bazin’s) with the Yan’an one was idiomatic of the cultural policy during the reign of the Chairman of the Chinese Communist Party (CCP) at the time, Hua Guofeng (1921-2008; in office 1976-81), who, despite significant public doubts and contestations, suggested the idea that peace, social order and prosperity would be restored as long as the CCP would turn the clock back to the Seventeen Years Period.15 But as I will point out later, the term “describing reality” also echoed the distance that film theorists of the 1920s and 1930s observed in their writings, an echo that would later become inspiring for the young film scholars that looked back to the 1920s with hopes of building a Chinese ontology.

Another methodological constraint faced by Chen and Zhong was the fact that film theory was discussed in China in the 1980s according to a scientific model. Euro-American discussions of ontology involve a critique of the critical language itself, and an examination of how linguistic definitions and ratiocination work hand-in-hand with the human construction of knowledge and understanding. In this sense, ontology cannot be derived from a series of observations without an active critique of the language itself, for human observation takes place within an epistemic space. However, because they use methodologies from the natural and social sciences, the discussions undertaken by Chen and Zhong are based primarily on principles that they observe from actual film productions and practices. Hence, their “theory” determines “what cinema is”—understood by Bazin, in his essay “The Myth of Total Cinema,” (1946) as a potentiality or a myth that has been gradually actualized and redefined historically—as a sum observation of the characteristics instantiated in individual productions. This is in sharp contrast to Bazin’s idea that “cinema has not yet been invented,” that is, that cinema as a potentiality is still in the historical process of being actualized as a human praxis.16

On top of these methodological problems, there was a sociopolitical problem—the re-emergence of the question of minzu xing [national character]. The search for a “national character” in Chinese cinema was not a new topic. In the 1920s, for example, producer-director-actor Gu Kenfu (c.1890s-1932) and director Hou Yao (1903-42) were deeply concerned with the effort of “rectifying” the image of Chinese characters and cultures in Hollywood and European films.17 In 1930, Lu Xun (1881-1936) was in fact the first intellectual who pointed out that Hollywood representations of Chinese characters and cultures cannot be easily “corrected” for three reasons: (1) the representations are informed by the capitalist and imperialist ideology; hence, one cannot change the cinematic representation without changing the deep-structural social relations under capitalism and imperialism; (2) the desire to change Hollywood is symptomatic of these critics’ subscription to the idea that the only way to represent the social reality of China is to convince the colonial powers to change their own beliefs; (3) Hollywood films attract the Shanghai audience not simply because they produce the affects of enjoying capitalist pleasures, but also because they produce a longing for those pleasures that are meant to be enjoyed by the colonizer only—a form of masochistic pleasure in misrecognizing one’s political impotence as an enjoyment.18 Lu Xun’s follower Zheng Zhengqiu (1889-1935), a studio executive and producer from the Mingxing Film Company, argued further that such masochistic pleasures are not necessarily imposed by means of Euro-American imperialism, but by the feudal infrastructure of the Chinese economy that makes capitalism and imperialism possible.19 In 1949 Ye Yiqun (1911-66) proposed that the only way to allow the “national character” to present itself on screen would be to permit cinema to represent truthfully the social reality of China.20 Likewise, in the 1950s Xia Yan proposed a thorough study of folk narrational forms and theatrical performances not only because they were indigenous to local cultures, but also because they stemmed from the everyday life of the workers and farmers.21

All of these quests for a “national character” can be understood as part of film critics’ and makers’ search for a form of reality that was critical of what they deemed the bourgeois ideology. Yet, in the 1980s the question of “national character” carried a more transcendental tenor. As Wang Fei argues, after the Cultural Revolution and Hua Guofeng’s policy of critiquing the Gang of Four but recognizing the philosophical ingenuity of Mao Zedong, it was—and still is—impossible to critique the CCP policy during the 1960s and 1970s. For Wang, what the CCP did in the late 1970s and 1980s was to avoid positing itself as the ultimate moral authority; but by continuing to draw a certain transcendental power from Mao, the public maintained its desire for a fetish that stands for the missing authority. Wang argues that with the power of Mao increasingly questioned, the transcendental and nebulous notion of the “nation”—as an imagined community built upon the collective trauma of the Cultural Revolution—replaced Mao and the CCP as the fetish.22 Seen this way, Chen and Zhong’s desire to come up with a Chinese ontology was in fact symptomatic of an eagerness to replace a class-based notion of the imagined community from the 1930s with a transcendental one in the early 1960s.

The Shadow Play Theory: A Critique of the Reconstruction by Chen Xihe and Zhong Dafeng

The ontological system that Chen and Zhong came up with between 1985 and 1986 is called the yingxi [shadow play] theory. In his article “Lun ‘Yingxi’” (“On the Shadow Play,” 1985), Zhong Dafeng argues that between the introduction of Lumière’s cinématographe to Shanghai in 1897 and the rise of the left-wing film movement in 1933, cinema was commonly known as the shadow play, a reference to the popular entertainment of the same name, which emerged around the Song (960-1279) and Yuan (1271-1368) dynasties.23 As a result, intellectuals and filmmakers who wrote about cinema in Shanghai in the 1920s built their theories upon the various meanings and implications of the term. For example, Zhong cites the memoire of the novelist and screenwriter Bao Tianxiao (1875-1973)—one of the major figures in the Yuanyang hudie pai (Mandarin Ducks and Butterfly School) of modern melodrama: “When the electric shadows [the common nomination for film and cinema] first arrived in China, it was also called the shadow play. We all used to say, ‘Let’s go see a shadow play.’ From this, we can deduce that the origin of cinema is the xi-ju [play-drama].”24 Zhong also quotes Hou Yao: “The shadow play is a kind of play-drama. A shadow play contains everything that a play-drama contains.”25 Another Mandarin Ducks and Butterfly School novelist and screenwriter Xu Zhuodai (1880-1958) likewise argues, “The shadow play is … after all, a play-drama. Although there are many forms of play, the play-drama is always a unified artistic [concept].”26 From these claims made by film critics, writes and directors in the 1920s, Zhong thus suggests that cinema between 1897 and 1933 was imagined not as a newly invented machine that radically transformed its spectator’s perception of reality; rather, cinema was considered as a variation of the art of shadow play, a new form of play, drama or theater that employed a modern technology to tell stories.27

The way Zhong establishes his argument is, to say the least, highly ambiguous. First of all, what does he mean by xi-ju? Notice that in my translation above, I tried to retain the ambiguity of the term by hyphenating xi [play] and ju [drama]. The former opens up a discursive space of playfulness, craft, game, imitation, mimesis, identification, affection and sensation, while the latter leads us to discussions of representation, action, form and structure, signification and semantics.28 Also, as a term upon which cinema ontology is supposed to be founded, should we focus on the level of the text, performance, theatrical architecture, spectatorship or social institution? Or on a more basic level, should we understand xi as the play itselfor the theater? At first glance Zhong does not seem to be clear whether the shadow play theory is about the ontology of the moving image or the institution of cinema itself. To complicate matters, Hou Yao’s categorization of cinema in relation to play and drama—an idea that would be further developed by Chen Xihe—is a highly debatable one. “A shadow play contains everything that a play-drama contains,” but that does not mean the reverse: “a play-drama contains everything a shadow play contains.”29 Hence, in order for Zhong to circumvent this problem, he has to find an answer to the reason why these film critics in the 1920s did not seem to be interested in cinema’s specificity—that is, the specificity of the photographic image itself. Such an idea should be quite familiar to scholars of Bazin, for Bazin pondered a similar question in his “Théatre et cinéma” (1951) essay. In the course of his discussion, Bazin agrees that both cinema and theater represent reality, but that cinema’s specificity lies in its being a photographic image. Hence, in the theater, the immediate presence of the actor often draws a boundary between the spectators and the dramatized or represented reality. With this, the spectators are drawn nearer or pushed further from such reality by an alternation between absorption and distantiation in their viewing process. The photographic image in the cinema, however, is immediately apprehended as reality without any human mediation.30

Zhong is in fact aware of the logical pitfall in Bazin’s argument. To address this problem, he proposes to study the shadow play theory not as a trans-historical system of thoughts, but as a way of thinking about cinema specific to the semi-colonial conditions of Shanghai in the 1920s. That is, for Zhong the shadow play theory does not explain what cinema is or has always been. It nonetheless tells us how film critics and makers in the 1920s thought about what cinema was, which in turn affected the way film forms were developed in Shanghai. In this sense, by imposing such a conceptual framework upon their theorization of cinema, Chinese cinema in the 1920s became the shadow play when filmmakers “borrowed the stage play’s methods of narration and visual representation” and used them in cinema—thus really turning cinema into a form of play. For Zhong, the Shanghai film critics collectively overlooked the photographic image for a reason. Zhong argues that these writers did acknowledge that the “shadow play,” unlike the stage play, uses the “image, not staged actions, as a means of representation.” Yet, the Shanghai critics concentrated less on cinema’s specificity than on those traits that are common to both cinema and the theater, that is, the textual, performative and aesthetic mediations that go into the realistic representation of life. In this light, the photographic image was not regarded as reality itself; rather, it was a jiqiao [technē], a new technology that the filmmakers can utilize to represent reality.31

The resulting ontology, according to Zhong and Chen, is called the xiju benti lun [ontology based on play-drama], as opposed to the yingxiang benti lun [ontology based on the (photographic) image] that Bazin proposed. For Zhong, Bazinian ontology is based on more than a century of the Euro-American fascination with reality, representation and simulacrum, and hence, cinema for Bazin is primarily a machine that reproduces reality, which happens to be used commercially as a narrational medium. Meanwhile, cinema in China was introduced as a commodity of semi-colonialism and feudalism, and hence, cinema for the film critics and makers in Shanghai in the 1920s was nothing but a form of entertainment.32

Zhong’s argument may sound irksome for us, but it is symptomatic of the socialist methodology that persuades him to find an ontological system that is not necessarily “universal,” but is instead historically conditioned. It also allowed him to establish a connection between his claim and the left-wing political film writing in the 1930s—commonly known as “the hard film theory.” The hard film theory can be traced back to Lu Xun’s translation and commentary on Iwasaki Akira’s article “Xiandai dianying yu youchan jieji” (“Cinema and the Bourgeoisie,” 1930), but is more commonly used to refer to a theoretical debate among the employees of the Mingxing Film Company and the pro-Communist Party Zhongguo dianying wenhua xiehui (Chinese Association of Cinematic Culture).33 In his 1930 article, Iwasaki argues that Hollywood and Japanese fiction films appear to be only “entertainment,” but that such entertainment produces a pleasurable affect among the spectators, which arouses in them a desire to consume those material and ideological pleasures that are represented onscreen and to continue supporting the capitalist system that makes such pleasures accessible.34 As I mentioned earlier, Lu Xun argues that Shanghai’s semi-colonial condition complicates Iwasaki’s reading. For him, those material and sexual pleasures represented by Hollywood and Japanese films were meant to be enjoyed by the colonizing powers only; they were in fact inaccessible to the Shanghai spectators who were posited as the subaltern in this asymmetrical political relationship. Hence, the Shanghai spectators were encouraged to enjoy not only those pleasures that were denied them by their colonizers, but also to turn their sense of shame and impotence—and the perpetual sense of longing for something unattainable—into a form of enjoyment.35 In this light, Zhong suggests that as cultural producers under imperialism, European, American and Japanese filmmakers and critics could “afford” to philosophize cinema in terms of representation, reality and simulacrum. In semi-colonial Shanghai, though, cinema was nothing but a commodity for public entertainment—one that in fact turned the displeasure of being colonized into a form of pleasure.

In fact, between the late 19th and early 20th centuries, cinema attracted those writers from the Mandarin Ducks and Butterfly School precisely because they could use it as a new medium to entertain their audience. It also attracted the writers and directors from the wenming xi [civilized play], a form of melodrama that emerged around the late 19th century and combined forms and performing techniques from local operas and European dramas. For these writers, Zhong argues, literature and cinema help the readers or spectators to cultivate their quwei, that is, the cultivation of quotidian, pleasurable and leisurely interests that appear to be detached from politics and the rest of the society, but are in fact made possible by the availability of excessive time (outside material labor) under capitalism. As a consequence, film critics and makers in the 1920s, including Hou Yao and Gu Kenfu (who came from the civilized theater), and Xu Zuodai and Bao Tianxiao (who were Mandarin Ducks and Butterfly writers) were primarily interested in perfecting the narrational devices and screenwriting techniques they inherited from their theatrical and literary works.36 Formally, Shanghai cinema in the 1920s therefore emphasizes the integrity of the stage, the two-dimensionality of a stage representation, the use of diffused lighting that emulates a “naturalistic” stage light, the allowance of seeing an actor enter and exit the frame fully, exaggerated performances and narratives that are structured not by means of a story’s cause-and-effect chain, but by the flow of entrances and exits of the characters in a series of loosely connected acts. In other words, these critics and makers worked under the semi-colonial conditions of Shanghai by maintaining cinema as a form of entertainment and cultivation of quwei, but in so doing, they also constructed a certain “national character” in their imagination, and in the production of film forms and modes of representation to express it.37

To put matters crudely, Zhong implies that Bazin’s preoccupation with the photographic image as the basis of his ontological discussion is a form of xiao youchan jieji quwei [petite-bourgeois cultivation] that is completely blind to the historical materialist conditions of the Shanghai audience in the 1920s. Yet, by the same token the shadow play theory overlooks the importance of the photographic image because cinema to these writers is nothing more than a narrational device for the purpose of entertainment. The underlying assumption of Zhong is that Bazinian ontology, like any other body of knowledge, is subject to its historical conditions and limits. Likewise, the Shanghai version of ontology based on play-drama is also bound by its own socio-historical edges.

Nonetheless, the political motto of the late-1970s and early 1980s was “chongchu Yazhou; zouxiang shijie” [Breaking into Asia; Marching Towards the World]. Zhong Dafeng and his colleague Chen Xihe had the ambition to elevate the shadow play theory from a historically conditioned view of cinema ontology to a philosophical system that could potentially compete with Bazinian ontology. Chen’s approach was to do it through a close reading of Hou Yao’s book Yingxi juben zuofa (Methods of Writing a Shadow Play, 1926). When Hou wrote this book in 1925, he was a contracted director in the Mingxing Film Company and a professor at the Southeast University in Nanjing.38 Methods is thus intended as a screenwriting manual, but Hou devotes the first chapter a discussion of the relationship between cinema, drama and life, thus inspiring Chen and later film scholars to reconstruct a system of ontological thinking.

In his short preface, Hou writes, “I advocate the principle of regarding the screenplay as the core of cinema. I believe that the only way to produce a good play is to have a good script. The shadow play is a kind of play. A script of the dianying [electric shadows, or film] is the soul of the electric shadows.”39 In response to this statement, Chen argues, “the author clearly sees the ‘play’ as the essence of the ‘shadow play.’ In other words, the electric shadows—i.e. cinema as an objective system—is built structurally and functionally upon the foundation of the ‘play.’” Hence, for Chen, Hou’s argument “departs from the “‘play’ as cinema’s ontological and epistemological basis.”40 In this sense, the “‘ying’ [shadow or image] is a method by which the ‘play’ can be brought to the point of completion.”41 To push his idea further, Chen observes that Hou discusses the different aspects of mise-en-scène in cinema in the order of the “plot, explication, dialogue, expressions” and eventually “composition and scenery.” In fact, Hou suggests that a director’s choice of scenery must either “harmonize” with the character and her or his action or establish certain spatial and atmospheric connections between scenes.42 On aesthetics and forms, Hou believes that the functions of a play is to “harmonize” and “aestheticize” life, hence, in such regards, he goes so far as to suggest what subject matters are best represented and avoided for these purposes.43 Chen thus argues that while Euro-American film theory connects aesthetics with forms, Hou associates aesthetics with representation and rensheng (life). Chen therefore argues that the Chinese view of cinema ontology is a holistic philosophical system that regards meixue [aesthetics] as a rensheng de taidu [attitude of life], and that cinema—as a form of play and drama—cultivates such an attitude of life not through the mediation of film forms, but through an attempt to represent and elevate the quality and aesthetics of life.44

Based on this observation, Chen proposes:

From the ontological perspective, the Chinese people see the “play” as the basis of cinema, whereas the Westerners [Bazin and his followers] see the “shadow” [image] as the cinematic foundation. For the Chinese people, the “shadow” is simply a method by which the drama can be brought to its completion, whereas for the Westerners, the “play” is a concrete state of the “shadow.” The Chinese people are preoccupied by ideas such as xiju xing [theatricality] and wenxue xing [literariness], whereas the Westerners are preoccupied by the difference between the long shot and montage…. [In short,] in China, cinema is being studied as a totality seen from a direct objective view; in the West, cinema is analyzed by means of abstraction and subjective observation.45

Such a clear-cut “East-versus-West” dichotomy is constructed, I argue, upon the problems with Chen’s understanding of Hou’s writing and Euro-American film theory in general. On the one hand, by setting out to find cultural differences in film theoretical thinking, Chen conveniently omits any discussions that may allow us to connect his ideas with Bazinian ontology. Hence, in the name of establishing a dialogue between Bazinian ontology and Hou’s ontology, Chen in fact starts from the assumption that the two schools of thinking are fundamentally different, thus deactivating any potential to allow negotiable elements to emerge. On the other hand, as Lagesse aptly points out, Chen’s understanding of Euro-American theory is based on Handerson’s presumed dichotomy between the Bazinian long shot and Eisensteinian montage, a view that reduces both theories to the level of film forms and aesthetics.46 It is on the basis of this presumed dichotomy that Chen and Zhong both come up with their own re-historicization of Chinese cinematic styles and film forms. The curious point is: Chen and Zhong both arrive at a proto-Bazinian notion of ontology, arguing that film forms are contingent upon the way reality is apprehended as consciousness.

Approaching Reality

In this sense, Chen and Zhong’s attempt to re-construct a Chinese cinema ontology represents a missed opportunity to establish a meaningful dialogue between Bazinian ontology and the early film thinking in Shanghai. Hou’s Methods was written as a screenwriting manual, and of course his primary objective in the book is to underline the importance of the screenplay and narrative structure in the creation of an entertaining, artful or even educational film. Yet, if we want to tease out the rather rarified discussion of cinema aesthetics, representation and life in the first chapter of Methods, and compare it with a remarkably similar one in the writings of Hou’s predecessor Gu Kenfu, we cannot simply take what these writers discuss on the textual surface as theoretical statements about ontology. Rather, we need to deduce what kind of cinema ontology is implied when they negotiate these problems.

Film criticism of this kind in the 1920s in fact responded to a political crisis. It was often written with a political agenda in mind: to “rectify” the (mis)-representation of Chinese characters, society and culture in Hollywood and European films, productions that were often referred to as the ru Hua pia [humiliation film; literally, films that humiliate the Chinese nation]. In 1921, for example, Gu Kenfu started a new film magazine called the Yingxi zazhi (Motion Picture Review). In his “Fakanci” (“Inaugurating Preface”), Gu argues that the ultimate objectives of initiating a debate in the Motion Picture Review were to “prevent hazardous films from being circulated” (i.e. representations of sexuality) and to “fight for the human dignity of the Chinese people in cinema.”47 Hence, by understanding cinema as a tool that draws the spectators near reality and allows them to reflect upon their own moral conditions, Chinese filmmakers could improve the aesthetic quality of their productions, which would eventually be used as a tool to inculcate the Euro-American audience with the reality of “Chinese life.” Similarly, in Methods Hou Yao argues that by identifying the proper sensuous excitements that could help cinematic spectators to “harmonize” their conflicting desires, cinema could aestheticize life by entertaining and elevating the human spirit. In so doing, these film images would serve as tools to rectify unfavorable representation of Chinese characters.48

As this scenario suggests, both Gu and Hou had a common objective: How could cinema represent the life of the Chinese people truthfully, first in Shanghai productions, then in Hollywood cinema? Underneath the seeming preoccupation with drama, narrative structure and screenwriting techniques, they share an underlying assumption that cinema is apprehended by the spectators as reality, and as a result, is taken as a form of evidence that contributes to an audience’s knowledge about a political community and its culture. Yet, neither Gu nor Hou were formally trained as rhetoricians. The question of what they mean by reality, and how such a reality forms a relationship with the spectators, thus leaves much room for further debate.

Nevertheless, Hou Yao’s theorization of the relationship between cinema and the theater can in fact be traced back to Gu’s in “Inaugurating Preface.” At first glance, it does seem that Gu considers the shadow play as a kind of play, drama or theater. In the opening paragraph of his article, he argues:

The moment we enter the theater and watch a tragedy being performed, tears will come out from many of us. If the play performed is a comedy, we will all laugh heartily. What is the reason? The reason is that a well-performed play can touch its spectators, ru shen ru qi jing [as though their bodies have entered the scene itself].49

By starting from an analysis of the spectators’ affects in the theater, Gu appears to hold the assumption that cinema is a form of theater. Yet, the core of the argument in this opening paragraph is not the theater itself, but the reason why such affects are produced by watching someone perform a dramatic action. Hence, the key to understanding what Gu wants to argue is the last sentence: “as though their bodies have entered the scene itself.”50 In other words, in front of the performance, the entire bodies of these spectators apprehend a state of consciousness, an idea that seems to be not entirely far from Bazinian ontology, with the exception of the use of the word “ru” [as though]. What Gu suggests here is therefore a distance between the state of consciousness one apprehends in lived reality, and the state of consciousness one arrests in front of a performance. What is it really?

Notice that in Gu’s analysis, he has carefully constructed a schema of how the readers should understand the word xi [play, drama, theater, performance]:

juchang (theater) → banyan (performance) of a beiju/xiju (tragedy/comedy; i.e. the dramatic text) → affects

In other words, the theater is an architectural or institutional apparatus in which a performance (of a dramatic text or human action) affects the spectators, as though they had apprehended a state of consciousness. Hence, the ontological link that Gu wishes to explore is not the dramatic text itself, but the relation between the performance, the affects and the state of consciousness. In this sense, unlike Chen and Zhong’s version of Chinese ontology, Gu’s ontological essay (if we call it this way) gives less emphasis to the dramatic forms and structure than the reason why such consciousness can be apprehended.51

In the following paragraph, Gu further narrows down his focus:

Play-drama is an entertainment. When one is entertained, one is easily affected. It is because at the moment of being entertained, one surrenders everything in the world and pays attention to only the object that entertains. [In such a condition], one does not have a guannian [concept; presupposition; point of view] to analyze one’s xin [heart; psychological state].52

What Gu suggests is that in the process of being entertained, the spectators have given up the state of consciousness they have grasped outside the performance, and instead take what is being performed and the affects it produces in their bodies as their own consciousness. For Gu, in the theater, the spectators are not preoccupied with conceptualizing what they see, sense, read and feel (for example, as a dramatic text or an aesthetic experience), but as an affective state that excludes reasoning and interpretation.

With this in mind, Gu then argues that human beings have been determined to improve the means by which one could achieve this state. Hence, if we map out the development of the different theatrical or dramatic forms in the world, we can see an evolutionary trend of moving towards a state of “bizhen” [approaching reality]. For Gu, in older theatrical forms, the performer knows her or himself as a performer, and likewise, the spectators know themselves as spectators. However, in more advanced theatrical forms:

The performer forgets her or himself as a “performer,” and instead believes her or himself as the “person in the dramatic action” by having the same emotions the character has. Meanwhile, the spectator also treats her or himself as the “person in the dramatic action.” Like being in the dramatic action itself, the spectator can be in an affective state a thousand times more powerful than in the older theatrical forms.53

Gu argues that in human history, theatrical innovations were made possible by means of technological improvements, literary ameliorations and scientific progress. However, the shadow play surpasses all other traditional forms by means of mechanical reproduction. Because of that, it saves the money, time, effort and cost of restaging the same play over and over again. It also preserves a performance that can put a large number and generations of spectators into the same affective state.54

Scholars in the history of Chinese art and literature can probably point out immediately that the term bizhen has been a subject of debate among artists, writers and critics for centuries. I do not intend to trace Gu’s use of this term along the lines of Chinese art criticism primarily because Gu probably took this term from a more vernacular understanding of it. However, there is one thing that we can learn from the historical debate itself. As Wen C. Fong suggests, the term bizhen, despite its multiple interpretations, never quite assumes the European meaning of “mimesis.” For Wong, in Chinese art criticism, the term bizhen does not necessarily imply that there is a proximity between the appearance of the painted image and the reality it represents; rather, it refers to an affective state that the painting is capable of producing in the sensorium of the beholder, one that either recalls the beholder’s affective state when he apprehended the image-consciousness in her or his lived reality before, or inserts this affective state into her or him as though she or he had been there.55

With this in mind, we can begin to detect a difference between Gu’s notion of cinema ontology and Bazin’s—certainly not that he believes that the shadow play is a form of drama, but that he carefully suggests a critical distance between the image-consciousness, as it is apprehended as an affective state, and the photographic image’s mimetic capability on the level of the appearance. Indeed, the rhetoric of Gu’s “Inaugurating Preface” is remarkably close to Bazin’s “Ontologie de l’image photographique” (1945), with the exception of one subtle but significant step: Bazin’s faith that the technology of mechanical reproduction has liberated painting from mimesis.56 For Gu, mechanical reproduction preserves the potentiality of a performance that would put future spectators into the same affective state—not necessarily because it reproduces reality itself. Instead, the performance itself imitates reality and a powerful imitation simply allows the spectators to get close to, or approach, reality.

The way Gu defines a successful performance is also unique. If the European notion of mimesis is commonly (mis)-understood as a form of “suspension of disbelief,” Gu’s notion of bizhen is based on a suspension of knowledge, that is, the knowledge that the performer is only a performer and the spectator is only a spectator. The character is therefore the vehicle that conveys the actor’s response to the affective state in a real (dramatic) situation—i.e. the emotions—to the dramatic action, which produces the same affective state among the audience. Yet, no matter how close—or in fact, how far—the representation is in relation to the appearance of the reality it seeks to represent, a successful performance can always allow the spectators to suspend their knowledge and succumb to their belief through their affective state.

In this light, in Methods of Writing a Shadow Play, Hou seems to argue confidently that the shadow play is a kind of play. Nonetheless, in his first chapter, he adds:

Life is best understood as a series of performances. It performs those phenomena such as happiness and sadness, union and separation, life, death and illness, strength and weakness, fullness and emptiness, lost and found, rise and fall. Seeing life as a drama, drama is like life itself.57

As this point suggests, by seeing drama as a “miniature of life,” the spectators can use a play to critique, adjust and aestheticize life itself.58 Yet, for Hou, the shadow play goes one step further because the photographic image does not imitate; it captures life and replays it as reality. As he puts it, “while other theatrical forms use the stage as a stage, the shadow play uses the universe as a stage.”59 In this respect Hou’s argument is in fact very close to the argument of his French contemporaries, the photogénie writers, who emphasize cinema’s liberation from mimesis—that is, the immediacy between reality and the spectators’ apprehension without any form of mediation.60 Hence, when Hou talks about play-drama as the core of cinema, he has in mind not an imitation of life, but life itself that can be captured and presented by the photographic medium as reality.

For both Gu and Hou, then, the problem with Hollywood’s representation of Chinese characters and culture does not necessarily lie in these films’ failure to imitate the appearance of Chinese life. What they are concerned with is whether cinematographic representation can be improved in such a way that both the Shanghai and Euro-American audiences could be put into a common affective state. In this sense, these common affective states could allow these two audiences—and in fact, film industries and communities—to work through what they perceive as a cultural difference.

Immediacy and Distance: The Temporal Dimension

My reading of Gu and Chen may help to establish a dialogue with one frequently-problematized dimension of Bazianin ontology—Bazin’s seeming reliance on the mechanical reproduction and “indexicality” of the photographic image. In fact, if one reads Bazin this way, one easily perceives the crisis brought about by the digital image as a problem primarily based on the potential or actual absence of the physical presence of a human being or physical object to which the image refers.61 In addition, the capability of the digital technology to offer a non-human perspective on spatial relationships (images that are too big or too small, too high or too low) problematizes our understanding of the anthropocentric notion of cinema—or at least, classical Hollywood cinema.62 Nonetheless, as Elsaesser argues, Bazin never once referred to photographic indexicality; rather, he treats photography and its capability of preserving and re-producing the “imprint” of a physical object or being as nothing more than a technē that opens up a new way of thinking about time—a problem most famously phrased by Bazin as “change mummified.”63

Seen this way, the notion of “change mummified” is perhaps a more fundamental register in Bazinian ontology than the question of photographical indexicality. If we read Bazin more carefully, what interests him is not simply the preservation of an object’s state of being in the photographic image, or the possibility of reactivating the potentiality of such a state of being by projecting the image on screen. What is at stake is the temporal dimension of this process of preservation and reactivation.64 What is captured in a photographic image is not a state of being but a process of becoming—and with it, a passage of time. The screening process, meanwhile, reactivates not only a state of being, but also a process of becoming in the past (that is, a process that has been dead) that is re-enacted as the present that opens up new potentialities in the future.65 The photographic image in fact captures and replays the passage from being in time to nothingness, and from time to the cessation of time. Along these very lines, in his essay “Death Every Afternoon” (1958), Bazin’s thoughts about the film The Bull Fight (Pierre Braunberger and Myriam, 1951), a documentary in which a scene captures a real death, inspire him to think further about the question of cinematographic time. He writes:

[M]usical time is immediately and by definition aesthetic time, whereas the cinema only attains and constructs its aesthetic time based on lived time, Bergsoninan “durée,” which is in essence irreversible and qualitative. The reality that cinema reproduces at will and organizes is the same worldly reality of which we are a part, the sensible continuum out of which the celluloid makes a mold both spatial and temporal. I cannot repeat a single moment of my life, but cinema can repeat any one of these moments indefinitely before my eyes. If it is true that for consciousness no moment is equal to any other, there is one on which this fundamental difference converges, and that is the moment of death. For every creature, death is the unique moment par excellence. The qualitative time of life is retroactively defined in relation to it. It marks the frontier between the duration of consciousness and the objective time of things.66

Similarly, for Deleuze, the “classical” cinematographic image (the movement image) is apprehended as movement unfolds in time—making the spectators conscious of their own process of becoming as chronometric time unfolds itself.67 Yet, such image-consciousness is also apprehended as an affect (affection image), and in this affective process, time is crystalized into a potentiality (crystal image) that folds back to the past and projects into the future.68 More recently, Giorgio Agamben introduced the notion of the kairos (the time it takes for time to end) into the discussion. Kairos is a concept of time mentioned in Paul’s Letter to the Romans, which describes the moment Saul was being called (vocated) by the voice of the Messiah (from the end of time, or from a non-state where time no longer exists) and is then thrown back to the earthly world to continue his secular vocations, as though he had never been called in the first place. Agamben argues that in ancient and medieval Judaic theology, the period of Paul’s life during which he continues to fulfill his earthly obligations as though nothing had happened is in fact understood as a form of kairos, for when Saul is called by the voice, time ends. In other words, there is no such thing as an afterward after the end of time, only a conceptual measurement—that is, the time it takes for time to end. For Agamben, this is not a form of chronometric time that unfolds linearly, but a measurement in depth, a distance between the chronometric and the abyss where time—along with consciousness itself—ceases to exist.69

Gu Kenfu does not go as far into the philosophical depth of time as this formulation. But when he discusses the benefits of the mechanical reproduction of cinema, he argues:

When a famous actor dies, a play [more specifically, the performance] is normally considered a juexiang [reflection from the end (of time)]. Yet, a film never dies. If one preserves it well, it will re-enact this performance in clarity for decades.70

Gu probably did not have a complex philosophical reason to make such a statement. But from the perspective of contemporary debates in film theory, if we consider that the performance of a dead performer would inevitably disappear into the depth of time, a memory that would only be reflected from such a depth through literary records or words of mouth, then we might also say that every performance in itself is a juexiang. After all, no performer can ever repeat or re-enact the same performance in time. What cinema can do is simply snatch a performance from the flow of time (a vocation) and preserve it for the purpose of future re-enactments.

Has the digital image put such a concept of time into a state of crisis? The current state of debate has a tendency to swing between two poles: an unequivocal yes or no. No, because commercially released feature films and many independent or amateur filmmakers still generally follow a certain expectation in perspectives, styles, narrational devices and aesthetics, which emulate the mode of perception—and as a result, the affective state—that we are used to seeing in the photographic image.71 Yes, because the digital camera no longer preserves an imprint of the physical object through optical perception; rather, its microchip analyzes and records the thermal energy emanated from light, which is then scrambled and stored as digital data. In this sense, each time the camera itself or a computer program shows the user an image, the process is no longer one of re-activating an indexical trace, but of recomposing the image by means of informational analysis. In addition, what seems to the user to be a simple process of touching up, animating or transforming an existing photographic trace is in fact interfaced by the computer as an emulation of its analogue predecessor. What the computer does is recompose each image with the available information.72 If we use a metaphor to make this technical process more understandable, the digital production process is analogous to storing the DNA of an individual, and the filmmaker repeatedly reconstructs the image of the human body or object in the editing and recomposing process. Likewise, what the spectators see is analogous to an avatar of the actor.

What I want to say here is not that the digital image has fundamentally changed our perception of time. It opens up a dimension in the way we think about time—not only in cinema, but also in our lived reality—in the first place. Does an avatar constructed out of the stored information of the thermal energy of a person assume the same subjectivity and identity as the body from which the DNA was drawn? Does the dramatic action of this avatar really render a reactivation of an action preserved from the past, or is it a different one in a different time? Does the information that is stored into a hard drive, ignorant of the human perception of space-time (unless it is interfaced via the computer), have the capability of re-enacting any form of space-time? Is the digital image simply a lifeless-timeless re-composition (not even a simulacrum)? Or does it activate a new spatio-temporal flow each time the image is re-composed in relation to our chronometric sense of time?

Let us look at another example. P. W. Singer points out a very interesting phenomenon in Wired for War (2009). He reports that in recent years, the US military has been sending robots to carry out dangerous missions in Iraq and Afghanistan. In a mission like that, the military officer sits in a control room in his U.S. office and carries out the bombing or assassination by monitoring a digitally simulated image on screen. With this, Singer argues, war becomes a 9-to-5 job. On the one hand, the whole point of such a design development is to minimize not only the injuries and death of the U.S. officers, but also the chances of their being traumatized by brutal killings. These officers could go home comfortably, play with their children and have dinner with their spouses without carrying a sense of guilt. On the other hand, military superiors often need to remind these officers that they are not playing a videogame; rather, these are serious missions that are to be carried out with a human dimension. What Singer discusses is an incongruity between the U.S. officer’s perception of a space-time that is carefully distantiated from her or his own—a suspension of the knowledge that such an image really happens and a belief that what you see is a parallel universe—and a real-time connection between the computer button that she or he presses in the office and the physical life that is being executed in this parallel universe. Singer argues that these officers are indeed traumatized by the moral implications of their action, that is, not only by their acts of killing, but also by the absence of any affects that would normally accompany the acts of killing.73

As we can see, then, the spatial-temporal dislocation between a recomposed image and the physical reality to which it is supposed to refer (via an interface that convinces us of their mutual referentiality) does have a profound impact on the way we understand how affects and image-consciousness are interrelated. Yet, what I want to emphasize is that the digital image does not necessarily signal an ontological shift from its analogue predecessor; rather, what it does is to remind us that both Bazinian ontology and the Shanghai critics’ discussion of cinema ontology are based on very particular concepts of space-time, and in turn, on underlying presumptions about the direct correlation between action and affect that require reconsideration and reconfiguration.


In this sense, the term “approaching reality” is a powerful one. It raises questions about the very sense of immediacy between image and apprehension that Bazin and his predecessors implicitly presume: namely, the sense that the photographic image is apprehended—in its immediacy and totality—as an image-consciousness. What the digital image opens up are in fact the multiple gaps and inconsistencies in the process of human apprehension and perception, and with these gaps, the suggestion that there are more possibilities and variations, more potentialities on one’s way to reality, than the photographic version of cinema ontology activates. In our approaches to reality, time does not simply begin or end, activated or deactivated; we might want to think about time afresh with concepts such as transposition, transference, reflection, inflection, inversion, reversion, shuffling, re-composition, counterpoint and re-sequencing. Indeed, if the photographic image was once considered able to mummify time, we may want to reconsider it now as time un-dead. In this sense, the un-dead image revisits us from a non-state in which time is suspended or irrelevant and we are at once obsessed by the lure of its vampiric eternity and the fear of its disappearance, a parallel universe that we have yet to find a vocabulary to analyze.


Victor Fan is Assistant Professor at the Department of East Asian Studies, McGill University. He graduated with a Ph.D. from the Film Studies Program and the Comparative Literature Department of Yale University, and an MFA in Film and Television Productions at the School of Cinema-Television, University of Southern California. His articles have been published in Camera ObscuraJournal of Chinese CinemasScreenFilm History: An International JournalCLCWeb: Comparative Literature and Culture, the anthology A Companion to Rainer Werner Fassbinder (ed. Brigitte Peucker) and the film magazine 24 Images: Cinéma, etc. He is also a contributor to the communal blog Printculture. He is now completing his book manuscript titled Cinema Sovereignty: How Does Chinese Cinema Emerge Out of a Collective Sense of Political Failure? In addition, Fan is also a filmmaker, composer and theater director. He will join the faculty of King’s College London, Department of Film Studies, in January 2013.