Source: Association for Chinese Animation Studies (5/5/2024)
What Happens to the Index in Animation? The Case of The Taking of Tiger Mountain
By Cassandra Xin Guan
In the opening sequence of The Taking of Tiger Mountain (Zhiqu Weihushan 智取威虎山 2014), an overseas Chinese student, “Jimmy,” walks into a karaoke parlor in Manhattan’s Chinatown trailing a suitcase. He mingles with a noisy group of young Asians, until the incongruous sound of Peking opera and the vision of a fur-clad actor gesturing before a painted snowy landscape interrupts the karaoke program. It is reconnaissance officer Comrade Yang Zirong astride an invisible horse in the 1970 film adaptation of the revolutionary model opera Taking Tiger Mountain by Strategy. [1]The room erupts into hilarity at the embarrassment of the singer, taken aback by this prank. Jimmy alone is entranced by the apparition on the flatscreen TV. Someone asks, “It’s your hometown, isn’t it?” Next, we see the young man sitting in a yellow cab en route to the airport. While the driver curses Yuletide traffic, Jimmy begins to watch a YouTube video of the model opera on his phone. The operatic soundtrack swells while the camera zooms intently into his face. Snow is falling in America and in the deciduous forests of Northeast China. Over aerial vistas of the hyperborean landscape, the title of the film appears followed by the name of the director and source material: The Taking of Tiger Mountain, a film by Tsui Hark 徐克, adapted from the 1955 novel Tracks in the Snowy Forest (Linhai xueyuan 林海雪原) by Qu Bo 曲波.
Thus begins a reality-bending experiment in the animation of history. Directed by a filmmaker so versatile with the genre that he earned the nickname “Spielberg of Asia,” The Taking of Tiger Mountain summons its audience on a surreal journey that traverses both space—from New York City to China’s Heilongjiang province—and time—from Christmas Eve, 2015 to Lunar New Year, 1946.[2] The central story of a small PLA platoon’s strategic ambush of a bandits’ lair, adapted from a film based on a model opera based on a socialist-realist novel based, loosely, on real events in the Chinese Civil War, is embedded in the master diegesis of the overseas Chinese student going home to reconnect with his roots. The focal character Jimmy who we meet in the present-day prologue turns out to be—spoiler alert—the descendent of a war orphan rescued by the Reds on their northern expedition. As the final line of the opening credits dissolves into a high-angle shot of a PLA column marching through the snowy forest, it is Jimmy’s voiceover that we hear on the soundtrack: “1946, less than a year after the Japanese surrender, China fell into a full-scale civil war…” With the premise of a heterodiegetic narrator who imagines the story of the model opera in his own head, The Taking of Tiger Mountain makes no attempt to conceal its own fictive status as a nostalgia- (as well as testosterone-) fueled fantasy about historical events. Unlike the typical homodiegetic narrator in Hollywood period films—think, for example, of the modern-day Rose Dawson Calvert in Titanic (1997)—Jimmy is not a witness of history, but the recipient of a family lore. In this essay, I argue his alienation from and fascination with the place and time of the diegesis determines the style and meaning of animation in the 3D action film.
Figure 1: A fantasy of repatriation aimed at the diasporic subject
As one would expect of a big-budget war film from the 2010s, The Taking of Tiger Mountain is a composite patchwork of live-action footage, motion-capture rendering, CGI effects, and 3D-enabled visualization. Instead of blending into a seamless spectacle, however, Tsui Hark’s self-reflexive montage of digital and analog animation techniques sets the barnstorming drama of military espionage and heroic rescue apart from other recent blockbuster Hong Kong-Mainland co-productions. Making sense of the critical strategy that organizes the postmodern ambush of a modern revolutionary text, I will venture to interpret the style of animation embedded in the consciousness of the focal character as an acknowledgement of the problem of making such a film as The Taking of Tiger Mountain in the first place. To begin with, the story of the Western-educated Sinophone expat who stumbles upon his revolutionary alter ego in a Manhattan karaoke bar is reminiscent of Tsui Hark’s self-reported experience of being spellbound by the 1970 film adaptation of Taking Tiger Mountain by Strategy, which he came across as a 23-year-old cinephile in New York City. Tsui Hark, the director of Iron Monkey (1993) and Once Upon a Time in China (1991), took his worldwide fan base by surprise not only by collaborating with mainland investors on a pro-party, pro-country “main-melody film” but by professing that he had always wanted to make a movie based on the model opera.[3] This is an acknowledgement worth dwelling on. As scholars of Chinese cinema like Wendy Xie have pointed out, Jimmy’s homebound journey in The Taking of Tiger Mountain reenacts the real-life trajectory of many “north-bound” Hong Kong filmmakers after 1997 and by extension the “return” of Hong Kong itself to mainland China.[4] That Tsui Hark uses animation not only to embellish the blockbuster spectacle but to allegorize the structure of fantasy implicit in the narrative of repatriation is an acknowledgement of the real contradictions besetting the traversal of such geographic as well as ideological antipodes.
Figure 2: The ghostly apparition of revolutionary art in a Manhattan karaoke bar
The film theory of animation tends to presuppose a globally synchronized break between analog and digital modes of representation.[5] In The Language of New Media (2001), Leo Manovich famously argued that the weakening of the indexical bond between image and referent through techniques such as compositing, frame-by-frame retouching, digital animation and image rendering brought computer-generated imagery (CGI) closer to the condition of drawing. He consequently likened digital cinema to a particular case of animation that uses live-action footage as one of its many elements.[6] Afterward, film theorists grappling with the increasingly digital nature of cinema have made the concept of animation central to their analyses of post-filmic moving image aesthetics. As Tom Gunning noted, “valorizing animation as the ‘anti-index’ played an essential role in shifting theoretical focus from a narrow obsession with photography.”[7] My reading of The Taking of Tiger Mountain suggests that the narrow and—as scholars like Kris Paulsen and Katherine Groo have shown—inaccurate association of the indexical sign with the mechanical process of photographic reproduction constrains the possibility of theorizing animation as an evolving configuration of media epistemology and aesthetic practices in a global context.[8] Tsui Hark’s self-reflexive uses of animation, specifically the indexical properties obscured by conventional understanding, pose a conceptual challenge to globalizing and periodizing narratives about new media ontology. The disidentification of animation with indexicality through its association with CGI and opposition to photography is flatly contradicted by the use of cartoon drawings to signify an indexical bond between the past and present.
Figure 3: The first appearance of the faded sketchbook on the train to Heilongjiang
The key evidence for my claim is the faded sketchbook that Jimmy contemplates on the train from Beijing to Harbin, a personal legacy that will lead him to the peak of Tiger Mountain. It contains pencil drawings made by Yang Zirong with dates and captions that anchor the imaginations of the modern-day subject in an antecedent reality. The sketchbook appears four times over the course of the film narrative. We first see it in the hands of Jimmy on the way to his ancestral home. A nifty piece of parallel editing has established the simultaneous movement of Yang Zirong and Jimmy toward the common destination of Leather Creek, conveyed respectively by a belching coal-powered steam engine and a noiseless high-speed bullet train. While taking a long-distance phone call from America, Jimmy directs our gaze onto an open page in the sketchbook with a drawing of a sleeping girl and the following inscription: “little white dove riding on the train near Harbin station. 1946 January 22, 4 o’clock in the afternoon, the engine car.” A single stationary cross-fade dissolves the present moment of seeing this image into the initial moment of its creation. From a close-up of an unidentified hand working away on the drawing, the camera zooms out and pans up to reveal the model, reposing in the exact posture depicted in the sketch. This simple device thus establishes continuity between past and present, fiction and reality, hero and spectator, even if the existence of a sketchbook in the historical diegesis strains credibility—what was the people’s reconnoiterer doing doodling away in the kind of brown-paper sketchbook that college art students like to keep?
Figure 4: Drawing establishes continuity between past and present
I would like to suggest the sketchbook was deliberately planted in the film as an instrument of navigation for the would-be time-traveler, anticipating her struggle to approach and apprehend an antecedent reality impossible to experience empirically. It appears at three other junctures in The Taking of Tiger Mountain. On the second occasion, we see Yang Zirong making a drawing of the war orphan, who will one day bequeath the sketchbook to Jimmy, his grandson, while overhearing his story on the soundtrack. On the third occasion, the sketchbook appears in the possession of the boy, as he watches Yang Zirong riding away from base camp. On the fourth and final occasion, we see the sketchbook back in the gloved hands of Jimmy as he surveys the surrounding vistas from the peak of Tiger Mountain in the year 2015. Merely seconds ago, but in another world, it was Yang Zirong who stood before this scene as dawn broke on New Year’s Day, 1946. Significantly, the cross dissolve that brings the spectator back to the present day takes place over an image of the same mountain range depicted in two different mediums: photography in the historical diegesis and drawing in the contemporary era.
Figure 5: Jimmy looking at Yang Zirong’s drawing of Tiger Mountain in situ
One possible interpretation of the importance Tsui Hark clearly attaches to the medium of drawing is in alignment with the theoretical narrative of cinema becoming an instance of animation.[9] From this point of view, the transformation of Comrade Yang Zirong into a comic artist, however anachronistic and weird, would seem to encapsulate Manovich’s assertion that “the history of the moving image makes a full circle,” meaning computer graphics has overtaken live-action filmmaking with the result that pre-filmic reality is becoming increasingly irrelevant to the ontology of the digital image.[10] In this light, drawing appears as a medium of fantasy, even psychosis, in the sense of a subject disassociating from reality and history. Yet the film itself contests this reading by insistently aligning the present with the past and animation with photography. If digital cinema has allegedly become a form of drawing due to the ubiquity of CGI, then how do we make sense of the return of animation as an indexical sign?
In a recent article that contains a lengthy discussion of The Taking of Tiger Mountain, Jason McGrath argues that the evolution of Chinese animation aesthetics has moved away from the organization of absence in traditional operatic art forms to a supersaturation of presence in contemporary digital entertainment. The scene in which Yang Zirong shoots a marauding tiger in the two film adaptations of Taking Tiger Mountain by Strategy is invoked by McGrath to illustrate the chasm between the suppositionality of the opera film, “in which much of a fictional diegesis is supposed or posited rather than being mimetically depicted,” and the phenomenological immediacy of the contemporary digital spectacle.[11] In the opera film, the actor who plays Yang Zirong stands on an empty stage with only a painted backdrop and a stylized whip for a prop. Through a series of conventionalized gestures, he manages to convey a full range of actions and emotions to an engaged and discerning audience: horseback riding, perceiving danger, and battling and shooting the tiger. In the 2014 remake, by contrast, even someone who has never seen a movie before cannot mistake the meaning of the scene. The magnificent CGI tiger—replete with digitally rendered whiskers that could be counted if one were to pause the action on the screen—leaves nothing to the imagination. In this comparison, the distinction made by Marshall McLuhan between hot and cold media comes to mind: the simulation of photorealistic action through computer imaging is arguably hot or “high definition” because it leaves the spectator with precious little to do but gasp, while traditional Peking opera is ostensibly cold or “low definition” because it provides less sensory data while simultaneously demanding greater participation by the audience.[12] In effect, setting the two tiger-fighting scenes side by side reinforces the theoretical tendency to define animation against the index, understood as a negative imprint of reality, standing in relation to the absent object as a blurred snapshot is to a person. By contrast, the computer-generated image “does not seem to claim any photographic/indexical link to the real world.”[13]
Figure 6: Two stylistic polarities in animation: a CGI tiger vs pencil drawings
But what happens if we shift our attention away from the CGI tiger to the pencil sketches? McGrath emphasizes the increasing integration of live-action and digital simulation in contemporary Chinese animation aesthetics. His reading of The Taking of Tiger Mountain as embodying a tendency diametrically opposed to the aesthetics of absence in the earlier opera film, however, ignores its most important suppositional element. The annotated drawings in Yang Zirong’s sketchbook have indexical status by virtue of being connected to the depicted people and events, but they lack the phenomenological immediacy of photographs. The content of the sketchbook resembles more than anything else inspirational sketches or story-board drawings used to develop characters and scenes prior to the production of narrative films. The minimum rendering of historical persons and places by way of simple outlines invites the contemporary spectator to complete the work of historical representation, filling in the blanks, as it were, with his or her own imaginative faculty. At the same time, the hand-written legends that accompany the drawings confirm the actual existence of the depicted objects—the film shows that the cartoon likenesses are likenesses of something that existed once upon a time. In this way, they transform the act of drawing into an indexical proposition.
Contrary to the reductive presentation of Peircean semiology in post-70s film theory—where the picture is often conflated with the icon and the photograph with the index—Peirce himself never identified the indexical sign with a specific medium or process of image reproduction. Rather, the index operates within a broader category of propositional signs (called Dicisign) that indicates the existence of a thing to an interpretant. A Dicisign is formed when the index forces us to look at something as an icon, or resemblance, as in the act of drawing a face on paper. Pierce stipulates that although the image of a human being may be purely iconic, giving no indication of an actual existence, “a man’s portrait with a man’s name written under it is strictly a proposition.”[14] It follows that the captioned sketches in Yang Zirong’s sketchbook certify the existential bond between image and referent while communicating to the film spectator the authenticity of this propositional mode of apprehending a past that cannot be experienced directly.
Figure 7: Drawing as indexical proposition
Elsewhere, Peirce writes that “a pure index simply forces attention to the object with which it reacts and puts the interpreter into mediate reaction with that object.”[15] Information about the object is conveyed only after the index interacts with the icon to generate a proposition about what has been. In contrast to the way realism has been construed by some followers of André Bazin as an effect of the ontological identity of the photograph with its prototype, the index in animation mediates absence through an aesthetics of supposition, forcing the spectator to imagine the absent referent as being virtually there. As a theoretical object, Yang Zirong’s sketchbook suggests that the entire weight of history as a fantasy that implicates the spectator rests on the open-ended structure of the indexical proposition, which needs to communicate little information about its referent apart from the fact of its existence. For this reason, it appears with full force in the graphic field of animation. Like the finger pointing toward some unseen object, the graphic marks that constitute the basis of animation mobilize the indexical connection without creating a full representation, whereas photographs, as Gunning has pointed out, are characterized by their “inexhaustible visual richness” and “resistance to signification.”[16] Accordingly, the subject viewing an album of faded family snapshots à la Barthes or Kracauer is overwhelmed by the complexity of the images, including details that escape normal observation. By contrast, we see that what Jimmy inherited from his revolutionary ancestor in the form of a proto-comic book is nothing but an invitation to imagine what had or might have been.
Figure 8: Reconstructing a nationalist aesthetics within the time-space of globalization
Another way of rethinking animation’s indexical force is through the concept of acknowledgement that Dan Morgan takes from Michael Fried to theorize how realist films can respond to their medium-specific condition through the construction of a style. If the task of criticism, as Morgan claims, is to “discover, from looking at a film what it is that its style is acknowledging,” then we could open up the definition of realism to works that attempt to do more than acknowledge the material condition of the medium.[17] What Tsui Hark achieves in The Taking of Tiger Mountain is to invent a style of animation that acknowledges the social reality of reconstructing a nationalist aesthetics within the time-space of globalization. Its essence is oneiric but not in a Freudian sense. Recursion rather than repression characterizes the relationship between past and present generated by the interplay between photography and drawing: Did I dream of history or did history dream of me? With the insertion of the comic sketchbook into the film diegesis, the phenomenological immediacy of the CGI spectacle is mediated by an indexical proposition that encapsulates what Philip Rosen calls “a subjective assigning of significance to the concrete real.”[18] At the same time, the subjective determination of meaning is complicated by a breakdown of communication between conflicting social ontologies. Perhaps the realist mode of animation can help us think through the social as well as physical ontology of the moving image. In any event, it is clear that Tsui Hark did not simply launch his assault of Tiger Mountain with the heavy artillery of digital special effects but contrived to take it by critical strategy. The movement of the artist’s pencil across the pages of history outlines a political field of action for cinema in the age of unlimited special effects.
One way to develop these issues in the context of contemporary Chinese cinema and popular culture is to think further about the realist dimension of the nationalist blockbuster, especially its effort to repatriate the deterritorialized subjects created by four decades of economic and cultural liberalization. In the film’s surreal epilogue, Jimmy arrives at the home of his grandmother in Heilongjiang, who has prepared a holiday meal for him to partake with a host of ancestral spirits. As characters from the diegetic world of the model opera troop into the grandmother’s dining room, the young bilingual would-be tech entrepreneur imagines an alternative ending to the storming of Tiger Mountain, one that involves a hidden airplane hangar and mortal combat over the precipice. In this temporal vortex, where past and present, Christmas and Lunar New Year, Communism and Capitalism coincide in an explicit fantasy of ideological reconciliation, animation is both a description of what happens to the subject, who “plays” the revolutionary model opera like a video game inside his consciousness, as well as the structure of fantasy that covers over the ruptures of history. As a political experiment in animation, The Taking of Tiger Mountain exploits the function of the indexical proposition to summon the ghosts of history, creating in the process a subject who can operate within the time-space of global capitalism and retreat into a nationalist imaginary.
[1] For discussion of the absent horse in the model opera work Taking Tiger Mountain by Strategy (1970), see Daisy Yan Du, Animated Encounters: Transnational Movements of Chinese Animation 1940s-1970s (Honolulu: University of Hawaii Press, 2019), 156.
[2] Patrick Frater, “Asia’s ‘Spielberg’ Returns,” Variety, 28 August 2005, accessed 23 March 2024, https://variety.com/2005/film/asia/asia-s-spielberg-returns-1117928220/
[3] Yiu-Wai Chu, Main Melody Films: Hong Kong Directors in Mainland China (Edinburgh: Edinburgh University Press, 2022), 29-30.
[4] Wendy Xie, “What’s Chairman Mao Got to Do with It? Nostalgia, Intertextuality and Reconstructing Revolutionary Myth in Tsui Hark’s The Taking of Tiger Mountain.” Cinéma & Cie 30 (Reinventing Mao: Maoisms and National Cinemas), 71-82.
[5] On the danger of thus erasing the sociohistorical and geopolitical dimensions of media, see Thomas Lamarre, The Anime Ecology: A Genealogy of Television, Animation, and Game Media (Minneapolis: University of Minnesota Press, 2018), 3.
[6] Leo Manovich, The Language of New Media (Cambridge, MA: The MIT Press, 2001).
[7] Tom Gunning, “Animating the Instant: The Secret Symmetry between Animation and Photography,” in Animating Film Theory, edited by Karen Redrobe (Durham, NC: Duke University Press, 2014), 37.
[8] Kris Paulsen, “The Index and the Interface,” Representations, Vol. 122, No. 1 (Spring 2013), pp. 83-109. Katherine Groo presented a paper called “Film History’s Pragmaticism” at the University of Chicago on 11 April 2024 as part of the Marva West Tan Lecture series.
[9] The anti-indexical status of animation is evident in Stephen Chow’s Kung Fu Hustle (2004), a commercially successful Hong Kong-Mainland co-production that blended traditional martial arts choreography with exaggerated CGI effects. The result is a kind of live-action cartooning that Roger Ebert described as “Jackie Chan and Buster Keaton meet…Bugs Bunny.” The actual comic book in that film is a vehicle of fantasy that doesn’t acknowledge an antecedent reality.
[10] Manovich, The Language of New Media, 302.
[11] Jason McGrath, “Suppositionality, Virtuality, and Chinese Cinema,” boundary 2 (2022) 49 (1): 263–292.
[12] Marshall McLuhan, “Understanding Media: The Extensions of Man (Cambridge, MA: The MIT Press, 1994), 23.
[13] McGrath, “Suppositionality, Virtuality, and Chinese Cinema,”286.
[14] Chares Sander Peirce, The Essential Peirce: Selected Philosophical Writings, Vol. 2 (Bloomington, IN: Indiana University Press, 1998), 282. I’m grateful to my friend Tyler Theus for pointing out to me the cited passages where Peirce discusses the index in relation to drawings and portraiture. It also owns to his commentary that I made the connection between graphic inscription and “pure index.”
[15] Ibid., 306-07.
[16] Tom Gunning, “What’s the Point of an Index? Or, Faking Photographs,” in Still Moving: Between Cinema and Photography, edited by Karen Redrobe and Jean Ma (Durham, NC: Duke University Press, 2008), 47.
[17] Dan Morgan, “Rethinking Bazin: Ontology and Realist Aesthetics,” Critical Inquiry, vol. 32, no. 3 (Spring 2006), pp. 472.
[18] Philip Rosen, “Subject, Ontology, and Historicity in Bazin” in Change Mummified: Cinema, Historicity, Theory (Minneapolis: University of Minnesota Press, 2001), 14.
Bio:
Cassandra Xin Guan 关昕 is Assistant Professor of Cinema and Media Studies at the University of Chicago. She was a Mellon Postdoctoral Fellow at MIT’s Center for Art, Science, and Technology from 2022-2024 and received a PhD in Modern Culture and Media from Brown University. Her work has been published in academic journals such as October, Screen, and Critical Inquiry. She is currently finishing a book manuscript, called Maladaptive Media: The Exigency of Life in the Era of Its Technical Reproducibility. The book explores how a radical form of animation aesthetics emerged during the interwar period across film, radio, and print that made visible a crisis in the relations between the living being and its environment.