Chan constructs almost all of his fight sequences with a high degree of stylization.

Many of Chan’s films are created and marketed for transnational reception —Rumble in the Bronx

Chan displays the broken ankle incurred while performing a “real” stunt, seen here in a signature out-take from Rumble in the Bronx.

Camera angles emphasize the danger of the leap, while other angles emphasize that it is Chan himself performing the stunt.

Slow-motion sequence demonstrating the sharpness of the weapons and the danger to Chan, from Fearless Hyena.

To avoid being crushed, Chan moves along unexpected planes by leaping into the space above—Rumble in the Bronx.

Violent dances in martial arts films

by Aaron Anderson

In a previous article, “Action in Motion: Kinesia in Martial Arts Films” (Jump Cut 42), I argued at length, drawing on the work of actor Steven Seagal, in favor of a kinesthesia-oriented analysis that includes physical sensation and individual body memory as shaping the pleasures of watching martial arts films. Here I take this analysis further, exploring some possibilities open through and implications of reading filmed martial-arts combat as dance.

Trying to analyze filmed violence has its traps. First is combat’s supposedly “natural” origin, analytically complicated by a long history of sociological writing that demonizes all violent spectacles Until recently most writing on the aesthetics of violent spectacles has tended to describe these spectacles in simple teleological terms as a step backward within a larger “civilizing process” (Guttmann). In this view, any displays of physical violence are simply described as undesirable remnants of a primitive, beast-like past which remain somehow anachronistically present in enlightened, modern civilization. This general unfavorable view of violent spectacles then becomes stated even more pejoratively by trends within the early history of aesthetic theory which tends to make very clear divisions between “high” art and “low” entertainment. In fact, within these debates the term “aesthetics of violent spectacles” may likely be an oxymoron. Now, more recent work in several intellectual fields questions such a simplified view of violent spectacles.

Among the many who have questioned this easy dismissal of violent spectacles are scholars from anthropology, history, literary theory, psychology, communications, and film criticism whose collective essays form the book Why We Watch: the Attractions of Violent Entertainment. Here, they essentially reverse the standard debate about violence in the media, which focuses on how to keep children from consuming massive amounts of preposterous violence. Instead, these writers ask, “Why are violent scenarios so alluring that their delivery and consumption is a major part of American (and international) popular culture?” (review back cover).

The conclusions found in the book are as diverse as the fields of the scholars asking the questions. Some of the book’s studies find a strong correlation between desire to watch violent entertainment and the viewer’s high need for physical arousal or excitement. Others focus on the viewers’ desires for old-fashioned virtues of morality and justice, need to establish and cultivate their gendered social identity, their release for expressing otherwise socially proscribed emotions, and the importance of context for viewers with regards to their concepts of reality. Each of these findings, in turn, suggests there is a reception continuum based on viewers’ perceptions of differences in the violent imagery’s “type or genre” and further differentiated by individual viewer preferences based on memory, body, and culture (Goldstein 212). After investigating this research and with these conclusions in mind, I then wish to analyze some of the U.S.-released films of Jackie Chan as dance. My goal here is to consider the ways in which concepts of movement may become fluid and thus negotiated within and between culture(s).

Jackie Chan’s films are particularly well-suited to this kind of investigation for a number of reasons: First, the fight sequences in almost all of Chan’s films are constructed with a high degree of stylization and thus help belie the “natural” origin of filmed fight sequences (an argument I began in “Action in Motion”).

Furthermore, many of Chan’s films are created and marketed for transnational reception, so that their cinematic construction allows for an investigation into the cultural aspects of movement-understanding; in particular, I wish to discuss the difference between Western notions of “feminizing” movement and Chinese concepts of Yin and Yang. Significantly focused on gender, modern discussions about the “meaning” of the human body in motion concern themselves with gender’s performative aspect. In this light, because Chan had a background in Peking (or Beijing) Opera performance and his training had a great subsequent influence on him, his films allow a degree of insight—however culturally removed or constrained—into a specific non-Western performance of macho women and malleable men, both of which roles would traditionally have been played in the Opera by males.

And another reason, last but not least, for choosing these films is my own interest in Chan’s choreography. I have been a practicing martial artist, stage and film fight director, and certified teacher of staged combat for a number of years, and Chan’s fight sequences are (almost universally) regarded among my colleagues as among the best in the business. I do not here try to identify what makes a fight sequence “good” or “bad,” but the richness of many of Chan’s sequences allows for a discussion of various qualities such as humor, rhythmic variation, interplay with the environment, and several other “aesthetic” considerations not always readily associated with staged fights but which most fight directors would claim is integral to much of what we try to do. Thus my analytical approach differs from most other analyses of martial arts or action films. Through this approach, I hope to stay away from discussions of the evils of filmed violence by asking instead not only, “Why do we watch?” but also, “What do we see when we watch?“

As I mentioned before I begin the analysis proper, I wish to describe the reception continuum suggested by research into the pleasures of watching violent entertainment. Here I wish to provide a theoretical background to the fundamental question I’m asking: How do we analyze human movement in moving-image media? Because such considerations are underdeveloped in film analysis at present, I find it important to explain the more abstract issues involved. Readers with a more general interest in martial arts films and Jackie Chan may wish to skip ahead to the “Jackie Chan and the Art of Fighting” subheading for a working out of the argument with examples since I would also like my essay to speak to a broad range of readers.

Levels of reality (kinesia revisited)

My own interest in the aesthetics of violent spectacles concerns learning how to read (and write) the human body in motion. This interest is also central to many areas of dance studies, through which one comes to understand that viewed human movement has the potential to communicate in phenomenological ways. At the core of all kinesthetic theory is the belief that in human experience the physical and the mental are never entirely separate. This basic phenomenological understanding may at first seem esoteric or at best unrelated to film studies. However, a number of film theorists do draw a basic connection between mental understanding and physical sensation of viewed movement. For example, in “Aesthetics in Action: Kung Fu, Gunplay, and Cinematic Expressivity,” David Bordwell compares Hong Kong action cinema’s practices to Sergei Eisenstein’s theories:

Clearly many Hong Kong filmmakers aim, as Yuen Woo-ping puts it, to make the viewer “feel the blow. Not only must the action be legible and expressively amplified; it must be communicated, as energy is communicated from one body to another; it must be stamped on the spectator’s senses. . . . All this is, once more, far closer to the tradition of Sergei Eisenstein than of Raoul Walsh or Steven Spielberg. Throughout his career Eisenstein, always a man of the theatre, emphasized that expressive movement was at the core of cinematic mise en scène. First, the filmmaker had to discover concrete actions which an actor could be trained to execute in simplified and stylized form. Then the filmmaker had to devise a way of framing (mise en cadre) and editing (montage) which would sharpen and further dynamize the expressive movements. And at key moments these techniques could cooperate, double, and intensify one another in an assault on the spectator’s senses. If the effort was successful, the force of the movement and its onscreen presentation would stir in the viewer’s body a palpable echo of the actor’s gesture. It is precisely expressive movement, built on an organically correct foundation, that is solely capable of evoking this emotion in the spectator, who in turn reflexively repeats in weakened form the entire system of the actor’s movements; as a result of the produced movements, the spectator’s incipient muscular tensions are released in the desired emotion. (88, citing Eisenstein and Tretyakov, Expressive Movement 187)

These ideas of “felt blows, palpable echoes of gestures, and reflexively repeated movements” conform to what dance theorist John Martin calls “metakinesis” (communication through movement) and “muscular sympathy” (the phenomenological “feeling” associated with this communication). In “Action in Motion” I described at length both these concepts in relation to filmed martial arts movement and so will not repeat the descriptions here. Instead, here I want to explore some of the implications raised by using dance analysis to understand filmed martial arts sequences.

Central to the idea of muscular sympathy is the innate knowledge that a body cannot conceivably be made to do anything that the body cannot do. Yet while this is necessarily true of live performance, film may manipulate human movement in ways that affect our kinesthetic understanding of that movement. Since editing techniques and special effects can create the illusion of bodies doing things and moving in ways they conceivably should not be able to, any bodily understanding of movement may take on new dimensions in relation to certain filmed images or movements.

For example, John Martin describes a kinesthetic “revulsion for abnormality” as feeling that may possibly be evoked by watching a human contortionist’s movements (12). Yet even the most abnormal movements by a contortionist pale in comparison to the movements made possible on film. Thus horror films frequently employ special visual effects to create grotesque monsters and horribly mangled bodies that still move and kill. However, although we can analyze the psychical sensation evoked in this example in terms of a kinesthetic “revulsion for abnormality,” we must also consider it in relation to our intellectual knowledge that the images have some degree of unreality about them. The physical thrill of watching this movement in essence forms part of the intellectual experience of the genre and vice versa.

Martial arts films likewise employ special cinematic effects to simulate movements difficult to achieve otherwise. For example, in an early combat sequence from Steven Seagal’s Out for Justice (as discussed in “Action in Motion”), special prosthetic devices allow Seagal visibly to break one adversary’s arm at the elbow and pin another to a wall with a meat-cleaver. Kinesthetic responses to these gruesome visuals share the same visceral revulsion for abnormality viewers experience in many horror films–the “oh shit” moaned aloud in theatres (Richmon 232). However, no matter what specific, personal kinesthetic response to special effects viewers have, their physical sensation of viewing will almost always be modified by their intellectual knowledge that the special effects are not “real.” In this way, several different heuristic levels of intellectual understanding can temper kinesthetic response to viewed martial movement on film. One such level implies the understanding that the movement being witnessed is actually being done to or by a human being. This is the level of live performance of actual combat with malicious intent, and our intellectual understanding of this facilitates a sort of kinesthetic baseline response.

One step away from our untempered, baseline physical response to real violence is our intellectual understanding that the violence itself is being performed in part for entertainment purposes, and that the consequences of the violence are somehow limited. Examples of this hypothetical level include responses to boxing matches and other martial arts competitions. While actual, substantive injuries and even deaths do occur in these competitions, we still know intellectually about the rules and regulations in place that attempt to limit these consequences. In kinesthetic terms, this means that our physical response to the martial movement may be tempered by our intellectual knowledge that the consequences to the movers are limited. Different yet similar feelings may be evoked when we watch a boxing match whose intention is sport than are evoked through if we watch violence with malicious intent. These physical feelings are differentiated in part by our understanding intellectually that similar movements may exist in different contexts.

The third type or level of intellectual understanding of movement rests on an understanding that the violence itself is entirely representational. On this level, we know intellectually that the violence itself has no actual or real consequences. On this level, when audiences understand that the violence exists solely for entertainment, their bodies respond accordingly. This —the level of the martial arts film—is the movement category with which I am presently concerned. Here, in part because of the staged fight’s distance from “real” violence, fight directors or choreographers use consciously created movement patterns and sound cues, similar to those described by Eisenstein and Bordwell, in order to heighten the kinesthetic response of the audience.

I present these levels here as a heuristic device to allow discussion of physical differences in the reception of different violent genres. However, there is also reason to suppose that these heuristic separations have a degree of “real” reference within audience psychology. A number of psychological studies have found differentiated responses to viewing disgusting or violent images based on viewers’ perceptions of reality versus non-reality and the presence or non-presence of framing devices within the film to mark this distinction. In an essay “When Screen Violence is Not Attractive,” Clark McCaulley sums up the results of much of this research, which I will summarize here for clarification (Why We Watch 144-62). A number of studies point to the basic conclusion that viewers make intellectual distinctions about the reality or non-reality of viewed violence, which then directly affects their physical and psychical response to viewing the material. For instance, studies found that viewers were very likely to be disgusted by documentary footage of animal or human mutilations but conversely likely to be entertained or amused by basically the same visuals when displayed in a horror film. Their response apparently derives in large part from signals in the “frame” of the depictions which help them contextualize the images. The viewer uses a contextualized, mentally constructed continuum ranging from reality to entertainment. In this example, the documentary footage’s frame signals the “reality” of the representation while the horror film’s frame signals “fiction” to the audience, thus allowing viewers to enjoy essentially the same spectacle.

Various things can act as framing devices to distance the viewer from the visuals’ “reality.” Even the quality of the sound track can serve as an effective framing or distancing device. For instance, these same studies found that if the gory documentary visuals had a musical accompaniment, viewers were much less likely to be disgusted by the mutilation. This same kind of distanciation necessarily holds true as well for violence in filmed fights where an “unrealistic” use of sound or the inclusion of background music signals a degree of distance from “real” gore or violence and frees the viewer to appreciate the movement as entertainment. Thus in film, sound works in conjunction with other framing devices to signal, “This is entertainment; relax, enjoy.” Paradoxically, though, music also heightens emotions and thus the intensity of sensation in horror films and fight scenes. Thus music may make the fight “feel” more immediate while at the same time distancing it from “reality.”

Closely related to this observation is the finding that one of two possible mechanisms allow a viewer to enjoy violent or otherwise negative spectacles:

One way is to assume that the dramatic distance of fiction can moderate emotional reactions such that they provide an enjoyable arousal jag at small cost in negative hedonic tone. The other way is to assume that emotions instigated by fiction and drama are qualitatively different emotions from their everyday counterparts, and that the dramatically instigated emotions are always enjoyable. (McCauley 159-60)

Neither of these theories necessarily contradict the other, and possibly elements of both mechanisms work to one degree or another whenever violence or gore become entertainment.

The most intriguing of the two theories, however, posits the possibility of qualitative difference. This theory has some precedence in one of the oldest written dramatic texts—the Natyasastra of Bharata (written in India sometime around 200-300 AD) which focuses on an emotion called rasa. Rasa is essentially a separate emotion for “aesthetic or imaginative experience” similar to “regular” emotions but parallel on a different reality. This aesthetic theory asserts that everyday emotions and dramatic emotions are not the same thing. Thus, the Natyasastra assigns terms to distinguish dramatically induced emotions from their everyday counterparts. Whichever of the two theories above (or some combination of both) that one accepts, the basic conclusion drawn from the evidence is the same: Watching “real” violence or gore will elicit a certain physical, emotional and psychical response, while watching “imaginative” violence or gore will elicit a different physical, emotional, and psychical response. This second response will either differ by degree and involve the same sensation, or it will elicit an entirely different yet parallel emotion. Either assumption involves some sort of differing physical sensation among viewers based on intellectual constructions of a continuum of “reality,” or viewers’ perceptions of differing distances from reality.

In order then to reach a kinesthetic understanding of movement-meaning within the context of this continuum, we must also take into account specific, individual differentiation based on personal bodily memory. The presence of individual, sedimented bodily memory (as discussed in “Action in Motion”) implies that viewer response is additionally differentiated by personal experience or situation. For example, a person witnessing a violent altercation between strangers will almost certainly experience a different physical response than will the same person witnessing a violent altercation involving immediate friends or family. Since—in terms of bodily understanding —immediacy or distance have both mental and physical referents, types of personal immediacy may include aspects related to one’s knowledge of his/her relation to the person being viewed as well as actual, physical presence. Thus, it can be said that physical response to viewed violence may be influenced by that violence’s immediacy to the viewer (in much the same manner that the addition of music distances viewers from the “reality” of gore and violence in the studies cited above).

In this context, filmed movement necessarily “feels” different from live movement. This difference in sensation can be understood in two basic ways: either as a lessened quantity of the same sensation, or as a qualitatively different experience. The possibility of a qualitatively different experience is the most intriguing for film studies, especially when considering the fact that mediated cinematic presentation frequently allows clearer sightlines and close-ups and thus a better view of the movement itself.

All of this implies a continuum of sensation in response to viewed violent spectacle based on perceptions of reality and immediacy, mediated by framing devices which signal differences in the intent of the viewed movement, and further differentiated by individual bodily and psychical memory. Thus, any analysis of violent spectacles—such as those which largely define the martial arts film genre—must account for the spectacle’s placement within this hypothetical continuum.

For instance (as argued in “Action in Motion”), part of the appeal of Steven Seagal’s films is the level of “reality” depicted in the hand-to-hand combat sequences. This level of reality forms a large part of the films’ marketing strategy and is closely tied to Seagal’s star persona. Thus, when Seagal advertises the fights in his films by making reference to his “really” having been in many real fights where his opponents ended up “hurt or carried away” (Richmond 306), he is, in effect, speaking directly to his film’s placement within the continuum of “real” to “fictional” violence.


Go to page 1 2 3 of this essay.

JC 44



Print version