Carter (2022).
There’s a scene in Carter that exemplifies how the film pushes the formal conceit of the “single-take” actioner to new extremes. Amnesiac protagonist Carter Lee (Joo Won), tasked with rescuing the kidnapped daughter (Kim Bo-min) of a renowned virologist (Jung Jae-young), walks with the girl into an alley. Suddenly, a motorcyclist races by, snatching her up. Another rider arrives, whom Carter kicks and sends into a pile of boxes. Arcing up to an overhead position, the camera observes from above as a third bike collides into the tipped-over second, flipping both vehicle and passenger onto the ground. As Carter steals one of the bikes and begins his pursuit, the camera drops to what seems to be mere inches from the pavement, somehow keeping pace as our hero accelerates to dangerous speeds. A yellow van suddenly passes over the still-low-to-the-ground camera from behind and makes a squealing U-turn, at which point the camera leaves Carter to follow this new player. With a flourish, the camera loops back under the van, tracking the vehicle’s movements from between its wheels. The chase spills over onto a highway, leading to increasingly complex stuntwork. In one moment, the girl, dangling off the side of the van, falls backwards after losing her grip. To save her, Carter leaps off his bike and gives her a midair push, like an athlete diving for a volleyball. His right shoulder briefly scrapes the asphalt before he snags a last-minute handhold near the van’s wheels.
This action-packed snippet captures how, unlike the literally earthbound perspectives of other recent single-take films like One Shot (2021) and Crazy Samurai Musashi (2021), the camera in Carter is ecstatically mobile, tracking its hero across assorted high-speed chases, chaotic melees, and vertiginous falls. The “one shot” illusion, created to give the sense the entire movie is staged and filmed in one single, edit-free take, is far from seamless. There are moments in which the use of green-screened backgrounds is so obvious that the image turns disorienting, the impression of three-dimensional physical space lost within the visual jumble of disaggregated composite image elements; elsewhere, cuts are “masked” so shoddily that the effect is akin to an outright jump cut. In a way, however, these “mistakes” magnify our sense of the film’s ambition, giving the impression of aspiration outpacing means.
Carter (2022).
As the camera swoops and weaves, the virtuosity of its movements underscores its status as its own, physical body. Even when the overtness of digital manipulation suggests that the “camera” is likely in part virtual—a “camera effect” generated in post-production—the single-take conceit and the manifest mobility of the implied camera continually activate the idea and impression of the camera as a physical entity physically moving through physical space. Especially when it seems to possess a mind of its own—peeling away from Carter to pursue the van, for instance, or performing movements that are overtly narratively extraneous, like repeatedly framing the underside of vehicles—the camera in Carter often feels like as much a body as the stunt-people it captures.
But, of course, the stunt-people are hardly sitting patiently, waiting for the camera’s gaze to alight on them. They are leaping, falling, climbing, dangling, fighting, driving, sprinting, shooting—daredevil feats of risk and coordination that, in scale and complexity, outpace most action scenes of the last few years. The scene described above is a case in point (and not even mentioned is the part where a brawl unfolds across the interior of three vans driving side-by-side, their open doors allowing combatants to freely navigate between them as one might contiguous train compartments). Another emblematic scene is a fight between two freefalling characters that ends with the victor parachuting onto the back of a moving truck, leading directly into the next set piece: a Mad Max-style car chase featuring gun-toting pursuers flanking and attempting to board the larger vehicle.
Carter (2022).
The examples abound. As the camera flies, swerves, and tumbles, so do the stunt performers it captures. Tracing an elaborate path through space, the camera conveys the sense of its own physical body, but its trajectory is always constrained by the tacit imperative of looking at the bodies in front of it and their spectacular movement. On the other hand (and this may seem an obvious point, but it bears mentioning), for stunt performers’ movements to appear spectacular, they need to be captured on camera to a significant extent. In other words, the movements of bodies themselves must defer to the camera’s own movement, to where the camera is pointing as a function of where it’s located in physical space. To an extent, this description applies to most live action cinema this side of abstract experimental film. All such cinema involves a dance of movement between the camera and what it captures, between where the camera points and is moving in relation to what it films and how these things are moving. What’s unique about Carter is the degree to which it recasts this relationship in terms of virtuosic physicality: not just where the camera happens to be placed vis-à-vis what it records, but the concerted, effortful physical coordination between two sets of bodies—the body of the camera, and the bodies of the stunt people it films. Furthermore, it’s because both camera and stunt people demonstrate spectacular movement—versus just one or the other—that their shared physicality is emphasized. Rather than using relatively subdued camera set-ups to capture dynamic bodies, or dynamic camera movements to capture relatively subdued bodies (e.g., many cases of “single-take” showmanship like Russian Ark [2002] or Enter the Void [2009]), Carter sets both poles in motion. It’s because both camera and stunt people are foregrounded as individual physical bodies with individual physical capacities that their alikeness as bodies and the work of physically choreographing them is emphasized.
The opening, single-take fight in Jung’s The Villainess (2017) begins with a first-person perspective (top) then switches to a third-person view (above).
Put another way, Carter continues a tendency in director Jung Byung-gil’s action cinema to project a sense of filmmaking itself as stuntwork: as involving not just the physicality captured by the camera, but of the camera itself. This conception collapses the distinction between first- and third-person perspectives, offscreen and onscreen, camera-back and camera-front into a single, shared physical arena through which camera and stunt-person alike traverse. Jung is not the first filmmaker to express such a “stunts” aesthetic. Tsui Hark and Gareth Evans, for example, are two action auteurs who have repeatedly highlighted dynamic physicality both of and in front of the camera. Jung, however, takes this physicalist style to rare extremes. This stylistic commitment is visible in his previous two films. Confession of Murder (2012) opens with a precarious foot chase featuring handheld camerawork that's kinetic to the point of disorientation. This scene pushes the expected equilibrium between camera-body and onscreen body (i.e., the expectation that each should defer to the other’s position and movement to some extent) to its limit. In The Villainess (2017), a single-take fight shot first-person abruptly shifts into a third-person perspective that, nonetheless, maintains the same dizzying, embodied camerawork. Here, the film primes the viewer to experience the camera (and, by extension, the camera operator) as a physical body. By having the same camerawork style carry over from first-person into third-person views—and by making the transition between the two so jarringly pronounced—the scene suggests that, even beyond cases where camera movement is explicitly tied to a character’s embodied perspective, the camera (and the camera operator) remains a physical entity sharing space with what lies before the lens.
This stunts ethos—in which the very art of filmmaking is reframed in the physicalist, choreographic terms of stunt performance—resonates with Jung’s own background. Trained at the stunt academy Seoul Action School, Jung ended up working largely behind the camera, but his stunt training seems to have seeped into his filmmaking, inspiring multiple scenes—and, with Carter, an entire film—in which the camera moves like a physical body, dismantling the classical division between form and content to express a shared, virtuosic physicality. Even when digital visual effects clearly play a mediating role, Jung’s films give the impression of cinema as not just filming but choreography, not just moving images but a physical doing. His is an alternate vision in which stunt people are not the below-the-line footnotes that mainstream discourse continues to treat them as. Rather, as virtuosic exerting bodies moving with coordinated dynamism through physical space, they become something of a model for cinema itself.
The Action Scene is a column that explores the construction of action set pieces in order to deepen appreciation for and spark discussion about action cinema.