Lady in the Lake (Robert Montgomery, 1947).
It is not quite right to call a point-of-view shot “subjective.”1 As Jean Mitry observes, there is an essential tension in the spectator’s relationship to a filmic point of view: “At the movies I am both in this action and outside it, in this space and outside this space. Having the gift of ubiquity, I am everywhere and nowhere.”2 If subjectivity designates an individual’s embodied movement through the world—structured by goals, obstacles, detours, means—then point-of-view shots only highlight the disjunction between seeing through someone else’s eyes and walking in their shoes.
Lady in the Lake (1947) was partly conceived with this equivalence in mind, advertised in trailers as a “revolutionary innovation in film technique,” starring Robert Montgomery and “You!” But if anything, this sustained experiment with a camera that purports to collapse viewer and character only accentuates the gap between a point of view and a genuine perspective. The act of seeing, one realizes, involves more than just occupying a point in space. For this reason, Montgomery’s bold attempt to replicate the first-person narration of Raymond Chandler’s novels ultimately fails: Being Philip Marlowe could not possibly be this boring.
But if point-of-view shots cannot be assimilated to character subjectivity, how should we understand their cinematic function? Since Lady in the Lake, cinema history has furnished several more extended uses of the point-of-view camera, with two notable instances in the past year alone: RaMell Ross’s Nickel Boys and Steven Soderbergh’s supernatural thriller Presence (both 2024). The range and variety of such experimentation should finally dispel the tacit assumption that point-of-view shots must do just one thing.
Consider the early, minute-long silent How It Feels to Be Run Over (1900), directed by Cecil M. Hepworth, which opens with a static view of a stretch of road. A horse-drawn coach trundles along and then veers off to the side, out of frame. Soon after, an automobile appears, swerving wildly about until it fills the screen, seeming to collide with the camera. The film is part of a cycle of Hepworth productions, including Explosion of a Motor Car (1900) and How to Stop a Motor Car (1902), featuring breakdowns and malfunctions in a then-nascent technology. (One imagines that a contemporary update might center on AI-driven catastrophes.) Unlike the others, How It Feels to Be Run Over places the viewer in the position of the unfortunate victim. In this regard, it is part of the broader early-cinema tradition of phantom rides and travel films, usually made by attaching a camera to the front of a moving vehicle, in which narrative is a secondary consideration; the primary concern is to present a fantastic, and sometimes phantasmatic, view of the world.
Hardcore Henry (Ilya Naishuller, 2015).
In Ilya Naishuller’s Hardcore Henry (2015), from over a century later, we discover that our point of view is attached to that of a cyborg who is awakened on an operating bed and soon finds himself navigating a plot involving a psychotic inventor with telekinetic powers. Despite the obvious differences from Hepworth’s film, one recognizes in it a similar tendency to explore a world rather than articulate a coherent narrative: Its main interest lies not in the convoluted machinations of its action-film plot, but in the range of unusual sights and situations it renders perceptible to the audience. In the first few minutes alone, we are treated to Henry ejecting himself from an aircraft, crash-landing on a busy overpass, and then flying off its railings onto the street below. At two points in the film, we even get a quasi-split-screen effect motivated by the misalignment of Henry’s augmented eyes. Call it: How It Feels to Be a Cyborg.
That Henry’s sight is only partly human also indicates how issues of point of view are inevitably bound up with questions concerning technologies of vision. It is eventually revealed that Henry’s viewpoint is, unbeknownst to him, being transmitted elsewhere—that he is effectively functioning as a moving surveillance camera. The further implication is that we, the audience, may be identified not with Henry, but with the presumed viewer of this livestreamed footage. It is a sort of visual rug-pull—a formal echo of the famous scene in Fritz Lang’s The 1,000 Eyes of Dr. Mabuse (1960) in which a two-shot of a couple talking in a hotel pulls back to reveal that we have been watching them on a surveillance monitor. Dave Kehr called this, from Lang’s final film, one of “the first overtly modernist flourishes in cinema.” It was thereafter impossible not to consider everyfilm image as being at least implicitly attached to a potential viewer. Found-footage films such as The Blair Witch Project (1999), Zero Day (2002), and My Life on Ice (2002) depend on collapsing an act of viewing within the film-world with the audience’s own experience. (Such films often include some indication of how the footage was found and thus can be seen.) So-called “screenlife” films such as Unfriended (2014) and Searching (2018) likewise hinge on the fact that the locked-down perspective of a flat screen can present virtual movement that’s just as real and captivating as any passage through physical space.
Dossier 51 (Michel Deville, 1978).
Even before Lang’s valedictory flourish, Hitchcock’s Rear Window (1954) had already demonstrated that the act of seeing is in certain respects a matter more epistemological than technological: more about the relations between the seer and what is seen than the specific mechanisms that make sight possible. Michel Deville’s Dossier 51 (1978) demonstrates this principle with especial perspicuity. Like Lady in the Lake and Hardcore Henry, the film is a sustained experiment in point-of-view shooting. Unlike those films, however, it sticks not just to the perspective of a single person but splits itself across the members of an intelligence team seeking to control a diplomat, Dominique Auphal: code name 51. Through the eyes of these agents, we learn of his wife’s extramarital affair, his troubled family life, his military service. We learn of his work habits, his moral rectitude, his former relationship with an anarchist when he was a university student.
What distinguishes Dossier 51 from other point-of-view experiments is its completely external perspective on the information it narrates—a perspective that for the duration of the film is ours as well. The disorienting quality of viewing the film derives from our complete inability to determine, from a veritable deluge of information, what is or is not of significance, what may or may not be useful to the ultimate goal of controlling the man known as 51. The brilliance of Deville’s uncanny conceit lies in how it harmonizes the viewer’s detachment with that of the intelligence agency: We identify with the agents not just because we share their fields of vision, but because we are similarly tasked with making sense of a situation which we do not yet understand. Dossier 51 is a key text of conspiracy and surveillance films because it takes the voyeuristic externality of the genre to something of a terminal point.
Enter the Void (Gaspar Noé, 2009).
Gaspar Noé employs the point-of-view camera to rather different ends in Enter the Void (2009). Initially filmed from the perspective of Oscar, an expat drug dealer who haunts the neon-light nightlife of Tokyo, the film quickly gets into more sensorially intense territory than anything discussed so far. After Oscar is shot, the film’s perspective seems to float apart from his body, offering us impossible out-of-body views, segueing into a dazzling slideshow of memories, and becoming increasingly unmoored from chronological spacetime. Noé has called Enter the Void a “psychedelic melodrama,” which makes sense if one views the film as he does, as “a dream of someone who read The Tibetan Book of the Dead, and heard about it before being [shot].” Authorial intention notwithstanding, it may be more productive to see Noé’s extended point-of-view experiment in the tradition of what Gene Youngblood in 1970 termed “expanded cinema,” a label that encompasses the Stargate sequence in 2001: A Space Odyssey (1968), Jordan Belson’s “cosmic cinema,” and James and John Whitney’s mandala-inspired films, among other technologically forward efforts. Youngblood’s term referred to how such artworks pointed the way toward—and, in a sense, simply were—veritable expansions of human consciousness. (“Expanded cinema isn’t a movie at all: like life it’s a process of becoming, man’s ongoing historical drive to manifest his consciousness outside of his mind, in front of his eyes.”)3 Even if we are suspect of his more bombastic technoutopian rhetoric, he was right to identify filmmakers’ approximations of a cosmic vision, otherwise inaccessible except, perhaps, by the use of psychedelic drugs.
Enter the Void thus draws an explicit link between the cinematic point-of-view camera and the experimental traditions to which expanded cinema belongs. Although it is not shot in the point-of-view style of Noé’s film, Lois Patiño’s Samsara (2023)—also inspired by the Bardo Thodol, as The Tibetan Book of the Dead is properly called—offers a recent example of the creative possibilities of this connection. The film’s two halves, set in Laos and Tanzania respectively, are shot conventionally, but they are bridged by a fifteen-minute passage that functions as a de facto point-of-view shot, in which we take the perspective of a dying woman as she moves into the afterlife. We are instructed by onscreen text to shut our eyes for the length of this journey into the void. For a while, we see only blackness, accompanied by a dense, eerie soundscape. But soon we get intermittent flashes of white light and fields of color which form into intricate, radially symmetric patterns not unlike those one might see in a film by Belson or the Whitneys. Going further even than Noé’s point-of-view psychedelia, Patiño makes the invitation to share a perspective an injunction to close one’s eyes—transforming the insides of our eyelids into an extension of the screen itself.
Nickel Boys (RaMell Ross, 2024).
In no two of these films does the point-of-view effect serve the same function. Aesthetic possibilities are not synonymous with technical ones. RaMell Ross’s Nickel Boys demonstrates just how much more terrain there is to explore within the constraints of this single formal conceit. The film unfolds from the perspective of one, and then two, Black teens, Elwood and Turner, attempting to survive a Jim Crow–era Florida reform school, but departs from the usual compositional tendencies of prior experiments. In Lady in the Lake, Montgomery employs relatively long shot durations, prioritizing spatial legibility and continuity of movement above all else. Ross freely dispenses with these expectations, employing shorter shots, tighter framing, and edits based not on dramatic legibility but sensorial rhyme. The result is dazzlingly impressionistic. An extreme close-up of a cake baked by Elwood’s grandmother becomes, by virtue of camera placement and distance, out of alignment with the present-tense wide shots that ostensibly frame it, the memory of every cake that she ever baked for him. More harrowingly, a scene where a young Elwood, shirtless, is prodded by the cane of a middle-aged white man, framed to momentarily anonymize those involved, becomes every time a Black man has been objectified by racist authority. Unfixed from their strict narrative function, the film’s images expand outward into a dense network of association.
Even when Nickel Boys begins to sketch something more like a continuous dramatic situation, the film uses its slipstreams of sensation to move fluidly between the boys’ present-tense perspectives, between past and future, between reported fact, harrowing memory, and half-remembered dream. The screenplay was adapted from Colson Whitehead’s novel, and it is a vestige of its literary origins that Elwood and Turner are positioned as opposites—on the one hand, the idealist who wants to change the institution from within; on the other, the cynic who would seem to look out only for himself. Nonetheless, the film’s split perspective conveys with moving lucidity the sense that no matter how many sights one shares with another, their ways of seeing will never be one’s own.
Presence (Steven Soderbergh, 2024).
The impossibility of truly occupying a first-person perspective other than one’s own is, in a much different context, the starting point of Steven Soderbergh’s Presence, which unfolds from the point of view of an invisible ghost. In the film’s opening minutes, the camera glides around an empty mansion, floating weightlessly through the staircases and rooms, later observing as a realtor attempts to sell the place to a family of four. The family, we soon learn, is attempting to escape from the fallout of some prior trauma. When they eventually move in, only the teen daughter, Chloe, seems able to sense this watchful presence.
After Hitchcock, it is impossible to make a movie without considering an audience’s awareness of generic convention. His Frenzy (1972) offers a perfect illustration of this filmmaker-viewer dynamic when, after tracking a serial killer and potential victim as they walk up the stairs of an apartment building, the camera declines to follow them inside and instead reverses direction, descending backward down the stairs, coming out onto the street, framing the building’s façade and the opaque apartment window seen from the outside. What we call Hitchcockian suspense depends on our awareness that while our perspective remains restricted to the exterior, the murder takes place inside. Indeed, in this post-Hitchcockian situation, and especially in films with Hitchcockian subject matter, it has become impossible to even move the camera without creating a sense of narrative expectation. Claude Chabrol’s Wedding in Blood (1973), for example, mainly follows a torrid affair between the deputy mayor and the mayor’s wife. But when, for the first time, the camera follows the mayor as he goes out to his car, the scene is imbued with suspense: there must be a reason, we think, that the camera is following him in this specific place, at this particular time. Filmmakers—and especially those working within recognizable genre conventions—have accordingly confronted something like the equivalent of the Euthyphro dilemma: Is the camera present because things are bound to happen, or are things bound to happen because the camera is present?4
With his latest experiment in leveraging the economic constraints of production into creative formal decisions, Soderbergh offers his own inventive response to this post-Hitchcockian conundrum: every shot of the film will comprise a long, fluid take that potentially identifies the camera with the ghost. The film’s unique charge, though, derives from the internal tensions of this very relationship. At every moment, there remains a lingering question about the extent to which the camera can simply be identified with the ghost, as opposed to the film itself. (In this respect, the film that Presence most directly recalls is Samuel Beckett’s unclassifiable Film [1965], in which an elderly man, played by Buster Keaton, is menaced by the camera itself.) Thus, when Chloe’s would-be boyfriend attempts to drug her by spiking a glass of orange juice, and the camera moves toward it, causing it to fall onto the floor, the event is equally attributable to the ghost (that is, it’s protecting Chloe) and the exigencies of story structure (that is, the narrative demands that the drugging cannot yet occur). Soderbergh’s novelty is to have created a situation where this tension is continually active and irresolvable.
In the nearly 80 years since Lady in the Lake, extended experiments with the so-called “subjective” camera have become a recognizable tradition. André Bazin once wrote that “inventions belong to those who master them.”5 Although D.W. Griffith didn’t invent the close-up, Bazin remarks, he nonetheless became a master of its use. The point-of-view camera, in contrast, does not yet seem to have been mastered. But if no single use of the technique ever comes to be the singular use of it, that is nothing to lament. It simply means that no instance of the point-of-view camera has exhausted the effect, and that it may be changed, altered, and renewed by subsequent uses. Like any great invention, it still has most of its life still before it.
- For more discussion on this point, see Daniel Morgan,”Where are we?: Camera movements and the problem of point of view,” New Review of Film and Television Studies 14, no. 2, 222–248. ↩
- Jean Mitry, Esthetique et psychologie du cinema. I. Les structures (Editions Universitaires, 1963), 179. Cited in George M. Wilson, Narration in Light: Studies in Cinematic Point of View (The Johns Hopkins University Press, 1986), 55. ↩
- Gene Youngblood, Expanded Cinema (P. Dutton & Co., Inc., 1970), 41. ↩
- “Is the pious loved by the gods because it is pious, Socrates wonders, “or is it pious because it is loved by the gods?” (Plato, Euthyphro, 10a.) ↩
- André Bazin, “The Technique of Citizen Kane,” in The André Bazin Reader, ed. and tr. Timothy Barnard (caboose, 2022), 47. ↩