Scorsese's enunciative presence

My previous post on The Impossible POV reminded me of the following piece on Scorsese I wrote years ago. Here it is, slightly touched-up, and with pictures.

Taxi Driver: Travis on the phone with Betsy

In a memorable shot from Taxi Driver, Travis Bickle, the cabbie-turned-assassin-turned-rescuing hero, is abandoned at a payphone by a rightward-tracking camera that chooses to present to its spectator a long, empty hallway instead of its protagonist's futile attempts to reconcile himself with a woman he has irreparably offended. Stephen Heath has observed that the camera movement in this shot threatens to undermine what he calls "narrativization" — the process by which the camera conceals its autonomy behind the narrative action it keeps before it, often as an alternative to the displacement of camera control onto character point-of-view. When camera movement becomes more intricate than character movement, when the camera abandons characters outside the frame or allows them to escape its confines, the autonomy of the camera is exposed.¹

This shot points to a significant element of Scorsese's style. The self-pronounced enunciative presence that this shot demonstrates is used throughout Scorsese's work to achieve both narrative and thematic purpose. It is used to provide thematic commentary, to protect the spectator from association with characters who would otherwise appear too harsh and unappealing, and even to create a new character, one that is never actually seen, but whose presence is strongly felt in the effects of the noticeably autonomous enunciative entity. Each of the three films examined here — Taxi Driver, Raging Bull, and The Last Temptation of Christ — demonstrates one or more of these uses.

Taxi Driver

This presence is employed in Taxi Driver to underscore the instability of the film's protagonist. Travis' inability to define for himself a successful social role is reflected in his inability to maintain camera control, that is, to act as disguise for the autonomous camera. Often the camera leaves Travis behind, as if he is too weak to narrativize it. At other times, camera control is removed from Travis' point-of-view, even at moments we expect that control to follow Travis' gaze.

Time to leave: Travis is abandoned by the camera

The above-mentioned shot is a perfect illustration of the former situation. Instead of staying with Travis, keeping him in frame to allow the spectator to see his reaction to Betsy's responses, the camera abandons him (who, even when in frame, is not centered, his placement in frame right forcing him to share the visual emphasis with two telephones), giving more emphasis to an empty hallway with no apparent narrative significance. In this way, Travis is marginalized to the extreme. The film that is, according to its title, about him finds him too insignificant an object to present. The camera's autonomy becomes another "character," emphasizing the door at the end of the hall, hinting to Travis that his attempts at reconciliation are futile, that he should leave the building, expressing a palpable impatience with a character whose hopeless attachment to a previous failure is postponing future narrative resolution.

Making its own action: the camera pans in the opposite direction of Travis' cab

Two other such examples demonstrate the camera's disrespect for the film's protagonist. As Travis returns to the garage after his first night of driving, the camera follows his car's entrance with a brief leftward pan, but then abruptly pans right, allowing Travis to drive out of frame left. The camera comes to rest as Travis' car enters and stops in frame right. Instead of simply following the path of Travis' car, that is, instead of following Travis' action, the camera creates its own action by moving in the opposite direction. Travis is seemingly unable to hold the camera's attention, and at the end of this shot is left hidden in his cab on the edge of the frame, literally marginalized. Then, after Travis' quasi-philosophical chat with Wizard, both Wizard and the camera — perhaps each sensing a danger in Travis that should not be waited upon — leave Travis behind, departing in Wizard's cab. From the back of the moving car, we see Travis walking away, disappearing into the background, becoming increasingly insignificant even to the camera that claims to be telling his story.

Annexed POV: the $20 bill on the front seat and two shots from behind Travis' head

Many of the moments that appear to preface Travis' point-of-view (subjective shots) are annexed by the autonomous camera, suggesting that Travis is not capable of properly "showing" the spectator anything "through his eyes." He is therefore unable to conceal the camera's control and the camera chooses to risk revealing itself and exposing the film's artifice (its construction as film) instead of trusting Travis with a point-of-view shot. After Travis returns to the garage after his first encounter with Iris (when Sport leaves a $20 bill on the front seat of the cab), he stares down at the money on the seat. This downward gaze is followed by a cut to the crumpled bill, a shot which appears to be Travis' point-of-view. The camera then tilts up to show Travis looking at the bill, the apparent subjective shot revealed as another view from the autonomous camera. As Travis sits in his cab waiting for Iris, right before buying some of her time, the camera shows what Travis is looking at only by shooting from behind his head, including him in his own "subjective shot," distrusting him even with his own vision.

Overhead view: the porn theater (note the caption in the magazine: "How You Spend Your Money Affects Your Sex Life"), Betsy's desk, and Travis' stall at the shooting range

There are also several instances where Travis refers to things on a table or desk in front of him, but instead of then seeing those things from his point-of-view, the camera removes itself to an overhead position, making it clear that it alone is presenting the objects. This happens at the porn theater's candy counter, at Betsy's desk when Travis "volunteers," and at the target range where Travis practices with his new guns. The camera thus reduces — if not eliminates — Travis' credibility as presenter, repeals his (assumed) assignment as center of spectator identification.

Removal from the traumatic scene: the overhead tracking shot out of Iris' room (left and center), the crane shot outside (right)

What this repeal ultimately allows, however, is for the spectator to watch Travis' actions without being implicated in them. This allows Scorsese to present to the spectator characters that are disturbing — indeed, psychotic — without forcing the spectator to be too closely associated with those characters. The spectator is therefore protected and able to experience a narrative that s/he might — if forced to experience every psychotic action through the protagonist's eyes — find repulsive. To this end, Scorsese often allows the camera further autonomy with the use of overhead shots (of Travis in bed, for example) and shots that pan across a room and reveal Travis' surroundings before they reveal his presence. These shots, unattached as they are to character point-of-view, call attention to a non-diegetic enunciative presence (i.e., the camera), but at the same time afford the spectator a safe and comfortable distance from the disturbing character. The camera thus makes it clear that Travis is being presented, not advocated. This is most important after the final shootout, when the camera removes itself — and the spectator — from the gruesome bloodbath in Iris' room with an overhead tracking shot. This shot begins a slow withdrawal from the building that ends in a crane shot elevated from the chaos on the street. At the moment of greatest disturbance, of Travis' most psychotic behavior, the spectator is spared the trauma of prolonged exposure to Travis' point-of-view and is allowed to evaluate the situation from a seemingly more "objective" viewpoint.

Raging Bull

On display: Jake caught in a camera's flash

Raging Bull extends this notion of safe spectatorial distance by presenting Jake La Motta, another disturbing character, in documentary fashion. The enunciative presence is extremely pronounced here, always creating the impression that Jake is on display, whether in the ring or on the stage. In this way, the enunciative presence works to uphold the presentation principle suggested by Scorsese's biblical postscript:

Whether or not he is a sinner, I do not know," the man replied. "All I know is this: once I was blind and now I can see."

Scorsese's emphasis is on presentation without apparent subjective commentary (i.e., without passing judgement). To this end, the film incorporates the various conventions of documentary filmmaking.

The image of normalcy: home movies

While present in ways throughout the film, this documentary style is most pronounced in the montage of fight stills and home movies that chronicles Jake's successes in the ring and the seemingly happy moments from his family life. This segment provides safe spectatorial distance from a disturbing character by calling attention to the film's artifice, its very construction as film spectacle. Intertitles, non-diegetic interruptions which give information about each fight, are interspersed with stills of those fights. The very use of stills is a dramatic contrast to the previous motion picture fight segments, and forces an awareness of an external enunciator, one that can prolong the image for an unnatural duration (one that can "stop time," the use of slow motion in the film having a similar effect). The home movies are the epitome of self-reflexivity: their color a striking contrast to the black-and-white of the narrative; their scratched surfaces, jump cuts, and bleeding intrusions of overexposure vivid testimony to their physical construction; their characters directly addressing the camera and acknowledging the hidden means of production; their happy moments an awkward contrast to the familial tensions shown in the narrative.

Like in Taxi Driver, we are often spared the discomfort of sharing Jake's point-of-view at disturbing moments. In the ring, we are often shown the point-of-view of Jake's opponents, forcing us to be "punched," perhaps, but protecting us from implication in the violence Jake perpetrates. When Jake first sees Vikki by the pool, his offscreen gazes are not followed by his point-of-view, but by continually moving shots that could only be from the perspective of an autonomous camera. In this way, the film protects the spectator from becoming too involved with Jake's actions. We are therefore not directly implicated in (i.e., not forced to experience from Jake's perspective, "through his eyes") the violence he commits against family members or against himself. We are more able to watch Jake, to watch the "objective" presentation of Jake, and not be instantly repelled.

The end: Jake's empty dressing room

Paradoxically, however, this safe spectatorial distance, a result of the apparent "objectivity" of the film, is dependent upon the extreme subjectivity of the camera. What creates this safe viewing distance is the camera's selectivity, the choice of images it presents in lieu of Jake's own subjectivity. Several segments begin with short "establishing montages," brief sequences of images of the location of subsequent action, often presented with the offscreen sound of that action already in progress. This allows the camera to declare its authority, to explain that it controls — that it promotes or hinders — the process of narrativization by choosing its own objects, that it has the power to restrict our view. While the restriction spares us implication in Jake's subjectivity, it nonetheless emphasizes the presence of the enunciative apparatus and states the truth behind this safe distance: even the film's objectivity is created, subjective. The film's closing segment provides an example: We hear Jake rehearsing his act over a sequence of establishing close-ups — of the Barbizon sign, the show placard, a lightbulb, bottles, a telephone, some clothes hangers — then see him finally in a static medium shot. He is allowed to leave the frame, the camera remaining static, choosing not to follow Jake onstage, but to end his story there, in the dressing room.

The Last Temptation of Christ

As Raging Bull amplifies the removal to safe spectatorial distance that is presented in Taxi Driver, The Last Temptation of Christ amplifies the notion, also present in Taxi Driver, that the autonomous camera can imply, if not become, its own character, one that interacts with the others already visible in the film. Not only does this film create a character out of the enunciative presence, it uses that presence/character to underscore one of the film's main themes.

Bird's-eye view: God's POV

The film opens with crane shot which descends into an arboreal location, accompanied by the sound of footsteps from an unseen source. The first cut is to an extreme high angle of Jesus sleeping on the ground, and is preceded by the screech of an unseen hawk on the soundtrack which suggests a raptor falling onto its prey. It is followed by two more cuts in to closer, ground-level framings. Jesus then explains in voice-over: "The feeling begins: very tender, very loving. Then the pain starts. Claws slip under the skin and tear their way up. Just before they reach my eyes, they dig in." The arrival of the pain in Jesus' head is thus preceded by a shot that appears to be the point-of-view of an unseen character, the footsteps suggesting that it is not merely the view of the de-narrativized (i.e., noticeably autonomous) camera. The screeching sound and the high angle shot which follow suggest the point-of-view of something high up, above the earth. Then, in the next segment depicting this terrible head pain, there is a fast track in towards the door to Jesus' home, the screeching hawk is heard again, followed by a shot of Jesus dropping to his knees, and then a shot from an extreme high angle (almost overhead) of Jesus writhing on the floor. He explains: "God loves me. I know he loves me. I want him to stop." With that statement, Jesus indicates that the autonomous camera (with its fast, de-narrativized tracking shots and extreme high angles) is representative of God's own point-of-view as he rushes toward Jesus to "get into his head." There are other high angle and overhead shots to follow, many preceded or accompanied by mention of God (as when Jesus prepares food with his mother, asking her, "You can't cast out God, can you?" or when he tells Jeroboam that God wants to "push me over," followed by a shot that sweeps over the side of a cliff, certainly detached from jesus' point-of-view, but not from God's.

Shared perspective: Jesus on the beach

There are further examples of the self-proclamation of the autonomous camera, these having to do with Jesus' own point-of-view. As he walks on the beach, pursued again by the sound of footsteps, he turns around quickly to see the source of the sound. The camera, which has been following behind him, pans around to reveal what constitutes his eyeline match. Then the camera tracks back, essentially reframing Jesus in his own subjective shot. As he is struck again with another head pain, the camera removes to another high angle, as it did in the first two such moments. (Once again, the high angle is preceded by the sound of a screeching hawk. Or perhaps it's a kestrel? See The Windhover by Gerard Manley Hopkins.) In this way, the points-of-view of Jesus and God are connected, linked together as the same perspective in one continuous mobile shot. Thus the later thematic notion that Jesus is in fact God (as he tells the high priests in the temple) is first represented visually, via the autonomous camera. The film continues to make this connection through Jesus' mediated point-of-view shots. Often, when Jesus speaks to another character (as when he chides those at the stoning), the exchange is presented in one shot which substitutes a pan (from Jesus to the other character) for shot/reverse shot cutting. In this way, the autonomous camera, representing God's point-of-view, infuses Jesus' subjectivity as well, mediating his "reverse shots" with the camera movement of God/autonomous camera.

Just a movie: projector trouble

This noticeable autonomy assumes greater thematic proportions at the very end of the film, when Jesus' exclamation "It is accomplished!" is followed by the intrusion of what appear to be broken pieces of film into the frame. This simulated projector malfunction references the external agency of production, breaking the diegesis with the reminder that this has all been "only a movie." This may serve to subdue accusations of blasphemy by reminding the squeamish spectator that this is merely one interpretation of the life of Jesus, an interpretation that only exists in this film. It implies something much more significant to the film's main thematic statement, however, and it does so by using a signifier of artifice (the broken pieces of film) to call attention to the autonomous enunciative presence (the fact that this is just a movie). Because Jesus and God have been connected to that autonomous camera/enunciator, they too are now implicated in the notion of artifice. If the autonomous enunciative presence has revealed itself as artifice with the self-reflexive broken film pieces, then Jesus and God are also thus revealed. The underlying implication here being that film and God and Jesus are all created by an external agency. Now Saul/Paul's words to Jesus return with extra significance: "I created the truth out of what people needed and what they believed. If I have to crucify you to save the world, then I'll crucify you. And if I have to resurrect you then I'll do that too, whether you like it or not." Then, the extra-filmic — Scorsese's own motivation for making the film: "So I can get to know Jesus better."² He made the Christ he needed for his own spiritual resolution, as did Paul, as does the world.

See also: I'm God, and Yes — I AM Talking to You

¹See Stephen Heath, "Narrative Space," Screen, Autumn 1976, 19-75; reprinted in Narrative, Apparatus, Ideology, Ed. Philip Rosen (New York: Columbia University Press, 1986): 379-420.

²From David Thompson and Ian Christie, Eds. Scorsese on Scorsese. (London: Faber and Faber, 1989), 120.

The impossible POV

Yellow Sky: Not in the sights, exactly

There's a shot in Yellow Sky (1948) that immediately distinguishes itself from all others in the film, when we see the male lead (an outlaw named Stretch, played by Gregory Peck) through the barrel of a rifle being aimed by a woman (Mike, played by Anne Baxter) who hasn't yet made up her mind about pulling the trigger. The shot's thick rifle barrel border fills all but the very center of the screen with darkness, a strong contrast to the brightly lit exterior shots of sun-baked desert which dominate the rest of the film. But the most striking thing about this shot is that, for all purposes, it proclaims itself as a point-of-view shot when in fact it is not. The framing of Stretch at the end of a long tube implies an owned specularity, like a shot of the moon as seen through a telescope. We are tricked for a moment into believing that the owner of this view is the shooter, when in fact it belongs to an unseen and unseeing object: the bullet.

The peril of such a trick is that it threatens to expose the camera as the autonomous presenter of the film. While not every shot in a feature film is a point-of-view shot (most are not), when a shot is first disguised as a point-of-view shot, then revealed to be otherwise, there is a noticeable disruption, a perceptual double-take, that hints at the film's construction. This of course works to varying degrees, depending upon the shot and the audience, and by pointing this out I am not suggesting that all shots should conceal the camera's presence as much as possible or that stylized camera work should never be used (if it never were, there would be no Expressionism or film noir, to say the least). Depending upon the type of shot and its context, though, the artifice of such shots can become more noticeable than for others, for better or worse.

Great Expectations: Pip comes to

Another such deception is the simulated "coming to" shot, which shows a blurry figure coming into focus as a character regains consciousness. A perfect example of this can be seen in the David Lean-directed Great Expectations (1946). After visiting Abel Magwitch's death bed, Pip staggers home through London streets and passes out on his bed. The next shot begins in hazy darkness, then slowly brightens and focuses on the shape of Joe Gargery. This appears in every way to be the point-of-view of the recovering Pip, a trick set up by the previous shot, which showed Pip's authentic point-of-view as he pushed through his door and fell into bed. Once Joe is in complete focus, however, we see that he is not looking into Pip's eyes — that is, directly into the camera lens — but rather offscreen. The shot is thus revealed not as Pip's actual point-of-view, but as an ornate fade.

First Men in the Moon: Images captured by a nonexistent camera

I mentioned another example in an earlier post, from the 1964 film First Men in the Moon. At the end there is a scene in which people on Earth watch a television broadcast from the moon, which shows a trio of astronauts escaping from a collapsing underground lunar city. The television broadcast shows three shots from three different camera angles, though no character is present on the moon to operate a camera, let alone three.

Of the three listed here, the rifle barrel shot is probably the most disruptive, due to its obvious gimmickry. I don't know how this shot was achieved, but it screams "special effect." The effect of the simulated "coming to" shot is perhaps a wash if one considers the alternative — it may have been more jarring if Bernard Miles (the actor playing Joe) had looked directly into the lens. (Lean could have presented the awakening with the camera trained on Pip, however, or in a medium two-shot, e.g., to keep the camera more concealed; obviously he and cinematographer Guy Green found value in the simulated POV.) However, like the lunar TV broadcast, this method of suggesting character point-of-view has become common enough that perhaps it is not as noticeable to all viewers.

See also:
Shoot the moon...somehow
Scorsese's Enunciative Presence