Lush life
The 4-minute opening shot of Paul Thomas Anderson’s Boogie Nights (1997) arrives at the end of a long history of refinement of the technique of the long take in Hollywood. It forms a narrative celebration of the low-rent glamour of the world it introduces, a pointedly moving version of those Dutch still lives that celebrate plenty, one that pivots on its own virtuosity and the massive visual apparatus within its power. (Having rewatched it this week, I keep thinking that its long swinging zoom, tilt and pan away from the neon club name is an inversion of the “No Trespassing” sign and crane shot from the beginning of Citizen Kane: come on in, the water’s nice.) The camera shows us more of the club and its production design than we’ll see at any other point in the film; its drifting overlap between the central characters, ending with the gay porn glance between soon-to-be wunderkind Eddie Adams (Mark Wahlberg) and director Jack Horner (Burt Reynolds), is almost beside the point, as it often is in Anderson’s films. The camera movement is unhurried but its path is filled with colour and lateral motion to the point, as Kevin B Lee notes, of “sensory overload”, the unfolding of time a matter of immersion in narrative information that’s often no such thing, only the steady burst of affect. It’s an appropriate opening to a film in mocking love with the hubris of filmmaking and its technology: what’s the real difference between making a $15 million biopic of a fictional porn star and Horner’s declared ambition to make artistic narrative porn? The concept of the long take recurs within the narrative itself, when Eddie acts for the first time (as “Dirk Diggler”) and the camera magazine runs out halfway through his take, recalling the production of Warhol’s Sleep and Empire, in which Warhol simply inserted a new cartridge when the old one finished, never moving the camera.
So much for all that. By the end of the movie, porn is shot on grainy, poorly-lit video, with which “you can just keep shooting”. The coda tying up the narrative threads - Jack and a sober Dirk reconciled and making films again in the new medium, the other members of the cohort out of porn and realising their mild ambitions - feels tacked-on, a dream following the horrific psychic and material descent of the film’s second half. It’s as if the promise of the long take - its intoxicating attention, its splendour of showing - dissipates at the exact moment it becomes technologically most possible, with Anderson unable to find a convincing substitute at the endpoint of his film’s simulation of the late 1970s. In that sense the ease - “look, no hands!” - and tension - what purpose does it serve exactly? - of the long take in Boogie Nights suggests it occupies a weird tipping-point in film history. For the long take has come to occupy a fetishistic place in recent online cinephilia, reinforced by its reappearance in Academy Award fodder like Alejandro Iñárritu’s Birdman (2014), Sam Mendes’ 1917 (2019), Sebastian Schipper’s Victoria (2017) and Alfonso Cuarón’s Gravity (2013). As Brad Stevens notes, while directors and actors make much of the analogue difficulty of realising these shots, in most cases they’re blatantly digital edits of multiple takes, hardly distinguishable from the outright CGI simulacra of The Adventures of Tintin. But Stevens goes on to essentially restate the case that André Bazin made for the long take in the films of Jean Renoir, before its current iteration as a panacea for “the fast-cutting associated with post-70s Hollywood”: that it returns film to “the social world” and “redeem cinema’s illusory nature by bringing it into closer contact with ‘reality’”. In Bazin, this claim that had nothing so much to do with the technical difficulties of the long take but with its relationship to time, which the editing strategies of classical Hollywood had obfuscated.
The oddity here is that, for all that contemporary long takes revel in their technological power - one aspect of Boogie Nights’ opening taken to its logical extreme - the idea of realism persists precisely in its eclipse, as a privileged relationship to space, and especially to time, that exists solely in form. The single-take aesthetic of Birdman is the most acute, and most smug and putrid, example of these prematurely resolved contradictions: a film that claims to present a clear-eyed demystification of Hollywood - the down-on-his luck former superhero actor, a mentally ill drunk, tilting at the windmills of Serious Theatre - done in a single act of vision that connects the spaces of art and pokes fun at its ideals, as the novel once mocked the chivalric romances read by Don Quixote in his pie-plate armour. Into this deflated heroic continuity Iñárritu inserts pieces of obvious non-realism - MCU fight scenes, a night-to-day transition as Keaton sits outside - as if to say that this realism can even accommodate the lyrical and fantastical, time and space shifting on a whim even as, in other passages, the camera movement slows and places the viewer as the observer of a fluid, unnarrativised time. (I think of the fire scene in Tarkovsky’s Zerkalo as the exemplar of this in the history of the long take.) In this, Birdman’s single take could be seen as an - admittedly outré - example of the new middlebrow aesthetic Joe Kennedy has described as “authentocracy”, which savours certain dull pleasures through the act of artworks and their audiences declaring itself disabused, positioned as smarter than the self-deluded saps who might still claim to believe in modernism, the force of form or the history of revolution.
This is the danger that lurks in the élan with which Boogie Nights crowds its opening with lurid presence, straining at the limits of the representation of a historical reality it at once celebrates and turns into fable: a taboo on the negative, which persisted as an element of the long take in cinematic modernism, going back to the grinding concrete realities of Antonioni or the bare force of wind in trees in Straub and Huillet. In its absence, “realism” converges with the temporal extremes of digital technology: why bother with a mere 100 minute HDV when you can watch endless Youtube playlists of men reviewing Hardee’s burgers in their cars, 5-hour Twitch streams of 20-hour games, or the always-new always-the-same refresh cycle of Twitter, vast rivers of time that wash away the “cinematic” like marks in the sand? The result is something like the opening of 1917, which resembles nothing so much as a Youtube Let’s Play video: the delicate-but-not-too-delicate music, the over-the-shoulder travelling two-shot, the character-setting exposition disguised as dialogue, Colin Firth’s general appearing like an NPC to provide a mission, map function and starting items, the frontal shot of the square trench arranged with space at the corners like a HUD interface. As with many other things, the new middlebrow fate of the long take is anticipated in a killer’s reverie in Don DeLillo’s Point Omega (2010), watching Douglas Gordon’s 24 Hour Psycho (1994), as narrative is sundered from time:
The original movie had been slowed to a running time of twenty-four hours. What he was watching seemed pure film, pure time. The broad horror of the old gothic movie was subsumed in time. How long would he have to stand here, how many weeks of months, before the film’s time scheme absorbed his own, or had this already begun to happen? […]
It takes close attention to see what is happening in front of you. It takes work, pious effort, to see what you are looking at. He was mesmerized by this, the depths that were possible in the slowing of motion, the things to see, the depths of things so easy to miss in the shallow habit of seeing.
People now and then casting shadows on the screen.
He began to think of one thing’s relationship to another. This film had the same relationship to the original movie that the original movie had to real lived experience. This was the departure from the departure. The original movie was fiction, this was real.