I love this story Howard Hawks told in an interview about the making of The Big Sleep:

“Jack Warner said, ‘I got such a great reaction from people, Howard. You ought to have more scenes with those two people.’ I said, ‘You talk to them. If they can’t behave themselves, if they have to get mushy all the time, I’m not going to stick around and make scenes with them.’”

Hawks, one of cinema’s most inconspicuous poets, talks about “making scenes” as if it were playing baseball or going fishing. To say he had a gift for making scenes would be an understatement. He could do it all day any day, making him the perfect director for the Hollywood system. If the censors wanted something cut, he cut it. If Jack Warner wanted more scenes, he added them. But his willingness to play ball with the powers that be in Hollywood was no compromise to his artistry. On the contrary, it inspired some of his grandest, most inventive gestures and a whole concept of cinema that was entirely his own: a cinema of moments.

Watch the two versions of The Big Sleep, the pre-release version made in 1944 and then the final version revised and released in 1946. In the pre-release version, that is, the one that inspired Warner to ask for more scenes, the movie works as a coherent whole. There are a few essential plot points missing from Raymond Chandler’s novel (such as references to the pornography business or “dirty picture racket” in Los Angeles), but for the most part we can follow the story with a logical continuity from scene to scene. The final version—18 minutes shorter—is a dream, a miracle, and something of a mystery. It shouldn’t work. It has no foundation in the sturdy narrative structuring of most Hollywood movies, showing us that plot and story are different things. It’s a display of Howard Hawks’s brilliance as a film grammatician and classifies him as one of cinema’s most reluctant theorists, comparable to Eisenstein, Godard or Tarkovsky in his insights on the medium as a practitioner rather than a viewer.

In making a film where narrative depth flattens into a series of scenes and sequences, Hawks shows us that cinema can be an incomprehensible art, an abstraction made of realities rendered and spliced together with a continuity of feeling rather than logic. A man opens a door only to be immediately shot down. Why? We hardly have a chance to ask before we’re sucked into the most cinematic of set-ups: the chase. The movie can go from the chaotic din of rapid gunfire to the eeriness of total silence. Humphrey Bogart is staking out a house, he hears a scream and a crash and runs into the house. The movie goes silent. We barely hear his footsteps. While Max Steiner’s score orates Bogey’s thoughts for most of the film, it hauntingly vanishes when he enters the house. It’s as if Bogart has, quite literally, lost his mind. He finds a woman, a body, a gun, a camera hidden in a sculpture. He’s lost and we’re lost. The moment lingers, illustrating Godard’s now paradigmatic quote that all you need to make a movie is “a girl and a gun.”

In this sense, The Big Sleep is a film of the 1940s made for the 1960s (when audiences and critics finally recognized Howard Hawks as a great artist). Its characters are characters, its sets are sets, its script is a script. Everything is artificial, veiled by a thick pea-soup backlot fog, signifying only itself. Hawks keeps his camera on the surface and the movie becomes a series of surfaces. He hardly cuts in for a point-of-view shot; instead, he sets his camera on the act of watching, the act of detecting. It is about the surface of watching, not the depth of seeing. It is about the state of waking sleep that overcomes the spectator in a theater.

We can see the hand of William Faulkner (who worked on the script with Hawks) in instances of sound and fury where the mundane takes on a magical incomprehensibility. But a famous bit of innuendo about horse racing between Bogey and Bacall (“a lot depends on who’s in the saddle”) was one Hawks penned himself, upon Warner’s behest. Certainly no one talks like this—that’s much of the charm of old movies—but few people even write like this. We are aware of the script in such strains of dialogue, not because it is inadequate but because it is so noticeably (and so damn well) written. Typically studios or directors add scenes in the service of guiding the viewer through the narrative more gently, often at the expense of thematic ambiguity. Hawks, however, adds scenes and the film becomes a narrative rollercoaster. It’s a confusing movie and most people can’t tell you what has actually happened after seeing it just once. If it works, and the general consensus by critics and laymen alike is that it does, Hawks’s comment on “making scenes” with Bogart and Bacall offers a valuable insight into his understanding of cinema. For Hawks, the scene, with all the intangible richness of a transient moment, is the dramatic and structural centerpiece of his films.

The film’s entertainment value, recognized on its release, hasn’t faded, but now we can see its place in history and the moment to which it was responding. Although we can read Hawks’ confusion at his own present (the chaos of war, the influx of media, and the dawn of the postmodern nightmare) the movie is a work of aesthetics, not of politics. That’s why Hawks could make movies about detectives, cowboys and gangsters as well as chemists, journalists and debutantes. He finds continuity in characters who live invisible pasts with their future prospects out of sight. But they are not vessels, not the vacant watchers of Hitchcock nor the architectural statues of Antonioni. They live and breath on moments, nods, winks, walks, chases; all woven together in a thread for detective Philip Marlowe to navigate as the arbiter of facts and decipherer of madness.