California Split

Historical Markers of the Modern Soundtrack

Developed for RMIT Media Arts

Altman's technique

In the early/mid 70s director Robert Altman developed a technique for shooting film - recording sound which achieved a particular type of dramatic realism that has since typified his directorial style. This style is one that centres on the actors' performances which - through improvisation and interaction - drive the narrative; constructing scenes more out of the presentation of characters than the description of action. Of importance to this approach was allowing the performers to work with one another undistracted by the technical procedure of the location shoot (with cameras, recorders, continuity people hanging around, etc.).

As developed by location shooting through the late 60s/early 70s, it was the boom operator who would often be moving around maniacally trying to record everyone's dialogue as they moved around. The actor's would be able to move freely around the camera-tracking, but the boom operator often got in the way. To overcome this, Altman started exploring radio-mikes, which required no wires and only a transmitter-receiver set-up which could be located at a recording desk away from the main action and movement. These small radio-mikes could be concealed quite easily on the body of the performer, allowing the performer to move around in any way whatsoever with the knowledge that all their dialogue would be picked up by the sound recordist. Furthermore, the camera crew themselves could then move further away and zoom in on the action, thus prevent further technical interference.

Basically, a shoot could then involve, say, three cameras and six radio-mikes, recording a scene involving six characters in the one large space. The scene could be played out in its entirety, with all six actors' speech recorded simultaneously yet separately on a multi-track recorder which would be synced up to the three cameras, with each camera framing the action from a different perspective. The whole scene would thus be recorded as a temporal-spatial continuum capturing the precise interaction of intensity, rhythm and energy of the actors' performance as a live event. Later in post-production, shots could then be selected from the three perspectives and intercut accordingly, plus the volume levels of the synch-speech tracks could be individually manipulated.

Ramifications for film language

This approach to filmmaking effects a particular mode of film language. In conventional narrative approaches to filmmaking, the inherently deconstructive process of breaking down the live event into a multiple fracturing of time, space and perspective determines a reconstruction of the scene in terms of focus - ie. the film itself will provide our focal points in order to establish a relationship with the viewer that will allow him/her to read the significance within any scene. In Altman's approach, there is the potential to totally remove that kind of focus, in that the totality of the live event is fully recorded without the multiple fracturing.

Thus, in his post-production, Altman has consistently chosen to absent that kind of directorial focus within the scenes, leaving us to focus in on the scene ourselves and sort out the scene's significance without any readable cues. Technically, this means that : 1. the numerous long-shots leave us to focus on action within the frame; 2. the absence of camera-tracking leaves us to make up the flow and movement of the significant action; and 3. the sound-mix of all characters' speech levels at the same volume forces us to aurally focus on the voice of the `main' characters in amongst other 'minor' characters. (Of course, directorial presence still exists, but comparatively we have to do a lot more work in focusing in and on the film than we would normally do.)

Sound-image fusion

In conventional realistic drama, mise-en-scene is usually visually articulated through production design, art direction and set decoration. The combination of depth-of-field photography with mise-en-scene will present the totality of the visual, allowing our eyes to wander across the frame and digest the detail, while the levels and components of the scene's sound will be presented often in a chronological selective form, ie. background traffic might initially be introduced only to be dropped in volume to then allow the characters' voices to cover it. A paradigm can be struck thus: 1. the visual (within the frame) is total/static while 2. the aural (in the sound-mix) is fragmented/linear.

Altman reverses this sound-image fusion by always establishing a scene aurally in its totality, while the first image will be an extreme close-up of a particular detail within the scene. The camera will then slowly zoom out (rarely track, occasionally pan) to reveal the visual totality of what we have been acoustically digesting.

Close analysis: CALIFORNIA SPLIT

General sound design

This film of Altman's contains no non- or extra-diegetic sound, in the sense that all sounds occurring within the film actually have an acoustic spatial location within the film. Due to the recording technique, we can assume that every sound we hear is likely to have occurred in the scene we are watching.

Music `score'

The musical narrative of the film - those raspy boogie piano songs - only temporarily functions as non-diegetic. Two striking aspects of this music are:

1. They sound live, ie. they don't carry the aural texture of a studio recording which privileges silence, separation and a vacuum within which the recording takes place. Those boogie songs themselves contain lots of background noise, and have a strangely loose and unstructured feel about them.

2. We eventually see the location of that music and those songs - the piano player in Reno - which we then relate to our previous experience of the accumulative musical narrative (as the songs were interspersed throughout the film).

For these reasons, even the musical score functions diegetically. However, this is not to say that the songs are merely live on-location recordings. Consider how we 'discover' the source and location of the songs we have been hearing throughout the film. This discovery functions as a centralizing of Reno within the narrative, in that it is where Gould and Segal both come together to reach a peak and then depart and go their own ways. Reno is, in a sense, the climax of the plot, or rather the zenith of those two characters' relationship with one another. The sequence of those songs mirrors the narrative by leading us to a different kind of peak - that of fusing image back to the sound of the music.

Noise & silence

Throughout CALIFORNIA SPLIT, noise (as a multiple of sound levels) and silence (as either an absence, an isolation or a softening of those sounds) work to distinguish the dominant flows of plot and character development. Basically, when there is a lot of sound going on (as at the track, card game and poker halls) the narrative conveys a continuum of character and action. When there is a noticeable type of silence (as in Segal's many reflective moments) the narrative conveys a change of character and action. It is almost as if the noise (of their surroundings and their own babbling) keeps both Gould and Segal from breaking from their compulsion to gamble, while the sudden silences initiate a pause wherein they can reflect on things and make a decision to change their situation. (Well, Segal does but Gould never stops talking!)

Vocal performance

The acting style of Gould and Segal well suits the recording technique for Altman's film in that both actors are garrulous, continually cutting into and talking over one another. This vocal interaction would make the film virtually impossible to post-dub the voices, thus leaving them to make full use of the lip-syncing of the multi track recording.

Even the many non-actors in the film convey an extreme realism by their vocal delivery which is also recorded live in the impossibility of post-dubbing. Their hesitant, fluctuating and often mumbled delivery conveys a richness that could not be acoustically recreated or simulated. (Consider how Brando's mumbles still convey a sense of drama and theatre, or how sporadic mumbling in Italian films is usually impossible to post-dub.)