For a fiction film, the story is written in advance of the shooting. In my documentaries the story is found in the editing. I have no idea before the shooting begins what the events, themes, ideas, or point of view of the film will be. This last statement is obvious, since before the six to twelve weeks of shooting I do not know what events will occur, what I will decide to shoot, or what words, gestures, emotions, and actions I will, by chance, find. If I knew all that, the film would not be worth making because there would be no surprises.
The purpose of the filming is to accumulate scenes, material from which a film can be edited. During the shooting I simply try to gather sequences that interest me for whatever reason—i.e., they are funny, sad, tragic; they reveal an aspect of character, illustrate an aspect of the division and exercise of power, point out the gap between ideology and practice, or show the work of the various professions, clients, or publics represented. The decision about what to shoot is always based on a shifting combination of judgment, instinct, and luck. After six to twelve weeks, I typically have eighty to a hundred and twenty hours of film from which a film has to be edited.
I begin the editing once the picture and sound are synchronized and a log is organized, summarizing and classifying each shot. The first thing I do is look at all the material and make an initial evaluation. I use a classification system based on the Guide Michelin: one, two, or three stars. Approximately sixty percent of the sequences are temporarily rejected during this first assessment. I know that much of the editing will be boring, so I start with sequences I particularly like. Often an event in real time will last for, say, forty-five minutes, of which forty-three minutes may be recorded on film. If I want to consider including the sequence, I have to reduce the forty-three minutes to a usable form without distorting what I take to be the purpose and meaning of the actual encounter. It is rare that I will find six consecutive minutes that I can directly extract. I have to select the moments (sometimes seconds, occasionally minutes) that I want to use and find a way of combining these separate shots into a form that respects the original event but seems as if it took place as it appears in the edited form. This means that the shots I choose, besides conveying the sense of the event, have to work together in film terms: i.e., have a coherence and rhythm in the selection, duration, and ordering of the shots chosen.
Over a period of five to six months I will edit in this way all the sequences that I think are candidates for inclusion in the final film. Until this process is completed, I do not formally think about the structure. As I go along, however, I begin to have the occasional thought about ways of relating individual sequences. I make notes about these possible combinations without trying them out. When I have edited all the possible sequences into a provisionally usable form, I begin work on the structure. I assemble the first draft of the final film in three or four days. I can do this relatively quickly because at this point in the editing I think I know the material quite well. However, it is always necessary throughout the editing to try to be constantly aware of the vast possibilities of self-deception and the vagaries of a process that lends itself quite easily to manic-depressive swings.
This first version is usually thirty to forty minutes longer than the final film. At this point in the process I begin to pay more attention to the rhythm of the film, the internal rhythm within a sequence and the external rhythm between the sequences. For example, a sequence as originally edited may have a beginning, a middle, and an end. When it is placed in relation to other sequences, the beginning may no longer be necessary because the same information (about character, physical location, or time) may have been suggested in a more appropriate form in another scene. The external rhythm is related to the shots that link the major sequences: that is, it may be necessary to have a minute of relative quiet after a very emotional scene, or several shots will need to be linked to suggest the passage of time or a change in location. The choice of shot, the direction of the movement within the shot, the time of day, the information conveyed by the people or objects—all these have to be evaluated both in relation to each other and to the sequences that come before and after. This, of course, is true for the internal as well as external editing of a sequence. Each sequence or group of related sequences has to be assessed in this way and it is also necessary to know the overall structural connection between all the sequences in the film (for example, the relationship between the first ten minutes of the film and the end). I have learned over the years to pay as much attention to the thoughts at the edge of my head—my associations to the material I am watching and hearing—as I do to the more overtly logical and deductive aspects of making editorial choices. Following these seemingly peripheral intuitive thoughts can lead to more startling and original combination of sequences with unanticipated benefits for the content, form, and structure of the film.
After I think I have completed the editing, I go back and look at all the rejected rushes. I often find solutions to editorial problems I had not resolved. I may find a better transition or a short scene that more successfully introduces a theme or person. These discoveries are made because at this, the final stage of editing, I have a much clearer idea of what I am looking for.
The editing is finally finished when I go through the film and try to explain to myself why each shot and sequence is in the film. I have to express in words both my rational and non-rational decisions. Since I like talking to myself, this is the last pleasure of the editing process.
Frederick Wiseman’s most recent documentary is State Legislature. All of his films have been released on DVD and are available at www.zipporah.com.
