What You Need for Transcripts

• Timecode—Transcripts should include timecode (hours, minutes, seconds) at the beginning of each interview and thereafter, every 30 seconds or so, the numbers interpolated after a convenient period and between sentences or paragraphs.
• Accurate transcriptions—These should be made verbatim with any ums, wells, you knows, and all the other verbal curlicues that people use as place markers while their mind assembles what they want to say next. Punctuation should follow the way a person speaks and never be adapted to make his presumed intentions clearer and more literary. The temptation to edit and clarify as you transcribe is ever present, but resist, because it makes extra work for the film editor later. Start a new paragraph when the speaker plainly starts a new thought; otherwise, keep typing in long paragraphs. Remember, it’s not literature and you don’t lay it out for easy reading. It’s a record of how someone spoke.

FIGURE 31-1 Flow chart illustrating how to turn transcripts into an assembly script.

• Line numbering—Word processing has a line-numbering feature (see box) used in legal documents that I could have used for an oral history film I once made. It had over 20 characters telling the convoluted story of the World War I antiwar conscientious objector movement. Had each interview transcript been line numbered, I could have known where even single lines of cut-up transcript came from—a lifesaver on a really big project.
• Using a database—For a film series drawing on a huge base of transcript or archive material, you might want to develop a resource like the one I am using for a large-scale biography. Using an all-purpose database program such as FileMaker® Pro, you can customize it so that, once materials are filed, you can search with incredible speed by key words, time period, or character—or whatever else you nominate as parameters. You can also display the material in different formats (lists, as cards, etc.) and by various priorities such as time period or characters in alphabetical order. Putting the data in the system does take time and effort, but later it hands you a huge information bank that you can use with efficiency and total thoroughness. If you’re working with a modest amount of source material, you may get by with yellow highlighter and Post-It® notes, or you can cut copies and paste them according to what follows.

Selecting the Materials

• Step A. From your typed-up transcript file (Frame 1) make two prints—one we shall designate the Master Transcript and the other the 1st Copy (Frame 2). Amass master copies in a binder titled Original Transcripts, and keep them apart. You are going to mark up the first copy of the master.
• Step B. Run the film interview on a DVD or editing machine, and follow the speakers’ words in the 1st Copy. When a section strikes you as effective—for whatever reason—put a rough vertical “preference mark” in the margin, as in Frame 3. At this stage just respond and don’t stop to analyze your reactions. There will be a time for that later. Some preferences will be for a story graphically told, others for some well-presented factual information, and still others will be in response to an intimate or emotional moment, be it humor, warmth, anger, or regret.
• Step C. Study each preferred section and find its logical in point and out point, using square brackets to indicate the start and finish of the section, as in Frame 4. Figure 31-2 shows an enlarged view of a typical chosen section with its in and out points bracketed. In the margin is written “6/Ted/3.” Decoded, this media ID means “Cassette 6, Ted Williams, section 3.” Use whatever format you like for section identities, but as with all filing systems stick with one system for the duration of the project and make improvements only when you start a new project.
  

  FIGURE 31-2 A selected section showing in and out brackets, media origin ID, and tag description.
  

Until you replay the selected section you won’t know whether you can execute your in and out points as marked. Maybe you can’t lose all those verbal warm-ups before the in point; maybe your out point leaves Ted on a rising voice inflection so his statement sounds strange and unfinished.

Section IDs will always allow you to quickly locate the section in the Marked Up Transcripts volume and find it in its parent take. Without some such method, finding where material originated in order to extend it can waste inordinate time during a heavily edited show. You can also of course use a computer search function on the transcript files or search via timecode in the edited version, but searches always take time and effort.

Figure 31-2 has one more notable feature: the handwritten tag description of the section’s use, “Ted’s descript. of Farmer Wills.” When you face dozens of slips of paper, tag descriptions allow you to initially sort and structure them.

• Step D. Make a photocopy of the marked sheets (Frame 5). You can now file the parent transcript copy (Frame 4) in shooting order in a binder titled Marked-Up Transcripts. Make an index at the front so you can quickly locate each character. Later, during editing, this file will be an important resource.
• Step E. Scissor the marked sections (Frame 5) into selected sections (Frame 6), ready for sorting and grouping in pursuit of the paper edit. Because you’ve identified each slip by subject or intended use and because it carries an ID, you know at a glance what it is, what it can do, and what its context is in the parent sheets, the Marked-Up Transcripts.

Now you are ready to construct a paper edit—really, a detailed sketch for the first assembly. The selected sections (Frame 6) will eventually be stapled to sheets of paper (Frame 7) as sequences, and the sequences will be assembled into a binder (Frame 8) holding the paper edit or plan for the first assembly of your movie.

This procedure may seem laborious, but if you are working on an oral history film, say, with a vast bank of spoken material, time spent organizing at the outset (indexes, graphics, guides, color coding, and so on) is effort generously rewarded later. I learned this when hired to edit the final game in the 1966 soccer World Cup documentary, Goal. There were 70,000 feet of mostly silent 35 mm film (that’s nearly 13 hours) shot from 17 camera positions. The only orientation was a shot of a clock at the beginning of each 1000-foot roll. My assistant Robert Giles and I spent a week up to our armpits in film, making a diagram of the stadium and coding each major event as it appeared in all the various angles. The game had gone into overtime due to a foul, and it was my luck to eventually establish that, during this most decisive moment, not one of the 17 cameras was rolling! Then, using sports reports and Robert’s far superior grasp of the game, we set about using an assortment of close shots to manufacture a facsimile of the missing foul. No one ever guessed we’d had to fake it.

The project was an editor’s nightmare. Had we not first taken the time to invent a decent retrieval system, the men in white coats would have carted us off in the legendary rubber bus.

Assembling the Paper Edit

To avoid your film turning into solid speech, first deal with your action sequences:

• Step F. Prepare to make a rough assembly of action sequences (those that show human processes with a beginning, middle, and end). Make a list of them and design an overall structure that moves them logically through time. Be conservative with your first structure. If you have a film about a rural girl going to the big city to become a college student, stay true at first to the chronology of events (as opposed to the events unfolding according to their importance, say). Afterwards, when you can better see how the mate rial plays, you might intercut her high school graduation, conversations with a teacher, discussions with her father, and leaving home with the development of her first semester as a college drama student. You would then have two parallel stories to tell, one in the present and one in the past. It is far too risky at this stage to assume you can do this in one giant step.
• Step G. View the action or behavioral material in time sequence more than once to establish its narrative possibilities firmly in your mind. Now you can go on to plan a safe, linear version, beginning from a paper edit that includes speech.
• Step H. Take up your chosen transcript sections, as discussed earlier in this chapter (see Figures 31-1 and 31-2. These slips of paper, coded “Jane” for Jane’s sequences so you can turn rapidly to parent copies in the marked-up transcripts, might look something like this:
  

1/Jane/1 Graduation speech

1/Jane/2 Dinner with boyfriend’s family

1/Jane/3 Conversation with English teacher

2/Jane/1 Conversation with Dad

You have the many pieces of action made into a preliminary assembly, ready to accommodate the dialogue sections of the film. It’s best to represent these as sequences rather than as individual shots, which would be too detailed and cumbersome. Three sequences would go on separate slips of paper, each with a media location (cassette or P2 card number, minutes, and seconds):

1/05:30 Exterior school, cars arriving

2/09:11 Preparations at podium, Jane rehearsing alone

4/17:38 Airport, Jane looking for bus

Kneeling on the floor, move the slips of paper around to try different orders and juxtapositions. Certain pieces of interview or conversational exchange belong with certain pieces of action, either because the location is the same or because one comments on the other. This comment may be literally a spoken comment or, better, implied through ironic juxtaposition (of action or speech) that makes its own point in the viewer’s mind.

An example would be a scene in which our student Jane faces having to make a graduation speech before the whole school, a prospect that scares her. To make a literal comment, one would simply intercut the scene with the interview shot later in which she confesses how nervous she feels. A nonliteral comment might take the same rehearsal and intercut her mother saying how calm and confident she usually is. A visual comment during the rehearsal might show that she is flustered when the microphone is the wrong height and that her hands shake when she turns the pages of her speech.

What’s the difference? The literal comment is show and tell because it merely illustrates what Jane’s thoughts and feelings already give us. The nonliteral comment is more interesting because it supplies us with conflicting information. Her mother rather enviously thinks her daughter can handle anything, but we notice signs that the girl is under a lot of strain. Either the mother is overrating the girl’s confidence or she is out of touch with her child’s inner life. This privileged insight, discreetly shared with the audience, gives us behavioral evidence that all is not well, that the girl is suffering, and alerts us to scrutinize the family dynamics more carefully.

The order and juxtaposition of material, we can say, have potent consequences. The way you eventually present and use the material signals your ideas about the people and the subject you are profiling and reveals how you intend to relate to your audience. In essence, you are like a lawyer juxtaposing mismatching pieces of evidence in order to stimulate the interest and involvement of the jury, your audience. Good evidentiary juxtaposition provides sharp impressions and removes the need to explain.

The mobility and flexibility of the paper edit system reveals initial possibilities and gets you thinking. You’ve moved your slips of paper around on a table or floor like the raw materials for a mosaic. Each juxtaposition gives you ideas and sets you thinking, just as you might from looking at the potent relationships between chess pieces laid out on a board.

Don’t worry if your paper edit is vastly too long and repetitious. This is normal; you can only make refined decisions from experimentally intercutting and screening the material. Once you’ve decided a reasonably logical order for the chosen materials, staple the slips of paper to whole, consecutively numbered sheets and place them in your paper edit binder.

Now that you have a rudimentary story, you can rule lines between sequences and group the sequences into scenes and acts. From this master plan you can begin making a loose, exploratory assembly in your editing computer.