Of the many projects I work on, documentaries and documentary-style productions are my favorite. I find these often more entertaining and certainly more enlightening than many dramatic features and shows. It’s hard to beat reality. Documentaries present challenges for the editor, but in no other form does the editor play more of a role in shaping the final outcome. Many of them truly typify an editor’s function as the “writer” through shot selection and construction.
Structure and style
There are different ways you can build a documentary, but in the end, the objective is to end up with a film that tells an engaging story in such a way that the audience comprehends it. Structurally a documentary tends to take one of these forms:
- Interview sound bites completely tell the story
- The “voice of God” narrator guides you through
- The “slice of life” story, where the viewer is a hidden observer
- Re-enactments of events through acted scenes or readings, a la The Civil War or The Blues
- The filmmaker as a first person guide, such as Werner Herzog
Sometimes, the best approach is a combination of all of these. You may set out to have the complete story told only through assembled sound bites, yet the story is never fully fleshed out. There, pieces of scripted narration will help clarify the story and bind disparate elements and thoughts together.
Story arc and character
The persons on screen are real, but to the audience they are no less characters in a film than a role performed by a dramatic actor. As an editor, the way you select sound bites and put them together – and the order in which these are presented throughout the film – establish not only a story arc, but also perceived heroes and villains in the minds of the audience. Viewers want a film with a logical start, building tension and ultimate resolution. Even when there is no happy ending, the editor should strive to build a story that leaves the audience with some answers or conclusion.
Remember to balance out your characters. In many interview-based stories, the same questions are posed to the various interviewees as the interviews are conducted. This is helpful to the editor, because you can balance out the different on-camera appearances by mixing up whose response you choose to use. That way, the same subject isn’t always to go-to person and you aren’t heavy with any single person. Sometimes it’s best to have one person start a thought or a statement and then conclude with another, assuming the two segments are complementary.
Objectivity
This is one of the myths taught in some film and journalism schools. The truth is that almost every documentary (and often many news stories) are approached from the point-of-view and biases of the writer, producer, director and editor. You can try to portray all sides fairly, but the choice of who is interviewed or which bites are selected reflects an often subconscious bias of the person making that decision. It can also appear lopsided simply based on which subjects decided to participate.
Sometimes the effects are subtle and harmless, as in reality TV shows, where the aim is to tell the most entertaining story. In the other extreme, it can become borderline propaganda for the agenda of the filmmaker. I’m not telling you what type of film to make – just to be aware of the inevitable. If there’s a subjective point-of-view, then don’t try to hide it. Rather, make it clearly a personal statement so the audience isn’t tricked into believing the filmmakers gave a fair shake to all sides.
The art of the interview
If your documentary tale is built out of interview clips, then a lot of your time as an editor will go into organizing the material and playing with story structure. That is, editing and re-arranging sound bites in a way to tell a complete story without the need for a narrator. Often this requires that you assemble sound bites in a way that’s quite different from the way they were recorded in linear time.
Enter the “Frankenbite”. That’s a term editors apply to two types of sound bite construction: a) splicing together parts of two or more sound bite snippets to create a new, concise statement; or b) editing a word or phrase from another part of the interview to get the right inflection, such as making a statement sound like the end of a sentence, when in fact the original part was really in mid-thought.
Personally I have no problem with any of this, but draw the line at dishonesty. It’s very important to listen to the interviews in their entirety and make sure that the elements you are splicing together aren’t taken out of context. You don’t want to create the impression that what is being said is the exact opposite of what the speaker meant to say. The point of this slicing is to collapse time and get the point across succinctly without presenting a full and possibly rambling answer. Be true to the intent and you’ll be fine.
Typically such edits are covered by cutaway shots to hide the jump cut, though some director stylistically prefer to show the jump cut that such edits produce. This can give a certain interesting rhythm to the cut that might not otherwise be there. It also clearly tells the audience that an edit was made. It’s a stylistic approach, so pick a path and stick with it.
The beauty of the HDSLR revolution brought about by Canon is that it’s easier (and cheaper) than ever to field two-camera shoots. This is especially useful for documentary interviews. Often directors will set up two 5D or 7D cameras – one facing the subject and the other at an angle. This gives the editor two camera angles to cut with and it’s often possible to assemble edited sound bites using cuts between the two cameras at these edit points. This lets you splice together thoughts and still appear like a live switch in a TV show – totally seamless without an obvious jump cut. I’ve been able to build short shows this way working 100% from the interviews without a single cutaway shot and still have the end result appear to the audience as completely contiguous and coherent.
Mine the unrehearsed responses. Naturally that depends on the talent of the interviewer and how much her or she can get out of the interviewee. The best interviewers will warm up their subject first, go through the pro forma questions and then circle back for more genuine answers, once the interviewee is less nervous with the process. This is usually where you’ll get the better responses, so often the first half of the recording tends to be less useful. If the interviewer asks at the end, “Is there anything else you’d like to add?” – that’s where you frequently get the best answers, especially if the subject is someone who is interviewed a lot. Those folks are used to giving stock answers to all the standard questions. If their answers can be more freeform, then you’ll tend to get more unique and thoughtful points-of-view.
Organizing non-timecoded source material
Archival footage frequently used in documentaries comes from a variety of sources, such as old home movies (on various film formats), VHS tapes and more. Before you ever start editing from these, they should be transferred with the best possible quality to a mastering format, such as Digital Betacam (for NTSC or PAL), HDCAM/HDCAM-SR (for HD) or high-quality QuickTime files (DNxHD, ProRes or uncompressed).
The point is to get these to a format, which can be organized and tracked through stages of the edit. This usually means some format that allows timecode, reel numbers or other file name coding to make it easy to find if the project takes years to complete. Remember that timecode and a 4-digit reel (or source) number lets you find any single frame within 10,000 hours of footage. To make this material easier to use during the offline editing stage of the project, you may elect to make low-cost/low-res copies for editing. For example, DVCAM if on tape or ProRes Proxy or DNxHD 36 for files. Doing so means that timecode and source/reel info MUST correspond perfectly between the low-res and hi-res versions.
Your still photo strategy
Photography and artwork are the visual lifeblood of documentaries that lack supporting film or video content. Ken Burns has elevated the technique of camera moves on still images to an art form. Clearly he’s a filmmaker known to the general public as much for this effect branded by Apple after his name, as his award-winning films. Yet, the technique clearly predates him and has gone by many terms over the years. A company I once worked for frequently called it “pictography”. Regardless of origin – the use of stills requires two elements: organization and motion.
There are numerous photo and still image organizing and manipulation applications, including Adobe Lightroom, Bridge, Apple iPhoto and Aperture. Each of these provides a method to catalog, rate and sort the photos. You’ll need the application with a good manipulation toolset to properly crop, color correction and/or fix damaged images. Lightroom is my personal preference, but they all get the job done.
Moves on stills can be accomplished in several ways: animated moves in software, a computer-assisted, motion control camera stand or simply a human operator doing real cameras moves. Often the last method is the simplest, fastest and best looking. If that’s your choice, print large versions of the stills, put them on an easel and set up a video camera. Then record a variety of moves at different speeds, which will become source “video” for your edit session.
Another popular method is to separate components of the image into Photoshop layers. Then bring these into After Effects and design perspective moves in which the foreground elements move or grow at a different rate than the background layer. This method was popularized in The Kid Stays in the Picture. The trick to pulling this off successfully is that the Photoshop artist must fill in the background layer to replace the portion cut out for the foreground person or object. Otherwise you see a repeated section of the foreground image or possibly the cut-out area.
Edit system organization
There are plenty of tools at your disposal, regardless of whether you prefer Avid, FCP 7, FCP X or something else. If this project takes several years with several editors and a potpourri of formats, then Media Composer is a good bet; however, Final Cut also has its share of fans among documentary editors. Make liberal use of subclips and markers to keep yourself straight. Tools like Boris Soundbite (formerly Get) and Avid ScriptSync and PhraseFind are essential to the editors who embrace them.
I tend to not use transcripts as the basis for my edits. Nevertheless, having an electronic and/or paper transcript of interviews available to you (with general timecode locations) makes it easy to find alternatives. That can be as simple as having a copy open in Word on the same computer and using the Find function. My point is that modern tools make it very easy to tackle a wealth of content without getting buried by the footage.
The value of the finishing process
I feel that even more so than on dramatic features, documentaries benefit for high-quality finishing services. These range from simple online editing to format conversion to color grading. Since original sources often vary so widely in quality, it’s important to get the polish that a trained online/finishing editor and/or colorist can provide. Same for audio. Use the services of talented sound designers, editors and mixers to bring the mix up a notch. Nothing screams “bad”, like a substandard soundtrack, no matter how striking the images are.
Clearances
It is important for the editor is to keep track of the sources and usage for stock images and music. These aren’t free. Many documentary producers seem to feel they can “sweet-talk” the rights holder into donating content out of a sense of interest or altruism. That’s almost never successful. So understand the licensing issues and be wary of using images and music – even on a temporary basis – that you know will be hard to clear or too expensive to purchase.
Make sure that you have an adequate system for tracking and reporting the use of stock material, so that it can be properly bought and cleared when the film is being finished. During the rough cut, stock footage and images will usually be low-res versions with a “burn-in” or watermark. When the time comes to purchase the final high-res images, most companies require that you request the exact range of the material used based on timecode. That material will be provided as files or on tape, but there’s no guarantee that the timecode will match. Be prepared to eye-match each shot if that’s the case.
©2011 Oliver Peters
A marker list can be exported as a tab delimited text document, which will open in most word processors or spreadsheet applications. This list will indicate comments, type, duration, color and timecode. Marker lists may be helpful as a reference for interviews in the absence of an actual transcript. Or they may be used to document notes for editorial changes.
Furthermore, the Find command (cmd-F) for the sequence lets you search by names and markers. Looking for a word you know you typed into a marker? Simply use Find in the sequence and it will take you to that spot. Of course, the more information you typed at the time, the better of a resource this becomes.
Get operates as a standalone application and is therefore independent of any written script or transcript. Think of it as a Spotlight-style search tool that is optimized for speech. Get can be installed for various languages and currently supports US English, UK English and Latin American Spanish. For now, Final Cut Pro is the only supported NLE, although Get can be used without FCP, as well.
After the initial indexing is completed, the rest of it is pretty simple. Typically you perform searches within Get’s user interface. Results will be grouped using various criteria, but normally Get will group search results according to a confidence indication. Search results that are listed as Very Confident will generally always match the search word or phrase, whereas Less Confident might not. One way to dial in search results is to use the Score Threshold pulldown menu. For example, if you search for a term and select a Score > 95 value, Get will return far fewer successful results than if you use a value of Score > 55. Lowering the score will achieve more total results, but also less accuracy.
Get’s Export routine is used to bring clips into a Final Cut Pro project. Individual clips or a group of clips can be exported from the Get interface into any existing FCP project. Get has the ability to access the project, so either clips can be sent directly into existing bins or new FCP project bins containing these search results can be created and exported right from within Get. When you return to the FCP project, the appropriate folder and clips will have been added. Each clip will also contain markers to identify the location of the search term. In addition, Get lets you export only markers, in the case where you are analyzing clips already in FCP.
I performed testing with an existing series of corporate videos produced about a family-owned Australian winery. The original audio was recorded on a Canon EOS 5D Mark II. Some audio was rather low and, of course, everyone spoke with an Australian accent. Get delivered good results using the US English installation, in spite of these challenges. Even specific Australian words, like less-common town names, were easily found. Without question, I’d have to say Get definitely works as advertised.
An NLE is nothing more than a large database. As such, there are many built-in tools to help you sort, search and find the necessary shots. The most obvious is a finder-style column sort. Highlight the column header and sort by ascending or descending values. Most do a single-column sort, but Avid Media Composer actually does a two-column sort. So for example, Scene might be a “primary” sort field and Take might be the “secondary”. Another fine Media Composer bin feature is Custom Sift. Set the criteria for one or more columns and Custom Sift will display the matches and hide the rest. When cutting a feature, I’ll have a Selects column and indicate preferred takes with an “X”. By sifting for any “X” in the Selects column, the bin will only display the few preferred takes instead of all the takes. Switch back to an un-sifted view and the bin shows all clips again.
Step three is to organize the sequences of people into new sequences by topic. As you’ve been listening to the comments, several themes will start to emerge. These may be predetermined – based on a set of questions that the interviewer was using – or it might come out of the natural on-camera discussion. Don’t get rid of your first set of selected sequences by person, in case you need to refer back to one of them. Depending on your NLE and style, build these by copy-and-pasting clips or by editing from one sequence into another.
As a sidebar, one interesting approach to all of this can be found at 
