How to watch a documentary critically: questions for the credits
I stay for the credits. Partly out of respect for the sound mixer and the colour grader, who rarely get applause, but mostly because the last few minutes of a documentary are often its most candid. That is where you learn who paid for the film, which broadcaster commissioned it, which foundation sits under "in association with", and whose archive supplied the footage that moved you forty minutes earlier.
A documentary is an argument built from real material. That does not make it a lie, and the habit I want to describe has nothing to do with the cynicism that treats every non-fiction film as propaganda. It is closer to the way a good editor reads a manuscript: with pleasure, with attention, and with a pencil in hand. Below are the questions I bring into a screening, roughly in the order they arrive, followed by a checklist short enough to carry home.
Start with who made it
Before there is a film there is a commission. A Canadian feature documentary is usually financed through some mix of broadcast licence fees, public funds and tax credits, with foundation or streaming money sometimes added on top. None of that is sinister. Each source brings expectations, though. A broadcaster wants a set running time and a certain kind of ending, a foundation wants its cause to look urgent, and a streamer wants you still watching after ten minutes.
So read the logos at the start and the cards at the end. Ask whether any funder has a stake in the subject. A film about a pipeline, a drug company, a political movement or a charity's work overseas is not disqualified because an interested party helped pay for it. You should still know, and a confident film will tell you without being asked. When the money is hard to trace, I treat that as information too.
Ask how the camera got in
Access is the hidden currency of documentary. Someone opened a door: a family, a police service, an armed group, a hospital, a ministry. Doors are seldom opened for nothing. The price may be time, trust, a promise to leave certain things out, or the quiet understanding that the crew will be welcome again only if the film is fair in the host's sense of the word.
Films made inside closed organizations are the hardest case. A crew embedded with a militant movement or a government may record things no outsider has seen, and that has real value, though it may also record precisely what its hosts wanted recorded. The film note on Hamas: Behind the Mask works through this problem with care, and I would read it before watching any access film about a political organization. The question itself is plain. What could the filmmakers not show, and did they tell you?
Look for what is missing
Every frame excludes something. What matters is whether the exclusions decide the conclusion for you. Who is interviewed, and who is only discussed? If a film about a factory closure hears from workers and economists but never from the company, perhaps the company refused, and a careful film says so on a title card. If it never mentions asking, you are watching half a conversation.
Absence works differently in films about violence, where the people at the centre often cannot appear because appearing would put them in danger. Crimes of Honour belongs to that difficult family, films about women who have been killed, hidden or made to speak at great personal risk. There, absence is the subject, and an honest film builds around it. Elsewhere it is a convenience. Learning to tell those two kinds of gap apart is most of the work.
Listen to who is talking
Narration is the film telling you what to think, and it is worth asking whose voice that is. The unseen narrator of older television documentary, calm and placeless, borrows an authority nobody granted. A filmmaker who speaks in the first person and appears on camera is at least visible, which makes their stake easier to judge. Neither style is more honest by default.
Pay attention to experts as well. A historian, a doctor or a former official in a well-lit office lends weight to whatever comes before and after the cut. Ask what they were actually asked, whether they speak to the precise claim the film is making, and whether anyone with a different reading of the same evidence was given a chair.
Notice reconstruction, and whether you were told
Staging is as old as the form. Robert Flaherty arranged much of what we see in Nanook of the North (1922), and documentary history since then has been, among other things, a long negotiation over how much arranging an audience will tolerate. Errol Morris turned re-enactment into an open style in The Thin Blue Line (1988). Nobody mistakes his stylized reconstructions for found footage, and that clarity is its own kind of honesty.
Trouble starts when reconstruction passes for observation. Watch for scenes that are too well covered: a private argument filmed from two angles in a room that clearly held one camera, or a subject "discovering" a letter while the lens waits in exactly the right spot. Small set-ups are ordinary practice, such as asking someone to walk through a door a second time. Re-creating an event and presenting it as if it unfolded in front of the camera is something else, and it ought to be labelled.
Turn the music down in your head
Music is the strongest and least examined tool in non-fiction. A low string drone under an interview tells you to distrust the speaker. A soft piano under the same face tells you to pity them. Neither instruction comes from the footage.
Try an experiment once, deliberately: during a scene that moves you, imagine it with only the room sound. Sometimes the scene holds, because the feeling was in the faces and the words. Sometimes it falls flat, and you realize the composer was doing the persuading. I have nothing against scores. I object to scores that make a case the material cannot make on its own.
Check the archive's papers
Archive footage carries an aura of truth because it is old and grainy and somebody else shot it. Yet archive is often used as illustration rather than evidence. A crowd from one protest stands in for another. A street from one decade dresses a story set in the next. Generic shots of soldiers, refugees or a skyline become wallpaper for whatever the narrator happens to be saying.
Good films identify their archive, with dates and places on screen or at least in the credits. When a film slides between archive and new material without signposts, ask whether it matters. Often it does not. When the archive is being used to prove something, it matters a great deal.
Remember that the edit is an argument
Far more is filmed than ever reaches the screen, and every cut is a decision about meaning. The reaction shot is the classic suspect: a face listening, a cut, and suddenly that face seems to respond to something it never heard. In interviews the related trick is the "Frankenbite", a sentence stitched together from separate answers. Reality television made it notorious. It can creep into documentary too.
You cannot see most edits, but you can hear some. Listen for breaths that do not match, for room tone that changes mid-sentence, for an answer that seems to begin before the question has quite ended. Pay attention to time as well. A film that covers three years in ninety minutes has chosen which months count, and the shape of those choices is its thesis.
Ask who benefits
This is the last question and the least comfortable. Who gains if you believe this film: the subjects, the filmmakers, a funder, a cause, a government, an industry? Benefit is not proof of bad faith. Most films that change minds benefit somebody, often people who deserve it. A film that lands exactly where its sponsors would want, without once showing you a fact that complicates the case, has earned a second viewing with the lights half up.
Then ask who carries the risk. A person who speaks on camera about a militia, a violent family or a corrupt employer may pay for that film long after the audience has forgotten it. In my experience, films that are careful with the people they expose tend to be careful with the truth as well. The two habits travel together.
A checklist for the walk home
I keep a version of this list in the back of my notebook. It is not a scoring system, and a great film can fail one or two items and stay great.
- Money: who commissioned and funded the film, and does any funder have a stake in the subject?
- Access: who let the camera in, on what terms, and does the film admit the limits of that access?
- Voices: who speaks, who is spoken about, and who declined to take part?
- Authority: whose voice narrates, and do the experts address the exact claim being made?
- Staging: are reconstructions labelled, and did any scene look suspiciously well covered?
- Music: would the key scenes survive without the score?
- Archive: is older footage identified by date and place, and is it evidence or decoration?
- Edit: do the interview answers sound intact, and how was time compressed?
- Benefit and risk: who gains if you believe the film, and who pays for having appeared in it?
Where to practise
The habit grows with use, and the cheapest place to practise in Canada is the National Film Board's online collection at nfb.ca, where much of the catalogue streams free. Choose two films on the same subject made decades apart and run the list on both. You will see how the rules of narration, music and access have shifted, and how differently each period decided what an audience needed to be told. For titles that reward this kind of attention, start with the film notes. Then, the next time the house lights come up, stay in your seat and read the money.
