Pitch a Verbatim Film Without Treating Recorded Words as a Finished Scene

Explain what stays verbatim, then explain what you construct around it. A film may retain a speaker's words while choosing a new body, room, rhythm, sequence, or image through which the audience encounters them. Those choices belong in the pitch because they are part of the film—not decoration around an untouched source.

“Every word is real” leaves several questions unanswered. Are we hearing the original recording or an actor? Is this a complete answer or a phrase extracted from it? Was the question removed? Does the image show the event being described, a later performance, or an association proposed by the filmmaker?

The useful promise is specific: what source material the project has, what part it retains, how it will be performed, and where the film adds an interpretation.

First, define the material you are retaining

Start with the source unit. That might be an entire exchange in an interview, a passage from a hearing transcript, a recorded voice message, or selected words drawn from several occasions. These are not interchangeable claims.

For example, “actors perform selected passages from the supplied interview transcripts” is different from “actors lip-sync to the original interview audio.” The first preserves a textual selection while creating a new vocal performance. The second retains recorded voices, including whatever timing remains in the selected audio, while creating a new visible performance. Either approach still requires decisions about selection and presentation.

Clio Barnard's account of making The Arbor makes those decisions concrete. She told BAFTA that she first shaped an audio screenplay with editor Daniel Goddard, then worked on the picture edit with Nick Fenton. The structure changed between those stages. She also described actors receiving playback through earpieces and miming rather than speaking the words aloud.[^1]

That is a maker's description of a method, not an assertion that this article has compared her source recordings against the finished film. Its value here is narrower: retained recorded voices and an authored screen structure are compatible, and the pitch can explain both.

For your project, name the actual materials held. Do not write “drawn from the original recordings” when you have only seen an extract quoted elsewhere. A transcript may be sufficient for the particular performance you propose, but it does not establish the voice, pauses, or sounds of a recording you have not obtained.

Also decide what “unchanged” means. It could refer to words within each excerpt, their order within an answer, their order across the whole interview, or the actual audio samples. A single broad claim should not stand in for all four.

Look at the surrounding exchange before choosing the line

Here is an original fictional exchange for a film about a repair workshop. The people, interview, disruption, and transcript are invented. There is no source audio, and the staging variants below have not been filmed.

Interviewer: What happened on the evening shift when the delivery was late?

Manager: We kept the lights on. Just in the front room, until the driver arrived. The repair benches were shut down as usual.

Interviewer: Was the workshop in danger of closing?

Manager: No. The driver had the wrong address. We waited about twenty minutes, took the boxes in, and went home.

Interviewer: Who stayed?

Manager: Me and Tessa. She checked the delivery against the order while I locked up.

The sentence “We kept the lights on” can refer here to an ordinary action during a short delay. The next sentence limits the place. The answer after that explicitly rejects a larger closure story.

Now extract only:

We kept the lights on.

Every word comes from the invented transcript. Yet the excerpt alone no longer supplies the limits that made the phrase clear. We have not changed the words, but we have changed what information accompanies them.

This does not mean the short excerpt can never be used. Perhaps the surrounding scene supplies the front room, the delivery, and the twenty-minute wait. Perhaps the film explicitly uses the sentence as a metaphor and makes that appropriation visible. The question is what the chosen construction asks the viewer to infer.

A more serious alteration would join separate fragments:

We kept the lights on. The workshop in danger of closing. Me and Tessa.

The middle fragment comes from the interviewer's question, not the manager's account, and the denial has disappeared. Even though the words occur somewhere in the exchange, the new sequence cannot responsibly be described as the manager's uninterrupted testimony.

Keep speaker attribution, question context, and omissions in your working record. When a selection changes who appears to be asserting something, a word-for-word check has reached the limit of what it can tell you.

One sentence, two proposed scenes

Let us keep the same five words and change the construction around them.

The practical account

An actor sits at the workshop's front desk in a visibly staged scene. An opening caption identifies the words as an excerpt from an interview, performed by an actor. The delivery slip and unopened boxes are props, not represented as documentary evidence.

The actor says, “We kept the lights on,” in an ordinary explanatory delivery. The sequence continues with the qualification about the front room. We see a hand checking the delivery against an order. The following answer retains the manager's statement that the workshop was not in danger of closing.

This version makes the line part of a small account of work after hours. It still invents the visible actions and the manner of performance. Calling it a reconstruction of the original interview would overstate what the text alone provides: the transcript does not tell us how the manager sat or where the interview took place.

A more exact description is “a staged performance of the selected exchange, with constructed workshop actions.” That tells the recipient how to read the sequence without claiming the proposed staging recovers the original event.

The institutional crisis

An empty building appears. A locked gate follows. After a long pause, the actor turns toward the camera and stresses we: “We kept the lights on.” The soundtrack falls away before “on.” Neither the front-room qualification nor the denial of closure is included.

The words are identical, but the proposed image sequence encourages a different reading: these people held an endangered institution together. The emphatic “we” may also suggest an absent or negligent “they.” Neither implication is established by the invented exchange.

The issue is not that an empty building is an impermissible image or an emphatic performance is inherently false. It is the relationship being claimed. Are we presenting the manager's account of a delivery delay, or using those words to make the filmmaker's broader argument?

If the project has separate evidence of a real institutional crisis, identify that evidence separately. Do not let the excerpt serve as its proof merely because the metaphor is convenient. If the sequence is openly associative, say so in the approach: the image is an editorial construction, not footage of the event described.

The second version may have an artistic reason to exist. It cannot acquire the first version's narrower evidential claim by being called verbatim.

Make performance an authored choice

The transcript tells us which words were written down. In our fictional example, it does not tell us which word the manager stressed, whether they smiled, whether they paused, or whether background sounds interrupted the answer. Adding those details supplies a performance rather than retrieving one.

Try the sentence with the emphasis marked in three places:

We kept the lights on.

We kept the lights on.

We kept the lights on.

As written performance directions, the first sets up a contrast between people; the second foregrounds continuation; the third makes the object of the action more prominent. These are proposed readings, not measured audience responses. An actor could complicate each one through timing, expression, or the surrounding scene.

The pitch should make the consequential direction intelligible. “A powerful delivery” does not explain whether the performer is reproducing recorded timing, interpreting transcript wording, addressing the viewer, or playing a character in a newly created situation.

An exact-audio method has different boundaries. You may retain the speaker's recorded hesitation but choose a visible face that interprets it. You may preserve an answer while placing a reaction shot beside it. What stays fixed in one channel does not make the other channel neutral.

In her interview with Jason Wood for Electric Sheep, Barnard described constructing fictional images around verbatim audio and said she wanted viewers to recognize the shaping of the story. That is an expressed artistic intention, not a guarantee of how every viewer responds.[^2] It offers a useful question for the deck: where will your own film let the audience recognize the construction?

The answer might involve a direct address, a visible recording setup, an explicit source label, or a formal distinction between material types. Select what belongs to your film. Do not add an explanatory device that claims to settle a problem the scene itself continues to create.

Keep a construction note that survives revision

For the workshop scene, the working note can be brief:

Keep the rejected selection beside the usable version in your working notes. That matters because a later edit could remove the denial while leaving the old description of the approach unchanged.

After a revision, check the specific relationship again. Did a new cut make a comment seem to answer a different question? Did rearrangement turn uncertainty into accusation? Did a close-up make an invented gesture appear to substantiate a claim about the original speaker?

Do not treat every omission as a breach of the method. Films select. The practical task is to know what each selection changes, then avoid promising more fidelity than that selection preserves.

The record should also distinguish editorial questions from use questions. Having an accurate transcript does not establish permission to reproduce or perform it. A participant's agreement to an interview does not, by itself, tell the reader what consultation or authorization exists for this proposed staging. State the actual arrangements and unresolved questions; seek appropriate review where the material requires it. Do not turn this formal approach note into a substitute for that work.

Write the pitch around the reason for this form

Why retain the wording at all? “Authenticity” is too broad to explain the choice. In our workshop film, the phrasing might matter because ordinary language lets us examine the difference between what a person says and what a film can make it seem to mean.

That purpose supports a particular treatment. Retain the question, the small qualification, and the denial. Let the actor perform the practical detail rather than manufacture a confession. Make the staging legible enough that the scene is not confused with an actual filmed interview.

The resulting pitch passage could read:

Actors perform selected interview exchanges in constructed workshop settings. The wording stays within each retained question and answer; the voices, visible actions, and spaces are newly staged. In the delivery scene, we keep the manager's front-room qualification and denial of closure, so a practical account is not recast as testimony about saving the business. The film's authorship sits in the arrangement and performance of these exchanges, not in a claim to reproduce the original encounters.

For this teaching example, that is a proposal based on invented material, not a report of permissions secured or scenes completed. A real project's passage would need to name its actual source basis and performance method.

The finished note does more than announce fidelity. It tells the recipient what remains quoted, what becomes a performance, and what the film is responsible for adding.

Sources

[^1]: BAFTA, “Clio Barnard: Interview”, posted September 14, 2011. Questions on narrative shape, production, and actors' lip-sync method. Inspected September 19, 2026. Cited as the director's account of the audio and picture workflow, not an independently inspected source-to-film comparison.

[^2]: Jason Wood, “The Arbor: Interview with Clio Barnard”, Electric Sheep, March 2, 2011. Opening question and response on fictional images, verbatim audio, and visible construction. Inspected September 19, 2026. The interviewer's credit appears at the end of the page; the posting account is not substituted for that credit.

Frequently asked questions

What should a verbatim-film pitch define about its source?

It should name the actual source unit and what part is retained. An entire exchange, a passage from a hearing transcript, a recorded voice message, and selected words from several occasions are not interchangeable. Unchanged might refer to words within each excerpt, their order within an answer, their order across the whole interview, or the actual audio samples. The pitch should also say how the material will be performed and where the film adds interpretation.

Why can an excerpt change meaning even when every word is accurate?

In the invented workshop exchange, We kept the lights on originally follows a question about a late delivery, and the next sentence limits the place to the front room. A later answer explicitly rejects the idea that the workshop was in danger of closing. Extract only those five words and every word is accurate, but the limits that made the phrase clear are gone. Joining fragments from the interviewer's question and the manager's answer can also change who appears to be asserting something.

What is the difference between actors performing transcript excerpts and lip-syncing to original audio?

Actors performing selected passages preserve a textual selection while creating a new vocal performance. Actors lip-syncing to original interview audio retain recorded voices, including timing in the selected audio, while creating a new visible performance. Both approaches still require decisions about selection and presentation. Clio Barnard described shaping an audio screenplay, then a picture edit, with actors receiving playback through earpieces and miming, but that is the maker's account rather than an independently inspected source-to-film comparison.

How can the same sentence support two different film claims?

The practical version stages an actor at a workshop front desk with a caption identifying the words as a performed excerpt, and it keeps the front-room qualification and the denial of closure. The institutional-crisis version shows an empty building and locked gate, stresses we, drops the qualification and denial, and encourages a reading about an endangered institution and an absent they. If the project has separate evidence of a real crisis, identify that evidence separately; if the sequence is associative, say so.

What should a construction note record, and what does it not settle?

A useful note records what is held or changed: wording, voice, setting, action, qualification, and disclosure. Keep the rejected selection beside the usable version so a later edit does not remove a qualification while leaving the old description unchanged. The note does not settle use or permission. An accurate transcript does not establish permission to reproduce or perform it, and appropriate review remains separate from the formal approach note.

Further reading

• Pair Archival Sound and Images Without Inventing an Event — When a pitch pairs a voice with images, this ledger helps you state the relationship the cut suggests without claiming an event was filmed.
• Build a Documentary Paper Edit Before Calling It a Rough Cut — Before calling a sequence cut, use this to see which held beats need addresses and which are still proposed, so nothing unseen is claimed as recorded.