Make a Montage Argue More Than “A Lot Happens”

By pitch.dogAdvanced13 min read20 September 2026

Advertising

A treatment page that says “a montage of the afternoon” has answered a scheduling question. It has not answered the editorial one. What does the cut sequence claim about the afternoon? A montage argues when the relation between fragments changes what they mean—when the viewer's understanding at the last shot is not simply the sum of the shots before it.

It can also merely accumulate, and that can be a legitimate choice. So the practical move is to write two versions from the same fragment list, one that accumulates and one that argues, and then choose. If both versions say the same thing, there is nothing to choose and the page is just a shot list with better grammar.

This article works through a single invented example end to end, because comparison needs material that stays still. No footage exists for it. No one shot it. It is a paper exercise: two proposed sequences over one finite set of fragments, annotated so the relationship between fragments is visible on the page rather than asserted in adjectives.

Name the change in understanding

Before writing the montage, write the proposition. The proposition is not a subject. Subject: preparations for a small gathering. Proposition: the host's visible calm is produced by work the guests never see. The gathering is the same; the proposition changes every adjacency decision that follows.

A compression montage can be good work. If the client needs thirty seconds to stand in for a nine-hour day, the montage's job is duration and texture—what the day felt like at speed, the lighting moving, the room filling. Compression doesn't need an argument because it isn't making one. The moment you claim a montage “shows how much care went in,” you have left compression and entered a proposition you now owe the edit.

A useful test before drafting: state in one sentence what the viewer believes after the last fragment that they did not believe after the first. If the sentence is “that a lot happened,” you have a compression sequence. If the sentence names a relation—cause, cost, contrast, contradiction—you have something to design toward.

Give each fragment a stated relation

Write the inventory first. For the gathering, a finite set:

Chairs carried in, arranged in a loose arc, one pushed back and set again
Tablecloth laid, corner smoothed flat
Flowers trimmed; one stem discarded; the remaining bunch rotated
A dish carried in from the kitchen, set down, foil adjusted
Guests arriving, coats to a chair back
Room full, plates in hands, two people laughing mid-sentence
Guests leaving, door held, goodbyes
Somebody loading plates into a bin bag

Now write each fragment's relation in the sequence, not its content. A fragment can establish a pattern, add pressure to one, contradict an impression, or redirect attention. “Add pressure” means: this shot makes the previous one cost something. Not every fragment needs a surprise. Repetition can make a difference perceptible—the second adjustment of the same chair is doing different work than the first because the first established the pattern it repeats.

Assembly A: accumulation

Ordered as above, with neutral captions in the treatment.

A1 Chair in place, adjusted. Sets the labor register.
A2 Tablecloth smoothed. Extends the register.
A3 Flowers trimmed. Extends it again.
A4 Dish carried in. Extends it.
A5 Guests arrive. The pattern now reads as before.
A6 Room full, plates, laughter. Pattern reads as after. The gathering happened in the interval.
A7 Guests leave.
A8 Plates to bin bag. The register returns; the day closes.

What does the last fragment mean here? That the work began, continued, and finished; the guests came and went in between. The sequence is honest and legible. Its proposition, if asked to state one, is: a gathering requires preparation and produces cleanup. That's a truism, not an argument. The two ends of the sequence are the same kind of event—labor—and nothing in the middle revises what the earlier labor meant. The music will have to carry the emotional claim, because the edit isn't making one.

This is why so many montage pages use lavish imagery: when the fragments don't change each other, the only available escalation is louder. If the treatment sentences are all “beautiful,” “warm,” “textured,” they're compensating for a missing relation.

Assembly B: counterpoint

Same eight fragments. Same footage, if this were real. Different ordering and one omission decision.

B1 Guests arrive, room filling fast (frag. 5–6 compressed at the top).
B2 The room full, plates, laughter (frag. 6).
B3 Flowers being trimmed—one stem discarded (frag. 3). Cut in while laughter is still on the track. The viewer now reads the flowers as having happened; the timeline has jumped backward and the pressure is that nobody in the room knows it did.
B4 Chair pushed back and set again (frag. 1).
B5 Dish carried in, foil adjusted (frag. 4).
B6 Room full again, a different pocket of it, laughter continuing (frag. 6 again—legitimate; repetition establishes that B3–B5 were taking place concurrently and unobserved).
B7 Guests leaving, door held, goodbyes (frag. 7).
B8 Plates into the bin bag (frag. 8). Interior, after.

Fragment 2, the smoothed tablecloth, is cut. Reason below.

What does B8 mean now? In B, the plates-to-bag shot follows a sequence in which every visible enjoyment was intercut with unseen preparation. The viewer has been given two timelines the guests on screen can't see unified: front-of-house and back-of-house, edited into one continuous present tense. B8 closes on the labor side, alone, after the door has shut. The proposition the arrangement is making: the room was good because someone was always working outside it—an attribution of cause the images themselves do not factually establish. More on that in a moment.

What changed, precisely

Three variables, and they're worth separating because a treatment draft usually conflates them.

Adjacency. In A, flowers sit between tablecloth and dish; each labor shot extends the same register because it's adjacent to labor. In B, the flowers are cut in immediately after the room full of people (B2) and immediately before the chair (B4). The counterpoint is that pair of cuts: inward at B2→B3, out of the room; outward at B5→B6, back into it. Same fragment, opposite work—in A the flowers sit inside the labor run, in B they open it.

Omission. Fragment 2 is cut from B. Not because the tablecloth is weak footage—because it is the only labor fragment in the set with no second beat in it. The chair is pushed back and set again; one flower stem is discarded and the bunch rotated; the foil on the dish is adjusted. Each of those second moves gives the viewer something to read inside the labor. The tablecloth is laid and its corner smoothed in one continuous gesture, so between the flowers and the chair it would have added duration to the back-of-house run without adding a decision. Cutting it lets B3 and B4 land harder, and keeps the run to the three fragments that each contain a turn. This is the fragment removed because it weakened the turn.

Repetition with a change. Fragment 6 appears three times in B: compressed into B1, plainly in B2, and repeated at B6. In A it appears once and marks a chronology. The decisive use is B6, where the lift of laughter after B3–B5 confirms that the labor interleaved between the two room-full shots was concurrent, not earlier. The recurrence isn't variety; it is the load-bearing evidence of the proposition.

What the edit cannot say

Here the reasoning has to slow down. B's arrangement encourages a viewer to read a causal relation: labor produced the enjoyment. That's a strong reading because the counterpoint puts difficulty adjacent to pleasure and the viewer supplies the cause. The images do not establish that. Nobody in fragments 3–5 is shown seeing the room, being thanked, or connecting their work to anyone's experience. The edit has created an attribution the material doesn't verify—the host, a friend, whoever trimmed those flowers, has not been shown to be the cause of anything on screen. The same caution applies to a real job: the edit proposes causal reading; what happened between people off camera is not something a montage can testify to.

An earlier version of this exercise tried to solve that by adding a ninth fragment—a person pausing in a doorway to watch the room, unseen. That's a different film. It would need to be planned, budgeted and covered, and it would shift the piece toward a point-of-view story rather than an editorial argument. The whole value of the two-assembly exercise is that the fragments stay fixed and only their relationship changes. If your counterpoint needs a new shot to justify itself, you're not choosing an argument in the edit; you're rewriting the shoot.

One more honest boundary: A and B are both coherent. The exercise does not prove B is better. It shows what proposition each one proposes, so the choice becomes explicit and can be made against the brief. A client whose message is “the care is in the details” may prefer a version with accumulation. A client whose message is “you won't see us working” has a proposition B actually argues.

Choose the turn and write the ending

Every counterpoint sequence needs one moment that revises the established alternation so the final fragment lands as a conclusion rather than a stop. In B the turn is fragment 7 to 8: guests leaving, door held, then the bin bag interior with no pleasure cut against it. The alternation between pleasure and labor has ended not because both are over, but because pleasure is no longer given its own shot; the labor continues alone. Fragment 8 is now no longer “cleanup” (A's reading). It is the thing that was always underneath the gathering, now visible alone. The viewer arrives at the last shot already holding the relation, so the shot doesn't need to illustrate anything.

That turn is why fragment 2 was cut. Restore it between the flowers and the chair and the back-of-house run becomes four consecutive labor shots—flowers, cloth, chair, dish—instead of three, so the return to the room after the dish reads as a block rather than a counterpoint. The alternation softens, and fragment 8's solitude is less isolated. A single omission protects the ending, and the omission worth making is the shot that would have added time to the run without adding a turn.

So the treatment page can say, in sentences rather than adjectives: the room fills; we intercut the work the room never sees; the guests leave; we stay with the work after they're gone. Each clause names a relation. None of them name a feeling the sequence is supposed to generate, because feelings belong to the viewer, and the page's job is to state the argument the edit is being asked to make so the client and the editor can agree before anyone is on set.

What to do with the two versions

Keep both on the page, briefly. Version A as a note—“compression option, no argument”—and Version B as the proposal, with its turn and its cut named. Showing the alternative costs four lines and makes the editorial choice legible as a decision rather than a reflex. It also protects you later: if someone asks in the edit why the tablecloth isn't there, the treatment already answered that question and the answer wasn't “we forgot.”

Frequently asked questions

What distinguishes a montage that argues from one that merely accumulates?

A montage argues when the relation between fragments changes what they mean—when the viewer's understanding at the last shot is not simply the sum of the shots before it. It can also merely accumulate, which can be a legitimate choice for compression: if the job is to stand in for a nine-hour day in thirty seconds, the montage's work is duration and texture. A useful test is to state in one sentence what the viewer believes after the last fragment that they did not believe after the first. If the sentence is 'that a lot happened,' it is compression; if it names a relation such as cause, cost, contrast, or contradiction, there is something to design toward.

Why write two versions from the same fragment list, and what is the proposition?

Writing two versions from the same fragment list, one that accumulates and one that argues, makes the editorial choice visible. If both versions say the same thing, there is nothing to choose and the page is just a shot list with better grammar. Before writing the montage, write the proposition. The subject is preparations for a small gathering; the proposition is that the host's visible calm is produced by work the guests never see. The proposition changes every adjacency decision. Assembly A's implied proposition—a gathering requires preparation and produces cleanup—is a truism; Assembly B proposes that the room was good because someone was always working outside it, an attribution the images do not factually establish.

What changes between Assembly A and Assembly B?

Three variables are worth separating: adjacency, omission, and repetition with a change. In Assembly A, each labor shot extends the same register because it is adjacent to labor. In Assembly B, the flowers are cut in immediately after the room full of people and immediately before the chair; the counterpoint is that pair of cuts—out of the room, then back into it—so the same fragment does opposite work. Fragment 2, the tablecloth, is cut from B. Fragment 6 appears three times in B: compressed at the top, plainly in B2, and repeated at B6, where the lift of laughter after B3–B5 confirms that the interleaved labor was concurrent, not earlier. The recurrence is the load-bearing evidence of the proposition.

Why cut the tablecloth fragment from Assembly B, and what does the turn do?

The tablecloth is the only labor fragment with no second beat in it. The chair is pushed back and set again; one flower stem is discarded and the bunch rotated; the foil on the dish is adjusted. The tablecloth is laid and its corner smoothed in one continuous gesture, so it would add duration to the back-of-house run without adding a decision. Cutting it lets B3 and B4 land harder and keeps the run to the three fragments that each contain a turn. Restoring it between the flowers and the chair makes the back-of-house run four consecutive labor shots instead of three, so the return to the room reads as a block rather than a counterpoint, softening the alternation and isolating fragment 8 less. The turn in B is fragment 7 to 8: guests leaving, door held, then the bin bag interior with no pleasure cut against it. The alternation ends because pleasure is no longer given its own shot; labor continues alone. Fragment 8 becomes the thing that was always underneath the gathering, now visible alone.

What can the edit not say, and does the exercise prove Assembly B is better?

Assembly B's arrangement encourages a viewer to read a causal relation: labor produced the enjoyment. The images do not establish that. Nobody in fragments 3–5 is shown seeing the room, being thanked, or connecting their work to anyone's experience. The edit has created an attribution the material does not verify; whoever trimmed those flowers is not shown to be the cause of anything on screen. The same caution applies to a real job: the edit proposes a causal reading, but what happened between people off camera is not something a montage can testify to. The exercise does not prove B is better. A and B are both coherent. It shows what proposition each one proposes, so the choice becomes explicit and can be made against the brief. A client whose message is 'the care is in the details' may prefer a version with accumulation; a client whose message is 'you won't see us working' has a proposition B actually argues.