How to Read a Multiple Exposure — whose work is the overlap?
When someone shows us a photograph with two images layered into one, we usually say it looks interesting. Then we move on to the next picture.
Multiple exposure may be the technique where our looking stops soonest. A frame that worked reads like a gift from chance; a frame that failed reads like mud. Either way we stop at "interesting" or "I don't get it."
This article treats multiple exposure not as a way of making, but as a problem of reading.
The short answer
The question that helps in front of a multiple exposure is not whether it overlapped well.
It is who decided the overlap.
The answer usually splits three ways.
- The overlap the photographer decided — what goes where, and how much of it
- The overlap the principle decided — exposure adds, so the brighter side wins whether or not anyone chose it
- The overlap the viewer read in — memory, absence, or a ghost seen in a half-transparent figure
All three mix inside one frame. Left mixed, you stop at "somehow good." Separated, judgment starts to move.
What this article is and is not
The author has not practiced multiple exposure. There is no deliberately overlapped frame of my own, and no failed one on my desk either.
So this article can offer three things:
- the arithmetic of exposure, which is verifiable
- historical uses that could be confirmed against primary sources
- distinctions usable in front of a picture
What it cannot offer is practical values. Which film, how many stops down, at what interval — for that, go to the records of people who shoot it. This article sits one step earlier, where you decide what you are aiming at.
It does not layer. It adds.
The principle matters first. Skip it and everything after becomes mere appreciation.
If you release the shutter again without advancing the film, the same surface is exposed twice. What happens is not a superimposition of two images. It is the addition of light.
From this, one decisive consequence follows.
The second exposure can only enter where the first one left shadow.
Nothing you add to a blown-out sky will register; the sky stays white. But inside a shadow that fell to black, the second exposure appears intact. The shadows of the first frame are the canvas for the second.
This asymmetry decides most of how a multiple exposure looks. Night scenes with figures, bare branches against sky, dark interiors with windows — these combinations recur in multiple exposure work not because photographers happen to share a taste, but because the principle permits little else.
One note on exposure itself. If you expose twice, each exposure has to be held back. The rough arithmetic says one stop under for each of two frames, but film's characteristic curve is not a straight line, and long exposures bring reciprocity failure. Expect the arithmetic not to hold. This is territory for practitioners' records.
Marey — he erased the subject in order to overlap it
Before multiple exposure was a work of art, it was an instrument.
In 1894 the French physiologist Étienne-Jules Marey published Le mouvement. The full text is available at Gallica, the digital library of the Bibliothèque nationale de France, and what follows comes from reading those pages directly.
Marey's problem was scientific. A person runs, walks, jumps — too fast for the eye to follow. Record successive positions on a single plate and the trajectory itself becomes readable as a diagram.
But expose one plate dozens of times and the image collapses. Because it adds.
Marey's answer was not better overlapping technique. It was to erase the subject.
The caption to a figure on p.136 of the same book states the method in one line:
Des images partielles, lignes brillantes sur un vêtement sombre, sont recueillies sur plaque fixe (25 images par seconde).
(Partial images — bright lines on a dark garment — are collected on a fixed plate, twenty-five images per second. Translation mine.)
Dress the subject in black. Attach bright lines along the limbs and joints. Expose one fixed plate twenty-five times a second. What remains on the plate is lines. The person does not appear.
This is the most thorough possible response to an additive medium. If the second exposure can only enter shadow, then make nearly the whole frame shadow. He subtracted in order to add.
The apparatus deserves a look too.
This "photographic gun" recorded twelve images a second at 1/720 second each. But it arranged those images side by side on a single plate. Exposing one plate many times is common to both methods; whether the images sit apart or on top of each other produces entirely different pictures.
The substance of multiple exposure is "several exposures on one surface." Whether they overlap is a matter of placement. That distinction returns later, when we look at finished work.
One caution. Marey's plates are beautiful, but they were not made for beauty. They are what came out of overlapping in order to see the invisible. Treating them as artworks is a rereading by later generations. That is not forbidden — but it should be conscious.
Spirit photography — when overlap looks like evidence
The same principle was put to a completely different use.
In 1860s Boston, William Mumler began selling "spirit photographs": portraits of living sitters with the faint figure of a dead relative behind them.
In 2005 The Metropolitan Museum of Art mounted The Perfect Medium: Photography and the Occult, co-organized with the Maison Européenne de la Photographie in Paris (at the Met, September 27 – December 31, 2005; some 120 photographs). According to the museum's announcement, the exhibition covered the 1860s through the Second World War, and the ghostly figures were usually produced by double exposure, or by a person stepping briefly into the frame during a long exposure.
The line that matters most here is what follows. Both believers and skeptics summoned the photographs as evidence.
This is the third layer of multiple exposure.
Regardless of what the photographer intended, the viewer reads the overlap. A half-transparent figure looks like someone who was not there. Its faint outline looks like the absence of a body.
But the figure is faint because the exposure was short, not because the body was immaterial. An image that can only enter shadow, and that received only part of the total exposure, comes out thin and lets the background through. That is a consequence of the principle, not a meaning.
This reading did not end in the nineteenth century. When we look at a multiple exposure today and feel that time is layered, or that it resembles memory, part of that feeling comes from our side of the picture, not from the frame.
Reading things in is not a fault. Looking at photographs always involves it. What makes the difference is knowing that the reading was yours. Without that distinction, an image stops at "evocative" and goes no further.
Callahan — taking the overlap on as form
Neither a scientific instrument nor spiritual evidence: some photographers treated overlap itself as form.
Harry Callahan (1912–1999) worked with multiple exposure over many years. The Art Institute of Chicago holds works with the term in the title.
- Multiple Exposure of Willow Leaves (1945, gelatin silver print, image 24.3 × 19.2 cm, Peabody Fund)
- Multiple Exposure Tree, Chicago (1956, gelatin silver print, image 5.5 × 5.8 cm, mount 25.3 × 19 cm, Gift of Matthew Horn)
Pause on the dimensions of the 1956 print. The image is 5.5 × 5.8 cm — about two business cards side by side — set in the middle of a 25.3 × 19 cm mount.
So this is not a work that shows the force of overlap at scale. The mount is far larger than the picture. It is something you approach and look into, lines accumulating inside a small area.
The choice of subject matters too. Willow leaves; a tree. Both are subjects made of thin lines with dark gaps between them.
Given how exposure adds, that is sound. The darkness between branches and leaves becomes the place the next exposure can enter. What Marey manufactured with a black garment, Callahan borrowed from the structure of a tree. The methods are unalike; the direction is the same. Both prepared shadow before overlapping.
I have not seen these prints in person. The dimensions and credit lines above come from the museum's published data. How far the lines separate, how firmly the shadows hold, what happens when you actually lean into a small print — that requires the print itself. The collection pages above are where to start.
Three overlaps, told apart
Here is the working form of the three examples. In front of a multiple exposure, ask four things in order.
| Question | What it tells you |
|---|---|
| Where did the second exposure land — in shadow, or over a blown highlight? | Whether the photographer used the principle or was pushed around by it |
| Was the subject reduced? Are there too many elements? | Whether there was a design before the overlap |
| Was the position of the overlap chosen, or is it simply where it fell? | How much intention is present |
| Whatever story you read there — where in the frame is its evidence? | How much of the work the viewer is doing |
The fourth matters most. If you felt that time was layered, look for the object in the frame that supports the feeling. If it is there, the work did it. If it is not, you did.
To be clear, "chance is fine" is a legitimate position. Designed overlap is not the only good kind. But choosing chance and having nothing but chance are different things. What this article offers is an eye for that difference.
If you want to try it on your own camera
For readers who want to shoot rather than only look, here is what could be confirmed. I do not practice this, so read the following as guidance on how to check, not as instructions.
A camera can do multiple exposures when film advance and shutter cocking can be separated. Without that separation, releasing the shutter always moves the film one frame.
Among current models:
- Lomography LC-A+ has an MX switch on the base. After the first exposure, flip it before advancing, and you can expose the same frame again. The operation is documented in the manufacturer's FAQ.
- Lomo MC-A also has a multiple exposure function. I looked at how its breadth of features relates to its weight in the comparison of three film cameras currently sold new.
- PENTAX 17's official feature page lists no multiple exposure mode (checked 28 August 2026). It names seven shooting modes, and multiple exposure is not among them. Two half-frame pictures sitting inside one 35mm frame is a different thing from a multiple exposure.
To check a camera you already own, look in the manual for "multiple exposure," "double exposure," or "MX." If nothing is listed, you can rewind and reshoot a roll, but the frame registration will not return exactly. It becomes a technique that includes the misalignment.
Note that I have not surveyed current film cameras exhaustively. For models not listed here, check the manufacturer's official specifications.
What actually differs between film and an app
Multiple exposure is now easy in apps and in-camera composites. Opacity, position, order — all adjustable while you watch, all reversible.
On film, it is settled the moment you release the shutter the second time. You see the result after development.
I do not want to turn this into an argument that the harder way is better. Difficulty is not quality.
What differs is whether the one-shot nature of the decision stays in the frame. A film multiple exposure contains a second shutter release made without knowing the result. An app composite contains a choice made while looking at the result. Both are legitimate; the events the pictures record are simply different.
Honestly, though: in most cases you cannot tell them apart from the finished image. Claiming that a film multiple exposure is recognizable on sight goes too far. What is knowable is a fact about the making — what the maker agreed to carry.
This line runs alongside what I considered in the "fake moon" problem in smartphone photographs. What is in the picture and how the picture was made have to be handled separately.
The conditions that separate choices
How to approach multiple exposure divides by purpose.
If you want to design the overlap, subtract before you add. Narrow the subject, decide which part of the frame stays as shadow, and hold each exposure back. Stripped of their differences, this is what Marey and Callahan were both doing. The hit rate rises; the size of the surprise falls.
If you want the density of chance, do not design. But budget for how few frames will hold. The question is whether you can live with one keeper per roll.
If you only want to read the work, you do not need to shoot at all. Separating the three layers — photographer, principle, self — can be practiced with a photobook open on a table. The posture is close to asking what the roughness is a result of, which I wrote about in how to read Daido Moriyama.
If an app is enough, it is enough. Do not look for a reason to use film in the quality of the result. If there is a reason, it lies in how the shape of the decision changes. If that does not interest you, there is no need to overlap anything.
As a set of movements, multiple exposure is simple: don't advance, press again. The difficult part is deciding, before you press, what you want to be left with afterward.
Sources, and what could not be confirmed
Primary sources consulted (all checked 28 August 2026):
- Étienne-Jules Marey, Le mouvement, Paris, 1894, published in full at BnF Gallica. This article draws on p.110 (Fig. 79, the photographic gun), p.120 (Fig. 90, sabre fencing), p.136 (Fig. 98, the caption on geometric chronophotography), and p.140 (Fig. 102). Translations from the French are mine.
- Art Institute of Chicago collection. Technique, dimensions, and credit lines for Harry Callahan's Multiple Exposure of Willow Leaves and Multiple Exposure Tree, Chicago.
- The Metropolitan Museum of Art, press release for The Perfect Medium: Photography and the Occult. Dates, number of works, and the explanation of double exposure.
- RICOH IMAGING, PENTAX 17 official feature page. The list of shooting modes.
- Lomography, official FAQ. The LC-A+ MX switch.
What could not be confirmed, and cautions:
- The Met press release and the Lomography page returned access restrictions when fetched directly on the check date. Their content was confirmed through search-retrieved excerpts of the page text. Please verify at the links above.
- Images of the Callahan works are not included here: the Art Institute's image delivery is restricted for these copyrighted works. They can be seen on the collection pages linked above.
- Practical exposure compensation, reciprocity failure by film stock, a full survey of current film cameras with the function, and the reception of multiple exposure in Japanese photographic history all remain unverified. For practical values in particular, go to the records of people who actually shoot this way.
- The author has not practiced multiple exposure. This is a way of reading built from the principle and from primary sources, not a conclusion drawn from shooting.