At 11 a.m. on November 11, 1918, an American crew near the Moselle River was pulling a strip of light-sensitive material past a beam of light while a tuning fork laid a hundred timing marks a second along its edge. What came off that machine is a photograph of a front going quiet, and the birdsong you hear over it on social media could never have reached it.
The image circulating with that audio is real. It is titled “The End of the War: A Graphic Record,” it is held by the Imperial War Museum, and it is one of the few physical artifacts that shows an active front and a silent one side by side. What it is not is an audio recording. Nothing about it can be played back, and the reason has more to do with how photographic emulsion works than with anything about audio.
What the Record Actually Shows
Look at the strip and you see six horizontal lines running left to right, with a dense row of tick marks along the top and bottom edges. Each of the six lines belongs to one microphone. The tick marks are a clock. On the left half, the lines break into ragged spikes, which is a front full of artillery. On the right half, they run nearly flat.
The caption printed with the record describes it as the last record by sound ranging of the artillery on the American front close to the River Moselle, made as 11 o’clock brought the general order to cease firing. It also accounts for the only interruption on the quiet side. Two small breaks in the second line, the caption says, are “probably being due to the exuberance of a doughboy firing his pistol twice close to one of the recording microphones.”
That is the whole document, and it is less continuous than it looks. The version everyone shares is an excerpt. The 1919 War Department volume that published it, Benedict Crowell’s America’s Munitions 1917-1918, says so directly in the caption: six seconds of sound recording are shown, and “the two minutes on either side of the exact armistice hour have been cut from the strip to emphasize the contrast.” The most famous picture of the moment the war ended does not contain the moment the war ended. It holds a few seconds of barrage, a few seconds of quiet, and an editor’s decision in the gap between them.
A Recorder With No Shutter and a Tuning-Fork Clock
The machine that made it was called a harp galvanometer, and it worked on a principle any photographer will recognize immediately. Six fine wires were strung parallel to each other in a magnetic field, roughly half a centimeter away from a plane of moving light-sensitive stock. A lamp and a small optical system threw the shadows of those wires onto a horizontal slit, with a cylindrical lens in front of the slit to narrow it, and through the slit onto the stock. When current from a microphone ran through one of the wires, the wire twitched, and the shadow twitched with it. That is the entire recording mechanism. No shutter, no diaphragm, no image of anything in the world, just six shadows and a transport.
The clock was mechanical and beautiful. A wheel with spokes spun in the path of the same light beam, chopping it into flashes and laying down a line on the stock about one-fiftieth of an inch apart. The spin rate was governed by a tuning fork, and the rate chosen was one hundred marks per second. Because the shadow of the twitch and the shadow of the time scale were exposed onto the same piece of material by the same lamp, the arrival time of a gun could be read to within a hundredth of a second with no synchronization problem at all.
Augustus Trowbridge, the Princeton physicist who organized the American service, wrote the medium down plainly in his 1920 account. “The photographic paper employed was of the width of the standard moving picture film,” he noted, “as this could be obtained quickly and at low cost both in Europe and America.” The British apparatus ran 35mm cine film through the same mechanism, so the two services were not working on the same medium at all. The honest answer to whether the armistice record is on photographic film is that the American one is photographic paper cut to the width of movie film. The width is what was borrowed, for the reason Trowbridge gives: it was the size you could get quickly and cheaply on both sides of the Atlantic in the middle of a war.
The apparatus also developed and fixed its own output automatically, which is worth sitting with. In 1918, a machine in a wrecked farmhouse was exposing a strip, processing it, and handing it to an operator wet, because the artillery needed the answer in minutes. A British sound ranging section carried three officers and eighteen other ranks, and the establishment included one photographer. His job was the darkroom.
The recorder ran in bursts of twenty or thirty seconds, triggered by forward observers who pressed a key when they heard guns in their assigned arc. A single strip typically caught the muzzle blast, the shell passing overhead, and the burst at the far end. Trowbridge put the turnaround from firing to a report in the artillery’s hands at one to two minutes.
Six Microphones Strung Across Five Miles
The strip was useless without geometry. Microphones were surveyed into position roughly 1,500 meters apart along an arc concave toward the enemy, giving a base about 7,500 meters long, slightly under five miles. Connecting them to the recording station took something like 40 miles of low-resistance wire, laid and repaired by linemen working in the open, because the vacuum tube amplifier existed but had not yet been adopted for this job.
Because the six microphones sat at known coordinates, the five intervals between arrival times on the strip were enough to solve for the gun. A March 1917 British General Staff report put the error from a single good observation at about 50 yards, dropping under 25 yards when several were averaged, and credited one section with 260 German battery locations in two months. The American section that ended the war on the Moselle logged 493 locations from a single base.
The man who made the British system work was William Lawrence Bragg, who had shared the Nobel Prize in Physics with his father in 1915 at the age of 25 and is still the youngest science laureate. He spent the war on this problem. The insight that saved it came from a farmhouse privy at La Clytte in Flanders, where Bragg noticed that when a British 6-inch gun fired a quarter mile away, anyone sitting over the only opening to the outside air was perceptibly lifted off the seat. The gun’s energy was in the very low frequencies, and the microphones then in use were listening in the wrong place entirely.
Why the Apparatus Could Not Hear a Bird
The fix came from Corporal William Sansome Tucker, a physicist from Imperial College who had been studying how moving air cools fine platinum wires. Guns firing near his tarred-paper hut pushed jets of cold air through the tears in the walls and onto his bunk, which gave him the idea. The first test stretched a thin wire over the mouth of an empty rum jar. The production version was a 23-liter tinplate cylinder with conical ends, one end sealed, a short open tube in the other, and a grid of fine platinum wire across a 4.5-centimeter aperture in a mica disk.
That is a Helmholtz resonator with a thermometer in its throat, and its resonance sat somewhere around 30 to 50 Hz. Bragg measured the characteristic frequencies of guns between 10 and 25 Hz, with the biggest guns at the bottom of that range. The device was built to be deaf to everything above it. When the crew first tried it, the shell wave from a passing projectile barely moved the galvanometer while the muzzle blast, in Bragg’s phrase, gave an enormous kick. That selectivity was the entire point, because the shell wave was the noise that had been ruining every earlier attempt.
Most songbirds work somewhere between about 1,000 and 8,000 Hz, two to three orders of magnitude above anything a Tucker microphone would respond to. Even if a bird had somehow driven the wire, the output would not have been a waveform. Air moving in either direction cooled the platinum equally, so the circuit rectified, producing a current that was always positive. The hundred marks a second sometimes get offered as a third reason, on the assumption that they are a sampling rate. They are not. The trace was continuous, and the marks were a ruler laid down beside it so arrival times could be read to a hundredth of a second. What kept a bird off this strip was the microphone, not the clock.
The instrument recorded arrival times of pressure fronts. It is far closer to a seismograph than to a phonograph, and it has no more audio content than a barometer trace.
The Man Who Built It Photographed Flies at 1,200 Frames a Second
The recorder came out of a photography lab, not an audio one. Lucien Bull was born in Dublin, joined Étienne-Jules Marey as an assistant in 1895, and took over the Marey Institute in Paris when Marey died in 1904. He spent those years on chronophotography, replacing Marey’s mechanical shutter with electric sparks so that illumination rather than a moving part set the exposure. By 1904 he had a stereoscopic spark drum camera taking pairs of frames at up to 2,000 per second, and he used it to film insect flight at 1,200 frames per second, the first serious analysis of how a wing actually beats.
He had also been building galvanometers for electrocardiography, adapting Willem Einthoven’s string galvanometer of 1901 to record a heartbeat photographically. When Charles Nordmann, an astronomer at the Paris Observatory serving at the front, worked out in 1914 that a gun could be located by timing its sound at separated points, he went to Bull for the recorder. Bull built the first fifty by hand before the Cambridge Instrument Company took over production. Trowbridge, writing in 1920, described the British service as one that “employed a photographic recording instrument devised by a British subject, resident in France, Mr. Lucien Bull.”
So the chain runs from Marey’s motion studies through spark-lit high-speed cinematography and the photographic electrocardiogram to an artillery instrument, and the constant across all of it is that a strip of emulsion moving at a known rate is the most precise timekeeper anyone had. That is what photographic material was doing for science in 1918. It was not making pictures. It was writing down when things happened.
Photographic sound on film was not waiting on any of this. Eugene Lauste patented a method of recording sound photographically onto film in 1906, exhibited a sound film in the United States in 1911, and had the system working to his satisfaction by 1913, at which point the war that produced the sound rangers stopped him. The two lines ran alongside each other rather than one leading to the other. What the 1920s added was equipment that worked in a theater: Lee de Forest demonstrated his variable-density Phonofilm in 1923, and RCA Photophone’s variable-area track arrived in 1928, both of them modulating a light beam onto moving 35mm stock and reading it back optically. No documented line runs from the sound ranging sections to either system, and the sound rangers were writing a position, not an intensity envelope. But the family resemblance is close enough that the confusion behind the viral clip becomes easy to understand. People know that sound has lived on 35mm film. They are looking at a strip of the same width that was doing something else entirely at the same time.
What You Are Actually Listening To
The audio is a 2018 commission. For its Making a New World season, the Imperial War Museum hired the studio Coda to Coda to build a piece called Armistice Soundwave, using the film’s timings plus research into what specific guns of the period sounded like. It was installed as a bone conduction sound bar, so visitors rested their elbows on a table, cupped their ears, and felt the barrage through bone rather than air. The birds at the end are a compositional decision standing in for the silence. The museum has consistently presented the work as an interpretation. That framing is what falls off when the clip is reposted.
The silence itself is documented, and by someone standing in the right sector. J. R. Hinman was an American sound ranger on the Moselle, and his 1919 unit history describes the last three minutes from the doorway of the section’s kitchen shack. “Suddenly the firing ceases,” he wrote. “There is silence — silence so great that it in itself is appalling. We look at the old French clock. We distinctly hear it tick off exactly six seconds — then it strikes the hour of 11.”
An eyewitness and a clock in the corner got the moment down to six seconds. The instrument down the line read intervals on its own trace to a hundredth of a second, which is the precision that located a gun rather than the kind that times an armistice, and the two minutes bracketing 11 o’clock were cut out of the picture before it was published. Between the two of them, the one that recorded when the war ended is the man in the kitchen doorway. Neither he nor the apparatus caught a bird, because neither was ever listening for one.
Lead image by unknown, public domain.
Estudio EA
