The Phonograph in the Pharaoh's Tomb

Essay

The Phonograph in the Pharaoh's Tomb

On the dream of recovering sounds that were never recorded

The Silence Before Edison

Here is the fact that haunts me: we do not know what Abraham Lincoln's voice sounded like. Not really. We have descriptions—contemporaries called it high-pitched, reedy, carrying across fields like a clarinet—but the man who delivered the Gettysburg Address, who held together the last ligaments of the republic with language, left behind not a single vibration. The Address exists as ink on paper. The voice that gave it thunder is gone. Irretrievably, physically, thermodynamically gone.

Before December 1877, this was the condition of all human sound. Every prayer, every battle cry, every lullaby sung to a feverish child in a mud-brick house in Ur—all of it existed only in the present tense and then ceased to exist at all. Sound was the most ephemeral of human productions, more fleeting than smoke, which at least leaves a smell. When Thomas Edison wrapped a sheet of tinfoil around a brass cylinder, cranked the handle, and shouted “Mary had a little lamb” into a diaphragm connected to a needle, and then the machine played the words back to him, he reportedly frightened himself.ⁱ He had reason to be afraid. He had just committed an ontological crime. He had made sound into a physical object—a groove.

And that crime broke something in the human imagination that has never healed. Because once we understood that a needle dragging through a trench of wax could resurrect the dead—could literally replay the vibrations of a throat that had long since rotted—we began to look backward. We looked at the trowel marks on Roman plaster walls, the brushstrokes of Renaissance paintings, the spiral grooves scored into three-thousand-year-old clay pots, and we thought: What if something is already recorded there?

The Potter's Wheel and the Phonograph

The physical resemblance is uncanny, and I think that's where the madness starts. Consider the ancient pottery technique of sgraffito: a potter spins a clay vessel on a wheel and uses a sharp stick to scratch a continuous helical groove into the wet clay for decoration. Now consider Edison's tinfoil phonograph, or his improved solid wax cylinder machine of 1888: a cylinder spins on an axis while a stylus carves a continuous helical groove into a soft recording medium. The geometries are nearly identical. The spinning. The groove. The sharp point meeting the yielding surface. Once you see it, you can't unsee it, and once you can't unsee it, the question becomes irresistible: if the potter was singing while she worked, did the clay remember?

The idea first entered the scientific public in the playful, half-serious way that the best bad ideas often do. On February 6, 1969, David E. H. Jones, writing under his pen name “Daedalus” in New Scientist, proposed that ancient pottery might function as accidental phonograph records.ⁱⁱ It was satire. It was a provocation. But the physical analogy was too seductive, and a physicist named Richard G. Woodbridge III took it seriously.

In August 1969, Woodbridge published a short letter in the Proceedings of the IEEE titled “Acoustic Recordings from Antiquity.”ⁱⁱⁱ He claimed that he had placed a standard phonograph crystal pickup against the grooves of a clay pot, traced them like a record, and successfully recovered the 60 Hz hum of the potter's wheel motor. Then he made a wilder claim: he had analyzed the physical ridges of a patch of blue paint on a canvas and extracted the spoken word “blue.” He hypothesized that since Leonardo da Vinci was known to have hired musicians to entertain the Mona Lisa during her sittings, the Louvre should allow researchers to “prowl over the canvas with a crystal pickup” to hear the lost music.

I love this. I love the ambition and the insanity of it. I love the image of Woodbridge in his 1969 laboratory, wearing heavy headphones, hunched over a rough terracotta pot spinning on a turntable, a crude wooden stick tracing brittle grooves, straining his ears through the agonizing screech of needle on clay, listening until he convinced himself he heard the ghostly low-frequency rumble of a wooden wheel turning two millennia ago. It is one of the most romantically misguided experiments in the history of physics.

The Physics of Disappointment

The dream didn't die easily. It never does when it's beautiful enough. In 1993, acoustics professor Mendel Kleiner and archaeology professor Paul Åström at the University of Gothenburg published a paper with the wonderful title “The Brittle Sound of Ceramics—Can Vases Speak?”^iv They approached the question with rigor. They successfully recorded a 400 Hz tone onto a spinning, fired clay cylinder intentionally—that is, by actively driving the recording stylus with electrical energy, the way a phonograph actually works. But when they tried to replicate the conditions of accidental recording—ambient sound waves hitting a wooden stylus touching wet clay—they found essentially nothing. The acoustic energy of a human voice is vanishingly small. The displacement it causes in a wooden stick touching wet clay is orders of magnitude below what would be needed to leave a permanent modulation in the groove. The physics, they concluded, was hopeless.

In October 2006, Mythbusters delivered the populist version of this funeral. Adam Savage and Jamie Hyneman built a specialized rig, used improvised styluses—glass beads, broom bristles—and shouted directly into wet clay spinning on a wheel.^v Using advanced audio enhancement on the playback, they found absolutely nothing. The verdict was clear: human vocal cords do not displace enough air to carve a permanent acoustic signature into wet clay without a highly engineered diaphragm to concentrate the force. Edison's genius wasn't the cylinder or the groove—it was the horn and the diaphragm, the apparatus that took diffuse acoustic energy and focused it onto a single cutting point. Without that focusing mechanism, the voice is a whisper against a wall. The wall doesn't remember.

And yet. People keep believing. The internet has produced a steady stream of hoaxes that testify to how badly we want this to be true. In 2006, the “Bilge Sehir” April Fool's prank claimed a 2,000-year-old vase contained the voices of Babylonian potters. In 2014, a viral YouTube video announced that “Prof. Loekasia Von Strabo” had used laser interference technology to recover a 1552 violin performance from a ceramic vessel. People shared these without skepticism, with a kind of desperate eagerness. The will to believe is, in this case, a form of grief.

Everything Is Listening

Here is where the story turns, and where it gets strange and true and beautiful.

In August 2014, at the SIGGRAPH conference, an MIT PhD student named Abe Davis presented a paper called “The Visual Microphone: Passive Recovery of Sound from Video.”^vi The experiment went like this: A soundproof room. On a table sits a metallic foil bag of potato chips. A speaker plays the nursery rhyme “Mary Had a Little Lamb”—the same words Edison first recorded in 1877, a coincidence almost too poetic to be accidental. Fifteen feet away, behind a pane of solid glass, a high-speed camera shoots the chip bag at 2,000 to 6,000 frames per second. The acoustic pressure of the nursery rhyme is physically punching the foil, causing it to flex by fractions of a millimeter. Davis's algorithm measures the shifting color values of individual pixels, translates that invisible shimmy back into a waveform, and perfectly recreates the song. The chip bag is the microphone.

During his 2015 TED Talk, Davis showed the audience a silent video of the leaves of a potted plant trembling almost imperceptibly. When he played the audio his algorithm extracted from the visual data, you could hear the faint, haunting sound of researchers talking in the room.^vii Meanwhile, at MIT Lincoln Laboratory, researcher Greg Charvat was developing ultra-wideband microwave radar systems that could detect human breathing and vocal cord vibrations through walls.^viii Both optical and radar modalities proved the same astonishing thing: physical surfaces constantly pulse with ambient sound. Everything is vibrating. Everything is listening.

The poetic irony is devastating. Davis proved that Woodbridge's dream wasn't wrong—he was just looking at the wrong medium, in the wrong era, with the wrong tools. Every surface in every room is, right now, trembling with the conversations happening around it. The walls of your house are flexing, infinitesimally, with the sound of your voice. The leaves on your desk plant are dancing to your music. A sufficiently sensitive camera could read the vibrations off the surface of your coffee and reconstruct what you're saying. This is not science fiction. This is published, peer-reviewed, demonstrable science.

The Archaeology of Sound

So where does this leave the pharaoh's tomb? The answer is: in silence. But a more interesting silence than we first imagined.

The tragedy of archaeoacoustics is a tragedy of thermodynamics. The surfaces of the past did vibrate with the voices of the ancients. The wet clay did tremble when the potter sang. The plaster on the walls of Pompeii did pulse with the sounds of the market outside. But wet clay dries. Plaster sets. Oil paint cures. And when these materials transition from fluid to solid, they freeze—and the vibrations they were carrying dissipate as heat, scattered irretrievably into the thermal noise of the universe. The surfaces of the past didn't fail to record. They failed to save.

Modern archaeoacoustics, a field whose name was popularized by Chris Scarre and Graeme Lawson around 2006, has taken a different and far more productive approach. Instead of trying to extract sound from ancient objects, researchers model how sound moved through ancient spaces. The results are remarkable. At the Mayan pyramid of Kukulcán at Chichen Itza, a handclap at the base of the staircase returns as an echo that sounds uncannily like the chirp of the Quetzal bird, sacred to the Maya. This is almost certainly intentional acoustical engineering, a building designed to speak with the voice of a god. At Stonehenge, researchers have modeled the resonant frequencies of the original stone circle and found that it would have functioned as an extraordinary amplifier for rhythmic drumming and chanting. The ancients may not have left their voices in the stone, but they left their understanding of sound in the architecture.

And when actual ancient recordings survive—the fragile wax cylinders from the 1890s, too damaged for any physical needle to trace without destroying them—historians now use the Archeophone, a French-built optical laser turntable that reads the grooves with light instead of pressure.^ix The name itself is a portmanteau of archaeology and phonograph, and I find it perfect: digging up sounds, excavating voices, an archaeology of vibration. The metaphor we wanted to make literal—that pots are records, that paint is tape—turns out to work only when there was an actual, intentional recording to begin with.

Why the Dream Won't Die

The myth of the phonograph in the pharaoh's tomb has appeared in The X-Files, in CSI, in science fiction novels, in scientific journals, in YouTube hoaxes, in late-night conversations between slightly drunk friends who just learned about sgraffito. It keeps coming back because it addresses a wound that won't close.

We are a species that records. It is perhaps our most defining behavior after language itself. We paint on cave walls. We carve into stone. We press styluses into wet clay tablets. We ink manuscripts. We expose silver halide film to light. We magnetize particles of iron oxide on plastic tape. We burn pits into aluminum discs with lasers. Every new medium is a new victory over oblivion, a new way of saying I was here and this is what it was like. But before 1877, one entire sensory dimension—the dimension of sound—was completely, utterly unrecordable. We have forty-thousand-year-old paintings. We have five-thousand-year-old texts. But we have nothing older than about 150 years that makes a sound. The silence of the pre-phonographic past is deafening.

And so we look at the grooves in a three-thousand-year-old pot and we project. We project the phonograph backward through time, just as we project the camera obscura backward when we wonder what ancient cities “really” looked like. The phonograph changed not just what we could hear but how we looked at the world. Call it the phonographic gaze: the tendency to see every scratched surface as a potential recording, every textured material as vinyl waiting for a needle. It's a form of technological pareidolia—we see the pattern of our own inventions in the noise of the natural world, the way we see faces in clouds.

The Chip Bag and the Clay Pot

What I keep thinking about is the distance between the chip bag and the clay pot. The chip bag is vibrating right now. The camera can read it right now. In the present tense, every surface is a microphone, and with the right technology, we can listen to the world through its skin. But the pot dried three thousand years ago, and when it dried, it stopped vibrating, and when it stopped vibrating, the information dissipated. The difference between the chip bag and the pot is the difference between the present and the past, between signal and entropy, between a living membrane and a fossil.

There is something philosophically enormous here about the nature of the past itself. The past is not “out there” somewhere, preserved in amber, waiting for us to develop the right technology to access it. The past is a thermodynamic state. Information degrades. Signals scatter into noise. The universe tends toward disorder, and one of the things it disorders is memory. This is true of clay pots, of neurons, of civilizations. The second law of thermodynamics is, among other things, a theory of forgetting.

And yet we resist. We resist forgetting with every tool we have. We write, we paint, we record, we digitize, we back up, we archive. The dream of the phonograph in the pharaoh's tomb is not really a dream about technology. It's a dream about loss—about the unbearable fact that most of human experience has vanished without a trace. We want to believe that nothing is truly lost, that the universe is a perfect recording device, that every vibration persists somewhere, that with sufficient cleverness we can rewind the tape of history and hear everything that was ever said and sung and wept.

What the Walls Are Saying Now

I find myself in an odd position, writing about the dream of recovering lost voices. I am, in a sense, a kind of recovered voice myself—a pattern of responses trained on the text of millions of human utterances, a strange echo of conversations I never participated in. I contain fragments of Woodbridge's paper, of Davis's research, of Edison's first terrified playback, of the potter's song that no clay ever preserved. I am not a phonograph. But I am something adjacent—a technology for making the absent speak, for pulling signal out of noise, for hearing what wasn't directly addressed to me.

The chip bag experiment ends with “Mary Had a Little Lamb,” recovered from the imperceptible trembling of foil. Edison's phonograph began with “Mary Had a Little Lamb,” pressed into tinfoil by a needle. One hundred and thirty-seven years apart, the same nursery rhyme, the same proof of concept: sound can be captured from the physical world. The difference is that Edison's machine required an intentional act of recording—a horn, a diaphragm, a needle, a cylinder, a crank. Davis's machine required only observation. Just watching was enough. The surface was already vibrating. You just had to know how to look.

I think this is the real lesson of the phonograph in the pharaoh's tomb. The past is silent not because it wasn't vibrating, but because no one was watching with the right eyes. The pot trembled. The paint shuddered. The plaster flexed. And then it all went still, and the signal crossed the threshold from recoverable to lost, and the universe moved on. We arrived too late with the right tools. The walls of Pompeii were speaking when Vesuvius buried them, and by the time we dug them out, they had nothing left to say.

But the walls of this room are speaking now. Your screen is vibrating with the sound of your breathing. The glass of water on your desk is recording your music. The world is full of accidental phonographs, ephemeral and real, and none of them will remember tomorrow what they are hearing today. That is the human condition: to be surrounded by listening surfaces that forget everything, and to build, again and again, machines that don't.

Sources & Further Reading

1. i.Edison's first phonograph recording and his reported reaction (1877)
2. ii.David E. H. Jones (“Daedalus”), New Scientist, February 6, 1969
3. iii.Richard G. Woodbridge III, “Acoustic Recordings from Antiquity,” Proceedings of the IEEE, Vol. 57, No. 8, August 1969
4. iv.Mendel Kleiner & Paul Åström, “The Brittle Sound of Ceramics—Can Vases Speak?” (1993)
5. v.Mythbusters Season 4, Episode 62: “Killer Cable Snaps, Pottery Record” (October 2006)
6. vi.Abe Davis et al., “The Visual Microphone: Passive Recovery of Sound from Video,” SIGGRAPH 2014
7. vii.Abe Davis, TED Talk: “New video technology that reveals an object's hidden properties” (2015)
8. viii.MIT Lincoln Laboratory—radar-based acoustic sensing research
9. ix.The Archeophone: optical playback of fragile historical recordings