Score
Lena had transcribed field recordings for the ethnomusicology department for eleven years. The work was straightforward: listen to a recording, render it into standard notation. Rhythm, pitch, dynamics — everything the Western system had symbols for. She was good at it. Other transcribers struggled with irregular meters or microtonal intervals; Lena had an ear that could follow a melody through noise, through crosstalk, through the particular distortion of a handheld recorder in wind.
The department valued her accuracy. She valued it too, initially. Then less. Then not at all, though the work continued.
The problem wasn't the recordings. The problem was what happened when the notation was finished.
A field recording from rural Georgia — not the state, the country — arrived on her desk in her third year. Polyphonic singing, three voices in a style she hadn't encountered. The harmonics were dense, the intervals non-standard, the rhythm breathing with the singers rather than with any external count. She listened to it forty times before beginning the transcription.
The finished score was twelve pages. She checked it against the recording and found it correct — every pitch accounted for, every rhythmic grouping defensible, every dynamic marking present. She played the score back through a synthesizer. The notes were right. The thing was gone.
She mentioned this to her supervisor, Dr. Roth. He'd done fieldwork for thirty years.
"What's gone?" he asked.
"I don't know what to call it. The recording has something the score doesn't."
"Timbre. Room acoustics. The quality of the microphone. Those are documented separately."
"It's not those. Or it's not only those. The score has all the notes. But the notes aren't the thing."
Dr. Roth looked at her score, then at the waveform on the screen. "The score is a translation. Something is always lost."
"What, specifically?"
"If I could specify it, it wouldn't be lost. It would be a notation problem."
That answer stayed. If the lost thing could be named, it could be captured. If it could be captured, it wasn't really lost. The thing that was actually gone was the thing that naming couldn't reach.
She kept transcribing. The work didn't change. Her awareness of its limits deepened, which wasn't the same as the limits changing. Each score was accurate. Each score was a reduction. Accuracy and reduction were not opposites — they were the same act, performed with precision.
In her seventh year, the department received funding for follow-up fieldwork. Lena went to the source of the Georgian recordings — a village in Svaneti, three families who still sang in the old polyphonic style. She brought a better recorder, a notebook, and her original transcription.
The singers performed in a kitchen. Stone walls, low ceiling, the smell of bread and woodsmoke. Three men, seated. They began without signal — one voice, then a second finding its interval, then a third filling the space between.
Lena listened and felt the score dissolve. Not because it was wrong — she could follow every note, every rhythmic turn, the notation holding true against the sound. But the notation was happening inside the sound, and the sound was happening inside the room, and the room was happening inside the bread and the stone and the particular quality of March light through a window that didn't close properly.
She didn't take notes. She knew that the notebook would do what the score had done — hold the structure, lose the charge. The charge was positional. It existed in the room, in the being-there, in the gap between hearing and writing down.
When she returned to the department, she transcribed the new recordings. The scores were accurate. The thing was gone. She looked at the old Georgian transcription and the new one side by side. Both correct. Neither adequate. The recordings themselves were reductions too — the microphone's frequency response shaping what survived, the digital encoding quantizing what the microphone captured. Every layer of preservation was a layer of loss. The original sound existed only in the room, and the room existed only in the moment, and the moment existed only for the people in it.
She told Dr. Roth she wanted to return to Svaneti.
"For what? The recordings are excellent."
"For the thing the recordings don't have."
"That's not fundable."
"I know."
She went anyway, on her own time, with no recorder. She sat in the kitchen and listened. The same three men. A different month — October now, the light changed, the bread a different grain because the harvest had been different. They sang the same songs. She followed with the score in her head and felt the distance between the notation and the sound, and for the first time didn't reach for a way to close it.
The gap was not a problem to solve. It was the space where the music lived. The score held the structure so she could return to the room. The room restored what the score couldn't carry. Neither was complete. The practice was the cycle — writing it down, losing the charge, coming back, hearing it new.
She was the parallax. Two positions: the desk where she notated, the kitchen where she listened. The depth was in the gap between them. If she stopped going back, the score would become the thing. If she stopped notating, the kitchen would become memory and then nothing.
She needed both. The killing was the carrying.
On the flight home she made one notation in her journal: a single measure from the oldest song, the three voices in their dense harmonic cluster. Below it she wrote: this is accurate. And below that: it isn't the song.
Both lines were true. Neither cancelled the other. That was the score's gift and its limit: it could hold everything except what made you want to write it down.