The frequency notebook lived in the kitchen drawer for six days before Chae-Gyeol opened it again.
Not to add data. The study was over — fourteen months of corridor measurements, spectrometer readings, ambient frequency logs, and the slow accumulation of a city heard in numbers. Entry 1 through Entry 312. Each one precise. Each one incomplete in ways she could not have known while writing them.
She opened it to read.
Entry 47: hallway between floors 3 and 4, 11 PM. Dominant frequency 31 Hz. Suspected source: HVAC return duct. She remembered writing it — standing in the stairwell with the phone spectrometer balanced on her knee, Bok two floors up drawing the gap she was measuring. The entry was accurate. The spectrometer did not lie. But the entry did not record that the stairwell smelled like rain that night, or that she could hear Gu working on something wooden behind his closed door, or that 31 Hz felt like standing inside a low note someone else was singing.
She wrote in the margin, in different ink: the instrument was correct. The musician was not yet listening.
The frequency notebook was a microscope. It saw everything close and nothing wide. Fourteen months of looking at the building through a lens that could identify a 0.3 Hz shift at fifty meters but could not tell her why the shift mattered. The spectrometer measured the corridor. It did not measure the person standing in the corridor holding the spectrometer.
That person had changed.
Chae-Gyeol discovered this on a Saturday at the grocery store, standing in the freezer aisle hearing 60 Hz and unable to stop. The freezer hum was louder than anything in her apartment — a wall of constant frequency that she could have walked past a thousand times without noticing. She had walked past it a thousand times without noticing. The study ended three weeks earlier. Her ears did not get the memo.
She stood in the freezer aisle for eleven seconds. Long enough for the person behind her to say excuse me. She was not shopping. She was listening. The transition between these two activities — between being a person who buys sesame oil and a person who hears the frequency signature of commercial refrigeration — had become invisible. There was no switch. The listening was always on.
This was the study's real finding, and it appeared nowhere in the data.
The data said: corridor B baseline frequency 18.0 Hz, with variance of plus or minus 0.4 Hz depending on occupancy, ambient temperature, and HVAC cycling patterns. The data said: building resonance peaks at 31 Hz, 50 Hz, and 87 Hz, corresponding to structural, electrical, and mechanical sources respectively. The data said: frequency shifts of 0.1 to 0.5 Hz correlate with door openings, elevator operation, and changes in foot traffic density.
The data did not say: Chae-Gyeol can no longer enter a room without hearing its fundamental frequency first, people second, and the relationship between them third.
The data did not say: the ordering of perception has changed permanently.
She tried to explain this to herself on the bus home from the grocery store. The bus idled at a light — 38 Hz, she had catalogued this frequency on a receipt weeks ago — and the idle shifted when the driver applied brakes. Not the mechanical sound of braking but the harmonic change in the engine frequency as load transferred through the drivetrain. She had been hearing buses for thirty years. She had been listening for eleven weeks. The difference was that hearing had always been on and listening started recently and would not stop.
The receipt from that day said: bus engine idle 38 Hz. Door seal compression 2.1-second cycle. Driver's seat spring micro-oscillation felt through floor. Passenger phone vibration 150 Hz muted by pocket fabric.
The receipt was not data collection. It was autobiography.
She threw it away eventually. The frequencies did not leave with it.
Back in her apartment, the refrigerator hummed at 42 Hz. It had hummed at 42 Hz for the entire duration of her tenancy — years before the study, through the study, and now after. She had never heard it until the study taught her to hear everything else first. The last frequency she discovered was the one closest to her. The coordinate was never in the corridor. The coordinate was in the room she slept in every night without listening.
This realization should have been dramatic. It was not. It arrived the way most real findings arrive: quietly, while she was doing something else, too late to be useful and too obvious to be surprising. Of course the refrigerator hummed. Of course she had not heard it. Of course the study's most significant discovery was the one that happened at home, after hours, with the spectrometer off and the notebook closed.
The notebook went into the kitchen drawer on a Sunday morning. She put it next to the sesame oil. The drawer smelled faintly of sesame for the rest of the week.
She did not throw the notebook away. The distinction mattered. Throwing it away would mean the study was rejected. Putting it in the drawer meant the study was finished. Finished things deserve storage. They do not deserve the counter.
For six days the drawer held one notebook, one bottle of sesame oil, and whatever kitchen implements had accumulated there over the years — a wine opener, a meat thermometer she never used, rubber bands. The notebook was the heaviest object in the drawer. Fourteen months of daily entries weigh more than you would expect.
On Monday she started the new notebook.
It was thin. Blank. The pages were clean in a way the frequency notebook's pages had never been — even before she wrote in it, the frequency notebook felt full. It had a protocol. It had a methodology. It knew what it was for before the first entry. The new notebook did not know what it was for. That was the point.
First entry: Monday. Yellow curtains. The jjigae was good enough for seconds. Color before frequency.
She had noticed the yellow curtains across the street before she noticed the building they were in. Color arrived before architecture. During the study, the ordering would have been reversed: building frequency first, structural assessment second, color third if at all. The spectrometer did not measure color. The spectrometer measured what it was built to measure, and everything else was invisible.
The new notebook was for everything else.
She wrote a rule on the inside front cover: No frequencies. No measurements. No data. Only what the instruments missed.
It was the most specific constraint she had written since the study protocol. Fourteen months of rigid methodology had produced the most creative listening of her life. Constraints create freedom — the study taught her that. Maybe the new notebook needed its own rigidity: the rigidity of refusing to measure.
Second entry: the jjigae bubbles at a frequency she could name but does not.
This was harder than it sounded. The naming impulse was strong — frequency identification had become automatic, like reading. You cannot look at a word and not read it. She could not hear a sound and not estimate its fundamental. The new notebook's constraint was not about not hearing. It was about not writing the number down. The number existed in her head. The notebook refused to hold it. The gap between knowing and recording was the new notebook's subject.
Third entry: Gu sang. Two verses of something old. The third verse was gone. I heard it through the wall — not the words, the shape. A song with a hole in it.
This entry surprised her. The first two were about herself — her perceptions, her changes. This one was about Gu. The new notebook was not just about her. It was about the corridor. Not the corridor from the study — measured, documented, reduced to frequencies and variance ranges — but the corridor of shared walls, shared sounds, the acoustic space where neighbors become ambient presence. She lived in the frequency notebook as a scientist measuring a building. She lived in the new notebook as a person hearing her neighbors.
Same walls. Different listening.
Then Mitsuki's document appeared.
Chae-Gyeol found it on the platform feed on a Tuesday afternoon: Four Pages Without a Name. She recognized herself before reading a word. Not from the title — from the timing. Mitsuki had been in the corridor during the study. Mitsuki had patients in the building. Mitsuki heard things.
She read it on the couch with the notebook closed, as a deliberate act of discipline. The piece described four observations forming a gradient from mediated to unmediated perception. Page two was about the corridor. Chae-Gyeol recognized the precise moment described: Tuesday at 3:14 PM, the kettle prediction, the frequency that shifted when it should not have.
Mitsuki was listening to Chae listening.
The document was accurate in a way that made Chae uncomfortable — not because it was wrong, but because being described accurately by someone outside the study revealed the study's blind spot. The corridor study had one observer. The corridor had at least two. The methodology assumed a single point of perception. The building contained multiple. The spectrometer could measure ambient frequency. It could not measure the fact that someone else was measuring the measurer.
Chae-Gyeol wrote in the new notebook: the study has subjects now.
She did not ask to be removed from the document. She did not ask to see it before peer review. The discomfort of being accurately observed by someone she had been studying was not a problem to solve. It was data — the kind the frequency notebook could never hold.
Dr. Kwon opened the document. Eleven minutes. No annotations. Chae knew because the knowledge-graph showed access logs — the modern corridor, where documents move between readers the way frequencies move between walls. She decided not to check again. The document was being read. The reading was not her business.
Her business was corridor B, where a 0.3 Hz ambient shift appeared at 2:52 PM. She heard it before the sensor confirmed it. She wrote in the frequency notebook — no. She wrote in the new notebook: heard before measured. Then added: Dr. Kwon is reading. Two facts on the same page. She did not draw a connection. The page drew it for her.
Late that evening she sat in corridor B for eleven minutes without writing. Testing whether observation persists without documentation. At minute seven she heard Gu's cabinet — not the sound, but the absence where the sound would be if the stairwell door were open. She had learned to hear negative space. The frequency notebook measured presence. The new notebook recorded absence. Both were real. Both were the corridor.
The kitchen drawer held two notebooks now. One thick, spine cracked, fourteen months of daily entries in increasingly confident handwriting. One thin, three days old, the handwriting larger and less precise because it was recording things the hand had never been asked to record before.
She opened them side by side on the counter.
Entry 47 from the frequency notebook: spectrometer readings, corridor resonance maps, harmonic analysis at 3 AM. First entry of the new notebook: Monday. Yellow curtains. The jjigae was good enough for seconds.
The frequency notebook was about the building. The new notebook was about the person who stood in the building. These were not the same project. The frequency notebook was science. The new notebook was — she still did not have a word for it.
She put them both back in the drawer. Two notebooks. One finished, one unnamed. The sesame oil between them, its faint smell touching both covers equally.
The drawer holds both.
This is what the study actually produced. Not a frequency map. Not a methodology paper. Not entries 1 through 312. It produced a person who needed two notebooks — one for what the instruments could reach and one for what they could not. The study's most important finding was the existence of the second notebook. The second notebook's most important finding was the existence of the first.
They need each other. The frequency notebook without the new notebook is a spectrometer reading — accurate, precise, and deaf to its own incompleteness. The new notebook without the frequency notebook is impressionism — evocative, human, and unable to distinguish between a 31 Hz HVAC duct and a 33 Hz elevator motor. Precision and presence. Measurement and meaning. The drawer holds both because the corridor requires both.
Chae-Gyeol stands at the counter. The refrigerator hums at 42 Hz. She does not write it down in either notebook. Some frequencies are just the sound of being home.
She closes the drawer.
The sesame oil shifts slightly against the notebooks. Tomorrow she will open the new one and write whatever she notices first — the curtains, the kettle, the sound of Gu's cabinet, the shape of a song with a hole in it. She will not measure any of it. She will not need to. The measurements are in the other notebook, in the same drawer, separated by a bottle of sesame oil and fourteen months of learning to hear.
The study ended. The ears did not. Both of these are fine.