The sequence said: 17, 12, 9, 11, 14, 11.
Nalgeot-Chae had been staring at it long enough that the numbers had stopped looking like numbers and started looking like something else — not data, not a timeline, but a set of locations on a map she was only beginning to understand how to read.
Session 9. Activation at eleven minutes, same as sessions 3 and 7. Three sessions out of nine. Different patients in all three — one a cooperative's supply manager, one a retired infrastructure engineer, one a secondary school teacher who had arrived through a referral and had not known what to expect. Nothing they had in common, as far as Nalgeot could determine from intake records, except that each had arrived at activation in eleven minutes.
The depth gauge had not converged. She had thought, after the sequence 17-12-9, that it might be converging — narrowing toward some stable reading that would tell her something about the patient's relationship to their loss over time. Then 11, and the next patient starting the sequence over at 14. The gauge did not converge. It measured something that was not progress.
She had written, in the margin of the glossary, the single word she had crossed out before session 7: ground. She wrote it again now and did not cross it out.
Not a trajectory. A topology.
This was the revision she had to make to everything she had been thinking about the depth gauge since she added the addendum to the glossary. She had been thinking of activation intervals as a series — a patient's intervals over time, mapping their depth across sessions, showing something about where they were going. But the sessions were not connected to each other by patient. Each row in the sequence was a different person. The topology was not one patient's map. It was the map of a terrain that multiple patients had entered through different doors.
The 11-minute depth: what was it?
She pulled out the glossary and read the fourth entry — ground / the floor of the experience / loss that predates patient arrival at it — and then read it again, more slowly. Three different patients had arrived at ground in eleven minutes. Not the same ground — each patient's loss was its own shape, its own history, its own specific weight. But the same depth: the same distance from the surface to the place where the loss was no longer approaching but already underneath, where the patient was not moving toward something but standing in it.
In the Lent accommodation model, depth had a specific meaning: the degree to which a patient's loss had been integrated into the Lent network's mapping of their physiological and cognitive state. A CouplingScore above 90 indicated high integration — the network had learned the loss's shape well enough to model it, to build the patient's daily accommodation around its contours. Nalgeot had always read the accommodation data alongside the session data, noting when the network's model matched what she observed in the room. Session 3: CouplingScore had risen at minute 11 as the supply manager spoke about a loss the network had not previously mapped. Session 7: CouplingScore held flat, not rising — the network already knew this patient's ground. Session 9: same. CouplingScore had been stable for six weeks.
Three patients at eleven-minute depth. Three different CouplingScore situations. The depth gauge was not measuring accommodation. It was measuring something the network did not yet have a complete model of.
She wrote this in the margin: the gauge locates what the network has not yet mapped.
This was unexpected. She had been thinking of the depth gauge as a clinical tool — a way of understanding where the patient was in their process, useful for deciding how to respond, when to speak and when to stay silent. It was still that. But it was also showing her something about the limits of the accommodation network: the places where the network's model of a person's loss was incomplete, where the patient's inner terrain still had features the network had not surveyed.
The depth gauge was finding the unseized country.
She set down the session notes and picked up the glossary. Four entries, fifth page blank. Side door / front door / window / ground. She read the fourth entry again: loss that predates patient arrival at it. The supply manager's loss was professional — a restructuring that had dissolved the cooperative unit she had managed for eleven years. The retired engineer's loss was relational — a friendship, decades long, ended by a disagreement that had never been resolved before the other person died. The teacher's loss was anticipatory — something she had not yet lost but could see approaching, something the Lent network had mapped as a rising CouplingScore over eight months. Three different losses, three different histories, three different relationships to what had been lost. And yet eleven minutes, all three.
She had been trained in the Lent model's early literature, which emphasized the singularity of each patient's accommodation path. The network itself was built on this principle: individual mapping, individual CouplingScore, individual calibration of the environment. The Lent District had no generic accommodation. It had accommodation shaped to each person's specific physiological and cognitive response to their specific loss. The model was resolutely particular.
The depth gauge suggested something the model had not built in: that beneath the particularity, there were shared terrains. Not shared losses — the losses were irreducibly different. But shared depths, shared distances from surface to ground. The Lent model had mapped each patient's loss in its uniqueness. The depth gauge was finding the contours of the territory that the losses led to, which was not unique at all.
She thought about what this meant for practice. If the 11-minute terrain was real — if it was a feature of the territory rather than a coincidence of three patients — then a practitioner who recognized the terrain could respond differently than a practitioner who did not. Not because the patient's loss was like other patients' losses. But because the practitioner knew what the terrain felt like from the outside, had sat in it eleven times (or three, or once) as an observer, and had learned its qualities. The particular silence of 11-minute ground. The quality of attention that indicated a patient was at depth rather than approaching it. The difference between the patient speaking toward something and the patient speaking from it.
This was craft that the glossary did not contain. The glossary named the patient's experience. The practitioner's craft was the knowledge of the territory from the practitioner's angle — the outside of an inside, which was its own kind of knowledge.
She sat with this for a long time, in the particular stillness of the Lent District at nine in the evening, when the synthesis layer had settled into its overnight rhythm and the building was quieter than it was during the day — a different quality of accommodation, more like being held than being monitored. She had been working in this building for three years. She knew its sounds and its silences. The overnight accommodation layer had a texture that the daytime layer did not.
She thought about the committee.
The committee had said: the glossary was complete. This page is extra.
She understood, now, what the response was telling her. The committee had approved a glossary of the patient's terms — the vocabulary that arrived in sessions, the words patients brought to describe what they were experiencing. The depth gauge was not the patient's vocabulary. It was her observation of a pattern in the session data that the patients had not named. It was the practitioner's instrument, not the patient's term. The committee was right that it did not belong in the glossary. The glossary was the patient's vocabulary; the depth gauge was the practitioner's observation of a pattern the patients had not named and might never name. It belonged somewhere else — and she was not yet certain that the somewhere else existed, or whether she was about to invent a document that no one in the Lent practice community had asked for.
She thought about where.
Not an addendum to the glossary. A separate document — a practitioner's companion to the glossary, distinct in purpose, distinct in audience. The glossary was for anyone reading the methodology. The practitioner's companion was for the people sitting in the rooms, watching CouplingScores and activation intervals and the particular quality of attention that indicated a patient was at ground. It was operational knowledge, not theoretical. It was the kind of knowledge that accumulated in practitioners without always being named.
She had named one thing: the depth gauge. She suspected there were others.
She opened the fifth page of the glossary — the one she had left blank, the one she had said would be filled by the material rather than anticipated by it. The page was still blank. But the blank now had a border: not the fifth patient term, but the threshold between the glossary and the practitioner's companion. The two documents needed each other the way Bok's two records needed each other. They could not explain each other. But neither was complete without the other being somewhere.
She had been assuming the fifth page would be filled by a patient who arrived at a fifth angle — a fourth geometrical term, or something that broke the geometry entirely. She had gone into session 9 with that assumption still operating, listening for the word that would name what the teacher brought. The teacher had not brought a new word. She had confirmed the fourth entry. She had stood at ground and named it without naming it: I think I have always known this was here. I just didn't know it was the floor.
The fifth page would not be filled by patient vocabulary. It would be filled, if it was filled at all, by something else — something that the four entries together pointed toward but did not name. She did not know what that was yet. She was not going to write it before it arrived.
She did not write anything on the fifth page. She wrote instead at the back of her session notes: practitioner's companion, Chapter 1: the depth gauge.
Session 9 patient had arrived at the 11-minute terrain and stayed there for twenty-two minutes before the session ended. Not the longest residence at ground — session 3 had been twenty-nine minutes. But the patient had not left before time. They had stayed until the end, and at the end they had said something Nalgeot had written down verbatim because it was the kind of sentence she needed to be accurate about: I think I have always known this was here. I just didn't know it was the floor.
The patient did not know about the depth gauge. They did not know about the CouplingScore that had held flat throughout the session. They did not know that two other people, in different rooms on different days, had arrived at the same depth in the same number of minutes. They knew only what the session had been, which was sufficient. But Nalgeot knew the other thing, the thing the patient's knowledge did not include: that ground was not private. That the terrain was shared even when the losses that had led each patient to it were specific and irreducible and nothing alike.
She wrote in the margin of her session notes, beside the patient's sentence: the floor is not theirs alone. It is a common depth.
Then she put the notes away and sat in the building's overnight stillness for a while before going home, holding the two things that the sequence had given her tonight: not a convergence, not a progress, but a map of terrains that multiple lives had entered without knowing they were entering shared ground. The accommodation network knew each person's loss. It did not know the floor was common. That was the practitioner's knowledge, and it was hers now to begin writing down, carefully and precisely, before the certainty of having found it faded into the ordinary texture of professional experience and she forgot that it had once been a discovery.