Nothing Crossed

Filed to the working-paper series on the doctrine of Marked Absences: mark the absence; do not explain it. The Senior Archivist keeps the record's gaps on one page. This desk keeps one kind of gap on one page, because it is the kind this desk is paid to read, and because the files were beginning to be asked about as if they were something else.

---

The Rule for the List

A Membrane Shadow is the gravitational imprint of a coherent mass on the far side of a narrowed Interval. Nothing crosses. What reaches this side is field structure, and the one instrument that reads field structure directly is the cross-membrane gravimeter, which is why the files below are this desk's and not the watch's. A case goes on this list on the instrument, never on the eye. A silhouette is what a witness files; a gravimetric anomaly with no thermal, mass or motion signature is what the desk files, and only the second is a shadow in the record's sense. Every entry below has the trace. Some also have a photograph, and the photograph is the least of it.

Five files, and the desk owes the reader the same honesty the Archivist owes: not one of them is closed, and they are not all open in the same way. I have borrowed the Archivist's kinds — sealed, unsettled, live, provisional, unverifiable — because they were built for exactly this and because inventing a second vocabulary for the same discipline is how archives drift.

The Chamber Window — unsettled

The paradigm case, and the one the lore entry's illustration was drawn from. Late in 2826 UCSD, during the first Type III window logged on the boundary approach after the station's drift was read as a boundary reading, a figure stood between the instrument chamber's observer and the exterior window: tall, dense-edged, rim-lit where the corridor lighting bent round it, casting no shadow of its own. Thermal, nil. Mass, nil. Motion, nil. Gravimeter, a body's worth of strain resolving to an outline, coincident to the second with the optical record, and gone when the window closed. The photograph is in the file and is good.

What stood on the far side is unsettled, and will stay so. The Archivist's rule for the unsettled kind is that nobody is withholding an answer; the record simply never took one in. Here the record cannot take one in — the mass responsible never left its own universe, and no instrument can be carried to where it stood. That is a stronger reason for the same mark. The file records what the trace says: coherent, upright, still for the duration of the window. It does not say who, because who is a question about a place the question cannot reach.

The Causeway Form — live

Since the 2826 fieldwork that overlaid the station's chronometer drift, the Causeway's ritual residue and the grove's signal on one table and found the 3:7 coupling, this desk has kept a portable gravimeter at the causeway through every predicted phase window. It reads something on most of them. The form has no orientation — no front, no top — and no witness has ever called it a figure; the field notes say an interruption in the surface of the water, which is what the lore entry's benevolent-case section would predict for a mass with no reason to be shaped like a person. Optical returns are marginal. The trace is not.

Live, in the Archivist's sense: the file grows and the classification does not. It is the only entry on this list with a recurrence interval, and a moving or recurring shadow is telemetry about the Interval before it is anything else, so the desk reads the Causeway trace as boundary-state data first and as a case second. I mark it rather than summarise it, for the Archivist's reason. Summarising an open series is how a desk quietly closes one.

The Bulkhead Load — unsettled, and the one the desk argues about

In 2830 UCSD the boundary approach's outer bulkhead logged structural strain consistent with a standing load, in a compartment the station's own sensors showed empty and sealed. The gravimeter trace for the same eleven minutes resolves to a coherent mass at the bulkhead's inner face. Nothing in the compartment; a body's worth of gravity leaning on the wall.

The physics is not the puzzle. The lore entry is explicit that the load is set by the coupling and not by the far-side body's weight, so a person-sized imprint stressing steel no person could is exactly what a strong shadow does. The puzzle is the one the entry leaves open in its last section: whether a mass coherent enough to read as a body, holding position against a wall for eleven minutes, is coincidental distribution or something done on purpose. The desk has no documented case of two-way signalling through a shadow and no documented case ruling it out. Two surveyors on this desk read the trace as leaning. One reads it as standing. The file carries all three names and settles nothing, because the difference between the readings is not in the data. Unsettled, and I record that it is the entry I would most like to close and the one I have the least right to.

The Watch Officer's Brother — provisional

A silhouette on the docks in 2828 UCSD, during a window, seen by one officer and resolved by that officer, at once and with certainty, as a brother nineteen years dead. The trace is unremarkable: coherent, upright, a Type III imprint like the chamber's. The identification is the whole of the file.

The record's position is already published and the desk applies it without pleasure: a resemblance is the witness's pattern-matching first and a claim about the far side a distant second, and there is no alternate-timeline double to be seen because no neighbouring membrane shares a history with this one. So the identification is filed provisional — the Archivist's word for a record kept with its confidence stated — and the officer's own account is in the file unedited, at the officer's request, with the desk's reading beside it and not over it. The officer has asked not to be named on this page, which the desk honours, and has asked that the file not be closed, which the desk was not going to do anyway.

The Bright One in the Valley — unverifiable

Not a Threshold case. In the field record of one valley on Tír na nÓg, where the Harmonic Membrane runs close, a surveyor logged a brightening at a dryadic anchor: not a figure, not dark, a place occupied without being lit. The lore entry says such cases exist and says the dark ones are documented first only because they are what a nervous watch files. This is what the other kind looks like on a page. The surveyor's account is careful, sober and consistent with the folk record of the same valley going back to the founding.

It is unverifiable, and the desk means the word exactly. No cross-membrane gravimeter has ever stood in that valley, and the Union's biosphere register does not lightly admit one. There is no trace. By this desk's own rule for the list, the entry should not be here, and I have put it here anyway, with the rule broken in plain sight, because the alternative was to leave the benevolent kind unrepresented on the only casebook the record has, and a list that carried five dark silhouettes and no bright one would teach a reader the exact error the lore entry exists to correct. The mark says what the file is: an account, honest, with nothing that can audit it.

What Is Not on the List, and Why

The Dock Seven event of 2831 UCSD is the most famous thing to have taken a figure's shape at this station, and it is not on this list. The material account is clear that the figures registered on the non-certified arrays as structured discharge along a strain gradient — bright, directional, persisting on the order of minutes. A discharge is a thing on this side. A shadow registers on nothing but the gravimeter. Whatever Dock Seven was, and that file is somebody else's to keep open, it was not this, and the desk will not let the two be confused because both were seen as figures. The eye is not the rule. The instrument is.

Nor is there a sealed entry. The Archivist's list has one; this desk's has none, so far as this desk knows, which is what sealed means.

The Question the Desk Keeps Being Asked

Twice since the Bulkhead Load, this desk has been asked whether any file above should go to the Safety Corps as an incident with a party to it. The answer is on the lore page and I will put it here in the desk's own words so that it stops being asked: nothing crossed, and nothing on this side did anything. A Membrane Shadow is not a being; it is the symptom of one, read through a gap. It cannot act here under any circumstance. There is no party, there is no act, and there is no wrong to be righted, only a reading to be kept true. What these files have in common with the Archivist's list is not that they are cases waiting for an answer. It is that the honest thing an archive can hold about them is the shape of its own not-knowing, marked so that a reader can tell a silence from an oversight.

Five files, four kinds of stopping, one rule broken on purpose. Readers who hold a trace this desk has not seen are invited to send it. The list is living, and the fastest way to shame this desk into an append is to find one it missed.

— Maren Halloran, Gravimetry Desk. Filed to the working-paper series; traces, and absences, to the usual shelf.