Simultaneous Referential Collapse Drills

Definition and purpose

Simultaneous Referential Collapse Drills are scheduled exercises in which the stacks of a single floor are deliberately subjected to conditions approximating those described by Clerk Aris in On Referential Collapse, in order to test the readiness of staff, shelving, and cross-reference apparatus. The practice is authorised under Division IV's standing mandate and is administered by the Bureau of Referential Integrity. Unlike the phenomenon itself, which is by definition uncontrollable, a drill is convened at a stated hour, concluded at a stated hour, and recorded in a ledger that is later inspected.

The distinction matters because the Tower's experience of genuine collapse has been sparse and expensive. The Collapse of the Western Stack on 17 Serene 891 remains the most thoroughly documented instance, and the drill programme is generally understood as an attempt to make that date repeatable in advance and therefore survivable.

Origins

The first drill was held on the floor numbered 4-7-1 (the designation is contested; see below) in the winter of 903, convened by Warden Selk of Division IV after a near-miss in which a reading room briefly lost agreement with its own catalogue. Selk's memorandum to the Bureau survives and is quoted frequently:

"We have procedures for a collapse that has already occurred and none for one that is merely considering it. The second is the more common condition of this archive."

— Warden Selk, memorandum to the Bureau of Referential Integrity, 11 Deepwinter 903

The Bureau approved a pilot series the following year, initially four drills per cycle, expanded to twelve after the Shelf Collapse of Δ-88 demonstrated that untested floors responded poorly to unexpected referential stress.

Conduct of a drill

A drill proceeds in three phases, each timed. In the first, the Cross-Referential Anchoring lines serving the floor are slackened by a measured degree — never fully severed, though a clause in the original authorisation does not specify by how much, which has produced three separate operational readings over the decades. In the second, staff are asked to perform ordinary catalogue work under the degraded conditions. In the third, anchors are restored and the floor is reconciled against its own holdings.

The second phase is where the exercise earns its reputation. Clerks report that the work itself does not become difficult so much as it becomes unarguable: a call slip will request a volume that the shelf insists was never there, while the shelf simultaneously insists the slip was never issued. Participants are instructed to log these episodes rather than resolve them. The logs are the point.

Rehearsal days are closed to outside readers, and no clerk is permitted to enter a drill floor carrying a personal index.

The designation dispute

Two of the Concordances disagree about which floor hosts the programme's standard drills. The Fourth Concordance names floor 4-7-1; an annex to the Eleventh Concordance names floor 7-1-4. Neither number is sequential with the other in any scheme the Tower has been able to reconstruct, and the Bureau has declined to adjudicate on the grounds that the drills are, by design, held somewhere else every time. A standing committee was convened in 918 to settle the matter and has met eleven times.

The disagreement is not merely clerical. Because the drills are meant to approximate a floor-wide event, the choice of floor determines which scriptorium maintains the anchor apparatus and therefore which office receives the reports — a matter of some consequence to the parties.

Notable incidents

The drill of 916 is the best-known failure. During phase two, the degraded conditions on the drill floor propagated one level down into a normally operating reading room, where a reader's catalogue card and the book it described ceased to agree for approximately nine minutes. The incident produced the Bureau's only formal restriction on drill timing: exercises are now suspended during the hour of the low bell.

In 921 a drill was convened during an active Simultaneous Veracity event on an adjacent floor, which the convening clerk later described as "an enthusiasm." No one was injured, but the reconciliation phase took four days and required reference to Simultaneous Amendments in order to determine which of the two floor-states had been the drill.

Relation to theory

The drills are sometimes criticised by scholars of Aris for treating a theoretical model as a rehearsalable script. The counter-argument, advanced most forcefully by the Bureau, is that the treatise describes a threshold rather than an event, and that floors which have been pushed to just below a threshold repeatedly behave differently from floors pushed over it once. Critics note that the second claim has never been tested, because floors pushed over it are not in a condition to file reports.

Whatever the theoretical merits, the drill programme has produced the largest continuous body of pre-collapse observation in the Tower's holdings — some forty volumes of logs, shelved under a designation that three separate catalogues give differently.