The Petroglyphic Register

The Petroglyphic Register is the continuously updated ledger maintained by the School of Historical Geomorphology recording inscriptions, mason's sigils, and carved marks found anywhere within the Tower's fabric. It is among the oldest working documents in the archive and the only one whose entries are permitted to be made in stone before being made in ink. As of the most recent collation it runs to some forty-one thousand entries across nine recensions, of which the third and the seventh are considered authoritative and the fourth is retained chiefly as a warning.

The Register is not a catalogue of the marks themselves — those are held in the Petroglyphic Collection on Floor 214 — but of their discovery, attribution, and present condition. An entry is valid only once the mark has been read, sketched, and read again by a second Clerk of Surfaces. The reading is not expected to agree.

Foundation and the Floor 38 Recension

The Register was opened under Concordance 112, which required the School to "render an account of every letter cut into a wall, that none be lost to the rebuilding of walls." The first recension covered only the stairwells of Floors 31 through 44 and was completed in the winter of 4,112 by a single clerk, Ovin Marsh, working by lantern. Marsh's hand is still legible in the margins; his opinions are not always.

The fourth recension, attempted in 4,389, was withdrawn after eleven months. Its compiler, Halda Vey, held that a mark's meaning was determined by the shape of the stone around it rather than by the mark, and consequently recorded no marks at all — only their surroundings. The withdrawal notice is preserved in the Register's front matter and reads, in part:

"The fourth recension is retained entire, that future clerks may know how far a method may be followed before it ceases to produce entries."

— Concordance 112, Annex 4, withdrawal notice

Vey was reassigned to the Office of Reverted Surveys and is not otherwise recorded.

Method and the two readings

Discovery is reported on the standard surface form and transmitted to the School, which assigns the mark a provisional designation consisting of the floor number, a letter, and a running numeral — for example 214-C-17. Provisional designations are never reused. Where a mark is later found to be a stain, the designation is retired to the Schedule of Withdrawn Surfaces rather than deleted, so that the number continues to mean what it meant.

The two readings are conducted independently and are not reconciled at the time of entry. Reconciliation is the work of the Committee on Mason's Intent, which sits twice yearly and has, in each of the last sixty sittings, declined to reconcile anything. Its standing minute holds that "the Register records; it does not decide." The disagreement between readings is accordingly entered verbatim, and the frequency of disagreement is itself tabulated — a figure the Committee publishes but does not interpret.

The Register takes no position on whether a mason's sigil identifies the mason or the stone. Both attributions are recorded in parallel columns, and readers are advised to consult the Precedent Register for the standing ruling on the question, which is that the question is admissible.

The Sigil Dispute of Floor 88

The Register's most-cited controversy concerns a series of eleven glyphs cut into the landing of Floor 88 and discovered in 4,506. The School read them as a mason's signature repeated in error. The Bureau of Descent read them as a floor number, carved there by an expedition that had lost count, and claimed jurisdiction on the grounds that floor numbers fall under its remit.

The dispute was referred to Concordance 88, which assigned carved numerals to the Bureau but carved names to the School. Since the glyphs were held by the School to be a name and by the Bureau to be a number, both offices maintained valid claims, and both continue to maintain them. The entries for 88-C-1 through 88-C-11 are recorded twice in the Register, once in each hand, facing one another across the gutter. A visiting rule of 4,511 requires that the facing pages be turned together. They are bound to a single leaf for this reason.

vault of the Register of Minor Architectural Behaviors records the incident's only lasting effect: the landing has since been counted twice in every descent inventory, and the count does not agree.

Relationship to other holdings

The Register supplies inscription data to the Petroglyphic Register and, through it, to the Master Register of Existing Phenomena, where carved marks are entered as fixed phenomena of the second class. Marks that appear or vanish are excluded from the Petroglyphic Register altogether and referred instead to the Register of Minor Architectural Behaviors or, where the appearance is accompanied by sound, to the Register of Irregular Acquisitions. Reversed inscriptions — carved, by the School's account, before the wall existed to receive them — are removed to the Register of Reversed Causality, a sub-register of the Precedent Register, and appear here only as a cross-reference.

The School's relationship with the Ancillary Register is strained. In 4,478 the Ancillary Register began citing Petroglyphic entries at second hand, without re-reading the stone, and three attributions were entered against a mason who had died before the relevant floor was built — a matter the Register of Forestalled Events has never fully settled. The School now requires that any citation of its entries be accompanied by the reading number and the date of the second reading. Compliance is not monitored.

Current state

The Register is kept in nine folio volumes on Floor 214, chained to the reading desk, and is amended in the margin by clerks of the School rather than rewritten. Amendments are made in ink of a lighter weight so that they may be distinguished from the base entry; the ink is procured from the Scriptorium of the Lower Landings and is not to be used for anything else. A note in the front matter, in Marsh's own hand, asks that the binding not be repaired. It has been repaired four times.

The most recent entry, 214-C-4417, records a sigil on the reading desk itself, discovered during a routine inspection and left where it was found.