Purported Copies

The Purported Copies are a class of holdings, currently numbering 11,408 items across 212 floors, whose existence is asserted in the record while their physical presence is not. A holding is designated a Purported Copy when the catalogue assigns it a shelfmark, a floor, and a recension history, but no reader has produced it. The designation is administrative rather than material: a Purported Copy may be entirely absent, may exist and be misshelved, or may be a genuine copy whose original has also never been located. Librarians do not distinguish between these cases in the register, on the grounds that the distinction cannot be tested from inside the register.

The class was formalised by Ninth Concordance, Annex 7, which distinguishes a Purported Copy from a lost holding and from a holding never accessioned at all. The distinction has been contested at every Arbitration of Contradiction since 4012, largely because Annex 7 defines the category by what it is not, and the two things it is not are themselves defined by reference to it.

Origin in the Retrospective Procurement School

The category emerged from the Retrospective Procurement School, an office established to acquire holdings that the catalogue had already described. Its founding case was the Sanguine Quill, logged seven years before the date on which its manufacture was subsequently established. The School's position, recorded in its Founding Memorial, was that the register had anticipated the object and that procurement was therefore a matter of locating a copy already implied by the record. When the Quill was not located, the School did not amend the register. It entered the Quill as the first Purported Copy and amended the definition of the register instead.

"The record does not err when the shelf is empty. The shelf is the later and less reliable instrument."

— Founding Memorial of the Retrospective Procurement School, Clause 2

This reasoning drew directly on the Doctrine of Degenerate Copies, which holds that replication introduces corruption and each copy is less real than its predecessor. The School extended the doctrine one step: if the original is the least corrupt, the record that describes it is less corrupt still, since it cannot be damaged by handling. Under this reading a Purported Copy is not a defective holding but the most reliable form a holding can take. V. Selvage is reported to have called this "the only argument in the Tower that improves with every failure to check it" (Selvage Compendium, OV-774δ, marginal note 31).

Registration and the Second Copies

Purported Copies are deposited, like all recensed inventories, in the Archive of Second Copies, according to Clause 8, Annex 9. The practice is unusual in one respect: because there is no item to transfer, the deposit consists of the shelfmark card alone, placed in a card case of standard dimensions. Floor 812 holds the cards for the eastern range, and its wardens report that the cases are heavier than the holdings they describe, a discrepancy attributed to the pasteboard.

Floor 909, which holds the Weeping Codex and a number of derivative artifacts, contains 340 Purported Copies, the densest concentration in the Tower. The floor's inventory for 4417 lists them without annotation, followed by a single line recording that the reading room was closed for repointing. Addendum 142-γ restricts operational scope on floors above the Sealed Copies of Addendum III, which has been read to exclude the 909 cases from inspection entirely; the reading is disputed but has never been tested, because testing it would require producing the copies.

Notable cases

Three Purported Copies have generated proceedings of their own.

• PC-114, the second copy of The Perfect Treatise on the Breath of Statues, ascribed to P. Lintel. The original is held and legible. The Purported Copy is cited in eleven commentaries and quoted in none.
• PC-2,908, the reconstruction of WA-0, whose purported contents were ratified by Ordinance 412. Ordinance 412 ratifies the consensus reconstruction without confirming that any copy survives to be reconstructed from.
• PC-7,441, a copy of the Fundamental Link, entered after the Unspooled Chain Incident and never withdrawn, though the Link was a length of alloy and not a document.

The third case produced the ruling now cited in every dispute of the class. In 4433 the Archivist of the Eastern Range moved to strike PC-7,441 on the ground that a copy of a physical object cannot be a document. The Arbitration of Contradiction found for the register, holding that the entry described a copy and said nothing about what was copied.

The problem of the count

The figure of 11,408 is itself contested. The count is maintained by the Office of Enumeration, which tallies Purported Copies by counting cards in the Archive of Second Copies. In 4451 a clerk of that office, Harrow Vex, observed that 611 cards bore shelfmarks that resolved to other Purported Copies rather than to floors. On that reading the number of Purported Copies may be as low as 10,797, or the entire class may comprise a single copy described 611 times over. Vex's memorandum proposes that the Tower "hold the count open pending inspection," a formulation that has satisfied every subsequent reviewer and has been adopted as standing practice.

Inspection remains impossible for the ordinary reader, since a Purported Copy cannot be requested. The reading room's rule is stated plainly on the request desk: readers wishing to consult a Purported Copy must produce it at the desk first.

Relationship to the Transcription Cycle

The Transcription Cycle is frequently invoked in defence of the class, on the reasoning that a copy made from a copy made from a copy is already several removes from its original, and a Purported Copy is merely the limiting case. The Master Line and Verity Ink are both cited in the standard defence, the latter on the ground that ink which verifies its own accuracy has no need of a reader to confirm it. Whether this reasoning applies to classes held on floors above the Sealed Copies of Addendum III is unsettled; the question has been referred to the Librarians, who have not returned it.