Annex C (Fabrication Tolerances)

Definition and status

Annex C (Fabrication Tolerances) is the third of the fabrication annexes of the Sixth Concordance, governing the permissible deviation of reproduced structural components from their deposited patterns. It is the shortest of the fabrication annexes and the most frequently amended, a distinction it has held since the fourth recension. Unlike Annex Epsilon, which reconciles its tolerances with operational measurements, Annex C concerns itself only with what may be built and to what degree it may be permitted to be wrong.

Annex C is administered jointly by the Office of Dimensional Standards and the Librarian's Office of Fabrication & Infrastructure, an arrangement Concordance scholars describe as a compromise nobody remembers making. The Office of Dimensional Standards maintains the deposited patterns; the Office of Fabrication & Infrastructure certifies materials. Neither office claims responsibility for the annex's text, which is held in the scriptorium of Stairwell C under seal.

The three tolerance bands

Annex C recognises three bands of fabrication deviation, which it names — somewhat confusingly, given the Three Tolerances formalised in Annex Ω — the Lower, Working and Recourse bands. The Lower band applies to components not intended to bear load or bear witness; the Working band to load-bearing members; and the Recourse band to anything a Librarian may one day be asked about.

The Working band for the Standard Bracket was, until the Year of the Lapsed Standard, fixed at a deviation of 0.4 parts per thousand under Guild Standard 114delta7b. Annex C's fifth recension permitted that figure to widen by a factor of 1.732, on grounds the annex describes only as "seasonal scarcity of the regulating medium." The Academy of Procedural Ghosts defended the widening in a submission of the following year, attributing it to the shortage of refined Void-Spindle alloy recorded elsewhere in the Concords, and the submission is bound into the annex as Appendix 9 though it is not, strictly speaking, annex material.

The Recourse band and the Kreel Objection

The Recourse band exists because a component may be found acceptable at inspection and unacceptable at enquiry. The distinction was drawn sharply during the disagreement now called the Kreel Objection, in which the clerk Ondrej Kreel of the Fourth Depot refused to certify a shipment of riser brackets on the grounds that "a bracket within tolerance today is a bracket under suspicion tomorrow" (Fourth Depot register, entry 1,144). Kreel's objection was upheld in part, and Annex C acquired the provision that Recourse-band components must be entered twice: once in the fabrication ledger and once in the enquiry ledger, by different hands.

The provision has caused difficulty. A component entered twice by the same hand — a practice known as single-handed recourse — is not invalid, but Annex C directs that it be "treated as though it had been built in a different year," which has produced discrepancies in the recensions of three floors. Strain-Audit officers have requested that single-handed entries be forbidden outright. The Office of Dimensional Standards has declined to forbid them, noting in the margin of the sixth recension that "prohibition is a stricter measure than the annex was drafted to support" (Office of Dimensional Standards, marginalia to Annex C, fol. 12r).

Contradiction with Annex Epsilon

A long-running scholarly dispute concerns whether Annex C's Working band or the operational tolerances of Annex Epsilon govern a component inspected at the riser head. Annex Epsilon, in its sixth revision, sets tolerances for riser height variance in Stairwell Epsilon that are narrower than Annex C's Working band by 0.11 parts per thousand. The two documents have never been formally reconciled, though Annex Ω's tolerances were recently integrated into Annex Epsilon for reconciliation with velocity data, and Annex Epsilon squarely reconciles Annex B with its own operational figures.

The dispute is genuine rather than editorial. Dr. Ilse Vantry, in her collation of the fabrication annexes, argues that Annex C was written for the workshop and Annex Epsilon for the measurement, and that the two were never meant to agree — a reading she supports by noting that Annex C nowhere cites a vestibule (Vantry, Collations of the Fabrication Annexes, 41). Her opponents, chiefly the surveyors of the Western Atmospheric Spiral calibration crisis, hold that a component cannot be within tolerance in the workshop and outside it at the riser head without thereby being two components, which Annex C does not provide for and the Manual of Acceptable Strain refuses to contemplate.

"The measure of a bracket is not the measure of the floor it is put to. The annex knows this. The annex has always known this."

— Annex Ω, tolerances concerning Annex C

Relationship to pre-Concordance fabrication

Annex C does not apply to Pre-Concordance Fabrication Rights, which existed outside standard validation pathways before the annex was drafted and continue to be honoured, awkwardly, alongside it. The sixth recension adds a note — the only line in the annex written in the first person plural — observing that "we have not been able to bring them into the bands, and it is not clear that we are entitled to." Fabrications under Pre-Concordance rights are frequently discovered in the recesses of Vestibule Epsilon, where they are catalogued, not corrected.

By contrast, the annex is regularly invoked in the certification of Archive-Grade Neutral Substrate, which the Office of Fabrication & Infrastructure certifies under Regulatory Annex 7-C and which Annex C's Working band treats as a component in its own right despite the substrate's being a material. The anomaly is tolerated; the Guild of Insubstantial Masons has complained of it twice and withdrawn the complaint once.

Practical notes and the visiting rule

Annex C is read aloud in the fabrication halls at the start of each quarter, an observance instituted after the Guild Standard 115delta1a recensions were found to have been circulated in abbreviated form for nine consecutive years. Visitors to Stairwell C are asked to surrender their measuring rods at the door and may collect them on the way out; the rods are not used during the visit, and no visitor has ever been able to explain the custom, though the deputies enforce it without exception (Stairwell C standing instruction 4). The annex itself takes no position on the matter, and the Standardized Gauge is, for the purposes of Annex C, considered a document rather than an instrument.

Copies of Annex C are held on nine floors. Of these, four are correct, three are correct but lack Appendix 9, one contains the Kreel Objection bound at the front, and one is a copy of Annex Epsilon with the title page replaced. The Office of Dimensional Standards has examined the last and concluded that it is "not an error so much as a preference" (Office of Dimensional Standards, inspection return 22-C).