governance: steward places REVIEWED-133 (PENDING-137)

Cell-constant markers: ratified as its own amendment, with the direction
of the change recorded. AUTHORIZED — (b) as corrected, three conditions;
(a) and (c) rejected. Closes a narrowing that had been in force by
reading since 2026-08-07 and undisclosed as an amendment until now.

Executed the same evening in studium-engine 2b30425.

Committed by the executor at the steward's direction; the content is the
jurist's and the steward's, and REVIEWED.md was written by neither the
executor's hand nor its judgement.

⚠ The entry records its own weakness in its Notes, and it should not be
read cold as an ordinary ruling: two of its three conditions were
corrected by the executor after the jurist ruled from a partial read of
the file it governs (it had read to line 70; the deciding date rows sit
at 127-128), the stratum distribution in condition 2 is executor
testimony from rows the jurist did not open, and governance-mcp.py's
selftest was FAILING while the ruling was being drafted from that
surface — five undeclared mutating calls in wake-digest.py from the
previous day's build, repaired at cc97888.

⚠ The header is the parenthetical house form `## REVIEWED-133
(PENDING-137) — …`, so the register-integrity control cannot resolve it
to a number: RE_ID requires a dash immediately after the identifier and
backtracks to the bare token `REVIEWED`. Fourth entry in that class after
130, 131 and 132. Harmless here; it means any future amendment to this
entry will report the same false orphan seen today. Recorded so it is not
rediscovered cold.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017vKkg2EJF1rGwdFdBogwqx

REVIEWED.md

@@ -2537,3 +2537,79 @@ dir reads as absence, not as safety.
 8. **Scope of this ruling.** The line numbers, the census table, the fixture contents and the append-and-recount demonstration are executor testimony from a script I cannot run. What I verified directly: REVIEWED-122 condition 5 verbatim, PENDING-110 option (c) verbatim, PENDING-145 and PENDING-146 verbatim, and the behaviour of `governance_item` at §7.
 
 **If AUTHORIZED:** Implement (a) per conditions 1–4. Report the enumeration in full, agreements included. Do not touch placed headers. Cross-reference PENDING-146 in the commit and in the item's kin list.
+
+## REVIEWED-133 (PENDING-137) — Cell-constant markers: ratified as its own amendment, with the direction of the change recorded
+**Date:** 2026-09-01
+**Decision:** AUTHORIZED — (b) as corrected, with three conditions. (a) and (c) rejected.
+**Rules on:** PENDING-137, per REVIEWED-121 point 2.
+
+**Notes:** The reasoning is sound and is not in question: a marker constant across a cell
+cannot stratify within it, and the literal reading leaves the fr/de stratification carrying
+no information. The item's own superseded-in-place correction — that an amendment is
+constituted by its disclosure and cannot be dated to a day on which none occurred — is
+adopted as the governing principle here. Recorded because it bears on how much weight this
+ruling can carry: two of its conditions were corrected by the executor after I ruled from a
+partial read of the file, and the third rests on a fact only the executor established.
+
+1. CONDITION — **Delete `undisclosed_days_in_force`; rule the endpoint as the file has it.**
+   `date_in_force: 2026-08-07` and `date_disclosed: 2026-08-14` stand, on the file's own
+   stated ground — 2026-08-14 is the first disclosure of the narrowing anywhere, and the
+   register entry is a disclosure even though the amendment is constituted at ratification.
+   The derived field is removed rather than corrected: a computed span pasted beside the two
+   inputs that produce it goes stale while its inputs stay right, which is the defect class
+   and not an instance of it. The span is read off the dates.
+   ⚠ **This corrects my own draft, which would have moved `date_disclosed` to the ruling date
+   and overturned a documented choice while presenting itself as refreshing a stale number.
+   I had read to line 70; the two date rows sit at 127–128.**
+
+2. CONDITION — **The amendment records the DIRECTION of the change, marked as inference.**
+   Under the literal reading, A:B ≈ 1:1 is unsatisfiable in fr and de; under the narrowing it
+   is satisfiable. The amendment converts an unpassable bar into a passable one, and it was
+   made after the cell was seen — the direction REVIEWED-116 point 1 guards. It is ratifiable
+   because the ground is structural (a constant carries no information) and would hold
+   identically had the effect run the other way.
+   ⚠ **The effect cannot be measured, and the disclosure must say so.** The fr cell records
+   only post-narrowing markers — F5 5/11, F8 3/11, F3 3/11, D-b 1/11 — and cross-lingual and
+   archaic register appear nowhere, because this narrowing is what removed them. The
+   before-state is therefore an **inference from the D-1 block's own grounds** (every fr pair
+   cross-lingual by construction; all of Mauss 1925 French), not a rerunnable measurement.
+   Record it in that form. **This is why the disclosure is required rather than optional: the
+   substrate can no longer supply what the ground conceals.**
+
+3. CONDITION — **The en-cell claim is attached to `taggable: false`, not left in the item.**
+   The decision asserts cross-lingual and archaic register *"would earn B in the en cell,
+   where they are not constant"* — untested in force, since no en pair has ever been tagged.
+   The note goes on the `taggable` line itself, so whoever flips it meets the untested claim
+   at the moment it takes force. Left in the item it is a condition honoured by remembering.
+
+4. **(a) REJECTED.** Folding this into PENDING-134's disclosure dates it wrongly and binds it
+   to a doctrine it does not depend on. REVIEWED-121 point 9's ruled resulting state names
+   **amendments**, plural; (b) satisfies that and (a) does not.
+
+5. **(c) REJECTED** on the item's own grounds, and listed only so the option set does not hide
+   the literal reading. It defeats the stated purpose of stratification, and nothing suggests
+   the reading itself is wrong.
+
+6. **This closes the PENDING-134 asymmetry, and that is part of its point.** PENDING-134 was
+   held four days and required an amendment with dated disclosure; this act governed the fr
+   cell undisclosed for seven. Both now enter by disclosed amendment, which is the stated
+   ground the item asked for.
+
+7. **Scope, and the limits are not incidental to this ruling.** PENDING-137 read verbatim; the
+   D-1 block verified against `corpus/v2-stratum-tags.yaml` lines 1–33 and found identical,
+   dropped `# ` markers included. **I did not read past line 70**, which is why condition 1
+   was wrong as first drafted. §6.2, §14.1 and REVIEWED-116 / REVIEWED-121's points are read
+   **only as quoted in this item** — no independent read. Condition 2 does not depend on
+   REVIEWED-116's wording; it stands on the disclosure's completeness. ⚠ The stratum
+   distribution in condition 2 is executor testimony from rows I did not open.
+   ⚠ **`governance-mcp.py --selftest` was failing at the time of the first draft** — five
+   undeclared mutating calls in `wake-digest.py`, from yesterday's REVIEWED-131 build, since
+   declared and passing at 62 controls (`cc97888`). The read surface this ruling was drafted
+   from was unverified while it was being drafted, and was repaired by the executor rather
+   than detected here.
+
+**If AUTHORIZED:** the three edits to `corpus/v2-stratum-tags.yaml` — delete line 132; extend
+the PENDING-137 entry at 127–131 with the direction and its epistemic status as siblings to
+the existing fields; attach the en-cell note at line 370 on the `taggable: false` line. Then
+place the amendment. Then PENDING-134's disclosure, whose before-state is now complete. Then
+`ratio_A_to_B` re-derived **once**, per REVIEWED-116 point 5.