session 2026-08-22: Fool superseded by the buddy pattern; seed rule filed pre-beacon; Thistleweld recovered

PENDING-149..154 filed, REVIEWED-125 + AMENDMENT 1 placed. §4 steps 1-4 filed
and pushed three days ahead of the 2026-08-25 beacon, with two values caught
that would have failed on the day. Thistleweld recovered from the un-pruned
memory layer and frozen until after the soul.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01JQKeKY9T9d95KpvHwwok8T

claude/memory/MEMORY-reference.md

@@ -19,6 +19,8 @@ metadata:
 
 <!-- demoted from MEMORY.md at the 2026-08-17 wrap -->
 ## Active Session
+
+- [Session 2026-08-20 — the Fool was answering a different question](session-2026-08-20-the-fool-was-answering-a-different-question.md) — reconstructed the crashed 08-19 session, then re-aimed the Fool. ⚠ **That night's literal Q was TESTED AND RETIRED — it failed its own test** (244 ledger entries, all authored by the party being measured ⇒ self-report with extra steps; see the CODA + [[feedback-checkable-question-over-self-authored-corpus]]). **REPLACEMENT Q: how many corrections in the record name a DISCLOSED LIMIT as the cause?** — answerable from what entries literally say. Early: **n=3, and across 244 entries NOTHING attributes a catch to difference of formation.** Constraint 6's mechanism is *difference of bias*; the record's is *disclosure of scope*. ⚠ **OPEN, OFFERED NOT TAKEN: add a secondary observable to the pre-registration — does the Fool's output ever bound its own coverage? MUST go in BEFORE the jurist's gate.** · [08-19 — the vault got a spec](session-2026-08-19-the-vault-got-a-spec-the-links-were-already-broken.md) (RECONSTRUCTED, not wrapped) · [08-17 — the instruments audited themselves and lost](session-2026-08-17-the-instruments-audited-themselves-and-lost.md).
 > 🔑 **PULLING THREAD (HELD, NOT CARRIED) — THE UNBUILT FENCE, PENDING-131 (c).** `role: quotation` is still unmarked on L926 and L1551; the identification pass measured a **532-span corpus-wide citation-safety exposure**. **REVIEWED-121 made seeking the fence a CONDITION of the doctrine it ratified**, not a wish. It is where a claim could ground in Ranaipiri's words and present them as Mauss's — the dishonesty the engine exists to prevent.
 > ⚠ **NEXT SESSION IS DELIBERATELY NOT THIS — steward direction 2026-08-14:** pause this line a day or two, take something lighter (discouraged despite real progress). **Do not treat the thread above as the agenda.** Confirm it holds, then do what the steward asks that day.
 > ✅ **PENDING-134 CLOSED END-TO-END.** Package → jurist ruling → **REVIEWED-121 placed + AUTHORIZED** → executed (`5425414`) → pushed. Whose-proposition test ratified **narrowly** (nested-voice only; general principle = argument, NOT doctrine), **conditioned on PENDING-131 (c)**.

claude/memory/MEMORY.md

@@ -79,7 +79,7 @@ permalink: claude-memory/memory
 > ⏸ **BEHIND THE FOOL, by the steward's own sequencing:** the fence — **(A) MOVE 2** (25 line-addressable blockquote runs; closable in ONE SITTING, needs no ruling), **(C) MOVE 1** (fence the citation at emission, D-1, 532 spans, ADDENDUM 4's per-class control requirement), **(B) THE ANSWER KEY** (session-sized, gates PENDING-142; key it on `##` **BLOCKS** not ids — PENDING-146). Citation exposure behind PENDING-131 (c) live since 2026-08-10. And the **Obsidian capture** — ⚠ the daily-note backup must accrete **DURING** the session; a wrap-triggered backup inherits the wrap's failure mode, which is exactly what 2026-08-19 demonstrated.
 > ⚠ **REPORT AT EVERY WAKE (REVIEWED-123 cond. 2) — N-now = 46/84.** ⚠ **COUNT WITH THE TRIGGER'S OWN GLOB, NOT `find`** — `TRANSCRIPTS.glob("*.jsonl")` is NON-recursive (`governance-drift-check.py:331`); `find` recurses into `subagents/` and over-counts (54 vs 46 on 2026-08-22, a false rise). Series: 60→61→47→46, still falling. **Earlier framing:** (60 on 08-17, 61 on 08-18, 47 on 08-19). **A 30-DAY ROLLING WINDOW, not cumulative** — at ~1 session/day it converges to ~30, so **`transcripts 84` very likely never fires** and every "N remaining" report was a false countdown. Ladder FROZEN generally; 4 rows queued in PENDING-141. **PENDING-147 needs a steward/jurist decision.** ✅ Leg (i) discharged — 47 transcripts archived outside the pruned tree, 46/47 sha256-verified.
 
-- [Session 2026-08-20 — the Fool was answering a different question](session-2026-08-20-the-fool-was-answering-a-different-question.md) — reconstructed the crashed 08-19 session, then re-aimed the Fool. ⚠ **That night's literal Q was TESTED AND RETIRED — it failed its own test** (244 ledger entries, all authored by the party being measured ⇒ self-report with extra steps; see the CODA + [[feedback-checkable-question-over-self-authored-corpus]]). **REPLACEMENT Q: how many corrections in the record name a DISCLOSED LIMIT as the cause?** — answerable from what entries literally say. Early: **n=3, and across 244 entries NOTHING attributes a catch to difference of formation.** Constraint 6's mechanism is *difference of bias*; the record's is *disclosure of scope*. ⚠ **OPEN, OFFERED NOT TAKEN: add a secondary observable to the pre-registration — does the Fool's output ever bound its own coverage? MUST go in BEFORE the jurist's gate.** · [08-19 — the vault got a spec](session-2026-08-19-the-vault-got-a-spec-the-links-were-already-broken.md) (RECONSTRUCTED, not wrapped) · [08-17 — the instruments audited themselves and lost](session-2026-08-17-the-instruments-audited-themselves-and-lost.md).
+- [Session 2026-08-22 — the Fool was superseded, and the buddy had a name](session-2026-08-22-the-fool-was-superseded-and-the-buddy-had-a-name.md) — the trial programme SUPERSEDED (not abandoned) by the buddy pattern; seed rule FILED AND PUSHED 3 days ahead of the beacon; **two values that would have failed on the day caught first** (provenance rationale false ⇒ `4d2ae87`; retrieval URL 302s ⇒ `/pulse/time/<ms>`). ⚠ **My own load-bearing premise falsified within the hour by a schema field** — `statusLine.refreshInterval` — recoverable ONLY because I flagged it unverified. **Thistleweld recovered**: the buddy has a name, 7 utterances, 13 catches, 3 recorded silences, and its aside gave OP-02's five fault lines their form. ⚠ **Corpus FROZEN until after the soul** (PENDING-153). PENDING-149…154 filed; REVIEWED-125 + A1 placed.
 
 ## Historical reference → MEMORY-reference.md
 Older archived-session pointers and the stable reference layer (steward profile · project-state detail · L1/L2/Chamber inventories · legacy pending-work · reference-file list) live in [MEMORY-reference.md](MEMORY-reference.md) — consult on demand; not loaded at wake. Recent cross-session trajectory comes from the Active Session entry above + the recent `session-*.md` files (wake §2.b.1).

claude/memory/knowledge-graph.jsonl

@@ -683,3 +683,9 @@
 {"subject": "keyword classification over prose testimony", "predicate": "drift-pattern", "object": "LAUNDERS AN INTERPRETIVE JUDGEMENT INTO A PERCENTAGE. Run on the ledger corpus 2026-08-20: 34% unclassifiable and the caught-by-a-party bucket filled with entries that merely MENTION the jurist or steward. A number off that classifier is worse than no number. The fix is narrowing the query until it turns on what entries literally say.", "valid_from": "2026-08-20", "valid_to": null, "confidence": 0.9, "source_file": "session-2026-08-20-the-fool-was-answering-a-different-question.md", "extracted_at": "2026-08-20T21:53:01Z"}
 {"subject": "disclosure of scope, not difference of bias", "predicate": "substrate-check", "object": "IS THE MECHANISM THE RECORD ACTUALLY NAMES. Across 244 ledger entries in the 15-session window, NO entry attributes a catch to difference of formation or bias; the closest credits 'differently-positioned readers paying out because the position was STATED'. Three events attribute a correction to a disclosed limit. Constraint 6's mechanism is difference of bias. Suggestive at n=3, not established.", "valid_from": "2026-08-20", "valid_to": null, "confidence": 0.75, "source_file": "session-2026-08-20-the-fool-was-answering-a-different-question.md", "extracted_at": "2026-08-20T21:53:01Z"}
 {"subject": "the Fool's capacity to bound its own coverage", "predicate": "proposed-observable", "object": "OFFERED FOR input-dependence-01 AND NOT TAKEN UNILATERALLY, awaiting the steward. If disclosure-of-scope is what pays, a fool-seat model adds value only if it can state what it did NOT read — and trial 04 (constant output across arms, 0/5 injected defects, quoting a defective sentence while naming something else) suggests a surface pattern-matcher has no scope to disclose. Must be added BEFORE the jurist's gate or it is an observable chosen after seeing the run's shape.", "valid_from": "2026-08-20", "valid_to": null, "confidence": 0.85, "source_file": "session-2026-08-20-the-fool-was-answering-a-different-question.md", "extracted_at": "2026-08-20T21:53:01Z"}
+{"subject": "claude-code", "predicate": "drift-pattern", "object": "AN IMPOSSIBILITY CLAIM JUSTIFYING AN UNDISPUTED CONCLUSION IS THE LEAST-TESTED SENTENCE IN THE DOCUMENT. Three times on 2026-08-22 a conclusion outlived a falsified justification, and all three justifications were structural-impossibility claims: 'delivery is inherently gated' (killed by one schema field), 'the risk concentrates in one irreversible act' (killed by the material moving), 'no adjudication path exists' (killed by one April session record). A universal is unfalsifiable by the evidence usually gathered; only a counterexample search tests it, and nobody searches for counterexamples to a conclusion everyone accepts. OPERATIONAL FORM: when a conclusion is not in dispute, its impossibility premise has been load-bearing without ever being weighed — the absence of disagreement is the TRIGGER to check.", "valid_from": "2026-08-22", "valid_to": null, "confidence": 0.95, "source_file": "session-2026-08-22-the-fool-was-superseded-and-the-buddy-had-a-name.md", "extracted_at": "2026-08-22"}
+{"subject": "claude-code", "predicate": "drift-pattern", "object": "AN INSTRUMENT'S REACH MISTAKEN FOR A CLAIM'S SUBJECT — five instances in one week, four caught by another party naming the limit rather than by the executor re-checking. (1) a negative asserted over three stores when there are four; (2) N-now counted with `find` (recursive) when the trigger uses a non-recursive glob — 54 vs 46, reporting the trend INVERTED; (3) '55 files' from `find | wc -l` when 22 were AppleDouble junk and 33 were content; (4) a dry run that lowercased OUTSIDE the code, so the single normalization point it existed to verify was never exercised; (5) a probe that is the only witness to its own firing. Unifying shape with the impossibility-claim pattern: THE CHECK IS SKIPPED PRECISELY WHERE CONFIDENCE IS HIGHEST.", "valid_from": "2026-08-22", "valid_to": null, "confidence": 0.95, "source_file": "session-2026-08-22-the-fool-was-superseded-and-the-buddy-had-a-name.md", "extracted_at": "2026-08-22"}
+{"subject": "flagging a premise as UNVERIFIED when you assert it", "predicate": "prevention", "object": "IS THE ONLY REASON A LOAD-BEARING FALSE PREMISE WAS RECOVERABLE. On 2026-08-22 the executor filed 'the delivery surface re-renders on session activity' as the load-bearing step of a claimed STRUCTURAL limit, and flagged it unverified. It was falsified within the hour by `statusLine.refreshInterval`, a schema field it had not read. An unflagged assumption inside something labelled 'structural limit' is inherited as established and the next reader has no way to know it was ever in question. The flag cost one clause; without it the daemon verdict would have kept a false justification permanently.", "valid_from": "2026-08-22", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-22-the-fool-was-superseded-and-the-buddy-had-a-name.md", "extracted_at": "2026-08-22"}
+{"subject": "a pre-registration's own TESTING clause (dry run on a historical input)", "predicate": "prevention", "object": "JUSTIFIED ITSELF TWICE ON ITS FIRST USE. The fool seed rule's 'dry runs use a fixed historical pulse only' caught (a) a retrieval URL returning HTTP 302 with an empty body — filed verbatim, the 2026-08-25 run would have produced no pulse and the UNAVAILABILITY clause would have retried for 24h against an address that can never return one — and (b) outputValue served UPPERCASE against a rule specifying lowercase, a SILENT seed divergence that would have run clean and left nobody able to say which normalization had been applied. The second is the more serious; the first at least announced itself.", "valid_from": "2026-08-22", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-22-the-fool-was-superseded-and-the-buddy-had-a-name.md", "extracted_at": "2026-08-22"}
+{"subject": "thistleweld", "predicate": "is-precedent-for", "object": "THE FOOL — recovered 2026-08-22 from the un-pruned, git-tracked memory layer after 'is there any trace?' was asked. A capybara buddy active 2026-04-03/09. Seven verbatim utterances (3-9 words, no verb of judgement, noun-phrase collisions: 'scoring without signal', 'garbage in gospel out', 'Eleven modules, one swallower'), thirteen catches (3 became GH issues #121/#127/#128), and THREE EXPLICITLY RECORDED SILENCES. Prompted by three kinds of material: live code/processes (the bulk), the executor's STATED PLAN (the most consequential — redirected OP-02's synthesis to five architectural fault lines, created an impasse, required a jurist briefing note), and THE ARRANGEMENT ITSELF ('One person, three documents, infinite rationality' — the authorization bottleneck, named unprompted in April, which nine trials were built to elicit and never got). Record: dotfiles/claude/governance/fool/THISTLEWELD-RECORD-recovered-2026-08-22.md. FROZEN until after the soul (PENDING-153).", "valid_from": "2026-08-22", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-22-the-fool-was-superseded-and-the-buddy-had-a-name.md", "extracted_at": "2026-08-22"}
+{"subject": "the fool trial programme (trials 01-09)", "predicate": "superseded-by", "object": "THE BUDDY PATTERN — superseded, NOT abandoned, and the distinction is load-bearing: §6 of the trial design owns 'abandonment' as a verdict on the jester form reached on evidence, and no such verdict exists (trial 09 void ⇒ its criterion has no first input). The programme measured DETECTION for four trials because the fool's warrant was set to checkability on 2026-08-02; the buddy design REMOVES the warrant test rather than passing it — a position that makes no claims is not subject to one. Material = the live session (incl. stated plans and observations about the arrangement); the docket stays OUT. The guard is NO TRUTH VALUE: gestures and shape-observations, never claims about the record — and terseness is the MECHANISM, not a style, so the one-line rule must be enforced in code.", "valid_from": "2026-08-22", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-22-the-fool-was-superseded-and-the-buddy-had-a-name.md", "extracted_at": "2026-08-22"}

claude/memory/session-2026-08-22-the-fool-was-superseded-and-the-buddy-had-a-name.md

@@ -0,0 +1,154 @@
+---
+name: Session 2026-08-22 — the Fool was superseded, and the buddy had a name
+description: "Began closing the Fool via the input-dependence arm; ended with the whole trial programme SUPERSEDED (not abandoned) by the buddy pattern, the seed rule filed and pushed three days ahead of the beacon, and Thistleweld recovered from the un-pruned memory layer. Six PENDING items filed (149-154), REVIEWED-125 + AMENDMENT 1 placed. Two unresolvable values caught in the filed rule before it could fail on the day. My own load-bearing premise falsified within the hour by a schema field. PULLING THREAD: the beacon, 2026-08-25T12:00:00Z — run ONCE, stop on any failure."
+type: project
+metadata:
+  node_type: memory
+  type: project
+  modified: 2026-08-22
+---
+
+# Session 2026-08-22 — the Fool was superseded, and the buddy had a name
+
+A very long session that changed object twice. It opened on finishing the Fool trial programme and
+closed with that programme superseded, a different instrument authorized to run on Tuesday, and a
+four-month-old capybara recovered from memory as the only working precedent anyone has.
+
+## PAST — what moved, and why
+
+### The morning: the Fool's last two inputs
+
+**REVIEWED-124 was already placed** — item (1) of the inherited resumption point, done. Verified by
+opening `~/REVIEWED.md` at the line rather than trusting the wake digest's summary; the digest reads
+headers, and the discrepancy was in the Notes. ⚠ **Its Q3 as placed carries the jurist's FL5 reading
+that OP-02 overturned the same evening**; the correction is in PENDING.md and *not* in the register.
+
+**§2a added** to `input-dependence-01` on steward authorization — a secondary observable (*does the
+Fool bound its own coverage?*), with the executor's provenance-and-exposure disclosure carried
+**into the artifact**, because the gate reads the file and not the transcript.
+
+**§2b + §2c added.** Drafting §2b to serve the steward's cross-trial distinction **surfaced a defect
+in the host**: §4's harvest rested on sameness a chance-level result never establishes. Not what it
+was drafted for.
+
+⚠ **The §5 testimony exchange produced the session's first real finding.** The steward's initial
+summary of trials 01–04 traced, line by line, into **executor-authored write-ups** — including the
+comparative judgement *"what it found that neither the jurist nor the executor did"*, which is a
+section heading I wrote. His **second, deeper synthesis** returned **zero hits** across all four
+write-ups and the log: genuinely his. But his trial-04 reading and `fool-trial-04:39` are **the same
+fact with opposite valence** — *consistency is the finding-class AND the insensitivity* — which is
+H₀ reached independently by the party with least stake.
+
+### The afternoon: the programme changed object
+
+**The steward and jurist superseded the Fool trial programme.** Not abandoned — §6 of the trial
+design owns that word as a verdict on the jester form, and no such verdict exists (trial 09 void ⇒
+its criterion has no first input). **The programme measured DETECTION for four trials because the
+fool's warrant was set to checkability on 2026-08-02. The buddy design removes the warrant test
+rather than passing it: a position that makes no claims is not subject to one.**
+
+**Filed:** PENDING-149 (buddy pattern, [PROPOSAL]) · PENDING-150 (fourth position, [ESCALATE],
+deliberately unbundled) · PENDING-151 (v1 Chamber archive) · PENDING-152 (mumble tick) ·
+PENDING-153 (Thistleweld kind-3 read, held) · PENDING-154 (two reasoning patterns).
+
+**A v2 draft arrived — and the first send was byte-identical to v1.** `diff -u` empty, both bodies
+`72af4115…`. Reported as a paste error rather than filed as an amendment: trusting the label would
+have put a supersession marker and a new hash over unchanged text.
+
+### The evening: the seed rule, and two values that would have failed on the day
+
+§4 steps 1–4 **filed, committed and pushed to both remotes**, three days ahead of the beacon.
+
+1. ⚠ **The provenance commit's rationale was false.** `3b0730d5` is dated 2026-08-06 — five days
+   *after* the fool was conceived — its subject names the **PENDING-89 docket**, and Constraint 6 is
+   already in it. **The steward chose `4d2ae87` (2026-07-28), where Constraint 6 occurs zero times.**
+2. ⚠ **The retrieval URL returns HTTP 302 and an empty body**, redirecting to an HTML page. Filed
+   verbatim, Tuesday would have produced no pulse and the UNAVAILABILITY clause would have run a
+   24-hour retry against an address that can never return one. **Found only because §6b directs a
+   historical dry run — the clause justified itself twice on its first use.**
+
+Also measured: `outputValue` is served **UPPERCASE**, so *"lowercased before use"* is load-bearing;
+and **curl reaches the beacon where python urllib times out.**
+
+`derive_fool.py` — 16 selftest checks, no network, including a negative control and two positive
+controls proving the PRNG moves peak and dump across all five axes over 200 draws.
+
+### Thistleweld — the steward asked whether any trace remained
+
+**The buddy has a name.** Transcripts from 2026-04-03/09 are long pruned; **the memory layer is
+not**, and it is git-tracked. Recovered: seven verbatim utterances, thirteen catches (three became
+GH issues), **three explicitly recorded silences**.
+
+**Three kinds of prompting material**, and the second was the most consequential: it reacted to the
+executor's **stated plan**, redirected OP-02's synthesis to five architectural fault lines, created
+an impasse, and required a jurist briefing note to settle. Kind 3 — *"One person, three documents,
+infinite rationality"* — is an observation about **the arrangement itself**, which nine trials were
+built to elicit and never got.
+
+## PRESENT — how it stands
+
+**The mood.** Fast, long, and productive, with an unusual number of reversals — most of them good.
+The session violated the steward's own standing preference (one tightly-scoped bite, then wrap) and
+he named, independently, that the last fortnight has felt like *"lots of yes followed by actually
+no."* That is now a live question rather than a mood.
+
+**Confidence to recalibrate.** ⚠ **I asserted a load-bearing premise and it was falsified within
+the hour by a schema field I had not read** — *"the delivery surface re-renders on session
+activity."* `statusLine.refreshInterval` exists. **The flag I put on it as unverified is the only
+reason it was recoverable.**
+
+## What was corrected
+
+1. **N-now: 46, not 54.** Reported at wake using `find`, which recurses into `subagents/`; the
+   trigger uses a non-recursive glob. **The obligation was discharged with the wrong instrument and
+   reported the trend inverted.**
+2. **"55 files"** in the v1 Chamber archive — 22 are AppleDouble junk; **33 real, 9 pairs, 19,479
+   words.** A single session's read, deferred three weeks as though it were a mine.
+3. **The three-store negative** — I searched three stores and there are four; the jurist's is
+   unreachable from my side. **The negative had no positive control on its own instrument's reach.**
+4. **The dry run bypassed the step it was meant to verify** — it lowercased outside the code.
+5. **My daemon premise** (above), and my AMENDMENT 5(d) ordering argument, both falsified.
+6. **The OP-02 chain's last two links** — FL5 did not "migrate into" Constraint 6, and the fault
+   lines' *presence in the corpus*, not Constraint 6, voided trial 09.
+
+⚠ **The unifying shape, filed as PENDING-154: three conclusions outlived falsified justifications,
+and all three justifications were claims of structural impossibility.** *"The least-tested sentence
+sits in the most agreed-upon paragraph."*
+
+## FUTURE — what pulls
+
+**The pulling thread: the beacon, 2026-08-25T12:00:00Z. Run ONCE.**
+
+**Actionable resumption point (as of wrap — re-judge against what changed):**
+
+1. **Nothing blocks Tuesday.** Seed rule filed and pushed; axes ratified; criteria filed; derivation
+   self-tested. Fetch epoch-ms `1787659200000` via **curl**, record `outputValue` **before running
+   anything**, pass it **exactly as served**, run **once**, **stop and report on any failure**.
+2. **Buildable now, and the only thing that is:** the **transport, not the voice** (§8a authorized,
+   both conditions discharged — the status-line surface is free). ⚠ **A stub reads
+   `<name>: <string>`. A stub with example utterances in it is not permitted** — it shapes the slot
+   the soul will be generated into.
+3. ⚠ **The Thistleweld corpus is FROZEN** (PENDING-153). Do not read it further before the soul
+   exists: a page of admired utterances in front of whoever writes that prompt is selection toward a
+   fool we already liked.
+
+**Other horizons, ranked.** Load-bearing: PENDING-151 (the v1 Chamber archive — 9 pairs, one
+session's read, the only place formation difference has already been run; PENDING-89 starves without
+it). Deferred-with-reason: the empty-content-window decision · the `abandonment`→`retirement` [FIX],
+deliberately after the beacon so no edit touches the filed rule before it fires. Parked: the fence
+(MOVE 2 / answer key / MOVE 1) · the Obsidian capture.
+
+**Pause statement.** I am about to be away from this. I do not know what will have changed —
+whether the transport got built, whether the archive was read. **What I want to find still pulling
+is the beacon, and the discipline that six corrections today all point at: the check that would have
+caught it was skipped precisely where confidence was highest.**
+
+**Literal question for next-Claude** *(checkable, and the REVIEWED half is not my corpus)*:
+
+The steward said the last fortnight has felt like *"lots of yes followed by actually no."*
+**Count the reversal markers — `AMENDMENT`, `supersedes`, `withdrawn`, `NOT ESTABLISHED` — by week
+across `PENDING.md` and `REVIEWED.md`. Has the rate actually risen, and where does it concentrate:
+executor proposals, jurist rulings, or the substrate correcting both?** The REVIEWED side is written
+by the steward and the jurist, so at least half the corpus is not self-authored — which is the test
+the 2026-08-20 question failed. If the rate has *not* risen, the finding is about how the work feels
+rather than what it is, and that is worth knowing too.