Phase 7 — Confidence Recalibration (Post-Phase-8 Addendum, 2026-05-16)

Run framing. Fourteenth full audit, subsequent-run mode. This addendum runs after Phase 8 per AUDIT_PLAN.md / CLAUDE.md §"Note on phase order": Phase 7's confidence-recalibration logic depends on Phase 8 outputs when synthetic-layer work occurred. Phase 7's initial pass (wiki/.audit/phase-7-initial-2026-05-16.md) deferred all confidence movements pending Phase 8 promotion outcomes.

Phase 8 outcomes summary. 9 candidate→live/supported promotions (8 live + 1 supported = stendhal-naturalness-practically-resolves-sartre-antithetic). 0 retires. 0 contestations. 0 concept-page upgrades (Step 8.5 deferred). 0 cite-back touches (Step 8.9 deferred).

Addendum walk

Phase 7 initial flagged 3 candidate confidence movements pending Phase 8 outcomes:

1. [[politics-mp]] (medium)

Phase 7 initial: addendum-candidate, consider upgrade to high if "general milieu" subsection from SnS lands at Step 8.5.

Phase 8 outcome: Step 8.5 deferred. The "general milieu" subsection does not land in this run.

Phase 7 addendum verdict: hold medium. No frontmatter change applied. The candidate confidence movement remains pending Step 8.5 application in a follow-up sweep.

2. [[pensee-de-survol]] (medium)

Phase 7 initial: addendum-candidate, consider upgrade to high if SnS Metaphysical in Man earliest-published-locus citation lands in Sources at Step 8.5.

Phase 8 outcome: Step 8.5 deferred. The SnS earliest-published-locus citation does not land in this run.

Phase 7 addendum verdict: hold medium. No frontmatter change applied. Same deferral as #1.

3. New stubs from Step 8.5 (a-focalism, stendhalian-naturalness, logos-of-the-sensible, lived-perspective)

Phase 7 initial: new-page confidence set at creation.

Phase 8 outcome: Step 8.5 deferred. None of these stubs are created in this run.

Phase 7 addendum verdict: no-op.

Additional addendum check — Phase 8 promotion downstream effects

The 9 Phase 8 promotions affect a number of concept-page Wiki-homes references. Confidence values on those concept pages may move iff the new claim status changes how the concept page is using the claim. Spot-check on the 6 concept pages most affected by the 9 promotions:

Concept page Pre-Phase-8 confidence Newly-live/supported claim attached Phase 7 addendum verdict
[[antithetic-critique-of-sartre]] high stendhal-naturalness (newly supported), sartre-too-antithetic-1945 (newly live) hold high — already high; supported claim strengthens but no rating change warranted
[[stendhal]] (entity) n/a (entity, no confidence per page format for entities; spot-checked) stendhal-naturalness (newly supported) n/a
[[coherent-deformation]] high conquering-language-as-coherent-deformation-of-language (newly live) hold high — already high; the new live claim is one of many supporting attestations
[[speaking-spoken-speech]] high pbp-as-middle-term-parole-parlante-parlee-1953-54 (newly live) hold high
[[sensible-ideas]] high pbp-as-primary-anchor-for-sensible-ideas-pre-vi (newly live) hold high — already high; this is a corrective claim that reframes the doctrine's MP-source anchor, not its philosophical content
[[perceptual-faith]] high (spot-checked) perceptual-faith-1946-1947-published-prefiguration (newly live) hold high

No concept-page confidence frontmatter changes warranted. The promotion pattern in this run is strengthening at the synthetic layer (claims.md) without requiring revisions at the concept-page layer, which is the expected effect of a well-tuned synthetic-layer that doesn't reach back to disturb its anchors.

Aggregate addendum verdict

0 confidence frontmatter changes applied in this addendum. 2 addendum candidates remain queued pending Step 8.5 follow-up work (politics-mp, pensee-de-survol). The Phase 8 promotion outcomes do not require any concept-page confidence movement because:

  • All promotions are within medium confidence at the claim layer (the new supported is supported / medium, the 8 new live are live / medium).
  • No new high-confidence claims that would tip downstream concept pages to higher confidence.
  • No contestations or retires that would tip downstream concept pages to lower confidence.

Density-bias-diagnosis crosscheck

The 2026-05-16 density-bias-diagnosis recorded 0 retirements in 36 days across 150 entries (then). This run added 0 retirements. The 0-retirement pattern persists at 156 entries / 40 days. Phase 7 addendum did not surface any candidate-for-retirement.

This is consistent with the diagnosis: the wiki's gates fire on promotion, not on retirement. The 156→156 entry-count stability across this run (9 status changes, 0 add/remove) is a healthy densification signal — exactly the consolidation behavior the density-bias-diagnosis recommended.

Stop conditions

None tripped. Audit pipeline approaches completion.

Outputs

  • This report (wiki/.audit/phase-7-addendum-2026-05-16.md)
  • 0 frontmatter changes applied
  • 2 addendum candidates queued for follow-up (politics-mp, pensee-de-survol)
  • Log entry: deferred to run close