TT Claude Memory

feedback_e_w_creep_recurrence_q027_first_v3_13_production_use

Properties1
descriptionQ-027 Wills.Bequest first-pass scorecard 2026-05-17 anchored on Wills Act 1837 / Law Commission UK / SD-JWT VC; treated Switzerland CC/CO 484+ + Scotland T&S(S)A 2024 as retrofits despite Q-015 λ.η Phase-1 3-jurisdiction lock. Rich called out drift: "INHERIT is an international standard". Pattern recurs despite `feedback_first_vertical_slice_symmetric_2026_04_24` + A-124 anti-E&W-creep discipline. v3.13 step 7.5 named-persona pass DID surface gap (persona P5 Jurisdiction-fidelity (c)-class) but I folded as Catala-rule fix not as c11 criterion-set invalidation — systemic framing error not surface fix. v3.14 amendment proposed adding step 7 international-scope sub-audit.

3 min read

E&W-creep recurrence on Q-027 first v3.13 production use (2026-05-17)

The rule

For any TT v2 standard work touching multi-jurisdiction scope, the criterion-set + option-set + substrate research MUST be authored international-scope-first, not E&W-first-with-retrofits. Specifically:

  • Criterion names MUST cite cross-jurisdictional fidelity, not a national statute (c11 must NOT be “Wills Act 1837 typology fidelity”; must be “Cross-jurisdictional typology fidelity (UK&W + Scotland + Switzerland …)”)
  • Substrate research tracks MUST be jurisdiction-symmetric (≥1 web track per Phase-1 jurisdiction, not 4 UK tracks + 1 EU track)
  • Pgvector queries MUST avoid jurisdiction-anchored keywords (“Wills Act ademption” is biased; “bequest typology cross-jurisdiction” is neutral)
  • Per-jurisdiction substrate must appear with parity in scorecard cells (Scotland not “retrofit”; Switzerland not “Phase-1.5+”)
  • Universal-vs-national-anchor classifier choice: prefer abstract universal classifier (e.g. TransferMechanismScheme) with per-jurisdiction SSSOM mapping_set over national-statute-anchored enum

Why:

  • INHERIT v2 is an international standard, not a UK standard. Treating UK as canonical baseline is repeatedly-flagged drift (24 April 2026 → 17 May 2026 = same drift in <4 weeks).
  • Q-015 λ.η LOCKED Phase-1 = UK&W + Scotland + Switzerland (3 jurisdictions Day-1); Texas/Florida/RON Phase-1.5+; faith pillars Phase-1.5+. The Phase-1 SYMMETRIC 3 includes a civil-law jurisdiction (Switzerland) — schema must accommodate civil-law model (universal succession via Erbeinsetzung + Vermächtnis as claim-against-heirs) not just common-law model.
  • A-124 module sequencing locks “symmetric-jurisdictional discipline preserved” per feedback_first_vertical_slice_symmetric_2026_04_24.
  • TT acquirer-DD package depends on the standard being demonstrably international, not UK-shaped.

How to apply:

  • Trigger: ANY v3.13 directive Q touching schema / classifier / rule / cross-module reference for a module covering multiple jurisdictions (i.e. all Wills / Probate / Trusts / Assets / Commerce / Transfer / Delegation / Credentials / Core work)
  • At v3.13 step 6 PRE-FLIGHT, explicitly state: “I have evidenced ≥1 jurisdiction-symmetric criterion + ≥1 jurisdiction-symmetric web track per Phase-1 jurisdiction”
  • At v3.13 step 7 option-set authoring, force ≥1 option that uses abstract universal classifier with SSSOM mapping_set
  • At v3.13 step 7 criterion authoring, force c11-equivalent criterion to name ALL Phase-1 jurisdictions (or use phrase “cross-jurisdictional fidelity”)
  • At v3.13 step 7.5 named-persona pass, force ≥1 persona to be a CIVIL-LAW practitioner (Swiss / German / French / Italian) — not just Scots-law mixed-tradition
  • v3.14 amendment proposed: step 7 explicit “international-scope sub-audit” — does each FR-flavour criterion name the international scope, not a national statute?

Diagnostic when drift appears:

  • IF any criterion mentions a national statute (Wills Act 1837 / Texas Estates Code / Florida Probate Code / Succession (Scotland) Act 1964) WITHOUT also naming the Phase-1 jurisdictional peers → drift
  • IF substrate research tracks are >50% one-jurisdiction → drift
  • IF pgvector query keywords are jurisdiction-anchored → drift
  • IF option-set lacks an abstract universal classifier option → drift
  • IF named-persona pass lacks a civil-law persona → drift

Q-027 recovery 2026-05-17

  1. ξ.ζ★ (E&W-anchored common-law 5-class BequestTypeScheme) withdrawn after Rich critique
  2. ξ.η’★ re-authored: universal TransferMechanismScheme 4 base values + 4 qualifier values + SSSOM 31-row mapping_set Day-1 bridging 3 Phase-1 jurisdictions + BequestFailureModeScheme 7 values including F7 jurisdiction-legal-rights-override (Scotland T&S(S)A 2024 + Swiss Pflichtteil + E&W 1975 Act) — passed inline S1 + S1b lock-BLOCKING spikes
  3. Rich confirmed separate Q-026.1 amendment first (jurisdiction:string → JurisdictionScheme uplift), then Q-027 ξ.η’★ lock
  4. v3.14 amendment authorship deferred but identified — step 7 international-scope sub-audit needed
  • [[feedback_first_vertical_slice_symmetric_2026_04_24]] — original anti-E&W-creep lock 24 April 2026
  • [[feedback_actively_use_t_files_in_scorecard_authoring]] — T-file citation discipline (sibling rule)
  • [[feedback_research_artefact_forward_traceability]] — forward-traceability discipline (sibling)
  • A-124 amendment — module sequencing + symmetric-jurisdictional discipline preservation
  • Q-015 λ.η lock — Phase-1 = 3 jurisdictions Day-1 (UK&W + Scotland + Switzerland)
  • feedback_e_w_creep_recurrence_q027_first_v3_13_production_use.md (this file) — pattern-naming for v3.14 amendment substrate

Graph View

Backlinks