Context
The representation stores the current longest even-palindromic suffix occurrence and the reversed context to its left. Typed receipts name the exact historical occurrence needed by a later transition, rather than only its word or endpoint.
Hypotheses and scope
- The alphabet is fixed and finite; the published construction specializes to the binary alphabet.
- Receipt closure is an abstract persistent-sequence statement. Its native implementation premise is separate and is stated by the complete Draft 5 Section 11.1 contract in T127/T128.
Proof or evidence
The proof derives the exact three-case word recurrence, mirror and selector receipt recurrences, bridge-shadow transport, and a fixed-operation closure theorem. Finite audits separately check 262,142 exhaustive word updates, 38,758 guarded shadow instances, and the stored selector against unbounded bridge-chase families.
Verification notes
Draft 1 was found defective: it used bridge shadow outside one hypothesis and did not store a bounded pointer to the selected occurrence D_a(P). Draft 2 adds a mirror-prefix lemma and one transition receipt per direct table entry. A subsequent cold read confirmed these repaired interfaces; finite checks corroborate but do not prove them.
Limitations
- Raw endpoints, immutable type-birth records, anchor-independent chunks, and bounded bridge chasing fail only as named representations; they are not general scaffold lower bounds.
- The proof depends on distinguishing palindrome types from historical occurrences.
Notes
The semantic receipt proof is retained, including Draft 2's mirror-prefix and typed-transition repairs. The native implementation obligation is separate and is stated in T127/T128. Representation obstructions are scoped to their named interfaces. Finite word audits are corroboration, not uniform proof; external specialist and novelty review remain open. See docs/integration/reconciliation/T125-T128.md for the current source edition, evidence boundary and seven-surface disposition.