Authority, Typed Lineage, and Atomic Re-entry in Contract-Preserving Recalibration

Abstract

Within Soft and Hard De-Attraction (SHDA), this paper separates the truth of operational objects from authority to act and specifies a quarantine, review and re-entry lifecycle for persistent-state artificial agents. Re-entry authority is conferred only at a linearizable commit that consumes the reservation and nonce; eligibility, reservation and approval states confer none. Given a suitable resolver, liabilities and failure history charged to semantic families survive aliases, merges, splits, epoch changes and rollback; restoration reapplies current ledgers, not historical authority. Under stated premises, Proposition G6 keeps confirmed, unresolved and reserved liabilities and actual consumption within every attributed cap, and Proposition G4 excludes, until revalidation, dependent authorizations after trusted recognition of a required invalidation and dependent effects after a durable receiver fence. Propositions G7 and G8 order leases, bound delegation chains but not aggregates, and bound hysteretic switching by score variation. An output contract fixes what each state and record licenses. Explicit-state studies close finite configurations, yield counterexamples for deliberate guard and ledger mutations, and retain overdue-duty and partial-effect states. In their timing models, local receiver checks cannot exclude stale effects before the fence without refusing valid work; atomic checks of the input's state (not the trusted log) and pre-fenced controlled changes exclude them by construction. The contribution is a typed authority interface over established transaction, capability, provenance and switching mechanisms, not production verification, liveness or measured superiority over a functioning runtime. Note on Version 2.0. This version replaces Version 1.0 (September 2026; about 9,000 words) and is a substantial revision (about 23,200 words). It states Propositions G4 and G6 to G8 under explicit premises (liability caps, dependent effects after invalidation, lease ordering and bounded delegation chains, hysteretic switching), adds an output contract fixing what each state and record licenses, and adds explicit-state studies with counterexamples for deliberate guard and ledger mutations and timing models for stale effects before a receiver fence. Files: the manuscript as PDF and a supplement archive (21 files) with the finite switching, authority, lineage and evidence-scope checks and the protocol witnesses; the full shared validation reports are in the supplement of the flagship record. The Version 1.0 file remains available in the previous version of this record. Publication role. Companion C develops the authority and re-entry branch of the Integrated Framework series on contract-preserving lower-to-upper recalibration (SHDA). Its interfaces connect Companion A's residual diagnosis, Companion B's disclosure and obligations, and the Technical Supplement's authoritative execution records; those items, the flagship and the technical working paper SHDA Algorithms for Scoped Evidence Reuse and Recalibration are archived separately. The paper is a typed authority interface over established transaction, capability, provenance and switching mechanisms; it reports no deployment result and claims neither production verification nor liveness. AI use disclosure. Generative AI (GPT-6.0, OpenAI; Claude Opus 5.5, Anthropic) was used substantively in preparing this work, including source comparison, drafting and editing, and, where applicable, mathematical and counterexample checks and the writing and running of supplementary code. The research questions, framework and final claims were directed and reviewed by the author, who takes full responsibility for the content, including the accuracy of all references and reported numbers. Repository metadata were prepared with assistance from Claude (Anthropic).