SΔϕ-18 — Human as the No-Arbiter Species: Provisional Institutions and Editability under Irreversibility (v1.0)

This paper provides a minimal functional definition of human within the Sofience–Δϕ Formalism by addressing the institutional reversal problem introduced in SΔϕ-17. SΔϕ-17 defined sacred/institution/law as cost devices stabilizing openness of relational futures (Ω_rel) without infinite validation cost, but raised a critical exit problem: stabilizers can reverse into coercive closure. SΔϕ-18 fixes the minimal mechanism required to detect and repair such reversal, and uses it to define “human” structurally rather than morally. Core claims: Axiom 18-0 (No external negotiator): in conflicts with the world or the other, there exists no ultimate external arbiter (¬∃ Arbiter_external). Humans are defined as systems that recognize this condition. Corollary (Provisionality): institutions must remain provisional cost devices, not ultimate arbiters; absolutization is a reversal risk. Axiom 18-1 (Institutional reversal): stabilizers can reverse into closure (Stabilizer(C) → Closure(C)), producing coercive persistence. Axiom 18-2 (Editability operator): the human function is modeled as applying edit operations to institutional cost devices when reversal is detected, restoring provisionality and re-opening relational futures. Proposition (Responsibility): responsibility is redesign capacity—modifying stabilizers so that closure becomes infeasible and future costs decrease across repetitions. The document is intentionally designed as a loop-closure item: it connects SΔϕ-16/17 institutional dynamics back to pre-commitment and openness stabilization (SΔϕ-15/17), closing one cycle of the series while enabling re-entry.