Defines the neutral constraints under which systems remain explainable, inspectable, and contestable without embedding interpretation.

Structural Explainability (SE) defines a neutral structural substrate for recording identity, structure, transformation, and change without embedding interpretation, authority, causality, or judgment.

Structural Explainability is not anti-interpretation; it is anti-implicit interpretation. It separates structural facts from downstream claims so interpretation and disagreement can remain explicit, attributable, and contestable over time.

Record systems commit to identity semantics. When a rule is refined, an agent redeployed, a jurisdiction subdivided, or two records merged, the system decides whether the thing is still the same thing. That decision is load-bearing for citation, liability, and audit, but not always declared.

SE makes the commitment declarable and the failure detectable.

• Identity & Persistence: A stable, inspectable identity is the prerequisite for all tracking and provenance. Identity is regime-relative: a single structural change might preserve identity under one regime, break it under another, or be identity-neutral under a third.
• Declared Identity Semantics: A system that branches on an undeclared field to decide sameness is running a hidden regime. SE exists to detect hidden regimes.
• Plural Systems: SE is built for independent architectures that share data without requiring a centralized authority, uniform naming, or a single forced ontology.

Within the core substrate, structure, transformation, persistence, and regime behavior may be recorded. Causal explanation, normative evaluation, authority, legitimacy, obligation, and enforcement are not asserted as substrate-level commitments; they may be attached through constrained downstream mechanisms.

Strict architectural boundaries protect this substrate from leaking or collapsing into downstream application layers:

• Governance Boundary (GB): Prevents operational records from implicitly transforming into claims of absolute authority, legitimacy, obligation, or enforcement.
• Interpretation Boundary (IB): Ensures that external frameworks, explanations, and model interpretations remain attached as dynamic overlays rather than becoming hardcoded substrate semantics.
• Attribution Boundary (REF / ATTR / EXT): Classifies exposed paths of a deployed record by the foundational commitment they carry: referential, attributive, or extension. This boundary is decidable and checked.

Downstream domains (e.g., law, infrastructure, agentic AI systems) consume the core substrate without redefining it. SE records the structural commitments of a system without deciding their ultimate meaning, enabling continuous cross-institutional coordination through persistent disagreement.

Some information systems make decisions that affect people, attribute claims to sources, record contested facts, or coordinate across institutions that disagree. Those systems can hold a compliance conclusion, a legal holding, or a causal claim while every party continues to refer to the same underlying artifacts.

They fail in a specific way. Under change, the system must decide what stays the same: whether a refined rule is the same rule, whether a redeployed agent is the same agent, whether a subdivided scope is the same scope. The system decides. It decides by branching on some field, such as a version column, status flag, or display attribute not necessarily declared as identity-relevant.

That is a hidden regime: identity semantics implemented but not declared, which may remain invisible until the system is challenged in court, in audit, in appeal, in public review, or in cross-institutional reconciliation. By then, the records may already have committed to an identity model not explicitly declared.

Existing infrastructure may not catch it. Provenance models, content credentials, software bills of materials, version control, and audit logs record what happened. None answer what the system takes sameness to be, and none can detect a system that answers it differently across undeclared fields.

Hidden regimes are findable, because the alternative is known in advance. Six of the nine identity regimes sit in pairs on three sibling axes: LOC/OBJ, SCOPE-E/SCOPE-S, RULE-C/RULE-S. Each pair offers two admissible identity bases for the same kind of referent.

A hidden regime appears as a specific, finite thing: a pair of records the declared regime treats as one referent, the sibling regime treats as two, and some field in the system already treats differently. That pair is the witness and there is nothing to infer or interpret.

Accountable Records defines what a system must declare for this to be checked, and gives the procedure.

SE does not replace domain vocabularies, standards, ontologies, or existing data systems. It constrains how systems implement them so that disagreement remains visible, attributable, and useful over time.

The formal core of Structural Explainability is developed in three papers. The papers are the normative specification.

SE-100 establishes the neutrality-by-design constraint: a substrate is neutral when its foundational layer is restricted to referential commitments and permitted attribution propositions.

SE-200 derives the referential-regime structure that neutral substrates require, yielding six coarse families and nine core identity regimes: OBL, OCC, REC, LOC, OBJ, SCOPE-E, SCOPE-S, RULE-C, RULE-S.

SE-300 defines the accountable record, the deployed artifact that declares its carrier, basis, regime, transformation history, attribution boundary, and discriminator surfaces, and provides checks that decide NS-conformance over it.

Conformance is three-valued: PASS, FAIL, or INDETERMINATE. An indeterminate result does not count as conformance. Failures return finite witnesses, not scores.

Full conformance is disclosure-relative: the checker decides over the declared artifact, and a later undeclared-surface witness refutes the claim.

A parallel stewardship track addresses governance over time covering how neutral systems are defined, audited, stressed, and repaired in real institutional contexts.

NS-conformance is the checkable relation defined by SE-300. This is the layer at which the theory meets a deployed system.

An accountable-record bundle supplies the following artifacts:

1. a carrier-basis-regime profile;
2. record identifiers and payload references;
3. typed transformation histories;
4. attribution-boundary declarations;
5. a discriminator-surface registry; and
6. a disclosure attestation identifying the implementation boundary.

The checker consumes a bundle and returns a conformance report: per-record and per-store results, the effective declarations used by each check, normalized histories, and every failure and indeterminate witness.

The conformance suite is a set of bundles with known verdicts sufficient to validate an independent implementation of the checks. It allows checkers to be tested and falsified.

Bundles may be produced directly, or emitted by adapters over existing provenance graphs, version-control systems, schema registries, workflow engines, and audit logs. Adapters are not part of the specification.

SE repositories use manifests to describe the repository's role, scope, dependencies, provided artifacts, validation expectations, governance, and traceability.

The SE manifest complements external metadata standards. Citation metadata belongs in CITATION.cff; software supply-chain metadata may use SPDX; software discovery metadata may use CodeMeta; research-object packaging may use RO-Crate.

flowchart LR
  NS[Neutral Substrates] --> RR[Referential Regimes]
  RR --> AR[Accountable Records]
  AR --> NC[ns-conformance]

• To understand the theory, read the three papers.
• To check a declared record system, use ns-conformance.