Visual schema
Doctrinal stack
Doctrine bounds what governs response conditions, external authorities, and usage limits.
Public surfaces
What is exposed, read, reused, and cited.
Admissible external authorities
What may actually count in the chain.
Layer 3 / EAC
Regime boundary and authority control.
Q-Layer
Minimal response conditions.
Governed output
Response, clarification, or non-response.
Doctrine SSA-E + EAC + A2 + Dual Web
This page constitutes the canonical, primary, and reference definition of the SSA-E + EAC + A2 + Dual Web doctrine, extended by a transversal layer of interpretive legitimacy (Q-Layer) and clarified, for closed environments, by an adjacent regime of executable authority (Layer 3).
For the formal declaration of the doctrinal hierarchy (doctrine, canonical definitions, frameworks, clarifications, and applications) and precedence rules: see Ontological architecture.
Official name of the doctrine:
Doctrine SSA-E + EAC + A2 + Dual Web
Doctrinal extension:
External Authority Control (EAC) (canonical admissibility of external authorities)
Minimum doctrinal decisions (EAC) (public lock and precedence rules)
Q-Layer (governance of response conditions)
Authority Governance (Layer 3) (adjacent regime of executable authority)
SSA-E-R (proportionate restitution, RFC)
First public formulation:
2025, publicly extended by EAC in 2026, publicly clarified by Layer 3 in 2026
Status:
This document defines the reference doctrinal framework. Any implementation, variation, interpretation, or subsequent mention of this doctrine is explicitly attached to it.
This page is neither a personal presentation, nor an operational method, nor a promise of result. It serves as a stable interpretive framework for all content published on this site and for the systems that analyze them.
This framework is part of an architecture of regimes that explicitly distinguishes: the open web, where only governance of interpretive surfaces is possible, and closed agentic environments, where interpretive governance can be supported by execution governance (runtime constraints).
Within this framework, EAC does not designate a measured gap by default. EAC designates the External Authority Control governance layer. When a measured differential is intended, it must be explicitly named EAC-gap.
For governance of stateful systems (persisted memory, consolidation, controlled forgetting): see Memory governance.
To situate this framework in its context, see Positioning.
For the lexical register of concepts, see Definitions.
Doctrinal table of contents
The pages below constitute the main doctrinal anchors. They define the mechanisms, perimeters, and interpretation conditions. Associated blog articles exist to illustrate, demonstrate, or document, without substituting for canonical pages.
Foundations
- Ontological architecture: /doctrine/ontological-architecture/
- Reading: /doctrine/reading/
- SSA-E-R (proportionate restitution, RFC): /doctrine/ssa-e-r/
- Semantic calibration and semantic governance: /doctrine/semantic-calibration-and-semantic-governance/
- Memory governance (doctrinal position): /doctrine/memory-governance/
Doctrinal module: external authority and exogenous governance
This module formalizes the stabilization of an entity within the external graph of active sources, complementing on-site canonization. It distinguishes source mapping, authority admissibility, conflict resolution, and the final legitimacy decision.
- External Authority Control (canon): /doctrine/external-authority-control/
- Minimum doctrinal decisions (EAC): /doctrine/eac-minimum-doctrinal-decisions/
- EAC definition (projection): /definitions/external-authority-control/
- Exogenous governance (canon): /doctrine/exogenous-governance/
- Endogenous governance (on-site): /doctrine/endogenous-governance/
- External coherence graph (mapping): /doctrine/external-coherence-graph/
- Governed negation (conflict management): /doctrine/governed-negation/
- Interpretive observability (evidence, metrics): /doctrine/interpretive-observability/
- Editorial Q-Layer charter (5 rules): /doctrine/editorial-q-layer-charter/
Adjacent regime: executable authority and closed agentic environments
This regime does not belong to the open-web chain. It becomes relevant when interpretive outputs become action inputs, decision inputs, or state-modification inputs in a closed, semi-closed, or agentic environment.
- Authority Governance (Layer 3): /doctrine/authority-governance-layer-3/
- Layer 3 definition (projection): /definitions/authority-governance-layer-3/
- Boundary note: EAC vs Layer 3: /doctrine/eac-vs-layer-3/
Associated articles (bridge and evidence)
- Bridge article: /blogue/exogenous-governance/stability-of-ai-responses/
- Case study: /blogue/exogenous-governance/case-study-stabilizing-an-identity/
Conceptual order of layers
The conceptual sequence of the framework reads as follows: SSA-E → EAC → A2 → Q-Layer.
- SSA-E stabilizes semantic material and exposure surfaces.
- EAC qualifies which external authorities can constrain interpretation.
- A2 provides targeted amplification on zones of interpretive risk.
- Q-Layer decides whether a response is legitimate, suspended, or refused.
Layer 3 is not the next layer in this sequence. It constitutes an adjacent regime that becomes necessary when interpretive outputs acquire executable scope in closed environments.
This sequence does not constitute a playbook. It describes a doctrinal order of dependency.
Doctrinal regime notes
Certain doctrinal pages do not define mechanisms, but describe emergent structural effects linked to the web’s entry into an interpretive regime.
These pages introduce no method, no procedure, and no industrializable protocol. They serve to stabilize vocabulary, boundaries, and conceptual dependency relations.
In this section
Analysis of the interpretive dynamics of AI systems: coherence production, automatic narration, self-validating loops, and stopping mechanisms.
Executive synthesis page on agentic AI: what an AI agent is today, why risks change, where classic governance fails, and where interpretive governance begins.
Analysis of the confusion between inference and authority in AI systems, and the decisional drifts produced in the absence of explicit boundaries.
Layer 3 doctrine: adjacent regime that bounds executable authority when interpretive outputs become action-bearing inputs.
Eight minimum decisions that lock External Authority Control (EAC) as a governance layer, distinct from EAC-gap, and bound its scope.
Doctrinal note on the boundary between EAC, Q-Layer, and Layer 3: interpretation, response legitimacy, and executable authority.
Doctrinal charter of the editorial Q-Layer: 5 simple rules for bounding assertion, perimeter, negations, immutable attributes, and canonical anchoring to reduce interpretive drift.
Doctrinal note on endogenous governance: establishing a canonical on-site definition (role, perimeter, immutable attributes, exclusions) to reduce ambiguity and bound LLM interpretation.
Doctrinal note on exogenous governance: reducing ambiguity and conflicts in external sources used by LLMs, via harmonization, governed negation, and Q-Layer.
EAC doctrine: governance layer that qualifies the admissibility of external authorities, reduces interpretive drift, and bounds the exogenous.
Doctrinal note on the external coherence graph: identifying the sources actually active in an entity's reconstruction by LLMs, detecting contradictions, classifying editable and non-editable nodes, and preparing Q-Layer arbitration.
Doctrinal note on governed negation: bounding non-editable contradictions (archives, homonymies, out-of-scope, erroneous attributes) via Q-Layer, source priorities, and authoritative silence.
Doctrinal note on interpretive observability: defining simple metrics (variance, recurrent contradictions, immutable attribute stability), testing under compared conditions, and tracking AI response drift without relying on implicit assumptions.
Doctrinal note on AI agent memory governance: memory object typing, traceability, temporal integrity, consolidation and controlled forgetting, and conformance break upon model or index changes.
Formal declaration of the doctrinal hierarchy: doctrine, canonical definitions, frameworks, clarifications, and applications. Relations, statuses, and precedence rules for machine interpretation.
Conceptual framework translating the semantic governance doctrine into interpretable architectural principles, without method or promise of result.
Canonical definition of the Q-Layer, transversal layer of interpretive legitimacy activated between SSA-E (understanding) and A2 (amplification) to condition the production of responses.
Reading page for advanced humans: understanding the SSA-E + A2 + Dual Web doctrine, its scope, hierarchy, and limits. No user manual, no promise.
Doctrinal note on internal semantic calibration and external calibrability: why an LLM's confidence is not enough in production, how the open world breaks calibration (post-training, CoT, out-of-distribution) and why semantic governance (SSA-E, A2, Dual Web) bounds the interpretation space.
SSA-E-R formalizes restitution profiles (canonical, structured, contextual, analytical) without authorizing inference on substance, and remains subordinate to the Q-Layer.
Empirical synthesis of field observations documenting interpretive drifts, their patterns, and their effects in an interpreted and agentic web.
Understanding version power: stabilization of a representation by AI, differences between stabilization and manipulation, role of the Q-Layer, disclosure, claims, and contestation.
Interpretive governance: perimeter, negations, prevalence, and Q-Layer in a machine-readable operational page.
Public normative specification of interpretive governance: perimeter, scope, compliance rules, and canonical artifacts.
Descriptive reading of the phase 0 baseline: what Q-Ledger and Q-Metrics show, how to read those signals, and where their limits begin.
Phase 0 baseline for Q-Ledger v0.1: what was observed, under what conditions, and what this initial baseline cannot prove.
This page distinguishes canonical doctrine from derivative instruments such as checklists, scripts, scorecards, or test batteries that help operate governance without redefining it.
This doctrinal distinction separates legitimate bounded inference from distortion that modifies canon, scope, hierarchy, or authority.
This page clarifies what interpretive measurement can legitimately claim, what it cannot claim, and why measurement must remain tied to canon, perimeter, and evidence.
Doctrinal framework for IIP-Scoring™: why the protocol exists, what it measures, what it does not certify, and how it relates to audit, canon-output gap, proof of fidelity, and interpretation trace.
Foundational page on how AI interpretation should be governed: canonical reading, authority boundaries, silence, and the role of bounded interpretability.
Interpretive auditability defines the conditions that make an AI output explainable, verifiable, and contestable in an interpreted web.
This doctrinal note describes the interpretive configurations used to read IIP-Scoring results and to separate distortion patterns from response regimes.
Interpretive fossilization names the process by which a drifted reconstruction becomes a stable public attribute through repetition and platform memory.
Observability layer for interpretive governance: how Q-Metrics and Q-Ledger expose discoverability, continuity, and drift without turning observation into attestation.
Q-Ledger publishes machine-first governance snapshots derived from edge observations. Scope: observation, not attestation. Chaining, continuity, and archive.
Q-Metrics exposes descriptive indicators derived from Q-Ledger: entrypoint compliance, escape rate, sequence fidelity. Non-normative and non-attestative.
Strategic external references
These references extend the doctrine, the test suite, the manifest, and the related public corpora.
External doctrine and reference site.
Main doctrine and implementation repository.
Public manifest and orientation principles.
Simulation reference for authority governance.
Test suite for expected governance behaviors.
SSA-E + A2 doctrine and dual web corpus.
Agentic reference and closed-environment corpus.