Reality Audit

Reality Audit is not a separate doctrine beside theory. It is the methodological form by which the requirements of actuality, reference, correspondence, traceability, and corrigibility are translated into examinative work.

The method may be applied to language, identity, models, representations, systems, institutions, authority, and practice. It may also be applied to Røyndalism itself. No method is corrigible if it audits only others while protecting its own premises from examination.

A Reality Audit does not decide in advance that something is false or captured. It distinguishes what is supported, what is only partly supported, what remains unresolved, what lacks sufficient ground, and what does not hold. Capture is one possible finding, not the premise of the inquiry.

The methodological architecture distinguishes functions that are often conflated. Audit, critique, finding, and correction are not synonyms.

A finding must arise from documented examination. Correction must answer proportionally to the finding. Critique without traceable grounds may be assertion; correction without a bounded finding may become imposition.

1. M1

   Audit

   Bounds the object, distinguishes object from claim and frame, and collects sources, observations, premises, and relevant counter-material.

2. M2

   Critique

   Tests whether references, inferences, classifications, models, and authority claims hold within the grounds provided.

3. M3

   Finding

   Formulates a bounded and reviewable assessment of correspondence, uncertainty, insufficiency, error, contradiction, or capture risk.

4. M4

   Correction

   Changes wording, scope, classification, model, procedure, role, or action to the degree warranted by the finding.

5. M5

   Review

   Checks whether the correction actually improved correspondence or created new hidden premises, losses, or errors.

These principles govern every Reality Audit, regardless of object or domain.

1. P1

   Actuality before frame

   The frame must serve inquiry into what is; it must not decide in advance what may count as actual.

2. P2

   Duty to distinguish

   Object, appearance, observation, claim, interpretation, model, and conclusion must remain as clearly distinguished as the matter permits.

3. P3

   Reference and correspondence

   An expression or model must remain traceable to what it refers to, and claim strength must answer to evidential strength.

4. P4

   Traceability

   Sources, selections, premises, transformations, calculations, and decisive steps must be documented so others can review them.

5. P5

   Corrigibility

   Claims, findings, and methodological choices must be changeable when better grounds or demonstrated errors require it.

6. P6

   Proportionality

   Findings and corrections must not extend beyond what the documented grounds can bear.

7. P7

   Role and responsibility clarity

   Who observes, interprets, decides, acts, and is affected must be made visible where this bears upon the matter.

8. P8

   Open uncertainty

   Unresolved questions and missing grounds must be marked as uncertainty rather than filled by authority, habit, or rhetorical confidence.

The procedure is a basic sequence, not a mechanical form. Simple matters may be documented briefly; complex or serious matters require more extensive source work, contradiction, and revision.

Each stage must leave an intelligible output. Where a stage cannot be completed, this must be recorded as a limitation on the finding.

1. 01

   Bound the object

   Define what is being audited, the relevant time and context, and what falls outside the inquiry.

   Audit mandate and scope.
2. 02

   Distinguish object, claim, and frame

   Identify what the claim concerns, the claim itself, and the concepts, models, or categories shaping its presentation.

   Structured claim and frame register.
3. 03

   Register the grounds

   Collect sources, observations, data, testimony, documents, premises, and relevant absences of material.

   Source and documentation register.
4. 04

   Test reference and inference

   Examine whether expressions refer to what they are used about and whether the conclusion follows without hidden leaps.

   Correspondence and inference test.
5. 05

   Test counter-material and alternatives

   Seek material that may bound, weaken, or alter the provisional account.

   Counter-test and uncertainty register.
6. 06

   Assess capture risk

   Examine whether the frame controls what may count as grounds, conceals contradiction, or protects itself from correction.

   Bounded capture-risk assessment.
7. 07

   Classify the finding

   State what is supported, unresolved, insufficient, or untenable, with visible strength and limitation.

   Reasoned audit finding.
8. 08

   Determine correction and review

   Select the least intrusive change proportionate to the finding and define how its effect will be checked.

   Correction and follow-up plan.

Finding categories are methodological classifications. They must not be used as moral labels for persons or as substitutes for the underlying documentation.

Several categories may apply to different parts of the same matter. A chain of claims may contain well-supported observations but an insufficient inference or a model assigned more ontological content than the grounds can bear.

The audit types are specialisations of the same method. They differ by object, characteristic failure modes, and the documentation required.

Forms of capture belong as risk patterns and possible findings within the relevant audit type. Identity capture is therefore not an alternative to Identity Audit, but a possible condition that Identity Audit may identify.

Application domains are not separate theories of truth. They are contexts in which the same methodological architecture is adapted to different sources, responsibilities, and effects.

The theory of world, society, and institution belongs under Theory. This section organises practical audit questions and cross-cutting chains of responsibility.

Research and science

What is observed, what is modelled, which premises are used, and how is the result assigned to actuality?

Data, model choice, measurement, uncertainty, replication, and communication.

Law and administration

Are fact, authority, role, assessment, and decision distinguished and documented?

Decisions, access, evidence, discretion, contradiction, and reasons.

Media and public discourse

What is shown, what is it said to show, what is omitted, and which frame governs the reading?

Headlines, images, sources, editing, labelling, and correction.

Education and knowledge communication

Are models and concepts taught as tools with limitations or as replacements for what they refer to?

Textbooks, assessment, disciplinary authority, practice, and open uncertainty.

Work and organisation

Do roles, reporting, and procedures answer to actual work, responsibility, and effect?

Health and safety, deviation, whistleblowing, metrics, management, and documentation.

Technology and artificial intelligence

Which data, classifications, and objectives ground the system, and what is wrongly assigned to the machine or its output?

Training, automation, traceability, authority, error, and human responsibility.

Health, welfare, and care

How are person, observation, professional assessment, records, resources, decisions, and actual effects held together?

Records, consent, prioritisation, professional judgment, standardised tools, and follow-up.

Politics, governance, and civic institutions

How are mandate, representation, knowledge grounds, interests, decision power, and public accountability made traceable?

Reports, consultations, budgets, indicators, emergency decisions, lobbying, and democratic oversight.

Correction is not punishment and not a rhetorical demand that the auditor be declared right. It is a bounded change intended to bring a claim, frame, model, system, or practice into better correspondence with documented grounds.

The choice of correction must consider harm, reversibility, due process, responsibility, and whether those affected can examine the grounds.

1. Clarify or narrow wording and claim strength.
2. Distinguish observation, interpretation, model, and conclusion more clearly.
3. Obtain missing sources, counter-material, or contradiction.
4. Correct calculation, classification, reference, or documentation.
5. Change model, procedure, allocation of roles, or decision path.
6. Withdraw or suspend a claim, decision, or practice where the grounds do not hold.
7. Schedule renewed review after correction has been implemented.

R0–R7 now exist in their first integrated public form. Overview, methodology, standard, procedure, findings, correction, audit types, applications, cases, registers, and resources have stable placement and reciprocal relations.

Structure version 1.0 means that the planned principal architecture is published. It does not mean that the method is final. Later substantive changes are to be assigned to the document they actually affect.