Evidence discipline keeps one kind of claim from passing as another. A documentary fact, an operational inference, a formal model, a symbolic reading, and direct experience each have their own way of being known. They can support a shared conclusion. A strong claim chain shows how they meet, then states that conclusion directly.
An evidence label identifies how a claim is known and what test can change it. It does not measure importance. Initiatic knowledge can reorganize a life. A documented fact can be trivial.
Evidence Postures
Documented claims rest on primary documents, public records, direct quotations, reproducible measurements, or stable bibliographic evidence. An authenticated document establishes its contents and provenance. The truth of each statement inside it requires its own support.
Supported claims rest on strong convergence among independent sources. Access, mechanism, or decisive records remain incomplete. Corroborated testimony and strong institutional inference belong here when the remaining limits are visible.
Modeled claims rest on a formal analogy, simulation, mathematical structure, or systems model. A model can explain a relation without establishing that every term has a literal counterpart in the world.
Correspondence identifies a structure that recurs across scales, traditions, or domains. Myth, ritual, esoteric teaching, fiction, and imaginal material can disclose this relation. Recurrence alone does not establish literal identity or historical causation.
Speculative claims remain hypotheses. Name the evidence that would raise or lower confidence in them.
Initiatic knowledge comes through disciplined practice and direct transformation. Phenomenology, contact, altered-state reports, dreams, synchronicity, and initiation belong here. Their tests concern experience and operation. They do not require public documents to have force.
Forecast extends a model forward from a named as-of date. It requires an immutable publication anchor, a bounded window, observable indicators, and a failure condition set before the result. Prediction Ledger records each outcome as a hit, partial, miss, or unresolved claim.
A single subject can require several postures. The existence of a government program can be documented. Its operational purpose can be supported, its mechanism modeled, and the operator’s knowledge initiatic. Keep these claims distinct even when they share a paragraph.
Confidence Scale
High confidence rests on direct documentation, strong corroboration, or a mechanism that repeatedly predicts results.
Medium confidence rests on convergent evidence with unresolved gaps, or a strong inference from a documented sequence.
Low confidence applies to a plausible claim that is under-sourced, mostly symbolic, mostly experiential, or dependent on one fragile source.
Open means a claim is worth tracking but cannot yet be graded without more evidence.
Assign confidence to a specific claim at a specific scope. “The document is authentic” can have high confidence while “its source attribution is correct” has low confidence. “The practice changes subjective state” can have high confidence while “the state accesses a nonlocal domain” remains open.
Source Quality
Source grade identifies the quality of a source. Claim posture identifies how a claim is known. A primary source is close to the event, but proximity does not guarantee honesty, completeness, or competence. A secondary source can give a more accurate account by comparing primary records and exposing their conflicts.
Prefer:
- the artifact over a report about it;
- the full transcript over a clipped quotation;
- contemporaneous records over later recollection;
- prospective prediction over retrospective fit;
- independent custody over circular citation;
- a source with something to lose over anonymous amplification;
- reproducible method over prestige.
Anonymous testimony, channeled material, fiction, myth, and dream remain usable. Give each its own posture. A nearby documented fact cannot authenticate it.
Comparative Reconstruction
Recurrence across sources becomes evidence when independent instruments recover the same unusual relation. Shared words and universal motifs carry little weight. A repeated sequence, safeguard, causal direction, or consequence carries more. Its force increases when the sources differ in genealogy, method, interest, and ways of failing.
Metaphysical Tomography develops this method. Record source independence, name the transformation between vocabularies, and preserve disagreements. Identify the strongest rival explanation. Test the reconstruction against material that was not used to build it. Shared ancestry and borrowing reduce the number of independent witnesses. Shared neurobiology can explain a recurring experience without deciding whether the nervous system generates or detects the structure it discloses.
AI-assisted synthesis follows the same rule. Retrieval can expand the source base. Fluent comparison cannot create independence. Count source families, record failed matches, and expose each translation that changes after the result is known.
Claim Chain Shape
A durable chain contains a thesis, mechanism, evidence, counter-reading, confidence, links, and open questions. The mechanism explains how the facts support the thesis. The counter-reading tests whether the thesis can withstand its strongest objection.
Show where the symbolic and concrete layers meet. A symbolic reading gains force through a concrete connection. This can be a legal instrument, institutional incentive, biological pathway, timeline sequence, financial architecture, technological affordance, or repeated operational output.
Use the smallest complete form:
claim
because mechanism
supported by evidence
against strongest rival
at stated confidence
revised if conditionName the condition that would require a downward revision. This makes the claim answerable to evidence without filling every sentence with uncertainty. State the thesis directly. Place the boundary at the joint on which it depends.
Counter-Reading
A counter-reading is the strongest rival explanation that can account for the evidence. A weak objection flatters the thesis. A strong rival shows where it must improve.
In high-strangeness and contact cases, attribution of a source must stay within the documented record. The direction and consequences of an event remain relevant when its source is unresolved.
Common Failures
- Adjacency: actors share a venue, funder, symbol, or date, and this connection is treated as proof that one commands the other.
- Authentication drift: a genuine document or video is treated as proof of its strongest interpretation.
- Register laundering: myth or testimony is presented as documented history.
- Analogy inflation: resemblance becomes a physical mechanism without a mapping or test.
- Survival as truth: a teaching is treated as correct because it is old or has endured.
- Unknown as confirmation: missing evidence is assigned to secrecy and counted as support.
- Scope creep: a result in one organism, population, scale, or condition becomes a universal law.
- Hedge fog: qualifiers accumulate until no claim remains to test.
Correct the structure of the claim. Clear distinctions let the synthesis speak with force.
References
Toulmin, Stephen. The Uses of Argument. Cambridge University Press, 1958.
Popper, Karl. The Logic of Scientific Discovery. Hutchinson, 1959.
Hill, Austin Bradford. “The Environment and Disease: Association or Causation?” Proceedings of the Royal Society of Medicine 58, no. 5 (1965): 295-300.
Guyatt, Gordon, et al. “GRADE Guidelines: 1. Introduction - GRADE Evidence Profiles and Summary of Findings Tables.” Journal of Clinical Epidemiology 64, no. 4 (2011): 383-394.