No tradition sees the whole object. Each receives reality through a human instrument, a practice, a language, a history, and an institution. The result is a projection: real information shaped by the angle and limits of the receiver.
Metaphysical tomography compares those projections to recover the architecture that produced them. Religions enter as situated instruments. Their doctrinal differences remain intact while a stricter question is asked:
What unusual relation survives translation across instruments that do not share the same vocabulary, method, history, or interest?
The method takes its name from tomography. A scanner reconstructs an interior from many partial projections recorded at known angles. The reconstruction gains definition through difference. Another angle contributes information because it adds a new cut through the object.
Traditions, sciences, arts, initiatic practices, anomalous reports, and institutional histories can work the same way. Each discloses one surface of a larger architecture. Their overlap reveals invariants. Their disagreements expose depth, distortion, and the limits of the model.
The Inverse Problem
The comparison runs backward from appearances to the structure capable of producing them.
underlying architecture
↓ transformed by an instrument
partial cultural projection
↓ compared across independent angles
recoverable invariant
The central problem is an inverse problem under incomplete information. Collecting more parallels only enlarges the noise until the source geometry is known.
Johann Radon showed that an object can be reconstructed from a sufficiently complete set of projections. The mathematical result carries an equally important warning. Finite projections do not determine a unique object. Restricted angles leave some features clear and others nearly invisible. Noise is amplified during reconstruction. A prior model is required to stabilize the result.
The metaphysical case is harder. Its projection angles arrive culturally constructed. The instruments interpret what they disclose. Their histories cross-contaminate one another. The analyst also belongs to the field being analyzed.
A useful accounting relation is:
observed teaching
= transformed invariant
+ inherited material
+ shared human structure
+ local history
+ noise
This accounting relation is a custody ledger with no numerical scale. Every comparison must determine which term is carrying the apparent match.
The method must be able to return a null result. Some projections will fail to fit any single object without arbitrary recoding. A procedure that always reconstructs unity has installed unity as its prior and then rediscovered it in the output.
Traditions Are Instruments
An instrument reveals and transforms at once. A telescope discloses distant light through the shape of its lens. A contemplative lineage discloses experience through doctrine, discipline, symbol, body, authority, and expectation.
Each instrument has an aperture.
Doctrine establishes the distinctions available to perception. Nondual fullness, emptiness, divine presence, covenant, subtle anatomy, and imaginal encounter organize experience differently.
Practice selects the accessible range. Prayer, meditation, breath, ordeal, dream incubation, plant medicine, ritual, ethical restraint, and spontaneous crisis open different windows.
Institution determines what can be said, preserved, initiated, forbidden, or forgotten. Canon, secrecy, patronage, persecution, translation, and professional legitimacy shape the surviving record.
Embodiment supplies a shared human carrier. Nervous systems, attachment, mortality, sexuality, sleep, illness, awe, grief, and agency detection generate recurring forms before any metaphysical referent is added.
History carries descent and borrowing. A motif repeated across five texts from one source remains one witness wearing five masks.
The operative instrument is a delimited school, practice, text family, or historical community. Names such as Buddhism, Christianity, and Hinduism contain internal traditions with incompatible doctrines and methods. Treating a civilization-sized category as one witness destroys the calibration before comparison begins.
Calibration begins by describing these transformations. Buddhist emptiness and Kabbalistic divine plenitude both exceed the ordinary ego while retaining incompatible meanings. The relation between the terms must be shown. What does each practice change? What does each ontology permit? What form of error does each recognize? What fruit counts as verification? Where do their consequences diverge?
Translation is earned when those questions have answers.
The Object Moves
Medical tomography assumes a stable object during the scan. Metaphysical reconstruction faces a layered and responsive object. Hermetic law may remain stable while ecologies, institutions, languages, bodies, and consensus conditions change. Practice changes the observer. Coordinated observation can change the shared field. A tradition can therefore record a different phase of the object as well as a different angle upon it.
Every projection requires a temporal address. Contradictions across ages may reveal distortion, historical development, a changed ecology, or a lawful turn of the coupled gyre. The clock as state measure prevents chronology from flattening those phase differences into one static inventory.
The deepest invariant may be a generative law governing transformation across time. The seven Hermetic principles carry that kind of invariance: they specify how manifestation differentiates, moves, recurs, retains consequence, and propagates across scale. A dynamic reconstruction asks which relations remain lawful while the visible world changes.
What Counts as Convergence
Convergence carries weight when different classes of evidence recover the same non-obvious relation.
William Whewell called this consilience of inductions. A hypothesis formed in one class of facts gains force when it explains facts of another class that did not help construct it. William Wimsatt gave the neighboring principle of robustness: a feature becomes more credible when it can be detected, derived, or produced through methods with different failure modes.
The operative phrase is different failure modes. Ten sources repeating one archive provide one angle. Meditation reports, textual genealogy, biological measurement, ritual structure, and institutional history can provide different angles because each carries a different error structure.
The strength of a convergence depends on four properties.
Independence. The sources come from distinct lineages, datasets, interpreters, and cultural transmission chains.
Specificity. The recovered relation is rare enough that its recurrence would be surprising. Water, light, ascent, death, rebirth, and cosmic conflict are nearly universal and therefore weak by themselves.
Constraint. The comparison preserves exclusions, sequence, causal direction, safeguards, and consequences. A relation that can match everything explains nothing.
Risk. The reconstruction identifies what should appear in material not used to build it and what would count against the model.
Shared vocabulary is the weakest convergence. Shared imagery is stronger only when provenance rules out borrowing. Shared operation is stronger again: distinct traditions prescribe different practices that reorganize attention, conduct, and identity in the same ordered relation. Shared consequence is stronger still: the practices create the same capacity, hazard, or institutional response despite different doctrines.
The Independence Ledger
Independence is graded, not declared.
Textual descent can often be detected through shared peculiarities and shared errors. Direct contact, translation, conquest, trade, intermarriage, migration, and common schooling join traditions that later appear separate. Deep mythic ancestry can reduce apparently distant stories to branches of one older narrative family.
Shared embodiment creates another common source. Human beings everywhere dream, grieve, detect agency, fear death, enter altered states, experience sleep paralysis, respond to rhythm, and build coalitions. Cross-cultural mystical experience can therefore reflect a shared detector, a shared generator, or both. Neurobiology explains part of the instrument before it decides what the instrument is receiving.
The ledger counts source families. Quotation count is irrelevant to independence.
five texts from one school = one clustered witness
two schools sharing an ancestor = partially independent witnesses
different methods and histories = stronger independence
different methods, histories,
and error structures = strongest available independence
Common ancestry changes evidentiary weight while leaving truth status open. A transmitted insight can remain true; all of its copies carry the weight of one lineage.
Disagreement Adds Depth
Flat perennialism keeps the agreements and discards the rest. Tomography requires the disagreements because they constrain the hidden object.
Plato describes education as a painful reorientation from shadows toward what makes them intelligible, followed by a return to the cave. Gnostic systems describe an order whose apparent rulers mistake derivative authority for the highest God; knowledge breaks their total claim. Kabbalah describes emanation, concealment, shattering, and repair within a creation that remains charged with divine relation. Kashmir Shaivism describes consciousness freely limiting itself into finite experience and recovering its own nature through recognition. Sufism describes heedlessness, remembrance, purification, annihilation of possessive selfhood, subsistence in God, and return to service.
The shared relation is precise enough to matter:
localized identity mistakes a partial order for the whole
↓
disciplined knowing changes the knower's relation to that order
↓
the result is tested through return, repair, or transformed conduct
The cause of localization remains contested. The world appears as exile, manifestation, covenantal field, pedagogical image, or divine self-concealment. Personal identity persists in some accounts and dissolves under analysis in others. Those contradictions prevent the reconstruction from claiming one universal cosmology.
They still recover one developmental architecture: false totality can be broken through transformed knowing, and realization is judged by what survives re-entry into life.
The disagreement locates the present limit. The operational sequence is recovered with higher confidence than the apex ontology that explains it.
False Convergence
The comparison space is enormous. Thousands of texts contain thousands of motifs. Unrestricted search guarantees impressive matches.
False convergence enters through several doors.
Parallelomania finds a resemblance, enlarges it, and then asserts a relation the evidence never established.
The Texas sharpshooter studies every available motif, circles the matches afterward, and presents the cluster as a prediction.
Universal symbolism assigns evidential weight to forms already expected from common bodies, landscapes, celestial cycles, and social structures.
Flexible recoding changes the translation whenever a source resists the model. Fullness becomes emptiness, a god becomes a state, a state becomes a frequency, and every contradiction becomes confirmation at another “level.”
Shared scientific costume mistakes modern vocabulary for ancient anticipation. An old text can describe vibration without describing electromagnetic frequency, quantum fields, or hertz.
AI-amplified pattern inflation searches a larger comparison space than any human reader, generates fluent bridges on demand, and conceals how many failed matches were discarded. Language-model synthesis is useful for retrieval and adversarial comparison. Fluency supplies no independence, provenance, or base rate.
The correction is structural. Predeclare the relation. Preserve failed matches. Count genealogical clusters. Name the transformation. State the rival. Hold something out.
The Prospective Test
Retrospective fit develops the reconstruction. Prospective fit tests it.
A serious test begins with a bounded training corpus. The analyst states the proposed invariant, the allowed translations, the forbidden findings, and the rival models before selecting a holdout. A source family not used to construct the model is then chosen through an external procedure. Independent readers code the held-out material without seeing the preferred conclusion.
The test scores:
- the predicted relational sequence;
- declared absences and contradictions;
- the number of flexible recodings required;
- the base rate of each feature;
- performance against rival reconstructions;
- whether historical contact explains the result.
The model fails when generic motifs perform as well as its specific relations, when success requires changing the code after reading the holdout, when the held-out tradition contradicts a required sequence, or when a simpler cognitive or genealogical model predicts the material better.
The ideal holdout is a newly translated or newly bounded source collection untouched by the model’s builders. A familiar tradition remains contaminated by prior knowledge. Until a clean test is run, the output remains a disciplined reconstruction with final proof outstanding.
What the Method Establishes
Metaphysical tomography grades competing reconstructions by how well they explain patterned agreements, transformations, and disagreements together. Every tradition remains answerable to the complete evidence.
The result may be one recovered architecture, several local architectures, a changing object, or no common object at the level tested. The null result protects the method from manufacturing the unity it was built to investigate.
Its strongest output is a graded invariant:
- recovered when independent instruments constrain the same unusual relation and prospective material confirms it;
- supported when the relation is specific and multi-method but the holdout test remains incomplete;
- modeled when the reconstruction organizes known material and exposes clear failure conditions;
- correspondence when the relation is illuminating but independence or mechanism remains unresolved;
- speculative when the match depends on contested provenance, flexible translation, or a single source family.
This hierarchy keeps direct knowledge, historical record, scientific measurement, and symbolic correspondence inside one inquiry without making them interchangeable.
The method also fixes a deeper principle of reading. No map contains the territory, but a map can contain a real cross-section of it. The task is to identify the transformation that produced the section and recover what remains invariant across the cuts.
That is how a large synthesis becomes cumulative. New material must change the reconstruction, sharpen a transformation, reveal a hidden dependency, or force the object to take a different shape.
Position
The traditions are neither obsolete superstitions nor mutually validating proof texts. They are inherited instruments aimed at a reality larger than any one instrument can contain.
Their overlap deserves confidence where the relation is independent, specific, constrained, and risky. Their contradictions deserve equal attention because they expose the angle of projection and the boundary of what has actually been recovered.
The result is disciplined consilience: one architecture made visible through irreducibly partial worlds.
References
Radon, Johann. “On the Determination of Functions from Their Integral Values along Certain Manifolds.” 1917. English translation in IEEE Transactions on Medical Imaging 5, no. 4 (1986): 170–176.
Herman, Gabor T., and Robert Davidi. “On Image Reconstruction from a Small Number of Projections.” Inverse Problems 24, no. 4 (2008): 045011. PMC.
Calvetti, Daniela, and Erkki Somersalo. “Inverse Problems: From Regularization to Bayesian Inference.” WIREs Computational Statistics 10, no. 3 (2018): e1427. DOI.
Meredith, William. “Measurement Invariance, Factor Analysis and Factorial Invariance.” Psychometrika 58 (1993): 525–543. DOI.
Whewell, William. The Philosophy of the Inductive Sciences. London: John W. Parker, 1840.
Wimsatt, William C. “Robustness, Reliability, and Overdetermination.” 1981. Reprinted in Characterizing the Robustness of Science, 2012.
Stegenga, Jacob. “Robustness, Discordance, and Relevance.” Philosophy of Science 76, no. 5 (2009): 650–661. DOI.
Mayo, Deborah G. Statistical Inference as Severe Testing. Cambridge University Press, 2018.
Lakatos, Imre. “Falsification and the Methodology of Scientific Research Programmes.” In Criticism and the Growth of Knowledge. Cambridge University Press, 1970.
Smith, Jonathan Z. “In Comparison a Magic Dwells.” In Imagining Religion. University of Chicago Press, 1982.
Sandmel, Samuel. “Parallelomania.” Journal of Biblical Literature 81, no. 1 (1962): 1–13. DOI.
Katz, Steven T., ed. Mysticism and Philosophical Analysis. Oxford University Press, 1978.
Smart, Ninian. Dimensions of the Sacred. University of California Press, 1996.
Mathison, Sandra. “Why Triangulate?” Educational Researcher 17, no. 2 (1988): 13–17. DOI.
Boyer, Pascal. “Cognitive Templates for Religious Concepts.” Cognitive Science 25, no. 4 (2001). DOI.
Griffiths, Roland R., et al. “Psilocybin Can Occasion Mystical-Type Experiences Having Substantial and Sustained Personal Meaning and Spiritual Significance.” Psychopharmacology 187 (2006): 268–283. PubMed.
Taylor, Steve. “A New Day for Perennialism: The Case for a ‘Soft Perennialism.’” Sophia 62 (2023): 731–759. DOI.