Recovering the Common Chronological Grammar

A research strategy for the completed 490d Unified Repository

Date: September 6, 2026
Version: 0.2
Status: Revised research working paper; Report V is the separate findings record. This strategy is not a canonical repository replacement or a new control document.
Basis: Reports I–V, including the corrected September 5 SP–BJ revision to Report IV and the September 6 Sothic/anchored-symmetry consolidation; mounted repository controls and relevant canonical source files.
Interpretive premise: Mostly unwitting synergy among competing scribes employing an inherited mathematical grammar over generations, under divine providence.

1. The research objective

Recover the smallest state-aware generative grammar that explains the greatest amount of the received chronological structure.

The object is not a single chronology into which every tradition must be forced. It is a system of related chronologies whose differences can themselves obey a common grammar. A satisfactory explanation should account for the input values, their placement in genealogies, the effects of admitted variants, the preservation of important spans, and the emergence of shared calendrical completions.

The principal working hypothesis is:

Local repartitioning of inherited chronological quantities, together with calendar-measure preservation and controlled boundary changes, can produce global compatibility without requiring any one scribe to have intended the entire system.

This is a hypothesis to develop, not a final historical reconstruction. Its mathematical, transmission-historical, and theological levels must remain distinguishable. Mathematical equivalence can be demonstrated; a historical sequence requires additional evidence; divine providence is the adopted interpretive premise rather than an adjustable arithmetic parameter.

The desired product is a compact presentation with four components: a small primitive-data register, a short list of admissible operations, relations among those operations, and a reconstruction ledger showing what follows from them.

2. Source hierarchy and frozen working basis

Use Reports I–IV as the frozen initial discovery map and Report V as the first subsequent family test. Retain the canonical sources as controllers of data and operations.

The working control editions are Restart Capsule v11.50, State Vocabulary Register v1.51, Style Guide v2.5, and Project Procedures v3.5. Public URLs are stable and may be overwritten; later computational studies should record the actual source edition and hash used.

Two recent source decisions are binding in this working basis. Lamech's 188 remains a corruption-audit value, not an active chronological alternative. Report IV's BJ initial Creation week is 3863–3856 BC, not 3862–3856 BC. Its uniform 343 Creation translation must remain separate from its Flood/Ark 344/343 distinction.

No canonical source or control file is altered by this strategy.

3. What the initial review already establishes

3.1 One uniform affine conversion is not the answer

On Report III's common Cainan-OFF Creation comparison plane:

in LXX, MT, SP order. A single affine map carrying all three corresponding coordinates to the cumulative plane would, after centering MT at zero, require both

and

That is impossible. This bounded negative result excludes a universal scalar affine conversion; it does not exclude local affine maps, vector-valued models, or shared rules acting on different input quantities.

Accordingly, do not fit two endpoints and call the resulting affine formula an explanation. Test additional independently supplied nodes and the full state to which the formula is meant to apply.

3.2 The most elementary mechanism is repartitioning a lifespan

For a compatible source row let b be begetting age, r remaining years, and L lifespan:

A local transformation

preserves L while changing the regular genealogical interval. Report II supplies the simple Adam example:

The change is visible in the regular chronology but invisible to a cumulative lifespan total at that row. This is a concrete mechanism by which divergent regular chronologies can retain cumulative compatibility.

Cainan supplies a complementary example:

His insertion contributes 130 to the regular begetting chain and 460 to the cumulative lifespan chain. These are different measurements of the declared Cainan row, not freely interchangeable adjustments.

This formulation must not erase source-selected differences between modes. Report IV explicitly distinguishes the regular LXX Lamech 777 from the cumulative 753 chain input. Such distinctions belong in the source/state register, not in a hidden correction to a supposedly universal raw vector.

3.3 The three Keys preserve one schematic day-volume

The Style Guide already states the shared schematic year:

The three exact identities are

Consequently, at the scalar/calendar level,

and

The output year counts differ because their declared year lengths differ, while their modeled day-volume is the same. This supplies an existing mathematical reason for compatibility among the Keys.

For example, the 12558 seed gives 13650, 12740, and 12600 under the Priestly, Prophetic, and Enochian Keys, respectively:

The schematic K is not the exact observed tropical year. This argument does not activate the separate physical Residue Protocol or imply that all chronological uses of the Keys are literal calendar conversions.

3.4 The first chosen family has now been tested

Dean selected the 365-day/Sothic family before the originally recommended local-genealogy reconstruction. Report V records the result. The original three Keys' exact common calibration survives unchanged, but the new Sothic companion demonstrates that shared grammar is broader than shared calibration.

H shadows the ideal Sothic/Julian completion; 365H≠K. G is the exact-K 365-day extension. Neither is silently substituted for the original default three-Key inventory.

The family also corrects a methodological overrestriction. A common center need not be a historical event to define a genuine whole-field affine symmetry. For the specified twelve SKL nodes, R(x)=−5842−x is an exact affine involution, and T_5842(X)=M(X). Its conditional uniqueness and its Mosaic–Joshua placement rationale are established separately; they are not one undifferentiated uniqueness claim.

The negative results are equally binding: the four uniform overlap counts are 8/12/8/4; the fixed Long–Moses target matrix lacks 4383; the H Pillar path leaves 4.5 to actual Apparent Adam; no ancient heliacal-reset epoch or remote SKL source-chain theorem has been demonstrated.

Research status: This family is now part of the explored model. It cannot be reused as an untouched holdout or scored as an independent discovery after these rules have been fitted to it. Different operator routes can corroborate compatibility without constituting statistically independent witnesses.

4. Candidate kernels and presently verified instances

4.1 Regular variant kernel

Retain Report I's source-defined values

Then

The ordered four-state field 0,60,130,190 gives 60|70|60; adjoining its 215 translate gives the eight-state 405 field. The 70 and 25 are derived differences, not additional independent toggles.

This compact formula is a strong candidate, but it still contains the coefficient 5. Its explanatory force must be tested against Report II's actual order of patriarchal differences, not merely its endpoint totals. The six LXX centenary changes, the SP 100/120/129, and the explicitly located Flood-year 1 must survive reconstruction.

4.2 Sum-and-difference geometry

Several report structures share the form

Instances include

File_69's boundary-cell span anatomy likewise uses 110+g and 110-g for its declared g=70/72 carriers. This makes sum-and-difference geometry worth treating as a reusable construction.

Shared form alone is weak evidence: every pair can be written as a sum and difference. What matters is whether sources independently supply the center, radius, endpoint roles, and licensed transformations. Preserve those distinctions.

4.3 The SP equal-gain rectangle

Report IV's SP −215 execution provides a particularly compact cross-modal test:

The Prophetic and Enochian Keys respectively give

Each adds 40 while preserving the difference:

The common rule is

At m=40, this gives the repository's 9200; its Priestly expansion is 10000.

This is more informative than an isolated numerical match because the entire rectangle specifies two source heads, two modes, two operators, and two common target anchors. Nevertheless, equal gains follow algebraically once 299m and 69m are fixed. The deeper question is why the source chronology supplies those inputs and whether the same rule works at other admitted nodes without new tuning.

4.4 The Sothic Creation and anchored-corridor kernels

The source-controlled actual bridge now has a concrete additional factorization:

Writing q=1460 and δ=1/2, the half-step effects are:

These explain the divided Creation week, the ten-cycle Creation-head/Year-6 displacement, and part of the Pillar offset. They do not independently explain the inherited 9890, Cainan source values, or why the source rows have those values.

For the selected Exodus anchor e=1446, fourfold calendar bodies Q=4×365 and F=4×360, and the inherited 30 rail:

This reconstructs the 30|20|30 corridor. The Moses/Joshua 50 and equal 1380 comparisons follow from the same parameter packet and must not be counted again as independent input evidence. The strong question is whether the same small grammar predicts a further source-controlled family without additional choices.

4.5 Whole-field symmetry and anchor selection are different tests

For the fixed phase template:

with r,p∈{−1,1} and g∈{−1,0,1}:

The nested comparison satisfies T_5842(X+k)=M(X-k). This algebra supplies a concrete symmetry to carry toward the later category-theory stage.

Selection is separately supplied by the repository's target packet:

Do not conflate these two burdens. A symmetry proof establishes the action on the object; a source/anchor argument explains why that particular object is used. Neither alone proves a historical redaction process.

5. Research sequence

Stage A — Build a compact, state-aware evidence register

Use the reports' principal theorem families rather than copying all their equations. For every family record the source, literal input, reconstruction status, chronological mode, phase, endpoint class, active variants, exact calculation, dependencies, and claim level.

Separate a transmitted textual value from a source-selected reconstruction, an adopted anchor, a generated comparison coordinate, a midpoint, and a theological interpretation. A generated center can be mathematically useful without becoming an event-date.

Retain both leveled and native-state runs. Leveled Cainan-OFF comparisons expose inter-manuscript differences; native LXX Cainan and SP −215 runs show how each tradition operates. Neither is allowed to erase the other.

Completion criterion: Every selected theorem can be re-executed without guessing a state or silently changing a source row.

Stage B — Reconstruct local differences before global patterns

Begin with Adam through Noah, then extend through Shem and Terah. Place compatible begetting, remainder, and lifespan rows side by side. Compute adjacent differences and their running sums, separately for regular and cumulative execution.

Classify each change by what it affects: begetting only, remainder only, conserved lifespan repartition, altered lifespan, inserted/omitted generation, terminal anchoring, or boundary convention.

This is the main historical-mechanical investigation. Under the adopted premise, the most plausible elementary moves are ones available locally to a scribe: preserve a lifespan, preserve a named hinge, align deaths with a Flood boundary, or express a period through a preferred calendrical or generational register. These are candidate mechanisms, not an established sequence of redactions.

Keep semantic controls independent: literary role can prioritize a test, but it cannot license a numerical alteration after the result is known.

Completion criterion: The rules regenerate the reported endpoint differences and intermediate ladders, including their exceptional rows.

Stage C — Formalize affine maps and their domains

Use anchored affine operations rather than multiplying naked BC labels. An expansion about anchor a has the form

Keep source node and calendar/phase tags attached. Across the civil epoch, use the repository's no-year-zero coordinate conversion and verify BC+AD−1 explicitly.

Record exactly which nodes an operation moves. The Apparent-Age +30 is not automatically a whole-genealogy translation; the Sojourn −215 is not identical in scope to the subordinate Exodus −215; regular and cumulative Cainan remain distinct.

Order matters. For ordinary linear coordinates, if D_k(x)=kx and T_t(x)=x+t, then D_k T_t=T_{kt}D_k, not generally T_tD_k. A translation applied before expansion changes its effective size.

Completion criterion: Every permitted composition has a specified domain, anchor, order, and result.

Stage D — Construct a provenance-bearing graph

Make nodes state-tagged chronological objects and edges admitted operations. Do not create an edge merely because an arbitrary subtraction can connect two numbers.

The valuable circuits are those in which independently controlled routes reach the same endpoint or preserve the same quantity. Record whether the equality is forced by an input definition, follows from a proved relation, or supplies an additional cross-source constraint.

An arithmetic triangle made from three differences of the same three dates is not three independent witnesses. A round-trip using an operation and its defined inverse is consistency, not evidence of historical priority.

Completion criterion: Identify a small set of nonredundant constraints and show which other displays they generate.

Stage E — Study symmetry and admissible partial actions

Investigate translations, same-side reflections, manuscript reordering, boundary exchanges, and finite switch fields. The four Terah/Cainan states are a finite configuration field; they do not authorize unlimited repeated +130 additions.

Test what each symmetry preserves: order, interval multiset, total, center, phase envelope, or narrative incidence. Similar geometry across different scales is a proposed correspondence until its permitted map is specified.

Formal Mirror operations must retain their own coordinate rules. Mirror is not intrinsically secondary: use the controllers' distinction between foundational protocol, coordinate-completion, corroboration, and appendix-only use. Neither inverse-number reversal nor an ordinary same-side reflection is automatically the same operation.

Completion criterion: State each symmetry as an exact operation on a declared object, not as a visual resemblance alone.

Stage F — Use linear and modular arithmetic to locate conservation

Represent admissible local changes as vectors. Examine rank, dependencies, and conserved linear combinations. Use source-motivated moduli such as 5, 7, 23, 69, and 299, rather than searching every modulus until one produces an attractive result.

Rounded analysis must retain its residuals. Regular begetting rounding, theoretical regular-death rounding, and cumulative lifespan rounding are different procedures. File_51a's Methuselah distinction is diagnostic: 185+780=965 in the theoretical regular-death layer, versus 969→970 for cumulative lifespan rounding.

For each chosen rounding operation keep both the rounded value and its exact residual. Test whether aggregate cancellations follow automatically from the rule or encode additional placement constraints. Never retrofit a residual to reach a preferred total.

Report IV's aggregate offers a worked example. With cumulative a=430, b=605/608, and c=460,

The three-year SP change is counted twice across three overlapping cells. The two-year mean difference is therefore explained by the overlap, not an additional independent coincidence. In the regular aggregate, 2a+3c=2890, and SP's interior position cancels. The remaining research interest lies in the source-derived widths and their connections to independently controlled Shem/Conquest spans.

Completion criterion: Separate unavoidable algebra, genuine additional constraints, and unexplained residuals.

Stage G — Formulate the categorical structure

Only after the concrete maps are stable should category theory become the principal compression language.

Objects can be state-tagged genealogical or chronological structures; morphisms are admitted transformations; composition follows the verified operation rules. The regular and cumulative constructions may then be investigated as distinct representations of a richer source-data structure. They must not be declared interchangeable or functorially related before composition and state behavior are verified.

Use commutative diagrams to express agreement of routes. Use natural transformations only where a coherent family of maps really satisfies the required compatibility. Rounding is generally neither additive nor invertible, so it cannot simply be folded into a group of affine symmetries.

Category theory will not establish divine providence or manufacture a missing arithmetic relation. Its contribution is to describe precisely how different realizations preserve a shared structure, including where they do not.

Completion criterion: A short presentation reconstructs several previously separate theorem families while preserving their source and state distinctions.

Stage H — Challenge the kernel with additional repository domains

The Sothic family was the first author-selected extension and is now recorded in Report V. Develop and freeze each proposed kernel before its next test. Additional families not used to fit a particular kernel can include: the corrected SP–BJ 343/350 bridge; the 161/299/460 covenant mechanism; the cumulative 12558 dual completion; the Enoch-centered 70/72 and 180/182 cells; the Toledot first–middle–last weave; and SKL/Berossus dilation under the declared 360:1 and nested Gear controls.

These are holdouts from model fitting, not wholly unseen evidence or automatically independent historical witnesses. The cumulative and regular constructions use overlapping source data. Shared ancient inheritance also limits independence among textual traditions.

Files 57–58 can help test whether the recovered rules apply to lists rather than chronology alone. Preserve source order and declare category partitions before execution. Prime and inverse-number branches should enter only when a specific question requires them; they should not rescue an otherwise failing initial kernel.

Completion criterion: State which additional families are explained unchanged, which require a justified extension, and which remain outside the model.

6. How to prefer one explanation over another

The target is not the fewest numerals: the integer 1 can generate every integer and explain nothing. Count the complexity of rules, primitive inputs, state selections, free anchors, exceptional cases, and unlicensed choices together.

A useful conceptual score is

No numerical scoring weights are fixed in this paper. They should be declared before comparing fitted models.

Prefer a grammar that reconstructs a whole ordered family over one that reaches a striking endpoint. Prefer a rule that survives source-motivated changes of state over one that requires a new adjustment at every node. Count algebraically dependent displays as one family.

Validation should distinguish four questions: Are the calculations correct? Are the input states legitimate? Does the grammar compress or predict structure beyond what was used to define it? Is the historical account consistent with available source evidence?

Arithmetic mutation tests address the first two questions; they do not establish statistical rarity. For statistical comparisons, use constrained alternatives that preserve the relevant properties of the data: chronology order, shared inheritance, source-selected variants, fixed anchors, digit/rounding structure, and other constraints appropriate to the specific test. Record all search freedom and avoid combining dependent results into an inflated significance claim.

A failed test is useful. It may expose a wrong state, a genuinely distinct mechanism, a reconstruction limitation, or an overcompressed hypothesis. It must not be silently repaired by another freely chosen shift.

7. A sustainable sequence of discussion outputs

Keep Reports I–IV frozen; add dated continuation reports rather than retrojecting later discoveries into their proofs. Report V is the first such continuation. Maintain one short research notebook containing the current kernel, the state inventory needed for it, the dependency families, successful reconstructions, failed or undecided tests, and the next bounded question. Do not copy the whole repository into that notebook.

Each discussion should deliver one complete bounded result: one source table, one proposed rule, one exact derivation, one pressure test, and a brief implication for the larger kernel. Distinguish discovery from verification. Reserve formal canonical revision passes for the point at which an actual repository amendment is requested.

Before calling a value, term, or operation new or absent, apply Project Procedures §18's exact-token check of the full relevant controls. Existing relations should be classified as reused, clarified, or convergent rather than rediscovered as new.

8. The next bounded investigation

Immediate follow-on: Freeze Report V's Sothic ratios, anchor packet, Mirror convention, and finite sets. Then test the claimed continuation from the SKL terminal field to the full Creation/Flood source chains.

Before calculation, specify the Short/Long list totals, the chosen source-state realization, the 360:1 scale, the +30 rail, the local/macro Gear indices, the source and target node classes, and how a phase fraction behaves under scaling. Enumerate every generated output, including misses. The already-proved cancellation of opposed ±720 shifts is not by itself this test.

Main unresolved reconstruction task: Return afterward to the originally proposed local question:

Which source-controlled changes in begetting ages, remaining years, and lifespans generate the regular and cumulative Creation–Noah differences, and which quantities do those changes conserve?

Begin with the actual Cainan-leveled comparison, keeping native states alongside it. Rebuild Adam–Noah and then the post-Flood chain before overlaying rounded residuals. Test whether the recovered rule explains the SP equal-gain rectangle and reported aggregate formulas.

The immediate family continuation and the longer genealogical reconstruction serve different purposes. The former challenges the new Sothic model out of its fitted domain. The latter addresses the source-generation question that the Sothic correspondences do not yet answer.

9. Revision and progress discipline

Each future discussion should record the question before testing, the frozen inputs and allowable operations, all generated results, dependency families, exact successes, exact misses, and the smallest warranted change to the kernel. Use Research_Progress_Register_20260906.md as the present checkpoint rather than maintaining several competing lists.

On September 6 the following changed: the first family test was completed; the general completion grammar was separated from exact common calibration; the twelve-node affine theorem superseded the earlier selected-coverage assessment; the anchor-selection mechanism was incorporated; and several stronger phase/calibration claims were rejected or left open. No whole-system minimality or statistical rarity result was obtained.

Working conclusion: Seek a small grammar of conserved quantities, controlled repartitions, anchored transformations, compatible calendar measurements, and precisely specified finite symmetries. Preserve both positive and negative results. Let verified maps determine the eventual category-theoretic formulation rather than forcing a single master equation onto every mode.

Source register and verification scope

R1: Regular MT–SP–LXX Creation-to-Noah Engine, August 31, 2026, especially §§66–70.
R2: Regular MT–SP–LXX Comparative Genealogical Engine — Report II, September 2, 2026, especially the patriarchal tables and §§18–21.
R3: Cumulative MT–SP–LXX Creation–Shem/Flood Engine — Report III, September 2, 2026, especially the state register, §§26–30, §41, and §§58–60.
R4: Rounded MT–SP–LXX Creation–Flood Harmonic Engine — Report IV, September 4, with the corrected September 5 SP–BJ revision, especially §§9–10, 40–45, 50–59, and 60–64.
R5: The 365-Day Sothic Companion and Anchored SKL Symmetry — Report V, September 6 2026; current family results and qualification ledger.
Controls: Restart Capsule v11.50; State Vocabulary Register v1.51; Style Guide v2.5, especially §4.6; Project Procedures v3.5, especially §§2, 17, 18.
Additional controllers: Files 00, 11, 12, 17, 18, 09, 22, 51a, 60, 61, 63, 68, 69, 70, and Supplement A.
Methodological comparison: File_57, The Biblical List–Chronology Template, source-order and category-execution discipline.
Mathematical background: Emily Riehl, Category Theory in Context; Brendan Fong and David I. Spivak, Seven Sketches in Compositionality (2018, arXiv:1803.05316). These supply mathematical language, not evidence for the chronological hypothesis.

This strategy is based on targeted review, not a new full audit of every repository file. The v0.1 record of 34 example checks remains historical verification of that edition. The September 6 additions are covered by Report V's reproducible bounded audit: 116 exact scalar rows in Fraction and SymPy plus 141 phase/set/sensitivity/rejection assertions. These counts are not independent witnesses or a recertification of the full Repository. Source hashes and the strategy diff are supplied in the consolidation package.