Independent review: complete SP Flood-cap family, C953–C983
Result: PASS, with two scope/locator clarifications already incorporated at C983. No arithmetic correction remains outstanding. The review reads the original File18 tables independently, compares the forward packet to those literals, recalculates the whole nine-row family, and checks the claims attached to the result. It does not treat the number of checks as evidence of statistical independence.
The controlling source is project_sources/38-File_18.Chronological_Data_Tables-20260919-141048-.md, SHA256 68301ab760bfbe2e874aedc5c8d943ced7a6219d34bc2859332a70713f9630d5. Relevant source sections are MT §2.1, SP §3.1 and §3.1.3, and main LXX §4.1. The independent script and exact results are evidence/independent_sp_cap_review.py and evidence/INDEPENDENT_C953_C983.json. Scope is the 31 journal actions C953–C983; C984 begins a separate Key-constraint block and is not reviewed here.
What the reconstruction explains
The forward packet holds the ordered SP completed begetting edges, the primary Noah placement and Noah-to-Flood relation, a baseline lifespan ledger, and the source-qualified minimum rule. It excludes the SP answer lifespan column. Those premises regenerate all nine SP births, capacities, lifespans, and branch-tagged deaths. This is useful explanatory compression: the three changed lifespans follow from one shared Flood boundary acting on the regular birth field, rather than remaining three unrelated substitutions.
It is a conditional forward reconstruction, not a blind prediction of unseen source material. The supplied study already contains the cap rule and SP architecture. The resulting agreement is a complete-family validation and shows how the Regular and Cumulative representations connect; it does not establish that an ancient editor constructed SP from MT. Indeed the main LXX baseline, retaining 753 for Lamech, gives exactly the same SP output. The cap discards the difference between the 777 and 753 inputs.
Counting and active support
Lamech's 53rd counted year supplies 52 completed units to primary Noah. It is resolved once in the birth chain; it is not a global one-year adjustment. The inclusive capacity is B−2893+1, equivalently B−2892, while the selected deaths of Jared, Methuselah, and Lamech remain at Flood start, 2893. The other six rows retain ordinary subtraction. C959 and C983 implement those distinct count tags correctly.
The capacities are 1307,1177,1072,982,912,847,785,720,653; the SP outputs are 930,912,905,910,895,847,365,720,653. The strict clipped support is Jared, Methuselah, Lamech, and the slack is 377,265,167,72,17,0,420,0,0. The minimum is the unique coordinatewise greatest feasible lifespan vector under the two upper bounds; the equivalent sum-maximization formulation is mathematical, with no claim about an ancient optimization procedure.
With k=Noah−Flood+1, retaining only the same strict support allows 584≤k<716. At 584, Mahalalel is tied at its capacity but is not strictly reduced. At 716, Jared ceases to be strictly reduced. The exact observed capped lives fix k=601; the total 5334+3k=7137 is an alternative use of the same data. The three pairwise synchronization equations have rank two. These distinctions prevent support, equal totals, and repeated cap equations from being counted as independent constraints.
Identifiability and lost information
The late-row incidence matrix has rank three and kernel (1,−1,0,0) on Jared/Enoch/Methuselah/Lamech begetting edges. Holding k=601, the capped lives recover Lamech 52, Methuselah 67, and Jared+Enoch 127. Holding the independently supplied Enoch 65 then recovers Jared 62. Without that premise, a Jared/Enoch exchange changes Enoch's birth but leaves every SP capped lifespan unchanged. The first five begetting columns do not enter the three active capacities. The rule therefore does not determine the entire genealogy.
At the other inverse boundary, an uncapped output fixes its baseline lifespan, whereas a capacity-valued output only supplies a lower bound. Even holding the three changed MT baseline lifespans as a multiset leaves four possible row assignments with the same cap output. These are finite ambiguity diagnostics; they are not admitted alternative chronologies or a target search. The equations establish conditional parameter identifiability and information loss, not historical direction.
Cumulative transmission, derivatives, frames, and rounding
The complete cumulative displacement is −488,−488,−488,−488,−488,−488,−373,−373,−124, with an implicit terminal zero at Noah. Adjacent differences recover the three local reductions. The distinct Eber and official Terah reductions contribute another −120, producing the inherited −608 total. The new achievement is explaining the pre-Flood support and mechanism; the total itself was already known.
The fixed-support derivative matrices are correct. Writing U for suffix summation and A for active-row selection, the lifespan and cumulative responses are A U and U A U. For absolute birth coordinates the full differential is dB=U db+1 dN; the simpler dB=U db requires fixed Noah placement, or relative B−N coordinates. C974 initially stated only the fixed Noah/Flood relation, and C983 explicitly supplies the needed placement qualification. The independent review recomputed all nine finite response columns and recovered all three matrices. The perturbations remain unadmitted sensitivity diagnostics, and threshold ties require branch care.
The admitted +215 paired frame translation preserves the entire cap field. The three local Noah/Shem paths each give 652 completed units and 653 inclusive units from Lamech to Flood start. They do not create companion birth stations upstream. C977's original source locator, §3.1.2, was imprecise: the three-path table is in §3.1. C983 records that correction. No upstream Gear transport is needed or claimed in this block.
Nearest-five rounding is monotone, so it commutes with minimum on the resolved count values. All nine rounded outputs and the retained strict support are correct; strict-support preservation is a property of this field, not a general theorem. Rounding the cap counts yields 845,720,655 for the three changed rows, whereas separately rounding their date boundaries before forming capacities yields 846,716,651. C980 correctly distinguishes these operations and does not promote the latter diagnostic into a canonical Rounded date table. The rowwise relative residuals 0,−1,+4 give the inherited +3 change, hence −608→−605.
The final classification is sound: a fixed-cap map is monotone, idempotent, and many-to-one. It extends the operation grammar beyond reversible frame changes. This remains a qualified operation on lifespan counts; it does not authorize recovery of discarded excess, new source variants, second digit inversion, or historical dependence claims. No quarantined primer was opened for this review.
Review disposition
- C974 derivative-placement qualification: accepted and recorded at C983.
- C977 local-path source locator: corrected by C983.
- Forward validation, conditional identifiability, support diagnostics, and historical-origin limits: adequately qualified throughout the block.
The block serves the Strategy's larger goal: it gives a complete, source-qualified route from ordered Regular ages through a shared event constraint into lifespan and Cumulative fields, and identifies exactly which information each representation retains or loses.