# SGTOE Rev33.1_S369 suite — PAGE-PRESERVING TEXT EXTRACTION — PART 3 (pp 201–300) (house, S369 2026-09-26). Source PDF: SGTOE_Rev33.1_S369_suite.pdf, 15800235 B, 433 pp, # sha256 b64ddcd9e16a6dcdc7ba41c97c5e1b8d087e72e82201ba581bc28db946779c34 (the sealed release PDF; the served copy is byte-identical). Tool: pdftotext -layout, one call per part split on form feeds. # Every page begins with a line '===== PDF PAGE n / 433 =====' — cite that n as the PDF page. Layout text is lossy (math symbols, tables); where a quote matters, # the tex file:line in the release bundle is authoritative. This extraction is a READING AID pinned to the PDF above; it is not a second edition of the suite. ===== PDF PAGE 201 / 433 ===== Theorem 6.1 (energy is T -even). On the certified 56 the compact clock generator K satisfies T K T −1 = +K (energy T -even), while the non-compact grading H = D56 /2 satisfies T H T −1 = −H (boost T -odd). The clock Hamiltonian is K, and U (θ) = exp(θK) is a bounded unitary flow. Remark 6.2 (coalition record: Defect-176). An earlier coalition proposal (the s109-era “Hamiltonian- bridge” model) promoted the grading to the Hamiltonian and imposed T H T −1 = −H as the energy relation. The in-house substrate audit s112b_hamiltonian_bridge.py (run; defects ∼ 10−6 ) shows that relation belongs to the boost D56 /2, not to a clock: the energy is the compact K and is T -even. The proposing model subsequently concurred. This is recorded as the convergent correction, not hidden. Honest scope of the time bridge (S112–S113), re-framed against Appendix I (Rev29 ). A T -even clock is necessary for a stable spectrum but is not by itself a proof of stable dynamics. Two different potentials must be kept apart; earlier revisions of this paragraph conflated them and drew a dynamical conclusion from the wrong one. (i) The ungauged invariant probe potential on the J3 (Os ) slice. There the golden point Jvac = diag(φ, 1, φ−1 ) has Hessian signature (15+, 12−), and the 12 negative direc- tions are the negative-norm directions of the split-octonion norm (sig = (4, 4)). Probe s113a_eta_stabilizer.py (run) establishes that the unique G2 -invariant quadratic on each octonion block is that indefinite norm, so no G2 /F4 -covariant scalar-potential term can change that signature: a covariant uplift η N stays (15+, 12−) for all η. Only a G2 -breaking compact (Euclidean) uplift reaches positive-definiteness, at the exact value ηc = 2φ2 (the Appendix-I “∼ φ2 ” figure, in that probe’s 12 Tr normalisation). All of this is a fact about the ungauged invariant toy, and it is exactly what s113a certifies — no more. (ii) The gauged dynamical statement supersedes the “saddle” reading. Appendix I retires the reading of (15+, 12−) as the dynamical verdict on Jvac . On the certified gauged E7(7) /SU (8) bed at the dyonic point ω = π/4 (Appendix I, Theorem “the golden vacuum is a tachyon-free Minkowski critical point at ω = π/4”; kernel s120c_golden_omega.py) the split directions are organised by the gauging, and the golden point is a tachyon-free V = 0 Minkowski critical point of Hessian signature (0− , 340 , 36+ ). There is no tachyon at the golden point on the gauged bed. The Rev16/Rev17 “saddle” language, which this appendix carried forward, is withdrawn as a dynamical claim and retained only as the statics of the invariant toy. (iii) What does not depend on the withdrawn reading. The (4, 4) Lorentzian-signature seed is a statement about the split-octonion norm and the uniqueness of the G2 -invariant quadratic on each octonion block; s113a certifies it directly, and it was never a consequence of Jvac being a saddle. It survives the withdrawal unchanged. Likewise the missing-16 portal failure (s112c) fails because no G2 /F4 -covariant invariant term can supply a definite quadratic on the octonion blocks — again a statement about the invariant family, not about the critical-point type of Jvac . Both conclusions stand as printed. (iv) Status. The time bridge stands as a clock/kinematic result. Dynamical stabilisation is not “open” in the sense this paragraph previously printed (an open item awaiting the gauged sector): the gauged sector has supplied a tachyon-free critical point. What remains open is narrower and is named in Appendix I — the lift of the residual σ-even flats, pending the certified field-dependent tensor Ciab , plus a separate no-scale dilaton input. Appendix I’s own operative status, stabilization partial (one named open tensor), governs; this appendix makes no independent stabilisation claim. 5 ===== PDF PAGE 202 / 433 ===== 7 The 32 → 64 doubling, honestly (S107a) The banked maximal commuting set of in-substrate involutions (the F4 -class flips) has order 25 = 32 (S106k). T normalises the E6 structure algebra: T D56 T −1 = −D56 , so T maps the centraliser centraliser(D56 ) = e6(6) ⊕ u(1) (dimension 79) to itself (residual ∼ 1 × 10−9 ), and e6(6) contains f4 and those 5 flips. Hence adjoining T genuinely doubles the flip group, 32 → 64. Because T 2 = −1 is central, the doubling is projective: ⟨T ⟩ = Z4 (a spinor line), so the strict (Z2 )6 = 64 of the abstract model becomes a Z4 -refined statement (the sixth/time line is a spinor, not a plain Z2 ). Provenance note (honest scope of §6). The 32 → 64 statement here rests on two certified facts: (i) T normalises e6(6) on the live 56 (s107a gate f, residual ∼ 10−9 ), and (ii) the in-substrate involutions have 2-rank 5 (= 32), established on the 27 in s106k. An explicit live-56 enumeration of the five commuting F4 -class flips with T adjoined — verifying ⟨flips, T ⟩ has order 64 directly — has not yet been run; it is recorded as future work. The doubling is therefore stated via normalisation plus the cited 2-rank, and the Z4 (rather than Z2 ) character of the new line is the load-bearing refinement. 8 One circle: the dial is the grading circle (S108b) √ The gauge-side dial moment-circle Cn = [µ1 , µ2 ]/κ (κ = 1/(3 2), Sessions 101d/104a) and the grading circle K of §4 have the identical spectrum on the 56, i · {±3 (×1), ±1 (×27)} (with Cn normalised by κ/2): the same su(2) content. They are conjugate inside the group — an explicit g ∈ E7(7) (built by a Lie-algebra Newton flow, every step the exponential of a genuine e7(7) element; symplectic defect ∼ 1.7 × 10−11 , det g = +1) satisfies g Cnnorm g −1 = K (residual ∼ 8.9 × 10−10 ). (5) Consequently g intertwines the whole circles, half-turns included: g (dial half-turn) g −1 = T (residual ∼ 7.2 × 10−9 ), and the half-turn squares to −1 on both. In the grading frame one dial moment carries the future 27 onto the past 27 (leak ∼ 7 × 10−14 ). Novelty label: physical interpretation not established here. Theorem 8.1 (one clock). The dial moment-circle and the grading circle are E7(7) -conjugate; hence the dial “moment” and the time-bridge T are the same half-turn, and the two 27s of the live 56 are the dial’s future and past windings. No new constant is introduced. This settles the two-27s relation: the moment-dial clock and the future/past grading clock are one circle in two frames. 9 Moment span versus per-particle thickness (S108c) The moment span is shared: T sweeps π of the 2π grading circle (span fraction exactly 0.5) and carries the future 27 onto the past 27, so one moment = 12 future + 12 past, with T 2 = −1 the universal “tick” — a single envelope for all 56 states. (The companion “past→past, future→future” four-tick return, T 4 = +1, is the −1 holonomy T accrues on reflection at the Det = 0 wall of 6 ===== PDF PAGE 203 / 433 ===== Appendix I; an external hostile review (S175, D905) confirms this reading is sound classical holonomy — a structural clarification of the turnover, not a source of any coupling.) The thickness is per-particle, at two layers: (a) Grading rate. Under exp(θK) a state’s phase advances at rate |H-eigenvalue| = |level|/2, so the singlets (level ±6, rate 3) tick three times as fast as the 27 matter (level ±2, rate 1) — a 3:1 ratio, not one shared rate. (b) Compton rate. Each fermion slot of the ladder is assigned ωf = ω0 φnf with nf = logφ (mf /me ), so its model proper-time thickness is τf = 1/ωf = (1/ω0 ) φ−nf . Heavier ⇒ thinner/faster; e.g. the muon ∼ 1/199 and the tau ∼ 1/3477 of the electron thickness; within one electron-moment the slot completes φnf internal turns. These are quantities of the construction, computed from external PDG masses; no proper time of a physical particle is being predicted (Rev29 ). Net: the moment is the shared frame ( 12 + 12 ); the thickness is the particle’s own (grading 3:1 times golden depth φn ). The PDG pole masses entering the muon/tau thickness figures are external inputs. 10 Engine-side capture: the winding ladder and electron-shell closure (Track-1, S164/S166/S167) The per-particle thickness τf = (1/ω0 ) φ−nf of §9 is one face of a single integer ladder that the engine workbook records in three tabs (Electron Shell Levels, S164; Winding→Mass (forward), S166; Exact Windings↔Energies, s264/S167) but that no earlier tex captured. This section folds that content so the paper and the engine agree. All of it is scale-free structure on the certified substrate; the one dimensionful ingredient (an absolute energy or mass) is, as of Rev25, precisely the framework’s single registered input — the depth-cylinder Bargmann class Ldepth /Lσ , registered with value and theorem chain in Appendix X (consistency: one dimensionful contact point here, one registered input there; they are the same object read in different frames) — and the occupation selector is Derived-Conditional (Appendix M, Rev25 section, {31, 39} live falsifier). Nothing here moves a public observable. (a) Electron-shell closure: the level integer is an orbital winding On the hydrogen Bohr ladder the de Broglie wavelength λn = h/(me vn ) with vn /c = α/n obeys, around the n-th shell circumference Cn = 2πrn , Cn #de Broglie wavelengths around shell n = = n exactly, (6) λn the scale-free residue of the identity a0 = λC /2πα (CODATA-rounding residual 1.4 × 10−9 ). Thus the shell integer n is an orbital winding number — how many times the internal phase closes around the orbit — and is firewall-safe: the ratio En /E1 = 1/n2 is scale-free, while the absolute En (eV) needs α and is fenced. This is the same closure mechanism as the ±30 four-part doubling (s249) and the golden rungs un = n ln φ (s251). 7 ===== PDF PAGE 204 / 433 ===== (b) The golden-ray boost rungs and the AX1 vacuum Write the boost frame J(u) = diag(eu , 1, e−u ). The rungs sit at rapidities un = n ln φ, so eun = φn exactly, and the golden vacuum Jvac = diag(φ, 1, φ−1 ) is the rung-1 boost. The single load-bearing identity is the relativistic-boost reading of the golden axiom (s251): cosh(2 ln φ) = 32 ⇐⇒ φ2 + φ−2 = 3 (the DET-7 condition; equivalent to AX1 for ρ > 1), (7) i.e. the golden vacuum is exactly the rapidity at which the boost’s cosh equals 3/2; “harmony of phases” is this boost, with vϕ vg = c2 and the rest clock Hrest = ω0 Pcode precessing at ω0 (its bilinear runs at the zitter rate 2ω0 inside the model; no laboratory observable is asserted, Rev29 ; s252). No new constant enters: φ is AX1. n un = n ln φ +mode φn −mode φ−n product (det) orbital winding 1 0.48121 1.618034 0.618034 1 1 2 0.96242 2.618034 0.381966 1 2 3 1.44364 4.236068 0.236068 1 3 4 1.92485 6.854102 0.145898 1 4 5 2.40606 11.090170 0.090170 1 5 6 2.88727 17.944272 0.055728 1 6 7 3.36848 29.034442 0.034442 1 7 Table 1: Exact windings↔energies (s264): the two null modes of the golden-ray boost J(u) carry +u (explosion) and −u (implosion); det = 1 locks them equal-and-opposite (product φn φ−n = 1), and T swaps them. (c) Rest mass is a T -conjugation invariant The two columns φ±n are the two metric-signature null modes of J(u); det = 1 keeps them synchronised and the live half-turn T (Theorem 4.1) maps one onto the other. Because T ∈ Sp(56, R) with T 2 = −1 is a conjugation, the loaded operator’s eigenvalues are invariant under it. Novelty label: physical interpretation not established here. Theorem 10.1 (the rest-mass rung is invariant under winding, s264). In this construction the rest-mass rung is carried by the T -conjugation–invariant spectral eigenvalue of the loaded clock operator ( Rev29: a representation of rest mass inside the model, not an identification of physical rest mass with a substrate eigenvalue). The orbital winding n (number of de Broglie waves closing the orbit) and the forward/backward density may vary across shells, but the rest-mass rung (beat-depth) is fixed — for the electron it is the anchor nf = 0 at every shell. “Wound up more/less” is orbital n, not the rest-mass rung. The elementary winding cell is the four-element set {±} × {φ, φ−1 } read off the Z/60 closure dial as golden harmonics 2 cos(2πk/60): the combinations (+, φ), (+, φ−1 ), (−, φ−1 ), (−, φ) sit at k = 6, 12, 18, 24 with values 1.618034, 0.618034, −0.618034, −1.618034. The det = 1 sync collapses the four lattice combos to two T -conjugate pairs, [(+, φ) ∼ (−, φ−1 )] and [(+, φ−1 ) ∼ (−, φ)]. 8 ===== PDF PAGE 205 / 433 ===== (d) The forward winding→mass map (descriptive; selector open) Read forward, the same ladder is the one-parameter-free mass estimate mpred = me φN , using only φ, the electron anchor me , and an integer winding N . The fit is clean for leptons and MS-bar-fuzzy for quarks (Table 2); the beat-depth nf = logφ (m/me ) is read backward from PDG and is descriptive. Crucially, the integer winding N — equivalently which rungs {0, 3, 11, 17, . . . } are occupied — is the selector: it is not forced here; its status since Rev25 is Derived-Conditional (Appendix M, three-layer selector; live falsifier {31, 39} — one status, stated Rev32.9). Electron 0, muon 11, tau 17 are near-integer “clean” rungs; the up quark sits alone at N = 3; the remaining quarks carry genuine MS-bar tension. particle q mPDG (MeV) nf = logφ m m e N [open] mpred = me φN residual % electron e− −1 0.51100 0.00 0 0.51100 0.0 muon µ− −1 105.658 11.08 11 101.691 −3.75 tau τ − −1 1776.93 16.94 17 1824.78 +2.69 up u + 23 2.16 3.00 3 2.165 +0.21 down d − 13 4.70 4.61 5 5.667 +20.58 strange s − 13 93.5 10.83 11 101.691 +8.76 charm c + 23 1272.9 16.25 16 1127.77 −11.40 bottom b − 13 4186.0 18.73 19 4777.33 +14.13 top t + 23 172600 26.45 26 138707 −19.64 Table 2: Forward winding→mass (S166), with the Rev32.2 comparator refresh. Charge q is the raw (DIAL-1) winding; mass is golden beat-depth. The heavy-quark inputs are PDG 2026 (mc (mc ) and mb (mb ) in MS, and the top direct/MC pole proxy), while mpred is the winding-model value. The displayed residuals are signed percentage offsets, not same-scheme tensions; N remains the open selector. Reconciliation (Track-1↔Track-2, s247–s254). The charge-spurion is orthogonal to the diagonal involution (s247); the de Broglie clock Hrest = ω0 Pcode and the golden-ray boost cosh(2 ln φ) = 3/2 are the kinematic content of the electron-shell tab (s251/s252/s254). What this section does not claim: no absolute mass or energy is derived (all α-/anchor fenced, A962); the integer winding/occupation selector is open (s263); the confinement √ string tension σ and related hadronic numbers are out of scope and live fenced in Paper 8. Engine = paper after this fold; no public observable moves. 11 The golden bridge: the braid word’s symplectic square root (Rev31; S276, A1512–A1513) The winding ladder of §10 carried its golden content numerically; this section records its algebraic root, proved in the post-closure clock round (A1512 ask-3, house-verified s1047–s1050) and placed here because it lives next to the s264 winding table: it is the same two-mode null-pair sector, now produced inside the frozen braid representation with no numerical φ input anywhere. Novelty label: new specialization proved here. Theorem 11.1 (the golden bridge, A1512). Within the frozen 56-frame let Wh = Q + 32 P + 32 γ1 − γ2 9 ===== PDF PAGE 206 / 433 ===== be the braid hyperbolic word and set A = Wh − Q, so that A2 − 3A + P = 0. [LIB2-015] Then A+P Hϕ = Q + √ 5 is an exact symplectic unimodular square root of the braid word, Hϕ2 = Wh , with spectrum spec Hϕ = φ−1 (12) ⊕ 1(32) ⊕ φ+1 (12) , boost generator G2 = P , Lorentzian form η = ΩG of inertia (12+ , 12− , 320 ), and a swap operator Tϕ with Tϕ2 = −I exchanging n ↔ −n. The twelve ± pairs are golden-ray winding doublets: the algebraic realization, inside the frozen braid representation, of the s264 two-mode null-pair sector with its det lock φn φ−n = 1. √ The 5 is not inserted: it is the discriminant of A2 − 3A + P = 0, i.e. of the trace-3 identity φ2 + φ−2 = 3 (the DET-7 condition, equivalent to AX1 for ρ > 1) that Appendix E obtains from its conditional selector theorem, in which DET-7 is a structural postulate (Rev32.9). The ladder’s golden rungs and the braid grammar are therefore one object: iterating the bridge is the ladder, and the half-step (the spinor double cover) is the φn column of the s264 table. Fence (carried verbatim from the round; binding). The occupation rule, the middle Peirce mode, the physical clock, and the energy spectrum are not derived here. And the occupation fence is now at theorem grade (A1513): the grammar-natural braid occupation class is closed negative — the algebra acts identically on all 12 active doublets (commutant dimension 144) and on all 16 middle pairs (256); only whole multiplicity blocks are selectable. An occupation law therefore requires a Hamiltonian- or boundary-typed selector; nothing internal to the braid grammar can pick individual rungs. This fence travels with the bridge, never separately: the integer winding N of Table 2 remains the open selector, exactly as before this fold. No public observable moves. 12 The two-sheet wormhole–black-hole throat (S230–S233) The split-octonion algebra Os carries both quaternion types: a division (4, 0) and a split (2, 2) subalgebra, giving two (1+3) sheets that share a common 3-plane and are exchanged by the Möbius Z2 (the second lattice). The physical metric is then forced per real-form sheet: the compact-H sheet selects the determinant (spin-factor) metric of signature (1, 3), while the split-H′ sheet selects the Koecher–Vinberg Hessian metric of signature (3, 1). That these two intrinsic metrics select opposite C-lines is the honest obstruction, named rather than hidden. The whole object is a Z2 ×Z2 : the throat involution σC (a signature-preserving Jordan automorphism, the WH–BH neck flip) and the real-form involution W (the signature-changing sheet flip) are distinct axes. The associated Wick rotation is the real-form flip E7(−25) ↔ E7(7) , i.e. SO(8) ↔ SO(4, 4) (this corrects an earlier SO(6, 2) reading of the throat axis; the SO(6, 2) of the gauge-boundary shadow selection in Appendix F is a separate object and stands). No public observable moves; this is structural geometry at conditional tier. 10 ===== PDF PAGE 207 / 433 ===== 13 Fences F1. The rotation is the substrate clock (occupation/phase); ω0 is an input frequency anchor; no seconds and no external-time dynamics are asserted. F2. No Standard-Model particle is identified; the 27/27 are substrate sectors. √ F3. No new constant: π, φ, κ = 1/(3 2), the grading, and ω0 (= q0clk in the Compton frame) are all intrinsic or already-declared inputs. F4. Interface fence (Rev29 ). Every object in this appendix — the clock circle, the moment, the half-turn T , the thickness τf , the winding rungs — is a constructed-layer object on the certified substrate. Constructed-layer objects are not asserted to be physical-layer objects: T models time reversal, the flicker frame calibrates against the electron Compton rate, and a thickness represents a proper time. Sentences of the form “X is the physical Y ” are retired throughout; where they survived earlier revisions they have been re-typed rather than deleted, and every number they carried is unchanged. 14 A clock candidate, killed: the DJ operator carries no golden content (Rev30, A1479/A1480) An operator DJ built from log P (J) was proposed as a golden clock and adjudicated directly against the fences of this appendix. It fails, and the reason is worth printing because it is a normalisation artifact rather than a subtle dynamical obstruction. The computation. For J = t1 e1 + t2 e2 + t3 e3 with all ti > 0, the operator log P (J) has eigenvalues log ti +log tj — multiplicity 1 for i = j and 8 for i < j. So log P (J) is linear in log t: it is the grading derivation of the frame and nothing more. In closed form DJ = LDφ with Dφ = diag(1, 0, −1). The kill. Normalising J = t e1 + e2 + t−1 e3 by 2 log t makes the t-dependence cancel identically. φ does not survive its own normalisation. DJ is the Peirce grading operator of the frame, carrying zero golden content and zero vacuum content: it cannot distinguish Jvac from any other regular element sharing that frame and ordering. A companion transfer-operator reading collapses the same way — the associated TJ is exactly diagonal, so its “theorems” (strict positivity, reflection positivity, self-adjointness, Gram exactness) are identities for a positive diagonal matrix and not theorems at all. All three objects offered are affine in the Peirce grading; they are one result presented three times. What this does and does not disturb. The time fences of this appendix are untouched and the adjudication was made against them: the tick remains phase structure, never frequency or time, and the B2-time no-go stands. Nothing here reopens a dynamical reading. One caution on provenance: the first refutation was made by type analysis — argument, not measurement — because no banked clock matrix could be located on disk; it was subsequently confirmed by direct computation of the closed form above, and it is the confirmation, not the type argument alone, that carries the result. Nothing in this section promotes an observable, moves a tier, or touches the engine. 11 ===== PDF PAGE 208 / 433 ===== 15 A second clock candidate, killed: the dilaton / SO(1, 1) grading clock (S283–S284, D737c) The DJ kill above has a sibling, adjudicated in the same programme of work against the same fences, and it fails for three independent reasons rather than one. It is printed here for the first time; the result has stood in the house census since S284 and reached no paper until Rev33.0. The candidate. The dilaton direction of the D = 5 → D = 4 reduction, with its SO(1, 1) grading, was offered as the generator of a physical clock — the most natural candidate on the list, because the grading already flows. The kill, in three parts, any one of which is sufficient. 1. The candidate quadratic form is sign-indefinite, with inertia (28, 28, 0). It is therefore not a positive generator and cannot serve as a Hamiltonian on a physical state space: what it generates is the noncompact grading overlay, not a reduced positive flow. 2. It is decoupled from the mass proxies, so even granted a flow it would not move the objects this suite reads. 3. It presupposes its own conclusion. The candidate is born inside the D = 5 → D = 4 spacetime reduction — that is, inside the very structure it was asked to derive. A clock recovered from a reduction that already assumes spacetime has derived nothing. Type. The route is closed as a derivation. Its geometric and intuitive use survives at interpretation tier only, and nothing in this suite may cite it above that. Why this kill is printed, and what it does not disturb. The fences of this appendix are untouched: the tick remains phase structure, the B2-time no-go stands, and no observable moves, no tier moves, and the engine is not touched. The point of printing a closed negative is that the closure is a result — the most natural clock candidate in the reduction is dead, and dead for a reason that generalises: a grading is not a clock, and neither is a coordinate. The full standing requirement for any future promotion — a positive reduced generator, a relational clock observable, covariance, a continuum reconstruction, unit registration, a readout theorem, and compatibility controls — is recorded in the house census (Trackers/PHYSICAL_TIME_WALL_CENSUS_S284.md); none of the eight is supplied by the parent Legendre transform, the compact K algebra, the page picture, the AdS seed or the finite OS control, alone or together. Provenance, stated because the delay is part of the record. The kill was derived and typed at S283–S284 and recorded in a tracker; it reached no component of this suite for sixty-nine sessions, and was recovered by the S264–S352 master sweep at S353 (Trackers/LOST_SHEEP_MASTER_SWEEP_S353.md, row S353-12). Its same-session twin, the continuum census, was cited by path in Appendix E at the time. One reached the papers and one did not, and nothing in the house noticed the difference for sixty-nine sessions. 16 Interior consolidation (Sessions 89–105) The time layer√ sits atop a certified interior built in Sessions 89–105: the exact dial su(2, R) algebra with √ κ = 1/(3 2) (S101d), the winding-lattice identity exp(T ·Cn ) = 1 (S102b), the clock form 2 3 (B−L) + 12 L̂ (S102e), the Hilbert/occupation completion (S104a), and the shadow-stratum 12 ===== PDF PAGE 209 / 433 ===== centraliser (S100h). These are recorded in the _S89–_S105 manifests and are summarised here rather than reproduced. Provenance footer. _S106_manifest, _S107_manifest (10/10), and _S108_manifest (10/10) all verify; replays byte-exact; state_check green; zero tex was written during Sessions 106–108 — this appendix is the first paper capture of the time layer. Posture: Definition A (derivation program in progress); the time layer is structural and moves no public observable. 13 ===== PDF PAGE 210 / 433 ===== Leibniz Quantum Beats Newton Appendix I — Dynamics of the Golden Vacuum: the Derived Dynamical Selection on the E7(7) Substrate, the σ-Odd Defect Mechanism, the Geometric Bounce, and the E7(7) Clock (Rev33.1) Derived σ-odd-defect selection (modulo the AX1 unit); Coleman–Weinberg cannot uniquely select golden in the tested σ-equivariant layers; stabilization partial (one named open tensor). Definition A — no public observable moves. Tom O’Sieg August 2026 Abstract The three external hostile reviews of the Rev14/Rev15 suite converged on one gap: the framework had no dynamics — it presented a fixed algebraic correspondence map with the golden vacuum Jvac = diag(ϕ, 1, ϕ−1 ) inserted by hand, inviting the “numerology” charge. This appendix folds the in-house dynamics layer (S120–S124) that answers the gap at the level it can now honestly reach, on the certified E7(7) 56/133 substrate. The headline is a derived dynamical vacuum: scanning the dyonic ω-deformation Θ(ω) of the shadow-selected SO(6, 2) gauging ( = SO∗ (8)), the electric runaway present at the origin cancels against its magnetic completion at exactly ω = π/4, where the golden vacuum Jvac is a tachyon-free Minkowski (V = 0) critical point with Hessian signature (0− , 340 , 36+ ) (§1; this supersedes the earlier compact-slice “saddle” reading, which is now understood as the statics of the ungauged invariant potential). On this dynamical bed the selection of ϕ is then derived, modulo one named unit: the one-loop Coleman– Weinberg (CW) supertrace Str M 4 is σ-even under the Jordan involution σ : ρ 7→ 1/ρ — bosonic (§2) and indexed-fermionic, the latter bein-independent — so CW is extremized at the symmetric point ρ = 1, never at golden, in every tested σ-equivariant layer; the σ-odd unit defect-source B is the sole selector, with its defect equation F (ρ) = ρ − ρ−1 = 1 identical to the golden axiom AX1 ρ2 − ρ − 1 = 0. A BPS / source-completion analysis (§3) derives the direction of B (the σ-odd grading K) and the completion + 12 ∥B∥2 (a manifest-square constant; the bare linear spurion alone gives anti-de Sitter V = − 12 ), with a physical radial mode of curvature γ 2 + 4, i.e. m2 = 7/4 at the unit point; the unit magnitude itself is not independently BPS-forced — it is equivalent to AX1 — so selection is derived modulo the AX1 unit (the framework’s core axiom), not “from nothing.” Stabilization is partial: of the 18 physical moduli the 9 σ-odd directions are lifted by the defect (8 portal +1 radial m2 = 7/4); the 9 residual are all σ-even, splitting as 8 CW-liftable flats whose lift lives entirely in a single named field-dependent tensor Ciab = 12 ∂a ∂b m2i (provably undecidable from σ-evenness and the point spectrum alone — the one deferred stabilization target, §4), plus 1 no-scale dilaton that is provably not CW-fixed P (a separate scale/lift input). The generation count appears as a clean Morse index χ(OP 2 ) = (−1)idx = 3 over the cells {0, 8, 16} (§5; the generation↔cell assignment is the non-finite A1 datum, fenced open). The earlier probe layers are retained: the overdetermined invariant routes to ϕ (§8), the classical geometric bounce (§9), the E7(7) clock connection with T 2 = −1 (§10), and the moment–annulus synthesis (§11). Everything is fenced and Definition A: no public observable moves, and the dimensionful mass scale remains an Input (the Honest Boundary). 1 ===== PDF PAGE 211 / 433 ===== Binding fence (read first). This appendix folds a derived selection result, not a completed dynamics. Concretely: (a) the dynamical vacuum (§1) is computed on the certified N = 8 E7(7) /SU (8) potential at the dyonic point ω = π/4 — real and tachyon-free — but the transverse 54 runaway of the non-compact gauging is a generic property of SO(6, 2), present already at the origin, fenced separately from golden; (b) the selection of ϕ is derived modulo the AX1 unit: the direction of the defect source and its completion are derived, the unit magnitude γ = 1 is equivalent to AX1 ρ2 − ρ − 1 = 0 (the framework’s banked golden axiom), not derived here from nothing; (c) stabilization is partial — the 8 σ-even residual flats’ lift is undecidable without the certified field-dependent tensor Ciab (do not claim “CW lifts the 8”), and the no-scale dilaton is a separate scale/lift input; (d) the bounce is a classical metric-completeness statement; (e) the dimensionful mass scale is not derived (Appendix X, Honest Boundary). Nothing here is Proved-tier and no public observable moves. The certified Ciab build is the single pre-registered next computation (§4, §14). Label convention. The Theorem/Proposition/Lemma environments in this appendix denote certified in-house computations — a Definition-A certified-computation tier: machine- verified on the certified E7(7) substrate and reproducible from the bundled kernels — not Proved-tier analytic theorems in the suite’s claim-language sense. No claim here is Proved- tier. Notation (σ) — four senses meet in this appendix, and this is the one that matters most. Throughout, σ means the Jordan frame involution σframe : ρ 7→ 1/ρ, the Peirce-frame transposition (1 3) with odd grading K = diag(−1, 0, 1) — and in particular every use of σ-even and σ-odd below refers to it. Three other objects share the letter and are not it: Appendix H’s throat involution σC ; Appendix F’s SO∗ (8) time-mirror σ = Ω; and the split-octonion conjugation σoct (e7 7→ −e7 ), the only registered up/down-type exchange, which is computed in the kernels √ and never appears in the typeset suite. A fourth, σ, is the string tension and not an involution. σframe ̸= σoct is computed, not assumed (house kernel s1183). 1 The derived dynamical vacuum: the dyonic ω = π/4 point The first dynamical content is a vacuum that is selected by a potential, not posited. We work on the certified N = 8 gauged-supergravity bed: the shadow-selection analysis of Appendix F fixes the gauge group to the non-compact SO(6, 2) (the unique stable completion of the missing-16), and the golden direction sits in the scalar 27 of the E6 branching 70 = 42 ⊕ 27 ⊕ 1 (kernel s120a, 13/13; the three primitive idempotents of the Peirce frame realize χ(OP 2 ) = 3 explicitly). Evaluating the certified Dall’Agata–Inverso N = 8 potential at the certified golden vacuum reproduces the published origin values {SO(8) : −6, SO(6, 2) : 0, SO(4, 4) : +2} and the SO(6, 2) shadow tadpole; the golden Jordan-27 direction is an exactly flat, V = 0 Minkowski valley of SO(6, 2) (ray scan V ≡ 0 to 2 × 10−15 , dV /dt ≡ 0 to 10−11 ), unique to that gauging (kernel s120b, sha 88c010a6). Novelty label: new specialization proved here. Theorem 1.1 (the golden vacuum is a tachyon-free Minkowski critical point at ω = π/4). Let Θ(ω) = cos ω soe (η) + sin ω somA (η) be the dyonic deformation of the SO(6, 2) embedding tensor ( = SO∗ (8)), consistent for all ω. The golden gradient is symmetric in ω and vanishes exactly at ω = π/4: the magnetic completion cancels the electric 54 runaway. At ω = π/4 the exact golden 2 ===== PDF PAGE 212 / 433 ===== vacuum Jvac = diag(ϕ, 1, ϕ−1 ) is a critical point in all three orientations (∥∇V ∥ ∼ 3 × 10−11 , portal tadpole ∼ 10−11 , V = 0 Minkowski) with Hessian signature (0− , 340 , 36+ ) — tachyon-free. (Kernel s120c_golden_omega.py, sha 54b2be21, reproducible; cross-checked against the external probe d806 modulo a grading-sign and a factor-2 η normalization; literature control Warner 1983 / Dall’Agata–Inverso confirms the non-compact origin runaway and the V ≡ 0 no-scale ray. A887.) This is the first vacuum in the suite that is a genuine critical point of a certified dynamical potential rather than an algebraically posited point, and it is the beachhead answer to the “no dynamics” killer. This supersedes the old “saddle” reading. Earlier drafts (Rev16/Rev17) reported Jvac as a saddle of signature (15+ , 12− ). That computation was correct but for the ungauged invariant potential on the J3 (Os ) slice (the 12 negative modes are the split (4, 4) octonion directions, G2 -protected — the Lorentzian-signature seed, not a removable instability; §14, Q1). The dynamical statement is Theorem 1.1: on the gauged E7(7) /SU (8) bed at the dyonic point the split directions are organized by the gauging and the golden point is tachyon-free Minkowski. The slice saddle is now read as the statics of the invariant toy, retained in §8–§14 for its kinematic content (the moment-graded dual collapse, A861). Fence (binding). The transverse 54 runaway of the non-compact SO(6, 2) is generic (present at the origin), so Jvac is a critical point but not a global minimum of the full scalar manifold; full stabilization is treated in §4. The 34 flat directions are tree-level moduli (Appendix-F shadow sector); their split into 16 gauge Goldstones +18 physical moduli, and the lift of the physical moduli, is the subject of §3–§4. Remark 1.2 (the dyonic vacuum validates the dual-collapse intuition; it does not re-open Wyler/α (S175)). The dyonic ω = π/4 vacuum of Theorem 1.1 is exactly the electric/magnetic (collapse ↔ expand) cancellation that the dual-orientation intuition predicts: two dual orientations meeting at π/4 kill the runaway and leave a tachyon-free, absolutely stable Minkowski point, with the Freudenthal round-trip (Jvac# )# = J vac closing the dual loop. This validates the stable-vacuum picture. It does not re-open Wyler or derive α: the stable vacuum fixes the modulus φ (via the AX1 self-duality F (ρ) = ρ − ρ−1 = 1), not the U (1)EM fibre coupling. Wyler’s number stays falsified (1/αW = 137.0360824 vs CODATA 137.035999, gap +6 × 10−7 ; App. N, Leg 1) and α stays fenced. Validation of the mechanism, not a derivation of α. 2 Why golden, derived: the σ-even/σ-odd selection dichotomy Given the dynamical bed, why the vacuum is golden and not symmetric is a selection problem. We resolve it into one structural dichotomy organized by the Jordan involution σ (the time-reversal 1 ↔ 3 of Appendix H), which acts on the radial coordinate as inversion σ : ρ 7→ 1/ρ and fixes the symmetric point ρ = 1. First, golden depth is not manufactured √ by the discrete scaffold. An explicit number-field analysis (kernel s123a, 10/10) shows ϕ ∈ / Q(i, 6) — the field generated by the pixel/Z4 -clock/χ = 3 data 3 ===== PDF PAGE 213 / 433 ===== √ √ √ √ (with q0KK → 6) excludes 5, since x2 −x−1 is irreducible over that field and Q( 5)∩Q(i, 6) = Q. The golden depth Rd = 1/(ϕ2d − 1) therefore reduces to the single irreducible axiom AX1, hosted (not derived) by the scaffold (3-way confirmed, A919–A922). Novelty label: new specialization proved here. Proposition 2.1 (Coleman–Weinberg is σ-even and cannot uniquely select golden in the tested σ-equivariant layers). The one-loop supertrace Str M 4 (ρ) along the golden ray is invariant under σ : ρ 7→ 1/ρ, hence extremized at the σ-fixed point ρ = 1, not at golden. This holds (i) bosonically (kernel s123b, 8/8: ∂ρ Str M 4 |ϕ ≠ 0, single-well argmin ρ ≈ 1); and (ii) for the indexed fermionic supertrace (kernel s124b, 12/12): σ is a verified J3 automorphism (Det, Tr preserved, A → C̄, B → B̄), so any σ-equivariant mass set gives Str M 4 σ-even and stationary at ρ = 1 — a bein-independent no-go (20 arbitrary equivariant masses all σ-even). Coleman–Weinberg is thus structurally unable to uniquely select golden in the tested σ-equivariant layers (bosonic, fermionic-norm, indexed-fermionic): the Aut(J3 )-equivariance premise is structurally forced (the A1 /A2 shift tensors transform equivariantly, so the no-go does not require the unbuilt complex 56-bein), while σ-evenness fixes ρ = 1 as a stationary point but does not by itself exclude σ-breaking field-dependent terms — which is exactly where the still-open Ciab tensor (§4) lives. Scope qualifier (Rev29 ). This proposition is a no-go within the tested σ-equivariant layers, stated against the supertrace-closure bookkeeping as it stood when the kernels cited above were run. That bookkeeping has since been reopened in-house as un-evaluated, and a separate mechanism result bearing on the CW lift has been recorded but not identified against it (status box immediately below). The proposition is therefore not to be read as a closed statement about the CW lift of the residual flats, in either direction. (A925, A934; corroborated 3-way A926–A929, and by the self-contained external run A944 max |Str M 4 (ρ)− Str M 4 (1/ρ)| = 7 × 10−15 .) 4 ===== PDF PAGE 214 / 433 ===== (Rev30 ) Y5/Y6 CW-lift: the three-way state resolves into TWO scoped rows — they were never the same object. Rev29 printed this as a three-way inconsistency whose third item was the open question “whether (1) and (2) name the same object”. Kernel s1035 (S269, 9/9 gates, payload 9111d769...e165) answered it: different objects. The S215 reopening does not contradict the S207 closure — different mass matrices, different theorems, and disjoint parameter dependence. The item is therefore split, not resolved by fiat. Row (i) — free-multiplet tier: closed, and it stands. The SS/Möbius-twisted one-loop supertrace of the free N = 8 multiplet (s508/s509, S207). Rebuilt independently at s1035 from multiplet degrees of freedom and the twist alone, with no reference to any vacuum spectrum: on-shell nB = nF = 128 from (1, 8, 28, 56, 70) with Str M 0 = 0; the finite Hosotani potential V ′′ = 256(2π)2 · 78 ζ(3) reproduced by independent series summation to < 10−3 ; twist typing qfermion = 12 , qboson = 0. Row (ii) — interacting golden-vacuum tier: open, and it stands as open. The Kallosh–Karlsson supertrace of the interacting de Wit–Nicolai spectrum at the golden vacuum (s578–s605, S214–S215), rebuilt at s1035 from vacuum spectra alone, with no reference to the twist. It does not close, and the obstruction is localized rather than diffuse: per-moment rescalings sk have spread ≈1.185 against a 1.05 tolerance, so no single rescaling closes it. The vacuum itself is clean — s605 finds 0 tachyons and 32 zero modes with the 24 Goldstone rank exact, so the point is Minkowski-stationary and the Kallosh–Karlsson theorem genuinely applies. The remaining walls are the vector shape (s578), the κ2 pin, and the scalar-metric convention immediately below. Name the scalar-metric convention wherever moments are printed. The two scalar conventions in circulation differ by an exact factor 4k at moment order k — verified at s1035 to 10−3 relative, giving ratios 4 and 16 at k = 1, 2. A moment quoted without its convention is therefore ambiguous by a factor of four or sixteen, and that ambiguity is exactly what kept row (ii) looking like a physics discrepancy when it is a normalization one. Every supertrace moment in this suite is to be read in the convention declared at its point of use; where no convention is declared, the moment is not yet a claim. Consequently: row (i) may be cited as closed at the free-multiplet tier and only there; row (ii) remains open and pre-registers the Ciab build (§4, §14) together with the convention declaration above. Neither row promotes anything, and Prop. 2.1 is unaffected. The blanket string “Y5/Y6 closed” must still not be printed anywhere in this suite — unqualified, it now conflates two objects that s1035 separated. The complementary σ-odd channel is the unit defect source. Its defect equation is the golden axiom: Novelty label: new specialization proved here. Lemma 2.2 (the σ-odd defect pins golden ⇔ AX1). The σ-odd cotangent source B on the moment- graded slice has defect function F (ρ) = ρ − ρ−1 . At unit weight F (ρ) = 1, which is ρ2 − ρ − 1 = 0, i.e. AX1, with positive root ρ = ϕ; the BPS completion keeps V = 0 and gives a physical radial mode m2 = 7/4 (kernel s123c; A929). Hence the σ-odd defect is the sole viable golden selector, and it selects golden exactly when the framework’s golden axiom holds. 5 ===== PDF PAGE 215 / 433 ===== Proposition 2.1 and Lemma 2.2 together are the selection mechanism: the symmetric (σ-even) loop physics cannot break to golden, and the only object that can — the σ-odd unit defect — breaks to golden precisely through AX1. 3 The BPS / source-completion theorem We now make the defect source quantitative on the J3 arena and state precisely what is derived and what is modular. Novelty label: new specialization proved here. Theorem 3.1 (derived defect direction, completion, and radial mode — modulo the AX1 unit). On the golden ray with bulk barrier − ln Det, the aligned source whose components are bi = 1/xi − xi equals the σ-odd grading K = diag(−1, 0, 1) at golden, and its direction is forced by the bulk barrier gradient together with the BPS condition V = 0. The completed potential is the manifest square 2 V = 12 G(J) − B , G(Jgolden ) = unit σ-odd K, so the completion term + 12 ∥B∥2 is a genuine constant (a Fayet–Iliopoulos / brane-defect self-energy); the bare linear spurion alone gives V = − 12 ∥B∥2 , an anti-de Sitter value. The radial mode is the physical R-ii fluctuation with raw curvature γ 2 + 4, normalized by Kradial = 20/7, giving γ 2 + 4 γ=1 7 m2 = −−−−→ , 20/7 4 of the 18 physical moduli. The unit magnitude γ = 1 is not independently and a rank-8 source lifts 8 p BPS-forced: ρ⋆ (γ) = (γ + γ 2 + 4)/2 selects a different point for every γ, and γ = 1 is the unique value for which ρ⋆ = ϕ, i.e. γ = 1 ⇔ F = 1 ⇔ AX1. (Kernel s124a, 17/17, sha 5f737000; 3-way panel A931 (Grok)/A932 (Gemini)/A933 (ChatGPT), all reproduced in-house. The ∝ ∥B∥2 form named by the text-only reviewers is the asymptotic limit of the exact law m2 = (γ 2 + 4)/(20/7); the +4 floor is the bulk-barrier curvature their leading order omits. The unit-magnitude correction is ChatGPT A933, verified in-house.) The honest end-state of the selection problem is therefore: derived are the σ-odd alignment, the + 12 ∥B∥2 completion (linear-alone → AdS), the physical radial mode m2 = (γ 2 + 4)/(20/7) (= 7/4 at the unit), and the rank-8 → 8-of-18 lift; not derived are the unit magnitude (equivalent to AX1), the specific N = 8 Fayet–Iliopoulos/Gukov–Vafa–Witten/brane identity of B, and the disposition of the residual flats (§4). Golden selection is a derived σ-odd-defect mechanism modulo the AX1 unit; stabilization stays partial — and stays partial for the reason recorded in §4 and in the Rev29 status box of §2: the certified field-dependent tensor Ciab has not been built, and the one in-house mechanism result that would bear on it is itself under an unresolved identification (Rev29 ). 6 ===== PDF PAGE 216 / 433 ===== One operator carries charge and the golden mass ladder (banked, kernel- verified). The hypercharge operator Y = Ldiag(1,1,−2) and the mass operator LJvac commute, [Y, LJvac ] = 0 (kernel s105b; commutator norm < 10−10 , with a live non-commuting control). Hence the structure-group charges {J12 : +1, J13 , J23 : − 21 } and the golden mass ladder √ {ϕ2 /2, 5/2, ϕ/2} are read off the same LJvac that fixes the vacuum. (Fences: Higgs-class P/CP order parameter, not the SM weak doublet; the ladder is dimensionless; a dimensionful lift needs a dimensionful input — the registered class Ldepth /Lσ (Appendix O; Appendix √ X input-count box) — and a tower normalisation, the pure number q0KK = 1/(2 6) of Ap- pendix G; no q0KK 7→ GeV map exists (Rev32.9, A1642 O7).) √ One number printed three ways (Rev32.6; A1567 §2.2, s1158). The middle rung 5/2 of the ladder above, the (1, 3) Peirce eigenvalue of Appendix B,√and the V3 rotation frequency Ω of Paper 5 are one number by the golden polynomial 2ϕ − 1 = 5, i.e. Ω = ϕ − 12 ; the identity is specific to the golden point (λ13 (r) = r − 21 ⇐⇒ r = ϕ; the s1158 controls at r = 3/2 and r = ϕ + 1/10 fail), and no registered map turns the rung into the rotation (Appendix N, Part N.E). 4 Residual moduli and the stabilization status: the one named open tensor Stabilization beyond selection is now reduced to a single, sharply specified object. The 27 splits under σ as 18 σ-even +9 σ-odd; the 18 physical moduli (after removing the 16 gauge Goldstones from the 34 flats, kernel s121a/A891) grade as 9 + 9 (kernel s124d, 15/15; A936). The σ-odd defect of §3 lifts all 9 σ-odd directions (8 portal +1 radial m2 = 7/4). The 9 residual flats are therefore all σ-even, splitting as 8 candidate CW-liftable flats +1 no-scale dilaton. The CW lift of the 8 is not a parity-block readout. At the σ-fixed point ρ = 1 the CW Hessian commutes with σ (∥[H, σ]∥ = 0), but at golden it does not (∥[H, σ]∥ = 633; golden is the σ-broken point), so one cannot read “CW lifts the σ-even residual” off a parity block (kernel s124f, 6/6; A938). CW does contribute curvature (∂ 2 Str M 4 = +276), but the definitive residual masses require the combined defect-B + CW indexed Hessian. Reducing that Hessian exactly (external runs A944/A945, reproduced in-house) gives CW 1 X F 2 Ciab = 21 ∂a ∂b m2i , ui = m2i .   Hab = (−1) n i 2ui ln(ui /µ ) + ui Ciab , 32π 2 i Because the residual coordinates are σ-even, ∂a ui |0 = 0, so the entire lift lives in the field-dependent tensor Ciab . The self-contained constraints (preserving δStr M 2 , δStr M 4 ) leave a rank-2 / 6- dimensional sign-free nullspace in which 8+ , 80 , 8− and mixed signatures are all feasible: Novelty label: new specialization proved here. Proposition 4.1 (the 8-flat lift is undecidable from σ-evenness and the point spectrum). σ-evenness together with the point spectrum and the low-order supertrace constraints do not determine the sign of the 8 × 8 residual Hessian; the lift is decided only by the certified field-dependent tensor Ciab in the E7(7) /SU (8) frame. Equivalently the on-disk 8-bein spans only so(8) (28/63 generators); the missing 35 are the complex part of the 56-bein (kernel s124i, 4/4; A941). This is the single deferred stabilization target: compute Ciab = 12 ∂a ∂b m2i on the 8 σ-even residual CW , and read the sign — does CW lift moduli in the certified E7(7) /SU (8) frame, contract with Hab 7 ===== PDF PAGE 217 / 433 ===== the 8 (full stabilization modulo the dilaton) or not? The build needs the complex 56-bein and is pre-registered (§14). The no-scale dilaton is a separate scale/lift input. The remaining σ-even direction is a no-scale dilaton, irreducible at the supergravity layer (N = 8 forces Str M 2k = 0). It is provably not CW-fixed: the one-loop minimum L⋆ = A/2B + 14 shifts with the renormalization scale µ, so its disposition is a standard scale/lift fence, distinct from the Ciab question above. Do not fold the dilaton into the “8 flats” count. Honest stabilization summary. Selected — yes (derived σ-odd defect mechanism modulo the AX1 unit; CW unable to uniquely select golden in the tested σ-equivariant layers). Fully stabilized — not yet: 9 σ-odd directions lifted; of the 9 σ-even residual, 8 have their lift sign pending the certified Ciab (Prop. 4.1) and 1 is a no-scale dilaton (separate input). Why it is still partial (Rev29 ): the deciding object is a missing build, not a missing argument. Ciab = 12 ∂a ∂b m2i in the certified E7(7) /SU (8) frame needs the complex 56-bein, and the on-disk 8-bein spans only so(8) — 28 of 63 generators, the missing 35 being exactly that complex part (Prop. 4.1). An in-house mechanism result bearing on the CW lift (the free-multiplet supertrace, row (i)) is banked and closed; the interacting golden-vacuum supertrace (row (ii)) is open. They were once printed as a three-way state; s1035 split them (Rev30 box, §2) and the PI ruled the split (Rev32.10, R141): different objects. Neither row is folded into the status line above. The operative status therefore remains stabilization partial. This is the strongest honest form of the answer to the no-dynamics killer. 5 The generation count as a Morse index The generation count is a clean topological index on the same Peirce frame. Novelty label: physical interpretation not established here. Proposition 5.1 (Ng = 3 as a Morse index). On the Cayley plane OP 2 realized by the three primitive idempotents of the golden frame, the cubic-norm Morse function has critical cells of index {0, 8, 16}, so X χ(OP 2 ) = (−1)idx = 1 − 0 + 1 − 0 + 1 = 3 = Ng , cells ordering-invariant across all 6 orderings (kernel s124c, 5/5; A935). This is a topological index, not the refuted “rank 3 ⇒ 3” route (A691–A694). The generation↔cell assignment, by contrast, is the non-finite A1 datum: the three idempotents form a single S3 automorphism orbit with identical intrinsic data, so all algebra-only labels are S3 -symmetric (0 bits) — the assignment needs the golden mass-ordering (dynamics), and is folded here as honest-open (kernel s124g, 4/4; A939). The count is algebra-derivable; the map is not. 6 Confinement as the spectral geometry of the cubic norm The same cubic norm N (J) = DetJ whose Morse cells count the generations (§5) also organizes the confinement sector. Three structural facts follow from N alone — its homogeneity, its polarization, 8 ===== PDF PAGE 218 / 433 ===== and the degeneracy of its Koecher–Vinberg Hessian GM N = −∂M ∂N ln N (the cone metric of Eq. (1) below) — independently of any particular numerical parametrization of that spectrum. Novelty label: new specialization proved here. Proposition 6.1 (a formal scaling identity for the cubic weight). For the scaling action S = c N (J) on J3 (Os ) (dimR = 27, deg N = 3) and real c > 0 — the sign branch is fixed here, before the identity: for c < 0 the real substitution J → |c|1/3 J gives |c|−9 instead, and c−9 below means the positive branch (Rev32.10, A1711-F04) — the substitution J → c1/3 J gives, formally, Z(c) = ec N dJ = c−9 Z(1), R hence d dimR J3 (Os ) c log Z(c) = − = −9, dc deg N constant in c. The integral does not converge on the real 27-dimensional space for either sign of c (N is odd), so the identity holds only for a regulator or integration domain that respects the scaling (a cone, or a contour); none is fixed in this appendix. The quantity c ∂c log Z is a scaling exponent, not the renormalization-group β-function, and no statement about the running of physical masses follows from it. The metric G = −∂∂ ln N is undefined on {N = 0}; the nine directions tangent to the null cone (Euler’s relation J ·∇N = 3N = 0 there) describe the limit of its degeneracy, not flat directions of G on the cone. (Retyped Rev32.9, A1641 F01: through Rev32.8 this was titled “the overall coupling does not run: βc = −9” and concluded that running must come from composite correlators. [LIB2-058] Algebraic identity; L-BetaCTrivial.) Novelty label: new specialization proved here. Proposition 6.2 (Faddeev evasion: the cubic invariant survives where the quadratic vanishes). At Jvac the one-loop correction to βc from any bilinear composite vanishes identically, g T H + g = 0: this is the mesonic, evaluation-type channel M † M built on the quadratic trace form. The baryonic channel is built instead on the cubic polarization of N — the εabc εijk contraction (colour × generation), i.e. the Λ3 → singlet / determinant trilinear. [LIB2-136] That contraction is antisymmetric within each index triple and therefore symmetric under a paired exchange of one colour and one generation index; the polarization of the cubic form N is a symmetric trilinear. A trilinear is not a quadratic form, so the g T H + g = 0 cancellation does not act on it. [LIB2-136] Whether its residue is nonzero, and its sign, is not computed here. Confinement is proposed to use the invariant of N that the quadratic degeneracy leaves untouched. (Rev32.9, A1671 F05: through Rev32.8 this read “totally antisymmetric” and “its residue is strictly positive”; neither was argued. Tensor-structure observation; L-FaddeevEvasion.) Novelty label: physical interpretation not established here. Proposition 6.3 (confinement as a spectral singularity, not strong coupling). Lifting the null-cone eigenspace by a mass ε gives a spectral trace T (ε) = A/ε, with A the trace of the cone Hessian over the lifted sector. [LIB2-135] At the self-consistent point ε = m2 = αA one has α T = α (A/αA) = 1 exactly, independent of α, fixing a confinement scale Λ2 = αA with the coupling weak throughout. [LIB2-135] The mechanism is a degeneracy-driven pole — closer to spectral localization than to asymptotic freedom — not a strong-coupling limit. It is proposed as the mechanism behind the G2 confinement scale ΛG2 that anchors the light-quark constituent masses (Paper 2, §3.4); ΛG2 itself is a calibration (App. X). Scope (Rev32.9, A1671 F06): the lift and T (ε) = A/ε come from the legacy rank-2 vacuum diag(ϕ2 , 1, 0), where N = 0 and G = −∂∂ ln N is undefined; that point is not related to the current vacuum diag(ϕ, 1, ϕ−1 ) (N = 1) by a normalization, and the transport of the null-eigenspace lift to the golden point is not supplied. (Mechanism sketch; L-HybridMassGap.) 9 ===== PDF PAGE 219 / 433 ===== Remark 6.4 (three companion facts about the same Hessian). The cone Hessian G carries three further structural facts from the same analysis, each a property of N rather than of a fitted spectrum. (i) The vacuum is a spectral background, not a saddle: Jvac is selected by the eigenvalue structure of G, not by ∇N = 0 (indeed ∇N = ̸ 0 there), which is why the expansion is organized by the Hessian spectrum rather than by a stationary phase. [LIB2-137] (ii) Colour isotropy: within each Peirce sector the eight SU (3)-colour components of an off-diagonal octonionic entry carry G-eigenvalues of equal magnitude and opposite signs in the raw block (±2ξk , fourfold each — Remark 6.5); in the reflected positive metric they carry one common eigenvalue, so it is the reflected metric that is colour-blind at tree level (Rev32.9, A1647 F01) (cf. the Casimir band law of App. X). [LIB2-086] (iii) Relevance: the cubic coupling has mass dimension [c] = (6 − d)/2 for a canonically normalised two-derivative kinetic term (a premise, stated Rev32.9), relevant in d = 4 exactly as in ϕ3 theory — the same homogeneity that fixes the scaling exponent in Prop. 6.1. [LIB2- 059] (Library L-SpectralBackground, L-ColourBlindExact, L-CubicRelevant; the legacy vacuum normalization diag(ϕ2 , 1, 0) is superseded by the current {ϕ, 1, ϕ−1 } frame, the structural statements unaffected.) √ Remark 6.5 (the flux-tube ratio σ = ΛG2 φ is an in-frame, α-free eigenvalue ratio (s284, S175)). On the current N = 1 cone the off-diagonal octonionic blocks of G at Jvac carry eigenvalues ±2ξk √ with ξ = (φ, 1, φ−1 ), so the golden flux-tube relation σ = ΛG2 φ is exactly the eigenvalue ratio 2ξ1 /2ξ2 = φ — a clean Hessian-eigenvalue ratio, recovered frame-current and independent of α (an upgrade from the thinly-banked legacy fence of Paper 8). It nonetheless merely restates the single scale ΛG2 (it reproduces no Λ-independent datum), so it is not an independent hadronic observable and nothing is promoted. The cone yields only this one φ-power family of α-free ratios; a second, genuinely independent in-frame hadronic observable does not follow (honest negative, s284). Remark 6.6 (why the confinement scale is blocked behind α (s284, S175)). The confinement scale cannot be derived without first un-fencing α, for two independent reasons. (i) By Prop. 6.1 the overall coupling βc = −9 is constant (pure homogeneity, − dim / deg = −27/3). Standard 2 dimensional transmutation requires a log-running coupling, Λ = µ e−1/(bg ) with b ̸= 0; a constant β generates no exponential scale, so G2 -running produces no ΛG2 of its own. The scale is instead fixed by the spectral-singularity self-consistency Λ2 = αA (Prop. 6.3) — i.e. by α (fenced, App. N) times a pure cone trace — so ΛG2 sits behind the α fence, not upstream of it. (ii) The cone supplies only the one φ-power family of α-free ratios (Remark 6.5), so no second, genuinely independent in-frame hadronic observable follows; the minimal two-observable falsifiable hadronic core is not reached from cone geometry alone. Both are honest negatives: nothing here promotes ΛG2 to a prediction, and α stays fenced. 10 ===== PDF PAGE 220 / 433 ===== Hadronic scope fence (binding). The framework models the confinement scale ΛG2 (Prop. 6.3; one ΛQCD -like calibration input, App. X) and the confinement mechanism (Props. 6.1–6.3, structural). [LIB2-135, LIB2-136] It does not attempt hadron spectroscopy: no glueball or full meson/baryon spectrum, no lattice comparison. The numerical values once attached to this sector — the spectral trace A = 144, the ambient Morse index 17/27, and the baryon/meson ratio MB /MM ≈ 6.7 — were computed in the legacy ambient-27 Hessian parametrization and predate the current {ϕ, 1, ϕ−1 }/OP 2 description. [LIB2-335, LIB2-354] An in-frame recompute on the current N = 1 cone (kernel s284, S175) is decisive: the ratio MB /MM ≈ 6.7 does not survive — the cone supplies only ϕ-power eigenvalue ratios and the PDG 938/140 = 6.72 is not a clean ϕ-power (nearest ϕ4 = 6.854, +2%) — so it is retired as a legacy null-cone artifact, not a frame-current relation. (The companion flux-tube relation √ σ = ΛG2 φ does survive in-frame and α-free, but only restates the one scale; Remark 6.5.) Both remain recorded, fenced, in Paper 8 (Junkyard, Fenced: out-of-scope work) and lie outside the falsifiable core; no public observable depends on them. 7 The probe Lagrangian and its scalar constraint The preceding dynamical results sit on the certified gauged bed. We retain the earlier probe model, whose kinematic content (the bounce, the clock, the moment-graded collapse) remains valid and feeds the synthesis. We work with a probe field J(t) ∈ J3 (Os ) with cubic norm N (J) = DetJ and Freudenthal quadratics Tr, Tr# , governed by a Wheeler–DeWitt-type scalar constraint H = 12 GM N (J) ΠM ΠN + λ (N − 1)2 + γ Tr(K ·J) ≈ 0, (1) with kinetic metric GM N = −∂M ∂N ln N (J) (the Koecher–Vinberg cone metric), K = diag(−1, 0, 1) the banked Peirce grading (the same K whose half-turn is the time-reversal T of Appendix H; the sign is fixed by stationarity of Jvac , so the largest eigenvalue ϕ sits at grading −1), and λ, γ stiffness 2 couplings. Equation (1) is the radial/diagonal restriction of the coset M4 Tr(Pµ P µ ) on E7(7) /SU (8), the (N − 1)2 term confining to Det = 1 and Tr(K ·J) the symmetry-breaking grading coupling discussed in §8. Why the grading term and not an invariant potential. A potential built purely from the characteristic-polynomial invariants (Tr, Tr# , N ) cannot select Jvac : those invariants are symmetric in the eigenvalues, so their minimum on the N = 1 cone is the symmetric point (1, 1, 1), never the golden multiset (a provable no-go on the diagonal slice — the diagonal-slice shadow of the σ-even no-go, Prop. 2.1). The grading K is not an invariant; it is the preferred (σ-odd) direction that breaks the eigenvalue-permutation symmetry, exactly the defect of §3. Two constraints on the invariants, by contrast, do pin the multiset; that is the independent Route A of §8. 8 The overdetermined ϕ-vacuum (invariant routes) Independently of the dynamics, the golden multiset is overdetermined by conditions the framework already carries. We give two routes to Jvac = diag(ϕ, 1, ϕ−1 ). 11 ===== PDF PAGE 221 / 433 ===== Route A — the invariant selector {N = 1, Tr = Tr# , det G = 7} Restrict to the inversion family J(r) = diag(r, 1, 1/r), so N = 1 automatically and Tr = Tr# (palindromic spectrum). Let G(J) be the Freudenthal/Gram form whose determinant on this family is det G = s2 − 9 with s = r2 + 1 + r−2 . Novelty label: new specialization proved here. Lemma 8.1 (det G = 7 ⇔ r = ϕ). On the inversion family diag(r, 1, 1/r) with r > 0, √ det G = 7 ⇐⇒ s = 4 ⇐⇒ r + 1r = 5 ⇐⇒ r = ϕ. Proof. det G = s2 − 9 = 7 gives s2 = 16, hence s √ = 4 (positivity). Then r2 + r−2√ = s − 1 = 3, −1 2 2 −2 −1 2 so (r√+ r ) = r + 2 + r = 5, i.e. r + r = 5. The positive root of r − 5 r + 1 = 0 is r = 5+1 2 = ϕ. Direct check: for r = ϕ, ϕ2 = ϕ + 1 and ϕ−2 = 2 − ϕ, so s = (ϕ + 1) + 1 + (2 − ϕ) = 4 and det G(Jvac ) = 16 − 9 = 7 exactly. The content of Lemma √ 8.1 is that the banked integer det√G = 7 (the same “7” in sin2 θ23 = 7/16 and |Vcb | = 1/(9 7); Appendix B) algebraically encodes 5, hence ϕ, with no ϕ placed by hand. The selector is built from E6 -invariants, so it lives on the gauge-invariant orbit; the diagonal Jvac is its gauge representative. Route B — the grading-coupled potential Take the diagonal-slice potential extracted from (1), V (J) = − ln DetJ + 12 TrJ 2 + γ Tr(K ·J), K = diag(−1, 0, 1). (2) Novelty label: new specialization proved here. Proposition 8.2 (unit coupling selects the golden vacuum). On J = diag(a, b, c), stationarity of (2) forces b = 1 and a common golden equation on the outer eigenvalues; at unit coupling γ = 1 the unique positive-definite minimum is Jvac = diag(ϕ, 1, ϕ−1 ), with DetJvac = 1 automatically and ϕ the positive root of c2 − c − 1 = 0. The Hessian restricted to the diagonal slice is positive-definite; on the full algebra the invariant-potential Hessian is the saddle of §14, Q1, resolved on the gauged bed by Theorem 1.1. The two routes are independent: Route A uses the cubic/Freudenthal invariants and imposes inversion; Route B uses the grading direction K and unit coupling. Neither inserts ϕ; both reduce “ϕ” to a single named structural condition (det G = 7, or γ = 1 ⇔ AX1, §3). This is the precise, fenced sense in which the golden vacuum is overdetermined. 12 ===== PDF PAGE 222 / 433 ===== The 7/3 amplitude bridge — characterized as open (S116). The integer det G = 7 is the dimension count dim Im Os = 7, and 7/3 = dim Im Os /rank J3 is a proved algebraic invariant. Its promotion to a heavy-quark S-matrix residue was scouted to a wall: no literature theorem maps a Jordan Gram determinant to an LSZ residue/OPE/crossing ratio; the special-geometry Yukawa Yijk = ∂i ∂j ∂k N contracted at Jvac either yields a traceless ladder with mean 8/3 (not 7/3) or, after canonical normalization, the degenerate spectrum 1 : 1 : 1; and an explicit canonical Peirce-trilinear LSZ computation returns mass-residue ratio 1 and inclusive colour ratio 3. The decisive obstruction is Lemma 8.3: “7” is a dimension, whereas a physical residue channel-sum is weighted by the split-signature metric, under which Im Os has signature (3, 4) and signed trace −1. The metric-weighted-residue route is therefore closed-negative (Lemma 8.3, sharpened at Rev27; kernel s902): no unitary residue over the Im Os channels can return 7. What stays open is only whether a non-residue, index-theoretic exceptional-field-theory realization exists; the 7 survives as the finite-carrier index (Appendix K), not as a signed amplitude. [LIB2-311] Novelty label: new specialization proved here. Lemma 8.3 (the 7 is a dimension, not a metric trace). On the split octonions Os with trace form of signature (4, 4), the imaginary subspace Im Os (the metric-orthogonal complement of the identity) has dim = 7 and signature (3, 4), hence signed trace −1. Consequently dim Im Os = 7 and Tr g|Im Os = −1 are distinct objects; under the equal-unit-weight channel prescription, an S- matrix residue weighted by g returns the√latter (an arbitrary coupling-weighted contraction ηi |ai |2 P does not — it reaches 7 at amplitude 7, s1171 T5; Rev32.9, A1652). The channel coefficients τ (e0 , ea , ea ) = (1, 1, 1, −1, −1, −1, −1) realize this explicitly: positive Hilbert count 7, signed split trace −1. Sharpening (Rev27, kernel s902; run1==run2, hash 547d9265). The signed residue over any sub- collection of g-orthonormal channels equals (#positive − #negative) and therefore lies in the closed range [−4, +3] (the negative and positive indices of the (3, 4) form); the full-space value is −1. [LIB2- 107] Since 7 ∈/ [−4, +3], i.e. |residue| ≤ 4 < 7, the integer 7 is unreachable as a signed (unitary) residue and is realized only as the unweighted multiplicity. [LIB2-107] Hence the metric-weighted LSZ promotion of 7/3 = dim Im Os /rank J3 is a structural no-go; the 7-as-index reading (the finite- carrier theorem, Appendix K) is the unique survivor. This closes the specific residue route on Im Os ; it does not assert that no other construction could realize 7/3 physically. [LIB2-107, LIB2-311] (Verified in-house on the suite’s omul/oconj and an independent Cayley–Dickson implementation; S116, A825–A830.) 13 ===== PDF PAGE 223 / 433 ===== OPE / colour-Casimir addenda (S116, observation register — not derivations). Three further recastings were tested and all return 7/3 only under an imposed selection, leaving the mass-boundary readout open: (i) OPE coefficient. Recast as a Wick/OPE coefficient C00 = i,j τ (e0 , ei , ej )2 over Im Os , one finds C00 = 7 and C00 /rank J3 = 7/3; P P but CAB = τ τ is a Gram matrix, positive-semidefinite by construction, so the 7 is again dim Im Os recovered by a sum of squares. (ii) Colour face. Numerically 7/3 = 1 + CF (SU (3)fund ) since a T a T a = CF I = 43 I, and 1 + CF (SU (N )) = 7/3 holds (algebraically, P up to permutation) only at N = 3 (3N 2 − 8N − 3 = 0, positive root N = 3). This gives the invariant a colour-current face, but the physical colour effect on mb /mτ is renormalization- group logarithmic (γm ∝ CF , exponentiated), not an additive tree factor 1 + CF . (iii) Same-line fence. The 7/3 readout survives only the same-line scalar-mass+Casimir insertion; the literal cross-line colour-adjoint exchange gives − 13 , the sequential/additive variants give 49/9 and 11/3, and the canonical U (3) current gives 3/2. Since the suite’s physical Yukawa is the cross-block Peirce trilinear, the 7/3-surviving same-line readout is structurally mismatched with the framework’s own Yukawa placement. Net: the colour face is suggestive and uniquely picks N = 3, but does not promote 7/3 to a derived mass coefficient. (S116, A831–A833.) The closure projector has a canonical form; the generation assignment reduces to two physical guards (S116–S117, A841–A843). The selecting operator can be written O = I + Π12 C2 (SU (3)) with CF = 43 , where the closure-current projector acquires an exact algebraic form Π12 = 4 Lc1 Lc2 — the unique off-diagonal Peirce block invisible to the c3 idempotent. In-house on s105b: rank Π12 = 8, idempotency ∥Π212 − Π12 ∥ = 0 (exact), c3 -invisibility, and orthogonality Π12 ⊥ Π13 , Π23 . Then O returns 7/3 on closure- current quarks and 1 on direct-carrier and colour-singlet anchors. What this does not do is derive which fermions occupy the c3 -invisible block — the fermion→Peirce-block generation assignment. A structural defect rule (A843) selects exactly b, c but leaves two load-bearing physical guards that are inserted, not derived (the colour-current guard and the non-confinement-anchor guard). Two no-go results (A841) show the residual assignment must be generation-topological: the full one-generation gauge tuple is identical for top and charm, so no gauge/Casimir-only function gives 7/3 on b, c but 1 on t; the cross-block Peirce signature is common to top, bottom and tau. This converges with the beat-order finding (A834) nb − nτ = logφ (7/3). Hence the heavy-quark 7/3 relation, the projector Π12 , and the quark beat-order selector are the same open object — a piece of the generation-count frontier (the A1 datum of §5). The readout side: a finite-operator form for the 7/3 edge (S117, A844–A845). Three candidate readouts are positively excluded, leaving a single finite-operator survivor (all reproduced in-house). (i) A universal field-level same-line self-energy weight is impossible (a brute scan over {0, 1}3 returns zero solutions). (ii) Ordinary kinetic Z-normalization gives the inverse: ZL = ZR = 7/3 ⇒ mphys = 37 m0 . (iii) The ordinary commutator/gauge- orbit Casimir vanishes. The survivor is the normalized double-anticommutator Casimir C+ (D) = 14 a {Ga , {Ga , D}} = CF D = 43 D, whence D + Pdefect C+ (D) = 73 D on the closure P edges and D on anchors (a post-EWSB audit forces the electromagnetic coefficient to zero). This upgrades the 7/3 edge from a hand-written scalar to a canonical finite self-adjoint operator, but does not prove the physical mass boundaries read C+ rather than the cross- block Yukawa singular value. P 3 remains settled-open; no observable moves. 14 ===== PDF PAGE 224 / 433 ===== The tested finite endpoint/self-adjoint/source layer is exhausted: the 7/3 edge is a cotangent source response, and the residual is non-finite (S117–S118, A846– A863; 27 runs, none breached). A sustained probe pushed the readout problem to its terminus on the finite layer it tests. (1) Operator form. The survivor C+ is the Hilbert gradient of the positive functional S+ [D] = 18 a Tr({Ga , D}† {Ga , D}) (A846) and equals P the double Jordan product a Ga ◦(Ga ◦D) = CF D (A847), matching the Peirce closure P law exactly. (2) A sharp negative bounding the class. The 7/3 is not a positive-parent- action minimizer: minimizing Srel gives cmin = 3/(4α + 3), so α = 1 returns the inverse 3/7; the factor survives only as an additive gradient insertion Deff = D + ∇Srel [D] = 73 D (A856–A857); integrate-out and kinetic-Z routes likewise return 3/7 (A859/A862). (3) The physical object. Restated symplectically, the 7/3 edge is the response of a cotangent Yukawa source B ∈ J12∨ with B = ∇S 7 edge = 3 D and rank span{By,z } = 8/8 (A862–A863) — the same defect object §3 derives the unit alignment of. A decisive no-go (A851) confirms the residual is generation-level (top and charm collide). Conclusion. The tested finite endpoint/self-adjoint/source layer is exhausted — form, readout, uniqueness, and physical object pinned, minimizer/integrate-out classes excluded. The remaining content is explicitly non-finite: the full E7(7) /J3 (Os ) fermion action with the source term Sedge [D] − ⟨B, D⟩ plus the FDR-1 generation assignment (A1). Across 27 runs (A843–A863) none breached and no public observable moved; the 7/3 label stays Structural/conditional, never a derived mass prediction. 9 The geometric bounce The kinetic metric GM N = −∂M ∂N ln N is the canonical Koecher–Vinberg metric on the symmetric cone h{N > 0}. Along the radial ray J(ρ) = eρ Jvac the cone-distance element behaves as dℓ ∼ √ i 2 d − 12 ln(1 − s) toward the boundary, so the metric length from the Det = 1 vacuum to the Det = 0 locus diverges: Z √ ℓ(Jvac → {Det = 0}) = − 2 d ln(1 − s) −→ ∞. (3) Novelty label: new specialization proved here. Proposition 9.1 (classical bounce). The Det = 0 singular boundary lies at infinite KV cone-distance from the Det = 1 vacuum. A free (geodesic) trajectory therefore never reaches the singular locus: the cone is metrically complete toward Det = 0. Singularity resolution is classical and kinematic, located at the boundary, not at the vacuum (the Det = 1 interior point). Equivalently, the regulated null wall W (N ) = N −2/3 + 2 N 1/3 − 3 (W (1) = 0, W ′ (1) = 0, W → ∞ as N → 0) has its minimum at the vacuum surface N = 1 and diverges at N = 0: VN (r) = Λ4 (e−2r + 2er − 3) has minimum at r = 0, VN′′ = 6Λ4 > 0, so any incoming radial trajectory bounces. This relocates the retired “Det = 0 as the vacuum” reading: the vacuum is the Det = 1 interior, Det = 0 is the wall it bounces off. Fence. Proposition 9.1 is classical singularity-resolution. The exponents and scale Λ in W (N ) are not derived. No quantum area operator, holonomy spectrum, or Hamiltonian in physical seconds follows (the LQG-quantization reading is dead-for-now: SO(4, 4) is non-compact, no compact area operator follows). 15 ===== PDF PAGE 225 / 433 ===== 10 The E7(7) clock connection and the quantum beat The time layer of Appendix H realised the past↔future moment as a genuine E7(7) element T with T 2 = −1. The probe supplies the connection 1-form: an E7(7) -valued clock connection Aclk µ = Qµ + aµ K, with Q the E6 × U (1) part and aµ the clock 1-form along K. Its holonomy is  I  I a = π2 =⇒ Uγ = exp π2 K = T, T 2 = −1.  Uγ = exp K a , (4) H This is the “quantum beat” of the title. The half-turn a = π/2 is one beat of the relational (Leibnizian) clock; T 2 = −1 is the Kramers/spinor double cover, so it takes two beats to return (T 4 = +1). Physically T is the wormhole↔black-hole time-orientation flip at the Det = 0 wall of §9: the bounce is where the clock turns over. Fence. T ∈ E7(7) , T 2 = −1 is certified (Appendix H, seed-independent). The clock connection Aclk = Q + aK is a Structural construction; the parameter β in β a = π/2 H is a clock-period convention (βK ), not the old βd = 11/(6π) derivation, which is dead as a derivation (K ∈ / g2 ). The reading of the turnover as a four-tick “past→past, future→future” return — T accruing a −1 phase on reflection at the Det = 0 wall (T 2 = −1, T 4 = +1, a Kramers/spinor double cover) — is validated as sound classical holonomy (the D905 review, S175): a genuine structural clarification of how the clock turns over at the bounce, but, being classical geodesic scattering rather than a Coleman event, it generates no dimensionless coupling (it is the fifth leg of the α fence, Appendix N). 11 The moment–annulus synthesis and the dual collapse Speculative / interpretive layer (S118) — fenced; no public observable moves. Moment ≡ annulus ≡ Z2 -twisted cylinder (A858/A860). The dial moment multiset on the 56, {−3 : 1, −1 : 27, +1 : 27, +3 : 1}, is exactly the annulus shell content 1⊕27⊕27′ ⊕1, and 60 = 30 + 30 splits as the two 27/27′ shells. The change of basis between the linear (charge) and geometric (mass) ladders is the exact Fibonacci identity φn = Fn φ + Fn−1 . The half-turn satisfies T 2 = −I on the full 56 (verified to 1.3 × 10−15 ; spin- 12 /Kramers). The synthesis is kinematic and substrate-grounded; the rest-mass depth selector stays dynamical — no parameter-free container rule over {16, 27, 60, 3, 30} reproduces the lepton depths {0, 11, 17} (A860; the isolated coincidence 27 − 16 = 11 = nµ has no τ parallel), so the selector is the A1/dynamics wall again, not packing combinatorics. (Fenced: Möbius–LQG quantization walled at D804; the dial circle is the internal moment circle, not a spatial KK circle.) The Koide half-depth link is a kinematic observation, explicitly fenced (S118, √ sharpened Rev18 per A865b). The Möbius double cover (×2) reproduces the m half-depth structure of the Koide relation as a change-of-basis observation on the grading circle, not a derived fermion propagator. The geometric origin of the factor 2 is the T 2 = −I Kramers cover; we do not claim it is bound to a physical Koide self-energy. Absent an explicit propagator derivation this link is held at Speculative/interpretive tier and contributes no scorecard datum — the standing fence against reading the ×2 as numerology. 16 ===== PDF PAGE 226 / 433 ===== The golden saddle collapses moment-graded, dual about the present (A861). The invariant-potential J3 (Os ) Hessian (the statics of §7, not the gauged vacuum of §1) is stationary, Det = 1, signature (15+ , 12− ). The 12 tachyonic directions are the three off-diagonal octonion blocks ×4, moment-graded under K = diag(−1, 0, 1): the past block a = (1, 2) has curvature −2φ, the present block b = (1, 3) has −4, the future block c = (2, 3) has −2φ2 . The collapse is graded past/present/future and dual about the grade-0 present, with the future direction steeper by exactly φ (a fenced arrow-of-time seed). This is statics only. The bang/crunch curvature difference is not the cosmological constant: the future−past difference is the dimensionless O(1) number −2 (ratio φ), whereas Λ ∼ 1.4 × 10−123 in Planck units — off by ∼ 123 orders, a dimensionless Hessian versus a dimensionful energy density. The asymmetry is a real derived structural result; its identification with Λ is an honest negative. 12 The weak-angle crossing (B4): a GeV-order, scheme-dominated note A separate threshold probe (B4) tested whether the running weak mixing angle crosses ϕ−3 = 0.2360680 at a fixed hadronic scale. The naive one-loop route fails (A870). With proper multi- threshold running the light quarks carry the rise, so the crossing migrates to the hadronic/IR scale: µcross is pushed sub-GeV (∼ 0.15 GeV, band [0.09, 0.55]), and the once-quoted 1.8 GeV lies outside the band (kernel s124h, 4/4; A940; an intermediate single-threshold estimate gave ∼ 1.04 GeV, A937). The honest statement is therefore: a ϕ−3 crossing exists, at sub-GeV/GeV order, scheme- and threshold-dominated — not a pinned µ⋆ = 1.8 GeV prediction. The sub-GeV region is non-perturbative (a hadronic R-ratio input is required); no observable moves. 13 The knot-neck geometry: figure-8 lepton neck and trefoil baryon neck (S233–S235) The figure-8 lepton neck The wormhole–black-hole neck is a Möbius Z2 -twisted bundle: σC reverses the fibre and the first Stiefel–Whitney class w1 ̸= 0. Its base of events is the figure-8 knot complement — hyperbolic, with volume 2.0298832 . . . — so Mostow rigidity fixes zero geometric moduli; the holonomy is the spin cover SL(2, C) → SO(3, 1) of PSL(2, C), with dilatation φ2 (hence ln = 2 ln φ, the banked τ ). The throat glues smoothly by an affine map; the seam ring is an intrinsic curvature singularity (R = ε/2ξ 2 , profile-independent) admitting no finite Israel thin shell. The one-loop gravitational determinant is analytic (Reidemeister) torsion: for the figure-8 monodromy det(I − tA) is the Alexander polynomial t2 − 3t + 1, and the knot determinant |∆(−1)| = 5 = pentagon = Z/5. The Euclidean no-boundary cap has action equal to the hyperbolic volume (dominant saddle); Perron–Frobenius φ2 dominance sets the arrow. Honest residues: the single external scale (E4 by design — a conformal, Mostow-rigid geometry cannot manufacture a dimensionful constant) and the exact quantum-gravity one-loop measure. 17 ===== PDF PAGE 227 / 433 ===== The trefoil baryon neck The baryon neck is the trefoil analogue: a Faddeev pole RB = φ, knot determinant 3 = triangle, and — unlike the figure-8 — two shape moduli (not Mostow-rigid). Static shape selection is a curvature- robust honest negative: in the curved KV geometry (K = − 41 ) the equilateral configuration is a curved-length maximum, with reduced shape Hessian in the circulant form g −2 1  1 −2 (eigenvalue ratio exactly 3, both modes tachyonic); curvature softens the instability by ≈ 8% but cannot flip its sign. Static-length functionals are therefore exhausted, and the remaining wall is the quantum source/propagator problem. Structurally, the spectrum closes in golden identities: E0 = φ2 , E1 = φ3 (from 2φ2 − 1 = φ3 ), R1 = φ, and µ3 R2 (µ3 ) = 2 − 2 , ρ2 = 12 , (5) 2φ with ρ2 derived given the identification contact-space ≡ two-cylinder twist space. This yields a twist-sector map (structural candidate, conditional): τ = +1 ⇒ µ3 = 2φ−2 ⇒ R2 = 2 − φ−4 = 1.854 (N (1710)-like); τ = −1 ⇒ µ3 = 0 ⇒ R2 = 2 (N (1880)-like/null). A house coverage theorem underwrites it: the cross-cubic trilinear has 64 nonzero unit channels lying on the 7 Fano quaternion lines, each single associative slice covering exactly 25%, the full vertex requiring the union = Os . The golden magnitude is pinned to a single exact target, κwall = κgolden = 9φ−2 /4 (derived given Eth ), with a measure-parity reading (democratic wall measure → null branch; quadratic/spectral → √ golden). Everything here is a spectroscopy discriminator, explicitly not a fit; σ/tube-tension stays fenced and no MeV appears. 14 Open leads and kill gates (pre-registered) The dynamics layer is deliberately incomplete. The decisive next computations are pre-registered; none is promoted here. Q1. Invariant-potential full-J3 (Os ) Hessian — statics, superseded dynamically. On the ungauged invariant potential the full 27 × 27 Hessian at Jvac has signature (15+ , 12− , 00 ): the 12 negative modes are the split (4, 4) octonion directions, per-block curvature ±2(xi + 1) = ±(2ϕ, 4, 2ϕ2 ). This is G2 -protected — the unique G2 -invariant block quadratic is the indefinite split norm, so no G2 /F4 -covariant scalar term lifts it; only a G2 -breaking compact uplift reaches positive-definiteness, at ηc = 2ϕ2 (kernels s112a, s113a, reproducible). Dynamically this is superseded: on the gauged E7(7) /SU (8) bed at ω = π/4 the golden point is tachyon-free Minkowski (Theorem 1.1); the slice saddle is the statics of the invariant toy, not the dynamical verdict. Q2. The certified Ciab build (the one deferred stabilization target). Compute Ciab = 1 2 2 ∂a ∂b mi on the 8 σ-even residual moduli in the certified E7(7) /SU (8) frame (needs the complex 56-bein; the on-disk bein is so(8)-only, 28/63, missing 35), contract with Hab CW , and read the 8 × 8 sign: does CW lift the 8 (full stabilization modulo the dilaton) or not? Undecidable without it (Prop. 4.1). Q3. Is γ = 1 forced beyond AX1? The unit magnitude is equivalent to AX1 ρ2 − ρ − 1 = 0 (§3); whether AX1 itself is forced by the octonion/Peirce self-similarity (the σ fixed point s = 1 + 1/s) is the standing deeper question. If only AX1 is needed, ϕ is selected modulo the framework’s single golden axiom. 18 ===== PDF PAGE 228 / 433 ===== Q4. The generation↔Morse-cell assignment (A1). The count χ(OP 2 ) = 3 is algebra-derived (§5); the assignment of the three families to the cells {0, 8, 16} needs the golden mass-ordering (dynamics) and is folded as honest-open. Q5. det G = 7 provenance. Derive det G = 7 from the E7 /annulus/missing-16 geometry, or it remains the deepest named Input. Rev25 update (the S216–S220 state of the deferred Ciab target). The stabilization item above was subsequently executed in the dynamics lane, and this appendix’s account is updated to the S218–S220 state rather than its original one. (i) An intermediate AMBER verdict (S215) was overturned by a convention fix (S216) — it is void, not a standing wound. (ii) The named wall — A-tensor projector precision (S217) — was killed at S218 by the exact rebuild (kernel s638: GAP = 0.0; Gate C 1.1 × 10−15 ). (iii) Gate H then ran the Coleman–Weinberg inertia readout: result (9, 0, 1) — against the pre-registered blind prediction (10, 0, 0) (A1329), scored per protocol as a recorded loss. The candidate mechanism therefore carries structural candidate wording (S220), neither promoted nor buried; the receipts, including the loss, are collected in Appendix Q. The dynamics tier itself stays held for Rev26; nothing in this appendix’s fences moves. 19 ===== PDF PAGE 229 / 433 ===== Frame fence on every Coleman–Weinberg Hessian number (S217 kernels s618/s619/s621/s623; landed S272). Three separate live-frame recomputations bear on how CW quantities in this lane may be quoted, and they do not all point the same way. (i) The sign is frame-robust. Recomputing Str(M 4 log M 2 ) in the live self-consistent frame (coset scalar, actual DWS E-frame vectors, live A21 = 2.0106) gives −45.66 (natural, cS = 12, cV = 1) and −45.89 (M 2 /M 4 co-calibrated, cS = 12.011, cV = 0.9971), against −43.89 and −45.61 from the earlier stale-input frame. The CW log is negative in all four live variants and the magnitude moves by ≲ 2 units. Unlike the sub-percent supertrace residual — which does flip sign stale↔live — the CW log is dominated by the heavy DN-scaled fermion sector, so the 0.2–0.8% input shift never flips it. (ii) The finite-difference route to H (1) is branch-limited, not merely noisy. The ten flats are exactly massless at the vacuum, so a displacement h lifts nearby modes across the 10−4 “massive” threshold and the sector counts [36, 48, 24, 8] do not stay fixed; shrinking h avoids the crossing but the outer Hessian amplifies the inner finite-difference noise as 1/h2 . There is no single h that is both branch-stable and low-noise. At flat 0 the nominal diagonal Haa = +48.8 carries branch_ok = False and a first-order Ward residual da StrM 2 = −1.78, so that value is contaminated by a branch discontinuity and is not quotable. This is the pre-registered risk (“same DWS branch counts at every sample”), realized. (iii) ⋆ The binding fence: two published flat bases are not the same sub- space. The principal cosines between the live flat basis and the earlier one are [ 1.000, 0.868, 0.693, 0.669, 0.626, 0.457, 0.424, 0.365, 0.143, 0.057 ] — agreement only in the first direction, and effectively orthogonal by the last. The kernel records same_subspace = False. A CW Hessian computed on the earlier basis is therefore an object on a different subspace, not a competing estimate of the same number, and the two must never be compared or averaged. On the live three-flat subset the fermion CW Hessian inertia is (3, 0, 0) with eigenvalues (−0.1927, −0.1727, −0.1421); the partial Ward gates are nonzero by construction (the scalar and vector sectors are absent), so this is a scoped diagnostic and nothing here is promoted. α and colour remain fenced, the scale remains the unit anchor U , and the engine is untouched. Dead / quarantined (do not revive without a new hook). (a) βd = 11/(6π) as a derivation — dead (clock-period convention only). (b) Area spectrum / LQG quantization — dead-for-now. (c) An RG flow from Det = 0 (UV) to Det = 1 (IR) “explaining” the t/b/c tensions — quarantined (conflicts with the firewall-locked A713 result). (d) Triality / the 3-axis as the fermion beat-depth selector — closed-negative (D805). (e) Selecting ϕ by minimizing a single invariant — gives the symmetric (1, 1, 1) (the σ-even no-go, Prop. 2.1); only the three-constraint Route A and the σ-odd defect work. (f) Coleman–Weinberg as a unique golden selector — closed-negative in the tested σ-equivariant layers (Prop. 2.1); do not re-open it as the selector. (g) “CW lifts the 8 σ-even residual” as a parity-block readout — barred (golden is σ-broken, Prop. 4.1); needs the certified Ciab . Status Derived selection layer, Definition A. Banked: the derived dynamical vacuum — golden = tachyon-free Minkowski critical point of dyonic ω = π/4 SO∗ (8) (Theorem 1.1, A887); the σ-even/σ- 20 ===== PDF PAGE 230 / 433 ===== odd selection dichotomy — CW unable to uniquely select golden in the tested σ-equivariant layers (Prop. 2.1), the σ-odd unit defect the sole selector with F = 1 ⇔ AX1 (Lemma 2.2); the BPS / source-completion theorem — direction, + 12 ∥B∥2 completion, and radial m2 = 7/4 derived, unit magnitude ≡ AX1 (Theorem 3.1); the generation count as a Morse index χ(OP 2 ) = 3 (Prop. 5.1); plus the retained probe layers (overdetermined invariant routes, geometric bounce, E7(7) clock T 2 = −1, moment–annulus synthesis). Honest end-state. Golden vacuum selection is a derived σ-odd-defect mechanism modulo the AX1 unit, with Coleman–Weinberg structurally unable to uniquely select golden in the tested σ-equivariant layers. Full stabilization is not closed: of the 18 physical moduli the 9 σ-odd are lifted by the defect, and the 9 σ-even residual split as 8 flats whose lift sign is pending the certified field- dependent tensor Ciab (provably undecidable from σ-evenness and the point spectrum — Prop. 4.1) plus 1 no-scale dilaton (a separate scale/lift input). “Selected — yes; fully stabilized — not yet.” This is the strongest honest form of the answer to the no-dynamics killer. Fences. All on the certified substrate but Definition A: the unit magnitude is AX1, not derived from nothing; the σ-even residual lift is not a parity-block readout (needs Ciab ); the no-scale dilaton is a separate input; the tested finite endpoint/self-adjoint/source layer is exhausted for the 7/3 edge but its residual is non-finite; B4 is a sub-GeV/GeV-order scheme-dominated crossing, not a 1.8 GeV claim; the bounce is classical; the mass scale is Input. No public observable moves. 21 ===== PDF PAGE 231 / 433 ===== Leibniz Quantum Beats Newton Appendix J — The Code Hamiltonian from the Parent’s Transversal Weyl Symmetry: the C3 × C3 Penalty Form and the Reynolds-Projector Unification of the One-Generation Yukawa Sector (Rev33.1) Derived at the symmetry level (s239/s240). NOT from a written SGTOE action — no Lagrangian yet imposes PX , PZ ; that identification is the lone residual. No public observable moves. Tom O’Sieg August 2026 1 Scope and standing assumptions This appendix records two results obtained at the symmetry level on top of the banked Peirce-monad qutrit code of A1039–A1041. We work in the parent Hilbert space ∼ C27 , H = C3A ⊗ C3B ⊗ C3C = the three tensor factors being the three Peirce monads, each modelled on the compact qutrit slice H3 (Cu ) ,→ J3 (Os ) identified in A1040. Let X, Z be the qutrit Weyl–Heisenberg clock and shift, satisfying the single-leg cocycle relation ZX = ωXZ with ω = e2πi/3 . Define the transversal (three-leg) operators SX = X ⊗ X ⊗ X, SZ = Z ⊗ Z ⊗ Z, and the cycle averages     PX = 13 I + SX + SX 2 , PZ = 13 I + SZ + SZ2 , Pcode = PX PZ . A1039/A1040 established (and s239 re-verified from the banked matrices) that [PX , PZ ] = 0, rank Pcode = 3, and that the rank-3 image im Pcode is the Peirce-qutrit code. The code Hamiltonian was, at that stage, a stated commuting-stabilizer choice, Hcode = (I − PX ) + (I − PZ ), with ker Hcode = im Pcode and a spectral gap ∆ = 1. The framework gates of A1039/A1040 (all_SGTOE_embedding_gates_pass and all_full_SGTOE_derivation_gates_pass) were correctly left False on exactly the point “the SGTOE parent action does not yet impose PX , PZ .” The two results below are honestly partial. Result 1 (s239) shows that the frustration-free penalty form of Hcode is forced (not a free choice) by a symmetry the parent already has — the transversal Weyl–Heisenberg C3 × C3 — with the symmetric point a = b = 1 a normalization (no proven X ↔ Z swap is exhibited here that would force a = b); and that the cubic (three-body) arity is forced by 1 ===== PDF PAGE 232 / 433 ===== the qutrit cocycle and coincides with the arity of the exceptional Jordan cubic. Result 2 (s240) shows that Pcode and the A1048 hypercharge projector PY =0 are the same operator class — the Reynolds (group-average) projector. That class result does not by itself produce the democratic one-generation Yukawa ray (Rev32.9, A1652 F01b: the “this forces the democratic ray” gloss printed here through Rev32.8 is deleted; see the multiplicity-one hypothesis of Proposition 3.2). Neither result touches a public observable. Both carry forward all standing fences: the compact unit u is unselected (the S 2 weak-plane family of A1041), Ngen = 3 is an input, and the single actual outcome is RED. 2 Result 1 (s239): the forced frustration-free penalty form of the transversal Weyl C3 × C3 2.1 Three-body arity is forced by the qutrit cocycle On a single qutrit, ZX = ωXZ (verified residual 6 × 10−16 ). On n transversal legs the cocycle multiplies: Z ⊗n X ⊗n = ω n X ⊗n Z ⊗n .     Hence the transversal clock and shift commute iff ω n = 1 iff 3 | n. Novelty label: new specialization proved here. (n) (n) Lemma 2.1 (Cocycle-forced cubic arity). SX = X ⊗n and SZ = Z ⊗n commute if and only if 3 | n; n = 3 is the minimal positive leg number that abelianises the transversal clock and shift. The gate values are unambiguous: at n = 1 and n = 2 the operators do not commute (at n = 2 (2) (2) the commutator norm is ∥[SX , SZ ]∥ = 5.20, the ω 2 phase), while at n = 3 they commute to ∥[SX , SZ ]∥ = 2.8 × 10−15 . An independent cross-check script reproduces the 5.20 (non-commuting) versus 2.8 × 10−15 (commuting) values. Therefore the three-monad interaction is not a modelling choice: it is the minimal leg count at which a commuting-stabilizer code of this transversal form can exist at all, and 3 is exactly the arity of the exceptional Jordan cubic trilinear. This discharges the prior open item B5 (“three monads = legs of one trivalent cubic vertex”). 2.2 The code is the Weyl-singlet sector The group ⟨SX , SZ ⟩ is an abelian C3 × C3 acting on C27 . Classifying C27 by the nine (j, k) isotypes via the cycle-average projectors 2 2 (X) (Z) = 31 ω −jm SXm , = 13 ω −km SZm , X X Pj Pk m=0 m=0 (X) (Z) each of the nine joint isotypes Pj Pk has dimension exactly 3 (27 = 9 × 3, the regular- representation structure), and the trivial–trivial isotype coincides with the code: (X) (Z) = Pcode ∥P(0,0) − Pcode ∥ = 1.9 × 10−15 .  P0 P0 Novelty label: new specialization proved here. 2 ===== PDF PAGE 233 / 433 ===== Proposition 2.2 (Code = Weyl singlet). Pcode is the (trivial, trivial)-isotype projector of the transversal Weyl–Heisenberg C3 × C3 ; equivalently the group-average projector onto the Weyl-singlet sector. The monad code is precisely the gauge-singlet subspace of the parent’s transversal Weyl symmetry. 2.3 The penalty form is forced (relative weight a normalization) Within the abelian ∗-algebra A = ⟨SX , SZ ⟩ (the C3 × C3 group algebra), consider the most general Hermitian penalty built from the two cycle averages with the code as zero-energy ground: H(a, b) = a (I − PX ) + b (I − PZ ), a, b > 0. Novelty label: new specialization proved here. Theorem 2.3 (Forced frustration-free Weyl-invariant penalty form). For all weights a, b > 0: 1. the ground space of H(a, b) equals im Pcode (rank 3), independent of a, b; 2. the spectral gap is ∆ = min(a, b); 3. each term is positive semidefinite and annihilates the code (I − PX )Pcode = (I − PZ )Pcode = 0 ,  so H(a, b) is frustration-free; 4. both C3 factors are required: dropping either term leaves a rank-9 ground space, not the rank-3 code. Consequently the functional form of the penalty — both terms (I − PX ) and (I − PZ ) required, frustration-free, with ground space exactly im Pcode — is forced by the transversal Weyl C3 × C3 symmetry. The symmetric representative Hcode = (I − PX ) + (I − PZ ) (a = b = 1) is the natural normalization of this family. Remark 2.4 (What is and is not fixed). Clauses (i)–(iv) establish a two-parameter family H(a, b), a, b > 0, all sharing the code ground space; they do not single out a = b = 1. Selecting the symmetric point as unique would require an additional X ↔ Z duality that actually swaps PX and PZ , together with an overall normalization. The involutions exhibited in §2.4 (Weyl inversion R, golden reversal J) fix PX and PZ individually and do not supply that swap. We therefore claim only that the penalty form is symmetry-forced and that a = b = 1 is its normalized symmetric point; the relative weight a : b is a normalization, not a derived number. Exhibiting the swap duality (and thereby promoting “form forced” to “Hcode unique”) is left open. The scan over (a, b) ∈ {(1, 1), (0.4, 1.7), (2.3, 0.6), (1, 3)} returns rank-3 code ground in every case with ∆ = min(a, b); the single-term Hamiltonians (I − PX ) and (I − PZ ) each have a rank-9 ground, confirming clause (iv). Thus the Weyl symmetry forces the functional form, and Hcode is its normalized symmetric point. The overall positive scale (the gap units) remains a normalization, not a derived number. 2.4 The primal/dual involution is a dynamical symmetry of Hcode The s236 primal/dual involution motif (T2) was previously a structural choice. Two distinct involutions of the qutrit parent are exact symmetries of the derived Hcode : 3 ===== PDF PAGE 234 / 433 ===== • Weyl inversion R : |i⟩ 7→ |−i mod 3⟩, with RXR = X −1 , RZR = Z −1 . Each cycle average is invariant under g 7→ g −1 (since {I, S, S 2 } = {I, S −1 , S −2 }), so R fixes PX , PZ . With Uσ = R ⊗ R ⊗ R one finds [Uσ , Hcode ] = 2.1 × 10−15 (the ghost-sign / round-trip leg). • Golden reversal J : |i⟩ 7→ |2 − i⟩, with JDφ J = Dφ−1 (the dual-VEV reciprocity Dφ ↔ Dφ−1 on the golden vacuum Dφ = diag(φ, 1, φ−1 )), and [UJ , Hcode ] = 2.8 × 10−15 (the dual-VEV leg). Novelty label: new specialization proved here. Proposition 2.5 (Structural → dynamical). The s236 primal/dual involution motif (ghost sign and dual VEV Dφ ↔ Dφ−1 ) is realised as an exact symmetry of the derived Hcode : [Uσ , Hcode ] ≈ 2 × 10−15 and [UJ , Hcode ] ≈ 3 × 10−15 . It is thereby upgraded from a structural choice to a dynamical symmetry, discharging the B4-pointer rung at the symmetry level. 3 Result 2 (s240): the Reynolds-projector unification of the one- generation Yukawa sector 3.1 Pcode and PY =0 are one operator class A Reynolds projector is the canonical projector onto the trivial-isotype (singlet) subspace of a finite or compact group G: 1 X Z PG = ρ(g) (finite G), PG = ρ(g) dg (compact G). |G| g∈G G It satisfies the defining identities P 2 = P, P † = P, and the group-absorption relation P ρ(g) = ρ(g) P = P. The code projector is exactly the C3 × C3 group average: 2 1 X Pcode = SjSk residual 1.9 × 10−15 ,  9 j,k=0 X Z idempotent (5.8 × 10−16 ), Hermitian (0.0), and group-absorbing (∥Pcode S − Pcode ∥ = 1.2 × 10−15 ). The A1048 hypercharge projector PY =0 = indicator(Ytotal = 0) is the U (1)Y group average onto ker(Ytotal ). Reconstructed from the banked Y16 grading and YHiggs = {± 12 }, it equals the shipped hypercharge mask exactly (err 0.0), is idempotent, satisfies PY =0 Traw = Tproj (err 0.0), and removes precisely the forbidden mirror half Ytotal ∈ {−1, +1}. Novelty label: standard theorem used. Theorem 3.1 (Reynolds unification). Pcode (transversal Weyl C3 × C3 ) and PY =0 (U (1)Y ) both satisfy the defining Reynolds identities P 2 = P, P † = P, Pρ(g) = P (group absorption). They are the same operator class — each is the Reynolds projector onto the singlet sector of its own group — but not the same operator: Reynolds projectors of different groups have different images, so a conclusion established for one is not thereby established for the other (Rev32.9, A1652 F01b). The loaded one-generation Yukawa sector is therefore Tloaded = Weyl-singlet Reynolds ◦ hypercharge-singlet Reynolds Traw ,   i.e. the raw exceptional cubic averaged over the parent symmetry and restricted to its singlet. 4 ===== PDF PAGE 235 / 433 ===== The shipped norms anchor the half-leak: ∥Traw ∥2 = 64, ∥Tproj ∥2 = 32 (leak ratio exactly 12 ), with the removed ∥·∥2 = 32 carried entirely by the forbidden Y = ±1 components. 3.2 Democracy is Reynolds-forced; splitting requires a compensator A Reynolds projector selects the singlet sector; it does not reweight within it (item 3 below is the load-bearing statement). Whether the selected sector can then be split depends on its multiplicity (Proposition 3.2). The banked data make this concrete: 1. the invariant coefficients are democratic, (1, 1, 1, 1), with spread 2.2 × 10−16 ; 2. democracy is genuine, not a unit-normalization artifact: the channel basis is orthogonal (off- diagonal Gram 0.0) but has unequal norms (Gram-diagonal spread 8.0, from distinct channel multiplicities), yet the cubic coefficients are all exactly 1; 3. re-projecting the democratic sum with PY =0 keeps (1, 1, 1, 1) (spread 2.2×10−16 ): the Reynolds projector selects, it does not reweight; 4. a deterministic non-invariant golden spurion (weighting channel f by φkf , k = [3, 0, 1, 2]) lifts the ray to coefficient spread 3.236, breaking democracy. Novelty label: standard theorem used. Proposition 3.2 (Compensator required). At trivial multiplicity one — when the singlet sector of the averaged symmetry G has multiplicity one on the four-channel span — no operation invariant under G can split the democratic (1, 1, 1, 1) ray of the one-generation Yukawa sector (Schur). Without that hypothesis the statement is false: diag(2, 1, 1, 1) is an invariant operator that splits the ray at multiplicity four (s1172 T2). Exhibition (Rev32.9, A1652 F01b): the multiplicity-one mechanism is exhibited by s243 on the model group S4 permuting the four channels (C4 = 1 ⊕ 3; its gate G4 is the m = 2 counter-model). The suite has not exhibited a registered group acting on the four physical channels with trivial multiplicity one: U (1)Y fixes all four channels, and |C3 × C3 | = 9 admits no orbit of size 4. The hypothesis is therefore a premise of this proposition, not a derived fact. A compensator outside the invariance is mathematically required to lift the ray; an explicit golden spurion is shown to do so (spread 3.236), naming golden Jvac as a viable candidate structure. This is the in-house half of the D885 question: it converts “is a compensator needed?” into “a compensator is mathematically required,” and identifies the golden Jvac as the candidate. It does not derive the four physical Yukawa coefficients — the spurion shows breaking is possible and supplies a candidate, not the physical hierarchy. 4 Result 3 (s241–s246): the written-action layer and the off- diagonal no-go theorems This section folds the S160–S161 written-action upgrades (in-house kernels s241–s246), which sharpen Results 1–2 from the symmetry level toward an explicit one-generation action and close several off-diagonal routes at theorem grade. Nothing here moves a public observable. 5 ===== PDF PAGE 236 / 433 ===== (s241) Manifest kinetic term. The code Hamiltonian is exhibited as a manifest sum of squares, Hcode = (I − PX ) + (I − PZ ), a frustration-free, manifestly positive-semidefinite kinetic form whose zero modes are exactly the joint ±1 eigenspace of the primal/dual projectors PX , PZ . This closes the W2 written-Lagrangian residual at the level of the kinetic term. (s242) Tiered one-generation Lagrangian and the quarantine theorem. Assembling the kinetic term with the cubic-invariant Yukawa of Result 2 gives a tiered one-generation Lagrangian in which the core democratic symmetry is realised exactly. Novelty label: new specialization proved here. Theorem 4.1 (Quarantine, s242). In the assembled one-generation action the within-generation hierarchy is confined entirely to the input tier: the democratic core is exact and symmetry-protected, and every hierarchy-generating quantity is an external input, not a derived coefficient. This is the constructive companion to Appendix K’s negative: the action isolates the undetermined coefficients rather than hiding them. (s244–s246) Off-diagonal no-go at theorem grade. Three further kernels close the off-diagonal selector routes. Novelty label: new specialization proved here. Theorem 4.2 (Joint frame×η Klein-four no-go, s244). No joint action of the frame rotation and the η-grading (a Klein four-group of candidate off-diagonal selectors) generates the within-generation splitting; the multiplicity-one fence forbids it and an explicit m = 2 counter-model saturates the bound. The nucleation analysis (s245) shows the “perfect-imperfection” defect O = 12 diag(1, 0, −1) is a seed, not a source: it can mark where a splitting would nucleate but cannot supply its magnitude. Kernel s246 locks the frame orientation σframe to the loaded vacuum Jvac , removing it as a free off-diagonal handle. Together these upgrade the Reynolds/Schur no-go (Appendix K, s243) to theorem grade. 5 Residual debt and firewall status Remark 5.1 (What is derived, what is not). Result 1 derives Hcode at the symmetry level: it is the unique frustration-free penalty enforcing the parent’s transversal Weyl C3 × C3 , with cubic arity forced by the qutrit cocycle, and with the primal/dual involution realised as a dynamical symmetry. Result 2 unifies the one-generation mass mechanism as a single Reynolds projection and proves democracy is forced. There is still no written microscopic SGTOE Lagrangian exhibiting PX , PZ as explicit operator terms. 6 ===== PDF PAGE 237 / 433 ===== Residual fence (the lone open piece). The remaining debt is to exhibit the transversal Weyl C3 × C3 as a symmetry of the written SGTOE parent Lagrangian. This is conditional on the standing A1039/A1040 identification of the qutrit clock/shift with the Peirce-triality structure of the Albert cubic J3 (Os ). Everything in this appendix is derived at the symmetry level; the written-Lagrangian identification of that Weyl symmetry is the single outstanding item. Carried fences, unchanged: the compact unit u is unselected (the S 2 weak-plane family of A1041, an unbroken Sp(1) no-go), the overall positive scale of Hcode is a normalization, Ngen = 3 is an input, and the four physical Yukawa coefficients are not derived (compensator required, not supplied). Debt narrowed, not closed. Firewall. No public observable moves. The single actual outcome remains RED / untouched: no collapse claim is made. The engine and presentation workbooks are untouched (the kernels read banked cold-storage npz only; RNG-free, run1=run2 byte- identical). Nothing in this appendix is promoted to a prediction. 7 ===== PDF PAGE 238 / 433 ===== Leibniz Quantum Beats Newton Appendix K — The Within-Generation Yukawa Coefficients are Irreducibly External to the Banked Parent Action: a Scoped Honest Negative from Five Independent Angles (Rev33.1) Derived: 4 SM Yukawa tensors + democratic normalization. Coefficients = irreducibly external (scoped to the banked framework). No public observable moves. Tom O’Sieg August 2026 1 Purpose and scope This appendix records a conditionally-closed honest negative. It is the sharpest statement the SuperGrokTOE program has reached on what the exceptional-Jordan parent action can and cannot fix about the Standard-Model (SM) Yukawa sector. The result has two halves, and keeping them separated is the whole point of the appendix. 1. What is derived (positive). A Der(Os )-invariant trilinear on the loaded Albert cubic (uniqueness of the invariant not claimed; see Theorem 2.1) projects onto exactly the four SM Yukawa tensors {Q uc H, L ν c H, Q dc H c , L ec H c }, with a single democratic common normalization (1, 1, 1, 1). [LIB2-111] This is a UV/matching statement; it is real and it is banked. 2. What is not derived (negative, scoped). The four distinct physical coefficients {yu , yd , ye , yν } — i.e. the entire within-generation mass hierarchy — do not follow from the banked parent action. [LIB2-118] Six independent analytic and numerical angles agree on this. A symmetry- breaking compensator lying outside the banked invariance structure would be required, and none of the banked enhanced symmetries supplies one. The negative is scoped, not universal. It is a no-go over the banked exceptional-Jordan / loaded- gauge parent action together with perturbative running and the banked enhanced-symmetry set. Non-perturbative and composite strong-coupling dynamics are not banked, hence are not excluded in general. The honest published claim is therefore exactly: the four SM Yukawa tensors plus their democratic normalization, full stop. No public observable moves; no dimensionful anchor is dropped; Ngen = 3 remains an input. 1 ===== PDF PAGE 239 / 433 ===== 2 What the framework does derive 2.1 The unique exceptional-cubic coupling Work over the split octonions Os (signature (4, 4); imaginary part Im(Os ) of signature (3, 4)) and the Albert algebra J3 (Os ). The derivation algebra Der(Os ) is the split form g2(2) , of dimension 14, and the G2 -invariant symmetric form on Im(Os ) is one-dimensional (proportional to the polarization η3,4 ). Novelty label: standard theorem used. Theorem 2.1 (An invariant trilinear). The cubic Yukawa coupling T (c, b, a) = Re c · (b · a)  is Der(Os )-invariant and nonzero. [LIB2-110] Uniqueness caveat: that T is the only such invari- ant trilinear is not established here — it would require computing the invariant multiplicity dim HomG2(2) (V1 ⊗ V2 ⊗ V3 , R) with the exact representations and scalar components specified. We use only invariance (verified below), not uniqueness; the democracy conclusion does not depend on uniqueness. [LIB2-110] The invariance is verified to machine precision: the maximal Leibniz residual obeys max T (D·, ·, ·) = 4.2 × 10−15 , max |T | = 12.8, P so T is genuinely nonzero and genuinely invariant. The conjugated form τ (x, y, z) = Re (x̄y)z̄ used   by an independent component-level check is the same coupling up to conjugation, and is likewise Der(Os )-invariant. 2.2 G2 transitivity forces democracy Novelty label: standard theorem used. Lemma 2.2 (Transitivity on a fixed-norm imaginary orbit, split form). Im(Os ) carries an indefinite (3, 4) form, so it is not the compact S 6 ; the compact identity G2 /SU (3) ∼ = S 6 may not be imported directly. The relevant group is the split real form G2(2) , which acts on Im(Os ) preserving the norm, with distinct orbit classes for positive-, negative-, and null-norm vectors. On a fixed positive-norm pseudo-sphere the Der-orbit tangent has dimension 6, and G2(2) acts transitively on that single norm-class orbit. The four gauge-allowed channels (the Y = 0 components of the census in Angle (b), §3.2) lie within one such orbit class, on which any G2(2) -invariant trilinear is constant. (Equality of a local tangent dimension with 6 is consistent with, but does not by itself prove, global transitivity; the latter is the standard split-G2 orbit statement, used here per-orbit-class.) Novelty label: new specialization proved here. Proposition 2.3 (Democratic readout). Evaluated across the seven G2 -rotated copies of a single channel, the cubic coupling T is constant (spread 3.6 × 10−12 ; all values = 1.044419). Consequently channels that differ only by their hypercharge assignment receive identical cubic weight. 2 ===== PDF PAGE 240 / 433 ===== Theorem 2.1 plus Lemma 2.2 give Proposition 2.3: an invariant trilinear, evaluated on an orbit on which the symmetry acts transitively, cannot distinguish points of that orbit. Premise, stated and not derived (Rev32.10, A1717-F02): Proposition 2.3 exhibits transitivity on the orbit of a single channel vector (its seven G2 -rotated copies); passing from that vector orbit to the four channel triples requires that the four normalized ordered triples lie on one common G2(2) orbit under a single group action. No registered-frame component certificate identifying the four triples under one action is exhibited here, so common triple-orbit membership is a premise of this subsection. Under that premise the four gauge-allowed channels enter with the common democratic ray (cu , cν , cd , ce ) = (1, 1, 1, 1). This is the G2 tight-frame value. It is a UV normalization statement, not a low-energy prediction. [LIB2- 113] Net of §2. The exceptional cubic derives the four SM Yukawa tensors {Q uc H, L ν c H, Q dc H c , L ec H c } and one democratic normalization (1, 1, 1, 1) — and noth- ing finer. Everything that remains (the four coefficient ratios = the fermion mass hierarchy) is external to the invariant cubic. 3 Six independent angles that the within-generation hierarchy is external The following six lines were obtained by distinct methods — abstract group theory, explicit component census, Reynolds-projector algebra, an enhanced-symmetry candidate sweep, one-loop renormalization-group (RG) running, and the parent-action cubic-cone kinetic metric. They converge on a single conclusion. 3.1 Angle (a): the 7–3 fence (s238) Distinct coefficients {yu , yd , ye , yν } would have to be supplied by the loaded U (1)Y , whose Casimirs are distinct, qd = 0.494, qu = 0.538, qν = 0, but U (1)Y ̸⊂ Der(Os ) = G2 . A G2 -invariant cubic cannot carry an object that lives outside G2 . The cubic therefore yields only the democratic value; any splitting requires G2 -breaking loading. This is precisely the standing k=1 LSZ / 7–3 running fence. 3.2 Angle (b): exact half-norm mirror leakage (A1048) At the explicit component level, the raw octonionic trilinear τ carries exactly half of its squared norm as Y = ±1 “forbidden mirror” leakage: |τraw |2 = 64.0, |τproj |2 = 32.0, leakage fraction = 21 (exact). The loaded U (1)Y Reynolds projector PY =0 removes this leak exactly — the projected tensor equals the mask-times-raw tensor (∥diff∥ = 0) and equals the four SM tensors (∥diff∥ = 0). [LIB2-111, LIB2-112] The component census confirms every forbidden component sits at total hypercharge ±1 3 ===== PDF PAGE 241 / 433 ===== (HQdc , HLec at +1; H c Quc , H c Lν c at −1) and every allowed component at Y = 0. Thus the raw J3 cubic is not the physical Yukawa tensor without the U (1)Y projection; the projection that makes it physical is precisely the loaded U (1)Y Reynolds average. That average removes the nonzero-total-hypercharge components and acts as the identity on the four allowed tensors, so it retains their common coefficients: this is preservation, not selection — the averaging by itself neither singles out equality nor excludes an unequal quartet on the same span (Angle (c); s1171 T1). 3.3 Angle (c): democracy is Reynolds-preserved (s239/s240; cross-ref. Appendix J) The monad-code penalty Hcode (Appendix J) is the symmetry-forced transversal-Weyl-singlet penalty form, and its projector Pcode (C3 × C3 ) lies in the same Reynolds-projector class as PY =0 (U (1)Y ). Both are group averages of different groups, so they are one class, not one operator (Appendix J). The Reynolds projector selects the singlet sector and does not reweight within it; PY =0 is the identity on the span of the four displayed tensors, which all carry total hypercharge 0 (s1171 T1). [LIB2-113] The democratic ray is protected at trivial multiplicity one (Appendix J, Proposition on the compensator; exhibition by s243 on a model group), and that hypothesis is a premise here, not derived: no registered group acts transitively on the four channels (U (1)Y fixes them; |C3 × C3 | = 9 admits no orbit of size 4). (Rev32.9, A1652 F01b/S330.7: through Rev32.8 this paragraph derived the ray from “a group average over a symmetry that acts transitively on the channels”, which is unavailable for both projectors.) Under that premise democracy is not an accident of normalization. [LIB2-113] To split (1, 1, 1, 1) one needs a compensator lying outside the invariance being averaged over; within the banked structure such a compensator is required, not optional. 3.4 Angle (d): the candidate-compensator sweep fails (A1049/A1050) A sweep of the banked enhanced-symmetry candidates was performed; each fails to produce the required complete within-generation hierarchy while respecting the projected channel census (Rev32.9, A1652: the dual-VEV bullet below does split up-type from down-type, so “fails to split democracy” was too broad): • F4 ⊃ G2 , E6 ⊃ F4 , B−L, Jvac : each either preserves the joint G2 × U (1)Y invariance (and so stays democratic) or reintroduces mixing with the already-projected-out mirror sectors (violating the A1048 tensor census). [LIB2-114] U (1)Y relabels the channels; it does not split them. • Dual-VEV (tan β = ̸ 1): splits up-type from down-type by O(10–100) but structurally cannot cross the quark↔lepton axis — it cannot separate mt from mu , nor mb from mτ . [LIB2-114] g 2 • Perturbative Z-factors: Zi ≈ 1 − 16π 2 C2 (Ri ) ln(Λ/µ) are bounded well below 10 ; reaching 3 10−5 would require non-physical exponentiation over trans-Planckian scales. The observed intra-generation split is ∼ 105 . • Jvac spurion: reaching ye /yt ∼ 3 × 10−6 needs depth ≈ Jvac 24 with φ−24 ≈ 9.6 × 10−6 — a dimension-∼ 28 insertion. This is numerology, not a leading-order mechanism. (Verified in-house: φ−24 = 9.645×10−6 ; exact depth for 3×10−6 is n ≈ 26.4.) Jvac governs the O(1) inter-generation beat-depth ratios well, but categorically fails the 105 intra-generation split. 4 ===== PDF PAGE 242 / 433 ===== 3.5 Angle (e): residue map rank-4; running alone fails (A1051) Write the field-residue map in log variables, y⋆ yf = q , ZLf ZRf ZHf with channel supports {Quc H}, {Qdc H c }, {Lec H c }, {Lν c H}, written with a single Higgs H and its conjugate H c (Paper 1 §6 names Hu , Hd ). The count below treats ZH and ZH c as formally independent: eight residues, rank 4, kernel dimension 4. Identifying ZH c = ZH merges the two Higgs columns and leaves seven residues at rank 4 and kernel dimension 3, so moving from one conjugate pair to two independent doublets takes the kernel dimension from 3 to 4, not from 4 to 3 (s1190; s1171 T4; Rev32.9, A1652). Rev32.10 and earlier printed that direction reversed and attached the eight-residue count to the single-Higgs label. The corresponding 4 × 8 incidence matrix has rank 4 (each channel carries a unique pivot Z-factor). [LIB2-115] Novelty label: new specialization proved here. Theorem 3.1 (Residue underdetermination). Because the residue map is rank 4 over eight inde- pendent Z-factors, the eight residues can realize any positive Yukawa quartet. [LIB2-115] The four coefficients are therefore underdetermined unless the parent action itself derives the residues and thresholds. If the residues stay Peirce-edge uniform, all four Yukawas remain equal (the golden edge product cancels: φ · 1 · φ−1 = 1, residual ∼ 10−16 ). Running alone cannot rescue the hierarchy. One-loop SM-plus-Dirac-ν running from the democratic unit ray lands at MZ on (yu , yd , ye , yν ) = (1.0207, 0.9958, 0.3959, 0.3581), whose ratios to the reference quartet are (1.02, 41.8, 38.8, 1.2 × 1012 ): the down/charged-lepton channels overshoot by ∼ 40× and the Dirac-ν channel by ∼ 1012 ×. Running produces O(1) spread, not the observed ∼ 105 intra-generation hierarchy. [LIB2-116] The Gram structure diag(12, 4, 12, 4) (color multiplicity) with unit coefficients confirms democracy is genuine, not a unit-normalization artifact. 3.6 Angle (f): the parent-action kinetic metric has relative rank zero (A1052) The deepest angle works directly with the parent action’s kinetic geometry rather than the cubic vertex. Take the cubic-cone metric GAB (J) = ∂A ∂B [− log N (J)] at Jvac = diag(φ, 1, φ−1 ) (N = 1; analytic metric matches the finite-difference Hessian to 1.7 × 10−6 ; raw signature (15, 12), det =1 tangent signature (14, 12), reflection-positive metric strictly positive with minimal eigenvalue φ−2 ). On the three off-diagonal Peirce edges the reflected positive metrics are component-uniform scalars, −1 12 = 2φ I8 , G+ 13 = 2I8 , G+ 23 = 2φI8 . G+ Every loaded Yukawa vertex draws one field from each edge (J12 (H) × J13 (ΨL ) × J23 (ΨcR )), so its canonical normalization is the common factor  + + + −1/2 −1/2 = 8 det Jvac = √1 ,  G12 G13 G23 8 5 ===== PDF PAGE 243 / 433 ===== the golden weights cancelling in the product (φ · 1 · φ−1 = 1). After yt = 1 this is again the ray (1, 1, 1, 1). Novelty label: new specialization proved here. Proposition 3.2 (Parent relative rank zero). The parent edge-residue map is rank 1 with relative rank 0: every channel sees the identical edge product, so the edge-uniform parent geometry cannot resolve Q from L inside J13 nor uc , dc , ec , ν c inside J23 . [LIB2-117] More generally, any two-derivative operator built by functional calculus from {N, Jvac , LJvac , UJvac , Θ} is scalar on each eight-dimensional Peirce edge. The full physical residue map (Theorem 3.1) has rank 4, relative rank 3; hence three independent loaded symmetry-breaking kinetic or threshold directions are needed to span a generic three-dimensional space of relative quartets. They are not required merely to split the four Yukawas: a more restrictive nonlinear law selects a particular quartet with fewer. y(t) = (1, et , e2t , e3t ) has four distinct entries for every t ̸= 0 and relative-log image of rank 1 (s1190), so one parameter suffices. This is the quantified form of Angle (c): "a compensator is required" becomes "the parent geometry has relative rank 0, so three independent external directions are needed to span the generic space." It is independently SHA-verified (in-house re-run, run1=run2; the physical-value gate is correctly negative), and its signature (15, 12) coincides with the Θ Jordan-automorphism signature of Appendix L. Convergence. Six independent angles — (a) the G2 /U (1)Y fence, (b) exact half-norm mirror leakage, (c) Reynolds-preserved democracy, (d) a failed candidate-compensator sweep, (e) a rank-4 underdetermined residue map plus failed running, and (f) a parent-action kinetic metric of relative rank zero — all return the same conclusion: the within-generation Yukawa hierarchy is irreducibly external to the banked parent action. 4 The scope fence Scope fence (read before citing this result). This is a no-go over the banked exceptional-Jordan parent action only: the G2 × U (1)Y invariants, perturbative one-loop running, and the banked enhanced-symmetry set {F4 , E6 , B−L, Jvac }. [LIB2-118] Non-perturbative and composite strong-coupling dynamics are NOT banked. They are therefore not excluded by anything above. [LIB2-114, LIB2-118] This is not a universal no-go: it does not claim that no conceivable parent could fix the coefficients. The honest published claim is exactly: the four SM Yukawa tensors {Quc H, Lν c H, Qdc H c , Lec H c } plus their democratic (1, 1, 1, 1) normalization — full stop. The four physical coefficient ratios (the fermion mass hierarchy) are recorded as irreducibly external to the banked parent, with a stated, falsifiable mechanism for what would have to be added. Remark 4.1 (What would close it). A genuine close would require the parent action to derive the eight Z-residues and thresholds (Theorem 3.1), or to supply a G2 -breaking compensator outside the banked invariance that survives the A1048 mirror-census — and to do so without re-introducing a dimension-∼ 28 spurion (Angle d). Until then the coefficient ratios remain an honest external input. 6 ===== PDF PAGE 244 / 433 ===== Remark 4.2 (Ledger status). This appendix moves the Yukawa-coefficient / 7–3 fence from “open blocker” to conditionally closed — honest negative (PI-approved this cycle), under the scope fence above. Xactual remains RED; Ngen = 3 remains an input; the engine and the rest of the paper suite are untouched; no public observable moves. Nothing is promoted on the strength of this negative. Novelty label: new specialization proved here. Theorem 4.3 (The 7/3 finite-carrier / trace-normalization theorem; S229). The heavy-quark trace ratio 7/3 = TrΘ (1 ⊕ 3 ⊕ 3̄)/Tr(3) is forced, not a chosen Markov normalization. On the reflected module 1 ⊕ 3 ⊕ 3̄ the most general trace weight is τa,b√= a Tr3 + a Tr3̄ + b Tr√ 1 , giving readout r = 2 + b/(3a). The Frobenius identity B([s, ui ], vi ) = 2 a = B(s, [ui , vi ]) = 2 b on the exact split-octonion (Zorn) product forces a = b; the twelve nonzero closure defects are all ∝ (a − b), so the constraint variety is exactly {a = b} and r = 7/3 is unique. [LIB2-053] Tiering (strict). This closes the last trace-weight freedom on the 7/3 road; it is not an LSZ residue, a low-energy mass theorem, or a physical export theorem. [LIB2-053] It is distinct from — and does not soften — the within-generation coefficient negative that is the subject of this appendix: the trace floor of 7/3 is now a theorem, while the physical export arrow (the loaded heavy-quark carrier of Appendix R’s “one carrier, three loadings” picture) remains the standing wall. Verified two independent ways in-house (verbatim byte-identical rerun; scratch re-implementation of the Zorn product); no public observable moves. Remark 4.4 (Theorem-grade upgrade, s243; off-diagonal closure cross-reference). The Reynolds/Schur underdetermination (Theorem 3.1) was subsequently upgraded to theorem grade by kernel s243: the residue map is multiplicity-one fenced, and an explicit m = 2 counter-model shows the bound is saturated rather than an artifact of the search. [LIB2-113] (That upgrade is a statement about the residue map of Theorem 3.1 and the model group of s243; the Registrar entry on Reynolds preservation, LIB2-113, asserts only that PY =0 is the identity on the four-channel span, and is not a certificate of this upgrade.) The complementary off-diagonal routes (a joint frame×η Klein-four action; the perfect-imperfection nucleation defect; and the frame-orientation handle) are closed at theorem grade in Appendix J’s written-action section (kernels s244–s246). The negative of this appendix is therefore not a single missed route but a closed family. 5 Summary The exceptional-Jordan parent action fixes the SM Yukawa structure completely — the four tensors and one democratic normalization — and fixes the within-generation coefficients not at all. The latter is not a failed search but a proved structural fact within the banked framework: G2 -invariance plus G2 -transitivity force democracy (Proposition 2.3); the splitting lives in the loaded U (1)Y ̸⊂ G2 ; the residue map is rank-4 underdetermined (Theorem 3.1); no banked enhanced symmetry supplies a compensator; the parent kinetic metric has relative rank zero (Proposition 3.2); and running alone misses by orders of magnitude. Six independent angles agree. The result is a clean, falsifiable stopping point, scoped honestly to the banked parent and claiming no more. 7 ===== PDF PAGE 245 / 433 ===== Leibniz Quantum Beats Newton Appendix L — Quantum-Mechanical Foundations on the Loaded Subimage: Born from Jordan–Gleason, the Tsirelson Bound from the Associative Envelope, and the Θ Jordan Automorphism (Rev33.1) Born ✓ & Bell ✓ conditional on the channel-count axiom; single-outcome Xactual remains RED. No public observable moves. Tom O’Sieg August 2026 Abstract This appendix collects the quantum-mechanical foundations of the SuperGrokTOE framework as they stand on the loaded subimage — the dimensionally-reduced, mass-bearing 4D effective category onto which Θ-compactification projects the otherwise scale-free exceptional parent. Three results are assembled. (i) The Born rule ω(P ) = Tr(ρ ◦ P ) is the unique non-contextual probability measure on the compact-positive order-unit space Cphys + , by the Bunce–Wright (Jordan–Gleason) theorem, precisely when the J3 trace form is positive-definite (27, 0); the master axiom Tr(ηΘ) = 7 is exactly that condition, with the raw split form being indefinite (15, 12). (ii) The map Θ = sign(η) is a Jordan automorphism of J3 (Os ) to machine precision, closing the Jordan-product/automorphism √ closure gap left open by the Born derivation. (iii) The Tsirelson bound ∥B∥ ≤ 2 2 follows from the associative envelope of the loaded C-subimage via the identity B 2 = 4I − [A1 , A2 ] ⊗ [B1 , B2 ], the exceptional parent being PR-box-compatible (under a chosen maximal tensor product, not by derivation); Euclidean JBW structure alone is shown to be insufficient. Everything here is conditional on the still-open channel-count axiom and on the H1a/H1b selectors, and — emphatically — the single-outcome problem Xactual remains RED. This appendix derives the statistical/relational structure of quantum mechanics (Born weights, Bell correlations, the quantum bound), not the selection of a single actual outcome. The measurement problem is not solved here. Contents 1 Setting: the loaded subimage and the firewall 1 2 The Born rule from Jordan–Gleason 2 2.1 The named theorem and its single discriminator . . . . . . . . . . . . . . . . . . . . . 2 2.2 The master axiom: Tr(ηΘ) = 7 as the positive-definiteness switch . . . . . . . . . . . 2 2.3 Composite (bipartite) Born . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3 The Θ Jordan automorphism: closing the closure gap 4 1 ===== PDF PAGE 246 / 433 ===== 4 The Tsirelson bound from the associative envelope 4 4.1 Bell lives on the associative subimage . . . . . . . . . . . . . . . . . . . . . . . . . . 4 4.2 The Tsirelson identity and the commutator bound . . . . . . . . . . . . . . . . . . . 5 4.3 Phase-3: the honest negative and the associativity firewall . . . . . . . . . . . . . . . 5 4.4 Physical narrative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 5 The firewall, stated as conditions 6 6 Rev28: the Xactual row splits 7 7 A positivity obstruction: the bare flow is not unitary in disguise (Rev30, A1482) 8 1 Setting: the loaded subimage and the firewall The framework’s kinematic arena is the rank-3 exceptional Jordan (Albert) algebra J3 (Os ), equipped with the split-signature metric η on the imaginary split-octonions Im(Os ). The flat (scale-free) parent is non-associative and indefinite; it admits neither a mass unit nor a bipartite composite. The physically realized sector is the loaded (massed, dimensionally-reduced) image obtained by the Θ-compactification Θ = sign(η), written Cphys + . Two distinct quantum structures live on this loaded image: the Born rule (the probability measure) and the Bell/Tsirelson structure (the bipartite correlation bound). The central claim of this appendix is that one mechanism — the Θ-loading, i.e. the channel-count axiom Tr(ηΘ) = 7 — underlies both. Before any derivation, we state the firewall once, prominently, because every result below is fenced by it. ⋆ FIREWALL — the measurement problem is NOT solved here. All results in this appendix derive the statistical / relational structure of quantum √ mechanics: the Born weights ω(P ) = Tr(ρ ◦ P ) and the Bell correlation bound ∥B∥ ≤ 2 2. They do not address the selection of a single actual outcome. The single-outcome quantity Xactual — which state of affairs is realized in an individual measurement — remains RED (unsolved). Nothing here is a hidden-variable theory, a collapse dynamics, or a derivation of definite outcomes. Born ✓ and Bell ✓ are CONDITIONAL (see Sec. 5); Xactual is not ✓. 2 The Born rule from Jordan–Gleason 2.1 The named theorem and its single discriminator The decisive input is the Jordan generalization of Gleason’s theorem. Novelty label: standard theorem used. Theorem 2.1 (Bunce–Wright / Jordan–Gleason, 1992). Let M be a Euclidean (formally real) JBW algebra of rank ≥ 3 with no type-I2 (spin-factor) summand. Then every state on the projection 2 ===== PDF PAGE 247 / 433 ===== lattice of M is of the form ω(P ) = Tr(ρ ◦ P ) for a unique density ρ. In particular the Born rule is the unique non-contextual measure. For J3 (Os ) the rank is 3 (past the I2 obstruction), so the only remaining hypothesis is that the algebra be formally real, equivalently that the Jordan trace form be positive-definite. The entire Born question thus reduces to a single discriminator: the signature of the J3 trace form. Novelty label: standard theorem used. Proposition 2.2 (Trace-form signature discriminator; s215). The compact algebra J3 (O) has trace-form signature (27, 0) (Euclidean / formally real). The raw split algebra J3 (Os ) has signature (15, 12) (indefinite, not formally real), decomposing as (3, 0)diag ⊕ 3 × (4, 4)off-diag . [P-002] The numerical content (s215, 5/5 gates, run1=run2): the octonion norm signatures are compact O = (8, 0) and split Os = (4, 4); the J3 trace forms are (27, 0) and (15, 12) respectively; rank-3/no-I2 holds; and the trace form is a frame function, i Tr(ρ ◦ ei ) = Tr(ρ), verified over 2000 J3 (C) frames P with max |Σ − 1| = 1.7 × 10−15 . 2.2 The master axiom: Tr(ηΘ) = 7 as the positive-definiteness switch The compactification Θ = sign(η) is designed to flip the sign of the negative-norm generators in the trace evaluation. Define the Θ-invariant subalgebra and its order-unit (OS) inner product Jphys = {A ∈ J3 (Os ) : ΘAΘ† = A}, ⟨A, B⟩Θ = Tr Θ(A ◦ B) . (1)  The positivity argument (A1024): on Jphys the insertion of Θ into the trace renders ⟨A, A⟩Θ > 0 for every nonzero A (“sum of squares of real components”), so A2 + B 2 = 0 ⇒ A = B = 0 — the formally-real criterion. Bunce–Wright then applies directly. The bridge from the Im(Os ) metric to the J3 trace form is the canonical 3-block decomposition, which is the load-bearing geometric step. Novelty label: new specialization proved here. Theorem 2.3 (Master axiom; canonical 3-block proof; A1024). The (15, 12) signature of J3 (Os ) traces to the (3, 4) signature of Im(Os ) via the three off-diagonal octonionic blocks: each contributes 4 negative dimensions, 3 × 4 = 12 negative dimensions in total, the diagonal 3 × 3 Hermitian block contributing (3, 0). Writing ηphys = ηΘ on Im(Os ), one has the biconditional chain Tr(ηΘ) = 7 ⇐⇒ ηΘ = I7 ⇐⇒ J3 trace form is (27, 0) ⇐⇒ Cphys + formally real ⇐⇒ Born unique. (2) Proof sketch (arithmetic of the 3-block decomposition). The diagonal block contributes (3, 0). In the raw split case each of the three off-diagonal octonionic blocks contributes (4, 4), for a total (3+12, 12) = (15, 12). The Θ-flip acts as +1 on the four compact directions and −1 on the four split directions of each off-diagonal block, converting each (4, 4) to (8, 0); the diagonal (3, 0) is unchanged. The total becomes (3+24, 0) = (27, 0). Rendering η positive-definite on Im(Os ) is exactly ηΘ = I7 , i.e. Tr(ηΘ) = 7 (the maximal trace of the 7 × 7 block, attained only when every eigenvalue of ηΘ is +1). Each link is reversible, giving the biconditional. 3 ===== PDF PAGE 248 / 433 ===== Remark 2.4 (Formal reality is a property of the product, not only the metric). The link “(27, 0) trace form ⇔ Cphys+ formally real” is stated above at the level of the trace form (signature). A positive Θ-twisted bilinear form is necessary for formal reality but is not by itself sufficient: formal reality ( x2i = 0 ⇒ xi = 0) is a property of the Jordan product, and the +1 fixed space of the P (15, 12) involution is not automatically a 27-dimensional compact Albert algebra closed under that product. A complete proof requires exhibiting the explicit Euclidean fixed-point subalgebra (or an appropriate isotope) and verifying its rank, cone, and projection lattice. We therefore read the boxed chain as established at the trace-form level and conditional (on that subalgebra construction) at the product level — consistent with the overall conditional status of the channel-count axiom Tr(ηΘ) = 7. Remark 2.5 (One axiom, three closures; s221). Via the s221 sharpening, the single axiom Tr(ηΘ) = 7 simultaneously closes (a) Born uniqueness (Bunce–Wright), (b) the channel-count normalization 7/3, and (c) gauge-emergence. It is the master geometric switch of the framework: Born-unconditional ⇐⇒ gauge-emergence. Remark 2.6 (The projector P+ is load-bearing; A1034). The honest form of the count is Tr(ηΘP+ ) = 7, with P+ the loaded positive-frequency/physical projector: it is not an unrestricted global trace. Without P+ the restricted count is 5 and the global J3 trace is 27; only the loaded projection isolates the 7. This is exactly why the axiom remains open (it is the LSZ-level statement that P+ (Rpole ) exhausts the 7-dimensional physical block), and why neither A1007, s226, nor κ = 1 alone forces it — explicit counterexamples R7 , R8 , R9 survive. The biconditional above should be read with P+ understood throughout. 2.3 Composite (bipartite) Born For Bell scenarios one needs Born on the bipartite system. Under the loaded frame Jphys is a Type I complex JBW factor (an associative complex matrix algebra B(Cn )). The tensor product of two formally-real associative JBW algebras is itself formally real, and the induced product inner product factorizes, ⟨A1 ⊗ B1 , A2 ⊗ B2 ⟩Θ = ⟨A1 , A2 ⟩Θ ⟨B1 , B2 ⟩Θ > 0, (3) so Bunce–Wright extends to the composite, giving the bipartite Born rule ω(PA ⊗PB ) = Tr ρAB (PA ⊗ PB ) . The BGW non-associativity and the indefinite (15, 12) signature are both stripped by the  loaded projection. Conditional, not unconditional. The Type I identification rests on Im(FΘ,load ) ⊆ {associator = 0}, which is asserted (leading conjecture, A1022 Q1) rather than derived from the algebra action. The composite Born is therefore banked conditional on that projection lemma and on the master axiom Tr(ηΘ) = 7. 3 The Θ Jordan automorphism: closing the closure gap The Born derivation of Sec. 2 left one analytic step open: the subalgebra Jphys = {A : ΘAΘ† = A} is closed under the Jordan product only if Θ is an automorphism of that product, Θ(A ◦ B) = (ΘA) ◦ (ΘB). This is now established numerically to machine precision (s226). Novelty label: new specialization proved here. 4 ===== PDF PAGE 249 / 433 ===== Theorem 3.1 (Θ is a Jordan automorphism of J3 (Os ); s226). The map Θ = sign(η) satisfies Θ(A ◦ B) = (ΘA) ◦ (ΘB) for all A, B ∈ J3 (Os ), (4) verified over a complete basis (729 basis pairs) with maximum error 3.1 × 10−16 (worst pair (i, j) = (3, 10)), and over 500 random pairs with maximum error 6.2×10−15 . Moreover Θ2 = I to 4.4×10−16 and Θ is symmetric (deviation 0.0), with signature (15, 12). All gates pass. Remark 3.2 (What this closes). Theorem 3.1 closes the Jordan-product/automorphism closure gap that A1023/A1024 flagged as the remaining analytic step of the Born derivation. With Θ a genuine Jordan automorphism, Jphys is a bona fide Jordan subalgebra, the OS inner product ⟨·, ·⟩Θ is well-defined on it, and the positivity argument of Sec. 2 is self-consistent. Note this is a numerical verification at machine precision, not yet a closed-form algebraic identity; the signature it reports, (15, 12), is that of the raw split form (it is the trace Tr(ηΘ) = 7 that selects the loaded (27, 0) image). 4 The Tsirelson bound from the associative envelope 4.1 Bell lives on the associative subimage The Bell scenario is well-posed only where a bipartite tensor product exists, i.e. where the algebra is associative. The discriminator is the split-octonion associator [x, y, z] = (xy)z − x(yz) (s214, 5/5 gates, run1=run2): • On the flat parent Os the associator is large, max∥[x, y, z]∥ = 59.4 over 2000 random triples — non-associative, no associative embedding, BGW (Bondi–Gardner–Williams) obstruction real, no composite, Bell ill-posed. • On the loaded C-subimage span{1, u} with u ◦ u = − 12 1 the associator vanishes (5.1 × 10−15 over 500 general elements) — associative, the tensor product exists, standard quantum Bell applies. √ • On the associative subimage the CHSH value is S = 2 2 (saturated), the classical LHV maximum is 2, the PR-box (4) is excluded, no-signalling holds (max |∆| = 0), and determinism holds. The split-octonion associator is the concrete BGW obstruction: nonzero on the exceptional parent, zero on the C-subimage. The same compactification that generates the 4D mass unit (radion/dilaton → KK scale) is the loading that projects onto the associative frame. 4.2 The Tsirelson identity and the commutator bound Let A1 , A2 (Alice) and B1 , B2 (Bob) be dichotomic observables, A2i = I, Bj2 = I, and let B = A1 ⊗ (B1 + B2 ) + A2 ⊗ (B1 − B2 ) be the CHSH operator. The load-bearing algebraic identity is (s225 gate G1, residual 8.9 × 10−16 ): Novelty label: standard theorem used. Theorem 4.1 (Tsirelson identity and bound; s225 / A1022). On an associative envelope, B 2 = 4I − [A1 , A2 ] ⊗ [B1 , B2 ]. (5) 5 ===== PDF PAGE 250 / 433 ===== From A2 = I one has the commutator bound ∥[A, A′ ]∥ ≤ 2 (analytic via [σx , σy ] = 2iσz , ∥2iσz ∥ = 2; numerically max = 1.9999992 over 5000 random involutions). Hence √ ∥B 2 ∥ ≤ 4 + ∥[A1 , A2 ]∥ ∥[B1 , B2 ]∥ ≤ 4 + 2 · 2 = 8, so ∥B∥ ≤ 2 2, (6) √ √ tight at 2 2 (attained on Φ+ with optimal observables, S = 2.82843 = 2 2). The PR-box (CHSH = 4) is excluded, since ∥B∥ = 4 would require ∥[A, A′ ]∥ = 4, violating the A2 = I commutator bound of 2. The classical bound is ≤ 2 (exhaustive). The commutator bound is the key: a Cauchy–Schwarz estimate gives only the trivial 4 (the PR-box value); the commutator identity is the correct and tight tool. 4.3 Phase-3: the honest negative and the associativity firewall √ A sharper question (Phase-3) is whether 2 2 can be forced from the Cphys+ axioms — Euclidean JBW plus reflection positivity — without presupposing Hilbert-space operators. The honest answer is negative (A1033). Novelty label: new specialization proved here. Theorem 4.2 (Phase-3 honest negative; A1033). √ Euclidean JBW structure together with the OS inner product is insufficient to force ∥B∥ ≤ 2 2. The obstruction is precise: without an associative envelope the product (A1 ⊗ B1 ) ◦ (A2 ⊗ B2 ) is undefined, so the identity B 2 = 4I − [A1 , A2 ] ⊗ [B1 , B2 ] cannot be formed. The maximal tensor product Cmax (compatible with no-signalling) contains extreme points saturating CHSH = 4. OS positive-definiteness secures the Born rule but does not constrain the state-space geometry to the quantum boundary. Novelty label: standard theorem used. Theorem 4.3 (Associativity √ firewall; the “if” direction; A1033). On an associative Type I complex JBW factor, ∥B∥ ≤ 2 2 is a theorem: the tensor product is associative, so (A1 ⊗ B1 )(A2 ⊗ B2 ) = (A1 A2 ) ⊗ (B1 B2 ) is well-defined; the B 2 identity follows algebraically; and by Gelfand–Naimark the √ is the self-adjoint part of a C -algebra, whose C -norm yields ∥[A, A ]∥ ≤ 2 associative JBW algebra ∗ ∗ ′ and hence ∥B∥ ≤ 2 2. This requires no√import of Hilbert operators: associativity generates the C ∗ -envelope. The biconditional “∥B∥ ≤ 2 2 iff Type I complex JBW factor” has its “if” direction proven; the “only if” is supported by Proposition 4.4 but not fully proven. Novelty label: new specialization proved here. Proposition 4.4 (Parent is PR-box-compatible under Cmax ; A1033). On the full exceptional J3 (Os ) (Albert algebra) the bipartite composite is not canonically defined: since Albert algebras embed in no associative algebra (BGW), there is no associative tensor product, and Bell is genuinely ill-posed on the parent itself (consistent with the BGW obstruction above: no composite, Bell ill-posed). One must therefore choose a composite. If one adopts the maximal tensor product Cmax (the most permissive no-signalling composite — a choice not fixed by the algebra), then Cmax has flat edges hosting extreme states ωPR with ⟨ωPR , B⟩ = 4. Caveat (do not overclaim): this makes the parent PR-box-compatible under Cmax ; it is not a derivation that the parent “is” a PR-box theory with CHSH = 4, because no canonical maximal composite follows from the mere absence of an associative embedding. “Bell undefined on the parent” and “CHSH = 4 on a chosen Cmax ” are compatible only with the composite named explicitly. 6 ===== PDF PAGE 251 / 433 ===== 4.4 Physical narrative The exceptional parent (scale-free, 5D/11D) is PR-box-compatible (under the adopted maximal tensor product Cmax ): it can reach CHSH up to the algebraic √ bound 4 only once Cmax is chosen as the composite. The descent to the quantum boundary 2 2 is the kinematic consequence of KK-mass loading (D874): the overall-scale modulus (radion/dilaton, the sign-S decompactification sector) supplies the mass unit, KK momentum phases require a complex structure, and the reduction projects the physical S-matrix onto the associative C-subimage. The supergravity backbone is the standard de Wit–Nicolai reduction E6(6) /U Sp(8) (42 scalars, 5D) → E7(7) /SU (8) (70 scalars, 4D) via a circle. In one sentence: the exceptional parent is PR-box-compatible under Cmax ; KK-mass √ loading selects the associative C-subimage; the associative envelope is the quantum boundary 2 2. 5 The firewall, stated as conditions We close by making explicit exactly what is conditional and what remains red. Conditions on which Born ✓ and Bell ✓ rest: 1. Channel-count axiom Tr(ηΘ) = 7 (the LSZ-open piece). This is the master switch (Thm. 2.3): it makes the J3 trace form (27, 0), hence Cphys + formally real, hence Bunce– Wright applies. It is not yet derived from first principles (the OS-norm shortcut was refuted in s201; the LSZ route is open). 2. The H1a + H1b selectors (s224 / A1025). Gauge-emergence — equivalent to Born-unconditional (s221) — decomposes into H1a (B−L removal needs an equivariant bosonic defect-lift Morse selector, a new portal) and H1b (SU(2)L selection needs separate structure; the hypercharge direction remains a loaded input). Both are open, so Born stays conditional. 3. The associative-projection lemma Im(FΘ,load ) ⊆ {associator = 0}, needed for the Type I identification of the Bell composite (A1022 Q1), is a leading conjecture, not a derived map. √ Given these, Born = Tr(ρ ◦ P ) is unique (Jordan–Gleason) and ∥B∥ ≤ 2 2 is a theorem on the associative subimage. ⋆ FIREWALL (restated) — Xactual is RED. The measurement problem is NOT solved. This appendix derives the statistical and relational structure of quantum mechanics: the Born weights and the Bell bound. It does not derive the selection of a single actual outcome. The single-outcome quantity Xactual remains unsolved (RED): nothing above tells you which eigenvalue is realized in an individual run. Born and Bell are ✓ conditional on the conditions in the gray box; Xactual is not ✓ under any condition stated here. No public observable moves. 7 ===== PDF PAGE 252 / 433 ===== Result Status Conditional on Born uniqueness (Jordan–Gleason) ✓ conditional Tr(ηΘ) = 7; H1a+H1b Θ Jordan automorphism ✓ (machine precision) — (verified) √ Tsirelson bound 2 2 ✓ conditional associative envelope; assoc. lemma PR-box excluded on subimage ✓ associative envelope Phase-3 from JBW alone × (honest negative) insufficient without associativity Single outcome Xactual RED (unsolved) not addressed 6 Rev28: the Xactual row splits The firewall stated in this appendix — between the conditional Born/Tsirelson structure derived here and the single-outcome (“actuality”) problem — is retained, and it is now theorem-backed from two directions. What changes at Rev28 is that the blanket form of the red row narrows. Novelty label: new specialization proved here. Proposition 6.1 (The actuality no-go, generalized). A probability law does not imply one realized outcome: the symmetric two-outcome state is swap-invariant while no character is, and the argument generalizes to any nontrivial transitive G-set by a fixed-point argument. Demanding a deterministic natural section of the actualization morphism is a category error for a stochastic law — probability theory returns a random element, not a deterministic function of its own law. Novelty label: new specialization proved here. Proposition 6.2 (The record type is derived). For the record process of Appendix T: finite-stage characters are exactly point evaluations, compatibility gives Spec C(ΩT ) ∼ = ΩT , and history, character and compatible boundary record are one object. The morphism 1 ⇝ ΩT exists in the Markov-kernel category. The law is atomless (p∗ = 5φ2 /16), so every point history has measure zero, and a realized history carries 1.168 bits per step: it is extensive record data, not one selector bit. The asymptotic-boundary rescue (tail/Poisson/Martin) is refuted for this chain — bounded harmonics are constants. Xactual — the split (Rev28). Xrecord-law/type : derived. Xrealized-value : not determined by the law — sample/record data, by theorem. The firewall of this appendix stays in force for the specific realized outcome; what narrows is the blanket claim that no actuality object exists. Remaining: one realized record (data), the conditional conserved-Z2 sector identification, and AX-MAP-HISTORY as a candidate only (named, not adopted — it is an atypical, measure-zero orbit, and adopting it changes the question). 7 A positivity obstruction: the bare flow is not unitary in disguise (Rev30, A1482) A recurring hope in this programme has been that a hyperbolic-looking flow might be recast as unitary dynamics by a better choice of polarization. For the bare quartic flow that hope is closed, and the argument is short enough to state completely. 8 ===== PDF PAGE 253 / 433 ===== The lemma. If A commutes with a compatible complex structure J and g = ΩJ > 0, then A is g-skew-adjoint and its spectrum is purely imaginary. The bare-I4 linearization has 54 real hyperbolic modes. Therefore no positive compatible polarization reinterprets the bare I4 flow as unitary dynamics. The conclusion is absolute in the sense that matters here: positivity cannot be bought by choosing a basis — the flow itself must be reduced. Provenance, stated because it is mixed. The lemma was checked in house numerically, against a randomly generated positive g (max |Re λ| < 10−8 , with AT g + gA = 0 confirmed), rather than proved symbolically here; and the count of 54 real modes is imported from the accompanying kernel rather than re-derived. The real form is part of the statement: signatures and mode counts in this family are consequences of the split declaration recorded in Appendix X, and quoting the 54 without it would repeat exactly the defect this era spent its rounds diagnosing. One fact in the same computation is real-form invariant and is worth recording beside the negative — the 26 flat directions are 26 in both real forms, in both the full and the tangent Hessians. Nothing here promotes an observable, moves a tier, or touches the engine. 9 ===== PDF PAGE 254 / 433 ===== Leibniz Quantum Beats Newton Appendix M — The Order-Five Orientation and the Categorical 60-Clock: M = 2h(E8 ), the Parity Grading, and the Galois-Rigid Localization of the Rest-Mass Occupation (Rev33.1; entered Rev21; occupation selector promoted at Rev25) Clock modulus and parity sector: positive/structural, conditional on the vacuum-to-projector bridge. Occupation O = {0, 3, 11, 17}: Derived-Conditional as of Rev25 (three-layer selector derivation, §7; live public falsifier {31, 39}). No public observable moves. Tom O’Sieg August 2026 Abstract This appendix folds the order-five / clock arc (in-house kernels s260–s283, sessions S165–S171, together with the externally-delivered, in-house-verified build-grade assessments A1085/A1086) into the lab notebook. The rest-mass occupation (bound as in Paper 0: occupation of golden mass-ladder rungs, not of atomic orbitals) — the set of integer beat-depths O = {0, 3, 11, 17} at which the charged leptons and the up quark sit on the golden ladder — is the framework’s deepest open selector. The S165–S171 arc recorded here did not derive O (current status: Derived-Conditional since Rev25 — the update at the end of this abstract; one status, stated Rev32.9). What it records is sharply delimited: (i) the rung parity is an exact universal Z2 grading (a theorem, M-independent); (ii) the clock modulus is derived target-independently two ways, both giving M = 60 = 2h(E8 ) — a Chinese-remainder carrier Z60 = Z3 × Z4 × Z5 and, independently, the binary-icosahedral McKay/E8 -Coxeter construction; and (iii) the occupation is localized exactly as a Galois-rigid cyclotomic-frame datum: Stab(Z/60)× (O) = {1}, so every Galois-covariant construction is provably blind to O, and the only target-independent source of the missing data is an anomaly-forced full-frame selector (fenced). Every positive statement here is conditional on the still-open vacuum-to-projector bridge of the golden Peirce construction, and on the banked input T 2 = −I. [LIB2-334] Rev25 update: the occupation selector has since been derived in all three layers (mod-2 Möbius parity, mod-5 torsion walls, mod-7 Fano/Artin adjacency; S224– S225, A1349/A1350-verified), and the occupation is promoted from Input/Open to Derived- Conditional (§7), with the hash-stamped period-two continuation {31, 39} standing as the live public kill condition. Contents 1 Setting and posture 2 2 The clock as a non-invertible-symmetry-reduced system; the parity theorem 2 1 ===== PDF PAGE 255 / 433 ===== 3 Categorification: Rep(D60 ) and the spectral obstruction 2 4 The clock modulus: M = 60 = 2h(E8 ), two independent ways 5 4.1 Chinese-remainder carrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4.2 McKay / E8 -Coxeter carrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 5 The order-five orientation (conditional) 5 6 Galois-rigid localization of the occupation 6 7 The occupation selector: Derived-Conditional (Rev25) 6 8 Status ledger 8 1 Setting and posture On the loaded subimage the charged-lepton and√light-quark rest masses sit, to within a few percent, at integer powers of the golden ratio φ = (1 + 5)/2 relative to the electron: mf nf = logφ , ne = 0, nu ≈ 2.99, nµ ≈ 11.07, nτ ≈ 16.94, me so that the nearest integers form the occupation set O = {0, 3, 11, 17} (kernel s263). The depth nf read off from a measured mass is descriptive; the integer selector that would say which rungs are occupied is the open object. What this appendix claims and does not claim. Claimed (structural, target- independent): the parity grading (§2), the clock modulus M = 60 = 2h(E8 ) and its odd-character spectrum (§4), and the exact Galois localization of O (§6). Conditional: the order-five orientation (§5) rests on the golden-Peirce vacuum-to-projector bridge, which is itself Open; and the central sign T 2 = −I is a banked input, not derived here. Rev25 status of the values: the occupation O = {0, 3, 11, 17} is Derived-Conditional — the selector derivation now exists in all three layers (§7) — conditional on the same standing fences as the rest of this appendix (the vacuum-to-projector bridge; the banked T 2 = −I), and falsifiable through the live {31, 39} continuation. 2 The clock as a non-invertible-symmetry-reduced system; the parity theorem The golden clockwork organizes as a ZM /Z2 non-invertible-symmetry-reduced (NISR) system (kernels s265–s267). Of the 28 inter-rung pairs, a conjugate-pair lemma splits them 15:13 into opposite-parity (all-order protected) and same-parity (soft) classes (s266); the electron (rung 0, even) is all-order protected from the odd rungs µ(11), τ (17). Novelty label: new specialization proved here. 2 ===== PDF PAGE 256 / 433 ===== Theorem 2.1 (Exact universal parity grading, s267). The rung parity is an exact Z2 grading of the clock, independent of the modulus M . The pointed (invertible) subcategory is Cpt = {0, 30} ∼ = Z2 . This promotes the s266 lemma to a theorem. It is a statement about structure (which rungs can mix to all orders), not about occupation. Kernel s269 provides the matching honest negative: parity is falsified as a flavor-mixing (CKM) rule — roughly two-thirds of the empirical mixing weight, including |Vcs | ≈ 0.973 and |Vtb | ≈ 0.999, sits in parity-forbidden entries. The grading governs selector stability, not mixing. 3 Categorification: Rep(D60 ) and the spectral obstruction The clock is categorified as C60 = Rep(D60 ), the Z2 -gauged NISR (kernel s275): it is group- theoretical with F -symbols equal to the D60 6j-symbols, has 33 simple objects with λ (FPdimλ )2 = P 120 = |D60 |, and its module categories lie on the di Francesco–Zuber ADE ladder. The spectral functor r : Irr(M) → Z60 /± images no module category onto O (kernel s276); the obstruction is that 11, 17 are coprime to 60 and the Galois group (Z/60)× places {11, 17} in a single eight-element jm orbit. The twist θ(j, m) = ζ60 is the Green’s-function pairing of the two signature windings (s277), and the modular S, T data are Galois-covariant. These are structured negatives — a covariant construction provably blind to O — not no-go theorems. Remark 3.1 (Three categorical occupation no-goes and the weak-gauge annihilation, s270–s273). A hostile audit added three further categorical attacks, all confirming the occupation is not bulk-forced. Condensation (s270): O = {0, 3, 11, 17} is not fusion-closed and carries non-bosonic components, so it cannot be a condensable algebra in the chiral folded ring. Boundary (s271): enumerating all 12 Lagrangian algebras (gapped boundaries) of the achiral double D(Z60 ) shows O is no boundary subgroup or rung projection. Dynamics (s272): any symmetry-breaking potential built from the bulk Jvac geometry lacks the per-rung tuning needed to target four coprime integers. Separately, the weak sector is not native to the diagonal frame (s273): the Eilenberg–Moore category of the Reynolds monad Pcode isolates the Weyl-singlet diagonal (+, +) frame but annihilates the off-diagonal (−, −) Peirce spaces where the non-commuting SU (2)L generators must reside, so SU (2)L is not extracted from the invariant frame. All four are structured negatives that reinforce the input status of the gauge group (Appendix N) and of the occupation; none touches the continuous-geometry mixing derivation of Paper 3, whose angles do not invoke the parity grading (the cross-cycle adjudication is recorded in Paper 3, Definition-A reconciliation, and the session notes). Remark 3.2 (No canonical order on Irr(Rep(D60 )): a two-lane blind nonexistence, s999–s1000). The freeze-first arc put the ordering question to two lanes blind — neither saw the other’s return, and both were adjudicated against a house table sealed beforehand (D1183/A1490, Gemini; D1184/A1491, Grok). Both proved nonexistence, each inside an explicitly declared class. The Grok class is stated with three clauses: maps Irr(X ) → R built solely from the fusion, F , braiding and annular data, free of continuous and discrete choices beyond D60 itself, and invariant under Aut⊗ (X ); the Gemini class is “canonical categorical invariants, no external imports”. The load-bearing argument is the same in both, and it is an invariance argument, not a counting one: canonical ⇒ Aut⊗ -invariant ⇒ constant on orbits ⇒ non-injective wherever a nontrivial orbit exists. The house supplied the witness the lanes asserted but did not exhibit (s999): φ7 : r 7→ r7 , s 7→ s is an automorphism of D60 — homomorphism and bijection checked exhaustively on all 120 elements — 3 ===== PDF PAGE 257 / 433 ===== inducing a tensor autoequivalence that permutes the 29 two-dimensional simples by j 7→ ±7j mod 60, with a genuine 2-cycle V5 ↔ V25 . One universal lemma of the Grok return is false as written, and the house corrects it here rather than repeating it. The return asserts “the orbits are larger than singletons.” Computing the full induced Aut-action on the 29 two-dimensional labels over all sixteen units k (s1000) gives orbit sizes [1, 1, 1, 2, 2, 2, 4, 4, 4, 8]: V10 , V15 and V20 are fixed by every automorphism. The proof needs only the existence of one nontrivial orbit, so the verdict is unaffected — and the resulting bound is in fact stronger than the lane claimed: with linear-character orbits [1, 1, 2], an Aut⊗ -invariant map takes at most 13 values on 33 simples, so non-injectivity is forced with room to spare. Scope, stated because the scope is the result. This is a nonexistence within the declared classes above; it is not a statement that no order of any kind exists. The subsidiary leg — that FPdim, degenerate on {1, 2} across 33 simples, is the only canonical real-valued invariant — is asserted, not proved, in both returns, and is recorded as such; the invariance leg carries the result alone. Two further steps of the Grok return (that no other modular-data invariant is injective without additional choices, and that the relevant centre eigenvalues are circle-valued or highly degenerate) are likewise asserted. Both returns were text-only, with no executable shipped; every number quoted here is house-computed. Adjudicator disclosure: the same house scorer adjudicated both lanes in one session with the Gemini return already in context, so blindness holds between the lanes while house-side independence is procedural rather than epistemic. Finally, the companion rigidity result of the same era — that 1 ⊂ Y ⊗ Y ∗ in any rigid fusion category makes reachability indiscrete — is not folded here: it was corroborated by neither lane, and remains a single-lane-plus-house result. What it costs the occupation programme. Together with the structured negatives above, this closes the categorical route to an occupation ordering: fusion supplies composition, braiding supplies exchange, and Fock occupation consumes a labelling rather than manufacturing one. None of the three supplies an energy order. The occupation’s input status (Appendix N) is reinforced, and nothing here promotes or moves any observable. Interpretive corollary, printed beside the fence only (Rev32.3, PI wording S300): the ⊗-layers are order-free by theorem; apparent motion is re-presentation along exchange isomorphisms; dynamics is priced extra structure. Remark 3.3 (What does carry an order, and what does not (Rev30, A1484/A1485)). The categorical no-go above is one half of the ordering question. The other half is positive, and it cost the house a sealed prediction to establish, so it is recorded exactly. An intrinsic partial order does exist. The weight-lattice dominance order on the weights of the 7 of G2 = Aut(Os ) is a genuine intrinsic partial order: 24 comparable ordered pairs and 9 incomparable pairs, so it is genuinely partial and not total, and it is built from root data alone — no n, no ℓ, no n + ℓ, no Z and no energy enters its construction. Read this correctly, because the sealed text and the physics point different ways. The house had sealed the claim that no filling-order carrier exists in the framework, and against the refutation condition as drafted — which asked for a total or partial order — that claim lost. It is recorded as a loss. But the object above is not a filling order, and nothing here says the framework contains one: a dominance order on weights is not a rule for traversing states. The loss is a drafting loss, and reading it as “the framework supplies a filling order” inverts it. Scope: the check used the 7 of G2 — one representation of one group — and is not a survey of the corpus. The corpus has still not been swept for ordering carriers by anyone; that remains an open task. 4 ===== PDF PAGE 258 / 433 ===== Fock occupation is not a source of order either. Nothing in the construction of the Fock space or of its number operators manufactures the labelling of the index set: that labelling is an external input. Fock occupation does not supply an ordering — it consumes one. Taken with the categorical result above, the position is that fusion supplies composition, braiding supplies exchange, and occupation consumes a labelling; none of the three supplies an energy order, and the missing object is a Hamiltonian or transfer law rather than a categorical invariant. A naming debt, acknowledged. Calling this set an “occupation” invites a false parity with physical subshell filling, and that is a debt against the whole corpus rather than a one-off. The binding fence for the term is given at its first use (Paper 0) and again in §7 of this appendix: occupation of golden mass-ladder rungs, not of atomic orbitals nor of states by particles. Nothing in this remark promotes an observable, moves a tier, or touches the engine. 4 The clock modulus: M = 60 = 2h(E8 ), two independent ways 4.1 Chinese-remainder carrier The central sign T 2 = −I forces a Z4 factor, the qutrit a Z3 , and the order-five orientation a Z5 ; on the carrier Z60 = Z3 × Z4 × Z5 the frozen, target-independent grading is U60 = U32 ⊗ U4−1 ⊗ U5−2 , primitive because 2 1 2 40 − 15 − 24 1 − − = = 3 4 5 60 60 is a unit mod 60, so U60 is a bijection onto Z60 (kernel s281). 4.2 McKay / E8 -Coxeter carrier Independently (assessment A1086, in-house-verified), the binary icosahedral group 2A5 ⊂ SU (2) has, under tensoring by its defining spinor, the affine E e8 McKay graph; deleting the tensor-unit node gives finite E8 , whose Coxeter transformation has order h(E8 ) = 30 with characteristic polynomial Φ30 . Its Clifford (Pin) lift QE8 satisfies Q30 E8 = I; paired with the banked operator T (T = −I) on 2 the common carrier ΣE8 ⊗ LT , U60 = QE8 ⊗ T, 30 U60 = −I, 60 U60 = I, with minimal polynomial x30 + 1 and characteristic polynomial (x30 + 1)(x2 + 1) = Φ24 Φ12 Φ20 Φ60 . The spectrum is exactly the 30 odd 60th roots of unity (labels 15, 45 doubled). Hence M = 60 = 2h(E8 ) on the McKay–Clifford carrier. The central product (2A5 × C4 )/⟨z = t2 ⟩ (order 240) contains no element of order 60, so the clock genuinely requires the McKay/Coxeter layer and not the central product alone. Remark 4.1. The two derivations are independent and consistent: the CRT route fuses the 3- and 5-structure as separate factors, whereas the Coxeter route fuses them inside a single E8 Coxeter element, with T supplying the doubling 30 → 60. This is the strongest internal convergence signal of the arc. The result is target-independent for the modulus and the odd-parity sector only; it does not touch occupation, and the even label 0 is absent from the spin clock (the tensor unit is a candidate origin, not a derived one). 5 ===== PDF PAGE 259 / 433 ===== 5 The order-five orientation (conditional) The native clock symmetries are all affine, contained in AGL(1, 60) (order 960, solvable); the Galois-breaking order-five element is the icosahedral A5 (simple, |A5 | = 60, with no element of order 60), whose binary cover is 2I = SL(2, 5) — the E8 McKay node (kernel s280). The order-five orientation is therefore an irreducible input relative to the affine clock. It is realized on the exceptional Jordan algebra (assessment A1085, in-house-verified two ways): golden Peirce reflections H0 , H1 on J3 (Os ) with v0 = (φ, 1, φ−1 ) give U5 = H0 H1 of exact order 5 and trace φ, with ⟨P, H0 ⟩ closing to an icosahedral group whose determinant-+1 subgroup is A5 ; the centralizer of U5 in A5 is exactly C5 (the C3 –C5 wall). The canonical lift 2I = SL(2, 5) has the exact order census {1:1, 2:1, 3:20, 4:30, 5:24, 6:20, 10:24}, a unique involution −I = T 2 , and every order-four element squaring to −I (kernel s281). Fences (carried verbatim). The order-five realization is conditional on the vacuum-to- projector bridge Jvac 7→ v0 7→ p0 of the golden-Peirce construction, which is a naturalness claim, not proven canonical. [LIB2-029] The central sign T 2 = −I (equivalently the 2I centre, the Dic30 relation) is a banked input: “the central sign is forced” holds given one is in 2I; choosing 2I is the input. 6 Galois-rigid localization of the occupation The decisive structural result narrows the occupation to a single, fully rigid cyclotomic frame. Novelty label: new specialization proved here. Theorem 6.1 (Maximal Galois irregularity, s283). Stab(Z/60)× (O) = {1} exactly. Consequently O requires the full cyclotomic frame Q(ζ60 ): it is fixed only after three further Z2 bits beyond the golden field are pinned (the ζ4 spinor, ζ3 qutrit, and ζ5 -square icosahedral signs). A blind golden/icosahedral √ index can be Galois-equivariant only under H = {u ≡ ±1 mod 5} ∼ = Z32 (order 8, the 5-fixer); but the H-orbits of 11 and 17 are disjoint, so O is not H-invariant and is unreachable by any blind golden index (kernel s282; this sharpens the earlier √ full-Galois √ statement of s279). The Galois closure of {3, 11, 17} has 24, 20, 10 labels over Q, Q( 5), Q(i, 5) respectively; no affine map ℓ 7→ aℓ + b cycles 3 → 11 → 17; and the targets are simple eigenvalues, so degeneracy supplies no rule (assessment A1086). Every Galois-covariant level is thus sealed. Remark 6.2 (Erratum √ to s282). Kernel √ s283 corrects the s282 wording: the residual gap is three Z2 bits, not a “single 5 sign” (the 5 sign is one of the three). 6 ===== PDF PAGE 260 / 433 ===== 7 The occupation selector: Derived-Conditional (Rev25) Superseded note (Rev25). Through Rev24 this section carried the red “missing theorem” box: construct a target-independent selector whose support is exactly O = {0, 3, 11, 17} — with the occupation values held as strictly Input/Open. [LIB2-029] That demand has now been met at derivation (not theorem-chain- complete) grade by the S224–S225 arc, and the occupation is promoted to Derived-Conditional. [LIB2-029] The original no-go results of this appendix stand unchanged — they proved a Galois-covariant construction cannot do it, and the selector that succeeded is indeed not Galois-covariant: it fixes the frame by anomaly-forced/monodromy data, exactly the type §6 said was the only one available. The three-layer derivation (in-house kernels s696-class, s704; hostile-lane verified A1349/A1350): 1. Mod 2 — Möbius parity. The rung parity is the exact universal Z2 grading of Theorem 2.1, realized as the Möbius (orientation) parity of the winding substrate (s696-class theorem). 2. Mod 5 — torsion walls. The order-five orientation’s torsion structure walls off the Z5 residues, fixing the ζ5 -sector frame data. 3. Mod 7 — Fano/Artin adjacency, now derived. The open spinor boundary is the support of one primitive Artin generator; σ1 7→ TArtin = ( 10 11 ) is the canonical B3 → P SL(2, Z) map (not a choice); mod 7, TArtin has order 7 (it is not the central-sign operator T = Tclock , Tclock 2 = −I, used elsewhere in this appendix — two operators, named apart at Rev32.9, A1637) and acts as z 7→ z+1 on F7 , forcing the boundary pair {0, 1} to cyclic distance 1. The Fano line completes as {0, 1, 3} = {0, s(0), s3 (0)} — the boundary pair plus the cube point — and the vector leg literally is T 3 , converging with the banked TS3 monodromy. Exclusions are typed, not assumed: d = 2 is Möbius-parity even (no open half-twist); d = 3 is the vector leg (a spinor endpoint there is a typing collapse). Verified by enumeration (A1350, kernel s711). The kill condition (live). The derived selector’s period-two continuation is {31, 39}, hash-stamped (7fd5775e...) before verification rounds and confirmed by verbatim re- run (A1349). [LIB2-029] It is the standing public falsifier: an established occupied rung outside the selector’s support — or a demonstrated failure of {31, 39} where the selector’s continuation is testable — kills the derivation and reverts the occupation to Input status. Conditional on: the vacuum-to-projector bridge (§5 fences, carried verbatim) and the banked T 2 = −I. [LIB2-029] Nothing else in the suite consumes the promotion; no public observable moves. Historical record (pre-Rev25): two fenced routes were held pointing at the then-open gap from opposite sides (opt-in; the anomaly-forced type of the first is the type the successful selector in fact has): • the G2 (2) extremal-CFT anomaly — G2 (2) = Aut(Os ) and J3 (Os ) is the carrier of the order-five orientation, so the gauge axis and the order-five axis are the same substrate; a central-charge- jm fixing CFT fixes the modular T -data (θ = ζ60 has trivial natural covariance, hence is of the right type); and 7 ===== PDF PAGE 261 / 433 ===== • a noncentral physical index — a Peirce/generation-coupled rank-three index with one zero-mode line in each of the three native blocks (3, 11, 17), under a tight blind contract (freeze before unblinding; rank derived, not stipulated; the same projector must govern the physical mass operator). √ Remark 7.1 (the oriented-Schur torsion κ = 3 and the all-orders firewall (s292–s294, S175; op- t-in)). The noncentral-index route above has a concrete, verified intermediate result that nonetheless leaves the occupation open. The orientation-odd entangling torsion the hunt needs is canoni- cal: eliminating the terminal complement of the paired-orientation parent vertex√by the √ oriented Schur √ complement forces Keff = i gR gJ β (RJ − JR) = κ K−− with spec = {− 3, 0, + 3}, i.e. κ = 3 — nonzero generically whenever both terminal bridges are present (the democratic-singlet gap cancels; coefficient theorem). But it lives detached from the current commutative vertex algebra (τz ⊗ Keff ⊥ C[τx ] ⊗ C[Dφ ]; cross-bridges ∼ 10−15 ), so installing it is new physics. The accompanying action-match firewall is an all-orders no-go gate with a sharp binary falsifier: if the physical R- and J-bridges land in orthogonal terminal sectors the interference vanishes (κ = 0) and the mechanism is dead. The frozen-index test — are the three odd sectors of Dent exactly {3, 11, 17}? — remains untested. This remark promotes nothing by itself; the occupation’s Rev25 status (Derived-Conditional, §7) rests on the three-layer selector derivation, not on this route (banked A1090). Remark 7.2 (independent corroboration: the adjacent-integer support floor (S245, s901/D1121)). The nonzero occupation {3, 11, 17} is re-derived, independently of the three-layer selector, by a clean minimax argument. Reading the continuous golden depths off the physical ratios gives (2.996, 11.080, 16.945); rounding each to its nearest integer and scoring a candidate support by its worst-case depth mismatch, {3, 11, 17} is the global minimax — objective floor 0.0795 (set by the muon offset nµ − 11), with the nearest competitor (3, 12, 17) far above at 0.588. This is corroboration, not a new promotion: it strengthens the Derived-Conditional support of §7 from a second, convention-light direction. It is fully consistent with the exact additive-χ completion of the mass ratios on this support being [closed-negative] at the ratio tier (Appendix O, and the closed-selector ledger of Appendix S): the integer support skeleton is robust; only the exact numerical completion on it fails. Nothing here moves the {31, 39} falsifier or any public observable. 8 ===== PDF PAGE 262 / 433 ===== 8 Status ledger Result Status Conditional on Parity Z2 universal grading (s267) ✓ theorem — Cpt = {0, 30} = Z2 ✓ — Rep(D60 ), 33 simples, FPdim2 = 120 — P ✓ M = 60 = 2h(E8 ) (two ways) ✓ target-indep. modulus + odd sector only Order-five orientation on J3 (Os ) ✓ conditional vacuum→projector bridge T 2 = −I central sign input (banked) choice of 2I Stab(Z/60)× (O) = {1} ✓ — Parity as CKM mixing rule (s269) × (honest negative) falsified Condensation/boundary/dynamics no-go × structured nega- O not bulk-forced (s270–s272) tive SU (2)L from Pcode monad (s273) × (annihilated) off-diagonal Peirce killed Occupation O = {0, 3, 11, 17} Derived- 3-layer selector (§7); {31, 39} Conditional live falsifier (Rev25) Adjacent-integer support floor {3, 11, 17} ✓ corroboration minimax floor 0.0795; inde- (s901) (S245) pendent of 3-layer selector G2 (2) / noncentral-index selector FENCED (opt- not run √ in) Oriented-Schur torsion κ = 3 (s292– FENCED (opt- detached vertex; firewall pend- s294) in) ing The clock modulus and parity are structural results on the categorical carrier; the occupation is Derived-Conditional as of Rev25 (three-layer selector, {31, 39} live falsifier), with the fences of §5 carried verbatim. The engine and all public observables are unchanged by this appendix. 9 ===== PDF PAGE 263 / 433 ===== Leibniz Quantum Beats Newton Appendix N — Selector Fences: Why the Fine-Structure Constant and the Compact Colour Form Are Inputs, Not Outputs (Rev33.1) Two honest negatives. α stays fenced (now five-legged); the compact colour real-form selector is named as input. No public observable moves. Tom O’Sieg August 2026 Abstract This appendix folds two settled-negative arcs into the lab notebook. Part N.A (kernels s255–s262, sessions S165–S166) records the attempt to un-fence the fine-structure constant α and its fivefold refutation: the Wyler bounded-domain route is numerically falsified; the golden field makes any near-137 target numerically dense, so closeness carries no information; “perfect-imperfection” (noble/KAM) stabilization is real mathematics that selects nothing; and an energy/dissipative principle locks to rationals, with the golden value unpinned only below the Aubry threshold (it pins through Kc ; s261); and the dynamical-scale-via-vacuum-decay route is closed (D905), since the ω = π/4 golden vacuum is absolutely stable (VN ≥ 0), so no nontrivial decay bounce exists (the trivial vacuum has zero action relative to itself). The conclusion is structural: a fixed geometric number for a running coupling is a category error absent a derived privileged scale, and the rest-mass integer selector is a discrete topological (winding) datum, not a continuous offset (under the Leg 4 deformation-class premise, Part N.A). Part N.B (kernels s253, s255, s257, sessions S164–S165) names the real-form / colour selector: the compact su(3) colour algebra cannot embed in the available compact part (maximal compact dimension 6 < 8), so the choice of real form is an input. Both results are honest negatives; α remains an input modulus and nothing is promoted. Contents 1 Part N.A — The fine-structure constant is fenced 1 1.1 Leg 1 — the Wyler route is falsified (s256) . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Leg 2 — golden numerology density (s258, s259) . . . . . . . . . . . . . . . . . . . . 2 1.3 Leg 3 — perfect-imperfection is robustness, not selection (s260, s261) . . . . . . . . 2 1.4 Leg 4 — the selector is topological (s262) . . . . . . . . . . . . . . . . . . . . . . . . 2 1.5 Leg 5 — the dynamical-scale route is closed (s175, D905/A1088/A1089) . . . . . . . 2 2 Part N.B — The compact colour real-form is an input 3 1 ===== PDF PAGE 264 / 433 ===== 3 Part N.C — The electroweak framing is loaded: the permanence theorem (S231– S232) 4 4 Part N.D — The dual ruling: what was killed, and what stayed open (Rev30, A1500) 5 5 Status ledger 6 6 Part N.E — Two fences added at Rev32.6 (A1566, A1567, A1568) 7 1 Part N.A — The fine-structure constant is fenced The framework declares α (the U (1)EM fibre scale) an input. [LIB2-236, P-008] Session S165 tested whether it could instead be derived. It cannot. Five independent legs establish the fence. 1.1 Leg 1 — the Wyler route is falsified (s256) The Wyler bounded-domain normalization, computed exactly, gives α−1 = 137.0360824 against the CODATA value 137.035999177 — a discrepancy of thousands of standard deviations. The associated multi-tuple template family is excluded en masse. The bounded-domain programme is retired as a derivation of α. [LIB2-236] 1.2 Leg 2 — golden numerology density (s258, s259) Scale-free ratios built from φ populate the neighbourhood of 137 densely; under a controlled density discipline (a Beatty/Wythoff admissibility count), a near-137 hit is expected by chance and therefore carries no selective information. [LIB2-092] The counts are the kernels’: a pool of 21 golden and substrate invariants, monomials up to complexity 6, 503 distinct values near 137 (s259; s258: 30 up to complexity 5 without the golden ray); the 1% band is reached at complexity 3 and the 1.5 ppm band is not reached at all (Rev32.9, A1647 U2). The “it is close” observation is a numerology-class artifact, not evidence. [LIB2-092] 1.3 Leg 3 — perfect-imperfection is robustness, not selection (s260, s261) The golden value is the most-irrational (noble) number and hence the KAM-most-stable torus, with critical coupling Kc ≈ 0.9716 in the standard-map sense (s260). But in a Frenkel–Kontorova / Peierls–Nabarro setting an energy/dissipative principle mode-locks to rationals; the golden value slides freely below the Aubry threshold Kc = 0.971635 and pins at and through it (s261: golden_unpinned_below_Kc, golden_pins_through_Kc; Rev32.9, A1647 F03 — through Rev32.8 this said “the unpinned, freely sliding extreme” without the regime). [LIB2-093] Noble stabilization is a robustness fact, not a mechanism that fixes a coupling. [LIB2-093] 2 ===== PDF PAGE 265 / 433 ===== 1.4 Leg 4 — the selector is topological (s262) The depth ladder 1 Rd = φ2d − 1 is a resolvent / Green’s function. The winding datum, named (Rev32.10, A1711-F05): the registered winding map is the occupation map of Appendix M, n : {ladder configurations} → Z, which integer rung of Rd is occupied; it is the selector in Paper 0’s bound sense and nothing else is meant by “winding” here. Its invariance is a premise about the deformation class, not a consequence of discreteness alone: the admissible class is continuous one-parameter shifts of the rung positions that preserve the ordering of the rung energies, and n is locally constant on that class. Where two rung energies cross the minimising label switches (the two-level control E0 (a) = a2 , E1 (a) = (1 − a)2 switches at a = 12 ), so a crossing is outside the admissible class by definition, and the statement below is conditional on that exclusion. A continuous spurion (the Reynolds offset) — the deformation class here is continuous one-parameter shifts of the rung positions — can shift positions on the ladder but cannot leap its discrete integer steps; therefore the integer selector (the occupation of Appendix M, in Paper 0’s bound sense: which integer rungs are occupied; Rev32.9, A1647 U1) is a discrete topological (winding) datum, not a continuous correction. [LIB2-094] 1.5 Leg 5 — the dynamical-scale route is closed (s175, D905/A1088/A1089) A final route would generate a privileged scale dynamically — a Coleman vacuum-decay rate ∼ e−Sbounce off the golden vacuum — and use it to pin α. It fails on three counts, the first decisive. (i) The golden vacuum is absolutely stable: the normalized radial potential VN (r) = Λ4 (e−2r +2er −3) has a unique critical point at r = 0 with VN (0) = 0, VN′′ (0) = 6Λ4 > 0 and VN (r) > 0 for all r ≠ 0, so there is nowhere lower to tunnel to and the Euclidean equation of motion has only the trivial solution: no nontrivial decay bounce exists in this radial problem (positive canonical kinetic term; a regular finite-action solution approaching r = 0; vanishing boundary term), and the trivial vacuum has zero action relative to itself (Rev32.9, A1647 F04: through Rev32.8 this read “Sbounce = 0 exactly”, which is the trivial vacuum’s own action, not a tunnelling exponent). [LIB2-095] The “bounce” of Appendix I is classical geodesic reflection off the Det = 0 wall (a metric-completeness fact), not a Coleman tunneling event; with no decay rate there is no dynamical scale. [LIB2-095] (ii) Even granting metastability, the O(4) bounce action SE ∝ (Mkin /Λpot )4 is Λ-dependent and requires a second input scale — never geometry alone. (iii) The bounce lives on the radial cone modulus, with no banked term linking it to the U (1)EM fibre kinetic coefficient α; a scale there would set a dimensionful quantity, not the dimensionless coupling. [LIB2-095] Convergent double refutation (Grok A1088 + Gemini A1089), leg (i) verified in-house (VN minimum = 0 at r = 0, no VN < 0). The companion four-tick reading — T accruing a −1 on reflection at the Det = 0 wall (T 2 = −1, T 4 = +1) — is validated by both reviewers as sound classical holonomy of the clock turnover, but it too generates no coupling. Ledger. α is an input modulus, fenced on five independent legs (Wyler-falsified; golden- dense; robustness-not-selection; running-coupling category error; dynamical-scale route closed). [P-008] It runs, so a fixed geometric value is ill-posed without a derived privileged scale — none exists in the banked framework. φ remains the single golden axiom (AX1); α is not derived from it. 3 ===== PDF PAGE 266 / 433 ===== 2 Part N.B — The compact colour real-form is an input The colour selector concerns which real form of the relevant Lie algebra carries the physical gauge content. Novelty label: new specialization proved here. Proposition 2.1 (Compact-colour obstruction, s253). The compact colour algebra su(3) (dimension 8) cannot embed in the compact part available to the loaded subimage, whose maximal compact dimension is 6 < 8. [LIB2-096, LIB2-297] Hence the physical colour content is not forced by a compactness criterion; the choice of real form is an input. [LIB2-096] The constructive side is recorded for completeness: the compact g2 (−14) is built from the split g2 (2) by an explicit Cartan involution (s255), and the conformal algebra so(4, 2) ∼ = su(2, 2) together with g2 (2) is located inside so(4, 4) (s257). These fix the ambient real-form geometry but do not remove the selector: the real form is chosen, not derived. [LIB2-297] Ledger. The real-form / colour selector is an input, on the same footing as the gauge-group assignment of the main text. Proposition 2.1 makes the obstruction explicit (a dimension count), so the input is named rather than hidden. No public observable moves. The carrier truncation is also chosen: a pre-registered null (s709, S225) The selector above is stated at the level of the real form. There is a second, finer place the same question can be asked, and it was asked under a pre-registered test whose result is recorded here for the first time. The loaded colour/ΩQ provenance is carried as conditionally selected, the condition being that the 56-dial — clock-weight multiplicities (1, 27, 27, 1) — may legitimately be truncated to one state per weight, loaded ∼ EQ = E−3 ⊕ E−1 ⊕ E+1 ⊕ E+3 . s709 tested whether that truncation is extracted or merely chosen, using the three measured translator velocities Rk (k = 3, 4, 5) acting on the weight-graded 56. Its four gates and their tiers were frozen before the run. Gate Result Reading C1 nonzero pass Every ak , bk has norm > 10−10 (norms 0.0283, 0.1216, 0.1409): the translators do connect the end states. C2 cross-translator agreement null sv2 /sv1 = 0.468 against pre-registered tiers forced < 10−10 and strong < 10−2 ; pairwise | cos | of 0.73, 0.62, 0.52. The three translators pick different lines in the 27. C3 ladder closure forced Top cos with e+3 equal to 1.0 at k = 3, 4, 5. C4 up/down consistency null | cos(bk , v2,k )| = 0.042, 0.00045, 0.0022 at k = 3, 4, 5: the down-ladder does not meet the up-ladder (row added Rev32.9, A1652; the frozen fourth gate was not printed through Rev32.8). 4 ===== PDF PAGE 267 / 433 ===== Verdict: S709_CARRIER_NULL_TRANSLATORS_INSUFFICIENT. What this adds to the fence. C2 was the gate that mattered: had the three translators picked the same line, one-state-per-weight would have been extracted rather than chosen, and the conditional under the loaded colour provenance would have discharged itself. It returned null, and not marginally — 0.468 against a 10−2 threshold. So the truncation is a choice, exactly as the real form above is a choice, and the conditional selection remains conditional. [LIB2-097] Note also what did not fail: C1 and C3 passed, so this is a specific negative about cross-translator agreement, not a general failure of the construction. [LIB2-097] No public observable moves either way. 3 Part N.C — The electroweak framing is loaded: the permanence theorem (S231–S232) √ The electroweak scale registration vEW / σ = φ105/8 (Appendix O, the σ-bridge section) rests on a framed-depth package DH = (PH , Nann , qH ) — a Higgs line, a winding level, and a quadratic refinement. The natural hope, that this package is selected frame-free, is closed negative at theorem grade, on the same footing as the α fence of Part N.A: Novelty label: new specialization proved here. Proposition 3.1 (Electroweak permanence; S231/S232). No frame-free package DH exists on the bare Freudenthal system F = R⊕R⊕J3 (Os )⊕J3 (Os ) under its automorphism group G0 = E7(7) . [LIB2- 098] Specifically: (i) load-ew-line — the 56 is real-irreducible, so no rank-one invariant projector (Higgs line) exists (Appendix F); (ii) load-ew-index — the level Nann = 13 is homed (= F7 = a8 ) but has no frame-free basepoint to attach to; (iii) load-ew-frame — the quadratic refinement is a torsor, not a function of the form; the two valid refinements (β = 1 and β = 7, q(1) = 1/4 vs 3/4) have equal |γ| = 1 and are exchanged by the metaplectic lift-sign q 7→ −q, so no intrinsic selector fixes β = 1. [LIB2-098] Ledger. The electroweak framing is a loaded input, on the same footing as the real- √ form/colour and α selectors. [LIB2-098] Consequently vEW / σ = φ105/8 is the unique visible framed-depth registration relative to the marked compact-positive carrier (Paper 4), not a frame-free scale theorem — the honest tier at which Appendix O foregrounds it. No public observable moves. 4 Part N.D — The dual ruling: what was killed, and what stayed open (Rev30, A1500) The final round of the freeze-first arc returned a dual ruling, and its two halves are routinely misquoted in opposite directions. This section states all four statuses together, because any one of them quoted alone is a false claim. 5 ===== PDF PAGE 268 / 433 ===== The four statuses, and they must travel together. (1) Row 1 is KILLED — but only as a sole parent datum, and only inside the declared class. [LIB2-099] The class was committed, not gestured at: the frozen-V56 category BeatFlipfr 56 , named before the target artifacts were loaded. Inside it, the orphan doublet is the carrier-level compact-reduction shadow of σsel — an exact frozen-carrier shadow and nothing more — not the missing external boundary datum. (2) GLOBAL Row 1 remains open, for genuinely external categories that do not factor through the orphan-image class. The kill does not reach them. (3) The conditional Row-2 subgate PASSES. (4) The ORIGINAL full Plan-v2 Row 2 remains open. The finite subgate is not the row and must never be quoted as if it were. Why Row 1 fails in its class. The vertical Ksel action moves L∂ with rank exactly ten, so the assignment gKsel 7→ gL∂ is not well-defined: changing the representative changes the alleged boundary image in every direction of a ten-dimensional orbit. The orbit dimension reached is 43 where 53 is required, leaving a residual fibre of 10, and the stabilizer arrives at dimension 20 where the dynamical requirement is 10. What the subgate did and did not establish. Four checks passed: an exact homogeneous reduction to the reduced action with 32 × 32 Hessian residual 0.0; the full ten-generator first-class constraint algebra, all 100 commutators and all 1000 Jacobi triples at residual 0.0; a gauge-reduced principal symbol spec σphys = 30(62) at k 2 = 30, with a wrong-sign control returning −30(32) ; and a positive finite transfer truncation (81 × 81, λmin = 4.27 × 10−5 > 0) together with a four-time reflection Gram (λmin = 4.55 × 10−5 > 0), against a time-reversed control at λmin = −1.0000039. None of that is a continuum result. No nontrivial stable renormalized four-dimensional continuum trajectory was constructed, which is precisely why status (4) is open. Two things that must never be printed as results. First, the reduced-sector gaps 0.3235, 0.3290, 0.3334 at a = 0.5, 0.4, 0.3 are a temporal-regulator sanity sequence, not a continuum trajectory. Throughout, a is typed as a regulator and not as a measured scale; reading the sequence as an extrapolation would manufacture a continuum limit that was not constructed, and both the lane and the house say so explicitly. Second — and this is the more instructive one — there was a control that succeeded, and it was refused. The object “orphan + L∂ ” does land on orbit dimension 53 with the required stabilizer and residual fibre 0: numerically perfect. It is also exactly the forbidden circular repackaging, because it reaches the target by importing the very datum it was supposed to supply. A round reporting only that success would have read as a solved row. It was typed as circular and not banked. The gate that separates the two outcomes is the non-circularity pre-test, and it is the reason the dispatch was built the way it was. The adjudication was reproduced across two hosts — 20/20 gates, 20/20 should-fire controls, semantic hash exact, with 3 of 398 result leaves differing and all three being environment stamps. Nothing was promoted. No observable, no tier and no comparator moved; α remains fenced (Part N.A) and the engine is unchanged. 6 ===== PDF PAGE 269 / 433 ===== 5 Status ledger Result Status Note α from Wyler domain (s256) × falsified 137.0361 vs 137.036, ≫ σ golden near-137 density (s258/9) × numerology closeness uninformative noble/KAM stabilization (s260/1) robustness only not a selector selector is topological (s262) ✓ structural winding, not offset dynamical scale via decay (s175/D905) × closed no nontrivial bounce; stable vacuum α overall FENCED (input) five-legged; runs compact su(3) embedding (s253) × obstructed max compact dim 6 < 8 real-form / colour selector input (named) chosen, not derived Both arcs are honest negatives scoped to the banked framework. They sharpen the input ledger (Appendix X) by naming why α and the colour real form are inputs, and they tie the rest-mass selector to the topological occupation of Appendix M. Nothing is promoted; no public observable moves. 6 Part N.E — Two fences added at Rev32.6 (A1566, A1567, A1568) The θ23 octant registration: fence (A1566/D1244; s501 retyped, s1156). s501 is a structural candidate/interface: the trefoil chirality supplies a possible odd Z2 carrier and the even sector is blind to the octant sign, but the chirality torsor {−1, +1} and the octant torsor {7/16, 9/16} admit exactly two Z2 -equivariant identifications (s1156: −1 7→ 7/16 and its mirror), and no registered action term, operator, intertwiner or readout map excludes the countermap. [LIB2-100, LIB2-241, LIB2-308] The internal identity sin2 θmismatch = 7/16 is theorem-grade; its attachment to the physical lower octant is Loaded-correspondence (Paper 3). [LIB2-100] The Rev32.5 claims “selector built”, “locked to εQ ” and “free bits 2 → 1” are retired. [LIB2-241] Reopen condition: a registered loaded odd object Ω23 with a theory-fixed sign in a coupling such as Slink = −κ χodd cos 2θ23 , plus a pre-registered mirror control proving the countermap impossible without using θ23 , δ, or observed charge-sign data. [LIB2-308] 7 ===== PDF PAGE 270 / 433 ===== Complex-ratio ladder: fence (A1567/D1246; A1568/D1248, s1159 v2). A1567 finds no registered phase-bearing ladder ratio: a census of 29 ladder/family rows against 39 phase/rotation rows across the shipped components registers no q = ϕ−1 eiα ; the one commut- ing contact [Y, Lvac ] = 0 (Appendix I, s105b) does not tie the U (1) angle to a rung. [LIB2-101] The missing object is typed: QL = SL eαJL with (JL |E )2 = −IE and [JL |E , SL |E ] = 0 on the ladder two-plane E, with E and its orientation named by the registering datum (as the antecedent D1247 states the target “on the ladder 2-plane”), and one source law fixing both the real rung step and a nonzero α; installing JL , choosing the angle, or selecting after comparison does not qualify. [LIB2-101, LIB2-309] The obstruction is now computed, not typed. For the two computed carriers, C(Lvac ) ⊂ e6(6) has dimension/signature/centre 30, (18, 12, 0), 2, while C(H) ⊂ e7(7) has 79, (43, 36, 0), 1. [LIB2-102] Both have positive- dimensional compact subalgebras, but Y , LK , H are noncompact and the Appendix-H K is not in C(H). [LIB2-103] Appendix G separately supplies a commuting compact Z2 , but leaves its angle independent of DKK . [LIB2-103] No one-law (JL , α) is selected; any attachment remains Loaded-correspondence. [LIB2-102] Loss condition (A1567), met at Rev32.5: no theory-selected compact U (1) in any ladder-scaling centralizer, or only ones whose angles stay free. Reopen: a registered compact datum commuting with the scaling, with its angle fixed without comparison. [LIB2-101, LIB2-309] The census is invariant under SU (8) → SU ∗ (8) (s1160; Appendix G). What a phase-bearing defect must deliver: the CP-cycle criterion (Rev32.9; A1576 check 3, house re-run to 10−12 and again at S341). After canonical kinetic normalization, with Hu = diag(a1 , a2 , a3 ), Tr[Hu , Hd ]3 = 6i (a1 − a2 )(a2 − a3 )(a3 − a1 ) Im (Hd )12 (Hd )23 (Hd )31 ,   the Jarlskog commutator invariant. A single off-diagonal link, or a two-link tree, gives mixing but no three-family Dirac CP violation (the invariant vanishes); a non-zero value needs a closed three-link product with a non-removable phase. Hence [Yu , Yd ] ̸= 0 is necessary only: a candidate defect for the Ω23 charter must export a registered, gauge-compatible response into Tr C 3 ̸= 0, and a half-period or compact rotation frequency is not a CP phase until it is so exported. Two limits of the criterion (S367; house kernel s1205): the invariant is proportional to the Jarlskog J ∝ s23 c23 , which is symmetric under θ23 → 90◦ − θ23 , so the criterion cannot decide the octant; and on today’s assembled up/down pair it evaluates to a multiple of sin δCKM of the formula-installed phase — a read-back, not an export — so the test applies to a future registered defect, not to today’s pair. 8 ===== PDF PAGE 271 / 433 ===== Why the target must be local: a parity lemma (Rev32.9; A1576 check 2, A1652; house kernel s1176). A real linear J with J 2 = −I and [J, S] = 0 exists only if every real eigenspace E of S is even-dimensional, since J preserves E and (det J|E )2 = det(−IE ) = (−1)dim E ≥ 0. The registered scaling spectra are Lvac on the 27 with multiplicities (1, 1, 1, 8, 8, 8) — three odd blocks — and H on the 56, from diag(1, 1, 1, 1, 1, 1, −3, −3)/2, with eigenvalues {−3, −1, +1, +3} and multiplicities (1, 27, 27, 1) — all four odd (s1176, 6/6 gates, both lemma controls). Hence no everywhere-defined JL commutes with Lvac on the full 27 or with H on the full 56: this is stronger than “the scaling centralizers contain compact subalgebras but do not select the phase law”. The phase target must be an even-dimensional invariant subcarrier — at most 24 of 27 and 52 of 56; these are parity bounds and select nothing — and the Hϕ spectrum (12, 32, 12) is exempt. No tier moves. 9 ===== PDF PAGE 272 / 433 ===== Leibniz Quantum Beats Newton Appendix O — The Charged-Lepton Rest Mass as a White/Black Self-Energy Residual: a Built-but-Open Operator Program (Rev23) — Closed as Ratios-Complete with One Registered Scale Input (Rev28; suite Rev33.1) A lab-notebook record of the S179–S192 rest-mass arc, closed at Rev25 by the S225–S226 scale-theorem chain (§10): the program is ratios-complete; the framework carries exactly one dimensionful input, counted by cohomology and registered in Appendix X. Definition A; α fenced; no public observable moves. Tom O’Sieg August 2026 Posture (binding). This appendix is an honest-open record of an active research program, not a results chapter. It documents what the framework’s charged-lepton rest-mass ma- chinery has and has not achieved across sessions S179–S192. Every statement is tagged [derived], [conditional], [input], [closed-negative], or [open]. No mass or phase value is promoted; the charged-lepton spectrum is reached only as a two-parameter fit; the framework’s published predictions are unaffected. The engine workbooks are unchanged across the whole arc (Machine 727b0bf01eb5, Outputs 7f3bc6f92b74 — the hashes of that era, S179–S192; the shipped pair is named in Appendix X); no public observable depends on anything below. 1 Scope and the one-line summary The framework fixes the charged-lepton mass ratios √ only through the Koide relation. What it derives is the shape relation — Q = 2/3 ⇔ r = 2, the circulant null condition (§3) — and not the value: the number 2/3 enters as an external selector (Rev29 ; WATCH box in §3). It does not derive the individual masses: those require two further numbers (the “gaps”) that, on all present evidence, are irreducible inputs of Yukawa type (§5). Between S179 and S192 the supporting operator architecture was built out in full and verified deterministically in-house (§4), and every physical route to deriving the two gaps was found to terminate at a single upstream object — a valid gauge-preserving S125 carrier, the same object that gates the Standard Model colour charges (§6). An independent AI-coalition review converged 2/2 on the conservative reading (§7): the shape is the result, the gaps are the input. 1 ===== PDF PAGE 273 / 433 ===== 2 The reframe: winding primary, mass a self-energy residual (S184) The organizing hypothesis of the arc replaces “select the mass” with “select the integer, then correct it.” A charged lepton occupies an integer winding N on the golden ladder, and its mass is written m = me φ N +δ , (1) with N primary and the mass detail carried by a smooth residual δ. The integer N is fixed by a spinor tension-balance closure that is winding-blind (det(φ+N , φ−N ) = 1 for every N ; Appendix H), which is precisely why every earlier attempt to dynamically select a preferred rung failed — there is no preferred rung to select. [conditional] reinterpretation: consistent with all data and it dissolves the “phantom rung” impasse, but it is not a derivation of δ. The integer windings themselves stay [input] (engine column H, SELECTOR=OPEN). The residual δ is read as a gauge self-energy correction to the propagator (kernels s346/s348): the leptons carry a Koide/EM self-energy (the winding-only Qpure = 0.6728 is refined to the physical Qreal = 2/3, with mean |δ| ≈ 0.045); the quarks carry an equal-and-opposite weak-isospin T3 self-energy (≈ ±12.4%, isospin-singlet part ≈ 0). This unifies two previously separate framework items under one reading but is shown only as self-energy structure, not computed from a propagator. [conditional]. The construction models the mediator as the white-hole/black-hole (WH/BH) dyonic two-sheet object: the electron is modelled as a T -bridged (4, 4) pair, BH(+E)/WH(−E), and the rest mass as the particle–hole half-gap the fermionic Möbius twist protects from cancellation (no negative rest mass; s308). [conditional]. Rev29 (F12): “is” is withdrawn throughout this reading. No identification of a physical mediator with the WH/BH object is claimed or shown; what exists is an operator model with the stated algebraic properties, and every downstream statement inherits the modelling step, not an identification. 3 The Koide shape: derived; the 2/9 phase: closed-negative Novelty label: standard theorem used. Proposition 3.1 (Koide as an exact circulant null condition, s430). For a Hermitian circulant √++ + co K−− on the 1 ⊕ 2 generation amplitude A = c0 I + ce K √ √ Q = 2/3 holds iff √ split, the Koide value the radial ratio is r = 2. The constant traces to 2 = 2/ dim J2 (Os ) (C = 2/10). [derived] √ as a shape/null condition — and the attribution is split: the biconditional Q = 2/3 ⇔ r = 2 for √ a circulant m amplitude is the known geometric form of Koide’s relation (R. Foot, arXiv:hep- ph/9402242, 1994; the circulant form is also Brannen’s, 2006), not new here. What this appendix derives is√the circulant structure itself on the triality-organized 1 ⊕ 2 split, and the tracing of the constant 2 to dim J2 (Os ) (s430). Hence the novelty label “standard theorem used” above, and the [derived] tag on the shape (Rev33.0, PI call S362). This is the genuine, parameter-free √ content: the circulant (democratic) structure, the triality organization, and Q = 2/3 ⇔ r = 2. What it fixes is the shape of the lepton triple, not its two independent scales. 2 ===== PDF PAGE 274 / 433 ===== WATCH (W02, Rev29 ) — the shape is derived, the value is an external selector, and the two are not the same√claim. Proposition 3.1 is a biconditional: given Q = 2/3 the radial ratio is forced to r = 2, and conversely. That is a statement about shape. It does not produce the number 2/3 — nothing in the circulant null condition selects that value over any other admissible Q, and the value is supplied from outside the algebra. Against this, Trackers/HONESTY_LEDGER.md:38 records the Koide selector K = 2/3 as RED — INPUT, external selection, and the electron closure is carried as Koide-consistent under that external rule throughout Appendices A/O/X. The tension is recorded here, not resolved: both statements are true of different objects, and no route from the algebra to the value 2/3 has been closed anywhere in the corpus. Remark 3.2 (the δ = 2/9 “phase = weight” route is closed-negative as a derivation (S185)). The empirical Koide-Foot angle δ = 0.222230 sits a fraction of a percent from 2/h∨ (F4 ) = 2/9 = 0.22222, which prompted an F4 = Der(J3 (Os )) “phase = weight” derivation candidate. It is [closed- negative] for three independent reasons: (i) the exact value 2/9 is 442σ from the CODATA pole ratio mµ /me (s356) — the “0.003%” framing hid the far tighter µ/e failure; (ii) F4 = Aut(J3 ) is isospectral (s357: it preserves the characteristic-polynomial multiset) and Jordan derivations have zero diagonal part, so no f4 element rotates the diagonal primitive-idempotent plane — the Koide 3-cycle is a discrete Weyl/triality element, not a continuous Cartan rotation; (iii) h∨ = 9 is not unique to F4 . Retained only: the Foot = SO(2) geometry and the 2/9 arithmetic. Not banked as a result. 4 The WH/BH operator architecture (A1169–A1175, S192) Over the A1169–A1175 arc the operator content behind (1) and Prop. 3.1 was constructed explicitly and reproduced in-house (all kernels deterministic, run1==run2). The architecture is complete and self-consistent; this is a real, cumulative advance. It is also, by itself, value-free. √ Remark  4.1(the m amplitude-square rule is derived, s434). From a quadratic heavy-mirror 0 A action A† µI the low-energy Schur mass is mlight = 2 ( µ + 4a2 − µ) → a2 /µ, so AA† is a derived 1 p 2 low-energy mass, not an ad-hoc fourth-order rule. [derived]. A single scalar gap µ cannot fit the spectrum (it needs µ = 1.53 for µ/e but 6.58 for τ /e), which is what forces the two-block structure below. 3 ===== PDF PAGE 275 / 433 ===== Object Content Status Finite parent action (A1169) nonfactorizing, Hermitian ⪰ 0; Schur complement [derived] (appended, carries √ the oriented torsion K−− ; M = AA† gives not extracted from √ m an origin S125 ) m mirror rule (A1171) Rem. 4.1 [derived] Qutrit terminal (A1171) minimal common terminal of both rank-3 bridges; [derived] (minimal) operator-Schmidt rank 2; compressed parent 192- dim Gauge-centre species C16 C16 = (−1)Nc = e2πiT3 = −ΓLR , the unique non- [derived] (A1171) identity Hermitian involution in the SM gauge centre Polar–Möbius quotient maps the two clock null lines to one parti- [derived] (A1172) cle/conjugate orbit per block on the full Fock 32 = 16 ⊕ 16 Vectorlike 16 ⊕ 16 block full-rank (32), symmetric, gauge-invariant to 5.8 × [derived] (A1172) 10−16 ; charged Dirac masses require the doubled rep 3|2 reciprocity (A1175) assuming an orthostochastic symmetric K = [conditional] on the 1 14 5 ( 4 1 ) gives µp − µd = 3/5, µd = 14/15, µp = assumed reciprocity; 23/15 still inexact Direct-mass static contract a quadratic √ Schur self-energy is a mass operator, [closed-negative] (A1170) not a m amplitude; the static near-match is Dyson-unstable One-body B − L conjugacy impossible; mismatch = 2PNodd =3 (rank 10) ⇒ a [closed-negative] (A1175) degree-3 cocycle is required 5 The two-gap fit and its (lack of) predictive content With the architecture fixed, the charged-lepton spectrum is reproduced by the outer mirror Hout = µd Pd + µp P⊥ , (2) where Pd is the democratic singlet and P⊥ the doublet. With the framework amplitude modes (the singular values ae , aµ , aτ of the loaded circulant A) held fixed, (2) reproduces [1 : 206.7683 : 3477.2283] = PDG to 3 × 10−16 — but via two free gaps µp = 1.5314, µd = 0.9313 fit to two ratios. [LIB2-133] An operator derivation plus a two-parameter fit to two numbers has zero predictive content for those numbers. [input]. [LIB2-133] Remark 5.1 (dual-route reconciliation with the compact Paper 2 ratio (owed, not√ closed)). √ Paper 2 carries a compact algebraic relation for the charged-lepton ratio mµ /mτ = (φ/ 5)8/3 2/10 (re- tiered Rev29 : Structural / Loaded-correspondence / Reproduced, not zero-parameter — the +0.37% residual is the fit gap on C; Appendix X). That relation and the two-gap operator above are a dual-route pair, not a contradiction: the compact formula is an exact algebraic expression whose attachment is loaded, while the operator here is the attempted physical mechanism, and the two are reconciled only once one proves that the compact ratio in fact determines (µd , µp ) (or exhibits why the two routes predict different objects). [LIB2-133] That proof is owed — routed to open items as a dispatch head — and until it lands neither route is a completed physical derivation of the individual charged-lepton masses. Neither route is demoted; the pairing is stated explicitly so the suite’s two charged-lepton accounts are not read as a silent current-state contradiction. 4 ===== PDF PAGE 276 / 433 ===== The conditional 3|2 theorem (table above) gives the clean rational µd = 14/15, µp = 23/15 (µp − µd = 3/5, Tr = 4), but it is conditional on an assumed reciprocity and is inexact: the data require µp − µd = 0.60012 (not 0.6) and Tr = 3.994 (not 4), so the exact (4, 3/5) law mispredicts at the 10−4 level (misses 5.9 × 10−5 , 3.3 × 10−4 ). A fair census ranks (4, 3/5) only 4th of 304 simple closures; 3/5 is merely the nearest simple rational to 0.60012. The residual 10−4 is localized (∼ 97%) to a single ∼ 0.16% renormalization of the mirror scale, with no parameter-free derived origin in hand. [open]. Remark 5.2 (the exact additive-χ completion of the ratios is refuted; the order skeleton and shape survive (S245, s894/D1119)). A natural attempt to replace the two-parameter fit above with a single closed completion writes each mass ratio as r = χ2 /(φ2N + χ2 ) on the occupancy depth support {0, 3, 11, 17} (Appendix M) and asks whether one universal χ reproduces the charged-lepton and up-type ratios exactly. It does not: an exact single-χ closure fails on every bijection of the support (0/24 admissible; lepton pairs r = −0.0795/ + 0.0518/ + 5217), and no residual-derived dressing rescues it without an independent derivation (a residual-fitted χ is a fit, not a completion; the tautology guard holds). [LIB2-106] [closed-negative] at the declared ratio tier. What survives is exactly the structural content that was never in question: the ordered ∼ φ-depth skeleton (the integer winding order of (1)) and the few-percent shape on it (central deviations u −0.21%, µ +3.90%, τ −2.62%). The exact completion is dead at the ratio tier on every support; the order and the shape are unaffected. [LIB2-106] A named watch item is banked (Appendix X): a banked-period-70 two- parameter harmonic a cos(6πN/70) + b sin(6πN/70) passes the 0.1% bar at max |∆d| = 3.04 × 10−4 but only fitted (refused adoption); were its two coefficients ever derived, this row reopens. [LIB2-248] The blanket “≤ 2-parameter completion is impossible” claim is therefore withdrawn; the honest statement is that surveyed one-parameter forms fail and no universal ≤ 2-parameter no-go is proved. [LIB2-106, LIB2-248] 6 The provenance wall: one valid S125 carrier Every route to deriving the two gaps (rather than inserting them) reduces to one demand: a valid gauge-preserving S125 action that realizes the reciprocal central channel, the rank-32 carrier with its degree-3 B −L cocycle, and the quantum Z2 gauging. The registered carrier is not such an action: FAIL_INVALID_MAJORANA_FOLD ∥M − M ∥/max = 1.39 × 10−5 vs 2 × 10−8 ; electric Ward 1.107; ⊤  gauge-carrier residual 0.635 . A direct in-house test makes the wall concrete: feeding the registered physical arrays into the basis-invariant terminal/channel runner returns a rank-defective positive operator (σmin = 8 × 10−15 ) — the reciprocal channel cannot even form, so the loaded action never reaches the gap-extraction step. The corrected, gate-passing array does not exist; constructing it is the open problem, and it is the same object that gates the SM colour charges. [open]. Remark 6.1 (what is closed on the wall: the minimal quantum Z2 contract (s445)). The diagonal Z2 generated by Gdiag = τz ⊗ C16 is gaugeable at the finite-representation level: a full-rank symmetry-preserving regulator exists (rank 192, σmin = 1, SM Ward 1.33 × 10−15 , det Gdiag = +1), the mixed-SM anomaly traces vanish (O(10−15 )), and ΩSpin 5 (BZ2 ) = 0 for an ordinary internal Z2 , so the old absolute ± local-system choice is replaced by the minimal flat-bundle sum. [derived] 5 ===== PDF PAGE 277 / 433 ===== (minimal closure; the propagating discrete gauge sector and any topological weight are left open). This is a genuine increment beyond the classical descent. 7 The in-house decider and the coalition verdict Running the gauged Z2 all the way through on the loaded parent action (kernel s446, in-house): the gauging plus the polar–Möbius quotient plus the zero-momentum Schur complement force the gap block to the two-parameter form µd Pd + µp P⊥ — the reality/CP-restricted commutant on the three outer modes is exactly two-dimensional, span{Pd , P⊥ }, and the mode-to-gap assignment is fixed (the democratic mode carries aτ , the doublet carries ae , aµ ). That much is derived. But the as-built loaded action has gap block = I exactly (µd = µp = 1), giving ratios 200.6 : 2201.6 versus PDG 206.8 : 3477.2 — the loaded action does not reproduce the leptons; the split must be inserted. Extracting the data-required gaps gives µd = 0.93128, µp = 1.53139, which satisfy neither Tr = 4 nor split = 3/5 exactly. The two gap values are free. [input]. Remark 7.1 (D930 closed 2/2; D931 dispatched). A cold-context coalition review (D930) closed 2/2 (two independent reviewers), both converging on the conservative reading: the parameter-free Koide shape is the genuine result; the two gaps are irreducible Yukawa-like inputs; “extraction from S125 ” is likely a category error (the appended minimal action is the physical carrier). The follow-on dispatch D931 asks for the one missing object — a valid corrected S125 carrier passing all six gates — or a proof that the obstruction is fundamental (i.e. identical to the colour-charge wall), which would settle the program by retiring the extraction demand. The three live readings (pre-registered). The evidence currently favors (i). (i) The values are input, the shape is the result — the defensible achievement is the parameter-free Koide shape plus the operator architecture; the two gaps are irreducibly input, and chasing them is a category error. (ii) The carrier is everything — a valid gauge-preserving S125 and a forward functional, once found, fix the gaps and the colour charges in one stroke. (iii) A different mediator — WH/BH radial-ratio dynamics is the wrong engine for the residual. The single decisive test is the D931 carrier construction (or its impossibility proof). 6 ===== PDF PAGE 278 / 433 ===== 8 Status ledger Result Status √ Koide shape Q = 2/3 ⇔ r = 2 (circulant null) [derived] as shape; value 2/3 [input] (Rev29 ) WH/BH operator architecture (A1169–A1175) [derived] (complete, value-free; models the mediator — Rev29 ) √ m amplitude-square mirror rule [derived] Minimal quantum Z2 gauging contract (s445) [derived] (minimal) Forced two-gap structure + mode assignment (s446) [derived] The two gap values µd , µp [input] Exact additive-χ ratio completion (§5.2) [closed-negative] (ratio tier; order + shape survive) Conditional 3|2 reciprocity (14/15, 23/15) [conditional], inexact δ = 2/9 “phase = weight” derivation [closed-negative] Static direct-mass contract; one-body B −L [closed-negative] Valid gauge-preserving S125 carrier [open] (= SM colour-charge wall) Integer windings N ; overall mass scale [input] Bottom line (Rev23 state; superseded in part by §10). The rest-mass program is a built, self-consistent, minimal operator architecture that derives the Koide shape and the forced two-gap structure, but derives no mass or phase value: the two gaps are inputs, the conditional 3/5 is not exact, and everything else routes to one unbuilt object — a valid gauge-preserving S125 carrier, the same gate as the SM colour charges. An honest, fully-mapped plateau. Nothing is promoted; the engine is unchanged; α stays fenced; no public observable moves. 9 Three recorded negatives from the S197/S208 reconciliation (Rev30) The S208 reconciliation ledger homes three of its Tier-C clean negatives in this appendix specifically. They were verified and grep-anchored at the time and then left unbanked for sixty-five sessions — not because they were doubted, but because none of them moves an observable. They are folded here. C2 — the charged action-level Dirac mass is forbidden A no-Q theorem forbids a charged Dirac mass at action level; the neutral 2 × 2 Majorana channel is not forbidden. The result is triple-confirmed — house kernels s364 and s391 plus an independent external derivation. This is the reason behind a statement the ledger above already makes but does not explain: the 16 ⊕ 16 vectorlike block (A1172) records that charged Dirac masses require the doubled representation. The no-Q theorem is why they require it. Without the doubling there is no gauge-consistent charged 7 ===== PDF PAGE 279 / 433 ===== Dirac mass to write down, so the doubled rep is not a modelling convenience — it is the only surviving channel. C8 — the selector Σ parity-split end-state, and what it honestly is Frozen at A1208, the selector Σ splits by parity into even = Urec + K∗ + gaps, odd = κ⊥ + ε. The honest end-state. Rest-mass together with mixing is a named section, not a derivation. The parity split organizes the content and tells you which pieces are recovered (Urec ), which are structural constants (K∗ , κ⊥ ), and which are gaps or residues; it does not produce a mass. Reading the split as a derivation is the error this entry exists to prevent, and it is the same plateau the status ledger above describes in different words. C9 — the winding-monotone driver is falsified by the leptons The hypothesis that Σwb acts as a winding-monotone rest-mass driver is falsified, 3/3: across the charged leptons the residuals alternate in sign rather than decreasing monotonically with winding. What is falsified, and what is not. Kernel s365 establishes an exact closed form for the white/black winding sums, Σw = φ−N and Σb = φN , reproduced to a maximum absolute error of 1.42 × 10−14 . That closed form stands. What fails is the physical reading laid on top of it — that monotone winding drives the rest-mass ordering. An exact identity and a false interpretation of that identity can coexist, and here they do; the record separates them so a later reader does not retire the identity along with the reading. This negative is also what the reframe in §2 already assumes in substance (winding primary, mass a self-energy residual); the negative itself had simply never been printed. 10 The scale endgame (Rev25): ratios-complete, exactly one reg- istered input The S225–S226 arc (assessments A1351–A1355; kernels s715–s748) closed the program’s scale question by theorem chain rather than by derivation of a value. The chain, in order: 1. Rescaling no-go (A1351, kernel-verified at deviation 0.0): the positive monoidal rescaling leaves every banked dimensionless output invariant — any claimed absolute-unit derivation from the banked structure alone has smuggled an input. Sharpened (A1352): monoidally natural boundary-trace functionals are homogeneous and also rescale freely. 2. Two-parameter no-go (A1353): two independent rescalings (λD , λσ ) leave every banked √ √ output invariant while me / σ = Lσ /Ldepth moves. Theorem: me / σ is underivable without a new non-homogeneous metric datum — an input is required, not merely missing. 3. Two doors (A1354): the missing metric hub splits into (i) the scale datum — non-homogeneous, Bargmann-class — and (ii) the charge coherence kernel — dimensionless, Green-function-class. Two different missing objects; only door (i) carries the unit. 8 ===== PDF PAGE 280 / 433 ===== 4. The count (A1355 + s741): H 2 (Sym; R+ ) = R, generated by the area cocycle;the census is 2 ncyl 2 unconditional (the declared-choice flag was removed in-house), and dim H = 2 = 2 = 1 — one H 2 class in the selected symmetry model Sym(Wind), whose census has ncyl = 2 (s697). Rev29 fence (F04): the earlier phrasing “one input because two cylinders” is withdrawn as an overclaim. The count is a theorem about the selected model; exhaustiveness is open — nothing here establishes that the two-cylinder census is the only admissible one, so the causal “because” is not carried. The input’s identity is the Bargmann central charge of the winding category — the classical pedigree is exact: as in Bargmann’s treatment of the Galilei group, the physical scale enters as a central extension class, not as a dynamical output. Declaration (Rev25). The rest-mass program is ratios-complete: every dimensionless ratio in its scope is derived or typed-open, and the program requires exactly one dimensionful input — the depth-cylinder class Ldepth /Lσ , registered with value, ladder position, and sealed null-model guard in Appendix X (whose Rev32 input-count convention box reconciles this one-class count with the two-calibration and three-anchor usage inventories printed elsewhere). The registration is an administrative act sitting on mathematics: the theorem chain above says an input of exactly this class must exist and that no second one may. Door 2, honestly. The charge coherence kernel (door ii) has a silhouette only: dimensionless, junction-resolved, configuration-dependent (“coherence multiplies by φ,” A1354); the radion-first- harmonic candidate for its organization was tested and killed (s743). It remains the named owner of the charge-magnitude defects (Paper 9, §8) and is not the scale input. Remark 10.1 (the one-input obligation — the hierarchy as a falsifiable prediction). dim H 2 = 1 forbids a second independent dimensionful input. Every other dimensionful ratio in the suite is therefore theorem-mandated derivable — in particular vEW √ = 553.3026 (logφ = 13.1250; not φ-menu-clean, consistent with a dynamical origin), σ which is now an obligation the framework must pay, not a hope: the electroweak hierarchy is converted into a falsifiable structural prediction (paid at conditional/loaded-depth tier in §11, √ vEW / σ = φ105/8 ). Exhibiting a second underivable dimensionful ratio would falsify the one-input theorem chain outright. The σ-bridge (vEW -derivation) program is the designated post-freeze flagship, alongside the island-of-stability program (Paper 7). Remark 10.2 (the Bergman kernel homes the φN ladder (s594/s595, S214)). A third object touches the rest-mass narrative and is recorded here at remark tier: the Bergman kernel of the golden domain evaluates exactly to K(golden) = 1920 φ10 /π 5 — the φN ladder appearing in a canonical reproducing-kernel normalization. Suggestive of where the ladder lives; no number is scored on it. Bottom line (Rev25). The scale question is finished mathematics: an input is required (two-parameter no-go), it is unique (cohomology count, unconditional census), its class is Bargmann, its value is registered with a sealed null guard (Appendix X) — and the √ same theorem that grants the one input demands that vEW / σ be derivable, making the hierarchy problem a prediction the framework must now pay or be falsified by. 9 ===== PDF PAGE 281 / 433 ===== √ 11 The σ-bridge scale registration (Rev26): vEW / σ = φ105/8 Posture (binding, Rev26). This section pays the obligation of Remark 10.1 at condi- tional / loaded-depth tier. It is the suite’s first foregrounded dimensionful result — but a framed registration, not a frame-free prediction: the value is fixed relative to the marked compact-positive Higgs carrier (the Marked Carrier Theorem, Paper 4), and frame-free selection of the electroweak line, level, and refinement is closed negative (the E7(7) no-go of Appendix F; the permanence obstruction). What is promoted is the derivation of the registered value, from bare obligation to a stated conditional registration on named loaded data; it is NOT a from-nothing prediction. No dimensionless observable moves, and α, colour, and VCKM = I are untouched. The framed-depth operator on the annular / beat-depth carrier evaluates to Ddepth = Nann + β8 = 13 + 18 = 105 8 , Nann = 13 = F7 = a8 , β = 1, (3) with the metaplectic eighth-tick 1/8 and Nann = 13 the eighth Fibonacci / annular index (homed at τ ⊗8 level n = 8; competitors 10, 12, 14 structurally excluded). This gives the electroweak registration vEW √ = φ105/8 = 553.30 (logφ = 13.1250), (4) σ paired with the companion dimensionless identity (exact, no scale content) m √ τ = 4 = |Jvac |2 = φ2 + 1 + φ−2 , Jvac = diag(φ, 1, φ−1 ), (5) σ so that vEW /mτ = φ105/8 /4, which the registered inputs reproduce to 0.17%. Remark 11.1 (what is derived, what is loaded). The permanence theorem (Appendix F) closes the frame-free route: on the bare Freudenthal system E7(7) fixes no Higgs line (load-ew-line) and no canonical quadratic-refinement basepoint (load-ew-frame, β = 1), and the level Nann = 13 is homed but unattached (load-ew-index). Hence φ105/8 is the unique visible framed-depth registration, relative to the marked compact-positive carrier of Paper 4 — a loaded-depth theorem candidate, not a frame-free scale theorem. Honest over-determination count (correcting an earlier house claim): the Ldepth /Lσ registration and the vEW route are mutually circular, so they do not constitute independent over-determinations of the scale — the framed registration carries no triangulation surplus from that pair. The one genuinely independent cross-check is the dimensionless √ vEW /mτ = φ105/8 /4 (using the exact identity mτ / σ = 4), which the registered inputs meet at √ 0.17%; the dimensionful vEW / σ = φ105/8 itself is a loaded-depth registration, not a zero-input prediction. Falsifier (obligation-as-threat). The one-input theorem chain (§10) forbids a second √ independent dimensionful input, so vEW / σ must be structurally fixed; φ105/8 is the value it is fixed to. The prediction is falsified by either (i) a second genuinely underivable √ dimensionful ratio in the suite’s scope, or (ii) a determination of vEW / σ inconsistent with √ φ105/8 beyond the σ registration uncertainty. The framework pays this or is broken by it. 10 ===== PDF PAGE 282 / 433 ===== 12 The authoritative charged-lepton ledger (entered Rev28; tiers as re-tiered Rev29) This section is the one authoritative statement of charged-lepton provenance in the suite. Papers 0, 2, 4, 7 and Appendix A point here; where any of them carries a shorter form, this table governs. Quantity Provenance Tier What is not claimed mτ external measured an- Input not derived; not a predic- chor (ΛG2 dimensional- tion transmutation anchor at the heavy-lepton √ 8/3 end) √ mµ /mτ (φ/ 5) · 2/10; prove- Structural / not proved and not zero- nance target-first [D616, A616, Loaded-corr. parameter: the +0.37% A618] / Reproduced residual is the fit gap on C (re-tiered Rev29 ) (required √ C = 0.140894 vs. 2/10 = 0.141421; Ap- pendix X) me QED-corrected Koide chain Koide- conditional on Koide; not under the external empirical consistent ⋆ ⋆ ⋆⋆ a J3 (Os )-internal deriva- rule K = 2/3 [A636, A637 — tion A637’s internal “proved at ev- ery step” rating is not the tier carried here; see App. A] Joint ledger anchor + structural ratio (re- algebraic mass cor- the three masses are not tiered Rev29; see the tier rows respondence jointly derived from above) + external-rule closure J3 (Os ) alone Dual route the compact ratio vs. the two- Reconciliation the two routes are not yet gap operator of §4/§5 owed shown to be the same ob- ject One-line posture. The charged-lepton sector is an algebraic mass correspondence, struc- turally pinned with mixed provenance, with a dual-route reconciliation owed. Rev29 : the phrases “closed”, “proved zero-parameter” and “zero-free-parameter” are withdrawn for the mµ /mτ ratio as well, which was the last place they applied; the ratio is Structural / Loaded-correspondence / Reproduced, and no quantity in this appendix may now be cited as zero-parameter. 11 ===== PDF PAGE 283 / 433 ===== Leibniz Quantum Beats Newton Appendix P — The Fifth Direction: Two-Sided Necessity, the Tick’s Exact Z4 Momentum-Like Grading, and the NS/R Parity Pattern of the Flicker (Rev33.1; entered Rev28, claim wording corrected Rev29 ) Five kernel-verified theorem-tier statements (S226: s734, s740, s741; A1356), with the Kaluza–Klein/Einstein–Bergmann identification confined to a labeled interpretive subsection. No observable moves; nothing dynamical is claimed. Tom O’Sieg August 2026 Posture (binding). This appendix promotes exactly five statements, each kernel-verified or hostile-lane-proved, and none of them an identification of the framework with a Kaluza–Klein spacetime. The identification narrative is quarantined in §3 (labeled interpretive) and earns no theorem status. The colour carrier question stays open (the loaded norm is missing); colour promotion remains fenced; structural statements only. Contents 1 What is being claimed 1 2 Theorem tier 2 3 Interpretive subsection (labeled; discussion tier — no theorem status) 3 4 Status ledger 4 1 What is being claimed The banked winding category carries an internal direction that behaves, in five independently verified respects, like a compact fifth coordinate: its clock weight carries an exact Z4 momentum-like grading, its parity flicker reproduces the NS/R parity pattern under the signed trace, its scale slot is a one-dimensional cohomology class in the selected symmetry model, a flat (cylinder-condition) reduction is dimensionally impossible, and no purely four-dimensional model reproduces its three signatures. Each statement is a theorem about banked objects; none asserts that spacetime has five dimensions. [LIB2-055] 1 ===== PDF PAGE 284 / 433 ===== Conflation fences (Rev29, families F02–F04). Three “is”-statements carried by earlier revisions are corrected here and are not restored anywhere below. (F02) The tick is not asserted to be a fifth momentum: what is proved is an exact Z4 momentum-like grading, and no compact-coordinate map exists — no internal coordinate x5 , no periodic identification, and no Fourier dual is constructed anywhere in the banked material. (F03) The flicker is not asserted to be the NS spin structure: the signed trace reproduces the NS/R parity pattern, and no Spin-bundle map exists — no bundle, no manifold, and no spin structure on one is exhibited. (F04) The H 2 count is a theorem about the selected symmetry model, not a causal derivation of the input count; exhaustiveness is open. 2 Theorem tier Novelty label: new specialization proved here. Theorem 2.1 (the tick carries an exact Z4 momentum-like grading; s740). The Z4 clock weight of the banked grading behaves as a momentum-like quantum number modulo 4: the grading law, additivity under composition, and reversal under the mirror are all exact on the full generator set. [LIB2-054] (Kernel-verified; deterministic; zero deviations.) Fence (F02, Rev29 ): this is a statement about a grading on banked objects, not an identification with a momentum conjugate to a compact coordinate. No compact-coordinate map is constructed — there is no internal x5 , no periodic identification, and no Fourier duality between the grading and any coordinate anywhere in the banked material. [LIB2-054] The word “momentum” is used for the formal properties (additivity, mirror reversal, Z4 valuedness) and for nothing else. Novelty label: new specialization proved here. Theorem 2.2 (the cylinder condition fails by dimension count; s740). The classical Kaluza–Klein cylinder condition (no dependence on the internal coordinate) is inconsistent with the banked content: the required invariant decomposition has dimension signature (1, 27, 27, 1), and the count does not close on the cylinder-invariant subspace. [LIB2-055] Any five-dimensional reading of the framework is therefore forced into the Einstein–Bergmann class (periodic internal dependence), not the Kaluza–Klein cylinder class. Moreover the loaded fiber admits no dynamical stabilization from the banked structure alone — the radion problem is met honestly, not hidden (s740; cf. the registered-input theorem chain, Appendices O/X). Novelty label: new specialization proved here. Theorem 2.3 (the signed trace reproduces the NS/R parity pattern; s734). The signed trace over the banked mirror (Lucas ±) ladder reproduces the antiperiodic (NS) / periodic (R) parity pattern of the three-torus sector TS3 , with the Freudenthal bit playing the role of the holonomy sign. [LIB2-084] Exact at both levels on the banked data (N = 24 check exact). Fence (F03, Rev29 ): what is exhibited is the parity pattern a spin structure would produce, not a spin structure. [P-016] No Spin-bundle map is constructed — no manifold, no principal bundle, and no lift of a frame bundle appears anywhere in the banked material, so “the flicker is the NS spin structure” is withdrawn. [LIB2-084, P-016] Novelty label: new specialization proved here. 2 ===== PDF PAGE 285 / 433 ===== Theorem 2.4 (one scale slot, counted; A1355 + s741). dim H 2 (Sym; R+ ) = ncyl  2 = 1 for the banked symmetry census (ncyl = 2): the generator is the area cocycle; the physical (both-flip) mirror preserves the class; single-cylinder flips (det = −1) are forbidden by consistency, and the census is unconditional — the earlier declared-choice flag on the Sym(Wind) identification was removed by in-house re-derivation with the actual banked generator set (s741). [LIB2-056] “Exactly one dimensionful input” is then a counting statement about this model: one H 2 class in the selected symmetry model, whose census has ncyl = 2. [LIB2-056] Fence (F04, Rev29 ): the earlier phrasing “one input because two cylinders” is withdrawn as an overclaim. The theorem computes dim H 2 given the Sym(Wind) census; it does not establish that two cylinders are the only possibility, and exhaustiveness is open. [LIB2-056] The census being unconditional within the identification is not the same as the identification being forced, and the causal “because” is not carried. [LIB2-056] Novelty label: new specialization proved here. Theorem 2.5 (no 4D counter-model without equivalent internal structure; A1356, hostile lane). A four-dimensional model that lacks the equivalent internal structure — a noncompact internal modulus, an internal H 1 (−; Z2 ) / w1 ̸= 0 twisted bundle, and an internal grading circle — cannot reproduce the three banked signatures, for three named reasons: (i) the parity ladder requires an independent internal H 1 (−; Z2 ) class — universal and particle-blind — which pure 4D supplies only as a spacetime loop (path dependence) or a hidden circle; (ii) compact 4D data yields elliptic holonomy (phases), never a positive hyperbolic φN dilation, absent a noncompact scale direction; (iii) the half-twist colour structure needs a w1 ̸= 0 transverse cycle that pure 4D does not possess. [LIB2-085] (Scored at the frozen binary; full credit.) Remark 2.6 (scope — the fifth direction can be traded, not eliminated). The theorem is a necessity of the structure, not a spacetime-dimension count: a four-dimensional model can reproduce the signatures if it imports a noncompact internal target modulus, a discrete/twisted internal bundle (w1 ̸= 0), or an internal grading variable — but doing so trades the fifth direction for the equivalent internal data rather than evading it. The claim is that the three signatures cannot arise from bare four-dimensional spacetime data with no such internal structure; they demand either a fifth direction or its internal proxy. [LIB2-085] Remark 2.7 (two-sided necessity). Theorems 2.1–2.2 argue from inside (the internal direction is really there, and it cannot be flattened away); Theorem 2.5 argues from outside (its signatures cannot be faked in 4D). The necessity argument is therefore two-sided — the strongest epistemic form available short of a scored observable, which the fifth direction does not yet have. That lack is exactly why §3 is interpretive. 3 Interpretive subsection (labeled; discussion tier — no theorem status) Label. Everything in this section is interpretive narrative: a reading of the theorems above, adopted for orientation and program design, not banked as a result. This is the discussion layer, and it is fenced. Read together, the five theorems say the framework has arrived at the classical Kaluza–Klein frontier in its Einstein–Bergmann form: holonomy (circulation) data is derived — charge signs, 3 ===== PDF PAGE 286 / 433 ===== parities, the nuclear circulation ladder of Paper 9 — while the modulus is not derived but counted: exactly one Bargmann/radion class, its value registered as the framework’s single dimensionful input (Appendices O/X). “Gauge data derived, modulus registered” has been the generic state of every KK theory since 1921; what is new here is that the radius problem is met by theorem (the count of Theorem 2.4, the honest no-stabilization statement of Theorem 2.2) instead of by hope. In this reading the moment-thickness of the time layer (Appendix H) is the depth length Ldepth itself; the universe is one section of the loaded bundle; and the tick-grading theorem echoes the 1926 Klein quantization argument in golden form (an echo, not an identification — see the F02 fence). If the identification ever earns more than narrative status, it will be because the two-circulation sector or the carrier norm produces a scored number of its own — until then, this section carries no weight in the ledger. Remark 3.1 (pencil paragraph (in-house, unreplicated — recorded, not promoted)). Two fresh in-house diagnostics are noted for completeness, both GRAY: s744 (signature appears to track stiffness: soft tree-Hessian modes, including the negative one-loop mode, load the (2, 2) sheet; translators and generic modes load (1, 3) — one session old, unreplicated) and s736 (n-face defects proportionally φ-quantized at ∼ 8% significance after trials — suggestive only). One clean negative is quotable: s743 — charge-face defects are not tick-organized; the radion-first-harmonic hypothesis is killed. None of the three enters any ledger. 4 Status ledger Statement Status Source Tick: exact Z4 momentum-like grading (no coord. derived (exact) s740 map) Cylinder condition fails; (1, 27, 27, 1); EB forced derived s740 Signed trace reproduces NS/R parity pattern (no derived (exact) s734 Spin-bundle map) dim H 2 = ncyl  2 = 1 in the selected model derived; exhaustive- A1355 + s741 ness open 4D-only counter-model refused (three walls) derived (hostile A1356 lane) Golden-KK/EB identification interpretive §3 s744/s736 diagnostics gray (unreplicated) tracker tier Radion first harmonic closed-negative s743 Bottom line (Rev29 wording). The fifth direction is necessary from both sides, its tick carries an exact Z4 momentum-like grading (with no compact-coordinate map), the signed trace of its flicker reproduces the NS/R parity pattern (with no Spin-bundle map), and its one missing number is counted — in the selected symmetry model, exhaustiveness open — and registered rather than fitted. The identification with an Einstein–Bergmann spacetime remains a labeled narrative — and will stay that way until it pays in a scored observable. 4 ===== PDF PAGE 287 / 433 ===== Leibniz Quantum Beats Newton Appendix P Sidebar — The Einstein–Bergmann–Bargmann–Pauli Five-Dimensional Program and Its Ledger Descendants (Rev33.1; drafted Rev29-WIP) A historical companion to Appendix P. Tier: Historical throughout — no theorem is proved here, no observable moves, and no physical five-dimensional spacetime is claimed. Every framework statement cited below carries its own tier in its own assessment; this sidebar adds commentary, not status. Sources: the 1938/1941/1943 papers; assessments A1452, A1462–A1466; kernels s942–s949. Tom O’Sieg August 2026 (sidebar drafted S256, Rev29-WIP) Posture (binding). This sidebar is Historical commentary. It promotes nothing. The correspondences drawn between Einstein’s five-dimensional program and the suite’s banked results are structural analogies with exact citations, not identifications: the suite’s fifth direction remains the internal tick/flicker carrier of Appendix P, and the identification with a physical Kaluza–Klein/Einstein–Bergmann spacetime remains interpretive and quarantined there. The S255–S256 results cited by assessment number (A1462–A1466) are fold-pending: their full appendix treatments do not yet exist in this revision, and the assessment files are the citable artifacts of record (Appendix Y discipline). Contents 1 Why this sidebar exists 1 2 The three acts, 1938–1943 2 3 The failure, retold as banked theorems 2 4 On the suite’s “reluctance” toward 5D 3 5 The Bargmann class: dimensionality from two structures 4 1 Why this sidebar exists The suite’s one registered dimensionful input is called the depth-cylinder Bargmann class (Appendices O/X). Appendix P confines the Kaluza–Klein reading of the fifth direction to a labeled interpretive 1 ===== PDF PAGE 288 / 433 ===== subsection and names Einstein–Bergmann in doing so. Those two names are not decoration. Between 1938 and 1943 Einstein, with Peter Bergmann, Valentine Bargmann, and finally Wolfgang Pauli, ran the closest historical ancestor of the program this suite is now executing: a finite-width fifth dimension, particles as regular field concentrations, and a scalar width modulus nobody could fix. The program failed, and Einstein said so plainly. The purpose of this sidebar is to record why it failed, in his own terms, and then to show that the S255–S256 mass-frontier arc has re-derived each of his failure modes as an exact banked theorem — while adding, with exact constructions, the two ingredients his setup could not contain. The wall he died on is real; the ledger has now proved it is real; and proving a wall real is different from, and better than, running into it. 2 The three acts, 1938–1943 Act I — the thick cylinder (1938). Einstein and Bergmann, “On a Generalization of Kaluza’s Theory of Electricity” (Ann. Math. 39, 683). Kaluza’s original theory imposed the cylinder condition — no field depends on x5 — by fiat, which reduced the fifth dimension to bookkeeping. Einstein and Bergmann relaxed fiat to periodicity: a fifth direction of genuine finite width, fields expandable in x5 -harmonics, spacetime as a sheet bounded by two hypersurfaces. In modern language they introduced the Kaluza–Klein tower; in this suite’s language they were the first to treat the thickness of the extra direction as physical data. Their stated hope was that particles would arise as stable, everywhere-regular concentrations of the five-dimensional field. Act II — the particle hunt (1941). Einstein, Bargmann, and Bergmann, “On the Five- Dimensional Representation of Gravitation and Electricity” (Theodor von Kármán Anniversary Volume). The search for non-singular particle-like solutions of the periodic theory. The result was deflationary: the construction kept collapsing back to general relativity plus Maxwell, with nothing left over to be the electron. Act III — the no-go (1943). Einstein and Pauli, “On the Non-Existence of Regular Stationary Solutions of Relativistic Field Equations” (Ann. Math. 44, 131): the five-dimensional vacuum theory admits no regular stationary particle-like solutions with mass. This theorem is the program’s tombstone. Einstein’s own post-mortem, stated in correspondence and in his later survey remarks, had four counts: the theory produced nothing beyond GR + Maxwell; it could not produce a particle; it said nothing about the quantum; and the g55 component — the width of the fifth dimension — was either frozen arbitrarily or ran loose as an unwanted scalar with no principle to fix it. A fifth count sat unspoken until decades later: a smooth five-dimensional circle cannot produce chiral fermions at all. 3 The failure, retold as banked theorems The correspondence below is the sidebar’s content. Each row pairs one of Einstein’s failure modes with the exact result that is its modern descendant in this ledger. The direction of the analogy matters: the ledger results are not about Einstein’s theory; they are theorems of this framework’s conditional 5D attachment which happen to close, with proofs, the questions his program left as frustrations. 2 ===== PDF PAGE 289 / 433 ===== Einstein’s failure mode Ledger descendant (exact, banked) Pure geometry yields no particle struc- Free compactification is hierarchy-blind ture; the periodic theory collapses to (A1462, s942); the free mouth-Real determinant GR + Maxwell (1941). is globally hierarchy-dead by strict concavity and weighted Jensen (A1465, s947); every universal spectral action selects only hierarchy-free rays (A1464/A1465). “Cheap geometry gives nothing” is now a theorem family, not a disappointment. No regular stationary particle solutions exist The evasion is exactly what the 1943 hypothe- (Einstein–Pauli 1943). ses exclude: a scalar sector with a potential and a topologicalp charge. The BPS √ kink wall 2 3 σ = v n tanh(v λ/2 y), tension 3 2λv , is de- rived, regular, and stable (A1464, s945) — a soli- ton living precisely outside the Einstein–Pauli no- go’s scope. A smooth S 1 gives vectorlike matter — no The scalar-twist censuses are zero-mode-dead, and chirality (the unspoken count). the mouth-Real matrix twist Tm = −τ1 produces exactly one branch-Real chiral zero channel with NS partner gap |p5 |R = 12 (A1463, s944). Chi- rality enters through the mouth exchange — an ingredient with no 1938 analogue. g55 : the width modulus can be neither fixed The radial-shadow theorem (A1466, s948): the nor banished. compact phase of the complexified X line is fixed by the center/semion structure, while the radial magnitude s is provably blind to center, semion, anomaly, Yukawa invariance, and wall topology. Einstein’s frustration is now a proved indepen- dence, with a countermodel pair. The wall is real. Particles as “bridges” (Einstein–Rosen 1935), The WH/BH mouth pair, algebraized: the killed by singularities and instability. branch-Real Majorana clause (A1452, one zero- dimensional clause, priced not derived). The bridge survives by not being a geometric bridge — conditional per Appendix P. Two ingredients Einstein lacked, one wall he could not name. The ingredients — chiral matter via the mouth twist, and a scalar/topological sector for solitons — are now exact constructions. The wall — the width modulus — is now an exact independence theorem, and it is the live head object (P0-24, Q2: does the parent action fix s?). 4 On the suite’s “reluctance” toward 5D It is fair to ask why a framework with a fifth-direction theorem, a KK-tower annulus appendix, and a depth-cylinder input has been slow to simply be a Kaluza–Klein theory. The answer is that the reluctance is Einstein’s failure, encoded as discipline. Einstein committed the interpretation first — 3 ===== PDF PAGE 290 / 433 ===== a real physical fifth dimension — and spent fifteen years hunting mathematics to justify it. The suite runs the arrow the other way: the internal statements (the tick’s exact Z4 momentum-like grading; the signed trace reproducing the NS/R parity pattern) are theorem-tier and closed as stated — and, per the Rev29 F02/F03 fences in Appendix P, they are not identifications: there is no compact-coordinate map and no Spin-bundle map. The spacetime identification is held interpretive in Appendix P, where it can wait indefinitely without collapsing. The step has, in fact, now been taken — in the only form the two-axis provenance rule permits: LMRSS is a named candidate 5D attachment, constructed-conditional, its caveat printed in every round that uses it (A1463 onward). This is the Einstein–Bergmann program resumed, with a ledger where he had only conviction. 5 The Bargmann class: dimensionality from two structures The deepest correspondence is the one the suite’s own nomenclature made before this sidebar was written. After leaving Einstein’s program, Bargmann proved (1954) that quantum mechanics represents the Galilei group only through a central extension — and the central charge is mass. A central extension is precisely the mechanism of the “T-bridge” intuition: a flat structure acquires one new dimension not by adding geometry but by a cocycle measuring the failure of two structures to compose flatly. The suite’s registered dimensionful input is the depth-cylinder Bargmann class: the extension class that welds the internal clock (flicker/NS) to internal translation (tick/p5 ) and turns a flat boundary into a cylinder. Einstein and Bergmann carried the width as a free parameter with no principle; Bargmann’s own later mathematics explains why kinematics alone could never have fixed it — central charges are inputs to a symmetry, not outputs of it. Whether the class is derivable from the parent action or remains the honest registered input is exactly the open radius-weld/X-magnitude question (P0-24, Q2/Q3). Einstein discovered that question the hard way between 1938 and 1943. This ledger has it stated, fenced, and gated. Fences (carried). No physical five-dimensional spacetime is claimed anywhere in this sidebar. The Einstein–Pauli evasion row does not assert that the BPS kink is a particle. The Majorana-clause row remains one priced zero-dimensional clause. The radial modulus s and the finite thresholds K3 , K1 remain underived; the mechanism of A1466 can span the hierarchy and has not been claimed to explain it. Historical attributions follow the primary papers cited in §2; secondary paraphrases of Einstein’s post-mortem are marked as such. 4 ===== PDF PAGE 291 / 433 ===== Leibniz Quantum Beats Newton Appendix Q — The Golden Vacuum’s Classical Spectrum: Machine-Zero Closures from the Dynamics Lane (Rev33.1; entered Rev25; bounded, conditional tier) A bounded window onto the S209–S223 dynamics arc: only the finished, machine-precision items (the s652c spectrum closure and the Gate-arc receipts, including one recorded blind-prediction loss). Everything here is conditional-tier; the live lane (κ(w) anatomy, B-dressing, sealed windows F/W) is explicitly out of scope. The dynamics revision proper is Rev26. Tom O’Sieg August 2026 Posture (binding). This appendix exists for cross-pollination: the lab notebook informs the coalition lanes, and these closures were invisible behind the dynamics fence. Scope is deliberately bounded: finished, deterministic, machine-zero items only. Every statement is conditional-tier — conditional on the κphys mechanism, whose status is structural candidate (the S220 wording), not derived. [LIB2-196] Nothing here is a scored world-facing observable; nothing moves the engine; the epistemic class is internal machine-zero closure, which is deliberately kept separate from the frozen-bar world-scored ledger of Paper 9. [LIB2-196] 1 The spectrum closure (s652c, S220) On the CDIZ locus xi = 1 with STU eigenvalue data e = (0.99471, 0.96943, 0.94410) (and s4 = 1 to 2 × 10−15 ), the golden vacuum’s entire classical spectrum closes as CSS charge sums of the four gravitino masses: Novelty label: new specialization proved here. Proposition 1.1 (conditional tier). All 8 gravitini, 28 vectors, 56 fermions (including the goldstini), and 70 scalars of the N = 8 classical spectrum at the golden point are reproduced at machine precision as CSS charge sums of the four gravitino masses on the CDIZ xi = 1 locus. [LIB2-196] Derived in-house before the lane’s matching return arrived — an independent-convergence receipt, not a fit. This is the arc’s crown jewel and the reason this appendix exists: a complete classical spectrum organized by one four-number seed, at machine zero, twice, independently. 1 ===== PDF PAGE 292 / 433 ===== 2 The Gate-arc receipts The supporting closure chain, all deterministic and manifested: • A-tensor exact rebuild (s638): GAP = 0.0; Gate C at 1.1 × 10−15 . • Gate D machine closure: R6 residual 1.05 × 10−15 . • Vector Ward identities: trace 1.4 × 10−14 ; derivative form 5.9 × 10−11 . • Gate H — Coleman–Weinberg inertia, with the loss on the books: the computed CW inertia signature is (9, 0, 1). The pre-registered blind prediction (A1329) was (10, 0, 0) and was scored per protocol as a recorded loss. It is kept here deliberately: the blind ledger’s credibility rests on its scored misses as much as its hits. 3 Two negatives at the golden boundary point (Rev30, A1481) Two independent results constrain what can be built at the golden boundary point z0 . Both are house-verified; both are negatives. (i) z0 is not a critical point of the bare quartic. Direct computation gives dI4 (z0 ) ̸= 0, with ∥dI4 ∥ = 8.944272 and the α-component exactly −4. [LIB2-104] The consequence is immediate and fatal to a class of proposals: a parent action whose candidate vacuum is not a critical point has no vacuum there. This does not touch the stationarity results proved elsewhere for the constrained objects of this appendix — it says that the bare quartic does not have a stationary point at z0 , and therefore cannot by itself be the action that selects it. [LIB2-104] (ii) No positive equivariant polarization exists in the split-G2 family. [LIB2-105] With dim Comm(Der(Os )) in End(Os ) equal to 2, and the split norm on Im Os of signature (3, 4), all four equivariant involutions of the form J = s P1 + t PIm give sig(G·J) ∈ {(4, 4), (5, 3), (3, 5), (4, 4)} — none positive. [LIB2-105] The no-go stands as stated. Scope, and one arithmetic caution. Both results are computed in the split real form and inherit the real-form declaration of Appendix X: signatures in this family are a consequence of a declared input, not an intrinsic feature, and must be quoted with that premise named. [LIB2-105] Relatedly, a parameter chain that has been reported as a narrowing — 196 → 52 → 20 — is arithmetic, not physics: neither endpoint is derived, and 20 is still a twenty-parameter family. Nothing in this section promotes an observable, moves a tier, or touches the engine. 4 Status language (strict) • Everything in this appendix: conditional tier, on the κphys candidate mechanism. [LIB2-104] • sign(c): structural candidate (S220 PI wording) — not settled, not promoted. • The κ(w) anatomy, the B-dressing program, and the sealed blind windows F/W: live lane, out of scope here — deliberately not papered while mid-stream. 2 ===== PDF PAGE 293 / 433 ===== • The dynamics revision proper — where this tier either promotes or dies — is Rev26, gated on sign(c)/κ(w) settling. Bottom line. One bounded window: the golden vacuum’s classical spectrum closes at machine zero from a four-number seed, the gate chain behind it is receipt-complete, and the one blind prediction it ventured lost and is recorded as lost. Conditional tier throughout; the live dynamics lane stays fenced until Rev26. 3 ===== PDF PAGE 294 / 433 ===== Leibniz Quantum Beats Newton Appendix R — The Interacting Vertex: from the Threshold Measure to the First Derived Relative Vertex Sign (Rev33.1) Rev26 dynamics frontier (S238–S240). Conditional/candidate tier throughout. The head wall “no parent action / physical projection map” attacked directly for the first time: the transfer question is now a package with an exact parts count — one merged loaded import, three continuous data, and one named non-equivariant placement kernel — and it has produced its first derived, zero-parameter vertex sign. Nothing here is a world-scored observable; the vertex sign is relative and amputated (a g → 0+ doorway ratio, structurally a trilinear antisymmetry), not an interaction strength or S-matrix prediction; the F4 wording fence (“mass shell”/“amplitude” stay out of public claims) applies to every statement below. Tom O’Sieg August 2026 Posture (binding). This appendix papers the S238–S240 arc at the frontier, not the observable tier. Its epistemic class is conditional (on the loaded boundary Bphys and its holonomy class [σCD ]) and candidate (the interpretive readings — dominoes ↔ generations ↔ arms ↔ placements — stay explicitly distinct and unmerged). Every quantitative closure below was independently reproduced in-house (house double Cayley–Dickson octonions, house group theory, independent eigensolvers), byte-identical to the executed external kernel, before being banked. No public observable moves in this appendix; α and colour stay fenced; VCKM = I; the engine is unchanged. The single derived number, the relative sign −1, is a sign of a bare/amputated doorway ratio, tiered below any width or scale claim. 1 The interacting contact measure Provenance. This section carries the S237–S238 measure rounds: the measure theorem under attack (A1386), the Horn D discriminator (A1387), the order-selection vertex (A1388), Σ analytic then executed (A1389, A1390), Pth (A1391) and the interacting measure (A1392). Round-by-round index: Appendix Y. The wall left standing after the geometry arc was the physical readout of the threshold-residue projection Pth . The finite-circumference spectral family is Hc,R = Eth + c (−i∂θ )2 on L2 (SR 1 ) ⊗ Pz R8 , X(θ + 2πR) = σCD X(θ), (1) with σCD the deck holonomy. Two exact results bracket its readout. Novelty label: new specialization proved here. 1 ===== PDF PAGE 295 / 433 ===== Theorem 1.1 (Finite-R Riesz edge projector; conditional on the S 1 family). At finite circumference the periodic Eth eigenvalue is isolated and Resz=Eth (z − Hc,R )−1 = PEth exactly; the odd sector carries no residue there. Thus Pth is exact mathematics if the physical readout is “take the isolated pole residue.” Sketch. On SR 1 the Laplacian (−i∂ )2 has discrete spectrum k 2 with k set by the σ θ n n CD deck twist; the even sector contains the constant mode k = 0, giving the isolated eigenvalue Eth with a rank-one spectral projector, while the twisted (odd) sector has kn ≠ 0 for all n, so no mode sits at Eth . The resolvent (z − H) = n |n⟩⟨n|/(z − Eth − −1 ckn ) has a simple pole at z = Eth only from k = 0; its 2 P residue is PEth , and the contour integral Eth (z − H)−1 dz/2πi reproduces P0 to < 10−10 (house H s842-V8). □ Novelty label: new specialization proved here. Theorem 1.2 (Free-measure no-go). On the free contact problem the readout does not follow from the spectral measure. As R → ∞ the edge-atom mass 1/(2πR) → 0 while the off-edge tower converges √ √ −1 to the threshold continuum ρ(E) = 2π c E − Eth , and ρ(E) 1 Z dE = √ (2) E 2 c Eth exactly — the continuum carries the full R → ∞ point resolvent; Eth is a branch point, not a pole. Atomization (route (a)) and the KV-flow route (c) are dead at shipped tier; the wall relocates onto the interacting width-dressed measure rather than softening. Sketch. The finite-R edge atom has weight equal to the constant-mode normalization 1/(2πR), which vanishes as R → ∞. The off-edge modes condense into the continuum ρ(E) dE with √ ρ(E) = (2π c)−1 (E −Eth )−1/2 (density of states of a 1D quadratic band). The point resolvent at the √ −1 √ origin is ⟨0|H −1 |0⟩ = ρ(E)E −1 dE; substituting and integrating E∞th E E − Eth dE = π/ Eth R R √ gives exactly 1/(2 c Eth ), i.e. the continuum alone saturates the full resolvent weight and the edge atom contributes nothing in the uncompressed limit — so no measure-atomization readout survives freely (house s842-V3, analytic + quadrature at three c-values). □ The dressing is furnished by the OOO source dictionary, now derived at the algebra level: B : z 7→ vec(Mz ) with B T B = 32 I8 , rank B = 8, MeTk Mek = 4 I8 (all singular values 2), (σCD ⊗ σCD ) B = B σCD , (3) four even source channels (e0 –e3 ) and four odd (e4 –e7 ). The structural hope that the odd channel is protected (χodd ≡ 0) is refuted: |χ− | = |χ+ | exactly (democratic split 16/16). The exact branch map {2φ−2 , 0} is therefore typed as a g → 0+ pole-residue limit, with gcrit = 2c/(πR) a genuine predicted failure regime, not an all-coupling theorem. Attraction derives conditionally on a positive exit gap (Schur complement −B † (Hexit − E)−1 B ⪯ 0, isotropic −(32/gap)I8 ), and the radial null closes for this dressing: B · rradial = 0 identically, so Pcontact = Pz (dressing-relative). 2 The chamber atlas and the decomposed transfer functor Provenance. The chamber functor round (A1394, D1096): the L3→L4 shortcut was killed there and the functor decomposed into typed data; the linear zero and exit gap come from A1393 (D1095). Appendix Y. 2 ===== PDF PAGE 296 / 433 ===== The boundary side organizes into a computed object: the quaternionic chamber atlas — seven chambers (one division-H marked, six split M2 (R)), 21 typed lines, edge automorphism orbits {6, 3, 12}, a domino decomposition {1} + {3 × 2} with a genuine C3 action, and σCD fixing every chamber setwise. Its symmetry group is identified exactly: Gch ∼ = S4 (chambers = tetrahedron edges; R6 = 1 ⊕ 2 ⊕ 3), (4) with order profile {1:1, 2:9, 3:8, 4:6}, trivial centre, A4 witness of order 12. The exit continuum aligns with the threshold on the nose: Novelty label: new specialization proved here. Theorem 2.1 (Exit-edge alignment; conditional on additive two-tube locality). The stable positive branches of the shipped wall Hessian are Ex+ = 2φ I4 and Ey+ = 2 I4 ; the additive two-tube pair edge is Eedge = 2φ + 2 = 2(φ + 1) = 2φ2 = Eth , (5) using φ + 1 = φ2 — the threshold is the sum of the two occupied golden tubes. Projecting the dictionary onto the stable sector, B+ = Pss B obeys B+ T B = 16 P + even , rank 4, χodd stable = 0 (odd protection restored, conditional on the positive-exit projection). Sketch. Diagonalizing the shipped wall Hessian on the split-x and split-y tubes gives stable positive branches 2φ and 2 (fourfold each). Their additive combination on the two occupied tubes is 2φ + 2; the golden identity φ2 = φ + 1 collapses this to 2φ2 , which is precisely the banked threshold Eth . For the doorway, Pss projects B onto the four stable even directions; B+ T B inherits the 32 I + 8 normalization halved on the retained even block (16 Peven , rank 4) and annihilated on the odd block, because the odd source channels live in the compact–split pairs carrying the negative transverse branch (house s849-V1/V2, exact to < 10−14 ). □ Two negatives sharpen the object. The raw transverse Hessian has inertia (8, 8, 0), so global positivity is false; below the edge the projected gap equals δ (energy-dependent, closing at threshold), with √ exact failure locus t⋆ = φ — the banked σ2 = 0 escape edge. And the induced magnitude cannot be discrete: gind + /gcrit = 8πR/(c∆) grows with R, so no single ∆ fixes the coupling; a renormalized subtraction is required and the g → 0+ reading stays a typed limit. The attempted chamber functor therefore resolves, with incidence forcing none of it, into four typed action-side data: Fphys = (6)  ch η+ , Tch→∂ , Csub , rC , under the standing rule (adopted verbatim as house law) that incidence multiplicities are not normalized transition amplitudes: incidence ̸⇒ positive metric; line counts ̸⇒ subtraction; shared line ̸⇒ locking amplitude. 3 The η+ weld Provenance. A1395 (D1097), where the weld landed conditional and two house overclaims were refuted in the same round (casualty register, Appendix Y §7). The first of the four data collapses onto the oldest standing import. Among all diagonal sign maps on Os exactly eight are automorphisms, and exactly one yields a positive twisted form: Novelty label: new specialization proved here. 3 ===== PDF PAGE 297 / 433 ===== Theorem 3.1 (The η+ weld; derived-conditional). The σCD -twisted conjugation form ⟨a, b⟩η = Re(ā σCD b) has Gram = I8 exactly on all of Os (both chamber types); the positive twisted metrics form one Aut-conjugacy class. Adopting it as the physical kinetic metric merges the newest import with the oldest: the boundary loads the holonomy class [σCD ], and η+ is the Cartan metric induced by that class — one conditional import in place of two. In the welded frame the even-only stable T B = 16 P doorway B+ + even is a theorem of the welded metric (given the positive-energy transfer rule), and the eight diagonal automorphisms coincide with the chamber-setwise-fixing kernel — the weld, the atlas, and the Y-spine probe meet on σCD . Sketch. A diagonal sign map diag(ϵ0 , . . . , ϵ7 ), ϵi = ±1, is an automorphism of Os iff it respects the split-octonion multiplication table; an exhaustive scan of the 256 sign patterns leaves exactly 8 automorphisms (house s855-V1). For each, form the twisted bilinear ⟨a, b⟩ = Re(ā σ b) and Gram it against the standard basis; exactly one — σ = σCD — is positive definite, with Gram = I8 identically. These eight automorphisms are exactly the kernel of the S4 chamber action (order 8, the setwise-fixing subgroup of s856), which is why the weld object and the atlas object coincide.□ 4 The first derived relative vertex sign Provenance. A1395 (D1097). The sign is amputated and F4 -fenced; it is not an amplitude. Appendix Y. What this is, structurally. The result below is the antisymmetry of a forced alternating trilinear under the coordinate swap x ↔ y: a fixed component and its argument-swapped partner differ by a sign, so their ratio is −1 and the normalization g3 and subtraction Csub divide out. It is a derived relative sign of the amputated doorway vertex, not an interaction strength, width, or amplitude — the value of naming it is that the sign is forced (not conventional) by the welded metric and the census, and it is the first vertex quantity the program derives at zero parameters. On the trace-free (off-diagonal) doorway the two lower invariants I1 = I2 vanish identically and the Albert cubic survives alone, forcing the cubic vertex Vg3 (x, y, z) = g3 Tdet (x, y, z) there. (The stronger full-27 uniqueness claim is refuted: 27 = 1⊕26 is reducible, with at least three independent invariant trilinears; the repair holds on the carrier that matters.) From this forced vertex the program banks its first derived interaction quantity: Novelty label: new specialization proved here. Theorem 4.1 (First derived relative vertex sign). In the house Fano basis, with the welded metric and the amputated g → 0+ doorway, A(4x , 5y ; 1z ) = −1 (7) A(5x , 4y ; 1z ) zero-parameter: the cubic normalization g3 cancels and the common subtraction Csub cancels (checked at δ ∈ {0.01, 0.1, 1}). Supporting census: the 64-channel census {1, 21, 42} is entirely σCD -even; the 16-channel stable selection and the 12-row Fano sign table are entry-exact and antisymmetric in the house basis. 4 ===== PDF PAGE 298 / 433 ===== Census and antisymmetry (house s855-V3/V4). The 64 nonzero trilinear channels partition as 64 = |{z} 1 + 21 |{z} + |{z} 42 , (8) scalar diagonal (0pp) Fano all σCD -even (the scalar rows carry +2). Restricting to the stable split⊗split sector leaves 16 channels; of these the 12 Fano-line rows form the sign table, each row exactly antisymmetric under the coordinate swap x ↔ y (A(ax , by ; cz ) = −A(bx , ay ; cz ) entrywise, verified on all 12). The quoted ratio is the (a, b) = (4, 5) instance of that antisymmetry; because it is a pure sign flip, the cubic normalization g3 and the common subtraction Csub divide out identically, which is what makes the number zero-parameter. The external kernel then checks the same −1 survives the amputated g → 0+ propagator dressing at δ ∈ {0.01, 0.1, 1}; the house reproduces the algebraic sign directly and byte-identically (result hash ddbede9d, bit-portable). The full-27 refutation (house casualty, s855-V5). The dispatch’s premise that the Albert cubic is the unique invariant trilinear on the full 27 is false: 27 = 1 ⊕ 26 is reducible, and the invariant-value matrix of {tr3 , tr·q2 , det -polarization} has rank 3 — at least three independent invariant trilinears (q2 here is the quadratic trace invariant tr(x2 ), a form, not a constant; it is unrelated to the constants q0KK , q0clk of Appendices G/H). The repair: on the trace-free / off-diagonal doorway the two lower invariants I1 = I2 vanish identically (self-test Txxx = 6N ), so Vg3 = g3 Tdet is forced there and the −1 is unaffected. Fence (binding, F4). The ratio is a bare/amputated doorway sign, not a finite-width S-matrix prediction. It is relative, not absolute: it is a sign, and every magnitude stays behind Csub . It becomes a candidate public falsifier only once an amplitude theorem lifts the F4 wording fence; until then “mass shell” and “amplitude” stay out of world-facing prose. This is nonetheless the deliverable both external reviewers demanded — “derive one vertex” — at its first honest rung. 5 The equivariant transfer, the invisible mode, and Π5←6 Provenance. A1396 (D1098, the ten-dimensional deficit: 10=10 dead beyond the orienta- tion fibre, the readout built at rank 7, g3 structurally free) and A1400 (D1100, the five-state closure refuted: the unique fully equivariant unital kernel is the uniform rank-1 U ). Both are respected, not overturned, by the Rev28 observation quotient ΠMOS 5←6 (Appendix T). Appendix Y. The smallest missing object — the chamber-to-boundary spectral transfer Tch→∂ — splits into a built half and a loaded remainder. First, the graded coincidence dies the right death: Novelty label: new specialization proved here. Theorem 5.1 (The 10=10 coincidence, adjudicated dead beyond the fibre). The graded dimensions match (5 + 5 both sides), but the representations do not: (2 ⊕ 3)S4 ∼ ̸ (1 ⊕ 2 ⊕ 2)D5 . Since = W0A4 = 0 and every homomorphism S4 → D5 factors through Z2 , the equivariant intertwiner space HomS4 (Kch , R[D5 ]φ ) = 0: there is no nonzero equivariant map, not merely no isomorphism. Only the σCD ↔ Möbius two-state orientation fibre survives as genuine common structure. 5 ===== PDF PAGE 299 / 433 ===== Sketch (house s858-U1–U3). Gch ∼ = S4 (order profile {1:1, 2:9, 3:8, 4:6}, trivial centre, A4 witness of order 12). The chamber deficit restricted to A4 has no invariants: W0A4 = 0. Any equivariant Kch → R[D5 ]φ composed with an S4 → D5 map would carry an A4 -invariant into the image; but the brute-force enumeration gives exactly 6 maps S4 → D5 (and 10 reverse), all of image order ≤ 2, so each factors through Z2 and annihilates A4 . Hence Hom = {0}. The 5+5 graded coincidence is real but carries only the Z2 orientation bit. □ Constructively, the equivariant half is done: with column order (six chambers, then trace) the 16-channel readout has Fano block Gram MFT MF = 8 I6 (rank 6; every chamber label separately visible, all five relative weights recoverable) and scalar entry 16, i.e. spectrum diag(8×6, 16), so the full stable map has rank 7. (The house prediction “rank ≥ 8” was wrong — a casualty — the labels are the six columns, not a separate payload.) What provably cannot pass is the five-placement AnnTL coordinate and the absolute mode. After one source normalization exactly one kernel direction remains, Nabs = g3 · Zsub (Csub ), δ log |g3 | = − δCsub (9) (rank 7 of the 9-parameter Jacobian) — “the one number the boundary cannot see.” And g3 is proven structurally free: the kinetic UNINORM cvec = 24 fixes the tensor normalization (2/243/2 = 0.01701035 . . .), not the action coefficient; a unit-channel convention |g3 | = 243/2 /2 = 58.78775383 . . . would be a new parent-action axiom, not a derivation. The remainder is named and typed: Named datum 5.2 (Π5←6 , loaded). The chamber-to-AnnTL placement/spectral kernel is neces- sarily non-equivariant: any construction must break S4 , so it is loaded data, not derivable atlas structure. It welds the transfer question to the S232 loaded-boundary frontier — the same loading that carries [σCD ]/η+ must carry Π5←6 , or nothing does. 6 Status language (strict) and the closing ledger • Theorems (in-house, executed, bit-portable): finite-R Riesz edge projector; free-measure no-go; exit-edge alignment Eedge = 2φ2 = Eth ; the even-only welded-metric doorway; the 10=10 equivariant no-go; g3 structurally free; and the relative sign −1 (amputated, F4-fenced). • Derived-conditional: the η+ weld (on the loaded holonomy class); attraction (on a positive exit gap); the odd-protection projection (on the positive-exit metric). • Named-open data (the wall, final form this arc): one rank-7 channel readout (built) + three continuous data {g3 , Csub , rC } (equivalently Nabs + rC + one convention) + the loaded kernel Π5←6 . Object Tier Note [σCD ] holonomy + η+ Cartan metric one merged conditional import weld, §3 additive locality; positive-energy transfer standing structural imports §1–2 g3 (cubic normalization) continuous, structurally free §5 Csub (additive subtraction) continuous; Nabs invisible §5 rC (locking form factor) continuous, open κ = 1 + |r|, tC (δ) = rC δ Π5←6 (placement kernel) loaded, non-equivariant §5 6 ===== PDF PAGE 300 / 433 ===== Bottom line. The head wall (“no parent action”) was attacked head-on and cracked along its typed seams. The transfer question is no longer one fuzzy object: it is a merged loaded import ([σCD ] holonomy welded to its induced metric η+ ), three continuous constants, one of which (Nabs ) is provably invisible from the boundary, a built rank-7 equivariant readout, and one loaded non-equivariant placement kernel Π5←6 . On the way it produced the program’s first derived, parameter-free (discrete, derived) relative vertex sign — the −1 — kept strictly amputated and F4-fenced. The “no dynamics” category is no longer a category; what remains is one loaded kernel and three constants, each with a name, a type, and a test. Conditional/candidate tier throughout; nothing here is a world-scored observable. 7