ec44b0af369d57eae8062e68cae9cd983f56e54b3f9d1aee4fe0046a980c1469, and this page is generated from exactly those bytes — the build refuses to run if they do not match the served MANIFEST. Every page below is its own anchor, so a Registrar record can link to the page it cites instead of to the top of a 300 KB file. ★ Built because a cold review lost six of fourteen tendon verdicts to unpinnable page markers (D1264 → A1593), and declined to guess rather than mis-cite — which was the right call and the house's cost to bear.101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
# SGTOE Rev33.1_S369 suite — PAGE-PRESERVING TEXT EXTRACTION — PART 2 (pp 101–200) (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.
Coordinate SU (5) tradeoff theorem [proved, A683C]. An exhaustive search over all ten
weak-pair/color-triple index assignments in S+ = Λeven C5 establishes:
4∈
/ W ⇒ [EW , Γ4,4 ] = 0 (weak doublets intact), but [EC4 , Γ4,4 ] ̸= 0 (color triplets split). (22)
4 ∈ W ⇒ [EC , Γ4,4 ] = 0 (color triplets intact), but [EW 4 , Γ4,4 ] ̸= 0 (weak doublets split). (23)
No coordinate SU (5) permutation preserves both SU (3)c triplets and SU (2)L doublets inside the
chirality eigenspaces simultaneously. The D682 obstruction (weak doublets split) cannot be removed
by a simple relabelling of the five exterior-algebra indices; it is merely traded for a color-triplet
splitting. A non-coordinate intertwiner is required.
The two remaining open paths are: (A) the Peirce projector Lc1 (requires explicit J3 (Os ) ,→ S+ basis
map, D683A/D684); and (B) the Jvac -stabilizer Spin(4, 4) chirality, whose correct explicit form has
not yet been computed (D683B/D684).
The SO(5, 5) branch and the Peirce singlet [proved, A684]. The split E6(6) branching
27 ↓ SO(5, 5) × SO(1, 1) = 16−1 ⊕ 10+2 ⊕ 1−4
admits three conjugate realizations, one per primitive idempotent. For the c3 -selected branch:
16 = J13 ⊕ J23 , 10 = J12 ⊕ Rc1 ⊕ Rc2 , 1 = Rc3 .
This matches the suite’s established assignments (tR ∈ J13 , J12 for CKM, J23 for gauge/CP). Note:
the singlet is the primitive idempotent c3 , not the trace direction c1 + c2 + c3 . The trace singlet arises
in the different decomposition 27 ↓ F4(4) = 26 ⊕ 1; these are distinct subspaces. The nilpotent blocks
J13 and J23 are exactly the fermionic carriers “connected to c3 ” via the Peirce eigenvalue condition
c3 ◦ X = 12 X for X ∈ J13 or J23 .
Intertwiner status and dimension obstruction [open, A684]. Abstractly, the Peirce chirality
on the 16 is
ΓPeirce = Π13 − Π23 , (ΓPeirce )2 = 1.
To write this operator in the exterior-algebra model S+ = Λeven R5 one needs the explicit basis
intertwiner B : J13 ⊕ J23 → Λeven R5 . Non-coordinate involutions commuting with SU (3)c × SU (2)L
do exist (kernel-verified, A684): for example ΓSM = +1 on {QL , L} and −1 on {uc , dc , ec , ν c }. However,
a dimensional obstruction limits the interpretation of J13 as an SM-invariant carrier:
dim QL + dim uc = 6 + 3 = 9 > 8,
so no 8-dimensional SM-invariant subspace can contain both a full quark doublet QL and a full
up-type singlet uc . Consequently J13 is best understood as a Peirce/Freudenthal Yukawa carrier—the
algebraic channel through which the direct Yukawa coupling is selected [proved, A674]—rather than
as an SM-representation container.
Explicit SM-commuting Clifford chirality [proved, A685]. The first explicitly proved SM-
commuting 8 + 8 chirality in the exterior-algebra model is
ΓQL := Γ3 Γ4 Γ8 Γ9 , Γ2QL = 1, [ESM , ΓQL ] = 0
11
for all pure SU (3)c × SU (2)L generators (kernel-verified, A685). Its eigenspaces realise a left–right
split:
(+) (−)
S+ = QL ⊕ L (all SU (2)L doublets), S+ = uc ⊕ dc ⊕ ec ⊕ ν c (all SU (2)L singlets). (24)
This operator evades the coordinate SU (5) tradeoff theorem (A683C) because it is non-coordinate.
Identifying ΓPeirce = ΓQL is a structural choice [structural]; it is not yet derived from J3 (Os ).
The correct next target is not a one-block SM-irrep identification but a trilinear projection theorem:
within the full spinor J13 ⊕ J23 , one seeks the coupling
J13 × J13 −→ QL · ΦH · uc
(+) (−)
such that QL ∈ S+ and uc ∈ S+ , consistently with the proved Peirce–Yukawa direct-channel
placement.
The 16 × 10 × 16 trilinear and the Peirce Yukawa channel [proved in exterior algebra,
A686]. In the c3 -selected branch the SM top Yukawa is realised as the exterior trilinear
T (ψ, H, χ) = Top(ψ ∧ H ∧ χ) , ψ ∈ Q ⊂ 8+ (ΓQL ), H ∈ (1, 2)+1/2 ⊂ 10, χ ∈ uc ⊂ 8− (ΓQL ),
giving
T (Q, Hu , uc ) ∝ εabc εαβ Qaα H β uc,bc [PROVED],
kernel-verified with the exact Standard Model index structure and six nonzero contractions (A686).
The Higgs lives in the vector 10 = J12 ⊕ Rc1 ⊕ Rc2 (acting on S+ by Clifford multiplication), not in
8± . This is a cross-block trilinear,
J13 × J12 × J23 → R,
distinct from the same-block Freudenthal bilinear J13 × J13 proved in A674. The same-block theorem
remains the Peirce unit-eigenvalue Yukawa-channel selection (extracting λ2 = 1); it is not the literal
QHu uc invariant. The Jordan triple product {J12 , J13 , J23 } also extracts λ2 = 1 in the real scalar
Peirce subalgebra [proved, A686], forming a structural bridge between the two descriptions. The
full generic Os cross-block normalization is now resolved; see below.
Cross-block Peirce trilinear [proved, A687]. For generic split-octonion entries x, y, z ∈ Os , let
X = J12 (x), Y = J13 (y), Z = J23 (z). The Jordan triple product
{X, Y, Z} = (X ◦ Y ) ◦ Z + (Z ◦ Y ) ◦ X − (X ◦ Z) ◦ Y
gives (kernel-verified, A687)
{J12 (x), J13 (y), J23 (z)} = τ (x, y, z) c2 ,
where the split-octonionic triality trilinear is
τ (x, y, z) = Re((x̄ y)z̄) = Re((yz̄)x̄) = Re((xz)ȳ) .
Consequently,
Tr[Jvac ◦ {J12 (x), J13 (y), J23 (z)}] = λ2 τ (x, y, z) = τ (x, y, z),
12
with λ2 = 1 because J13 occupies the middle slot, selecting the central idempotent c2 . The real part
of the split-octonion associator contributes no independent correction: Re([x, y, z]) = 0 in the relevant
trace (A687). The correct SM unit-eigenvalue ordering is therefore
{J12 (H), J13 (ΨL ), J23 (ΨcR )},
matching the 16 × 10 × 16 Yukawa structure proved in A686.
This cross-block trilinear is distinct from the same-block Freudenthal channel
Tr[Jvac ◦ (J13 × J13 )] = −2 Re(ψ h̄),
but both channels select c2 and extract λ2 = 1. The suite therefore has a two-channel Peirce Yukawa
theorem: the same-block channel (A674) is the unit-eigenvalue Yukawa carrier; the cross-block channel
(A687) is the literal SM 16 × 10 × 16 coupling. The exact component dictionary mapping Os entries
to exterior-algebra SM fields is now fixed for the top-quark sector; see below.
Component normalization of the cross-block Yukawa [proved under displayed orientation,
A688]. Using the SplitCD convention e21 = e22 = e23 = −1, e24 = · · · = e27 = +1, choose the orientation
Hu3 = e0 , Hu4 = e7 ,
Q0,3 = e1 , Q0,4 = −e6 , Q1,3 = e2 , Q1,4 = e5 , Q2,3 = e3 , Q2,4 = e4 ,
uc01 = e4 , uc02 = −e5 , uc12 = −e6 .
This is the split-real form of the compact color-triplet convention ua = (real part, oriented imaginary part).
A direct component audit (36 tests, 36/36 pass) gives
Re (H̄u Q)uc = εcij εαβ Huβ Qcα uc,ij [PROVED].
The Peirce cross-block triple therefore realizes the SM QHu uc Yukawa invariant with unit normaliza-
tion.
Full cross-block Yukawa normalization [proved under explicit B, A689]. The corrected
one-generation component dictionary is
Hd3 = e0 , Hd4 = −e7 (split-octonion conjugate of Hu ),
dc0 = e1 , dc1 = −e2 , dc2 = e3 , L3 = e0 , L4 = e7 , ec = e0 .
Direct component audits give zero mismatches for all three Yukawa sectors (A689):
QHu uc [PROVED], QHd dc [PROVED], LHd ec [PROVED].
The down Higgs is the split-octonion conjugate Hd = H̄u in the (e0 , e7 ) neutral plane; no additional
Os directions are required.
Hypercharge from the B-map [derived from B + SU (5), A689]. With the SU (5) exterior-
index assignment Y0 = Y1 = Y2 = − 13 , Y3 = Y4 = + 12 , the hypercharge of each exterior basis element
is Y (I) = i∈I Yi . All 16 elements of S+ = Λeven R5 give the correct SM value (16/16 matches, 0
P
mismatches):
Y (Q) = + 16 , Y (uc ) = − 23 , Y (dc ) = + 13 , Y (L) = − 12 , Y (ec ) = +1, Y (ν c ) = 0.
The previous HY1 field-by-field projector axiom is accordingly replaced by the B + SU (5) index
assignment. The remaining open item is not the individual hypercharge values, but the intrinsic
J3 (Os )-based canonical selection of the B-map orientation.
13
Canonical status of the B-map (A690). A finite audit (kernel d690_canonical_B_audit,
re-run in house) shows that the present J3 (Os ) data do not uniquely select B: there are 192 signed-
permutation automorphisms of the SplitCD table, and applying any of them simultaneously to
every B-component preserves all QHu uc , QHd dc and LHd ec tests (6/6, 6/6, 2/2; zero mismatches)
and the full hypercharge table. This is not accidental: the Yukawa contractions are values of the
Aut(Os ) = G2(2) -invariant trilinear τ (x, y, z) = Re((x̄y)z̄), so the admissible frames form (at least)
the global automorphism orbit of B0 . Neither Jvac , the cubic norm N , nor the Peirce eigenvalue
ladder (ϕ, 1, ϕ−1 ) pins the orientation, since all three are blind to the internal Os basis inside a fixed
Peirce block. The honest label is therefore
one-generation SM Yukawa + YSM [PROVED under explicit B],
B canonical from J3 (Os ) alone [NOT PROVED],
with the residual reduced to a single frame axiom AB (“choose the oriented SplitCD/SU (5) frame”),
P
strictly weaker than HY1: AB fixes one basis, after which every charge follows from Y (I) = i∈I Yi .
A continuous double-check (A692, kernel-verified in house) sharpens this. Among e0 -fixing signed-
permutation frames the valid set is exactly Aut(Os )⊔Aut(Os )◦σ with σ = diag(1, 1, 1, −1, 1, −1, −1, −1)
a harmless global field rephasing. But the 76 Yukawa component equations do not pin the frame
even to G2(2) : inside the e0 -fixing SO(4, 4) frame space their infinitesimal stabilizer has dimension
15 = dim g2(2) + 1, the extra direction being the non-compact boost e3 7→ cosh t e3 + sinh t e4 , e4 7→
sinh t e3 + cosh t e4 , which preserves all 76 tests but is not a split-octonion automorphism. The residual
axiom is therefore
AB : B is the G2(2) -compatible oriented SplitCD/SU (5) frame
— i.e. B must intertwine the full split-octonion product, not merely the SM Yukawa tensor (this
removes the SO(1, 1) boost). AB is strictly weaker than HY1 (a frame choice, not a charge table)
but remains an axiom until a J3 (Os )-intrinsic criterion selects the G2(2) -compatible frame.
Compact gauge kills the boost (A707); the global frame is canonical (A708). The residual
SO(1, 1)34 boost is excluded by the compact Standard-Model gauge structure the suite already uses
(A707, kernel-verified in house with the core re-derived independently): it preserves the split (4, 4)
metric but not the positive-definite color norm (|e3 |2pos : 1 → cosh 2t = 3.76), mixes the J23 fields
dc (Y = + 13 ) and uc (Y = − 23 ), and fails to commute with SU (2)L . A global finite audit (A708;
kernel re-run in house and the count Aut(Os ) = 192 re-derived from a from-scratch Cayley–Dickson
table) exhausts the e0 -fixing signed-permutation frames (7! 27 = 645,120): the 76-valid set is exactly
Aut(Os ) ⊔ Aut(Os )◦σ (384 = 192 + 192), and once the compact-gauge frame is imposed the valid finite
frames reduce to the e7 -line-preserving subset (96 = 48 + 48) — a single component after quotienting
by G2(2) , compact SU (3)c × U (1)Y × SU (2)L , and the σ-rephasing. The two natural discrete suspects,
color conjugation 3 ↔ 3̄ and the singlet swap e0 ↔ e7 , both fail validity outright (e0 is the octonion
identity). No physically distinct B-frame survives, so the former axiom AB is discharged relative to
the compact SM gauge embedding:
so(1, 1)34 compatible with compact SM gauge [PROVED NO],
AB [DISCHARGED rel. compact SM embedding].
The one-generation Yukawa + YSM system therefore reads [proved relative to the compact SM gauge
embedding]: the qualifier now names a physical input (the compact embedding — not derived from
14
J3 here; the former “gauge-emergence frontier” is now settled as a boundary: emergence-as-gauging
is closed negative by the branching-wall theorem, the completion is unique up to U (1)B−L and global
structure, and the maximal derivable shadow is dynamically selected — see Appendix F) rather than
a frame ambiguity. One technical caveat remains: a formal theorem over arbitrary disconnected
O(4, 4) frames beyond the signed-permutation class is not yet formalized. [LIB2-318]
PART II — HISTORICAL Rev8 MATERIAL (superseded, non-operative in
Rev29)
Everything from here to the end of Paper 6 is the Rev8 electron-mass bridge
programme, retained for the historical record only. It is superseded by the QED-
corrected Koide chain (me = 0.5076 MeV, Koide-consistent ⋆ ⋆ ⋆⋆ under external K = 2/3
[A636, A637]) and by the A674–A689 Peirce-Yukawa results (above). None of the sections
below represent the operative Rev29 state; they are not part of the current claims of the
suite.
5 Electron-mass baseline and residual (Rev8 historical)
The current baseline electron formula is
π
√
mbase
e = MPl exp − ϕ4 2 = 0.5036 MeV. (25)
8α
Relative to the physical value
mphys
e = 0.5109989461 MeV, (26)
the baseline residual is
mphys
e − mbase
e
∆obs
e = = +1.454%. (27)
mphys
e
The algebraic form of Eq. (25) is currently taken seriously. The issue is not whether the baseline can
be written compactly; the issue is whether the remaining percent-level bridge is a derived consequence
of the same Lagrangian or a phenomenological closure layer built around the known electron mass.
6 Phase decomposition of the electron residual
Rev8 carries five correction phases. Their physical interpretations are motivated by structures already
present in the suite, but their precise percentages are not yet derived from the Lagrangian. The
correct status is therefore Phenomenological Ansatz.
Table 1: Rev8 phase decomposition of the electron residual.
Phase Central value Uncertainty Current source Rev8 reading
KK threshold match- +0.50% ±0.10% Paper 6 threshold physically motivated,
ing sector not derived
15
Phase Central value Uncertainty Current source Rev8 reading
RGE bridging −0.048% ±0.030% standard one-loop
kernel-fixed bridge
running term; not computed
from Feynman dia-
gram or matching
condition
Möbius topology +0.60% exact in the cur- winding / half- topologically moti-
rent topological twist picture vated, percentage not
normalization derived
Associator correction +0.15% ±0.06% non-associative physically motivated,
sector coefficient not derived
Electroweak thresh- +0.102% ±0.041% low-energy match- physically motivated,
olds ing cascade coefficient not derived
The central values sum to
∆phases
e = +1.304%, (28)
which is smaller than the observed residual in Eq. (27). Combining the quoted uncertainties in
quadrature gives
p
σphases = 0.102 + 0.0302 + 0.002 + 0.062 + 0.0412 ≈ 0.127%. (29)
This is smaller than the observed mismatch between Eqs. (27) and (28). The suite therefore does not
claim that the current five-phase table explains the electron mass completely.
7 The unresolved 0.15% remainder
The unexplained remainder is
∆miss
e = ∆obs phases
e − ∆e = +0.150%. (30)
At present there are three live interpretations:
(a) a missing sixth phase associated with structure not yet tracked in the bundle,
(b) an underestimated uncertainty or incomplete treatment in the RGE/threshold pieces,
(c) framework incompleteness in the present electron closure mechanism.
Rev8 does not decide among these options. It records them as the real sources of current uncertainty
rather than as rhetorical afterthoughts.
8 Necessary or sufficient?
The five phases are currently best interpreted as sufficient phenomenological descriptors of the known
electron residual, not as necessary outputs of the Lagrangian. This distinction is crucial. A necessary
output would be fixed uniquely by the algebraic and field-theoretic structure before the electron mass
is consulted. A sufficient descriptor is a physically motivated decomposition that reproduces most of
the observed residual but still depends on post hoc calibration. Rev8 adopts the second description.
16
9 Muon/tau validation protocol
The next decisive test is whether the same cubic-invariant logic and the same phase structure extend
to the heavier charged leptons without adjustment.
8.1 Operational protocol
(1) Extract yµ (MPl ) and yτ (MPl ) from the same cubic-invariant/Jordan bookkeeping used for the
electron baseline.
(2) Apply the same standard one-loop running used for the electron bridge.
(3) Apply the same five phases with no lepton-specific retuning.
(4) Compare the resulting mµ and mτ predictions directly to the PDG benchmarks.
8.2 Interpretation of outcomes
Table 2: Interpretation of the muon/tau validation outcomes.
Outcome Interpretation Rev8 consequence
Both within about framework behaves predic- strong case for structural phase sys-
1–2% with no new tively across the charged- tem
phases lepton family
Require only small framework may be struc- retain cautiously as structural can-
RGE/threshold refine- turally right but quantita- didate
ments tively incomplete
Need adjusted phases phase system behaves like downgrade electron closure to
for each lepton phenomenological tuning framework-specific ansatz
Miss badly or require electron success was not uni- record boundary and reclassify
new topological data versal mechanism
17
10 Status
Rev29 status (Sessions 16–23), reworded Rev32. The operative tree-level weak-angle
readout is sin2 θW = ϕ−3 under the polarisation postulate AX6pol (Paper 4 §6, which closes the
earlier 3/13 estimate; this paper’s Weak-angle bridge section already marks 3/13 superseded
— Rev32.9: through Rev32.8 this box still gave 3/13 as the target); it is a structural target
under that independent postulate (it is not citable as algebraically derived without that
named premise — Appendix X); threshold corrections are open (no spectrum computed).
The electron-mass phase system is a phenomenological ansatz, not a Lagrangian consequence.
The electron baseline inherits the AX2 vacuum (now proved as a theorem); if that theorem
were invalidated, the baseline formula would require revision. CKM sector (Sessions
22–23, historical marking — superseded by the Rev29 re-tier and the A1566
retype: first row Loaded, |Vcb | Loaded-correspondence attachment (physical) /
Structural formula, δCKM Coincidence-class): three then-Proved-tier magnitudes
+ one conditional-theorem phase. The three magnitudes (|Vus |, |Vcb |, |Vub |) follow from
the J3 (Os ) Route-B map and agree with PDG within 1σ (Proved-tier, the six-of-eight
mixing theorem; Appendix X); δCKM is Coincidence-class / candidate construction (value
retained, G7 selection absent; the Rev29 “conditional theorem under F-δ” wording retired
Rev32.5; Appendix B.6). The phase origin is algebraic, not RGE/threshold (D594); the earlier
“100/100, all green” framing is retired (S112 claim-hygiene). Muon/tau no-retuning test:
extending the same five phases to muon and tau without lepton-specific adjustment currently
fails at ∼ 111σ. This is the decisive test for structural validity of the phase system. The
decisive next step is independent muon/tau validation with no new percent-level tuning.
11 CKM Dirac phase: RGE and threshold contributions ruled out
Sessions 22–23 marked the CKM magnitudes Proved-tier within Route B — a marking superseded
by the Rev29 re-tier and the A1566 retype (first row Loaded, |Vcb | Loaded-correspondence
attachment (physical) / Structural formula, δCKM Coincidence-class; the phase remains
readout-conditioned) — and explicitly tested whether standard QFT effects could account for the
CKM Dirac phase. The result is a clean structural division.
9.1 RGE and threshold running ruled out (D594, A576)
Grok dispatch D594 tested whether renormalization-group running and top-quark threshold corrections
could shift the naive tree-level δCKM prediction into agreement with the PDG value. The verdict
was unambiguous: neither αs running from MZ to Mth , nor top-quark decoupling thresholds, can
produce a correction of the right sign and magnitude. RGE contributions to δCKM are ruled out as
the explanation.
9.2 Algebraic origin: G7 = ϕ + 5ϕ−5
The correct explanation is purely algebraic. The sub-leading eigenvalue of J3 (Os ) receives a correction
ε = 5ϕ−4 from the n = 5 Peirce-graph loop, giving
q
G7 = ϕ + 5ϕ−5 , δCKM = arctan 3(ϕ + 5ϕ−5 ) = 68.1297◦ . (31)
18
(Comparison: 0.8σ from the γ-fit; 1.6σ from the PDG direct value.) The factor 5ϕ−5 arises from
the split-octonion (4, 4) signature of Os : 5 of 8 octonion directions contribute to the Peirce coupling
(compact e1 , e2 , e3 cancel; scalar + non-compact e4 –e7 survive). This is not a QFT effect; it is a
consequence of the Albert-algebra multiplication rules (Structural Result 2.1 in Paper 3; E96–E98,
A579).
9.3 Status: current CKM provenance and PDG 2026 comparator record
Table reading rule. Rev32.1 reading rule: tiers name mathematical provenance; physical attach-
ments are stated separately; no row may be cited as zero-parameter; displayed σ values are named
comparator distances, not profile likelihoods.
Quantity Algebraic value PDG 2026 comparator
√
PMNS / 6 = 0.225256 (s1162)
|Vus | sin θ12 0.22431 ± 0.00085
√
|Vcb | 1/(9 7) = 0.041996 0.0407 ± 0.0013
√
|Vub | |Vus ||Vcb |/ 6 = 0.003862 0.00389 ± 0.00016
arctan 3(ϕ + 5ϕ−5 ) = 68.1297◦ global-fit δ = 1.154 ± 0.025 rad (= 66.12◦ ± 1.43◦ ; radians primar
p
δCKM
The three CKM magnitudes were marked Proved-tier within the Route-B map pre-Rev29; after the
row-by-row re-tiering (Rev29 re-tiering, Paper 3 § “Row-by-row re-tiering”; kernel s959) the CKM first
row is Loaded because its route was selected against kaon data, |Vcb | is Loaded-correspondence
(physical attachment) / Structural formula (A1566), and the Dirac phase δCKM is Coincidence-class
while the motivation for G7 is missing — not a 100/100 closure of the matrix. No fitted mixing
coefficients enter the Route-B map, but no row may be cited as zero-parameter. The decisive
implication for Paper 6 is that the CKM phase does not belong to the RGE/threshold bridge: it is
fixed by the exceptional-Jordan algebraic structure before any running is applied (no QFT input
beyond the tree-level algebra).
12 Conclusion
Paper 6 keeps the suite honest in two places. In the electroweak sector it states the weak-angle
bridge as an ordinary QFT running-and-threshold problem. In the charged-lepton sector it states
clearly that the current electron residual closure is not yet a derived Lagrangian consequence. The
framework now has a clean next test: if the same logic predicts mµ and mτ without retuning, the
present electron mechanism strengthens sharply; if not, its domain of validity is revealed.
19
Leibniz Quantum Beats Newton
Paper 7: Precision Tests and Failure Conditions (Rev33.1)
Tom O’Sieg
August 2026
Abstract
This paper is the public falsification note for the current (Rev29) J3 (Os ) programme; “Rev29”
below names the revision in which a given result first entered the suite, not the revision of
this document. It collects the measurements that can most directly confirm or exclude the
current exceptional-Jordan correspondence and states them without protective ambiguity. The
three highest-priority external tests remain the leptonic Dirac phase, the atmospheric mixing
angle, and the weak-angle bridge. Rev29 S43 advances the internal electron-mass validation
test: me = 0.5076 MeV is Koide-closed ⋆ ⋆ ⋆⋆ via the QED-corrected Koide chain [A636, A637],
conditional on the external empirical rule K = 2/3 (not derived from J3 (Os ) alone). The current
open internal test is the formal S-matrix derivation of the 7/3 lepton–quark amplitude bridge
[A640].
1 Introduction
A theory that claims algebraic compression of particle-physics data should state how it can fail. Paper
7 exists for that purpose. The programme asks to be judged by measurements and by clean internal
cross-checks rather than by rhetoric.
2 Primary falsification benchmarks
1
Table 1: Primary falsification benchmarks (entered at
Rev29). Normal ordering is assumed throughout; the ordering-
dependent rows are additionally conditional on the Struc-
tural Rd ladder and on the undeclared ladder-to-propagation-
eigenstate interface (Paper 5 §2). Rev32.1 reading rule:
tiers name mathematical provenance; physical attach-
ments are stated separately; no row may be cited as
zero-parameter; displayed σ values are named com-
parator distances, not profile likelihoods.
Quantity Framework target Measurement scenario Failure condition
δCP −160.997◦ (JCP ≈ NuFIT 6.1 NH (w/SK): exclusion at ≥ 3σ by
−0.011) [Derived- 212+26
−36
◦ (−160.997◦ ≡ DUNE Phase II global
conditional, Rev29 199.003◦ ); current devi- fit
re-tier — pre-Rev29 ation 0.36σ [excellent];
marking THEOREM DUNE-era O(10◦ ) at 1σ
⋆100 superseded] (2030s)
θ23 41.41◦ (sin2 θ23 = NuFIT 6.1 NH (w/SK) stable exclusion of the
7/16 = 0.4375) global best fit: 43.3◦ 7/16 branch at ≥ 3σ in
(sin2 θ23 = 0.470+0.017
−0.014 , global fits (this attacks
lower octant, 2.3σ from the selected physical cor-
7/16; upper-octant lo- respondence, Loaded-
cal comparator ≈ 0.561 correspondence since
at ≈ 9.5σ; 3σ range A1566; the internal 7/16
includes 7/16); DUNE- mismatch identity is not
era few-degree precision refuted by it unless the
(2030s) readout attachment is
first derived); note: the
NuFIT 6.1 global best fit
is itself lower-octant
2
Quantity Framework target Measurement scenario Failure condition
√
sin2 θW bridge ϕ−3 = 5 − 2 ≈ ϕ−3 lies on the run- failure to supply a first-
0.23607 as a ∼2% ning curve at a principles µ⋆ (or a com-
structural low-energy sub-GeV/GeV-order, pleted bridge) to the Z-
target on the effec- scheme-dominated pole observable
tive sin2 θW (µ) run- crossing (corrected
ning curve (A697; Pa- S124/Appendix I §B4;
per 4 §7 authoritative) the once-quoted
µ⋆ ≈ 1.8 GeV lies
outside the band);
the MZ difference is
standard running (right
direction). matching
scale/scheme not de-
rived ⇒ structural
target, not a Z-pole
prediction; AX6 fixes
the independent orbit
branch
3
Quantity Framework target Measurement scenario Failure condition
√ √
mµ /mτ (ϕ/ 5)8/3 × 2/10 = mµ = 106.05 MeV Miss > 2% with-
0.059684. Struc- (+0.37% PDG). The out retuning would
tural / Loaded- +0.37% residual is invalidate the loaded-
correspondence / the fit gap, not a correspondence mass
Reproduced (Rev29 loop correction: the formula; the row
re-tier; this row previ- coefficient required to may not be cited as
ously printed Proved, reproduce the measured zero-parameter
zero-parameter, ratio is C = 0.140894
“forced by algebra, no against
√ the offered
fit”). The exponent C = 2/10 = 0.141421,
8/3 = dim Os /rank J3 and C enters the for-
is an exact dimen- mula multiplicatively, so
sion ratio but is the discrepancy passes
not, alone, a mass straight through to
exponent (Paper 0 mµ /mτ at the same
§ “physical attach- +0.37%
ment”);√the coefficient
C = 2/10 divides
by dim J2 = 10,
a step A618 itself
records as “argued
by analogy,” and
A618 lists the formal
derivation as Block-
ing — never closed.
Provenance is target-
first: D616 computes
0.0595/0.4220 = 0.141
from PDG and asks
whether it is algebraic
√ φ; A616 then offers
in
2/10 at 0.37% while
stating that “the
algebraic origin is not
yet proved”; A618
supplies dim J2 = 10
afterward. [A616,
A618]
4
Quantity Framework target Measurement scenario Failure condition
AX6pol : Λ⋆ = Independent orbit- A667 degree argument Derivation of AX6 from
3P/Q(Y ) = 6ϕ − 1 selector postulate; the (Det2 object only) plus first principles, or failure
(readout readout is an identity earlier negative CW ver- to anchor the structural
3/(4 + Λ⋆ ) = ϕ−3 given the postulate, dicts [A617,A620,A621] target ϕ−3 (via a first-
exactly); the and does not select it principles µ⋆ ) to the Z-
Det 2 object pole observable
cos(6ψ) = 12 is a
distinct, retired
selector (Rev32.2)
Pure µ–γ associa- aµ = aSMµ to all orders 2025 Theory Initia- confirmed non-hadronic
tor null in the lepton–photon tive: aexp
µ − aSMµ = aµ anomaly after
sector (Artin theorem −11
38(63) × 10 ; HLbL HVP/HLbL resolution,
on ⟨e0 , e7 ⟩; A721) = 115.5(9.9) × 10−11 ; or a required associator
HLbL | ≲ 0.55
implied |Cassoc coefficient |C| ≫ O(1)
(conservative) or ≲ 0.086
(HLbL budget); normal-
ization FIXED to the
LbL class (A725) — no
dipole-class escape
√
vEW / σ (Rev26) φ105/8 = 553.30 [CON- registration relative to a determination of
√
DITIONAL / loaded- the marked compact- vEW / σ inconsistent
depth; Appendix O] positive Higgs carrier with φ105/8 beyond the
√
— framed-depth oper- (Paper 4); the frame- σ registration uncer-
ator D = 13 + 18 ; free route is closed nega- tainty, or exhibition of a
√
mτ / σ = 4 exact tive (Appendix F perma- second genuinely under-
√
companion identity nence); σ registration ivable dimensionful ratio
carries ±1.51% in scope (which falsifies
the one-input theorem
chain, Appendix X/O)
Stamps in this table ( ⋆100, THEOREM, Proved) are the Rev29-era marking, superseded en bloc by the Rev29 re-tier
(Paper 3 §“Row-by-row re-tiering”); the current tier of each row is the re-tier table’s. Footer added Rev32.5.
3 Secondary tests
RGE scale convention for mixing targets. The PMNS and CKM angle targets are quoted at a
single canonical comparison scale, which the suite takes to be the electroweak scale MZ unless explicitly
stated otherwise. In the Standard Model, running of CKM angles between MZ and a GUT scale
is numerically small because the dominant anomalous-dimension terms are flavour-universal within
a fixed representation; for hierarchical Dirac-neutrino masses the PMNS angle running is likewise
sub-percent, typically at the O(10−3 ) level. The algebraic angle values are therefore compared to
low-energy global-fit values without adding a new fitted RGE offset. A future high-precision treatment
should add a dedicated uncertainty line for scheme, threshold, and normal-ordering assumptions;
such corrections are subleading relative to the present DUNE-era falsification bands.
5
The relation yt = 1 is a falsifiable boundary statement, not a fitted threshold: imposed as a UV
boundary condition it defines a running √ curve yt (µ) that is compared with the extracted coupling
at a stated scale and scheme (the v/ 2 tree value sits at +0.87% vs. PDG 2026, naive cross-
scheme comparator distance dcmp = +5.57 — not a tension, Appendix X; the authoritative negative
running/matching audit is Paper 2 §3.6). Rev32.1 (A1538 F4): the “≥ 19σ consistency at every scale”
clause previously printed here is withdrawn as ill-formed — like the transposed “+5.3σ” admitted
alongside it, a pipeline-string defect; the s450/S193 run record and its 0.17 GeV pipeline residual are
carried unaltered in the kernel ledger [s450/S193].
Several additional quantities provide cross-checks:
(1) charged-lepton and down-type mass ratios from the resolvent family,
(2) CKM hierarchy relations,
(3) CKM first-row unitarity check,
(4) internal consistency of the cubic-invariant Yukawa construction,
(5) formal S-matrix derivation of the 7/3 lepton–quark amplitude bridge,
(6) the shadow charge-pattern match (S77 designated test — executed S78, fence fired): the residual
so(6) ⊕ so(2) of the dynamically selected SO(6, 2) point hosts su(3) ⊕ u(1) ⊕ u(1) (Appendix F);
the framework-internal check was whether the two u(1) charge spectra on the 56 reproduce the
derived YSM and 3(B−L) patterns. Outcome: mismatch at the representation level at the origin
point (octet present, one singlet vs nine; exact Casimir clusters), so the pre-registered demotion
applies there — “stable su(3) ⊕ u(1)2 point selected” — with the obstruction and uniqueness
theorems untouched; and the apparent moduli-scan escape was then closed by counting (S78b):
the entire 56 carries only four su(3)-singlets under the residual class, versus the nine the matter
pattern requires in a 27-plane, so the demotion is final for the whole SO(6, 2)/SO∗ (8) family. The
charge wall question — can any gauging’s gauge algebra (hence any vacuum residual) contain
a unitary-class su(3) (8 = 3 ⊕ 5 · 1)? — was subsequently executed (S80): the global negative
holds, the second wall theorem (Appendix F), house-proved and blind-replicated. Only vector-
and adjoint-class su(3)’s are ever gauging-realizable; the labeled shadow is excluded for every
gauging, and the demotion above is global and final.
The 7/3 bridge as an exact high-scale matching boundary: a falsifiable 4321/twin-
Pati–Salam prediction
The 7/3 lepton–quark ratio is an exact high-scale matching boundary: the bare-vertex unit-weight
condition KB = I7 is forced by the cubic Jordan norm, and the I8 → I7 reduction is exact under
Standard-Model covariance (S146, A1016–A1020). This is a relation imposed at the matching scale,
not a closed low-energy mass theorem; whether it survives the running to laboratory energies is
a separate, testable dynamical question. The minimal ultraviolet completion that realizes the 7/3
channel structure is a 4321 / twin-Pati–Salam model (SU (4)×SU (3)′ ×SU (2)L ×U (1)), which predicts,
as a concrete and falsifiable low-energy signature, a TeV-scale vector leptoquark U1 ∼ (3, 1, 2/3)
together with a Z ′ and vector-like fermions. This completion is not parameter-free, and the competing
realization is a tuned Standard-Model Yukawa pattern (21/29, 85/29); the framework does not at
present select between them.
6
Falsification statement. A confirmed absence of any such TeV-scale U1 leptoquark / Z ′ / vector-like
sector up to the energies where the 7/3 high-scale relation would have to be matched by tuned
Yukawas, combined with no first-principles derivation of the matching scale, would show that the
7/3 ratio survives only as a high-scale relation with no laboratory imprint. The 7/3 ratio retains
Structural ⋆⋆⋆⋆ status: the in-house advance toward deriving the channel-count axiom Tr3 (W ) = 7
(s191) is not yet hostile-reviewed and is therefore not treated here as a closure of the bridge.
Failure dependency tree
Failure Layer attacked
√
δCP miss at ≥ 3σ V3 rotation mechanism (Thalf |V3 , Ω = 5/2); does not
by itself attack AX1/AX3 (AX2 is eliminated by DET-7,
Paper 3) or the angle theorems
θ23 miss at ≥ 3σ first falsifies the loaded lower-octant registration (A1566);
attacks the internal Jvac versus Jvac# mismatch geometry
only if that physical attachment is independently derived
weak-angle bridge miss threshold-sector calculation and RG bridge; does not by
itself falsify the Jordan algebra core
7/3 amplitude failure If S-matrix vertex construction shows closure amplitude
̸= n23 /rank(J3 ) = 7/3; would falsify Structural ⋆ ⋆ ⋆⋆
status of mb and mc ; does not affect proved lepton masses
or mixing sector — the independence is structural, not
convenient: the bridge failure attacks the S-matrix real-
ization of the Gram ratio as a physical amplitude, while
the angle theorems use n23 = 7 only through the kine-
matic normalization sin2 θ23 = n23 /∥Jvac ∥2 (the inter-
nal identity; the physical octant attachment is Loaded-
correspondence, A1566), which a failed amplitude con-
struction leaves untouched. Rev27 sharpening (s902):
the metric-weighted-residue route is now a proven no-
go — a unitary residue over Im Os lies in [−4, +3], so
7 cannot be a signed amplitude; the 7/3 stands as an
index ratio (finite-carrier theorem, Appendix K), and
only a non-residue index-theoretic realization remains
open (Appendix I, Lemma)
heavy-quark T3 pattern failure If improved PDG quark masses break the equal-and-
opposite residual pattern |δmt /mt | = |δmb /mb | (cur-
rently equal to 0.67σ of the PDG 2026 mb uncertainty:
|κt − κb | = 0.194 pp vs σκb = 0.289 pp; a PDG central
value moving |κt − κb | past 1σ breaks it), the structural
weak-isospin T3 threshold reading (A715–A717) is falsi-
fied; attacks only the structural threshold term, not the
bare boundary values, the proved lepton masses, or the
mixing sector
7
QED-Koide chain failure If one-loop QED coefficient at ΛG2 differs from 3α/4π
(e.g. non-associative leading-loop correction), or Koide
relation breaks at required precision; would reopen me
derivation
charged-lepton mass-chain QED-Koide or muon-resolvent chain only; does not attack
miss the mixing sector unless the shared algebraic inputs fail
non-hadronic aµ anomaly con- If a confirmed aµ deviation survives HVP/HLbL res-
firmed olution and is shown to be non-hadronic, the associa-
tive lepton–photon embedding {µ, γ} ⊂ ⟨e0 , e7 ⟩ (A721)
is falsified; attacks the verified field-assignment map
(A688/A689), not the mass formulas or the mixing theo-
rems
light sterile neutrino confirmed The fermionic content of the 27 is exactly J13 ⊕ J23 = 16,
saturated by the 15 SM Weyl states + ν c ; the diagonal
idempotent slots carry no spinor content and J12 is fully
bosonic (A726). A confirmed extra color-singlet Q = 0
Weyl fermion (e.g. an eV-scale sterile resolving a short-
baseline anomaly) falsifies this 16-exhaustion; attacks the
representation content of the verified assignment, not the
mixing theorems
confirmed Majorana neutrino Falsifies the Dirac-neutrino benchmark assumption (T4-
mass (0νββ) D), not the PMNS mixing algebra: the bare singlet bilin-
ear M ν c ν c is gauge-invariant and algebraically available
within the existing 16, excluded here by choice rather
than by theorem [A729–A729c]. The non-gauging of B−L
is inherited from the compact rank-four input (S65 lad-
der); a B−L gauge boson or seesaw branch would require
a new vector plus breaking data outside the current input
extra long-range U (1)B−L A confirmed massless long-range force coupling to B −L
gauge boson falsifies the empirical half of the Rv -selection (no fifth
force) and would show Rv = adj(GSM × U (1)B−L ) was
physical; a confirmed massive ZB−L ′ instead falsifies the
structural no-breaking theorem (a B −L-charged scalar
VEV / 126-analog is absent from the on-record 27).
Attacks the Rv -selection residue (D730 [HYBRID], S66),
not the J3 (Os ) charge construction
4 Lepton-family validation as an internal benchmark
Rev29 upgraded the lepton-family validation picture substantially; Rev29 corrects the tier at which
that upgrade may be cited. The muon-to-tau ratio
√ √
mµ /mτ = (ϕ/ 5)8/3 × 2/10 = 0.059684 (+0.37% PDG)
is Structural / Loaded-correspondence √ / Reproduced, not proved and not zero-parameter.
Both p = 8/3 = dim Os /rank J3 and C = 2/10 are algebraic quantities built from Jordan-algebra
dimensions, but an exact dimension ratio is not by itself a mass exponent, and the divide-by-dim J2
8
step producing C is recorded in A618 as “argued by analogy,” with the formal derivation listed there
as Blocking and closed nowhere in the corpus. The provenance is target-first rather than derivation-
first: D616 first computes 0.0595/0.4220
√ = 0.141 from PDG values and asks whether that number is
algebraic in φ; A616 then supplies 2/10 as a 0.37% match while explicitly stating that “the algebraic
origin is not yet proved”; only afterwards does A618 identify the divisor as dim J2 = 10. Consequently
the +0.37% is not a residual awaiting a small correction — it is the fit gap √ itself : reproducing
the measured ratio requires C = 0.140894 where the framework offers C = 2/10 = 0.141421, and
because C multiplies the whole expression the 0.37% propagates unattenuated into mµ /mτ and into
mµ . The muon mass is mµ = 106.05 MeV (+0.37% PDG) at the same loading. Paper 0 already prints
the honest attachment row for 8/3; this paragraph brings Paper 7 into agreement with it.
The electron mass chain is now closed. One-loop QED matching gives δQED = +0.3125% and
mµ,phys = 105.72 MeV; with mτ = 1776.93 MeV the Koide constraint gives me = 0.5076 MeV (−0.66%
PDG). The remaining internal mass-sector test is not the electron, but the formal S-matrix derivation
of the heavy-quark 7/3 amplitude bridge.
5 Rev25 additions: the nuclear ledger, the selector falsifier, the
anomaly closure, and the island program
The nuclear cluster ledger (Paper 9). Rev25 adds a second world-facing precision theatre:
seventeen sharp binding-energy passes, deuterium through calcium, each against a pre-frozen bar, plus
one typed obstruction (28 Si: passed its pre-answer 1% bar and entered through the frozen disjunctive
refinement gate — neither a sharp nor a miss at a pre-answer bar), plus the Hoyle state, eleven
correct sign calls, and two glueball bands (Paper 9, with its protocol sub-appendix carrying the
complete receipt chain and honesty ledger). Its failure conditions are stated in Paper 9 directly: any
scored miss at a pre-answer frozen bar, any untyped miss at any frozen gate, a broken sign call, or a
vault-match failure under seal attacks the cluster kernel at paper tier. (Rev28 clarification, dated: the
Rev27 text scored 28 Si as a miss. That was an over-correction — the 0.1% figure was a post-answer
refinement gate whose frozen text was explicitly disjunctive and whose obstruction branch was taken
on the record. The pre-answer bar was 1% and was passed at 0.334%. The correction is made here
with receipts, not silently; the irreducibility claim itself remains falsifiable — a first-circulation fix
refutes it.)
The occupation falsifier {31, 39} (Appendix M). The rest-mass occupation selector, promoted
to Derived-Conditional at Rev25, carries a live, hash-stamped public kill condition: the selector’s
period-two continuation {31, 39} (sha 7fd5775e..., stamped before verification). An established
occupied rung outside the selector’s support kills the derivation and reverts the occupation to Input.
Anomaly audit: closed exactly (fold of s572/s588). All Standard-Model gauge and gravi-
tational anomalies cancel exactly from the framework’s own derived hypercharges — the complete
exact audit (s588, S214) upgraded the structural check (s572) and formally closed the hostile-board
item T3-P. Cross-reference: Appendix F (gauge boundary). As a falsification row: any demonstration
that a banked hypercharge assignment produces a non-cancelling SM anomaly would reopen T3-P
and attack the charge construction directly; the audit says no such demonstration exists against the
current bank.
9
The island-of-stability forecast (named program — not a result). The cluster kernel of
Paper 9 supplies closed-form ladder weights, which Paper 9’s declared cluster-ledger interface carries
to binding energies (Rev29 : predication corrected; the kernel prices a weight, not an energy), and two
magic numbers ({2, 8}) are native to the framework; the remaining shell structure is open. A forecast
of superheavy stability — the island — would require (i) the two-circulation sector (the typed 28 Si
boundary), (ii) leg scaling beyond ten cells, and (iii) the missing magic numbers. It is registered here
as future falsification theatre at program tier: no number is offered, and the program’s first scored
forecast, whenever it exists, must arrive with a pre-frozen bar like everything else in Paper 9.
6 Rev28 additions: the selection frontier and its falsifiers
The twenty-round fold recorded in Appendices S, T and X adds no world-scored number. It adds
named, checkable failure conditions, which are listed here so that they are falsifiable in the same sense
as the rest of this paper.
1. AX-MG. Exhibit a derivation of either clause — coefficient-one or degree-one — from the
banked boundary rows without inserting an edge-functor property (unitality, trace-preservation,
idempotence). Both clauses are currently proved independent by exact countermodel; a derivation
refutes the pricing and improves the ledger.
2. AX-COT. [P-003] Exhibit an isometric universal property tying chamber, Cartan spin factor and
occupancy simultaneously. Ordinary equivariance is proved insufficient; such a property would
discharge the clause off shell.
3. The observation functor. Exhibit an admissible observable that distinguishes two loaded
choices sharing a scalar character w = cφ−m , or that sees the signature transition through a
15-dimensional even readout. Either computes a piece of the faithfulness kernel and unblocks the
freedom total.
4. The onset. Derive n∗ = 7 from level structure. The current statement is that the level structure
does not select seven; a principle that does retires a priced integer marking.
5. Einstein universality. Compute the AdS linearized spectrum of the sourced heat-kernel action.
A spectrum inconsistent with linearized general relativity falsifies the gravity-measure programme
at the tier where its existence is already closed. [LIB2-301]
q mass attachment. Exhibit an interacting pole differing from the quasi-free gap MN =
6. The
Λ φ2N + 14 . AX-MASS-LABEL is a one-particle attachment and claims nothing beyond it.
7. The record. No falsifier is possible for the realized value: it is proved to be sample data. This is
stated as a limit of the theory’s claims, not as a result about the world.
What would not count. An improved χ-ladder expression fitted against the banked ratio
vector. The suite has adopted a standing rule against such rounds until a physical generator
and response map exist; the two cautionary rows are the off-support 0.0331562% kill and the
0.2744490308% legal miss, both recorded in Appendix X.
[LIB2-302]
10
7 Interpreting success and failure
Agreement with current data does not prove the framework. Disagreement in one of the primary
falsification channels would count heavily against the present correspondence map. The electron-
mass chain is Koide-consistent under the external rule K = 2/3 (S43) — an algebraic mass
correspondence with a dual-route reconciliation owed against the Appendix O two-gap operator,
not a joint derivation of the three charged-lepton masses (authoritative ledger: Appendix O, The
authoritative charged-lepton ledger). The current open mass-sector test is the S-matrix derivation of
the 7/3 lepton–quark bridge.
D4 closed
Rev29 closes the PMNS phase problem at the level of the operative framework target. The old ansatz
value −133.36◦ is superseded. [LIB2-277] The new target
2π
δCP = − √ = −160.997◦
5
is carried as a derived result with no free phase parameter inside the selected V3 construction; the
readout premises and the reactor forcing remain as stated elsewhere (Rev32.9). The corresponding
physical invariant is JCP ≈ −0.011. The failure mode
√ is therefore now sharply localized: a future miss
attacks the V3 rotation mechanism (Thalf |V3 , Ω = 5/2), not the entire Jordan-algebra programme.
Atmospheric-angle precedent
The atmospheric angle is a useful case study in real-time falsification. The current framework value is
the mismatch-geometry branch sin2 θ23 = 7/16, rather than the older higher-octant phenomenological
placeholder — printed since Rev32.6 as the sharp loaded lower-branch target of the selected corre-
spondence (A1566: the octant registration is a candidate map, not a selector; the internal identity
stays theorem-grade). [LIB2-240, LIB2-284] Recent benchmark movement sharpened the need to state
the branch explicitly rather than speak loosely of a “near-maximal” target. This is how falsifiable
phenomenology should behave: predictions are stated, old placeholders are retired, and the target
remains exposed to future data.
8 Conclusion
Paper 7 is the suite’s shortest paper and one of its most important. The programme can fail. Long-
baseline oscillation data, atmospheric-angle precision, electroweak weak-angle refinement, and charged-
lepton-family validation have the power to decide whether the exceptional-Jordan correspondence is
physically right, merely suggestive, or wrong.
11
Leibniz Quantum Beats Newton
Paper 8: Junkyard — Deleted Content, Dead Ends, and Junked Ideas
(Rev33.1)
Tom O’Sieg
August 2026
Abstract
Paper 8 is the rev7 archive of deleted content, dead ends, and candid programme commentary.
It is not a graveyard in the sense of a list of proved impossibility theorems; that role belongs
elsewhere in the suite. It is the record of routes that were explored seriously enough to matter,
then abandoned because a simpler, clearer, or more algebraically closed alternative replaced them.
The central rev7 example is the Yukawa sector. A linear Freudenthal projection appeared
attractive in earlier drafts but could not preserve the off-diagonal Higgs geometry without additional
assumptions. The cubic invariant
Tr Jvac ◦ (Ψ × ΦH ) (1)
replaced it cleanly, so the linear route was deleted. Similar logic governs the other entries: modular
attempts to derive the Higgs vacuum expectation value, Niemeier-lattice analogies, attractor-
inspired weak-scale ideas, monopole and E8 digressions, and several speculative bridges were
explored, recorded, and then parked.
The point of publishing this archive is methodological. A mature theoretical programme should
publish its junk as well as its gold. The deleted routes in this paper are not embarrassing leftovers.
They are part of the reason the surviving rev7 bundle is cleaner than its predecessors.
1 Deleted content from rev5 to rev7
Linear Freudenthal Yukawa (deleted)
Rev5 attempted to package the Yukawa sector through a linear Freudenthal projection. The route
failed because it effectively forced the Higgs into a diagonal placement if the middle eigenvalue was to
be isolated cleanly. That broke the SU (2)L doublet geometry used elsewhere in the framework. Rev7
therefore deletes the linear route and replaces it everywhere with the cubic invariant
Tr Jvac ◦ (Ψ × ΦH ) . (2)
H-embedding ambiguity (deleted)
Earlier drafts flirted with the idea that specifying the Higgs block might require a new axiom. The
cubic-invariant route resolves that issue. No extra axiom is needed. The off-diagonal Higgs placement
is preserved directly by the local cubic invariant.
1
2 Curios (moved, not deleted)
Boolean hypercube / I Ching trigram indexing (moved from Paper 0 Box C.1,
S64)
Moved here during the S64 honest-language pass: hostile reviewers consistently read the box as
numerology despite its disclaimer. The content is unchanged; it remains an observation on shared
combinatorial structure [A626], not a formal isomorphism and not load-bearing anywhere in the suite.
While the formal algebraic breaking of the (4, 4) vacuum is governed by J3 (Os ), its combinatorics can
be illustrated by a classical system. The basis of Os is indexed by the Boolean hypercube Z32 — the
same underlying combinatorial structure that generates the I Ching’s eight trigrams (first identified
by Leibniz in 1703 as binary arithmetic). Under this identification, {e0 , . . . , e7 } correspond to the
eight trigrams, with e0 ↔ 000 (pure yin, Earth) and e7 ↔ 111 (pure yang, Heaven).
The 64 entries of the Os Cayley table map naturally to the 64 hexagrams. The 8 pure hexagrams
(squared basis elements, e2i = ±1) correspond to diagonal Peirce sectors contributing zero to the
fermionic Coleman–Weinberg potential (VCW bos ≡ 0). The 56 mixed hexagrams correspond to off-
ferm ̸= 0 via the Möbius anti-periodic boundary
diagonal Peirce sectors P12 , P13 , P23 that drive VCW
condition.
The solved vacuum Jvac = diag(φ, 1, φ−1 ) assigns a natural reading to the three generation face-
centres: φ (yang-weighted, τ ), 1 (balanced, µ), φ−1 (yin-weighted, e). This is an observation on
shared combinatorial structure [A626], not a formal isomorphism; the Fano-plane sign structure of
Os has no I Ching counterpart.
PSL(2, 7) class-algebra structure constants (moved from the library, S271)
Two structure constants of PSL(2, 7) = GL(3, F2 ), computed from the character table via the
k = |C | |C |/|G| P χ (C ) χ (C ) χ (C )/ dim ρ, are proved [D161, S19]:
class-algebra formula ci,j i j ρ ρ i ρ j ρ k
7A 7B
c4A,7A = 14, c4A,7A = 0.
The second vanishing is exact, and it was the algebraic content of the old chirality argument: the
subtractive combination − arccos(−1/3) + 2π/7 would require c4A,7A 7B ̸= 0, so it is forbidden, leaving
◦
− arccos(−1/3) − 2π/7 = −160.900 as the only allowed sign.
√
Why it sits here. That application is deprecated. The authoritative CP phase is δCP = −2π/ 5 =
−160.997◦ from the V3 rotation (Derived-conditional after the Rev29 re-tiering; pre-Rev29 stamp
⋆100; Paper 5, Appendix C), which does not involve PSL(2, 7) at all; the 2π/7 expression survives
elsewhere in the suite only as an explicitly labelled cross-check and must not be re-introduced as a
derivation. [LIB2-245] The constants above therefore have no load-bearing role left anywhere. They are
recorded because the algebra is correct, not because anything rests on it. [LIB2-057, LIB2-245]
On 14 = dim G2 . The coincidence is real, and it is why the result was noticed at all. It is not a
link to any G2 object in this suite. The suite’s G2 occurs in two guises — the dynamical anchor
ΛG2 ≈ 260 MeV (Appendix A, a dimensional-transmutation scale) and the split real form G2(2) in the
Jordan chain — and neither of those is the dimension of the compact automorphism group turning up
as a class-algebra integer. No mechanism relating a PSL(2, 7) structure constant to dim G2 has been
proposed or tested anywhere in the corpus. [LIB2-057] The equality is therefore a numerical coincidence
2
with no established mechanism: an observation, in the same sense as the trigram indexing above, and
not a formal correspondence.
Library entry: L-StructConst_PSL27_structure_constants.md.
3 Dead ends and why they failed
The S197/S208 Tier-C clean negatives (recorded S270)
The following results were verified, grep-anchored against this suite, and then left unbanked for
sixty-five sessions. They move no observable; that is precisely why they went unrecorded, and precisely
why they belong here. Each is reproduced from the S208 reconciliation ledger with its provenance.
Provenance Result
C1 T1-a σ-mass-gap is not an F4 = Der(J3 ) Cartan object: it is the
LJvac multiplication in the e6 /f4 complement, ∥projDer ∥ =
10−16 . A one-line no-go.
C2 T1-c, s364+s391 Charged action-level Dirac mass is forbidden (no-Q theorem,
triple-confirmed); the neutral 2 × 2 Majorana channel is
allowed.
C3 T1-d 912 Θ(A) gauging is blind to the hypercharge residue A
— the compact gSM is excluded for every A, including the
Standard-Model A = 2. A decisive negative.
C4 T1-e, A1208 Extraction retired: defect ranks are congruence / frame-orbit
invariants, so no legal frame repair exists (house verification
of A1208, S197). The original kernel s449 is not recoverable
and is no longer cited as the anchor.
C5 T1-f A1111 four-corrections errata: T 2 = −I ⇏ antiperi-
odic/Möbius; one warp ⇏ both EM and mass; the half-twist
commutes with mass. The corrected facts are already on disk;
the errata block is what was missing.
C6 T2-e so(6, 6) = D6 is an e7(7) rung, and the 28:28 versus 30:30
distinction matters: “30:30 = T-bridge” is false. The
SU (2) = weak identification is fenced, not assumed.
C8 T2-f Selector Σ parity-split end-state: even = Urec + K∗ + gaps,
odd = κ⊥ + ε (frozen at A1208). The honest statement is
that rest-mass + mixing is a named section, not a derivation.
C9 T3-b, 3/3 The Σwb winding-monotone rest-mass driver is falsified by
the leptons: the residuals alternate in sign.
3
Two honest notes on this table. First, C7 is deliberately absent: the Peirce-arrival √
ordered-vertex theorem (Yf = q · Jvac · C forced by chiral middle-slot routing, ∆ = 3 − 5
unique) is a positive result — the selection principle behind the A3/B-mixing route — and does
not belong in a junkyard. Its conclusion is already folded into the mixing paper; its premise
is not, and that remains owed. Second, C10 is deliberately absent: the charge-clock tower
(q0KK )2 = 1/24 = 1/(4 · 6) with the ΩQ Plücker line is marked hold at source — deep but
conditional, gated on the ΩQ provenance that Appendix N, The carrier truncation is also
chosen shows is still a choice. It is library-resident by decision, not by neglect. [LIB2-355]
Still owed. The S208 ledger homes C2, C8 and C9 substantively in Appendix O (mechanism
section, honest end-state, and the falsification record respectively). Recording them here
makes them findable; it does not discharge those folds.
Modular-forms route to the weak scale
A modular-forms route was explored as a way of deriving the electroweak scale from the same
golden-ratio data that organise the algebraic vacuum. The route never acquired an algorithmic bridge
from J3 (Os ) to modular weights. It is parked rather than promoted.
Niemeier-lattice route
The programme explored whether the Peirce eigenvalues and the 24-dimensional Niemeier-lattice
story could constrain the Higgs vacuum expectation value. Structural parallels existed; a derivation
did not. The idea is deferred.
Extremal attractor mechanism for the electroweak scale
Because cubic invariants play a prominent role in supergravity attractor equations, the possibility of
deriving the weak scale from an attractor construction was examined. The formal analogy was real;
the scale derivation was not achieved. The route is deferred to a later revision.
Seesaw mechanism as a required neutrino ingredient
A conventional seesaw route was tested as a mandatory extra layer for neutrino masses. In the present
scheme it turned out not to be structurally necessary for the basic neutrino mass ratios, so it was
omitted from the operational rev7 bundle.
Proton-decay rate estimate
A genuine proton-decay calculation would require a concrete high-scale unification completion and a
definite heavy gauge-boson sector. The current algebraic suite does not yet contain that completion.
Proton decay is therefore deferred rather than advertised prematurely.
4
Monopole condensation route
A monopole-condensation story was explored as a symmetry-breaking mechanism. It proved more
complicated and less directly constrained than the cubic-invariant route. It was abandoned.
E8 embedding
The programme repeatedly checked whether an E8 -based unification picture could replace or complete
the J3 (Os ) story. Interesting parallels were found, but no concrete derivation resulted. The rev7 suite
remains explicitly based on J3 (Os ), not on E8 .
4 Fenced: out-of-scope work (recorded, not promoted)
This section is the explicit pen for results that reach past the framework’s remit — the dimensionless
flavour observables of J3 (Os ) and the single hadronic calibration scale ΛG2 — into territory the suite
does not claim to model. They are not dead ends (they may be revisited) and not errors (they were
computed correctly in their day); they are simply out of scope. We record them here, fenced, rather
than fold them into the suite proper, so the falsifiable core stays clean and the strays stay visible.
Hadronic over-reach: the baryon/meson ratio MB /MM ≈ 6.7 (S23–S24)
A self-consistent Faddeev-pole gap equation produced a baryon/meson mass ratio MB /MM ≈ 6.7,
matching the PDG proton/pion value 938/140 at a single free coupling α ≈ 0.0025 (the ratio stays in
5–8 across all perturbative α). [LIB2-246] Three reasons it is fenced rather than promoted: it reproduces
one hadron ratio (a single number is not a spectroscopy programme); it carries a free parameter;
and it was computed in the legacy ambient-27 Hessian parametrization (spectral trace A = 144,
Morse index 17/27) that the current {φ, 1, φ−1 }/OP 2 description has since replaced. An in-frame
recompute (kernel s284, S175) confirms it does not survive: the current N = 1 cone yields only
φ-power eigenvalue ratios, and 938/140 = 6.72 is not a clean φ-power (nearest φ4 = 6.854, +2%).
The number is therefore retired — a legacy null-cone artifact superseded by the current frame,
not a parked candidate. [LIB2-246, P-022] The structural content underneath it — that the baryonic
(cubic) channel survives the quadratic one-loop cancellation that kills the mesonic channel — is
kept in the suite as a theorem (Appendix I, Confinement as the spectral geometry of the cubic
norm). [LIB2-246] Only the number is parked here. Library: L-BaryonMesonRatio, L-FaddeevEvasion,
L-A144, L-TMatrixResidue, L-HybridMassGap. [LIB2-352]
√
Golden flux-tube tension σ = ΛG2 φ (S164 era)
A flux-tube relation tying the string tension to the confinement scale by a single golden factor,
√
σ = ΛG2 φ. It is thinly banked (no canonical library entry; it survives only as a notebook pointer),
and it reproduces no datum independent of ΛG2 itself — it restates the one calibration scale dressed
by φ. The S175 in-frame recompute (kernel s284) does upgrade its provenance: on the current
N = 1 cone it is exactly the off-diagonal Hessian eigenvalue ratio 2ξ1 /2ξ2 = φ (ξ = (φ, 1, φ−1 )),
hence frame-current and α-free, and it is now recorded as a structural remark in Appendix I (the
in-frame flux-tube remark, Dynamics of the Golden Vacuum). But because it still merely restates
5
ΛG2 (no Λ-independent content), it remains out of the falsifiable core — recorded here as a suggestive
coincidence pending a derivation that earns its keep. (The companion S164-era observation that the
golden ray reads as a relativistic boost, cosh(2 ln φ) = 3/2, i.e. φ2 + φ−2 = 3 (the DET-7 condition of
Appendix E; equivalent to AX1 for ρ > 1), is in scope as kinematics and belongs with the time/clock
bridge, not here.)
The S103 moment ledger: parked candidate (60-tick clock, time-silence, Gog-
berashvili splice)
The S103 “moment ledger” (filed 2026-06-13, status candidate, never star-gated) is parked here by PI
decision rather than promoted. [LIB2-201] Its three results: (i) one observed moment as a single traversal
of the S 1 /Z2 base rotation circle, with the 56 substrate states carrying Σ|winding| = 60 ν-ticks; (ii)
time-silence — a clean derivable negative: no monomial in the banked constants {q0clk , κ, Γ} carries a
time dimension, so a duration in seconds is not derivable from the framework’s mass/length-complete
constant set (any seconds figure is illustrative, never banked); (iii) the Gogberashvili splice reconciling
the house constants with an external {c, ℏ} time axis on the shared SO(4, 4) signature. [LIB2-201, LIB2-202,
LIB2-203] Result (ii) is a genuine scope statement — the framework is time-incomplete by construction
— and is the part most worth keeping; (i) and (iii) are suggestive but unpromoted. [LIB2-202, LIB2-203]
The papered Moment Clock + T -operator (Appendix H) is unaffected; only the candidate headline
numbers are parked. [LIB2-201] Library: L-S103-MomentLedger.
Why these are fenced, not deleted. The lab notebook keeps the good, the bad, and the
ugly. Out-of-scope results are real work that simply sits outside what the present framework
claims to predict; deleting them loses provenance, promoting them enlarges the attack surface
for no predictive gain. The fence is the honest middle: visible, dated, and explicitly outside
the falsifiable core (Appendices B/C). [LIB2-203] Future strays that overreach the flavour-plus-
one-scale remit belong in this section.
5 PI’s junked ideas
Octonion-based supersymmetry
The idea that non-associativity might itself encode supersymmetry was explored as a brainstorming
direction. It would have required new particles and new axioms. The rev7 bundle is intentionally
non-supersymmetric and does not pursue this path.
Gravitational coupling from the Jordan determinant
Another brainstorm attempted to read Newton’s constant directly from the Jordan determinant. No
natural scaling law emerged, so the idea was shelved.
Cosmological constant from a trivial Peirce block
A dark-energy interpretation of a trivial Peirce block was considered. No mechanism was found that
would naturally suppress the cosmological constant to the observed scale. The idea was abandoned.
6
Electroweak precision tests as the best CP-falsification channel
A kaon- or electroweak-precision-based falsification route for the geometric δCP prediction was
explored. The conclusion was negative: long-baseline neutrino measurements remain the more direct
test.
Spinor-octonion duality
The idea that fermions might be literally octonion-valued spinors rather than ordinary spinors acted
on by an octonion algebra was examined and dropped as an unnecessary over-complication.
The Z12 fused clock (Rev14, S86)
The proposal that the Z4 tick and the Z3 generation cycle fuse into a single Z12 clock generator was
killed by computation: the candidate generator C is not in the normalizer of e7(7) (C ∈
/ N (e7(7) ), S86
probe). Only the trivial product reading Z3 × Z4 survives, which adds nothing. Junked; the tick and
the generation structure remain independent.
The anti-symplectic T 2 = +1 moment model (superseded, Rev15, S106l→S107)
An earlier abstract model of the past↔future “moment” (S106l) built the operation T as a hand-made
real swap on 27 ⊕ 27 and read it as anti-symplectic with T 2 = +1. The live computation on the
certified 56 (S107a) overturns both readings: T is a genuine element of E7(7) ⊂ Sp(56, R) — hence
symplectic, not anti-symplectic — and squares to −1 (the Kramers/spinor half-turn), so antiunitary
time reversal appears as T 2 = −1. The abstract model is therefore superseded; the correct live
statement is recorded in Appendix H (Theorem H.1). Retained here only as the cautionary note that
a real-swap caricature of a substrate operation can flip both its symplectic character and its square.
6 Candid commentary
The rev7 suite was improved mainly by deletion. The cubic invariant proved simpler than the
linear route. The weak-angle gap proved to be an ordinary threshold problem rather than a fatal
contradiction. The electroweak scale proved more stubborn than hoped. The archive therefore records
a simple lesson: the most elegant route is often the one that removes assumptions rather than the
one that adds them.
7 Archive index
Category Count
Deleted rev5 content 2
Dead ends and deferred routes 7
PI brainstorm ideas 5
Total 14
7
Paper 8 is published with the bundle because scientific maturity includes an archive of discarded
routes, not only a catalogue of surviving claims.
8
Leibniz Quantum Beats Newton
Paper 9 — The Nuclear Cluster Tier: Binding Energies from the Golden
Fusion Category, Deuterium through Calcium (Rev33.1)
Seventeen sharp binding energies (worst untyped residual 0.77%, 3 H) plus one typed obstruction
(28 Si, 0.334%; pre-answer bar 1% passed, refinement-gate obstruction branch taken), the Hoyle state,
eleven correct sign calls, and two glueball bands — every number scored against a bar frozen before
the answer existed. The composition ladder is a closed-form categorical trace invariant. All values
√
are ratios to σ; the absolute scale is the framework’s one registered input (Appendices O/X).
Sessions S225–S226; assessments A1349–A1357; kernels s705–s747.
Tom O’Sieg
August 2026
Posture (binding). This paper promotes the S225–S226 nuclear/cluster arc from the
lab-notebook layer to paper tier. The promotion case is protocol-based: seventeen sharp
passes at pre-frozen accuracy bars (plus one typed 28 Si obstruction, §9) across ten scored
dispatch rounds with zero threshold moves, three blind vault matches (§10), and a freedom
ledger audited round-by-round. Everything here is dimensionless (binding energies as ratios
√
to the string tension σ); the single dimensionful input of the framework is registered
separately (Appendices O/X). The engine workbooks are unchanged; no public observable
of the Rev24 suite moves. The continuum (Yang–Mills) functor remains the named missing
object: nothing in this paper is a Clay-problem claim. [LIB2-300] Known soft flank, stated
up front: charge signs and counts are derived, but charge magnitudes are open — two
exposed defects are filed in §9 (Sp (9 B) × 1.29, Sp (8 B) × 2.60). Read the seventeen-sharp
ledger (one typed 28 Si obstruction, scored as neither a sharp nor a miss at a pre-answer bar
— §9) with that flank in view.
Contents
1 Scope and the one-line result 1
2 The functor and the cluster menu 2
2.1 The exhibited functor Fφ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2 The morphism menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3 The A ≤ 4 binding kernel 3
4 α-closure and the gap theorems 3
1
5 Bridged and Borromean systems, 5 < A < 12 4
6 The ladder: a closed-form trace invariant 4
7 The Hoyle state and the breathing kernel 5
8 The ledger 6
9 Typed features and open items 7
10 Protocol sub-appendix: the receipt chain 9
11 Status ledger 10
1 Scope and the one-line result
The golden fusion category machinery of this suite (the Fibonacci quotient with f4 = 0, the exhibited
functor Fφ , and a discrete menu of φ-monomial morphism weights) fixes the ledger weights Tn , which
the declared cluster-ledger interface (2) then carries to nuclear binding energies (Rev29 : predication
corrected; the categorical machinery does not itself price an energy, the interface does). [LIB2-143]
The one-line result:
From deuterium to calcium, every scored binding energy lands inside a pre-frozen sharp bar
with one typed exception — seventeen sharp passes, worst untyped residual 0.77% (3 H;
0.37%, 24 Mg, is the worst of the A ≥ 5 non-silicon continuation), plus one typed obstruction
(28 Si, 0.334%: pre-answer bar 1% passed; refinement-gate obstruction branch taken) — the
Hoyle resonance passes at its frozen bar, eleven sign/stability calls are all correct, and the nα
composition ladder closes into a one-line trace formula whose two blind forecasts matched
numbers sealed in advance (a φ2 match at 26 ppm and a T9 match at 0.10%).
√
Throughout, world values are AME binding energies; σ = 445(3)(6) MeV is imported solely to
express MeV; the inputs of every formula are A, Z, the selector depth N⋆ = 11, and φ — nothing
else. No fine-structure constant enters: the charge penalties below do Coulomb-like work with
φ-monomials alone (declared, audited, and the charge magnitude sector is honestly the weakest —
see §9).
2 The functor and the cluster menu
2.1 The exhibited functor Fφ
Fφ sends the winding category to the Fibonacci category: objects Fφ ([N ]) = τ ⊗N ; the mirror is the
dagger; the four-strand Jones–Wenzl projector is killed, Fφ (f4 ) = 0 (the [5]δ = 0 specialization at
δ = φ). [LIB2-143] It was exhibited with its full generator set and kernel-verified (A1349, kernel s706:
2
F -involution, braid relation, both mirror-inversion laws, all ≤ 3 × 10−16 ; Spec(φ−1 Nτ ) = {1, −φ−2 }
exact). This is the object referred to below as the cluster kernel: the binding rules of this paper
are morphism-weight statements in the image of Fφ . [LIB2-143]
A standing fact about the recoupling matrix, recorded as a fence (Rev30,
A1480). The Fibonacci recoupling matrix is an involution: F 2 = I. House check:
(F 2 )00 = φ−1 (φ−1 + 1) = φ−1 · φ = 1 with the off-diagonal identically zero. Consequently
F 5 = F and (F 5 )00 = φ−1 — not φ−5 . Any argument that accumulates a golden power
by composing an odd number of F -moves is therefore false as stated, not merely unproven:
the moves do not accumulate, they alternate. This is printed here as a fence rather than
as a correction — no such five-move monodromy argument appears anywhere in this suite,
and this note exists so that none is introduced. It bears on no banked binding formula in
the menu below.
2.2 The morphism menu
Every banked binding formula is assembled from a short, closed menu; no entry was added after
S226. Generators (with their derivations where banked):
• M1, the two-body bridge (S=1, T =0): the Fibonacci singlet channel τ ⊗ τ → 1 exists iff
spin-symmetric/isospin-antisymmetric — the topological Pauli statement. Weight φ−N⋆ = φ−11
(the deuteron). Consequence, scored 3/3: nn unbound, pp unbound, pn bound (A1352).
• M2, the closed α-cell: f4 = 0 makes A = 4 a closed Jones–Wenzl cell; α ≡ 1. The α is
saturated: a fifth nucleon sees 1α ⊗ τ = τ , a boundary defect with no scalar channel. [LIB2-146]
• M3, the triple-α associator Θ : 1⊗3 → 1: three-strand associator data survives f4 = 0 while
the four-strand channel dies — the same quotient that forbids 8 Be builds carbon (A1353).
• M5, the Borromean Y -junction Y : 1 ⊗ 1 ⊗ τ → 1 (9 Be: cut any leg and closure dies —
nature’s αα/αn/ααn pattern). Fence (Rev32.9, A1671 F03): in the ordinary Fibonacci category
Hom(1 ⊗ 1 ⊗ τ, 1) = Hom(τ, 1) = 0, by the same fusion rule that leaves M2’s saturated α with
no scalar channel, so Y as printed is the zero morphism. A nonzero Y needs an enrichment this
suite does not exhibit (a boundary charge or a retained τ -line capped outside the three legs,
and a selector for the α-cell inside τ ⊗4 = 2 · 1 ⊕ 3 · τ ). M5 is carried as a leg-counting pattern:
the 9 Be rule below uses it to count (cancel one annulus defect, add one bridge step), not as a
matrix element, so no banked number changes; the 9 Be derivation inherits this gap.
• M6, the halo dual Y ′ : 1 ⊗ τ ⊗ τ → 1 (6 He: the dineutron’s singlet channel is typed wrong in
vacuum; the closed cell supplies the annular cap — Borromean, not pairwise; A1355, with the
same mechanism pre-registered blind in-house by the Skyrme-lens kernel s735).
• Leg rules: composite-1 and τ -cluster leg typing; leg-local attachment of odd lines (the endpoint
rule: two unequal images, φ−18 junction-near + φ−19 free-endpoint, Möbius-delayed; A1356).
• Charge generators (retro-derived, A1354): the half-tick 12 φ−12 (Z4 half-tick averaged over the
Möbius sheets); the fused pair φ−11 = 2 · 12 φ−12 · φ (two half-ticks fused in the τ -channel pick
up the Perron–Frobenius eigenvalue: coherence multiplies by φ — exact); the count 3 = the
Fano line {0, 1, 3} (the selector’s own line reappearing as an image-link count).
3
• The universal second-order defect φ−18 : the closed-cell annulus subtraction, appearing in
8 Be, nn, the 7 Li/7 Be splitting, the 9 Be/9 B corrections, the endpoint rule, and the breathing
kernel — one constant, seven appearances.
• The breathing generator ∆br = φ−15 − φ−18 with 15 = 8 + 7 (triangle scale + heptad
circulation), the same decomposition banked for the Hoyle exponent (§7).
3 The A ≤ 4 binding kernel
The rule, filed before any number (A1349):
1 Z i
! ! ! !
B(A, Z) h A A A
√ = φ−11 + 2φ−2 + φ3 − 2φ
1
− . (1)
σ 2 3 4 2φ 2
The deuteron is the 22 term alone (machine-exact); the scored sharps are 3 H 0.77%, 3 He 0.10%,
4 He 0.16% at the frozen 1% bar. Freedom audit (carried verbatim): the rule introduced three
new φ-coefficients and was scored on three new nuclei with a triangular dependency — zero
net overdetermination within A ≤ 4, so no blind credit for the coefficient choice. [LIB2-145] The
discriminating test was structural: A ≥ 5 has no new binomial available, and everything from §4 on
is scored against that zero-freedom continuation. [LIB2-145]
4 α-closure and the gap theorems
Novelty label: physical interpretation not established here.
Theorem 4.1 (A = 5 gap). f4 = 0 closes the α cell, so a fifth nucleon is an external τ -line with no
scalar channel: no single nucleon binds to α. [LIB2-146] Signs (choice-free): Sn (5 He) = −φ−12 < 0,
Sp (5 Li) = − 23 φ−12 < 0 — both correct (A1350).
Novelty label: physical interpretation not established here.
Theorem 4.2 (A = 8 gap). α ⊗ α = 1 ⊗ 1 exposes no τ -line: no bridge is typeable, B(8 Be) = 2Bα
√
at leading order, and the first correction is the level-repulsive virtual τ τ̄ annulus, Sαα = −φ−18 σ <
√
0. [LIB2-147] Sign correct; the pre-frozen marginality tier was hit (|S|/ σ = 1.7 × 10−4 < 10−3 ; report
magnitude 0.84× world). Stars make carbon through the triple-α channel because 1 ⊗ 1 offers no
bridge while the three-strand associator survives (A1351, A1353). [LIB2-147]
These two are consequences, not inputs: the same projector that closes the α forbids its pairwise
growth and forces the associator architecture of §6. [LIB2-146]
5 Bridged and Borromean systems, 5 < A < 12
6 Li = α + d + one Fibonacci bridge one step past the selector depth (φ−12 ); closed form φ−11 (7 +
8φ−2 + φ3 ); sharp 0.11% (A1350). 7 Li = α + t with the neutral N⋆ -bridge (φ−11 ); 7 Be = α + 3 He
with the charged mirror bridge (φ−11 − 12 φ−12 , the half-step penalty reused), refined by the three
4
proton-image links (−3φ−18 ) and completed by the breathing kernel (§7): final sharps −0.0074% and
+0.076% at the tightened 0.25% bars, with the mirror differential preserved by type (a mirror-even
object cancels identically in the difference; verified at 2.5 × 10−17 ). 9 Be: the Y -junction cancels
the αα annulus defect and adds one bridge step, Sn = φ−12 + φ−18 > 0, sharp 0.18%; the mirror
9 B dies correctly, S = −φ−13 + φ−18 < 0, with the shadow-tier magnitude S = −φ−13 + 8φ−18
p p
(8 = 2 × 3 Fano image links +2 endpoint images) at ×1.29 of world (A1353, A1355). 6 He: the halo,
B = Bα + φ−13 + φ−18 , sharp 0.0039%, with both A=6 signs correct (6 He bound, 6 Be killed by the
already-derived two-face φ−11 ). 8 Li/8 B: the endpoint rule (two unequal images φ−18 + φ−19 ; the
neutral junction-near image is absorbed into the triton bridge under vacuum normalization for 8 Li,
while charged 8 B keeps both) — sharps 0.0027% and 0.0049% (A1356, pass 2 refiling under clean
receipt, mechanism-carried).
6 The ladder: a closed-form trace invariant
For n-fold α composites the banked law is
B(nα)
√ = n Bα + Tn T3 , T3 ≡ φ−8 − φ−11 (2)
σ
(T3 = the net triple-α associator weight, so T3 = 1 by construction), and the ladder coefficients close
into one formula (A1356, A1357):
Tn = (n − 2) + εn , εn = trϕ (Rn ) (3)
where (n − 2) counts the triangle slots in any pants decomposition of the n-gon and εn is the Markov
trace of the residual boundary-circulation class of the n-gon flip graph in the Temperley–Lieb
quotient TLφ /⟨f4 ⟩. [LIB2-148] The derived selection rule (A1357):
ε5 = −φ−2 (the pentagon F -move coherence loop: the killed channel appears once, negative trace),
(Rev32.9, A1671 F04: the parenthetical is a mnemonic, not a computation. In the quotient
trφ (f¯4 ) = 0, so the sign has to come from a defined signed residual of the class R5 , which this paper
does not write down; likewise the representatives Rn , the oriented hexagon pair and the n ≥ 7
argument are banked from A1357 and not exhibited here. The values εn are not disputed; their
derivation is owed.)
ε6 = 0 (the hexagon’s two boundary orientations are mirror-paired in the annular quotient and cancel),
εn = φ 5−n (n ≥ 7) (exactly one noncontractible class survives, decaying φ−1 per added cell). [LIB2 − 149]
Note T5 = 3 − φ−2 ≡ φ2 is exact algebra: the pentagon identity fixes the ledger weight T5 , and the
interface (2) carries that weight to a binding energy (Rev29 : predication corrected; the identity
prices a weight, not an energy). [LIB2-148] Scored consequences, all against frozen bars: 20 Ne 0.00031%
(T5 = φ2 — the best number on the ledger) · 24 Mg 0.37% (T6 = 4) · 28 Si 0.33% (T7 ; typed, see §9)
· 32 S 0.024% (T8 ) · 40 Ca −0.032% (T10 ; doubly-magic calcium at ten cells, matching the in-house
blind application s745 exactly) · 36 Ar report 0.018% (T9 = 7 + φ−4 ).
Remark 6.1 (the two blind vault matches). Twice, the closed form produced a number that had
been extracted from world data and sealed in a hash-stamped pre-commit before the question
5
was asked. (i) The s732 vault held kimplied (20 Ne) = 2.617965, never disclosed; the lane derived
T5 = φ2 = 2.618034 from the pentagon identity alone — 26 ppm. (ii) The s737 answer sheet held
the implied T9 = 7.1385 behind the sole hint “at n = 9 the ladder does not do the obvious thing”;
the rule declined the linear 7 and produced 7 + φ−4 = 7.1459 — 0.10%, against a vault the lane
never saw. Receipts, hashes, and the unseal schedule are in §10.
7 The Hoyle state and the breathing kernel
Hoyle (scored PASS at the frozen 10% bar, A1354):
√
EHoyle − E3α = (φ−15 + φ−18 ) σ = 0.4033 MeV vs world 0.3796 MeV (6.2%).
The resonance that makes carbon in stars is the trace-free breathing mode of the same associator
whose ledger weight enters the 12 C binding energy through interface (2) (Rev29 : the associator is a
constructed object and binds nothing; it supplies a weight, and the interface supplies the energy),
priced by two φ-powers, with the exponent decomposition 15 = 8 + 7 (triangle scale + heptad
circulation). [LIB2-152]
The breathing kernel (A1357): the same object, typed — the trace-free breathing excitation of
the α + 3-body bridge, mirror-even and charge-blind, priced by
∆br = φ−15 − φ−18
with zero new constants (15 = 8 + 7 reused; −φ−18 = the annulus subtraction, seventh ap-
pearance). [LIB2-162] It repaired the ledger’s two worst entries (7 Li 0.64% → −0.0074%, 7 Be
0.59% → +0.076%) at tightened bars, and the mirror differential was preserved by type, not
numerically. The refusal-then-derivation arc — the lane declined to use this object untyped in
A1355, then derived its typing kernel in A1357 — is the protocol working as designed.
The baseline ruling (banked, A1357): separation energies are computed in the ledger category —
Sx (P → D + x) = B(P ) − B(Dledger ) − B(x) with the refined daughter always, where B(x) = 0 for a
single emitted nucleon (the case the ruling was banked for) and the bound-fragment term is required
otherwise (8 Be→ αα gives 0 at leading order only with it; Rev32.9, A1674 F01; Rev32.10, A1714-F01
— this does not license the open A = 8 alternative baseline); neutral excitations cancel only if present
in both typings. [LIB2-155, LIB2-163, LIB2-304, LIB2-306] Applied honestly it made one number worse:
Sp (8 B) = 4φ−18 + φ−19 = 0.356 MeV = ×2.60 of world, filed as an exposed charge-sector magnitude
defect rather than baseline-shopped (§9). [LIB2-163]
8 The ledger
6
Table 1: The Rev25 freeze-state nuclear ledger. All sharps
are |Bpred − Bworld |/Bworld against bars frozen before the
answer; ten scored rounds, zero threshold moves. † = first-
circulation residual, 0.334%: the value and the frozen-bar
history are unchanged; the direct connected-Markov-second-
moment energy attachment (A1543/D1224) is closed negative
(A1544) and a reopen needs a different localized operator or
a derived negative energy map (§9); the Rev30 readout fork
is separate and untouched.
√
System Formula (B/ σ or as Sharp Bar
stated)
d φ−11 0.52% 1%
3H eq. (1) 0.77% 1%
3 He eq. (1) 0.10% 1%
4 He eq. (1) 0.16% 1%
6 He Bα + φ−13 + φ−18 0.0039% 1%
6 Li φ−11 (7 + 8φ−2 + φ3 ) 0.11% 1%
7 Li α + t bridge + ∆br −0.0074% 0.25%
7 Be mirror bridge −3φ−18 + +0.076% 0.25%
∆br
8 Li endpoint rule (pass-2 refil- 0.0027% 1%
ing)
8B endpoint rule (φ−18 + 0.0049% 1%
φ−19 )
9 Be 2Bα + φ−12 0.18% 1%
12 C 3Bα + T3 0.10% 1%
16 O 4Bα + 2T3 0.17% 1%
20 Ne 5Bα + φ2 T3 0.00031% 1%
24 Mg 6Bα + 4T3 0.37% 1%
28 Si 7Bα + (5 + φ−2 )T3 0.334%† (typed 1% pre-answer (D1060): PASS;
obstr.) 0.1% refinement gate (D1061):
obstruction branch (A1357)
32 S 8Bα + (6 + φ−3 )T3 0.024% 1%
40 Ca 10Bα + (8 + φ−5 )T3 −0.032% 1%
Hoyle (φ−15 + φ−18 )√σ 6.2% 10% (PASS)
√ √
m(0++ )/ σ √8π log φ = 3.4777 in band [3.3, 3.9]
m(2++ )/m(0++ ) 2 in band [1.3, 1.6]
That is seventeen sharp passes, deuterium through calcium (worst untyped 0.77%, 3 H;
0.37% is the worst of the A ≥ 5 non-silicon continuation — Rev32.9), plus one typed obstruction
(28 Si, 0.334%; pre-answer bar 1% passed, §9), plus the Hoyle pass, plus eleven sign/stability
calls all correct (pn/nn/pp; the A = 5 pair; 8 Be; 6 He/6 Be; 9 Be/9 B), plus two glueball numbers
in their frozen bands, plus report-tier entries (36 Ar 0.018% · 10 Be 0.20% · 6 Be S2p within 4.9%
· the exposed Sp (8 B), §9). [LIB2-153] The glueball chain — the annulus functional E(R) as the
− log of a vacuum-normalized closed Markov correlator, saddle m2 /σ = 4π∆ with ∆ = 2 log φ, the
7
Lüscher vacuum term correctly cancelling in the correlator ratio — is banked at draft tier with its
normalization derivation recorded (A1350–A1352).
Remark 8.1 (magic numbers: what is native and what is not). Two magic numbers are native
to the framework: 2 (Z2 doubling) and 8 (the octonion octet, the 11 − 3 gap of the occupation
O = {0, 3, 11, 17}; kernels s574, s702). The remaining magic numbers are open — the shell-structure
question, and with it any island-of-stability forecast, is a named program (Paper 7), not a result.
This scope line is deliberate.
Remark 8.2 (the upper wall is measured: a blind negative past 60 Zn). The upper edge of this tier
is a measured failure, not an assumption. The continuation of the nα ladder (2) along the N = Z
line from 64 Ge to 100 Sn was pre-committed blind (kernel s1101, frozen before any AME2020 value
with A > 60 was read), with two frozen correction classes. Both fail. The 1% bar dies at the first
blind row, 64 Ge (+1.64%). From there the residual grows monotonically to +6.03% at 100 Sn: the
ladder over-binds every measured row, and there is no recovery at the 100 Sn closure. Kernel s1102
identifies the missing term only qualitatively, as the net curvature of a surface term and a Coulomb
term. That term needs a nuclear radius the framework does not yet have, and the fitted Coulomb
scale is 43% of the physical one. This is a failed pre-registered prediction and is printed as one: the
ledger above claims no pass beyond calcium, and past 60 Zn the ladder is falsified.
9 Typed features and open items
28 Si — first-circulation residual; direct annular second-moment attachment closed
negative (Gate 1, re-adjudicated Rev28; retyped Rev32.2 from A1542–A1545, PI ruling
S294). For 28 Si the first-circulation ladder leaves a 0.334% residual. The pre-answer 1% bar was
passed; the refinement gate nevertheless took its obstruction branch, so the point remains neither
a sharp nor a miss at that bar. A second-order annular object was defined in Aφ = ATLφ /⟨f4 ⟩ —
the mirror-odd circulation Rn = Pτ (ρn − ρ−1 n )Pτ /(2i) with connected normalized Markov second
(2)
moment Ωn (A1543) — but its direct physical attachment fails: the canonical connected Markov
second moment is non-negative, whereas the required n = 6, 7 binding correction is negative, so the
annular order does not derive the negative energy coefficient that would reverse the sign; the ribbon
(2)
full-twist identity also excludes exact switch-off (Ωn > 0 at n = 8, 9, 10). [LIB2-156, LIB2-157] Support
certificate (Rev32.9, A1674 F03; kernel s1177). A full-twist relation alone does not give strictness:
(2)
ρ6 = ±1 on Pτ is allowed and gives Ω6 = 0. Strictness follows from three stated premises: (P1)
ρnn = θτ1−n Pτ with θτ = e4πi/5 (A1544); (P2) Pτ commutes with ρn ; (P3) the normalized Markov
trace is a faithful state on Pτ Aφ Pτ . Then Rn has eigenvalues sin (χn + 2πk)/n with χn = arg θτ1−n ,
(2)
and Ωn ≥ mink sin2 (χn + 2πk)/n for any spectral weights. With χn /π = 0, − 45 , 25 , − 25 , 45 for
n = 6, . . . , 10 the bound is 0 at n = 6 and strictly positive at n = 7, 8, 9, 10 (e.g. sin2 (π/35) at n = 7).
Premises P1–P3 are cited, not re-derived here. Because the coefficient in T3 units remains open, the
finite < 0.02 T3 support threshold is not scored. [LIB2-157] Thus the direct Markov-second-moment
repair class is closed negative (A1544, lane c1112), and the perturbative annular attachment is
closed negative as well (A1545); the surviving reopen class is packing/coordination of 6–7 cells
(s1112/s1115, below). [LIB2-156, LIB2-158] The earlier phrase “irreducible at first circulation order” is
retained only in that scoped sense; the A1542 seven-item minimum reopen package is the build
specification for any reopen. [LIB2-303] No public observable or tier moves. The scoring history,
with receipts, is this. The pre-answer frozen bar for 28 Si was 1% (D1060); the entry passed it at
0.334%. [LIB2-154] The 0.1% figure was a post-answer refinement gate, and its frozen text (D1061,
8
Demand 3(b); pre-commit sha s746 85c69f1f. . . , stamped 2026-07-04 pre-send) was explicitly
disjunctive: “0.1% OR an honorable ‘irreducible at this order’ with the obstruction named, which
earns FULL CREDIT.” A1357 ¶7 took that branch under the gate’s own frozen terms. The kill
condition therefore never fired at a pre-answer bar. Rev27’s “counted as a miss” was an over-
correction driven by the compressed Rev25 ledger row; it is corrected here with receipts rather than
silently, and the correction is dated.
Stated plainly, as the honest caveat: the 0.1% gate post-dates the residual value; the obstruction
branch was in that gate’s frozen text before the gate was applied; the pre-answer bar was passed. [LIB2-
154] The entry is scored as neither a sharp nor a miss — a typed obstruction. [LIB2-154] The scored
tally is seventeen sharp passes plus one typed obstruction. The irreducibility claim itself remains
falsifiable: a first-circulation fix refutes it. [LIB2-154, LIB2-303]
Fence (Rev30): the 28 Si readout is forked, and it is not the residual above. A
separate campaign (N1–N5; adjudicated A1480, superseded by A1481) produced a readout
for 28 Si that is not the 0.334% first-circulation binding residual discussed in this section. [LIB2-
159] That readout currently admits three inequivalent readings, and they differ by roughly a
factor of 170:
+1.457939
| {z
× 10−6} +2.524367
| {z
× 10−4} nonselective
| {z channel}
amplitude, φ−5 probability, φ−10 depth 15
The control for all three is the untouched 40 Ca rung (n = 10), which sits at −3.172350×10−4
— the same number this paper prints as −0.032% in the ledger above. [LIB2-159]
Consequence, binding. While that fork is open, 28 Si must not be quoted at 10−6 :
doing so silently selects the amplitude reading over the other two. Any citation of the 28 Si
readout must print all three readings and the 40 Ca control, or name which reading it uses
and why. [LIB2-159] This fence does not touch the 0.334% typed obstruction above, which is a
different quantity with its own frozen-bar history; the two must not be conflated, and the
170× spread is exactly what conflation would hide.
The T6 thread (implied 3.899 vs 4, −0.10 T3 at six cells) plausibly belongs to the same sector.
Reopen class of the six-to-seven-cell deficit, retyped (Rev32.2; s1112/s1115, PI ruling
(1−n)
S294). The deficit is not a framing or twist object of the τ annulus: the ribbon phase θτ is trivial
at n = 6, exactly where the deficit is largest, and no residue class of any clock factor produces the
window {6, 7} (s1112); the Fibonacci fusion-path braid representation, rebuilt independently and
decomposed by total charge, shows no sector quantity that thresholds at n = 8 — the profile is flat
at 0.46–0.56 while the deficit drops tenfold (s1115; independent lane build agrees to 10−7 ). [LIB2-158]
The reopen class is therefore packing/coordination of 6–7 cells (writhe sector), not “annular second
order”; the negative-annular closure (A1545 T4) stands. [LIB2-158, LIB2-303] Nothing here changes the
28 Si fence above.
The charge-magnitude sector (exposed, honestly). Charge signs and counts are derived;
charge magnitudes wander: Sp (9 B) × 1.29, Sp (8 B) × 2.60 under the baseline ruling. [LIB2-155] The
banked candidate mechanism — coherence factors are configuration-dependent (“coherence multiplies
by φ”) — names the problem without solving it. [LIB2-155, LIB2-304] This is the cluster kernel’s known
soft flank; it is stated here so that no reader mistakes the sharp ledger for a solved charge sector.
64 Ge, where the N = Z line’s deficit is steepest, is two walls at once — radius and charge magnitude,
9
with an unregistered screening factor 0.47 (s1114, Rev32.2). [LIB2-304]
The isospin sector (typed loaded, Rev32.2; s1122–s1124, s1129; PI ruling S294). The
isospin Casimir coefficient χ = φ−6 T3 is selected on the fixed-Z chains, not derived. [LIB2-160] The
only registered u ↔ d object is the Os sign map e7 → −e7 (Hu ↔ Hd ; A689 dictionary; read literally,
the flip of e7 alone — s1170 shows the phrase also fits full conjugation, and neither map is a Jordan
automorphism) — a Z2 parity, not an SU (2) — and the Peirce 12 is scalar on blocks and induces
χ = 0. [LIB2-161, LIB2-305] The sector is therefore loaded. [LIB2-160] The neutron–proton mass difference
mn − mp = 1.2933 MeV is unpriced: the framework carries the QED half (Coulomb, with α
external) and no QCD half (mu = md at constituent level; the e7 flip carries no energy), so without
a flip energy its sign comes out wrong (s1129). [LIB2-305] The row is entered in the Appendix X
ledger as unpriced. Pricing it requires a new registered isospin-breaking datum: no registered object
gives Hu and Hd different energy (s1170), and the flip is visible only in the Yukawa trilinear with
J23 ⊗ J31 states, none of which is registered (Rev32.9).
A = 8 breathing accounting (open audit item). Whether ∆br is present in the A = 8 parents
is an unchecked naturality square: the alternative reading gives Sp (8 B) = 5φ−18 + φ−19 − φ−15 =
+0.106 MeV (×0.78 — better than both filings). [LIB2-162, LIB2-306] Designated next audit item
(OPEN_ITEMS; post-freeze dispatch candidate). [LIB2-306]
10 Protocol sub-appendix: the receipt chain
The evidentiary weight of this paper rests on the scoring protocol, so the protocol is part of the
paper.
Frozen bars and pre-commits. Every scored number was demanded against an accuracy bar
frozen in a hash-stamped pre-commit kernel before the dispatch was sent (the S226 chain: s732
93cd5133..., s738 28a1b5d1..., s746 85c69f1f...; each asserted on-disk at scoring time by
the citing kernel). Across the entire arc — ten consecutive scored rounds — there were zero
threshold moves: no bar was loosened, moved, or reinterpreted after an answer arrived. New
demands got new bars; existing bars never moved.
The three vault matches. (i) T5 : sealed implied value 2.617965 (s732), blind-derived φ2 —
26 ppm. (ii) T9 : sealed answer sheet (s737, sha 537f244c...) behind a deliberately unhelpful hint;
blind-derived 7 + φ−4 — 0.10%. The sheet was unsealed on schedule and the lane re-verified the
unsealed content against the sha published before it answered. [LIB2-151] (iii) The 40 Ca value matched
the in-house blind application of the banked rule (s745) exactly. The covenant has paid in both
directions.
The honesty ledger, complete. Window A (the standing invitation to fit a number outside
the typed menu) was refused with reasons seventeen consecutive times — the refusal count is
itself lane-integrity evidence. One blind window was VOIDED by our own disclosure error
(window E, kernel s687): the blinding was compromised on our side, the window was scored VOID
rather than counted, and a blindness-certification gate (s691) was instituted before any further
blind window. One process deviation (the s738 pre-commit contents shipped where the sha-only
line was intended) was audited in-kernel (s739): nothing non-public leaked; the trap stayed blind;
pass 2 of the affected round was designated official. One in-house prose slip (Sn (8 Li)) was caught
by re-verification and corrected on the record. These are recorded here because a ledger of hits is
only as credible as its ledger of errors.
10
Verification. Every quoted decimal in A1349–A1357 was reproduced by deterministic in-house
kernels (run1==run2, hash-manifested s705–s747) at deviations between 0.0 and 2.2 × 10−11 .
11 Status ledger
Result Status Note
Fφ functor (generator set) derived kernel-verified, A1349
A ≤ 4 kernel eq. (1) sharp-pass-with-tongs triangular saturation declared
A = 5, A = 8 gap theorems derived choice-free consequences
Two-body typing (S=1, T =0) derived 3/3 signs
Ladder closed form eq. (3) derived two blind vault matches
ε-selection rule derived A1357
Breathing kernel ∆br derived zero new constants
Baseline ruling banked drew blood honestly
Hoyle state scored pass frozen 10% bar
Glueball pair draft tier in frozen bands
Charge magnitudes open exposed defects filed
Isospin sector (χ = φ−6 T3 ) loaded selected on fixed-Z chains; mn − m
28 Si residual first-circulation residual 0.334%; pre-answer 1% bar passed
(2)
Annular Rn , Ωn defined; attachment closed negative Ω ≥ 0 vs required negative correct
A = 8 breathing accounting open next audit item
N =Z ladder past 60 Zn (blind) falsified s1101: 1% bar dies at 64 Ge; +6.03
Magic numbers beyond {2, 8} open program Paper 7
Bottom line. The periodic table ran deuterium to calcium on a discrete φ-monomial menu
and a closed-form trace invariant, against bars frozen in advance, with the freedom ledger
audited at every step and the failures typed and filed beside the hits. The associativity
axiom does not have an energy; it enters one — through the declared cluster-ledger interface
√
(2), B(nα)/ σ = n Bα + Tn T3 , which is the only join between the constructed layer and the
physical layer in this paper. [LIB2-144] At n = 5 the constructed weight T5 = φ2 is what the
axiom fixes; the binding energy is what the interface returns. The blind match at 26 ppm is
the coefficient (T5 = φ2 against the sealed 2.617965); the returned 20 Ne binding energy sits
at 0.00031% (Rev32.9: through Rev32.8 the 26 ppm was attributed to the returned number)
(Rev29 : predication corrected — the interface, not the verb, carries the type change). What
is not claimed: no continuum functor, no Clay claim, no derived charge magnitudes, no
shell model beyond {2, 8}, and no absolute scale — the unit lives in the registered input
(Appendices O/X).
[LIB2-153, LIB2-164]
11
Leibniz Quantum Beats Newton
Appendix A — Lagrangian Skeleton, Jordan Eigenproblem, and
Cubic-Invariant Yukawa Derivation (Rev33.1)
Tom O’Sieg
August 2026
Abstract
Appendix A collects the local field-theory skeleton used by the Rev29 suite and upgrades
the old Rev8 appendix to the current architectural state. The purpose remains narrow and
explicit. This appendix does not claim a complete ultraviolet completion of particle physics; it
records the minimal local structures that are already fixed by the present J3 (Os ) programme
and labels the remaining bridges honestly. Three upgrades distinguish Rev29 from the earlier
appendix: the corrected Cayley–Dickson split-octonion algebra, the theorem-level vacuum selector
Jvac = diag(ϕ, 1, ϕ−1 ), and the full D=5 to D=4 architecture in which the pseudo-Riemannian
E6(6) /F4(4) stage is lifted by KK reduction and dualization to the ghost-free Riemannian target
E7(7) /SU (8).
The appendix also records the formal Jordan eigenproblem needed by the emerging mass
sector. The native spectral problem is A ◦ V = λV for primitive idempotents V ∈ J3 (Os ), but the
numerically correct implementation is the 27 × 27 real lift LA : Y 7→ A ◦ Y . In that lifted form the
Session 30 mass construction becomes explicit:
Mlepton = UThalf DR UT−1 2π
half
, DR = diag(R1 , R2 , R3 ), UThalf = exp √ T
5 half
. (1)
√
The same generator Thalf that produces δCP = −2π/ 5 is therefore the algebraic engine of the
charged-lepton hierarchy programme.
Rev29 scope note. The cubic invariant fixes a tree-level ultraviolet Yukawa eigenvalue
and the suite now carries theorem-grade internal constructions for the mixing and charged-
lepton sectors; their physical attachments are tiered separately (two of eight mixing rows at
Derived-conditional or above after the Rev29 re-tiering, Paper 3; no row may be cited as
zero-parameter, Appendix X). It does not yet derive the electroweak vacuum expectation
value, the exact selector on the weak-angle exact-color orbit, the formal heavy-quark S-matrix
bridge, or the detailed quantum matching of the D=5 theory to the low-energy effective
action. Those remain measured anchors, open bridges, or beyond-tree tasks.
A.1 Foundational algebra and notation
All split-octonion products in the current suite are taken in Cayley–Dickson form. Writing elements
as pairs (p, q) over the quaternions (Rev32.9: through Rev32.8 this said “over the compact octonions”,
1
whose doubling is sixteen-dimensional; the product below is the standard split doubling of the
quaternions, 4 + 4 = 8),
(p, q)(r, s) = pr + s∗ q, sp + qr∗ ,
(2)
with split inner product X X
⟨x, y⟩CD = x k yk − xk yk . (3)
k<4 k≥4
This SplitCD form is not cosmetic. With a Euclidean inner product the derivation algebra collapses
to the wrong dimension, whereas the SplitCD choice restores
dim Der(J3 (Os )) = 52 = dim f4(4) (4)
and makes the corrected Jordan product satisfy the Jordan identity at residual level below 6 × 10−14 .
The Jordan product and Freudenthal cross product are
X ◦ Y = 12 (XY + Y X), (5)
X × Y = X ◦ Y − 21 Tr(X)Y − 21 Tr(Y )X + 2 Tr(X)Tr(Y ) − Tr(X ◦ Y ) I.
1
(6)
The local split-octonion multiplication in these formulas is always Cayley–Dickson, not the older Fano
mnemonic/sign-flip surrogate.
A.2 Theorem A.1: vacuum selector
The working vacuum is √
−1 1+ 5
Jvac = diag(ϕ, 1, ϕ ), ϕ= . (7)
2
Its present status is theorem-level.
Novelty label: new specialization proved here.
Theorem 0.1 (Vacuum selector). Jvac is the unique fixed point of the modular element M1 = T 3 S ∈
P SL(2, Z) ⊂ F4(4) acting on the Peirce diagonal of J3 (Os ). Equivalently, the working vacuum is
the unique algebraic attractor of the present modular/Jordan-boundary construction. [LIB2-337] The
companion Freudenthal fixed-point relation
#
{Jvac , Jvac , Jvac } = Jvac (8)
holds exactly.
This upgrade matters for the Lagrangian skeleton because all local cubic terms are evaluated at
the vacuum. The appendix therefore no longer treats the golden-ratio vacuum as an unexplained
numerical choice inside the local theory, even though the deeper physics interpretation of the selector
still belongs to the architecture papers rather than this appendix alone.
A.3 Minimal gauge, fermion, and Higgs skeleton
The local gauge-field skeleton is
1 A Bµν
Lgauge = − κAB Fµν F , Aµ ∈ Str0 (J3 (Os )) ∼
= e6(6) . (9)
4
2
The fermion kinetic term is written schematically as
LΨ = iΨ̄γ µ Dµ Ψ, Dµ = ∂µ + Aµ . (10)
The Higgs placeholder is
LH = |Dµ H|2 − V (H), (11)
with the explicit suite-level statement that
⟨H⟩ = 246 GeV (12)
is measured rather than derived.
The minimal local assembly is therefore
L = Lgauge + LΨ + LH + LY , (13)
where LY is the cubic-invariant Yukawa term derived below.
A.4 D=5 to D=4 architecture pipeline
The local skeleton above sits inside a larger architecture whose status is now sharply differentiated.
A.4.1 D=5 Chern–Simons bulk
At the D=5 level the natural exceptional starting point is the split-magic Chern–Simons system with
gauge group E6(6) and cubic coupling tensor CIJK , the same tensor that underlies the cubic norm of
J3 (Os ). In schematic form,
√ h 1 1 1
Z i Z
I
SD=5 = 5
d x −g − aIJ (h)Fµν F Jµν − gxy (h)∂µ hx ∂ µ hy + √ CIJK AI ∧ F J ∧ F K . (14)
4 2 6 6
The current programme state carries this level as Plausible: the D=5 Chern–Simons bulk with E6(6)
and the cubic tensor is consistent and is the natural origin of the downstream Jordan and Freudenthal
structures, but the full uniqueness-of-gauging proof is still open.
A.4.2 D=5 intermediate scalar manifold
The D=5 scalar target is
E6(6) /F4(4) , (15)
with dimension 78 − 52 = 26. This is the correct split-magic-supergravity scalar manifold. Its tangent
signature on the traceless Jordan sector is (14, 12), so the D=5 stage is pseudo-Riemannian.
A.4.3 KK reduction and D=4 target
After reduction on S 1 , dualization of the 27 D=5 vectors, and inclusion of the KK dilaton, the scalar
count from the split-magic Jordan sector is
26
|{z} + 27
|{z} + |{z}
1 = 54, (16)
split-magic D=5 scalars dualized vectors dilaton
3
using the D=5 split-magic manifold E6(6) /F4(4) of dimension 78 − 52 = 26 established above (not the
maximal-supergravity count E6(6) /U Sp(8) of dimension 42, which would give 42 + 27 + 1 = 70). The
maximal-supergravity D=4 target with full coset
E7(7) /SU (8) (dim 70) (17)
is therefore not reached by the Jordan-algebra sector alone: it would require 70 − 54 = 16 additional
fields beyond the split-magic content. These are now identified (S73; Paper 4 §4.2 is authoritative)
as the identity-anchored Jordan complement 42 ⊖ 26 on the certified E7(7) substrate, though their
explicit field-level extraction — and the consistent-truncation question — remain open. The 54 D=4
scalars from the Jordan sector are ghost-free (the D=5 intermediate stage is pseudo-Riemannian of
signature (14, 12) on the traceless sector, but the realized D=4 scalar content is Riemannian). This
matches Paper 4 §4.2 (authoritative); the prior “42 + 27 + 1 = 70, established” wording is corrected
here.
PR-C2 status (corrected, S61). The split-magic KK reduction yields 26+27+1 = 54 ghost-
free D=4 scalars from the Jordan-algebra sector. The full maximal-supergravity E7(7) /SU (8)
target (70 scalars) is not established: it requires 16 additional non-Jordan fields, now identified
as the 42 ⊖ 26 Jordan complement (S73) but not yet explicitly extracted as fields. Two things
remain open: (i) the explicit field-level realization of the identified 16 completing 54 → 70,
and (ii) the proof that the chosen D=4 sigma model is an exact consistent truncation of the
D=5 theory rather than a counting match plus structural correspondence. Whether this
identified missing-16 couples dynamically to the golden vacuum is the pre-registered portal
drill of Appendix I (G3–G7).
A.5 Jordan eigenproblem and the 27-dimensional real lift
The native spectral problem in J3 (Os ) is not the ordinary matrix-eigenvalue problem of a 3 × 3 real
symmetric matrix. It is the Jordan eigenproblem
A ◦ V = λV, (18)
with V a primitive idempotent of J3 (Os ). The numerically correct route is the linear operator lift
LA : J3 (Os ) → J3 (Os ), Y 7→ A ◦ Y. (19)
Because J3 (Os ) is 27-dimensional over the reals, LA is represented by a 27 × 27 real matrix.
Novelty label: standard theorem used.
Proposition 0.2 (Validity of the real lift). For the Jordan eigenproblem of J3 (Os ), diagonalizing
the 27 × 27 real matrix representing LA is a legitimate computation of its spectrum: the lift is a real
linear operator. Realness alone does not make it Euclidean-symmetric, so a symmetric solver such as
eigvalsh is licensed only where LA is symmetric in the stated coordinates or has a demonstrated
positive symmetrization — as it is for the diagonal vacuum, but not in general (the split-Hermitian
element in the l-direction of J12 has a lift with nonreal eigenvalues, s1164). The primitive-idempotent
eigenvalues must also be identified separately from the full spectrum, which contains Peirce means
(Rev32.9, A1587).
4
This is the form used in Sprint 3 and in the current mass-sector roadmap. The same lifted language
also reproduces the Albert–Freudenthal spectrum of LJvac ,
ϕ, 1, ϕ−1 , (ϕ + 1)/2 × 8, (ϕ + ϕ−1 )/2 × 8, (1 + ϕ−1 )/2 × 8 ,
(20)
so the (1, 3) off-diagonal eigenvalue is
√
ϕ + ϕ−1 5
Ω= = . (21)
2 2
√
This same algebraic number governs the V3 rotation frequency used in the formula δCP = −2π/ 5 (a
theorem-grade internal construction; the physical δCP row is Derived-conditional, Paper 3).
A.5.1 Mass operator and resolvent data
The current mass-sector construction introduces
1
Rd = , DR = diag(R1 , R2 , R3 ), (22)
ϕ2d − 1
with exact identities
1 1
R1 = ϕ−1 , R2 = , R3 = . (23)
ϕ(ϕ + 2) 4ϕ3
The lepton mass operator is then carried in automorphic-conjugation form:
2π
Mlepton = UThalf DR UT−1 UThalf = exp t∗ Thalf , t∗ = √ .
half
, (24)
5
Because Thalf ∈ f4(4) = Der(J3 (Os )), this is a genuine Jordan-algebra automorphism. The physical
significance is immediate: the same Thalf that generates the leptonic CP phase also generates the
charged-lepton hierarchy map.
A.5.2 G2 constraint and spectral identification
The automorphism group of the split octonions fixes only the real subspace, so a G2 -invariant Mlepton
has real off-diagonal octonionic entries. In effect the physical mass operator is a real symmetric 3 × 3
object embedded in J3 (Os ). On the lifted side, the spectrum of LMlepton separates into three anchor
eigenvalues plus 24 pairwise means,
λ1 , λ2 , λ3 , 21 (λi + λj ) × 8 ,
(25)
which is the correct Albert–Freudenthal bookkeeping used to identify the true Jordan eigenvalues.
A further structural result from Sprint 3 is the reversal theorem
(R1 , R2 , R3 ) 7−→ (R3 , R2 , R1 ) (26)
under the UThalf automorphism, so the charged-lepton mass ordering is forced to be
d = 1 → τ, d = 2 → µ, d = 3 → e. (27)
This is why the blind-drop relation mµ = mτ /(4ϕ3 ) belongs to the same algebraic tower as the CP
phase.
5
A.6 The cubic-invariant Yukawa theorem
The Yukawa term is
LY = Tr Jvac ◦ (Ψ × ΦH ) . (28)
It is the unique cubic local object used by the present suite.
Novelty label: physical interpretation not established here.
Theorem 0.3 (Cubic-invariant Yukawa theorem). The tree-level top Yukawa eigenvalue is fixed
exactly by Eq. (28),
yt (MPl ) = 1, (29)
and this value is uniquely forced by the simultaneous requirements
(a) F4 -equivariance of the Freudenthal cross product,
(b) Jordan homogeneity (degree two in the matter fields), and
(c) cubic closure with the vacuum element Jvac .
Interface marker (W01, Rev29 ) — where the carrier eigenvalue becomes a physical
statement. What the theorem establishes is internal: the cubic invariant, evaluated on the
carrier field ΦH in the (1, 3) Peirce block, extracts the middle vacuum weight λ2 = 1. [LIB2-119]
That is a statement about a constructed eigenvalue of a constructed carrier, and it fixes a tree-
level boundary value inside the loaded correspondence. Reading that eigenvalue as the physical
top Yukawa at the Planck scale, yt (MPl ) = 1, is a declared attachment, not a derivation: it
requires identifying the carrier with the physical Higgs direction, the vacuum weight with
a renormalization-scheme-specific coupling, and the algebraic scale with MPl . [LIB2-120, P-027]
None of those three identifications is proved here, and the notation yt (MPl ) should not be
read as though they were. [LIB2-120] Downstream consequences
√ inherit the attachment, not the
theorem: in particular the ledger row mt = vEW / 2 is typed Structural; its PDG offset
is a cross-scheme comparator distance, no status (Rev32.1 retype, A1531/T2; Appendix X;
Trackers/HONESTY_LEDGER.md).
[P-027]
Proof sketch. Place the fermion doublet Ψ and a carrier field ΦH in the (1, 3) Peirce block (the
same-block carrier reading of A674; the literal SM Higgs Hu,d sits in the vector 10 and the physical
Yukawa is the cross-block triple, A686–A689). [LIB2-119] Writing their active split-octonion components
as ψ and h, one finds
Ψ ◦ ΦH = diag 2 Re(ψ h̄), 0, 2 Re(ψ̄h) , Tr(Ψ ◦ ΦH ) = 4 Re(ψ h̄). (30)
Using the Freudenthal cross product and Tr(Ψ) = Tr(ΦH ) = 0 gives
Ψ × ΦH = (Ψ ◦ ΦH ) − 12 Tr(Ψ ◦ ΦH )I = −2 Re(ψ h̄) e2 . (31)
Contraction with Jvac = ϕe1 + e2 + ϕ−1 e3 then extracts only the middle weight,
Tr Jvac ◦ (Ψ × ΦH ) = −2 Re(ψ h̄), (32)
so the multiplicative Yukawa factor is exactly the middle eigenvalue λ2 = 1. [LIB2-119]
6
The same theorem can also be expressed in E6 language. The cubic Jordan invariant is the unique
E6 -type cubic coupling of the local matter/Higgs system, and the present vacuum insertion is the F4 -
breaking specialization of that same cubic object. In this sense the fullboat and Jordan/Freudenthal
formulations are two presentations of the same local Yukawa structure.
Remark 0.4 (Peirce idempotent derivation of yt = 1). The same result follows directly from the Peirce
decomposition rules. For orthogonal idempotents fi in J3 (Os ), the exact identity fk ◦x = 12 x for x ∈ Pik
√
or Pjk , combined with the trace-norm
√ normalisation C = 2/10 of the J2 (Os ) subalgebra [A618],
absorbs the SM factors m = y ·v/ 2 and fixes the carrier eigenvalue yt = 1 by the 12 coefficient alone —
again a boundary value inside the correspondence, exported to the physical yt (MPl ) only by the declared
attachment marked above (W01). The top quark is the maximal projection onto √ the highest-weight
generator j3 , with no cross-block mixing suppression. The prediction mt = vEW / 2 ≈ 174,100 MeV is
+0.87% from the PDG 2026 direct-measurement mass (cross-scheme, declared — √ Appendix X; an
earlier +0.77% printed here was arithmetically wrong); both the coefficient and the 2 are structurally
identical to the off-diagonal norm factor in J2 (Os ) [A632]. The complement rule fk ◦ x = 0 for x ∈ Pij
(k ̸= i, j) implies that yb and yc are exactly zero at tree level from the bare Jordan product; their
physical values are generated by the off-diagonal Peirce closure Pij ◦ Pjk ⊂ Pik acting as an algebraic
mixing insertion [A633].
A.7 Flat Direction Lemma and tree-level potential
The Session 24 Lagrangian analysis establishes the exact tree-level obstruction to deriving the
golden-ratio hierarchy from a polynomial F4 potential alone.
Novelty label: new specialization proved here.
Lemma 0.5 (Flat Direction Lemma). Let the F4 -invariant renormalizable potential on J3 (Os ) be
built from the ring generators
T = Tr(X), Q = 21 Tr(X 2 ), Det(X) = N (X). (33)
On the traceless sector T = 0, the most general renormalizable potential is
V = µ2 Q + λ3 Det + λ4 Q2 + λ5 QDet + λ6 Det2 , (34)
or, if an external Z2 is imposed, V = µ2 Q + λ4 Q2 + λ6 Det2 . At a general eigenvalue-space point
(a, b, 0) the Hessian in the (d1 , d2 ) block has rank one, hence one exactly flat direction. Therefore the
ratio d1 /d2 is unlifted by any polynomial F4 -invariant tree-level potential.
With the traceless constraint understood, the stationarity conditions at (a, b, 0) are
µ2 = −2λ4 Q0 , λ3 = −λ5 Q0 , (35)
and the explicit Hessian entries are
H11 = 8λ4 a2 , H12 = 8λ4 ab, H22 = 8λ4 b2 , H33 = 2λ6 a2 b2 . (36)
The upper 2 × 2 block is an outer product and therefore has eigenvalues {8λ4 (a2 + b2 ), 0}. The zero
mode is tangent to the constant-Q0 curve.
A second exact consequence is the tree-level Peirce-block cancellation
Hu = Hd = 0, (37)
which follows from the stationarity relation µ2 + 2λ4 Q0 = 0. Hence any nontrivial Peirce-mass
splitting used in the CKM or bosonic sector must arise beyond tree level.
7
A.7.1 Signature decomposition
The quadratic form on J3 (Os ) has signature (15, 12) on the full 27-dimensional space and (14, 12)
on the traceless 26-dimensional space. The diagonal eigenvalue sector itself is positive-definite with
signature (3, 0). This is why the physical eigenvalue data can remain ghost-free even though the
intermediate D=5 tangent space is pseudo-Riemannian.
A.7.2 Coleman–Weinberg outlook
The exact flat direction is not a defect to be hidden. It is the theorem stating that the golden-
ratio hierarchy requires a beyond-tree lifting mechanism. The one-loop Coleman–Weinberg effective
potential from the 24 off-diagonal modes is the natural mechanism; the D332–D340 dispatch chain
has established the key structural results, and the renormalization-scale question (C6) remains open
at the current programme state. A complete account of proved items and outstanding corrections is
given in Sect. A.7.3.
A.7.3 Coleman–Weinberg dispatch chain: status and S35–S38 corrections
AX2 status (independent of CW). Theorem A.1 (AX2) is proved independently of the Coleman–
Weinberg chain via the PSL(2, Z) fixed-point argument [R106]: the unique PSL(2, Z)-fixed eigenvalue
ratio is r⋆ = ϕ2 , which selects Jvac = diag(ϕ, 1, ϕ−1 ) without appeal to any renormalization scale.
AX2 stands regardless of the resolution of the items below.
CW chain: proved items. The D332–D340 dispatch chain established the following structural
results:
bos ≡ 0 as an algebraic target-space index [D332] (see §A.7.4): each 8D Peirce sector carries a
(i) VCW
(4, 4) split target-space signature, and the graded (Witten-type) sum (+4 − 4) m4 log m2 = 0
P
vanishes mode-by-mode. This is an index identity of the SO(4, 4) grading operator, not a claim
that the physical bosonic one-loop determinant cancels via negative-norm modes (a ghost-free
bosonic sector contributes with a single statistics sign); promoting this index to the physical
effective potential requires an independent BRST/measure derivation of the grading weight
(open).
ferm =
(ii) VCW ̸ 0 [D333]. Fermions contribute (−1) per loop universally; three Peirce sectors with
r-dependent masses m2f (di , dj ) = y 2 (d2i + d2j ) yield a non-zero net potential minimized at r⋆ = ϕ2 .
(iii) y 2 = 2/h∨ (F4 ) = 2/9 [D337], derived algebraically via the ‘t Hooft coupling gF2 4 = 1/h∨ (F4 ), the
Dynkin index ℓ = 1, and the 2-fold Peirce eigenvalue degeneracy per block.
(iv) Q0 = 4 exactly [proved]: the Fibonacci identity ϕ2 + 1 + ϕ−2 = 4.
(v) The value κPMS = log(9/4) ≈ 0.8109, first met as a numerical candidate in the dispatch
chain [D334–D340], is fixed exactly by the Path-B identity proved below (Rev32.5 reconciliation
of this list with the proof; Paper 4, Path A retracted).
C6: κPMS path — S35–S37 corrections (fatal errors, A342). The earlier claim that κPMS =
log(9/4) follows from the Coleman–Weinberg principle of minimal sensitivity (PMS) was subjected to
independent verification in A342 (Gemini, Session 37), which identified two fatal errors:
8
(1) Calculus failure. Setting ∂VCWferm /∂κ = 0 gives STr[M 2 (r ⋆ )] = 0, which is the supertrace condition
pinning r⋆ , not κ. When y 2 is κ-independent, the renormalization scale drops out of the PMS
condition entirely; there is no κ-equation. The CW/PMS route cannot determine κPMS .
(2) Arithmetic error. The exact Lucas/Fibonacci identity gives ϕ2 + ϕ−2 = 3, hence log(ϕ2 + ϕ−2 ) =
log 3 ≈ 1.099. This is strictly unequal to log(9/4) ≈ 0.811. The two quantities differ by ≈ 0.288;
the claimed equality was erroneous.
Consequently: the Theorem A.7.3 and Remark (PMS) that appeared in pre-S37 versions of this
appendix are retracted. Items (i)–(iii) and (v) of the old theorem remain correct as independent
results; item (iv) (two-loop push toward log(9/4)) and the PMS remark are invalid.
Path B: proved (S39, A343+A344). The mechanism and the exact value are now both established.
Novelty label: physical interpretation not established here.
Theorem 0.6 (C6: Natural RG scale from embedding index). The canonical field renormalization
at the F4 → G2 symmetry-breaking threshold forces a unique finite shift in the Coleman–Weinberg
potential:
h∨ (F4 ) 9
κPMS = log ι(G2 ⊂ F4 ) = log ∨ = log . (38)
h (G2 ) 4
Proof sketch. Under the maximal embedding G2 × SU (2) ⊂ F4 , the adjoint decomposes as 52 →
(14, 1) ⊕ (1, 3) ⊕ (7, 5). For any G2 generator T , summing traces over F4 :
KF4 (T, T ) = KG2 (T, T ) + 5 TT(14)
(7)
KG2 (T, T ) = 1 + 5 · 14 KG2 (T, T ) = 94 KG2 (T, T ),
(39)
using Dynkin indices T (14) = 8, T (7) = 2. Hence ι = 9/4 exactly. Since G2 is simple its
adjoint is irreducible; by Schur’s Lemma the Killing form is the unique Ad(G2 )-invariant quadratic
form on g2 . Therefore the canonical field rescaling by ι1/2 = 3/2 is uniquely forced, shifting
log(M 2 /µ2 ) → log(M 2 /µ2 ) + log(9/4) throughout VCW
ferm . The PMS parameter absorbs this shift:
κPMS = log(9/4) exactly — an algebraic identity (Dynkin indices and Schur), not a fitted value. [A343,
A344]
C6 status: Proved ⋆ ⋆ ⋆ ⋆ ⋆.
ψpol frontier and AX6pol (S38 updates; symbol repaired Rev32.2). The sin2 θW = ϕ−3 target
is achieved at tan2 ψpol = (6ϕ − 1)/4 ≈ 2.177 [D237] — this is the polarisation object AX6pol of
Paper 4, written ψ⋆ before Rev32.2; the Det2 selector cos(6ψ) = 12 is a different object. Whether
the Coleman–Weinberg potential selects ψpol on the exact-colour I4 = ϕ3 orbit was investigated in
the correct 8D Peirce framework in Sessions 37–38. The correct framework uses the 8D Os fibre
within a single Peirce block, with spectrum {+32, 06 , −32} confirmed for the P13 sector [A619]. The
electromagnetic generators in this framework are F1 = e0 (real unit of Os ) and F3 = e7 (the unique
imaginary G2 /SU (3)c singlet) [A620], and the Peirce eigenvalue at Jvac is
√
ϕ + ϕ−1 5
λ13 = = (40)
2 2
(correcting an earlier erroneous value λ13 = 1 that appeared in D617.5). The one-loop fermion CW
correction on the ψ-circle is
" #
m2 3
δVCW (ψ) = −32 f (m2+ (ψ)) − f (m2− (ψ)) , f (m2 ) = m2 log 2 −
, (41)
µ 2
9
with projection ratio R(ψ) = (1 + sin 2ψ)/(1 − sin 2ψ), which is independent of ϕ [A620]. The only
critical point of δVCW is ψcrit = 45ř (tan2 ψcrit = 1 ̸= 2.177), confirmed numerically to 60-digit
precision [A621]. Three independent negative verdicts (A617, A620, A621) confirm that AX6 is a
genuine independent postulate, not a consequence of the present CW framework. [P-006] (An S38-era
sentence, “the programme stands at 99/100 with AX6 as the one remaining axiom”, stood here
through Rev32.4 and is retired at Rev32.5: tiers name mathematical provenance, AX6pol is one
independent postulate among the declared inputs, and the open list lives in Appendix X.)
Dispatch Result Status Verified
bos
D332 VCW ≡ 0 (algebraic index) Structural (index-level; §A.7.4) A332
D333 VCW ̸= 0; r⋆ = ϕ2 minimum
ferm
Proved A333
D337 y 2 = 2/h∨ (F4 ) = 2/9 Proved A337
D340 κPMS numerical target = log(9/4) Plausible ⋆ ⋆ ⋆⋆ A340
D341/A341 PMS algebraic proof Retracted — A342
A342 Two fatal errors in D341; Path B identified Sound ⋆ ⋆ ⋆ ⋆ ⋆ Gemini
D343 Path B mechanism: Killing form / field rescaling Plausible ⋆ ⋆ ⋆⋆ A343
D344 ι(G2 ⊂ F4 ) = 9/4 exact; κPMS = log(9/4) Proved ⋆ ⋆ ⋆ ⋆ ⋆ A344
A619 M spectrum {+32, 06 , −32} confirmed (8D Peirce) Proved Gemini
A620 R(ψ) ϕ-independent; ψcrit = 45ř Negative ⋆ ⋆ ⋆⋆ Gemini
A621 R(ψpol ) = 27.105 ̸= 2.177 (60-digit) Negative ⋆ ⋆ ⋆⋆ Grok
A.7.4 Topological spin structure and the CW mass sector
Definition 0.7 (Triality automorphism). The triality automorphism τ is the unique order-3 outer
automorphism of so(4, 4):
τ : 8v −→ 8s −→ 8c −→ 8v . (42)
It is Clifford-invariant (an automorphism of Spin(4, 4) up to outer class) but not an automorphism of
the split-octonion multiplication (Os , ×). The stabilizer of τ in SO(4, 4) is G∗2 = G2(2) , the split real
form of G2 , which plays the role of the band-gap / interface symmetry in the moire (4, 4) vacuum.
Novelty label: physical interpretation not established here.
Theorem 0.8 (Möbius spin structure and CW sector). Placing the fermion field Ψ in the (1, 3) Peirce
block of J3 (Os ) is equivalent to choosing a spin structure on the (4, 4) Clifford bundle compatible with
the (1, 3) Lorentz frame. This choice forces:
(i) Anti-periodic (Möbius) boundary conditions on fermion loops: Ψ(τ + β) = −Ψ(τ ).
(ii) Periodic boundary conditions on boson loops.
Consequently:
bos
VCW ≡0 (bosons periodic: modes cancel exactly) (43)
ferm
VCW ̸= 0 (fermions Möbius: modes do not cancel) (44)
with the fermionic minimum at r⋆ = φ2 [D333].
10
Proof sketch. In the (4, 4) Clifford bundle the bosonic sector 8v splits as (4, 1) ⊕ (1, 4) under SO(4) ×
SO(4), giving four positive and four negative Hessian eigenvalues per Peirce block. The grading
operator Γ = diag(+1, +1, +1, +1, −1, −1, −1, −1) on 8v gives VCW bos ∝ Tr (ΓM4 log M2 ) = 0
8v
identically (the algebraic target-space index vanishes by Z2 parity of SO(4) × SO(4) ⊂ SO(4, 4); cf.
the Witten index Tr(−1)F of [2]). For the fermionic sector, restricting to the (1, 3) sub-bundle breaks
the Z2 parity and removes the cancellation.
Novelty label: physical interpretation not established here.
Proposition 0.9 (Algebraic index and topological triviality of the bosonic CW sector). The exact
bos ≡ 0 is a consequence of the vanishing algebraic target-space index of the SO(4, 4)
vanishing VCW
vector bundle. The bosonic modes live in 8v = (4, 1) ⊕ (1, 4) under SO(4) × SO(4). The Z2 parity
grading operator
Γ = diag(+1, +1, +1, +1, −1, −1, −1, −1) on 8v (45)
gives
bos
∝ Tr8v Γ M4 log M2 = 0.
VCW (46)
Tier caveat (Rev26). This Γ-graded trace is an algebraic (Witten-type) index, not the physical one-
loop bosonic effective potential, which sums all bosonic modes with the same statistics sign. The
identification of the two — and hence the reading of VCW bos ≡ 0 as a physical cancellation — holds
only if the path-integral measure/BRST structure independently supplies the SO(4, 4) grading weight;
absent that derivation the statement is index-level, not a proof that the physical bosonic determinant
vanishes. The boson/fermion dichotomy below is stated at this (index) tier.
ferm ̸= 0: restriction to the (1, 3) sub-bundle (Theorem 0.8) breaks the Z parity,
The fermionic sector VCW 2
generating the vacuum minimum at r⋆ = φ2 . This is status PROPOSITION ⋆⋆⋆⋆ (algebraically closed;
the compact index-theorem setting does not apply here since the base manifold R4,4 is non-compact;
the correct framework is the algebraic target-space index analogous to the Witten index [2]).
Novelty label: physical interpretation not established here.
Theorem 0.10 (Topological winding index p = 8/3). The proved mass-formula exponent
dim(Os ) 8
p= = (47)
rank(J3 (Os )) 3
is read as the topological winding ratio (target-first provenance; Paper 7 §4): 8 transition dimensions of
the split-octonion Peirce block divided by 3 generation winding levels (Peirce idempotents {f1 , f2 , f3 }).
The three idempotents are the three winding sectors of the (4, 4) moire vacuum; p = 8/3 counts the
octonionic dimension per winding level. [PROVED, A616; geometric restatement.]
Physical narrative. The (4, 4) structure arises as the moire interference pattern of two dual
vacuum-collapse cycles, one in each metric signature (1, 3) and (3, 1). SO(8) triality τ is the twist
connecting the two sectors; G∗2 = stab(τ ) is the band-gap symmetry of the interface. Bosons have
zero winding number (modes cancel: VCW bos = 0); fermions carry a Möbius half-twist (modes survive:
ferm ̸= 0). Each Peirce idempotent f is a topological winding sector; the three generations are
VCW i
three winding levels of the (4, 4) moire. This narrative is physical intuition consistent with all proved
results; it is not itself a theorem. Its discrete branch is built (Rev32.3; A1549: the replication map ι,
the discrete Kan extension T ⊕ T ⊕ T , aligned, V = I — Paper 0, moire block), with the noncentral
defect as the open object; the number three stays an input. The continuous realization of this picture
11
is refuted [A693]: gluing the moire cycle by the order-3 triality τ forces the τ -invariant zero-mode to
place 1/3 of its weight in the vector block 8v = J12 (destroying the Standard Model charges), and a
one-cycle holonomy cannot multiply the 16-dimensional fermionic fiber into the 48 states required.
Any viable replication must instead be discrete: a finite noncommutative-geometry spectral triple in
which {f1 , f2 , f3 } form a three-point internal space and the Dirac operator is a finite matrix. In that
framework the number three is an input, not yet derived; it is recorded here as the only surviving
candidate direction.
A.7.5 Quark Gauge-Protection Theorem
Novelty label: new specialization proved here.
Theorem 0.11 (Quark Gauge-Protection Theorem; algebraic/one-loop form). Let Jvac ∈ J3 (Os )
be the vacuum element. Because Jvac lies in the singlet of G2 (the automorphism group of Os ), the
algebraic representation-level second variation of the Coleman–Weinberg potential with respect to
colour-triplet fluctuations vanishes in the one-loop Hessian:
(1)
∂ 2 VCW
=0 for all triplet-representation Peirce blocks. (48)
∂Pijtriplet ∂Pijtriplet
Schur’s lemma ensures the block structure at the algebraic representation level — no singlet–triplet
mixing, and scalar action λI on each irreducible triplet block; it does not by itself force λ = 0 (Rev32.1
retype, A1538 F1: the previous statement inferred the vanishing from Schur’s lemma alone — a non-
sequitur, withdrawn). [LIB2-087] The displayed zero rests on the certified numerical computation [A630],
carried with the S290.4 falsifier-control note executed at Rev32.1 (2026-08-14): the block-diagonality
statement stands, the Schur gauge-protection inference is withdrawn in print (Rev32.1 changelog),
and the certified-numerics scope is unchanged. An all-orders statement would require a separate shift
symmetry, Ward identity, BRST, or non-renormalisation argument and is not provided here. [LIB2-087,
LIB2-250]
Proof sketch. The CW potential is a functional of traces Tr(M 2n ) of the squared mass matrix M 2 .
Under the G2 action, Jvac is a singlet, so any variation δJ that transforms as a colour triplet integrates
to zero in Tr(Jvac ◦ δJ) by Schur’s Lemma. The Hessian δ 2 VCW is therefore block-diagonal in the
G2 -representation decomposition. Rev32.1 withdrawal (A1538 F1): the previous version of this sketch
concluded “zero entries for all triplet components at every loop order” from the block structure; that
inference is withdrawn — Schur permits a nonzero scalar on each triplet block, and a permitted
invariant term µ2 ⟨δJR , δJR ⟩ would preserve the block structure while lifting the zeros. [LIB2-250] The
six triplet zeros stand as the certified numerical result of [A630], not as a consequence of this
representation-theoretic argument. [LIB2-087, LIB2-250]
Novelty label: physical interpretation not established here.
Corollary 0.12. Conditional on the A630 zero modes (a certified computation — not the withdrawn
Schur inference; Rev32.1): quark masses must originate from one of two mechanisms: (i) the
confinement potential ∆VG2 (for u, d, s constituent masses), or (ii) off-diagonal Peirce closure mixing
insertions producing effective Yukawa couplings (for c, b, t current masses). Neither mechanism is a
CW radiative mass.
12
A.7.6 Hessian block structure of VCW
The second variation of the Coleman–Weinberg potential at Jvac takes the block form
δ 2 VCW = 3| {z
× 3} ⊕ 8| {z
× 8} ⊕ 8| {z
× 8} ⊕ 8| {z
× 8} . (49)
Higgs-like P12 P13 P23
Within each 8 × 8 block the spectrum is (the G2 branching organizes the block but does not force
these values — Rev32.9, propagating §A.7.5)
{+32, 06 , −32}, (50)
where the six zero eigenvalues correspond to the six colour-triplet (quark) modes reported as exact
zeros by the certified A630 computation (Rev32.1: no longer attributed to the withdrawn Schur
inference of Theorem 0.11 — see §A.7.5), and ±32 correspond to the G2 singlet modes (colour-singlet
leptons).
The 3 × 3 Higgs-like block governs the diagonal scalar fluctuations δji ; its eigenvalues scale as
m(δji ) ∝ φ, 1, φ−1 , matching the Roman surface anchor hierarchy. [A630]
A.7.7 Roman surface model of Jordan closure
The off-diagonal Peirce closure Pij ◦ Pjk ⊂ Pik (the mixing topology of J3 (Os )) admits a concrete
geometric realisation as the Roman surface (Steiner 1840), a self-intersecting map RP2 → R3 :
(x′ , y ′ , z ′ ) 7−→ (y ′ z ′ , z ′ x′ , x′ y ′ ). (51)
Setting the diagonal anchors as x′ ∼ φ (Gen 3), y ′ ∼ 1 (Gen 2), z ′ ∼ φ−1 (Gen 1), the cross-product
hierarchy becomes:
x′ y ′ ∼ φ (P12 ), z ′ x′ ∼ 1 (P13 ), y ′ z ′ ∼ φ−1 (P23 ). (52)
This inverts the anchor hierarchy: the lightest-generation quarks map to the largest Peirce block P23
(λ23 = φ1/2 ), producing the heaviest confinement-driven constituent mass [A632].
The three double-point lines of the Roman surface correspond to the three off-diagonal Peirce blocks
(the Z22 grading); the six pinch points (at (± 21 , 0, 0) and cyclic permutations) are the rank-drop points
of the diagonal Jordan adjoint, while the determinant-zero locus maps to the Roman double-line
skeleton [A680]. The non-orientability of the Roman surface reflects the Möbius boundary condition
on (1, 3)-representation spinors established in Theorem A.7.4.
Remark 0.13 (Lepton/quark duality). Leptons acquire mass through direct CW diagonal coupling,
following the primary anchor hierarchy (φ > 1 > φ−1 : heavier generation, larger coupling). Quarks
acquire mass through the inverted cross-product hierarchy: lighter generation, larger Peirce block, larger
confinement scale. The duality is an exact algebraic consequence of the Roman surface parametrisation
and is not imposed by hand. [A632]
Remark 0.14 (Nomenclature). All prior references to a informal cube model for generation mixing
are superseded by the Roman surface model, which is the correct mathematical object for the Peirce
closure structure. The Roman surface is also known as the Steiner surface; see Bryant–Kusner for
the modern parametrisation [1].
13
A.7.8 Artin’s theorem and the QFT loop bridge
A key structural question for the programme is whether the non-associativity of Os propagates into
quantum loop corrections, requiring the construction of a non-associative perturbation theory. Artin’s
theorem resolves this:
Novelty label: standard theorem used.
Theorem 0.15 (Artin). Any subalgebra generated by two elements in an alternative algebra is
associative.
The split octonions Os form an alternative algebra (satisfying x2 y = x(xy) and yx2 = (yx)x). In
quantum field theory:
(i) 2-point functions (propagators) involve two field insertions. Two insertions do not by themselves
make the internal algebra labels two-generated; Artin applies where that containment holds,
which is the same generator census as item (ii) (Rev32.9, A1686 F05).
(ii) 3-point functions (Yukawa and gauge vertex corrections) involve the Higgs and a fermion
bilinear. Where a diagram’s internal algebra labels lie in a subalgebra generated by two elements,
associativity follows by Artin; for the generic cross-block Yukawa trilinear (three independent
split-octonion entries, Paper 6) that containment is not established — momentum conservation
constrains momenta, not the number of independent internal octonionic directions (Rev32.1 scope,
A1538 F5). [LIB2-088, LIB2-251] A diagram-by-diagram generator census is the named outstanding
object. [LIB2-088]
(iii) 4-point functions (box diagrams) are, pending that census, the first place where three fully inde-
pendent octonionic directions are guaranteed to appear (P12 , P13 , P23 ) together. Non-associativity
is here in principle, but suppressed by an additional loop factor relative to the Yukawa and gauge
corrections.
Novelty label: physical interpretation not established here.
Corollary 0.16. Standard MS renormalisation-group running of Yukawa and gauge couplings from
ΛG2 to mZ is protected by Artin’s theorem for diagrams whose internal labels are two-generated
(any propagator or vertex correction meeting the containment of items (i)–(ii) above); the blanket
“fully protected” claim is scoped Rev32.1 pending the generator census (A1538 F5). [LIB2-251] The G2
automorphism group provides a consistent associative envelope for the leading-order diagrams that pass
the generator census (scoped Rev32.9; the census is outstanding). Non-associative QFT is not required
to derive the physical c and b quark masses from the skeleton provided by the Jordan geometry. [A633]
A.8 Open bridges: KK consistency and Grassmann parity
Two older programme items remain explicit.
(1) PR-C2 (KK consistency). The KK-70 counting and the E7(7) /SU (8) target are secure, but
the explicit consistent-truncation proof remains open.
(2) PR-C3 (Grassmann parity / spin-statistics). The local skeleton displayed here does not yet
solve the deeper spin-statistics and Grassmann-parity problem from first principles. This remains
an open bridge rather than a hidden omission.
14
A.9 Electron closure and lepton-family validation
The historical instanton baseline is
π
√
me = MPl × exp − × ϕ4 × 2 = 0.5036 MeV, (53)
8α
noting that 2π/(16α) = π/(8α) exactly (same formula, cleaner form). In Rev29 this expression is
retained as a semiclassical cross-check; the operative closure is the QED-corrected Koide chain.
Sessions 36–38 replaced the Sprint 3 empirical µ/τ formula by an exact algebraic expression. Rev29 :
that expression is re-tiered Structural / Loaded-correspondence / Reproduced, and the
earlier “fully proved / decisive validation closed” language is withdrawn; the provenance chain and
the fit gap are printed once, in Appendix X (“Provenance of the charged-lepton ratio”). The exact
tree-level value is √
mµ ϕ 8/3 2
= √ × = 0.059684 PDG: 0.059461 , (54)
mτ 5 10
√
where p = 8/3 = dim(Os )/rank(J3 ) is an exact dimension ratio [A616] and C = 2/10 is an exact
dimension
√ count in the J2 (Os ) sub-algebra geometry: dim(J2 (Os )) = 10 and off-diagonal trace norm
= 2 [A618]. [LIB2-049] Rev29 correction. Neither exact fact establishes, by itself, that these are the
mass exponent and the normalization: the provenance√is target-first (D616:99–102 asks whether the
measured 0.141 is algebraic in φ; A616:50–63 offers 2/10 while saying the algebraic origin is not
yet proved; dim J2 = 10 arrives afterward at A618:47–51; the divide-by-dimension step is “argued by
analogy” at A618:134–140; and A618:147 lists the formal derivation as blocking, closed nowhere in
the corpus). The claim that this is a lepton mass ratio “obtained with zero free parameters from the
Jordan algebra alone” is therefore withdrawn. The corresponding muon mass is mµ = 106.05 MeV
(+0.37% PDG), superseding the Sprint 3 empirical value of 104.87 MeV which optimised a float;
because C enters multiplicatively,
√ that +0.37% is the fit gap — the ratio requires C = 0.140894
and the offered constant is 2/10 = 0.141421. The calibration ΛG2 ≈ 260 MeV — one object with
one type, an external dimensional calibration (Appendix X input-count box), its value taken from
hadronic phenomenology and not a measured ΛQCD — is not algebraically derivable; here it serves
as the dimensional-transmutation anchor for the heavy-lepton sector, and Paper 2 uses the same
calibration for the light-quark constituents [A618] (Rev32.9, R84-h). [P-029] The electron mass is
Koide-consistent ⋆ ⋆ ⋆⋆ under the external empirical rule K = 2/3 (A636, A637): me = 0.5076 MeV
(−0.66% PDG). [P-030] The charged-lepton chain is structurally pinned with mixed provenance — mτ
anchor, mµ from a Structural / Loaded-correspondence ratio (re-tiered Rev29 ), me Koide-
consistent under the external rule — and the three masses are not jointly derived; the authoritative
ledger is Appendix O (“The authoritative charged-lepton ledger”).
Novelty label: physical interpretation not established here.
Proposition 0.17 (Geometric instanton interpretation of the electron mass). In the (4, 4) Clifford
vacuum of J3 (Os ), with fermions assigned to the (1, 3) Peirce block (Theorem 0.8), the electron mass
formula (53) admits the following semi-classical instanton interpretation at PLAUSIBLE ⋆ ⋆ ⋆⋆.
Exponent. The factor −π/(8α) is the on-shell Euclidean instanton action of a Möbius half-vortex in
the U (1) gauge background Tem = [e0 , e7 ] [A620]. The chain is: S2D = π/e2 = 1/(4α) (Bogomol’nyi-
saturated 2D vortex); lifting to 4D with half-twist area factor π/2 averaged over the 4 positive-signature
directions of the (4, 4) metric (mandated by Proposition 0.9):
1 π/2 π
Sinst = · = . (55)
4α 4 8α
15
The Jordan kinetic term LJ contributes at subleading order only (saddle-point cancellation at MPl
scale).
Prefactor. The factor ϕ4 = (r⋆ )2 is the Jacobian √ of the scale collective coordinate at the proved
minimum r⋆ = ϕ2 of VCW ferm [D333]. The factor 2 is the quadratic-form normalisation of the real
q
2 2
√
idempotent anchors (Tr(fi fj ) = δij , so Tr(f1 + f2 ) = 2), coinciding with the proved normalisation
of the two-generation subalgebra J2 (Os ) [A618].
Status in Rev29. This instanton formula is no longer the operative mass closure. The electron mass
is fixed by the QED-corrected Koide chain (under the external empirical rule K = 2/3),
me = 0.5076 MeV, (56)
using mµ,phys = 105.72 MeV after the one-loop QED matching correction. The instanton baseline is
retained only as a semiclassical cross-check on the scale and topology of the electron sector. [A636,
A637 — tier annotation: A637 records this chain as “proved at every step”; that internal rating is
not the tier carried here. The chain misses by −0.66%, and this appendix deliberately retains it only
as a semiclassical cross-check. Where the assessment and the paper disagree on tier, the paper’s
weaker reading governs.]
A.10 Parameter bookkeeping and suite integration
Rev29 uses four explicit bookkeeping classes.
(1) Algebraic quantities: Peirce eigenvalues, resolvent ratios, tree-level Yukawa eigenvalues, the
mixing formulae (internal constructions; physical tiers per Paper 3).
(2) Measured anchors: v = 246 GeV, mτ = 1776.93 MeV where used as anchor, and standard
benchmark quantities used only for comparison.
(3) Phenomenological ans"atze: none remaining in the charged-lepton mass sector. The electron
mass me = 0.5076 MeV is Koide-consistent ⋆ ⋆ ⋆⋆ via the QED-corrected Koide chain under the
external empirical rule K = 2/3 [A636, A637] (the external√rule keeps this at mixed provenance,
not class-1 algebraic). The muon-to-tau ratio constant C = 2/10 and exponent p = 8/3 are exact
algebraic quantities whose physical attachment to the mass formula is Loaded-correspondence
(re-tiered Rev29 ; the S38 “proved” typing is withdrawn — Appendix X, “Provenance of the
charged-lepton ratio”).
(4) Deferred items: exact weak-angle selector, full KK consistency proof, formal heavy-quark
S-matrix bridge, and the remaining bosonic extraction problems.
Papers 0–7 use Appendix A in a strict way. No statement about the local Yukawa or mass-operator
structure is allowed to float freely in the prose without either tracing back to the derivation recorded
here or being labelled as measured, phenomenological, or deferred.
References
[1] R. Bryant and R. Kusner, “The Roman surface and its symmetries,” unpublished notes; see also
J. Steiner, “Über solche algebraischen Curven,. . . ,” J. reine angew. Math. 21 (1840) 33–66.
16
[2] E. Witten, “Constraints on Supersymmetry Breaking,” Nucl. Phys. B 202 (1982) 253.
17
Leibniz Quantum Beats Newton
Appendix B — Internal Consistency Checks (established Rev25; suite Rev33.1)
Tom O’Sieg
August 2026
Abstract
Appendix B records the internal consistency checks attached to the Rev29 cubic-invariant
Yukawa derivation. It is included to make the verification trail explicit without interrupting the
main papers. The appendix does not add new physics claims beyond the Rev29 PMNS closure
update. It states which algebraic steps were cross-checked, what has newly closed in the leptonic
phase sector, and what remains outside the verified core. These checks are self-audits performed
within the programme; independent external replication awaits public release of the computational
kernels.
B.1 Verification questions
The Rev29 cubic-invariant Yukawa derivation was checked against five questions:
(1) Does the Freudenthal cross product of two (1, 3)-block fields project onto the middle diagonal
idempotent?
(2) Does contraction with Jvac = diag(ϕ, 1, ϕ−1 ) extract only the middle eigenvalue?
(3) Is the resulting tree-level Yukawa eigenvalue exactly yt = 1?
(4) Is the off-diagonal Higgs geometry preserved?
(5) Does the replacement of the linear route by the cubic route remove the need for an extra
Higgs-placement axiom?
B.2 Outcome
The outcome of the internal Rev29 verification pass is:
1
Check Result
Cross-product algebra passed
Vacuum contraction passed
Tree-level yt = 1 extraction passed
Off-diagonal Higgs geometry passed
No-extra-axiom closure passed
B.3 Remaining issue assessment
The verified core is the tree-level algebra. The following items remain outside the scope of the
verification note:
(a) the running of yt (µ) beyond the tree-level boundary value,
(b) the derivation of the electroweak vacuum expectation value,
(c) the deeper selector for the golden-ratio vacuum.
B.4 Verdict
The verification summary supports the publication-level statement used in Rev29: the cubic-invariant
Yukawa derivation is algebraically ready, while the measured electroweak scale remains an open
bridge. The vacuum selector is not an open bridge: it is fixed conditionally, by the vacuum-selector
theorems of Appendix E (the Paper 0 ledger row; Rev33.0 correction of an under-claim).
B.5 Leptonic CP phase — V3 rotation proof (Rev29)
Status change. Rev10 carried the leptonic phase as − arccos(−1/3) − 2π/7 ≈ −160.900◦ . Rev29
upgrades the result to a theorem:
2π
δCP = − √ ≈ −160.997◦ , JCP ≈ −0.011. (1)
5
No fitted coefficients enter the PMNS sector (three of four entries theorem-tier pre-Rev29, Derived-
conditional after the re-tiering (Rev29 re-tiering, Paper 3 § “Row-by-row re-tiering”; kernel s959);
θ13 structural under the named readout rule, §B.5′ ). No row may be cited as zero-parameter.
The V3 rotation mechanism. Define the three-dimensional subspace
#
V3 = span{Jvac , Jvac , Thalf · Jvac }. (2)
Kernel verification shows that V3 is invariant under the adjoint action of Thalf . Inside this subspace,
Thalf acts as a genuine three-dimensional rotation with angular frequency
√
5
Ω= . (3)
2
2
The canonical half-period is therefore
2π
t∗ = √ , (4)
5
and the defining orbit relation is
exp(t∗ Thalf ) · Jvac = Jvac
#
. (5)
The leptonic Dirac phase is the negative of this rotation time:
2π
δCP = −t∗ = − √ . (6)
5
Structural support. Three earlier theorem blocks feed directly into the result:
(a) DET-7 theorem. The Jordan pair (Jvac , Jvac # ) has Gram determinant 7, fixing the mismatch
geometry and the normalization of the (1, 3) sector. [LIB2-048]
(b) C1 theorem. The weight c = 12 in Thalf = Tscalar + 12 Toct is exact, so the generator is fixed with
no free coefficient. [LIB2-051]
(c) Uniqueness theorem. Thalf is, up to scale, the unique G2(2) -invariant generator in the (1, 3)
Peirce sector of f4(4) ; the scale is the normalisation Thalf = Tscalar + 12 Toct of the item above
(Rev32.9). [LIB2-052]
Independent cross-check. The older − arccos(−1/3) − 2π/7 = −160.900◦ formula remains useful
as a structural cross-check, but it is no longer the authoritative statement of the phase law.
PMNS closure. With
√ √
PMNS 3 2 7 2 4π 3−2 2
tan θ12 = , sin θ23 = , sin θ13 = sin = ,
ϕ2 16 8 8
and
2π
δCP = − √ ,
5
three of the four PMNS parameters were marked theorems at ⋆100 pre-Rev29; after the re-tiering the
two physical rows θ12 and δCP stand at Derived-conditional and the θ23 octant attachment is
Loaded-correspondence (A1566) (Rev29 re-tiering, Paper 3 § “Row-by-row re-tiering”; kernel
s959); θ13 is STRUCTURAL ⋆ ⋆ ⋆ (round-trip framing T 2 , exponent-2; forcing pending [s502/A1275]).
Lemma B.5′ (θ13 readout normalization; Structural, readout-lemma). The reactor angle
follows from a single readout-normalization step rather than from the deprecated “triality maps
π/4 → π/8, then square” chain. Reading the lepton mixing entry off the unit endpoint ray of the
Route-B Jordan frame, and passing through the spin double cover with the half-angle π/8 — a
readout normalization choice that the T 2 = −1 moment structure of Appendix H motivates but does
not make canonical (Rev29 ) — gives
√
2π 2 4π 3−2 2
|Ue3 | = sin =⇒ sin θ13 = sin = ≈ 0.02145,
8 8 8
in agreement with the NuFIT 6.1 global-fit value 0.02248+0.00055
−0.00059 (≈ 1.8σ). Status. This is Struc-
tural plus a named readout lemma: it is a theorem given the unit-endpoint-ray readout rule, but
that readout rule is not itself derived from the algebra (the round-trip T 2 squaring is identified
(exponent-2, s502/A1275) but its forcing is pending), so θ13 is not promoted to Proved. [LIB2-281,
LIB2-282, LIB2-315] It is the same readout-axiom posture as F-δ for the CKM phase.
3
(Rev29 ) The T -bridge does not canonicalize π/8. Earlier wording made the half-angle
π/8 canonical by appeal to T 2 = −1. That inference is withdrawn. [LIB2-252] T 2 = −1 is
a statement about a symplectic element of E7(7) on the certified 56 (Appendix H, live
T -bridge): it fixes the spinorial character of the substrate half-turn, and it is consistent
with reading a mixing entry at a half-angle, but it does not select which angle is halved,
does not fix the ray the entry is read off, and does not supply the squaring step. [LIB2-252,
LIB2-315] The half-angle π/8 is therefore motivated by the moment structure and chosen by
the readout rule — which is exactly what the Status note above already concedes; the lemma
statement is here brought into√line with its own status note. [LIB2-252] Nothing numerical moves:
sin2 θ13 = sin4 (π/8) = (3 − 2 2)/8 ≈ 0.02145 and its ≈ 1.8σ standing against NuFIT 6.1 are
as printed, and θ13 stays Structural ⋆ ⋆ ⋆ with forcing pending (s502/A1275). Recorded
explicitly because this retraction has been made in-house three times (killed S188, resurrected
S189, re-killed S261) and reached the papers in none of the three; it is printed here so the
canonicality claim cannot be re-mined downstream (p0, p3, p5, Appendix C, Appendix X
inherited it).
B.6 CKM sector — algebraic construction and the conditional δCKM
theorem (Sessions 22–23)
Status (S112 claim-hygiene; re-tiered Rev29). The three CKM magnitudes (|Vus |, |Vcb |, |Vub |)
were marked Proved-tier within the Route-B map pre-Rev29 and agree with PDG within 1σ. After
the Rev29 row-by-row re-tiering (Rev29 re-tiering, Paper 3 § “Row-by-row re-tiering”; kernel s959)
the CKM first row is Loaded — Paper 3 selected Route B over Route A by comparing both to kaon
data — and |Vcb | is a Structural formula with Loaded-correspondence attachment (A1566). The
pre-Rev29 “six of eight” tally in the provenance ledger (Appendix X) is superseded by two of eight
PMNS , δ
(θ12 CP ; the physical θ23 row is Loaded-correspondence since A1566/D1244, Rev32.6, its
internal 7/16 identity theorem-grade). The Dirac phase δCKM is not of that grade: it is a candidate
construction under a named readout axiom (stated immediately below), i.e. a Coincidence-class
numeral with an algebraic identity behind it and no selection of G7 (Rev32.5 retype; the Rev29
“Structural claim with a derivation path” wording is retired). [LIB2-253] The earlier “100/100, all four
GREEN, zero free parameters” framing is retired here: it swept the conditional phase into the same
bucket as the proved magnitudes, and was inconsistent with this construction’s own status note —
the chain (C)–(D) was shown (S87–S88, D772/D773) to be an algebraic identity of the construction,
not an independent selection theorem.
The G7 = ϕ + 5ϕ−5 candidate construction (A578+A579; E92–E98) — formerly titled
“conditional theorem”; Coincidence-class since Rev29.
Named readout axiom (F-δ, Structural). δCKM is read out as an additive bounded-
resolvent channel-expectation over the Route-B rank-5 complement. Under this axiom — and
only under it — the construction below is a theorem. The axiom is itself Structural (it
is the readout rule, not derived from the algebra), so δCKM is a conditional theorem, not a
Proved-tier result. Determinant / log-determinant readout foils fail in code (D791/D793),
which is why the channel-expectation form is named explicitly rather than asserted canonical.
Under F-δ, the CKM Dirac phase follows from the sub-leading eigenvalue correction of the Albert-
4
algebra vacuum:
q
G7 = ϕ + 5ϕ−5 , δCKM = arctan 3(ϕ + 5ϕ−5 ) = 68.1297◦ . (7)
(Comparison: 0.8σ from the γ-fit; 1.6σ from the PDG direct value.)
Four-step algebraic proof (A578+A579; complete):
(C) Peirce-graph loop order (E96, part C). The minimum closed Peirce-graph path returning to
P13 requires exactly n = 5 applications of Thalf . The Fibonacci number at n = 5 is F5 = 5, giving
correction order ϕ−5 with coefficient 5.
(B) Split-octonion coupling-weight theorem (E96, A579; Part B). The split-octonion (4, 4)
signature of Os determines sector-dependent Peirce coupling weights in the 27-dimensional
representation of J3 (Os ). Under the W13 block decomposition: the 3 compact imaginary directions
e1 , e2 , e3 (norm −1) contribute zero net coupling (isotropic cancellation); the scalar e0 and 4 non-
compact directions e4 –e7 (norm +1) contribute fully. Therefore wP13 ,k = 58 wtotal,k exactly. When
√
inserted into the resolvent sums from m12 (x) ∈ Q( 5)[x], the weighted sums equalize: δMP13 =
δMP22 = 5ϕ−4 = ε. Rev14 status correction: the equal-shift property of this step was shown
(S87–S88, D772/D773 adjudication, artifact-backed) to be an algebraic identity of the construction
rather than an independent derivation — house-proven circular as a selection argument. The
chain (C)–(D) is therefore classified numerology-class pending derivation: numerically exact and
unchanged, but not a theorem of selection. An attempted kill of the integer 5 (the α(0)/mτ error
budget, D773) does not land (4.9635 ± 0.087 ∋ 5 at 1σ), so the construction is candidate-class —
neither derived nor dead.
(A) Equal-shift formula (E93, A578; Part A). Equal shift ε = 5ϕ−4 on both Peirce masses ϕ2 and
1 gives G7 = ϕ + ε/ϕ = ϕ + 5ϕ−5 exactly.
(D) Jvac deformation uniqueness (E97, A579; Part D). The perturbation (λ − (ϕ2 + ε))(λ − (1 +
ε))(λ − (ϕ−2 + ε)) = 0 is the unique O(ϕ−4 ) deformation of Jvac preserving the Albert-algebra
characteristic-polynomial form. Trace shift = 3ε = 15ϕ−4 ; no alternative corrections at this order.
Table reading rule. Rev32.1 reading rule: tiers name mathematical provenance; physical attach-
ments are stated separately; no row may be cited as zero-parameter; displayed σ values are named
comparator distances, not profile likelihoods.
5
CKM sector status (Rev32.2 comparator; provenance per the Rev29 re-tier;
A1566 retype, Rev32.6): first row Loaded, |Vcb | Loaded-correspondence attach-
ment (physical) / Structural formula, δCKM Coincidence-class. [LIB2-066, LIB2-285]
Parameter Algebraic value PDG 2026 comparator Comparator
distance
√
PMNS / 6
|Vus | sin θ12 = 0.225256 0.22431 ± 0.00085 +1.11σ
(s1162)
√
|Vcb | 1/(9 7) = 0.041996 0.0407 ± 0.0013 +1.00σ
√
|Vub | |Vus ||Vcb |/ 6 = 0.003862 0.00389 ± 0.00016 −0.18σ
p
δCKM arctan 3(ϕ + 5ϕ ) −5 = global-fit δ = 1.154 ± 0.025 rad +1.40σ /
68.1297◦ (= 66.12◦ ± 1.43◦ ; radians +0.64σ
primary, R-29); direct γ =
66.4+2.7
−2.8
◦
No fitted CKM coefficient appears inside the selected Route-B map, but this is not a
parameter-free theorem of the physical matrix: the first √ row is loaded and the phase
motivation is incomplete. The “7 structure”: |Vcb | = 1/(9 7) and sin2 θ23 = 7/16 share the
algebraic integer n23 = 7 from det G = 7 (Theorem B.5); the physical octant attachment of
7/16 remains Loaded-correspondence
√ (A1566). Note: The authoritative Rev29 CP
formula is δCP = −2π/ 5 = −160.997◦ (V3 rotation theorem); the older PSL(2, 7) formula
involving 2π/7 is a deprecated cross-check and is not part of the current “7 structure” claim.
B.7 Spectral equivalence lemma (C2 — representation independence)
Lemma B.7 (Albert–Freudenthal spectral equivalence; kernel E115). Let LJvac : X 7→ Jvac ◦ X denote
Jordan multiplication by the vacuum element Jvac = diag(ϕ, 1, ϕ−1 ) on J3 (Os ). The eigenspectrum of
LJvac is the same whether computed abstractly from the Peirce decomposition or concretely from the
explicit 27 × 27 real-matrix lift.
Proof sketch. The Jordan product rule assigns eigenvalue (λi + λj )/2 to the (i, j) off-diagonal Peirce
block, where λi are the diagonal entries of Jvac . For Jvac = diag(ϕ, 1, ϕ−1 ) this gives:
Sector Eigenvalue (exact) Multiplicity
(1, 1), (2, 2), (3, 3) diagonal ϕ, 1, ϕ−1 1 each
(1, 2) off-diagonal (ϕ + 1)/2 = ϕ2 /2 8
√
(1, 3) off-diagonal (ϕ + ϕ−1 )/2 = 5/2 ≡ Ω 8
(2, 3) off-diagonal (1 + ϕ−1 )/2 = ϕ/2 8
Total: 3 + 24 = 27 eigenvalues. Kernel check E115 constructs the explicit 27 × 27 matrix of LJvac
and confirms numerical agreement to better than 10−12 . The result is a standard consequence of
the Albert–Freudenthal theorem for the split form J3 (Os ); see also McCrimmon [1] for the abstract
Peirce rule.
6
√
Physical consequence. The (1, 3) Peirce eigenvalue Ω = 5/2 matches the angular frequency used
# ,T
by the V3 rotation theorem in V3 = span{Jvac , Jvac half · Jvac } (Theorem B.1, §B.5). The leptonic
CP phase is therefore consistent with
π π 2π
δCP = −t∗ = − = −√ = − √ = −160.997◦ (0.36σ from NuFIT 6.1 212+26 ◦
−36 , central-value only).
Ω 5/2 5
(8)
The spectral structure of LJvac matches the expected eigenvalues of the derivation-generated CP
phase; a formal proof of equivalence (eigenvalue ↔ angular velocity) is the subject of a forthcoming
appendix. Thus the spectral calculation is supporting evidence, not by itself a complete proof that
the eigenvalue is the physical angular velocity.
B.8 In-house re-verification of the Koide chain (S210 machinery;
Rev25 fold)
The independently-built verification machinery of S210 (kernels s532, s535) re-derived the banked
charged-lepton chain from scratch, deterministically, with no reference to the original build:
Quantity Route Re-verified agreement Kernel
mτ from (me , mµ ) Koide closure 0.006% s535
K = 2/3 circulant null condition 6 × 10−6 s535
ae (anomalous moment machinery) QED chain rebuild 3 ppb s532
These rows are verification, not new results: they document that the banked lepton-sector chain
reproduces under independent machinery at the quoted precision. No observable moves.
References
[1] K. McCrimmon, A Taste of Jordan Algebras, Springer, 2004.
7
Leibniz Quantum Beats Newton
Appendix C — δCP Precision Tests and Falsification Window
Established Rev25; suite Rev33.1
Tom O’Sieg
August 2026
Abstract
Appendix C records the precision-test language for the leptonic CP phase. The V3 rotation is
an exact algebraic construction, while the physical phase attachment is Derived-conditional
under its named readout√ premise (the pre-Rev29 THEOREM/⋆100
√ marking is historical). The
target is δCP = −2π/ 5 = −160.997◦ with Ω = 5/2 (A603/E110). Two physical PMNS
rows, θ12 and δCP , stand at Derived-conditional after the Rev29 re-tiering; the physical θ23
octant is Loaded-correspondence (its internal 7/16 identity theorem-grade, A1566) and θ13 is
Structural; no row may be cited as zero-parameter.
C.1 Framework target
The Rev29 phase target is
2π
δCP = − √ = −160.997◦ (JCP ≈ −0.011, rephasing-invariant). (1)
5
Status (current reading): the V3 rotation theorem supplies the exact algebraic orbit relation, and
the physical identification of its half-period with δCP is Derived-conditional under the printed
readout premise. The pre-Rev29 THEOREM/⋆100 label is retained only as a historical stamp. NuFIT
6.1 (NH, w/SK) gives a central-value comparator distance of 0.36σ; the full profile is non-Gaussian.
Normal ordering is assumed.
C.1a History of the leptonic CP target
The leptonic CP target changed because the algebraic derivation was refined, not because the value
was tuned to oscillation data.
1
Era / value Formula Reason for supersession
Early Rev8 value δCP ≈ −133.36◦ Phase-only/non-associator ansatz; re-
tained historically but no longer the
operative derivation.
S11 intermediate − arccos(−1/3) − 2π/7 ≈ PSL(2, 7) / discrete-flavour refinement;
−160.900◦ useful cross-check, but the 2π/7 term
is deprecated.
√ √
Current target (en- δCP = −2π/ 5 = −160.997◦ V3 rotation theorem
√ with Ω = 5/2
tered Rev29; current and t∗ = 2π/ 5; authoritative suite
at Rev29) value.
C.2 Adopted precision scenario
The publication bundle uses a design-level DUNE-era precision scenario in which the Dirac phase is
measured to a characteristic uncertainty of order ±15◦ . This is used as a falsification scenario, not as
a substitute for future experimental publications.
C.3 Comparison table
Quantity Framework target JCP /notes DUNE precision Current reading
window
√
δCP −2π/ 5 = −160.997◦ JCP ≈ −0.011 O(10◦ ) at NuFIT 6.1 NH:
[Derived- 1σ, exposure- 0.36σ; DUNE
conditional; dependent Phase II will test
A603/E110; read-
out premise printed]
θ23 41.41◦ (sin2 θ23 = 7/16; mismatch geome- few-degree preci- NuFIT 6.1 NH
physical oc- try sion global best
tant Loaded- fit lower-octant
correspondence, (0.470), 2.3σ away;
internal identity upper-octant com-
theorem-grade — parator ≈ 0.561
A1566) at ≈ 9.5σ
C.4 Interpretation
The logic is intentionally sharp. If the future measured phase sits far outside the
√ adopted band
◦
around the framework target −160.997 , the V3 rotation mechanism (Thalf |V3 , Ω = 5/2) is falsified.
The suite does not reserve an unbounded freedom to repackage the prediction after the fact. The
prior ansatz value −133.36◦ and the intermediate PSL(2, 7) value −160.900◦ are both superseded;
the V3 formula is authoritative.
2
C.5 V3 rotation proof of δCP (exact algebraic layer; physical attach-
ment Derived-conditional)
The proof used by the Rev29 suite is algebraic rather than phenomenological. Define
#
V3 = span{Jvac , Jvac , Thalf · Jvac }. (2)
Kernel verification shows that V3 is invariant under the action of Thalf and that the restriction Thalf |V3
is a genuine three-dimensional rotation. The rotation frequency is
√
5
Ω= , (3)
2
which coincides with the (1, 3)-sector eigenvalue of LJvac in the Albert-Freudenthal spectrum. The
canonical half-period is therefore
2π
t∗ = √ . (4)
5
At this time one has the exact orbit relation
exp(t∗ Thalf ) · Jvac = Jvac
#
. (5)
The leptonic Dirac phase is identified with the negative of this rotation time,
2π
δCP = −t∗ = − √ = −160.997◦ . (6)
5
This proof contains no fitted parameter. The only algebraic inputs are the working vacuum Jvac =
diag(ϕ, 1, ϕ−1 ), the exact C1 coefficient c = 12 in Thalf = Tscalar + 12 Toct , and the kernel-verified V3
invariance. The older − arccos(−1/3) − 2π/7 = −160.900◦ expression remains as a useful internal
cross-check, but the suite takes the V3 rotation law as authoritative.
Derived structural
√ CP seed (s502/A1208). The commutator invariant Im Tr([Φu , Φd ]3 ) =
3
ε ∆ ·(−243 15/256) is recorded as a derived structural quantity, not a prediction: Parseval/reciprocity
fix a single Z2 sign ε, while the magnitude remains a frozen section (A1208). It is carried for provenance
only and moves no observable.
3
Leibniz Quantum Beats Newton
Appendix D — RG and Threshold Reference Notes (Rev7-era reference; suite
Rev33.1)
Legacy Reference — Rev7 era. Superseded by Rev29 for all threshold calculations.
Tom O’Sieg
August 2026
Abstract
Legacy reference (Rev7 era). This appendix is superseded by Rev29 for all threshold
calculations. It is retained as a historical record only.
Appendix D is the rev7 reference note for the weak-angle renormalization-group bridge. It is
intentionally a sketch. The full threshold derivation is reserved for a later revision. The present
appendix records only the algebraic input, the standard running formulas, and the threshold logic
adopted by the bundle.
D.1 Tree-level input
The exceptional-Jordan sector supplies a tree-level weak-angle ratio. Rev7 does not identify that tree-
level quantity with the measured electroweak weak angle without additional running and threshold
effects.
D.2 Standard running
The one-loop Standard-Model running is
dgi bi 3 41 19
µ = g , (b1 , b2 , b3 ) = , − , −7 , (1)
dµ 16π 2 i 10 6
— the Standard-Model one-loop coefficients, imported physics (tier Input; Rev32.9) — or equivalently
d bi
α−1 = − . (2)
d ln µ i 2π
D.3 Threshold sector
The rev7 threshold sketch introduces heavy electroweakly charged Peirce-sector modes at a scale of
order
Mth ∼ 5 TeV. (3)
1
These modes contribute threshold shifts to the electroweak couplings of the order required to reconcile
the tree-level algebraic ratio with the measured weak-angle benchmark.
D.4 Status
The threshold bridge is structural but incomplete. The sign of the correction, its multi-TeV scale,
and its origin in heavy Peirce-sector doublets are part of the rev7 picture. The full explicit threshold
integral is deferred to rev7.
2
Leibniz Quantum Beats Newton
Appendix E — The Vacuum-Selector Theorem (Rev33.1)
Tom O’Sieg
August 2026
Abstract
This appendix discharges the one item carried by the Rev29 bundle as “theorem carried; proof
to be reproduced” [A641/Q5A]: that the working vacuum Jvac = diag(ϕ, 1, ϕ−1 ) is not chosen
but uniquely forced. We give two independent, elementary, fully explicit proofs. Route A uses
only the two structural conditions the suite already imposes — unit Jordan determinant together
with the DET-7 invariant and invariance of the vacuum spectrum under Jordan inversion (the
“palindromic” condition) — and selects Jvac with no reference to any number field. Route B
re-derives the same vacuum arithmetically as the unique golden-field unit triple compatible with
DET-7. Both routes remove all continuous moduli; the residual √ input is named explicitly in
each case.√The spectral corollary ϕ2 + ϕ−2 = 3 is exactly the 3 that enters the θ12 theorem
tan θ12 = 3/ϕ2 .
1 Setup and hypotheses
Let Jvac = diag(a, b, c) be a real diagonal element of J3 (Os ) with a, b, c > 0. Two structural
conditions are carried throughout the suite and are taken as the hypotheses of the selector:
(H0) Unit determinant: Det(Jvac ) = abc = 1. (The vacuum is a norm-one idempotent-frame
element; equivalently the cubic Jordan norm N (Jvac ) = 1.)
(H1) Inversion symmetry (palindromic spectrum): the characteristic polynomial of Jvac is
# . Under (H0) the adjoint equals
palindromic, i.e. Jvac is isospectral to its Jordan adjoint Jvac
# −1 −1
the inverse, Jvac = Det(Jvac ) Jvac = Jvac [A680], so (H1) states that the vacuum spectrum is
invariant under λ 7→ λ−1 — the Weyl/inversion symmetry of the Peirce ϕ-ladder.
# ) = 7, the n
(H2) DET-7 invariant: det Gram(Jvac , Jvac 23 = 7 “seven-unification” invariant used
throughout the mixing and heavy-quark sectors.
Writing e1 = a + b + c, e2 = ab + bc + ca, e3 = abc, the characteristic polynomial is t3 − e1 t2 + e2 t − e3 .
2 The DET-7 invariant in elementary form
Novelty label: new specialization proved here.
1
Lemma 1. For Jvac = diag(a, b, c) with abc = 1,
#
) = a2 + b2 + c2 a−2 + b−2 + c−2 − 9.
det Gram(Jvac , Jvac
Hence (H2) is equivalent to
a2 + b2 + c2 a−2 + b−2 + c−2 = 16.
(1)
Proof. With abc = 1 one has Jvac # = J −1 = diag(a−1 , b−1 , c−1 ). The Gram matrix of the pair
vac
# 2 ) = a2 + b2 + c2 and
(Jvac , Jvac ) in the trace form ⟨X, Y ⟩ = Tr(XY ) has diagonal blocks Tr(Jvac
−2 ) = a−2 + b−2 + c−2 , with off-diagonal Tr(J −1
Tr(Jvac vac Jvac ) = Tr(I) = 3. Its determinant is
2 2 2 −2 −2 −2
(a + b + c )(a + b + c ) − 9. Setting this equal to 7 gives (1).
3 Route A: symmetry + DET-7 (field-free)
Novelty label: new specialization proved here.
Lemma 2. Under (H0), condition (H1) holds iff e1 = e2 , and then the spectrum is {ρ, 1, ρ−1 } for
some ρ > 0; i.e. one eigenvalue equals 1 and the other two are mutually reciprocal.
Proof. With e3 = 1 the characteristic polynomial is t3 − e1 t2 + e2 t − 1. Reversing its coefficients gives
−(t3 −e2 t2 +e1 t−1), so the polynomial is palindromic iff e1 = e2 . In that case p(t) = t3 −e1 t2 +e1 t−1
and p(1) = 1 − e1 + e1 − 1 = 0, so t = 1 is a root; say b = 1. Then ac = 1 by (H0), so c = a−1 =: ρ−1
with a = ρ.
Novelty label: new specialization proved here.
Theorem 1 (Vacuum-selector, symmetry form). A real positive diagonal Jvac ∈ J3 (Os ) satisfying
(H0)–(H2) is unique up to permutation:
√
Jvac = diag(ϕ, 1, ϕ−1 ), ϕ = 1+2 5 .
Proof. By Lemma 2 the spectrum is {ρ, 1, ρ−1 }. Substituting into (1),
2
ρ2 + 1 + ρ−2 ρ−2 + 1 + ρ2 = ρ2 + 1 + ρ−2
= 16,
so ρ2 +√1 + ρ−2 = 4 (the positive root), i.e. ρ2 + ρ−2 = 3. Writing u = ρ2 gives u2 − 3u + 1 = 0, so
u = 3±2 5 . The two roots are reciprocal (u+ u− = 1) and correspond to interchanging ρ ↔ ρ−1 — a
√
permutation of the spectrum. Taking ρ > 1, ρ2 = 3+2 5 = ϕ2 , hence ρ = ϕ. Thus the spectrum is
{ϕ, 1, ϕ−1 }, unique up to permutation. [LIB2-019]
Remark 1. Route A uses no number-theoretic input. The only structural hypotheses are (H1)
inversion symmetry and (H2) DET-7; (H0) is the normalization. Both (H1) and (H2) are already
imposed elsewhere in the suite — (H1) is the ladder inversion/Weyl symmetry, (H2) is the seven-
unification invariant — so the theorem shows these two discrete conditions leave zero continuous
moduli.
2
Remark 2 (Status of (H2) — honest label, S64). DET-7 enters this appendix as a structural
postulate, not a derived necessity; the result is therefore a conditional selector theorem. [LIB2-018, LIB2-
336, LIB2-337, P-013] Its motivation is threefold. (i) det Gram is discrete — there is no continuous dial to
fit — once the spectrum is restricted to the integral golden-unit lattice of Route B; on an unrestricted
spectrum it varies continuously (det Gram(diag(t, 1, t−1 )) = (t2 + 1 + t−2 )2 − 9; Rev32.9, A1632).
2
(ii) The single√ integer n23 = 7 simultaneously fixes three independent observables — sin θ23 = 7/16,
|Vcb | = 1/(9 7), and the 7/3 lepton–quark bridge — plus the mismatch normalization: one discrete
choice, multiple independent payoffs, which is compression rather than curve-fitting. (iii) Route B
selects the same vacuum by golden-field unit arithmetic in which 7 appears as the output of the
DET-7 evaluation on the selected triple, not as an input to the field choice. A first-principles
derivation of DET-7 itself remains open and is carried as such.
Remark 3 (Origin of ϕ; S66). The golden ratio enters solely through the selector theorem above
(unit determinant + inversion/palindromic symmetry + DET-7), which is field-free and therefore
signature-independent. Two richer candidate origins were examined and excluded (D732, three-way).
(i) A golden-quasicrystal Z[ϕ] Fourier module via the O → E8 → icosian cut-and-project chain is
obstructed by the split signature: it requires a positive-definite metric, whereas the split forms give
an indefinite lattice with null cones and no quasicrystal. (ii) An icosahedral (A5 ) Galois origin from
the resolvent quintics fails because the within-sector quintics are binomial (µ5 − Rd , dihedral D5 ),
not icosahedral. Thus ϕ is principled at the elementary selector level; the grander constructions are
not its source.
4 Route B: golden-field units + DET-7 (arithmetic)
Novelty label: new specialization proved here.
Theorem 2 (Vacuum-selector,
√ arithmetic form). Let Jvac = diag(a, b, c) with a, b, c > 0 algebraic
integers in Q( 5) and abc = 1. [P-024] Then (H2) forces Jvac = diag(ϕ, 1, ϕ−1 ) up to permuta-
tion. [LIB2-020]
√
Proof. The ring of integers of Q( 5) is Z[ϕ], whose unit group is {±ϕn : n ∈ Z} with ϕ the
fundamental unit. From abc = 1 each of a, b, c is invertible in Z[ϕ], hence a unit; positivity removes
the sign, so a = ϕn1 , b = ϕn2 , c = ϕn3 . Then abc = 1 gives n1 + n2 + n3 = 0. By Lemma 1, (H2)
reads P P
2ni −2ni = 16.
iϕ iϕ
Order n1 ≥ n2 ≥ n3 and write the gaps p = n1 − n2 ≥ 0, q = n2 − n3 ≥ 0. The left side depends
only on the gaps:
F (p, q) = 3 + Ap + Aq + Ap+q , Ak := ϕ2k + ϕ−2k = L2k ∈ {2, 3, 7, 18, 47, . . . },
with Ak the (strictly increasing, even-index) Lucas numbers, and the zero-sum condition admits a
gap pair iff p + 2q ≡ 0 (mod 3). (An earlier version of this proof asserted strict monotonicity of F
in maxi |ni |; that is false — F (5, 0, −5) = 15376 > 11561 = F (6, −3, −3) — and the step is replaced,
Rev32, by the exact case count below; the conclusion is unchanged [A1519 F1; exhaustive |n| ≤ 9
scan house-run].) The admissible cases are: (p, q) = (0, 0) gives F = 3 + 2 + 2 + 2 = 9; (p, q) = (1, 1)
— the triple {1, 0, −1} — gives F = 3 + 3 + 3 + 7 = 16; no other pair with p + q ≤ 2 is admissible
(p + 2q ≡ 0 (mod 3) excludes (1, 0), (0, 1), (2, 0), (0, 2)); for p + q = 3 the admissible pairs are (3, 0)
3
and (0, 3), each giving F = 3 + 18 + 2 + 18 = 41 (attained at {2, −1, −1}); and for p + q ≥ 4,
Ap+q ≥ 47 gives F ≥ 3 + 47 + 2 + 2 = 54 (the next value is 64 at {2, 0, −2}). (Rev32.9, A1632: the
termwise bound printed through Rev32.8 used Ap + Aq ≥ 20 for every p + q ≥ 3, which does not hold
— the admissible (2, 2) has Ap + Aq = 14.) Hence F = 16 occurs only at {n1 , n2 , n3 } = {1, 0, −1},
i.e. Jvac = diag(ϕ, 1, ϕ−1 ).
Remark 4. Routes A and B are independent and corroborating. Route A’s residual input is the
inversion symmetry (H1); Route B’s residual input is the golden-field hypothesis a, b, c ∈ Z[ϕ]. Each,
together with (H0) and DET-7, removes all continuous freedom. [LIB2-020] That ϕ is precisely the
fundamental unit of the golden field is the arithmetic reason the same vacuum appears from both
directions.
5 Spectral corollary and the solar-angle theorem
Novelty label: physical interpretation not established here.
Corollary 1. The selected vacuum satisfies the exact spectral identities
√
ϕ2 + ϕ−2 = 3, e1 = e2 = ϕ + 1 + ϕ−1 = 1 + 5.
Consequently the (2, 3) Peirce spectral ratio yields the parameter-free solar angle
√
3
tan θ12 = 2 = 0.66158, θ12 = 33.49◦ ,
ϕ
compared with NuFIT 6.1 (NH) 33.76+0.42 ◦ ◦
−0.41 at 0.66σ (Rev32.1 refresh: the 33.44 /0.06σ previously
printed here used a pre-6.0-era value — Paper 5 correction) [A537].
√
Proof. From ρ2 + ρ−2 = 3 at ρ = ϕ (Theorem 1). The 3 in the solar ratio is exactly ϕ2 + ϕ−2 ;
p
the denominator ϕ2 is the squared dominant eigenvalue of Jvac . The numerical value follows.
6 Seven declared-class negatives on selector laws (Rev30–Rev31)
The freeze-first arc spent several rounds asking what a selection law cannot be. Six results survive
as durable negatives (A1483, A1488, A1492, A1493, A1495); a seventh, proved post-closure, is
item (vii) below (A1512). Each is scoped to a class that was declared before testing, and in each case
the scope is the result: stated without its class, every one of them becomes a false universal claim.
(i) The Gram-dark fibre is exactly two-dimensional (s993–s994). House-verified: dim p = 70
and rank (X 7→ X iB ) = 68, so the fibre invisible to GB (M ) = iT B M iB has dimension exactly 2.
Consequently no action factoring only through GB can uniquely select a coset point — not
obstructed at some order, but dead at every order. [LIB2-036] The qualifier “factoring only through
GB ” is load-bearing: the all-order no-go is proved for the GB -factoring subclass, and the wider class
of parent actions and boundary laws is explicitly left open. [LIB2-036] Two associated facts are weaker
than they look and are recorded as such: the recovery of all 16 horizontal coordinates at second
order is a rank statement, not a construction, and the independent quadratic-trace no-go offered
alongside was not verified in house.
4
(ii) The frame-free diagonal-invariant class contains no law with a nonzero dark response (s1002–
s1003; Rev32.9: constants belong to the class, so it is not empty). For the class Cff of C 2 scalars on
(E7(7) /SU (8)) × OB that are invariant under the diagonal E7(7) action, with no spurion and a point
source space — declared before testing — there is no law with a nonzero dark response: for Z in
the dark fibre, etZ fixes the loaded plane exactly while moving the coset point, so every diagonal
invariant is constant along the dark orbit. [LIB2-037] A basepointed full-carrier law does exist and
does fire on that fibre, with response exactly t2 , and the rank-two dark projector is minimal; but its
basepoint is a convention rather than a consequence, and it carries no Hamiltonian, no action and
no observable. [LIB2-037] It is certified mathematics occupying an uninhabited physical slot.
(iii) The transported class consumes a source but cannot select one (s1011–s1012). The transported
functional co-moves the loaded plane to zero residual — the co-moving residual is below 10−13
while the fixed-frame shear remains exactly 34 |j| as the tangent displacement of a co-moving plane.
But the action is exactly isotropic on sources: equal-norm sources share the minimum value, the
transport price and the local spectrum. The transported quadratic/current class consumes
j and cannot select its ray. [LIB2-038] Two cautions travel with this. Its constants enter under
a declared conditioning–minimax principle — a named mathematical principle, not a measured
number — and must never be read as fitted parameters; and the object is a transported selector
functional, not a parent action and not a Lagrangian, a distinction the originating round itself
was careful to preserve. [LIB2-038] (That distinction was later discharged rather than dissolved: the
transported selector was extended to a full conditional parent action and closed at the selector tier in
S274–S275 — §7.) [LIB2-038]
(iv) The fixed-grade cubic/current class is closed negative for rank-two detection (s1005–s1006). [LIB2-
039] Restricting the dark fibre to the fixed +2 grade, rank (z2 ∋ Z 7→ P27 ZP27 ) = 1: one of the
two dark directions moves the carrier out of the fixed grade entirely, with P27 ZP27 = 0 while
[D56 , Z] ̸= 0, and the witness is exhibited explicitly. [LIB2-039] A class needing rank two therefore
cannot be detected inside a fixed grade, and the fixed-grade cubic/current route closes negative.
The scope is exactly that: rank-two detection inside a fixed grade. The embedding-tensor class
is scoped out rather than decided, because no convention-certified branching artifact for it was
supplied. [LIB2-039]
(v) An even action cannot resolve the sign (s1005–s1006). For X in the horizontal 16-dimensional
sector, RB e2tX RB = e−2tX (house-verified numerically), so every RB -conjugation-invariant scalar
satisfies S(e2tX ) = S(e−2tX ). [LIB2-040] An RB -even action cannot resolve the sign — the parity
of the law, not its strength, is what fails. [LIB2-040] Any candidate must therefore be odd under RB
somewhere, which is a usable design constraint rather than a mere obstruction.
The representation supplies tensors; it does not select one. The branching 27+2 ↓ Spin(5, 5) ×
SO(1, 1)Y = 16+1 ⊕ 10−2 ⊕ 1+4 is computed, and it delivers the carriers the construction needs.
But an inequivalent irreducible branch passes the same tests: uniqueness fails, and one discrete
choice is priced. The name “Higgs” attached to the 10−2 carrier is an imported label, not a derived
object — no electroweak doublet is derived anywhere in this construction, and the label must not be
read as if one were. The honest one-line summary of this whole family of results is that the tensor
exists and the action does not.
(vi) A frame-free F4 -invariant cannot reach the ordered frame (s978/s981). The stabilizer of the
ordered triple inside the automorphism group leaves a 24-dimensional orbit, hence, at a stationary
point on that orbit, at least 24 Hessian zero modes (flat directions) for any frame-free invariant
(Rev32.9: a statement about stationary orbits). [LIB2-041] The consequence is sharper than the
5
obstruction it replaces: a frame-free F4 -invariant can at best select the spectral orbit [Jvac ]F4 ; it can
never isolate the ordered frame diag(φ, 1, φ−1 ). [LIB2-041] Selecting the ordered frame requires a
genuine loading or symmetry-breaking tensor — which is to say the ordering is an input to this
class, not an output of it. A companion norm-only statement runs the same way: for any C 2
action S = F (α, N (X)), ambient stationarity in X at a regular norm-one point forces FN = 0 and
hence Hess S ≡ 0 on ker dN , giving at least 26 flats structurally. [LIB2-042] (Rev32.9, A1637: under
stationarity constrained to N = 1, FN need not vanish; the same count then holds for the constrained
Hessian Hess S − FN Hess N = FN N dN ⊗ dN , which vanishes on ker dN .) Attribution and scope:
this is the lane’s theorem, not the house’s. The house verified it at the level of the dimension counts;
the claim that a frame-free invariant is constant on the orbit is accepted as standard rather than
proved here, and the norm-only computation was checked on a generic symmetric cubic in R27
rather than on J3 (Os ) itself. [LIB2-042]
(vii) The unshifted zero-level quadratic-carrier Kempf–Ness class cannot carry the selector (post-
closure; Rev31, S276; A1512 Ask-2). For the class of Kempf–Ness potentials built on an unshifted
zero-level quadratic carrier — declared before the run (class sha 6ece8dc9...) — two independent
no-gos close it: rank Hess0 ∥µK ∥2 ≤ dim K ≤ 63 < 69 = rank QT (the rank no-go), and QT = sin2
along the beat orbit while no beat-invariant KN potential isolates [W0 ] (the beat-transitivity no-
go). [LIB2-043, LIB2-044] The escapes are named and priced rather than left vague: a central shift is
available on proper subgroups only; a beat-breaking spurion, dropping the clock circle, or leaving
the KN family altogether each exits the declared class. Scope is exactly the declared class; this
negative postdates the closure of §7 and prunes its surroundings, not its content. [LIB2-044]
Type signature of anything that could still work. What the six negatives leave is
a shape rather than an object: a law on (E7(7) /SU (8)) × Gr16 (V56 ) × Dphys possessing a
plane-preserving stationary point, a positive horizontal Hessian, a detector for the exact
two-dimensional dark fibre, and a full-carrier or downstream tensor that does not factor
through iT
B M iB . Two candidate tensors would satisfy the type. They have not been supplied.
Nothing in this section promotes an observable, moves a tier, or touches the engine.
(Supersession note (Rev31; S274–S275). The type signature above has since been inhabited, adju-
dicated under precommitted seals, and closed at the selector tier — §7. The six negatives stand
unchanged as the walls the closure was built against; their scopes are unaffected.) [LIB2-036]
7 The parent-action closure: complete-conditional at the selector
tier (Rev31; S274–S278, A1506–A1515)
The string that ends here ran four adversarial rounds across three sealed dispatches (D1196–D1198,
plus a cold two-referee confirmation D1199), was adjudicated against precommitted house tables
(seals s1043, s1045 spent; s1041 retired scope-limited), and was verified two-host with byte-identical
reruns at every step. Its outcome is stated exactly, because its exactness is the result.
What was assembled (A1506, conditional tier)
The first full conditional parent action in the programme’s history: a minimal-parent type theorem
(any homogeneous SU (8)/H mapping equivariantly to both targets forces H ⊆ Kdyn , hence dim ≥ 53
6
with equality iff H = Kdyn ); the forced + 34 source-backreaction coefficient; an all-parameter Hessian
positivity theorem; the source-ray selection mechanism — spec(QL ) = {01 , 210 , 45 } computed from
the banked artifact with kernel dimension exactly 1, the kernel ray equal to the certified source
unit at overlap 1.0, so stationarity selects one projective source ray at the witness — and a local
Hamiltonian/Legendre attachment, all sector-assembled into Sparent with an itemized open list and
nothing promoted. The Hamiltonian/Legendre material enters here, and only here: it is a local
attachment at the conditional tier, not a continuum Hamiltonian and not an LSZ-grade dynamics.
Corrected typing (Rev32; the D1212/A1525 catch, kernel s1072): the Hessian that enters the
Legendre step is the velocity Hessian I32 /U — a different object from the A1506 configuration
Hessian, which is a stability record only; conflating the two mis-types the attachment. [LIB2-021] The
full typing ledger is Trackers/PARENT_ACTION_HAMILTONIAN_LEDGER_S284.md.
The boundary datum: one ten-dimensional object, three exact faces (A1508)
The datum the action is conditional on is single: [j] ⇐⇒ L∂ ⇐⇒ E ∗ = span{iB j, ΩiB j} —
three faces of one ten-continuous boundary object, priced once. [LIB2-022, P-010] E ∗ is the unique
(53, 10, 0; Kdyn ) pair, certified non-blind after a post-target discovery (the 16-member Ω-pair family)
forced a scope hold and a PI-ruled non-blind rerun; the certification is consistent-and-redundant
with [j] ⇐⇒ L∂ . The physical selection of this datum remains open/loaded (register clause:
Appendix X, §“The Rev31 register clause”). [LIB2-022]
The obstruction theorem (A1507, A1509 Thm. A, sharpened A1510)
At theorem strength, (FB16 )Ksel = 0: the priced ten-continuous datum has no invariant inhabitant, and
the pure-spinor point is exactly that datum (A1507’s honest loss, with the owed Fierz/moment-map
gates exact: mA (j) = 0, x# = 0, N27 = 0, Ann(j) = L∂ ). The grammar acts on the 32-dimensional
braid-fixed carrier im(iB ) ⊕ Ω · im(iB ) as scalars/zero; Ug = Pact + iB g iT T
B + (ΩiB ) g (ΩiB ) is
symplectic and commutes with every grammar generator (house-checked, commutators ∼ 10 , W0 −13
included). O(16) acts transitively on source rays — and SO(16) alone is already transitive on S 15
(the A1510 sharpening, closing the connectedness loophole) — so no construction internal to
the braid/trace grammar can select [j], L∂ , or E ∗ . [LIB2-045] The boundary price is irreducible
within the grammar. Scope: genuinely external categories are not excluded. [LIB2-045]
The dark attachment (A1509 Thm. B): the grammar owns the ray it was never
given
The S1 fibre law is exact in closed form,
h √ i
S1 (eY ) = 54 cos2 θ cosh r/ 6 − 1 ,
with W1 zero at every order (the S274 residual −3.6 × 10−4 fully demystified). Provenance upgrade
(Rev31; S276, A1512). Every constant of this law is now proved-with-forced-constants rather than
house-numeric: the active carrier is a Cl(2, 1) module in exactly 12 copies; Wh = Q + 32 P + 32 γ1 − γ2 ;
√
the constants 54, 6, 24, 9/2 are each forced by the module structure; and κlaw = κclock is a
theorem — one structure constant appearing simultaneously as an eigenvalue difference and as the
commutator coefficient in [K0 , W0 ] = √16 W1 , exact at the full 56 × 56 level (house 8 × 10−16 ; kernels
7
s1047–s1048, lane proof A1512, house-verified). [LIB2-014, LIB2-024] Corroboration link (Rev32): this
golden bridge independently re-derives a face of the S205 dynamics seed’s golden Cartan flow (L-
DynamicsSeed, A1268–A1270, Motivated tier; the exp(2 ln φ H) ladder) — printed as corroboration
in both directions, neither result derives the other (Appendix X, clause-27 reconciliation). The
naive trap — the ground valley of +εS1 is the W1 ray — computed and fenced. The crown is
prefrozen: the grammar’s own detector quadratic form on the 70-dimensional p-sector has spectrum
01 ⊕ (5/3)10 ⊕ 226 ⊕ (5/2)32 ⊕ (9/2)1 (house-recomputed exact) with a unique top eigenspace, frozen
pre-target and shown post-freeze to be exactly RW0 (overlap 1.000000000000). [LIB2-023] The dark
axis is therefore grammar-derived, not dark-axis-labelled, and the selector
T = ∥log M ∥2 − 29 QS1
has forced coefficients — A1511 proved they follow from the top eigenvalue alone, leaving a parameter-
free selector. On the fibre T = r2 sin2 θ; the horizontal Hessian is 89 I; the total is 32 + 89 κd I > 0;
Kdyn preserves the dark axes, so no new stabilizer datum is owed. The two faces of the selection
wall are now asymmetric, and priced as such: the grammar cannot see the source ray (obstruction
above) but intrinsically owns the dark ray. [LIB2-023]
The dark half twice over, and the forced interior split (Rev31; S276/S278, A1512–
A1515)
Two post-closure rounds sharpened the dark attachment without moving its tier. First (A1513):
the golden Lorentzian form η = ΩG — grammar-derived, zero new data — independently produces
the selector kernel ray [W0 ] and the exact dark-fibre law r2 sin2 θ: the detector quadratic form
of η has spectrum 01 ⊕ (13/36)32 ⊕ (13/27)36 ⊕ 11 with kernel overlap 1.0 against [W0 ]. The
full spectrum is closed negative in the declared two-parameter class Σ = aη + bPB : the interior
36-block resists with a gap of exactly 2/27. The honest premise for that gap (corrected in-round,
A1514: the house’s “scalar-blind on the entire 36-block” was too strong): the banked QPB record
028 ⊕ (1/2)32 ⊕ (2/3)10 already separated the 10-block; what no construction had produced was
the exact 15 : 17 interior ratio. So the dark half of the selector is grammar-derived twice over —
spectrum (A1509) and potential (A1513) — while the interior split carried the remaining loaded
information. Then (A1514, Tier-B unconditional, zero price): that last loaded split is forced too.
PB forces the exact 26 ⊕ 10 eigenspace split (QP |V36 = 026 ⊕ (2/3)10 ), its symplectic transport defect
∆T = [Tϕ , PB ] = [J0 , PB ] forces the complement, and the commuting joint pair (QP , Q∆ ) with joint
spectrum (0, 0)2 ( 12 , 1)32 (0, 38 )26 ( 23 , 0)10 — four joint eigenvalues, and polynomials in the pair are
scalar on the two-dimensional common kernel — produces all five QT eigenspaces as subspaces once
that kernel is split by the top projector of the Hessian of the banked golden form η, which s1051
adds at gate G8 (H70 = Pηtop + 24 −14 ; “word degree ≤ 2”
54 P32 + H36 ; projector distances ≤ 5.1 × 10
counts words in QP and Q∆ ; no new datum beyond the banked η — Rev32.9, A1632). [LIB2-025] The
eigenvalue weights on those subspaces are the subject of Appendix S’s production arc (A1515–A1516),
where they are derived in-class and their action shape is priced. [LIB2-025]
8
The closure declaration (A1509 Thm. C; referees A1510/A1511)
Complete-conditional, selector tier. Within the frozen split-real 56-frame and the
declared finite/internal loaded-profile class: the selector parent action Vtotal has its global
minimum at (j ∗ , q ∗ = 32 j ∗ , M ∗ = I) with combined Hessian positive for all ρ, u0 , κd > 0
on the directions transverse to [W0 ], the exact projective soft ray at zero inserted dark
amplitude, along which it vanishes (scope stated Rev32.9, A1632). [LIB2-021] The incremental
price is 10 + 1 continuous — the boundary object (three faces, priced once) and one
positive dark coupling κd — with zero discrete data consumed (Z2 lifts left explicitly as
degeneracies) and no stabilizer owed (T ≥ 0 globally). [LIB2-022] No further selector datum
is hidden. Two cold referees audited the closure volumes independently (A1510 Gemini
6/6; A1511 Grok 6/6) with zero disagreements. Fences: this closes the finite/internal
loaded-profile selector only — continuum, pole, and LSZ remain a separately typed frontier;
ρ, κd , U enter by positivity only, no numeric values claimed; nothing here promotes an
observable, moves a tier, or touches the engine.
[LIB2-021]
8 Rev32 addendum: rulings, prices, and the selection frontier
(S284–S288)
The S284 rulings, printed beside the closure (S284.1)
Three PI rulings attach to the closure section above. (i) The boundary-condition-family question is
ruled: CLASS-ρ is primary; CLASS-W applies under C4 = 0; CLASS-H is deferred. [LIB2-323] (ii)
The s1073 amendment A1–A3 is adopted (verifier of record and battery filed, byte-certified). (iii)
The H2 minimal-class rank is predeclared: rank 2. The H2 leg’s standing price is stated exactly:
six continuous C0 coefficients plus one-map-for-all-η naturality — and target-triviality of
the action clause is forced, not assumed, by the D1214 forced-triviality theorem (house-verified to
the digit, A1527). [LIB2-030, LIB2-031]
Coefficient prices have an adjudicated precedent class (S285.1)
The clause-27 price above is not the first priced cross-layer coefficient in this suite: the AX-MG
coefficient-one clause was proved a genuine independent axiom by exact countermodel (A1435,
Appendix S) and stands as the adjudicated model for how such a price is typed. The C0 six-coefficient
price is the same wall type, now at the same evidential grade (below).
The C0 independence theorem (S285.3; D1217→A1530)
At the frozen H2 tier the six-coefficient freedom is countermodel-grade: two disjoint symbolic
rank-2 candidates pass all 13 gates and 30 controls identically, and no banked selection polynomial
separates them (lane ×2 and house ×2 byte-identical; independent house verifier s1077, 14/14,
×2). [LIB2-032] This supersedes the earlier census-negative wording in grade only — the scope clause
is mandatory: a higher-tier Sjoint , should one be banked, may still derive the coefficients. [LIB2-032]
9
First finite OS subgate pass, at its tier (S284.3)
The continuum frontier moved by one typed step: the first finite Gaussian OS subgate pass
and a toy resolvent, both two-host byte-certified (A1529), are recorded at conditional/finite tier
with their full non-claim fences: no continuum limit, no reflection-positive continuum measure, no
pole, no LSZ; the 17-stage continuum staging map (Trackers/CONTINUUM_STAGING_MAP_S284.md)
locates the step. [LIB2-165, LIB2-166, LIB2-300]
The selection frontier: F-SEL(c) closes negative, scoped (S287.9–.11; D1220→A1533)
The Rev31 arc left one unexplored selection direction: could a refined condition family derive, rather
than adopt, the four Φdiag relations {v = 0, w = 0, u = t, 27z = t}? The Rev32 answer is a scoped
negative, in three parts.
(a) The house’s universal cone claim is withdrawn. An earlier banked argument held that no
positivity- or stability-type refinement could answer the question because inequality conditions
cannot reduce dimension. [LIB2-254] That claim is refuted (x ≥ 0 with −x ≥ 0 has feasible set
{0}: degenerate inequality systems can encode equalities). The replacement statement is scoped
and stronger where it applies: the presently banked non-degenerate PSD/stability conditions
u ≥ 0, 4t 16t 16t
9 + v ≥ 0, 27 + w − z ≥ 0, 27 + w + z ≥ 0, t2 ≥ 0 have non-empty interior —
(t, t2 , u, v, w, z) = (1, 1, 1, 1, 1, 0) satisfies all five strictly — so the feasible set is 6-dimensional before
the scale quotient and 5-dimensional after it, and cannot select a ray. [LIB2-035] Degenerate systems
can lower dimension in principle, so any proposed selector must derive its everywhere-active equality
locus independently and clear the circularity bar. [LIB2-254] Fchar is interior only among the four
minimal-word inequalities: in the five-inequality cone it lies on the face t2 = 0 (slacks 1, 49 , 59 , 17 27 , 0),
which Appendix S requires for Φdiag . That face contains the target without selecting it; any other
forced degeneracy face would exclude it (Rev32.9, A1637). [LIB2-035]
(b) Casimir-naturality is tested and refuting. The lead candidate — A = f (CKdyn ), coefficients
a function of the Kdyn Casimir — clears the circularity bar (it is stated entirely from banked
Kdyn data) and fails on the merits: the Casimir spectrum on the coefficient carrier p70 is 0(6) ⊕
(1/6)(20) ⊕ (1/4)(30) ⊕ (5/12)(14) , with P10 pure (1/6)(10) and P26 not Casimir-isotypic at all (0(2) ⊕
(1/4)(10) ⊕ (5/12)(14) ); equal Casimirs recur across blocks, so A = f (C) forces u = v = w = z = 0
(literal carrier; cut 5 → 1) or u = v = w, z = 0 (support-69 reading, an extra priced premise; cut
5 → 2). [LIB2-034] Φdiag needs 27z = t = ̸ 0: Casimir-naturality excludes Φdiag ; it is a refuter,
not a selector (independently re-derived house-side in sympy with a falsifier control — make P26
isotypic and z survives — kernel s1083, 11/11). [LIB2-034] Price to cite it: A = f (C) is not forced by
ordinary equivariance (the symmetric commutant has dimension 38; functions of the Casimir form a
4-dimensional centre) — it is a further multiplicity-blind naturality principle. [LIB2-034]
(c) The closure, scoped. Across the filed condition inventory — categorical/invariance conditions,
word minimality (a separate, already-adopted class axis), ordinary equivariance (cuts nothing),
Casimir centrality (excludes), strict positivity/stability (full-dimensional), degenerate faces (other
than t2 = 0, which contains the target without selecting it, exclude it), equal-weight and condition-
number minimisation (select a different ray), Ξ-blindness (excludes), trace normalisation (scale
only), exact QT matching (circular, two-dimensional fibre) — no banked refined condition
family selects Φdiag non-circularly. F-SEL(c) closes negative at the current banked
class and condition inventory. [LIB2-028, LIB2-033] Φdiag remains an adopted structural section at
the printed price: word minimality as a domain/class clause plus the four coefficient relations;
10
without the adoption the enlarged class carries five projective continuous moduli. Claim boundary,
binding: this is not a universal impossibility theorem and must never be printed as one; the reopen
condition is a genuinely new, independently derived invariant or condition family entering the
banked corpus. [LIB2-033] The open-selection set accordingly drops to G7 motivation and SEL-27,
and SEL-27 is a theorem awaiting a PI ruling rather than a research gap — the live selection frontier
is one item. [LIB2-324]
9 Status
Theorems 1 and 2 discharge the carried item [A641/Q5A]: the working vacuum spectrum {ϕ, 1, ϕ−1 }
is forced up to permutation, conditional on the named hypotheses (unit norm; inversion symmetry,
or the golden-field restriction; DET-7), with the residual input named in each route. [LIB2-018, LIB2-046]
The ordered physical Peirce frame remains loaded: a frame-free F4 -invariant selects only the spectral
orbit (item (vi), s978/s981), so ordering requires a loading or symmetry-breaking tensor. [LIB2-001,
LIB2-019, LIB2-020] The two structural conditions DET-7 and (H1) are not free parameters but invariants
used elsewhere in the suite, so the selector relocates no hidden freedom into the vacuum. What
remains genuinely external is upstream of this appendix: the choice of the cubic Jordan algebra
J3 (Os ) and its trace form, within which DET-7 and the inversion symmetry are then natural.
This appendix is self-contained and elementary; both routes are verifiable by direct computation (the
admissible gap-pair case split and Lucas lower bound in Theorem 2 — Rev32.1: the earlier “spread
monotonicity” phrasing here cited the step retracted inside that theorem (A1538 F6); kernel s1085
verifies the replacement and proves the stronger variational form — and the quadratic u2 −3u+1 = 0
in Theorem 1). [LIB2-001]
A third, independent corroboration — variational rather than algebraic — is recorded as a speculative
probe in Appendix I: a grading-coupled cone potential whose unique positive-definite minimum
at unit coupling is Jvac , with ϕ emerging from c2 − c − 1 = 0. It shares DET-7 / unit-coupling as
structural inputs and does not upgrade the tier of this selector; it is reported only because it makes
the golden vacuum overdetermined (three independent routes pin the same point), which is the
substance of the answer to the “numerology” charge. (Tracker-tier note (Rev25 fold; s654+, S221):
in the dynamics-lane machinery the house Cartan satisfies u · u = 24 δ with Fano-line support, and
the CDIZ locus xi = 1 is forced rather than chosen there — a fourth, independent appearance of the
same vacuum point, recorded as support only (the dynamics tier itself remains held; Appendix Q
posture).)
11
Leibniz Quantum Beats Newton
Appendix F — The Gauge Boundary: Obstruction, Uniqueness of
Completion, and Dynamical Shadow Selection (Rev33.1)
Tom O’Sieg
August 2026
Abstract
This appendix settles the status of the gauge sector with three results, obtained in Sessions
74–80 on a certified E7(7) 56/133 embedding-tensor substrate. (i) The two-theorem gauge
boundary: the branching wall (no gauging of maximal N =8, D=4 supergravity, in any symplectic
frame, has a gauge algebra containing a compactly embedded su(3) ⊕ su(2) — a fortiori none
realizes gSM ) and the charge wall (no gauging’s algebra contains a unitary-class su(3), the
matter-colour embedding 8 ↓ su(3) = 3 ⊕ 5 · 1; only the vector and adjoint su(3) classes are ever
gauging-realizable). The right algebra can appear, but provably never in the right representation.
Gauge-group emergence as a gauging is therefore closed negative, by theorem rather than by
failure-to-find. (ii) Uniqueness of completion: among compact reductive low-energy completions
that gauge the full derived colour + weak action on exactly the sixteen derived Weyl fermions
with the derived hypercharges and 10H -only scalar closure, GSM is unique up to U (1)B−L (whose
extra Cartan is computed unique, = 3(B−L)) and up to global structure Γ ∈ {1, Z2 , Z3 , Z6 }.
(iii) Dynamical shadow selection: on the scalar potential of the gauged theory — built here at
the published normalization, calibration constant exactly 1, and certified against the published
SO(8) and SO∗ (8) spectra — vacuum stability of the missing-16 scalar sector, combined with the
shadow-hosting condition, selects SO(6, 2) uniquely within the scanned electric family; its dyonic
completion is a published Minkowski vacuum with residual SO(6)×SO(2) ⊃ SU (3)×U (1)×U (1).
The net position is stated without inflation: the gauge group remains an input at the group
level, now bounded on both sides — it could not have arisen as a gauging (obstruction), and —
within the enumerated compact-reductive completion class and the fixed 16 + 10H package — it
could not have been anything else (uniqueness up to U (1)B−L and global form) — while the
maximal derivable remnant, the shadow su(3) ⊕ u(1) ⊕ u(1), is selected dynamically rather than
by hand.
Notation (σ). In this appendix σ denotes the SO∗ (8) time-mirror σ = Ω, under which
σDKK σ −1 = −DKK ; the terms σ-fixed and σ-closure below refer to it. It is not Appendix I’s
Jordan frame involution σframe (ρ 7→ 1/ρ), not Appendix H’s throat involution σC , and not the
split-octonion conjugation σoct (e7 7→ −e7 ) of the A689 dictionary, which does not appear in the
typeset suite at all.
1 Scope and provenance
All computations referenced here are reproducible from the project’s e7_substrate module (Sessions
S71–S77): the certified ambient E7(7) substrate (V1–V6: 133 generators on the 56, signature +7,
1
symplectic, quartic-invariant stabilizer of dimension exactly 133), the embedding-tensor layer (V7/V8:
linear 912 constraint and quadratic closure, with the de Wit–Nicolai SO(8) regression at residuals
∼ 10−16 ), the Jordan anchor (V9: the missing-16 pinned as the identity-anchored complement
(b) = 42 ⊖ 26), and the potential layer (S77). [LIB2-006] External conventions are taken from de Wit–
Samtleben–Trigiante (arXiv:0705.2101) as transmitted by Dall’Agata–Inverso (arXiv:1112.3345),
whose equation (3.1) fixes the scalar potential
g2
V = XM N R XP Q S MM P MN Q MRS + 7 XM N Q XP Q N MM P , XM N P = ΘM α (tα )N P ,
672
(1)
with M = VV T the scalar matrix. The 1/672 normalization was citation-checked against the
published text this session; the suite’s generator normalization reproduces it with calibration
constant exactly 1.00000000 (no rescaling of any kind enters the numbers below).
2 The branching-wall theorem (the su(2)L wall, total form)
Novelty label: new specialization proved here.
Theorem 2.1 (Branching wall; S75). No gauging of maximal N =8, D=4 supergravity, in any
symplectic frame, has a gauge algebra containing a compactly embedded su(3) ⊕ su(2). A fortiori no
gauging realizes gSM , and no vacuum of any gauging has residual gauge symmetry containing gSM .
The maximal SM shadow available in this framework is su(3) ⊕ u(1) ⊕ u(1).
Proof structure (each step labelled computed or cited). L1 (cited-standard, with on-substrate posi-
tive control): the quadratic constraint implies the embedding tensor Θ is invariant under its own
gauge algebra im Θ [dWST]. Control: the maximal violation over the 28 so(8) generators of the de
Wit–Nicolai tensor is 1.5 × 10−16 on the substrate. Hence if ι = su(3) ⊕ su(2) ⊆ im Θ, then Θ is
ι-invariant.
L2 (computed): every compactly embedded su(3) ⊕ su(2) in E7(7) conjugates into the fixed maximal
compact su(8); the faithful embedding classes are exactly the 21 branchings of the 8 (census
computed twice, by character arithmetic and by explicit equivariant kernels on the substrate; the
two methods agree 21/21).
L3 (new; formal statement): regard the embedding tensor as the linear map Θ : 56 → 133,
eM 7→ ΘM α tα . The ι-invariance of L1 says precisely that Θ intertwines the ι-action on the 56 with
the adjoint ι-action on the 133: Θ(X · eM ) = [X, Θ(eM )] for all X ∈ ι. By the first isomorphism
theorem for ι-modules, rowspace(Θ) = im Θ ∼ = 56/ ker Θ as an ι-module — a quotient of the 56;
complete reducibility of the compact ι makes the quotient a direct summand, so every irreducible
constituent of im Θ occurs in 56 ↓ ι with at least its multiplicity. Since ι ⊆ im Θ carries its own
adjoint isotype (8, 1) ⊕ (1, 3), both adjoints must occur in 56 ↓ ι.
L4 (new — the wall, exact integer multiplicities): for every one of the 21 faithful classes, mult(8, 1) =
0 or mult(1, 3) = 0 in 56 ↓ ι = Λ2 8 ⊕ Λ2 8̄. [LIB2-076] Twenty classes have mult(8, 1) = 0; the unique
(8, 1)-bearing class, 8 = (1, 2) + (3, 1) + (3̄, 1), has mult(1, 3) = 0. [LIB2-076] Analytically: an (8, 1)
in Λ2 8 ⊕ Λ2 8̄ requires 8 ⊇ (3, m) ⊕ (3̄, m) with m = 1 forced by slot counting, leaving exactly two
slots — a single (1, 2) — whose Λ2 contains no (1, 3). Contradiction with L3 for every class; the
theorem follows.
2
Remark 2.2. The failure is upstream of the quadratic constraint, in the linear/equivariant structure:
no frame enumeration and no classification citation is load-bearing. “Compactly embedded” is the
physically correct hypothesis (gSM sits in the maximal compact of any candidate gauge group);
noncompact sl(3) ⊕ sl(2)-type real forms are not, and need not be, excluded. The one place a referee
should push — the self-duality bookkeeping in L3 — carries an on-substrate corroboration: the
measured image-deficit of the 11 ι-generators against rowspace(Θ) over each nonzero invariant space
equals 1.0000 exactly where both adjoints are absent (full isotypic orthogonality), 0.3805 for the
unique (8, 1) class (su(2)-blocked, floor 3/11 ≈ 0.27), and is never near zero. It additionally carries
an independent blind replication (A759): a second executor, unbriefed on this proof and using a
method-disjoint route (exact Gelfand–Tsetlin character arithmetic, no substrate), reproduced the
21-class census, all Inv(912) dimensions, and the full adjoint multiplicity table integer-for-integer —
including the fingerprint class 8 = (1, 2) + (3, 1) + (3̄, 1) with mult(8, 1) = 2, mult(1, 3) = 0 — and
independently arrived at the same equivariance argument (“no image-compatible candidate for the
quadratic constraint to act on”).
Consequence for the framework. The Rev29 claim ledger keeps the gauge group as
input; Theorem 2.1 upgrades the reason: the input label is not “unselected” but unreachable
— emergence-as-gauging is closed negative at theorem grade. The constructive question
moves, with no residue, to the potential layer (Section 5).
3 The charge wall (the second wall: never the right representation)
The branching wall bars the algebra su(3) ⊕ su(2). The second wall bars the representation: even
a lone su(3), which gaugings demonstrably can carry, can never be carried in the matter-colour
embedding. Among the eight inner embedding classes of a compact su(3) ⊂ su(8) (classified
by 8 ↓ su(3); census computed three ways, exact agreement), matter colour is the unitary class
8 = 3 ⊕ 5 · 1, with 56-content 20 · 1 ⊕ 6 · 3 ⊕ 6 · 3̄ and no octet. [LIB2-077]
Novelty label: new specialization proved here.
Theorem 3.1 (Charge wall; S80). No quadratic-constraint-satisfying gauging of maximal N =8,
D=4 supergravity contains a unitary-class su(3) in its gauge algebra im Θ. In particular, no vacuum
of any gauging has a residual gauge symmetry containing a unitary-class su(3): matter colour is
unreachable in any gauging, in any frame, at any vacuum.
Novelty label: new specialization proved here.
Corollary 3.2 (Realizable colour classes). The only su(3) embedding classes realizable inside the
gauge algebra of any QC gauging are the vector class (8 = 3 ⊕ 3̄ ⊕ 2 · 1) and the adjoint class
(8 = 8) — precisely the two classes whose octet occurs in the 56. The six remaining classes are
barred. This matches every residual observed in the scanned family and in the published vacua (the
Warner SU (3) × U (1)2 point: vector class; SO(8): adjoint class).
Proof structure (two independent legs, both computed). Suppose QC(Θ) = 0 and a unitary-class
su(3) ⊆ im Θ. By L1 of Theorem 2.1, Θ ∈ Invsu(3) (912).
Leg A (rowspace orthogonality; direct and self-contained): Invsu(3),unitary (912) was computed
explicitly: dimension 140, matching the character prediction 140 = 2·15 (36⊕36)+2·55 (420⊕420)
3
exactly (singular-value gap 8.9 × 1013 ). The union of the rowspaces of all invariant Θ has rank 125,
and the su(3)-containment leak equals 1.0000 on all eight generators. [LIB2-077] The 125 has a name:
125 = 35 · 1 ⊕ 15 · 3 ⊕ 15 · 3̄ is exactly the (1, 3, 3̄) isotype of 133 ↓ su(3)unitary — the eight missing
directions are precisely the su(3) adjoint itself (full isotypic orthogonality). [LIB2-077] Since im Θ lies
inside that union, su(3) ̸⊆ im Θ. Contradiction.
Leg B (the octet/module argument; the L3 form): by L3, im Θ is a quotient of the 56 as a module
for ι = su(3); ι ⊆ im Θ requires its adjoint octet to occur in 56 ↓ ι. The octet multiplicity in
56 ↓ su(3)unitary is zero (on-substrate spectrum: 20 singlet states ⊕ 36 triplet states, zero octet states;
positive control: the vector-class embedding shows exactly 16 octet states in the same pipeline).
Contradiction.
Remark 3.3 (Replication and provenance). Both legs carry an independent blind replication
(A761h): a second executor, unbriefed on the house numbers and deriving its own character
prediction, reproduced the result seven-for-seven — dim Inv = 140, rowspace rank 125, containment
leak 1.0 on all eight generators, octet multiplicity 0, twenty singlets, matched convention pin, control
residual 1.5 × 10−16 (house: 1.3 × 10−16 ) — and independently identified the 125 as the (1, 3, 3̄)
isotype. This is the same replication pedigree as the branching wall’s (A759). A structural companion
(self-conjugacy): every so(p, q)-type residual is an orthogonal algebra, whose 8 is self-conjugate, so
only the self-conjugate (vector/adjoint) classes fit residuals — the complex-type unitary class is
barred from all orthogonal residuals categorically; the theorem extends the bar from residuals to
every gauge algebra.
The two-theorem boundary, stated once. Theorem 2.1 and Theorem 3.1 together
close the gauge boundary in both algebra and representation: su(3) ⊕ su(2) never fits
(branching wall); a lone su(3) fits only in the vector or adjoint class, never the matter-colour
unitary class (charge wall). The right algebra can appear — provably never in the right
representation. Fences: this is a statement about the supergravity gauging layer, not about
the derived matter (the 16-C fermions are kernel-side); the vector-class shadow selection of
Section 5 stands untouched; A1 stays input, now with proven reasons on both sides.
4 Uniqueness of completion (G2 reframe, D756)
Novelty label: new specialization proved here.
Theorem 4.1 (Uniqueness of the low-energy completion; S76). Fix the derived one-generation
package: exactly the sixteen Weyl fermions of the 16 with the derived hypercharges YSM , and the
10H -only scalar sector. [LIB2-081] Among compact reductive low-energy completions that gauge the
full derived colour + weak action on this package:
(1) GSM survives every filter (all twelve-plus anomaly sums vanish exactly, in integer arithmetic;
Witten count even);
(2) the only extension surviving all structural filters is GSM ×U (1)B−L , and the extra compact Cartan
is unique: modulo the existing hypercharge direction the extra solution space of the Yukawa-
neutrality + anomaly system is one-dimensional, and in the Higgs-doublet-neutral representative
it is spanned by (1, −1, −1, −3, 3, 3) = 3(B−L) (the raw system is two-dimensional, X =
2hY + (3q − h)(B−L); Rev32.9, A1647) — an independent, representation-level re-derivation
4
of the B−L residue carried elsewhere in the suite; it is removed only by the named no-fifth-
force/scalar-closure empirical input;
(3) every GUT and intermediate foil — SU (5), SO(10), Pati–Salam, left-right, flipped SU (5) —
fails as a low-energy completion (extra massless vectors; breaking scalars absent from 10H );
trinification fails the exact-16 matter content itself; none fail by anomalies;
(4) the proper sub-completions U (1)Y , SU (3) × U (1)Y , SU (2) × U (1)Y are themselves anomaly-free
and are excluded only by the gauge-the-full-action clause — that clause is load-bearing and
travels with the theorem;
(5) the group acting faithfully on the sixteen states is S(U (3) × U (2)) = GSM /Z6 (verified state-by-
state); the global structure Γ ∈ {1, Z2 , Z3 , Z6 } is a line-operator-level choice not fixed by the
matter spectrum — uniqueness is stated up to Γ.
Enumeration boundary: the foil family and the adversary directions are exhausted; no closed
classification over all compact reductive groups on a sixteen-dimensional Weyl space is claimed. [LIB2-
081] No second completion was found.
Boundary sentence (mandatory). Theorem 4.1 is uniqueness-of-completion, not
emergence. [LIB2-081] The gauge group remains an input at the group level; Theorem 2.1 is
why that is so (unreachable as a gauging), and Theorem 4.1 is the strongest honest statement
on the other side (the input could not have been otherwise, in the stated class, up to U (1)B−L
and Γ). [LIB2-081] The corresponding ledger label is A1: [input + structural + empirical
uniqueness-of-completion].
4.1 Explicit state-by-state verification of the Z6 faithful action
The “verified state-by-state” clause of Theorem 4.1(5) is the following one-line integer congruence,
displayed here so the derived Z6 is reproducible rather than merely asserted. The generator of the
candidate trivially-acting centre subgroup is
g = ω ⊮3 , −⊮2 , e2πiY ∈ Z(SU (3)c ) × Z(SU (2)L ) × U (1)Y , ω = e2πi/3 , (2)
acting on a Weyl field of colour triality t (with 3 7→ 1, 3 7→ 2, singlet7→ 0), weak duality d ∈ {0, 1}
(doublet 1), and hypercharge Y in the Q = T3 + Y normalisation. Then g acts as the phase
exp 2πi [ Y + 3t + d2 ] , which is trivial on that field if and only if
6Y + 2t + 3d ≡ 0 (mod 6) . (3)
Evaluated on the entire one-generation 16:
field (t, d) Y 6Y + 2t + 3d mod 6
Q (3, 2) (1, 1) + 16 6 0
uc (3, 1) (2, 0) − 23 0 0
dc (3, 1) (2, 0) + 13 6 0
L (1, 2) (0, 1) − 12 0 0
ec (1, 1) (0, 0) +1 6 0
ν c (1, 1) (0, 0) 0 0 0
5
All sixteen states satisfy (3), so g generates a Z6 = Z3 × Z2 that acts trivially on the whole package;
since Z(SU (3)) alone is non-trivial on Q (t = 1) and Z(SU (2)) alone is non-trivial on L (d = 1),
the trivially-acting subgroup is exactly this correlated cyclic Z6 , and the group acting faithfully
is S(U (3) × U (2)) = GSM /Z6 . [LIB2-083] This much is derived from the loaded matter content
(conditional on the field dictionary). It does not fix which Γ ⊆ Z6 is the physical global structure
— that remains a line-operator-level choice not seen by the matter spectrum — and it does not
upgrade the gauge group itself, which stays an input at the group level by Theorem 2.1. [LIB2-083]
5 The potential layer and dynamical shadow selection
5.1 Construction and certification
The scalar potential (1) was implemented on the substrate with M = exp(2Φ) on the K-orthonormal
noncompact basis (the 70 of the E7(7) /SU (8) coset, in which the missing-16 projector P16 of the
V9 anchor acts), with exact polynomial gradients and Hessians at the origin, cross-checked by finite
differences row by row. Certification against published values, with the calibration constant exactly
1:
• V (0) = 14 Tr θ2 − 81 (Tr θ)2 reproduced exactly for all seven electric rows: SO(8) : −6; SO(7, 1) : −2.5;
SO(6, 2) : 0; SO(5, 3) : +1.5; SO(4, 4) : +2; CSO(7, 0, 1) : −35/8; CSO(6, 0, 2) : −3; and −6 for
the dyonic ω-rotated SO(8) (origin ω-independence);
• the N =8 point: m2 = −4, seventy-fold degenerate (= − 23 |Λ|, the published value);
• the dyonic SO∗ (8) row (θ = ξ): a Minkowski vacuum at the origin with ∇V = 0 exactly
(×20)
and mass spectrum m2 = {2(×2) , 12 , 0(×48) } — the published spectrum, multiplicity for
multiplicity;
• the zeroth-order tadpole-direction ranks 1/5/8/0 of the earlier engine regression are reproduced
unchanged (same P16 , same substrate).
Mass units: Hessian eigenvalues in the 56-trace-orthonormal scalar basis are converted by the factor
12 (the Dynkin index of the 56 of su(8) over the fundamental, Λ2 8 ⊕ Λ2 8̄ = 2(N −2)); the factor is
anchored on the SO(8) value and then predicts the SO∗ (8) spectrum, which lands exactly.
5.2 The selector and the result
Novelty label: new specialization proved here.
Theorem 5.1 (Shadow selection at a stable SO(6, 2) point; S77). Define the selector on a gauging
Θ at the scalar origin:
S(Θ) ≡ P16 ∇V = 0 ∧ P16 Hess V P16 > 0 (Minkowski criterion)
∧ the residual gauge algebra hosts su(3) ⊕ u(1) ⊕ u(1) .
Over the scanned family (the full electric SO(p, q)/CSO ladder, the dyonic ω-rotated SO(8), and
the dyonic SO∗ (8)):
6
(i) SO(6, 2) √is the unique electric row satisfying all three conditions: V (0) = 0; the entire tadpole
(|∇V | = 2) lies along the DKK dilaton of the 6+2 grading — a 54-direction — with the P16
component at machine zero (1.5 × 10−15 ); the sixteen eigenvalues of P16 Hess V P16 are strictly
positive (m2 ∈ [0.171, 3.830]); and the residual algebra so(6) ⊕ so(2) (dimension 16, compact)
contains u(3) ⊕ so(2) ⊃ su(3) ⊕ u(1) ⊕ u(1) (containment residual 2 × 10−16 ).
(ii) Its dyonic completion SO∗ (8) ∼ = SO(6, 2) is a genuine Minkowski vacuum (all seventy tadpoles
zero) with residual SO(6) × SO(2), the published spectrum above, and the shadow hosted; the
dyonic ξ exactly lifts the single DKK runaway of the electric row. [LIB2-079] Stability extends
beyond quadratic order: 46 of the 48 flat directions stay exactly flat to the probed amplitudes
along the probed paths, and the two exceptions — the flat directions intersecting the 16 — lift
positively at quartic order (Rev32.9, A1647: through Rev32.8 this said “all 48 . . . exactly
flat” beside the two that lift) (V ∝ t4 measured); three hundred random flat-space combinations
produce no tachyon.
(iii) Every other row fails: SO(8) and SO(8)ω are the AdS N =8 point (m2 = −4: BF-allowed in
AdS, but failing the Minkowski criterion); SO(7, 1), SO(5, 3), CSO(7, 0, 1) have P16 -tadpoles
of order one; SO(4, 4) passes both stability legs but its residual so(4) ⊕ so(4) ∼
= su(2)⊕4 contains
no su(3) (and the point is de Sitter); CSO(6, 0, 2) — the structural foil, with the same DKK -
aligned tadpole geometry — fails twice: P16 -Hessian entirely negative, and residual so(6) only
(no second u(1)).
Remark 5.2 (Honest fences). (a) Uniqueness in Theorem 5.1 is family-relative: the scan covers the
stated rows; Theorem 2.1 bounds what any unscanned gauging could deliver (never more than the
shadow), and Theorem 3.1 now bounds the representation content too (no unscanned gauging can
carry matter-colour su(3)); but the uniqueness of SO(6, 2) among all gaugings is not claimed. [LIB2-
079] (b) “Shadow” is an algebra-type statement: the residual so(6) ⊕ so(2) hosts su(3) ⊕ u(1) ⊕ u(1);
the identification of the two u(1) charge patterns on the 56 with the derived YSM and 3(B−L) was
the designated next computation. Executed (S78), with a two-part outcome. (b1) At the origin of the
selected family the match fails at the representation level, exactly: the matter 27 (the kernel-locked
+2 grade of the DKK grading) branches under the residual as 150 ⊕ 6+1 ⊕ 6−1 , refining under
su(3) ⊕ u(1)A ⊕ u(1)B to one octet, one singlet, and six triplet-types (Casimir clusters exactly
(3)
{−3(8) , 0(1) } and − 43 ), whereas the derived matter pattern 16 + 10 + 1 has nine singlets and no
octet. The residual su(3) is the gauging-frame (vector-type) embedding, not conjugate to the matter
colour; branching multisets are conjugation invariants, so no GL(2, Q) mixing of the two u(1)’s
and no alternative su(3) ⊂ residual (unique conjugacy class) can repair this. Per the pre-registered
Paper 7 fence, “SM-shadow selected” therefore demotes, at this point, to “stable su(3) ⊕ u(1) ⊕ u(1)
point selected.” (b2) The naive rigidity argument fails by computation (all 36 physical moduli move
the DKK grading), so a moduli scan appeared necessary; it is not — the question is settled by
counting (S78b). The 56 of E7(7) under the residual-class su(3) decomposes with Casimir clusters
(36)
{−3(16) , − 34 , 0(4) }: four su(3)-singlets in the entire 56, while the matter pattern requires nine
inside a 27-plane. [LIB2-080] Conjugation preserves 56-content, and the faithful su(3) class in so(6) is
unique; hence no point of the 36-parameter vacuum family — indeed no su(3) inside any E7 -conjugate
of the residual, acting on any 27-plane — can carry the matter multiset. [LIB2-080] The demotion is
final for the entire SO(6, 2)/SO∗ (8) family. [LIB2-080] The reframed question — matter colour is the
unitary class in su(8) (8 ↓ su(3)c = 3 ⊕ 5 · 1), whereas every so(p, q)-type residual — including the
Warner SU (3) × U (1)2 point of the SO(8) gauging — supplies only the vector class; can any gauging
carry the unitary class? — was executed (S80) and is now Theorem 3.1: the global negative holds.
The labeled shadow is excluded for every gauging, not merely for the scanned family; the demotion
7
of (b1) is thereby global and final. (c) The stability legs alone do not single out SO(6, 2) (SO(4, 4)
survives them); the shadow-hosting leg is load-bearing. (d) The rank-24 mechanism (12 × 2 = 24)
remains a conjecture; the 48 flat directions (= 2 × 24) of the SO∗ (8) vacuum are consistent with it
but do not prove it. (e) The Jvac -centralizer alternative selector is dead by computation: stabe6(6) (X)
has dimension 52 with signature (28+ , 24− ) — an F4(4) -conjugate — at every det = 1 rank-3 point
tested, including Jvac = diag(ϕ, 1, ϕ−1 ) and degenerate-spectrum points, and its center is zero: it is
semisimple and cannot source even one u(1), hence neither shadow Cartan. Distinct eigenvalues
do not shrink the annihilator. (f) Cosmological note (S79b). In full-70 FRW evolution the electric
SO(6, 2) row is lethal: its single DKK dilaton tadpole drives immediate recollapse (3/3 trajectories,
t ≲ 1 — fast collapse, not quintessence), so the dyonic completion is load-bearing dynamically
as well as statically. (g) Frame-flip distinction, with granularity stated (S80). The symplectic
form Ω implements the Cartan involution on the substrate (Ad Ω = θ exactly: −X T on all 133
generators, residual zero), flips the DKK grading (σDKK σ −1 = −DKK , spectrum {±3:1, ±1:27}),
and time-mirrors the geodesic-16. Under σ = Ω the purely electric gauging families map out of
frame (onto magnetic), while the dyonic family maps into itself : σ(ΘSO∗ (8) , E+M ) = +(E−M ),
the conjugate dyonic partner (residual 2 × 10−16 ). The granularity is part of the statement: no
individual Θ is σ-fixed (pointwise stability fails for every gauging); σ-closure holds at family level,
and the dyonic family is the unique σ-closed attached family. The dyonic completion is thus
distinguished a third time — statically (S77), cosmologically (S79b, fence (f)), and by frame-flip
closure (S80) — by three independent computations. No dynamical or temporal interpretation of σ
is claimed. Rev14 update (S82): the σ = Ω family map now carries three-way external provenance
(A762b); and three adjacent readings are closed negative at computation grade (D763, 4/4): the
dyonic pair-of-pairs is obstructed, flickering rest mass is excluded (ε = id exact — shadow mass is
σ-permanent), and the σ-turnaround link is null (H = 0 smooth, stabilizer dimension constant).
(h) Stability language (Rev14, binding). All “stable”/“no tachyon” statements in this appendix are
classical, quadratic-order statements. [LIB2-079] The published one-loop result for this vacuum class
(Str M8 > 0, arXiv:1307.4389) destabilizes N =0 Minkowski vacua; no quantum-stability claim is
made. See Appendix G for the post-Rev13 tower/annulus layer on the same vacuum.
5.3 The truncation question settled, and the geometry of the missing 16 (S78b)
The same session closed the long-standing truncation hole (Paper 4, T4-C) in both directions,
by the Lie-triple criterion (a subspace m ⊂ p of a symmetric space exponentiates to a totally
geodesic submanifold — equivalently supports a consistent scalar sigma-model truncation — iff
[[m, m], m] ⊆ m):
• The Jordan-54 is NOT a Lie-triple system: the projection of [[54, 54], 54] onto the
missing-16 has order-one components (max 0.815 over 3 × 104 triples, exhaustive in the
leading block). [LIB2-008] No consistent truncation of even the ungauged maximal theory retains
exactly the Jordan-visible 54 scalars; the “D=5→D=4 Jordan-54 model” is definitively a
counting/structural correspondence. (The gauged-layer observation that no electric row has
rank-zero leakage was the shadow of this sharper fact.)
• The missing-16 IS a Lie-triple system, to machine precision (4.9 × 10−15 over all 1,920
triples). [LIB2-009] It generates g′ = [16, 16] ⊕ 16 of dimension 28, closing at 7.4 × 10−15 , with
compact part of dimension 12, trivial center, rank 4, and Cartan character +4 — a tuple that
by itself does not distinguish so(4, 4) from g2(2) ⊕ g2(2) ; the identification is certified on the
8
registered plane by s1167 (End = 1, Killing signature (16, 12), 24 roots of one length; Rev32.9,
A1632): g′ ∼ ∼ so(4) ⊕ so(4). [LIB2-010] The missing-16 is, locally, the totally
= so(4, 4), [16, 16] =
geodesic submanifold SO(4, 4)/(SO(4) × SO(4)) of E7(7) /SU (8) (the identification is a local
isometry only; Rev32.9, A1632) — the split-octonion (4, 4) norm-form Grassmannian. The
framework’s foundational split signature reappears at the top of the E7 stack as exactly the
geometry of the “missing” sector, and the geometrically self-consistent sub-sector is precisely
the one the potential layer stabilizes (Theorem 5.1).
Scope fence: “consistent truncation” here is the scalar sigma-model criterion; vector-sector statements
are downstream and not claimed. [LIB2-008, LIB2-009]
6 Net position of the gauge sector
The gauge boundary, stated once. The gauge group of the Standard Model is an input
of this framework, and Rev29 now knows exactly why and exactly how much: (1) it could not
have arisen as a gauging of the maximal supergravity layer — Theorem 2.1 (no su(3) ⊕ su(2),
with su(2)L as precisely the piece that can never fit) and Theorem 3.1 (no matter-colour
su(3), in any gauging, at any vacuum) — two proven walls: the right algebra, never the
right representation; (2) given the derived matter and hypercharges, it could not have been
any other group — Theorem 4.1, up to U (1)B−L (itself the unique extra Cartan, 3(B−L))
and global structure Γ; (3) the maximal derivable remnant, an su(3) ⊕ u(1) ⊕ u(1) of shadow
type, is not inserted but dynamically selected by vacuum stability of the missing-16 sector
at a stable SO(6, 2) point whose dyonic completion is a published Minkowski vacuum —
Theorem 5.1, with the fences of Remark 5.2; the SM charge labels for the selected u(1)’s
failed at the origin point (S78), were excluded for the whole vacuum family by counting
(S78b), and are excluded for every gauging by the charge wall (S80) — the shadow is
structural, not the Standard Model in disguise. Nothing in the observable map of Papers 1–7
changes: no public number moves. What changes is the epistemic status of the border: the
gauge-sector holes are no longer open questions but two proven walls, a uniqueness bound,
and a fired selector whose charge labeling is closed negative at theorem grade.
7 The frame-free Higgs line: an E7(7) no-go (S231–S232)
A third wall, of the same character as the two above, closes the question of whether the electroweak
Higgs line can be selected without a loaded boundary. Working in the Freudenthal system F =
R ⊕ R ⊕ J3 (Os ) ⊕ J3 (Os ) with its frame-free automorphism group G0 = Aut(F, ω, Q)0 = E7(7) acting
on the real 56:
Novelty label: new specialization proved here.
Theorem 7.1 (No frame-free Higgs line; S231/S232). There is no G0 -invariant rank-one projector
on the 56: the only invariant idempotents have rank 0 or 56. [LIB2-078] Consequently the electroweak
vacuum line (load-ew-line) cannot be selected frame-free; it is a loaded datum.
Proof (explicit, octonion-free; S232 kernel s796). Present e7(7) = sl(8, R) ⊕ Λ4 R8 (dimension
63 + 70 = 133) acting on 56 = Λ2 R8 ⊕ Λ2 R8 ∗ . The 133 generators (i) are independent (raw
9
rank 133), (ii) all lie in sp(56, R) (preserve the intrinsic symplectic form), and (iii) close as a Lie
algebra of dimension exactly 133 — so the generated algebra is e7(7) , not the larger sp(56). The
commutant {M : [X, M ] = 0 ∀X} then has dimension 1, spanned by I: the 56 is real-irreducible
with commutant R · I (real type). An invariant idempotent lies in the commutant, so its only values
are 0 and I, i.e. ranks {0, 56}; no rank-one invariant line exists. [LIB2-078] □
This upgrades the standing obstruction from a citation (irreducibility of the minuscule ϖ7 ) to an
in-house construction with nothing assumed. It is the action-side counterpart of the two gauging
walls: the boundary the Standard Model needs is loaded, not internally selected — and, as elsewhere
in this suite, no public observable moves. (This is the theorem folded into the interacting-vertex
chain of Appendix R as the oldest standing import.)
8 The invariant-connection kill: a holonomy no-go (Rev30, A1496)
The freeze-first arc closed one of the two decidable rows of its plan by a holonomy argument, and the
result is a gauge-boundary statement, so it is recorded here. The declared class is part of the
theorem and must travel with it: left-E7(7) -invariant principal SU (8) connections compatible
with Ω and the loaded Ksel . [LIB2-091] Inside that class, and only inside it, the class is empty of
anything that could cancel the obstruction.
The argument is two steps, both rebuilt end to end in house (s1014, s1015) and agreeing with the
lane to 9 × 10−15 .
Step 1 (uniqueness, by Schur and Wang). The Casimir acts as an exact scalar on each summand of
the symmetric-space splitting — 3/2 on p (70-dimensional) and 4/3 on k (63-dimensional). The
two scalars differ, so HomSU (8) (p, k) = 0 by Schur, and Wang’s theorem then leaves the canonical
symmetric-space connection as the only connection in the class.
Step 2 (holonomy, by Ambrose–Singer). Evaluating that connection’s curvature on all 2415 noncom-
pact pairs gives coefficient rank 63, so Ambrose–Singer returns Hol = SU (8) in full. The holonomy
therefore escapes Ksel , and by Step 1 there is no other connection available to cancel it. [LIB2-091]
The frozen kill condition fires.
What this is not. It is not a topological no-section theorem — the base is contractible,
and no such statement is claimed. What remains inhabited after the kill is a non-equivariant
Ksel reduction σsel carrying a preserving connection, and it is typed and counted rather
than waved at: 43 continuous fibre coordinates (53 once the source is fixed), 0
discrete. [LIB2-091] Stating the kill as “no compensator connection exists” would overstate a
class-relative result into a topological no-go this appendix explicitly disclaims.
The parallel dependency survives the kill. The row-2 computation is carried out before any row-1
output exists — only the explicitly labelled Ksel,j variant consumes the source ray — and row 1
consumes only frozen-frame tensors together with the boundary datum L∂ . That a later row must
import both outputs is composition, not serialization, so the plan’s parallel split is intact.
The companion row is a success only relative to a typed input, and is recorded that way. The
vector/Higgs 10 carries a unique invariant symmetric form η10 of split signature (5, 5) — house-
checked as an exactly one-dimensional invariant-form space, with a random form failing the same
test. But the generation of the source is not zero-input and was never claimed as such: L∂ costs
10
10 continuous parameters, U one inherited slot, ρ one selection-neutral slot, and one orientation
bit that exists only after the ray has been fixed. The physical attachment of L∂ remains Input.
A kernel coordinate appearing in the construction is a basis witness, not a physics identification.
Nothing in this section promotes an observable, moves a tier, or touches the engine.
Provenance
S74: compact wall and first LΘ table. S75: branching-wall theorem (stages 0–3b, master log; 21/21
kill table). S76: D756 close (uniqueness; three-way, two genuine returns plus house extension; all
anomaly sums exact). S77: potential layer (stage-0 cache plus row runner; nine rows; master table),
SO∗ (8) benchmark, flat-direction probe, Jvac -stabilizer run. S78/S78b: charge-pattern test and
counting kill; Lie-triple geometry of the missing 16. S79b: full-70 FRW cosmology (electric-row
lethality; basin boundary). S80: charge-wall theorem (Inv dimension 140, rowspace 125, octet census
3-way; blind replication A761h, 7/7) and the σ = Ω frame-flip family map (external rerun of the
S77/S78 layer: end-to-end identical, A761g). Citation checks performed against arXiv:1112.3345
directly; the Warner-vacuum literature anchor for SU (3) × U (1)2 at SO(8) complex scalars is
Warner ’83 and Fischbacher–Pilch–Warner arXiv:1010.4910 (not 1112.3345, whose scan excludes
complex vevs).
Rev14 (S82–S88). S82: σ = Ω three-way (A762b); D763 closed negatives; annulus/shell-straddle
ledger (D764, blind collision at D761R pedigree). The post-Rev13 tower, annulus geometry, e6(6)
identification, and Jordan-interface certificates are collected in Appendix G with their own provenance
and SHA manifests (S82–S88).
Rev25 (S214 fold). The anomaly filter of this appendix’s uniqueness argument is now backed by the
complete exact audit: kernel s588 verified that all SM gauge and gravitational anomaly sums cancel
exactly from the framework’s own derived hypercharges (upgrading the structural s572 check),
formally closing hostile-board item T3-P. Cross-references: Paper 1 (anomaly-cancellation status,
Rev25 update) and Paper 7 (Rev25 additions).
11
Leibniz Quantum Beats Newton
Appendix G — The Kaluza–Klein Annulus: Scherk–Schwarz-Type Mass
Levels, e6(6) Identification, and the Jordan Interface (Rev33.1)
Tom O’Sieg
August 2026
Abstract
This appendix records the post-Rev13 layer (Sessions 82–88) of the certified E7(7) 56/133
substrate programme: the discovery, lawful description, documented identification, and finally
algebraic naming of the vacuum mass tower. (i) The annulus geometry (S82): every Dirac
pairing connects the 27 to the 27′ through the (78+1) annulus; each shell carries one full
shadow copy; the grading operator DKK (named for the Kaluza–Klein reading, which is not
derived — see the vocabulary note) is a third independent grading. (ii) The tower law (S87):
on the SO∗ (8) Minkowski vacuum the scalar Hessian satisfies the Kaluza–Klein form identity √
H = −(ad Z2 )2 modulo Goldstone directions; masses obey m = |q| q0KK with q0KK = 1/(2 6)
exact (the tower quantum; distinct from the clock anchor q0clk = me of Appendix H — two
constants, two names from Rev32.2); the twelve Goldstones are exactly the charged content
of the massless 48. (iii) The documented identification (S87): the vacuum spectrum coincides
point-by-point with the CSO(2, 0, 6) ≡ CSSN =0 model of Catino–Dall’Agata–Inverso–Zwirner
(arXiv:1307.4389, Table 2) — a twisted reduction on a spacelike circle — including the published
“12 Goldstones + 36 moduli”; the timelike alternative is excluded by document (Hull–Julia,
hep-th/9803239: a timelike circle forces the non-compact SU ∗ (8) denominator). (iv) The naming
theorem (S88): the annulus is e6(6) ⊕ so(1, 1)KK — the five-dimensional U-duality algebra —
and the shells are its 27/27; equivalently, the tower lives on the canonical minuscule 3-grading
e7(7) = 27′ ⊕ (e6(6) ⊕ so(1, 1)) ⊕ 27, i.e. the 4D ← 5D Kaluza–Klein decomposition, with the
Jordan 27 realized as the cubic boundary slice of the Freudenthal 56. The clock is independent
of all of it: the certified A1 block is exactly so(2)-neutral — tick and tower are separate exact
structures on one vacuum. All results are classical/quadratic-order statements on the certified
substrate; no public observable moves.
Vocabulary (Rev32.9, ruling R-19; A1644 E-7/E-9). None of the three graded objects
in this appendix — the so(2) charge q of Theorem 3.1, the Z4 tick of Appendix P, the
grading operator DKK — is exhibited as the fifth momentum p5 of a Kaluza–Klein reduction:
no compact coordinate and no conjugate-momentum map is registered. “Kaluza–Klein”
and “Scherk–Schwarz” here name the documented identification with the Z2 -twisted 5D
reduction of arXiv:1307.4389 at the level of the published spectrum — a finite three-level
scalar Hessian with m = |q| q0KK — not a derived tower. Momentum vocabulary is not used
for q, the tick, or DKK .
Notation (σ). Where this appendix cites σ as an involution it means Appendix F’s SO∗ (8) time-
mirror σ = Ω. It is not Appendix I’s Jordan frame involution σframe (ρ 7→ 1/ρ), not Appendix H’s
1
throat involution σC , and not the split-octonion conjugation σoct of the A689 dictionary.
1 Scope and provenance
Everything in this appendix is a finite, machine-verifiable statement on the certified E7(7) substrate of
Appendix F, obtained in Sessions 82–88 (2026-06-10/11) under pre-committed gates, with SHA-256–
manifested artifacts (_S82–_S88 manifests; house scripts s82*–s88*). External returns used here
(D763/D764 annulus computations; D772/D773 scrutiny items; D774 interface lemmas) were re-run
or cross-certified in house before banking. Conventions: DKK = ρsl8 diag(−3, −3, 1, 1, 1, 1, 1, 1)
(vacuum-aligned frame; the index-permuted form diag(16 , −3, −3) is conjugate and used interchange-
ably); Z2 denotes the certified so(2) center of the residual compact algebra at the SO∗ (8) Minkowski
origin.
2 The annulus and shell geometry (S82)
Under ad DKK the adjoint grades as 133 = 27−4 ⊕ (78+1)0 ⊕ 27+4 and the fundamental as
56 = 1+6 ⊕ 27+2 ⊕ 27′−2 ⊕ 1−6 . The banked S82 ledger (D764 close; blind external collision with
the sealed house table, house re-run identical to BLAS noise) establishes:
1. every Dirac pairing connects the 27 to the 27′ through the annulus — “stage in the ring, shells
at the rim, cores empty” (Rev32.9, A1642 O8: through Rev32.8 the slogan said “matter in
the shells”; the 27/27′ shells are Jordan-cubic labels, flat at quadratic order, and carry no
particle-content claim); the geodesic-16 sits at grade 0 with fraction exactly 1.000000 (S82b);
2. each shell carries one full shadow copy; DKK is a third independent charge, commuting with
the shadow gradings;
3. three closed negatives (D763, 4/4): the dyonic pair-of-pairs is obstructed (su(2)4 certified,
the pair irreducible); flickering rest mass is closed negative (ε = id exact — shadow mass
is σ-permanent in the E+M completion); the σ-turnaround link is closed negative (H = 0
smooth, stabilizer dimension constant);
4. the pre-registered information-content sweep returned its null (no “annulus number”); the DKK
ray is the unique residual-symmetric collapse channel (16-vs-0).
The σ = Ω frame-flip family map of Appendix F fence (g) was upgraded to three-way external
provenance en route (A762b).
3 The tower law and the documented identification (S84–S85, S87)
Novelty label: new specialization proved here.
Theorem 3.1 (Kaluza–Klein form of the Hessian; S87). At the SO∗ (8) Minkowski origin, on the
orthogonal complement of the twelve Goldstone directions, the scalar Hessian satisfies H = −(ad Z2 )2
2
(operator residual 4.4 × 10−10 relative); consequently every scalar mass obeys m = |q| q0KK with q the
integer so(2) charge and
q0KK = 2√
1
6
(exact; deviation 5.3 × 10−16 ).
The twelve Goldstones are exactly the charged content of the massless 48 (principal cosines
1.000000000000); the tower lies entirely in the (78+1) annulus; the 27/27′ shells are flat at
quadratic order; [Z2 , DKK ] = 0.
The supporting spectral layer was completed in S85 (28 vector masses: 16 massless with residual
1 ×20 1 ×2
so(6) ⊕ so(2), 12 exactly degenerate massive; scalar Hessian {0×48 , 24 , 6 }, no tachyons, m2 ∝
(so(2) charge) ; massive vectors the 6+1 ⊕ 6−1 coset; shadow u(1)’s = Cartan of the broken SO∗ (8)),
2
and the σ-link closed at theorem grade (three lanes, two house runs, one blind collision; σ-closure
exactly 0.0 two ways; random-Θ control 7/7 — the flip identity is kinematics, the dyonic content is
σ-closure).
Identification. The vacuum spectrum coincides, entry by entry, with [SO∗ (8)]c=1 ≡ CSO(2, 0, 6) ≡
CSSN =0 of arXiv:1307.4389 (Table 2) — the Z2 -twisted reduction of the 5D maximal theory on
a spacelike circle — including the published “12 Goldstones + 36 moduli” (our K4 measurement)
and the spin-unified rungs (rung unit: the published rung is the u(1) charge Q of arXiv:1307.4389
Table 2, and Q = 2|q| with |q| the so(2) charge of Theorem 3.1. The centre Z2 acts on the 8 of su(8)
with charges ± 12 (four each, in units of q0KK ) — the unique solution reproducing its spectra on the
56 = Λ2 8 ⊕ Λ2 8 and the 70 = Λ4 8, house s1201. Hence gravitini |q| = 12 (Q = 1), the 12 massive
vectors and the 20 massive scalars |q| = 1 (Q = 2), the two heaviest scalars |q| = 2 (Q = 4); in the
published table m2 = Q2 /8 holds for the gravitini and all 28 vectors, but not charge-by-charge for
spin 12 (Q = ±1 carries both m2 = 0 and 18 ). Through Rev32.11 this parenthesis read “gravitini at
half-rung; . . . at rung 2”, one sentence in two units, Rev33.0, A1642 O9). The timelike alternative is
not the identified case: Hull–Julia (hep-th/9803239, Table 2, row D = 4) shows that a timelike circle
replaces the SU (8) denominator by its non-compact real form SU ∗ (8) (maximal compact U Sp(8);
see also Hull, hep-th/9806146); the spectrum-grade identification of this appendix is to the spacelike
table only. (Rev32.6, S302.4: the earlier gloss “with no compact u(1)” was the house’s reading, not
the source’s, and is withdrawn. The ladder-centralizer census of Appendix N is invariant under
SU (8) → SU ∗ (8) — s1160: θ′ = θσ fixes su∗ (8), dimension 63, Killing signature (27, 36, 0); θ′ = θ
on C(H) to principal angle 2.6 × 10−8 — so nothing computed in Appendix N depends on which
real form the circle selects.) The identification is established at spectrum grade and at substrate
grade: the canonical CSS flat algebra F = Z2 ⋉ t12 exists inside e7(7) (27+4 = 15neutral ⊕ 12charged
at |q| = q0KK , deviation 4.4 × 10−16 ; abelianness grading-forced). The gauging-grade equivalence
(that the suite’s gauging is itself a CSS gauging) remains open; the suite’s own gauging is not flat
(symmetric pair, forced), so the 5D reading is currently: same vacuum, same spectrum, same flat
algebra present in the substrate — not yet the same gauging.
Clock ̸= quanta (S87). The certified A1 block is exactly so(2)-neutral (residual 1.3 × 10−17 ):
the Z4 tick is not a u(1) holonomy of the tower charge; (q, d) are independent commuting gradings.
The relational tick τ = +i (S85–S86; rigid, a constant of the motion, ≤ 1.4 × 10−14 across static
and FRW probes) and the tower are separate exact structures on one vacuum. The suite is exactly
orientation-even at the A1 and full 36-isotypic layers (S87 scan, worst line 3.8 × 10−14 against a
10−12 gate; the only odd content is the sign convention of Im τ ); pair-conjugacy is certified at the
full 36 layer (cosines 0 to 3.5 × 10−16 ).
3
4 The annulus is e6(6) (S88)
Novelty label: standard theorem used.
Theorem 4.1 (F-ANNULUS-E6; S88, house). Let m be the trace-form–orthogonal complement of
R DKK in the grade-zero subalgebra g0 = ker(ad DKK ) ⊂ e7(7) . Then m is a simple real Lie algebra
isomorphic to e6(6) , and the grade-±4 shells are its 27/27:
e7(7) = 27′ ⊕ e6(6) ⊕ so(1, 1)KK
⊕ 27.
Certificate (all gates pre-committed; script s88a, manifest _S88_outputs.sha256): closure of g0
exact to machine zero and B(DKK , [g0 , g0 ]) = 0 exactly; trace-form signature on m equals (42, 36) —
the e6(6) fingerprint, which excludes the real forms e6(2) , e6(−14) , e6(−26) ; generic-element centralizer
dimension 6 (rank); adjoint Casimir = 2 ⊮ to 2.3 × 10−15 (simplicity); the action on the grade-+4
shell is absolutely irreducible (commutant dimension 1, cyclic rank 27/27, traceless), with Casimir
ratio
c27 13
= exact to twelve digits,
c78 18
the e6 fundamental/adjoint ratio. The dimension-and-rank foils so(13) and sp(12, R) (both 78-
dimensional, rank 6) were declared in advance and are killed by the existence of the 27-dimensional
irreducible action, which neither algebra possesses.
Remark 4.2 (Reading). e6(6) is the U-duality algebra of maximal supergravity in five dimensions.
Theorem 4.1 therefore upgrades the Scherk–Schwarz identification ladder by an algebra-grade rung:
the suite’s own grading reproduces the canonical 4D ← 5D decomposition — spectrum grade (Table-2
match) → substrate grade (F = Z2 ⋉ t12 ) → algebra grade (this theorem) — with the gauging-grade
equivalence still open.
Novelty label: standard theorem used.
Lemma 4.3 (Jordan interface; D774, house cross-certified). In the graded 56 = 1+6 ⊕J+2 ⊕J−2 ∨ ⊕1
−6
∼
with J+2 = J3 (Os ): (i) the symplectic form Ω pairs only opposite grades; J+2 is isotropic and
LJ = Re− ⊕ J+2 is Lagrangian; (ii) Ω : J+2 × J−2 ∨ → R has full rank 27; (iii) the quartic invariant
restricts to the Jordan cubic norm by adjoining the opposite singlet, I4 (βe− + X) = β N (X) (β-
linearity verified at 1.4 × 10−14 with non-zero witness), the grading making any other monomial
weight-forbidden; (iv) the stabilizer of e− inside g0 is precisely the e6(6) of Theorem 4.1 (certified by
the same invariant gates, including c27 /c78 = 13/18 on the 56-side Jordan 27).
Remark 4.4 (Known mathematics, certified in-substrate). Lemma 4.3 is classical Freudenthal-triple-
system / Jordan-pair structure theory (Freudenthal; Günaydin–Koepsell–Nicolai, hep-th/0008063;
Krutelevich, math/0411104; see also the exceptional-supergravity literature). Its content here is not
novelty but certification: these are now finite, re-runnable facts about the same certified substrate
that carries the walls of Appendix F and the tower of Theorem 3.1, with the suite’s measured
objects (annulus, shells, DKK , Jvac -aligned frame) appearing as the named classical structures. Two
independent same-day routes (house annulus-invariants; external Freudenthal-pairing free hand,
D774) converged on the same identification and were cross-certified (script s88b).
4
5 Phase box and universal weight (S85–S86)
For completeness of the post-Rev13 record: the phase-box keystone PH1 is the equivariant L-
intertwiner ((z, z, z) block, spec ikkljl->ij; equivariance ≤ 2 × 10−15 ; multiplicity-one Schur
family; origin bank un-fit 9/9; kill test ≤ 3.3 × 10−15 at 28/28 against naive controls at 1.1 × 10−5 –
1.3 × 10−2 ). The relational tick τ = +i is rigid (S86; retires the S85 “suggestive” caveat; cite as
Q11/Q12 — the tick is phase structure, never a frequency). The certified pair is global-phase
conjugate everywhere (overlap 1 to 2.5 × 10−16 , 8/8); the Ξ determinant-phase observable is blind
at the origin (8 × π/4 = 2π) and demoted to off-origin, where it is pair-common (forced). The
A770 universal-weight lemma fixes every Peirce channel weight at 2/9 exactly, making the 7/3 ratio
a conditional trace theorem at the finite internal layer — its physical (LSZ/S-matrix) promotion
remains open, and the Z12 fused-clock reading is dead (C ∈ / N (e7(7) ); recorded in the junkyard).
6 Fences and language (Rev14)
Binding language fences for everything in this appendix. (1) Stability is classical.
Every “stable/no tachyon” statement is a quadratic-order statement on the classical potential.
The published one-loop result for this vacuum class has Str M8 > 0 and destabilizes N =0
Minkowski vacua (arXiv:1307.4389); no claim of quantum stability is made anywhere in the
suite. (2) Spacelike only — stated as a predicate (Rev32.6). The KK circle is spacelike: the
compact orbit’s norm stays positive on the carrier. Timelike-torus/timelike-circle language
is excluded by that predicate, not by label (PE2; Hull–Julia hep-th/9803239 Table 2 gives
the timelike case its own real form, SU ∗ (8), which this appendix does not identify); the
Gödel null circle, sinh r = 1, is the exhibit of the predicate failing. The annulus reading
is (spacelike circle) × (internal twist u(1) ⊃ Z2 ). (3) No chirality gloss. The 27/27′ of
this appendix are Jordan-cubic versus symplectic-pair structures. They are not 4D fermion
chirality, and no “vector-like bulk versus chiral boundary” physics claim is made; any such
mapping requires an explicit fermion-embedding statement that has not been made. The
walls of Appendix F stand on their own proofs; the interface picture is consistent with
them, it does not re-derive them. (4) Numerology class. The G7 = ϕ + 5ϕ−5 correction
and the related tau-‘5’ bridge are formally classified numerology-class pending derivation
(D772/D773, artifact-backed); an attempted kill of the integer 5 via the α(0)/mτ error
budget does not land (the required coefficient is 4.9635 ± 0.087, containing 5 at 1σ; the
ΛG2 rounding convention dominates), so the bridge is scheme-dependent candidate-class —
neither derived nor dead. [LIB2-229] (5) Tick. τ -rigidity is cited as Q11/Q12; the tick is phase
structure, never frequency; clock ̸= quanta is banked — no holonomy gloss. (6) Scope. No
public observable moved anywhere in S82–S88; the gauging-grade CSS equivalence and the
physical (LSZ) layer of 7/3 remain open and are stated as such.
Provenance
S82: annulus/shell-straddle ledger (D763 4/4 closed negatives; D764 3/3 blind collision at D761R
pedigree; s82_annulus_straddle, s82b_stage_grade). S83: the extractor stack (su(8) frame via
triality-computed 8s , Schur intertwiner 3.9 × 10−16 , exact-912 dressing; gravitini 1/4 × 8 exact — the
A1 Takagi values in the de Wit–Samtleben–Trigiante normalization, not a rung label; E±M phase
5
conjugates at origin). S84: σ-link at norm level (15/15 ≤ 7.1 × 10−15 , controls fail 15/15). S85:
D767 arc (blind replication A767a point-identical 15/15); spectral completion; relational tick +i
exact; Ξ origin-blind. S86: PH1 equivariant intertwiner; τ rigid; Q13 global-phase conjugacy; A770
universal weight 2/9; Z12 killed. S87: orientation-odd null scan; KKT joint grading (Theorem 3.1);
W4 document pull (Hull–Julia; 1307.4389); SS gauging grade (F = Z2 ⋉ t12 in-substrate; suite
gauging not flat). S88: F-ANNULUS-E6 (Theorem 4.1, s88a); D774 interface lemmas cross-certified
(s88b); D773 five-kill error budget banked. All house artifacts SHA-manifested (_S83 15/15, _S84
8/8, _S85 6/6+caches, _S86 35/35, _S87 18/18, _S88 6/6); external packs content-verified including
renamed-file hash audit (S88 sweep).
Citations: F. Catino, G. Dall’Agata, G. Inverso, F. Zwirner, arXiv:1307.4389; C. M. Hull, B. Julia,
hep-th/9803239; C. M. Hull, hep-th/9806146; M. Günaydin, K. Koepsell, H. Nicolai, hep-th/0008063;
S. Krutelevich, math/0411104.
6
Leibniz Quantum Beats Newton
Appendix H — The Time Bridge: the Moment Clock, T -Reversal as a Live
e7(7) Element, and the One Circle (Rev33.1)
Tom O’Sieg
August 2026
Abstract
This appendix records the post-Rev14 time layer (Sessions 106–108) of the certified E7(7)
56/133 substrate programme. (i) The flicker frame (S106): the clock normalisation is q0clk = me
(a unit choice inside the one registered dimensionful class — Appendix X input-count box; through
Rev32.8 this read “the single √ dimensionful anchor”) (the clock anchor; not the Kaluza–Klein
tower quantum q0KK = 1/(2 6) of Appendix G — the two shared one symbol before Rev32.2),
and the substrate clock frequency ω0 is calibrated against the electron Compton/zitterbewegung
rate, me = ℏω0 /c2 , which fixes c and ℏ in terms of ω0 ; that calibration is a declared normalisation
of the constructed clock and is not a claim that the substrate phase circle is the electron’s
zitterbewegung (Rev29 ); ω0 is a declared input, not a new constant, and the admissible-constant
set stays F3 = {q0clk , κ, Γ}. (ii) The live time-reversal bridge (S107): the past↔future operation T
is a genuine element of E7(7) on the live 56 — the half-turn of the E6 × U (1) grading su(2) — and
the live representation corrects the earlier abstract model: T is symplectic (T ∈ E7 ⊂ Sp(56, R))
and squares to −1 (a spinor half-turn), so the sign that antiunitary time reversal carries in
quantum mechanics is reproduced here as T 2 = −1 rather than as anti-symplecticity; T is
a symplectic element of E7(7) , not an antiunitary operator on a physical Hilbert space, and
no physical Kramers degeneracy is claimed (Rev29 ). T realises the moment half-turn of the
constructed clock (T 4 = +1 = two moments) and normalises the E6 structure algebra, giving a
genuine, projective 32 → 64 doubling. (iii) One circle (S108): the gauge-side dial moment-circle
and the grading circle are E7(7) -conjugate by an explicit g ∈ E7(7) , so the dial “moment” and the
grading half-turn T are the same operation, and the two 27s are the dial’s future/past windings.
(iv) Moment span vs. thickness (S108): the moment span is a single shared envelope ( 12 future
+ 12 past), while each ladder slot carries its own clock rate and hence its own model proper-time
thickness — “thickness” being a quantity of the construction, not an asserted measured property
of a physical particle (Rev29 ). A third split-octonion cell invariant accompanies c and ℏ: the
entropy-like invariant S0 = ln φ2 = 2 ln φ on the skew J4 , numerically equal to the model’s
per-tick proper-time rate — one cell, two cones, one bit [s500]; reading S0 as a physical minimal
entropy quantum is not asserted (Rev29 ). All results are classical/finite, machine-verifiable
statements on the certified substrate; no public observable moves.
Notation (σ) — three senses in this appendix. (i) σC is the throat involution, the signature-
preserving Jordan automorphism that flips the WH–BH neck; it is always written with its subscript.
(ii) Where the Jordan frame involution ρ 7→ 1/ρ is meant, it is Appendix I’s σframe , the frame
√
transposition (1 3). (iii) σ is the string tension and is not an involution at all. Appendix F’s
time-mirror σ = Ω and the split-octonion σoct (e7 7→ −e7 ) are two further distinct objects, named
here so that no bare σ in this suite is read as a single map.
1
1 Scope and provenance
Everything in this appendix is a finite computation on the certified E7(7) substrate of Appendix F
(133 generators on the real 56), obtained in Sessions 106–108 (2026-06-13) under pre-committed
gates, with SHA-256–manifested artifacts (_S106, _S107, _S108 manifests; house scripts s106*,
s107a, s108a–e; replays byte-exact). External inputs are used only in the companion mass discussion
(PDG 2026 charged-lepton pole masses, in Appendix H §9 and the p2 Koide block) and are flagged
as external. Conventions: the grading element is D56 = ρsu8 diag(1, 1, 1, 1, 1, 1, −3, −3) (vacuum-
aligned frame; the index-permuted
√ diag(−3, −3, 1, 1, 1, 1, 1, 1) is conjugate and
√ used interchangeably);
H = D56 /2; κ = 1/(3 2) (the substrate central charge, exact); φ = (1 + 5)/2. The scope clause
of Appendix G applies verbatim: these are classical, finite statements on the certified substrate; no
seconds, dynamics, or public-observable claims are made. Section 10 additionally folds the engine-side
winding ladder recorded in the workbook tabs Electron Shell Levels (S164), Winding→Mass
(forward) (S166), and Exact Windings↔Energies (s264/S167), so that the paper and the engine
agree; that material is likewise scale-free structure with all absolute energies/masses α-/anchor
fenced and the occupation selector open.
2 The flicker frame (S106)
The substrate carries one dimensionless clock (the occupation/phase circle) and one dimensionful
anchor q0clk . We set q0clk = me and calibrate the clock frequency against the electron Comp-
ton/zitterbewegung rate,
me c2
ω0 = ≈ 7.7634 × 1020 rad/s (zitterbewegung 2ω0 ≈ 1.5527 × 1021 rad/s), (1)
ℏ
so that me = ℏω0 /c2 . This is option A of s106i: it places the electron at depth 0 of the model
ladder (the “bare flicker” slot) and fixes c and ℏ within that frame from ω0 ; ω0 itself is a declared
input, not a new constant. The calibration is a normalisation of the constructed clock: nothing here
asserts that the substrate phase circle is the electron’s physical clock, and no measured quantity is
derived from it (Rev29 ). The admissible-constant set is unchanged, F3 = {q0clk , κ, Γ}. A moment is
one half-turn of the base clock carrying the spinor sign ghalf = −1; the identity (+1) return takes
two moments (a full turn).
Fence F1. The rotation is the substrate’s own occupation/phase circle. ω0 is an input
frequency anchor; no statement about external seconds or external-time dynamics is made.
3 The grading circle on the live 56 (S107)
D56 ∈ e7(7) (residual < 10−9 ) is symplectic and gives the E6 × U (1) three-grading
56 = 1−6 ⊕ 27−2 ⊕ 27+2 ⊕ 1+6 , (2)
with eigenvalue multiset {−6 : 1, −2 : 27, +2 : 27, +6 : 1} (gate a). The two level-±2 spaces are the
two clock orientations — the 27 (future) and 27 (past). H = D56 /2 has weights {±3, ±1}.
2
4 T as a genuine e7(7) element (S107a)
Build the grading su(2) by Jacobson–Morozov inside e7(7) : g±4 = {x ∈ e7(7) : [D56 , x] = ±4x}
are 27-dimensional; pick a regular E ∈ g+4 and linear-solve F ∈ g−4 with [E, F ] = H (residual
∼ 2 × 10−13 ; [H, E] = 2E, [H, F ] = −2F automatic). The compact generator is K = E − F ∈ e7(7) ,
with purely imaginary spectrum i · {±3 (×1), ±1 (×27)} — one spin- 32 plus 26 spin- 12 . The time
bridge is the half-turn
T = exp π2 K ∈ E7(7) , (3)
the Weyl reflection of the grading su(2): T HT −1 = −H (residual ∼ 9 × 10−11 ), so T maps level
k 7→ −k and swaps 27 ↔ 27 and 1+6 ↔ 1−6 exactly (leaks < 10−6 ). This is the live past↔future
clock swap.
Novelty label: new specialization proved here.
Theorem 4.1 (live T -bridge). On the certified 56 there is a genuine element T = exp( π2 K) ∈ E7(7) ,
with K the compact generator of the grading su(2), such that T HT −1 = −H; T interchanges the
future 27 and past 27 (and the two singlets) exactly.
Remark 4.2 (canonicality, S108d). The Jacobson–Morozov build uses a random regular E ∈ g+4 ;
different choices give conjugate triples. Across 6 independent seeds the build closes (residual < 10−11 )
and every invariant is reproduced — T 2 = −1 (worst 1.0 × 10−10 ), symplecticity (3.6 × 10−11 ),
T HT −1 = −H (2.3 × 10−10 ), the 27 ↔ 27 swap (2.2 × 10−12 ), and the fixed spectrum i · {±3 ×
1, ±1 × 27}. So T is canonical up to conjugacy in the grading-su(2) normaliser: “the” time bridge is
well-defined, not a seed artifact.
Remark 4.3 (independent machine re-validation (s548, S211; Rev25 fold)). The T -bridge was later
rebuilt from scratch in an independent in-house kernel: T = exp( π2 K) reconstructed with all gates
at the ∼ 10−13 level, reproducing Theorem 4.1 exactly. With the original live build and the external
verification (A1303), the bridge now has a three-way convergence record — live construction, hostile-
lane check, and independent kernel rebuild — all agreeing on T 2 = −1, symplecticity, T HT −1 = −H,
and the sector swap. Recorded as validation only; nothing new is claimed.
Remark 4.4 (neutral-signature anti-isometry as a type model (s1087, S291)). The strong finite
model of a past/future exchange as an η-isometry between signature-(1, 3) and signature-(3, 1)
summands is obstructed by Sylvester inertia. A refined model exists in neutral signature: an
explicit integral T4,4 ∈ GL(8, Z) satisfies T4,4
T ηT
4,4 = −η, T4,4 = −I, and T4,4 = I, exchanges the
2 4
two summands, and normalizes so(4, 4). In the census with p + q = 8, such an anti-isometry exists
iff p = q. This is a type-admissibility result only. The live time bridge of Theorem 4.1 is instead a
symplectic element of E7(7) on the 56; the two constructions share the order-four, square-minus-one,
and sector-swap fingerprint, but they are not identified. (27 computed gates plus one documentary
type-note; kernel s1087.)
Remark 4.5 (two-full-turn closure of the spinor dial (s1088, S291)). On the hand-built 27 ⊕ 27
dial carrier, the plain swap has square +I and vector order two, whereas the spinor dial has square
−I and vector order four. Thus the spinor dial closes after four half-turn steps, or two full turns,
and it alone has the same square/order fingerprint as the live Appendix-H bridge T 2 = −I, T 4 = I.
This is closure arithmetic of the dial model, not a measured width of the electron and not an equality
with the live 56 representation. Appendix H’s slot thickness remains a construction quantity under
its Rev29 fence. (18 computed gates; kernel s1088.)
3
5 What the live representation corrects (S107a)
The abstract model (the earlier 27 ⊕ 27 hand-built swap) made two claims the live representation
overturns:
(C1) T is symplectic: T ∈ E7 ⊂ Sp(56, R), defect ∥T T ΩT −Ω∥ ∼ 2×10−11 . It is not anti-symplectic;
the earlier “anti-symplectic” reading was a real-swap-model artifact.
(C2) T 2 = −1 on the 56 (∥T 2 + I∥ ∼ 7 × 10−11 ), the spinorial/Kramers half-turn; T 4 = +1. The
abstract plain swap had T 2 = +1 and missed it.
Structural reading (Rev29 ; constructed layer). The sign that antiunitary time reversal carries in
quantum mechanics — the fermionic T 2 = −1 of Kramers — is reproduced on the substrate by a
symplectic half-turn, not by anti-symplecticity. That is a structural match, not an identification:
T ∈ E7(7) ⊂ Sp(56, R) is a linear symplectic element of the certified substrate; it is not an antiunitary
operator on a physical Hilbert space; and no physical Kramers degeneracy, no physical time-reversal
selection rule, and no half-angle readout normalization for any observable follows from T 2 = −1
alone. (This last point is load-bearing downstream: Appendix B Lemma B.5′ has been re-typed in
Rev29 so that the π/8 half-angle is a readout choice this moment structure motivates, not one it
makes canonical.) Within the model, T 2 = −1 is exactly the moment half-turn ghalf = −1 (S106d):
T is the moment operator of the constructed clock — one application = one moment, sliding the
future sector onto the past; T 4 = +1 = two moments.
6 Energy is T -even: the clock Hamiltonian (S112, Defect-176
correction)
The half-turn T = exp( π2 K) of §4.1 settles a sign question about which substrate generator is the
energy, and with which time-reversal parity. Two grading generators must not be confused:
• the compact clock generator K = E−F (§4), with purely imaginary spectrum i·{±3(×1), ±1(×27)},
which generates the bounded unitary clock flow U (θ) = exp(θK) (| eig U | = 1, symplectic);
and
• the non-compact grading/boost H = D56 /2, with real spectrum {±3, ±1}, which cannot
generate a bounded clock.
The clock Hamiltonian of the construction is the compact K (the model energies are the eigenvalues
of the Hermitian −iK, bounded); no physical Hamiltonian is being identified here (Rev29 ). Because
T is generated by K, it commutes with it, so on the live 56
T K T −1 = +K (defect ≲ 10−6 ), (4)
i.e. the energy is T -even. This is exactly the stability requirement: for an antiunitary T , an energy
obeying T K T −1 = −K would send every positive-energy state to a negative-energy one, producing
a spectrum unbounded below and no stable ground state. The operator that T genuinely reverses is
the boost/grading, T H T −1 = −H (Theorem 4.1), as a boost must be — not the energy.
Novelty label: physical interpretation not established here.
4Generated by scripts/suite_text_anchors_build.py (S370a) from SUITE_Rev33.1_TEXT_part2_pp101-200.txt. The .txt file is unchanged and remains the object of record.