Historical excerpt: IDWT_10_CP_Completion.md

Imported 2026-09-08. Historical evidence only; claims are not endorsed by V2. This is a local extract, not a link to or dependency on another repository. Original full-file SHA-256: d4178a06e2aa7be1057114f6008415d5ae173d5437486ad2cf1afefbfe50e8c5.

Original source lines 21–284

§1. Spectral Flow Coefficient from the T2 Kernel

1.0 Relation to Part 9 Spectral Phase Computation

Part 9 T8 computed the spectral phase triangle integral directly on the product state $\Psi_{\rm sect} = \bigotimes_d \chi_{n_d,d}$ in coupling space $(g_{66}, g_{10,10})$ and found $\gamma = 0$, $\mathcal{F}_{6,10} = 0$ exactly — consistent with $\delta_{\rm CP}^{(\rm tree)} = 0$, because real normalized product-state modes have zero spectral connection. Part 9 identified the open item as: compute $\delta_{\rm CP}$ from the PMNS holonomy on the entangled lepton coupling state (the U(3) determinant bundle), not the product state.

The present approach addresses this by parameterizing the entanglement via the reactor angle $\theta_{13}$. As $\theta_{13}$ increases from 0, the T2 cross-sector kernel couples $\nu_1$ to the $\tau$ and $\mu$ sectors, creating genuine sector-sector entanglement — this is precisely the non-product structure that Part 9 identified as necessary. The charged-lepton Dirac operators $D_{\mathbb{CP}^3}(\theta_{13})$ and $D_{\mathbb{CP}^5}(\theta_{13})$ are perturbed by this coupling, and their relative spectral flow captures the topological content of the determinant bundle $\det U_{\rm PMNS}(\theta_{13})$. This is a different route to the same integral: instead of evaluating $\int_A \mathcal{F}_\mathcal{M}$ on the coupling triangle in $(g_{66}, g_{10,10})$ space, we compute the APS spectral flow along the physical path in $\theta_{13}$. The two are related by the APS theorem for families: $c_1(\det L) = {\rm sf}(D; {\rm path})$.

1.1 Setup

The cross-sector kernel (T2) is the unique $\mathrm{U}(d)\times\mathrm{U}(d')$-invariant degree-4 interaction:

$$K(\xi_d, \xi_{d'}) = g_{dd'}\,(\xi_d \cdot \xi_{d'})^2.$$

The physical PMNS angles are generated by the $d=5 \leftrightarrow d=6$ and $d=5 \leftrightarrow d=10$ components of this kernel, with coupling amplitudes $v_5 v_6$ and $v_5 v_{10}$ respectively. Because $v_6 = v_{10} = 1/2$ (T9c), the $d=6$ and $d=10$ sectors enter symmetrically at tree level — this is the μ-τ symmetry of the μ–τ symmetric limit.

The reactor angle $\theta_{13}$ parameterizes the displacement from the μ–τ symmetric limit. From T6 Step 4:

$$\sin^2\theta_{13} = g_{55} \cdot \delta_{23} \cdot \ln\frac{S(n_\tau, 10)}{S(n_\mu, 6)}$$

where $\delta_{23} = \sin^2\theta_{23} - 1/2 > 0$. This is a cross-sector log-ratio, entering through the $d=5 \leftrightarrow d=10$ coupling amplitude $v_5 v_{10}\sin\theta_{13}$. As $\theta_{13}$ increases from 0, the kernel component coupling $\nu_1$ to the τ sector switches on proportionally to $\sin\theta_{13}$.

The perturbed sector Dirac operator. The kernel term for the $d=5 \leftrightarrow d=10$ coupling, projected onto the leading occupied modes ($n_{\nu_1} = 10$, $n_\tau = 23$), acts as a perturbation of the $d=10$ Dirac operator:

$$D_{\mathbb{CP}^5}(\theta_{13}) = D_{\mathbb{CP}^5}^{(0)} + \delta D_{\mathbb{CP}^5}(\theta_{13})$$

where the perturbation $\delta D_{\mathbb{CP}^5}$ arises from the $\ell=0$ component of the kernel (T2, §1/d decomposition) acting at the $d=5 \leftrightarrow d=10$ sector boundary. The coupling amplitude at angle $\theta_{13}$ is:

$$v_{5,10}(\theta_{13}) = v_5 v_{10} \sin\theta_{13} = \tfrac{1}{2}\sqrt{g_{55}}\,\sin\theta_{13}.$$

For the $d=6$ sector, an identical perturbation applies with $d=6$ in place of $d=10$ and the complementary amplitude $\cos\theta_{13}$ (unitarity of the $U_{e3}$, $U_{e2}$ column). The physical family is therefore parameterized by $\theta_{13} \in [0, \theta_{13}^{\mathrm{phys}}]$.

1.2 The Determinant Line Bundle in the Sector Mode Basis

The PMNS matrix element $U_{e3} = \sin\theta_{13}\,e^{-i\delta}$ is the overlap between the electron flavor state $|\nu_e\rangle$ — the electron being the $d=6$ $\mathbb{CP}^3$ sector excitation of $\Psi_\infty$ at mode index $n=13$, a genuine 6D object — and the third mass eigenstate $|\nu_3\rangle$. In the IDWT mode basis:

$$U_{e3} = \sum_{d \in \{6,10\}} \langle \chi_{n_e, d} \,|\, K_{d,5}(\theta_{13}) \,|\, \chi_{n_{\nu_1}, 5} \rangle$$

where $K_{d,5}(\theta_{13})$ is the kernel matrix element between sector $d$ and the neutrino sector at coupling parameter $\theta_{13}$. The phase of $U_{e3}$ is the phase of this overlap integral as a function of $\theta_{13}$.

The determinant line bundle of $U_{\rm PMNS}$ is the complex line bundle $L^{\det} \to [0, \theta_{13}^{\mathrm{phys}}]$ whose fiber at $\theta_{13}$ is the complex number $\det U_{\mathrm{PMNS}}(\theta_{13})$. The first Chern class $c_1(L^{\det})$ is the total phase winding of $\det U_{\mathrm{PMNS}}$ as $\theta_{13}$ traverses the physical interval — this is the CP phase.

1.3 The Spectral Flow Coefficient from the IDWT Mode Structure

Claim: For the one-parameter family $D_{\mathbb{CP}^n}(\theta_{13})$ defined by the IDWT T2 kernel perturbation, the spectral flow per unit $\theta_{13}$ (in units of $\pi$) equals $c_1(\mathbb{CP}^n)$.

Derivation. The classical mode function space for the d-sector is $L^2(\mathbb{CP}^n, S_d)$, where $S_d$ is the spinor bundle. The perturbation $\delta D_{\mathbb{CP}^n}$ at first order in $\sin\theta_{13}$ is:

$$\delta D_{\mathbb{CP}^n} = v_5 v_{d'} \sin\theta_{13} \cdot \Pi_n$$

where $v_{d'} = v_{10}$ for the $d=10$ perturbation and $v_{d'} = v_6$ for the $d=6$ perturbation, $\Pi_n$ is the projection onto the $n_\tau$ (or $n_\mu$) mode in sector $d'$, entering through the $\ell=0$ kernel component (coefficient $1/d$ from T2). This is a rank-1 perturbation.

For a rank-1 perturbation of a Dirac operator on $\mathbb{CP}^n$, the spectral flow is determined by the index of the unperturbed operator twisted by the line bundle of the perturbation. The perturbation $\Pi_n$ is a section of the line bundle $O(n-1)$ on $\mathbb{CP}^n$ (since the mode $\chi_{n,d}$ is a degree-$(n-1)$ section of the tautological bundle). Open (🔶): the identification of $\Pi_n$ as a section of $O(n-1)$ relies on the assertion that $\chi_{n,d}$ is a degree-$(n-1)$ holomorphic section; the explicit mode-bundle correspondence, representation map, and basis construction from the IDWT mode functions have not been worked out. The following index calculation assumes this identification. By the Atiyah-Singer index theorem for the twisted operator:

$$\mathrm{ind}(D_{\mathbb{CP}^n} \otimes \mathcal{O}(n-1)) = \chi(\mathbb{CP}^n, \mathcal{O}(n-1)) = \binom{n-1+n}{n} = \binom{2n-1}{n}$$

The spectral flow per unit $\theta_{13}$ is carried entirely by the zero modes, not by a bulk crossing density. The $\mathcal{O}(k)$ mode tower on $\mathbb{CP}^n$ — the sector's own mode functions, labelled by the $\mathrm{SU}(n{+}1)$ isometry — has the exact spectrum (from the isometry Casimir; files/idwt.py STEP 132)

$$\lambda_l^2 = l\,(l+k+n), \qquad l = 0, 1, 2, \ldots$$

with zero modes exactly at the ground level $l=0$, a level of dimension $\binom{n+k}{n}$, and a strictly positive gap $\lambda_1^2 = k+n+1$ to the rest of the tower. For charge $k=1$ this gives exactly $n+1$ zero modes, gap-protected by $\lambda_1^2 = n+2 > 0$. (In bundle language the $l=0$ level is $H^0(\mathbb{CP}^n, \mathcal{O}(k))$; the count $n+1$ at $k=1$ equals $c_1(\mathbb{CP}^n)$. The canonical bundle $K_{\mathbb{CP}^n} = O(-n-1)$ has no global sections for $n \geq 1$ and is not the relevant object here.)

The flow coefficient follows by counting the sector's own modes, with no density estimate. Each charge-1 zero mode advances its phase by the base loop once per cycle — unit charge is the mode's grading under the loop $\mathrm{U}(1)$, the overall phase of the $n+1$ sector coordinates — so the combined phase of the $n+1$ zero modes, the object identified with the CP phase (§1.2), advances $n+1$ times the base. The gap $\lambda_1^2 = n+2$ keeps the zero-mode count stable for any perturbation amplitude below the gap (the coupling $v_0\sin\theta_{13}$, with $\|\Pi_n\|^2 = 1$ in the IDWT weighted $L^2$ norm, is far sub-gap), so the gapped tower contributes zero net flow and the rate is the exact zero-mode count:

$$\frac{d}{d\theta_{13}} \mathrm{sf}(D_{\mathbb{CP}^n}; \theta_{13}) = v_0 \times \dim H^0(\mathbb{CP}^n, O(1)) = v_0 \times (n+1) = v_0 \times c_1(\mathbb{CP}^n).$$

Integrating from 0 to $\theta_{13}$ and accounting for the periodicity of the Dirac spectrum on $\mathbb{CP}^n$ (period $\pi$ in coupling angle, not $2\pi$ — see §3 for the derivation of this factor), the total spectral flow is:

$$\mathrm{sf}(D_{\mathbb{CP}^n}; 0 \to \theta_{13}) = c_1(\mathbb{CP}^n) \times \frac{\theta_{13}}{\pi}.$$

For $\mathbb{CP}^3$ ($d=6$): $c_1(\mathbb{CP}^3) = 4$, so $\mathrm{sf}_6 = 4\theta_{13}/\pi$. For $\mathbb{CP}^5$ ($d=10$): $c_1(\mathbb{CP}^5) = 6$, so $\mathrm{sf}_{10} = 6\theta_{13}/\pi$.

Relative spectral flow:

$$\Delta\eta = \mathrm{sf}_{10} - \mathrm{sf}_{6} = (6-4)\frac{\theta_{13}}{\pi} = \frac{2\theta_{13}}{\pi}.$$

The $\Delta\eta$ formula follows from the Chern class difference $\Delta c_1 = -2$, expressed via the IDWT mode structure of the T2 kernel. The coefficient 2 is $c_1(\mathbb{CP}^5) - c_1(\mathbb{CP}^3) = 6 - 4 = 2$, where $c_1(\mathbb{CP}^n) = n+1$ gives $c_1(\mathbb{CP}^5) = 6$ and $c_1(\mathbb{CP}^3) = 4$ — entirely determined by the sector chain endpoint (T15a) and the step in the Hopf chain (T3 Rule B). ✅

---

§2. Sign of $U_{e3}$ at the μ–τ Symmetric Limit from the T6 Formula

2.1 The μ–τ Symmetric Limit Matrix and the $U_{e3}$ Column

At the μ–τ symmetric limit ($\theta_{13} = 0$), the PMNS matrix has the exact form:

$$U_{\mathrm{sym}} = \begin{pmatrix} \sqrt{2/3} & \sqrt{1/3} & 0 \\ -\sqrt{1/6} & \sqrt{1/3} & \sqrt{1/2} \\ \sqrt{1/6} & -\sqrt{1/3} & \sqrt{1/2} \end{pmatrix}$$

with $U_{e3} = 0$. The physical PMNS is obtained from $U_{\mathrm{sym}}$ by a perturbation that rotates in the $(1,3)$ plane by the reactor angle $\theta_{13}$ and introduces the CP phase $\delta$.

In the standard PDG parameterization (phase convention: $\delta$ enters via $e^{-i\delta}$ in $U_{e3}$), the perturbation away from the μ–τ symmetric limit gives:

$$U_{e3} = \sin\theta_{13} \cdot e^{-i\delta}.$$

As $\theta_{13} \to 0^+$, the $U_{e3}$ element vanishes. The sign of the real perturbation $\delta U_{e3}^{(1)} = \partial U_{e3}/\partial\theta_{13}|_{0^+}$ determines whether $\delta \to 0$ or $\delta \to \pi$ in this limit.

2.2 The Sign from the T6 Formula

From T6 Step 4, the $\theta_{13}$ formula is:

$$\sin^2\theta_{13} = g_{55} \cdot \delta_{23} \cdot \ln\frac{S(n_\tau,10)}{S(n_\mu,6)}$$

where:

• $g_{55} = 96/g_{22} = 0.13287 > 0$
• $\delta_{23} = \sin^2\theta_{23} - 1/2 > 0$ (since $\theta_{23} > \pi/4$ by T6; this follows from $S(n_\tau,10) > S(n_\mu,6)$)
• $\ln(S(n_\tau,10)/S(n_\mu,6)) = \ln(S(23,10)/S(35,6)) > 0$

Therefore $\sin^2\theta_{13} > 0$, which is tautological. What matters is the sign of $U_{e3}$ itself, not $\sin^2\theta_{13}$.

The T6 derivation proceeds via the rank-1 coupling structure (T2 Step 1 of T6):

$$W[\alpha, i] \propto \sqrt{S(n_\alpha, d_\alpha)} \cdot \sqrt{S(n_{\nu_i}, 5)},$$

where $\alpha$ labels the charged lepton flavor and $i$ labels the neutrino mass eigenstate. The PMNS matrix in the basis of mass eigenstates is:

$$U_{\alpha i} = W[\alpha, i] / \left(\sum_j |W[\alpha, j]|^2\right)^{1/2}.$$

The τ-$\nu_3$ dominance at $d=10$ and the sign of $U_{e3}$.

The key structural fact (T5, T6) is that $|n_\tau - n_{\nu_3}| = 1$ — the τ and $\nu_3$ mode indices are adjacent. The coupling $W[\tau, 3] = \sqrt{S(n_\tau,10)} \cdot \sqrt{S(n_{\nu_3},5)}$ is therefore the dominant off-diagonal kernel element. By the column unitarity of $U_{\rm PMNS}$, the dominant element $U_{\tau 3}$ being large and positive (both S-values positive, both square roots positive) forces the remaining elements in the third column to be small. In particular, $U_{e3}$ is the element in the $(e, \nu_3)$ position.

The sign is fixed by the orientation of the $d=6 \leftrightarrow d=5 \leftrightarrow d=10$ triangle in the kernel sum. Explicitly, from the rank-1 T2 kernel with the IDWT mode ordering $n_{\nu_1} < n_{\nu_2} < n_{\nu_3}$ and the $d=5$ sector structure ($S^5$, no chirality — the neutrino sector is vector-like):

The coupling matrix element $W[e, 3]$ involves the $d=6$ (electron sector, $\mathbb{CP}^3$) coupling to the third neutrino mass eigenstate ($n_{\nu_3} = 22$ in the bare sector, $22+\text{correction}$). The electron mode $n_e = 13$ sits in $d=6$, and the mixing proceeds through the cross-sector kernel which enters with a relative minus sign for the $(1,3)$ element when the μ–τ symmetric-limit basis rotation is applied. This arises because the generation law for $n_{\nu_3}$ proceeds via inclusion-exclusion (Part 2 §9c, T11b): $n_{\nu_3} = n_{\nu_1} + n_{\nu_2} - n_u$, which is the one neutrino mode index not derived from a primary Pascal evaluation. The inclusion-exclusion combination enters the rank-1 coupling matrix with a relative sign, forcing $U_{e3}$ to approach zero from negative values as $\theta_{13} \to 0^+$.

Formal statement. The (1,3) matrix element of the T6 coupling matrix W, evaluated in the μ–τ symmetric limit with the $d=5$ inclusion-exclusion mode structure, has a definite sign:

$$\lim_{\theta_{13}\to 0^+} \frac{\partial}{\partial\theta_{13}} U_{e3}^{(real)}(\theta_{13}) < 0.$$

This relative sign is explicit. In the oscillator realization (level $n\leftrightarrow$ Hermite degree $k=n-1$), the additive tower edges $n_c=n_a+n_b$ are degree-1 condensate matrix elements $\langle\chi_{n_c}|\xi|\chi_{n_a}\chi_{n_b}\rangle$ with positive top-of-band coefficient ($+\tfrac12$), whereas $n_{\nu_3}=n_{\nu_1}+n_{\nu_2}-n_u$ is the one edge with a net degree-lowering and so carries the inclusion-exclusion (cumulant) minus sign of the shared seed $n_u$. The electron, being seed-built ($n_e=n_c-n_u-n_s$), overlaps the subtracted leg, so $W[e,3]$ inherits the minus relative to the positive additive edges (idwt.py STEP 41). The overall branch ($\delta\to\pi$ rather than $\delta\to 0$) remains fixed by the PDG phase convention as noted in §4.

A negative real perturbation to $U_{e3}$, combined with the PDG convention $U_{e3} = \sin\theta_{13}\,e^{-i\delta}$, gives:

$$\mathrm{arg}(U_{e3}) \to \pi \quad \text{as } \theta_{13} \to 0^+$$

i.e., $\delta = \pi$ in the μ–τ symmetric limit. ✅

2.3 Cross-check via Neutrino Mass Ordering

The inclusion-exclusion sign can be cross-checked against the mass ordering. The correction $\delta_{\nu_3} = +1/35$ (Part 2 §9d) is positive — it increases $m_{\nu_3}$. A positive mass correction to the heaviest neutrino is consistent with the normal ordering ($m_{\nu_1} < m_{\nu_2} < m_{\nu_3}$), which IDWT uniquely predicts (T11b, T11d). The same inclusion-exclusion structure that gives $\delta_{\nu_3} > 0$ gives $\partial U_{e3}/\partial\theta_{13}|_{0^+} < 0$ — both signs are driven by the relative orientation of $n_{\nu_3}$ in the T6 coupling matrix, and they are mutually consistent. ✅

---

§3. Factor-of-2 Normalization from Mode Function Inner Products

3.1 The Question

The spectral flow formula $\mathrm{sf}(D_{\mathbb{CP}^n}; 0\to\theta_{13}) = c_1(\mathbb{CP}^n) \times \theta_{13}/\pi$ uses $\theta_{13}$ in units where a full rotation of the coupling matrix is $2\pi$, but the Dirac spectral period is $\pi$. The question: is the natural parameter in the T2 kernel perturbation the full angle $\theta_{13}$, or $\theta_{13}/2$?

This matters because it determines whether the spectral phase correction to $\delta_{CP}$ is $2\theta_{13}$ or $\theta_{13}$:

$$\phi_{\mathrm{spec}}^{\mathrm{det}} = -\pi \times \Delta\eta = -\pi \times \frac{2\theta_{13}}{\pi} = -2\theta_{13}$$

versus $-\theta_{13}$ if the spectral flow parameter were $\theta_{13}/2$.

3.2 The IDWT Mode Function Normalization

In the IDWT weighted L² norm (Part 8 §9):

$$\|\Psi\|_w^2 = \sum_{d \in D} \sum_{n \geq 1} S(n,d) |c_{n,d}|^2,$$

the mode function inner product relevant to the spectral flow computation is:

$$\langle \chi_{n_{\nu_1}, 5} \,|\, \chi_{n_\tau, 10} \rangle_w = \delta_{n_{ν_1}, n_\tau} \cdot S(n_{\nu_1}, 5)^{1/2} \cdot S(n_\tau, 10)^{1/2} \times [\text{sector overlap}].$$

For the cross-sector kernel, the sector overlap integral is determined by the $\ell=0$ component of $(\xi_d \cdot \xi_{d'})^2$:

$$\int_{\Xi_d \times \Xi_{d'}} (\xi_d \cdot \xi_{d'})^2 \cdot |\chi_{n_d, d}|^2 \cdot |\chi_{n_{d'}, d'}|^2 \, d\mu_d \, d\mu_{d'} = \frac{1}{d'} \int |\xi_d|^2 \, d\mu_d \int |\xi_{d'}|^2 \, d\mu_{d'}$$

using the $\ell=0$ coefficient $1/d$ from T2 (the same coefficient that generates mass scales in T9).

3.3 The Coupling Angle in the T2 Kernel

The T2 kernel coupling amplitude between the $\nu_1$ mode and the τ mode, as a function of $\theta_{13}$, is:

$$g_{5,10}(\theta_{13}) = v_5 \cdot v_{10} \cdot \sin\theta_{13} = \frac{1}{2}\sqrt{g_{55}} \cdot \sin\theta_{13}.$$

Under the APS theorem, the spectral flow is sensitive to the half-angle of the rotation in the mode function space, not the rotation angle itself — this is the standard spectral phase doubling. Specifically, a Dirac operator $D(\theta) = D_0 + \sin\theta \cdot V$ with weighted L² inner product $(\cdot,\cdot)_w$ returns to $D_0$ at $\theta = \pi$ (not $2\pi$), because $\sin\pi = 0$. The Dirac spectral period is therefore $\pi$ in the coupling angle $\theta$.

The identification: The coupling angle $\theta$ in $D(\theta) = D_{\mathbb{CP}^n}^{(0)} + \sin\theta \cdot \Pi_n$ is exactly the PMNS reactor angle $\theta_{13}$ — not $\theta_{13}/2$. This follows from the T2 kernel structure: the kernel perturbs $D_{\mathbb{CP}^n}$ with amplitude $\sin\theta_{13}$, and the spectral flow counts crossings as $\theta_{13}$ increases from 0 to $\pi$. The physical value $\theta_{13}^{\mathrm{phys}} \approx 8.55^\circ \ll \pi/2$, so the coupling parameter is linearly in the small-angle regime.

Consequence: The spectral period is $\pi$ in $\theta_{13}$, giving a spectral phase per period of $c_1(\mathbb{CP}^n) \times 1 = c_1(\mathbb{CP}^n)$. The spectral phase for a path from 0 to $\theta_{13}$ is:

$$\phi_{\mathrm{spec}}^{(n)} = -c_1(\mathbb{CP}^n) \times \frac{\theta_{13}}{\pi} \times \pi = -c_1(\mathbb{CP}^n) \times \theta_{13},$$

and the relative phase between $d=10$ and $d=6$:

$$\phi_{\mathrm{spec}}^{\mathrm{det}} = -[c_1(\mathbb{CP}^5) - c_1(\mathbb{CP}^3)] \times \theta_{13} = -(6-4)\theta_{13} = -2\theta_{13}.$$

The factor of 2 is $c_1(\mathbb{CP}^5) - c_1(\mathbb{CP}^3) = \Delta c_1$ reversed in sign — exactly the topological integer identified in T8. ✅

3.4 Mode Function Inner Product Consistency Check

Verification that the mode norm is 1 in the relevant computation. The IDWT inner product weight $S(n,d)$ cancels in the spectral flow computation because the spectral flow counts zero-mode crossings of the Dirac operator, which are normalization-independent. Explicitly: a zero crossing occurs when $D_{\mathbb{CP}^n}(\theta_{13})$ has an eigenvalue that passes through zero. This is a topological count, invariant under rescaling of the mode functions. Therefore the factor $\|\Pi_n\|^2 = 1$ used in §1.3 is consistent with the IDWT weighted norm — both are equal to 1 because zero crossings are topological. ✅

---

§4. T8 Closed: The CP Phase Theorem

Theorem T8 (CP Phase from Spectral Flow). 🔶

The leptonic CP-violating phase is:

$$\boxed{\delta_{CP} = \pi + (N_c - 1)\,\theta_{13} = \pi + 2\theta_{13}}$$

where $\theta_{13}$ is the reactor angle, $N_c = \chi(\mathbb{CP}^2) = 3$ is the number of quark colors, and the formula holds to leading order in $\theta_{13}$ in the APS spectral flow of the one-parameter family $D_{\mathbb{CP}^n}(\theta_{13})$.

Basis-dependence note. The derivation uses the PDG standard parameterization with phase convention $e^{-i\delta}$ in $U_{e3}$, the μ–τ reflection symmetry as the zero-phase reference, and the specific orientation $\partial U_{e3}^{(\mathrm{real})}/\partial\theta_{13} < 0$ from the inclusion-exclusion sign of $n_{\nu_3}$. These choices select one of the two physically equivalent phase branches ($\delta$ and $2\pi-\delta$). The IDWT prediction $\delta = \pi + 2\theta_{13}$ is structural in the sense that the combination $\pi + 2\theta_{13}$ arises from $\Delta c_1 = -2$ and the μ–τ symmetric-limit boundary; however, the overall sign of the $2\theta_{13}$ shift (which branch) depends on the sign convention for the perturbation in §2. A basis-independent statement: $|\delta - \pi| = 2|\theta_{13}|$; the prediction is which branch IDWT selects.

Proof summary. The CP phase is the argument of the PMNS matrix element $U_{e3} = \sin\theta_{13}\,e^{-i\delta}$. It has two contributions:

(i) Boundary condition at the μ–τ symmetric limit. At $\theta_{13} = 0$, $U_{e3} = 0$. As $\theta_{13} \to 0^+$, the T6 coupling matrix gives $\partial U_{e3}^{(\mathrm{real})}/\partial\theta_{13} < 0$ from the inclusion-exclusion sign of $n_{\nu_3}$ (§2). This places $\delta = \pi$ at the μ–τ symmetric-limit boundary.

(ii) Spectral flow correction. As $\theta_{13}$ increases from 0 to its physical value, the charged-lepton sector Dirac operators $D_{\mathbb{CP}^3}$ and $D_{\mathbb{CP}^5}$ are perturbed by the T2 cross-sector kernel at amplitude $\sin\theta_{13}$ (§1). The relative spectral flow between $d=10$ and $d=6$ is $\Delta\eta = (c_1(\mathbb{CP}^5) - c_1(\mathbb{CP}^3)) \times \theta_{13}/\pi = 2\theta_{13}/\pi$ (§1.3). The APS theorem connects this to a spectral phase $\phi_{\mathrm{spec}} = -2\theta_{13}$ on the determinant line bundle of $U_{\mathrm{PMNS}}$ (§3). Since $\phi_{\mathrm{spec}} = -2\theta_{13} < 0$ and $\delta = \pi$ in the μ–τ symmetric limit, the physical phase is:

$$\delta_{CP} = \pi - (-2\theta_{13}) = \pi + 2\theta_{13}.$$

The $d=5$ sector ($S^5$) contributes zero spectral flow because $S^5$ has no complex structure and hence no Chern classes — there is no holomorphic line bundle over $S^5$ to generate spectral flow. The $d=3$ and $d=4$ quark sectors enter only through $N_c$ via T15. □

Numerical prediction:

$$\delta_{CP} = 180° + 2 \times 8.555° = \mathbf{197.11°}$$

$$\sin\delta_{CP} = -\sin(17.11°) = -0.2942$$

$$J_{\mathrm{PMNS}} = J_{\mathrm{max}} \times \sin\delta_{CP} = 0.033351 \times (-0.2942) = -0.009813$$

Status: T8 remains 🔶. The formula $\delta_{CP} = \pi + 2\theta_{13} = 197.11^\circ$ is a concrete, testable prediction with strong topological motivation ($\Delta c_1 = -2$, $\delta = \pi$ in the μ–τ symmetric limit). Content gaps and their state: (i) the spectral flow coefficient is now derived by exact zero-mode counting — $n+1$ charge-1 ground modes, each advancing its phase once per base loop, gap-protected by the exact mode tower $\lambda_l^2 = l(l+n+1)$ (§1.3; files/idwt.py STEP 132) — replacing the earlier density estimate; the remaining rigor for (i), that the T2 rank-1 perturbation implements the phase loop, is the formalization item (vi); (ii) the sign $\partial U_{e3}/\partial\theta_{13}|_{0^+} < 0$ follows from the inclusion-exclusion (cumulant) structure of $n_{\nu_3}$ — now made explicit as the relative sign via the condensate matrix-element computation (§2), with the full T6-coupling-matrix derivation including the branch convention not yet assembled at the same rigor; (iii) the equivalence between spectral flow along the $\theta_{13}$ path and the Part 9 spectral-phase coupling-space integral has not been proved formally. Four additional rigorous-mathematics gaps also remain: (iv) Fredholm continuity of the one-parameter operator family $D_{\mathbb{CP}^n}(\theta_{13})$ is not proved; (v) the spectral-flow definition used is heuristic (eigenvalue counting) rather than the rigorous APS definition (η function regularization); (vi) the family index construction (determinant line bundle with connection) is not formalized; (vii) regularity conditions on the mode functions and domain under the $\theta_{13}$ perturbation are not verified. Closing (i)–(iii) would move T8 to 🔵; (iv)–(vii) are needed for a ✅ proof.

4.1 Quark sector: the leading CKM CP phase vanishes by sector structure

The same spectral-flow mechanism applied to the quark sector gives a qualitatively different result, and the difference is geometric. The leptonic phase is nonzero because the charged leptons span two distinct Kähler sectors — the electron and muon in $\mathbb{CP}^3$ ($d=6$, $c_1=4$) and the tau in $\mathbb{CP}^5$ ($d=10$, $c_1=6$) — so different generations accumulate different spectral flow, and the relative phase $\Delta c_1 = 2$ survives. The quark generations have no such spread. All three up-type quarks (u, c, t) are $d=4$ sector excitations on $\mathbb{CP}^2$, and all three down-type quarks (d, s, b) are $d=3$ sector excitations on $S^3$.

Two facts then fix the leading-order phase. First, $S^3$ is a real sphere: it carries no complex structure and no holomorphic line bundle, so its spectral flow is zero (the same reason §4 gives for $S^5$). Second, the $\mathbb{CP}^2$ spectral flow is common to all three up-type generations. Consequently the spectral-flow phase of every CKM element $V_{u_i \to d_j}$ is the same up-sector-minus-down-sector value,

$$\arg V_{u_i\to d_j}^{(\mathrm{spec})} \;=\; c_1(\mathbb{CP}^2) - c_1(S^3) \;=\; 3 - 0,$$

independent of the generation indices i and j. A phase that is constant across the matrix is a global rephasing of the up-type relative to the down-type fields; it cancels in every rephasing-invariant. The Jarlskog invariant is therefore zero at leading order:

$$\boxed{J_{\mathrm{CKM}}^{(\mathrm{leading})} = 0.}$$

This is the structural origin of the tree-level result $\rho = \eta = 0$ (Part 5): the CKM phase is not merely numerically small at tree level — the holonomy mechanism that generates the leptonic phase is switched off for quarks, because the quark generations of each type occupy a single sector. The result requires only the sector assignments and the rephasing identity, not the spectral-flow coefficient of §1.3, so it is insensitive to the gaps (i)–(vii) that keep T8 at 🔶.

The observed quark CP violation, $J_{\mathrm{CKM}} \approx 3\times10^{-5}$, is therefore necessarily a subleading effect, coming from the holomorphic-versus-real mismatch in the kernel overlap between $\mathbb{CP}^2$ up-modes and $S^3$ down-modes. Writing the explicit $\mathbb{CP}^2$ holomorphic $\mathcal{O}(k)$ up-sections and real $S^3$ down-harmonics in shared Hopf coordinates (2026-06-24; files/idwt.py STEP 6) yields three derived results:

(i) Charge selection (derived): The kernel $(\xi_3\cdot\xi_4)^2$ carries Hopf charge $\in\{-2,0,+2\}$; a chiral $\mathcal{O}(k)$ up-mode overlaps a real $S^3$ down-mode only if $k\le 2$. Only $u$ ($k=2$) couples; $c,t$ ($k=19,71$) give exactly zero. The quark CP phase sits on $V_{ub}$ alone — derived from first principles.

(ii) Factorization theorem (derived): All clean holomorphic-versus-real single-frequency overlaps are real ($\arg\in\{0,\pi\}$) and factorize, giving $J=0$ from those alone. The sole non-factorizable source is the $b$-quark beat at $n=16$: $Y_b=Y_{L1}+e^{i\chi}Y_{L2}$ with a continuous internal phase $\chi$.

(iii) Real-beat theorem (derived): The bottom-quark beat is sustained by the framework's real Jacobi coupling $\sqrt{b_{16}b_{17}}$ and real quartic density term (Part 2 §8), which phase-lock $\chi\in\{0,\pi\}\Rightarrow J=0$ from real dynamics alone. A nonzero $J$ requires a complex Fubini–Study holonomy $\phi_{\rm link}$ on the $16\leftrightarrow17$ link: $\chi=\pi-\phi_{\rm link}$, $\sin\phi_{ub}=\sin\phi_{\rm link}$.

Magnitude and phase. The link phase $\phi_{\rm link}$ is a spectral-flow holonomy of the same determinant-line-bundle mechanism that fixes the lepton phase (§1.3): $\chi$ is the relative phase between the two Dirac eigenmodes $n=16,17$, so its rate is the anticanonical degree $c_1(\mathbb{CP}^n)=n+1$, not the matter-section degree $\mathcal{O}(n-1)$. For the up-sector that rate is $c_1(\mathbb{CP}^2)=N_c=3$ — the same value carried by the leading-order phase $c_1(\mathbb{CP}^2)-c_1(S^3)=3-0$ above — so a single-section reading with degree $k_u=2$ (which would give $45^\circ$) is excluded. The $16\leftrightarrow17$ link advances one node of the $k_0=n_s^2=16$ beat, giving $\phi_{\rm link}=c_1(\mathbb{CP}^2)\cdot2\pi/k_0=67.5^\circ$. This is the predicted unitarity-triangle angle $\gamma$; the PDG value is $\gamma\approx65.7^\circ$, a $+1.8^\circ$ ($\sim3\%$) discrepancy. The one-node step $f=1/k_0$ follows from the lock structure: the real-beat theorem locks $\chi$ at the beat nodes, the beat carries $k_0=16$ nodes per cycle (the established beat index, Appendix E §49), and the holonomy transport is homogeneous along the cycle (constant-curvature connection, so the phase is proportional to the arc fraction) — hence the phase accrued between adjacent locks, the quantity entering $\chi = \pi - \phi_{\rm link}$, is $1/k_0$ of the full loop. Robustness: varying $f\in\{1/15,1/16,1/17\}$ moves $\gamma$ across $\{72^\circ,67.5^\circ,63.5^\circ\}$ and changes $J$ by at most $\pm3\%$. The Jarlskog itself is $J=|V_{us}V_{cb}V_{ub}V_{cs}|\sin\gamma$; $|V_{us}|$ and $|V_{cb}|$ are IDWT-derived, and the $|V_{ub}|$ magnitude — the apex — is candidate-derived (🔶, files/idwt.py STEP 136) by the same conditional-Born counting law applied to the charge-selected hop: $R_{ub}^2 = n_u/S(n_s,3) = 3/20$, the pairing selected uniquely by data. This gives $|V_{ub}| = 3.67\times10^{-3}$ (PDG $(3.82\pm0.24)\times10^{-3}$, $-0.6\sigma$) and an all-IDWT Jarlskog $J = A^2\lambda^6\bar\eta = 3.06\times10^{-5}$ against the PDG global fit $J = (3.18\pm0.15)\times10^{-5}$ ($-0.8\sigma$); with the measured $|V_{ub}|$ instead, $J = 3.16\times10^{-5}$ ($-0.1\sigma$).

Status. The leading-order vanishing $J_{\rm CKM}^{(\rm leading)}=0$ is ✅. The charge-selection ($V_{ub}$ localization), the factorization and real-beat theorems, and the holonomy-degree identification — the link rate is $c_1(\mathbb{CP}^2)=3$, not the matter-section degree, with the $45^\circ$ reading excluded — are derived sub-results (sub-🔵 within the overall 🔶). The hierarchy $J_{\rm CKM}\ll J_{\rm PMNS}$ is explained structurally. The predicted phase $\gamma=c_1(\mathbb{CP}^2)\cdot2\pi/k_0=67.5^\circ$ agrees with PDG $65.7^\circ$ to $\sim3\%$, but its magnitude stays 🔶: the one-node step $f=1/k_0$ is derived from the node-lock structure (adjacent-lock transport, above), the $|V_{ub}|$ apex magnitude is not yet derived, and the result cannot exceed its parent T8 (🔶), on which the spectral-flow mechanism rests. T8 itself is unchanged.

---