Knot-Theoretic Bootstrap

provisional

Overview

This derivation explores a topological route to the bootstrap fixed-point conjectures (Bootstrap Conjectures 7.1–7.2) and the continuous-discrete duality (Continuous-Discrete Duality Conjecture 4.1).

The core observation. The framework’s minimal observers are $U(1)$ phase loops (Loop Closure). In a spatial manifold, loops can knot and link. Two observers connected by a relational invariant (Type III interaction) are topologically linked — their loops are intertwined in a way that cannot be undone by smooth deformation. Three mathematical structures already present in the framework have direct counterparts in knot theory and topological quantum field theory:

1. Wilson loops (Electromagnetism Proposition 2.2) are the natural observables of Chern-Simons theory.
2. The Chern-Simons 3-form already appears in the framework’s strong CP resolution (Strong CP Step 3c).
3. The Cayley-Dickson gauge chain $U(1) \to SU(2) \to SU(3)$ (Bootstrap Division Algebras) maps to a specific finite sequence of Chern-Simons gauge theories, terminated at the octonions by the zero-divisor obstruction.

The program. If the observer network is a framed link in a 3-manifold, then:

• The geometry functor $G: \mathbf{Obs} \to \mathbf{Geom}$ is Dehn surgery: a framed link determines a 3-manifold via the Lickorish-Wallace theorem.
• The bootstrap fixed-point $\mathcal{U} \cong \mathcal{R}(\mathcal{U}, \mathcal{U})$ becomes self-consistent surgery: the link that produces the manifold (via surgery) is the same link that the manifold contains.
• The coherence measure may be identified with (or derived from) the Chern-Simons partition function, which computes knot and link invariants as path integrals over gauge fields.

Why bidirectionality helps. The continuous-discrete duality requires each layer to constrain the other. Surgery theory provides exactly this: the Lickorish-Wallace theorem is bidirectional (every closed oriented 3-manifold has a surgery presentation, and every surgery presentation determines a manifold), and the Kirby calculus determines when two different link presentations give the same manifold. The bidirectional constraint reduces the solution space, potentially to a unique fixed point.

Status. This is a stub. The connections identified below are structurally suggestive and grounded in existing framework derivations, but no formal proofs are offered. The goal is to establish the contact points between the framework’s existing machinery and the topological tools, and to identify the concrete steps needed to formalize the program.

Derivation

Step 1: Observer Loops as Framed Links

Preliminary disclaimer (spatial embedding). The framework’s observer loops are primarily abstract $U(1)$ phase loops on state space, not embedded curves in a spatial manifold. The spatial-embedding picture used throughout Step 1 — and downstream — presupposes the embedding prescription of Open Gap 3 (where each observer is assigned a location $\gamma_\mathcal{O} \subset M$ consistent with the self-referential geometry produced by the network’s own surgery structure). All statements in this step are conditional on that prescription being formalized.

Definition 1.1. A framed link in a 3-manifold $M$ is a finite collection of disjoint embedded circles $L = \gamma_1 \cup \cdots \cup \gamma_n$, each equipped with a framing (a choice of non-zero normal vector field along the curve, or equivalently an integer — the framing number).

Observation 1.2 (Observers are loops). Every observer $\mathcal{O} = (\Sigma, I, \mathcal{B})$ has a $U(1)$ phase loop from Loop Closure. The minimal observer has $\Sigma \cong S^1$. Under the embedding prescription of Open Gap 3, this abstract phase loop is realized as an embedded circle $\gamma_\mathcal{O} \subset M$ in the spatial manifold.

Proposition 1.3 (Framing from loop closure via the normal-bundle almost-complex structure). Given an observer loop $\gamma_\mathcal{O} \subset M$ embedded in an oriented 3-manifold, the $U(1)$ phase $\phi: \gamma_\mathcal{O} \to U(1)$ with winding number $w \in \mathbb{Z}$ (Axiom 3) induces a canonical integer framing of $\gamma_\mathcal{O}$ with framing number $w$, via a normal-bundle almost-complex structure derived from the bootstrap.

Proof.

(i) Normal bundle and its orientation. The tangent bundle $T\gamma_\mathcal{O}$ is a rank-1 subbundle of $TM|_{\gamma_\mathcal{O}}$. The normal bundle $N(\gamma_\mathcal{O})$ is the quotient $TM|_{\gamma_\mathcal{O}} / T\gamma_\mathcal{O}$, with rank 2. The orientations of $M$ (ambient) and $\gamma_\mathcal{O}$ (loop) induce a canonical orientation on $N(\gamma_\mathcal{O})$ via the short exact sequence $0 \to T\gamma_\mathcal{O} \to TM|_{\gamma_\mathcal{O}} \to N(\gamma_\mathcal{O}) \to 0$.

(ii) Canonical almost-complex structure on $N(\gamma_\mathcal{O})$. Any oriented rank-2 real vector bundle admits a canonical almost-complex structure $J: N \to N$ satisfying $J^2 = -\text{id}$ — the $90°$ rotation in the oriented plane. For $N(\gamma_\mathcal{O})$, this is well-defined fiber-by-fiber: in any oriented 2D fiber, $J$ rotates vectors by $\pi/2$ in the positive orientation.

(iii) Integer framing from phase winding. Choose any nowhere-zero section $s_0: \gamma_\mathcal{O} \to N(\gamma_\mathcal{O})$ (the “reference framing”; for a null-homologous loop in $M$, there is a canonical choice via a Seifert surface, but any nowhere-zero section suffices). The phase $\phi(\tau) = e^{2\pi i w \tau}$ for $\tau \in [0, 1]$ induces the section:

$s_\phi(\tau) = \cos(2\pi w \tau) \cdot s_0(\tau) + \sin(2\pi w \tau) \cdot J(s_0(\tau))$

This is nowhere zero (since $\|s_\phi(\tau)\| = \|s_0(\tau)\| \neq 0$) and therefore defines a framing of $\gamma_\mathcal{O}$. The framing number (the signed linking of $\gamma_\mathcal{O}$ with its push-off along $s_\phi$, relative to $s_0$) equals $w$.

By Axiom 3 (loop closure), $w \in \mathbb{Z}$. Therefore the induced framing is integer-valued. $\square$

Remark (Bootstrap-native $J$ at each level). The almost-complex structure $J$ is not imposed on the normal bundle from outside — at each bootstrap level, the framework’s algebraic structure supplies a canonical $J$:

• Level 1 ($\mathbb{C}$): The minimal observer’s phase space $\Sigma = S^1$ is itself a $U(1)$-circle; its ambient 2D real plane carries the complex-multiplication structure of $\mathbb{C}$ intrinsically. The normal-bundle $J$ pulls back from this intrinsic $\mathbb{C}$-structure via the embedding prescription (Open Gap 3). At this level, $J$ is manifestly canonical.
• Level 2 ($\mathbb{H}$): Quaternions carry three complex structures $I, J, K$ with $IJK = -1$. Picking one of the three selects a 2D complex direction. The framework’s electroweak structure already picks out a canonical embedding $\mathbb{C} \hookrightarrow \mathbb{H}$ (the hypercharge direction $i \in \text{Im}(\mathbb{C}) \subset \text{Im}(\mathbb{H})$, see Coupling Constants Theorem 1.1). This embedding supplies the canonical almost-complex structure for framings at level 2.
• Level 3 ($\mathbb{O}$): Octonions have seven independent imaginary directions, but the framing $J$ is tied to a single direction — the hypercharge direction $e \in \text{Im}(\mathbb{O})$ fixed globally by the electroweak structure at level 2 (Coupling Constants Theorem 1.1). The 2-plane $\text{span}(1, e) \subset \mathbb{O}$ inherits the canonical $J$ from the chain $\mathbb{C} \hookrightarrow \mathbb{H} \hookrightarrow \mathbb{O}$ via this hypercharge embedding. Since the level-3 bootstrap-induced connection is $SU(3) = \text{Stab}_{G_2}(e)$-valued (Observation 1.3c), parallel transport preserves this 2-plane and its $J$, and the framing is globally consistent. The naively expected $G_2$-monodromy obstruction does not arise.

Observation 1.3b (Octonionic framing obstruction at the color-force level — resolved via Observation 1.3c). At bootstrap level 3 ($\mathbb{O}$), one might worry that the almost-complex structure needed for framing lives in a $G_2/SO(4)$ moduli space: if the holonomy of the bootstrap-induced connection around $\gamma_\mathcal{O}$ had non-trivial $G_2$ monodromy, there would be no globally consistent choice of $J$ and Proposition 1.3 would fail to produce an integer framing.

Resolution. The concern does not apply. The framework’s electroweak structure globally fixes a preferred imaginary direction $e \in \text{Im}(\mathbb{O})$ (the hypercharge direction) before level-3 structure forms, reducing the residual structure group from $G_2$ to $\text{Stab}_{G_2}(e) \cong SU(3)$. The framing $J$ is tied to the single direction $e$ — not to a full quaternionic subalgebra setwise — and every element of $SU(3)$ preserves $e$ pointwise, hence preserves $\text{span}(1, e) \subset \mathbb{O}$ and the multiplication-by-$e$ map on it. No $G_2$ monodromy can rotate the 2-plane or flip $J$. See Observation 1.3c for the formal argument.

Observation 1.3c (Level-3 framing is well-defined for all physical observers). The framing construction of Proposition 1.3 produces a globally well-defined integer framing at bootstrap level 3 for all physical observers, conditional on the embedding prescription of Open Gap 3 identifying $N(\gamma_\mathcal{O})$ with $\text{span}(1, e) \subset \mathbb{O}$ consistently along the loop.

Proof.

(i) $J$ is tied to a single direction $e$, not to a setwise quaternionic subalgebra. The chain $\mathbb{C} \subset \mathbb{H} \subset \mathbb{O}$ supplying the normal-bundle $J$ at level 3 is built from the single hypercharge direction $e \in \text{Im}(\mathbb{O})$ (Coupling Constants Theorem 1.1, Color Force Proposition 3.2). The 2-plane $\text{span}(1, e) \subset \mathbb{O}$, with $J$ = multiplication-by-$e$, is the inherited almost-complex structure. No additional choice of orthogonal imaginary units (which would be required to fix a full quaternionic subalgebra $\mathbb{H} \subset \mathbb{O}$ setwise) enters the framing construction. The distinction is substantive: $\text{Stab}_{G_2}(\mathbb{H}\,\text{setwise}) \cong SO(4)$, whereas $\text{Stab}_{G_2}(e) \cong SU(3)$ — and it is $SU(3)$ that matches the framework’s color gauge group.

(ii) The level-3 connection is $SU(3)$-valued. Electroweak symmetry breaking (at level 2) selects the hypercharge direction $e \in \text{Im}(\mathbb{O})$ globally. Because level 2 precedes level 3 in the bootstrap chain, the level-3 bootstrap-induced connection is restricted to the residual stabilizer $\text{Stab}_{G_2}(e) \cong SU(3)$ (Color Force Theorem 3.1), not to the full $G_2$.

(iii) $SU(3)$ preserves $(\text{span}(1, e), J)$. By definition, every $\varphi \in \text{Stab}_{G_2}(e) = SU(3)$ satisfies $\varphi(1) = 1$ and $\varphi(e) = e$. So $\varphi$ acts as the identity on $\text{span}(1, e)$ and commutes with multiplication-by-$e$ (i.e., with $J$). Parallel transport around any loop $\gamma_\mathcal{O}$ using the $SU(3)$-valued connection therefore fixes both the 2-plane and $J$ on it. No $G_2$ monodromy exists at level 3, and no non-associative bracketing ambiguity enters: elements of $\text{span}(1, e)$ lie in the associative subalgebra $\mathbb{R}[e] \cong \mathbb{C}$, and the $SU(3)$-action on this 2-plane is trivially associative.

(iv) Confinement handles the color non-singlet case. Individual quarks (fundamental $\mathbf{3}$ of $SU(3)$) live in $\mathbb{O}/\mathbb{H}$, and their parallel transport suffers the non-associative bracketing ambiguity of Color Force Step 2. By Confinement Theorem 3.1, only color-singlet states have well-defined parallel transport, and only they appear as asymptotic observers. The $G_2$-monodromy question therefore does not arise for non-singlets: they cannot form isolated observer loops in the first place. For color singlets, the singlet projection annihilates the associator, and the transport on $\text{span}(1, e) \subset \mathbb{H}$ — the associative subalgebra — is bracketing-independent by (iii).

Combining (i)–(iv): for every physical observer $\mathcal{O}$ at level 3, the framing construction of Proposition 1.3 produces a globally well-defined integer framing. $\square$

Remark (residual dependency on Open Gap 3). The argument is conditional on the embedding prescription identifying $N(\gamma_\mathcal{O})$ with $\text{span}(1, e) \subset \mathbb{O}$ consistently along $\gamma_\mathcal{O}$ — i.e., the identification must be covariantly constant with respect to the $SU(3)$-valued bootstrap connection. This is a specific requirement placed on the embedding prescription; see Open Gap 3 for the larger program of formalizing that prescription.

Observation 1.4 (Relational invariants as linking). When two observers $\mathcal{O}_1$ and $\mathcal{O}_2$ interact via a Type III interaction, generating a relational invariant $I_{12}$, their loops become topologically linked. The linking is irreducible: it cannot be undone by smooth deformation of either loop in the ambient manifold, matching the irreducibility of the relational invariant (Relational Invariants Theorem 4.1).

Proposition 1.5 (Relational coherence = linking number). The relational coherence between two observer loops in a spatial 3-manifold equals the absolute linking number times the coherence quantum:

$\boxed{\mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2) = |\text{Lk}(\gamma_1, \gamma_2)| \cdot \hbar\omega_0}$

Proof sketch. The argument connects the linking number (a topological invariant) to the boundary-crossing count (from the holographic entropy bound) via three existing results.

(i) Linking number as intersection count. By the definition of linking number, $\text{Lk}(\gamma_1, \gamma_2)$ is the algebraic (signed) intersection number of $\gamma_2$ with any Seifert surface $\Sigma$ bounded by $\gamma_1$ ($\partial\Sigma = \gamma_1$). This is a standard result in knot theory: the count is independent of the choice of $\Sigma$.

(ii) Each intersection is a boundary crossing. In the geometric picture, the Seifert surface $\Sigma$ plays the role of the boundary of $\mathcal{O}_1$‘s coherence domain — the self/non-self divide for observer 1. Each point where $\gamma_2$ pierces $\Sigma$ is a relational invariant crossing this boundary. By Area Scaling Proposition 1.2 (boundary mediation), all shared information between $\mathcal{O}_1$ and $\mathcal{O}_2$ is mediated by such crossings. By Area Scaling Proposition 3.1, each crossing contributes one independent bit of relational information. By Bootstrap Corollary 2.3 and Coherence Conservation Corollary 5.5a, each bit equals one $\hbar\omega_0$ unit.

(iii) The stable crossing count is $|\text{Lk}|$. The physical number of crossings in the stable (ground-state) configuration equals the topological minimum $|\text{Lk}|$, by coherence conservation. Two explicit assumptions are required:

• Assumption A (Sign-independent coherence contribution). Each crossing contributes $+\hbar\omega_0$ to coherence regardless of its topological sign. Coherence is a non-negative measure (C2 of Coherence Conservation); the $\pm$ sign of a crossing is a topological label (orientation of the crossing relative to the Seifert surface), not a coherence sign. A pair of cancelling crossings therefore carries $2\hbar\omega_0$ of coherence, not zero.
• Assumption B (Isotopy preserves coherence; ground state minimizes count). Smooth isotopy of the embedded link is a coherence-preserving dynamics: ambient isotopies do not create or destroy relational invariants. The ground state of the two-loop system is the configuration that minimizes the unsigned crossing count subject to the topological linking invariant.

Given these:

• No spontaneous pair creation: creating a cancelling pair of crossings from nothing would add $2\hbar\omega_0$ of coherence (Assumption A), violating Axiom 1 (coherence cannot be created).
• No pair annihilation: destroying an existing cancelling pair would remove $2\hbar\omega_0$ of coherence (Assumption A), also violating Axiom 1.
• Ground-state minimization: by Assumption B, the stable configuration realizes the topological minimum unsigned crossing count, which is $|\text{Lk}(\gamma_1, \gamma_2)|$ (standard knot-theory result: cancelling-sign pairs can always be isotoped away to reach the minimum). At the minimum, all crossings have the same sign, so unsigned count = $|\text{signed count}| = |\text{Lk}|$.

Combining (i)–(iii): $\mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2) = |\text{Lk}| \times \hbar\omega_0$. $\square$

Consistency checks:

• Non-negativity: $|\text{Lk}| \geq 0$ and $\hbar\omega_0 > 0$, so $\mathcal{C} \geq 0$. ✓ (Required by C2 of Coherence Conservation.)
• Symmetry: $|\text{Lk}(\gamma_1, \gamma_2)| = |\text{Lk}(\gamma_2, \gamma_1)|$, so $\mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2) = \mathcal{C}(\mathcal{O}_2 : \mathcal{O}_1)$. ✓
• Vanishing for unlinked pairs: $\text{Lk} = 0 \Rightarrow \mathcal{C} = 0$. ✓ (Unlinked loops can be separated without cutting either, so they share no irreducible relational structure.)
• Entanglement identification: Entanglement Theorem 2.1 identifies relational coherence with von Neumann entanglement entropy. In topological quantum computing, the linking of anyonic worldlines is the entanglement between them — a known result that matches this identification.
• Integer quantization: $|\text{Lk}| \in \mathbb{Z}_{\geq 0}$, so $\mathcal{C} \in \mathbb{Z}_{\geq 0} \cdot \hbar\omega_0$. ✓ (Required by Corollary 2.3.)
• Topological invariance = coherence conservation: the linking number is preserved by ambient isotopy (smooth deformations); coherence is preserved by admissible dynamics (Axiom 1). These are structurally the same conservation law.

Open gap in the proof. Step (ii) identifies the Seifert surface $\Sigma$ bounded by $\gamma_1$ with the boundary of $\mathcal{O}_1$‘s coherence domain. The Entropy derivation defines the coherence domain as a subset of the relational invariant graph (Definition 2.1), not a spatial surface. The identification of the graph-theoretic domain boundary with a Seifert surface in the embedding manifold is a geometric interpretation that requires the full loop-embedding picture of Step 1 to be formalized. Once observer loops are rigorously embedded in the spatial manifold (Open Gap 3 of this derivation), this identification becomes a theorem rather than an interpretation.

Step 2: The Cayley-Dickson Gauge Chain as a Chern-Simons Sequence

Observation 2.1 (Gauge groups are CS gauge groups). The Bootstrap Division Algebras derivation forces the gauge group sequence:

Each gauge group $U(1)$, $SU(2)$, $SU(3)$ admits a well-defined Chern-Simons theory with integer level $k$. The sequence terminates at the octonions because sedenions have zero divisors (Bootstrap Division Algebras Theorem 7.1), meaning the tower of CS theories is finite — exactly three non-trivial levels.

Observation 2.2 (Wilson loops as observer phase measurements). The Wilson loop $W(\gamma) = \mathcal{P}\exp\!\left(-ig\oint_\gamma A_\mu\, dx^\mu\right)$ computes the holonomy of the gauge field around a closed curve $\gamma$. In the framework, each observer IS a phase loop, and its interaction with the gauge field is precisely its Wilson-loop holonomy. The identification:

$\text{observer phase measurement} = \text{Wilson loop expectation value in CS theory}$

is the natural dictionary between the framework’s observer-loop language and the CS TQFT language.

Observation 2.3 (CS level and integer quantization). In Chern-Simons theory, the level $k$ is a positive integer required for gauge invariance under large gauge transformations. In the framework, coherence is quantized in $\hbar\omega_0$ units (Bootstrap Corollary 2.3). The CS level $k$ and the coherence integer count $N^{(n)}$ are both integer-valued parameters that control the “resolution” of the theory at each level.

Proposition 2.4 (CS level ratios from the division algebra chain). The Chern-Simons level at each bootstrap level is inversely proportional to the real dimension of the corresponding division algebra:

$k_1 : k_2 : k_3 = \frac{1}{\dim_\mathbb{R}\,\mathbb{C}} : \frac{1}{\dim_\mathbb{R}\,\mathbb{H}} : \frac{1}{\dim_\mathbb{R}\,\mathbb{O}} = \frac{1}{2} : \frac{1}{4} : \frac{1}{8} = 4 : 2 : 1$

Proof. The standard relationship between a Yang-Mills coupling $\alpha = g^2/(4\pi)$ and the Chern-Simons level is $k = 1/\alpha$ (in conventions where the YM action is $(1/4g^2)\int\text{tr}(F \wedge *F)$). The Coupling Constants derivation (Structural Postulate S1) constrains the gauge coupling ratios at the algebraic crystallization scale:

$\alpha_1 : \alpha_2 : \alpha_3 = \dim_\mathbb{R}\,\mathbb{C} : \dim_\mathbb{R}\,\mathbb{H} : \dim_\mathbb{R}\,\mathbb{O} = 1 : 2 : 4$

(larger algebras → stronger couplings, matching the empirical ordering $\alpha_3 > \alpha_2 > \alpha_Y$). Applying $k = 1/\alpha$ at each level inverts the ratios:

$k_1 : k_2 : k_3 = \frac{1}{\dim_\mathbb{R}\,\mathbb{A}_1} : \frac{1}{\dim_\mathbb{R}\,\mathbb{A}_2} : \frac{1}{\dim_\mathbb{R}\,\mathbb{A}_3} = 4 : 2 : 1. \quad\square$

Status. The CS-level ratios $k_1 : k_2 : k_3 = 4 : 2 : 1$ follow from S1 and the standard $k = 1/\alpha$ identification. The absolute integer values are not fixed by S1 alone — any integer multiple $(k_1, k_2, k_3) = (4m, 2m, m)$ for $m \in \mathbb{Z}_{>0}$ satisfies the ratio constraint. Fixing the absolute normalization would require an independent principle such as the Verlinde / holographic-bound correspondence (Open Question 2.6) or a direct derivation of the bootstrap absolute coupling.

The smallest positive integer assignment is $m = 1$, giving $(k_1, k_2, k_3) = (4, 2, 1)$. Downstream propositions reference this smallest case for concreteness; all such statements are conditional on this absolute-normalization choice.

Proposition 2.5 (Finite representation content at each level). At any CS level $k$, $SU(N)$ Chern-Simons theory has a finite set of integrable representations determined by the Verlinde formula. Conditional on the smallest integer assignment $(k_1, k_2, k_3) = (4, 2, 1)$ from Proposition 2.4:

• $U(1)$ at $k = 4$: charge lattice with periodicity 4 — 4 distinct charge classes
• $SU(2)$ at $k = 2$: exactly $k + 1 = 3$ integrable representations (spins $0, \frac{1}{2}, 1$)
• $SU(3)$ at $k = 1$: exactly $\binom{k+2}{2} = 3$ integrable representations (trivial, fundamental, antifundamental)

The finiteness of the representation content at each level is a topological constraint from the CS structure, matching the framework’s expectation that each bootstrap level has finite degrees of freedom (from the holographic bound, Area Scaling Proposition 6.2 of Bootstrap). For larger normalizations $(k_1, k_2, k_3) = (4m, 2m, m)$ the rep counts grow accordingly.

Open question 2.6 (Verlinde formula vs. holographic bound). For $SU(N)$ CS at level $k$ on a surface of genus $g$, the Verlinde formula gives the dimension of the Hilbert space. The framework’s holographic bound gives $A/(4\ell_P^2)$ as the maximum number of independent states on a boundary surface. If these are the same finite-dimensionality condition, then the CS level is fixed by the holographic bound — giving both a direct link between the topological (CS) and gravitational (area-scaling) sectors and the missing absolute-normalization principle for Proposition 2.4. For genus 1 (torus) with $SU(2)$ at $k = 2$ (smallest assignment): Verlinde dimension = 3, so $A/(4\ell_P^2) = 3$, giving $A = 12\,\ell_P^2$ as the minimum torus area supporting $SU(2)$ gauge structure. Whether this correspondence holds — and whether the smallest integer assignment is the physically correct one — is an open question tracked as a forward direction.

Step 3: The Coherence Lagrangian and the Chern-Simons Action

Observation 3.1 (The CS 3-form is already in the framework). The Strong CP derivation (Step 3c) writes the topological charge as:

$\nu = \frac{1}{8\pi^2}\int_{S^3} \text{CS}(A), \qquad \text{CS}(A) = \text{tr}\!\left(A \wedge dA + \tfrac{2}{3}A \wedge A \wedge A\right)$

The framework constrains this term (octonionic non-associativity forces $\theta = 0$), but the Chern-Simons structure is present in the formalism.

Proposition 3.2 (Topological sector of canonical 4D Yang-Mills is labeled by the spatial Chern-Simons functional). The gauge sector of the Coherence Lagrangian is:

$\mathcal{L}_{\text{gauge}} = -\frac{1}{4}F_{\mu\nu}F^{\mu\nu}$

In the canonical (Hamiltonian) formulation on a spacetime $\mathbb{R} \times M$ where $M$ is a spatial 3-manifold, gauge-orbit winding sectors are labeled by the Chern-Simons functional of the spatial connection $A_i$. The full 4D Yang-Mills theory decomposes into:

• Dynamical content: propagating gauge bosons, governed by the full 4D kinetic term. This is the part responsible for forces between particles.
• Topological content: the classification of gauge orbits on $M$ by their winding number, governed by the Chern-Simons functional of the spatial connection $A_i$:

$W_{\text{CS}}[A] = \frac{1}{4\pi}\int_M \text{tr}\!\left(A \wedge dA + \tfrac{2}{3}A \wedge A \wedge A\right)$

Under a large gauge transformation with winding number 1, $W_{\text{CS}}$ shifts by 1. The quantity $W_{\text{CS}} \bmod 1$ is gauge-invariant and labels the topological sector. The relationship between 4D Yang-Mills and 3D Chern-Simons is not an approximation — the topological sector is exact [Witten, 1988; Jackiw, 1984].

Proposition 3.3 ($\theta = 0$ makes the topological sector maximally accessible). The Strong CP derivation forces $\theta = 0$ exactly (octonionic non-associativity obstructs non-trivial winding). At $\theta = 0$, the gauge vacuum is an equal-weight superposition over all topological sectors: $|0\rangle = \sum_n |n\rangle$. No sector is suppressed by a phase factor. The topological structure of the gauge vacuum — including the linking invariants of Wilson loops computed by CS theory — is maximally “visible” rather than modulated by an arbitrary phase.

Proposition 3.4 (Wilson-loop linking from the topological sector). In the topological sector of the coherence Lagrangian’s gauge vacuum, the expectation value of two Wilson loops is a knot/link invariant computed by Chern-Simons theory. At CS level $k$, $SU(N)$ CS gives:

• $U(1)$ at level $k$: $\langle W_{q_1}(\gamma_1)\, W_{q_2}(\gamma_2)\rangle_{\text{top}} \propto \exp\!\left(2\pi i\, q_1 q_2\, \text{Lk}(\gamma_1, \gamma_2)/k\right)$, depending only on the linking number modulo $k$.
• $SU(2)$ at level $k$: the Jones polynomial evaluated at $q = e^{2\pi i/(k+2)}$.
• $SU(3)$ at level $k$: the HOMFLYPT polynomial evaluated at a specific root of unity determined by $k$.

Conditional on the smallest integer assignment $(k_1, k_2, k_3) = (4, 2, 1)$ from Proposition 2.4:

• U(1) / electromagnetism at $k = 4$: linking numbers detected modulo 4; the charge lattice has 4 classes.
• SU(2) / weak at $k = 2$: Jones polynomial at $q = e^{2\pi i/4} = i$ — the fourth root of unity, giving the Kauffman bracket at $A = e^{i\pi/4}$.
• SU(3) / color at $k = 1$: HOMFLYPT at the root of unity corresponding to minimal SU(3) CS, which at $k = 1$ has only three integrable reps (trivial, fundamental, antifundamental).

For larger integer assignments $(4m, 2m, m)$ the specific roots of unity and polynomial evaluations scale with $m$.

The underlying relationship between CS theory and knot invariants is a known mathematical result (Witten 1988) applied to the Lagrangian the framework derives. What is conditional — and what requires the absolute-normalization principle of Open Question 2.6 — is the specific integer level $m$ at which the invariants are evaluated.

Observation 3.5 (The coherence measure has a topological-gauge component). The full coherence measure $\mathcal{C}$ receives contributions from all sectors of the coherence Lagrangian: matter (Fisher kinetic term), gauge ($F^2$ term), and gravity (Einstein-Hilbert $R$ term). The topological component of the gauge sector — governed by the CS functional on spatial slices — provides the linking-invariant content. This is the component where Proposition 1.5 (relational coherence = linking number $\times\,\hbar\omega_0$) would live. The dynamical component (propagating gauge bosons) and the other sectors (matter, gravity) contribute additional, non-topological coherence that is distinct from the linking structure.

Observation 3.6 (Specific knot invariants at the smallest integer assignment). Under Proposition 2.4’s smallest integer assignment $(k_1, k_2, k_3) = (4, 2, 1)$, the Jones polynomial at $q = i$ (from $SU(2)$ at $k = 2$) has known mathematical properties — it is related to the representation theory of the quantum group $U_q(\mathfrak{su}(2))$ at a 4th root of unity (the Kauffman bracket variable $A = e^{i\pi/4}$), where the representation category is modular and has exactly 3 simple objects corresponding to spins $0, 1/2, 1$. If this modular category is the correct description of the observer network’s topological content at bootstrap level 2, it provides a concrete finite algebraic structure governing the weak-sector linking of observer loops. Resolution of the absolute-normalization question (Open Question 2.6) would either confirm this assignment or redirect the specific invariants accordingly.

Step 4: Surgery as the Geometry Functor

Observation 4.1 (Lickorish-Wallace theorem). Every closed oriented 3-manifold can be obtained by Dehn surgery on a framed link in $S^3$ [Lickorish, 1962; Wallace, 1960]. Conversely, every framed link in $S^3$ determines a unique closed oriented 3-manifold. This provides a bidirectional dictionary:

$\text{framed link} \xleftrightarrow{\text{surgery}} \text{3-manifold}$

Observation 4.2 (Kirby calculus). Two framed links give the same 3-manifold if and only if they are related by a finite sequence of Kirby moves (handle slides and stabilizations) Kirby, 1978. This provides the equivalence relation on the discrete side that corresponds to diffeomorphism on the continuous side.

Conjecture 4.3 (Geometry functor = Dehn surgery). The geometry functor $G: \mathbf{Obs} \to \mathbf{Geom}$ maps the observer network (a framed link $L$ in the spatial manifold) to the 3-manifold $M_L$ obtained by Dehn surgery on $L$. The bootstrap fixed-point equation $\mathcal{U} \cong \mathcal{R}(\mathcal{U}, \mathcal{U})$ becomes:

$M_L \text{ contains } L \text{ as its observer network, and surgery on } L \text{ reproduces } M_L$

This is self-consistent surgery: the manifold produced by the link is the manifold the link inhabits.

Why this is a fixed-point equation. Start with a candidate manifold $M_0$. Embed the observer network as a framed link $L_0 \subset M_0$. Perform surgery on $L_0$ to get a new manifold $M_1 = \text{Surgery}(L_0)$. Embed the observer network as a framed link $L_1 \subset M_1$. Iterate. The fixed point is the pair $(M^*, L^*)$ where $\text{Surgery}(L^*) = M^*$ and $L^* \subset M^*$ is the observer network in $M^*$. If the iteration converges, the fixed point is the self-consistent geometry.

Step 5: The ER=EPR Bridge as Topological Tubes

Observation 5.1 (Wormhole throat as linking tube). The ER=EPR derivation establishes that the wormhole throat area between two entangled observers satisfies $A = 4\ell_P^2\,\mathcal{C}(I_{12})$, where $\mathcal{C}(I_{12})$ is the coherence content of the relational invariant. In the knot-theoretic picture, two linked loops have a tubular neighborhood of the linking region — a handle connecting the two loops’ neighborhoods. The cross-sectional area of this tube scales with the linking number. With the identification of Proposition 1.5 ($\mathcal{C} = \text{Lk} \cdot \hbar\omega_0$), the ER=EPR throat area becomes:

$A = 4\ell_P^2 \cdot \text{Lk}(\gamma_1, \gamma_2) \cdot \hbar\omega_0$

This is geometrically natural: more linking = thicker tube = bigger wormhole throat.

Step 6: Modular Equivariance of the Horizon CS Theory

The CS theory of Step 3 lives on the framework’s horizon structure — the null boundary $\partial M_A$ of each observer’s projection, topologically $S^2 \times S^1$ (Horizon Gauge Shell Proposition 3.1 plus Axiom 3’s time compactification). This step identifies the modular symmetry that acts on the CS partition functions.

Proposition 6.1 (Möbius automorphism group of the horizon). The conformal $S^2$ factor of $\partial M_A$ admits the orientation-preserving conformal automorphism group $\mathrm{PSL}(2, \mathbb{C})$ — the Möbius group. Combined with $T_A$-periodic translations on the $S^1$ time factor, the full horizon automorphism group is $\mathrm{PSL}(2, \mathbb{C}) \times U(1)$.

Argument. Classical conformal geometry: $\mathrm{Conf}^+(S^2) = \mathrm{PSL}(2, \mathbb{C})$ [Schottenloher 1997, §2]. The dS static-patch metric (Observer-Projected Spacetime Theorem 3.1) restricted to the Killing horizon $r = L_A$ induces the round conformal structure on the $S^2$ factor [Anderson–Chruściel 2005]. The $S^1$ factor is the $T_A$-periodic time direction from Axiom 3. $\square$

Proposition 6.2 (Chern–Simons modular equivariance). The horizon CS theory at levels $k_1 : k_2 : k_3 = 4 : 2 : 1$ (Proposition 2.4) is equivariant under $\mathrm{SL}(2, \mathbb{Z})$ — the arithmetic subgroup of $\mathrm{PSL}(2, \mathbb{C})$ — via modular transformations on the boundary torus.

Argument sketch. The Chern–Simons partition function on a 3-manifold with $T^2$ boundary is a modular form under $\mathrm{SL}(2, \mathbb{Z})$ [Witten 1989, §4]. CS quantization at integer level produces Verlinde characters that transform under $\mathrm{SL}(2, \mathbb{Z})$ via specific modular matrices [Moore–Seiberg 1989]. For the framework’s CS levels, the Verlinde algebras are finite and explicitly modular-covariant. $\square$

Remark 6.3 (Distinguished involutive elements). Two involutions on $\mathrm{SL}(2, \mathbb{Z})$ are physically relevant: the S-transformation $\tau \to -1/\tau$ (modular inversion) and the charge-conjugation element $C$. The CPT operator of CPT Theorem acts on horizon data as an involution preserving the integer residue (cf. Observer Holographic Equivalence Corollary 4.5); identifying which specific $\mathrm{SL}(2, \mathbb{Z})$ element this corresponds to is open — see Open Gap 6 below. Resolution requires computing CS modular matrices at the framework’s specific levels and matching their action on Verlinde characters to the CPT action.

Remark 6.4 (Why the framework’s horizon content is integer-valued). Observer Holographic Equivalence Proposition 4.1 establishes that null surfaces carry only integer/topological data. The modular equivariance here is the natural automorphism group action on that integer data: once the horizon’s CS description is fixed (integer levels, Verlinde characters), the only automorphisms it can admit are those preserving integer quantization — i.e., $\mathrm{SL}(2, \mathbb{Z})$ rather than the full continuous $\mathrm{PSL}(2, \mathbb{C})$. This makes the modular structure structurally forced rather than a supplementary choice.

Rigor Assessment

Structurally grounded (from existing rigorous derivations):

• Observation 1.2: Observers are $U(1)$ loops (from Loop Closure, rigorous)
• Observation 2.1: Gauge groups from Cayley-Dickson tower (from Bootstrap Division Algebras, rigorous)
• Observation 2.2: Wilson loops in the framework (from Electromagnetism, rigorous)
• Observation 3.1: CS 3-form in the framework (from Strong CP, rigorous)

Grounded in known mathematical results, applied to the framework:

• Proposition 1.3: Canonical framing via normal-bundle almost-complex structure. The construction is standard differential geometry on oriented rank-2 real vector bundles over oriented loops; the integer framing follows from Axiom 3 (loop closure). Bootstrap-native almost-complex structures at all three levels are manifest: level 1 from $\mathbb{C}$, level 2 from the hypercharge embedding $\mathbb{C} \hookrightarrow \mathbb{H}$, level 3 from the chain $\mathbb{C} \hookrightarrow \mathbb{H} \hookrightarrow \mathbb{O}$ via the electroweak-fixed hypercharge direction $e$ (Observation 1.3c). The naively expected $G_2$-monodromy obstruction does not arise because the level-3 connection is $SU(3) = \text{Stab}_{G_2}(e)$-valued and $SU(3)$ preserves $\text{span}(1, e)$ pointwise.
• Proposition 3.2: The topological sector of the coherence Lagrangian’s gauge sector is governed by Chern-Simons theory on spatial slices (standard canonical YM structure [Witten 1988, Jackiw 1984], applied to the specific Lagrangian derived in Coherence Lagrangian Theorem 3.1)
• Proposition 3.3: θ = 0 (from Strong CP) makes the topological sector maximally accessible
• Proposition 3.4: Wilson-loop expectation values in the topological sector are knot/link invariants (standard CS-TQFT result), with specific CS levels from Proposition 2.4
• Observation 1.3c: Level-3 framing resolution — the framework’s electroweak structure globally fixes the hypercharge direction $e$, reducing the level-3 structure group to $SU(3) = \text{Stab}_{G_2}(e)$, which preserves the 2-plane $\text{span}(1, e)$ carrying $J$ pointwise. Uses the stabilizer computation of Color Force Theorem 3.1 and the confinement result of Confinement Theorem 3.1. Conditional on the embedding prescription of Open Gap 3 identifying $N(\gamma_\mathcal{O})$ with $\text{span}(1, e)$ consistently along the loop.

Derived from the coupling ratio rule (rigorous for ratios; absolute integer levels conditional on Open Question 2.6):

• Proposition 2.4: CS level ratios $k_1 : k_2 : k_3 = 4 : 2 : 1$ follow from Structural Postulate S1 of Coupling Constants (structural consequence of the Cayley-Dickson tower, with $\alpha_i \propto \dim \mathbb{A}_i$) and the standard $k = 1/\alpha$ identification. The absolute integer multiplier $m$ in $(k_1, k_2, k_3) = (4m, 2m, m)$ is not determined by S1 alone.
• Proposition 2.5: Finite representation content at each level follows from the Verlinde formula applied at the CS levels from Proposition 2.4 (standard TQFT). Specific rep counts are conditional on the smallest integer assignment $m = 1$.
• Proposition 3.4: Wilson-loop knot invariants at specific roots of unity (standard CS-TQFT result [Witten 1988]) — the specific root-of-unity evaluations are conditional on the smallest integer assignment.

Structurally suggestive (consistent with existing derivations, not yet derived):

• Observation 1.4: Relational invariants as topological linking
• Observation 3.5: The coherence measure decomposes into topological-gauge + dynamical + gravitational components

Rigorous (classical mathematics):

• Proposition 6.1 (Möbius automorphism group of the horizon): classical conformal geometry of $S^2$ plus the dS static-patch induced metric on the Killing horizon.
• Proposition 6.2 (CS modular equivariance): standard CS-TQFT result ($\mathrm{SL}(2, \mathbb{Z})$ equivariance of CS partition functions on $T^2$-boundary 3-manifolds) applied to the framework’s integer CS levels.

Proof sketch (grounded in existing derivations, two explicit assumptions, one geometric identification remains):

• Proposition 1.5: Relational coherence = |Lk| × ℏω₀. The argument chains area-scaling boundary counting (Propositions 1.2 + 3.1), integer quantization (Corollary 2.3), and coherence conservation (Axiom 1) through the standard knot-theory fact that the linking number is the minimum geometric intersection number. Two explicit assumptions (A: sign-independent coherence contribution; B: isotopy preserves coherence and the ground state minimizes crossing count) are stated in the proof. The one remaining gap is the identification of the observer’s coherence-domain boundary with a Seifert surface — a geometric interpretation that becomes a theorem once the loop-embedding picture (Open Gap 3) is formalized.

Conjectural (no proof, requires new formal work):

• Conjecture 4.3: Geometry functor = Dehn surgery
• Open Question 2.6: Verlinde formula = holographic bound

Open Gaps

1. Absolute normalization of the CS levels. Proposition 2.4 gives only the ratios $k_1 : k_2 : k_3 = 4 : 2 : 1$. The absolute integer multiplier $m$ in $(k_1, k_2, k_3) = (4m, 2m, m)$ is not determined by S1 alone. Fixing $m$ would determine the specific rep counts in Proposition 2.5, the specific roots of unity in Proposition 3.4, and the concrete knot invariants throughout. Two plausible principles could fix $m$: (i) a Verlinde/holographic-bound correspondence (Open Question 2.6), or (ii) an independent bootstrap derivation of one absolute coupling (which would also close Open Gap 1 of Coupling Constants).

2. Verlinde formula vs. holographic bound (Open Question 2.6). At CS level $k$ with gauge group $SU(N)$, the Verlinde formula gives the Hilbert-space dimension on a genus-$g$ surface. The framework’s holographic bound $A/(4\ell_P^2)$ gives a finite count of states on a boundary surface. If these are the same constraint, it would directly link the topological (CS) and gravitational (area-scaling) sectors and fix the absolute integer multiplier $m$ in Proposition 2.4. Checking this requires computing the Verlinde dimensions at candidate $(k_1, k_2, k_3) = (4m, 2m, m)$ for physically relevant surfaces and comparing to the holographic bound evaluated on the same surfaces.

3. Embedding of observer loops. The framework’s observer loops are currently abstract $U(1)$ phases, not embedded curves in a spatial manifold. Making the knot-theoretic program concrete requires a canonical embedding prescription — determining WHERE in the spatial geometry each observer’s loop sits. The self-referential nature of this (the geometry is produced by the loops’ surgery, so the loops’ positions are determined by the geometry they produce) is exactly the fixed-point equation. Additional requirement from Observation 1.3c: at bootstrap level 3, the embedding prescription must deliver a covariantly-constant identification of the normal bundle $N(\gamma_\mathcal{O})$ with $\text{span}(1, e) \subset \mathbb{O}$ (where $e$ is the globally-fixed hypercharge direction), so that the inherited framing $J$ is well-defined along the loop. The $SU(3)$-valued bootstrap connection automatically preserves $\text{span}(1, e)$, so the constraint reduces to the embedding being $SU(3)$-equivariant.

4. Linking number vs. relational coherence for non-minimal observers. Proposition 1.5 identifies $\mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2) = \text{Lk}(\gamma_1, \gamma_2) \cdot \hbar\omega_0$ for minimal $U(1)$ loops. For composite observers at higher bootstrap levels, the “loop” is a more complex embedded object (satellite knot, cable link). The linking number generalizes to satellite linking numbers and Milnor invariants. Does the framework’s relational coherence at higher levels match these higher invariants?

5. Surgery convergence. The self-consistent surgery iteration (Step 4) must converge. Under what conditions does iterated surgery on a framed link in successive manifolds converge to a fixed point? This is a question in 3-manifold topology that may have known answers or may require new results.

6. Specific $\mathrm{SL}(2, \mathbb{Z})$ element for CPT. Proposition 6.2 establishes that the CS theory is modular-equivariant; Remark 6.3 notes that the CPT operator of CPT Theorem acts on horizon integer data as a specific element of $\mathrm{SL}(2, \mathbb{Z})$. Identifying the element (S-inversion $\tau \to -1/\tau$, charge conjugation $C$, product $ST$, or other) requires explicit computation of Verlinde-character modular matrices at the framework’s CS levels ($k_n = 4m, 2m, m$) matched to CPT’s action on observer structure (per CPT Theorem Theorem 4.1 Step A). Connects to Open Gap 1 (absolute multiplier $m$). Difficulty: MODERATE.