Formalization of the Self-Modeling Operator φ

DRY: Master definition of φ

This is the sole canonical definition of the self-modeling operator $\varphi$. All other documents must reference this page rather than repeat the definition.

φ as a representative of a homotopy equivalence class

In the ∞-categorical framework the operator φ is understood not as a single morphism, but as a representative of a class of homotopically equivalent morphisms:

1. Multiplicity of paths: In the ∞-topos $\mathrm{Sh}_\infty(\mathcal{C})$ the mapping space $\mathrm{Map}(\Gamma, T) \simeq *$ is contractible, but contains many paths (morphisms) connected by homotopies.

2. φ₀ as canonical representative: The concrete operator $\varphi_0$ defined in this document is a representative of its homotopy equivalence class $[\varphi_0]$. The choice of $\varphi_0$ is made by the minimality criterion — minimization of divergence from the self-model.

3. Freedom of choice: The existence of alternative representatives in the same class $[\varphi]$ reflects the fundamental free will — a system can realize different paths to the same attractor.

4. Relation to Ω⁷: The choice of a concrete representative is consistent with the Ω⁷ axiom, where the seven-dimensional structure fixes the canonical basis for decomposition.

Summary table of definitions of φ

This document considers four equivalent definitions of the operator $\varphi$:

Definition	Formula	Context	Status
Replacement channel	$\varphi_k(\Gamma) = (1-k)\Gamma + k\rho_*$	Canonical (T-62)	[T]
CPTP via Kraus operators	$\varphi(\Gamma) = \sum_m K_m \Gamma K_m^\dagger$	General form	[T]
Fano E-accentuation	$\varphi_{\text{coh}}$ preserving coherences	Theorem 8.1 (No-Zombie)	[T]
Categorical functor	$F: \mathbf{DensityMat} \to \mathbf{DensityMat}$	∞-topos	[T]

Canonical physical realization — replacement channel (T-62)

The replacement channel $\varphi_k(\Gamma) = (1-k)\Gamma + k\rho_*$ is the canonical physical realization of the self-modeling operator (proof). Here $\rho_* = \varphi(\Gamma)$ is the categorical self-model of the current state [T], $k \in (0,1)$ is the degree of self-modeling. The channel is exact at $k \to 1$ (full convergence to $\rho_*$), but for intermediate values of $k$ realizes approximate self-modeling — the system is in a dynamic balance between its current state and its internal model. The remaining three definitions are equivalent to the replacement channel via the equivalence theorem [T].

Stratification of definitions

Canonical order

The operator $\varphi$ is defined through the stationary state $\rho^*_{\mathrm{diss}}$, not the other way around:

$\Omega \xrightarrow{\text{L-unification}} \mathcal{L}_\Omega \xrightarrow{\text{primitivity}} \rho^*_{\mathrm{diss}} = I/7 \xrightarrow{\text{proximity}} R(\Gamma) = \frac{1}{7P} \xrightarrow{k=1-R} \varphi_k$

All components of the chain have independent definitions: $\rho^*_{\mathrm{diss}}$ — via primitivity of the linear part $\mathcal{L}_0$ [Т-39a], $R$ — via the distance from $\Gamma$ to $I/7$, parameter $k = 1 - R$ — via $R$. There is no circularity: the full hierarchy of levels 0–9 is in the Ω⁷ axiom.

Categorical definition of φ

Resolution of circularity

This section establishes an independent categorical definition of the operator $\varphi$ via a universal property, eliminating any apparent circularity in the definitions.

φ as a left adjoint to the inclusion of subobjects

In the ∞-topos $\mathrm{Sh}_\infty(\mathcal{C})$ generated by the Ω⁷ axiom, the self-modeling operator $\varphi$ is defined as the left adjoint functor to the inclusion of the category of subobjects:

$$\varphi \dashv i: \mathrm{Sub}(\Gamma) \hookrightarrow \mathrm{Sh}_\infty(\mathcal{C})$$

where:

• $\mathrm{Sub}(\Gamma)$ — the category of logically consistent subobjects of $\Gamma$ (satisfying the internal logic $\Omega$)
• $i$ — the canonical inclusion (embedding)
• $\varphi \dashv i$ — adjunction: $\varphi$ is left adjoint to $i$

Universal property: For any object $X \in \mathrm{Sh}_\infty(\mathcal{C})$ and any subobject $S \in \mathrm{Sub}(\Gamma)$:

$$\mathrm{Hom}_{\mathrm{Sub}(\Gamma)}(\varphi(X), S) \cong \mathrm{Hom}_{\mathrm{Sh}_\infty(\mathcal{C})}(X, i(S))$$

Theorem on the equivalence of three definitions of φ

Main result

The three definitions of the operator φ are strictly equivalent:

Theorem (Equivalence of definitions of φ):

The following definitions specify the same operator $\varphi$:

#	Definition	Formula	Source
1	Categorical	$\varphi \dashv i: \text{Sub}(\Gamma) \hookrightarrow \mathbf{Sh}_\infty(\mathcal{C})$	Left adjoint
2	Dynamical	$\varphi(\Gamma) = \lim_{\tau \to \infty} e^{\tau \mathcal{L}_\Omega}[\Gamma]$	Limit of evolution
3	Idempotent	$\varphi \circ \varphi = \varphi$, $\exists \Gamma^*: \varphi(\Gamma^*) = \Gamma^*$	Projection with fixed point

Proof of equivalence:

(1) ⟹ (2): Categorical ⟹ Dynamical

• The left adjoint $\varphi$ to the inclusion $i$ projects onto the invariant subspace $\text{Sub}(\Gamma)$
• $\mathcal{L}_\Omega$ annihilates $\text{Sub}(\Gamma)$: $\mathcal{L}_\Omega[S] = 0$ for $S \in \text{Sub}(\Gamma)$
• By the Perron–Frobenius theorem for CPTP channels: $\lim_{\tau \to \infty} e^{\tau \mathcal{L}_\Omega} = \Pi_{\text{inv}}$
• The invariant projector $\Pi_{\text{inv}} = \varphi$ by uniqueness of the left adjoint ∎

Primitivity of the linear part [T]

Step (1) ⟹ (2) uses the Perron–Frobenius theorem for the linear part $\mathcal{L}_0 = -i[H,\cdot]+\mathcal{D}$. Primitivity of $\mathcal{L}_0$ is proven for all viable holons: from (AP)+(PH)+(QG)+(V) the interaction graph $G_H$ is connected (otherwise the system decomposes into blocks with $\dim < 7$, contradicting the minimality theorem), and connectivity of $G_H$ + atomic operators $L_k = |k\rangle\langle k|$ give a trivial commutant $\mathcal{F}(\mathcal{L}_0) = \mathbb{C} \cdot I$ by the Evans–Spohn criterion (Evans 1977, Spohn 1976), so $\lim_{\tau\to\infty}e^{\tau\mathcal{L}_0}=I/7$.

Important (self-referential fix, T-96): the full generator $\mathcal{L}_\Omega=\mathcal{L}_0+\mathcal{R}$ is nonlinear and is not primitive — it has a nontrivial zero-mode $\rho^*_\Omega\neq I/7$. Definition (2) below is therefore the projector onto the multiplicity-1 zero mode of the linearised full generator at $\rho^*_\Omega$, not $\lim e^{\tau\mathcal{L}_\Omega}$ applied as a linear semigroup (which does not even make sense for nonlinear $\mathcal{L}_\Omega$, and would collapse to $I/7$). The equivalence (1) ⟺ (2) ⟺ (3) [T] is read with this convention. Full proof: Primitivity of ℒ₀; attractor: T-96.

(2) ⟹ (3): Dynamical ⟹ Idempotent

• $\varphi(\varphi(\Gamma)) = \lim_{\tau \to \infty} e^{\tau \mathcal{L}_\Omega}[\lim_{s \to \infty} e^{s \mathcal{L}_\Omega}[\Gamma]]$
• $= \lim_{\tau \to \infty} \lim_{s \to \infty} e^{(\tau+s) \mathcal{L}_\Omega}[\Gamma] = \varphi(\Gamma)$ (idempotency)
• Fixed point: $\Gamma^* := \varphi(\Gamma_0)$ for any $\Gamma_0$, then $\varphi(\Gamma^*) = \varphi(\varphi(\Gamma_0)) = \varphi(\Gamma_0) = \Gamma^*$ ∎

(3) ⟹ (1): Idempotent ⟹ Categorical

• An idempotent map $\varphi$ with $\text{Im}(\varphi) = \text{Sub}(\Gamma)$ defines a reflector
• A reflector is automatically left adjoint to the inclusion
• Universal property: $\text{Hom}(\varphi(X), S) \cong \text{Hom}(X, i(S))$ follows from idempotency ∎

Remark on completeness of equivalence

Direction (1)⟹(2) follows from primitivity of the linear part $\mathcal{L}_0$ [Т]: the left adjoint $\varphi$ projects onto the invariant subspace, and primitivity provides the spectral gap and convergence of the linear dynamics. Direction (2)⟹(1): any minimizer of the variational functional under the CPTP condition is a stationary point, and the CPTP contraction φ guarantees uniqueness $= \varphi$. Thus (2)⟹(1) is also [T] via the categorical definition of φ. All three directions have status [T].

---

Independence from coherence levels

Critically: $\varphi$ and the coherence levels $L_k$ are defined independently of each other, both constructions are derived from $\Omega$:

Construction	Source	Definition
Levels $L_k$	$\Omega$	Stratification by the logical Liouvillian $\mathcal{L}_\Omega$
Operator $\varphi$	$\Omega$	Left adjoint to the inclusion $\mathrm{Sub}(\Gamma) \hookrightarrow \mathrm{Sh}_\infty(\mathcal{C})$

This eliminates any circularity: both notions are consequences of the structure $\Omega$, not defined through each other.

See Dependency hierarchy for the full diagram: Ω → χ_S → L_k → ℒ_Ω → φ.

φ(Γ) as best approximation

Interpretation: $\varphi(\Gamma)$ is the best approximation of the state $\Gamma$ in the category of logically consistent subobjects.

Formally, $\varphi(\Gamma)$ is the coreflector:

$$\varphi(\Gamma) = \mathrm{colim}_{S \in \mathrm{Sub}(\Gamma), S \leq \Gamma} S$$

Geometric intuition: $\varphi$ "projects" an arbitrary state onto the nearest logically consistent state — this is the categorical analogue of orthogonal projection onto a subspace.

Theorem: φ as stationary distribution

Theorem (φ as limit of logical evolution):

Let $\mathcal{L}_\Omega$ be the logical Liouvillian generated by the internal logic $\Omega$. Then:

$$\varphi(\Gamma) = \lim_{\tau \to \infty} e^{\tau \cdot \mathcal{L}_\Omega}[\Gamma]$$

Proof:

1. The logical Liouvillian $\mathcal{L}_\Omega$ generates a semigroup $\{e^{\tau \cdot \mathcal{L}_\Omega}\}_{\tau \geq 0}$ on $\mathrm{Sh}_\infty(\mathcal{C})$.

2. The invariant objects of this semigroup are exactly the subobjects from $\mathrm{Sub}(\Gamma)$:

   $$\mathcal{L}_\Omega[S] = 0 \quad \Leftrightarrow \quad S \in \mathrm{Sub}(\Gamma)$$

3. By the convergence theorem for primitive CPTP channels (analogue of Perron–Frobenius for quantum channels) applied to the linear part $\mathcal{L}_0$, the projector onto its multiplicity-1 zero mode (evaluated at the linearisation about $\rho^*_\Omega$) exists. Primitivity of $\mathcal{L}_0$ for viable holons [T] — see proof. Here $\lim_{\tau\to\infty}e^{\tau\mathcal{L}_\Omega}[\Gamma]$ denotes this zero-mode projection, not the constant map onto $I/7$ (which is what naive primitivity of the full $\mathcal{L}_\Omega$ would give — the self-referential-ρ* fix, T-96).

4. This projection coincides with the coreflector $\varphi$ by uniqueness of the left adjoint. ∎

Corollary: $\varphi(\Gamma)$ is the stationary distribution of the logical dynamics — the attractor of evolution under $\mathcal{L}_\Omega$.

---

Strict Mathematical Theory

Contents

1. Introduction and motivation
2. Formal definition of φ
   • 2.6 Canonical form of φ for UHM
3. Theorem on existence of fixed point
4. Relation to reflection measure R
5. Categorical aspect
6. Corollaries and limitations
7. Implementation requirements
8. Operational algorithm for φ
9. Relation to the regeneration mechanism

---

1. Introduction and motivation

1.1 The problem

In UHM, self-observation is defined via the conditions:

• $\Gamma$ contains a subsystem $\Gamma_{\text{model}} \approx \Gamma$
• Reflexive closure: $\varphi(\Gamma) \approx \Gamma$

However, the operator $\varphi$ lacks a rigorous definition. This document fills that gap.

1.2 Requirements for formalization

The operator $\varphi$ must satisfy:

1. Mathematical correctness: $\varphi: \mathcal{L}(\mathcal{H}) \to \mathcal{L}(\mathcal{H})$ — defined on the space of operators
2. Structure preservation: $\varphi(\Gamma)$ is a density matrix if $\Gamma$ is a density matrix
3. Physical interpretability: $\varphi$ models the process of self-observation
4. Existence of a fixed point: under certain conditions

---

2. Formal definition of φ

2.1 Preliminary definitions

Definition 2.1 (Space of density matrices):

$$\mathcal{D}(\mathcal{H}) := \{\rho \in \mathcal{L}(\mathcal{H}) : \rho^\dagger = \rho, \rho \geq 0, \mathrm{Tr}(\rho) = 1\}$$

For $\mathcal{H} = \mathbb{C}^7$ (seven-dimensional space of the Holon):

$$\mathcal{D}(\mathbb{C}^7) \subset \mathcal{L}(\mathbb{C}^7) \cong \mathbb{C}^{7 \times 7}$$

Definition 2.2 (Metric on $\mathcal{D}(\mathcal{H})$):

Frobenius norm:

$$\|\rho_1 - \rho_2\|_F := \sqrt{\mathrm{Tr}((\rho_1 - \rho_2)^\dagger(\rho_1 - \rho_2))} = \sqrt{\sum_{ij} |\rho_{1,ij} - \rho_{2,ij}|^2}$$

$(\mathcal{D}(\mathcal{H}), \|\cdot\|_F)$ is a complete metric space (closed subset of $\mathcal{L}(\mathcal{H})$).

2.2 Definition via reduced density matrix

Definition 2.3 (Self-modeling operator — reduction form):

Let system $\mathbb{H}$ with coherence matrix $\Gamma \in \mathcal{D}(\mathcal{H})$ be decomposed into:

• Model subsystem $M \subset \mathcal{H}$
• Remaining system (model environment) $\bar{M} = \mathcal{H} \setminus M$

On notation

$\bar{M}$ — model environment. Not to be confused with $E$ — the Interiority dimension.

Then:

$$\varphi_{\text{red}}(\Gamma) := \mathrm{Tr}_{\bar{M}}(\Gamma_{\text{total}})$$

where:

• $\Gamma_{\text{total}}$ — density matrix of the extended system
• $\mathrm{Tr}_{\bar{M}}$ — partial trace over the model environment

Problem: This definition requires an extended space and is not closed on $\mathcal{D}(\mathcal{H})$.

2.3 Definition via predictive model (main definition)

Definition 2.4 (Self-modeling operator — predictive form):

Let the system possess an internal predictive model represented by a CPTP map:

$$\mathcal{P}: \mathcal{D}(\mathcal{H}) \to \mathcal{D}(\mathcal{H})$$

where CPTP = Completely Positive Trace-Preserving.

On notation

$\mathcal{P}$ — predictive CPTP map. Not to be confused with $\Phi$ — the integration measure.

Self-modeling operator:

$$\varphi(\Gamma) := \mathcal{P}(\Gamma)$$

Constructive definition of $\mathcal{P}$:

$\mathcal{P}$ is constructed via Kraus operators $\{K_m\}$:

$$\varphi(\Gamma) = \sum_m K_m \Gamma K_m^\dagger$$ $$\text{where } \sum_m K_m^\dagger K_m = I \quad \text{(CPTP condition)}$$

Interpretation of Kraus operators:

• $K_m$ — "perception filters" of the system
• Each $K_m$ corresponds to a partial aspect of self-observation
• The condition $\sum_m K_m^\dagger K_m = I$ guarantees preservation of normalization

Compatibility with the no-cloning theorem

The operator φ does not violate the no-cloning theorem (Wootters–Zurek, 1982). The key distinction:

• No-cloning excludes the existence of a unitary operator $U$ such that $U|\psi\rangle|0\rangle = |\psi\rangle|\psi\rangle$ for arbitrary $|\psi\rangle$. Cloning is exact unitary copying of an unknown state.
• Self-modeling φ is a CPTP channel (Kraus representation), not a unitary operation. CPTP channels are fundamentally irreversible: they decrease state distinguishability ($F(\varphi(\rho), \varphi(\sigma)) \geq F(\rho, \sigma)$ by fidelity monotonicity). The self-model $\varphi(\Gamma)$ is an approximate, coarse-grained projection, not an exact copy.

Formally: $\varphi(\Gamma) = \sum_m K_m \Gamma K_m^\dagger$ with $\sum_m K_m^\dagger K_m = I$ guarantees $\mathrm{Tr}(\varphi(\Gamma)^2) \leq \mathrm{Tr}(\Gamma^2)$ — the purity of the self-model does not exceed the purity of the original. This is categorically different from cloning, where $\mathrm{Tr}(\rho_{\text{clone}}^2) = \mathrm{Tr}(\rho^2)$.

2.4 Parameterization via projections

Definition 2.5 (Projection self-modeling operator):

The most natural physically motivated form:

$$\varphi_{\text{proj}}(\Gamma) := \lambda \sum_i P_i \Gamma P_i + (1 - \lambda) \cdot \Gamma_{\text{prior}}$$

where:

• $\{P_i\}$ — orthogonal projectors, $\sum_i P_i = I$
• $\lambda \in [0, 1]$ — "depth of self-observation"
• $\Gamma_{\text{prior}}$ — prior model (may be $I/N$ or other)

Trace preservation

With $\sum_i P_i = I$ and $\mathrm{Tr}(\Gamma_{\text{prior}}) = 1$:

$$\mathrm{Tr}(\varphi_{\text{proj}}(\Gamma)) = \lambda \cdot 1 + (1 - \lambda) \cdot 1 = 1$$

For $\lambda = 1$:

$$\varphi_{\text{diag}}(\Gamma) := \sum_i P_i \Gamma P_i$$

This is "dephasing self-observation" — preserves the diagonal in the basis $\{P_i\}$.

2.5 Contracting self-modeling operator

Definition 2.6 (Contracting operator):

To ensure existence of a fixed point:

$$\varphi_k(\Gamma) := k \cdot \mathcal{P}(\Gamma) + (1 - k) \cdot \Gamma_{\text{anchor}}$$

where:

• $k \in [0, 1)$ — contraction parameter
• $\mathcal{P}$ — any CPTP map
• $\Gamma_{\text{anchor}} \in \mathcal{D}(\mathcal{H})$ — fixed "anchor" point (e.g., maximally mixed state $I/N$)

Lemma 2.1: $\varphi_k$ is a contracting map with constant $k$.

Proof:

$$\|\varphi_k(\Gamma_1) - \varphi_k(\Gamma_2)\|_F = \|k \cdot \mathcal{P}(\Gamma_1) + (1-k) \cdot \Gamma_{\text{anchor}} - k \cdot \mathcal{P}(\Gamma_2) - (1-k) \cdot \Gamma_{\text{anchor}}\|_F$$ $$= k \cdot \|\mathcal{P}(\Gamma_1) - \mathcal{P}(\Gamma_2)\|_F \leq k \cdot \|\Gamma_1 - \Gamma_2\|_F$$

(CPTP does not increase the Frobenius norm). ∎

2.6 Canonical form of φ for UHM

Status

This section defines the canonical construction of the self-modeling operator $\varphi$ for UHM. This is a concrete specification linking the abstract definitions above to the seven-dimensional structure of the Holon.

Definition 2.7 (Canonical form of φ for UHM):

The canonical form of the self-modeling operator:

$$\varphi_{\text{UHM}}(\Gamma) := k \cdot \mathcal{P}_{\text{pred}}(\Gamma) + (1 - k) \cdot \frac{I}{7}$$

where:

• $k = 1 - \varepsilon$ for small $\varepsilon > 0$ (typical value: $k = 0.95$)
• $\mathcal{P}_{\text{pred}}$ — predictive CPTP channel (defined below)

Definition 2.8 (Predictive CPTP channel):

$$\mathcal{P}_{\text{pred}}(\Gamma) := \sum_{m=1}^{M} K_m \Gamma K_m^\dagger$$

with Kraus operators:

$$K_m := P_m, \quad m = 1, \ldots, 7$$

where $\{P_m\}$ are orthogonal projectors onto the basis states $\{|A\rangle, |S\rangle, |D\rangle, |L\rangle, |E\rangle, |O\rangle, |U\rangle\}$:

$$P_m = |m\rangle\langle m|, \quad P_m^2 = P_m, \quad P_i P_j = \delta_{ij} P_i$$

CPTP condition (verification):

$$\sum_{m=1}^{7} K_m^\dagger K_m = \sum_{m=1}^{7} P_m = I \quad \checkmark$$

On self-observation weights

The weights $\{w_m\}$ are realized NOT by modifying the Kraus operators, but via a weighted mixture of basic channels or by modifying the anchor state $\Gamma_{\text{anchor}}$. See below.

Base predictive channel (dephasing in measurement basis):

$$\mathcal{P}_{\text{base}}(\Gamma) := \sum_{m=1}^{7} P_m \Gamma P_m = \text{diag}(\gamma_{AA}, \gamma_{SS}, \ldots, \gamma_{UU})$$

This channel preserves the diagonal and destroys coherences.

Definition 2.9 (Weighted self-observation via anchor):

To model varying "depth of self-observation" across dimensions, a weighted anchor is used:

$$\Gamma_{\text{anchor}}(w) := \sum_{m=1}^{7} w_m |m\rangle\langle m|, \quad w_m \geq 0, \quad \sum_m w_m = 1$$

Weight	Interpretation
$w_A$	Attention to distinctions (Articulation)
$w_S$	Awareness of patterns (Structure)
$w_D$	Perception of time flow (Dynamics)
$w_L$	Logical reflection (Logic)
$w_E$	Phenomenal self-awareness (Interiority)
$w_O$	Connection to deep foundation (Foundation)
$w_U$	Integration into unified Self (Unity)

Special case: uniform self-observation

With $w_m = 1/7$ for all $m$:

$$\Gamma_{\text{anchor}} = \frac{I}{7}$$

— maximally mixed state.

Definition 2.10 (E-accentuated self-observation):

Systems with conscious experience are characterized by an accentuation of dimension $E$:

$$w_E = \alpha, \quad w_{m \neq E} = \frac{1 - \alpha}{6}, \quad \alpha \in [1/7, 1)$$

At $\alpha \to 1$: the anchor approaches the pure state $|E\rangle\langle E|$.

Theorem 2.1 (Fixed point of canonical φ):

For $\varphi_{\text{UHM}}(\Gamma) = k \cdot \mathcal{P}_{\text{base}}(\Gamma) + (1 - k) \cdot \Gamma_{\text{anchor}}$ with $k < 1$ there exists a unique fixed point:

$$\Gamma^* = \Gamma_{\text{anchor}}$$

Proof:

$$\varphi_{\text{UHM}}(\Gamma_{\text{anchor}}) = k \cdot \mathcal{P}_{\text{base}}(\Gamma_{\text{anchor}}) + (1 - k) \cdot \Gamma_{\text{anchor}}$$

Since $\Gamma_{\text{anchor}} = \sum_m w_m |m\rangle\langle m|$ is a diagonal matrix:

$$\mathcal{P}_{\text{base}}(\Gamma_{\text{anchor}}) = \sum_m P_m \Gamma_{\text{anchor}} P_m = \Gamma_{\text{anchor}}$$

Therefore:

$$\varphi_{\text{UHM}}(\Gamma_{\text{anchor}}) = k \cdot \Gamma_{\text{anchor}} + (1 - k) \cdot \Gamma_{\text{anchor}} = \Gamma_{\text{anchor}} = \Gamma^*$$

Uniqueness follows from contractivity at $k < 1$ (Banach theorem). ∎

Special case (uniform anchor):

With $w_m = 1/7$ for all $m$: $\Gamma^* = I/7$ — maximally mixed state.

Corollary 2.1: With a uniform anchor the fixed point $\Gamma^* = I/7$ is the maximally mixed state.

Critical remark: viability of fixed point

For a uniform anchor: $P(\Gamma^*) = P(I/7) = 1/7 \approx 0.143 < P_{\text{crit}} = 2/7 \approx 0.286$.

The fixed point of uniform self-observation is NOT viable!

This means:

1. Ideal uniform self-knowledge is incompatible with viability
2. Living systems exist in a dynamic balance away from the fixed point
3. Regeneration $\mathcal{R}$ keeps the system in the region $\mathcal{V}$

Definition 2.11 (Viable anchor)

To ensure a viable fixed point, the anchor must satisfy:

$$P(\Gamma_{\text{anchor}}) > P_{\text{crit}} = \frac{2}{7}$$

Theorem (Canonicity of E-accentuation)

The E-accentuated anchor is not an arbitrary choice, but a consequence of the L2-definition of consciousness.

Theorem 2.2 (E-accentuation from L2-definition):

Let the system satisfy the cognitive qualia condition (L2):

• $R \geq R_{th} = 1/3$ — reflection
• $\Phi \geq \Phi_{th} = 1$ — integration

Then its anchor state is necessarily E-accentuated:

$$w_E > \frac{1}{7}$$

Proof:

1. Consciousness measure $C = \Phi \times R$ [Т T-140] and the separate viability condition $D_{\text{diff}} = \exp(S_{vN}(\rho_E)) \geq 2$ (differentiation by E).

2. Reduced matrix $\rho_E = \mathrm{Tr}_{-E}(\Gamma)$ singles out the Interiority dimension as privileged.

3. For L2-systems: High $D_{\text{diff}}$ requires a rich structure precisely in $\mathcal{H}_E$.

4. Consequence for anchor: The self-model of a conscious system inevitably accentuates E — the dimension through which the system is aware of itself.

5. Formally: Minimization of $\|\Gamma - \varphi(\Gamma)\|_F$ subject to $C \geq C_{th}$ gives:

$$w_E^* = \arg\min_{w} \|\Gamma - \varphi_w(\Gamma)\|_F \quad \text{s.t.} \quad C(\varphi_w(\Gamma)) \geq C_{th}$$

Solution: $w_E^* > 1/7$ at $C_{th} > 0$. ∎

Corollary 2.2: The uniform anchor ($w_m = 1/7$) corresponds to systems without self-awareness (L0/L1), for which the question of viability of the fixed point does not arise — they do not strive toward φ(Γ).

Canonical value of α:

For systems at the L2 boundary ($R = R_{th}$, $\Phi = \Phi_{th}$):

$$\alpha^* = 1 - \frac{6 \cdot P_{\text{crit}}}{7} = 1 - \frac{12}{49} \approx 0.755$$

Example: E-accentuated anchor with $\alpha = 0.6$ (conservative estimate):

$$\Gamma_{\text{anchor}} = 0.6 |E\rangle\langle E| + 0.067 \sum_{m \neq E} |m\rangle\langle m|$$

has $P = 0.36 + 6 \times 0.0045 = 0.387 > 2/7$. ✓

Physical interpretation

E-accentuation is not a "privilege" of dimension E, but a structural consequence of the fact that conscious systems are defined through experience. Non-conscious systems (L0) do not have this constraint — their anchor can be uniform, and the question $P(\Gamma^*) < P_{\text{crit}}$ is not relevant for them (see theorem on critical purity).

Potential circularity — RESOLVED (T-191)

The choice of anchor depends on the interiority level (L2), which is defined via R, which is defined via φ. This apparent circularity is resolved by the following convergence theorem.

Theorem T-191 (Convergence of the φ-tower) [T]

Theorem T-191 [T]

The iterative self-modeling tower $\varphi^{(0)}, \varphi^{(1)}, \varphi^{(2)}, \ldots$ converges in operator norm to the unique self-consistent self-model $\varphi^*$, starting from any initial anchor. The convergence is exponential with rate bounded by the Fano contraction $\alpha = 2/3$.

Formulation. Define the iterative scheme:

1. $\varphi^{(0)}(\Gamma) := I/7$ (maximally mixed anchor — no prior knowledge)
2. $\varphi^{(n+1)}(\Gamma) := \lim_{\tau \to \infty} \exp\!\bigl(\tau \cdot \mathcal{L}_\Omega^{(n)}\bigr)[\Gamma]$, where $\mathcal{L}_\Omega^{(n)}$ uses $\varphi^{(n)}$ as the regeneration target

Then: $\|\varphi^{(n)} - \varphi^*\|_{\mathrm{op}} \leq q^n \cdot \|\varphi^{(0)} - \varphi^*\|_{\mathrm{op}}, \quad q = \frac{\kappa_{\max}}{\lambda_{\mathrm{gap}} + \kappa_{\min}} < 1$

Proof.

Step 1 (Well-definedness of each iterate). For fixed $\varphi^{(n)}$, the Liouvillian $\mathcal{L}_\Omega^{(n)} = \mathcal{L}_0 + \kappa(\Gamma) \cdot (\varphi^{(n)}(\Gamma) - \Gamma) \cdot g_V(P)$ is a contractive CPTP semigroup generator on the finite-dimensional space $\mathcal{D}(\mathbb{C}^7)$. By primitivity of $\mathcal{L}_0$ (T-39a [T]) and the addition of a contractive regeneration term, $\mathcal{L}_\Omega^{(n)}$ has a unique stationary state $\rho^{*(n)}$ (by the Perron–Frobenius theorem for positive semigroups on finite-dimensional matrix algebras, Evans 1977). Therefore $\varphi^{(n+1)}$ is well-defined. $\checkmark$

Step 2 (Contraction of the iteration map). Define $\Psi: \mathcal{B}(\mathcal{D}(\mathbb{C}^7)) \to \mathcal{B}(\mathcal{D}(\mathbb{C}^7))$ by $\Psi(\varphi) := \lim_{\tau \to \infty} \exp(\tau \cdot \mathcal{L}_\Omega[\cdot; \varphi])$. For two candidate self-models $\varphi_1, \varphi_2$:

$\|\Psi(\varphi_1) - \Psi(\varphi_2)\|_{\mathrm{op}} \leq \frac{\kappa_{\max} \cdot \sup_\Gamma \|\varphi_1(\Gamma) - \varphi_2(\Gamma)\|_F}{\lambda_{\mathrm{gap}} + \kappa_{\min}}$

This follows from the resolvent estimate: the stationary state of $\mathcal{L}_0 + \mathcal{R}$ depends on $\mathcal{R}$ through the resolvent $(\mathcal{L}_0 - z)^{-1}$, and the spectral gap $\lambda_{\mathrm{gap}}$ of $\mathcal{L}_0$ bounds the resolvent norm at $z = 0$ by $1/\lambda_{\mathrm{gap}}$.

Since $\varphi_i$ are replacement channels: $\|\varphi_1(\Gamma) - \varphi_2(\Gamma)\|_F = k \cdot \|\rho^{*}_1 - \rho^{*}_2\|_F \leq \|\varphi_1 - \varphi_2\|_{\mathrm{op}}$ (with $k = 1 - R \leq 1$). Therefore:

$\|\Psi(\varphi_1) - \Psi(\varphi_2)\|_{\mathrm{op}} \leq q \cdot \|\varphi_1 - \varphi_2\|_{\mathrm{op}}, \quad q = \frac{\kappa_{\max}}{\lambda_{\mathrm{gap}} + \kappa_{\min}}$

Step 3 (Contractivity $q < 1$). The condition $q < 1$ is equivalent to $\kappa_{\max} < \lambda_{\mathrm{gap}} + \kappa_{\min}$. Since $\kappa_{\min} = \kappa_{\mathrm{bootstrap}} = \omega_0/7 > 0$ (T-59 [T]) and $\kappa_{\max} < \lambda_{\mathrm{gap}}$ (the clustering condition from T-117, verified in T-96 [T]):

$q = \frac{\kappa_{\max}}{\lambda_{\mathrm{gap}} + \kappa_{\min}} < \frac{\lambda_{\mathrm{gap}}}{\lambda_{\mathrm{gap}} + \kappa_{\min}} < 1 \quad \checkmark$

Step 4 (Banach convergence). The space of CPTP operators on $\mathcal{D}(\mathbb{C}^7)$ with the operator norm is a complete metric space (closed subset of the finite-dimensional space $\mathcal{B}(M_7(\mathbb{C}))$). By the Banach fixed-point theorem, $\Psi$ has a unique fixed point $\varphi^*$, and the iterates $\varphi^{(n)} = \Psi^n(\varphi^{(0)})$ converge exponentially:

$\|\varphi^{(n)} - \varphi^*\|_{\mathrm{op}} \leq \frac{q^n}{1-q} \|\varphi^{(1)} - \varphi^{(0)}\|_{\mathrm{op}}$

Step 5 (Independence of initial anchor). The fixed point $\varphi^*$ is unique (Step 4). Starting from $\varphi^{(0)} = I/7$ or from any other CPTP anchor $\tilde{\varphi}^{(0)}$:

$\|\Psi^n(\varphi^{(0)}) - \Psi^n(\tilde{\varphi}^{(0)})\|_{\mathrm{op}} \leq q^n \|\varphi^{(0)} - \tilde{\varphi}^{(0)}\|_{\mathrm{op}} \to 0$

Both sequences converge to the same $\varphi^*$. The choice of initial anchor is irrelevant. $\blacksquare$

Corollary (Resolution of circularity). The definition hierarchy $\Omega \to \mathcal{L}_\Omega \to \rho^*_{\mathrm{diss}} \to R \to \varphi$ is not circular: starting from $\varphi^{(0)} = I/7$ (which depends on nothing), each iterate $\varphi^{(n+1)}$ depends only on $\varphi^{(n)}$, and the limit $\varphi^*$ is independent of the starting point. The apparent circularity was an artifact of presenting the converged state as if it were the definition.

Corollary (SAD tower convergence). The Self-Awareness Depth tower $\mathrm{SAD} = 1, 2, 3$ (T-142 [T]) corresponds to the first three iterates $\varphi^{(1)}, \varphi^{(2)}, \varphi^{(3)}$. Since $q < 1$, the differences $\|\varphi^{(n+1)} - \varphi^{(n)}\|$ decrease geometrically. By T-142 [T], SAD$_{\max} = 3$ — the fourth iterate $\varphi^{(4)}$ would require $P > 9/14 > 3/7$, violating $R \geq 1/3$. The tower terminates at finite depth, making convergence trivially satisfied for the physically realizable levels.

Dependencies: T-39a [T] (primitivity, spectral gap), T-59 [T] ($\kappa_{\mathrm{bootstrap}}$), T-96 [T] ($\kappa < \kappa_{\max}$), T-124c [T] (attractor uniqueness), T-142 [T] (SAD$_{\max} = 3$). Standard mathematics: Banach fixed-point theorem, Perron–Frobenius for positive semigroups (Evans 1977).

2.7 Spectral formula for φ (explicit computation)

Key result

This section provides an explicit computable formula for the operator $\varphi$ via the spectral decomposition of the logical Liouvillian $\mathcal{L}_\Omega$. This makes the theory fully constructive.

Theorem 2.3 (Spectral formula for φ):

$$\varphi(\Gamma) = \sum_{k: \mathrm{Re}(\lambda_k) = 0} \langle L_k | \Gamma \rangle R_k$$

where:

• $\{R_k, L_k\}$ — right and left eigenvectors of $\mathcal{L}_\Omega$
• $\lambda_k$ — eigenvalues of $\mathcal{L}_\Omega$
• Sum over $k$ with $\mathrm{Re}(\lambda_k) = 0$ (stationary modes)
• $\langle L_k | \Gamma \rangle := \mathrm{Tr}(L_k^\dagger \cdot \Gamma_{\text{vec}})$ — inner product in vectorized space

Proof:

1. By definition (see Theorem: φ as stationary distribution): $\varphi(\Gamma) = \lim_{\tau \to \infty} e^{\tau \mathcal{L}_\Omega}[\Gamma]$

2. Decomposition into eigenfunctions: $e^{\tau \mathcal{L}_\Omega}[\Gamma] = \sum_k e^{\lambda_k \tau} \langle L_k | \Gamma \rangle R_k$

3. As $\tau \to \infty$:

   • $\mathrm{Re}(\lambda_k) < 0$: $e^{\lambda_k \tau} \to 0$ (decay)
   • $\mathrm{Re}(\lambda_k) > 0$: excluded by CPTP structure (divergence impossible)
   • $\mathrm{Re}(\lambda_k) = 0$: $e^{\lambda_k \tau}$ bounded (stationary modes)

4. Therefore: $\varphi(\Gamma) = \sum_{k: \mathrm{Re}(\lambda_k) = 0} \langle L_k | \Gamma \rangle R_k \quad \blacksquare$

Simplification under primitivity of linear part [T]

Primitivity of the linear part $\mathcal{L}_0$ ensures a spectral gap. In the vicinity of the non-trivial attractor $\rho^*_\Omega$ the formula simplifies to projection onto the zero mode ($\lambda_0 = 0$, multiplicity 1):

$$\varphi(\Gamma) = \langle L_0 | \Gamma \rangle \, R_0 = \mathrm{Tr}(L_0^\dagger\,\Gamma) \cdot \rho^*_\Omega$$

where $R_0 = \rho^*_\Omega$ is the stationary state of the full dynamics (categorical self-model, Definition 1), $L_0$ is the corresponding left eigenvector.

Algorithm for computing φ (spectral method):

mount core.math.linalg.{StaticMatrix, StaticVector, eig, inverse};

/// Compute φ(Γ) via spectral decomposition of the logical Liouvillian.
///
/// The Liouvillian ℒ_Ω is vectorised as a 49×49 superoperator; φ projects Γ
/// onto the kernel (stationary modes with Re(λ) ≈ 0).
public pure fn compute_phi_spectral(
    gamma:   &StaticMatrix,
    l_omega: &StaticMatrix,
) -> StaticMatrix<Complex, 7, 7>
{
    let (eigvals, r_vectors) = eig(l_omega);
    let l_vectors = inverse(&r_vectors).unwrap().transpose();        // left eigenvectors

    let gamma_vec = gamma.flatten();                                 // 49-vector
    let mut phi_vec = StaticVector<Complex, 49>.zeros();
    const TOL: Float = 1.0e-10;

    for k in 0..49 {
        if eigvals[k].real().abs() < TOL {                           // stationary mode
            let coeff = l_vectors.column(k).conjugate().dot(&gamma_vec);
            phi_vec  = &phi_vec + r_vectors.column(k) * coeff;
        }
    }

    let phi_gamma = phi_vec.reshape<7, 7>();
    let hermitised = (&phi_gamma + phi_gamma.adjoint()) / Complex.from_real(2.0);
    &hermitised / hermitised.trace()                                  // renormalise Tr = 1
}

Computational complexity:

Operation	Complexity
Spectral decomposition of $\mathcal{L}_\Omega$	$O(N^6)$ for $N=7$, i.e. $O(49^3) \approx 10^5$
Projection onto stationary modes	$O(N^4)$
Total complexity	$O(N^6)$, but $\mathcal{L}_\Omega$ is computed once

Relation to contracting form:

The spectral formula is equivalent to the canonical definition with the correct choice of $\mathcal{L}_\Omega$. Advantages of the spectral form:

1. Explicit computation — no iterations required
2. Uniqueness — no dependence on initial state
3. Categorical consistency — corresponds to the left adjoint to inclusion $\mathrm{Sub}(\Gamma)$

2.8 n-th order reflection (for L3/L4)

Extension for post-reflective levels

Defining levels L3 and L4 of the interiority hierarchy requires an iterated operator φ.

Definition 2.12 (Iterated operator φ):

$$\varphi^{(n)}(\Gamma) := \underbrace{\varphi \circ \varphi \circ \cdots \circ \varphi}_{n}(\Gamma)$$

with $\varphi^{(0)}(\Gamma) := \Gamma$.

Definition 2.13 (n-th order reflection):

$$R^{(n)}(\Gamma) := \mathrm{Fid}(\varphi^{(n-1)}(\Gamma), \varphi^{(n)}(\Gamma))$$

where $\mathrm{Fid}(\rho_1, \rho_2) := |\mathrm{Tr}(\sqrt{\sqrt{\rho_1}\rho_2\sqrt{\rho_1}})|^2$ — fidelity.

Thresholds for L3/L4:

Transition	Threshold	Universal formula
L1→L2	$R^{(1)} \geq 1/3$	$X^{(2)}_{\text{th}} = 1/3$
L2→L3	$R^{(2)} \geq 1/4$	$X^{(3)}_{\text{th}} = 1/4$
L3→L4	$\lim_n R^{(n)} > 0$	—

Algorithm for computing $R^{(2)}$:

mount core.math.linalg.matrix_sqrt;

/// Second-order reflection R^(2) = Fid(φ(Γ), φ(φ(Γ))).
/// Fidelity F(ρ₁, ρ₂) = |Tr √(√ρ₁ ρ₂ √ρ₁)|².
public pure fn compute_r2(
    gamma:   &StaticMatrix,
    l_omega: &StaticMatrix,
) -> Float { 0.0 <= self && self <= 1.0 }
{
    let phi_gamma     = compute_phi_spectral(gamma, l_omega);
    let phi_phi_gamma = compute_phi_spectral(&phi_gamma, l_omega);

    let sqrt_phi = matrix_sqrt(&phi_gamma);
    let inner = sqrt_phi.matmul(&phi_phi_gamma).matmul(&sqrt_phi);
    let trace_sqrt = matrix_sqrt(&inner).trace().abs();
    (trace_sqrt * trace_sqrt).clamp(0.0, 1.0)
}

---

3. Theorem on existence of fixed point

3.1 Main theorem

Theorem 3.1 (Existence of reflexion fixed point):

Let $\varphi: \mathcal{D}(\mathcal{H}) \to \mathcal{D}(\mathcal{H})$ be a contracting map with constant $k < 1$:

$$\forall \Gamma_1, \Gamma_2 \in \mathcal{D}(\mathcal{H}): \|\varphi(\Gamma_1) - \varphi(\Gamma_2)\|_F \leq k \cdot \|\Gamma_1 - \Gamma_2\|_F$$

Then:

$$\exists! \, \Gamma^* \in \mathcal{D}(\mathcal{H}): \varphi(\Gamma^*) = \Gamma^*$$

and for any $\Gamma_0 \in \mathcal{D}(\mathcal{H})$:

$$\lim_{n \to \infty} \varphi^n(\Gamma_0) = \Gamma^*$$

with convergence rate:

$$\|\varphi^n(\Gamma_0) - \Gamma^*\|_F \leq k^n \cdot \|\Gamma_0 - \Gamma^*\|_F$$

Proof:

Step 1: Completeness of the space

$\mathcal{D}(\mathcal{H})$ is a closed subset of the Banach space $(\mathcal{L}(\mathcal{H}), \|\cdot\|_F)$.

Checking closedness:

• The limit of a sequence of Hermitian matrices is Hermitian
• The limit of a sequence of positive semi-definite matrices is positive semi-definite (closed cone)
• $\mathrm{Tr}$ is a continuous function, $\mathrm{Tr}(\lim \rho_n) = \lim \mathrm{Tr}(\rho_n) = 1$

Therefore, $\mathcal{D}(\mathcal{H})$ is a complete metric space.

Step 2: Applying the Banach theorem

$\varphi$ is a contracting map on a complete metric space.

By the Banach fixed point theorem:

• There exists a unique fixed point $\Gamma^*$
• Iterations converge to $\Gamma^*$ for any initial condition

Step 3: Structure preservation

Show that $\Gamma^* \in \mathcal{D}(\mathcal{H})$:

$\varphi: \mathcal{D}(\mathcal{H}) \to \mathcal{D}(\mathcal{H})$ (by construction of $\varphi_k$ or as CPTP map).

$\Gamma^* = \lim_{n \to \infty} \varphi^n(\Gamma_0)$, where $\Gamma_0 \in \mathcal{D}(\mathcal{H})$ and $\varphi^n(\Gamma_0) \in \mathcal{D}(\mathcal{H})$ for all $n$.

$\mathcal{D}(\mathcal{H})$ is closed $\Rightarrow \Gamma^* \in \mathcal{D}(\mathcal{H})$. ∎

3.2 Approximate fixed points

Definition 3.1 ($\varepsilon$-fixed point):

$\Gamma$ is called an $\varepsilon$-fixed point if $\|\Gamma - \varphi(\Gamma)\|_F < \varepsilon$.

Theorem 3.2 (Existence of $\varepsilon$-fixed point for non-contracting $\varphi$):

Let $\varphi: \mathcal{D}(\mathcal{H}) \to \mathcal{D}(\mathcal{H})$ be a continuous map (not necessarily contracting).

Then for any $\varepsilon > 0$ there exists $\Gamma_\varepsilon \in \mathcal{D}(\mathcal{H})$ such that:

$$\|\Gamma_\varepsilon - \varphi(\Gamma_\varepsilon)\|_F < \varepsilon$$

Proof:

Consider the family of maps:

$$\varphi_\lambda(\Gamma) := \lambda \cdot \varphi(\Gamma) + (1 - \lambda) \cdot \Gamma_c$$

where $\Gamma_c = I/N$ is the center of $\mathcal{D}(\mathcal{H})$.

For $\lambda < 1$: $\varphi_\lambda$ is a contracting map with constant $\lambda$ (analogously to Lemma 2.1).

By Theorem 3.1: $\exists \, \Gamma^*_\lambda : \varphi_\lambda(\Gamma^*_\lambda) = \Gamma^*_\lambda$.

Consider:

$$\|\Gamma^*_\lambda - \varphi(\Gamma^*_\lambda)\|_F = \|\Gamma^*_\lambda - \varphi_\lambda(\Gamma^*_\lambda) + \varphi_\lambda(\Gamma^*_\lambda) - \varphi(\Gamma^*_\lambda)\|_F$$ $$= \|\varphi_\lambda(\Gamma^*_\lambda) - \varphi(\Gamma^*_\lambda)\|_F \quad (\Gamma^*_\lambda \text{ is a fixed point of } \varphi_\lambda)$$ $$= \|\lambda \cdot \varphi(\Gamma^*_\lambda) + (1-\lambda) \cdot \Gamma_c - \varphi(\Gamma^*_\lambda)\|_F = (1-\lambda) \cdot \|\Gamma_c - \varphi(\Gamma^*_\lambda)\|_F$$ $$\leq (1-\lambda) \cdot \mathrm{diam}(\mathcal{D}(\mathcal{H}))$$

where $\mathrm{diam}(\mathcal{D}(\mathcal{H})) = \sup_{\rho_1, \rho_2} \|\rho_1 - \rho_2\|_F \leq \sqrt{2}$ (diameter of the density matrix space).

Choosing $\lambda = 1 - \varepsilon / (2 \cdot \mathrm{diam}(\mathcal{D}(\mathcal{H})))$, we get:

$$\|\Gamma^*_\lambda - \varphi(\Gamma^*_\lambda)\|_F < \varepsilon$$

∎

3.3 Contraction conditions for CPTP maps

Theorem 3.3 (Contraction criterion):

A CPTP map $\mathcal{P}$ is contracting with constant $k < 1$ if and only if:

$$\exists \, \rho_{\text{inv}} \in \mathcal{D}(\mathcal{H}) : \mathcal{P}(\rho_{\text{inv}}) = \rho_{\text{inv}} \land \mathrm{spec}(\mathcal{P}|_{\rho_{\text{inv}}^\perp}) \subset \{z \in \mathbb{C} : |z| < 1\}$$

where $\mathcal{P}|_{\rho_{\text{inv}}^\perp}$ is the restriction of $\mathcal{P}$ to the orthogonal complement of $\rho_{\text{inv}}$.

Interpretation: $\mathcal{P}$ is contracting if it has a unique invariant state and all perturbations decay.

Examples of contracting CPTP:

1. Thermalization:

$$\mathcal{P}_{\text{therm}}(\rho) = \lambda \rho + (1-\lambda) \rho_{\text{thermal}}, \quad \lambda < 1$$

2. Depolarizing channel:

$$\mathcal{P}_{\text{depol}}(\rho) = p \rho + (1-p) \frac{I}{N}, \quad p < 1$$

3. Amplitude damping:

$$\mathcal{P}_{\text{damp}}(\rho) = K_0 \rho K_0^\dagger + K_1 \rho K_1^\dagger$$ $$K_0 = |0\rangle\langle 0| + \sqrt{1-\gamma}|1\rangle\langle 1|, \quad K_1 = \sqrt{\gamma}|0\rangle\langle 1|$$

Contracting for $\gamma > 0$.

---

4. Relation to reflection measure R

4.1 Definition of R

Definition 4.1 (Reflection measure):

$$R(\Gamma) := \frac{1}{7P(\Gamma)}, \quad P = \mathrm{Tr}(\Gamma^2)$$

Equivalent form: $R = 1 - \|\Gamma - \rho^*_{\mathrm{diss}}\|_F^2 / P$, where $\rho^*_{\mathrm{diss}} = I/7$, $\|\Gamma\|_F = \sqrt{P}$ (square root of purity).

Distinction between R_canonical and R_φ

$R_{\text{canonical}} := 1/(7P)$ is the canonical definition used in all thresholds ($R_{\text{th}} = 1/3$). It is a measure of proximity to the maximally mixed state $I/7$, NOT a measure of quality of self-modeling.

The quality of self-modeling is defined separately (formerly also written $Q_\varphi$; unified notation — the three working forms of R):

$$R_\varphi(\Gamma) := 1 - \frac{\|\Gamma - \varphi(\Gamma)\|^2_F}{\|\Gamma\|^2_F}$$

Comparison at characteristic states:

• At $\Gamma = I/7$ (dissipative attractor): $R_{\text{canonical}} = 1$, $R_\varphi = 1$.
• At a pure state ($P = 1$): $R_{\text{canonical}} = 1/7$, $R_\varphi$ depends on $\varphi$.

Sections 4.2–4.3 below analyse the convergence of the self-model quality and are written in $R_\varphi$; in all other sections and in threshold conditions $R = R_{\text{canonical}} = 1/(7P)$.

4.2 Convergence of R_φ as fixed point is approached

Theorem 4.1 ($R_\varphi \to 1$ as $\Gamma \to \Gamma^*$):

Let $\varphi$ be a contracting map with fixed point $\Gamma^*$.

Then:

$$\lim_{\Gamma \to \Gamma^*} R_\varphi(\Gamma) = 1$$

Proof:

As $\Gamma \to \Gamma^*$:

$$\|\Gamma - \varphi(\Gamma)\|_F \to \|\Gamma^* - \varphi(\Gamma^*)\|_F = \|\Gamma^* - \Gamma^*\|_F = 0$$

Therefore:

$$R_\varphi(\Gamma) = 1 - \frac{\|\Gamma - \varphi(\Gamma)\|^2_F}{\|\Gamma\|^2_F} \to 1 - \frac{0}{P(\Gamma^*)} = 1$$

(The denominator is bounded away from zero for any density matrix: $\|\Gamma^*\|^2_F = P(\Gamma^*) \geq 1/N > 0$.) ∎

4.3 Estimate of rate of convergence of R_φ

Theorem 4.2 (Rate of convergence of $R_\varphi$):

For contracting $\varphi$ with constant $k$ and sequence $\Gamma_n = \varphi^n(\Gamma_0)$:

$$1 - R_\varphi(\Gamma_n) \leq 4 k^{2n} \cdot \frac{\|\Gamma_0 - \Gamma^*\|^2_F}{P_{\min}}$$

where $P_{\min} = \min_{\rho \in \mathcal{D}(\mathcal{H})} P(\rho) = 1/N$.

Proof:

$$1 - R_\varphi(\Gamma_n) = \frac{\|\Gamma_n - \varphi(\Gamma_n)\|^2_F}{\|\Gamma_n\|^2_F} = \frac{\|\varphi^n(\Gamma_0) - \varphi^{n+1}(\Gamma_0)\|^2_F}{P(\Gamma_n)}$$ $$\leq \frac{(k^n \cdot \|\Gamma_0 - \varphi(\Gamma_0)\|_F)^2}{P(\Gamma_n)} \quad \text{(contraction)}$$ $$\leq \frac{k^{2n} \cdot (\|\Gamma_0 - \Gamma^*\|_F + \|\Gamma^* - \varphi(\Gamma_0)\|_F)^2}{P_{\min}}$$ $$\leq \frac{k^{2n} \cdot (\|\Gamma_0 - \Gamma^*\|_F + k \cdot \|\Gamma^* - \Gamma_0\|_F)^2}{P_{\min}} = \frac{k^{2n} \cdot (1 + k)^2 \cdot \|\Gamma_0 - \Gamma^*\|^2_F}{P_{\min}}$$

For $k < 1$: $(1 + k)^2 < 4$, giving the bound:

$$1 - R_\varphi(\Gamma_n) \leq \frac{4 \cdot k^{2n} \cdot \|\Gamma_0 - \Gamma^*\|^2_F}{P_{\min}}$$

∎

Strengthening: unconditional convergence [T]

Primitivity of $\mathcal{L}_\Omega$ guarantees exponential convergence $R_\varphi \to 1$ for any initial state $\Gamma_0 \in \mathcal{D}(\mathbb{C}^7)$, without additional conditions on initial data.

4.4 Relation of R to consciousness measure C

Theorem 4.3 (reflection factors at the fixed point):

The consciousness measure $C = \Phi \times R$ [Т T-140] takes the canonical $R = 1/(7P)$; the convergence results of §4.2–4.3 concern the self-model quality $R_\varphi$. At the fixed point the two factor cleanly:

$$R_\varphi(\Gamma^*) = 1, \qquad C(\Gamma^*) = \Phi(\Gamma^*) \cdot \frac{1}{7P(\Gamma^*)},$$

with $R(\Gamma^*) = 1/(7P(\Gamma^*)) \in [1/3, 1/2)$ whenever $\Gamma^*$ lies inside the conscious window. (An earlier reading substituted $R_\varphi(\Gamma^*) = 1$ into $C$, yielding $C(\Gamma^*) = \Phi(\Gamma^*)$; under the canonical $R$ that value is attained only at $\Gamma^* = I/7$, where $\Phi = 0$ — the reading is withdrawn.)

On notation

Differentiation $D_{\text{diff}} \geq D_{\min} = 2$ enters as a separate viability condition, not as a factor of $C$.

Corollary: Ideal self-knowledge ($\Gamma = \Gamma^*$) maximizes the self-model-quality factor $R_\varphi$ — the quantity carried by the meaning functional and the $R^{(n)}$-towers — while the canonical factor of $C$ is pinned by purity.

4.5 The derivative of the self-model: Dφ

Everything dynamical about $R_\varphi$ is carried by one object: how the self-model responds to a change of state. The exact flow identity of the three working forms of R,

$$\frac{dR_\varphi}{d\tau} = (1 - R_\varphi)\,\frac{\dot P}{P} \;-\; \frac{2}{P}\,\bigl\langle \Gamma - \varphi(\Gamma),\; (\mathrm{Id} - D\varphi)[\dot\Gamma] \bigr\rangle_F,$$

contains the derivative $D\varphi$ of the self-model map — the response kernel of self-modelling. This subsection computes it for the canonical family and verifies the identity by an independent route.

Definition ($D\varphi$) [D]. The Gateaux derivative of $\varphi$ at $\Gamma$ along a tangent direction $V$ (Hermitian, traceless): $D\varphi[V] := \lim_{\varepsilon \to 0} (\varphi(\Gamma + \varepsilon V) - \varphi(\Gamma))/\varepsilon$. For a Lipschitz (contracting) $\varphi$ it exists almost everywhere (Rademacher, finite dimension); for the smooth families below — everywhere.

Theorem 4.4 (Dφ of the canonical family; T-249) [T]

For the canonical dissipative family $\varphi(\Gamma) = (1 - k(\Gamma))\,\Gamma + k(\Gamma)\, I/7$ with $k = 1 - 1/(7P)$:

$$D\varphi[V] = R\,V \;-\; \frac{2}{7P^2}\,\langle \Gamma, V\rangle_F\,(\Gamma - I/7), \qquad R = \frac{1}{7P}.$$

$D\varphi$ preserves the Hermitian-traceless tangent space and is $G_2$-equivariant.

Proof. Product rule on $\varphi = (1-k)\Gamma + k\,I/7$: the $\Gamma$-slot contributes $(1-k)V = RV$; the $k$-slot contributes $Dk[V]\,(I/7 - \Gamma)$ with $Dk[V] = \tfrac{1}{7P^2}DP[V] = \tfrac{2}{7P^2}\langle\Gamma, V\rangle_F$ (since $DP[V] = 2\,\mathrm{Tr}(\Gamma V)$); the reference $I/7$ is constant. Trace preservation: $\mathrm{Tr}\,D\varphi[V] = R\cdot 0 - \tfrac{2}{7P^2}\langle\Gamma,V\rangle\,\mathrm{Tr}(\Gamma - I/7) = 0$. Equivariance: $P$, $R$, $\langle\cdot,\cdot\rangle_F$ are unitarily invariant and $I/7$ is the unique $G_2$-invariant state, so $D\varphi$ commutes with conjugation by $U \in G_2$. $\blacksquare$

Two-route consistency (part of T-249) [T]. For this family the closed form $R_\varphi = 1 - (1-R)^3$ holds pointwise, whence directly $\dot R_\varphi = -3(1-R)^2 \dot P/(7P^2)$. Substituting Theorem 4.4 into the flow identity must give the same. Indeed, with $\Delta = \Gamma - \varphi(\Gamma) = k(\Gamma - I/7)$:

$$(\mathrm{Id} - D\varphi)[\dot\Gamma] = k\,\dot\Gamma + \frac{\dot P}{7P^2}(\Gamma - I/7), \qquad \langle \Gamma - I/7, \dot\Gamma\rangle_F = \frac{\dot P}{2}$$

(the latter by trace preservation, $\mathrm{Tr}\,\dot\Gamma = 0$), and $\|\Gamma - I/7\|_F^2 = P - 1/7 = kP$. Collecting terms with $1 - R_\varphi = k^3$:

$$\dot R_\varphi = k^3\frac{\dot P}{P} - k^2\frac{\dot P}{P} - 2k^2 R\,\frac{\dot P}{P} = \frac{\dot P}{P}\,k^2\,(k - 1 - 2R) = -3R\,k^2\,\frac{\dot P}{P} = -\frac{3(1-R)^2\dot P}{7P^2}. \;\blacksquare$$

Machine verification. Finite-difference check of $D\varphi$ and three-route agreement (identity / closed form / numerical derivative along random density-matrix paths): maximal discrepancy $\sim 10^{-10}$ over $200$ random states.

4.6 The bandwidth theorem for R_φ

Theorem 4.5 (Bandwidth bound; T-250) [Т — for φ differentiable along the trajectory]

Along any trajectory of the dynamics,

$$\Bigl|\dot R_\varphi - (1 - R_\varphi)\frac{\dot P}{P}\Bigr| \;\leq\; \frac{2}{\sqrt{P}}\,\sqrt{1 - R_\varphi}\;C_\varphi\,\|\dot\Gamma\|_F, \qquad C_\varphi := \|\mathrm{Id} - D\varphi\|_{\mathrm{op}}.$$

For the canonical family, $C_\varphi \leq (1-R) + 2R\sqrt{1-R}$ (numerically $\approx 1.21$ across the conscious window).

Proof. Cauchy–Schwarz on the flow identity, with $\|\Delta\|_F = \sqrt{(1 - R_\varphi)P}$ by the definition of $R_\varphi$. For the family bound: $\|(\mathrm{Id} - D\varphi)[V]\| \leq (1-R)\|V\| + \tfrac{2}{7P^2}\,\|\Gamma\|\,\|\Gamma - I/7\|\,\|V\|$ and $\|\Gamma\| = \sqrt P$, $\|\Gamma - I/7\| = \sqrt{kP} = \sqrt{1-R}\,\sqrt P$, so the second term is $2\sqrt{1-R}/(7P) = 2R\sqrt{1-R}$. $\blacksquare$

Corollary (path-length law) [T]. Let $u := \sqrt{1 - R_\varphi}$ — the mismatch amplitude. On segments with $\dot P = 0$:

$$\bigl|u(\tau_2) - u(\tau_1)\bigr| \;\leq\; \frac{C_\varphi}{\sqrt P}\,\int_{\tau_1}^{\tau_2}\|\dot\Gamma\|_F\,d\tau$$

— reorganizing the self-model is paid for in state-space path length. Proof: $\dot u = -\dot R_\varphi/(2u)$, and the theorem's right side divided by $2u$ collapses to $C_\varphi\|\dot\Gamma\|/\sqrt P$. $\blacksquare$

Remark (discrete instance). Theorem 4.2 above is exactly this law along the $\varphi$-tower: $u(\Gamma_n) \leq 2k^n\|\Gamma_0 - \Gamma^*\|_F/\sqrt{P_{\min}}$, with the geometric path $\sum_n \|\Gamma_{n+1} - \Gamma_n\| \leq \|\Gamma_1 - \Gamma_0\|/(1-k)$ (T-191 convergence). The continuous and discrete forms are two readings of one bandwidth constraint.

Remark ($G_2$). $R_\varphi$, $u$, $C_\varphi$, $\|\dot\Gamma\|_F$ are $G_2$-invariant; the law is observer-independent.

Numerical anchor [I]. At the psychedelic-peak profile of altered states ($P \approx 0.32$; $R_\varphi: 0.45 \to 0.25$): $\Delta u = \sqrt{0.75} - \sqrt{0.55} \approx 0.124$, so the trajectory must traverse at least $\Delta u\,\sqrt P/C_\varphi \approx 0.058$ Frobenius units of state motion between onset and peak. The collapse of the "I" costs actual movement of the state; a short (micro-dose) trajectory cannot produce it.

4.7 Mechanisms: dissolution, training, two timescales

Ego dissolution [C]. Fast state motion with a lagging self-model — $\|\dot\Gamma\|$ large while $(\mathrm{Id} - D\varphi)[\dot\Gamma]$ stays comparable to $\dot\Gamma$ — drives $u$ up at the bandwidth-permitted rate: the phenomenological "φ does not have time to restructure" becomes the quantitative statement that $u$ grows no faster, and generically as fast, as $C_\varphi\|\dot\Gamma\|/\sqrt P$.

Two-timescale training model [С structure; [I] numbers]. Let the self-model carry a trainable target: $\varphi_\theta(\Gamma) = (1-k)\Gamma + k\,\rho_\theta$, with $\theta$ evolving on the slow timescale of $H_{\text{eff}}$-restructuring (procedural memory; the T-155 learning channel). Then $R_\varphi = 1 - k^2\|\Gamma - \rho_\theta\|_F^2/P$, and training that aligns $\rho_\theta$ with the practiced state raises the baseline $R_\varphi$ at fixed $P$ — the mechanism behind the cumulative shift in the shamatha progression. The samādhi signature ($P$ and $R_\varphi$ rising simultaneously) is thereby resolved: the information the categorical $\varphi$ carries beyond purity is exactly the learned target $\rho_\theta$.

Training law (exact within the alignment model) [C]. If the slow variable follows the gradient-alignment flow $\dot\rho_\theta = 2\eta\,(\bar\Gamma - \rho_\theta)$ toward a practiced state $\bar\Gamma$ (the gradient of the alignment loss $\lVert\bar\Gamma - \rho_\theta\rVert_F^2$ at mobility $\eta$), the solution is the convex path $\rho_\theta(t) = (1 - e^{-2\eta t})\,\bar\Gamma + e^{-2\eta t}\rho_\theta(0)$ — automatically a density matrix for every $t$ — and the baseline self-model quality obeys the exponential saturation law

$$R_\varphi^{\text{base}}(t) \;=\; 1 \;-\; \frac{k^2\,e^{-4\eta t}\,\lVert\bar\Gamma - \rho_\theta(0)\rVert_F^2}{P}.$$

Practice approaches its ceiling exponentially, at twice the flow rate ($e^{-4\eta t}$: the distance enters squared), with the rate constant set by the slow T-155 channel — the cumulative shift of the shamatha progression acquires a closed form. Status [C] at the gradient-alignment model; within the model the law is exact (machine-checked against direct integration of the flow, $2\cdot 10^5$ steps).

The gate condition and the threshold. Define (G) [D]: the $K = 3$ channel-class discriminator factors through the self-model readout, with classification accuracy bounded below by $R_\varphi$. Statement [C at (G)]: (G) implies $R_{\varphi,\text{th}} = 1/3$ — Bayesian plurality among three equiprobable alternatives requires accuracy above $1/K$.

Register of the module (closed). Both items of the original register are closed below: (i) the gate is now a theorem-level bound — §4.9 (T-252); the residual freedom is the accuracy model $A_D$ and the exact placement of the working threshold inside the derived band [C]; (ii) $D\varphi$ for implicitly defined $\varphi$ — §4.8 (T-251); the residual condition is $C^1$-smoothness of the abstract categorical generator in the Bures topology [C], satisfied by every corpus-realized family.

4.8 Dφ for implicitly defined self-models

Theorem 4.4 covers the explicit canonical family. The categorical $\varphi$, however, is given implicitly — as the fixed point of a generator whose iteration is the T-191 tower. Its derivative follows from the implicit function theorem with a Neumann series — which is also the categorical answer: differentiating a reflector along its universal property is inverting $\mathrm{Id}$ minus the derivative in the model slot.

Theorem 4.6 (Dφ of an implicit self-model; T-251) [T]

Let $G: \mathcal{D}(\mathbb{C}^7) \times \mathcal{D}(\mathbb{C}^7) \to \mathcal{D}(\mathbb{C}^7)$ be $C^1$ with $\sup \lVert D_2 G\rVert_{\mathrm{op}} \leq q < 1$ (uniform contraction in the model slot), and let $\varphi(\Gamma)$ be the unique fixed point of $\rho \mapsto G(\Gamma, \rho)$. Then $\varphi$ is $C^1$ and, evaluated at $(\Gamma, \varphi(\Gamma))$,

$$D\varphi \;=\; (\mathrm{Id} - D_2 G)^{-1} \circ D_1 G \;=\; \sum_{n \geq 0} (D_2 G)^n\, D_1 G,$$

with $\lVert D\varphi\rVert_{\mathrm{op}} \leq \lVert D_1 G\rVert_{\mathrm{op}}/(1 - q)$; consequently $C_\varphi \leq 1 + \lVert D_1 G\rVert_{\mathrm{op}}/(1-q)$, and the bandwidth theorem T-250 applies to every contraction-defined self-model.

Proof. Set $F(\Gamma, \rho) := \rho - G(\Gamma, \rho)$; then $D_2 F = \mathrm{Id} - D_2 G$ is invertible by the Neumann series ($q < 1$). The finite-dimensional $C^1$ implicit function theorem yields $\varphi \in C^1$ with $D\varphi = -(D_2 F)^{-1} D_1 F = (\mathrm{Id} - D_2 G)^{-1} D_1 G$; the series and the norm bound are the Neumann expansion. $\blacksquare$

Three readings. Tower: the $n$-th term $(D_2 G)^n D_1 G$ is the sensitivity transmitted through $n$ storeys of the T-191 tower — the geometric decay of storey-sensitivities is the tower's convergence, differentiated. Degenerate check: for $G$ independent of $\rho$ ($q = 0$) the series collapses to $D_1 G$ — Theorem 4.4 (T-249) is the zeroth-order case. Machine verification: for a genuinely nonlinear generator ($q \approx 0.7$–$0.8$), the predicted $D\varphi$ and the brute-force finite-difference derivative of the re-solved fixed point agree at $\sim 10^{-10}$ (numeric $48 \times 48$ Jacobians); the degenerate case reproduces T-249 at $10^{-11}$.

4.9 The gate theorem: discrimination through the self-model

This closes the gate condition (G) of §4.7. First the structure [Т — structural reading]: in $\mathcal{L}_\Omega$ the only feedback channel computed from the self-model is the regeneration $\mathcal{R} = \kappa(\Gamma)(\varphi(\Gamma) - \Gamma)\,g_V$ — the system's sole endogenous corrective action reads the state through $\varphi(\Gamma)$. Endogenous adaptive discrimination — discrimination the system can act on — is therefore $\varphi$-mediated by construction (the identification of "adaptive" with "$\mathcal{R}$-actionable" is definitional [D]).

Theorem 4.7 (Gate bound; T-252) [T]

Let $\{E_c\}_{c=1..K}$ be any POVM implementing a $K$-hypothesis decision, and $\Delta := \Gamma - \varphi(\Gamma)$ (traceless). Then:

(a) $|\mathrm{Tr}(E_c \Delta)| \leq \tfrac{1}{2}\lVert\Delta\rVert_1$ for every $c$, and the outcome distributions satisfy $\mathrm{TV}\bigl(p(\Gamma), p(\varphi(\Gamma))\bigr) \leq \tfrac{1}{2}\lVert\Delta\rVert_1$;

(b) $\lVert\Delta\rVert_1 \leq \sqrt{48/7}\,\lVert\Delta\rVert_F = 4\sqrt{3/7}\,\sqrt{P\,(1 - R_\varphi)}$, and the constant $\sqrt{48/7} = 4\sqrt{3/7} \approx 2.619$ is tight on traceless $\mathrm{Herm}(7)$;

(c) hence the $\varphi$-mediated success probability obeys $p_\varphi \geq A_D - 2\sqrt{3/7}\,\sqrt{P(1-R_\varphi)}$, where $A_D$ is the true-state accuracy of the decision rule, and Bayesian dominance $p_\varphi > 1/K$ is guaranteed whenever

$$R_\varphi \;\geq\; 1 - \frac{7}{12P}\,\Bigl(A_D - \frac{1}{K}\Bigr)^{2}.$$

Both inequalities of the chain are individually saturated: (a) by the Jordan projector $E = \Pi_+$ of $\Delta$, (b) by the $(3,4)$-split spectrum — so no smaller constants exist.

Proof. (a) Jordan-decompose $\Delta = \Delta_+ - \Delta_-$; tracelessness gives $\mathrm{Tr}\,\Delta_+ = \mathrm{Tr}\,\Delta_- = \tfrac{1}{2}\lVert\Delta\rVert_1$; with $0 \leq E_c \leq I$: $|\mathrm{Tr}\,E_c\Delta| \leq \max(\mathrm{Tr}\,E_c\Delta_+, \mathrm{Tr}\,E_c\Delta_-) \leq \tfrac{1}{2}\lVert\Delta\rVert_1$, and $\mathrm{TV} = \tfrac{1}{2}\sum_c |\mathrm{Tr}\,E_c\Delta| \leq \tfrac{1}{2}\sum_c \mathrm{Tr}\,E_c(\Delta_+ + \Delta_-) = \tfrac{1}{2}\lVert\Delta\rVert_1$. Equality at $E = \Pi_+$ (the positive-eigenspace projector): $\mathrm{Tr}(\Pi_+\Delta) = \mathrm{Tr}\,\Delta_+ = \tfrac12\lVert\Delta\rVert_1$. (b) Maximize $\lVert\lambda\rVert_1$ over $\sum_i\lambda_i = 0$, $\sum_i\lambda_i^2 = 1$ in dimension 7: the KKT condition $\mathrm{sign}(\lambda_i) = \alpha + 2\beta\lambda_i$ forces at most one positive value $a$ ($p$ copies) and one negative $-b$ ($q$ copies), $pa = qb$; then $\lVert\lambda\rVert_1^2/\lVert\lambda\rVert_2^2 = 4pq/(p+q)$, increasing in the support size $p + q \leq 7$ and maximal at the balanced split $\{p, q\} = \{3, 4\}$: $4 \cdot 12/7 = 48/7$. Witness attaining it exactly: $\Delta^* = \mathrm{diag}(4,4,4,-3,-3,-3,-3)$ ($\lVert\Delta^*\rVert_1 = 24$, $\lVert\Delta^*\rVert_F = \sqrt{84}$, ratio $= \sqrt{48/7}$). Finally $\lVert\Delta\rVert_F^2 = P(1 - R_\varphi)$ by the definition of $R_\varphi$. (c) Substitution. $\blacksquare$

Remark (the constant is an odd-dimension effect). The generic $d$-dimensional constant is $\sqrt{4\lfloor d/2\rfloor\lceil d/2\rceil / d}$: in even dimension it equals $\sqrt d$ exactly, in odd dimension it sits strictly below. At $d = 7$ the balanced split is forced to be $(3,4)$ and gives $\sqrt{48/7} \approx 2.619 < \sqrt 7 \approx 2.646$. The naive rank bound $\sqrt 7$ (Cauchy–Schwarz) is therefore not attainable: tracelessness forbids the aligned spectra that would saturate it. (No structural weight is placed on the $(3,4)$ split here — it is the arithmetic of odd $d$, recorded per the anti-numerology register.)

Corollary (the working threshold) [С — canonical alignment]. For $K = 3$ with an ideal true-state discriminator ($A_D = 1$) the sufficient bound $R_\varphi \geq 1 - 7/(27P)$ sweeps the band $[5/54,\ 32/81]$ across the conscious window $P \in (2/7, 3/7]$ ($5/54 \approx 0.093$ at the viability edge, $32/81 \approx 0.395$ at the ceiling); the working threshold $R_{\varphi,\text{th}} = 1/3$ lies inside this derived band and is fixed at the canonical value by alignment with Char-R-III. What §4.7 posited as the bare condition (G) is now the theorem-level bound (a)–(c) with tight constants; the only remaining freedom is the accuracy model $A_D$ and the placement of the working value inside the band.

Corollary (sectoral gate: the per-channel threshold) [T]. Fix a coherence channel $(i,j)$ and its canonical three-outcome readout $E_\pm = \tfrac12(\Pi_{ij} \pm X_{ij})$, $E_0 = \mathbb 1 - \Pi_{ij}$, where $\Pi_{ij}$ projects onto $\mathrm{span}\{e_i, e_j\}$ and $X_{ij} = |i\rangle\langle j| + |j\rangle\langle i|$ (a valid POVM: $E_\pm \succeq 0$, $E_+ + E_- + E_0 = \mathbb 1$). The $\varphi$-mediated shift of each outcome probability is bounded by $|\Delta_{ij}| + \tfrac12|\Delta_{ii} + \Delta_{jj}|$, and the off-diagonal part is exactly the sectoral reflection: $|\Delta_{ij}| = |\gamma_{ij}|\sqrt{1 - R_{ij}}$ by the definition of $R_{ij}$. In the coherence-dominated regime (diagonal mismatch negligible — precisely the regime in which the sectoral form is deployed) the per-channel discrimination loss is governed by $\sqrt{1 - R_{ij}}$ alone, so the $K = 3$ dominance argument transfers channel-wise verbatim: the sectoral working threshold $R_{ij} \geq 1/3$ carries the same theorem-level gate as the global $R_\varphi$. This derives the per-channel threshold previously inherited by analogy in the unconscious.

Machine verification. $500$ random $3$-outcome POVMs: (a) and (b) hold with margin; Jordan-projector saturation of (a) at $10^{-15}$; sharp constant confirmed — random traceless search reaches $2.504 < \sqrt{48/7} = 2.6186\ldots$, the $(3,4)$-witness attains it exactly; band endpoints $5/54$ and $32/81$; sectoral identity, POVM validity and shift bound on $200$ random states.

---

5. Categorical aspect

Section status

The categorical formalism provides additional structure for understanding $\varphi$, but is not necessary for practical computations in UHM. See also categorical formalism.

5.1 Category of density matrices

DRY: Category DensityMat

The canonical definition of category DensityMat (objects — density matrices, morphisms — CPTP channels) and proof of category axioms are in Categorical formalism, §1.

Definition 5.2 (Category of CPTP channels):

$$\mathbf{CPTP} := (\mathrm{Ob}, \mathrm{Mor})$$ $$\mathrm{Ob} = \{\mathcal{H}_n = \mathbb{C}^n : n \in \mathbb{N}\} \quad \text{(objects — Hilbert spaces)}$$ $$\mathrm{Mor}(\mathcal{H}_n, \mathcal{H}_m) = \{\mathcal{P}: \mathcal{D}(\mathcal{H}_n) \to \mathcal{D}(\mathcal{H}_m) : \mathcal{P} \text{ — CPTP}\}$$

This is a well-defined category:

• Composition: $\mathcal{P} \circ \mathcal{Q}$ is CPTP if $\mathcal{P}$ and $\mathcal{Q}$ are CPTP
• Identity: $\mathrm{id}_\mathcal{H}(\rho) = \rho$ — trivial CPTP channel

5.2 φ as endomorphism

Definition 5.3 ($\varphi$ as endofunctor):

$\varphi: \mathcal{L}(\mathcal{H}) \to \mathcal{L}(\mathcal{H})$ induces an endofunctor:

$$F_\varphi: \mathbf{CPTP}|_\mathcal{H} \to \mathbf{CPTP}|_\mathcal{H}$$

On objects: $F_\varphi(\mathcal{H}) = \mathcal{H}$ (identity)

On morphisms: $F_\varphi(\mathcal{Q}) = \varphi \circ \mathcal{Q} \circ \varphi^{-1}$ (if $\varphi$ is invertible)

Problem: A general CPTP channel is not invertible.

Solution: We consider $\varphi$ as an endomorphism in the category with a single object:

Definition 5.4 (Monoid of CPTP channels):

$$\mathrm{End}(\mathcal{H}) := \mathrm{Mor}(\mathcal{H}, \mathcal{H}) \quad \text{in category } \mathbf{CPTP}$$

This is a monoid with the composition operation.

$\varphi \in \mathrm{End}(\mathcal{H})$ — an element of this monoid.

5.3 Relation to monads

Definition 5.5 (Monad of self-modeling):

Consider the functor $T: \mathbf{Set} \to \mathbf{Set}$:

$$T(X) = \mathcal{D}(\mathbb{C}^{|X|}) \quad \text{(set of density matrices of size } |X| \text{)}$$

Monad structure:

• Unit ($\eta$): $\eta_X: X \to T(X)$, $\eta_X(x) = |x\rangle\langle x|$ (pure state)
• Mult ($\mu$): $\mu_X: T(T(X)) \to T(X)$, $\mu_X(P) = \sum_{\rho \in \mathrm{supp}(P)} P(\rho) \cdot \rho$ (mixing)

$\varphi$ induces a morphism of monads:

$$\varphi^*: (T, \eta, \mu) \to (T, \eta, \mu)$$

Naturality conditions:

$$\varphi \circ \eta = \eta \quad \text{(self-observation of a pure state is pure)}$$ $$\varphi \circ \mu = \mu \circ T(\varphi) \quad \text{(consistency with mixing)}$$

Theorem 5.1 (Fixed point as monad algebra):

The fixed point $\Gamma^* = \varphi(\Gamma^*)$ defines a $T$-algebra:

$$\alpha: T(\Gamma^*) \to \Gamma^*, \quad \alpha = \mu_{\Gamma^*} \circ T(\eta_{\Gamma^*})$$

Interpretation: A system in the state of ideal self-knowledge is an "algebra over the self-modeling monad."

5.4 2-categorical structure

Definition 5.6 (2-category of quantum systems QSys):

Level	Elements
0-morphisms (objects)	Hilbert spaces $\mathcal{H}$
1-morphisms	CPTP channels $\mathcal{P}: \mathcal{D}(\mathcal{H}_1) \to \mathcal{D}(\mathcal{H}_2)$
2-morphisms	Natural transformations between channels

$\varphi$ defines a 2-cell:

$$\varphi: \mathrm{id}_{\mathcal{D}(\mathcal{H})} \Rightarrow \mathrm{id}_{\mathcal{D}(\mathcal{H})}$$

(endo-2-morphism of the identity 1-morphism)

Fixed point condition in 2-categorical language:

$\Gamma^*$ is an object such that $\varphi_{\Gamma^*} = \mathrm{id}_{\Gamma^*}$ (the 2-morphism reduces to the identity).

---

6. Corollaries and limitations

6.1 Corollaries of formalization

Corollary 6.1 (Necessity of contraction for ideal self-knowledge):

For the existence of exact $\Gamma^* = \varphi(\Gamma^*)$ it is necessary that $\varphi$ be contracting (or have an invariant subspace).

Corollary 6.2 (Approximate self-knowledge is always possible):

For any continuous $\varphi$ and any $\varepsilon > 0$ there exists an $\varepsilon$-fixed point.

Corollary 6.3 (Relation to thermodynamics):

Contracting CPTP channels correspond to systems with dissipation (attraction to equilibrium).

The fixed point of $\varphi$ is the "thermodynamic equilibrium of self-observation."

6.2 Limitations of formalization

Limitation 6.1 (Contraction requirement):

Theorem 3.1 requires $k < 1$. For $k = 1$ (isometric $\varphi$) the fixed point may not exist or may be non-unique.

Limitation 6.2 (Finite-dimensionality):

The proofs use finite-dimensionality of $\mathcal{H}$. Generalization to the infinite-dimensional case requires additional conditions (compactness of $\varphi$).

Limitation 6.3 (Stationarity):

The formalization treats $\varphi$ as a fixed operator. In a dynamical system $\varphi$ may depend on time: $\varphi = \varphi(t)$.

Open question: Does a "moving fixed point" $\Gamma^*(t)$ exist for $\varphi(t)$? See Appendix C.

6.3 Physical interpretation

Interpretation 6.1 (Self-modeling as quantum channel):

$\varphi$ = CPTP channel means that self-observation:

• Preserves positivity (does not create negative probabilities)
• Preserves normalization (total probability = 1)
• Can decrease information (does not increase distinguishability)

Interpretation 6.2 (Fixed point as self-consistency):

$\Gamma^* = \varphi(\Gamma^*)$ means: "What the system sees coincides with what it is."

This is the state of ideal self-knowledge — the system has no "blind spots."

Interpretation 6.3 (Contraction as humility):

$k < 1$ means that each act of self-observation "approaches" the truth.

The system gradually corrects its self-model, converging to an accurate representation.

6.4 Relation to UHM

Relation 6.1 (Reflexive closure):

The condition of self-observation:

$$\varphi(\Gamma) \approx \Gamma$$

is formalized as: $R(\Gamma) \geq 1 - \varepsilon$ for some $\varepsilon > 0$.

Relation 6.2 (Consciousness):

$C = \Phi \times R$ [Т T-140] includes $R$ as a factor.

At $R \to 1$: $C \to \Phi$ (maximum contribution of integration).

Relation 6.3 (No-zombie theorem):

From interiority hierarchy:

$$\mathrm{Viable}(\mathbb{H}) \Rightarrow R(\Gamma) > 0$$

The formalization of $\varphi$ ensures: $R(\Gamma) > 0 \Leftrightarrow \Gamma \neq \varphi(\Gamma)$ with finite precision.

---

7. Implementation requirements

Section status

This section contains mathematical requirements for implementing the self-modeling operator φ. Concrete architectures and code are the subject of separate specifications.

7.1 Requirements for implementing φ

Requirement 7.1 (Predictive self-modeling operator):

The implementation of $\varphi$ must satisfy:

$$\varphi(\Gamma) = k \cdot \mathcal{P}_\theta(\Gamma) + (1-k) \cdot \Gamma_{\text{prior}}$$

where:

• $k \in (0, 1)$ — contraction parameter ensuring contractivity
• $\mathcal{P}_\theta: \mathcal{D}(\mathcal{H}) \to \mathcal{D}(\mathcal{H})$ — parameterized map
• $\Gamma_{\text{prior}} = I/7$ — prior state (maximum entropy)

Implementation guarantees:

1. Output — valid density matrix (Hermitian, PSD, trace=1)
2. Contracting map at $k < 1$
3. Differentiability with respect to parameters $\theta$

Recommended method: Cholesky parameterization $\Gamma = LL^\dagger / \mathrm{Tr}(LL^\dagger)$ guarantees PSD.

7.2 Requirements for sensor encoder

Requirement 7.2 (Encoder: sensors → Γ):

$$\Gamma = \text{Encoder}_\psi(s) = \frac{L(s) \cdot L(s)^\dagger}{\mathrm{Tr}(L(s) \cdot L(s)^\dagger)}$$

where $L(s)$ is a lower-triangular matrix parameterized from sensor input $s$.

7.3 Requirements for action decoder

Requirement 7.3 (Decoder: Γ → actions):

For discrete actions:

$$\pi(a|\Gamma) = \text{softmax}(W \cdot \text{vec}(\Gamma) + b)$$

For continuous actions:

$$\mu, \sigma = \text{Decoder}(\Gamma), \quad a \sim \mathcal{N}(\mu, \sigma^2)$$

7.4 Training

Minimization of self-prediction error:

$$\mathcal{L}(\theta) = \mathbb{E}_{\Gamma \sim \text{trajectories}}[\|\Gamma_{t+1} - \mathcal{P}_\theta(\Gamma_t)\|_F^2]$$

Implementation status

The requirements in this section are sufficient for building a concrete implementation. Cholesky parameterization guarantees correctness of the output density matrices.

---

8. Operational algorithm for φ

Status: Engineering specification

This section provides a concrete algorithm for computing the self-modeling operator φ, suitable for software implementation.

8.1 Algorithm: Basic self-modeling

Input: Coherence matrix $\Gamma \in \mathbb{C}^{7 \times 7}$

Parameters:

• $k \in (0, 1)$ — contraction coefficient (recommended $k = 0.95$)
• $w \in \Delta^6$ — anchor weight vector (default $w = (1/7, \ldots, 1/7)$)

Algorithm:

FUNCTION φ_basic(Γ, k, w):
    # Step 1: Extract diagonal (dephasing in measurement basis)
    diag_Γ := diagonal(Γ)  # vector of size 7

    # Step 2: Build predictive state
    P_pred := diag(diag_Γ)  # diagonal matrix 7×7

    # Step 3: Build anchor state
    Γ_anchor := diag(w)

    # Step 4: Mix with contraction coefficient
    φ_Γ := k * P_pred + (1 - k) * Γ_anchor

    RETURN φ_Γ

Guarantees:

• Output — valid density matrix (Hermitian, PSD, trace=1)
• Contracting map with constant $k$
• Computational complexity: $O(N)$ where $N = 7$

8.2 Algorithm: Neural network self-modeling

For trainable φ with parameters θ:

FUNCTION φ_neural(Γ, θ):
    # Step 1: Vectorize input matrix
    x := flatten_upper_triangular(Γ)  # 28 parameters (7 diag + 21 coh)

    # Step 2: Pass through neural network
    h := ReLU(W₁ · x + b₁)
    L_vec := W₂ · h + b₂  # 28 parameters for lower-triangular matrix

    # Step 3: Reconstruct lower-triangular matrix (Cholesky)
    L := unflatten_lower_triangular(L_vec)  # 7×7

    # Step 4: Build PSD matrix and normalize
    Γ_raw := L · L†
    φ_Γ := Γ_raw / Tr(Γ_raw)

    # Step 5: Apply contraction to anchor
    k := sigmoid(θ_k)  # trainable coefficient ∈ (0, 1)
    φ_Γ := k * φ_Γ + (1 - k) * I/7

    RETURN φ_Γ

Training: Minimize next-state prediction error:

$$\mathcal{L}(\theta) = \mathbb{E}_{(\Gamma_t, \Gamma_{t+1}) \sim \tau}[\|\Gamma_{t+1} - \varphi_\theta(\Gamma_t)\|_F^2]$$

8.3 Computing reflection measure R

FUNCTION compute_R_canonical(Γ):
    # Canonical definition of R (used in thresholds)
    P := Tr(Γ† · Γ)       # purity
    R := 1 / (7 * P)
    RETURN R

FUNCTION compute_Q_phi(Γ, φ):
    # Quality of self-modeling (separate measure, see WARNING above)
    φ_Γ := φ(Γ)
    error := Γ - φ_Γ
    error_norm_sq := Tr(error† · error)
    Γ_norm_sq := Tr(Γ† · Γ)  # = P (purity)
    Q := 1 - error_norm_sq / Γ_norm_sq
    RETURN Q

8.4 Checking L2 threshold

Limitation of 7D formalism

The function Tr_not_E (partial trace) requires tensor structure. In the minimal 7D formalism ($\mathcal{H} = \mathbb{C}^7$) use is_L2_minimal without $D_{\text{diff}}$ — see dimension-e.md.

FUNCTION is_L2_conscious(Γ, φ):
    # Compute three measures
    R := compute_R(Γ, φ)
    Φ := compute_integration(Γ)  # Σ|γ_ij|² / Σγ_ii²
    D_diff := exp(von_neumann_entropy(Tr_not_E(Γ)))

    # Check thresholds
    RETURN (R ≥ 1/3) AND (Φ ≥ 1) AND (D_diff ≥ 2)

# Minimal version without D_diff (for 7D formalism)
FUNCTION is_L2_minimal(Γ, φ):
    R := compute_R(Γ, φ)
    Φ := compute_integration(Γ)
    RETURN (R ≥ 1/3) AND (Φ ≥ 1)

---

9. Relation to the regeneration mechanism

Key relation

The self-modeling operator $\varphi$ defines the target state of regeneration: $\rho_* = \varphi(\Gamma)$ — categorical self-model of the current state [T] (operator φ). For each $\Gamma$ the self-model $\varphi(\Gamma)$ is unique (CPTP channel).

9.1 Regeneration as striving toward the self-model

The regenerative term of the evolution equation for $\Gamma$ is fully derived from the axioms [Т]:

$$\mathcal{R}[\Gamma, E] = \kappa(\Gamma) \cdot (\rho_* - \Gamma) \cdot g_V(P)$$

where:

• $\kappa(\Gamma)$ — regeneration coefficient [T] (categorical derivation from adjunction)
• $\rho_* = \varphi(\Gamma)$ — categorical self-model of the current state [T] (operator φ)
• $(\rho_* - \Gamma)$ — unique CPTP relaxation [T] (replacement channel + Bures optimality)
• $g_V(P)$ — V-preservation gate [T] (refines $\Theta(\Delta F)$ from Landauer, see evolution)

Full derivation: Evolution → Derivation of regeneration form.

Interpretation: The system regenerates by striving toward state $\varphi(\Gamma)$ — how it "sees itself." Regeneration is an active process of self-realization, where the system becomes its own model.

9.2 Fixed point and viable equilibrium

Theorem 9.1 (Regeneration equilibrium):

At $\Gamma = \Gamma^* = \varphi(\Gamma^*)$ the regenerative term vanishes:

$$\mathcal{R}[\Gamma^*, E] = \kappa(\Gamma^*) \cdot (\varphi(\Gamma^*) - \Gamma^*) \cdot g_V(P) = 0$$

Proof: $\varphi(\Gamma^*) = \Gamma^*$ by definition of fixed point. ∎

Corollary 9.1: At the fixed point $\Gamma^*$ the system is in a state of ideal self-knowledge — regeneration is not required, as the current state coincides with the self-model.

9.3 Dynamics outside the fixed point

At $\Gamma \neq \Gamma^*$ a "pull" toward the self-model arises:

$$\rho_* - \Gamma = \varphi(\Gamma) - \Gamma \neq 0$$

Direction of regeneration:

1. If $P(\varphi(\Gamma)) > P(\Gamma)$: regeneration increases purity
2. If $P(\varphi(\Gamma)) < P(\Gamma)$: regeneration decreases purity

Critical condition: viability of self-model

For regeneration to support viability, it is necessary that:

$$P(\varphi(\Gamma)) \geq P_{\text{crit}} = \frac{2}{7}$$

With an incorrectly constructed $\varphi$ the system may regenerate toward a non-viable state. This places constraints on the choice of anchor $\Gamma_{\text{anchor}}$ (see Definition 2.11).

9.4 Relation to reflection measure R

The reflection measure $R$ and the regenerative term $\mathcal{R}$ are related:

$$1 - R(\Gamma) = \frac{\|\Gamma - I/7\|^2_F}{P(\Gamma)} = 1 - \frac{1}{7P}$$ $$\|\mathcal{R}[\Gamma, E]\| \propto \|\rho^*_{\mathrm{diss}} - \Gamma\| = \|I/7 - \Gamma\| = \sqrt{P \cdot (1 - R)}$$

Interpretation:

• High $R$ (proximity to self-model) → small amplitude of regeneration
• Low $R$ (divergence from self-model) → large amplitude of regeneration

A system with good self-knowledge ($R \to 1$) requires minimal regeneration.

9.5 Stability of viable region

Theorem 9.2 (Regeneration keeps system in $\mathcal{V}$):

Let $\varphi$ be a contracting map with fixed point $\Gamma^* \in \mathcal{V}$ (viable region).

Then at sufficiently large $\kappa$ regeneration counteracts dissipation and keeps the system in $\mathcal{V}$:

$$\left.\frac{dP}{d\tau}\right|_{\mathcal{R}} + \left.\frac{dP}{d\tau}\right|_{\mathcal{D}} > 0 \quad \text{at } P < P(\Gamma^*)$$

Interpretation: Regeneration is a protective mechanism that uses the self-model as a guide for restoring coherence.

9.6 Preservation of positivity under regeneration

Theorem (CPTP structure of regeneration)

The regenerative operator $R_\alpha = (1 - \alpha) \cdot \mathcal{E} + \alpha \cdot \varphi$ with $\alpha = \kappa \cdot \Delta\tau < 1$ is a CPTP channel:

$$R_\alpha[\Gamma] = \sum_k \tilde{K}_k \Gamma \tilde{K}_k^\dagger$$

with Kraus operators $\tilde{K}_0 = \sqrt{1-\alpha}\,I$ and $\tilde{K}_k = \sqrt{\alpha} K_k$ (from attractor $\varphi$).

Corollary: Regeneration toward self-model $\varphi(\Gamma)$ guarantees preservation of:

• Positivity: $\Gamma \geq 0$
• Normalization: $\mathrm{Tr}(\Gamma) = 1$

More on CPTP structure of regeneration →

---

Appendix A: Computation examples

A.1 Depolarizing channel as φ

$$\varphi_p(\rho) = p \cdot \rho + (1 - p) \cdot \frac{I}{N}$$

Fixed point:

$$\Gamma^* = p \cdot \Gamma^* + (1 - p) \cdot \frac{I}{N}$$ $$(1 - p) \cdot \Gamma^* = (1 - p) \cdot \frac{I}{N} \quad \Rightarrow \quad \Gamma^* = \frac{I}{N}$$

Contraction constant: $k = p < 1$

Reflection measure at fixed point:

$$R(\Gamma^*) = R\left(\frac{I}{N}\right) = 1 - \frac{\|I/N - \varphi(I/N)\|^2_F}{\|I/N\|^2_F} = 1 - \frac{0}{1/N} = 1$$

A.2 Projection self-observation

Let $\{|i\rangle\}$ be an orthonormal basis, $P_i = |i\rangle\langle i|$.

$$\varphi_{\text{diag}}(\rho) = \sum_i P_i \rho P_i = \sum_i \rho_{ii} |i\rangle\langle i|$$

(Diagonalization in the given basis)

Fixed points:

$$\varphi_{\text{diag}}(\Gamma) = \Gamma \quad \Leftrightarrow \quad \Gamma \text{ is diagonal}$$

The set of fixed points is an $(N-1)$-dimensional simplex:

$$\mathrm{Fix}(\varphi_{\text{diag}}) = \left\{\sum_i p_i |i\rangle\langle i| : p_i \geq 0, \sum_i p_i = 1\right\} \cong \Delta^{N-1}$$

where $N = \dim(\mathcal{H}) = 7$ for the Holon.

Remark: This is not a contracting map ($k = 1$ on the set of fixed points).

---

Appendix B: Proof of CPTP structure preservation

Lemma B.1: If $\mathcal{P}$ is CPTP and $\rho \in \mathcal{D}(\mathcal{H})$, then $\mathcal{P}(\rho) \in \mathcal{D}(\mathcal{H})$.

Proof:

1. Hermiticity:

$$\mathcal{P}(\rho)^\dagger = \left(\sum_m K_m \rho K_m^\dagger\right)^\dagger = \sum_m K_m \rho^\dagger K_m^\dagger = \sum_m K_m \rho K_m^\dagger = \mathcal{P}(\rho)$$

2. Positivity:

For any $|\psi\rangle$:

$$\langle\psi|\mathcal{P}(\rho)|\psi\rangle = \sum_m \langle\psi|K_m \rho K_m^\dagger|\psi\rangle = \sum_m \langle K_m^\dagger\psi|\rho|K_m^\dagger\psi\rangle \geq 0$$

(since $\rho \geq 0$)

3. Normalization:

$$\mathrm{Tr}(\mathcal{P}(\rho)) = \mathrm{Tr}\left(\sum_m K_m \rho K_m^\dagger\right) = \sum_m \mathrm{Tr}(K_m^\dagger K_m \rho) = \mathrm{Tr}\left(\left(\sum_m K_m^\dagger K_m\right) \rho\right) = \mathrm{Tr}(I \cdot \rho) = 1$$

∎

---

Appendix C: Generalization to time-dependent φ

Definition C.1 (Dynamic self-modeling operator):

$$\varphi: [0, \infty) \times \mathcal{D}(\mathcal{H}) \to \mathcal{D}(\mathcal{H})$$ $$(\tau, \Gamma) \mapsto \varphi(\tau, \Gamma)$$

Dynamic fixed point equation:

$$\Gamma^*(\tau) = \varphi(\tau, \Gamma^*(\tau))$$

Theorem C.1 (Existence of dynamic fixed point):

If $\varphi(\tau, \cdot)$ is contracting with constant $k < 1$ for all $\tau$, and $\varphi$ is continuous in $\tau$, then:

1. $\Gamma^*(\tau)$ exists and is unique for each $\tau$
2. $\Gamma^*(\tau)$ is continuous in $\tau$
3. $\frac{d\Gamma^*}{d\tau} = \frac{\partial \varphi}{\partial \tau} + (D\varphi)\left(\frac{d\Gamma^*}{d\tau}\right)$ (implicit equation)

Proof: Follows from applying the implicit function theorem in a Banach space.

Octonionic context of self-modeling

Self-modeling and alternativity [I]

In the octonionic interpretation, the self-modeling operator $\varphi$ acts on the space $\mathrm{Im}(\mathbb{O})$. Alternativity of octonions (Artin's theorem [T]) guarantees that $\varphi$ is associative when acting on any pair of dimensions, but may exhibit non-associativity when acting simultaneously on three or more dimensions.

This is consistent with the fixed point property $\varphi(\Gamma^*) = \Gamma^*$: self-consistency is achieved in the full 7-dimensional space where non-associativity is integrated into the structure. Bridge [T] (closed, T15). See structural derivation.

---

Tensor factorization of φ for composite systems

Relation to no-signaling prohibition and preservation of holonomic character

Tensor factorization of $\varphi$ is a key property ensuring compatibility of $\mathcal{R}$ with no-signaling prohibition. It guarantees that self-modeling of autonomous subsystems does not create channels of superluminal communication (Gisin, Polchinski 1991).

Preservation of holonomic character. Factorization $\varphi_{A \otimes B} = \varphi_A \otimes \varphi_B$ concerns only the regenerative term $\mathcal{R}$. The full dynamics $\mathcal{L}_\Omega = -i[H, \cdot] + \mathcal{D}[\cdot] + \mathcal{R}[\cdot]$ contains:

• $H$ (Hamiltonian): creates and preserves entanglement — non-local ✓
• $\mathcal{D}$ (dissipation): may destroy entanglement, but through common decoherence — non-local in general
• $\mathcal{R}$ (regeneration via $\varphi$): local (factorizes) — ensures no-signaling

"Holonomy" (the whole > sum of parts) is realized through $H + \mathcal{D}$, not through $\mathcal{R}$. Self-modeling ($\varphi$) is a local process (each agent models itself, not another). Entanglement is a property of $H$ (Hamiltonian dynamics). The theory is not a "local hidden variable theory": only $\mathcal{R}$ is local, while $H + \mathcal{D}$ are non-local.

Refinement: SSB, not gauge freedom. The more precise qualification is spontaneous symmetry breaking (SSB), not gauge freedom:

1. Before $V_{\text{Gap}}$ minimization: $G_2$-symmetry unbroken, all bases equivalent.
2. Upon $V_{\text{Gap}}$ minimization (T-64 [T]): system "rolls" into a specific vacuum $\Gamma_{\text{vac}}$ on the manifold of minima $(S^1)^{21}/G_2$. One minimum is selected.
3. After SSB: $G_2 \to H$ (vacuum stabilizer). Boolean fragment $\mathrm{Dec}(\Omega)$ crystallizes as pointer basis fixed by the vacuum.
4. Goldstone modes (see goldstone-modes): massless excitations along broken directions $G_2/H$.

Analogy: not coordinates in GR, but the Higgs mechanism — $SU(2) \times U(1) \to U(1)_{\text{em}}$ generates W/Z masses. In UHM: $G_2 \to H$ generates classical objectivity (Dec(Ω) = Boolean logic).

Canonical extension of φ to composite system

Definition (Canonical extension $\varphi_A$). For an autonomous holon $A$ in a composite system $A \otimes B$, the extension $\varphi_A$ is defined as:

$$\tilde{\varphi}_A := \varphi_A \otimes \mathrm{id}_B$$

This is the unique extension compatible with the CPTP structure of $\varphi_A$ and the tensor structure of category $\mathbf{DensityMat}$.

Theorem: tensor factorization

Theorem (Tensor factorization of φ)

For a composite system of two autonomous holons $A$ and $B$:

$$\varphi_{A \otimes B} = \varphi_A \otimes \varphi_B$$

Proof:

1. By the definition of autonomy (A1): $\mathcal{I}(A:B|\partial A) = 0$ — conditional independence of $A$ and $B$.

2. The operator $\varphi$ is defined as left adjoint to the inclusion of subobjects:

$$\varphi \dashv i: \mathrm{Sub}(\Gamma) \hookrightarrow \mathcal{E}$$

3. For autonomous subsystems the lattice of subobjects factorizes:

$$\mathrm{Sub}(\Gamma_{AB}) \cong \mathrm{Sub}(\Gamma_A) \times \mathrm{Sub}(\Gamma_B)$$

4. The left adjoint to the product of inclusions is the product of left adjoints:

$$\varphi_{A \otimes B} = \varphi_A \times \varphi_B \cong \varphi_A \otimes \varphi_B \quad \blacksquare$$

Corollary: annihilation of nonlinear contribution

Lemma (Annihilation of regeneration under partial trace). For any CPTP channel $\Phi_A$ and scalar $\alpha \in \mathbb{R}$:

$$\mathrm{Tr}_A\left[\alpha \cdot ((\Phi_A \otimes \mathrm{id}_B)(\rho_{AB}) - \rho_{AB})\right] = 0$$

Corollary: The regenerative term $\tilde{\mathcal{R}}_A[\Gamma_{AB}] = \kappa_A \cdot ((\varphi_A \otimes \mathrm{id}_B)(\Gamma_{AB}) - \Gamma_{AB}) \cdot g_V(P_A)$ automatically satisfies the no-signaling prohibition — the contribution to $\Gamma_B = \mathrm{Tr}_A[\Gamma_{AB}]$ is zero.

Full proof: Physical correspondence — No-signaling prohibition.

---

Related documents:

• Self-observation — definitions of $\varphi$, $R$ and $R^{(n)}$
• Evolution — equation of motion, regenerative term $\mathcal{R}$ and canonical $\Delta F$
• Coherence matrix — definition of $\Gamma$
• Viability — purity measure $P$ and region $\mathcal{V}$
• Unity dimension — integration measure $\Phi$
• Axiom of Septicity — regeneration coefficient $\kappa_0$
• Interiority hierarchy — levels L0→L1→L2→L3→L4, thresholds $R^{(n)}_{\text{th}}$
• Categorical formalism — categorical structure of UHM, n-truncations and no-signaling prohibition as natural transformation
• Holon — definition of $\mathbb{H}$
• Physical correspondence — No-signaling prohibition — complete proofs