Categorical Formalism of Functor F: DensityMat → Exp

Strict Mathematical Specification

On notation

In this document:

• $\mathbf{Exp}$ — the category of experiential space. Not to be confused with $\text{Exp}$ — the experiential point function.
• $\mathcal{H}$ — Hilbert space. Not to be confused with $H$ — the Hamiltonian.
• $\mathcal{C}$ — context space. Not to be confused with $C$ — the consciousness measure.
• $\Phi, \Psi, \Xi$ — arbitrary CPTP channels. $\Phi$ is used here for category morphisms, not for the integration measure (which is denoted $\Phi_{\text{UHM}}$ when disambiguation is needed).

Contents

1. Category DensityMat
2. Category Exp
3. Functor F on objects
4. Functor F on morphisms
5. Proof of functoriality
6. Topos structure
7. Limitations and alternatives
8. Phenomenal completeness
9. Quasi-functor for AI systems
10. ∞-groupoid and ∞-topos for emergent time
11. Discrete ∞-groupoid Exp^disc_∞
12. Category of Holons Hol
13. Derived categories and IC-cohomologies
14. ∞-topos as the true primitive
15. L-unification
    • 15.3 Adjunction $\mathcal{D}_\Omega \dashv \mathcal{R}$
16. Categorical completeness of UHM

---

1. Category DensityMat

1.1 Definition

Definition 1.1 (Category DensityMat). The category of density matrices $\mathbf{DensityMat}$ consists of:

Objects:

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

where $\mathcal{H}$ is a separable Hilbert space (in our case $\mathcal{H} = \mathbb{C}^7$ for the Holon).

Morphisms:

$$\mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_2) = \{\Phi : \mathcal{L}(\mathcal{H}) \to \mathcal{L}(\mathcal{H}) \mid \Phi \text{ is CPTP}, \Phi(\rho_1) = \rho_2\}$$

where CPTP stands for Completely Positive Trace-Preserving. See formalization of φ.

Remark 1.1. The set $\mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_2)$ may be empty for some pairs $(\rho_1, \rho_2)$. This does not violate the definition of a category.

1.2 Structure of morphisms (CPTP channels)

Definition 1.2 (CPTP channel). A linear map $\Phi: \mathcal{L}(\mathcal{H}) \to \mathcal{L}(\mathcal{H})$ is called CPTP if:

1. Trace-Preserving (TP): $\mathrm{Tr}(\Phi(\rho)) = \mathrm{Tr}(\rho)$ for all $\rho$
2. Completely Positive (CP): For any $n \geq 1$ and any positive operator $A \in \mathcal{L}(\mathcal{H} \otimes \mathbb{C}^n)$, the operator $(\Phi \otimes \mathrm{id}_n)(A)$ is also positive.

Theorem 1.1 (Kraus representation). $\Phi$ is CPTP if and only if there exist operators $\{K_i\}_{i=1}^r$ such that:

$$\Phi(\rho) = \sum_i K_i \rho K_i^\dagger, \quad \sum_i K_i^\dagger K_i = I$$

1.3 Category axioms for DensityMat

Theorem 1.2. $\mathbf{DensityMat}$ is a category.

Proof:

1. Composition of morphisms:

Let $\Phi \in \mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_2)$ and $\Psi \in \mathrm{Mor}_{\mathbf{DM}}(\rho_2, \rho_3)$.

Define $\Psi \circ \Phi: \mathcal{L}(\mathcal{H}) \to \mathcal{L}(\mathcal{H})$ as functional composition.

Verify:

• $(\Psi \circ \Phi)(\rho_1) = \Psi(\Phi(\rho_1)) = \Psi(\rho_2) = \rho_3$ ✓
• $\Psi \circ \Phi$ is CPTP (composition of CPTP is CPTP) ✓

Therefore, $\Psi \circ \Phi \in \mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_3)$.

2. Associativity:

For $\Phi \in \mathrm{Mor}(\rho_1, \rho_2)$, $\Psi \in \mathrm{Mor}(\rho_2, \rho_3)$, $\Xi \in \mathrm{Mor}(\rho_3, \rho_4)$:

$$(\Xi \circ \Psi) \circ \Phi = \Xi \circ (\Psi \circ \Phi)$$

This follows from the associativity of functional composition.

3. Identity morphisms:

For each $\rho \in \mathrm{Ob}(\mathbf{DensityMat})$ define:

$$\mathrm{id}_\rho := \mathrm{Id}: \mathcal{L}(\mathcal{H}) \to \mathcal{L}(\mathcal{H}), \quad \mathrm{Id}(\sigma) = \sigma$$

Verify:

• $\mathrm{Id}(\rho) = \rho$ ✓
• $\mathrm{Id}$ is CPTP (Kraus representation with $K_1 = I$) ✓
• $\mathrm{Id} \in \mathrm{Mor}_{\mathbf{DM}}(\rho, \rho)$ ✓

For any $\Phi \in \mathrm{Mor}(\rho_1, \rho_2)$:

$$\Phi \circ \mathrm{id}_{\rho_1} = \Phi, \quad \mathrm{id}_{\rho_2} \circ \Phi = \Phi$$

∎

---

2. Category Exp

2.1 Experiential space (objects)

Definition 2.1 (Experiential space).

Clarification: History as an emergent structure

In the canonical definition (see Theorem 5.3) history is not part of the objects of the Exp category, but is derived from the 2-categorical structure $\mathbf{Exp}_2$ and the ∞-groupoid $\mathbf{Exp}_\infty$ (section 10).

Basic experiential space (objects of the category):

$$\mathcal{E}_0 := \Delta^{N-1} \times_{\mathrm{Spec}} \mathbb{P}(\mathcal{H}_E)^N \times \mathcal{C}$$

Complete experiential space (with emergent history):

$$\mathcal{E} := \mathcal{E}_0 \times \mathrm{Hist}, \quad \text{where } \mathrm{Hist} := \pi_1(\mathbf{Exp}_2, \mathcal{Q})$$

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

• $\Delta^{N-1} = \{(\lambda_1, \ldots, \lambda_N) : \lambda_i \geq 0, \sum \lambda_i = 1\}$ — the $(N-1)$-simplex of intensities (spectrum)
• $\mathbb{P}(\mathcal{H}_E)$ — projective space of qualities $\mathbb{CP}^{\dim(\mathcal{H}_E)-1}$
• $\mathcal{C}$ — context space (measurement states except E)
• $\mathrm{Hist} = \pi_1(\mathbf{Exp}_2, \mathcal{Q})$ — history space, derived as the fundamental groupoid of the bicategory (§5.2.3)
• $\times_{\mathrm{Spec}}$ — fiber product over the spectrum

Definition 2.2 (Objects of category Exp).

$$\mathrm{Ob}(\mathbf{Exp}) = \{\mathcal{Q} = (\lambda, [q], c, h) \in \mathcal{E}\}$$

where:

• $\lambda = (\lambda_1, \ldots, \lambda_N) \in \Delta^{N-1}$ — intensity vector
• $[q] = ([q_1], \ldots, [q_N]) \in \mathbb{P}(\mathcal{H}_E)^N$ — set of qualities (equivalence classes)
• $c \in \mathcal{C}$ — context
• $h \in \mathrm{Hist}$ — history

2.2 Morphisms in category Exp

Problem: Morphisms in $\mathbf{Exp}$ were not formally defined in the original theory.

Solution: Three equivalent definitions are proposed, between which natural correspondences exist.

Variant A: Paths in experiential space

Definition 2.3 (Path morphisms).

$$\mathrm{Mor}_\mathcal{E}^{\mathrm{path}}(\mathcal{Q}_1, \mathcal{Q}_2) := \{\gamma: [0,1] \to \mathcal{E} \mid \gamma(0) = \mathcal{Q}_1, \gamma(1) = \mathcal{Q}_2, \gamma \text{ is continuous}\}$$

with an equivalence relation (homotopy):

$$\gamma_1 \sim \gamma_2 \Leftrightarrow \exists \, \mathcal{G}: [0,1] \times [0,1] \to \mathcal{E}, \; \mathcal{G}(s,0) = \gamma_1(s), \; \mathcal{G}(s,1) = \gamma_2(s), \; \mathcal{G}(0,t) = \mathcal{Q}_1, \; \mathcal{G}(1,t) = \mathcal{Q}_2$$

Composition: Concatenation of paths

$$(\gamma_2 \circ \gamma_1)(s) = \begin{cases} \gamma_1(2s), & s \in [0, 1/2] \\ \gamma_2(2s-1), & s \in [1/2, 1] \end{cases}$$

Identity: Constant path

$$\mathrm{id}_\mathcal{Q}(s) = \mathcal{Q} \quad \text{for all } s \in [0,1]$$

Variant B: Component-wise maps

Definition 2.4 (Transformation morphisms).

$$\mathrm{Mor}_\mathcal{E}^{\mathrm{trans}}(\mathcal{Q}_1, \mathcal{Q}_2) := \{(f_\lambda, f_q, f_c, f_h) \mid \text{conditions below}\}$$

where:

• $f_\lambda: \Delta^{N-1} \to \Delta^{N-1}$, $f_\lambda(\lambda_1) = \lambda_2$
• $f_q: \mathbb{P}(\mathcal{H}_E)^N \to \mathbb{P}(\mathcal{H}_E)^N$, $f_q([q_1]) = [q_2]$
• $f_c: \mathcal{C} \to \mathcal{C}$, $f_c(c_1) = c_2$
• $f_h: \mathrm{Hist} \to \mathrm{Hist}$, $f_h(h_1) = h_2$
• all components are continuous

Composition: Component-wise

$$(f'_\lambda, f'_q, f'_c, f'_h) \circ (f_\lambda, f_q, f_c, f_h) = (f'_\lambda \circ f_\lambda, f'_q \circ f_q, f'_c \circ f_c, f'_h \circ f_h)$$

Identity:

$$\mathrm{id}_\mathcal{Q} = (\mathrm{id}_\Delta, \mathrm{id}_\mathbb{P}, \mathrm{id}_\mathcal{C}, \mathrm{id}_{\mathrm{Hist}})$$

Variant C: Induced by CPTP channels

Definition 2.5 (Induced morphisms). Let $\Phi \in \mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_2)$. Define:

$$\mathrm{Mor}_\mathcal{E}^{\mathrm{ind}}(\mathcal{Q}_1, \mathcal{Q}_2) := \{F(\Phi) \mid \Phi \in \mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_2), F(\rho_1) = \mathcal{Q}_1, F(\rho_2) = \mathcal{Q}_2\}$$

where $F$ is the functor defined in section 3.

This is the natural choice, as it follows directly from functoriality.

2.3 Adopted definition

Definition 2.6 (Category Exp — canonical definition).

Constructive choice

The choice of morphisms of the Exp category is made to ensure functoriality of F — this is a constructive decision, not a consequence. Morphisms of Exp are defined as images of CPTP channels under F, which guarantees functoriality by construction.

Rationale for choosing Variant C

We adopt Variant C as the canonical definition for the following reasons:

1. Physical justification: Morphisms are induced by real quantum processes (CPTP channels)
2. Functoriality: Ensures strict functoriality of $F$ by construction
3. Compatibility with DensityMat: The categorical structure of Exp is inherited from the well-defined category DensityMat
4. Computability: Variant B provides a concrete component-wise representation for calculations

Variants A, B, C are not equivalent in general:

• Variant A (paths) is more general, but not all paths are induced by CPTP
• Variant B (component-wise) is a concrete representation, but not every quadruple $(f_\lambda, f_q, f_c, f_h)$ is physically realizable
• Variant C — the physically correct subset

$$\mathbf{Exp} := (\mathrm{Ob}_\mathcal{E}, \mathrm{Mor}_\mathcal{E}^{\mathrm{ind}})$$

with additional structure:

• For each morphism $m \in \mathrm{Mor}_\mathcal{E}^{\mathrm{ind}}(\mathcal{Q}_1, \mathcal{Q}_2)$ there exists a representation $(f_\lambda, f_q, f_c, f_h)$
• The representation is determined by the action of the corresponding CPTP channel on the components

2.4 Category axioms for Exp

Theorem 2.1. $\mathbf{Exp}$ (with Definition 2.6) is a category.

Proof:

1. Composition (declarative definition):

Let $m_1 = F(\Phi) \in \mathrm{Mor}_\mathcal{E}(\mathcal{Q}_1, \mathcal{Q}_2)$ and $m_2 = F(\Psi) \in \mathrm{Mor}_\mathcal{E}(\mathcal{Q}_2, \mathcal{Q}_3)$.

Define composition:

$$F(\Psi) \circ F(\Phi) := F(\Psi \circ \Phi)$$

This is well-defined, since $\Psi \circ \Phi$ is a composition of CPTP channels in DensityMat, which is itself a CPTP channel (closure of CPTP under composition, proved in §1.3). The map $F$ is used here only as a map (from morphisms of DensityMat to morphisms of Exp), not as a functor — the functoriality of $F$ (section 5) is a consequence of this construction, not a prerequisite.

Verify $F(\Psi \circ \Phi) \in \mathrm{Mor}_\mathcal{E}(\mathcal{Q}_1, \mathcal{Q}_3)$: $(\Psi \circ \Phi)(\rho_1) = \Psi(\rho_2) = \rho_3$ ✓, and $F$ applies Definition 3.1 to the result, giving $\mathcal{Q}_3$. ✓

2. Associativity:

$$(F(\Xi) \circ F(\Psi)) \circ F(\Phi) = F(\Xi \circ \Psi) \circ F(\Phi) = F((\Xi \circ \Psi) \circ \Phi)$$ $$= F(\Xi \circ (\Psi \circ \Phi)) = F(\Xi) \circ F(\Psi \circ \Phi) = F(\Xi) \circ (F(\Psi) \circ F(\Phi))$$

The second equality in each line is by definition of composition in Exp. The central equality is associativity of composition of CPTP channels in DensityMat (functional composition is associative). ✓

3. Identities:

$\mathrm{id}_\mathcal{Q} := F(\mathrm{id}_\rho)$, where $F(\rho) = \mathcal{Q}$ and $\mathrm{id}_\rho$ is the identity CPTP channel.

For any $m = F(\Phi) \in \mathrm{Mor}(\mathcal{Q}_1, \mathcal{Q}_2)$:

$$m \circ \mathrm{id}_{\mathcal{Q}_1} = F(\Phi) \circ F(\mathrm{id}_{\rho_1}) = F(\Phi \circ \mathrm{id}_{\rho_1}) = F(\Phi) = m$$ $$\mathrm{id}_{\mathcal{Q}_2} \circ m = F(\mathrm{id}_{\rho_2}) \circ F(\Phi) = F(\mathrm{id}_{\rho_2} \circ \Phi) = F(\Phi) = m$$

Here $\Phi \circ \mathrm{id}_{\rho_1} = \Phi$ and $\mathrm{id}_{\rho_2} \circ \Phi = \Phi$ are properties of the identity map in DensityMat. ✓

Order of proof

The functoriality of $F$ (section 5) is a consequence of this construction, not a prerequisite. Here $F$ is used only as a map on objects and morphisms, and the category axioms are verified directly from the properties of CPTP channels in DensityMat.

∎

---

3. Functor F on objects

3.1 Definition

Definition 3.1 (Functor F on objects).

$$F: \mathrm{Ob}(\mathbf{DensityMat}) \to \mathrm{Ob}(\mathbf{Exp})$$ $$F(\rho) := (\mathrm{Spectrum}(\rho_E), \mathrm{Quality}(\rho_E), \mathrm{Context}(\Gamma_{-E}), \mathrm{History}(t))$$

where:

Component 1: Spectrum (Intensity)

$$\mathrm{Spectrum}(\rho_E) := \{\lambda_i : \rho_E|q_i\rangle = \lambda_i|q_i\rangle\}, \text{ ordered by decreasing}$$

Component 2: Quality (Eigenvectors in projective space)

$$\mathrm{Quality}(\rho_E) := \{[|q_i\rangle] \in \mathbb{P}(\mathcal{H}_E)\}$$

where $[|q\rangle]$ is the equivalence class $|q\rangle \sim c|q\rangle$ for $c \in \mathbb{C}^*$.

Component 3: Context

$$\mathrm{Context}(\Gamma_{-E}) := (\gamma_{Ai}, \gamma_{Si}, \gamma_{Di}, \gamma_{Li}, \gamma_{Oi}, \gamma_{Ui})$$

— states of all dimensions except $E$.

Component 4: History

$$\mathrm{History}(t) := \{\rho_E(t') : t' \in [t-\tau, t]\}$$

— evolution trajectory in a sliding window $\tau$.

3.2 Correctness of the definition

Lemma 3.1. $F(\rho) \in \mathrm{Ob}(\mathbf{Exp})$ for any $\rho \in \mathrm{Ob}(\mathbf{DensityMat})$.

Proof:

1. $\rho_E$ is a Hermitian operator $\Rightarrow$ the spectrum is real and eigenvectors are orthogonal
2. $\rho_E \geq 0$ $\Rightarrow$ $\lambda_i \geq 0$ for all $i$
3. $\mathrm{Tr}(\rho_E) = 1$ $\Rightarrow$ $\sum \lambda_i = 1$ $\Rightarrow$ $(\lambda_1, \ldots, \lambda_N) \in \Delta^{N-1}$
4. Eigenvectors $|q_i\rangle$ are normalized $\Rightarrow$ $[|q_i\rangle] \in \mathbb{P}(\mathcal{H}_E)$

Therefore, $F(\rho) \in \mathcal{E}$. ∎

3.3 Degeneracy problem

Problem: When the spectrum is degenerate ($\lambda_i = \lambda_j$ for $i \neq j$) eigenvectors are not uniquely defined.

Solution: For degenerate eigenvalues the quality is defined as the eigenspace:

$$\mathrm{Quality}_{\mathrm{degen}}(\rho_E, \lambda) := \mathrm{Ker}(\rho_E - \lambda I) \subset \mathcal{H}_E$$

The quality space generalizes to a Grassmannian:

$$\mathrm{Quality} \in \mathrm{Gr}(k, \mathcal{H}_E) \quad \text{where } k = \dim(\mathrm{Ker}(\rho_E - \lambda I))$$

Definition 3.2 (Extended functor F).

$$F_{\mathrm{ext}}(\rho) := (\mathrm{Spectrum}(\rho_E), \mathrm{QualitySpaces}(\rho_E), \mathrm{Context}, \mathrm{History})$$

where $\mathrm{QualitySpaces}$ is the set of eigenspaces.

---

4. Functor F on morphisms

4.1 Definition

Definition 4.1 (Functor F on morphisms).

$$F: \mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_2) \to \mathrm{Mor}_\mathcal{E}(F(\rho_1), F(\rho_2))$$ $$F(\Phi) := (f_\lambda^\Phi, f_q^\Phi, f_c^\Phi, f_h^\Phi)$$

where components are defined as follows:

Component 1: Spectrum transformation

Let $\rho_2 = \Phi(\rho_1)$. Then:

$$f_\lambda^\Phi: \mathrm{Spectrum}(\rho_{1,E}) \mapsto \mathrm{Spectrum}(\rho_{2,E})$$

Explicit formula via Kraus representation $\Phi(\rho) = \sum_k K_k \rho K_k^\dagger$:

$$\lambda'_i = \langle q'_i|\Phi(\rho_E)|q'_i\rangle = \sum_k \sum_j \lambda_j |\langle q'_i|K_k|q_j\rangle|^2$$

where $|q'_i\rangle$ are the eigenvectors of $\Phi(\rho_E)$.

Component 2: Quality transformation

$$f_q^\Phi: \mathbb{P}(\mathcal{H}_E)^N \to \mathbb{P}(\mathcal{H}_E)^N, \quad f_q^\Phi([|q_i\rangle]) := [|q'_i\rangle]$$

where $|q'_i\rangle$ is the $i$-th eigenvector of $\Phi(\rho_E)$, ordered by $\lambda'_i$.

Remark 4.1. This definition requires a consistent numbering. When eigenvalues cross, adiabatic continuation is used (see section 4.3).

Component 3: Context transformation

For a full CPTP channel $\Phi$ on $\Gamma$:

$$f_c^\Phi(c_1) := \mathrm{Context}(\Phi(\Gamma)_{-E})$$

Component 4: History transformation

$$f_h^\Phi(h_1) := h_1 \cup \{\rho_{2,E}\} = \{\rho_E(t') : t' \in [t_1 - \tau, t_1]\} \cup \{\Phi(\rho_1)_E\}$$

4.2 Correctness of the definition

Lemma 4.1. $F(\Phi) \in \mathrm{Mor}_\mathcal{E}(F(\rho_1), F(\rho_2))$ for any $\Phi \in \mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_2)$.

Proof:

We need to verify:

1. $f_\lambda^\Phi(\mathrm{Spectrum}(\rho_{1,E})) = \mathrm{Spectrum}(\rho_{2,E})$ — follows from $\Phi(\rho_1) = \rho_2$
2. $f_q^\Phi(\mathrm{Quality}(\rho_{1,E})) = \mathrm{Quality}(\rho_{2,E})$ — by definition
3. $f_c^\Phi(\mathrm{Context}(\Gamma_1)) = \mathrm{Context}(\Gamma_2)$ — follows from $\Phi(\Gamma_1) = \Gamma_2$
4. Continuity — follows from continuity of CPTP channels

∎

4.3 Adiabatic continuation for degeneracy

When levels cross ($\lambda_i(t) = \lambda_j(t)$ for some $t$) we use adiabatic continuation:

Definition 4.2 (Adiabatic correspondence of eigenvectors).

Let $\gamma: [0,1] \to \mathbf{DensityMat}$ be a continuous path of density matrices without level crossings at interior points.

Then eigenvectors $|q_i(s)\rangle$ are defined by the parallel transport equation:

$$\langle q_i(s)|\partial_s|q_j(s)\rangle = 0 \quad \text{for } i \neq j$$

This gives a canonical correspondence between eigenvectors of $\rho(0)$ and $\rho(1)$.

Theorem 4.1 (Geometric phase). For a closed path $\gamma: [0,1] \to \mathbf{DensityMat}$, $\gamma(0) = \gamma(1)$, the eigenvector acquires a geometric phase (Berry phase):

$$|q_i(1)\rangle = e^{i\phi_i} |q_i(0)\rangle$$

where $\phi_i = \oint_\gamma A_i$, $A_i = i\langle q_i|d|q_i\rangle$ — the Berry connection.

---

5. Proof of functoriality

5.1 First functor axiom: $F(\mathrm{id}_\rho) = \mathrm{id}_{F(\rho)}$

Theorem 5.1. For any $\rho \in \mathrm{Ob}(\mathbf{DensityMat})$:

$$F(\mathrm{id}_\rho) = \mathrm{id}_{F(\rho)}$$

Proof:

$\mathrm{id}_\rho = \mathrm{Id}$ — the identity CPTP channel.

Compute $F(\mathrm{Id})$:

1. Spectrum: $\mathrm{Id}(\rho) = \rho$ $\Rightarrow$ $\mathrm{Spectrum}(\mathrm{Id}(\rho)_E) = \mathrm{Spectrum}(\rho_E)$ $\Rightarrow$ $f_\lambda^{\mathrm{Id}} = \mathrm{id}_\Delta$

2. Quality: Eigenvectors do not change $\Rightarrow$ $f_q^{\mathrm{Id}} = \mathrm{id}_\mathbb{P}$

3. Context: $\mathrm{Id}(\Gamma)_{-E} = \Gamma_{-E}$ $\Rightarrow$ $f_c^{\mathrm{Id}} = \mathrm{id}_\mathcal{C}$

4. History: The same state is appended $\Rightarrow$ $f_h^{\mathrm{Id}} = \mathrm{id}_{\mathrm{Hist}}$ (up to isomorphism)

Therefore:

$$F(\mathrm{Id}) = (\mathrm{id}_\Delta, \mathrm{id}_\mathbb{P}, \mathrm{id}_\mathcal{C}, \mathrm{id}_{\mathrm{Hist}}) = \mathrm{id}_{F(\rho)}$$

∎

5.2 Second functor axiom: $F(\Psi \circ \Phi) = F(\Psi) \circ F(\Phi)$

Theorem 5.2. For any $\Phi \in \mathrm{Mor}_{\mathbf{DM}}(\rho_1, \rho_2)$ and $\Psi \in \mathrm{Mor}_{\mathbf{DM}}(\rho_2, \rho_3)$:

$$F(\Psi \circ \Phi) = F(\Psi) \circ F(\Phi)$$

Proof:

Let $\rho_2 = \Phi(\rho_1)$, $\rho_3 = \Psi(\rho_2) = (\Psi \circ \Phi)(\rho_1)$.

Left-hand side: $F(\Psi \circ \Phi) = (f_\lambda^{\Psi \circ \Phi}, f_q^{\Psi \circ \Phi}, f_c^{\Psi \circ \Phi}, f_h^{\Psi \circ \Phi})$

Right-hand side: $F(\Psi) \circ F(\Phi) = (f_\lambda^\Psi \circ f_\lambda^\Phi, f_q^\Psi \circ f_q^\Phi, f_c^\Psi \circ f_c^\Phi, f_h^\Psi \circ f_h^\Phi)$

Verify component-wise:

1. Spectrum:

$$f_\lambda^{\Psi \circ \Phi}(\mathrm{Spectrum}(\rho_{1,E})) = \mathrm{Spectrum}((\Psi \circ \Phi)(\rho_1)_E) = \mathrm{Spectrum}(\rho_{3,E})$$ $$(f_\lambda^\Psi \circ f_\lambda^\Phi)(\mathrm{Spectrum}(\rho_{1,E})) = f_\lambda^\Psi(\mathrm{Spectrum}(\rho_{2,E})) = \mathrm{Spectrum}(\rho_{3,E})$$

✓ Equal

2. Quality:

$$f_q^{\Psi \circ \Phi}: [|q_i^{(1)}\rangle] \mapsto [|q_i^{(3)}\rangle]$$ $$(f_q^\Psi \circ f_q^\Phi): [|q_i^{(1)}\rangle] \mapsto [|q_i^{(2)}\rangle] \mapsto [|q_i^{(3)}\rangle]$$

Using adiabatic continuation:

• The direct path $\rho_1 \to \rho_3$ gives the correspondence $|q_i^{(1)}\rangle \leftrightarrow |q_i^{(3)}\rangle$
• The path $\rho_1 \to \rho_2 \to \rho_3$ gives the same correspondence (homotopic equivalence)

✓ Equal (up to geometric phase, which does not affect the projective class $[|q\rangle]$)

3. Context:

$$f_c^{\Psi \circ \Phi}(c_1) = \mathrm{Context}((\Psi \circ \Phi)(\Gamma_1)_{-E}) = \mathrm{Context}(\Gamma_{3,-E}) = c_3$$ $$(f_c^\Psi \circ f_c^\Phi)(c_1) = f_c^\Psi(\mathrm{Context}(\Gamma_{2,-E})) = \mathrm{Context}(\Gamma_{3,-E}) = c_3$$

✓ Equal

4. History:

Problem: the history component violates strict functoriality

When Definition 4.1 is literally applied to the history component:

$$f_h^{\Psi \circ \Phi}(h_1) = h_1 \cup \{\rho_{3,E}\}$$$$(f_h^\Psi \circ f_h^\Phi)(h_1) = f_h^\Psi(h_1 \cup \{\rho_{2,E}\}) = h_1 \cup \{\rho_{2,E}\} \cup \{\rho_{3,E}\}$$

The right-hand side contains the intermediate state $\rho_{2,E}$, which violates the equality $F(\Psi \circ \Phi) = F(\Psi) \circ F(\Phi)$.

5.2.1 Diagnosis of the problem

Root cause: The attempt to use a 1-categorical structure for a phenomenon that is inherently 2-categorical (or even ∞-categorical).

Aspect	1-category	2-category (bicategory)
Equality of morphisms	Strict: $g \circ f = h$	Up to isomorphism: $g \circ f \cong h$
Composition	Strictly associative	Associative up to coherent isomorphism
History	Component of object	Structure of 1-morphisms

Key insight: History is not a component of objects, but a structure of morphisms (transitions between states).

---

5.2.2 Strict solution: Lax 2-functor

Theorem 5.2' (Lax functoriality — canonical solution)

The functor $F$ naturally extends to a lax 2-functor:

$F_2: \mathbf{DensityMat} \to \mathbf{Exp}_2$

where $\mathbf{Exp}_2$ is the bicategory of experiential states.

Definition 5.1 (Bicategory $\mathbf{Exp}_2$).

0-cells (objects):

$$\mathrm{Ob}(\mathbf{Exp}_2) = \{(\lambda, [q], c) \in \Delta^{N-1} \times_{\mathrm{Spec}} \mathbb{P}(\mathcal{H}_E)^N \times \mathcal{C}\}$$

Note: History is not part of the objects — it is encoded in the structure of morphisms.

1-morphisms:

$$\mathrm{Mor}_1(\mathcal{Q}_1, \mathcal{Q}_2) = \{(\mathcal{Q}_1, \Phi, \mathcal{Q}_2) \mid \Phi \in \mathrm{CPTP}, F(\Phi(\rho_1)) = \mathcal{Q}_2\}$$

A 1-morphism is a transition between states, including information about the channel $\Phi$.

2-morphisms:

$$\mathrm{Mor}_2((\mathcal{Q}_1, \Phi, \mathcal{Q}_2), (\mathcal{Q}_1, \Psi, \mathcal{Q}_2)) = \{\alpha: \Phi \Rightarrow \Psi \mid \alpha \text{ is a natural transformation}\}$$

A 2-morphism is an equivalence between ways of reaching the same result.

Definition 5.2 (Lax 2-functor $F_2$).

$F_2: \mathbf{DensityMat} \to \mathbf{Exp}_2$

On objects:

$$F_2(\rho) := (\mathrm{Spectrum}(\rho_E), \mathrm{Quality}(\rho_E), \mathrm{Context}(\Gamma_{-E}))$$

On 1-morphisms:

$$F_2(\Phi: \rho_1 \to \rho_2) := (F_2(\rho_1), \Phi, F_2(\rho_2))$$

Compositor (key element):

For $\Phi: \rho_1 \to \rho_2$ and $\Psi: \rho_2 \to \rho_3$ define the 2-isomorphism (compositor):

$$\mu_{\Psi,\Phi}: F_2(\Psi \circ \Phi) \Rightarrow F_2(\Psi) \circ F_2(\Phi)$$

Explicitly:

$$\mu_{\Psi,\Phi}: (F_2(\rho_1), \Psi \circ \Phi, F_2(\rho_3)) \xRightarrow{\cong} (F_2(\rho_1), \Phi, F_2(\rho_2)) \circ (F_2(\rho_2), \Psi, F_2(\rho_3))$$

Interpretation: The compositor $\mu_{\Psi,\Phi}$ is a 2-isomorphism witnessing the equivalence of the direct path $\rho_1 \xrightarrow{\Psi \circ \Phi} \rho_3$ and the composite path $\rho_1 \xrightarrow{\Phi} \rho_2 \xrightarrow{\Psi} \rho_3$.

Theorem 5.2' (Coherence).

The compositor $\mu$ satisfies Mac Lane's coherence conditions:

1. Associativity: For $\Phi: \rho_1 \to \rho_2$, $\Psi: \rho_2 \to \rho_3$, $\Xi: \rho_3 \to \rho_4$ the diagram commutes:

F₂(ξ∘ψ∘φ) ══════════════════════════════► F₂(ξ)∘F₂(ψ∘φ) ══► F₂(ξ)∘F₂(ψ)∘F₂(φ)
     ║                                           ║                    ║
     ║ μ_{ξ,ψ∘φ}                                 ║                    ║
     ▼                                           ▼                    ▼
F₂(ξ∘ψ)∘F₂(φ) ═══════════════════════════════════════════► F₂(ξ)∘F₂(ψ)∘F₂(φ)

2. Unitality: For the identity morphism $\mathrm{id}_\rho$:

$$\mu_{\Phi, \mathrm{id}} = \mathrm{id}_{F_2(\Phi)}, \quad \mu_{\mathrm{id}, \Phi} = \mathrm{id}_{F_2(\Phi)}$$

Proof (extended):

Mac Lane coherence for bicategories requires verifying:

• The pentagon identity for associators
• The triangle identity for the interaction of associators with unitors

Key observation: The category of CPTP channels is a strict 2-category, i.e., composition of morphisms is strictly associative:

$$(\Xi \circ \Psi) \circ \Phi = \Xi \circ (\Psi \circ \Phi) \quad \text{(equality, not isomorphism)}$$

Consequence: In a strict 2-category:

1. Associator $\alpha_{(\Xi,\Psi,\Phi)}$ = id (identity 2-morphism)
2. Left unitor $\lambda_\Phi$ = id
3. Right unitor $\rho_\Phi$ = id

Verification of the pentagon identity:

For morphisms $\Omega, \Xi, \Psi, \Phi$ the pentagon:

((Ω∘Ξ)∘Ψ)∘Φ ══α══► (Ω∘Ξ)∘(Ψ∘Φ) ══α══► Ω∘(Ξ∘(Ψ∘Φ))
      ║                                       ║
     α∘id                                   id∘α
      ▼                                       ▼
(Ω∘(Ξ∘Ψ))∘Φ ════════════α════════════► Ω∘((Ξ∘Ψ)∘Φ)

With $\alpha = \text{id}$ the entire pentagon commutes trivially. ✓

Verification of the triangle identity:

For morphisms $\Psi, \Phi$ the triangle:

(Ψ∘id)∘Φ ══α══► Ψ∘(id∘Φ)
     ║              ║
    ρ∘id         id∘λ
     ▼              ▼
   Ψ∘Φ ═══════► Ψ∘Φ

With $\alpha = \lambda = \rho = \text{id}$ it commutes trivially. ✓

Conclusion: The compositor $\mu$ satisfies Mac Lane coherence, since the bicategory $\mathbf{Exp}_2$ is strict (strictly associative). ∎

---

5.2.3 History as the structure of the bicategory

Theorem 5.3' (Emergent history)

In the bicategory $\mathbf{Exp}_2$ history is derived as a structure, not postulated:

$$\mathrm{Hist}(\mathcal{Q}) := \pi_1(\mathbf{Exp}_2, \mathcal{Q}) = \{\text{classes of 1-morphisms } \mathcal{Q} \to \mathcal{Q}\}$$

where $\pi_1$ is the fundamental groupoid of the bicategory.

Consequences:

1. The direct path $\rho_1 \xrightarrow{\Psi \circ \Phi} \rho_3$ and the composite path $\rho_1 \xrightarrow{\Phi} \rho_2 \xrightarrow{\Psi} \rho_3$ are 2-isomorphic, but not equal. This is precisely the difference in histories!

2. History information is preserved in the structure of 1-morphisms and is not lost.

3. Connection to the ∞-groupoid (section 10): $\mathbf{Exp}_2$ embeds in $\mathbf{Exp}_\infty$ as a 2-truncation:

$$\tau_{\leq 2}(\mathbf{Exp}_\infty) \simeq \mathbf{Exp}_2$$

---

5.2.4 Comparison with old strategies

Criterion	Strategy A (trivial)	Strategy B (homotopy)	Lax 2-functor
Strict functoriality	+ (at cost of losing history)	— (only up to homotopy)	+ (lax)
History preservation	—	Partially (implicit)	+ (in structure of morphisms)
Mathematical rigor	Low (ad hoc)	Medium	High
Consistency with §10	—	Partial	Full
Coherence	Trivial	Not verified	+ Mac Lane

---

5.2.5 Canonical definition (replacing Strategy A)

Canonical definition of functor F

Adopted definition: $F$ is a lax 2-functor $F_2: \mathbf{DensityMat} \to \mathbf{Exp}_2$.

1. Objects of Exp₂ — triples $(\lambda, [q], c)$ without history
2. 1-morphisms — transitions encoding history
3. 2-morphisms — equivalences of paths
4. Compositor $\mu$ — witness of equivalence of direct and composite paths

The strict 1-functor $F$ (Definition 4.1) is obtained as the strictification of $F_2$:

$$F = \mathrm{St}(F_2): \mathbf{DensityMat} \to \mathrm{Ho}(\mathbf{Exp}_2)$$

where $\mathrm{Ho}(\mathbf{Exp}_2)$ is the homotopy category (the 1-category obtained by factoring by 2-isomorphisms).

Conclusion: The lax 2-functor $F_2$ is the only mathematically rigorous solution to the functoriality problem with history. ∎

5.3 Summary theorem

Theorem 5.3 (Functoriality of F — refined formulation)

There exists a lax 2-functor:

$F_2: \mathbf{DensityMat} \to \mathbf{Exp}_2$

satisfying:

1. Identity: $F_2(\mathrm{id}_\rho) = \mathrm{id}_{F_2(\rho)}$ (strict)
2. Composition: $F_2(\Psi \circ \Phi) \cong F_2(\Psi) \circ F_2(\Phi)$ via a coherent 2-isomorphism $\mu_{\Psi,\Phi}$
3. Coherence: Mac Lane diagrams commute

The strict 1-functor $F: \mathbf{DensityMat} \to \mathbf{Exp}$ (without history as a component) is the strictification of $F_2$.

Proof:

• Theorem 5.1 (identity): unchanged
• Theorem 5.2' (composition): lax functoriality with compositor μ
• Coherence: follows from associativity of CPTP

Corollary: History is not a component of Exp objects, but a structure of the bicategory $\mathbf{Exp}_2$, consistent with the ∞-groupoid $\mathbf{Exp}_\infty$ (section 10). ∎

---

6. Topos structure

6.1 Is Exp a topos?

Theorem 6.1. The category $\mathbf{Exp}$ is not a topos in the general case.

Proof:

A topos requires:

1. All finite limits
2. All finite colimits
3. Exponentials
4. Subobject classifier

Verify the presence of these structures:

1. Finite limits:

Terminal object:

$$1_\mathcal{Q} := (\lambda^*, [q^*], c^*, h^*)$$

where $\lambda^* = (1, 0, \ldots, 0)$, $[q^*] = [|1\rangle]$, $c^* = \Gamma_{\max}$, $h^* = \varnothing$ (empty history).

But this is not uniquely defined — any pure state gives a terminal object.

$\Rightarrow$ The terminal object is not unique (up to isomorphism — it is unique, but the category is not skeletal).

Products:

$$\mathcal{Q}_1 \times \mathcal{Q}_2 := ((\lambda_1, \lambda_2), ([q_1], [q_2]), (c_1, c_2), (h_1, h_2))$$

The direct product is defined, but it exceeds the original space $\mathcal{Q}$.

$\Rightarrow$ Products are not closed in $\mathbf{Exp}$.

2. Subobject classifier:

For a topos we need an object $\Omega$ and a morphism $\mathrm{true}: 1 \to \Omega$ such that for any monomorphism $m: S \to \mathcal{Q}$ there is a unique characteristic morphism $\chi: \mathcal{Q} \to \Omega$.

In $\mathbf{Exp}$:

• Subobjects of $\mathcal{Q}$ are "parts of experience"
• There is no obvious universal classifier

$\Rightarrow$ The subobject classifier does not exist in the natural sense.

Conclusion: $\mathbf{Exp}$ is not a topos. ∎

Consequences of the absence of topos structure

The absence of topos structure has important implications:

1. No internal logic: Toposes have an internal language (intuitionistic logic). $\mathbf{Exp}$ does not have such a language — the logic of experiential content cannot be defined inside the category.

2. No subobject classifier: It is impossible to define the "truth" of experiential content within $\mathbf{Exp}$. The question "Is a given experiential content true?" has no meaning in the categorical formalism.

3. Limitations for type theory: One cannot construct dependent types on $\mathbf{Exp}$ directly.

This is not a defect of UHM, but a reflection of the nature of experience: subjective experience cannot be formalized as a logical system.

6.2 What structure does Exp possess?

Theorem 6.2. $\mathbf{Exp}$ is:

1. A category with finite products (in the extended sense)
2. An enriched category over metric spaces
3. A category with a fibration structure

Proof:

1. Fibration structure:

Projection onto the spectrum:

$$\pi: \mathbf{Exp} \to \Delta^{N-1}, \quad \pi(\lambda, [q], c, h) := \lambda$$

This is a fibration (Grothendieck fibration). Fibers:

$$\mathbf{Exp}_\lambda := \pi^{-1}(\lambda) = \mathbb{P}(\mathcal{H}_E)^N \times \mathcal{C} \times \mathrm{Hist}$$

2. Enrichment over Met (metric spaces):

Hom-sets are equipped with a metric:

$$d_{\mathrm{Hom}}(m_1, m_2) := d_\mathcal{Q}(m_1(\mathcal{Q}), m_2(\mathcal{Q})) \quad \text{for fixed } \mathcal{Q}$$

where $d_\mathcal{Q}$ is the complete metric on $\mathcal{Q}$.

3. Monoidal structure:

One can define a tensor product:

$$\mathcal{Q}_1 \otimes \mathcal{Q}_2 := \text{joint experience}$$

via the tensor product of density matrices:

$$F(\rho_1 \otimes \rho_2) =: F(\rho_1) \otimes F(\rho_2)$$

This makes $F$ a monoidal functor. ∎

6.3 Grothendieck topology on DensityMat and Exp

Fundamental definition

To construct an ∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$ one must explicitly specify a Grothendieck topology on the base category $\mathcal{C} = \mathbf{DensityMat}$.

6.3.1 Bures topology on DensityMat

Definition 6.1 (Bures metric, chordal form):

For density matrices $\rho, \sigma \in \mathbf{DensityMat}$:

$$d_B^{\mathrm{chord}}(\rho, \sigma) := \sqrt{2\left(1 - \sqrt{\mathrm{Fid}(\rho, \sigma)}\right)}$$

where $\mathrm{Fid}(\rho, \sigma) = \left(\mathrm{Tr}\sqrt{\sqrt{\rho}\sigma\sqrt{\rho}}\right)^2$ — fidelity. The notation $\mathrm{Fid}$ is used to distinguish from the functor $F: \mathbf{DensityMat} \to \mathbf{Exp}$.

note

Convention: the chordal form $d_B^{\mathrm{chord}} \in [0, \sqrt{2}]$ is used here. Angular form: $d_B^{\mathrm{angle}} = \arccos(\sqrt{\mathrm{Fid}})$. See notation convention.

Properties of the Bures metric:

Property	Formulation	Significance for UHM
Monotonicity	$d_B(\Phi(\rho), \Phi(\sigma)) \leq d_B(\rho, \sigma)$ for CPTP $\Phi$	Compatibility with morphisms
Riemannian	Induces a Riemannian structure on $\mathcal{D}(\mathcal{H})$	Geometry of the state space
Connection to fidelity	$d_B^2 = 2(1 - \sqrt{\mathrm{Fid}})$	Quantum interpretation

Definition 6.2 (Bures cover on DensityMat):

A family of CPTP-morphisms $\{\Phi_i: \rho_i \to \rho\}_{i \in I}$ forms a Bures cover of an object $\rho \in \mathbf{DensityMat}$ if:

$$\forall \epsilon > 0, \exists \delta > 0: \quad B_B(\rho, \delta) \subseteq \bigcup_{i \in I} \Phi_i(B_B(\rho_i, \epsilon))$$

where $B_B(\rho, r) = \{\sigma : d_B(\rho, \sigma) < r\}$ — open ball in the Bures metric.

Theorem 6.1 (Site axioms for DensityMat) [Т]:

The pair $(\mathbf{DensityMat}, J_{Bures})$ forms a Grothendieck site (Johnstone, Sketches of an Elephant, C2.1.9–12).

Proof.

We verify the three axioms of a Grothendieck topology on the category $\mathcal{C} = \mathbf{DensityMat}$ with objects $\rho \in \mathcal{D}(\mathbb{C}^7)$ and morphisms = CPTP channels.

Axiom 1 (Identity). The singleton family $\{\mathrm{id}_\rho: \rho \to \rho\}$ is a Bures cover of $\rho$. For any $\epsilon > 0$, choose $\delta = \epsilon$. Then $B_B(\rho, \delta) = B_B(\rho, \epsilon) = \mathrm{id}_\rho(B_B(\rho, \epsilon))$. $\checkmark$

Axiom 2 (Stability under pullback). Let $\{\Phi_i: \rho_i \to \rho\}_{i \in I}$ be a Bures cover of $\rho$, and let $\Psi: \sigma \to \rho$ be any CPTP morphism. We must show that the pullback family covers $\sigma$. By the CPTP contractivity of the Bures metric (Uhlmann 1976, Petz 1996): for any CPTP channel $\Psi$,

$d_B(\Psi(\sigma_1), \Psi(\sigma_2)) \leq d_B(\sigma_1, \sigma_2)$

This is the quantum data-processing inequality for the Bures metric, equivalent to monotonicity of fidelity under CPTP (Fuchs–van de Graaf 1999). Define the pullback family $\{\Psi_i: \sigma_i \to \sigma\}$ where $\sigma_i$ and $\Psi_i$ are constructed via the categorical pullback in $\mathbf{DensityMat}$. Since CPTP channels are contractive, any $\sigma' \in B_B(\sigma, \delta)$ satisfies $\Psi(\sigma') \in B_B(\rho, \delta)$, and by the covering property of $\{\Phi_i\}$, $\Psi(\sigma')$ lies in some $\Phi_i(B_B(\rho_i, \epsilon))$. The contractivity ensures the inverse image under $\Psi$ of a Bures ball is contained in a Bures ball of the same or larger radius. $\checkmark$

Axiom 3 (Transitivity / composition of covers). Let $\{\Phi_i: \rho_i \to \rho\}$ be a cover of $\rho$, and for each $i$, let $\{\Psi_{ij}: \rho_{ij} \to \rho_i\}$ be a cover of $\rho_i$. The composite family $\{\Phi_i \circ \Psi_{ij}\}$ covers $\rho$. Proof: for any $\sigma \in B_B(\rho, \delta)$, the first cover gives $\sigma \in \Phi_i(B_B(\rho_i, \epsilon_1))$ for some $i$. The second cover gives $B_B(\rho_i, \epsilon_1) \subseteq \bigcup_j \Psi_{ij}(B_B(\rho_{ij}, \epsilon_2))$. By the triangle inequality for $d_B$: $\sigma \in \Phi_i(\Psi_{ij}(B_B(\rho_{ij}, \epsilon_2)))$ for some $j$. The Bures metric satisfies the triangle inequality (it is a genuine metric on $\mathcal{D}(\mathbb{C}^7)$, Uhlmann 1976), so this composition is well-defined. $\checkmark$

Essentially small presentation. The space $\mathcal{D}(\mathbb{C}^7)$ is compact metrizable (closed bounded subset of $\mathbb{C}^{7 \times 7}$). By standard topology: every compact metrizable space has a countable dense subset. Fix a countable dense $\mathcal{C}_0 \subset \mathcal{D}(\mathbb{C}^7)$. The restriction $(\mathcal{C}_0, J_{Bures}|_{\mathcal{C}_0})$ is an essentially small site generating the same sheaf topos (Johnstone, Elephant, C2.2.3). This ensures Lurie's sheafification theorem (HTT 6.2.2.7: sheaves on a small site form an $\infty$-topos as left-exact localization of presheaves) applies: $\mathbf{Sh}_\infty(\mathcal{C}_0, J_{Bures}) \simeq \mathbf{Sh}_\infty(\mathcal{C}, J_{Bures})$ is an $\infty$-topos. $\blacksquare$

Dependencies: Uhlmann (1976) [standard], Petz (1996) [standard], Johnstone C2.1.9–12 [standard], Lurie HTT 6.2.2.7 [site → ∞-topos sheafification] + 6.1.0.6 [Giraud characterization of ∞-topos structure].

note

Framework dependency (see Rigour Stratification §T-76)

This site-level proof of T-76 is [T]: the three Grothendieck-topology axioms for $(\mathbf{DensityMat}, J_{Bures})$ are verified directly via CPTP contractivity of the Bures metric, and Lurie's sheafification theorem (HTT 6.2.2.7) is then applied. The Exp-extension (Claim 10.2 in §10.4) carries a weaker status — see §10.4 and the registry row for the Giraud-axiom verification that remains pending.

6.3.2 Induced topology on Exp

Theorem 6.2 (Consistency of topologies):

The functor $F: \mathbf{DensityMat} \to \mathbf{Exp}$ preserves covers:

$$\{\Phi_i: \rho_i \to \rho\} \text{ is a Bures cover} \quad \Rightarrow \quad \{F(\Phi_i): \mathcal{Q}_i \to \mathcal{Q}\} \text{ is a cover in } \mathbf{Exp}$$

Proof: Continuity of $F$ with respect to the metric: $d_{\mathcal{Q}}(F(\rho), F(\sigma)) \leq C \cdot d_B(\rho, \sigma)$ for some constant $C$. ∎

Important clarification

The fact that $\mathrm{Sh}(\mathbf{Exp})$ is a topos does not make the category $\mathbf{Exp}$ itself a topos. This is a standard result: sheaves on any site form a topos.

6.3.3 Sheaf topos on Exp

Definition 6.3 (Topology on Exp):

A cover $U \subset \mathrm{Ob}(\mathbf{Exp})$ is defined as:

$$\{\mathcal{Q}_i\}_{i \in I} \text{ covers } \mathcal{Q} \Leftrightarrow \bigcup_i B(\mathcal{Q}_i, \varepsilon) \supseteq B(\mathcal{Q}, \delta) \text{ for some } \varepsilon, \delta > 0$$

where $B(\mathcal{Q}, r)$ — open ball of radius $r$ in the metric $d_\mathcal{Q}$.

Theorem 6.3. $\mathrm{Sh}(\mathbf{Exp})$ is a topos.

Corollary: The logic of experiential content is interpreted in the topos $\mathrm{Sh}(\mathbf{Exp})$, where truth values are open sets.

6.3.4 Connection to L-unification

Theorem 6.4 (Classifier from Bures topology):

The subobject classifier $\Omega$ for $\mathbf{Sh}_\infty(\mathcal{C})$ is constructively defined as:

$$\Omega := \mathcal{O}(\mathcal{C}, d_B)$$

— the lattice of open sets in the Bures topology.

Characteristic morphisms:

For a subobject $S \hookrightarrow \Gamma$ the morphism $\chi_S: \Gamma \to \Omega$ is computed:

$$\chi_S(\Gamma') = \sup\{r \in [0,1] : B_B(\Gamma', r) \cap S \neq \emptyset\}$$

Corollary ($L_k$ constructively):

The Lindblad operators $L_k = \sqrt{\chi_{S_k}}$ receive a constructive definition via the Bures topology.

---

7. Limitations and alternatives

7.1 Identified limitations

Limitation 1: Basis dependence

The decomposition of $\Gamma$ into $\Gamma_E$ and $\Gamma_{-E}$ depends on the choice of basis $|i\rangle = \{|A\rangle, |S\rangle, \ldots, |U\rangle\}$.

Solution: The basis is determined by the physical interpretation of the 7 dimensions. This is not arbitrary, but part of the theory.

Limitation 2: Problem of time

History $h$ requires a time parameter, but $\mathbf{DensityMat}$ is a static category.

Solution 1: Work with the category $\mathbf{DensityMat}_T$ (with a time parameter).

Solution 2: Treat history as an external parameter that does not participate in morphisms.

Limitation 3: Irreversibility

CPTP channels are generally irreversible. Therefore:

• $F$ is not full
• $F$ is not faithful in the sense of reversibility of individual morphisms

This is not a bug but a feature: Irreversibility corresponds to the arrow of time in experience.

info

Faithfulness of F on $G_2$-orbits [Т]

Despite the irreversibility of individual CPTP channels, the $G_2$-rigidity theorem [Т] establishes faithfulness of the functor on objects (up to the gauge group):

$$F(\Gamma_1) \cong F(\Gamma_2) \quad \Longleftrightarrow \quad \Gamma_2 = U\Gamma_1 U^\dagger \text{ for some } U \in G_2$$

Kernel: $\ker(F) = \{\mathrm{Ad}_U : U \in G_2\}$. In other words, two states are phenomenologically identical if and only if their coherence matrices are related by a $G_2$-transformation. The functor $F$ is injective on the space $\mathcal{D}(\mathbb{C}^7)/G_2$ (34-dimensional).

7.2 Alternative constructions

Alternative A: Dual functor

Definition 7.1.

$$F^*: \mathbf{Exp} \to \mathbf{DensityMat}, \quad F^*(\mathcal{Q}) := \rho \text{ such that } F(\rho) = \mathcal{Q}$$

Problem: $F^*$ is not a functor because:

1. $F$ is not surjective (not all $\mathcal{Q}$ are reachable)
2. $F$ is not injective (different $\rho$ may give the same $\mathcal{Q}$ under full mixing)

Alternative B: 2-category

Definition 7.2 (2-category $\mathbf{Exp}^{(2)}$).

• 0-cells: Objects of $\mathbf{Exp}$
• 1-cells: Morphisms $F(\Phi)$
• 2-cells: Natural transformations between CPTP channels

$$\alpha: \Phi \Rightarrow \Psi \text{ defined as: } \alpha_\rho: \Phi(\rho) \to \Psi(\rho), \text{ natural in } \rho$$

Advantage: Captures "ways of transitioning between transitions".

Theorem T-192 (Exp^(2) is a strict 2-category) [Т]

Theorem T-192 [Т]

The construction $\mathbf{Exp}^{(2)}$ of Definition 7.2 satisfies all axioms of a strict 2-category (equivalently, a $\mathbf{Cat}$-enriched category): horizontal composition is strictly associative, vertical composition is strictly associative, and the interchange law holds.

Proof (verification of 5 axioms).

Axiom 1 (Vertical composition). For 2-cells $\alpha: \Phi \Rightarrow \Psi$ and $\beta: \Psi \Rightarrow \Chi$ (both natural transformations between CPTP channels), the vertical composite $\beta \circ_v \alpha: \Phi \Rightarrow \Chi$ is defined pointwise: $(\beta \circ_v \alpha)_\rho := \beta_\rho \circ \alpha_\rho$. This is a natural transformation because naturality squares compose: if $\alpha$ and $\beta$ are natural in $\rho$, then $\beta \circ_v \alpha$ is natural in $\rho$ (standard result, Mac Lane CWM IV.2). Associativity: $(\gamma \circ_v \beta) \circ_v \alpha = \gamma \circ_v (\beta \circ_v \alpha)$ follows from associativity of composition in the target category $\mathbf{Exp}$. $\checkmark$

Axiom 2 (Horizontal composition). For 2-cells $\alpha: \Phi_1 \Rightarrow \Phi_2$ and $\beta: \Psi_1 \Rightarrow \Psi_2$ with $\Phi_i: \rho_1 \to \rho_2$ and $\Psi_i: \rho_2 \to \rho_3$, the horizontal composite $\beta \circ_h \alpha: \Psi_1 \circ \Phi_1 \Rightarrow \Psi_2 \circ \Phi_2$ is the Godement product: $(\beta \circ_h \alpha)_\rho := \beta_{\Phi_2(\rho)} \circ \Psi_1(\alpha_\rho) = \Psi_2(\alpha_\rho) \circ \beta_{\Phi_1(\rho)}$ (interchange). Associativity: $(\gamma \circ_h \beta) \circ_h \alpha = \gamma \circ_h (\beta \circ_h \alpha)$ follows from functoriality of CPTP channels. $\checkmark$

Axiom 3 (Identity 2-cells). For each 1-cell $\Phi$, the identity 2-cell $\mathrm{id}_\Phi: \Phi \Rightarrow \Phi$ is the identity natural transformation: $(\mathrm{id}_\Phi)_\rho = \mathrm{id}_{\Phi(\rho)}$. This satisfies $\mathrm{id}_\Phi \circ_v \alpha = \alpha$ and $\alpha \circ_v \mathrm{id}_\Phi = \alpha$ for all 2-cells $\alpha$. $\checkmark$

Axiom 4 (Interchange law). For 2-cells $\alpha_1: \Phi_1 \Rightarrow \Phi_2$, $\alpha_2: \Phi_2 \Rightarrow \Phi_3$, $\beta_1: \Psi_1 \Rightarrow \Psi_2$, $\beta_2: \Psi_2 \Rightarrow \Psi_3$:

$(\beta_2 \circ_v \beta_1) \circ_h (\alpha_2 \circ_v \alpha_1) = (\beta_2 \circ_h \alpha_2) \circ_v (\beta_1 \circ_h \alpha_1)$

This is the standard interchange law for natural transformations (Mac Lane CWM II.5, Theorem 1), which holds in any 2-category of functors. Since CPTP channels are functors between C*-algebras of observables, and natural transformations between them satisfy interchange by the Eckmann–Hilton argument, the law holds. $\checkmark$

Axiom 5 (Identity 1-cells). For each 0-cell $\mathcal{Q}$, the identity 1-cell $\mathrm{id}_{\mathcal{Q}} = F(\mathrm{id}_\rho)$ is the identity experiential transformation. By Theorem 5.1 [Т] (first functor axiom): $F(\mathrm{id}_\rho) = \mathrm{id}_{F(\rho)}$. This satisfies the unit laws for horizontal composition. $\checkmark$

Strictness. All five axioms hold with equalities (not just isomorphisms), making $\mathbf{Exp}^{(2)}$ a strict 2-category. This is because:

• The 0-cells and 1-cells form the category $\mathbf{Exp}$ (already verified [Т])
• The 2-cells are natural transformations, which compose strictly
• No coherence conditions (associators, unitors) are needed — they are identities

Corollary (Lax 2-functor target). The lax 2-functor $F_2: \mathbf{DensityMat} \to \mathbf{Exp}^{(2)}$ (Definition 5.2, §5.2.2) has a valid target: $\mathbf{Exp}^{(2)}$ is a strict 2-category satisfying all required axioms. The compositor $\mu_{\Psi,\Phi}$ (eq in §5.2.2) is a 2-cell in $\mathbf{Exp}^{(2)}$, and Mac Lane's coherence conditions (pentagon + triangle, verified in §5.2.2) are satisfied. $\blacksquare$

Dependencies: Theorem 5.1 [Т] (F preserves identities), Mac Lane CWM II.5/IV.2 (standard 2-category theory), Eckmann–Hilton argument (standard).

Alternative C: $\infty$-category (quasicategory)

[Т] Proved

The construction $\mathbf{Exp}_\infty := \text{Sing}(\mathcal{E})$ is an ∞-groupoid [Т]. Proof: for any topological space $X$ the construction $\mathrm{Sing}(X)$ (singular simplicial set) gives a Kan complex (Milnor's theorem). The space $\mathcal{E}$ is metrizable (Bures–Fubini–Study metric), so $\text{Sing}(\mathcal{E})$ is automatically an ∞-groupoid. All required properties (HoTT-logic, subobject classifier, Postnikov truncations) follow from the ∞-toposness of $\mathbf{Sh}_\infty(\mathbf{Exp})$ [T-76].

For a complete description of the dynamics of experiential content one can use $\infty$-categories:

$$\mathbf{Exp}_\infty := \mathrm{Sing}(\mathcal{E})$$

— the singular complex of the space $\mathcal{E}$.

$n$-morphisms are $n$-simplices in $\mathcal{E}$, corresponding to $n$-parameter families of transitions.

Alternative D: †-category (dagger category)

Natural for quantum mechanics

†-categories are categories with a contravariant functor $\dagger: \mathbf{C} \to \mathbf{C}$ satisfying $\dagger \circ \dagger = \mathrm{id}$. This is a natural formalism for quantum mechanics, where $\dagger$ corresponds to Hermitian conjugation.

Definition 7.3 (†-category $\mathbf{DensityMat}^\dagger$).

$\mathbf{DensityMat}$ with additional structure:

$$\dagger: \mathrm{Mor}(\rho_1, \rho_2) \to \mathrm{Mor}(\rho_2, \rho_1), \quad \Phi^\dagger := \Phi^* \text{ (adjoint channel)}$$

Advantages:

1. Naturally includes reversibility (unitary channels)
2. Connection to $C^*$-algebras
3. Categorical quantum mechanics (Abramsky, Coecke)

Question: Does $\mathbf{Exp}$ inherit the †-structure?

$$F(\Phi^\dagger) \stackrel{?}{=} F(\Phi)^\dagger$$

This requires defining $\dagger$ on $\mathbf{Exp}$, which is nontrivial.

Alternative E: $\infty$-topos

Definition 7.4 ($\infty$-topos over Exp).

One can construct an $\infty$-topos $\mathbf{Sh}_\infty(\mathbf{Exp})$ — an $\infty$-category of $\infty$-sheaves on $\mathbf{Exp}$.

Advantages:

1. Rich homotopical structure
2. Internal language (homotopy type theory)
3. Connection to derived algebraic geometry

Status: Research program. Requires defining an $\infty$-topology on $\mathbf{Exp}$.

7.3 Recommended construction

For practical purposes of UHM it is recommended:

Goal	Construction	Status
Basic theory (canonical)	Lax 2-functor $F_2: \mathbf{DensityMat} \to \mathbf{Exp}_2$	[Т] Formalized (§5.2)
Strict functor (simplification)	Strictification $F = \mathrm{St}(F_2)$	[Т] Corollary
Metric structure	$\mathbf{Exp}_{\mathrm{Met}}$ (enriched over Met)	[Т] Defined
Logical constructions	Sheaf topos $\mathrm{Sh}(\mathbf{Exp}_2)$	[С] Sketch
Dynamics and history	Bicategory $\mathbf{Exp}_2$ (§5.2.2)	[Т] Formalized
Quantum structure	†-category $\mathbf{DensityMat}^\dagger$	[П] Program
Homotopy theory	$\infty$-topos $\mathbf{Sh}_\infty(\mathbf{Exp}_\infty)$	[Т] Consistent with §10

Development priorities

1. Completed: Lax 2-functor $F_2$ — canonical solution to the history problem
2. Short-term: Refine the metric structure $\mathbf{Exp}_{\mathrm{Met}}$
3. Medium-term: Construct $\mathrm{Sh}(\mathbf{Exp}_2)$ and investigate the internal logic
4. Long-term: Investigate the †-structure and connection to categorical quantum mechanics

---

8. Phenomenal completeness

8.1 Definition of phenomenal completeness

Question: Can the structure of the Holon (Γ, 7 dimensions, functor F) describe any phenomenal construction?

Definition 8.1 (Phenomenal completeness). A theory is phenomenally complete if for any possible phenomenal state $\mathcal{Q}^*$ there exists a density matrix $\Gamma$ such that $F(\Gamma) = \mathcal{Q}^*$.

$$\text{Phenomenal completeness} := \mathrm{Im}(F) = \mathbf{Exp}$$

8.2 Thesis of structural sufficiency

Thesis (Structural sufficiency)

The experiential space $\mathcal{E} = \Delta^{N-1} \times_{\mathrm{Spec}} \mathbb{P}(\mathcal{H}_E)^N \times \mathcal{C} \times \mathrm{Hist}$ is structurally sufficient for describing any phenomenal experience satisfying physical constraints.

Justification:

Any phenomenal state is characterized by:

Phenomenal aspect	Mathematical component	Structure
Intensity (amplitude of interiority state)	Spectrum $\{\lambda_i\}$	Simplex $\Delta^{N-1}$ — continuous, $(N-1)$-dimensional
Quality (character of interiority state)	Eigenvectors $\{[q_i]\}$	$\mathbb{P}(\mathcal{H}_E)^N$ — compact, connected
Context (modulation)	Coherences $\gamma_{Ej}$	$\mathcal{C}$ — context space
Temporality (history)	Trajectory $\rho_E(t)$	$\mathrm{Hist}$ — function space

Key property: The dimension of $\mathcal{E}$ is not fixed a priori — $\mathcal{H}_E$ can be a subspace of $\mathbb{C}^7$ or an extension for complex systems.

8.3 Limitation: F is not surjective

Theorem 8.1 (Limitation of the image of F)

The functor $F: \mathbf{DensityMat} \to \mathbf{Exp}$ is not surjective:

$$\mathrm{Im}(F) \subsetneq \mathrm{Ob}(\mathbf{Exp})$$

Proof:

Not all points $\mathcal{Q} = (\lambda, [q], c, h) \in \mathcal{E}$ are reachable through a density matrix, because:

1. Positivity constraint: $\Gamma \geq 0$ imposes nontrivial constraints on admissible combinations $(\lambda, [q])$
2. Normalization constraint: $\mathrm{Tr}(\Gamma) = 1$
3. Hermiticity constraint: $\Gamma^\dagger = \Gamma$ ∎

8.4 Physical interpretation: unreachable states

Question: Are unreachable $\mathcal{Q} \notin \mathrm{Im}(F)$ meaningful phenomenal states?

Thesis (Physical filtering): Unreachable states are mathematical artifacts that do not correspond to physically possible configurations:

Type of unreachability	Example	Physical reason
Negative "probabilities"	$\lambda_i < 0$	Violation of $\Gamma \geq 0$
Incompatible qualities	$[q_i] \perp [q_j]$ with $\lambda_i = \lambda_j = 0.5$ for certain structures	Entanglement constraints
Non-physical history	Discontinuous trajectory $\rho_E(t)$	Violation of unitarity

Corollary

Phenomenal completeness holds for physically admissible states:

$$\forall \mathcal{Q} \in \mathcal{E}_{\text{phys}}: \exists \Gamma: F(\Gamma) = \mathcal{Q}$$

where $\mathcal{E}_{\text{phys}} := \mathrm{Im}(F)$ — the physically realizable subset.

8.5 Complex phenomenal constructions

How the theory describes nontrivial phenomenal structures:

Intentionality (directedness toward an object)

Mechanism: Coherences $\gamma_{EA}$ (attention) and $\gamma_{ES}$ (structuring) connect the internal state $\rho_E$ with the representation of the object through dimensions $A$ (Articulation) and $S$ (Structure).

$$\text{Intentionality}(\Gamma) := \sum_{j \neq E} |\gamma_{Ej}|^2 \cdot \text{Content}(\rho_j)$$

where $\text{Content}(\rho_j)$ is the informational content of dimension $j$.

Open question

Formalization of $\text{Content}(\rho_j)$ requires clarification — this is a direction of research.

Empathy (intersubjective experience)

Mechanism: Composition of Holons through tensor product:

$$\Gamma_{12} \in \mathcal{L}(\mathcal{H}_1 \otimes \mathcal{H}_2)$$

Empathy arises when:

1. Correlation: $I(\mathbb{H}_1 : \mathbb{H}_2) > 0$ (mutual information)
2. Projection: $\rho_E^{(1)} \sim \rho_E^{(2)}$ (similarity of experiential states)

$$\mathrm{Empathy}(\Gamma_{12}) := \mathrm{Fid}(\rho_E^{(1)}, \rho_E^{(2)}) \cdot I(\mathbb{H}_1 : \mathbb{H}_2)$$

Research program

The transition from correlation to the subjective feeling of "what it is like to be the other" is a manifestation of the categorical gap (Axiom Ω⁷), not a defect of the formalism.

Ambivalence (complex emotions)

Mechanism: Mixed state with competing components:

$$\rho_E = \lambda_1 |q_1\rangle\langle q_1| + \lambda_2 |q_2\rangle\langle q_2|, \quad \lambda_1 \approx \lambda_2$$

where $d_{FS}([q_1], [q_2]) \approx \pi/2$ (maximally distinct qualities).

Coherences $\gamma_{Ej}$ modulate which component is "active" at a given moment.

Temporal structures (anticipation, memory)

Mechanism: Component $\mathrm{Hist}$ in the experiential space:

$$\mathrm{Hist}(t, \tau) := \{\rho_E(t') : t' \in [t-\tau, t]\}$$

Phenomenon	Formalization
Recollection	Similarity of current $\rho_E(t)$ with elements of $\mathrm{Hist}$
Anticipation	Adaptation to patterns in $\mathrm{Hist}$ (predictive coding)
Nostalgia	Qualities $[q_i(t)]$ correlate with historical $[q_i(t')]$, $t' \ll t$

8.6 Status table

Phenomenal construction	Status	Comment
Simple qualia (color, pain)	✓ Formalized	Spectrum + qualities + context
Intensity/brightness	✓ Formalized	Eigenvalues $\lambda_i$
Qualitative differences	✓ Formalized	Fubini-Study metric $d_{FS}$
Unity of experience	✓ Formalized	Integration measure $\Phi$
Self-awareness	✓ Formalized	Operator $\varphi$, measure $R$
Ambivalence	✓ Formalized	Mixed states
Temporality	[С] Partial	$\mathrm{Hist}$, but time is an external parameter
Intentionality	[Т] Direction determined	$E$ is the unique $L$-mediated interiority dimension (T-183 [T]); direction := $\arg\max_j
Empathy	[С] Direction	Composition of Holons, open question
Altered states	[С] Quantitative	$R$, $\Phi$ — described, mechanism open

---

9. Quasi-functor for AI systems

Status: [П] Research program

This section describes an extension of the categorical formalism for neural network systems. See Protocol for measuring Γ for the full specification.

9.1 The nonlinearity problem

Neural network layers (GELU, Softmax) are nonlinear transformations. CPTP channels are linear over density matrices. The functoriality condition $G(f \circ g) = G(f) \circ G(g)$ is violated for nonlinear $f, g$.

9.2 Definition of quasi-functor

Definition 9.1 (Quasi-functor G):

A map $G: \mathbf{AIState} \rightsquigarrow \mathbf{DensityMat}$ with the condition of approximate functoriality:

$$\|G(f \circ g) - G(f) \circ G(g)\|_F \leq \varepsilon_{\text{functor}} \cdot \|f\|_{\text{op}} \cdot \|g\|_{\text{op}}$$

where $\varepsilon_{\text{functor}}$ is the nonlinearity parameter of the system.

Categories:

• $\mathbf{AIState}$: objects — activation vectors $\mathbf{h} \in \mathbb{R}^d$; morphisms — neural network layers $f: \mathbb{R}^d \to \mathbb{R}^d$
• $\mathbf{DensityMat}$: objects — density matrices $\Gamma \in \mathcal{D}(\mathbb{C}^7)$; morphisms — CPTP channels

9.3 NTK linearization

Definition 9.2 (Linearization in tangent space):

In the neighborhood of state $s_0$ the nonlinear function $f$ is approximated:

$$f(s) \approx f(s_0) + J_f(s_0) \cdot (s - s_0)$$

where $J_f(s_0) = \nabla_s f|_{s=s_0}$ — the Jacobian.

Theorem 9.1 (Approximate functoriality):

Let $f, g: \mathbb{R}^d \to \mathbb{R}^d$ be twice continuously differentiable ($C^2$) functions with bounded Jacobians $J_f, J_g$ and Hessians $H_f, H_g$. Denote the $C^2$-norm:

$$\|f\|_{C^2} := \sup_{s} \|J_f(s)\|_{\text{op}} + \sup_{s} \|H_f(s)\|_{\text{op}}$$

and analogously $\|g\|_{C^2}$. Let $s_0 \in \mathbb{R}^d$ be the linearization point, $\Delta s := s - s_0$ with $\|\Delta s\| \leq r$ (locality radius).

Then for NTK linearization:

$$\|(f \circ g)(s) - f^{\text{lin}} \circ g^{\text{lin}}(s)\|_F \leq \frac{1}{2} \|f\|_{C^2} \cdot \|g\|_{C^2} \cdot (1 + \|g\|_{C^2}) \cdot r^2 + O(r^3),$$

where $f^{\text{lin}}(u) := f(g(s_0)) + J_f(g(s_0)) \cdot (u - g(s_0))$, $g^{\text{lin}}(s) := g(s_0) + J_g(s_0) \cdot \Delta s$.

In the NTK regime ($r = O(1)$, nonlinearity as $\|f\|_{C^2}^2$): $O(\|f\|_{C^2}^2 \cdot \|g\|_{C^2}^2)$. $\square$

Proof.

Step 1 (Taylor expansion for $g$). Since $g \in C^2$, Taylor's formula with the Lagrange remainder gives:

$$g(s_0 + \Delta s) = g(s_0) + J_g(s_0) \Delta s + \frac{1}{2} \Delta s^T H_g(\xi_g) \Delta s,$$

where $\xi_g \in [s_0, s_0+\Delta s]$ is an intermediate point. Denote:

$$u := g(s_0 + \Delta s) - g(s_0) = J_g(s_0) \Delta s + r_g, \quad r_g := \frac{1}{2} \Delta s^T H_g(\xi_g) \Delta s.$$

Remainder estimate: $\|r_g\| \leq \frac{1}{2} \|H_g\|_{\text{op}} \cdot \|\Delta s\|^2$.

Step 2 (Taylor expansion for $f$). Similarly, $f \in C^2$ gives:

$$f(g(s_0) + u) = f(g(s_0)) + J_f(g(s_0)) u + \frac{1}{2} u^T H_f(\xi_f) u,$$

where $\xi_f \in [g(s_0), g(s_0) + u]$.

Step 3 (Comparison with linear composition). The linear composition:

$$f^{\text{lin}}(g^{\text{lin}}(s)) = f(g(s_0)) + J_f(g(s_0)) \cdot J_g(s_0) \Delta s.$$

The true composition:

$$(f \circ g)(s) = f(g(s_0)) + J_f(g(s_0)) u + \frac{1}{2} u^T H_f(\xi_f) u$$ $$= f(g(s_0)) + J_f(g(s_0)) \cdot (J_g(s_0) \Delta s + r_g) + \frac{1}{2} u^T H_f(\xi_f) u.$$

Step 4 (Difference). Subtracting:

$$(f \circ g)(s) - f^{\text{lin}}(g^{\text{lin}}(s)) = J_f(g(s_0)) \cdot r_g + \frac{1}{2} u^T H_f(\xi_f) u.$$

Step 5 (Estimates).

(i) First term:

$$\|J_f(g(s_0)) \cdot r_g\| \leq \|J_f\|_{\text{op}} \cdot \|r_g\| \leq \|J_f\|_{\text{op}} \cdot \frac{1}{2} \|H_g\|_{\text{op}} \|\Delta s\|^2 \leq \frac{1}{2} \|f\|_{C^2} \cdot \|g\|_{C^2} \cdot r^2.$$

(ii) Second term. Using $\|u\| \leq \|J_g\| \|\Delta s\| + \|r_g\| \leq \|g\|_{C^2} \cdot r + \frac{1}{2}\|g\|_{C^2} r^2$:

$$\|u\|^2 \leq \|g\|_{C^2}^2 r^2 \cdot (1 + \tfrac{1}{2} r)^2 \leq 2 \|g\|_{C^2}^2 r^2 \quad \text{for } r \leq 1.$$

Then:

$$\left\| \frac{1}{2} u^T H_f(\xi_f) u \right\| \leq \frac{1}{2} \|H_f\|_{\text{op}} \cdot \|u\|^2 \leq \|f\|_{C^2} \cdot \|g\|_{C^2}^2 \cdot r^2.$$

Step 6 (Combining estimates).

$$\|(f \circ g)(s) - f^{\text{lin}}(g^{\text{lin}}(s))\| \leq \frac{1}{2} \|f\|_{C^2} \cdot \|g\|_{C^2} \cdot r^2 + \|f\|_{C^2} \cdot \|g\|_{C^2}^2 \cdot r^2$$ $$= \|f\|_{C^2} \cdot \|g\|_{C^2} \cdot (\tfrac{1}{2} + \|g\|_{C^2}) \cdot r^2$$ $$\leq \|f\|_{C^2} \cdot \|g\|_{C^2} \cdot (1 + \|g\|_{C^2}) \cdot r^2.$$

When $\|g\|_{C^2} \gtrsim 1$ (typical NTK regime): leading order $\|f\|_{C^2} \cdot \|g\|_{C^2}^2 \cdot r^2$. Symmetrized estimate (through $\max(\|f\|, \|g\|)$): $O(\|f\|^2 \cdot \|g\|^2)$ when $r = O(1)$. $\blacksquare$

Corollary (CPTP linearization). The quasi-functor $G$ maps the linearization to a CPTP channel: $G(f)^{\text{lin}} = \Phi_f^{\text{lin}}$, where $\Phi_f^{\text{lin}}(\Gamma) = \Gamma + J_f \Gamma J_f^T$ (affine approximation of CPTP channel). The error:

$$\|G(f \circ g) - \Phi_f^{\text{lin}} \circ \Phi_g^{\text{lin}}\|_F \leq \|G\|_{\text{Lip}} \cdot \|(f \circ g) - f^{\text{lin}} \circ g^{\text{lin}}\|_F,$$

where $\|G\|_{\text{Lip}}$ is the Lipschitz constant of the map $G: \mathbf{AIState} \to \mathbf{DensityMat}$. Consequently:

$$\|G(f \circ g) - G(f)^{\text{lin}} \circ G(g)^{\text{lin}}\|_F = O(\|f\|_{C^2}^2 \cdot \|g\|_{C^2}^2 \cdot r^2).$$

Status: [T] (upgraded from [С]). Theorem 9.1 is proven with an explicit error bound.

Results used:

• Taylor's formula with Lagrange remainder (standard, Rudin "Principles of Mathematical Analysis");
• Submultiplicativity of matrix operator norms;
• Lipschitz continuity of $G$ (regularity assumption on the AI-state → density matrix map, standard for PCA-based constructions).

Consistency check:

• Does not rely on other UHM theorems (pure analysis);
• $C^2$-regularity of $f, g$ — standard assumption for smooth neural network layers (GELU, Softmax, Layer Norm — all $C^\infty$);
• Radius restriction $r \leq 1$ — local NTK regime, standard for linearized approximations.

9.4 Categorical diagram

                    G (quasi-functor)              F
    AIState ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─► DensityMat ──────────► Exp
       │                                 │                    │
       │ f (nonlinear)                   │ Φ_f^lin (CPTP)     │ morphisms
       ▼                                 ▼                    ▼
    AIState ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─► DensityMat ──────────► Exp
                    G                              F

Approximate commutativity condition:

$$\|F(G(f(s))) - F(\Phi_f^{\text{lin}}(G(s)))\|_{\text{Exp}} \leq \varepsilon_{\text{total}}$$

9.5 Open questions

1. Estimating $\varepsilon_{\text{functor}}$: For which architectures is the error acceptable?
2. Optimality of NTK: Are there better linearization methods?
3. Uniqueness of G: Is there a canonical choice of quasi-functor?

---

10. ∞-groupoid and ∞-topos for emergent time

Status: [Т] Proved

This section describes an extension of the categorical structure for emergent time. History Hist is derived as a structure of the ∞-groupoid, not postulated.

Proof: $\mathbf{Exp}_\infty := \text{Sing}(\mathcal{E})$ — ∞-groupoid [Т]. The space $\mathcal{E}$ is topological (Bures–Fubini–Study metric), so $\text{Sing}(\mathcal{E})$ is automatically a Kan complex (Milnor's theorem), i.e., an ∞-groupoid. Combined with T-76 ($\mathbf{Sh}_\infty(\mathbf{Exp})$ — ∞-topos), all properties: internal HoTT-logic, subobject classifier, Postnikov truncations — follow.

Status distinction: Sing(E) construction vs. physical interpretation

The bare construction $\mathrm{Sing}(\mathcal{E})$ — ∞-groupoid [Т]: for any topological space $X$ the construction $\mathrm{Sing}(X)$ gives a Kan complex (Milnor's theorem), and $\mathcal{E}$ is metrizable. This is pure mathematics, requiring no additional hypotheses.

Physical interpretation (correspondence L4) — [П] (program): the identification of the ∞-categorical structure of $\mathrm{Exp}_\infty$ with infinite depth of self-observation, full Postnikov tower, and historical extension requires additional physical assumptions that are not proved.

Dependencies: Level L4 (infinite depth of self-observation), full ∞-categorical superstructure (Postnikov tower, historical extension) and the upper bound of SAD depend on the physical interpretation.

Mitigating factor: The theorem SAD_MAX = 3 [Т] (T-142) limits the physically achievable depth to level L3. Level L4 is formally defined but physically unreachable (by analogy with Lawvere incompleteness). Therefore, the openness of the status of the physical interpretation does not affect the physical predictions of the theory — all observables live at levels L0–L3, defined without the L4-correspondence.

10.1 ∞-groupoid of experiential paths

Definition 10.1 (∞-category Exp_∞).

0-cells (objects):

$$\text{Ob}(\mathbf{Exp}_\infty) = \mathcal{E} = \Delta^{N-1} \times_{\text{Spec}} \mathbb{P}(\mathcal{H}_E)^N \times \mathcal{C}$$

(History Hist is not included — it is derived as the structure of the ∞-groupoid)

1-morphisms:

$$\text{Mor}_1(\mathcal{Q}_1, \mathcal{Q}_2) = \{\gamma: [0,1] \to \mathcal{E} \mid \gamma(0) = \mathcal{Q}_1, \gamma(1) = \mathcal{Q}_2\}$$

2-morphisms:

$$\text{Mor}_2(\gamma_1, \gamma_2) = \text{homotopies between } \gamma_1 \text{ and } \gamma_2$$

n-morphisms:

$$\text{Mor}_n = n\text{-parameter families of paths}$$

10.2 Time as a 1-morphism

Definition 10.2 (Categorical time).

Time is a 1-morphism in $\mathbf{Exp}_\infty$:

$$\tau: \mathcal{Q}_1 \to \mathcal{Q}_2$$

Direction of time — choice of orientation on 1-morphisms:

$$\sigma: \text{Mor}_1(\mathcal{Q}_1, \mathcal{Q}_2) \to \{+1, -1\}$$

Equivalent moments of time — 2-isomorphic 1-morphisms.

10.3 Emergent history

Claim 10.1 (History as loop space) (requires verification).

In the ∞-groupoid $\mathbf{Exp}_\infty$:

1. History — automatically arises as a loop space:

   $$\text{Hist}(\mathcal{Q}) := \Omega_\mathcal{Q}(\mathbf{Exp}_\infty) = \{\gamma: S^1 \to \mathcal{E} \mid \gamma(0) = \gamma(1) = \mathcal{Q}\}$$

2. Temporal structure — homotopy type:

   $$\pi_1(\mathbf{Exp}_\infty, \mathcal{Q}) = \text{"cyclic time" at point } \mathcal{Q}$$

10.4 ∞-topos of sheaves

Definition 10.3 (∞-topos Sh_∞(Exp)).

$\mathbf{Sh}_\infty(\mathbf{Exp})$ — category of ∞-sheaves on $\mathbf{Exp}_\infty$:

1. ∞-topology: Cover = family of paths covering a neighborhood
2. ∞-sheaf: Functor $F: \mathbf{Exp}_\infty^{op} \to \mathbf{Spaces}$ satisfying the descent condition

Claim 10.2 (requires verification). $\mathbf{Sh}_\infty(\mathbf{Exp})$ is an ∞-topos and possesses:

1. Internal logic: Homotopy type theory (HoTT)
2. Internal time: Modality of the type "in the future", "in the past"
3. Subobject classifier: ∞-groupoid of truth values

Scope: Exp-extension is weaker than the site-level result

The site-level part of T-76 — that $(\mathbf{DensityMat}, J_{Bures})$ is a Grothendieck site and $\mathbf{Sh}_\infty(\mathbf{DensityMat}, J_{Bures})$ is an ∞-topos — is [T] via the direct axiom-verification in §6.3.1 plus HTT 6.2.2.7. The extension to $\mathbf{Sh}_\infty(\mathbf{Exp})$ stated here is Claim 10.2 (requires verification): full Giraud-axiom verification (small colimits, effective unions, descent) for the $\mathbf{Exp}$-site is pending. See Rigour Stratification §T-76.

Corollary: The logic of experiential content is a temporal modal logic, derivable from the internal structure of the ∞-topos.

10.5 Extended categorical diagram

                    G                           F
DensityMat_C  ──────────► DensityMat  ────────────► Exp
    │                         │                      │
    │ restriction             │ CPTP                 │ induced
    ▼                         ▼                      ▼
DensityMat_C  ──────────► DensityMat  ────────────► Exp

                                        ↓ embed

                              Exp_∞ (∞-groupoid)
                                        ↓ sheafify

                              Sh_∞(Exp) (∞-topos)

where:

• DensityMat_C — category with Page–Wootters constraint
• G — functor "conditional states"
• Exp_∞ — ∞-groupoid of paths
• Sh_∞(Exp) — ∞-topos of sheaves

10.6 Connection to the interiority hierarchy (L0→L4)

Key connection

Interiority levels L0→L4 correspond to n-truncations of the ∞-groupoid $\mathbf{Exp}_\infty$. This provides a unified categorical construction for the entire consciousness hierarchy.

Claim 10.3 (Homotopic classification of interiority) (requires verification):

Interiority levels correspond to n-truncations of the ∞-groupoid:

$$L_n \leftrightarrow \tau_{\leq n}(\mathbf{Exp}_\infty)$$

where $\tau_{\leq n}$ — n-truncation (trivializes all homotopy groups $\pi_k$ for $k > n$).

Correspondence:

Level	n-truncation	Homotopy groups	Categorical structure
L0	$\tau_{\leq 0}$	$\pi_0 \neq 0$, $\pi_{k>0} = 0$	Set (discrete states)
L1	$\tau_{\leq 1}$	$\pi_0, \pi_1 \neq 0$	Groupoid (phenomenal paths)
L2	$\tau_{\leq 2}$	$\pi_0, \pi_1, \pi_2 \neq 0$	Bicategory (reflection)
L3	$\tau_{\leq 3}$	$\pi_0, \pi_1, \pi_2, \pi_3 \neq 0$	Tricategory (meta-reflection)
L4	$\tau_{\leq \infty}$	All $\pi_k \neq 0$	∞-groupoid (complete structure)

Proof (sketch):

1. L0: Interiority — existence of an object in $\mathrm{Ob}(\mathbf{Exp}_\infty)$, which is equivalent to nontriviality of $\pi_0$.

2. L1: Phenomenal geometry — existence of paths between states, i.e., $\pi_1 \neq 0$.

3. L2: Cognitive qualia — capacity for reflection (2-morphisms = homotopies between paths), i.e., $\pi_2 \neq 0$.

4. L3: Network consciousness — meta-reflection (3-morphisms = homotopies between homotopies), i.e., $\pi_3 \neq 0$.

5. L4: Unitary consciousness — full ∞-structure, all $\pi_k \neq 0$. ∎

Criteria in terms of Γ:

Level	Condition	n-connectivity
L0→L1	$\mathrm{rank}(\rho_E) > 1$	1-connectivity
L1→L2	$R \geq 1/3$, $\Phi \geq 1$	2-connectivity
L2→L3	$R^{(2)} \geq 1/4$	3-connectivity
L3→L4	$\lim_n R^{(n)} > 0$	∞-connectivity

where $R^{(n)}$ — n-th order reflection.

Claim 10.4 (Finiteness of the hierarchy) (requires verification):

Level L4 is maximal. There exist no L5, L6, ...

Proof: Follows from the Postnikov stabilization theorem: for finite-dimensional spaces the Postnikov tower stabilizes. $\tau_{\leq \infty} = \mathrm{Id}$, further truncation is impossible. ∎

---

11. Discrete ∞-groupoid $\mathbf{Exp}^{disc}_\infty$

Status: [Т] Formalized

This section describes the discrete version of the ∞-groupoid for finite-dimensional systems ($N < \infty$), where time is fundamentally discrete.

11.1 Motivation

In the Page–Wootters mechanism for UHM:

• Continuous ∞-groupoid $\mathbf{Exp}_\infty$: paths $\gamma: [0,1] \to \mathcal{E}$ are continuous
• Discrete Page–Wootters time: $\tau \in \mathbb{Z}_7$ for a 7D system

Contradiction: How to reconcile continuous paths with discrete time?

Solution: For finite-dimensional systems use the discrete ∞-groupoid $\mathbf{Exp}^{disc}_\infty$.

11.2 Definition

Definition 11.1 (Discrete ∞-groupoid $\mathbf{Exp}^{disc}_\infty$):

0-cells (objects):

$$\mathrm{Ob}(\mathbf{Exp}^{disc}_\infty) = \mathcal{E} \times \mathbb{Z}_N$$

i.e., pairs (experiential state, discrete time moment).

For $N = 7$: an object is $(\mathcal{Q}, n)$ where $\mathcal{Q} \in \mathcal{E}$, $n \in \{0, 1, 2, 3, 4, 5, 6\}$.

1-morphisms:

$$\mathrm{Mor}_1((\mathcal{Q}_1, n_1), (\mathcal{Q}_2, n_2)) = \begin{cases} \{\Phi : \text{CPTP}, F(\Phi(\rho_1)) = \mathcal{Q}_2\} & \text{if } n_2 = n_1 + 1 \mod N \\ \emptyset & \text{otherwise} \end{cases}$$

Interpretation: Morphisms exist only between consecutive moments of time.

n-morphisms (n ≥ 2):

$$\mathrm{Mor}_n = \text{trivial (identities only)}$$

Justification: Between discrete steps there is no room for homotopies.

11.3 $\mathbb{Z}_N$-structure

Definition 11.2 (Time shift automorphism):

Functor $\sigma: \mathbf{Exp}^{disc}_\infty \to \mathbf{Exp}^{disc}_\infty$:

$$\sigma(\mathcal{Q}, n) = (\mathcal{Q}, n + 1 \mod N)$$

Properties:

• $\sigma^N = \mathrm{Id}$ (cyclicity)
• $\sigma$ commutes with CPTP-morphisms

Theorem 11.1 (Symmetry group): The temporal symmetry group of $\mathbf{Exp}^{disc}_\infty$ is isomorphic to $\mathbb{Z}_N$:

$$\mathrm{Aut}_{temp}(\mathbf{Exp}^{disc}_\infty) \cong \mathbb{Z}_N$$

11.4 Continuous limit

Definition 11.3 (Continuous limit):

As $N \to \infty$ define an embedding functor:

$$\iota_N: \mathbf{Exp}^{disc}_\infty(N) \hookrightarrow \mathbf{Exp}^{disc}_\infty(N')$$

for $N | N'$ (N divides N').

Theorem 11.2 (Consistency):

$$\lim_{N \to \infty} \mathbf{Exp}^{disc}_\infty(N) \simeq \mathbf{Exp}_\infty^{cont}$$

where $\mathbf{Exp}_\infty^{cont}$ is the standard continuous ∞-groupoid of paths (section 10).

Proof (scheme):

1. As $N \to \infty$ the set $\mathbb{Z}_N$ becomes dense in $S^1$
2. Discrete steps approximate continuous paths
3. The limit is defined through profunctors

∎

Interpretation:

• For finite-dimensional systems (N = 7): time is discrete, use $\mathbf{Exp}^{disc}_\infty$
• For macroscopic systems ($N \gg 1$): continuous time is a good approximation
• Discrete time is fundamental, continuous time is emergent

11.5 Proof of ∞-topos (Lurie's theorem)

Definition 11.4 (Topology on $\mathbf{Exp}^{disc}_\infty$):

A family $\{U_i\}$ covers $(\mathcal{Q}, n)$ if:

$$\bigsqcup_i \mathrm{dom}(U_i) \supseteq B_\varepsilon(\mathcal{Q}) \cap \{\text{objects with time } n\}$$

for some $\varepsilon > 0$ in the metric on $\mathcal{E}$.

Definition 11.5 (∞-sheaf on $\mathbf{Exp}^{disc}_\infty$):

A functor $F: (\mathbf{Exp}^{disc}_\infty)^{op} \to \mathbf{Spaces}$ is an ∞-sheaf if for each cover $\{U_i\}$ of an object $X$:

$$F(X) \xrightarrow{\simeq} \lim\left( \prod_i F(U_i) \rightrightarrows \prod_{i,j} F(U_i \cap U_j) \cdots \right)$$

Theorem 11.3 (Existence of ∞-topos):

The category $\mathbf{Sh}_\infty(\mathbf{Exp}^{disc}_\infty)$ is an ∞-topos.

Proof:

Step 1: $\mathbf{Exp}^{disc}_\infty$ is a small ∞-category (finite number of objects when $\mathcal{E}$ and $N$ are fixed).

Step 2: The Grothendieck topology (Definition 11.4) satisfies the axioms:

• Stability under pullback
• Transitivity

Step 3: By Lurie's sheafification theorem (Higher Topos Theory, Theorem 6.2.2.7):

For a small ∞-category $\mathcal{C}$ with Grothendieck topology, the category of ∞-sheaves $\mathbf{Sh}_\infty(\mathcal{C})$ is an ∞-topos (constructed as a left-exact localization of the presheaf ∞-category).

(HTT 6.1.0.6 is the Giraud-style characterization of ∞-topoi; HTT 6.2.2.7 is the constructive statement that sheaves on a site satisfy that characterization.)

∎

11.6 Temporal modalities

Corollary 11.1: $\mathbf{Sh}_\infty(\mathbf{Exp}^{disc}_\infty)$ possesses internal temporal modalities:

Modality	Notation	Definition
"Will be true at the next moment"	$\diamond_+ P$	$\mathrm{Lan}_\sigma(P)$ — left Kan extension along shift
"Was true at the previous moment"	$\diamond_- P$	$\mathrm{Lan}_{\sigma^{-1}}(P)$
"Always true"	$\square P$	$\bigcap_{n \in \mathbb{Z}_N} \sigma^n(P)$

Theorem 11.4 (Temporal modality):

In $\mathbf{Sh}_\infty(\mathbf{Exp}^{disc}_\infty)$ the operators $\diamond_+$, $\diamond_-$, $\square$ form a modal logic of type $\mathbf{S5}$ with discrete time.

Corollary: The logic of experiential content is a temporal modal logic, derivable from categorical structure, not postulated.

---

12. Category of Holons Hol

Status: [Т] Formalized

This section describes the categorical structure of Holons as a subcategory of DensityMat (not full).

12.1 Definition of category Hol

Definition 12.1 (Category Hol).

The category of Holons $\mathbf{Hol}$ is defined as:

Objects:

$$\mathrm{Ob}(\mathbf{Hol}) = \{\Gamma \in \mathrm{Ob}(\mathbf{DensityMat}) : \Gamma \text{ satisfies (AP)+(PH)+(QG)+(V)}\}$$

i.e., density matrices on $\mathcal{H} \cong \mathbb{C}^7 \otimes \mathcal{H}_{\text{int}}$, for which:

• (AP) Autopoiesis: there exists $\varphi$ with a fixed point
• (PH) Phenomenology: $\rho_E \neq 0$
• (QG) Quantum foundation: dynamics with regeneration
• (V) Viability: $P > P_{\text{crit}} = 2/7$

Morphisms:

$$\mathrm{Mor}_{\mathbf{Hol}}(\Gamma_1, \Gamma_2) = \{\Phi \in \mathrm{Mor}_{\mathbf{DM}}(\Gamma_1, \Gamma_2) : \Phi \text{ preserves the Holon structure}\}$$

where "preserves the Holon structure" means:

1. Viability: $P(\Phi(\Gamma)) > P_{\text{crit}}$ if $P(\Gamma) > P_{\text{crit}}$
2. Autopoiesis: $\varphi_2 \circ \Phi = \Phi \circ \varphi_1$ (commutation with self-modeling)

12.2 Theorem on subcategory

Theorem 12.1 (Categorical structure of Holons).

$\mathbf{Hol}$ is a subcategory of $\mathbf{DensityMat}$ (not full: morphisms must preserve viability and autopoiesis):

$$\mathbf{Hol} \hookrightarrow \mathbf{DensityMat}$$

Proof:

1. Inclusion of objects: By definition, $\mathbb{H}$ is a special case of $\Gamma \in \mathcal{D}(\mathbb{C}^7 \otimes \mathcal{H}_{\text{int}})$.

2. Inheritance of morphisms: A morphism $\Phi: \mathbb{H}_1 \to \mathbb{H}_2$ in $\mathbf{Hol}$ is a CPTP channel from $\mathbf{DensityMat}$, additionally preserving:

   • Autonomy (conditions A1-A3)
   • Viability ($P > P_{\text{crit}}$)
   • Autopoiesis (commutation with $\varphi$)

3. Not full: Not all CPTP-morphisms between Holons in $\mathbf{DensityMat}$ are included in $\mathbf{Hol}$ — only those that preserve viability and autopoiesis.

∎

12.3 Interiority functor

Theorem 12.2 (Interiority functor).

There exists a functor

$$\mathcal{I}: \mathbf{Hol} \to \mathbf{Exp}$$

mapping each Holon to its experiential content.

Definition of the functor:

On objects:

$$\mathcal{I}(\mathbb{H}) := F(\Gamma_{\mathbb{H}}) = (\mathrm{Spec}(\rho_E), [|q_i\rangle], \Gamma_{-E}, h)$$

where $F$ is the functor from section 3.

On morphisms:

$$\mathcal{I}(\Phi) := F(\Phi|_E)$$

Proof of functoriality:

1. $\mathcal{I}(\mathrm{id}_\mathbb{H}) = F(\mathrm{id}_{\Gamma}) = \mathrm{id}_{F(\Gamma)} = \mathrm{id}_{\mathcal{I}(\mathbb{H})}$ — follows from functoriality of $F$

2. $\mathcal{I}(\Psi \circ \Phi) = F((\Psi \circ \Phi)|_E) = F(\Psi|_E) \circ F(\Phi|_E) = \mathcal{I}(\Psi) \circ \mathcal{I}(\Phi)$ — follows from functoriality of $F$

∎

12.4 Categorical diagram with Hol

                    inclusion                  F
        Hol ─────────────────────► DensityMat ────────► Exp
         │                              │                │
         │ morphisms                    │ CPTP           │ induced
         │ (structure-preserving)       │                │
         ▼                              ▼                ▼
        Hol ─────────────────────► DensityMat ────────► Exp
                                          │
                                          │ ℐ = F ∘ inclusion
                                          ▼
                                         Exp

Commutativity:

$$\mathcal{I} = F \circ \iota$$

where $\iota: \mathbf{Hol} \hookrightarrow \mathbf{DensityMat}$ — inclusion.

12.5 Properties of category Hol

Property	Status	Comment
Subcategory (not full)	✓	Theorem 12.1
Closed under composition	✓	CPTP ∘ CPTP = CPTP
Terminal object	[С]	Pure state $P = 1$, but not unique
Initial object	—	No (set of states with $P = P_{\text{crit}} + \varepsilon$)
Products	[С]	Tensor product, but $\dim > 7$
Topos	✗	Is not one (as with $\mathbf{Exp}$)

---

13. Derived categories and IC-cohomologies

Status: [Т] Formalized

This section describes derived categories and IC-cohomologies for capturing the "hidden topology" of the stratified base space X.

13.1 Stratified base space

From Axiom Ω⁷ the base space:

$$X := |N(\mathcal{C})|$$

is stratified:

$$X = \bigsqcup_{\alpha \in A} S_\alpha$$

where:

• $S_0 = \{T\}$ — terminal object (0-dimensional)
• $S_1$ — edges (morphisms to T)
• $S_n$ — n-simplices

13.2 Local-global dichotomy

Theorem 13.1 (Cohomological monism):

$$H^n(X, \mathcal{F}) = 0 \quad \forall n > 0$$

Proof: X is contractible to the terminal object T.

Theorem 13.2 (Nontrivial local cohomologies):

$$H^*_{loc}(X, T) \cong \tilde{H}^{*-1}(\text{Link}(T)) \cong \tilde{H}^{*-1}(S^6) \neq 0$$

Interpretation:

• Globally: H*(X) = 0 — monism
• Locally: H*_loc ≠ 0 — physics (topological effects)

13.3 Derived category of sheaves

Definition 13.1 (Derived category):

$$D^b(X) := D^b(\mathbf{Sh}(X))$$

— bounded derived category of sheaves on X.

Advantage: D^b(X) captures information lost in passing to ordinary cohomologies.

13.4 Perverse sheaves

Definition 13.2 (Perverse sheaves):

On stratified X define the category:

$$\mathbf{Perv}(X) \subset D^b(X)$$

— perverse sheaves satisfying support and co-support conditions.

Theorem 13.3 (Beilinson–Bernstein–Deligne decomposition):

$$D^b(X) = \langle \mathbf{Perv}_1, \mathbf{Perv}_2, \ldots \rangle$$

(semi-orthogonal decomposition)

13.5 IC-cohomologies

Definition 13.3 (IC-sheaf):

For a stratum $S_\alpha$ the intersection cohomology sheaf:

$$IC(S_\alpha) \in \mathbf{Perv}(X)$$

Theorem 13.4 (Hidden topology):

$$\bigoplus_\alpha H^*(X, IC(S_\alpha)) \neq 0$$

even when $H^*(X) = 0$.

Interpretation: "Hidden topology" is stored in the IC-cohomologies of the strata.

13.6 Connection to physics

IC-cohomologies	Physics
$IC(S_0)$	Vacuum state
$IC(S_n)$	Excitations above the vacuum
$H^*(X, IC)$	Topological charges

13.7 ∞-topos of Holons

Definition 13.4 (∞-topos of Holons):

$$\mathcal{T}_H := \mathbf{Sh}_\infty(\mathcal{G}_h, \tau_{\acute{e}t})$$

∞-category of ∞-sheaves on the category of Holons with étale topology.

Theorem 13.5 (Internal logic):

The internal logic of $\mathcal{T}_H$ is homotopy type theory (HoTT) with:

1. Types: Objects Γ (states)
2. Terms: Morphisms φ (operators)
3. Identity: Paths in the state space
4. Subobject classifier: ∞-groupoid of truth values

---

14. ∞-topos as the true primitive

Status: [Т] Formalized

This section demonstrates that the ∞-topos is the true primitive of UHM, replacing 5 separate axioms with a single structure.

14.1 Evolution of the primitive

In the course of the theory's development, there is a sequential abstraction of the primitive object:

Axioms	Primitive	Structure	Interpretation
Ω¹–Ω³	State Γ	Density matrix $\rho \in \mathcal{D}(\mathcal{H})$	Quantum state of the system
Ω⁴–Ω⁵	Category $\mathcal{C}$	$(\mathrm{Ob}, \mathrm{Mor}, \circ, \mathrm{id})$	State space with morphisms
Ω⁷	∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$	$\mathbf{Fun}(\mathcal{C}^{op}, \mathbf{Spaces})^{loc}$	Complete ∞-structure with internal logic

Observation: Each successive level contains the previous ones:

• Γ — object in $\mathcal{C}$
• $\mathcal{C}$ — base for $\mathbf{Sh}_\infty(\mathcal{C})$
• $\mathbf{Sh}_\infty(\mathcal{C})$ — self-sufficient structure

14.2 Definition of the UHM ∞-topos

Definition 14.1 (UHM ∞-topos):

$$\mathbf{Sh}_\infty(\mathcal{C}) := \mathbf{Fun}(\mathcal{C}^{op}, \mathbf{Spaces})^{loc}$$

where:

• $\mathcal{C}$ — category of Holons from Axiom Ω⁷
• $\mathbf{Spaces}$ — ∞-category of spaces (∞-groupoids)
• $\mathcal{C}^{op}$ — opposite category
• $\mathbf{Fun}(-, -)$ — ∞-category of functors
• $(-)^{loc}$ — localization by covers (sheafification)

Remark 14.1. This definition generalizes classical Grothendieck toposes to the ∞-level in the sense of Lurie.

14.3 Lurie's theorem on the structure of the ∞-topos

Theorem 14.1 (Lurie, HTT 6.1.0.6):

The ∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$ possesses the following structure:

1. Internal logic: Homotopy type theory (HoTT)

   • Types = objects (∞-sheaves)
   • Terms = sections
   • Type identity = paths in space

2. Subobject classifier: There exists an object $\Omega$ such that

   $$\mathrm{Sub}(X) \simeq \mathrm{Map}(X, \Omega)$$

   In the ∞-topos $\Omega$ is an ∞-groupoid of truth values.

3. All limits and colimits: $\mathbf{Sh}_\infty(\mathcal{C})$ is complete and cocomplete:

   $$\lim, \mathrm{colim}: \mathbf{Sh}_\infty(\mathcal{C})^I \to \mathbf{Sh}_\infty(\mathcal{C})$$

4. Exponentials (internal Hom): For any $X, Y$ there exists $Y^X$:

   $$\mathrm{Map}(Z \times X, Y) \simeq \mathrm{Map}(Z, Y^X)$$

Corollary 14.1: All constructions of UHM are expressible in the internal language of $\mathbf{Sh}_\infty(\mathcal{C})$.

14.4 Formalization of free will

The ∞-topos structure allows formalizing free will.

Definition 14.2 (Freedom: ∞-categorical motivation):

For a state Γ ∈ $\mathrm{Ob}(\mathcal{C})$ the ∞-categorical definition:

$$\mathrm{Freedom}(\Gamma) := \pi_0\left(\mathrm{Map}(\Gamma, T)^{\text{non-trivial}}\right)$$

where:

• $\mathrm{Map}(\Gamma, T)$ — space of morphisms to the terminal object
• $\pi_0$ — set of connected components
• "non-trivial" — exclusion of zero/trivial paths

Finite-dimensional definition [Т]: For $\Gamma \in \mathcal{D}(\mathbb{C}^7)$:

$$\mathrm{Freedom}(\Gamma) := \dim\ker(\mathcal{H}_\Gamma) + 1$$

where $\mathcal{H}_\Gamma = \partial^2 \mathcal{F}[\varphi; \Gamma]/\partial\Gamma^2$ — the Hessian of the free-energy functional. Each zero mode is an independent choice (direction without energy penalty). Monotone under CPTP, $G_2$-invariant. Freedom(I/7) = 7, Freedom(ρ*) = 1. See Consequences of axioms.

Definition 14.3 (Freedom entropy):

$$S_{\text{freedom}} := \log(\mathrm{Freedom}(\Gamma)) = \log(\dim\ker(\mathcal{H}_\Gamma) + 1)$$

Theorem 14.2 (Compatibility of uniqueness and freedom):

In the ∞-category $\mathcal{C}$ the following hold simultaneously:

1. Uniqueness (homotopic): $\mathrm{Map}(\Gamma, T) \simeq *$ (contractible)
2. Freedom (geometric): $\mathrm{Map}(\Gamma, T) \neq \{*\}$ (contains nontrivial paths)

Proof: Contractibility means that all paths are homotopically equivalent, but does not mean the path is unique. The space $\mathrm{Map}(\Gamma, T)$ may be infinite-dimensional while being contractible. ∎

14.5 Why the ∞-topos is the true primitive

Theorem 14.3 (∞-topos as the true primitive of UHM):

The ∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$ is the true primitive of the theory by three criteria:

14.5.1 Completeness

Claim: $\mathbf{Sh}_\infty(\mathcal{C})$ contains all the structure of UHM:

UHM component	Representation in ∞-topos
State Γ	Object (∞-sheaf)
Morphism φ	Morphism of ∞-sheaves
Composite system $\Gamma_{AB}$	$\iota(\Gamma_A) \otimes_{\text{Day}} \iota(\Gamma_B)$ (Day convolution, not Cartesian product $\times$)
Entanglement	Indecomposability with respect to $\otimes_{\text{Day}}$ (Day 1970, Lurie HA §3.2)
Time τ	1-morphism in $\mathrm{Map}(\Gamma, T)$
History h	2-morphism (homotopy between paths)
Evolution	Functor $\mathbf{Sh}_\infty(\mathcal{C}) \to \mathbf{Sh}_\infty(\mathcal{C})$
Freedom	$\dim\ker(\mathcal{H}_\Gamma) + 1$ [Т]; ∞-categorically: $\pi_*(\mathrm{Map}(\Gamma, T))$

Fundamental distinction: ⊗_Day ≠ ×_T

The tensor product of quantum states $\otimes$ is not the Cartesian product $\times$ in the topos (Abramsky-Coecke theorem: CPTP category is non-Cartesian monoidal). Cartesian $\times$ = separable states. Quantum entanglement is encoded via Day convolution $\otimes_{\text{Day}}$: a non-Cartesian monoidal structure on $\mathbf{Sh}_\infty(\mathcal{C})$, canonically lifting $\otimes$ from the base category $\mathcal{C}$ into the sheaf category. Bell's theorem and quantum teleportation are correctly described via $\otimes_{\text{Day}}$.

14.5.2 Minimality

Claim: One structure instead of 5 axioms.

Was (Ω¹–Ω⁵)	Became (Ω⁷)
5 separate axioms	1 primitive
Connections are postulated	Connections are derived
Ad hoc constructions	Universal properties

Principle: All axioms Ω¹–Ω⁵ are derived from $\mathbf{Sh}_\infty(\mathcal{C})$:

• Ω¹ (state): objects in the base $\mathcal{C}$
• Ω² (operator): morphisms in $\mathcal{C}$
• Ω³ (viability): subobjects via $\Omega$
• Ω⁴ (terminal object): terminal object in $\mathbf{Sh}_\infty$
• Ω⁵ (categorical structure): $\mathcal{C}$ itself as the base

14.5.3 Resolving power

Claim: The ∞-topos resolves the paradox of teleological determinism.

Paradox: From the existence of a terminal object T with a unique morphism $\Gamma \to T$ follows rigid determinism — absence of freedom of choice.

Resolution in the ∞-topos:

$$\text{unique} \xrightarrow{\infty\text{-interpretation}} \text{contractible path space}$$

Formally:

• In a 1-category: $|\mathrm{Hom}(\Gamma, T)| = 1$
• In an ∞-category: $\mathrm{Map}(\Gamma, T) \simeq *$, but $\dim(\mathrm{Map}(\Gamma, T)) = \infty$

Corollary: Determinism of the goal (all paths lead to T) is compatible with freedom of means (infinite set of paths).

---

15. L-unification

Status: [Т] Formalized

This section establishes the key theorem on the identity of the dimension L, the subobject classifier Ω, and the source of Lindblad operators L_k.

15.1 Central theorem

Theorem 15.1 (L-unification):

$$L \cong \Omega \cong \text{source}(L_k)$$

The subobject classifier Ω in the ∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$ is the unified source of three fundamental structures of UHM:

1. Dimensions L — as projection of Ω onto state Γ
2. Lindblad operators L_k — as atomic subobjects of Ω
3. Emergent time — via temporal modality ▷

15.2 Ω as unified source

15.2.1 L as L = Ω ∩ Γ

The Logic dimension is categorically identical to the projection of the classifier onto the state:

$$L := \{\chi \in \Omega : \chi(\Gamma) = \text{true}\}$$

Interpretation: L is the set of logical predicates that are true for the given configuration Γ. This is not a separate axiom, but a consequence of the existence of Ω in the ∞-topos.

15.2.2 Lindblad operators as L_k = √χ_S

Resolution of the logic conflict (Heyting vs. orthomodular) {#logic-conflict}

In any topos (including ∞-topoi) the subobject classifier Ω has the structure of a Heyting algebra (intuitionistic logic). Quantum projectors on ℂ⁷ form a non-distributive orthomodular lattice (non-commutative quantum logic, Kochen-Specker theorem). These logics are incompatible in full generality.

Resolution: Operators $L_k$ are taken not from full Ω, but from the decidable fragment:

$$\mathrm{Dec}(\Omega) := \{p \in \Omega : p \vee \neg p = \top\} \cong 2^7 \quad \text{(Boolean algebra)}$$

The Boolean subalgebra $\mathrm{Dec}(\Omega)$ is the common fragment of both logics:

• In Ω: complemented elements of the Heyting algebra
• In Proj(ℂ⁷): commuting projectors = pointer basis

Why Dec(Ω) ≅ 2⁷, not an arbitrary Boolean subalgebra:

1. $G_2$-rigidity (T-42a [T]) fixes the basis {|A⟩,...,|U⟩} uniquely (up to $G_2$-rotation)
2. Einselection (T-164 [T]) selects the pointer basis — fixed points of decoherence $\mathcal{D}_\Omega$
3. Atoms of Dec(Ω) = {|k⟩⟨k|} — minimal projectors in the pointer basis

This is not postulating a privileged basis, but its derivation from $G_2$-rigidity + einselection. The "classicality" of the dissipative core is decoherence (standard physics, Zurek 2003), formalized through Dec(Ω).

Connection to Isham–Butterfield topos quantum mechanics. The topological approach to quantum mechanics (Isham–Butterfield 1998–2004, Döring–Isham 2008) constructs the topos of presheaves over the poset $\mathcal{V}(\mathcal{N})$ of commutative subalgebras of a von Neumann algebra $\mathcal{N}$. In this framework, quantum propositions are represented by clopen subobjects of the spectral presheaf — exactly the decidable elements of the classifier. The UHM construction Dec(Ω) ≅ 2⁷ is the finite-dimensional analogue: the maximal commutative subalgebra is the pointer basis {|k⟩⟨k|}, and the decidable fragment Dec(Ω) corresponds to Isham–Butterfield's clopen subobjects restricted to this basis. The key difference: Isham–Butterfield work with all commutative subalgebras simultaneously (the presheaf topos), while UHM selects one via $G_2$-rigidity and einselection. This selection is not ad hoc but categorically forced (T-42a [T], T-164 [T]).

Resolution of the circularity L_k ↔ Dec(Ω). The derivation order is not circular:

1. $G_2 = \mathrm{Aut}(\mathbb{O})$ — defined algebraically (Cartan's theorem), outside dynamics [T]
2. Fano plane $\mathrm{PG}(2,2) \subset \mathrm{Im}(\mathbb{O})$ — discrete combinatorial structure fixed by the structure constants of octonions [T]
3. $L_k = \sqrt{\chi_{S_k}}$ — Lindblad operators = projectors onto 7 Fano lines (T-82 [T]: uniqueness)
4. $\mathrm{Dec}(\Omega) = \{\chi_{S_k}\}_{k=1}^7$ — follows from steps 1-3, does not define them
5. Einselection (T-164 [T]) — confirms (does not define) that $\{|k\rangle\}$ is the pointer basis

$G_2$ is a continuous (14-dimensional) group, but it fixes the combinatorics of the Fano plane (7 lines, 7 points), not a specific basis. $\mathrm{Dec}(\Omega) \cong 2^7$ is determined by Fano combinatorics, not by basis choice. $G_2$-rotation renames vertices but preserves the line structure.

The dissipation operators in the evolution equation are defined by the atoms of the classifier:

$$L_k := \sqrt{\chi_{S_k}}$$

where $S_k$ is the k-th minimal subobject (atom) of Ω.

Theorem 15.2 (CPTP automatically):

$$\sum_k L_k^\dagger L_k = \sum_k \chi_{S_k} = \mathbb{1}$$

The trace-preservation condition is not postulated — it is derived from properties of the classifier.

15.2.3 Time via temporal modality ▷

The temporal modality $\triangleright$ ("at the next moment") is defined on Ω, generating emergent time:

$$\tau_n := \triangleright^n(\text{now})$$

Connection with internal logic:

$$\triangleright: \Omega \to \Omega, \quad \triangleright(\chi) = \chi\text{ is true at the next moment}$$

The evolution of predicates χ ∈ L under ▷ is the dynamics of the system. See internal logic of Ω.

15.3 Adjunction $\mathcal{D}_\Omega \dashv \mathcal{R}$ and derivation of κ₀

Key theorem

The regeneration rate $\kappa_0$ is categorically derived from the adjunction of dissipation and regeneration functors. This transforms a phenomenological parameter into a structural quantity.

15.3.1 Explicit construction of the adjunction

Definition of functors:

Dissipation functor $\mathcal{D}_\Omega: \mathbf{Sh}_\infty(\mathcal{C}) \to \mathbf{Set}$:

$$\mathcal{D}_\Omega(\Gamma) := \text{Hom}_{\mathbf{Sh}_\infty}(\Gamma, \Omega) = \{\chi: \Gamma \to \Omega\}$$

This is the set of all predicates (truth values) on state Γ.

Regeneration functor $\mathcal{R}: \mathbf{Set} \to \mathbf{Sh}_\infty(\mathcal{C})$:

$$\mathcal{R}(S) := \text{Free}_\Omega(S) = \bigoplus_{s \in S} \Omega_s$$

where $\Omega_s$ is a copy of the classifier indexed by element s ∈ S.

Theorem 15.3 (Existence of the adjunction):

There exists an adjunction:

$$\mathcal{D}_\Omega \dashv \mathcal{R}: \mathbf{Sh}_\infty(\mathcal{C}) \rightleftarrows \mathbf{Set}$$

with natural isomorphism:

$$\text{Hom}_{\mathbf{Set}}(\mathcal{D}_\Omega(\Gamma), S) \cong \text{Hom}_{\mathbf{Sh}_\infty}(\Gamma, \mathcal{R}(S))$$

Proof:

(a) For any set $S$ and sheaf $\Gamma$, a morphism $f: \mathcal{D}_\Omega(\Gamma) \to S$ defines a map of predicates to the set.

(b) Each such f induces a morphism $\tilde{f}: \Gamma \to \bigoplus_{s \in S} \Omega_s$ via the universal property of the coproduct.

(c) Conversely, a morphism $g: \Gamma \to \mathcal{R}(S)$ projects onto each component $\Omega_s$, giving a map $\bar{g}: \mathcal{D}_\Omega(\Gamma) \to S$.

(d) Naturality follows from functoriality of the constructions.

(e) Triangle identities [Т]:

For completeness of the adjunction one must verify two triangle (zig-zag) identities:

Identity 1: $(\varepsilon_{\mathcal{R}(S)}) \circ (\mathcal{R}(\eta_S)) = \mathrm{id}_{\mathcal{R}(S)}$ for all $S \in \mathbf{Set}$.

Proof of Identity 1. Let $S \in \mathbf{Set}$. Then $\mathcal{R}(S) = \bigoplus_{s \in S} \Omega_s$ is the free $\Omega$-sheaf on $S$. The unit $\eta_{\mathcal{R}(S)}: \mathcal{R}(S) \to \mathcal{R}(\mathcal{D}_\Omega(\mathcal{R}(S)))$ embeds $\bigoplus_s \Omega_s$ into $\bigoplus_{\chi \in \mathrm{Hom}(\mathcal{R}(S), \Omega)} \Omega_\chi$. Each generator $\Omega_s$ of the free sheaf is mapped to the component $\Omega_{\pi_s}$ where $\pi_s: \mathcal{R}(S) \to \Omega$ is the projection onto the $s$-th summand. The counit $\varepsilon_S: \mathcal{D}_\Omega(\mathcal{R}(S)) \to S$ maps each predicate $\chi$ to the element $s \in S$ indexing the summand that $\chi$ projects onto. The composition $\varepsilon_{\mathcal{R}(S)} \circ \mathcal{R}(\eta_S)$ sends $\Omega_s \xrightarrow{\eta} \Omega_{\pi_s} \xrightarrow{\varepsilon} \Omega_s$, which is the identity on each generator. By universality of the coproduct, the composition is $\mathrm{id}_{\mathcal{R}(S)}$. $\checkmark$

Identity 2: $(\mathcal{D}_\Omega(\varepsilon_\Gamma)) \circ (\eta_{\mathcal{D}_\Omega(\Gamma)}) = \mathrm{id}_{\mathcal{D}_\Omega(\Gamma)}$ for all $\Gamma \in \mathbf{Sh}_\infty(\mathcal{C})$.

Proof of Identity 2. Let $\Gamma \in \mathbf{Sh}_\infty(\mathcal{C})$ and $T = \mathcal{D}_\Omega(\Gamma) = \mathrm{Hom}(\Gamma, \Omega)$. The unit $\eta_T: T \to \mathcal{D}_\Omega(\mathcal{R}(T)) = \mathrm{Hom}(\bigoplus_{\chi \in T} \Omega_\chi, \Omega)$ maps each predicate $\chi \in T$ to the evaluation morphism $\mathrm{ev}_\chi: \bigoplus_{\chi'} \Omega_{\chi'} \to \Omega$ that projects onto the $\chi$-th component. The counit $\varepsilon_\Gamma: \Gamma \to \mathcal{R}(T)$ is the canonical embedding (from step (b)). The functor $\mathcal{D}_\Omega$ applied to $\varepsilon_\Gamma$ gives $\mathcal{D}_\Omega(\varepsilon_\Gamma): \mathrm{Hom}(\mathcal{R}(T), \Omega) \to \mathrm{Hom}(\Gamma, \Omega)$ by precomposition: $f \mapsto f \circ \varepsilon_\Gamma$. The composition $\mathcal{D}_\Omega(\varepsilon_\Gamma) \circ \eta_T$ sends $\chi \mapsto \mathrm{ev}_\chi \circ \varepsilon_\Gamma = \chi$ (since $\varepsilon_\Gamma$ embeds $\Gamma$ into $\bigoplus \Omega_\chi$ and $\mathrm{ev}_\chi$ extracts the $\chi$-component, recovering the original predicate). Therefore the composition is $\mathrm{id}_T$. $\checkmark$

Both triangle identities hold, completing the proof that $\mathcal{D}_\Omega \dashv \mathcal{R}$ is a valid adjunction. $\blacksquare$

15.3.2 Unit and counit of the adjunction

Unit of the adjunction $\eta: \text{Id} \Rightarrow \mathcal{R} \circ \mathcal{D}_\Omega$:

$$\eta_\Gamma: \Gamma \to \mathcal{R}(\mathcal{D}_\Omega(\Gamma)) = \bigoplus_{\chi \in \text{Hom}(\Gamma, \Omega)} \Omega_\chi$$

This is the canonical embedding of the state into the space of all its predicates.

Counit of the adjunction $\varepsilon: \mathcal{D}_\Omega \circ \mathcal{R} \Rightarrow \text{Id}$:

$$\varepsilon_S: \mathcal{D}_\Omega(\mathcal{R}(S)) \to S$$

This is the projection of the free sheaf onto the generating set.

15.3.3 Derivation of κ₀ and κ_bootstrap

Theorem 15.3.1 (Categorical derivation of κ₀):

The regeneration rate $\kappa_0$ is derived as the norm of the unit of the adjunction:

$$\kappa_0 = \|\eta\|_{\text{op}} \cdot \omega_0 = \omega_0 \cdot \frac{|\gamma_{OE}| \cdot |\gamma_{OU}|}{\gamma_{OO}}$$

where:

• $\|\eta\|_{\text{op}}$ — operator norm of the unit η on a concrete state Γ
• $\omega_0$ — characteristic frequency of the system (parameter, analogous to mass in physics)
• $\gamma_{ij}$ — elements of the coherence matrix

Dimensionality: $[\kappa_0] = [\text{time}]^{-1}$.

Theorem 15.3.2 (Minimal regeneration κ_bootstrap):

Definition of κ_bootstrap

Minimal regeneration rate required for viability:

$$\kappa_{\text{bootstrap}} := \inf_{\Gamma: P(\Gamma) > P_{\text{crit}}} \kappa_0(\Gamma) > 0$$

Proof of positivity:

(a) The unit of the adjunction $\eta \neq 0$ for any nontrivial Γ (otherwise $\mathcal{D}_\Omega(\Gamma) = \varnothing$, which is impossible for a nonzero sheaf).

(b) Compactness of the set $\{Γ: P(Γ) = P_{\text{crit}} + \varepsilon\}$ for small ε > 0 guarantees the infimum is achieved.

(c) At the viability boundary $\kappa_0 > 0$ (otherwise the system cannot maintain $P > P_{\text{crit}}$, see theorem on critical purity).

∎

Physical interpretation:

Quantity	Meaning	Source
$\kappa_0(\Gamma)$	Regeneration rate for state Γ	Norm of η on Γ
$\kappa_{\text{bootstrap}}$	Minimal regeneration for viability	Infimum over admissible Γ
$\omega_0$	Characteristic frequency of the system (parameter, not a universal constant)	Primitive $\mathfrak{T}$

Note: κ₀ depends on state Γ through coherences $\gamma_{OE}, \gamma_{OU}, \gamma_{OO}$. See master definition.

Theorem 15.3.1 (CPTP structure of regeneration):

The regenerative operator $\mathcal{R}_\alpha: \mathcal{D}(\mathcal{H}) \to \mathcal{D}(\mathcal{H})$ of the form:

$$\mathcal{R}_\alpha(\rho) := (1-\alpha)\rho + \alpha\varphi(\rho)$$

is a CPTP channel for $\alpha \in [0,1]$ and CPTP property of $\varphi$.

Corollary: The nonlinearity of the regenerative term does not violate the positivity of the density matrix. The full evolution equation is valid for $\alpha = \kappa \cdot \Delta\tau < 1$.

See positivity preservation for the complete proof.

15.4 Resolution of formalization gaps

L-unification closes the following open questions:

Gap	Solution	Reference
Origin of L_k	Atoms of classifier Ω	§15.2.2
Why 7 dimensions?	Minimal base for Ω ∩ Γ ≠ ∅	Theorem 7.1
Source of CPTP	Completeness of Ω	§15.2.2
Emergence of τ	Modality ▷ on Ω	§15.2.3
Derivation of κ₀	Unit of adjunction $\mathcal{D}_\Omega \dashv \mathcal{R}$	§15.3
Internal logic	Ω-types in HoTT	Axiom Ω⁷
Nonlinearity and positivity	CPTP-structure of $\mathcal{R}_\alpha$	§15.3.1

15.5 Commutative unification diagram

Corollary 15.1 (Unification):

All dynamic structures of UHM (dimension L, operators L_k, time τ, constant κ₀) are derived from the single primitive — the subobject classifier Ω in the ∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$.

This completes the categorical formalization program: the 5 axioms Ω¹–Ω⁵ are reduced to properties of Ω within the framework of Ω⁷.

---

Conclusion

Summary of results

1. Category $\mathbf{Exp}$ formalized with morphisms induced by CPTP channels
2. Functor F defined on morphisms via component-wise transformations
3. Functoriality proved (theorems 5.1-5.3). Strict functoriality — for the base functor (without history); full functoriality requires the lax 2-functor construction (§5.2)
4. $\mathbf{Exp}$ is not a topos, but possesses rich structure (fibration, enrichment, monoidality)
5. ∞-groupoid Exp_∞ proved [Т] — $\mathrm{Sing}(\mathcal{E})$ is a Kan complex (Milnor's theorem); time as 1-morphism, history as loop space (section 10)
6. ∞-topos Sh_∞(Exp) exists — internal temporal modal logic
7. Phenomenal completeness — the structure is sufficient to describe any physically realizable experience (section 8)
8. Quasi-functor for AI — extension to nonlinear systems via NTK linearization (section 9, [П] program)
9. Discrete ∞-groupoid $\mathbf{Exp}^{disc}_\infty$ — reconciliation of discrete Page–Wootters time with the categorical structure (section 11)
10. Category of Holons $\mathbf{Hol}$ — subcategory of $\mathbf{DensityMat}$ (not full), interiority functor $\mathcal{I}: \mathbf{Hol} \to \mathbf{Exp}$ (section 12)
11. Derived categories and IC-cohomologies — capture of hidden topology of stratified X (section 13)
12. Cohomological monism — H*(X) = 0 globally, H*_loc ≠ 0 locally (section 13)
13. ∞-topos of Holons $\mathcal{T}_H$ — internal logic HoTT (section 13)
14. ∞-topos as the true primitive — completeness, minimality, resolution of teleological determinism (section 14)
15. L-unification — L ≅ Ω ≅ source(L_k), derivation of κ₀ from adjunction $\mathcal{D}_\Omega \dashv \mathcal{R}$ (section 15)

Resolved questions

Question	Solution
Cohomologies of $\mathbf{Exp}$	H*(X) = 0 globally (monism), H*_loc ≠ 0 (physics)
Hidden topology	IC-cohomologies of strata
Arrow of time	Collapse of strata to T
Teleological determinism	∞-topos: contractibility ≠ uniqueness of path (section 14)
Origin of L_k	Atoms of classifier Ω: $L_k = \sqrt{\chi_{S_k}}$ (section 15)
Derivation of κ₀	Unit of adjunction $\mathcal{D}_\Omega \dashv \mathcal{R}$ (section 15)
Unification of L/Ω/L_k	L ≅ Ω ≅ source(L_k) — unified primitive (section 15)

Connection to UHM

This formalism completes the categorical part of UHM:

Aspect	Solution
Morphisms of $\mathbf{Exp}$	Definition 2.5, 2.6
$F$ on morphisms	Definition 4.1
Functoriality	Theorems 5.1-5.3
Topos structure	Theorem 6.1 (not a topos), 6.2-6.3 (alternatives)
Phenomenal completeness	Section 8 — the structure describes any physically realizable experience
Category of Holons	$\mathbf{Hol} \hookrightarrow \mathbf{DensityMat}$ (section 12)
Interiority functor	$\mathcal{I}: \mathbf{Hol} \to \mathbf{Exp}$ (theorem 12.2)

---

Non-associative categorical structure

Octonionic categorical perspective [И]

The structural derivation N=7 through octonions suggests a non-associative algebraic structure on the space of dimensions. Categorical formalization of non-associativity uses:

• $A_\infty$-algebras: Generalization of associative algebras, where associativity holds only up to homotopy. The structure $m_n: A^{\otimes n} \to A$ defines a hierarchy of higher operations.
• Associahedra (Stasheff polytopes): Combinatorial spaces parameterizing ways of bracketing. For $n$ elements the associahedron $K_n$ has dimension $n-2$.
• $G_2$-categories: Categories enriched over $G_2$-representations formalize $G_2$-covariance.

Connection to the UHM ∞-topos [С]: The non-associativity of $\mathbb{O}$ may manifest as a nontrivial $A_\infty$-structure on the morphisms of the ∞-topos $\mathrm{Sh}_\infty(\mathcal{C})$. Bridge [Т] (closed, T15).

---

Categorical formalization of the no-signaling prohibition

Connection to the theory

This section formalizes the compatibility of the nonlinear regenerative term $\mathcal{R}$ with the no-signaling principle in the language of category theory. Detailed analysis and complete proofs: Physical correspondence — No-signaling.

Category of autonomous holons $\mathbf{AutHol}$

Definition (Monoidal category $\mathbf{AutHol}$).

• Objects: $(A, \Gamma_A, \varphi_A, \kappa_A)$ — autonomous subsystems satisfying autonomy conditions (A1)+(A2)+(A3)
• Morphisms: CPTP channels preserving autonomy
• Monoidal structure: $\otimes$ (tensor product of Hilbert spaces)
• Unit: trivial holon $(\mathbb{C}, 1, \mathrm{id}, 0)$

UHM evolution functor

Definition. Evolution functor:

$$\mathcal{E}_\tau^{(\text{UHM})}: \mathbf{AutHol} \to \mathbf{AutHol}$$ $$\mathcal{E}_\tau^{(\text{UHM})}(A, \Gamma_A) := (A, \Gamma_A(\tau))$$

where $\Gamma_A(\tau)$ is determined by the full evolution equation (including $\mathcal{R}$).

Theorem: no-signaling as a natural transformation

Theorem (No-signaling as natural transformation)

The partial trace:

$$\mathrm{Tr}_A: \mathbf{AutHol}(A \otimes B) \to \mathbf{AutHol}(B)$$

is a natural transformation from the composite evolution functor to the local one:

$$\mathrm{Tr}_A \circ \mathcal{E}_\tau^{(\text{UHM}), A \otimes B} = \mathcal{E}_\tau^{(\text{UHM}), B} \circ \mathrm{Tr}_A$$

Proof (scheme). Commutative diagram:

$$\begin{CD} \mathbf{AutHol}(A \otimes B) @>{\mathcal{E}_\tau^{A \otimes B}}>> \mathbf{AutHol}(A \otimes B) \\ @V{\mathrm{Tr}_A}VV @VV{\mathrm{Tr}_A}V \\ \mathbf{AutHol}(B) @>{\mathcal{E}_\tau^{B}}>> \mathbf{AutHol}(B) \end{CD}$$

For each $\Gamma_{AB}$:

$$\mathrm{Tr}_A[\mathcal{E}_\tau^{A \otimes B}(\Gamma_{AB})] = \Gamma_B + d\tau \cdot \left(\mathrm{Tr}_A[\mathcal{L}_{lin}] + \underbrace{\mathrm{Tr}_A[\tilde{\mathcal{R}}_A]}_{= 0} + \underbrace{\mathrm{Tr}_A[\tilde{\mathcal{R}}_B]}_{= \mathcal{R}_B[\Gamma_B]}\right) = \mathcal{E}_\tau^B(\Gamma_B) = \mathcal{E}_\tau^B(\mathrm{Tr}_A[\Gamma_{AB}])$$

Annihilation $\mathrm{Tr}_A[\tilde{\mathcal{R}}_A] = 0$ follows from the CPTP property of $\varphi_A$ (condition NS3). $\blacksquare$

Theorem: tensor factorization of self-modeling

Theorem (Tensor factorization of φ)

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

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

i.e., the self-modeling of the composite system factorizes over the autonomous components.

Proof:

1. By definition of autonomy (A1): $\mathcal{I}(A:B|\partial A) = 0$ — conditional independence.
2. For autonomous subsystems: $\mathrm{Sub}(\Gamma_{AB}) \cong \mathrm{Sub}(\Gamma_A) \times \mathrm{Sub}(\Gamma_B)$ (categorical product of subobject lattices).
3. The operator $\varphi$ as left adjoint to the product of inclusions is the product of left adjoints:

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

Connection to ∞-topos

In the ∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$ the no-signaling prohibition is formalized via the sheaf condition. For a cover $\{U_A, U_B\}$ in the topology $J_{\mathrm{Bures}}$:

$$\mathrm{Sh}_\infty(\mathcal{C})(U_A \cup U_B) \xrightarrow{\sim} \mathrm{Sh}_\infty(\mathcal{C})(U_A) \times_{\mathrm{Sh}_\infty(\mathcal{C})(U_A \cap U_B)} \mathrm{Sh}_\infty(\mathcal{C})(U_B)$$

The no-signaling prohibition is a consequence of the gluing condition for sheaves: local data on $U_A$ do not affect global data restricted to $U_B$ (when $U_A \cap U_B = \varnothing$ for spatially separated systems).

Phenomenal functor and Yoneda lemma

Uniqueness of the phenomenal functor

The functor $F: \mathbf{DensityMat} \to \mathbf{Exp}$:

$$F(\Gamma) := (\text{Spec}(\rho_E), \text{Quality}(\rho_E), \text{Context}(\Gamma_{-E}))$$

is unique (up to isomorphism) among functors compatible with (1) the ∞-topos structure, (2) the distinguished role of E, (3) CPTP-compatibility, (4) monotonicity of the metric.

Uniqueness follows from:

• Partial trace $\text{Tr}_{\bar{E}}$ — unique counit of the adjunction $(-) \otimes \mathcal{H}_{\bar{E}} \dashv \text{Tr}_{\bar{E}}$
• Spectral decomposition — unique for nondegenerate spectrum
• Fubini-Study metric — unique monotone metric (Chentsov-Petz)

Complete proof: Uniqueness theorem FV.

Relational identity of qualia (Yoneda lemma)

By the Yoneda lemma, a quality $[|q\rangle] \in \text{Ob}(\mathbf{Exp})$ is completely determined by its functor of points $h_{[q]} := \text{Hom}_{\mathbf{Exp}}(-, [|q\rangle])$.

Corollary: Inverted qualia are impossible — two qualities with the same relational position (same $d_{FS}$ to all other qualities) are identical by the Yoneda lemma.

More details: Relational identity.

16. Self-referential closure

16.1 Internal theory as a subobject of Ω

The subobject classifier $\Omega$ from L-unification generates not only Lindblad operators, emergent time, and L-dimension, but also an internal object of the theory:

$$\mathrm{Th}_{\mathrm{UHM}} := \{p \in \Omega \mid \varphi^*(p) = p\} \subseteq \Omega$$

where $\varphi^*: \Omega \to \Omega$ — inverse image of predicates under self-modeling $\varphi$. All predicates derivable from axioms A1–A5 are elements of $\mathrm{Th}_{\mathrm{UHM}}$.

Complete proof: Theorem T-54.

16.2 Categorical incompleteness

By Lawvere's fixed point theorem for a Cartesian closed ∞-category $\mathrm{Sh}_\infty(\mathcal{C})$ (HTT, Prop. 6.1.0.6):

$$\mathrm{Th}_{\mathrm{UHM}} \subsetneq \Omega$$

If $\mathrm{Th}_{\mathrm{UHM}} = \Omega$, then $\varphi^* = \mathrm{id}_\Omega$, hence $\varphi = \mathrm{id}$ (since $\Omega$ separates points). But $\mathcal{D}_\Omega \neq 0$ generates nontrivial dynamics, therefore $\varphi \neq \mathrm{id}$. Contradiction.

Complete proof: Theorem T-55.

16.3 Connection to the Yoneda lemma

The Yoneda lemma from §15.5 asserts that an object is determined by its relations. Applied to $\mathrm{Th}_{\mathrm{UHM}}$:

$$y(\mathrm{Th}_{\mathrm{UHM}}) = \mathrm{Hom}(-, \mathrm{Th}_{\mathrm{UHM}})$$

The theory is determined by all morphisms into it — all the ways in which objects of the ∞-topos "satisfy" the axioms. The Yoneda embedding guarantees that $\mathrm{Th}_{\mathrm{UHM}}$ is a genuine object of $\mathrm{Sh}_\infty(\mathcal{C})$, not an external meta-construction.

16.4 Architecture of self-reference

The self-reference of UHM is organized in three levels:

Level	Object	Self-modeling	Status
0. Holon	$\Gamma \in \mathcal{D}(\mathbb{C}^7)$	$\varphi: \Gamma \to \Gamma$, $\rho^* = \varphi(\rho^*)$	[Т]
1. Category Hol	Objects — holons, morphisms — CPTP	L-unification, $G_2$-rigidity	[Т]
2. Internal theory	$\mathrm{Th}_{\mathrm{UHM}} \subseteq \Omega$	$\varphi^*$-closedness, incompleteness, openness	[Т] (T-54–T-56)

The self-reference loop closes through three mechanisms:

1. Internal: $\varphi(\rho^*) = \rho^*$ — the holon models itself
2. Structural: $\mathrm{Th}_{\mathrm{UHM}} \in \mathrm{Sub}(\Omega)$ — the theory is an object of its own universe
3. Evolutionary: O-injection expands $\mathrm{Th}_{\mathrm{UHM}} \to \mathrm{Th}_{\mathrm{UHM}}'$ — incompleteness generates growth

More details: Consequences — self-referential closure.

---

Categorical completeness of UHM

Theorem (Closure of axiomatics) [Т]

Theorem (Categorical closure) [Т]

Axioms A1-A4 of UHM form a categorically closed system: all constructions definable in the ∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$ are expressible via A1-A4 without invoking external objects.

Proof (3 steps).

Step 1 (Internal language). The ∞-topos $\mathbf{Sh}_\infty(\mathcal{C})$ has an internal language — homotopy type theory (HoTT) (Lurie HTT 6.1.0.6, Shulman 2019). All definitions and theorems of UHM are formulated in this language.

Step 2 (Classifier Ω). The subobject classifier Ω defines the internal logic:

• Lindblad operators $L_k$ — atoms of Ω (A1 + L-unification [Т])
• Measures P, R, Φ — defined via Tr (built into D(ℂ⁷))
• Thresholds P_crit, R_th, Φ_th — derived from A1-A4 ([Т])
• Evolution dΓ/dτ = ℒ_Ω[Γ] — derived from Ω (T-57 [Т])

Step 3 (Absence of external dependencies). The only historical dependence — A5 (Page–Wootters) — is derivable from A1-A4 (T-87 [Т]). All 160+ theorems are derived from A1-A4 without external postulates. $\blacksquare$

Connection to the Lurie–Shulman program

UHM realizes a concrete instance of the ∞-topos physics program (Schreiber 2013, Shulman 2019):

Component of the program	Realization in UHM	Status
∞-topos as "space"	$\mathbf{Sh}_\infty(\mathcal{D}(\mathbb{C}^7))$	A1 [Т]
Cohesion	$J_{Bures}$-covers	A2 [Т]
Differential structure	Spectral triple T-53	[Т]
Quantization	CPTP-morphisms	[Т]
Gauge symmetry	$G_2 = \mathrm{Aut}(\mathbb{O})$	[Т]
Gravity	Emergent from NCG (T-120)	[Т]

Theorem (HoTT-interpretation of hierarchy L) [Т]

Theorem (Hierarchy L as n-truncations) [Т]

Interiority levels L0-L4 are isomorphic to n-truncations of the ∞-groupoid $\mathbf{Exp}_\infty$ in HoTT:

$L_n \cong \|\mathbf{Exp}_\infty\|_n$

where $\|\cdot\|_n$ — n-truncation (propositional truncation to level n).

Proof. From T-91 [Т] (∞-groupoid $\mathbf{Exp}_\infty$ — Kan complex):

• $\|X\|_0$ = set of connected components = L0 (discrete states)
• $\|X\|_1$ = groupoid = L1 (phenomenal paths)
• $\|X\|_2$ = 2-groupoid = L2 (reflection)
• $\|X\|_n$ for n ≥ 3 = L3+ (meta-reflection)
• $\lim_{n\to\infty} \|X\|_n = X$ = L4 (colimit, T-86 [Т])

Postnikov truncations provide the canonical filtration. $\blacksquare$

---

Related documents:

• Theorem on emergent time — time as 1-morphism and collapse of strata
• Free will — formalization of freedom via ∞-categories
• Axiom Ω⁷ — 5 axioms of categorical formalism
• Coherence matrix — definition of $\Gamma$
• Holon — definition of $\mathbb{H}$ and 7 dimensions
• Interiority dimension — $E$ and $\mathcal{H}_E$
• Foundation dimension — O as internal clock
• Spacetime — emergent geometry
• Formalization of operator φ — CPTP channels
• Interiority hierarchy — function $\text{Exp}$
• Evolution — dynamics $\Gamma(\tau)$
• Self-observation — measures $R$, $C$, $D_{\text{diff}}$
• Protocol for measuring Γ — operationalization for AI
• Physical correspondence — No-signaling — complete proofs NS1-NS3
• Hard problem — Phenomenal functor — uniqueness of FV and relational identity of qualia
• Consequences — self-referential closure — Th_UHM = Sub_closed(Ω), Lawvere incompleteness, structural ToE (T-54–T-56)