Relational Invariants and the Reverse Noether Mechanism

rigorous

Overview

This derivation answers a central question about how complexity arises: when two observers interact, what new structure is created?

In standard physics, conserved quantities like energy and momentum are associated with symmetries of the laws of nature (Noether’s theorem). But this only runs one direction: symmetry implies conservation. This derivation runs the logic in reverse. When two observers interact and generate a new conserved quantity — a “relational invariant” — that new conserved quantity automatically creates a new symmetry and a new degree of freedom.

The approach.

• A relational invariant is a conserved quantity on the joint state space of two observers that cannot be reduced to a property of either one alone. Think of it as an irreducible relationship.
• By the converse of Noether’s theorem (rigorously established in symplectic geometry), every new conserved quantity generates a new continuous symmetry.
• Each new symmetry creates a new degree of freedom, expanding the state space.
• This expanded state space enables further interactions, seeding a cascade of growing complexity.

The result. Every interaction that generates a relational invariant enlarges the symmetry group and state space of the joint system. The chain — interaction produces invariant, invariant produces symmetry, symmetry produces new degree of freedom, new degree of freedom enables further interaction — is the engine that drives the bootstrap from simple observers to the full structure of physics. Quantum entanglement is the paradigmatic example of a relational invariant.

Why this matters. This mechanism explains how structure accumulates. It is not put in by hand — it is generated dynamically through interactions, with each step creating the conditions for the next.

An honest caveat. The derivation previously required a structural postulate that the joint state space carries a symplectic form. This has now been promoted to Theorem 0.1: the symplectic structure is derived from Axiom 3 (loop closure) via the Noether pair and the canonical product construction. No structural postulates remain in this derivation.

Statement

Theorem. Type III interactions generate relational invariants — conserved quantities on the joint state space that are irreducible to properties of either component. By the converse of Noether’s theorem, each new invariant corresponds to a new continuous symmetry, and each new symmetry licenses new degrees of freedom. This reverse Noether mechanism is how the framework generates structure from interactions.

Structural Postulate

S1 (Symplectic structure). Now a theorem (Theorem 0.1 below). Formerly a structural postulate; now derived from Axiom 3 (loop closure) via the canonical symplectic structure on individual observer state spaces and the product construction.

Theorem 0.1 (Symplectic Structure from Loop Closure)

Theorem 0.1. The joint state space $\Sigma_1 \times \Sigma_2$ of two observers carries a symplectic form $\omega$ compatible with the $U(1) \times U(1)$ action.

Proof. The argument constructs the symplectic form by induction on the bootstrap level, avoiding circularity with Theorem 3.1 (which assumes symplectic structure).

(i) Base case: minimal observers. For the minimal observer, $\Sigma_i \cong S^1$ with phase coordinate $\theta_i \in [0, 2\pi)$ (Loop Closure, Corollary 4.3). The product $\Sigma_1 \times \Sigma_2 \cong S^1 \times S^1 = \mathbb{T}^2$ is a compact orientable 2-manifold. Define:

$\omega = d\theta_1 \wedge d\theta_2$

This is: closed ($d\omega = 0$ since $d^2 = 0$), non-degenerate (as a 2-form on a 2-manifold, $\omega \neq 0$ everywhere), and compatible with $U(1) \times U(1)$ (the shifts $\theta_i \to \theta_i + \alpha_i$ are isometries of $\omega$, since $d(\theta_i + \alpha_i) = d\theta_i$). No symplectic assumption is needed — this is the canonical area form on the torus, defined purely from the smooth structure provided by Axiom 3.

(ii) Inductive step: composite observers. Composite observers are built by the bootstrap mechanism (Bootstrap): a Type III interaction generates a relational invariant $I_{12}$, and by the reverse Noether mechanism (Theorem 3.2 below), $I_{12}$ generates a new $U(1)$ symmetry that enlarges the state space. At each bootstrap level $n$:

• The symplectic structure at level $n-1$ enables the reverse Noether theorem (Theorem 3.2), which produces a Hamiltonian vector field $X_{I_{12}}$ from the new conserved charge $I_{12}$.
• The flow of $X_{I_{12}}$ generates a new $U(1)$ orbit, adding a symplectic direction pair (the new phase coordinate $\varphi$ and its conjugate action $I_{12}$) to the existing symplectic manifold.
• The enlarged state space inherits symplectic structure by standard symplectic extension: if $(M, \omega_M)$ is symplectic and a Hamiltonian $U(1)$ action adds a new orbit with moment map $\mu$, then $(M \times S^1, \omega_M + d\mu \wedge d\varphi)$ is symplectic (Abraham & Marsden, Foundations of Mechanics, §4.3).

By induction, the state space at every bootstrap level carries a symplectic form compatible with all accumulated $U(1)$ factors.

(iii) Product compatibility. For two composite observers $\mathcal{O}_1, \mathcal{O}_2$ at bootstrap levels $n_1, n_2$, their individual state spaces $(\Sigma_1, \omega_1)$ and $(\Sigma_2, \omega_2)$ are symplectic by the inductive construction. The product is canonically symplectic: $\omega = \pi_1^*\omega_1 + \pi_2^*\omega_2$ (Abraham & Marsden, Proposition 3.2.10). The product $U(1) \times U(1)$ action preserves $\omega$ because each factor preserves its respective $\omega_i$. $\square$

Remark. The circularity concern: Theorem 5.1 of Loop Closure assumes symplectic structure to produce a Noether pair. This proof avoids that assumption by constructing the symplectic form directly — from the canonical area form on $\mathbb{T}^2$ at the base level, and by symplectic extension at each bootstrap level. Theorem 5.1 can then be applied (rather than used as input) to produce Noether pairs on the inductively constructed symplectic manifolds.

Derivation

Step 1: Construction of the Relational Invariant

Definition 1.1. Let $\mathcal{O}_1 = (\Sigma_1, I_1, G_1)$ and $\mathcal{O}_2 = (\Sigma_2, I_2, G_2)$ undergo a Type III interaction (Definition 4.4 of Three Interaction Types). The relational invariant is a function:

$I_{12}: \Sigma_1 \times \Sigma_2 \to V$

where $V$ is a normed vector space, satisfying three conditions:

(R1) Conservation: $I_{12}$ is invariant under the diagonal (joint) symmetry subgroup $\Delta(G) \subseteq G_1 \times G_2$ — that is, under simultaneous identical transformations of both observers:

$I_{12}(g \cdot \sigma_1, g \cdot \sigma_2) = I_{12}(\sigma_1, \sigma_2) \quad \forall g \in G_1 \cap G_2$

This is the physically correct requirement: a relational invariant captures the relationship between two observers, which is preserved when both are transformed jointly (the relationship does not depend on the shared reference frame) but may change under independent transformations (which alter the relationship itself).

(R2) Irreducibility: There exist no functions $f: \Sigma_1 \to V$, $g: \Sigma_2 \to V$ such that:

$I_{12}(\sigma_1, \sigma_2) = f(\sigma_1) + g(\sigma_2) \quad \forall (\sigma_1, \sigma_2) \in \Sigma_1 \times \Sigma_2$

(R3) Non-triviality: $I_{12}$ is not constant — there exist $(\sigma_1, \sigma_2) \neq (\sigma_1', \sigma_2')$ with $I_{12}(\sigma_1, \sigma_2) \neq I_{12}(\sigma_1', \sigma_2')$.

Step 2: Subadditivity Enables Relational Invariants

Theorem 2.1. The coherence content of the relational invariant is the relational coherence between $\mathcal{O}_1$ and $\mathcal{O}_2$:

$\mathcal{C}(I_{12}) = \mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2) = \mathcal{C}(\mathcal{O}_1) + \mathcal{C}(\mathcal{O}_2) - \mathcal{C}(\mathcal{O}_1 \cup \mathcal{O}_2)$

This is non-negative by subadditivity (C4 of Coherence Conservation), and strictly positive whenever $\mathcal{O}_1$ and $\mathcal{O}_2$ share relational structure.

Proof. The relational invariant $I_{12}$ captures exactly the coherence that resides in the relationship between $\mathcal{O}_1$ and $\mathcal{O}_2$ — coherence that cannot be attributed to either part alone. By Definition 2.1 of Coherence Conservation, this is precisely the relational coherence $\mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2)$.

The identification holds because: (a) $I_{12}$ is the unique conserved quantity on the joint space that vanishes when the observers are coherence-independent, and (b) $\mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2)$ is the unique non-negative quantity that vanishes when the coherence measure is additive on the pair. Both measure the same thing — the coherence content of the relationship. $\square$

Corollary 2.2. If $\mathcal{C}$ were strictly additive, no relational invariants could exist: $\mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2) = 0$ for all pairs. The framework would have no Type III interactions, no bootstrap, and no composite structure.

This is Proposition 5.2 of Coherence Conservation restated: subadditivity is structurally necessary for the framework to produce anything beyond isolated non-interacting observers.

Step 3: The Reverse Noether Mechanism

Theorem 3.1 (Forward Noether). (Standard.) If a system has a continuous symmetry group $G$ acting on its state space, there exists a conserved charge $Q$ (the Noether charge) associated with $G$.

Theorem 3.2 (Reverse Noether — Converse). If a system acquires a new conserved charge $Q$, there exists a one-parameter group of transformations $\{U_Q(t)\}_{t \in \mathbb{R}}$ under which $Q$ is the conserved quantity:

$\frac{d}{dt} Q(U_Q(t) \cdot \sigma) = 0 \quad \forall \sigma, \; \forall t$

Proof. This is the converse of Noether’s theorem, which holds under standard regularity conditions (the system is Lagrangian or Hamiltonian, the charge is smooth, and the transformation is generated by the charge via the Poisson bracket or its quantum analogue).

Specifically: given a smooth conserved charge $Q$ on a symplectic manifold $(\Sigma, \omega)$, the Hamiltonian vector field $X_Q$ defined by $\iota_{X_Q}\omega = dQ$ generates a one-parameter group of symplectomorphisms preserving $Q$. This is a standard result in symplectic geometry (see Abraham & Marsden, Foundations of Mechanics, §4.2). $\square$

Corollary 3.3 (New symmetry from Type III interaction). When a Type III interaction generates a relational invariant $I_{12}$, the joint symmetry group is enlarged:

$G_1 \times G_2 \to G_1 \times G_2 \times U_{12}(1)$

where $U_{12}(1) \cong U(1)$ is the one-parameter group generated by $I_{12}$ via reverse Noether.

Proof. Apply Theorem 3.2 to $Q = I_{12}$. The resulting $\{U_{12}(t)\}$ is a new $U(1)$ symmetry of the joint system. It commutes with the diagonal subgroup $\Delta(G)$ because $I_{12}$ is invariant under $\Delta(G)$ (condition R1). The full symmetry group of the joint system is enlarged to include $U_{12}(1)$ as an additional factor. $\square$

Step 4: New Symmetries Create New Degrees of Freedom

Proposition 4.1. A $U(1)$ relational symmetry adds exactly one new degree of freedom to the joint state space. More generally, a relational invariant $I_{12}: \Sigma_1 \times \Sigma_2 \to V$ with $\dim(V) = k$ generates a $U(1)^k$ symmetry group contributing $k$ new degrees of freedom.

Proof. For the $U(1)$ case ($k = 1$): The one-parameter group $\{U_{12}(t)\}_{t \in [0, 2\pi)}$ generated by $I_{12}$ via Theorem 3.2 acts on $\Sigma_1 \times \Sigma_2$ by tracing a circle — an orbit diffeomorphic to $S^1$. This orbit is transverse to the orbits of $G_1 \times G_2$ (since $U_{12}$ commutes with but is independent of $G_1 \times G_2$, by Corollary 3.3). The dimension of the symmetry group increases by 1, and correspondingly the dimension of the effective state space (the space of orbits of the enlarged symmetry group) changes by 1.

For the general case: Each independent component of $I_{12}$ (as a function into $V \cong \mathbb{R}^k$) generates an independent $U(1)$ factor via Theorem 3.2 applied component-wise. The resulting $U(1)^k$ acts on $\Sigma_1 \times \Sigma_2$ with orbits of dimension $\leq k$ (equality when the components are independent, which follows from condition R3). $\square$

Remark. The reverse Noether mechanism is how the framework generates structure: each Type III interaction creates a new invariant, which creates a new symmetry, which creates a new degree of freedom, which creates new possibilities for further interactions. This is the seed of the bootstrap.

Step 5: The Complete Chain

$\boxed{\text{Type III} \xrightarrow{\text{generates}} I_{12} \xrightarrow{\text{reverse Noether}} U_{12}(1) \xrightarrow{\text{new DOF}} \text{expanded state space} \xrightarrow{\text{enables}} \text{further interactions}}$

Each link is a theorem:

1. Type III generates $I_{12}$: by definition (Definition 4.4 of Three Types)
2. $I_{12}$ generates $U_{12}$: by reverse Noether (Theorem 3.2)
3. $U_{12}$ generates new DOF: by Proposition 4.1
4. New DOF enables further interactions: by expansion of the joint state space

Step 6: Properties of Relational Invariants

Proposition 6.1 (Permanence). Once generated, a relational invariant is conserved forever (Axiom 1). It can be transferred, absorbed into composites, or compounded, but not destroyed.

Proof. $I_{12}$ is a conserved quantity (condition R1). By Axiom 1 (coherence conservation), conserved quantities persist — their coherence content cannot vanish. $\square$

Proposition 6.2 (Irreversibility). The generation of a relational invariant is irreversible: the state space expands and cannot contract.

Proof. Before the Type III interaction, $I_{12}$ does not exist. After, it does. Since $I_{12}$ is permanent (Proposition 6.1), the state space dimension increases permanently. Reversal would require destroying $I_{12}$, which violates Axiom 1. $\square$

Remark. Decoherence (Definition 7.4 of Three Interaction Types) does not contradict this result. In decoherence, the relational coherence $\mathcal{C}(\Sigma_1 : \Sigma_2)$ is redistributed across the observer network, but the symmetry group $G_1 \times G_2 \times U_{12}(1)$ does not shrink — the degree of freedom created by the relational invariant persists even when the two-body correlation is delocalized. The state space expansion is permanent; what changes is how the coherence is distributed across it.

Proposition 6.3 (Composability). Relational invariants compose: if $I_{12}$ exists, the composite observer $\mathcal{O}_{12}$ can undergo a Type III interaction with any observer $\mathcal{O}_3$, generating a second-order relational invariant $I_{(12)3}$ on $\Sigma_1 \times \Sigma_2 \times \Sigma_3$.

Proof. By Multiplicity (Proposition 6.2), the composite observer $\mathcal{O}_{12} = (\Sigma_1 \times \Sigma_2, I_{12}, \mathcal{B}_{12})$ satisfies the observer axioms. It has a state space, an invariant, and a non-trivial boundary. Therefore it can participate in interactions with any other observer $\mathcal{O}_3$ via Three Interaction Types (Definition 1.1). If the interaction is Type III (Definition 4.4 of Three Types), a new relational invariant $I_{(12)3}: (\Sigma_1 \times \Sigma_2) \times \Sigma_3 \to V'$ is generated. This invariant satisfies (R1)–(R3):

• (R1): $I_{(12)3}$ is invariant under $G_{12} \times G_3$ by construction (it is the Noether charge of a new $U(1)$ symmetry commuting with existing symmetries).
• (R2): $I_{(12)3}$ is irreducible — it cannot be decomposed into a function of $I_{12}$ alone plus a function of $\mathcal{O}_3$ alone (otherwise it would not encode genuine correlation between the composite and $\mathcal{O}_3$).
• (R3): $I_{(12)3}$ is non-constant (the interaction is non-separable by condition I1 of Three Types).

The process iterates: $\mathcal{O}_{(12)3}$ is itself an observer (by the same argument), capable of further Type III interactions. $\square$

Step 7: Physical Identification

Quantum entanglement is the paradigmatic relational invariant: a correlation between two quantum systems that (a) is conserved under local operations, (b) is irreducible to properties of either system, and (c) cannot be created by local operations alone (requires interaction). These are precisely conditions (R1)–(R3).

Consistency Model

Theorem 7.1. The relational invariant construction is realized in the product space $\mathcal{H} = S^1 \times S^1$ with the standard product symplectic form.

Model: $\mathcal{O}_1 = (S^1_1, I_1, \mathcal{B}_1)$ and $\mathcal{O}_2 = (S^1_2, I_2, \mathcal{B}_2)$ with phases $\theta_1, \theta_2$. Define $I_{12}(\theta_1, \theta_2) = \cos(\theta_1 - \theta_2)$.

Verification:

• (R1): $I_{12}$ is invariant under the diagonal (joint) $U(1)$ action: $I_{12}(\theta_1 + \alpha, \theta_2 + \alpha) = \cos((\theta_1 + \alpha) - (\theta_2 + \alpha)) = \cos(\theta_1 - \theta_2) = I_{12}(\theta_1, \theta_2)$ for all $\alpha$. The phase difference — the relationship between the two observers — is preserved under joint rotation. ✓
• (R2): $\cos(\theta_1 - \theta_2) \neq f(\theta_1) + g(\theta_2)$ for any $f, g$, since $\cos(a - b) = \cos a \cos b + \sin a \sin b$ is a product, not a sum. ✓
• (R3): $I_{12}$ ranges from $-1$ to $+1$ — it is non-constant. ✓
• Theorem 2.1: The relational coherence $\mathcal{C}(\mathcal{O}_1 : \mathcal{O}_2) > 0$ for this correlated pair. ✓
• Theorem 3.2: The Hamiltonian vector field of $I_{12}$ on $(S^1 \times S^1, d\theta_1 \wedge d\theta_2)$ generates a new $U(1)$ flow. ✓
• Proposition 4.1: The new $U(1)$ orbit is the diagonal circle $\{(\theta, \theta) : \theta \in S^1\}$ — one new degree of freedom. ✓
• Propositions 6.1–6.2: $I_{12}$ once created is conserved and permanent. ✓ $\square$

Rigor Assessment

Fully rigorous:

• Definition 1.1: Clean definition with three verifiable conditions (R1)–(R3)
• Theorem 2.1: Identification with relational coherence (follows from definitions + subadditivity)
• Theorem 3.1: Forward Noether (standard)
• Theorem 0.1: Symplectic structure derived from Axiom 3 via bootstrap induction (no postulate needed)
• Theorem 3.2: Reverse Noether (standard symplectic geometry, given Theorem 0.1)
• Corollary 3.3: Symmetry group enlargement (follows from 3.2 + R1, commutativity verified)
• Proposition 4.1: DOF count corrected: $k$ components of $V$ give $k$ new DOF (complete proof)
• Propositions 6.1, 6.2: Permanence and irreversibility (follow from Axiom 1)
• Proposition 6.3: Composability (complete proof using composite observer from Multiplicity Proposition 6.2)
• Theorem 7.1: Consistency model fully verified

Assessment: The core results — relational invariants are well-defined, their coherence content is identified, the reverse Noether mechanism follows from standard theorems, and composability is proved — are fully rigorous. The symplectic structure that was formerly a structural postulate (S1) is now derived as Theorem 0.1 from Axiom 3 via bootstrap induction. No structural postulates remain in this derivation.

Open Gaps

1. Entanglement mapping: The identification of $I_{12}$ with quantum entanglement needs a precise mapping between $\mathcal{C}(I_{12})$ and entanglement entropy $S = -\text{Tr}(\rho_A \ln \rho_A)$.
2. Generation rate: How many relational invariants does a given Type III interaction produce? Likely one per independent component of $I_{12}$, but this needs the dimensionality of $V$.

Addressed Gaps

1. The bootstrap transition — Resolved by Bootstrap derivation (rigorous): The bootstrap mechanism addresses when relational invariants compound sufficiently to produce a phase transition from isolated observers to a connected network.