# Derivation 13: Einstein Field Equations from the Rational Field Equation

## Theorem

The self-consistency equation N(p/q) = N_total × g(p/q) × w(p/q, K₀F[N])
on the Stern-Brocot tree, at critical coupling K = 1 in the continuum
limit, produces the Einstein field equations as its unique output:

    G_μν + Λ g_μν = 8πG T_μν

No other field equation is consistent with the four conditions the
continuum limit satisfies: (a) metric theory on 3+1 manifold,
(b) self-consistency origin, (c) second-order in metric derivatives,
(d) general covariance. Uniqueness follows from Lovelock's theorem (1971).

---

## Part I: Exact ADM from Kuramoto (closing the weak-gradient gap)

### The first evolution equation is exact

The weak-gradient derivation (Derivation 12) showed ∂γᵢⱼ/∂t =
-2N𝒦ᵢⱼ + DᵢNⱼ + DⱼNᵢ with the flat connection. The extension to
the exact (nonlinear) case requires no additional assumptions beyond
the locked-state conditions already used.

**Why.** The Christoffel symbols Γᵏᵢⱼ of γᵢⱼ = Cᵢⱼ/C₀ are
expressible in terms of Kuramoto three-point correlations:

    Tᵢⱼₗ = ⟨∂ᵢ∂ⱼθ ∂ₗθ⟩ + ⟨∂ⱼθ ∂ᵢ∂ₗθ⟩

via the standard Levi-Civita formula:

    Γᵏᵢⱼ = (γᵏˡ/2C₀)(Tₗᵢⱼ - Tᵢⱼₗ - Tⱼᵢₗ) + (conformal terms from C₀)

This is a **tautology of Riemannian geometry**: given any smooth
positive-definite symmetric tensor field, its Levi-Civita connection
exists and is unique. The metric compatibility Dₖγᵢⱼ = 0 holds
identically by the fundamental theorem of Riemannian geometry. No
dynamical input is needed.

The nontrivial content is that the Kuramoto dynamics preserves this
structure. The first ADM equation holds exactly under four conditions:

| Condition | Status |
|-----------|--------|
| C₀ = 1 (or absorbed into lapse) | Gauge choice |
| ⟨∂ᵢθ⟩ = 0 (centered ensemble) | Ensemble symmetry |
| ⟨cos(ψ-θ) ∂ⱼθ⟩ = ∂ⱼψ | Locked state (K ≈ 1) |
| ⟨sin(ψ-θ) ∂ⱼθ⟩ = 0 | Locked state (antisymmetry) |

The first two are kinematic. The last two are dynamical consequences
of the locked state at K ≈ 1. **No weak-gradient assumption is
needed.** The connection is exact; the approximation enters only
through the locked-state condition on ensemble statistics.

### The 𝒦ᵢⱼ evolution equation (second ADM equation)

Starting from 𝒦ᵢⱼ = ⟨∂ᵢθ cos(ψ-θ) ∂ⱼθ⟩, differentiate under
the ensemble average using the Kuramoto equation. The result
decomposes into five classes of terms:

**Term 1: -DᵢDⱼN** (double covariant derivative of lapse)

Arises from the exact relation ∂ᵢN = ⟨sin(ψ-θ) ∂ᵢθ⟩ (which follows
from differentiating r = ⟨cos(ψ-θ)⟩ with centered ensemble). The
second derivative gives:

    ∂ᵢ∂ⱼN ≈ NᵢNⱼ/N - 𝒦ᵢⱼ    (locked state, zeroth order in φ)

So -DᵢDⱼN ≈ 𝒦ᵢⱼ - NᵢNⱼ/N + Γᵏᵢⱼ∂ₖN.

**Term 2: N ³Rᵢⱼ** (Ricci tensor of spatial metric)

Kinematic — follows from the definition γᵢⱼ = Cᵢⱼ/C₀. The Ricci
tensor is expressible in terms of Kuramoto three- and four-point
correlations (derivatives of Tᵢⱼₗ). No dynamical input or locked-state
approximation needed. This is the phase stiffness — the curvature of
coherence.

**Term 3: N(K𝒦ᵢⱼ - 2𝒦ᵢₖ𝒦ᵏⱼ)** (extrinsic curvature self-interaction)

Products of two-point correlations contracted with γⁱʲ. Requires
mean-field factorization: ⟨cos²φ ∂ᵢθ ∂ⱼθ⟩ ≈ ⟨cosφ ∂ᵢθ ∂ₖθ⟩
γᵏˡ⟨cosφ ∂ₗθ ∂ⱼθ⟩. This holds for Gaussian fluctuations about the
locked state (thermodynamic limit or small-fluctuation regime).

**Term 4: ℒ_β 𝒦ᵢⱼ** (Lie derivative along shift)

Shift transport: Nᵢ⟨cos²φ ∂ⱼθ⟩ + (i↔j). At locked state, this
becomes 2NᵢNⱼ/N, matching the advection of extrinsic curvature by
the spatial flow.

**Term 5: Matter terms**

From the ω(x)-dependent correlations. With ω = √(4πGρ), these
give -8πGN(Sᵢⱼ - ½γᵢⱼ(S-ρ)). The precise coefficient requires
normalizing θ as θ/√(4πG) (canonical scalar field).

### Summary of conditions

| ADM Term | Kuramoto origin | Locked state? | Additional? |
|----------|----------------|---------------|-------------|
| ∂γ/∂t = -2N𝒦 + DN | Coherence tensor time derivative | Yes | Centered ensemble |
| -DᵢDⱼN | ∂ᵢ⟨sinφ ∂ⱼθ⟩ | Yes | None |
| N ³Rᵢⱼ | Phase stiffness (kinematic) | **No** | None |
| N(K𝒦-2𝒦²) | Correlation products | Yes | Mean-field factorization |
| ℒ_β 𝒦 | Shift transport | Yes | None |
| Matter | ω-dependent correlations | Partial | Scalar field normalization |

---

## Part II: Uniqueness via Lovelock's theorem

### The four premises

**Premise (a): Metric theory on 3+1 manifold.**

The coherence tensor Cᵢⱼ = ⟨∂ᵢθ ∂ⱼθ⟩ is symmetric (pointwise
multiplication is commutative). At K = 1, every oscillator is locked,
making Cᵢⱼ positive-definite. The ADM construction gives a 4D
spacetime metric ds² = -N²dt² + γᵢⱼ(dxⁱ + βⁱdt)(dxʲ + βʲdt).

**Premise (b): Self-consistency origin.**

The field equation comes from the fixed-point condition N(p/q) =
N_total × g(p/q) × w(p/q, K₀F[N]). The order parameter r appears
on both sides. In the continuum limit, this becomes the Kuramoto
self-consistency integral, whose spatial and temporal integrability
conditions produce the ADM constraints.

**Premise (c): Second-order in metric derivatives.**

The Kuramoto equation is first-order in ∂t and second-order in ∇².
Since γᵢⱼ is quadratic in ∂θ, one ∂t on θ becomes effectively ∂t²
on γ, and ∇² on θ becomes ∇² on γ. No higher derivatives appear.
At K = 1, the phase field θ is smooth (analytic), and the derivative
expansion truncates at second order without fractal corrections from
tongue boundaries.

**Premise (d): General covariance.**

The coherence tensor Cᵢⱼ transforms as a rank-2 tensor under
coordinate changes (it is defined as a correlation of phase
gradients, which are covariant). The Levi-Civita connection,
Riemann tensor, and all derived objects inherit covariance. The
Kuramoto self-consistency condition is coordinate-independent (it
involves the order parameter magnitude |r|, which is a scalar).

### Application of Lovelock's theorem

**Lovelock's theorem (1971):** In four dimensions, the most general
symmetric, divergence-free rank-2 tensor constructed from the metric
and its first and second derivatives is:

    ℰ_μν = α G_μν + β g_μν

where G_μν = R_μν - ½Rg_μν is the Einstein tensor and α, β are
constants.

With α = 1 (choice of units) and β = -Λ:

    **G_μν + Λ g_μν = 8πG T_μν**

The cosmological constant Λ is the uniform background frequency of
the Kuramoto ensemble: Λ = 3(H₀/c)². The matter tensor T_μν comes
from the natural frequency distribution ω(x) = √(4πGρ(x)).

**Uniqueness.** Lovelock's theorem is an if-and-only-if result. No
other tensor satisfies all four conditions. Therefore no other field
equation can arise from the Stern-Brocot continuum limit at K = 1.

---

## Part III: The one assumption beyond the framework

**Assumption A1:** The spatial manifold dimension equals the minimum
self-sustaining loop size N = 3 (from Derivation 6).

This is motivated by the framework (each independent oscillator
direction becomes a spatial direction) but constitutes an independent
geometric assumption linking discrete combinatorics to continuum
dimensionality. All other steps follow from the Kuramoto dynamics,
the ADM dictionary, and established mathematics (Noether, Lovelock).

If A1 is relaxed:
- d = 2: Einstein tensor vanishes identically (no propagating gravity)
- d = 3: standard GR (the theorem's conclusion)
- d ≥ 5: Lovelock's theorem allows additional terms (Gauss-Bonnet,
  etc.). The framework predicts d = 3 spatial dimensions, excluding
  these.

---

## Part IV: Remaining refinements

Three items sharpen the result but do not affect the theorem:

1. **Gauss-Codazzi verification.** The Ricci coefficient in ∂t𝒦ᵢⱼ
   is exactly N (not some other function) because of the Gauss-Codazzi
   embedding equations. Verifying these for the Kuramoto embedding
   confirms the coefficient without relying on Lovelock.

2. **Mean-field factorization.** The 𝒦² terms use ⟨ABCD⟩ ≈ ⟨AB⟩⟨CD⟩.
   Corrections are connected four-point cumulants, which produce
   non-Einstein terms suppressed by 1/N_total (thermodynamic limit).
   In the N_total → ∞ continuum limit, these vanish.

3. **Matter normalization.** The factor 8πG requires normalizing the
   phase field as θ/√(4πG). This is a convention, not a derivation
   of G from the framework.

---

## Status

**Theorem established.** The Einstein field equations with cosmological
constant are the unique output of the Stern-Brocot fixed-point equation
at K = 1, in the continuum limit, under the ADM-Kuramoto dictionary,
given the locked-state conditions and A1 (d = 3).

The uniqueness is not a property of the dictionary — it is a property
of the mathematics (Lovelock). The dictionary maps Kuramoto to ADM.
Lovelock says ADM can only produce Einstein. Therefore Kuramoto at
K = 1 can only produce Einstein. QED.

**Connection to Derivation 19 (Klein bottle):** This derivation
assumes the Stern-Brocot tree with periodic (torus) topology — the
full tree, all modes, no topological constraint. Derivation 19 shows
that imposing Klein bottle topology (XOR parity filter + twist)
collapses the field equation to 4 modes at denominator classes {2,3}.
The fractions that survive numerically match quark charges and gauge
group ranks, but this identification is conjectural.

The open question for this derivation: what happens to the K=1
continuum limit when the XOR constraint is imposed? The standard
result (Einstein from Lovelock) assumes the full mode spectrum is
available. If the continuum limit of the XOR-filtered tree produces
additional field equations beyond Einstein — specifically gauge field
equations with the structure constants of SU(2) and SU(3) — the
Klein bottle identification becomes structural. If it produces only
Einstein with restricted mode content, the numerical matches in D19
are coincidence involving the simplest fractions on the tree.

This is the load-bearing computation that separates the established
results (topology → 4 modes → dimension loop) from the conjectural
ones (modes → particle physics).

---

## Proof chain

This derivation is Proposition P8 (the capstone) in
[**Proof Chain A: Polynomial → General Relativity**](PROOF_A_gravity.md).