# Derivation 1: Born Rule from Basin Measure

## Claim

The Born rule (P = |ψ|²) is the basin measure of the synchronization
cost landscape under dissipative convergence. It is not axiomatic — it
is the generic outcome of dissipative dynamics on a cost landscape with
quadratic structure near attractors.

## Setup

Consider a system with N possible outcome states {s₁, …, s_N}, each
corresponding to an attractor in the synchronization cost landscape.

The system's state is described by a complex amplitude vector ψ, where
ψ_k encodes both the proximity to attractor k and the phase relationship
with the environmental mean field.

The synchronization cost functional near attractor k is:

    C_k(ψ) = C_k⁰ + α_k |ψ - ψ_k|² + O(|ψ - ψ_k|³)

where C_k⁰ is the cost at the attractor and α_k > 0 is the curvature
(stiffness of the basin).

## Dissipative convergence

The dynamics are dissipative — the system loses energy to the
environmental mean field. The equation of motion is:

    dψ/dt = -γ ∇_ψ* C(ψ)

where γ > 0 is the dissipation rate and ∇_ψ* is the Wirtinger derivative
(gradient with respect to the conjugate variable, appropriate for complex
amplitudes).

Near attractor k, this gives:

    dψ_k/dt = -γ α_k ψ_k

with solution:

    ψ_k(t) = ψ_k(0) exp(-γ α_k t)

The amplitude decays exponentially. The system converges to whichever
attractor has the largest initial basin occupation.

## Basin measure

The key question: given initial conditions distributed over the cost
landscape, what fraction ends up at each attractor?

For a quadratic cost landscape, the basin of attractor k has volume
proportional to |ψ_k(0)|² in the phase space. This follows from:

1. **Liouville's theorem for dissipative systems**: Phase space volume
   contracts uniformly along each attractor's basin. The contraction
   rate is 2γα_k per complex dimension.

2. **The basin boundary** is determined by the cost landscape's saddle
   points. For a system with quadratic basins and uniform curvature
   (α_k = α for all k), the basin of attractor k in amplitude space
   has measure:

       μ(B_k) = |ψ_k|² / Σ_j |ψ_j|²

   This is exactly the Born rule.

3. **Non-uniform curvature**: If α_k varies between attractors, the
   basin measure becomes:

       μ(B_k) = |ψ_k|² / α_k  ×  [Σ_j |ψ_j|² / α_j]⁻¹

   The standard Born rule is recovered when all attractors have equal
   curvature — i.e., when the cost landscape is symmetric under
   permutation of outcomes. This is the case for a measurement apparatus
   that treats all outcomes equivalently (a fair detector).

## Why |ψ|² and not |ψ|⁴ or |ψ|

The exponent 2 comes from the cost functional being quadratic near
attractors. This is generic:

- Linear cost (|ψ|) would give non-differentiable dynamics at the
  attractor — physically, the convergence would never complete smoothly.

- Higher-order cost (|ψ|⁴, |ψ|⁶) is possible but unstable to
  perturbation — any small quadratic term dominates near the attractor
  where |ψ| is small.

- Quadratic is the lowest-order smooth, stable cost structure. It is
  the generic case, not a special one. The Born rule holds because
  quadratic basins are structurally stable.

This is analogous to why Gaussian distributions appear everywhere:
the central limit theorem follows from quadratic structure near the
mean of any smooth distribution. The Born rule is the central limit
theorem of cost landscapes.

## Connection to lattice results

The Stribeck lattice (a chain of friction oscillators coupled by elastic springs, showing mode-locking and bifurcation thresholds analogous to synchronization transitions) has two basins: stick (subharmonic) and slip
(fundamental). The bifurcation threshold is the saddle point between
them. Below threshold, the system occupies the slip basin entirely.
Above threshold, it transitions to the stick basin.

The basin measure — the fraction of phase space that drains into each
attractor — determines which regime dominates. At the bifurcation point,
the basin volumes are equal (50/50). This is the lattice analog of
equal-amplitude superposition: |ψ_stick|² = |ψ_slip|² = 0.5.

The sharp transition in the lattice (not gradual) is consistent with
the Born rule's discreteness — you get one outcome, not a mixture.
The "measurement" is the nonlinear coupling driving the system past
the bifurcation point.

## Connection to Arnold tongue geometry

The quadratic basin structure above was assumed. It can be *derived*
from the circle map dynamics (the circle map is the canonical discrete map for driven synchronization, defined by theta_{n+1} = theta_n + Omega - (K/2pi)sin(2pi*theta_n)) that underlie the spectral tilt
derivation (see [`born_rule_tongues.py`](https://github.com/nickjoven/harmonics/blob/main/sync_cost/derivations/born_rule_tongues.py)).

Every Arnold tongue boundary is a **saddle-node bifurcation**: a
stable fixed point (attractor) and an unstable fixed point (repeller)
collide and annihilate. Inside the tongue, they sit apart by a
distance Δθ on the circle. Near the boundary:

    Δθ  =  √(4ε / πK)

where ε is the depth inside the tongue (distance from the boundary
in Ω-space) and K is the coupling strength.

This is exact, not approximate. The exponent is 1/2:

    Δθ  ∝  ε^{1/2}       (confirmed numerically to 6 decimal places)

Therefore:

    Δθ²  ∝  ε             (squared separation is LINEAR in the parameter)

The tongue width — the range of Ω values that mode-lock to a given
rational p/q — is proportional to Δθ²_max. This means:

    P(p/q)  ∝  tongue_width  ∝  Δθ²_max  ∝  |ψ|²

The identification: |ψ_k| is the basin separation Δθ_k at attractor k.
The probability P_k is the parameter-space measure (tongue width).
The Born rule P = |ψ|² follows from the parabolic geometry of
saddle-node bifurcation.

**Why the exponent is exactly 2**: Saddle-node is the *generic*
codimension-1 bifurcation for maps on the circle. Its normal form
is x² + μ = 0, giving x = ±√μ. The square root is structurally
stable — small perturbations don't change the exponent. No other
exponent is generic. This answers the open question: quadratic basins
aren't assumed, they're the universal geometry of mode-locking.

## Status

This derivation recovers the Born rule from two independent routes:

1. **Basin measure** (above): quadratic cost landscapes give |ψ|²
   weighting. The exponent 2 is the lowest-order smooth stable cost.

2. **Tongue geometry** (born_rule_tongues.py): Arnold tongue boundaries
   are saddle-node bifurcations with Δθ ∝ √ε, giving P ∝ Δθ² = |ψ|².
   The exponent 2 is the universal normal form, not a choice.

Route 2 answers the open question from route 1: the quadratic
structure IS derivable from the self-consistency of the circle map.
The map's fixed-point equation (θ determines the field that determines θ)
produces saddle-node bifurcations at every tongue boundary, which
produce |ψ|².

**Resolved**: Quadratic basin structure follows from saddle-node
universality at Arnold tongue boundaries. Not assumed — derived.

**Open**: The uniform curvature condition (α_k = α) recovers the
standard Born rule. Non-uniform curvature gives a modified rule.
Are there physical systems where the modification is measurable?
Candidates: systems near the quantum-classical boundary where the
cost landscape is asymmetric.

---

## Proof chain

This derivation is Proposition Q2 in
[**Proof Chain B: Polynomial → Quantum Mechanics**](PROOF_B_quantum.md).