β is a coordinate, not a parameter
1. The claim
The inverse temperature β — the single dial of the Maslov-Gibbs Einsum — is a coordinate on the adèlic β-plane: the product $\mathbb{R} \times \prod_p \mathbb{Q}_p$ of the real line and all p-adic completions of ℚ. Every major computational model in mathematics and physics corresponds to a specific location on this plane. Planck’s constant, kinematic viscosity, Black-Scholes volatility, Sinkhorn regularisation, and the temperature parameter of the Viterbi algorithm are all the same object — coordinates on the positive real axis of the β-plane — seen from different fields.
2. Why it matters
Most unification claims in mathematics say that two constructs are analogous. This claim says something stronger: that a single equation,
\[\pi_k(\beta) = \frac{e^{-\beta E_k}}{\sum_j e^{-\beta E_j}}\]produces every major computational regime as an exact consequence of where β sits in its domain — not as an approximation or analogy, but because the partition function $Z_\beta$ has a zero, a branch point, or a limiting behaviour at that location.
If the claim is true:
- Every algorithm has a “dual” obtained by analytically continuing β
- The gap between classical and quantum computing is a statement about p-adic integrality, not about coherence times or engineering
- Population inversion (lasers), PT-symmetric quantum mechanics, and p-adic computation are not exotic specialities — they are coordinates on the same map as Boltzmann statistics
- Ostrowski’s theorem provides a completeness guarantee: the adèlic β-plane is not a convenient parameterisation, it is the only one
3. The map
The positive real axis: Forge ISA and Origami ISA
The positive real axis $0 < \beta < \infty$ is the most familiar territory.
- β → 0 (the Ambient): All states equally weighted. Smooth Hodge theory. No computation, maximum entropy, hot reservoir.
- β finite, real (Forge ISA): Gibbs distribution. Statistical mechanics, simulated annealing, the natural gradient. The snap event at $\beta^\star(\rho) = \frac{3}{8}\ln\frac{1}{1-\rho}$ marks the phase transition from exploratory to committed computation.
- β → ∞ (Origami ISA): The tropical limit. The Gibbs distribution collapses to a delta function on the ground state. Discrete logic, dynamic programming, shortest paths, the $(\max, +)$ semiring.
The imaginary axis: Meld ISA
Wick-rotating β to $it$ replaces Boltzmann weights $e^{-\beta E}$ with quantum phases $e^{-itE}$. The partition function becomes a quantum amplitude. This is not an analogy — it is the same equation with the same index structure, evaluated at an imaginary argument.
The Clifford group is the quantum tropical semiring: its matrix entries live in $\mathbb{Z}[\zeta_8, 1/\sqrt{2}]$, the ring of 2-adically integral elements of the 8th cyclotomic field. The T gate — the first genuinely non-Clifford operation — has matrix entry $e^{i\pi/4}$, which is 2-adically non-integral. The magic gap between Clifford and universal quantum computation is a p-adic integrality condition, visible only once β is treated as a coordinate on the full adèlic plane.
The six classical dualities (Paper 443)
Each of the following famous pairs of equations is the same β-deformation on the positive real axis:
| Pair | Soft side | Hard side | Bridge variable |
|---|---|---|---|
| 1 | Schrödinger | Hamilton-Jacobi | ħ = 1/β (imaginary axis) |
| 2 | Heat equation | Burgers shocks | ν (viscosity) ∝ 1/β |
| 3 | Black-Scholes | HJB / intrinsic value | σ² (volatility) ∝ 1/β |
| 4 | Sinkhorn OT | Kantorovich OT | ε (regularisation) = 1/β |
| 5 | Forward-backward | Viterbi | 1/β (temperature) |
| 6 | Quantum group $SU(2)_k$ | Classical $SU(2)$ | $q = e^{i\pi\beta}$ |
In every case the bridge variable plays exactly the role that Planck’s constant plays in quantum mechanics. The ML engineer’s softmax temperature, the fluid dynamicist’s viscosity, and the financial mathematician’s volatility are all the same coordinate.
The negative real axis: population inversion
When β < 0 the Boltzmann weights invert — the highest-energy state receives the largest weight. This is a population inversion: the precise thermodynamic condition for stimulated emission in laser physics. The Ambient (β = 0) is the infinite-temperature boundary between absorption (β > 0) and emission (β < 0).
The complex plane: PT-symmetric quantum mechanics
Complex $\beta = \sigma + it$ interpolates between the thermal and quantum regimes. The exceptional points of PT-symmetric Hamiltonians — where two eigenvalues coalesce and the Hamiltonian becomes non-diagonalisable — are branch points of $Z_\beta$: loci where $Z_\beta = 0$ and $\frac{1}{\beta}\log Z_\beta$ has a branch cut. The Berry phase accumulated around such a branch point is the TWIST opcode of the Forge ISA.
The p-adic fibres: Ostrowski closes the map
Ostrowski’s theorem (1916): every non-trivial absolute value on ℚ is equivalent either to the standard Archimedean absolute value (giving ℝ) or to a p-adic absolute value (giving ℚ_p for some prime p). The adèlic β-plane $\mathbb{R} \times \prod_p \mathbb{Q}_p$ is therefore not a choice — it is the unique complete parameter space for the MGE.
The p-adic completions attach at β = 0 (the Ambient), not at β → ∞. The tropical limit is real-arithmetic, not p-adic. This distinction matters: an earlier version of this programme incorrectly identified the two, and Papers 543 and 201 (v16) explicitly correct it.
Palmer’s Invariant Set $I_U$ — the measure-zero fractal subset of Hilbert space on which the universe evolves (Palmer 2009) — has a natural adèlic interpretation: it is the adèlically integral slice, the set of states whose p-adic valuations are non-negative for every prime p. Cantor-like from the real perspective, but p-adic measure one.
4. The evidence
| Paper | What it establishes |
|---|---|
| #201 | MGE Dodecagon: 12 constructs as exact β limits; three geometric invariances; Turing completeness as corollary |
| #443 | Six classical dualities unified as β-deformations; conductor = 2/β for quantum group fields |
| #419 | Positive real axis in full: snap event β★, vorton, Carnot efficiency η = 0.1825 (FMO) |
| #454 | Imaginary axis: Meld ISA, Wick rotation, quantum algorithms |
| #543 | Full adèlic β-plane: negative β, PT-symmetry, p-adic, Ostrowski, Palmer $I_U$, Clifford = quantum tropical |
| #445 | Quantum groups at $q = e^{i\pi\beta}$; conductor = 2/β; Ising/Fibonacci anyons at specific β values |
5. What would falsify it
- A computational regime that cannot be expressed as the MGE at any point in the adèlic β-plane
- A non-trivial absolute value on ℚ not covered by Ostrowski’s theorem (impossible by the theorem itself, but a gap in the mapping would be damaging)
- Evidence that the T gate’s 2-adic non-integrality is not the correct explanation of the Clifford/magic boundary — e.g. a magic state with 2-adically integral entries, or a Clifford operation with non-integral entries
6. Open questions
- The obstruction-dodging theorem (Paper 543 §9): every H^k obstruction trivialised at level k+1 or a different completion. Is this a general theorem or does it fail for some obstructions?
- Negative β computing: Is population inversion the only computational use of β < 0, or does the Forge ISA at negative β describe a distinct computational regime (anti-annealing, uphill optimisation)?
- p-adic algorithms: The p-adic fibres of the β-plane give ultrametric computation. What algorithms are natural here that are hard on the real axis?
- The β-plane as moduli space: Is the adèlic β-plane the moduli space of all semiring deformations of the tropical semiring? If so, the MGE is the universal family over this moduli space.