The ASA Framework

A map of how the pieces fit together.


The ASA grew in layers, each one built on the last. They are not synonyms.

Layer Name What it is Origin
1 MGE The core operation: Maslov-Gibbs Einsum Paper 201
2 TRS The β-deformation framework built on MGE Paper 202
3 ISA trilogy Three compilers, one for each temperature Papers 258, 419, 454
4 H^k ladder Complexity and cohomology classification Papers 396–398, 420
5 ASA The full programme: all of the above plus all applications Paper 219

Layer 1 — MGE (Maslov-Gibbs Einsum)

The single operation at the heart of everything:

\[\pi_k(\beta) = \frac{e^{-\beta E_k}}{\sum_j e^{-\beta E_j}}\]

The domain of β is the full adèlic β-plane — real, imaginary, negative, complex, and p-adic values each give a different computational regime:

β Regime What it computes
0 The Ambient Uniform weights; smooth Hodge theory
real, finite Forge ISA Gibbs distribution; statistical mechanics
real → ∞ Origami ISA Tropical (min, +) semiring; hard discrete selection
imaginary ($it$) Meld ISA Quantum amplitudes; Wick rotation
negative Population inversion Lasers; anti-Boltzmann weighting
p-adic Ultrametric p-adic computation; attaches at β = 0

Ostrowski’s theorem closes the map: every non-trivial absolute value on ℚ is either Archimedean (giving ℝ) or p-adic (giving ℚ_p), so the adèlic β-plane is the complete parameter space for the MGE. The MGE is what makes any discrete model differentiable: replace argmin with this, and gradients flow everywhere.

→ Papers 201 (MGE), 543 (adèlic β-plane)


Layer 2 — TRS (Topological Resonance Synthesis)

TRS is the β-deformation framework built on MGE. The name describes the underlying picture: Lie groups serve as the tape of a generalised Turing machine. The Chladni resonance patterns on the group manifold encode the computational state; the topology of those nodal lines — their H⁰/H¹/H² skeleton — is the computation. The MGE makes that skeleton differentiable.

  • Topological — the nodal-line topology of Chladni eigenmodes on the group manifold
  • Resonance — the Chladni eigenmodes themselves; also Wigner-Racah angular momentum resonances
  • Synthesis — the compiler that synthesises across classical/statistical/quantum via β

TRS is not the same as the ISA trilogy, not the same as the H^k ladder, and not the same as the ASA. It is the specific β-deformation layer.

Paper 202


Layer 3 — The ISA Trilogy

Four temperature regimes, three operative ISAs:

ISA β Arithmetic Paper
Ambient 0 Uniform / Hodge theory 417
Origami β → ∞ Tropical (min, +) 258
Forge 0 < β < ∞ Real Gibbs 419
Meld β = it Complex amplitudes 454

Each operative ISA (Origami, Forge, Meld) uses the same five opcodes — SPLIT, SPLAT, FLIP, FLOP, TWIST — running at a different temperature. The β → it substitution (Wick rotation) turns the Forge ISA into the Meld ISA and turns statistical mechanics into quantum mechanics.

The ISAs are not the same as the H^k ladder (Layer 4). The H^k ladder classifies problem type by the cohomological obstruction that makes it hard; any ISA can in principle compute any H^k problem, but at different cost. The Origami ISA naturally handles H⁰ problems; the Forge ISA excels at H¹; the Meld ISA is required for H² (quantum) problems. But the correspondence is one of natural fit, not restriction.

Eight independent mathematical communities have each, working separately, been forced to the same five opcodes. Paper 455 explains why: Shum’s theorem (1994) identifies the free ribbon pivotal category on one self-dual object, whose generators are forced by the topology of framed tangles.

→ Papers 258, 419, 454, 455

The chain complex: H^k is genuine cohomology

The H^k tiers are not merely a grading — they form a genuine chain complex. The boundary map ∂: C^k → C^{k+1} is assembled from the SPLIT and SPLAT opcodes with Koszul signs; it satisfies ∂² = 0 as a direct consequence of the Frobenius algebra axiom (SPLAT∘SPLIT = id). This is the same condition as the Pentagon identity, the MIP*=RE self-test, and Khovanov’s categorification of the Jones polynomial. The ORBIT count is the Euler characteristic of the complex; the full Poincaré polynomial is a strictly stronger invariant. At the H² level, BIND = the Kuperberg G₂ spider vertex (CMP 1996) — the complete diagrammatic axiomatisation of non-Abelian holonomy in the ISA.

→ Papers 357 (Pentagon/MIP), 571 (chain complex), 572 (G₂ spider = BIND)*

The 731-ISA: beyond the Pentagon

The three-ISA trilogy operates within the associative world — the Pentagon identity SPLAT∘SPLIT = 0 holds throughout. There is a fourth ISA that breaks this:

ISA Adds Breaks Algebraic home
Origami / Forge / Meld ℝ, ℂ (associative)
731-ISA BIND, SPIN Pentagon identity 𝕆 (octonions, non-associative)

BIND is the frog vertex — the non-abelian fusion that requires G₂ symmetry and implements the Fano associator obstruction. At j=½ it is the T-gate; at higher spin it accesses genuinely non-associative territory inaccessible to any standard quantum gate set. SPIN is the G₂ triality automorphism — the order-3 outer automorphism of Spin(8) that cyclically permutes its three 8-dimensional representations — and is the elementary injection gate for associamancy (the fourth level of the quantum resource hierarchy; see Paper 407).

A gate set is triality-complete if it contains SPIN. The 731-ISA is triality-complete; the Origami/Forge/Meld trilogy is not.

The 731-ISA is the computational realisation of the bottom-right cell of the Freudenthal-Tits magic square — the 4×4 table of exceptional Lie algebras built from pairs of division algebras (ℝ, ℂ, ℍ, 𝕆). The diagonal of that table is the division algebra ladder expressed as Lie algebras: G₂, F₄, E₆, E₈. The full non-associative frontier is described in the Non-Associative Frontier explainer.

→ Papers 207, 258, 263, 407, 405


Layer 4 — The H^k Complexity Ladder

The ISAs compute things whose complexity is classified by sheaf cohomology:

Level Symbol Name Computation Complexity
0 H⁰ Islands Classical / bilateral PSPACE
1 Flatland Statistical / triangular BPP
2 The Deep Quantum / tetrahedral BQP

H⁰ = problems solvable by looking at each part independently. H¹ = problems requiring triangular (three-party) consistency. H² = problems requiring tetrahedral (four-party) coherence — the level at which quantum algorithms live and at which the 2008 financial crisis became unresolvable.

The snap threshold β* = (3/8)ln(1/(1−ρ)) marks the transition between H⁰ and H¹ regimes for any network with edge density ρ.

→ Papers 396, 397, 420


Layer 5 — The ASA (Adelic Simplicial Architecture)

The full research programme: all five layers plus their applications across quantum computing, nuclear and molecular spectroscopy, financial risk, climate economics, and pure mathematics. Organised into five portfolios:

Portfolio Theme
A — Core Engine MGE, TRS, non-associative calculus
B — Foundations Algebra, topology, category theory
C — Hardware & AI ISA trilogy, registers, QEC
E — Chemistry & Physics Bonding theory, nuclear, amplituhedron
F — Quantum Foundations Magic, self-tests, paradoxes
G — Finance & Economics EconIAC, gauge theory, risk

The adelic structure — using all completions of ℚ simultaneously (real + p-adic) — is captured by the adèlic β-plane (Paper 543): the real axis governs the Origami/Forge/Meld ISA trilogy, while the p-adic completions attach at β = 0 (the Ambient). Ostrowski’s theorem guarantees this is the complete picture.

Paper 219 (An Adelic Invitation)


Where EconIAC fits

EconIAC is the Portfolio G application of the ASA. It uses all five layers:

  • MGE — Gibbs relaxation makes every argmax differentiable; β is the rationality temperature
  • TRS — the β-deformation of discrete economic models (Nash equilibria, order statistics, Leontief minimum) into smooth, calibratable functions
  • Forge ISA — the compiler for statistical/probabilistic economic computation at finite β
  • H^k ladder — H⁰ = bilateral risk (Basel), H¹ = triangular risk (XVA, convexity), H² = systemic risk (2008 crisis, unresolvable cascades)
  • ASA — the full framework that places the Keynesian multiplier, FMO photosynthesis efficiency, and Shor’s algorithm on the same ladder

Reading order

New to the framework? Start here:

  1. The Maslov-Gibbs Einsum (Paper 201) — the single equation; the dodecagon of twelve unified constructs; Turing completeness as a corollary
  2. Planck’s Constant in Disguise (Paper 443) — the six classical dualities as β-deformations; no prerequisites
  3. The Adèlic β-Plane (Paper 543) — the complete parameter space: why ℏ, viscosity, volatility, and quantum amplitudes are all the same coordinate
  4. In Praise of Tetrahedra (Paper 386) — the geometric seed; accessible to anyone
  5. Eight Derivations (Paper 455) — why the five opcodes are inevitable

Then follow the Start Here routing page for your field.