The ISA Family
All named ISAs share the same five opcodes (LABEL 🏷️ / ORBIT 🔄 / TWIST 🌀 / BIND 💎 / FLIP 👁️) — they differ only in the value of the inverse temperature β and the arithmetic they run over.
| ISA | β location | In one phrase | Paper |
|---|---|---|---|
| Origami | all β (umbrella) | Five-opcode open standard; tropical at β→∞, quantum at β=it | 631 |
| Forge | 0 < β < ∞ (real Gibbs) | Free-energy routing; MGE soft threshold; snap at β* | 419 |
| Meld | β = it (imaginary) | Complex amplitudes; full quantum mechanics | 454 |
| Raven | β ≈ β* (physiological) | Biological proofreading; enzyme catalysis; kinetic QEC | Raven |
| Motive | abstract parent | Five primitive opcodes; ERASE = second law | Motive |
| Hum | β = it/ℏ (QFT) | QFT vacuum; EMIT opcode; amplituhedron as ORBIT | 620 |
| Pentagon | coherence theorem | Monoidal coherence; five sides = five opcodes | 622 |
| Rising Sea | full ℂ_β plane | Every ISA as a fibre over the β-plane | 621 |
Full opcode reference: The ISA Opcodes · β-plane geometry: Forge & Meld · Non-associative frontier (BIND at 𝕆-rung): 731-ISA
The Five Pillars
Everything else in the framework is an application of these five ideas.
1. β is a coordinate
Classical logic, statistical mechanics, and quantum mechanics are not three separate theories. They are the same theory evaluated at different values of a single parameter β — the inverse temperature.
| β | Arithmetic | Physics |
|---|---|---|
| β → ∞ | (max,+) tropical | Classical logic, discrete optimisation |
| 0 < β < ∞ | Gibbs / ℝ | Statistical mechanics, annealing |
| β = it/ℏ | Unitary / ℂ | Quantum interference, Feynman path integral |
| β = complex | Fibred | Full β-plane; p-adic places at each prime |
The Maslov–Gibbs Einsum (MGE) is the operation that makes β a differentiable coordinate: discrete combinatorial models become smooth functions of β, and the snap at β* is a genuine phase transition — not a metaphor. Planck’s constant, viscosity, volatility, softmax temperature, and the quantum-group deformation parameter q = e^{iπβ} are all the same object seen from different fields.
Key papers: 201 (MGE), 443 (Planck in disguise), 454 (Meld / β=it), 543 (β-plane), 597 (soft thresholds)
2. The H^k stratification
Every computation has a cohomological address. The three tiers are not a taxonomy — they are a theorem: the Pentagon identity d² = 0 forces exactly this structure.
| Tier | Cohomology | Opcodes | Character |
|---|---|---|---|
| H⁰ | Classical / bilateral | LABEL, ORBIT, FLIP | Symmetry sectors, discrete orbits |
| H¹ | Gauge / triangular | + TWIST | Phase accumulation, convexity, stabiliser QC |
| H² | Systemic / entangled | + BIND | Entanglement, correlation, fault-tolerant QC |
The triangle and the tetrahedron. H⁰ is a closed triangle: three nodes, three edges, everything balances bilaterally. H¹ is what happens when you try to fill that triangle with a consistent phase and find you can’t — there is a Berry phase, a convexity correction, a residual that won’t cancel. H² is four triangles glued into a tetrahedron: the interior is genuinely non-trivial, and evaluating it requires the 6j symbol. The Pachner move that replaces one tetrahedron with four — and the identity that says both give the same amplitude — is the Pentagon identity, and it is the same equation as d² = 0.
In finance. Three banks A→B→C→A form an H¹ cycle: bilateral netting cannot dissolve it, but it is not yet catastrophic. The 2008 crisis arrived when those H¹ cycles became globally inconsistent — H² ≠ 0 — and no bilateral deal could fix it. Only a systemic intervention (central bank, not counterparty) operates at H². The Pentagon identity is the no-arbitrage condition; its failure is a financial crisis.
In machine learning. Each attention head is an H⁰ orbit. Multi-head attention accumulates H¹ phase across heads. The grokking phenomenon — where a network suddenly generalises after a long plateau — is an H² snap: the loss landscape crosses β* and the circuit topology discretely changes. LLM temperature is literally β⁻¹; sampling at T > 0 is Gibbs sampling at finite β.
The Weyl chamber and magic. The same cohomological boundary that separates classically simulable (H¹, Clifford) from universal (H², magic) quantum circuits also separates DFT-tractable from DFT-failing molecules. The Weyl chamber stratification and the Grassmannian angle θ_G are the same geometric object seen from quantum computing and chemistry respectively. This is not analogy: Paper 595 proves they carry the same Bredon H² class (Euler characteristic 2).
The Pentagon identity (d² = 0) is simultaneously: the HJM no-arbitrage condition · the Biedenharn–Elliott identity · the MIP* verifier constraint · the H² = 0 financial stability condition. One equation, four theorems.
Key papers: 420 (H^k complexity ladder), 421, 469 (ISA completeness), 470 (hot logic), 472 (Shor lifting), 595 (Weyl chamber homology), 596 (Weyl–DFT accelerator)
3. The five opcodes are universal
Every computation — at any β, in any domain — decomposes into five operations. This is not a design choice; it follows from the structure of Čech cohomology on a sheaf.
| Opcode | Symbol | Cohomological role |
|---|---|---|
| LABEL 🏷️ | ⊢ | δ⁻¹: assign a symmetry sector |
| ORBIT 🔄 | 𝒪 | H⁰: enumerate group orbits |
| FLIP 👁️ | ⌁ | Sheaf dualisation / time-reversal |
| TWIST 🌀 | ∮ | H¹: gauge transformation / phase |
| BIND 💎 | ⋈ | H²: Pachner surgery / entanglement |
The same five opcodes describe a ribosome, Shor’s algorithm, a yield curve, and an enzyme — at twenty orders of magnitude in physical scale. This is not analogy: the 6j symbol is H¹ of the representation sheaf in every case.
Key papers: 258, 370, 455 (eight derivations), 631 (Origami open ISA manifesto)
4. The Fano crystal is the universal phase detector
Whether an orbit is closed (H² = 0, stable) or open (H² ≠ 0, unstable) is the single binary that governs:
- Photosynthetic efficiency (FMO: closed orbit = η = 0.1828 Carnot bound)
- Financial contagion (systemic risk: open H² orbit = cascade)
- Quark confinement (QCD: open orbit = confined)
- Quantum error correction (closed orbit = code space preserved)
- Enzyme catalysis (G-step: orbit closure = reaction proceeds)
The Fano plane (7 points, 7 lines, the smallest projective plane) is the minimal structure that realises H² non-trivially. The β* snap — the phase transition at the critical inverse temperature — is the moment an orbit closes. It appears at the same Grassmannian angle θ_G ≈ 20° across all four domains.
Key papers: 317, 325, 357 (MIP* = RE), 563, 570, 595, 596, 602
5. Four-body interactions are irreducible
Every colour you have ever seen — the blue of the sky, the green of a leaf, the orange of a flame, the entire visible spectrum of every star — is ultimately governed by spectroscopy. And spectroscopy has been understood since the 1930s to rest on a single irreducible mathematical object: the 6j symbol, the recoupling coefficient for three angular momenta combining into a fourth. Wigner, Racah, and Weyl worked this out before the Second World War. The 6j is not a 2-body interaction dressed up in notation. It is genuinely, irreducibly a 4-body object — four particle lines meeting at a vertex — and no combination of pairwise forces can reproduce it.
This surprises most people. We are trained to think that nature is pairwise at heart: Newton’s gravity, Coulomb’s law, the bilateral loan, the two-qubit gate. But the 6j symbol has been sitting in the spectroscopy textbooks for ninety years, quietly ruling out that picture for any system with non-Abelian symmetry.
BIND is that object. In the Origami ISA, BIND is the opcode that encodes the non-trivial associator — the F-matrix of a fusion category, the octonion associator, the 6j recoupling coefficient. It appears identically in:
| System | BIND incarnation | Known since |
|---|---|---|
| Atomic / molecular spectroscopy | Racah 6j symbol; every spectral line | Racah 1942 |
| Nuclear physics | Tensor force $S_{12}$; mandatory in every nucleus from the deuteron up | Wigner 1933 |
| Topological quantum computing | Non-Abelian anyon F-matrix; fault-tolerant braiding | Kitaev 2003 |
| Systemic financial risk | H² snap event; 2008 global financial crisis | — |
| Nitrogen fixation (FeMoco) | 4-Majorana coupling in the iron-sulfur cluster | — |
| Non-Abelian anyons | Pentagon equation for fusion categories | Moore–Seiberg 1989 |
BIND cannot be built from TWIST or FLIP applied twice. It is not a 2-body interaction in disguise. The reason DFT fails for strongly-correlated molecules, the reason Clifford circuits cannot achieve universal quantum computation, and the reason the 2008 crisis could not be unwound bilaterally are all the same theorem: you hit the H¹→H² boundary and BIND is on the other side.
The irreducibility of 4-body interactions is not a new discovery. It is a ninety-year-old fact from spectroscopy that has not yet been fully absorbed by the rest of science. Thermyon names it, and makes it computable across every domain where it appears.
Key papers: 258, 357, 447, 455, 571, 572 (ISA chain complex and G₂ spider)
The ISA Family
All named ISAs share the same five opcodes (LABEL 🏷️ / ORBIT 🔄 / TWIST 🌀 / BIND 💎 / FLIP 👁️) — they differ only in the value of the inverse temperature β and the arithmetic they run over.
| ISA | β location | In one phrase | Paper |
|---|---|---|---|
| Origami | all β (umbrella) | Five-opcode open standard; tropical at β→∞, quantum at β=it | 631 |
| Forge | 0 < β < ∞ (real Gibbs) | Free-energy routing; MGE soft threshold; snap at β* | 419 |
| Meld | β = it (imaginary) | Complex amplitudes; full quantum mechanics | 454 |
| Raven | β ≈ β* (physiological) | Biological proofreading; enzyme catalysis; kinetic QEC | Raven |
| Motive | abstract parent | Five primitive opcodes; ERASE = second law | Motive |
| Hum | β = it/ℏ (QFT) | QFT vacuum; EMIT opcode; amplituhedron as ORBIT | 620 |
| Pentagon | coherence theorem | Monoidal coherence; five sides = five opcodes | 622 |
| Rising Sea | full ℂ_β plane | Every ISA as a fibre over the β-plane | 621 |
Full opcode reference: The ISA Opcodes · β-plane geometry: Forge & Meld · Non-associative frontier (BIND at 𝕆-rung): 731-ISA