What is a braid?
Three strands hanging vertically. When two cross, that's a braid generatorA single crossing of two adjacent strands. The atomic operation of braid theory. Every braid is built from sequences of generators.. The crossing IS the information — not the material the strands are made of, not their thickness, not their color. Just: which crossed over which.
Blue over red.
Red over green.
Over and under matter
These two look almost the same. They're not. In σ₁, blue goes over red. In σ₁¹, red goes over blue. Different braids. Different topologyThe study of properties preserved under continuous deformation — stretching, bending, twisting, but not cutting or gluing. ‘Rubber-sheet geometry.’. You cannot smoothly deform one into the other.
This is why topology is powerful: the crossing direction is a discrete, absolute fact that survives any continuous deformation. You can stretch, wiggle, or compress the strands — the over/under pattern doesn't change.
Braids compose
Stack crossings top to bottom. Two crossings make a braid word. Three make a longer word. The word encodes the complete braiding pattern.
Two critical braid words, both with two crossings, but with very different futures:
σ₁σ₂ — same direction
Both crossings go the same way (left strand over right). The strands rotate around each other.
σ₁σ₂⁻¹ — alternating
Crossings alternate direction (left over, then right over). The strands fold over each other.
Periodic vs. pseudo-Anosov
Extend each pattern. Watch what happens.
Periodic: σ₁σ₂σ₁σ₂σ₁σ₂
After 3 full cycles, the strands return to their starting positions. This braid repeats. TEPO = 0.
Golden: σ₁σ₂⁻¹σ₁σ₂⁻¹σ₁σ₂⁻¹
The strands never return. Each cycle stretches the pattern further. This is pseudo-AnosovA braid that stretches material exponentially with each iteration. The mathematical opposite of periodic (which returns to its starting state). Named after mathematician Dmitri Anosov.. TEPOTopological Entropy Per Operation. Measures how much a braid stretches space per crossing. TEPO = 0 means periodic. TEPO = log(φ) is the minimum for a genuinely complex 3-strand braid. = log(φ).
The periodic braid is like twisting a rope — it winds up but can unwind. The golden braid is like kneading dough — each fold builds on the previous one, and you can never unfold it back.
Where the golden ratio comes from
Imagine stretching taffy with three hooks. Each σ₁ folds the taffy one way. Each σ₂⁻¹ folds it the opposite way. Because the folds alternate direction, each new fold builds on the previous two folds, not just the last one.
Count the layers of taffy after each crossing:
| Crossings | Layers | Growth ratio |
|---|---|---|
| 0 | 1 | — |
| 1 | 1 | 1.000 |
| 2 | 2 | 2.000 |
| 3 | 3 | 1.500 |
| 4 | 5 | 1.667 |
| 5 | 8 | 1.600 |
| 6 | 13 | 1.625 |
| 7 | 21 | 1.615 |
| 8 | 34 | 1.619 → φ |
1, 1, 2, 3, 5, 8, 13, 21, 34 — the Fibonacci numbersThe sequence 1, 1, 2, 3, 5, 8, 13, 21... where each number is the sum of the previous two. They appear whenever a process reaches back two steps.. Each is the sum of the previous two. The ratio of consecutive Fibonacci numbers converges to φ = 1.618...
The golden ratio appears because alternating braid crossings create Fibonacci growth. Each fold reaches back two steps, satisfying F(n) = F(n-1) + F(n-2). The growth rate of that recursion is φ.
Watch φ appear
Each crossing alternates σ1, σ2−1. The braid never repeats, and the length of the tracked strand follows the Fibonacci recursion. The ratio of successive lengths is what converges to φ.
Press Play. The recursion has not started yet.
TEPO = log(φ) ≈ 0.481 is the minimum possible stretching rate for a non-repeating 3-strand braid. Less than log(φ) means the braid is periodic. At or above log(φ) means it's genuinely complex (pseudo-Anosov). The golden ratio is the exact boundary between simple and complex braiding.
Topological protection
Now the physical payoff. Look at this braid:
I'm deforming it. Wiggling the strands, stretching them, compressing them. The crossing pattern doesn't change. Blue still goes over red at the top. Red still goes over green at the bottom. The topology — the over/under sequence — survives any smooth deformation.
If you store information in that crossing pattern, the information is protected against local noiseInformation stored in topology survives any smooth deformation — stretching, wiggling, compressing. Only cutting (a drastic, non-local action) can change it.. Small perturbations can't change it. Thermal fluctuations can't change it. The only way to change the crossing pattern is to cut a strand — a drastic, non-local action.
This is Depth 1: Protection. The system has topological properties that resist local perturbation. Topological insulatorsMaterials that conduct electricity on their surface but not their interior, with the surface conduction protected by topology. Predicted 2005, observed 2007. work this way — their surface conduction is protected by topology, not by material purity.
Topological computation
If crossing patterns can encode information, can they encode computation?
Yes. Kitaev (1997) showed1 that braiding certain particles — anyonsParticles that exist only in 2D systems and obey neither Bose nor Fermi statistics. Braiding them around each other performs quantum logic gates. Named by Frank Wilczek. — around each other performs quantum logic gates. The computation is encoded in the braid pattern, and the braid pattern is topologically protected. Errors can't creep in because errors are local, and the computation is global.
This is Depth 2: Fault-tolerant computation. Not just protecting stored information, but protecting active computation. The system doesn't just resist noise — it thinks through it.
Not all anyons can do this. Ising anyonsAnyons at Chern-Simons level k=2. Too simple for universal quantum computation — they can only produce a limited set of quantum gates. (k=2) are too simple. Fibonacci anyonsThe simplest anyons that achieve computational universality. Their quantum dimension is φ. At Chern-Simons level k=3 in the SU(2) theory. (k=3) are the simplest that achieve computational universality. Their quantum dimensionA number characterizing how the Hilbert space grows when you add an anyon to the system. For Fibonacci anyons, d = φ. It determines the computational power. is φ — the same golden ratio, from the same braid group algebra.
Where φ appears
The golden ratio is the tracer for this skeleton. When φ shows up in a physical system, it means B₃ braiding is active. Four fields found it independently:
Active Nematics
Liquid crystal defects braid in confined geometries. The stretching rate of defect trajectories = log(φ). Measured computationally. Geometry selects the braid class.
Fibonacci Anyons
Quantum dimension of Fibonacci anyons = φ. From the fusion algebra of SU(2) at level k=3. The simplest non-abelian anyon with computational universality.
Quasicrystals
Long-to-short tile ratio in Penrose tilings = φ. From the algebraic structure of 5-fold symmetry, connected to the icosahedral group.
Cardiac (predicted)
If spiral wave tips in the heart braid according to B₃, the golden ratio should appear in their trajectories. Untested. This is an open falsification target.
Four fields. Same number. Same algebra underneath. No shared citations between the first three discoveries. Because φ either shows up or it doesn't, it gives this skeleton a relatively crisp falsification target. A caution, though: φ appears in many unrelated settings (growth patterns, continued fractions, aesthetics), so its presence is suggestive rather than proof — the stronger claim is that the golden ratio arises specifically from B₃ braid algebra, which is what the mechanism above predicts.
Is “topology” one skeleton?
Everything above treats topology as one thing: braids, crossings, invariants that survive deformation. That is the picture this page teaches, and for the systems it teaches it on, it holds. But a sweep across the twelve multi-skeleton systems in the matrix turned up a problem, and it belongs on this page more than any other, because this is the page that makes the claim.
The systems scored as “topological” are not all topological in the same sense. There are at least four different classifying invariants involved.
| Invariant | Systems | What it classifies |
|---|---|---|
| Mermin homotopy defectsDefects in an ordered medium are classified by the homotopy groups of the order-parameter manifold G/H. Mermin, Rev. Mod. Phys. 51, 591 (1979). πn(G/H) |
Nematics, cardiac, BZ, tokamak, brain, development | Defects of a broken symmetry |
| Topological orderPhases distinguished not by a local order parameter but by ground-state degeneracy on a surface and by the braiding statistics of their excitations. No symmetry is broken. modular tensor category |
FQHE, anyons | Ground-state degeneracy and anyon statistics. No symmetry broken |
| Knot and link invariants of a 1-D embedding |
Protein folding | Entanglement of a chain in 3-space |
| Cut-and-projectA quasicrystal is a periodic lattice in a higher dimension, sliced at an irrational angle. The invariant is the shape of the window used to take the slice. from a higher-dimensional lattice |
Quasicrystals | Aperiodic order. More symmetry broken than a crystal |
These do not classify the same objects and do not agree on when two systems are equivalent. A Mermin defect requires a broken symmetry; topological order requires that none be broken and that no local order parameter exist. A knot invariant is indifferent to whether anything ordered at all. This series currently records the first-versus-second split as a depth — a score of 1 against a score of 2 — which says they are the same kind of thing in different amounts. It is a difference of kind, recorded as a difference of degree.
So either these four are shown to be aspects of one classification, or “topology” is not one skeleton and the S1 column is at least two columns wearing one label. That is unresolved. It is argued in full, with the sweep it came from and four animations of the four invariants, on Broken Symmetry and the S1×S2 Edge — which also reports that where topology on this list is shallow, it is generally the defect classification of a broken symmetry rather than an independent structure.
Seven communities, one mathematics
Like criticality (FACETSA cross-domain survey cataloguing how the mathematics of criticality was independently developed across roughly a dozen scientific fields. Preprint, arXiv:2601.22389, with R. Macomber — under review, not yet peer-reviewed.), topological mathematics was independently rediscovered by communities that don't read each other's journals:
Knot theory, braid groups, Jones polynomialA knot invariant discovered by Vaughan Jones (1984) that connects knot theory to statistical mechanics. Fields Medal work that bridged pure math and physics.. Pure abstraction.
Fractional quantum Hall effect. Found topological order in a 2D electron gas.
Topological quantum field theory. String theory. Connected topology to physics axiomatically.
Topological quantum computation. Realized braiding = fault-tolerant quantum gates.
Topological insulators.2 A new class of materials with topologically protected surface states.
Topological defects govern dynamics in liquid crystals, biological tissues, bacterial colonies.
The mathematicians didn't teach the physicists. The condensed matter people didn't teach the computer scientists. The active matter people came at it from biology. Seven approaches to the same mathematics, arriving from seven directions.
Recap
- Skeleton
- Topological / braided mathematics — braid groups, knot invariants, modular categories.
- Depth 1: Protection
- Properties that survive any local perturbation. Information stored in topology is physically indestructible short of a drastic global change.
- Depth 2: Computation
- Using topological protection to compute fault-tolerantly. Braiding = logic gates. Errors can't creep in because errors are local and the computation is global.
- Tracer: φ = 1.618...
- The golden ratio. Signature of B₃ braid algebra. Appears as TEPO, quantum dimension, or tile ratio. A relatively crisp check — but because φ also arises in unrelated settings, its presence is suggestive rather than conclusive on its own.
- Key insight
- Alternating braid crossings create Fibonacci growth. The growth rate is φ. This is why the golden ratio appears in every system where three-strand braiding is active.
- Convergence
- 7 communities independently discovered this mathematics. None cited the others. This is a FACETS-scale convergence story, undocumented as such.