← all systems · see it in The System
Dijkgraaf-Witten model (topological gauge theory with a finite gauge group; walked as an idea)
Topological gauge theories 'for the special case of a finite group' (Dijkgraaf and Witten 1989/1990): the only degree of freedom is the topology of a flat G-bundle; the action is a class alpha in H3(BG, U(1)) = H4(BG, Z); 'the path-integral is reduced to a finite sum', Z(M) = (1/|G|) x the sum over homomorphisms pi1(M) -> G of the weights. With alpha trivial it is untwisted discrete gauge theory (BF with a finite group; Kitaev's quantum double).
Properties published scienceour proposal
Sourced fields, and our own readings.
| aliases | Dijkgraaf-Witten model (BF with a finite group) |
| born at | June 1989 (IASSNS-HEP-89/33); published 1990 |
| born precision | a month |
| boundary | Inside: the theory as written and held: a finite gauge group G, a class alpha in H3(BG, U(1)), the finite sum over flat bundles, its lattice form, the state spaces and invariants it yields, its anyon algebra (the twisted quantum double), and its reformulations. Outside: the matter and devices it describes (topological orders, SPT phases, anyons, the surface code on processors), the scientists and literature that hold it, and the broader Chern-Simons, BF and TQFT families. |
| nested level | An idea of mathematical physics (rung 11), held by sys:science; the finite-group case of Chern-Simons theory and a twisted BF theory; Schwarz-type° TQFT (metric-free action; my reading of the TQFT article's two classes); realised as engineered states on quantum processors (rung 12) from 2021. |
| representation | True |
| scale | social |
Measures (13) published science
| preprint date | June 1989 (IASSNS-HEP-89/33, THU-89/9) date |
| publication | Commun. Math. Phys. 129(2), 393-429 (1990) citation |
| Z(S3) | 1/|G| dimensionless |
| Z(S2 x S1) | 1 dimensionless |
| Z(RP3), G = Z2, non-trivial cocycle | 0 (= (1/2)(1 + (-1))) dimensionless |
| Hilbert-space dimension bound | dim H(Sigma_g) <= |V_g| = |Hom(pi1(Sigma_g), G)/G| dimension |
| D4 order realisation: qubits | 27 qubits |
| D4 order realisation: fidelity per site | exceeding 98.4% (0.990 after readout correction) fidelity |
| D4 order: ground states on a torus | 22 states |
| toric code topological entanglement entropy | near the expected value of ln 2 nats |
| surface code distance on Willow (2025) | 7 code distance |
| dihedral group order in the Z2^3 duality | 8 elements |
| time from preprint to first processor realisation of the toric-code ground state | about 32 (June 1989 to the April 2021 preprint of Satzinger et al.) years |
Cross-scale patterns seen here (6) our proposal
Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Dijkgraaf-Witten model (topological gauge theory with a finite gauge group; walked as an idea). Each links to every other system that shows it.
- same role, different builds (structural)
One topological order, two constructions: a Z2 x Z2 x Z2 theory with a special 3-cocycle and an ordinary gauge theory with the dihedral group of order eight (dual, 1995); built in 2024 by gauging a Z2^3 SPT and identified as D(D4) order. - seed (partial)
A finite group plus one 3-cocycle unpacks a whole quantum field theory; group cohomology gives SPT phases "in any dimensions with any on-site symmetry groups", as group theory gives 230 crystal structures. - templating (partial)
The toric code is the simplest quantum double; the same (G, alpha) recipe runs from Z2 to D4 and across dimensions. - barrier-gates-complexity (surface)
Topological protection: computation by anyons "fault-tolerant by its physical nature"; below-threshold error suppression on Willow. - the whole sets the local reference (surface)
Every local patch is flat and the same; only the global topology (holonomy round non-contractible loops) carries information. - who opens the door (surface)
The model fixed the expected values ahead of time (ln 2 entropy; 22 ground states for D4); experiments arrived into those slots.
Valence: what it trades, and with whom (4) published science
Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.
- Science (as a human social and cultural system) via sys:written-record · Dijkgraaf-Witten model (topological gauge theory with a finite gauge group; walked as an idea) gain
- Chern-Simons gauge fields (levels k, -k) via sys:spacetime · Dijkgraaf-Witten model (topological gauge theory with a finite gauge group; walked as an idea) gain
- Conformal field theory via sys:mental-representation · Dijkgraaf-Witten model (topological gauge theory with a finite gauge group; walked as an idea) gain
- BF theory (topological gauge theory; walked as a projected model) via sys:spacetime · Dijkgraaf-Witten model (topological gauge theory with a finite gauge group; walked as an idea) gain
Partners and parts published science
Built from / builds (6)
- Holonomy (parallel transport round closed loops; its trace is the Wilson loop)
- Finite Hilbert space H_F (fermions)
- Triangulated 3-manifold
- Principal bundle
- Finite group (the gauge group)
- Group cohomology (the twist alpha)
Sits inside / holds (3)
- Science (as a human social and cultural system)
- Topological quantum field theory
- BF theory (topological gauge theory; walked as a projected model)
expressed by (7)
- Generation/Synthesis
- Recombination/Composition
- Scaffolding/Enablement (Meta-Tooling)
- Templating / Standardization
- Interoperability/Transduction (Bridging)
- Encapsulation & Modularity (Abstraction)
- Technique-Based Identification (a.k.a. Technique-Based Detection)
floor filled by (18)
- Science (as a human social and cultural system)
- Holonomy (parallel transport round closed loops; its trace is the Wilson loop)
- Chern-Simons gauge fields (levels k, -k)
- Finite Hilbert space H_F (fermions)
- Anyon
- Triangulated 3-manifold
- BF theory (topological gauge theory; walked as a projected model)
- Principal bundle
- Finite group (the gauge group)
- Group cohomology (the twist alpha)
- Twisted quantum double model
- Surface code
- Topological order
- Symmetry-protected topological phases
- Quantum processors that realise its states
- Robbert Dijkgraaf
- Edward Witten
- Superconductor (the superconducting state of matter)
represents (7)
- Anyon
- Superconductor (the superconducting state of matter)
- Surface code
- Topological order
- Symmetry-protected topological phases
- Quantum processors that realise its states
- the model itself is the idea; no rung 0-1 structure to represent (its physical readings are engineered, rung 12)
Kinds (5)
- Chern-Simons gauge fields (levels k, -k)
- BF theory (topological gauge theory; walked as a projected model)
- Kitaev quantum double model (untwisted)
- Twisted quantum double model
- Dijkgraaf-Witten theory in 3+1 dimensions
Notes published science
- In 1995 a thesis showed that a twisted gauge theory built from three copies of the two-element group gives the same particles as an ordinary gauge theory of the eight-element symmetry group of a square; in 2023-24 that exact equivalence was built on 27 trapped-ion qubits and used to make non-Abelian anyons.
- Dijkgraaf and Witten called these theories "somewhat esoteric" and "completely finite": the whole quantum path integral becomes a finite sum you can do by hand, and on a three-sphere it is just one over the size of the group.
- With the right twist, two possible histories cancel exactly, so the theory gives zero on one shape of space (RP3): the twist is a hidden phase that can make whole states vanish.
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
anyons of the twisted quantum double (projected model)
- L1: Excitations: magnetic vortices (fluxes), global charges and dyons, with topological Aharonov-Bohm interactions (residual discrete gauge theory)
- L2: Algebra: the quantum double D(H), deformed by the 3-cocycle into the quasi-Hopf twisted quantum double D^alpha(G); ground states of the Hamiltonian model characterised by it
- L3: Braiding and identity: braiding statistics characterise the order one to one; dualities (string-nets to twisted quantum doubles; twisted Z2^3 = D(D4))
- L4: Global read-out: degenerate ground states on a torus (22 for D4 order), topological entanglement entropy (ln 2 for the toric code)
finite-group gauge theory on a closed manifold (projected model)
- L1: Fields: a principal G-bundle over a closed oriented manifold M, G a finite group; every bundle flat, so the only freedom is its topology, seen as holonomy round non-contractible loops (a homomorphism pi1(M) -> G up to conjugation)
- L2: Action: a class alpha in H3(BG, U(1)) = H4(BG, Z) (torsion for a finite group); the weight W(gamma) = <gamma* alpha, [M]>; alpha trivial is the untwisted theory
- L3: Sum: Z(M) = (1/|G|) x the sum over Hom(pi1(M), G) of W(gamma); a finite sum, diffeomorphism-invariant; Z(S3) = 1/|G|, Z(S2 x S1) = 1, Z(RP3) = 0 for Z2 with the non-trivial class
- L4: Lattice form: a triangulation with ordered vertices; group elements on 1-simplices with vanishing lattice field strength (flat); the action a product of alpha over the oriented tetrahedra
- L5: Outputs: a Hilbert space for each surface, dim H = Z(Sigma x S1), at most the number of flat bundles (smaller when alpha is non-trivial: a global anomaly); gluing (sewing) axiom; three dimensions treated, other dimensions straightforward
the Dijkgraaf-Witten model as held by science (mental representation)
- L1: Papers: Dijkgraaf and Witten preprint (IASSNS-HEP-89/33, June 1989), Commun. Math. Phys. 129 (1990) 393-429; Dijkgraaf then a postdoc at the IAS alongside Witten
- L2: Motive and framing: classify Chern-Simons actions by H4(BG, Z); finite groups as the extreme case where the classification is the whole story; an elementary illustration of the functorial (Segal) axioms; "somewhat esoteric"
- L3: Reformulations: Freed and Quinn (finite sum, modular structure, any dimension); Hamiltonian twisted quantum double (Hu, Wan, Wu); continuum TQFTs in 2+1 and 3+1 D (Putrov, Wang, Yau); BF with twist terms
- L4: Uses in condensed matter: group-cohomology classification of bosonic SPT phases (H^(d+1)[G, U(1)]); gauging an SPT gives a twisted discrete gauge theory (Levin and Gu); residual discrete gauge theories after symmetry breaking (de Wild Propitius)
- L5: Tests on hardware: toric code ground state on a superconducting processor (2021, entropy near ln 2); D4 non-Abelian order from a gauged Z2^3 SPT on 27 trapped-ion qubits (2024, all 22 ground states); a distance-7 planar code on Willow (2025)