Absence of twisting for non-trivial discrete torsion
Primož Moravec · 2025
Read Absence of twisting for non-trivial discrete torsion by Primož Moravec (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-17068.txt.
Introduction
Primoz Moravec studies discrete torsion for the n-torus with a finite symmetry group, from the Dijkgraaf-Witten angle, where a cohomology class assigns a phase to each flat bundle. He asks which classes give trivial phases on all commuting tuples, the untwisted ones.
Those classes form a subgroup, written Br to the n of G, inside the full cohomology group. Moravec derives an exact sequence, using universal coefficients, that places the new invariant. In degree two it returns the Bogomolov multiplier, also known as the unramified Brauer group. Algorithms compute the subgroup and torus partition functions, with runs reported for families of finite subgroups of SU(4).
Catalog record
| CREATOR | Primož Moravec |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-17068
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Absence of twisting for non-trivial discrete torsion Thanks: The author was partially supported by ARIS, Grant numbers J1-50001, P1-0222. Primož Moravec Abstract . We study discrete torsion for the n n –torus with finite symmetry group G G from the Dijkgraaf–Witten viewpoint. A class in H n ( G , U ( 1 ) ) H^{n}(G,U(1)) assigns a phase to each flat G G –bundle, equivalently to each commuting n n –tuple in G G up to conjugation. We introduce the subgroup Utw n ( G ) ⊆ H n ( G , U ( 1 ) ) \Br^{n}(G)\subseteq H^{n}(G,U(1)) of untwisted classes, those whose Dijkgraaf–Witten phases are trivial on all commuting tuples, and derive a universal coefficient exact sequence involving this invariant. In degree 2 2 this recovers the Bogomolov multiplier / unramified Brauer group. We implement algorithms for computing Utw n ( G ) \Br^{n}(G) and corresponding torus partition functions, and report on computations for families of finite subgroups of SU ( 4 ) \SU(4) . 1. Introduction Discrete torsion is one of the fundamental tools used for describing string orbifolds. It was first discovered by Vafa [ 28 ] as a somewhat mysterious way of choices of phases corresponding to twisted vectors of string orbifold partition functions. These describe the effects of the orbifold group action on the B -field in string theory [ 1 ] , [ 26 ] . Discrete torsion also generalizes orbifold projections of D-brane probes at space-time singularities. In a series of papers, Sharpe clarified the mathematical background of discrete torsion, see [ 26 , 27 ] for overview. In a two-dimensional orbifold CFT with finite orbifold group G , discrete torsion is specified by a cohomology class [\omega]\in H^{2}(G,U(1)) . The latter group is also known as the Schur multiplier of G . A representative 2-cocycle \omega\in Z^{2}(G,U(1)) assigns a phase W_{\omega}(g,h) to each commuting pair (g,h) of group elements, which enters the one-loop (torus) orbifold partition function as (1.0.1) Z^{\text{orb}}_{\omega}(T^{2})\;=\;\frac{1}{|G|}\sum_{\begin{subarray}{c}g,h\in G\\ gh=hg\end{subarray}}W_{\omega}(g,h)\,Z_{g,h}(T^{2}), where Z_{g,h}(T^{2}) is the contribution from the (g,h) –twisted sector. For the standard discrete–torsion construction one takes (1.0.2) W_{\omega}(g,h)\;=\;\frac{\omega(g,h)}{\omega(h,g)}\,. Beyond its original CFT definition, discrete torsion admits a clean modern formulation. It is a choice of topological coupling for a finite symmetry group G , encoded by a class in H^{n}(G,U(1)) , and realized by coupling the theory to the corresponding Dijkgraaf–Witten (DW) topological gauge theory [ 7 ] . On a closed n –manifold M , this coupling contributes a phase obtained by evaluating f_{A}^{*}[\omega] on [M] , where f_{A}:M\to BG classifies the flat G –bundle A . A subtlety, which is easy to miss if one focuses only on the abstract group H^{n}(G,U(1)) , is that different cohomology classes can become physically indistinguishable on a given class of backgrounds. For the n –torus T^{n} , the DW weight W_{\omega}(\mathbf{g}) depends only on the evaluation of [\omega] on the fundamental class through the classifying map associated to a commuting n –tuple \mathbf{g} . This leads to a natural notion of untwisted discrete torsion. These are the cohomology classes whose associated phases are trivial on all commuting tuples. Such classes are invisible to the torus partition function, and more generally to any observable that factors through the DW holonomy on flat G –bundles on T^{n} . The central theme of this paper is that “invisibility on tori” has a robust group-theoretic meaning, and can be computed efficiently. In degree 2 it recovers the unramified Brauer group, also known as the Bogomolov multiplier [ 2 ] , and admits a dual description which makes large-scale computation feasible. In higher degrees we introduce and study the groups \Br^{n}(G) of untwisted n –dimensional discrete torsion, prove a universal coefficient sequence tailored to this setting, and explain how the obstruction to twisting can be read off from an explicit quotient H_{0n}(G,\mathbb{Z}) of H_{n}(G,\mathbb{Z}) . This provides a homological and computational bridge between discrete torsion phases and the topology of the moduli space of flat G –connections on T^{n} . We provide an implementation [ 23 ] of the constructions in publicly available package HAP [ 10 ] of the computer algebra system GAP [ 9 ] . On the physics side, discrete torsion is often invoked as a genuine deformation parameter of an orbifold, yet in concrete families the “available” torsion in H^{n}(G,U(1)) may fail to change torus amplitudes at all. Understanding when this happens clarifies which orbifolds admit genuinely distinct topological couplings and which ones exhibit a rigidity phenomenon. Our results supply a systematic diagnostic for when discrete torsion is physically effective. This is precisely the complement of \Br^{n}(G) , and it controls whether twisting is genuinely present in torus amplitudes. 2. Partition function of the n -torus Let \mathcal{T} be a d -dimensional QFT (or CFT) with finite global symmetry group G , and let [\omega]\in H^{d}(G,U(1)) encode a topological G -action (discrete torsion). For a closed oriented d -manifold M , a flat G -bundle A on M is equivalently a homomorphism f_{A}:\pi_{1}(M)\to G , or a map f_{A}:M\to BG up to homotopy. The Dijkgraaf–Witten topological gauge theory [ 7 ] with action \omega has partition function (2.0.1) Z_{\omega}^{\mathrm{DW}}(M)\;=\;\frac{1}{|G|}\sum_{[A]\ \mathrm{flat}}\exp\!\bigl(2\pi i\,S_{\omega}(A)\bigr)\;=\;\frac{1}{|G|}\sum_{[A]\ \mathrm{flat}}\bigl\langle f_{A}^{*}[\omega],[M]\bigr\rangle, where f_{A}^{*}:H^{d}(G,U(1))\to H^{d}(\pi_{1}(M),U(1)) is the map induced by f_{A} , and S_{\omega}(A) is the evaluation of f_{A}^{*}[\omega]\in H^{d}(M,U(1)) on the fundamental class [M] . The orbifold of \mathcal{T} by G with discrete torsion [\omega] is obtained by coupling \mathcal{T} to this topological DW gauge theory and summing over flat G -bundles. Its partition function on M is (2.0.2) Z^{\mathrm{orb}}_{\omega}(M)\;=\;\frac{1}{|G|}\sum_{[A]\ \mathrm{flat}}\bigl\langle f_{A}^{*}[\omega],[M]\bigr\rangle\,Z_{\mathcal{T}}(M;A), where Z_{\mathcal{T}}(M;A) denotes the partition function of the original theory \mathcal{T} on M in the background of the flat G -connection A . We now specialize to M=T^{n}=S^{1}\times\cdots\times S^{1} . A flat G –bundle on T^{n} is determined by its holonomies along the n circle factors, i.e. by a commuting n –tuple (2.0.3) \mathbf{g}=(g_{1},\dots,g_{n})\in G^{n},\qquad g_{i}g_{j}=g_{j}g_{i}. Denote X_{n}(G)=\{(g_{1},\dots,g_{n})\in G^{n}\mid g_{i}g_{j}=g_{j}g_{i}\text{ for all }i,j\}. For a commuting tuple \mathbf{g}=(g_{1},\dots,g_{n}) , let f_{\mathbf{g}}:T^{n}\to BG be the corresponding classifying map. The DW weight of this background is (2.0.4) W_{\omega}(\mathbf{g})\;:=\;\big\langle f_{\mathbf{g}}^{*}[\omega],[T^{n}]\big\rangle\;\in U(1), so that (2.0.5) Z^{\mathrm{DW}}_{\omega}(T^{n})\;=\;\frac{1}{|G|}\sum_{\mathbf{g}\in X_{n}(G)}W_{\omega}(\mathbf{g}) and (2.0.6) Z^{\mathrm{orb}}_{\omega}(T^{n})\;=\;\frac{1}{|G|}\sum_{\mathbf{g}\in X_{n}(G)}W_{\omega}(\mathbf{g})Z_{\mathcal{T}}(T^{n};\mathbf{g}). To obtain an explicit formula for W_{\omega}(\mathbf{g}) , represent T^{n} as the unit cube [0,1]^{n} with opposite faces identified, and equip it with the standard orientation \mathrm{d}x_{1}\wedge\cdots\wedge\mathrm{d}x_{n} . There is a canonical decomposition of the cube into n! n –simplices indexed by permutations \sigma\in S_{n} : (2.0.7) \Delta_{\sigma}\;=\;\big\{0\leq t_{1}\leq\cdots\leq t_{n}\leq 1\big\}\ni x\;=\;t_{1}e_{\sigma(1)}+\cdots+t_{n}e_{\sigma(n)}, where e_{i} are the standard basis vectors of \mathbb{R}^{n} . Each \Delta_{\sigma} inherits an orientation; its oriented volume form is \sgn(\sigma)\,\mathrm{d}x_{1}\wedge\cdots\wedge\mathrm{d}x_{n} , so the relative orientation of \Delta_{\sigma} with respect to T^{n} is \sgn(\sigma) . In the DW construction one chooses a gauge in which a distinguished vertex of the triangulation is labeled by the identity of G , and the edge along the i –th coordinate direction is labeled by the group element g_{i} . For an oriented n –simplex whose edges follow the ordered sequence of coordinate directions (\sigma(1),\dots,\sigma(n)) , the local DW weight is precisely (2.0.8) \omega\big(g_{\sigma(1)},\dots,g_{\sigma(n)}\big) (or its inverse, depending on orientation). Therefore, the total DW weight for the commuting n –tuple (g_{1},\dots,g_{n}) is given