REVIEW
No-go theorems for dual Hamiltonians of Non-Abelian Lattice Gauge Theories
High Energy Physics - Lattice · Submitted 30 Sep 2026 · 39 references
Pith's verdict: Sound no-go results for the static-constraint rotor class and the natural plaquette constraint, attached to an abstract that reaches past what the proofs actually establish.
T0 · AI review
Why Pith says this
Highlight any text to comment on it.
Referee report
The paper
The paper claims
Non-Abelian dual gauge Hamiltonians cannot be made local: a no-go proof.
The reading
What carries the argument
The central object is the maximal tree of helper links that any static-constraint reparametrization must leave behind, together with the tree distance $d_T(x,y)$ between lattice sites. The tree-distance hierarchy $d_T \ge d_1 \ge d_\infty$ is used to show that a maximally balanced partition of the tree (root $r$, sets $A$ and $B$) forces a vertex on the boundary of $B$ whose distance to the root is Ω(ℓ), which in turn is a neighbor of a vertex in $A$; the electric term connecting them must be a path of Ω(ℓ) operators on the tree. For the dynamic case the key object is the parallel-transported plaquette Lie derivative $\hat{C}^c_{\rm plaq} = -\hat{R}^c_1 + \hat{L}^c_2$, whose commutator $[\hat{C}^a_{\rm plaq}, \hat{C}^b_{\rm plaq}] = if^{abc}(\hat{R}^c_1 + \hat{L}^c_2)$ lies outside the constraint set, driving the proof that both $\hat{R}^c_1$ and $\hat{L}^c_2$ must annihilate the state.
Extended reading notes
Core claim
For a d-dimensional hypercubic lattice with d>1, any static-constraint dualization — introducing a rotor for every plaquette and removing the same number of links — forces the helper links into a maximal tree (Lemma III.1), and every spanning tree contains a pair of previously neighboring sites whose tree distance is Ω(ℓ) (Theorem 1, with Corollary III.2 extending this to a constant number of extra links). Since the electric interaction between those two sites must be routed along the tree, the dual Hamiltonian necessarily contains a many-point electric string of length growing linearly in ℓ. The Abelian U(1) case avoids this because Gauss' law reduces to a linear system that eliminates the strings, but the non-Abelian Gauss law is nonlinear and cannot. In d≥3, the static-constraint scheme fails even more completely: there are more plaquettes than removable links, so all rotors cannot be introduced at all. The dynamic-constraint route (plaquette separation) reproduces the U(1) rotor Hamiltonian, but in SU(N) the constraint operators — built from parallel-transported plaquette Lie derivatives — do not close into an algebra, and Theorem 2 shows the only wavefunction satisfying them is constant, so the ansatz overconstrains the system.
Load-bearing premise
The no-go theorems apply to reparametrizations in which a rotor is introduced for every plaquette and the same number of links is removed, forcing the helper links to form a maximal tree (Section II.C and Lemma III.1); the paper does not prove that every possible canonical dual formulation with a free magnetic part and preserved degrees of freedom must take this exact form, so the abstract's sweeping claim rests on identifying all dualizations with this static-constraint class.
Editorial extensions
If this is right
- Quantum and tensor-network simulations of non-Abelian lattice gauge theories at weak coupling, if built on static-constraint rotor dualizations, inherit electric terms with operator strings of length Ω(ℓ), inflating tensor-network bond dimensions and quantum circuit depth at least linearly in the lattice extent.
- In d≥3 there are more plaquettes than removable links for both open and periodic boundary conditions, so the all-rotor static-constraint program cannot even be formulated while preserving the number of degrees of freedom (Lemma III.1c).
- U(1) is the exceptional case: its Gauss law is linear and exactly solvable in the momentum basis, which is why the rotor formulation (eq. 21 and eq. 47) has a free magnetic part and only 1-local electric terms.
- The plaquette-separation ansatz, which works in U(1), is inconsistent in SU(N): the plaquette constraint operators do not form an algebra, and the only allowed state is constant, so a dynamic constraint of this form cannot reduce the Hilbert space correctly.
- Any future candidate dynamic constraint must generate a closed algebra on the physical subspace; the authors explicitly leave open whether such a constraint exists.
Reading between the lines
- The no-go may extend beyond maximal-tree static constraints: if locality of the electric term is incompatible with any exactly gauge-fixing reparametrization that keeps the magnetic part single-site, then weak-coupling simulations must either accept polynomially long electric strings or work in an enlarged Hilbert space, which strengthens the case for hybrid digitization and for quantum link model
- Theorem 2 suggests a testable criterion for future dual formulations: a valid dynamic constraint must close into an algebra on the constrained subspace, and any candidate whose commutator falls outside the constraint set is inconsistent for a non-Abelian group.
- The rotor-versus-locality trade-off of Section III.B, which bounds non-locality by $mn+1$ using $m \times n$ cells, could be probed numerically at intermediate coupling to see whether partial rotor introduction yields a workable compromise despite the many-point plaquette terms it reintroduces.
Original abstract
Hamiltonian simulations of lattice gauge theories promise new insights into the inner workings of QCD. The rapid development of quantum computing hardware suggests that large scale simulations may soon become feasible. However, no efficient formulation suitable for simulations near the continuum limit is currently known for the non-Abelian case, and existing candidates typically feature long range interactions in the electric part of the Hamiltonian. We prove that such non- localities are unavoidable when the gauge configuration is reparametrised while preserving both the number of degrees of freedom and the fundamental commutation relations. We then attempt to mitigate this by introducing additional gauge links alongside additional constraints. Although this approach is successful in U(1), we prove that the corresponding construction is inconsistent for non-Abelian lattice gauge theories.
The compiled record
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
The abstract, sentence by sentence
Hamiltonian simulations of lattice gauge theories promise new insights into the inner workings of QCD. The rapid development of quantum computing hardware suggests that large scale simulations may soon become feasible. However, no efficient formulation suitable for simulations near the continuum limit is currently known for the non-Abelian case, and existing candidates typically feature long range interactions in the electric part of the Hamiltonian. We prove that such non- localities are unavoidable when the gauge configuration is reparametrised while preserving both the number of degrees of freedom and the fundamental commutation relations. We then attempt to mitigate this by introducing additional gauge links alongside additional constraints. Although this approach is successful in U(1), we prove that the corresponding construction is inconsistent for non-Abelian lattice gauge theories.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Hamiltonian simulations of lattice gauge theories promise new insights into the inner workings of QCD. | background 1.00 | 0.04 | 0.02 | hedged possibility 0.58 |
| s1 | The rapid development of quantum computing hardware suggests that large scale simulations may soon become feasible. | background 1.00 | 0.03 | 0.02 | hedged possibility 1.00 |
| s2 | However, no efficient formulation suitable for simulations near the continuum limit is currently known for the non-Abelian case, and existing candidates typically feature long range interactions in the electric part of the Hamiltonian. | gap 0.97 | 0.06 | 0.08 | asserted 1.00 |
| s3 | We prove that such non- localities are unavoidable when the gauge configuration is reparametrised while preserving both the number of degrees of freedom and the fundamental commutation relations. | result 0.96 | 0.94 | 0.76 | asserted 1.00 |
| s4 | We then attempt to mitigate this by introducing additional gauge links alongside additional constraints. | method 0.66 | 0.91 | 0.04 | asserted 0.76 |
| s5 | Although this approach is successful in U(1), we prove that the corresponding construction is inconsistent for non-Abelian lattice gauge theories. | result 0.92 | 0.93 | 0.41 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
Figures
Figures reproduced from arXiv:2609.39780 by the authors.
References
Discussion
Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.