Nonperturbative Gauge Dynamics
Nonperturbative gauge dynamics studies phenomena that cannot be certified by a finite weak-coupling expansion: confinement, dynamically generated scales, strongly coupled spectra, chiral symmetry breaking, topological sectors, and infrared phases. “Strong-coupling QCD” is an important subset, not an alias for the field. This guide compares research programs and their evidence; it excludes canonical definitions, solver tutorials, and any claim that one confinement diagnostic is universal across matter content and global form.
Evidence cutoff. 11 August 2026.
Required background. Confinement definitions and their non-equivalence separates spectral, force-law, symmetry, and operational criteria; the non-Abelian mass gap states the actual continuum spectral question.
Helpful background. Confinement with dynamical quarks prevents importing pure-gauge order parameters into QCD, one-form symmetries organize exact line-operator diagnostics, lattice gauge ensembles connect regulated measurements to observables, and functional-method validation exposes closure and branch errors.
Infrared phases require more than one diagnostic
Section titled “Infrared phases require more than one diagnostic”| Program | Typical objects and observables | Strongest validation | Persistent limitation |
|---|---|---|---|
| Euclidean lattice gauge theory | Wilson loops, static energies, glueball and hadron spectra, condensates, topology | volume and lattice-spacing sequences; action and collaboration comparisons | sign problems and analytic continuation restrict density and real time |
| semiclassics and compactification | monopoles, bions, instanton constituents, transseries | calculable weak-coupling regimes linked by controlled deformations | adiabatic continuity to the target theory must be established, not assumed |
| Schwinger–Dyson and functional RG | propagators, vertices, effective potentials, running couplings | symmetry identities, regulator/truncation scans, benchmark models | hierarchy closure and gauge dependence can dominate |
| anomaly and generalized-symmetry constraints | line operators, anomalies, allowed infrared realizations | theorem-level matching and consistency across deformations | constraints usually do not select a unique phase or spectrum |
| Hamiltonian and tensor-network approaches | spectra, string breaking, real-time quenches | convergence with basis/bond dimension and Euclidean cross-checks | continuum and higher-dimensional scaling remain expensive |
Wilson’s strong-coupling lattice argument exhibits an area law in a controlled corner and founded the modern regulator program Wilson 1974, foundational. It is not a proof of the four-dimensional continuum Yang–Mills mass gap. Likewise, an infrared solution of a truncated functional hierarchy is evidence only after renormalization, symmetry identities, branch selection, and observable reconstruction are checked. A modern synthesis of continuum nonperturbative gauge-theory methods further shows why gauge-dependent intermediate quantities must be separated from physical observables Capri et al. 2022, synthesis.
Negative results sharpen the question
Section titled “Negative results sharpen the question”Elitzur’s theorem forbids spontaneous breaking of a local gauge redundancy; gauge-fixed condensates therefore cannot serve as ordinary gauge-invariant order parameters Elitzur 1975, obstruction. With fundamental Higgs matter, analyticity can connect regions conventionally called Higgs and confinement without a thermodynamic phase boundary Fradkin and Shenker 1979, no-go/qualification. With dynamical fundamental quarks, center one-form symmetry is explicitly broken, so an asymptotic Wilson-loop area law is not the QCD criterion; string breaking and the absence of colored asymptotic states become central.
Anomaly matching and generalized symmetries can exclude a trivially gapped symmetric infrared, but they rarely prove which allowed alternative is realized Gaiotto et al. 2015, generalized-symmetry framework. Semiclassical continuity on a compactified geometry can reveal mechanisms and ambiguity cancellation, yet a phase transition along decompactification invalidates the extrapolation. Lattice agreement at fixed spacing or within one scale-setting procedure does not certify the continuum.
Evidence, errors, and benchmarks
Section titled “Evidence, errors, and benchmarks”The key evidence units are dimensionless spectral ratios, finite-size scaling functions, renormalized line or local observables, Ward identities, and cross-method quantities with an explicit translation. Error budgets include sampling and autocorrelation, finite volume, cutoff effects, renormalization, excited-state contamination, topology freezing, analytic continuation, functional closure, and deformation dependence. Shared gauge ensembles or common scale inputs place nominally separate analyses in the same evidence family.
Benchmark ladders should begin with exactly solvable or high-precision lower-dimensional systems, continue through pure-gauge spectra and thermodynamics, and only then address QCD-like dynamical matter. In a compact gauge model, reproduce the known confinement diagnostic and its volume and continuum scaling before transferring the method. For strong-coupling gap equations, vary the vertex closure, enforce the relevant Ward or Slavnov–Taylor identities, and compare regulator dependence against a benchmark observable. At finite density, compare methods only in a shared validity window, with matched parameters and observables, and show explicitly where reweighting, series convergence, truncation, or analytic continuation ceases to be controlled.
Entering the field
Section titled “Entering the field”Choose a theory with gauge group, global form, matter representation, dimension, temperature/density, and observable fixed. State which meaning of “confinement” or “mass generation” is under test. Then reproduce a benchmark with at least two regulator/truncation values and one independent diagnostic. The computational field theory pathway and uncertainty and negative-results module provide the practical entry.
This guide omits nuclear bound-state structure and phenomenological hadron spectroscopy except where they validate a gauge-dynamics method. It also avoids assigning a single mechanism to QCD confinement: monopoles, vortices, center symmetry, dual superconductivity, and resurgence can be related in special regimes without being generally equivalent.
Evidence scope and related assessments
Section titled “Evidence scope and related assessments”The finite search used INSPIRE, arXiv, journal/DOI records, lattice-collaboration resources, and targeted searches for counterexamples to order-parameter claims. Sources public through 11 August 2026 were eligible. Reassess when a continuum-extrapolated result resolves a major cross-method discrepancy, a rigorous obstruction changes the allowed infrared phases, or a proposed continuity fails.
Continue to the four-dimensional Yang–Mills mass gap, confinement with dynamical matter, controlled finite-density QCD, or the functional-equations method map.
References
Section titled “References”- M. A. L. Capri et al., “Nonperturbative Aspects of Gauge Theories,” Physics Reports 945 (2022) 1–141. arXiv.
- S. Elitzur, “Impossibility of Spontaneously Breaking Local Symmetries,” Physical Review D 12 (1975) 3978–3982. DOI.
- E. Fradkin and S. H. Shenker, “Phase Diagrams of Lattice Gauge Theories with Higgs Fields,” Physical Review D 19 (1979) 3682–3697. DOI.
- D. Gaiotto, A. Kapustin, N. Seiberg, and B. Willett, “Generalized Global Symmetries,” JHEP 02 (2015) 172. DOI.
- K. G. Wilson, “Confinement of Quarks,” Physical Review D 10 (1974) 2445–2459. DOI.