STRATA
Skip to study
XLVII

Quantum field theory and analysis

Yang–Mills existence and mass gap

Can four-dimensional quantum Yang–Mills theory be constructed rigorously, with a strictly positive energy threshold above its vacuum?

Clay: Unsolved

Status and source review: 2026-11-19

State the mathematical question precisely.

For every compact simple gauge group G, construct a nontrivial quantum Yang–Mills theory on R⁴ satisfying axiomatic requirements at least as strong as the specified Wightman or Osterwalder–Schrader frameworks, and establish a mass gap Δ > 0. A classical gauge-field solution or a finite lattice calculation does not meet this formulation.

Identify what a complete argument must establish.

  1. Specify the physical observables and Hilbert space, recover relativistic locality and positivity, and construct a nontrivial continuum theory.
  2. Control removal of the short-distance regulator and the infinite-volume limit; estimates must remain meaningful in physical units.
  3. Prove that the energy spectrum has no positive values below a fixed Δ, without assuming the desired gap in a compactness or reconstruction step.

Trace the results that changed the problem.

  1. Yang and Mills introduce non-Abelian gauge fields.

    Their framework connects local internal symmetry with a gauge connection; the quantum existence question later becomes a separate constructive problem.

  2. Short-distance behavior becomes calculable.

    Gross and Wilczek's asymptotic-freedom calculation concerns ultraviolet behavior. It does not establish long-distance mass generation.

  3. Lattices and Euclidean reconstruction supply complementary tools.

    Wilson preserves gauge symmetry on a lattice; Osterwalder and Schrader specify when Euclidean correlations reconstruct a relativistic theory.

  4. A general spectral-gap algorithm encounters an undecidability boundary.

    Cubitt, Perez-Garcia and Wolf construct undecidable families of two-dimensional spin models. Their result does not determine the status of this particular four-dimensional gauge theory.

Separate established results from the remaining claim.

Strong coupling gives a controlled lattice regime.

Wilson's strong-coupling expansion yields a confinement mechanism on the regulator lattice. The regime lacks the required continuum rotational symmetry; extrapolating it to a continuum limit needs additional arguments.

Inspect the upstream dependencies.

Make each proposed approach answer for its obstacle.

These are research directions and diagnostic tests. Their inclusion does not establish novelty, feasibility at full scale, or a solution.

Investigate a uniform correlation-decay certificate.

Choose a gauge-invariant observable and seek an exponential decay estimate with constants controlled through both limiting procedures. Record the physical units, regulator dependence and vacuum subtraction explicitly.

Obstacle. A fitted exponential over a finite distance window can coexist with slower asymptotic decay. A proof also needs convergence and nontriviality; decay alone cannot manufacture a quantum theory.

Try the next discriminating test.

For a restricted lattice model, derive a rigorous bound and identify exactly where its constants cease to remain uniform. This is a tractable obstruction audit, not a claimed new mass-gap proof.

Test whether a perturbative estimate survives increasing volume.

Start with a solvable gapped Hamiltonian, introduce a bounded interaction, and track the dependency of a spectral lower bound on system size.

Obstacle. The crude norm of an extensive perturbation grows with volume. A valid small-system estimate can therefore become useless even when the actual model remains gapped.

Try the next discriminating test.

Compare local stability estimates with the global operator-norm bound on a spin chain; then identify which additional gauge-specific arguments a Yang–Mills application would require.

Work through a restricted calculation.

Positive gaps at every finite size do not imply a uniform gap.

Consider the exactly diagonal matrices H_N = diag(0, 1/N), for positive integers N.

Every matrix has gap Δ_N = 1/N > 0, while inf_N Δ_N = 0. Exact rational checks at N = 1, 2, 4, 8, 16 reproduce this sequence; the limit follows directly from 1/N.

This is a logical counterexample to an inference from finite spectra. These matrices are not a Yang–Mills model, and no conclusion about its actual gap follows.

Compare this example with the interactive proof diagnostics

Follow the downstream consequences conditionally.

If a computation observes a positive gap.

It supplies evidence for the specified regulator, volume and observable. Uniform limits and axiom verification remain separate obligations.

Locate the actual quantum connection.

This problem is directly about quantum field theory. Spin-chain spectra, quantum simulation and Hamiltonian complexity provide useful test environments, but a spin-model analogy is not an equivalence to four-dimensional Yang–Mills.

Follow quantum mechanics through the historical map

Inspect the supporting sources.

  1. The Millennium Prize Problems

    Clay Mathematics Institute · Current official categories and original seven-prize fund.

  2. Quantum Yang–Mills Theory

    Arthur Jaffe and Edward Witten / Clay Mathematics Institute · Existence of nontrivial four-dimensional QFT for every compact simple gauge group and positive mass gap.

  3. Confinement of quarks

    Kenneth G. Wilson / Physical Review D · 1974 gauge-invariant lattice construction and strong-coupling confinement mechanism.

  4. Ultraviolet Behavior of Non-Abelian Gauge Theories

    David J. Gross and Frank Wilczek / Physical Review Letters · 1973 short-distance asymptotic freedom calculation.

  5. Axioms for Euclidean Green's Functions II

    Konrad Osterwalder and Robert Schrader / Communications in Mathematical Physics · 1975 reconstruction conditions extending and correcting the 1973 paper.

  6. Undecidability of the Spectral Gap

    Toby Cubitt, David Perez-Garcia and Michael M. Wolf / arXiv · Undecidability across specially constructed families of translation-invariant two-dimensional lattice Hamiltonians.