research32 min studyLesson 709 in Physics

Gribov copies and global gauge fixing

Distinguish local Faddeev-Popov invertibility from global orbit uniqueness and interpret restricted gauge regions carefully.

Before you begin

On this page

The question for today

What you’ll understand

Explain how multiple gauge representatives coexist with nonzero local Jacobians.
Distinguish a first Gribov region from a fundamental modular region.
Use a finite nonlinear example without treating it as an actual gauge-field solver.

Gauge fixing is locally a coordinate choice on gauge orbits, but a global configuration space need not admit one simple coordinate chart. Gribov copies are distinct gauge-equivalent connections satisfying the same chosen condition. They expose a limitation of extending a local Faddeev-Popov prescription across the full non-Abelian field space. The issue does not invalidate a controlled perturbative patch; it requires care when its local formulas are promoted to statements about nonperturbative measures and physical states.

Local invertibility does not exclude distant copies

If the gauge-condition variation operator is invertible on its declared domain, nearby orbit motion changes the condition and a local representative can be isolated. The word nearby is essential. A nonlinear equation can have several distant roots, each with a nonzero derivative. Their existence is compatible with the implicit-function theorem. The same distinction separates a local gauge-coordinate calculation from a global assertion that each orbit has been counted once.

Global obstructions depend on the manifold, bundle, allowed transformations, and boundary conditions. Residual constant transformations or stabilizers of special connections require separate treatment even before considering more distant copies. A gauge chosen on a finite periodic lattice, a continuum Euclidean domain, and an asymptotic scattering region can have different admissible transformations. Calling every failure a zero determinant conflates discrete distant intersections with an infinitesimal unfixed direction.

G[A^U]=0\ \text{and}\ G[A]=0\quad\text{can hold for distinct representatives even if}\quad \det M[A]\ne0

This statement concerns the declared gauge group and boundary domain, with redundant residual modes removed. A nonzero regulated determinant is a local statement, not a global uniqueness theorem.↔ Scroll wide equations horizontally.

Restrict a region without declaring the problem solved

In Euclidean Landau gauge, the Faddeev-Popov operator can be treated as Hermitian with suitable boundary conditions. The first Gribov region consists of transverse connections where its nontrivial eigenvalues are positive. Its boundary is reached when the lowest such eigenvalue vanishes. Removing trivial global modes is essential for this definition: otherwise a zero mode can be present everywhere and would obscure the intended horizon condition.

Positivity removes infinitesimal copies in the relevant local sense, but distinct copies can still occur inside the region. A fundamental modular region instead selects absolute minima of the connection norm along each orbit, with degenerate boundary representatives identified. It lies in the closure of the first region and has a more complicated global structure. Confusing a local minimum with an absolute minimum is the finite-dimensional version of mistaking the first region for a complete unique gauge selection.

\Omega=\{A:\partial_\mu A_\mu=0,\ M[A]>0\},\qquad \lambda_{\min}(M)=0\ \text{on its horizon},\qquad \Lambda\subseteq\overline\Omega

The positivity and lowest eigenvalue refer to a regulated operator with trivial residual modes excluded. Lambda denotes the absolute-minimum fundamental modular region, including necessary boundary identifications.↔ Scroll wide equations horizontally.

Keep global restrictions compatible with identities

The BRST change-of-variables argument used an invariant measure and no contributing boundary. Restricting integration to a region can alter how a BRST transformation acts at its boundary. One must therefore reexamine the measure, Ward identities, and physical-state construction rather than retaining their local formulas without qualification. Different nonperturbative formulations handle these issues differently, and a region restriction alone is not a derivation of every confinement observable.

Finite examples can test the logical inference without reproducing gauge-field geometry. A positive scalar derivative plays the role of a local orientation test, but it has no color, spacetime spectrum, or functional measure. Finding two positive-derivative roots is enough to refute the claim that positivity always implies global uniqueness. Conversely, a numerical root finder confined to a monotone interval can succeed perfectly while saying nothing about solutions outside its declared domain.

F(\alpha)=\sin(2\alpha),\qquad M_{\rm toy}(\alpha)=F'(\alpha)=2\cos(2\alpha)

This is a dimensionless circle-valued coordinate example, not a Yang-Mills gauge condition or a Faddeev-Popov spectrum. It isolates the difference between local and global reasoning.↔ Scroll wide equations horizontally.

Work it through

Multiple roots with positive local orientation

On $0\le\alpha<2\pi$ , solve the toy condition $F(\alpha)=\sin(2\alpha)=0$ . Evaluate its derivative at every root and count how many roots remain if positive derivative is required.

The roots in the half-open interval are $0,\pi/2,\pi,3\pi/2$ . The endpoint at two pi is excluded because it is the same circle coordinate as zero.
The derivative is $2\cos(2\alpha)$ and gives values $2,-2,2,-2$ at those four roots. Every root has a nonzero local Jacobian, so each is locally regular.
Requiring a positive derivative keeps both zero and pi. There are still two distinct retained representatives in this finite analogy, demonstrating why a positivity test alone does not establish global uniqueness.

The result

Four locally regular roots exist, and two survive positivity; local orientation and global single counting are different requirements.

The living notebook

Explore the model

Interactive lab

Interactive lab · root-finding

A bracket selects a domain

Reset and set target $c=1$ and upper bound $U=2$ . Predict the positive root one, reached exactly after one Step at the midpoint. Compare a fresh Reset run at $c=4,U=2$ : the positive endpoint root two has zero bracket bound immediately. The model only solves $x^2-c$ on $[0,U]$ and excludes its negative root; it does not find Gribov copies or Faddeev-Popov eigenvalues.

Quadratic / midpointCurrent bracketPositive root

Midpoint after 0 refinements: 1.5
Bracket error bound: 1.5
Computed residual · x² − c: 0.25

Target c2Initial upper bound U3

t = 0 s

Bisection of x² − c on [0,U], capped at 50 iterations for floating-point resolution. The error bound is the maximum distance from the displayed midpoint to either bracket endpoint (half the width in exact arithmetic). Compensated square evaluation preserves the bracket when ordinary multiplication rounds a residual to zero. A truly exact endpoint or midpoint collapses the bracket. This is one monotone quadratic, not a general root solver. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.

Look forA certified result inside a selected bracket is compatible with additional solutions outside that domain.

Pause & reason

Check your understanding

Interactive checks need JavaScript. You can still read the explanation.

Read the checkpoint explanation

The first region imposes transverse gauge and local positivity. Absolute minimization along an orbit, together with boundary identifications, adds global information and can remove copies left by local positivity.

Your turn

Try it yourself

For the separate dimensionless map $F(\alpha)=\alpha^3-\alpha$ on the real line, find its zero-condition roots, derivative at each, and the points where the local map becomes singular. Does positive derivative choose one root?

Take a moment to make a prediction or work it out on paper. Open a hint when you need one.

Give me a hint

Factor the cubic to find roots, and solve the derivative equation separately; a derivative zero need not be a zero-condition root.

Compare with the solution

The zero roots are minus one, zero, and plus one. Their derivatives $3\alpha^2-1$ are two, minus one, and two. Thus positive derivative retains both endpoint roots. Local singular points occur at $\alpha=\pm1/\sqrt3$ , where the derivative vanishes; these are not roots of the zero condition. This example separates multiple regular intersections from singular coordinate points and is not a claim about a specific gauge-field configuration.

Keep the idea

Take it with you

Gribov copies concern global orbit intersections, not only zero local determinants.
First-region positivity is weaker than global absolute-minimum selection.
Nonperturbative restrictions require rechecking the assumptions of BRST identities.

Read further

Original sources and deeper explanations to follow at your own pace.

Optional progress, saved in this browser when available. Revisit any lesson at any time.

Lesson not marked complete.