Skip to content

From Symmetry Breaking to Goldstone Theorem

Spontaneous symmetry breaking is the bridge from symmetry as classification to symmetry as low-energy dynamics.

In finite quantum mechanics, a symmetry often gives labels, degeneracies, and selection rules. In many-body physics and quantum field theory, a continuous global symmetry can be exact in the equations while the vacuum or phase fails to display it. Under suitable assumptions, that mismatch forces low-energy collective excitations: Goldstone modes.

The compressed route is:

continuous global symmetry⟹Noether current⟹broken vacuum⟹Goldstone mode.\text{continuous global symmetry} \quad\Longrightarrow\quad \text{Noether current} \quad\Longrightarrow\quad \text{broken vacuum} \quad\Longrightarrow\quad \text{Goldstone mode}.

This page explains the field-theory bridge. The conceptual prerequisites are Spontaneous Symmetry Breaking Preview and Goldstone Modes Preview. Goldstone Modes in Many-Body Systems owns nonrelativistic type-A/type-B counting and condensed-matter diagnostics; the compact reference statement is Goldstone Theorem Preview.

From Symmetric Hamiltonian to Broken State

Section titled “From Symmetric Hamiltonian to Broken State”

Let a continuous group GG act by unitary transformations on the theory. In ordinary Hamiltonian language, exact symmetry means

U(g)HU(g)−1=H.U(g)HU(g)^{-1}=H.

Spontaneous breaking means that the relevant state or vacuum is not invariant under the full group. Instead, it preserves only a subgroup H⊂GH\subset G:

G⟶HG \longrightarrow H

The broken directions are the elements of G/HG/H. They label ways to move the order parameter without changing the energy at the perfectly symmetric level.

The finite-system caveat remains important. A finite system with a unique ground state usually preserves the exact symmetry as a ray. The broken-vacuum picture becomes sharp in an infinite-volume, thermodynamic, or field-theoretic limit.

An order parameter is an operator or field O\mathcal O whose expectation value diagnoses the symmetry breaking. A generator QaQ_a is broken if the vacuum expectation value of the transformed operator is nonzero:

⟨Ω∣δaO∣Ω⟩≠0.\langle\Omega|\delta_a\mathcal O|\Omega\rangle \ne 0.

Using the charge action

δaO=iℏ[Qa,O],\delta_a\mathcal O = \frac{i}{\hbar}[Q_a,\mathcal O],

the same condition can be written as

⟨Ω∣[Qa,O]∣Ω⟩≠0.\langle\Omega|[Q_a,\mathcal O]|\Omega\rangle \ne 0.

This is the field-theory upgrade of saying that the state is not invariant under the generator. The charge does something physically visible to the chosen vacuum.

For a continuous global symmetry in field theory, the charge is the spatial integral of a current density:

Qa(t)=∫d3x ja0(t,x),∂μjaμ=0.Q_a(t) = \int d^3x\,j_a^0(t,\mathbf x), \qquad \partial_\mu j_a^\mu=0.

Thus the broken-charge condition is really a statement about the integrated current:

∫d3x ⟨Ω∣[ja0(t,x),O(0)]∣Ω⟩≠0.\int d^3x\, \langle\Omega|[j_a^0(t,\mathbf x),\mathcal O(0)]|\Omega\rangle \ne 0.

That integral cannot be nonzero if the current correlation is short-ranged in every direction and has no low-energy spectral support. The theorem-level argument makes this intuition precise: current conservation, locality, and a nonzero order-parameter commutator force a massless or gapless contribution.

This is why Goldstone theorem is not just a picture of a tilted potential. It is a statement about current, charge, vacuum, and spectral weight.

In relativistic field theory, a broken continuous global symmetry is often visible as a pole in a current-order-parameter correlation function. Schematically,

∫d4x eip⋅x⟨Ω∣T jaμ(x)O(0)∣Ω⟩∼pμFap2+i0\int d^4x\, e^{ip\cdot x} \langle\Omega| T\,j_a^\mu(x)\mathcal O(0) |\Omega\rangle \sim \frac{p^\mu F_a}{p^2+i0}

near p2=0p^2=0, when O\mathcal O overlaps with the broken direction. The pole represents a massless Goldstone boson.

The exact numerator, normalization, and operator choice are convention-dependent. The invariant lesson is that the current couples to a gapless excitation:

⟨Ω∣jaμ(0)∣πb(p)⟩∝pμ.\langle\Omega|j_a^\mu(0)|\pi_b(p)\rangle \propto p^\mu.

The state ∣πb(p)⟩|\pi_b(p)\rangle is the field-theory particle or collective excitation associated with the broken generator. In condensed-matter language it may be a phonon, magnon, phase mode, or sound-like branch rather than a relativistic particle.

A common field-theory cartoon uses a complex scalar order parameter with potential

V(ϕ)=λ(∣ϕ∣2−v22)2.V(\phi) = \lambda \left( |\phi|^2-\frac{v^2}{2} \right)^2.

The minima satisfy

∣ϕ∣=v2.|\phi|=\frac{v}{\sqrt2}.

Choosing one phase breaks the global U(1)U(1) symmetry. Writing fluctuations as

ϕ(x)=12[v+h(x)]eiπ(x)/v,\phi(x) = \frac{1}{\sqrt2} \left[ v+h(x) \right] e^{i\pi(x)/v},

the radial field hh is typically massive, while the phase field π\pi is massless in the exact global-symmetry limit.

This picture is useful, but it is not the theorem. The theorem does not require a literal two-dimensional potential drawn on a page. It requires the symmetry, broken vacuum, locality, and current assumptions.

In many relativistic examples, the number of Goldstone bosons equals the number of broken generators:

NGoldstone=Nbroken.N_{\rm Goldstone} = N_{\rm broken}.

Nonrelativistic many-body systems can be subtler. Broken generators can form conjugate pairs in the ordered state, producing fewer Goldstone modes than broken generators. A common counting refinement uses

ρab=−iV⟨[Qa,Qb]⟩,\rho_{ab} = -\frac{i}{V} \langle[Q_a,Q_b]\rangle,

and, under suitable assumptions,

NNG=Nbroken−12rank⁡ρ.N_{\rm NG} = N_{\rm broken} - \frac12 \operatorname{rank}\rho.

This formula is included only as orientation. The detailed counting theorem belongs to many-body and field-theory treatments. For this bridge page, the key lesson is that continuous broken global symmetry produces gapless collective structure, while the exact number and dispersion can depend on the system.

Goldstone theorem applies to broken physical global symmetries. A gauge transformation is a redundancy in the description, not an operation that maps one physical state to a distinct physical state.

This distinction changes the spectrum. In a theory with a local gauge field, the would-be Goldstone field can combine with the gauge field. In particle-physics language this is the Higgs mechanism; in superconductors it is tied to electromagnetic response and the Anderson mechanism.

The safe bridge statement is:

broken global symmetrygives Goldstone modes,\text{broken global symmetry} \quad \text{gives Goldstone modes,}

but

gauge redundancyrequires a different analysis.\text{gauge redundancy} \quad \text{requires a different analysis.}

The global-to-gauge distinction is developed in From Phase Symmetry to Gauge Theory.

If the symmetry is explicitly broken, the current is no longer exactly conserved:

∂μjμ≠0.\partial_\mu j^\mu \ne 0.

Then the Goldstone theorem’s exact gapless conclusion need not hold. If the explicit breaking is small, the mode may remain light but acquire a small gap. Such a mode is called a pseudo-Goldstone mode.

This is the field-theory version of a familiar idea from approximate symmetries: a small symmetry-breaking term turns an exact selection or degeneracy statement into an approximate one. The difference is that here the low-energy spectrum itself remembers the almost-flat broken direction.

The assumptions matter. Continuous symmetry breaking can be obstructed by strong fluctuations in low dimensions. Coleman-type and Mermin–Wagner-type results show that certain low-dimensional relativistic or thermal systems cannot spontaneously break continuous symmetries under their stated hypotheses.

Long-range interactions, gauge fields, finite density, finite temperature, boundaries, and disorder can also change the simple conclusion or its observable form.

These caveats do not make the Goldstone theorem vague. They make it honest: the theorem is powerful because its assumptions are specific.

When moving from ordinary quantum mechanics to field theory, ask:

  • What is the continuous global symmetry group GG?
  • What is the conserved current jaμj_a^\mu?
  • What state, phase, or vacuum is being considered?
  • Which subgroup HH leaves it invariant?
  • Is there an order parameter O\mathcal O with ⟨δaO⟩≠0\langle\delta_a\mathcal O\rangle\ne0?
  • Is the symmetry exact, approximate, explicitly broken, or gauged?
  • Is the system relativistic, nonrelativistic, finite-density, finite-temperature, or low-dimensional?

Those questions are the operational bridge from symmetry labels to Goldstone physics.

  • Calling any asymmetric finite-system state a spontaneously broken phase.
  • Applying Goldstone theorem to broken discrete symmetries.
  • Forgetting that the theorem needs a continuous global symmetry.
  • Treating gauge redundancy as a physical global symmetry.
  • Counting Goldstone modes by broken generators without checking nonrelativistic charge commutators.
  • Ignoring explicit breaking and then expecting exactly gapless modes.
  • Treating the Mexican-hat diagram as the theorem rather than an illustrative model.
  • Forgetting that current conservation and locality are part of the field-theory argument.
  • J. Goldstone, “Field theories with superconductor solutions,” Il Nuovo Cimento 19, 154-164, 1961.
  • J. Goldstone, A. Salam, and S. Weinberg, “Broken symmetries,” Physical Review 127, 965-970, 1962.
  • Y. Nambu, “Quasi-particles and gauge invariance in the theory of superconductivity,” Physical Review 117, 648-663, 1960.
  • P. W. Anderson, “Plasmons, gauge invariance, and mass,” Physical Review 130, 439-442, 1963.
  • S. Coleman, “There are no Goldstone bosons in two dimensions,” Communications in Mathematical Physics 31, 259-264, 1973.
  • H. Watanabe and H. Murayama, “Unified description of Nambu–Goldstone bosons without Lorentz invariance,” Physical Review Letters 108, 251602, 2012.
  • S. Weinberg, The Quantum Theory of Fields, Volume II: Modern Applications, Cambridge University Press, 1996.
  • M. E. Peskin and D. V. Schroeder, An Introduction to Quantum Field Theory, Addison-Wesley, 1995.
  • P. Coleman, Introduction to Many-Body Physics, Cambridge University Press, 2015.
  • A. Altland and B. Simons, Condensed Matter Field Theory, 2nd ed., Cambridge University Press, 2010.
  1. Broken charge criterion.

Suppose δO=(i/ℏ)[Q,O]\delta\mathcal O=(i/\hbar)[Q,\mathcal O] and ⟨Ω∣δO∣Ω⟩≠0\langle\Omega|\delta\mathcal O|\Omega\rangle\ne0. What does this say about the action of QQ on the vacuum?

Solution

It says that the charge acts nontrivially in the vacuum sector. If the vacuum were invariant under QQ in the relevant sense, the expectation value of a pure symmetry variation would vanish. A nonzero value of

⟨Ω∣[Q,O]∣Ω⟩\langle\Omega|[Q,\mathcal O]|\Omega\rangle

means that the symmetry transformation changes an observable order parameter in the vacuum. That is the broken-generator signal.

  1. Goldstone versus radial mode.

In the complex-scalar parametrization

ϕ(x)=12[v+h(x)]eiπ(x)/v,\phi(x) = \frac{1}{\sqrt2} \left[ v+h(x) \right] e^{i\pi(x)/v},

which field is the Goldstone candidate and why?

Solution

The phase field π(x)\pi(x) is the Goldstone candidate. It moves the order parameter along the circle of symmetry-related minima. A uniform change of that phase is a symmetry transformation, so the exact global-symmetry limit gives no potential energy cost for the uniform phase direction. The radial field h(x)h(x) changes the magnitude and is typically massive.

  1. Why does explicit breaking gap a Goldstone mode?
Solution

Explicit breaking tilts or lifts the formerly flat symmetry direction. In current language, the current is no longer exactly conserved. In effective-field language, a potential term appears for the would-be Goldstone coordinate. Small oscillations then have a nonzero restoring force at zero momentum, so the mode acquires a gap. If the breaking is weak, the result is a pseudo-Goldstone mode.

  1. Why is a gauge theory not handled by naive Goldstone counting?
Solution

Goldstone theorem counts broken physical global symmetries. Gauge transformations are redundancies, so gauge-related configurations do not represent distinct physical states. When a charged condensate is coupled to a gauge field, the would-be Goldstone degree of freedom can combine with the gauge field and alter the spectrum. One must analyze gauge-invariant observables and constraints rather than naively counting broken gauge generators.