Quantum Noether Principle

The quantum Noether principle is the operator version of the symmetry-conservation link: a continuous unitary symmetry of the Hamiltonian has a generator, and that generator is conserved when it has no explicit time dependence.

In its most common form,

$$U(\alpha) = e^{-i\alpha G/\hbar}, \qquad U(\alpha)HU^\dagger(\alpha)=H \quad \Longrightarrow \quad [G,H]=0.$$

Then the Heisenberg equation implies

$$\frac{dG_H}{dt}=0$$

when $G$ has no explicit time dependence. This is not the full classical or field-theoretic Noether theorem. It is the ordinary quantum-mechanical commutator form of the same pattern.

Classical Theorem Preview

Classical Noether theorem relates continuous symmetries of an action to conserved quantities. In point-particle mechanics, spatial translation symmetry gives momentum conservation; rotational symmetry gives angular-momentum conservation; time-translation symmetry gives energy conservation.

For a continuous global symmetry of a local classical field action, the on-shell statement is local: a Noether current $j^\mu$ obeys

$$\partial_\mu j^\mu=0.$$

The conserved charge is an integral of the density,

$$Q = \int d^3x\,j^0(t,\mathbf x),$$

when the boundary flux is controlled. The bridge from quantum generators to field-theory charges and currents is treated in From Quantum Generators to Noether Currents.

This page stays with ordinary quantum mechanics: Hilbert spaces, unitary transformations, generators, Hamiltonians, and commutators.

Quantum Generator Relation

A continuous additive symmetry subgroup is represented by a one-parameter unitary group:

$$U(\alpha+\beta)=U(\alpha)U(\beta), \qquad U(0)=I.$$

Under the usual continuity and domain assumptions, it has a self-adjoint generator $G$:

$$U(\alpha) = e^{-i\alpha G/\hbar}.$$

In finite dimensions, for small $\delta\alpha$,

$$U(\delta\alpha) = I-\frac{i}{\hbar}\delta\alpha\,G +O(\delta\alpha^2).$$

The generator $G$ is the observable associated with motion along the symmetry direction. Momentum is the generator of translations; angular momentum is the generator of rotations; the Hamiltonian is the generator of time translations.

Symmetry of the Hamiltonian

The transformation is a symmetry of a Hamiltonian $H$ when the Hamiltonian is invariant:

$$U(\alpha)HU^\dagger(\alpha)=H$$

for all $\alpha$ in the symmetry group. Differentiating this condition at $\alpha=0$ gives

$$[G,H]=0.$$

This is the key operator step. It is derived in first-order form in Infinitesimal Transformations and used systematically in Commutators and Conservation Laws.

The statement also has a converse under standard assumptions. If $G$ is self-adjoint and $[G,H]=0$ in the appropriate operator sense, then the unitary family generated by $G$ commutes with the dynamics, so it is a symmetry. For unbounded self-adjoint $G$ and $H$, the precise sufficient condition is that their spectral projections commute. A formal commutator on a common domain does not establish this by itself.

Conservation of the Generator

For an observable $A$ in the Heisenberg picture,

$$\frac{dA_H}{dt} = \frac{i}{\hbar}[H_H,A_H] + \left(\frac{\partial A}{\partial t}\right)_H.$$

Taking $A=G$ gives

$$\frac{dG_H}{dt} = \frac{i}{\hbar}[H_H,G_H] + \left(\frac{\partial G}{\partial t}\right)_H.$$

If $[H,G]=0$ and $G$ has no explicit time dependence, then

$$\frac{dG_H}{dt}=0.$$

Equivalently, the probability distribution of measurement outcomes of $G$ is constant under the dynamics. Conservation does not mean that every state is an eigenstate of $G$.

Examples

The standard examples are the quantum versions of the familiar Noether associations.

Continuous symmetry	Generator	Conservation statement
spatial translations	momentum $\mathbf P$	translation invariance implies $[\mathbf P,H]=0$
rotations	angular momentum $\mathbf J$	rotational invariance implies $[\mathbf J,H]=0$ componentwise
time translations	Hamiltonian $H$	time-independent dynamics conserves energy
global internal phase rotations	number or charge $N,Q$	invariance conserves the corresponding charge

These entries are schematic. External fields, boundary conditions, explicit time dependence, and domain choices can modify the precise symmetry statement.

Translation Example

In one dimension, translations are represented by

$$T(a)=e^{-iaP/\hbar}.$$

If

$$T(a)HT^\dagger(a)=H$$

for every displacement $a$, then differentiating at $a=0$ gives

$$[P,H]=0.$$

For

$$H=\frac{P^2}{2m}+V(X),$$

this requires $V(X)$ to commute with $P$, which holds for a constant potential on the translated region. Thus the free particle has momentum conservation because it has translation symmetry.

Rotation Example

For rotations about a fixed axis $\hat{\mathbf n}$,

$$U(\theta) = \exp\left( -\frac{i}{\hbar}\theta\,\hat{\mathbf n}\cdot\mathbf J \right).$$

If the Hamiltonian is invariant under this one-parameter subgroup, then

$$[\hat{\mathbf n}\cdot\mathbf J,H]=0.$$

If the Hamiltonian is invariant under all spatial rotations, then each component of total angular momentum commutes with $H$:

$$[J_i,H]=0.$$

For a central potential, this is the reason energy eigenstates can be organized by angular momentum quantum numbers.

Time Translation Example

For a time-independent Hamiltonian,

$$U(t)=e^{-itH/\hbar}.$$

The Hamiltonian generates time translations. Since $[H,H]=0$,

$$\frac{d}{dt}\langle H\rangle=0$$

when $H$ has no explicit time dependence.

This example can be misleading if stated too quickly. A time-dependent Hamiltonian still generates infinitesimal time evolution, but it does not usually define a one-parameter time-translation symmetry. In driven systems, energy need not be conserved.

Phase Symmetry Example

For a global internal phase symmetry,

$$U(\theta)=e^{-i\theta N},$$

where $N$ is dimensionless. If

$$U(\theta)HU^\dagger(\theta)=H$$

for all $\theta$, then

$$[N,H]=0.$$

This is an internal transformation whose phase depends on the number sector: an $n$-particle component acquires $e^{-in\theta}$. It is different from multiplying every vector by the same phase $e^{i\theta}$, which is generated by the identity and yields no nontrivial conserved quantity. The quantum number associated with $N$ is conserved. In many-body language this is the seed of particle-number or charge conservation. In field theory, the corresponding conserved charge is often written as the spatial integral of a density.

What the Principle Does Not Say

The quantum Noether principle has a precise scope.

It applies to continuous unitary symmetries. A discrete symmetry such as parity can constrain matrix elements and label states, but its geometric group has no infinitesimal parameter at the identity. If parity commutes with $H$, the parity observable is nevertheless conserved; a discrete symmetry does not need a Noether argument for that conclusion.

It does not apply to antiunitary symmetries in the same form. Time reversal is essential in quantum mechanics, but it is not generated by exponentiating a self-adjoint observable through a real continuous unitary parameter.

It does not say that any unitary operator gives a conserved quantity. The unitary must be a symmetry of the Hamiltonian.

It does not say that a conserved expectation value in one special state proves an operator symmetry. The operator commutator or full invariance condition is stronger.

It does not turn gauge redundancy into an ordinary physical symmetry. Global symmetries can carry charges; gauge redundancies impose constraints and identify descriptions.

Relation to Good Quantum Numbers

When a conserved generator $G$ commutes with $H$, energy eigenstates can often be chosen to diagonalize $G$ as well. The eigenvalue of $G$ is then a good quantum number.

For example, in a central potential,

$$[H,L^2]=0, \qquad [H,L_z]=0.$$

Energy eigenstates can be labeled by angular momentum quantum numbers, subject to degeneracies and the choice of a commuting set. The practical state-labeling story is developed in Simultaneous Eigenstates and Good Quantum Numbers.

Exercises

1. For finite matrices, let $U(\alpha)=e^{-i\alpha G/\hbar}$ and suppose $U(\alpha)HU^\dagger(\alpha)=H$ for all $\alpha$. Derive $[G,H]=0$.

Solution

For small $\alpha$,

$$U(\alpha) = I-\frac{i}{\hbar}\alpha G+O(\alpha^2), \qquad U^\dagger(\alpha) = I+\frac{i}{\hbar}\alpha G+O(\alpha^2).$$

Then

$$UHU^\dagger = H-\frac{i\alpha}{\hbar}GH +\frac{i\alpha}{\hbar}HG +O(\alpha^2).$$

The first-order term must vanish, so $GH-HG=0$. Therefore $[G,H]=0$.

2. A Hamiltonian $H=P^2/(2m)+V(X)$ is invariant under all translations on the line. What does the quantum Noether principle say, and what does it imply about $V$?

Solution

Translations are generated by $P$, so translation invariance implies

$$[P,H]=0.$$

Since $[P,P^2/(2m)]=0$ and $[P,V(X)]=-i\hbar V'(X)$, the commutator vanishes only when $V'(X)=0$ on the translated region. Thus $V$ is constant there, and momentum is conserved.

3. Explain why a time-dependent Hamiltonian need not conserve energy even though the Hamiltonian generates time evolution.

Solution

The Hamiltonian always generates infinitesimal time evolution in closed quantum mechanics. Energy conservation, however, requires time-translation symmetry. If $H$ depends explicitly on time, then the dynamics is not invariant under shifting the time origin in the same way, and

$$\frac{d}{dt}\langle H\rangle = \left\langle\frac{\partial H}{\partial t}\right\rangle$$

in the closed-system case. This need not vanish.

4. Why is parity not covered by the quantum Noether principle in the same way as translations?

Solution

Parity is a discrete transformation. The two-element spatial parity group has no small real parameter through the identity and therefore no geometric infinitesimal generator. One can mathematically embed its operator in a continuous family, but that is not a continuous family of spatial reflections. Parity can still be a symmetry and can still constrain spectra and matrix elements, but its consequences are not the Noether-style generator-conservation relation.

References

• L. E. Ballentine, Quantum Mechanics: A Modern Development, 2nd ed., World Scientific, 2014.
• H. Goldstein, C. Poole, and J. Safko, Classical Mechanics, 3rd ed., Addison-Wesley, 2002.
• E. Noether, “Invariante Variationsprobleme,” Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen, Mathematisch-Physikalische Klasse, 235–257 (1918).
• J. J. Sakurai and J. Napolitano, Modern Quantum Mechanics, 3rd ed., Cambridge University Press, 2020.
• S. Weinberg, The Quantum Theory of Fields, Vol. I: Foundations, Cambridge University Press, 1995.