Hidden Symmetry

A hidden symmetry is an exact symmetry or conserved algebra that is not manifest in the first description of a system. It often appears only after one discovers additional operators that commute with the Hamiltonian, close a larger algebra with the visible generators, and organize degeneracies that the obvious geometric symmetry does not explain.

The word “hidden” is relative to a representation. A symmetry may be hidden in coordinates, visible in ladder operators, hidden in a differential equation, or visible after a change of variables. What matters is not psychological surprise; it is the existence of exact operators with definite commutation relations.

Basic Operator Criterion

Let $H$ be a Hamiltonian. A hidden conserved quantity is an operator $A$ such that

$$[H,A]=0,$$

but $A$ is not part of the manifest symmetry set one started from. If several such operators exist, they may close an algebra with the visible generators.

For example, angular momentum satisfies

$$[H,L_i]=0$$

for a central potential. That is not hidden; it is the manifest rotational symmetry. A new conserved vector $\mathbf A$ built from $\mathbf R$, $\mathbf P$, and $\mathbf L$ would be hidden if it is conserved only for a special potential and relates sectors not connected by rotations alone.

What Hidden Symmetry Explains

Hidden symmetry usually explains one of three things:

• degeneracy larger than the manifest symmetry requires;
• unexpectedly simple spectra;
• algebraic solvability beyond separation of variables.

It does not mean every degeneracy has a hidden symmetry. Some repeated energies are fine-tuned accidents. The diagnostic page is Accidental Symmetry; the hidden-symmetry question is whether one can actually write conserved operators that organize the pattern.

Coulomb Problem and the Runge–Lenz Vector

For the spinless Coulomb Hamiltonian

$$H = \frac{\mathbf P^2}{2\mu} - \frac{\kappa}{R}, \qquad R=\lvert\mathbf R\rvert,$$

rotational symmetry gives conservation of orbital angular momentum:

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

Rotations explain the degeneracy among $m$ states at fixed $\ell$. They do not explain why different $\ell$ values inside a fixed principal shell share the same ideal Coulomb energy.

The extra conserved quantity is the quantum Laplace–Runge–Lenz vector. With a common Hermitian ordering convention, it can be written

$$\mathbf A = \frac{1}{2\mu} \left( \mathbf P\times\mathbf L - \mathbf L\times\mathbf P \right) - \kappa\frac{\mathbf R}{R}.$$

For the Coulomb Hamiltonian,

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

Together, $\mathbf L$ and $\mathbf A$ form a larger algebra. Schematically,

$$[L_i,A_j] = i\hbar\sum_k\epsilon_{ijk}A_k,$$

so $\mathbf A$ transforms as a vector under rotations, while

$$[A_i,A_j] = - \frac{2i\hbar H}{\mu} \sum_k\epsilon_{ijk}L_k.$$

In the bound-state sector, where $H<0$, a rescaled version of $\mathbf A$ combines with $\mathbf L$ into an $\mathrm{SO}(4)$ algebra. That enlarged structure organizes the degeneracy across different $\ell$ values. The detailed hydrogen counting belongs to Degeneracy of the Hydrogen Atom.

Why This Is Not Ordinary Rotation

The angular momentum operators change $m$ within a fixed $\ell$ multiplet:

$$L_\pm:\quad \ell,m \mapsto \ell,m\pm1.$$

They do not change $\ell$. The hidden Coulomb generators relate states inside the same principal shell in a way that goes beyond physical-space rotations. That is why the degeneracy between, for example, $2s$ and $2p$ states is not a mere consequence of spherical symmetry.

This distinction is one of the most useful tests in central-potential problems:

$$\text{rotational symmetry} \Rightarrow \text{$m$ degeneracy},$$

but

$$\text{special hidden structure} \Rightarrow \text{extra degeneracy across other labels}.$$

Isotropic Oscillator Hidden Structure

The isotropic harmonic oscillator also has more algebraic structure than ordinary rotations alone suggest. In $d$ dimensions,

$$H = \hbar\omega \left( \sum_{i=1}^{d}a_i^\dagger a_i + \frac d2 \right).$$

The bilinears

$$E_{ij} = a_i^\dagger a_j$$

commute with $H$:

$$[H,E_{ij}]=0.$$

They move excitation quanta between Cartesian directions while keeping the total excitation number

$$N=\sum_i a_i^\dagger a_i$$

fixed. These operators generate a $\mathfrak u(d)$ algebra:

$$[E_{ij},E_{kl}] = \delta_{jk}E_{il} - \delta_{il}E_{kj}.$$

This algebra explains why all states with the same total $N$ have the same isotropic oscillator energy. If the frequencies become unequal, the bilinears no longer commute with the Hamiltonian in general, and the extra degeneracy is lost.

How Hidden Symmetry Is Found

There is no universal mechanical recipe, but several clues recur.

First, look for degeneracy larger than the manifest symmetry predicts. Hydrogenic $\ell$ degeneracy is the classic example.

Second, look for conserved quantities in the corresponding classical problem. Quantization must handle operator ordering and domains carefully, but classical constants of motion often point toward quantum operators.

Third, test commutators directly:

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

Fourth, check closure. A collection of conserved operators is more than a list when their commutators close into a recognizable algebra.

Fifth, test perturbations. If a small term preserving only the manifest symmetry splits the extra degeneracy, the hidden symmetry was special to the original Hamiltonian.

Hidden Versus Dynamical Symmetry

The terminology is not perfectly uniform across physics. A useful working distinction is:

• a hidden symmetry is often represented by conserved operators that commute with $H$ and organize degeneracy within energy eigenspaces;
• a dynamical or spectrum-generating symmetry may use operators that connect states of different energies or organize an entire spectrum.

The same physical system can have both descriptions in different contexts. For example, the Coulomb problem is often described with an $\mathrm{SO}(4)$ bound-state symmetry and also with larger spectrum-generating structures in more advanced treatments.

This page uses “hidden symmetry” for the conservative quantum-mechanical claim: exact conserved operators reveal structure not manifest in the first symmetry description.

Hidden Symmetry Is Physical, Not Gauge Redundancy

A hidden symmetry maps physical states or observables in a way that explains spectral or dynamical structure. It is not merely a redundancy in coordinates or gauge variables.

Gauge transformations require a separate interpretation: they relate different descriptions of the same physical state. Hidden symmetries, by contrast, usually act within the physical Hilbert space and can relate distinct states in a degenerate multiplet.

Common Mistakes

• Calling any unexplained degeneracy a hidden symmetry before constructing conserved operators.
• Treating the hydrogen $\ell$ degeneracy as ordinary rotational degeneracy.
• Forgetting operator-ordering issues when importing classical constants of motion into quantum mechanics.
• Assuming hidden symmetry survives arbitrary perturbations.
• Confusing hidden physical symmetry with gauge redundancy.
• Using “hidden,” “accidental,” and “dynamical” interchangeably without stating the operator algebra.

Cross-Links

• Exact Symmetry
• Accidental Symmetry
• Dynamical Symmetry
• Degeneracy Lifting
• Commutators and Conservation Laws
• Central Potentials and Rotational Symmetry
• Degeneracy of the Hydrogen Atom
• Hydrogen Atom
• Quantum Harmonic Oscillator
• Degenerate Perturbation Theory
• Textbooks

References

• M. Bander and C. Itzykson, “Group theory and the hydrogen atom (I),” Reviews of Modern Physics 38, 330-345, 1966.
• L. D. Landau and E. M. Lifshitz, Quantum Mechanics: Non-Relativistic Theory, 3rd ed., Butterworth-Heinemann, 1977.
• R. Shankar, Principles of Quantum Mechanics, 2nd ed., Springer, 1994.
• J. J. Sakurai and J. Napolitano, Modern Quantum Mechanics, 3rd ed., Cambridge University Press, 2020.
• M. Moshinsky and Y. F. Smirnov, The Harmonic Oscillator in Modern Physics, Harwood Academic, 1996.

Exercises

1. Conserved hidden operator.

Show that if $[H,A]=0$ and $H|\psi\rangle=E|\psi\rangle$, then $A|\psi\rangle$ is either zero or another state with the same energy.

Solution

Using the commutator,

$$H(A|\psi\rangle) = AH|\psi\rangle = E(A|\psi\rangle).$$

Thus $A|\psi\rangle$ is an eigenstate with the same energy whenever it is nonzero. If it is linearly independent of $|\psi\rangle$, it gives a degeneracy or moves within an already degenerate subspace.

2. Why is $2s$-$2p$ degeneracy not ordinary rotational degeneracy?

Solution

The $2s$ state has $\ell=0$, while the $2p$ states have $\ell=1$. Ordinary rotation generators $L_i$ act inside a fixed $\ell$ representation and change only the orientation labels $m$. They do not turn an $\ell=0$ state into an $\ell=1$ state. The ideal Coulomb degeneracy between $2s$ and $2p$ therefore requires structure beyond ordinary rotations.

3. Oscillator bilinears.

For the isotropic oscillator

$$H = \hbar\omega \left( \sum_i a_i^\dagger a_i + \frac d2 \right),$$

show that $E_{ij}=a_i^\dagger a_j$ commutes with $H$.

Solution

Let

$$N=\sum_k a_k^\dagger a_k.$$

The oscillator Hamiltonian is $\hbar\omega(N+d/2)$, so it is enough to show $[N,a_i^\dagger a_j]=0$. Using

$$[N,a_i^\dagger]=a_i^\dagger, \qquad [N,a_j]=-a_j,$$

we get

$$[N,a_i^\dagger a_j] = [N,a_i^\dagger]a_j + a_i^\dagger[N,a_j] = a_i^\dagger a_j - a_i^\dagger a_j = 0.$$

Therefore $[H,E_{ij}]=0$.

4. Unequal frequencies.

Why do the oscillator bilinears $a_i^\dagger a_j$ generally stop commuting with the Hamiltonian when the frequencies $\omega_i$ are unequal?

Solution

For unequal frequencies,

$$H = \sum_k \hbar\omega_k \left( a_k^\dagger a_k+\frac12 \right).$$

The bilinear $a_i^\dagger a_j$ raises the excitation number in direction $i$ and lowers it in direction $j$. The energy change associated with this move is

$$\hbar(\omega_i-\omega_j).$$

Thus

$$[H,a_i^\dagger a_j] = \hbar(\omega_i-\omega_j)a_i^\dagger a_j,$$

which vanishes only when $\omega_i=\omega_j$ or when the operator is diagonal with $i=j$.