Wigner’s Theorem Foundational Significance

Wigner’s theorem links an operational symmetry of pure states to the geometry of Hilbert space. A bijection of rays that preserves all transition probabilities is implemented by a unitary or antiunitary operator. The full theorem states the exact assumptions and proof; the foundational lesson is that physical symmetry is defined on states and probabilities before choosing vector representatives.

Symmetry of rays

Normalized vectors differing only by a phase factor represent the same pure-state ray. For normalized representatives the transition probability is

$$T(\psi,\phi)=|\langle\psi|\phi\rangle|^2.$$

Suppose a bijection $f$ of all rays preserves $T$. Wigner’s result supplies an operator $U$ such that $f([\psi])=[U\psi]$, where $U$ is either complex-linear and unitary or conjugate-linear and antiunitary. The operator representing the same ray map is only fixed up to a global phase.

These qualifications matter. A probability-preserving embedding need not be onto; a transformation preserving only one measurement’s statistics is much weaker; and a many-to-one noisy process is not a Wigner symmetry. The theorem already assumes quantum projective state space. It is not a derivation of Hilbert space from probability alone.

Complex conjugation as a concrete antiunitary

Let $K$ conjugate coefficients in a fixed orthonormal basis. Then

$$K(a\psi+b\phi)=a^*K\psi+b^*K\phi, \qquad \langle K\psi|K\phi\rangle=\langle\phi|\psi\rangle.$$

Squaring the overlap removes conjugation, so $K$ preserves transition probabilities. On a qubit Bloch vector it maps $(r_x,r_y,r_z)$ to $(r_x,-r_y,r_z)$. This reflection is not a proper rotation; it illustrates why unitary rotations alone do not exhaust ray symmetries.

A time-reversal symmetry uses an antiunitary representation appropriate to the system. For a spinless particle, basis conjugation can reverse momentum. For spin, an additional unitary factor is needed. Specifying a symmetry also requires its action on observables and the Hamiltonian; declaring a map antiunitary is not enough to prove a given dynamics is time-reversal invariant.

Symmetry is not a universal processing device

On density matrices the conjugation example acts as transposition, $\rho\mapsto\rho^T$. Transposition is positive on an isolated system but not completely positive. Applying it to one half of $|\Phi^+\rangle=(|00\rangle+|11\rangle)/\sqrt2$ yields an operator with eigenvalue $-1/2$ on the singlet subspace. It therefore cannot be a deterministic channel on an arbitrary subsystem entangled with an external reference.

There is no conflict with Wigner’s theorem. A symmetry transforms the description of an entire physical situation, including its temporal or geometric interpretation. A quantum channel describes a realizable local operation while arbitrary ancillary systems are left intact. These are different mathematical requirements.

Group composition and phases

If ray transformations form a group, their unitary representatives can satisfy $U(g)U(h)=e^{i\alpha(g,h)}U(gh)$. Observable transition probabilities do not detect the phase, but consistency of composition restricts it and leads to projective representations. Continuous transformations connected to the identity lie on the unitary side; antiunitary symmetries appear in a disconnected component. The theorem classifies probability-preserving symmetries, while continuity and group structure add further conclusions.

Exercise: a one-measurement imitation

A qubit dephasing channel removes off-diagonal matrix elements in the computational basis. Does it preserve probabilities for that basis? Is it a Wigner symmetry?

Solution

It preserves the two diagonal probabilities. But it sends the distinct pure states $|+\rangle$ and $|-\rangle$ to the same mixed state $I/2$. It neither maps all pure rays bijectively to pure rays nor preserves all transition probabilities, so the theorem does not apply.

References

• Bargmann, Valentine, “Note on Wigner’s Theorem on Symmetry Operations,” Journal of Mathematical Physics 5, 862–868 (1964), doi:10.1063/1.1704188 — precise symmetry theorem.
• Wigner, Eugene P., Group Theory and its Application to the Quantum Mechanics of Atomic Spectra, Academic Press (1959) — ray symmetries and their physical role.