Stone–von Neumann Theorem Foundational Significance

The Stone–von Neumann theorem explains why the usual position and momentum realizations of a finite canonical system are equivalent when their precise representation assumptions agree. It does not prove that every quantum theory has one representation, that quantization is unique, or that wave mechanics and particle ontology are identical. The full theorem contains the exact hypotheses and proof structure.

Uniqueness of a specified Weyl system

For one canonical pair, use position and momentum translations

$$U(a)=e^{-iaP/\hbar},\qquad V(b)=e^{ibQ/\hbar}.$$

In the standard Schrödinger representation they satisfy

$$U(a)V(b)=e^{-iab/\hbar}V(b)U(a).$$

The theorem concerns strongly continuous, irreducible unitary representations of the corresponding Weyl relations for finitely many canonical pairs, with the same fixed nonzero central parameter. Such representations are unitarily equivalent. Regularity supplies generators; irreducibility excludes extra multiplicity; finite degrees of freedom is a substantive restriction.

A formal equation $[Q,P]=i\hbar I$ on an unspecified domain is not an adequate replacement for these hypotheses. Unbounded operators require domain control, and exponentiation to a Weyl system can fail under weakened assumptions.

Equivalent representations preserve predictions

If $W$ intertwines two representations, then $Q'=WQW^\dagger$ and $P'=WPW^\dagger$ with their domains transformed as well. States and observables must be transformed together. For bounded $A$, or for an unbounded observable with a well-defined finite expectation,

$$\rho'=W\rho W^\dagger, \qquad A'=WAW^\dagger, \qquad \operatorname{Tr}(\rho'A')=\operatorname{Tr}(\rho A).$$

Thus the Fourier-transform passage between position and momentum descriptions changes coordinates in the representation, not experimental probabilities. Transforming a state while silently leaving its measurement operator unchanged would describe a different experiment.

This equivalence is a statement about the chosen observable algebra. It does not tell us whether a wavefunction is ontic, what constitutes an outcome or which variables a hidden-variable account may add.

What is not fixed by the theorem

The Hamiltonian is not specified by canonical kinematics alone. For example, $P^2/(2m)+V(Q)$ and a different potential share the same representation but have different dynamics. Quantization of general classical products also raises ordering choices: equivalent canonical representations do not choose a unique operator for every classical function.

Configuration-space topology may change the relevant algebra. A particle on a circle should not be assumed to possess a globally defined angle–momentum Weyl pair of the line type. Infinite systems and quantum fields can have inequivalent representations with physical significance; the finite-degree theorem does not extend by merely letting the number of pairs grow. See Infinite Systems and Inequivalent Representations for the precise boundary.

A finite-dimensional check

No finite matrices can obey $[Q,P]=i\hbar I$ exactly for $\hbar\ne0$. Taking the trace gives $0=i\hbar d$, a contradiction. A finite numerical truncation of an oscillator is consequently an approximation to a canonical system, not another exact irreducible Weyl representation to which uniqueness applies. Its boundary defects must be controlled for the states under study.

This example shows why the phrase “finite degrees of freedom” does not mean “finite Hilbert-space dimension.” One canonical coordinate has an infinite-dimensional Schrödinger space.

Exercise: adding an unused label

Take the standard representation on $L^2(\mathbb R)$ and tensor it with $\mathbb C^2$, with $Q=Q_0\otimes I$ and $P=P_0\otimes I$. Which uniqueness hypothesis fails?

Solution

Irreducibility fails. The bounded operator $I\otimes\sigma_z$ commutes with every Weyl operator, and its eigenspaces are invariant. The extra label gives two copies of the same representation. This is a multiplicity issue, not a contradiction of the irreducible uniqueness theorem.

References

• Folland, Gerald B., Harmonic Analysis in Phase Space, Princeton University Press (1989) — the Weyl representation and its uniqueness.
• Reed, Michael, and Barry Simon, Methods of Modern Mathematical Physics, Volume II: Fourier Analysis, Self-Adjointness, Academic Press (1975) — operator and domain qualifications.