Interpreting the Klein–Gordon Equation

The same Klein–Gordon differential equation appears in classical field theory, a restricted one-particle description, and quantum field theory. Its mathematical solutions do not by themselves decide which quantities are probabilities, charges, or detector responses. The interpretation depends on the state space, observables, and approximation being used.

Required background. The Klein–Gordon Inner Product defines the solution-space norm; External Potentials states the background approximation. Helpful background. Locality and Causality Warnings separates correlations from response.

Three uses of a scalar equation

Description	Meaning of the scalar object	What must be supplied
Classical field	A commuting field amplitude with Cauchy data.	An action or energy normalization and classical sources or boundaries.
Free one-particle sector	A vector in a selected positive-frequency solution space.	A positive Hilbert product and operators representing the intended measurements.
Quantized scalar field	An operator-valued distribution acting on a many-particle state space.	Commutation relations, a state, observables, and any interactions.

Moving between these descriptions is a physical construction, not a change of name for $|\phi|^2$. The free one-particle description can be consistent and useful, while still being insufficient for an experiment that changes particle number.

Positive classical energy and indefinite scalar charge

For the following free example use $\hbar=c=1$ and a standard classical normalization. A real scalar has energy density

$$\mathcal H=\frac12\left[ (\partial_t\phi)^2+|\nabla\phi|^2+m^2\phi^2\right]\geq0.$$

This is different from the antisymmetric complex-scalar charge bilinear, which vanishes when the field is real. In a periodic box, the homogeneous real solution $\phi=A\cos(mt)$ has $\mathcal H=m^2A^2/2$, while its KG charge is zero. It contains both frequency signs, but it does not have zero classical energy or a negative classical energy instability.

The distinction persists for a complex classical field: its free energy is a positive quadratic expression, while its phase-symmetry charge distinguishes the two frequency sectors. Neither expression is forced to be a Born position density. “Negative frequency,” “negative KG norm,” and “negative energy of a physical excitation” are different statements. Negative-Energy Solutions explains the spectral and quantization distinctions.

What the positive-frequency restriction achieves

In free Minkowski spacetime, the positive-frequency KG product is positive and invariant under proper orthochronous transformations. It therefore supports a one-particle Hilbert space. Its positive integrated norm does not imply that the original local KG density is nonnegative. The explicit interference calculation is in Probability-Density Problems.

A position observable can be constructed on a fixed spacelike slice, but its relation to a local relativistic detector is additional physics. Localization Problems gives the Newton–Wigner construction and its qualifications. A detector click probability must be calculated from a specified measurement model; it cannot be identified with a signed local charge density.

In a time-dependent background, the conserved pairing still exists, but an instantaneous or asymptotic frequency split needs its own definition. Mixing between chosen mode bases is not yet a complete prediction of a detector count. A field-theory particle-production calculation also specifies the initial state and the particle notion used in the final region.

What quantization changes

Continue in natural units $\hbar=c=1$ and write $\omega_{\mathbf p}=\sqrt{\mathbf p^2+m^2}$. For a free charged scalar in a box, quantization assigns annihilation operators to one frequency sector and antiparticle creation operators to the other. With canonical mode normalization, the normal-ordered free energy and charge take the schematic forms

$$H=\sum_{\mathbf p}\omega_{\mathbf p} (a_{\mathbf p}^\dagger a_{\mathbf p} +b_{\mathbf p}^\dagger b_{\mathbf p}),\qquad Q_{\rm electric}=q\sum_{\mathbf p} (a_{\mathbf p}^\dagger a_{\mathbf p} -b_{\mathbf p}^\dagger b_{\mathbf p}).$$

Both kinds of excitation have positive energy, while their charges are opposite. The displayed $H$ subtracts the free vacuum constant; it is not a statement about the gravitational significance of vacuum energy. A neutral real scalar has one species and no continuous phase charge, yet still has particle excitations and energy.

These formulas summarize what the quantized framework supplies; a c-number negative-frequency solution has not already performed that construction. External-field scalar equations remain useful ingredients in quantum-field calculations, but classical sources and first-quantized amplitudes alone do not specify a Fock-space vacuum.

Causal propagation and measurement

The complete scalar Cauchy problem has finite propagation speed. Vacuum correlations can be nonzero at spacelike separation, and a positive-frequency amplitude can have spatial tails. Neither observation replaces a response calculation. For a prescribed source use the retarded inverse in Klein–Gordon Propagators; for quantum local measurements examine the appropriate observable commutators and measurement coupling.

These distinctions also delimit applications. Scalar Coulomb levels describe a spin-zero external-source model, not electron hydrogen. A conserved scalar charge does not prove that pair production is negligible, because pairs may carry zero net added charge. Accuracy within a fixed particle sector requires an estimate of the omitted channels for the chosen observable and duration.

Exercises

1. A real free solution has zero KG charge. Does it have to vanish? Give a finite-volume counterexample and its energy density.

Solution

$\phi=A\cos(mt)$ in a periodic box has zero antisymmetric charge bilinear and positive constant energy density $m^2A^2/2$. Charge is not the norm of a real classical field configuration.

2. A process adds one particle and one antiparticle of a charged scalar. What happens to total electric charge and particle count?

Solution

The added charge is $q-q=0$, while the count rises by two. Charge conservation is fully compatible with pair creation; it cannot establish fixed particle number.

3. A source-response calculation produces a signal before the source was switched on. The calculation used $iD_F$ and has the correct delta normalization. What needs to change?

Solution

The boundary condition needs to change to the retarded inverse $G_R$. Correct source normalization does not impose retarded support. Independently prepared homogeneous initial data must also be accounted for when interpreting the complete solution.

References

• N. D. Birrell and P. C. W. Davies, Quantum Fields in Curved Space, Cambridge University Press, 1982, doi:10.1017/CBO9780511622632 — scalar solution spaces and particle notions.
• W. Pauli and V. Weisskopf, “Über die Quantisierung der skalaren relativistischen Wellengleichung,” Helvetica Physica Acta 7, 709–731, 1934 — quantization of the charged scalar field.
• D. Tong, Quantum Field Theory, University of Cambridge lecture notes, 2006–2007, §2, Free Fields — energy, charge, and scalar field quantization.