Quantum Emission and Light Speed | Generated by AI

Home 2026.07

Question: My friend refines the point: “light speed invariance” (光速不变) means that in any local reference frame — whether that of the emitter or the observer — a light beam is measured as constant c; this is a basic postulate. Going deeper, though, the emission process is a quantum process, while relativity describes the macroscopic world — these are different scales that haven’t been unified, so there’s some genuine tension here.

Answer:

The first part is correct and is exactly the content of the second postulate — both frames, emitter and observer, measure c. But the second move — “emission is quantum, relativity is macroscopic, quantum and relativistic physics aren’t unified, so this creates tension for the postulate” — is a category error. It’s conflating two very different open/closed problems in physics:

What’s actually unsolved: General Relativity (gravity, curved spacetime) + Quantum Mechanics = quantum gravity. Genuinely unsolved. String theory, loop quantum gravity, etc. are attempts.

What’s actually solved and extremely well tested: Special Relativity + Quantum Mechanics = relativistic Quantum Field Theory (QED, QCD, the full Standard Model). This is not an open problem. QFT is built from the ground up to be Lorentz-invariant — the Lagrangian for the electromagnetic field, L = -¼F_μν F^μν, is manifestly Lorentz-invariant before you even quantize it. Quantizing it (canonical quantization or path integral) doesn’t introduce any tension with SR; it’s how you get the photon as a massless spin-1 excitation of that field, with dispersion relation E = pc baked in — which is the statement “photon speed = c in every inertial frame.” There’s no seam between “quantum emission” and “relativistic propagation” here; they’re two aspects of one Lorentz-invariant field theory.

Concretely, why the quantum-ness of emission doesn’t matter to the postulate:

For any massless particle, the on-shell condition from Lorentz-invariant kinematics is:

E² = (pc)² + (m c²)²,  m = 0  ⟹  E = pc  ⟹  v_group = ∂E/∂p = c

This holds identically whether the photon comes from a classical antenna, an atomic transition (discrete quantum jump), or a particle decay. The “quantumness” of the emission event only affects when and in what direction a photon is emitted (governed by matrix elements / Fermi’s golden rule) — it says nothing about the propagator’s dispersion relation once the photon is on-shell and free. Quantization changes the statistics of emission, not the kinematics of propagation.

And this isn’t just theory — it’s exactly what’s been tested with quantum-emitted photons from ultra-relativistic sources:

So the empirical record already directly covers “does quantum emission break the invariance,” and the answer is no — to parts in 10⁻⁴ or better even with sources moving at 99.975% of c.

Where the friend’s intuition is pointing at something real, just misapplied: the unresolved SR/QM-vs-GR tension is real and deep (quantum gravity), but it’s not “macro vs micro physics,” it’s specifically “curved spacetime dynamics vs quantum field theory in flat spacetime.” SR + QM is not a frontier problem; it’s the best-tested framework in all of physics (QED’s prediction of the electron’s anomalous magnetic moment agrees with experiment to ~10 significant figures). If you want to point your friend at the actual live seam, it’s holography / AdS-CFT, string theory, loop quantum gravity — not “light emission is quantum so maybe c-invariance is shaky.”

References:

Back Donate