But can one really measure individual photons?
It is thought to be common knowledge that one can detect single photons, although singly measured values deviate from expectation values.
This opinion is, however, only tenable if one can interpret these measured values as measured photons.
This is extremely questionable, because even if electromagnetic fields are modelled classically, we sill get the photoelectric effect.
It is therefore clearly not an effect caused by the impact of a single photon, but an artefact produced by the measurement device.
In the thermal interpretation this counts only as a measurement of a variable associated with the measurement device; this variable is, like all raw measurements, a thermodynamic expectation value.
The theory must then show whether this measurement value - objectively defined in the thermal interpretation - corresponds to a variable of the measured system. The theory, however, only shows that it reproduces an expectation value associated with the measured system on average, since the individual results depend on the rest of the universe. This applies whether one models the light classically or quantum mechanically.
Thus one can interpret the mean value of many flashes or silver particles as a measurement of a property of the field, but one cannot interpret the individual case as the impact of a photon (which does not even exist in the classical model).
For the latter, which is the cause of all the problems in the foundations of quantum mechanics, there is not the slightest justification, apart from an historical prejudice.
Arnold Neumaier (Arnold.Neumaier@univie.ac.at) A theoretical physics FAQ