DrChinese said:
Second, of course I don’t believe that the apparatus is entangled with photon B. They are “obviously” in a product state. No experiment is gonna show otherwise.
Assume Photon A is observed to be |- >. We already know photon B will be observed to be |+>. Exactly what spin State is the measurement apparatus now in, post measurement? Writing down an entangled state does not make it entangled.
Ah, I see now the source of confusion. When you say they are in a product state, you are talking about
one photon B and
one apparatus. Since the apparatus has shown a definite known state, its state effectively collapsed, which effectively destroyed the entanglement. In that sense, even BM accepts that there is an effective collapse and hence an effective destruction of entanglement.
However, that is not what we are talking about. We are
not talking about
one B and
one apparatus. We are talking about a
statistical ensemble. The entanglement is observed by measuring correlations, and for that purpose we need to repeat the measurement many times, so we are really dealing with an ensemble. In the ensemble the apparatus sometimes shows one outcome, and sometimes the other. Hence the state of the ensemble does not describe only one of the outcomes, but
both outcomes simultaneously. In other words, the system, viewed as a statistical ensemble, is in the superposition. It is this superposition at the ensemble level which is responsible for the entanglement. In other words, the ensemble is not in a product state, even though any individual B+apparatus system is.
To understand all this properly, Bohm is not very important. Much more important is Ballentine, who emphasizes that the quantum state is a description of a statistical ensemble, not of a single system. Without understanding that, the DCES experiment cannot be understood properly. More generally, the ensemble perspective is crucial whenever we do post-selection (as we do in the DCES experiment), because the
post-selection changes the ensemble. Without post-selection, we don't see any entanglement in the DCES experiment, do you agree? Hence, at an instrumental level, we may say that the post-selection creates entanglement. (Mathematically, post-selection is represented by a projection, and the projector is not a unitary operator.) That's how even the apparatus becomes entangled.
Or to make the long story short, at the operational instrumental level, entanglement is a property of an ensemble, not of an individual system. At the ensemble level the apparatus does not show only one definite outcome, so the system is not in the product state. Instead the system is in the superposition, namely, in an entangled state.