atyy said:
I don't totally agree with bhobba's way of stating the measurement problem, but I think I understand it. What he means is that if we just have a unitarily evolving wave function, we clearly have no outcomes unless there are some additional assumptions. So the job of any interpretation that solves the measurement problem is to add assumptions that produce the outcome. For example, many worlds adds the assumption that blah, blah, blah ...
"Many worlds"? Never heard of it.
I'll put my question another way. Rather than asking how measurement creates a proper mixed state, shouldn't we ask whether it actually does so?
Why do we need outcomes at all? We know that measurements create the
appearence of outcomes in the macro world, but the improper mixed state that bhobba referred to (and which does follow from unitary evolution) would, to my simple mind, solve this part of the measurement problem. I don't see the need to insist that the
appearence of a mixture must be explained by the mixture being proper. In fact, in the last paper bhobba referred me to, Hensen makes it perfectly clear (section 1.2.3.) that they (case 2 and 3) are indistingishable at the level of observation since they yield the same probabilities. So I'm left asking what role is played by a proper mixed state that cannot be played by an improper one? If decoherence can provide an improper mixed state in which the observer
apparently sees a particular outcome, why does this not solve the problem?
edit - Hensen's paper says that it superficially solves the outcome problem (bottom of page 39) but I don't understand the grudging "superficial". Nor the subsequent rejection of the ignorance interpretation.