Dear Prof. Singh,
Thank you very much for reading and commenting.
You said "I have always been puzzled by the significance and interpretation attached to the delayed choice experiment. Does it tell us something new about the conventional view of quantum theory, which we already did not know without this experiment?"
I think that Wheeler's delayed choice experiment doesn't say something that is not implicit in the known experiments. Its great merit is, I think, pedagogical: it emphasizes a feature which, otherwise, is ignored and rolled from one corner of the mind to another, to avoid confronting it. The essence is that what happened in the past depends on how we prepare the measurement device now. Wheeler liked the spectacular conclusion of the observer participance, probably because he liked his conclusion of 'it from bit'. I think we can limit this to the experimental setup, rather than extending it to the observer (although the observer chooses what to observe). He wanted to conclude that this proves there's no 'it', and 'it' is inferred from the 'bit', while I prefer to restore reality, the 'it'. My claim is that 'it' is something that prevents 'bits' from contradicting one another, a 'reality check'. But, the price to restore realism is to make it dependent on the context, and by this I mean future measurements. I like to look at this as a 4D universe, in which events at various positions and moments in time constrain one another (global consistency). In quantum phenomena, when we develop the events in time, the 4D constrains manifests as if the present depends on what we measure in the future (delayed initial conditions).
Does this dependence of the past on the future measurements persist in realistic approaches like GRW and deBroglie-Bohm? I think it does, and I would refer here to Bell's and Kochen-Specker's theorems. Some claim realistic approaches like dBB are ruled out by such theorems. I don't think so, but the price is the same: to admit that the experimental setup constrains the past. Otherwise, the dynamics of GRW and dBB is not contradicted. In fact, I think that even the unitary evolution, as in the Schrodinger's equation, can be maintained without discontinuous collapse, if we accept that the initial conditions are delayed, or that they have to include the future experimental setup (superdeterminism). No reference here to observers, but only to measurement device. Restoring unitary evolution in the theory (without tricks like "unitary evolution is preserved, if we include all the branches corresponding to the different outcomes") is much more difficult than in modified dynamics or hidden variables, because unitary evolution is much more rigid. But even if there is no proof that unitary evolution is preserved, at least we know that it is not obligatory to be violated - the collapse is not necessarily discontinuous.
Thank you for the link to he conference Quantum theory without observers III. Currently I am watching The Quantum Landscape 2013.
Best regards,
Cristi