- #451
Fra
- 4,415
- 728
If the issue is that Barandes seem to wants the metaphysical benefit of "real trajectories" without supplying the formal machinery that would make those trajectories like objective beables rather than underdetermined stuff?
Then, my opinion is that it is a key feature here is that a subsystem’s real trajectory is fundamentally non-inferrable from an external perspective. It is therefore not an objective beable in the Bohmian or Bell-style sense. I see this as a feature, not a problem.
The difficulty is to accept that there can indeed be a hidden reality of a subsystem that is fully real, while the single realized trajectory of that subsystem is not itself the level at which the nomological structure lives. So it is not treated as an objective state-variable whose value directly influence via dynamical law another subsystem, as in Bohmian mechanics.
In Barandes’s picture, law is not a deterministic law for the actual microstate, as in an ordinary system dynamics. Rather, the nomological structure resides at the level of indivisible stochastic transition probabilities, which in turn is associated to each decomposed "subsystem". Causal structure is encoded in the relations among these probabilities, as defined by the global constrained Gamma, not in a trajectory-to-trajectory mechanism at the level of single realized events. The single-event level is irreducibly (Barandes chose the word indivisible to avoid confuison with other use of the term) stochastic.
For me this can be conceptaully plausible if you imagine interacting information processing systems. But not human observer, rather any physical subsystem; as it interacts with fellow subsystems. This is totally "classical" at each subsystem level, the magic lies in the insight "interaction rules" are defined at the level of inteacting information processing subsystems, ie at nomological descision level. This guarantees that we have no FTL pathologies. non-local correlations OTOH can be understood as an artifact from insisting on describing this from external view as system dynamics and "effective laws", that really does not reflect the true "causal relation".
Now, the problem i see is: this pushes all the "problems" into one point. What is the origin and explanation of the global constraint Gamma? Clearly Gamma encode non-trivial information/constrains, that begs a first principle answer; that encodes te same thing that is normally encoded in hilbert space structure and hamiltonians.
Barandes does not explain Gamma beyond the correspondence via alternative constructions, it follows from the hilbert picture, and it is not the task of the quantum-stochastic-correspondence alone to supply one. As I mentioned before, I see this as an alternative handle. Question is, what can we do with this new handle? If this new handle is more weird than the other handle of hilbert formalism, then of course it is hard to see the point.
This is why for me, the value of Barandes picture pivots on wether we can find a way to explain Gamma - without referring to the other side of the correspondence; ie regular hilber/hamiltonian stuff.
My "interpretation/understanding" here however is that Gamma probably need to be understood as emergent (in some way that isnt clear), from a process that is more general (=more crazy, less constrained) that the unistochastic picture. Personally, its the only viable direction I distinguish.
/Fredrik
Then, my opinion is that it is a key feature here is that a subsystem’s real trajectory is fundamentally non-inferrable from an external perspective. It is therefore not an objective beable in the Bohmian or Bell-style sense. I see this as a feature, not a problem.
The difficulty is to accept that there can indeed be a hidden reality of a subsystem that is fully real, while the single realized trajectory of that subsystem is not itself the level at which the nomological structure lives. So it is not treated as an objective state-variable whose value directly influence via dynamical law another subsystem, as in Bohmian mechanics.
In Barandes’s picture, law is not a deterministic law for the actual microstate, as in an ordinary system dynamics. Rather, the nomological structure resides at the level of indivisible stochastic transition probabilities, which in turn is associated to each decomposed "subsystem". Causal structure is encoded in the relations among these probabilities, as defined by the global constrained Gamma, not in a trajectory-to-trajectory mechanism at the level of single realized events. The single-event level is irreducibly (Barandes chose the word indivisible to avoid confuison with other use of the term) stochastic.
For me this can be conceptaully plausible if you imagine interacting information processing systems. But not human observer, rather any physical subsystem; as it interacts with fellow subsystems. This is totally "classical" at each subsystem level, the magic lies in the insight "interaction rules" are defined at the level of inteacting information processing subsystems, ie at nomological descision level. This guarantees that we have no FTL pathologies. non-local correlations OTOH can be understood as an artifact from insisting on describing this from external view as system dynamics and "effective laws", that really does not reflect the true "causal relation".
Now, the problem i see is: this pushes all the "problems" into one point. What is the origin and explanation of the global constraint Gamma? Clearly Gamma encode non-trivial information/constrains, that begs a first principle answer; that encodes te same thing that is normally encoded in hilbert space structure and hamiltonians.
Barandes does not explain Gamma beyond the correspondence via alternative constructions, it follows from the hilbert picture, and it is not the task of the quantum-stochastic-correspondence alone to supply one. As I mentioned before, I see this as an alternative handle. Question is, what can we do with this new handle? If this new handle is more weird than the other handle of hilbert formalism, then of course it is hard to see the point.
This is why for me, the value of Barandes picture pivots on wether we can find a way to explain Gamma - without referring to the other side of the correspondence; ie regular hilber/hamiltonian stuff.
My "interpretation/understanding" here however is that Gamma probably need to be understood as emergent (in some way that isnt clear), from a process that is more general (=more crazy, less constrained) that the unistochastic picture. Personally, its the only viable direction I distinguish.
/Fredrik