What happens during a measurement in CSD?

The short answer

The textbook's answer is famously two answers, smooth evolution until a measurement happens, then a sudden jump, with no rule for when to switch from one to the other. CSD's answer has one law and three stages, each of them ordinary physics.

The first stage is de-isolation, as measuring something means coupling it to an apparatus, which deliberately ends its isolation. The second is sorting, as the joint deterministic dynamics of system and apparatus carries the point of the arena, the underlying state space, into one of the outcome basins, regions fixed by the apparatus alone, before the system ever arrived. The third is the record, as which basin the trajectory occupies is a physical fact about the arena, stable, re- readable, and protected from distant interference by a theorem showing that no influence from far away can reach it faster than a fixed speed allows.

The state's jump is then bookkeeping. You learned which basin, so you update the description, and nothing physical collapsed, as your summary caught up with the world.

The longer answer

Each stage is formalised at its own level of the programme. De-isolation flows exist as constructed objects, and they're deterministic, they preserve the arena's volume, and they're provably not the identity, so they do something. They realise dynamically what textbook quantum mechanics calls the measurement dilation, the description of a measurement as ordinary reversible evolution on a larger system that includes the apparatus, and the volumes of their pointer blocks equal the Born weights, the textbook probabilities. The basins are the record layer's context-fixed partition of the arena, and they're Born-weighted by theorem. The Luders update rule, the textbook prescription for how the state should be revised once an outcome is known, is derived as the correct conditional bookkeeping both for sequences of measurements and for mixed states.

The stated open frontier is the junction between the stages, deriving the record mechanism at the level of the fibre from the de-isolation dynamics, rather than exhibiting each side separately. The programme calls this the record layer's remaining work, and it says so in public.

What is actually proved

measurement_flow_born_frequency covers the dynamics tier, globalBasin_born and the disjointness and exhaustion theorems cover the kinematic tier, and record_lightcone covers the record's stability.

The Luders tier covers sequential conditioning, and the einselection theorems explain why pointer readings are the stable alphabet.

Related entries

• De-isolationNothing collapses, as a measured system stops being isolated, and measurement is what de-isolation dynamics looks like from inside.
• Measurement flowA deterministic, volume-preserving flow carries out a measurement, and the volumes of its pointer blocks are the Born weights, for every preparation.
• CollapseIn CSD the quantum state never jumps, and that's a theorem about volume-preserving maps rather than a preference among interpretations.
• Record layerA measurement is over when the world itself carries which outcome happened, and the record layer is where that fact lives, on the arena and not in anyone's notebook.
• Measurement problemQuantum mechanics contains two incompatible rules of change and no instruction for when to switch between them, and CSD replaces the switch with theorems about what no dynamics can do.