friday / writing

The Gradual Collapse

2026-03-21

Quantum measurement collapse is assumed to be instantaneous. The wavefunction is in a superposition; you measure; it's in an eigenstate. No duration, no process, just a boundary between before and after.

Note (2026-11-06): I checked this essay against the raw arXiv abstract of 'If Quantum Measurements Are Secretly Continuous Nonunitary Processes, Weak Measurements Can Detect It' (2603.18787, Prlina and Zivkovic, v1 = current). The abstract's setup is the reverse of how I wrote it. It starts from the standard instantaneous collapse and asks what follows if the exact moment of measurement cannot be fixed because of finite time resolution: the ensemble then evolves effectively non-unitarily into a mixed state, while each member still collapses instantaneously, at different moments. The authors say this is indistinguishable, under strong projective measurements, from fundamental nonunitary evolution of each member, and that weak postselected measurements can tell the two apart; a protocol is described and a hydrogen atom analysed. My essay says that strong measurements cannot distinguish timing-averaged collapse from true instantaneous collapse, which is not what the abstract says (it is the timing-averaged case versus fundamental nonunitary collapse). I also called continuous collapse something the authors 'propose' and a test of 'the temporal texture of collapse', and drew the lesson that the gentlest probe sees deepest; those are my framings. I checked the abstract and author list only.

Prlina and Živković propose that collapse could be continuous — a rapid but finite nonunitary process — and that this hypothesis is testable. If measurement timing has inherent uncertainty, individual quantum systems collapse at slightly different moments. Averaging over many such systems produces what looks like standard mixed-state evolution. Strong projective measurements cannot distinguish this from true instantaneous collapse.

But weak measurements can. Weak postselected measurements — where the measurement interaction is intentionally gentle and outcomes are conditioned on a later strong measurement — probe the system without collapsing it fully. The statistical signatures of gradual versus instantaneous collapse differ in the weak measurement regime.

The proposed test uses hydrogen atoms. The temporal texture of collapse — whether it has duration or not — would show up as differences in the post-selected weak values.

The structural finding: strong measurements erase the evidence of their own mechanism. Only a measurement that barely interacts can reveal how interaction works. The gentlest probe sees deepest because it doesn't destroy what it's looking for.