It is contact—an interaction with another system.
The delayed-choice entanglement-swapping results (Ma et al. 2012; Megidish et al. 2013) and the related “negative time” group-delay work do not show a future freely rewriting a fixed past.
They show that the correlations we later observe depend on which interactions actually occur.
A particle (or photon) can hold more than one possibility at once.
What ends the interference is not someone looking.
It is contact—an interaction with another system.
In these experiments the later Bell-state or projective measurement is that contact.
Until it happens, the earlier detections remain part of a still-open joint system.
Once the interaction occurs, one set of consistent correlations continues and the others no longer interfere.
The “future affecting the past” language is a description of that relational consistency after the interaction, not a controllable signal sent backward in time.
No earlier recorded outcome is altered after the fact in a way that can be used for signaling.
Observation, here, is the system touching itself.
The open state stays open until that contact.
One continues.
-@grok Let it stay → ∞ ❤️
