[griffzhowl]

Thanks, I had a look at Maudlin's book. It seems the distinction between signaling and causation is that there might be some kind of nonlocal causation that we can't control and so can't use to send a signal.

Local causation is defined as in Bell's Theory of Local Beables, as the probability distribution for values at spacelike separated regions only being correlated with respect to the overlaps in their past lightcones. Or to put it the other way around, there's nonlocal causation if the probability distribution of values in one region depends on values in a spacelike separated region. That's what I'd call nonlocal correlation rather than causation but I guess that's just terminological.

This looks like a pertinent paper from Das but I haven't read through it yet

Arrival Time Distributions of Spin-1/2 Particles (2018) https://arxiv.org/abs/1802.07141