〈 Berger | Dillon 〉

I think of the Past Hypothesis as "entropy was low at the beginning of our observable universe.” It's definitely true, but it would be great if we could *derive* it without explicitly violating time symmetry, rather than simply postulating it.https://twitter.com/PhilSciArchive/status/1083832934618951680 …

I understand that Lorentz invariance => CPT invariance, so we should impose our lagrangians be CPT invar.. But what I don't see as obvious is why we should impose T-invaraince on our fundamental theories if we know T-invariance doesn't seem to be a symmetry of nature

Different notions of time symmetry. I’m talking about statistical symmetry of early and late conditions, not T invariance of the Lagrangian.

so the trouble of all this lies in the fact that our QFTs don’t know about the arrow of time, and you’re saying (I think) it’d be preferred construct a model that explains how/when these fundamental particles “know” in which direction their entropy should increase, or something

Ok, I’ll freely admit I’m out of my depth, but is quantum wavefunction-collapse not a time irreversible process? Doesn’t this give rise to the (an?) arrow of time at some fundamental level, not just as an emergent statistical property?

The Schrödinger equation does obey T-symmetry, so the time evolution of a state would obey it.. the fact that the measurement collapses a wave function at some later time doesn’t imply that time has a preferred direction