Thanks! That was very helpful.
That’s just the thermodynamic arrow of time, again. Most “arrows of time” that people can think of end up being the thermodynamic one.
There are three distinct arrows of time known in physics: The thermodynamic arrow of time, which has gotten plenty of discussion in this thread; the cosmological arrow of time, which is just an observation that the Universe is expanding but which doesn’t have any real fundamental significance; and the particle-physics arrow of time, which as you mentioned involves the weak force, but which is so incredibly subtle and small an effect that it’s never yet been directly observed. Any other arrow of time must just be a variation of one of those, and it’s almost always a variation of the thermodynamic one.
I might be missing the obvious, but could you explain how?
More spread out = higher entropy.
OK. So I’ve had a few beers. I might be a bit dense right now. And what you’re saying is obviously correct. But I’m still not sure if that means that the electromagnetic arrow of time is the same as the thermodynamic one – for instance, in Boltzman’s H-theorem, the derivation of the second law essentially rests on the assumption of retarded (or at least, not advanced) interactions in requiring a molecular chaos, i.e. uncorrelated particle movement before a collision.
I’d think you could even argue that using this view, you could do away with Loschmidt’s paradox entirely – you can’t have pre-correlated particle movements with retarded interactions, and thus, a velocity reversal in such a way that the second law is violated ought not be possible.
Again, I might just not have a clue what I’m talking about, and I hope I’m not engaging in a highjack of this thread, but I’d really like my ignorance fought on this subject, as you folks like to put it round here.
Sure you could. It’s just that the correlations would have to either be coincidental, or the result of some contrivance to make them be correlated in the right way. But the coincidence is highly unlikely, and the contrivance would involve enough entropy arranged the other way that the “forward” process would be a violation of the Second Law.
Hrmies. I fear I’m not expressing myself very clearly. I’ll blame the language barrier, beer, and nicotine withdrawal when I look like a fool in the morning, but, to rephrase, what exactly is wrong with Cramer’s view as expressed in this paper that the thermodynamic arrow of time might just as well be a consequence of the electromagnetic one as the other way around?
All the energy jolting together in such a way to bring shards of glass together into a whole glass does not violate any law of physics. But it does violate the laws of statistics (as much as you can call them laws). You would indeed be able to identify a film of this as running backwards, but that’s because it’s so incredibly unlikely, not impossible. If you “zoomed in” on any small group of particles in this film you wouldn’t see anything suspicious and wouldn’t know it was being run backwards.
Brian Greene addresses this paradox in his 2004 book The Fabric Of the Cosmos: Space, Time, And The Texture Of Reality.
He starts by taking this statistical argument (that higher entropy states are vastly more likely, both in the past and the future) to its logical conclusion: The universe is much much more likely (mind-blowingly – thanks Pasta) to be at a highly disordered state at all times. Thus, it is much more likely that the universe just recently dipped down from a state of high disorder to its present low-entropy state (rather than being at a low-entropy state all along) and that all our memories otherwise are wrong. A sort of scientific Last-Thursdayism. Not only that, but as each instant passes by, it is mind-blowingly more likely that our memories of the last instant are wrong, and that this well-ordered universe we see now just popped into being last instant, instead of two instants ago. In other words, statistically, it is much more likely that our memories are wrong and we are just unlikely bits of matter temporarily flung together before the universe falls back to its much more probable state of disorder in an instant or two.
In this universe, we can obviously see that science is impossible. However, not only can we not trust our memories or experiments that gave us our knowledge of scientific laws, we can’t eve trust the assumptions and reasoning that we just used to lead us to this conclusion. I’m not sure if it’s supposed to be reductio ad absurdum, hence our premise that the universe was more likely to be disordered in the past must be wrong, or the simple practicality that we can’t even do science at all without assuming that the past was more ordered than the present or future.
Just thought I’d share that. I can quote bits of Greene’s writing on this topic if requested. Frankly, this sounds eerily like the ‘proofs’ of the parallel postulate, with their ‘obviously crazy’ (non-Euclidean) conclusions. I’m not sure how to fix it though.
I forgot to add that he (Greene) chucks the whole shebang out the window by postulating an initial condition at t=0 of minimal entropy.
I’m not sure that there is anything to warrant that assumption, though, except that it allows us to do science without the entire house of cards collapsing.