Ergodicity and the perception of time

[T S Bailey]

Boltzmann once stated that if the universe is ergodic then, in the event that the universe returns to a low entropy state, observers residing within the universe would always say that the lower entropy state was "the past" and the higher entropy state was "the future". My question is: I have heard creationists use the 2nd law of thermodynamics to claim that life cannot possibly arise in a universe which is always tending to higher entropy states. They are then corrected that it is only the total entropy which increases or stays the same in an isolated system, entropy may decrease in one area as long as that decreased entropy results in increased entropy somewhere else. So wouldn't the same be true in an ergodic universe? Perhaps the total entropy of the universe is decreasing, this should not mean that there are no systems within the universe where the entropy is still increasing, ie memories are still being formed.

 

[jfizzix]

The universe is very big, and very far from thermodynamic equilibrium.
The laws of thermodynamics are only equipped to describe systems in equilibrium, though they may also show just how far away from equilibrium a system is.

As far as the laws of probability goes, it is astoundingly and overwhelmingly more likely than not that the Universe's entropy will increase toward some maximum.
As an example, you can consider a room with a wall in the middle, and air only on one side. If you poke a hole in the wall, air will rush through until both sides have roughly the same amount. There's nothing in the equations governing the trajectories of each molecule that forbids the reverse from happening, but you can contemplate the odds that at random every molecule of air in the room, just happens to have a leftward trajectory, so that all the air rushes to one side. On the scale of a dozen atoms, this is rather unlikely, but not exactly out of the question if you were to watch the room for a few hours. On the scale of a room full of sextillions of atoms (i.e., at atmospheric pressure), it's mind-bogglingly small. On the scale of the Universe, the odds are practically unimaginable.

Long story short, there's nothing truly forbidding patches of the universe from running seemingly backward in time, or even that the rest of the universe beyond our local patch could happen to behave that way, but if we're just talking about this happening due to random chance (and not relativistic time warps and such), then it is by a large margin one of the least likely things that could ever happen.