Has the total entropy of the universe remained constant throughout its history?

[chrisina]

something has been puzzling me for a while :
is it possible to think that the total entropy of the universe (ie that of all the microstates of matter, radiation AND vacuum) has remained constant throughout its history. I other words, that the vacuum has transferred it's entropy to what makes the universe today ?

 

[marcus]

a uniform distribution of matter corresponds to a very low entropy geometry (gravitational field)

it means the geometry is perfectly flat or uniformly curved----this is extremely unlikely

however from a point of view of classical matter thermodynamics---the behavior of gases at different pressure and temperature and stuff like that---the uniform distribution of matter is very HIGH entropy and by far the most likely: if you release some gas into a room it will spread out to fill the whole room at uniform density and will not all concentrate over in one corner.

therefore the evolution of the universe presents a subtle thermodynamics.

the only person I know who has attacked this problem with full force is Roger Penrose.
I do not like the answer he has come up with but at least he has conjectured SOME answer.
there is an online video lecture he gave at Perimeter Institute if you like video lecture.
I saw him give the same talk live at UC Berkeley. His hand-drawn slides are visually impressive and his delivery is entertaining.

For Penrose, the entropy of the universe is always inreasing because of gravitational CLUMPING. The field starts out nearly smooth and becomes more and more BUMPY as stars and galaxies collect

but finally all the stars die and all the crâp collects into black HOLES and these black holes eventually EVAPORATE and then there is nothing but empty space with uniformly distributed RADIATION in it

and so the ultimate dominant form of entropy is this huge entropy of radiation left over from evaporating black holes and all the background radiation but at least the gravitational field is SMOOTH AGAIN!

and so (here Penrose becomes a mythologist or scenario-imagineer, as important scientists sometimes do become) in some patch of perfectly smooth gravitational field A NEW UNIVERSE BEGINS not with a bang but perhaps simply with a redefinition of scale because where there is no matter perhaps there is no scale. He waves his hands here, in the age-old way of storytellers around the nomad campfire in the desert.

We always want to be told that a new universe begins somehow. It is part of the hero myth that is our STORY. Linde also includes that in his story at the University of Paris, as a thrilling punchline at the end.

 

[marcus]

both Sean Carroll and Lee Smolin have scenarios which are also
reproductive cosmology scenarios
in which the entropy is made to continually increase
by the universe making babies

in both cases the baby is made by a black hole (as seen by an observer in the old) which is an inflationary big bang (as seen by an observer in the new)
and there is no problem of where all the matter comes from because inflation takes care of that.

the difference between Carroll and Smolin is that Smolin was about 10 years earlier (1994) and uses astrophysical black holes and makes falsifiable predictions

Carroll on the other hand is more recent 2005-2007 and uses black holes caused by quantum hiccups and does not as far as I can see make concrete predictions that can be tested by astronomical observation.

the good thing about Carroll is that he is kind of a weathervane of the shifting winds of fashion and so his coming out with a Reproductive Cosmology scheme is an indication that Reproductive Cosmology is becoming acceptable. The key thing, though, is having a scheme which leads to some kind of testable hypothesis that observational astronomers can get to work on.

 

[chrisina]

But think about this (Penrose is the greatest), if spacetime is quantized, the smoother is the sphere, the higher is the information entropy on its surface.

 

[Jim Kata]

Sorry, I'm new here, and I don't know near as much Physics as all you guys, but I had a question very similar to the one in this post. So, I'm just going to add it on. There's all these questions about dark matter, and dark energy, and if dark matter and dark energy are true it would imply that we only understand 5% of what makes up the universe. My question relating to this post is could entropy be used to explain the accelerating expansion of the universe? On kind of an off topic I heard in passing about Rovelli's and Conne's concept of the thermodynamic time. I didn't really read their theory, but my question is does anyone think that the second law could explain causality or vice versa?

 

[Fra]

I think there are many interesting reflections around this, but it seems there are several different ways of digging the way forward and clearing the fog.

At least in my personal opinon, if you are interested in entropy reasoning, and want to take it beyond the classical (successful) context of stat mech and chemistry, a very exciting and promising route is by means of information geometry, various "maxium entropy" methods, and bayesian reasoning.

Ariel's papers is pleasant reading
http://www.albany.edu/physics/ariel_caticha.htm

Is some pages which contains various links to the topic of Max Ent methods.
http://www.cscs.umich.edu/~crshalizi/notebooks/max-ent.html

IMO, a deeper usage of entropy should be avoided unless the meaning of entropy is analyzed, and one realizes that the more sensible versions where entropy is relative. Which ultimately leads into information theoretic reasonings on learning.

Then, I think the basic machinery can be applied to many different things.

The above links may not contain anything directly applicable to cosmology, but I think the basic question posed there, should be kept in mind when applying the ideas, because often classical entropy methods are really special cases as they contain a few assumptions, for example uniform priors that are dependent on the representation (model or coordinate system).

/Fredrik