Age of universe -- Probability why it is so young?

[logico]

The weak anthropic principle is a truism: given that we exist, a universe capable of supporting our existence must exist. But that does not mean that given the existence of this universe, we must exist. More importantly (re the original post) if the universe is in a highly improbable state, that requires explanation, irrespective of the fact that we observe that state. If we observe an ice cube in a container of boiling water, we know (ultimately from statistical mechanics) that this is a very young system. In that case we explain its existence by human agency. The universe we live in is in an overwhelmingly more improbable thermodynamic state. (Personally, I baulk at attributing it to a random fluctuation - pretty much on the same grounds as Feynman, 'The Character of Physical Law'. As an atheist, I find this a major problem; or, more positively, a key datum.)

 

[Chalnoth]

The universe we live in is in an overwhelmingly more improbable thermodynamic state. (Personally, I baulk at attributing it to a random fluctuation - pretty much on the same grounds as Feynman, 'The Character of Physical Law'. As an atheist, I find this a major problem; or, more positively, a key datum.)

The entropy issue is an open area of research among theorists, though there are a number of early-universe models which don't have issues with entropy.

One of them is the Carroll-Chen model, which assumes a small but non-zero cosmological constant with empty space as a pseudo-equilibrium state, and that the small temperature that results from that cosmological constant in empty space occasionally produces new inflating regions of space-time. In this model, the very low entropy of our early universe isn't such a huge drop in entropy from the "parent universe" because of the low entropy density of the parent.
http://arxiv.org/abs/hep-th/0410270

The theory isn't perfect, and there are dozens of other possibilities. But yes, the low entropy of our early universe is something that needs to be explained, and the weak anthropic principle doesn't help in this particular case. The reason it doesn't help here is that the problem with the low entropy of the early universe is the Boltzmann Brain issue: we have observers either way, and we need to explain why we observe a real universe.

But the weak anthropic principle may well be a (nearly) complete solution to the cosmological constant problem. Note that its (rough) measured value was predicted on anthropic grounds nearly a decade prior. All that is needed to complete the solution is a theory which allows for a huge variety of different values of the cosmological constant (string theory is one candidate here).

 

[logico]

One of them is the Carroll-Chen model, which assumes a small but non-zero cosmological constant with empty space as a pseudo-equilibrium state, and that the small temperature that results from that cosmological constant in empty space occasionally produces new inflating regions of space-time. In this model, the very low entropy of our early universe isn't such a huge drop in entropy from the "parent universe" because of the low entropy density of the parent.

This model, like your suggested string solution, or like the Many Worlds Interpretation of quantum mechanics, smacks of a fudge philosophically. Basically, if x is highly improbable in the set (x, y, ...), simply devise a vast (even infinite) hidden set-of-sets within which the set where x is chosen from x, y, z... is almost inevitable. Such arguments assume, I think, that what is "almost inevitable" is "highly probable" - but what exists with certainty or high probability is not selected from a large set with high probability.

But, quite apart from that concern, I think that all these multiple hidden universe theories are in conflict with the commonsense principles Newton established for the simplicity/unity of explanation and reliance on observation.

 

[Chalnoth]

This model, like your suggested string solution, or like the Many Worlds Interpretation of quantum mechanics, smacks of a fudge philosophically. Basically, if x is highly improbable in the set (x, y, ...), simply devise a vast (even infinite) hidden set-of-sets within which the set where x is chosen from x, y, z... is almost inevitable. Such arguments assume, I think, that what is "almost inevitable" is "highly probable" - but what exists with certainty or high probability is not selected from a large set with high probability.

What you've said here really has nothing to do with the model.

But, quite apart from that concern, I think that all these multiple hidden universe theories are in conflict with the commonsense principles Newton established for the simplicity/unity of explanation and reliance on observation.

That's probably because you haven't bothered to take the models on their own terms.

There may be problems with the Carroll-Chen model, but this isn't one of them.

 

[logico]

Chalnoth

You are saying that the very low entropy universes produced in the C-C model are the norm? My apologies - I thought there were a whole range of types, and that very low entropy universes would therefore be very rare.

Leaving only the problem of why the parent universe has extremely low entropy. I am well aware that it is presumed to be in a quasistable state, but all such metastabilities in nature require explanation. It's not as simple as, "Oh, hey - metastable state, no prob." I have been thinking about these "hidden stabilities" (as Jacob Bronowski called them in his famous passage on the thermodynamics of life) for many years, and familiarity has not bred contempt.

Still, as I just said, where is the great explanatory power of a theory which explains a very low entropy universe

 

[logico]

... (Whoops) ...which we observe by hypothesising an even more extreme low entropy uiniverse which we do not and perhaps in principle can never observe? I think that has been tried before by Moses...

 

[Chalnoth]

In that model, the parent universe has very high entropy. It just has low entropy density. The low entropy density is an intrinsic factor of the small cosmological constant.

Regardless, it's one model among a very large number of them. Reality might be kinda sort of similar to this model, but it's unlikely to be entirely accurate (purely from counting: there are too many speculative early-universe models for anyone of them to be likely).

 

[Chalnoth]

... (Whoops) ...which we observe by hypothesising an even more extreme low entropy uiniverse which we do not and perhaps in principle can never observe? I think that has been tried before by Moses...

Not quite. The proposal is that of a pseudo-equilibrium state. That state is the likely far future of our own universe: empty space with a small cosmological constant.

Proposing an equilibrium state from which the universe occasionally fluctuates is one of the most reasonable ways to solve this problem. And the far future of our own universe is the only sensible choice for such an equilibrium state. But it has, historically, been fraught with difficulties, as most such proposals suffer from the Boltzmann Brain problem.

 

[logico]

In that model, the parent universe has very high entropy. It just has low entropy density. The low entropy density is an intrinsic factor of the small cosmological constant.

Regardless, it's one model among a very large number of them. Reality might be kinda sort of similar to this model, but it's unlikely to be entirely accurate (purely from counting: there are too many speculative early-universe models for anyone of them to be likely).

Yes; I have reread a couple of summaries. This is a vast universe which is basically a hard vacuum with a very low radiation density: then the baby universes arise as quantum (rather than Boltzmann) fluctuations, and somehow inherit the low entropy density but (as I guess is true of all inflationary models) not the emptiness. I can't say I like it any more than Boltzmann's idea - which at least did not involve new entities and mechanisms. That's my Newtonian bias talking. (The latest I read on string theory, if it can be got to work, it will require scores of new independent parameters. For me, that CAN'T be the way to make a universe.)

More much later - got to go. (BTW if someone invents a new speculative cosmology, does that really make the others less likely...)

 

[Chalnoth]

Yes; I have reread a couple of summaries. This is a vast universe which is basically a hard vacuum with a very low radiation density: then the baby universes arise as quantum (rather than Boltzmann) fluctuations,

Not quite. These are thermal fluctuations. Quantum mechanics plays a role (because inflation is fundamentally quantum-mechanical). But they're best thought of as thermal fluctuations in the Boltzmann sense, except with an underlying quantum theory.

I can't say I like it any more than Boltzmann's idea - which at least did not involve new entities and mechanisms. That's my Newtonian bias talking.

Huh? It's not possible to have a sensible theory of how our region of space-time came to be without invoking new physics. The fundamental inconsistency of classical thermal fluctuations being a source of our universe makes that theory impossible.

(The latest I read on string theory, if it can be got to work, it will require scores of new independent parameters. For me, that CAN'T be the way to make a universe.)

String theory has one free parameter, the string tension. There are no other parameters.