PDA

View Full Version : [Ed] Inflation vs Cyclic Model


Skwinty
14th April 2009, 12:01 PM
This is an excerpt from the LCDM woo thread.
I would like to discuss these models outside of that thread as I don’t believe that either of the models qualify in any manner as woo.

Originally Posted by Sol Invictus
Their idea is to introduce a pre-big bang phase, i.e. a phase during which the universe is contracting. When it gets very dense it "bounces", meaning the contraction reverses and the universe begins to expand. It turns out that one can construct models in which, during the contracting phase, the conditions are such as to generate some scale-invariant perturbations. Then with certain assumptions about the bounce, those perturbations form appropriate initial conditions (similar to those produced by inflation) to feed into standard Lambda-CDM and match observations today.

There are a number of very serious problems with this idea. The worst is that not only do we not know of a force that would reverse the contraction, the required properties of such a force make it almost certainly impossible. It's a little hard to explain this, but essentially any force which grows strong enough to resist gravity when the universe is so dense must have extremely bizarre properties, and those properties appear to rule it out (i.e. it would have huge effects even today, which we do not observe). There is some controversy about that, but my impression is that most people in the field do not think such a bounce is possible (or more accurately it might be possible, but it will require some totally unknown and new physics that somehow avoids the problems, and since it's so uncertain how that would work one cannot trust any model that relies on it).

I'll also add that this is an old idea, one that's been proposed many times in many forms since 1915. It never succeeds, because it always runs into this same problem.

I have read “The Endless Universe” as well as the papers below.
http://arxiv.org/abs/astro-ph/0404480v1 (http://arxiv.org/abs/astro-ph/0404480v1)
http://www.physics.princeton.edu/~steinh/cyclicFAQS/index.html (http://www.physics.princeton.edu/~steinh/cyclicFAQS/index.html)
http://arxiv.org/PS_cache/hep-th/pdf/0408/0408083v1.pdf (http://arxiv.org/PS_cache/hep-th/pdf/0408/0408083v1.pdf)
http://arxiv.org/abs/astro-ph/0204479v1 (http://arxiv.org/abs/astro-ph/0204479v1)
The points I would like to raise are as follows.
1.I agree that earlier models of a cyclic universe had loads of problems as indicated by Tolman and others.
2.I agree that string and M-theory are still in their infancy and are not complete as yet, as far as a theory of everything.
3.However, the cyclic model as posited by Steinhardt and Turok seems to overcome or explain away the problems experienced by earlier models.
4.The cyclic model agrees with the predictions of the inflation model: these being near scale invariance of temperature (COBE), the flatness of space through numerous ground based experiments and WMAP, matter and dark matter distribution in CBMR (Adiabaticity) and the random noise characteristics of this distribution (Gaussianity). The jury is still out on the gravitational wave predictions, hopefully the Planck mission will produce the goods one way or the other.
5.The cyclic model does away with inflationary energy and replaces it with dark energy, it is parsimonious with no ad hoc additions and seems to be the only serious contender to inflation.

Any comments appreciated.

sol invictus
14th April 2009, 12:38 PM
I have read “The Endless Universe” as well as the papers below.
http://arxiv.org/abs/astro-ph/0404480v1
http://www.physics.princeton.edu/~steinh/cyclicFAQS/index.html
http://arxiv.org/PS_cache/hep-th/pdf/0408/0408083v1.pdf
http://arxiv.org/abs/astro-ph/0204479v1

Please note that every one of those papers is by the same two authors - who happen to be the most vigorous proponents of these ideas.


3.However, the cyclic model as posited by Steinhardt and Turok seems to overcome or explain away the problems experienced by earlier models.


I don't think so, and I believe that opinion is shared by most experts in the field. The problems are still there, they have not been solved - at best they have been swept under the rug. If you look, you'll find many of papers referencing the ones you listed and pointing out various problems with them.


4.The cyclic model agrees with the predictions of the inflation model: these being near scale invariance of temperature (COBE), the flatness of space through numerous ground based experiments and WMAP, matter and dark matter distribution in CBMR (Adiabaticity) and the random noise characteristics of this distribution (Gaussianity). The jury is still out on the gravitational wave predictions, hopefully the Planck mission will produce the goods one way or the other.


That's true only if you accept that the matching through the bounce makes sense. If you do, you have something that isn't worse than inflation in that regard.

5.The cyclic model does away with inflationary energy and replaces it with dark energy, it is parsimonious with no ad hoc additions and seems to be the only serious contender to inflation.

That's not accurate. Both (standard) inflation and dark energy can be explained with the same scalar field with a relatively simple potential. But in order to get a cyclic model to work, one needs both such a scalar with a very fine-tuned potential AND a new - and highly problematic - force that causes the bounce. More broadly, you also need to posit a whole new stage of the evolution of the universe (contraction) for which there is zero observational evidence. And if you look at the detailed, quantitative models people have proposed to actually try and realize these ideas (this (http://arxiv.org/abs/hep-th/0702154), for example) you'll find all sorts of extra ingredients.... and a model that fails anyway for a several independent reasons.

It's an interesting idea, and I'm glad people are working on it - but I'm very skeptical they will succeed without some major breakthrough.

Skwinty
14th April 2009, 01:09 PM
Thanks Sol,
I will read the paper you link to. With regard to the papers I quoted, the latest one is 2004. I have yet to find a newer one by Steinhardt and Turok.
In fairness to the authors, they do make it clear that there are still problems with the model, but they are positive that these issues will be resolved.All in all a fascinating subject, how so much can be predicted about events that happened billions of years ago

Tubbythin
14th April 2009, 04:27 PM
Any comments appreciated.

That yellow highlighting is hideous. (Other than that, no. I'd be way out of my depth.)