A paper appeared on the arXiv last night entitled Primordial Inflation Explains Why the Universe is Accelerating Today by Rocky Kolb of Fermilab, together with Sabino Matarrese, Alessio Notari and Antonio Riotto. There’s also a Fermilab press release about it today.
I’m no expert on the subject, and would love to hear the opinion of someone who is. As near as I can figure out the idea is that what is really responsible for the effects that have been ascribed to a cosmological constant is a “cosmological perturbation” of the gravitational field. This is supposed to be a perturbation that expanded during the inflationary period so that its wavelength is now larger than the Hubble radius. According to the authors, this predicts a different magnitude vs. red-shift relation than the standard cosmological constant does, so their idea should in principle be testable.
If they’re right, this certainly will cause a huge problem for the whole “Landscape” business, which has advertised as its greatest success the “prediction” of a non-zero cosmological constant of the right order of magnitude.
Well sometimes important ideas emerge in an incomplete or confused form but this paper of Kolb et al. is going along the directions of the sort of thing you really want–a clean conservative scientific resolution of the cosmic acceleration/dark energy problem. They are approaching the problem in a sober way and hopefully they can develop and refine more details. Seems to have generated much interest already. If it works out then the whole landscape business will look increasingly silly.
Well, we had some discussion about this paper today in our lunchtime seminar. The consensus seems to be that they might be on to something, but there are a few things in the paper that don’t quite make sense. For one, around equation (7) they make the claim that their formula for \Psi(x,t) is valid to all orders in perturbation theory. Yet the longer paper that they refererence only gives a result to second order.
Of course, this is largely a function of the length of the paper (which is clearly intended for PRL), a lot of details get glossed over.
Still, the cosmo people here were interested, but skeptical. One group member also had an email from a big time cosmologist who said this was an ineresting idea…
It’d be nice if there was a longer paper, it’s hard to get much from a PRL. For those who want to look into it more, it seemed fairly clear that you should track back at least one layer to their previous work.
This is the only blog that has much interest because it’s about something other than the owner.
-drl
there is only one math blog by a professional mathematician and it is here http://www.matrix.ua.ac.be/ There are a number of other blogs by math instructors here
http://talldarkandmysterious.ca/
and here
http://learningcurves.blogspot.com/
This is an unrelated comment, but I’m just curious if you know any mathematicians who keep blogs? In general, I’d be interested in anything. Specifically, I’m interested in Algebraic Geometry, etc.
Just curious.
Well Q has a point – this hair-pulling over the CC, inflation, dark matter and other miraculous genies seems silly in view of Mk205 and NGC7603 inter alia.
-drl
It is very nice to see this mixture of Ideas here.
It will be nice to see what Sean has to say about this further.
Quantoken, this whole blog is built around an idea Peter does not believe.
Click on my name below to see why the experimental evidence for an “accelerating” universe is questionable and extremely weak, plus some useful links to the original ARXIV papers.
Peter does not allow discussion of ideas he does not believe, but said simply providing a link and discuss it some where else is fine. So go and see for yourself:
http://quantoken.blogspot.com
Quantoken
I would be happy, of course, if the true LAMBDA was zero after all, the expansion explained different, and we would return to the original C.C. problem “why it’s exactly zero”. But one must distinguish wishful thinking from reality. Longer-than-horizon fluctuations are always problematic.
If the effects like that are true, then these trans-Hubble oscillations may effectively be able to change many kinds of “constants” within the Universe. They’re frozen, they don’t change, yet the expectation values of the scalars may have an effect.
Cabi blogs on this issue too:
http://www.mit.edu/people/cabi/blog/2005/03/universe-accelerated-beyond-horizon.html#comments
Also Stephon Alexander pre-announces, in his blog
http://qd.typepad.com/17/, some advance in the question of the cosmological constant:
“Now I’m out of the cave with an answer.
Whats the answer?
I’ll tell ya tomorrow.”
For a different view about accelerated expansion and problem of cosmological constant see No cosmological constant after all?
Matti Pitkanen
Even though I posted my blog on this issue I wanted to join the discussion here by pasting my opinion:
“I should say that I did not like the proposed solution. We always thought that solution of the acceleration of the universe puzzle will also solve two other great mysteries. Namely; why we do not see a cosmological constant in the Einstein-Hilbert action although the known symmetry principles allow it, and why the QFT vacuum condensates (ground state) do not gravitate; or why they cancel out. If the acceleration is due to super-Hubble fluctuations, I think we will be in a worse trouble. At least they are offering observable differences from a pure cosmological constant and nature will say the last word.”
These sorts of issue are very sensitive to gauge freedom issues in GR, and he is explicitly fixing a gauge in his model. That raised alarm flags in my head.
Regardless this is jsut as ugly as the landscape in principle, it again sticks us at a very unusual spot in phase space.
Hi Sean,
Thanks for sharing the quick opinion, I hope we’ll hear more when you’ve had a chance to look at this more carefully.
As to string theory. Sure, there are plenty of string theorists who aren’t happy with the landscape, and the disappearance of the CC wouldn’t change things for them. But they still would face the perennial two huge problems of
1. how to break Q (supersymmetry generator), but not Q^2 (Hamiltonian operator).
2. finding some argument why there aren’t at least 10^500 solutions to the theory, ruining its ability to predict anything.
By now a large number of prominent string theorists have publicly announced that they believe string theory has to have an exponentially large number of consistent ground states. They have argued this is a point in the theory’s favor since it allows the anthropic explanation of the value of the CC. If this disappears, and the existence of so many ground states becomes purely a bad thing for the theory, it will be interesting to see if they change their mind about them.
Sean wrote:
there’s no way that the situation outside our Hubble patch can influence our observed dynamics.
But just because these modes have wavelength larger than Hubble radius, that doesn’t mean they’re entirely outside our Hubble patch — they still have some energy density inside our Hubble patch.
How the time-dependence arises is a bit of a mystery to me, but it is just scaling like a(t). They seem to claim to have derived this from Einstein’s equations. (They don’t seem to say phi is time-dependent, just that it has a time-dependent effect on the metric, parametrized by Psi.)
As a non-cosmologist, I was hoping you would have all the answers. Can you get Kolb to explain it all?
Hi Peter,
This paper has caused some discussion around here already. You’ve got the gist of it, the crucial point seems to be that these super-hubble perturbations have some time dependance (constant perturbations could be scaled away). You can see this in equation (6), if the super-hubble part of the potential was constant, you’d get the standard result.
It’s going to be the subject of our Friday lunch discussion, so I might be able to tell you what the experts around here think.
Oops, I left a link to my home page instead of my blog. For a minute there I was thinking like a scientist instead of a blogger!
My first impression, upon an admittedly brief reading, is that this mechanism won’t work — there’s no way that the situation outside our Hubble patch can influence our observed dynamics. But I have to think about it more to understand exactly what they are claiming.
Of course, if any cosmological observation shows that the dark energy is not a cosmological constant, people will abandon the landscape idea. String theorists, like everyone else, will go back to trying to understand a way to get a vanishing vacuum energy without unbroken supersymmetry. You try to build models that are consistent with the data, that’s the way it works.
“The landscape” is not the same as “string theory”; it is an idea within string theory that may or may not be right. String theorists themselves argue back and forth about whether the landscape is the right way to think about things; the fact that there is so little experimental guidance makes it a difficult task, but not necessarily an impossible one.
Common-sense suggests that super-Hubble fluctuations generically mimic an anisotropic expansion. While supernova data are consistent with an isotropic acceleration. Is this a problem? The text below eq. (12) seems a (plausible but not sound) attempt of addressing it.
Pingback: Preposterous Universe