What caused the shift of interest in quantum cosmology?

[atyy]

I disagree that you can't talk about possible predictions from a not (yet - per its proponents) consistent theory. I see string theorists doing the equivalent all the time, and I applaud it. String theory is not yet a theory, but you say "if there is a consistent theory in here, we can argue it has certain properties, leading to the following predictions". Call them conditional predictions (based on some assumptions about how the theory will develop). I think it is highly advantageous to do such exercises while developing the theory, at all stages of its development.

Yes, but marcus was saying predictions in the sense of predictions from a complete theory, such that the theory can be falsified.

Second, strings are in a completely different league from loops, with regards to its consistency checks and definite yielding of gravity.

 

[marcus]

There is a risk that we get bogged/distracted by legalistic quibbling and miss the real interest of the topic.

Atyy harks back to a remark Ashtekar made in 2009 in a recorded threeway discussion in the ILQGS and refers to some stuff I am not sure is still meaningful. Maybe we have to discuss history a little.

The first time I recall Rovelli laying out the theory in concise form and saying "This is the theory" was in October 2010. The presentation was like nothing we had seen 2009 or earlier.

There was a much more tentative presentation in April 2010 which explored this approach for the first time, referring to some unpublished lectures by Bianchi as the source, but no flat statement "This defines the theory". The key steps only occur on pages 11 and 12, and in a provisional way---thinking it out, so to speak, while presenting the ideas.
That was http://arxiv.org/abs/1004.1780. "A new look..."
At the start of the paper this new approach was cautiously called "a possible perspective".

Then starting in October we got a series of "final version" type papers that actually lay the theory out in a definite form in one or three equations and declare that to be the theory.
http://arxiv.org/abs/1010.1939 "A simple model..."
http://arxiv.org/abs/1012.4707 "LQG, the first 25 years".
http://arxiv.org/abs/1102.3660 "Zakopane Lectures on Loop Gravity"

The first of these still looks a bit sketchy to me, it is a short paper aimed at mathematicians, but it does give the theory in a short list of Feynman rules. I prefer to go by the December 2010 and February 2011 formulations which are equivalent and spelled out in more detail.

Previous presentations were always more exploratory and less declaratory. There is a risk involved in "laying it on the line" like this. Rovelli et al are now taking that risk. I think the reasonable thing is to acknowledge and respect this. (Not confuse it with some remark Ashtekar made in 2009 )

At the same time there was a move into cosmology beginning with the March 2010 paper by Bianchi Rovelli Vidotto that actually used the new formulation which would only appear a month later in http://arxiv.org/abs/1004.1780. The cosmology paper was still very tentative and titled "Towards a Spinfoam Cosmology" http://arxiv.org/abs/1003.3483...
This comes across as a minor exploratory paper. It does not even derive the Bounce resolution of the singularity at start of expansion. However it becomes more important in light of a number of followup cosmo papers using the "final version" formulation of LQG to get cosmology stuff like Bounce, and introduce stuff like inflation and a positive cosmological constant. The Bounce (which is considered by phenomenologists to be testable) was derived to first order approximation by Benedetti Marciano later in 2010. Of course I expect to see more work on that. I would like to see their result confirmed and extended to higher order. But there is already a sense that the theory (as presented definitively in the Zakopane lectures) is testable. There is too much activity by phenomenologists in the Loop Cosmo area for this to be easily dismissed.

 

[marcus]

I disagree that you can't talk about possible predictions from a not (yet - per its proponents) consistent theory. I see string theorists doing the equivalent all the time, and I applaud it. String theory is not yet a theory, but you say "if there is a consistent theory in here, we can argue it has certain properties, leading to the following predictions". Call them conditional predictions (based on some assumptions about how the theory will develop). I think it is highly advantageous to do such exercises while developing the theory, at all stages of its development.

Good point! It is natural and advantageous. Also at least in my personal judgment the Bounce resolution of singularity has turned out to be such a robust prediction at every stage of Loop's development that if it now were ruled out by observation (a very real possibility) this would wipe LQG as we know it. You may disagree but it looks to me as if Rovelli has put it on the line and would have difficulty backing out.

See the December review http://arxiv.org/abs/1012.4707 , the section on cosmology.
A large credibility bet has been placed, or so I think.

But you can see it differently and that is fine too. As you say, it makes sense to test predictions even while work goes on developing the theory!

I hope that they get funding for the space mission CMBPol, which is what it would take according to that paper by Wen Zhao. (Tell me if you want the link.)

 

[atyy]

The bounce has not been derived (even approximately) from the new theory. Battisti and Marciano's http://arxiv.org/abs/1010.1258 works with the theory Vidotto and Rovelli used before their spin foam cosmolgy paper.

 

[marcus]

Battisti and Marciano's http://arxiv.org/abs/1010.1258 works with the theory Vidotto and Rovelli used before their spin foam cosmolgy paper.

No, the Battisti and Marciano bounce paper is explicitly based on this spinfoam cosmology paper by Rovelli and Vidotto:
http://arxiv.org/abs/0911.3097
On the spinfoam expansion in cosmology
Carlo Rovelli, Francesca Vidotto
6 pages
(Submitted on 16 Nov 2009 (v1), last revised 17 Feb 2010 (this version, v2))
We consider the technique introduced in a recent work by Ashtekar, Campiglia and Henderson, which generate a spinfoam-like sum from a Hamiltonian theory. We study the possibility of using it for finding the generalized projector of a constraint on physical states, without first deparametrising the system. We illustrate this technique in the context of a very simple example. We discuss the infinities that appear in the calculation, and argue that they can be appropriately controlled. We apply these ideas to write a spinfoam expansion for the "dipole cosmology".

This is the 17 Feb 2010 version. I doubt there is anything that does not carry over to their March 2010 paper, or any serious conflict in the results. The "dipole" spinfoam model used is exactly the same in both.

Also see Rovelli's December review http://arxiv.org/abs/1012.4707 page 15.

In a fast-developing field authors cannot always use the very latest work. So Battisti Marciano explicitly say they based their research on this paper and they also reference the more recent one. Here is the March 2010 one they cite as their reference [11]---it is part of the same line of development:

http://arxiv.org/abs/1003.3483
Towards Spinfoam Cosmology
Eugenio Bianchi, Carlo Rovelli, Francesca Vidotto
8 pages
(Submitted on 17 Mar 2010)
We compute the transition amplitude between coherent quantum-states of geometry peaked on homogeneous isotropic metrics. We use the holomorphic representations of loop quantum gravity and the Kaminski-Kisielowski-Lewandowski generalization of the new vertex, and work at first order in the vertex expansion, second order in the graph (multipole) expansion, and first order in 1/volume. We show that the resulting amplitude is in the kernel of a differential operator whose classical limit is the canonical hamiltonian of a Friedmann-Robertson-Walker cosmology. This result is an indication that the dynamics of loop quantum gravity defined by the new vertex yields the Friedmann equation in the appropriate limit.

I see that in the meantime there was another Spinfoam Cosmology paper by Battisti Marciano and Rovelli, that came out in February 2010.

 

[atyy]

No, the Battisti and Marciano bounce paper is explicitly based on this spinfoam cosmology paper by Rovelli and Vidotto:
http://arxiv.org/abs/0911.3097
On the spinfoam expansion in cosmology
Carlo Rovelli, Francesca Vidotto
6 pages
(Submitted on 16 Nov 2009 (v1), last revised 17 Feb 2010 (this version, v2))
We consider the technique introduced in a recent work by Ashtekar, Campiglia and Henderson, which generate a spinfoam-like sum from a Hamiltonian theory. We study the possibility of using it for finding the generalized projector of a constraint on physical states, without first deparametrising the system. We illustrate this technique in the context of a very simple example. We discuss the infinities that appear in the calculation, and argue that they can be appropriately controlled. We apply these ideas to write a spinfoam expansion for the "dipole cosmology".

That paper is not cited by http://arxiv.org/abs/1010.1258. Ref 16 is a paper by Rovelli and Vidotto, but not that one.

 

[marcus]

In that case I must apologize. I evidently misread a reference or an arxiv number. I will check to see.

There is this paper by Benedetti Marciano and Rovelli which seems to be on the canonical side:
http://arxiv.org/abs/0911.2653
Triangulated Loop Quantum Cosmology: Bianchi IX and inhomogenous perturbations
Marco Valerio Battisti, Antonino Marciano, Carlo Rovelli
In the conclusions they say:
"Finally, notice that the results of this paper, together with the link between LQC and spin-foams derived in [41], might provide a path to connect cosmological spin-foam models from the cosmological sector of LQG."

And their reference [41] is to the paper I mentioned, on the spinfoam side. So it looks like as of November 2009 the path connecting canonical and spinfoam Loop cosmology was not firmly established.

Reference [41] is:
[41] A. Ashtekar, M. Campiglia and A. Henderson, Loop quantum cosmology and spin foams, Phys. Lett. B 681 (2009) 347-352, arXiv:0909.4221 [gr-qc]; C. Rovelli and F. Vidotto, On the spinfoam expansion in cosmology, arXiv:0911.3097 [gr-qc]; A. Ashtekar, M. Campiglia and A. Henderson, Casting loop quantum cosmology in the spin foam paradigm, arXiv:1001.5147 [gr-qc]
===========================

And I see that it is this Benedetti Marciano Rovelli paper that the Benedetti Marciano "
Bounce" paper indicates it is based on.
http://arxiv.org/abs/1010.1258
Big Bounce in Dipole Cosmology
Marco Valerio Battisti, Antonino Marciano
5 pages
(Submitted on 6 Oct 2010)
"We derive the cosmological Big Bounce scenario from the dipole approximation of Loop Quantum Gravity. We show that a non-singular evolution takes place for any matter field and that, by considering a massless scalar field as a relational clock for the dynamics, the semi-classical proprieties of an initial state are preserved on the other side of the bounce. This model thus enhances the relation between Loop Quantum Cosmology and the full theory."

So this "Dipole Cosmology" paper is neither fish nor fowl. It is not the simplified LQC (symmetry reduced) nor yet is it the full spinfoam LQG theory. It seems halfway in between.
Closer to the full theory but still "old LQG".

 

[atyy]

In that case I must apologize. I evidently misread a reference or an arxiv number. I will check to see.

There is this paper by Benedetti Marciano and Rovelli which seems to be on the canonical side:
http://arxiv.org/abs/0911.2653
Triangulated Loop Quantum Cosmology: Bianchi IX and inhomogenous perturbations
Marco Valerio Battisti, Antonino Marciano, Carlo Rovelli
In the conclusions they say:
"Finally, notice that the results of this paper, together with the link between LQC and spin-foams derived in [41], might provide a path to connect cosmological spin-foam models from the cosmological sector of LQG."

And their reference [41] is to the paper I mentioned, on the spinfoam side. So it looks like as of November 2009 the path connecting canonical and spinfoam Loop cosmology was not firmly established.

Reference [41] is:
[41] A. Ashtekar, M. Campiglia and A. Henderson, Loop quantum cosmology and spin foams, Phys. Lett. B 681 (2009) 347-352, arXiv:0909.4221 [gr-qc]; C. Rovelli and F. Vidotto, On the spinfoam expansion in cosmology, arXiv:0911.3097 [gr-qc]; A. Ashtekar, M. Campiglia and A. Henderson, Casting loop quantum cosmology in the spin foam paradigm, arXiv:1001.5147 [gr-qc]
===========================

And I see that it is this Benedetti Marciano Rovelli paper that the Benedetti Marciano "
Bounce" paper indicates it is based on.
http://arxiv.org/abs/1010.1258
Big Bounce in Dipole Cosmology
Marco Valerio Battisti, Antonino Marciano
5 pages
(Submitted on 6 Oct 2010)
"We derive the cosmological Big Bounce scenario from the dipole approximation of Loop Quantum Gravity. We show that a non-singular evolution takes place for any matter field and that, by considering a massless scalar field as a relational clock for the dynamics, the semi-classical proprieties of an initial state are preserved on the other side of the bounce. This model thus enhances the relation between Loop Quantum Cosmology and the full theory."

So this "Dipole Cosmology" paper is neither fish nor fowl. It is not the simplified LQC (symmetry reduced) nor yet is it the full spinfoam LQG theory. It seems halfway in between.
Closer to the full theory but still "old LQG".

Yes, that's my reading too (well, close enough).

Roughly, I think Vidotto works in 2 lines, which are not yet connected. The first is http://arxiv.org/abs/0805.4585 , which in Battisti and Marciano's follow-up can be argued to have a bounce. The second is http://arxiv.org/abs/1003.3483 , which really tries to go from EPRL (actually KKL, I think).

Incidentally, Rovelli wrote in http://arxiv.org/abs/1012.4707v4 p19 "But for the moment, I see no definite prediction that could be used to falsify the theory."

 

[marcus]

Yes, that's my reading too (well, close enough).

Roughly, I think Vidotto works in 2 lines, which are not yet connected. The first is http://arxiv.org/abs/0805.4585 , which in Battisti and Marciano's follow-up can be argued to have a bounce. The second is http://arxiv.org/abs/1003.3483 , which really tries to go from EPRL (actually KKL, I think).

Incidentally, Rovelli wrote in http://arxiv.org/abs/1012.4707v4 p19 "But for the moment, I see no definite prediction that could be used to falsify the theory."

So now I have to re-think what the physics reasons are for the recent growth of interest in Loop quantum cosmology.

The "old LQG" is what has the firm prediction of a bounce, and so is testable.

On the other hand the new formulation spinfoam Loop gravity of the Zakopane lectures is based on principles, and is presented simply and concisely. It is mathematically very much to my taste---kind of beautiful and clean. The minimum of extra baggage.

Also it has a reassuring similarity to lattice QCD, and some strong analogies with Feynman diagrams of QED.

Just to be clear I could call it Zakopane Loop gravity. Would that be OK? Do you have a different terminology you want to suggest? Whatever we call it, you know what I mean: the vintage 2011 Loop.

So the appealing features are divided. Old Loop has the testability and a concrete version of the universe that people can work with and tack inflation scenarios onto etc. etc.
The new Zakopane Loop has conciseness and elegance of a sort that I think appeals to mathematicians. And so far the bridge connecting them is not complete.

Still, there is enough there that it may explain the rise in interest and research activity.

 

[atyy]

So now I have to re-think what the physics reasons are for the recent growth of interest in Loop quantum cosmology.

The "old LQG" is what has the firm prediction of a bounce, and so is testable.

On the other hand the new formulation spinfoam Loop gravity of the Zakopane lectures is based on principles, and is presented simply and concisely. It is mathematically very much to my taste---kind of beautiful and clean. The minimum of extra baggage.

Also it has a reassuring similarity to lattice QCD, and some strong analogies with Feynman diagrams of QED.

Just to be clear I could call it Zakopane Loop gravity. Would that be OK? Do you have a different terminology you want to suggest? Whatever we call it, you know what I mean: the vintage 2011 Loop.

So the appealing features are divided. Old Loop has the testability and a concrete version of the universe that people can work with and tack inflation scenarios onto etc. etc.
The new Zakopane Loop has conciseness and elegance of a sort that I think appeals to mathematicians. And so far the bridge connecting them is not complete.

Still, there is enough there that it may explain the rise in interest and research activity.

Yes, that's closer to what I think. The minor difference is that I don't stress testability and bounciness in LQC as much. For me, the main achievement of canonical LQC, circumscribed as it is, is dynamics, which canonical LQG did not have for a very long time, and which spin foams were supposed to remedy.

Since you noted the similarity of the Zakopane formalism to QCD, let me entice you to think about its connection to AdS/CFT. I know the Zakopane formalism tries to be more background independent than that, but in which case, it should contain AdS/CFT as a special case.

It's interesting that Rovelli seems to have accepted KKLI ( http://arxiv.org/abs/0909.0939 ) since he uses it in his work with Vidotto and cites it in the Zakopane lectures as a paper (together with EPRL and FK) defining the new vertex (http://arxiv.org/abs/1102.3660, comments preceding Eq 60). Yet he doesn't accept KKLII (KKLII http://arxiv.org/abs/0912.0540 , http://arxiv.org/abs/1010.4787 Rovelli's "anti" KKLII paper: http://arxiv.org/abs/1011.2149 ).

So, since the Zakopane lectures include KKLI, does that mean the current official Rovellian LQG is manifoldy?

 

[marcus]

It's interesting that Rovelli seems to have accepted KKLI ( http://arxiv.org/abs/0909.0939 ) since he uses it in his work with Vidotto and cites it in the Zakopane lectures as a paper ...
So, since the Zakopane lectures include KKLI, does that mean the current official Rovellian LQG is manifoldy?

The KKL spinfoam "for all LQG" presents a valuable way ( which does not require manifold) to define a spinfoam vertex with higher valence. I think R. gladly accepted that part of it.

The paper also has some manifold and embedding related stuff but I don't think Rovelli drew on it. The vertex definition is what I think is valuable. Surround the vertex with a spinnetwork boundary---this construction can be described without reference to embedding--then evaluate the little spin network. I remember R. going through that step by step, so it would be explicit and obvious to everybody that you don't need a manifold to do what L. was talking about to the vertex. Don't recall where, but likely in the December review and/or the Zakopane lectures.

Lewandowski clearly likes to keep manifolds in the picture, he talks about knots etc. But you can buy his spin foam higher valence vertex ("for all LQG") without buying the rest of the package.

So in answer to your question I would say no. The Zako formulation is not manifoldy.
===================

About the other thing, I don't understand what's going on with the second "KKL" paper you mentioned and the Ding Han Rovelli paper http://arxiv.org/abs/1011.2149 responding to it. That business about the extra degree of freedom, and "generalized" spinfoams with operator labels. I don't even get that the Ding Han Rovelli paper is outright ANTI this "operator spinfoam" generalization that Lewandowski is proposing. I see them as just tentatively checking it out, poking and probing some, without any sense of urgency about adopting it.

I recall Lewandowski gave one or two ILQGS talks about his ideas, so you could hear people's reactions---the tone of the Q and A. I'd better check to see when that was...
...late 2010.

Tuesday, Nov 2nd
Jerzy Lewandowski, Warszaw
Title: New results: operator spin foams, SL(2,C)
PDF of the slides (766k)
Audio [.wav 42MB], Audio [.aif 4MB].

Tuesday, Oct 26th
Jerzy Lewandowski, Warszaw
Title: LQG with all the degrees of freedom
PDF of the slides (1MB)
Audio [.wav 33MB], Audio [.aif 3MB].

 

[atyy]

Jerzy is pronounced "Yeryk"?

 

[marcus]

Jerzy is pronounced "Yeryk"?

"Yeryk" seems like how one would pronounce Jurek.

I think his legal name is Jurek Lewandowski, but his friends could call him Jerzy. Like George and Georgy (Georgey, spelling?).
Another example of the name is: Jurek "Jerzy" Kowalski-Glikman.
I'm not sure how to pronounce the nickname.

 

[marcus]

Recently we've been noticing increased research activity in the phenomenology of Loop Cosmo. especially as regards predictions of features in the CMB. E.g. in the power spectra of either primordial gravitational waves (GW) or temperature fluctuations.

As a sample, here are 40 Loop pheno papers which have appeared since 2008:
http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=FIND+%28DK+QUANTUM+GRAVITY%2C+LOOP+SPACE+OR+DK+QUANTUM+COSMOLOGY%2C+LOOP+SPACE%29+AND+%28DK+PRIMORDIAL%2C+FLUCTUATION+OR+DK+INFLATION+OR+DK+COSMIC+BACKGROUND+RADIATION%29+AND+DATE+%3E+2007&FORMAT=www&SEQUENCE=citecount%28d%29

I was curious about the beginnings of this trend and looked back at early papers. One that stood out for me was this 2008 paper by people at Nottingham and Cambridge. It predicts a blue tilt to the GW spectrum: distinctively different from the level or even red tilt spectra predicted from various inflation scenarios.

http://arXiv.org/abs/0810.0104
The gravitational wave background from super-inflation in Loop Quantum Cosmology
E. J. Copeland, D. J. Mulryne, N. J. Nunes, M. Shaeri
8 pages, 3 figures
(Submitted on 1 Oct 2008)
"We investigate the behaviour of tensor fluctuations in Loop Quantum Cosmology, focusing on a class of scaling solutions which admit a near scale-invariant scalar field power spectrum. We obtain the spectral index of the gravitational field perturbations, and find a strong blue tilt in the power spectrum with n_t ≈ 2.
The amplitude of tensor modes are, therefore, suppressed by many orders of magnitude on large scales compared to those predicted by the standard inflationary scenario where n_t ≈ 0."

This prediction has subsequently been reiterated by others under various detailed assumptions--it's not unusual to see it pointed out. So this seems like a longstanding and robust prediction to look for whenever missions like CMBPol become able to study the GW imprint on the CMB ancient light.
===========================

A NEW TYPE OF LOOP COSMO SIGNATURE appeared recently in a paper by Nelson and Wilson-Ewing:
They find that IF circles appear in the CMB (not yet certain that they do) then according to the Loop theory there should be a different size distribution from what you would get with Penrose CCC. With Loop there should be more larger circles than with Penrose CCC, and indeed with other nonsingular alternatives.
http://arxiv.org/abs/1104.3688
Pre-Big-Bang Cosmology and Circles in the Cosmic Microwave Background
William Nelson, Edward Wilson-Ewing
21 pages, 3 figures
(Submitted on 19 Apr 2011)
"We examine the possibility that circles in the cosmic microwave background could be formed by the interaction of a gravitational wave pulse emitted in some pre-big-bang phase of the universe with the last scattering surface. We derive the expected size distribution of such circles, as well as their typical width and (for concentric circles) angular separation. We apply these results in particular to conformal cyclic cosmology, ekpyrotic cosmology as well as loop quantum cosmology with and without inflation in order to determine how the predicted geometric properties of these circles would vary from one model to the other, and thus, if detected, could allow us to differentiate between various pre-big-bang cosmological models. We also show that the angular width and the sine of the angular radius of such circles are inversely proportional. This relation can be used in order to determine whether or not circles observed in the cosmic microwave background are due to energetic pre-big-bang events."

They don't predict that circles will be found, but it is claimed here that according to Penrose's model they should all be less than 6 degrees angular size, if found. By contrast, if they occur in the Loop context there should be larger circles as well---e.g. 30-50 degrees angular radius---according to the analysis in this paper. If circles are seen, therefore, one can ask if they are all small?---this would disfavor Loop and favor alternatives such as Penrose CCC.
On the other hand if larger circles were found this would disfavor Penrose's idea and favor Loop Cosmology.

It's an interesting idea. However the authors do not go so far as to predict that circles will be found if Loop is right. There is still a missing piece to the puzzle, a mechanism to explain how GW could propagate through the bounce.
They stress that because of the conditional nature of what they are saying, it should be considered a "signature" of Loop, not a prediction.

 

[marcus]

Actually Nelson and Wilson-Ewing say the above so clearly in their conclusions that I should simply quote:

==quote N and W-E http://arxiv.org/abs/1104.3688 ==
In this paper, we have shown how to

​

We have also studied four specific cosmological models which have a pre-big- bang epoch:

• conformal cyclic cosmology,
• ekpyrotic cosmology and
• loop quantum cosmology with and without inflation.

We have shown that the probability distribution of the circle size varies from one cosmological model to another and could potentially be used in order to differentiate between them based upon observations. We must stress that we are not predicting the presence of these circles in the CMB —rather, what we have done is to show how, assuming the circles are present, their geometric properties would differ from one cosmological model to another.

In particular, in conformal cyclic cosmology and the ekpyrotic universe, one would expect there to be only very small circles with an angular radius of at most 6◦, while in LQC without inflation one would expect larger circles as well whereas for LQC with inflation (and also all other pre-big-bang models that have an inflationary era), one should expect a distribution of large circles with an average radius of 57◦.

Also, in all of these models one expects the width and the radius of the circles to be inversely proportional. In fact, this relationship offers a simple way to test whether the circles found in [1] are due to extremely energetic events in a pre-big-bang epoch or not: if the smaller circles are also wider, this would provide strong evidence in support of the viewpoint presented in [1]. If not, then it seems more likely that the circles are simply statistical flukes or, perhaps, are due to a completely different mechanism. If the circles described in [1] are indeed imprints from pre-big-bang events, their geometric properties will give significant insight into the dynamics of the pre-big-bang era of our universe and also the nature of the quantum gravity effects that were undoubtedly important in the early universe.
==endquote==