338. Before the Big Bang: video of the day
ASTRO — By Dmitry Podolsky on April 6, 2009 at 11:04 am
Save This Post as PDF
Dmitry Podolsky has got his PhD from Landau Institute for Theoretical Physics. He currently works as postdoc at Case Western Reserve University. He is also one of the editors of NEQNET.
BBC News’ Stephen Sackur interviews Sir Roger Penrose about cyclic universe theory and the problem of initial state.
The quality is not terrific, but the interview itself is fun. By the way, do you know why Penrose is uncomfortable with inflationary paradigm?
P.S. As Lubos pointed out, this is just a part of the full interview, the whole interview (20+ min) can be found on the BBC www site.
12 Comments
Hi Dmitry, the link to the full video (23 minutes) and my long discussions about it,
http://motls.blogspot.com/2009.....nrose.html
Hi Lubos,
The last sentence have just blown me away – I understood what I was uncomfortable with reading recent papers by the most famous Penrose’s collaborator.
Cheers,
Dmitry.
Hi Dmitry, does it mean that without realizing, I have contributed to Podolskology? Maybe it’s not that different from Lubology. Except that I am still not 100% certain which collaborator you mean.
If you mean Stephen H., in my opinion, he is doing kind of the conceptual stuff of a similar way as others do. It’s not filled with new spectacular results, as far as I can say, but what I find rather amazing is that Stephen H. seems to take the right sides (=the same as I do, after a more careful analysis) all the time.
Hi Lubos,
I have no slightest idea what you are talking about
Yes, I meant Hawking, in particular, his recent attempts to invent newer measure for eternal inflation (namely addition of the volume weighting to the HM instanton). All this was said one thousand times already and repeated another thousand times, I am not sure why he pretends that what he does is really interesting 
Maybe, some experts in eternal inflation will explain it to me at some point…
Cheers,
Dmitry.
Hi Dmitry, Podolskology is a portion of psychology that studies how Podolsky thinks. Try to use Podolskology in practice and guess what Lubology means.
I don’t have a feeling that Hawking tries to pretend that what he’s doing is excessively interesting. It’s still interesting because of the name of Hawking, isn’t it?
Many other other people are doing the same not-too-very-interesting stuff about the measures so I wouldn’t be picking Hawking as a selected target.
“But before making my critical assessment, I should make clear that my remarks do not tell us that inflationary cosmology is wrong. They merely provide strong reasons to doubt most of the initial motivations behind the inflationary idea.”
R. Penrose The road to Reality p753
Basically, he thinks that inflation is correct, but that it doesn’t solve some of the problems it was invented to solve. Certainly it doesn’t solve the problem of initial conditions — inflation only starts if the initial geometry is very very special. Also, the inflationary party line has many circular arguments: see Penrose’s brilliant discussion of the horizon “problem” for example.
Dear Dr. Who,
That’s an interesting statement – I always thought that inflationary trajectory is an attractor in the phase space, so, almost arbitrary initial conditions lead to inflation and solution of homogeneity and flatness problems. Perhaps, not so many initial conditions lead to the observed level of primordial perturbations (
), but the set of suitable initial conditions is definitely larger than measure zero. So, what do you exactly keep the word “certainly” in your statement?
Unfortunately, I don’t have access to the book, could you explain his argumentation in some details?
Cheers,
Dmitry.
Dear Dr Who,
what you write sounds bizarre if not downright illogical.
The main reason why inflation is likely to be correct remains the fact that it does solve the well-known problems of the standard Big Bang cosmology. If someone doesn’t think that inflation makes all these things more natural, it is not quite clear to me why he would believe inflation.
Also, and this paragraph is related to the previous one, it is untrue that inflation only starts if the initial geometry is very special. Quite on the contrary, inflation is a process that can create a large smooth Universe out of pretty much any initial conditions. What is necessary is that the underlying theory has a flat neighborhood of a maximum of the potential, that rolls towards the minimum. And then it’s just enough that the scalar field, the inflaton, visits the neighborhood of this maximum in a small region of space. This region begins to inflate.
What you write suggests that Penrose has not yet understood the basic mechanisms of inflation at the “undergrad” level.
Best wishes
Lubos
Dear Dimitry, you might like to read
http://arxiv.org/abs/hep-th/0410270
Basically, for inflation to work, the inflaton has to be in the “potential dominated” state. That is an extremely special state with very few of the inflaton degrees of freedom activated. See the above paper [particularly the section on unitarity] to understand why this undermines any claim that inflation allows us to ignore initial conditions. You might also enjoy
http://arxiv.org/abs/astro-ph/0210527
about the way inflation “passes the buck”, though I don’t agree with his conclusions.
Penrose’s point is simply that equilibrium is the one state of a system that does not need to be explained: one only has to explain *deviations* from equilibrium. So concern about the universe “not having had time to equilibrate” are misplaced, *unless* you already understand why the universe was not in an equilibrium state at the very beginning. But really you should get hold of “The Road to Reality”, Penrose is much clearer and more convincing than I can be.
Dear Dr Motl: please read
http://tinyurl.com/c4xay9
before making any more comments about Penrose.
Dear Dr. Who,
Ok, I don’t think we are on the same frequency here – if you take a look on the list of my publications on SPIRES, you’ll find that there is no need to explain to me that “Basically, for inflation to work, the inflaton has to be in the “potential dominated” state”.
Also, you seem to miss the point of my earlier statement (not mine really – it is known for decades) that inflationary trajectory (“potentially dominated” state using your terminology) is attractor. Basically, if you think a bit (or at least read on Wikipedia) what attractor in the phase space is, you’ll understand that the state you claim to be special is basically almost general
Somewhat disappointed,
Dmitry.
Sorry, I didn’t mean to be rude. I think you just have to read Penrose or Carroll/Chen. Their point is precisely that the “attractor” argument is wrong.
Dear Dr. Who,
As you surely know, “attractor” argument goes as follows. Take a scalar field with non-trivial potential and pick almost arbitrary initial conditions for
and 
, let the scalar field roll and you will find that almost generally it will enter the regime where potential energy term dominates over kinetic one. This can be easily checked by, say, numerics (and was since mid 1980s), and strictly speaking the conditions that Carroll/Chen consider necessary
are just sufficient.
Do you (or Penrose) question that?
I’ve just finished reading the discussion of unitarity in the Carroll/Chen paper. To be honest, it sounds somewhat funny to me. Ok, generic collapse does not coincide with deflation – sure
Does it have to? Consider a different process in the end of de Sitter stage, when particles are produced from the inflaton condensate. Turn evolution backwards – do these particles get reassembled back into a homogeneous condensate? Or forget about inflation all together and consider dynamics of spontaneous symmetry breaking. If you turn evolution backwards, does fragmented condensate restore coherence? 
Generally, one can consider a dynamical system featuring attractor in its phase space. I think, it is rather well known that attractor becomes repeller if you turn evolution backwards.
To conclude, I am sure Penrose should be smarter than that.
Cheers,
Dmitry.
Trackback responses to this post