Successors to string scuffle (physical assets/liabilities?)

[marcus]

5 years ago you could think of CDT and AsymSafe as part of a group of approaches called "Loop and allied QG". The numbers of people were so small it seemed like a single non-string QG community. Rovelli convened a Loop-and-friends Nonperturbative QG workshop at Marseille in the spring of 2004. Some 50-60 people participated. Loll included. So CDT was part of that first conference. The next year both Loll and Martin Reuter gave invited talks at Loops 2005 Potsdam. Both CDT and AsymSafe were again represented at Loops 2007.

Now CDT has grown and is appearing as a potential rival to Loop. 5 years ago CDT was only being done by Renate Loll's group at Utrecht. Now the CDT computer work is being done in at least two other places, Perimeter and UC Davis. People show up at conferences from places like Poland and Iceland to give CDT papers. Loll administers a large grant from the ESF (european science foundation) able to support postdoc contracts, meetings, workshops etc.

AsymSafe is also appearing as a potential rival. Important landmarks were the talk at CERN by Steven Weinberg and the organization of the Perimeter conference to be held in November.

Horava QG is another 4D approach that is getting a lot of attention.

Herbert Hamber is making a strong pitch for massive computer calculations along Lattice QCD lines to implement his version of Regge QG.

Carlo Rovelli is this week presenting a "new look" LQG, and the abstract stresses its analogy with Lattice QCD---and mentions renormalization.

You may want to add others to the list of contenders.

Any of these could become a star or premier successor to string. All can be seen as jockeying for position---scuffling for a place in the sunlight that has opened up. What physics differences do you see? What are the strong/weak points?

Sometimes one cannot tell if some feature is an advantage or disadvantage, it is just a difference. These distinctive features should still be pointed out.

All these approaches seem to echo the themes of renormalization, and continuity with the tried and true, the "good old" (as Weinberg put it) Quantum Field Theory. Several if not all the recent presentations refer at some point to the success of Lattice QCD.

 

[marcus]

Herbert Hamber gave a talk at Perimeter earlier this year, the video is online.
He was questioned by Lee Smolin and Laurent Freidel, among others. A lot of audience reaction.

It was essentially the same talk as the one he gave to the 800-some participants of Marcel 12 in Paris this summer. Except better, more slides with more detail, if I'm not mistaken.
He's aggressively saying that his simple 4D approach to lattice gravity is correct and CDT is not right and Loop is not right---and there is no need to make up extra stuff like strings.
He has access to a big computer and has run dedicated programs that take 2 months. He presents his approach as the natural heir to Lattice QCD. To the extent this is to be taken seriously, we definitely need to pay close attention.

I'll get the link to that PIRSA talk by Hamber.
Here is the link:
http://pirsa.org/09050006

 

[ensabah6]

hi do you have a link for this new loop? loop or sf?

 

[tom.stoer]

Any of these could become a star or premier successor to string.

I hesitate to call it "successor to string" because string theory was (is?) much more ambitious than "only" quantizing gravity. None of the approaches you mentioned has the potential to unify all known interactions and to replace strings. They should be compatible with a large class of interactions, but are not predictive in the sense that they single out specific interactions.

OK, string theory failed or is trapped in a multiverse of dead-ends, but originally it was heading for true unification and uniqueness. In that sense no program should be called a successor.

But (!) my secret hope is that "new LQG" with braiding could do the job to derive particles = their symmetries and interactions from topological and/or pure geometrical considerations. That's why I voted for "new LQG".

 

[ensabah6]

I hesitate to call it "successor to string" because string theory was (is?) much more ambitious than "only" quantizing gravity. None of the approaches you mentioned has the potential to unify all known interactions and to replace strings. They should be compatible with a large class of interactions, but are not predictive in the sense that they single out specific interactions.

OK, string theory failed or is trapped in a multiverse of dead-ends, but originally it was heading for true unification and uniqueness. In that sense no program should be called a successor.

But (!) my secret hope is that "new LQG" with braiding could do the job to derive particles = their symmetries and interactions from topological and/or pure geometrical considerations. That's why I voted for "new LQG".

Do we know for a fact there is a force unification? If nature does not unify forces at high energies, neither should our theories

 

[tom.stoer]

Do we know for a fact there is a force unification? If nature does not unify forces at high energies, neither should our theories

Unification does not necessarily mean that it is something like the old-fashioned GUT SU(5) <= U(1) * SU(2) * SU(3). It can be something like an explanation of a specific symmetry structure based on a deeper and in in some sense simpler model. That's my expectation.

 

[Bob_for_short]

The problem of all current theories is in a self-action ansatz. I find it failed physically and mathematically. I think our theories need reformulation and it is very close to what we have now but simpler. We have just to admit that interacting particles form compound systems so what we observe is quasi-particles, not particles. Nobody argues that in a solid state the energy-momentum is distributed amongst quasi-particles. But who said, for example, that the electron, permanently coupled to the quantized electromagnetic field, is elementary? Only those who could imagine it decoupled. Then, to couple it, one introduces a self-action. At the same time we can look at photons as at quasi-particles of a compound system and understand their "decoupling" as decoupling separated variables in one compound system. Such an approach excludes the self-action from the very beginning (rather than perturbatively, as in renormalization prescription), preserves the energy-momentum conservation laws and correctly describes the physical phenomena (soft radiation, Lamb shift, etc.) I believe this physical and mathematical approach has to replace current attempts so full of problems, patches, and patches of patches. In this sense it will be a successor.

 

[xepma]

Do we know for a fact there is a force unification? If nature does not unify forces at high energies, neither should our theories

Ofcourse there should be some form of unification. General Relativity and QFT are two completely different ballparks - they don't match up at all. Something has to give in.

 

[tom.stoer]

Do we know for a fact there is a force unification? If nature does not unify forces at high energies, neither should our theories

I have to admit: no fact!

One of the biggest problems we face today is that we have no direct results from experiments forcing us to develop a theory beyond the standard model + GR. No facts, "only" mathematical and aesthetical reasons.

 

[marcus]

Here is the abstract for Herbert Hamber's talk.
Quantum Gravitation and the Renormalization Group
Herbert Hamber
"In my talk I will provide an overview of the applications of Wilson's modern renormalization group (RG) to problems in quantum gravity. I will first discuss the development of the RG for continuum gravity within the framework of Feynman's covariant path integral approach. Then I will discuss a number of issues that arise when implementing the path integral approach with an explicit lattice UV regulator, and later how non-perturbative RG flows and universal non-trivial scaling dimensions can in principle be extracted from these calculations. Towards the end I will discuss recent attempts at formulating RG flows for gravitational couplings within the framework of a set of manifestly covariant, but non-local, effective field equations suitable for quantum cosmology."
May 13, 2009 -
http://pirsa.org/09050006/

He comes across as somewhat arrogant and aggressive against the other competing approaches like Loop and including also Loll's CDT. I think this is all right---he is just playing hardball with his close competitors. And until recently I think his research was not so visible as either of those. Since the talk is at Perimeter and Lee Smolin and Laurent Freidel are in the audience, they are among those asking questions.

The slides are much the same as the ones he used for the invited plenary talk in Paris on 14 July, in a session chaired by Ashtekar, where Laurent Freidel and Juan Maldacena also gave talks.
===========================

Rovelli gave his "new look" LQG talk today at the Corfu school. It seems to me that LQG gets redefined from time to time. The current version is apt to be slightly different and we won't know in what way until these 5 one-hour talks are online. Here for reference is the abstract of the lecture series that started today:

Carlo Rovelli

Covariant loop quantum gravity and its low-energy limit

"I present a new look on Loop Quantum Gravity, aimed at giving a better grasp on its dynamics and its low-energy limit. Following the highly successful model of QCD, general relativity is quantized by discretizing it on a finite lattice, quantizing, and then studying the continuous limit of expectation values. The quantization can be completed, and two remarkable theorems follow. The first gives the equivalence with the kinematics of canonical Loop Quantum Gravity. This amounts to an independent re-derivation of all well known Loop Quantum gravity kinematical results. The second the equivalence of the theory with the Feynman expansion of an auxiliary field theory. Observable quantities in the discretized theory can be identifies with general relativity n-point functions in appropriate regimes. The continuous limit turns out to be subtly different than that of QCD, for reasons that can be traced to the general covariance of the theory. I discuss this limit, the scaling properties of the theory, and I pose the problem of a renormalization group analysis of its large distance behavior."

http://www.physics.ntua.gr/corfu2009/qg.html
http://www.physics.ntua.gr/corfu2009/Program/3rdWeekSchool.html

 

[tom.stoer]

Covariant loop quantum gravity and its low-energy limit

"... general relativity is quantized by discretizing it on a finite lattice, ... equivalence with the kinematics of canonical Loop Quantum Gravity.

That's remarkable. It became clear in the past years that naive lattice discretization and quantization (spin-foams) does not yield a theoy which is equivalent to canonical LQG. The reason is that LQG (the cylindrical functions) is based on graphs instead of simplices. For every discretitzation (triangulation) one can construct a dual graph, but the opposite is not true: there are graphs for which a dual triangulation does not exist.

One can understand this quite easily. Take a triangulation and construct it's dual graph: associate one point within each cell (simplex) with a vertex of a graph; this is the dual graph. Now connect two arbitrary vertices (which are not nearest neighbours according to the graph) with a new edge. It's immediately clear that this new edge cuts through several simplices prohibiting to construct a dual triangulation. In the dual triangulation one would have to associate an n-1 simplex (= a face) with the new non-local edge. This new face would cut other simplices.

Spin-foams which are based on a naive triangulation seem to be not rich enough to have GR as low-energy / large-distance limit.

This is remarkable as the construction of the cylindrical functions starts with a kind of epsilon-regularization which relies on "cells" as well. It is not clear to me at which point the space of cylindrical functions becomes richer than a space defined for simplices only.

Smolin wrote a paper in which he tried make use of these non-local edges. According to him they cause a mismatch of micro-causality (as defined by the graph) and macro-causality (as defined by a smooth Riemann manifold emerging in an effective low-energy limit). In the latter one should see a residual effect of the non-local edges which is basically equivalent with the effects of a cosmological constant. It would be interesting if Smolin's rather generic arguments can be derived in an LQG or spin foam context. This should clarify if the cosmological constant is an input for LQG (as it used to be a couple of years ago when they studied quantum-deformations of the local Lorentz symmetry of the vierbein) or a prediction of LQG (or some generalization).

I would like to speculate that turning the input into a result but keeping the framing / braiding of spin networks as an option has the power to let particles emerge as topological configurations of braids.

 

[marcus]

You go further with this than I can. I am frustrated by our not having the media from these talks yet. I sense from how the abstract is worded that Rovelli is going to attempt a reformulation, or a new look of sorts. He may have new results to talk about as well.

But I find I simply have to wait patiently to find out about that.

In the meantime, I would like us to be able to list a short list of media which make the most convincing case for the other possible entries.

With Horava, I would like it if there were a video talk by Horava himself, or by Robert Brandenberger, which would make the strongest possible case. I think it is significant that Horava cosmo has a bounce. Brandenberger has a paper about this.

This could be seen as an important distinction. Reuter and Loll (AsymSafe and CDT) do not have a bounce. Causal Sets (if anyone is interested in that) and Fotini's Graphity do not have a bounce. You could think of it as an asset, something that will get the attention of researchers. Or as a liability, something to eventually falsify?

Of these contenders I think only Loop and Horava have the possibility of a bounce cosmo.

However AsymSafe has the asset of a very natural inflation, obtained by the running couplings. Weinberg pointed that out in his CERN video.

Hamber gives a strong confident, almost arrogant, pitch. Also Loll gives a terrific presentation. They both explain the advantages of their two styles of Lattice QG so clearly that I don't even want to repeat the points.

My take is there is a patch of stage spotlight, or of sunlight, opening up, and these different approaches are vying for a piece of the action. As they should!

Perimeter has already said Weinberg and Horava. They have two conferences in November planned: for AsymSafe and for Horava gravity. The GR19 conference in Mexico City has designated Robert Brandenberger as chair of the parallel session on Mathematical Cosmology (that says a strong Horava cosmology presence).

I think all these approaches SHOULD compete for the attention of the research community, especially the young researchers.

So I am going to try to assemble a few links to what I think are strong presentations of leading contenders. People can if they want watch several videos and see who they bet on taking a place in the sun (or if you like the stage limelight.)

 

[marcus]

Here is a provisional choice of one single strong advocacy of each approach (where an up-to-date presentation is available)

CDT
Obviously Loll at the Planck Scale conference.
http://www.ift.uni.wroc.pl/~rdurka/planckscale/index-video.php?plik=http://panoramix.ift.uni.wroc.pl/~planckscale/video/Day1/1-4.flv&tytul=1.4%20Loll
Video: "Causal Dynamical Triangulations and the Quest for Quantum Gravity"

AsymSafe
The last 12 minutes of Weinberg's CERN talk can't be beat.
http://cdsweb.cern.ch/record/1188567/
Video: "The Quantum Theory of Fields: Effective or Fundamental?"
To save time jump to minute 58.

Horava QG
Don't have a video lecture. May not have one until November conference.

new look Loop
Waiting for the Corfu School talks to be posted online.

Hamber Regge QG
Hamber does a great job on PIRSA
http://pirsa.org/09050006/
Video: "Quantum Gravitation and the Renormalization Group"

Condensed matter approach: geometry emerges from graph.
In my view, Fotini M. makes the most persuasive presentation. Not Wen exactly, but same general condensed matter idea.
http://pirsa.org/09030018/
Video: "Quantum Graphity: a Model of the Emergence of Locality in Quantum Gravity"

If anybody else can suggest other similar but better presentations of the same material, that are on line, please do. The idea is to have a minimal concentrated introduction to each one, so someone could watch all 4 or 5 talks and possibly pick one or several approaches as having more going.

 

[atyy]

Wen is my favourite - the whole Sakharov, Visser, Volovik style - even though it is very unlikely to succeed - it is hard to put chiral fermions and interactions on the lattice, because of the Nielsen-Ninomiya theorem. There are some work-arounds in lattice QCD, but Wen has said that while he has some idea of how to get chiral fermions, he doesn't know how to do chiral interactions. I love it because of its playful style and it's tower of turtles philosophy. My vote for serious modelling still goes to string theory. Wen can be seen in action at http://pirsa.org/08110003/ .

 

[marcus]

Wen is my favourite - the whole Sakharov, Visser, Volovik style - even though it is very unlikely to succeed ...

In that case, please go on record with a point for Wen!

 

[marcus]

I hesitate to call it "successor to string" because string theory was (is?) much more ambitious ...

I see your point. I wasn't thinking of any of these approaches pursuing exactly the same program goals as string! I see the attention of researchers shifting in a 4D direction and these various contenders competing to some extent to "take up the slack" in research interest.

Steven Weinberg described the situation in a helpful way, I thought. String might not be needed, he suggested, and gravity might be brought together with the rest essentially within the framework of "good old" quantum field theory. In his case the idea was Asymptotic Safety, but there are other ways of handling 4D gravity within something more like traditional quantum fleld theory---not so drastically inventive of new degrees of of freedom.

Anyway I don't think anyone is suggesting that these new approaches are "better" in some sense than string. We don't know the ultimate value of any type of research ahead of the results. And no one is saying that those who continue doing string should stop! It's just a fact that interest has declined or is no longer so intensely focused. Former string folks are finding other areas to research. The field is no longer so much in the limelight. So there is slack, and the natural question is what other areas of theory will take up the slack.

 

[tom.stoer]

I see your point. I wasn't thinking of any of these approaches pursuing exactly the same program goals as string! I see the attention of researchers shifting in a 4D direction and these various contenders competing to some extent to "take up the slack" in research interest.

Steven Weinberg described the situation in a helpful way, I thought. String might not be needed, he suggested, and gravity might be brought together with the rest essentially within the framework of "good old" quantum field theory. In his case the idea was Asymptotic Safety, but there are other ways of handling 4D gravity within something more like traditional quantum fleld theory---not so drastically inventive of new degrees of of freedom.

Anyway I don't think anyone is suggesting that these new approaches are "better" in some sense than string. We don't know the ultimate value of any type of research ahead of the results. And no one is saying that those who continue doing string should stop! It's just a fact that interest has declined or is no longer so intensely focused. Former string folks are finding other areas to research. The field is no longer so much in the limelight. So there is slack, and the natural question is what other areas of theory will take up the slack.

I agree. I don't think that the idea of unification based on one physical entity (string?) is wrong. What was wrong (or at least not successful) with strings is the idea of doing everything like in old-fashioned quantum field theory w/o taking into account other lines of thought. With that I mean "fixed background", "perturbation series", ... Perhaps there's a chance to re-start with strings but from the very start incorporate these lessons learned.

 

[BigFairy]

society eventually demanded accountability, and purposefully rightfully so.

 

[Bob_for_short]

...What are the strong/weak points?

These attempts are not phenomenological, i.e., they are not physical. They are some mathematical exercises. They proceed from some artificial constructions and are in fact nothing but soap operas: many watch them today and will forget tomorrow.

 

[marcus]

the comparison with "soap opera" is funny! But I think in fact that we can't know the future. Some of these approaches may either teach us something valuable or continue into a successful development. And of course some may not. In any case for now I want to choose one single strong advocacy for each approach (wherever a current presentation is available). Here is an up-date of the earlier list:

CDT
Loll at the Planck Scale conference is tops--best available single lecture on the subject.
http://www.ift.uni.wroc.pl/~rdurka/planckscale/index-video.php?plik=http://panoramix.ift.uni.wroc.pl/~planckscale/video/Day1/1-4.flv&tytul=1.4%20Loll
Video: "Causal Dynamical Triangulations and the Quest for Quantum Gravity"

AsymSafe
The last 12 minutes of Weinberg's CERN talk can't be beat.
http://cdsweb.cern.ch/record/1188567/
Video: "The Quantum Theory of Fields: Effective or Fundamental?"
To save time jump to minute 58.

Horava QG
A video lecture by Horava himself. Fixed camera though. We may get something better after the November conference.
http://online.itp.ucsb.edu/online/adscmt_m09/horava/rm/flash.html
Video: "Quantum Gravity with Anisotropic Scaling"

new look Loop
Waiting for the Corfu School talks to be posted online.
For the time being here's Rovelli's talk at Strings 2008.
http://cdsweb.cern.ch/record/1121957?ln=en
http://indico.cern.ch/getFile.py/access?contribId=30&resId=0&materialId=slides&confId=21917

Hamber Regge QG
Hamber does a great job on PIRSA
http://pirsa.org/09050006/
Video: "Quantum Gravitation and the Renormalization Group"

Condensed matter approach: geometry emerges from graph.
In my view, Fotini M. makes the most persuasive presentation. This is not Wen exactly, but same general condensed matter idea.
http://pirsa.org/09030018/
Video: "Quantum Graphity: a Model of the Emergence of Locality in Quantum Gravity"

However Atyy has suggested a 2008 PIRSA video of Xiao-Gang Wen. So let me put that link up too.
http://pirsa.org/08110003/
Video: "The Emergence of Photons, Electrons, and Gravitons from Quantum Qbit Systems"
==============

Just a comment about the Horava presentation. We are all aware of the enormously successful 1940s theory of QED (Quantum Electrodynamics). Horava says in his speech that QED can ( "morally" I suppose ) be considered to be a part of string theory and thus, in the same spirit, his new 4D QG (the Horava-Lifgarbagez) can be viewed as a part of string theory. He is giving the talk at Santa Barbara KITP with David Gross in the audience and has some nice things to say about both David Gross and string. So one knows not to think of Horava 4D gravity as breaking away from the string program and community, in any sense. It is "really" a part of string theory just as Feynman's QED is, and Newton's gravity theory as well, one might add.

 

[marcus]

Of the six contenders, the six approaches we are comparing, four have what Steve Carlip referred to as "spontaneous dimensional reduction". At small scale the dimensionality of spacetime (measured by a diffusion process) goes down from 4D to around 2D.

In his talk at the Planck Scale conference, the four approaches he indicated as having this feature were CDT, AsymSafe, Horava, and Loop.
==quote from Carlip's second slide==
Accumulating bits of evidence that quantum gravity simplifies at short distances

• Causal dynamical triangulations
• Exact renormalization group/asymptotic safety
• Loop quantum gravity area spectrum
• Anisotropic scaling models (Horava)

Are these hints telling us something important?
==endquote==

Again, of these 6 approaches we are looking at, two have an application to cosmology that replaces the cosmo singularity with a bounce.
These two are Horava and Loop.

Robert Brandenberger posted a paper this year which showed that if a special field was put in and things were just right you could get Horava cosmology to bounce. I don't know how contrived it actually is, or fine-tuned, or conversely how robust the result is. In any case it's worth following to see how it plays out.

These features of dimensional reduction and comological bounce are not necessarily assets. They are just distinguishing features, that make some of the approaches stand out.

Here is the video of Steve Carlip's talk
http://www.ift.uni.wroc.pl/~rdurka/planckscale/index-video.php?plik=http://panoramix.ift.uni.wroc.pl/~planckscale/video/Day1/1-1.flv&tytul=1.1%20Carlip
It was the opening talk of the Planck Scale conference---the topic of spontaneous dimensional reduction as Planck scale is approached is currently of much interest and he is a long-time major figure in QG, so it made sense for the organizers to put this talk at the top of the list.

=================
EDIT TO REPLY TO NEXT:
MTd2 I think that paper of Carlip's that you pointed out for us today is a classic. It is short, reasonably simple, yet may have far-reaching consequences. I saw him give this same talk in person last week at Horava's "string seminar", as well as having watched the video from Planck Scale. It's an exciting paper and I'm very glad to learn of it's being posted on arxiv.

 

[MTd2]

This is the article that corresponds to the talk Carlip gave on Planck 2009, it was made available today on arxiv:

http://arxiv.org/abs/0909.3329

Spontaneous Dimensional Reduction in Short-Distance Quantum Gravity?

Steven Carlip
(Submitted on 17 Sep 2009)
Several lines of evidence suggest that quantum gravity at very short distances may behave effectively as a two-dimensional theory. I summarize these hints, and offer an additional argument based on the strong-coupling limit of the Wheeler-DeWitt equation. The resulting scenario suggests a novel approach to quantum gravity at the Planck scale.

 

[crackjack]

Isn't it too early to talk of successors to string theory already??

Alternatives have always been around... for a good reason... and will continue to be around. After all, apart from the die-hard alternative theories' fans who work on those irrespective of the success/failure of string theory, the less luminary string fans also need something to work on during drought times within string theory. But, calling this as the end of the string civilization seems immature, doesn't it?

 

[tom.stoer]

Marcus,

why not considering supergravity?

- there are indications that it could be renormalizable
- there are versions possibly rich enough to contain the standard model
- SUGRA is a theory on its own and does not necessarily need strings

 

[marcus]

Hi Crackjack, the poll question was: "Which of these potential string successors seem most promising?"

Since we don't know the future, we can't tell if the current shift of interest (e.g. into Horava QG) will be temporary or permanent. The list of competing ex-string research programs was meant to be open-ended:

You may want to add others to the list of contenders.

Any of these could become a star or premier successor to string. All can be seen as jockeying for position---scuffling for a place in the sunlight that has opened up. What physics differences do you see? What are the strong/weak points?
...

Tom, let's add SUGRA to the list of contenders. I keep seeing Kelly Stelle's name on speaker lists, might he serve as a persuasive advocate? Is there an introduction/overview to recommend?

 

[tom.stoer]

... let's add SUGRA to the list of contenders. ... Is there an introduction/overview to recommend?

Don't know; I was just reading some papers regarding on-shell finiteness; very technical; I do not understand all details.

 

[Bob_for_short]

Isn't it too early to talk of successors to string theory already??
Alternatives have always been around...

I am afraid it's about time to pay attention to other approaches since not all alternatives have similar funding as the string community.

But, calling this as the end of the string civilization seems immature, doesn't it?

In my opinion, this direction has never been alive. It's a Frankenstein's direction and many others are alike. Why do I think so? Because they are based on illusions, fantasies, not on phenomenology. Many think it is sufficient to take a carpet, shake it and then the waves of matter/gravity will "emerge". Many hopes and promising results were and are connected with such an approach. The problem seems just in taking the right carpet. If it is sufficiently “rich” and with a good pattern on it, that might work, many think. And each time a carpet is shaken, two dead bodies appear underneath: those of physics and mathematics.

Just a simplified example of killing physics and mathematics, from which everything went bad:

First we had an electron in an external field F_ext, its equation was m_ea =F_ext. A very good, Newton-like equation, no problem to solve it. Then, knowing that acceleration causes radiation and thus the energy-momentum losses, we make a step forward – we add a phenomenological term δm_ra to the right hand side: m_ea =F_ext - δm_ra. Such a theory development preserves the second-order equation character and does not bring any mathematical and conceptual problem. The coefficient δm_r (kind of added mass due to radiative friction) is supposed to be small because the radiative friction effect is rather small. Now, whether δm_r should be added to m_e or not? If yes, then such combining may be called a mass renormalization: m_R = m_e + δm_r. If no, then the two masses should be considered separately. In general case the answer is “NO”.

Indeed, in an external electric field we may join two terms and consider the radiating electron as a little bit heavier than non radiating. But in a gravity filed, for example, m_eg counterbalanced with a Hook force (weighting, static experiment), only m_e enters in equations: (m_e + δm_r )a = m_eg - kx = 0. So in general case both masses should enter the equation, for example, m_Ra = eE_ext + m_eg - kx. Thus no mass renormalization actually happens or needed - each mass coefficient has it own physical meaning.

Now, what happened in the Classical Electrodynamics development? H. Lorentz, in order to preserve the energy-momentum conservation, advances a self-action ansatz: m_ea =F_ext + eF_rad. It resulted in adding two terms to the right hand side: eF_rad = δm_ra + (2e²/3c³)a_dot. As soon as the term δm_r was too big, it was “joined” with m_e with saying a rubbish like “only their sum is observable”. I have shown above that not only their sum appears in the equations but m_e separately too, so it was not a mass renormalization. It was term discarding. In other words, it was proposing to try another equation, not the original m_ea =F_ext + δm_ra + (2e²/3c³)a_dot. Discarding the term with δm_r is postulating another equation, not the mass renormalization, let us not fool ourselves.

Besides, the term δm_r was not connected with radiative friction but with "electrostatic electron mass" - nothing in common with the radiative friction effect. Finally, the Lorentz-Abraham equation has non-physical (runaway) solutions that contradict the motivation of preserving the energy-momentum conservation laws. I.e., the self-action ansatz turned out in a complete fiasco. Nevertheless it was implemented in QED.

Similar discarding is carried out in renormalizable QFTs for that same reason: the perturbative corrections to the fundamental constants are not necessary: if finite, they worsen agreement with experiments, if infinite – all the more. And each time (in each order) renormalizations mean postulating new equations. This kind of theory patching on go is not innocent. Discarding terms from equations or solutions is not mathematically legitimate. Physically, obtaining corrections to constants means too bad guess of interaction (including self-action). So persisting in “renormalizations” (discarding) we kill physics and mathematics. Unfortunately, many think we can live with it and many try to get a “renormalizable” version of ToE or more specific theory. You should note that everything turns around it – justification of cut-off or getting rid of UV divergences due to super-symmetry, - because it is an important obstacle in practical calculations.

I analysed this problem and found another, physically justified way of preserving the energy-momentum conservation laws. I explained my vision of the problem solving in “Reformulation instead if Renormalizations” and other works and I think this direction is worth developing. In such an approach one naturally obtains physically meaningful results from the very beginning and no mathematical problems spring up.

Vladimir Kalitvianski.

 

[tom.stoer]

Don't restrict you view to old-fashioned QFT.

What we call renormalization (regularization) has at least some physical relevance as it generates a scale in an scale-free classical theory; so there is at least something that works physically correct.

On-shell finiteness in SUGRA is different from just regularization. It means finiteness! Therefore it could be a hint to omitted unphysical off-shell Greens functions.

If SUGRA and/or string theory succeeds with the proof of n-loop finiteness then this would be a major step forward.

What I always stress is that perturbative finiteness may not be taken as seriously as is usually the case. Many people think that infinities in the perturbation expansion or non-renormalizability point towards the fact that the whole theory is not well-defined. That could be wrong! Instead it's an indication that the perturbation expansion fails, not the theory. Look at the Taylor expansion of exp(-1/x²) and you will see what I mean :-)

 

[Bob_for_short]

If SUGRA and/or string theory succeeds with the proof of n-loop finiteness then this would be a major step forward.

So you will be surprised and happy with my finite-from-the-very-beginning construction then.

...Many people think that infinities in the perturbation expansion or non-renormalizability point towards the fact that the whole theory is not well-defined. That could be wrong! Instead it's an indication that the perturbation expansion fails, not the theory. Look at the Taylor expansion of exp(-1/x²) and you will see what I mean :-)

I say similar things, with a stress on the initial approximation quality. As well, I provide much more relevant examples of expansions. My problem now is to get funding for continuing my research.

What we call renormalization (regularization) has at least some physical relevance as it generates a scale in an scale-free classical theory; so there is at least something that works physically correct.

I am afraid it is a drawback to get scales in a scale-free theory. You introdice them artificially. Besides, without divergences and necessity to make a cut-off there is no need in scales. As I said, there are finite and physically justified theoretical constructions where all that "renormalizing" ideology is absent.

On-shell finiteness in SUGRA is different from just regularization. It means finiteness! Therefore it could be a hint to omitted unphysical off-shell Greens functions.

Imagine, you obtain finite corrections to the fundamental constants m and e. They will worsen a very good agreement with experimental data obtained in the first Born approximation. Thus, finite or infinite, they must be discarded which is not legal mathematical operation. It is better to reformulate a theory in such a way that no corrections to the fundamental constants arise. For example, if a perturbation term is of a kinetic nature, it brings corrections to masses. If it is of a potential nature, it does not bring such correctoins. You can find details in my publications.

 

[tom.stoer]

I am afraid it is a drawback to get scales in a scale-free theory. You introdice them artificially. Besides, without divergences and necessity to make a cut-off there is no need in scales. As I said, there are finite and physically justified theoretical constructions where all that "renormalizing" ideology is absent.

But you can measure the QCD scale :-)

 

[Bob_for_short]

But you can measure the QCD scale :-)

I do not know about what you are speaking. Anyway, such a scale measuring relies upon a theoretical picture. If it contains natural scale things, than it is OK. If it originates from cut-offs, it is just a self-fooling (the worst kind of self-action ).

 

[Bob_for_short]

Don't restrict your view to old-fashioned QFT.

As I said, the problem appeared even before QFTs. The QFTs contain this problem because they inherited a self-action ansatz. So fixing it opens good perspectives in correctly describing particle physics. UV and IR problems can be removed at one stroke.

 

[marcus]

Just a simplified example of killing physics and mathematics, from which everything went bad...
I analysed this problem and found another, physically justified way of preserving the energy-momentum conservation laws. I explained my vision of the problem solving in “Reformulation instead of Renormalizations” and other works...

Vladimir Kalitvianski.

So you will be surprised and happy with my finite-from-the-very-beginning construction then...
...My problem now is to get funding for continuing my research.

You can find details in my publications.

I'm not sure how appropriate or helpful it is to discuss these in this particular thread, but in case anyone wants to look at some of Bob Kalitvianski's published and unpublished research papers, there are three on arxiv:
http://arxiv.org/find/all/1/au:+Kalitvianski/0/1/0/all/0/1
Bob may have given links to these papers and/or others elsewhere.

 

[Bob_for_short]

Thanks, Markus, I could not give any references because ZapperZ forbids it. I expressed my opinion about drawbacks of the existing approaches and backed my opinion with my results. Now I cease discussing my own works in this thread. Those who are interested can do it in my "Independent Research" thread "Reformulation instead of Renormalizations" and on my weblog http://vladimirkalitvianski.wordpress.com.

Bob_for_short.

 

[marcus]

...I don't think that the idea of unification based on one physical entity (string?) is wrong. What was wrong (or at least not successful) with strings is the idea of doing everything like in old-fashioned quantum field theory w/o taking into account other lines of thought. With that I mean "fixed background", "perturbation series", ... Perhaps there's a chance to re-start with strings but from the very start incorporate these lessons learned.

To restart the program afresh, in a nonperturbative and explicitly background-independent manner. (I mean here background independent = not building on spaces with fixed prior geometry.) Your post contained this radical idea and seems to call for a second look.

What you suggest seems like tall order (almost too challenging), but it reminds me of some recent quotes from Gell-Mann (on his 90th birthday!) As I recall he said, in effect, why do they go on ducking the real issues?

Peter Woit quoted this short passage from a long interview in New Scientist:
"I am puzzled by what seems to me the paucity of effort to find the underlying principle of superstring theory-based unified theory. Einstein didn’t just cobble together his general relativistic theory of gravitation. Instead he found the principle, which was general relativity, general invariance under change of coordinate system. Very deep result. And all that was necessary then to write down the equation was to contact Einstein’s classmate Marcel Grossmann, who knew about Riemannian geometry and ask him what was the equation, and he gave Einstein the formula. Once you find the principle, the theory is not that far behind. And that principle is in some sense a symmetry principle always."

Several of your posts on this thread have contained interesting observations. I'm thinking particularly of post #11. I've been either distracted or insufficiently prepared to respond properly. In post #11 you indicate that Loop is currently in confusion, but that it may be a "fertile" one. Should spinfoams correspond to triangulation (as in usual Lattice practice) or shouldn't they? Lewandowski says they shouldn't. Rovelli (but we haven't heard his latest talks) seems undecided. A recent note of his on black holes, with Krasnov, had a spin network node with a huge number of edges linked to it. If that kind of spin network is allowed, and taken as final state of a spinfoam path, then very general spinfoams must also be allowed...
There is also (in my mind at least) a puzzle about renormalization and the running constants. Maybe the theory is already UV finite and should have no dealings with this. But Krasnov has offered some ideas of how the spinfoam approach can "come to terms" with renormalization.