What is Ed Witten's stance on string theory?

[ensabah6]

There were "rumors" that Ed Witten has de-emphasized string theory in his personal research quite possibly discouraged by the landscape and recent criticisms such as PW and LS. One quote to show this is Witten's "I hope string theory has some relevance to nature".

At Strings 2007 he presented his work on 3D QG. He did not attend Strings 2008. He did not present a paper at Strings 2009, but gave a public lecture which was about nonstring physics. His talk this summer at Cern was about prospects for new physics away from the high energy frontier.

Ed Witten gives lecture in 2010
http://www.yourphysics.org/resources/videos/lectures/page_44292.html

He clearly affirms his belief String theory is the best hope for TOE and that he remains personally invested in this project.

Peter Woit discusses it here
http://www.math.columbia.edu/~woit/wordpress/

 

[MTd2]

He uploaded many papers this year to arxiv.org, but many of them showed up on math-ph, not on hep-th, even though they were about string theory. So, this is not something surprising.

 

[marcus]

At Strings 2007 he presented his work on 3D QG. He did not attend Strings 2008. He did not present a paper at Strings 2009, but gave a public lecture which was about nonstring physics. His talk this summer at Cern was about prospects for new physics away from the high energy frontier.

This is not rumor, it is simply fact. In the case of the 2009 Cern talk, "physics away from the high energy frontier" was even part of the title. A couple of days ago, I listened to the IOP lecture which Woit gave a link to and which you say "dispels rumors". I can't say it changed my assessment of where prominent figures (such as Witten) are.

I think one has to look at the facts and decide for oneself what to make of them.

 

[marcus]

Ed Witten gives lecture in 2010
http://www.yourphysics.org/resources/videos/lectures/page_44292.html

He clearly affirms his belief String theory is the best hope for TOE and that he remains personally invested in this project.

That lecture video is one that I linked to two days ago. Here is my 27 July post in the "up to?" thread:

Witten recently gave a public lecture (as recipient of this year's Isaac Newton medal--a prize awarded in the UK) in which he presented a bright picture of the promise of string research.

The online video is here:
http://www.yourphysics.org/resources/videos/lectures/page_44292.html
The hour lecture is in two 30 minute parts.

In case anyone is curious, here are links to the 21 papers he has posted
on arxiv during the past five years---2006 to 2010.
http://arxiv.org/find/grp_physics/1/au:+Witten_E/0/1/0/2006,2007,2008,2009,2010/0/1

A short audio interview (about 5 minutes) with a journalist is here:
http://audio.theguardian.tv/audio/k.../gdn.sci.100712.ad.Science-Weekly-podcast.mp3
The interview occurs as a segment of a longer program, between minutes 15:30 and 20:20.
He presents the anthropic multiverse idea.

It may not matter very much, what that one researcher is doing and what he says in public. But in any case I think the best thing is to look at available facts and decide for yourself what the person in question is up to.

I also linked to a press interview done after the IOP Newton medal lecture.

As for "personal investment", that is your interpretation. I doubt that that any of the above will "dispel" what you call "rumors".

What is at issue here are facts, not people's "spin" or interpretation.

 

[marcus]

He uploaded many papers this year to arxiv.org, but many of them showed up on math-ph, not on hep-th, even though they were about string theory. So, this is not something surprising.

I gave a link to Witten's output of 21 papers 2006-2010 in the post I just quoted. Here is the link again:
http://arxiv.org/find/grp_physics/1/au:+Witten_E/0/1/0/2006,2007,2008,2009,2010/0/1

This list includes math-ph and papers also listed in math.

Please tell me which of these 21 papers you think are explicitly about superstring physics---modeling reality in, for instance, 10 dimensions---rather than about, for example, the Langlands program or 3D quantum gravity.

You say "many" of these were "about string theory". I'd appreciate it if you'd let me know which of the 21 papers constitute this "many", so that I can see what you mean by "about".

 

[MTd2]

This one, for example. See the first paper that cited it:

http://arxiv.org/abs/1001.2933

 

[marcus]

Let's not waste our time over piece-meal. You said "many". Out of those 21 papers I would like to see your list of "many".

Say around half. Like 10. What is a list of 10 of those paper written since 2006 which are explicitly about string theory?

=====================

But I think you were kidding. The paper of Witten's you picked as an example was about 3D, and Chern-Simons, and when I looked to see who cited it (as you suggested) the paper at the top of the list was by Karim Noui and Catherine Meusburger! You snuck a LQG connection in on me!
http://arxiv.org/cits/1001.2933
http://arxiv.org/abs/1007.4615

Here is where Noui-Meusburger are coming from:
http://arxiv.org/abs/0809.2875
The relation between 3d loop quantum gravity and combinatorial quantisation: Quantum group symmetries and observables
C. Meusburger, K. Noui
(Submitted on 17 Sep 2008)
"We relate three-dimensional loop quantum gravity to the combinatorial quantisation formalism based on the Chern-Simons formulation for three-dimensional Lorentzian and Euclidean gravity with vanishing cosmological constant..."

So in January 2010 Witten (himself no longer doing much with "extra dimensions" physics models) posts something on 3D QG and CS, and *boom*
it gets picked up by Noui-Meusburger and connected with the Loop QG effort.

 

[MTd2]

Well, the first one that cited it was:

http://arxiv.org/abs/1002.0500

But you noticed one I didn't see. The second one is nice too.

 

[marcus]

Well, the first one that cited it was:

http://arxiv.org/abs/1002.0500

But you noticed one I didn't see. The second one is nice too.

Holy sh*t! You spotted a spin foam one citing Witten!
http://arxiv.org/abs/1002.0500
Spin Foam State Sums and Chern-Simons Theory
Aleksandar Mikovic, Joao Faria Martins
8 pages. Talk presented by A. Mikovic at "Chern-Simons Gauge Theory: Twenty years after" conference, Bonn, 3-7 August 2009
(Submitted on 2 Feb 2010)
"We review the spin foam state-sum invariants of 3-manifolds, and explain their relationship to manifold invariants coming from the Chern-Simons theory. We also explain the relationship between the known invariants of spin networks by using the Chain-Mail formalism of J. Roberts. This formalism can be understood as a quantum-group regularization of the BF theory path integrals."

Some other pertinent facts:
https://www.physicsforums.com/showthread.php?p=2817879#post2817879

 

[MTd2]

I can give you a connection between Spin Foams and Superstrings.

Lorentz Group is SO(3,1) ~ SU(2)XSU(2). SU(2) contain E(8) as a weyl group, that is, the icosahedron

So there is a kind of E(8)xE(8) group inside the Lorentz Group. Since we are talking about spin foams, we have at least 2 links for every node, so one E(8), for the past and to the future. Here we are talking about a weyl group, not a compact group, we cannot talk about particle content. Unless we make it compact it somehow by allowing the states of both sides to entangle. Thus a node in a spin foam would mimic a R(8) lattice. E copies from the past and 3 to the futures, would yield a group large enough to get the standard model in a similar way to the F-Theory.

 

[marcus]

Anyway to get back on topic, research actions--how a scientist behaves--speak louder than words in media interview or public lecture.

Professor Witten is not especially unique, among those who have at various times served as leaders in string research, in the sense that little or nothing he has done in the past five years serves to contradict the widespread impression that string is stalled or "going south". This impression, on the part of knowledgeable people, is not a "rumor" in the sense that you need to hear it passed along---anyone can see it for themselves.*

It seems to be somewhat simple-minded to pick on Witten---it gives his example too much significance. One could just as well focus on the direction taken by other prominent but younger people:

For example Nima Arkani-Hamed. Or Erik Verlinde and Petr Horava. Whatever they might say in public statements covered by the media matters far less than the directions actually taken in their research.

*There are several sorts of more or less hard evidence of "going south" that are not difficult to fish up for oneself, if anyone is interested. Here are some examples:
1. Citation counts to recent string papers have fallen precipitously
2. Compare Strings 2005 (Toronto) and Strings 2007 (Madrid) with the recent showing at Strings 2010 (College Station TX). All are online.
3. Dramatic growth in Loop Gravity and Loop Cosmology publication rates since 2005. Both doubled within spans of a few years, and continue.
4. String publication appears to be stagnant or declining.
Some links and other evidence of this are given here:
https://www.physicsforums.com/showthread.php?p=2817879#post2817879

 

[negru]

Funny you should name Arkani-Hamed, given the whole amplitudes stuff was born out of string theory developments. And given that part of Horava's motivation is driven by understanding some gauge/string aspects.

The thing is that people aren't classifying string theory as a separate research area from everything else anymore. Via ads/cft, twistor strings, klt relations (btw, how are those as an example of string theory calculations?), we know you can't have gauge theory or gravity without strings. Even if you go about finding ads/cft without strings (as suggested in a different topic), you soon realize that strings naturally come in anyway. We know strings are there and aren't going anywhere, so many string theorists are looking to understand the theory from a different perspective. For example, by looking at YM, CS, or other stuff.

 

[marcus]

Hi Negru, you mention things that can serve to illustrate my point of view.
I suppose someone might say that LQG-spinfoam research could not be considered a rival to string research because LQG uses twistors, Chern-Simons, holographic ideas, and has offered an approach to proving a form of the AdS/CFT conjecture.

In other words it might be said that lines of theoretical research should not be distinguished if they use the same mathematical tools. Are "mathematical cousins" so to speak.

That is not my view. I do make distinctions.
AFAIK twistors were introduced by Penrose and were not "string-inspired". They have become important in the LQG work of Freidel, Livine, and others. There is current ongoing work. Rovelli gives refs in his April 2010 review. Twistors also have been tried in string research but lately I haven't seen as much action in that direction as there was a few years back. Maybe you can cite some 2010 papers?

AFAIK Chern-Simons (CS) theory is a 3D theory which Witten introduced in work on knots, the Jones polynomial, perhaps his work on 3D gravity (which has been a longterm interest). I do not consider CS to be string-inspired simply because Witten developed it.
There are a whole bunch of CS papers in LQG. Recently (2009, 2010) has helped to refine and understand the LQG black hole entropy result. See papers by K. Noui, A. Perez and others.

In 2008 a LQG-theorist at Perimeter named Laurent Freidel posted a paper on AdS/CFT. In low dimension it offered an explicit procedure for reconstructing the (3D) bulk from the (2D) boundary. There may be other LQG work involving AdS/CFT. Holographic principles, more generally, come up. Ashtekar has a LQG proof of the Bousso entropy bound conjecture that extends previous work. The proof works where previous ones fail.
Freidel: "Reconstructing AdS/CFT" http://arxiv.org/abs/0804.0632
Ashtekar Wilson: "The covariant entropy bound and loop quantum cosmology" http://arxiv.org/abs/0805.3511

In any case string theory does not own those math tools. They do not carry the string brand. So if physicists work on them (CS, twistors, AdS/CFT) they are not necessarily doing string theory.

I think it is important to point that out. Otherwise the discussion can get very sloppy. Perhaps you would agree?

In case anyone is interested, here is the Astekar et al abstract:
"We examine Bousso's covariant entropy bound conjecture in the context of radiation filled, spatially flat, Friedmann-Robertson-Walker models. The bound is violated near the big bang. However, the hope has been that quantum gravity effects would intervene and protect it. Loop quantum cosmology provides a near ideal setting for investigating this issue. For, on the one hand, quantum geometry effects resolve the singularity and, on the other hand, the wave function is sharply peaked at a quantum corrected but smooth geometry which can supply the structure needed to test the bound. We find that the bound is respected..."

 

[marcus]

About Arkani-Hamed, I could be mistaken but I see his stuff from after 2007 as not much string content.
There was a November 2008 conference at Princeton (organized, I think, by Paul Steinhardt) where Nima had some disparaging things to say about the string Landscape, and about the usefulness of string in particle theory. So I checked his output from 2008, 2009, 2010. Maybe someone can explain to me that his papers are "really" string. That his focus has "really" not changed.
What is the Simplest Quantum Field Theory? http://arxiv.org/abs/0808.1446
A Theory of Dark Matter http://arxiv.org/abs/0810.0713
LHC Signals for a SuperUnified Theory of Dark Matter http://arxiv.org/abs/0810.0714
The S-Matrix in Twistor Space http://arxiv.org/abs/0903.2110 (This looks thoroughly 4D, no Xtra dims.)
A Duality For The S Matrix http://arxiv.org/abs/0907.5418
The Grassmannian Origin Of Dual Superconformal Invariance http://arxiv.org/abs/0909.0483
Unification of Residues and Grassmannian Dualities http://arxiv.org/abs/0912.4912

 

[atyy]

This guy claims AdS/CFT is part of string theory "In this sense, this approach has similarities with the philosophy of the Maldacena duality in string theory: a nonperturbative theory is dual to a more quantum field theory." http://relativity.livingreviews.org/Articles/lrr-2008-5/

 

[negru]

Like I said, this new S-matrix program was started up by twistor string theory. If you can use string theory to find out new things about old things (like YM) what's the problem? Isn't that the purpose? That's why I said string theory isn't separate anymore. It's there, and everyone uses it how they see fit. For example, if someone uses string theory to make a prediction about YM, and then people work on what that means for YM, are they doing string theory, or not? Technically no, but who cares? If some people then migrate to YM, is this a sign of string theory’s downfall?

Maybe you can get ads/cft without strings (and you can, at least heuristically - see eg Polchinski at TASI 2010), but the full equivalence includes strings, it would be silly to leave them out. If LQG people are working on ads/cft from their perspective, that's good for them, since whatever they find will automatically have implications for string theory.

 

[chrispb]

About Arkani-Hamed, I could be mistaken but I see his stuff from after 2007 as not much string content... Maybe someone can explain to me that his papers are "really" string. That his focus has "really" not changed.

I've read a majority of his twistor/S-matrix papers recently (most all of the ones except the dark matter ones you pointed out), and I can say with confidence that they are very much NOT string papers. He's been focusing on computing the N=4 SYM and N=8 SUGRA (but primarily the former)'s S-matrix using a twistor-y approach that reproduces known results from Feynman diagram calculations.

Some of his work can be CHECKED by the string people via AdS/CFT, but what he does is very much a part of the S-matrix program in QFT, with no direct reference to strings.

 

[marcus]

I've read a majority of his twistor/S-matrix papers recently (most all of the ones except the dark matter ones you pointed out), and I can say with confidence that they are very much NOT string papers. He's been focusing on computing the N=4 SYM and N=8 SUGRA (but primarily the former)'s S-matrix using a twistor-y approach that reproduces known results from Feynman diagram calculations.

Some of his work can be CHECKED by the string people via AdS/CFT, but what he does is very much a part of the S-matrix program in QFT, with no direct reference to strings.

Chris, Thanks for confirming the impression I gleaned from a quick look!

Someone else mentioned AdS/CFT which a lot of people associate in their minds with string theory. Some interesting comments on AdS/CFT were recently made in another thread:

Actually, maybe it is AdS/CFT that destroyed string theory (from the inside), because it turned out that AdS/CFT makes sense even without string theory, so many string theorists suddenly started to do AdS/CFT without doing string theory.

But then again, some say that whatever they do, they call it string theory. For example, strings do not play a fundamental role in M-theory, but they still call it string theory. In this way, string theory cannot collapse; it can only change its form.

AdS/CFT is basically due t'Hooft from the late 70's. The actual approach is also due to t'Hooft, published a few months before Maldacena, but it didn't include supegravity as an example, if I am not mistaken.

Personally I don't know the history but have often heard that the pre-string origin of the holo idea go back at least to 't Hooft. What I pointed out, on that topic, is that a LQG guy seems to have (corrected the conjecture and also) proved it in the 2D and 3D case. Finding a way to explicitly reconstruct the 3D bulk from the 2D boundary.
Freidel: "Reconstructing AdS/CFT" http://arxiv.org/abs/0804.0632
See post #13 above for some other examples of this kind of thing.

 

[negru]

The whole ads/cft without strings is pretty silly, seriously. I know some people's happiest day will be the day string theory dies, but promoting so much non-sense on a "physics" forum is troublesome at least.

Look, even if parts of ads/cft can work without strings, there are some objects on the gauge theory side which are dual to strings, not pure gravity. If you want ads/cft to be completely true, you need to account for ALL objects on both sides.

 

[tom.stoer]

One must not confuse AdS/CFT with the holographic principle itself. The holographic principle does only say that there should be a surface representation of a bulk theory. AdS/CFT is an exampe where this seems to work, but that does not mean that AdS/CFT has something to do with nature; and it does not say that there cannot be other frameworks where the holographic principle could make sense.

As far as I can see LQG has nothing to do with strings. In LQG the 3d Chern-Simons theory arises rather naturally on a BH horizon. And the holographic principle which is based on a kind of counting of microstates (it's not exactly that) seems to work. But that has nothing to do with strings nor with AdS/CFT. This is fine as if the holographic principle is a deep principle then it should not be restricted to a certain framework.

Regarding strings, LQG and other theories this means that the holographic principle seems to have the potential to be a true principle of nature even if certain theories which incorporate this principle may not be theories of nature.

Regarding you last argument that there are dualities in AdS/CFT with strings (not pure gravity) can very well be interpreted as a weak point AdS/CFT - at least as long as there is no physical (experimental) evidence for strings.

 

[ensabah6]

One must not confuse AdS/CFT with the holographic principle itself. The holographic principle does only say that there should be a surface representation of a bulk theory. AdS/CFT is an exampe where this seems to work, but that does not mean that AdS/CFT has something to do with nature; and it does not say that there cannot be other frameworks where the holographic principle could make sense.
.

We know nature is Desitter so I've wondered why all the excitement?

 

[tom.stoer]

Because you can do a lot of explicit calculations, you have more sound proofs than for other string-related stuff, you can apply AdS/CFT to quark gluon plasma, ...

 

[negru]

How about this?
Chern-Simons Gauge Theory As A String Theory
http://arxiv.org/abs/hep-th/9207094

What do you mean the fact that parts of ads/cft need strings is a weak point? You can use strings to predict things about SYM, CS, and other gauge theories (and people have done so with great success). Strings ARE there, whether you like it or not, and whether you'll ever see them actually wiggling about as you naively expect. String theory, in this regard, is equivalent to gauge theory. As long as you have gauge theories, you'll have string theory. (so far only susy YM, true, but there are many reasons to expect that susy is not crucial).

Btw, you might need to refine what "physical evidence" for strings means. If I use string theory to predict, or explain, something about QCD, via a gauge/string duality, would that count as evidence? We don't know what string theory is, like I said before, so any calculation involving strings is just as good as the next.Just to clarify, gauge/string dualities might have nothing to do with "traditional" string theory. Even if you go to whatever huge energies and never find any strings, that means nothing for eg ads/cft (because we already know that ads/cft is correct and it includes strings). I personally suspect that what we have so far (gauge/string duality, usual compactifications etc) are just hints that some underlying string theory exists. I wouldn't bet that either of these approaches will be the right one in the end.

 

[ensabah6]

Because you can do a lot of explicit calculations, you have more sound proofs than for other string-related stuff, you can apply AdS/CFT to quark gluon plasma, ...

Maybe nature is AdS and astronomers are wrong?

Is there a proposal for holography for DS?

 

[tom.stoer]

This discussion - and many others here - are the reason why I started this thread https://www.physicsforums.com/showthread.php?t=419450