Causes of loss of interest in String program

[marcus]

There has never been so much defensiveness towards string theory as it is now. The reason it is that there has been no progress on it. No experimental data for it. No blue sky.

I think you could say "never so much defensiveness on the part of the string community" as there is now. "Towards" doesn't sound idiomatic.

Anyway I think that's right. If you think back to 2004 say, there was this huge joyous complacency and youthful enthusiasm. Even after the KKLT and the Anthropic Landscape shock in 2003 there was still great confidence and conviction.

If it helps remind you, look at the Paris Strings 2004 website and think back.
http://strings04.lpthe.jussieu.fr/
I can't say I understand the reasons for the change---it might be those you mentioned, or might be something else (a missing ingredient?)

 

[atyy]

I think you could say "never so much defensiveness on the part of the string community" as there is now

But I will argue with marcus.

 

[MTd2]

I think you could say "never so much defensiveness on the part of the string community" as there is now

Yes, that is what I meant :)

 

[MTd2]

I can't say I understand the reasons for the change---it might be those you mentioned, or might be something else (a missing ingredient?)

The ultimate reason is too much effort for no gain at all. I am not sure, but another reason, perhaps, is in front of you. Just look at how much electronics evolved, with very little theoretical advance on the fundamental side. That makes everyone else envy and frustrated.

 

[martinbn]

Well, even without new experimental data, it has to at least match old data - the Einstein equations! Any approach that doesn't get this cannot be said to be quantum gravity. As far as I know, string theory is the only approach to quantum gravity that gets this at the moment.

A side question, for which I apologize, could you direct me to a place where this is done. I mean a paper (or anything else) where it is shown how to derive Einstein's equations from string theory.

 

[marcus]

Remember that the Einstein equation involves no prior geometry. Geometry is fully interactive with matter. (No fixed background geometry specified.)
I believe Einstein sometimes called this the "principle of reciprocity."

A side question, for which I apologize, could you direct me to a place where this is done. I mean a paper (or anything else) where it is shown how to derive Einstein's equations from string theory.

As I recall what stringy math mainly does in this direction is derive a graviton on a fixed prior geometry.
Anyway a spin 2 field which could be interpreted as a graviton. No one else seems to have answered. This is to the best of my memory. Not a full background-independent realization of GR, but a nice step in that direction!

 

[negru]

There has never been so much defensiveness towards string theory as it is now. The reason it is that there has been no progress on it. No experimental data for it. No blue sky.

The reason is that until recently we haven't had computer administrators at columbia write books and blogs on the failure of string theory.

Just some of the replies in this thread show how much poison people like the above have introduced.

There is no more argument of whether string theory is useful or will survive. Because of things like ads/cft, the klt relations, string theory is here to stay, and that's it.

 

[suprised]

negru, don't waste your time on this, they will never want to understand. Better talk to scientists.

 

[petergreat]

OK folks, here is the revised tabulation. No one has explained why there were so many papers for so long and then suddenly, in most cases, so few:

          1995-1998      1999-2002      2003-2006      2007-2010

Strominger     23             14             22              4
Maldacena      27             33             24              9 
Polchinski     21             17             11              4
Harvey,J       16             15              9              2
Duff,M         24             17              8              5
Gibbons,G      17             29             11              2
Dijkgraaf      18             11              9              7
Ooguri         31             18             13              8
Silverstein,E  16             15             16             10
Seiberg,N      19             16             22              1

Obvious explanation: Aren't these people getting old?

 

[marcus]

Obvious explanation: Aren't these people getting old?

Yes! that's what was wrong with my first attempt to look at decline of abs numbers of string&brane-related papers. Most everybody I thought of was already middle-aged!
I didn't want rigorous measure, just an informal spot check. And as it turns out this does not say anything about the AMOUNT of research, but about the focus. Using the keywords "string model" and "brane model".

However Suprised kindly suggested some names of YOUNGER people for me to look at. So I did the same quick check on their publications. Gaiotto, Alday, Neitzke, Nekrasov.
Their productivity is presumably on the upswing.

Remember this is not measuring INTEREST which I've defined for the purposes of this thread to be measured by current citations (annual standings of recent papers in the Spires Top Fifty) and conference attendence.

If there is a trend away from specifically studying strings&branes that could be interesting though (could indicate increased mathematical abstraction and sophistication---less of a concrete buildingblock approach---could. It really remains to be interpreted, if we see a trend there)

So Peter, here is an improved table with the YOUNG researchers added (still just a very informal spot check for a trend, not to make a fuss over either way)

What I am measuring here is number of papers that DESY classifies as "string model" and "membrane model". The interesting thing, which I would like explained, is why (from this informal sample of people) there used to be a lot of papers in that category and now are less. Maybe there is a very simple explanation.

I felt I should redo the table in part because Suprised suggested I look at younger theorist's output in the same light as I did the others earlier. The table is not especially important to the discussion. Since we are on a new page, here it is:

          2003-2004      2005-2006      2007-2008      2009-2010
Gaiotto         5              9              5              4
Alday           1              5              4              0
Neitzke         3              2              1              0
Nekrasov        4              3              1              2  
Strominger     15              7              4              0
Dijkgraaf       4              5              3              4
Polchinski      8              3              2              2
Maldacena      17              7              5              4 
Gibbons,G       6              5              1              1
Harvey,J        4              5              2              0
Ooguri          7              6              3              4
Silverstein,E   9              7              5              5

In connection with this table I am not asking some other question which you or others may have in mind. As I said this tabulation is not about "interest in string theory" broadly interpreted. Nor is it about "string techniques" etc etc. or what "should" be classified as string.

What I think it may have to do with is what I called "increased sophistication".

If you want to filter out the age effect then just look at the first four---they are the young rising stars suggested to me by Suprised. Interestingly this trend away from involvement with the concrete buildingblocks and towards the abstract may also be visible with them. You have to decide for yourself---small point, not worth spending a lot of time on.

 

[MTd2]

There is no more argument of whether string theory is useful or will survive. Because of things like ads/cft, the klt relations, string theory is here to stay, and that's it.

Show me any technique that uses string theory perform better than the older approaches in terms of accuracy. Fitting cross sections in a qualitative way of heavy ion collisions is not enough because it could be a simple matter of algebraic numerology without content. If that was enough, MOND would be much more widely accepted because it fits galaxy rotations better then ACDM models, which is the standard theory.

So, it doesn't matter much the theorists are from IAS, Princeton, Rutgers or Harvard. Physics must run in labs too. Real experimental scientists don't care about string theory, because it cannot be ruled out. I am not even thinking about Peter Woit. I am thinking more about, say, Tommaso Dorigo. He designs experiments and runs them. So, everyday, he accomplishes more, in real science, than all string theorists together, doing string theory. Well, not only him, but anyone any random biologist, growing, tediously, bacteria in warm oven plates.

 

[negru]

I wasn't arguing whether theoretical physics is worth exploring. I just said that string theory is completely and irreversibly embedded in what we already know. It doesn't mean that string theory is the ultimate theory, but if there is anything else out there, it will have to explain everything that string theory already uncovered.

 

[MTd2]

Unfortunately, this kind of deep theoretical explanation is let aside by almost all string theorists. It is a rare case where I agree with Lubosh.

 

[negru]

Out of curiosity, which is the paper you're most excited about?

Yeah like Haelfix said it's Nima's stuff. I'm more interested in the gravity side in all of this though, since after understanding gravity from this perspective it will be easier to jump to non-planar YM. And then comparing the different aspects of all these theories I strongly believe will allow us to generalize to many other QFT's. I also believe it can be pushed beyond scattering amplitudes, and then the implications would obviously dwarf things like ads/cft. And through ads/cft it will mean a lot to string theory as well. I mean yeah there are a lot of "if"s, but this is what I find most promising.

 

[marcus]

This is to remind everybody that for the purposes of this tread we are using current citation standings as a measure of the current level of interest and activity in the field.

Also I said attendance at the annual conference---that's another conventional gauge used by people in academic walks of life. We tracked that over the past 10 years and it looked roughly flat.

Citations aren't a perfect gauge but they do tend to show the value of a paper to other researchers. If they refer to it a lot in their own work then the cite-count goes up. If the community is producing a lot of exciting papers then they get cited and a lot of those papers will make the annual Spires Top 50 list. We tracked that, and it showed a decline. But at this point there seems no need to make a big fuss over that. I may be able to get some other "citations to current research" measures to look at for comparison. But that's not the main thing!

What I really hope we can dredge up in this thread is constructive ideas of directions the program could go, that would increase its vitality, significance, interest (whatever you want to call it).

 

[negru]

And since it's a new page, I guess it's time to repost your analysis right?

 

[marcus]

And since it's a new page, I guess it's time to repost your analysis right?

I wasn't thinking of doing that, but thanks for reminding me! The charts are not the important thing, but they help focus attention.

Maybe I will put up a new table. You said in your last post that you were especially interested in the gravity side.

I'm more interested in the gravity side in all of this though, since after understanding gravity from this perspective it will be easier to jump to non-planar YM...

Maybe we should look at papers that DESY classifies "quantum gravity" and track the citation standings for both String and Loop programs.

Papers making the QG top thirty in terms of lifetime citations
Year                    2000      2002      2004      2006      2008      2010
Broadly String-related   8         4         3         5         5         7
Loop                     6        11        14        10        14        14

The String numbers may be slightly inflated because I counted everything I could Randall-Sundrum, a cosmology paper that only had one section about possible stringy dark energy, among other ideas, AdS/CFT of course and so on. But you see there is no decline to speak of.

 

[negru]

What I really hope we can dredge up in this thread is constructive ideas of directions the program could go, that would increase its vitality, significance, interest (whatever you want to call it).

Like I mentioned in many threads the whole scattering amplitudes business is a very fruitful direction people are going in. Witten, Maldacena and Arkani-Hamed among others have devoted quite a lot of their time to this topic. Whether that's "actual" string theory, pretty much nobody really cares. Sometimes it pays off to make small detours. And with Ads/CFT, working on YM is just the same as doing string theory. The whole point of Ads/CFT is to work on whichever side you wish. Only very twisted minds would interpret this as a decline in string theory research.

 

[negru]

Papers making the QG top thirty in terms of lifetime citations
Year                    2000      2002      2004      2006      2008      2010
Broadly String-related   8         4         3         5         5         7
Loop                     6        11        14        10        14        14

The String numbers may be slightly inflated because I counted everything I could Randall-Sundrum, a cosmology paper that only had one section about possible stringy dark energy, among other ideas, AdS/CFT of course and so on. But you see there is no decline to speak of.

I don't know Marcus, I kinda see an upper trend since 2004. The number of papers more than doubled since then! Wow! While Loop has been stagnating all the way since 2008/ I think things are looking of bad for Loops.

 

[marcus]

... the whole scattering amplitudes business is a very fruitful direction people are going in...

Thank you! You are offering a constructive suggestion. How about URL for a recent sample paper so we can be clearer on what you're talking about?

As encouragement to others ("surely you can do better than this") I will toss out a few of my ideas.

I think it would substantially liven up the program for it:

1. To get a math representation of the geometry of the universe.
2. To resolve the big bang singularity (so we could test some predictions by CMB observation.)
3. To represent geometry as fully interactive with matter (not fixed on ahead of time.)
4. To have the measurement process be somehow included in the model.

If other ideas have come up in the thread, and I've forgotten them, I hope someone will remind me.

 

[negru]

Maybe I will put up a new table. You said in your last post that you were especially interested in the gravity side.


Maybe we should look at papers that DESY classifies "quantum gravity" and track the citation standings for both String and Loop programs.
.

And if you had even the SLIGHTEST idea of what I was talking about, you'd know that what you concluded/understood was just completely retarded. Or maybe you were just looking for another excuse to post some loop vs string count...If you think that table you posted had just ANY connection with reality, or any meaning whatsoever, you are just completely out of it.

I honestly think I'm done with this thread. The amount of misinformation and misinterpretation of facts you are spewing is just unbelievable.

Thank you! You are offering a constructive suggestion. How about a recent sample paper so we can be clearer on what you're talking about?

MArcus, we have been talking about these things for ages...like I said, check Nima's, Maldacena's, Bern's, etc papers, that's what I'm talking about.

 

[negru]

Just to make it clear. For most of the time, the QG keyword is meaningless. For one, almost no string theorists put their papers under QG on the arxiv. The reason is that the QG is over flooded with garbage LQG papers. In fact I don't know a single physicist who even opens up the QG section to check recent papers. Even worse, if they are suggested to even look at the QG for a particular paper, they would ask: what is it doing under QG, if it's any good?
I have no idea how desy people do it, but I'm pretty sure it works about the same way.

 

[marcus]

I haven't evaluated (for myself) how good a job DESY does with the category "quantum gravity". I've found their classification job professional consistent and useful in other cases, but who knows how they've done with this category?

I'd be delighted to hear anyone's evaluation good or bad, so if anyone wants please do take a look at the lists:

2000:
http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=dk+quantum+gravity+and+date+%3D+2000&FORMAT=WWW&SEQUENCE=citecount%28d%29
String 8 Loop 6

2002:
http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=dk+quantum+gravity+and+date+%3D+2002&FORMAT=WWW&SEQUENCE=citecount%28d%29
String 4 Loop 11

2004:
http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=dk+quantum+gravity+and+date+%3D+2004&FORMAT=WWW&SEQUENCE=citecount%28d%29
String 3 Loop 14

2006:
http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=dk+quantum+gravity+and+date+%3D+2006&FORMAT=WWW&SEQUENCE=citecount%28d%29
String 5 Loop 10

2008:
http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=dk+quantum+gravity+and+date+%3D+2008&FORMAT=WWW&SEQUENCE=citecount%28d%29
String 5 Loop 14

2010:
http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=dk+quantum+gravity+and+date+%3D+2010&FORMAT=WWW&SEQUENCE=citecount%28d%29
String 7 Loop 14

People may wish to see what I have included as "broadly string-related". I have included some papers which resorted to Sugra4 or Sugra8 without mentioning strings or branes. Also I included paprs with Randall-Sundrum contructions, and whatever AdS stuff showed up. And Bousso's 2002 entropy bound, which does not depend on string (the result was extended by Ashtekar using LQC so as not to fail at the big bang singularity).

The whole tabulation thing is of no great importance, but anyway there it is. Some people may wish to make pronouncements like, in a manner of speaking, "DESY keywords are no damned good!" or, so to say, "quantum gravity papers are all rubbish!" But I think that going forward into the future, the quantum gravity research category will actually be quite important, and DESY generally does a pretty good job on other things, on the whole, and somehow it matters when some papers make the Top 30 in citation standings. It is an (imperfect) measure of quality and performance. Any comment?

 

[negru]

The whole tabulation thing is of no great importance, but anyway there it is. Some people may wish to make pronouncements like, in a manner of speaking, "DESY keywords are no damned good!"

Marcus you have been proven with CONCRETE examples that DESY keywords are, in fact, no good.

 

[marcus]

Not so, I think In any case it should be possible for anyone to inspect these lists for themselves, especially if you are at all familiar with the literature.

 

[Sardano]

I don't know Marcus, I kinda see an upper trend since 2004. The number of papers more than doubled since then! Wow! While Loop has been stagnating all the way since 2008/ I think things are looking of bad for Loops.

I have checked, and I agree. I think it has to do with the impossibility of making any advance in the resolution of the Hamiltonian constrain, and the impossibility to reproduce the BH entropy. I think it is related too with the failure of the "big assumption" of the theory: background independence: they have been forced to abandon background independence in order to make any advance (I am talking about LQC), and even there they haven't been able to deal with any "physical matter content" (they only deal with toy models). Besides, Enrique Alvarezs arguments and others in the same line, make LQG a dead end.

We also have to realize that LQG is a theory that does not take into account that GR is not a perfect theory and maybe subject to corrections at the classical level (the ones predicted by ST, so the right action to quantize wouldn't be the EH action). They have forgotten that we only have a few tests of GR, and they have forgotten that at cosmological scales GR is probably not right. I still remember this Rovellis lame paper about "why the cosmological constant is not a problem" or something like that, trying to desperately justify thar GR is perfect at it is and LQG is not wrong. What a shame!

LQG came from generation of relativistic physics that were obssesed with GR, and they couldn't acept that maybe GR should be corrected. They forgot about the truly 20th century theories: the QFT that are being tested every day in particles accelerators.

Anyway, if you put all this things together, it is comprehensible that people is losing interest in LQG program. However, if you think there are other causes of loss of interest in the LQG program you are welcome to participate.

 

[marcus]

One beneficial side-effect of this thread, for me, is that I have gotten to know more about highly talented and productive younger string researchers. Suprised gave us some suggestions of people's output to look at (Gaiotto, Neitzke, Alday, Nekrasov...). And from there I have looked around some.

I would like to examine more of the young people's work for the trend to more abstract stuff (AdS etc) and stuff not EXPLICITLY involving superstrings and the like. A lot of work goes now in the direction of Sugra4 and Sugra8 some of which may be "related" or "inspired" but is not necessarily explicit string/brane.

So I decided to do this check again, but refine it using keywords
(DK superstring OR string model OR membrane model OR M-brane)

And after looking around I found that I had particularly high regard for a young researcher at AEI Potsdam named Niklas Beisert. So let me run that check with his work now to see if there is any trend.

Notice that this may not pick up ALL the papers that whoever might insist it "should" but we can do the same test repeatedly over time and see if there is a noticeable change. Again everybody should realize that in this case we are not gauging INTEREST in string broadly interpreted. We are just making a very rough informal spot check to see if we can see some change in the makeup or emphasis.

It has to do with what I was talking about earlier: increased sophistication or abstraction.
Or some other kind of shift of focus that we have yet to correctly identify.

Anyway this is an outstanding guy, I think, so I want to see what's going on keyword-wise with his research.

https://www.physicsforums.com/showthread.php?p=3274499#post3274499http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=a+Beisert+and+%28dk+string+model+OR+membrane+model+OR+superstring+OR+M-brane%29+and+date+%3E+2002+and+date+%3C+2005&FORMAT=WWW&SEQUENCE= (5)

http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=find+a+Beisert+and+%28dk+string+model+or+membrane+model+OR+superstring+OR+M-brane%29+and+date+%3E+2004+and+date+%3C+2007&FORMAT=WWW&SEQUENCE= (6)

http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=find+a+Beisert+and+%28dk+string+model+or+membrane+model+OR+superstring+OR+M-brane%29+and+date+%3E+2006+and+date+%3C+2009&FORMAT=WWW&SEQUENCE= (2)

http://www-library.desy.de/cgi-bin/spiface/find/hep/www?rawcmd=find+a+Beisert+and+%28dk+string+model+or+membrane+model+OR+superstring+OR+M-brane%29+and+date+%3E+2008+and+date+%3C+2011&FORMAT=WWW&SEQUENCE= (1)

          2003-2004      2005-2006      2007-2008      2009-2010
Beisert         5              6              2              1
Gaiotto         5              9              5              4
Alday           1              5              4              0
Neitzke         3              2              1              0
Nekrasov        4              3              1              2  
Strominger     15              7              4              0

Be cautious about attributing any meaning to this, of course. In string-broadly-interpreted (including papers mainly focused on AdS stuff and/or Sugra) these young people's output is defintely not declining. They are highly productive. But we can see a decline of explicit involvement in categories "superstring, string model, brane model, M-brane" for whatever reason. It could even be due (but I strongly doubt this) to a systematic error by the DESY librarians! One always has to be cautious. And as I've suggested, it could show a shift towards a different idea of what the main focus of string program is: a shift away from the old focus. I'm reserving judgment for the time being.

 

[marcus]

As a reminder of ideas that came up in the thread so far---for revitalizing the String program. What I'm hoping for are suggestions from other people along these and different lines. Those offered so far might not necessarily be viable or appropriate to the problem.

1. incorporate measurement process in the model
represent experiment/observation mathematically in the picture---so it is less about what nature is supposed to be made of and more about how it responds to measurement.
2. get a quantum cosmology we can test
mathematical representation of the geometry of the universe. Science has always been driven by this quest, from its beginnings in the 6th century BC.
3. resolve the singularity
IMHO probably this is the main job of quantum cosmo in our time
4. clearer focus and direction
the risk of having too much recital about how this and that is "string-inspired" and "string-related"---the String program does not own SUGRA and supersymmetry. they are shared resources that exist independently without string. Likewise dualities (other theories have them as well). To have a clear direction it is not enough to pursue Sugra and AdS-stuff and invoke some remote connection
5. geometry fully interactive with matter.
the need to break free from prior geometry. Einstein's reciprocity principle of no one-way influences in nature. No degrees of freedom which act on others but are not themselves acted upon.
6. avoid getting too elaborate and Baroque
two people in the thread mentioned Ptolemaic epicycles and Aristotelian physics. I don't know if that is a fair analogy. There was a clever exchange about it. Hermann Nicolai used the word Baroque in his 2009 talk at the Max Born symposium. So it is a concern of insiders. "becoming increasingly Baroque, if you follow the literature." Of course he is still writing string papers, as he should be! But he frankly and honorably avows his misgivings.

================
EDIT: Hi Fzero, nice to see a post from you here! I will reply briefly simply by editing this. Indeed it is nice that string math is serviceable and versatile. It is useful in several contexts (as you mentioned) and can provide the most economical way to derive something---at least that we know of as yet.
You need not imagine that I find this upsetting! I welcome it

Speaking of dualities (as you were) you might be interested in this:

http://arxiv.org/pdf/1010.2481
Einstein gravity as a 3D conformally invariant theory
Henrique Gomes, Sean Gryb, Tim Koslowski
27 pages. Published in Class.Quant.Grav.28:045005,2011

Each of the three authors is giving a talk at Loops 2011 this month in Madrid.

EDIT: brief reply to Fzero and Haelfix. Thanks to both for these thoughtful responses! (#239 and 240) Food for thought.

 

[fzero]

4. clearer focus and direction
The risk of having too much recital about how this and that is "string-inspired" and "string-related"---the String program does not own SUGRA and supersymmetry. they are shared resources that exist independently without string. Likewise dualities (other theories have them as well). To have a clear direction it is not enough to pursue Sugra and AdS-stuff and invoke some remote connection

Marcus, it is usually very clear to an expert what the origins of an idea are. It is not too much trouble to read though the literature on, say, SUSY gauge theories and decide what techniques are being used. No matter how upsetting it may be to you, the fact remains that there are results that either can only be derived using string theory techniques or are much simpler to derive from string methods as opposed to brute force field theory computations.

You also seem to have various misunderstanding about the existence of supergravity theories on their own. It is extremely unlikely that supergravities are UV complete, though maximal supergravities seem to be much less divergent than we would have thought. Therefore all SUGRAs should be thought of as effective field theories related to some superstring theory.

Finally, you have a vast misunderstanding of how basic research is done. There is no focus and direction provided by any central figure. Researchers are generally free to choose topics on their own, without interference from anyone. There are small exceptions, such as grad students and postdocs who might be expected to work closely with their advisor/supervisor. Even these are rare, as in larger groups, postdocs are very independent and the main thing stopping grad students from being so is maturity in the discipline.

Even funding and job security are not huge impediments to freedom of choice in research topics. Grant and hiring committees tend to do a good job of evaluating research on it's academic value. If you do excellent work you will get jobs and funding, whether you are working on AdS/CFT or CY compactifications. There is no one that will "officially" or otherwise tell someone publishing important papers on AdS/CFT that they should be studying cosmology.

Also, as I have mentioned previously, only a small amount of research that is begun actually gets published. It is impossible to determine what areas people have worked on if they decided not to publish for whatever reason.

 

[Haelfix]

As to the 'resolve the singularity point'. We've been over this before!

It might be the case that it is simply a bad question! String theory is a UV completion of General Relativity, and thus by definition new degrees of freedom become important.

Singularities of the classical metric tensor in particular backgrounds (read FRW universes) might simply not be a good description of the correct physics, in the same way that divergences in fluid mechanics are not necessarily physical because atomic physics becomes important at some scale. In any event, the assumptions of the classical singularity theorems by Penrose et al, are clearly violated in quantum gravity and so it is unclear whether they persist or not.

A better question is whether string theory has its own singularities and how it resolves them.