Causes of loss of interest in String program

MTd2

This is a more likely explanation. But with a caveat: people are getting tired.

negru

For a rigorous analysis to prove that statement you should keep track of total number of citations in het+qg. Also h-index. In the past few years more and more garbage is showing up on the arxiv, from instituions i'll call sketchy at best.

I'd like to see evidence that papers and citations that in past came from string theory are now coming from alternatives. Just counting isn't enough. My own impression is that string people are doing SCFT's(amplitudes, localization, index stuff, a/c/f theorems), while newcomers especially from Europe are doing the alternative stuff. Also jobs are going to phenomenology related stuff, which is of course natural because of the LHC.

I'd also be curious to find out what job situation is in CMT to compare.

mitchell porter

What is your opinion of grand unified theories and supersymmetric field theories?

Aidyan

How do you know what data threshold is needed? The data might well be already in front of our eyes but we can't see it because probably we don't want to give up our still too classical mindset. That's what happened to those who insisted on epicycles like Tycho Brahe, or to Poincaré who had all data and couldn't see relativity as Einstein did, or to Einstein himself with QM, just to mention some. I don't think it is only about available data, but about a message from nature we still don't want to swallow.

You can't compare with the actual state of affairs the exceptional individual cases of some people who showed up from time to time throughout a centuries long period and that otherwise had almost no scientific progress. Organized science began with Galilei, Newton, perhaps even later. And since then, and also before that, there is nothing in history like the effort in terms of people, research and money set behind a single project like string theory *AND* with no concrete results.

As I wrote, I'm not in a position to express too technical judgements apart of my doubts from the historical perspective (and BTW, if someone wants to help me: http://www.physicsforums.com/showthread.php?t=594293 ). I spent several years on applied physics and only recently turned my interests to more theoretical issues. But this happened just because of the "string crises". Intuitively I always perceived string theories as nonsense. I can only see in it the attempt of human mind to impose its naive classical aristotelian and anthropomorphic understanding of the macro-cosmos on a micro-cosmos that obviously refuses to be imprisoned in such narrow limits. But on the other side I always told myself: "those guys know better than you, they have certainly good reasons to believe in it, let me see....". But now, enough is enough.... As to GUT they appeal certainly much more to my aesthetic sense. However, since they didn't produce concrete results either, this might be a lesson too: beauty is not a criterion. But something tells me that nucleons aren't stable, as predicted by several GUTs. I would keep an eye open on experiments looking for that.

mitchell porter

This is why I asked your opinion of GUTs and supersymmetric QFT - to see if you thought that they differ from string theory in this regard.

Let's recall another historically unprecedented situation: the existence of a single theory which does explain almost all of physics. This is the standard model, which has existed since the 1970s and has only needed the addition of neutrino masses, and a dark sector about which there is almost no data, to remain valid.

GUTs and supersymmetry and string theory have all grown up in the era of standard model dominance. GUTs are held to explain certain features of the standard model, like the hypercharge assignments; supersymmetry is supposed to give us dark matter, GUT coupling unification, and stabilization of the Higgs mass. Specific theoretical constructions give us, not the exact particle masses, but ratios between them with the right order of magnitude.

Nonetheless, none of these beyond-standard-model theories has yet become the new standard. It is a mathematical fact that there are innumerable possibilities to explore, even just within the framework of supersymmetric GUTs, because there are innumerable possible field theories which reduce to the standard model at accessible energies.

String theory has also turned out to contain innumerable possibilities, but they do have one new feature (apart from containing gravity): these distinct stringy possibilities do not come with continuously adjustable parameters, unlike field theories. Therefore, they are potentially more predictive than field theory. Unfortunately, like QCD, in practice it has proven very difficult to extract the predictions. The ability to calculate in string theory does progress, but this progress takes years to occur, and requires new mathematics.

There is continuity between the field-theoretic research program of unification and supersymmetry, and the research program of string theory, because the field-theory limit of string theory is typically a grand-unified supersymmetric theory. It has also been discovered that some field theories are simply equivalent to string theory on the specific corresponding background; the strings are essentially flux-lines in the field theory, and the extra dimensions emerge from scalars. It's likely that QCD itself is equivalent in this way to string theory on a particular background.

So reality does look a lot like string theory, because string theory looks like gauge theory plus gravity, and increasingly we also learn that gauge theories, like the standard model, look like string theory! It may be that some of the dominant physical hypotheses about how string theory works are misguided. Perhaps there's no supersymmetry, or no supersymmetry until ultra-high energies; perhaps the "extra dimensions" are algebraic rather than geometric. But it's also still very possible that the central hypotheses of the field are entirely correct. We might be living in a heterotic compactification with weak-scale supersymmetry, neutrino masses coming from the GUT scale, and so on.

When you say string theory has "no concrete results", I can't agree. What it has given us is a very large number of models which incorporate and complete the field theories that non-string theoretical physicists were already using, and which have the potential to explain the quantities which are just input parameters for something like the standard model. We know that string theory can get close to reality in various ways. One has every reason to hope it can go all the way.

MTd2

It was not casual. We lost a lot of information on those people. Only the very best survived because, as I said, the availability of recording media was extremely scarce.

negru

Just an example off the top of my head, Heckman and Vafa compute the CKM matrix to within something like 1% from F-theory http://arxiv.org/abs/0811.2417

arivero

If you are happy with 50 years without concrete results, then my previous example, the quest for quadratures of areas and cubatures of volumes, holds. Note that Kepler http://www.matematicasvisuales.com/e...oliometry.html Nova Stereometria doliorum vinariorum is from 1615, Cavalieri atoms, "Geometría indivisibilibus continuorum quadam nova ratione promota" are from 1635 and the final "concrete" results, Newton' "Analysis per aequationes número terminorum infinitos" and then the Principia, are from from 1669 and 1687. During this time, most of the Natural Philosophers (ie MathPhys), from Glasgow to Rome, were putting a lot of resources on this, and before Newton all they got was to confirm the Greek results and to add some extra examples.

Now I agree, that String Theory is near to break this record (and we will see if it gets something "concrete" out of it).

Aidyan

With "concrete" in the case of ST I obviously meant "experimental evidence" which supports ST against other candidates, not just mathematical developments.

That's precisely what I find unconvincing. There are "various ways" to build a general theory that reduces to a previous one and yet turns out to be wrong. Especially if one is free to chose among a "very large number of models". Bohr's atomic model (which did "not come with continuously adjustable parameters" too) got 'close to the reality' in some respect. And Sommerfeld could refine it making it even closer to reality. But soon broke down because it couldn't account for the spectra of atoms much beyond Hydrogen. One can shows that it is possible to describe the observed planets trajectory on the sky with arbitrary precision all the way in a geo-centric model by adding epicycles to epicycles (very reminiscent of today's perturbative approaches....). But then Galilei brushed all this aside by observing the phases of Venus. Where would physics have ended by insisting on these paths because "reality does look a lot like..."? As long as you don't have the observational anomaly that can be explained only by one theory alone where all the others fail and that makes a minimum amount of testable predictions where all the other models predict something else, the argument of coming "close to reality in various ways" is week.

As far as I know very concrete results were already obtained by a single philosopher like Archimedes with his exhaustion method. Cavalieri made some progress in between, and he furnished also some results. And Leibniz, Descartes and Newton followed giving us integral and differential calculus. These were steps where someone could do something out of it, not just a complicate theory about the world that was unclear if it was correct or not.

What is there more "concrete" than estimating the prize of wine barrels? But it is interesting that to justify ST's supposed "slow success" one has to resort to dubious examples that are more than four centuries old. Do you have an idea in what miserable material, cultural and even more academic and scientific conditions was the world at those times? I live in the city where Galileo observed for the first time the Milky Way, the sun spots, and Jupiter's satellites. The historical documents tell that there was almost nothing here, apart from a hill with few houses, and that thing they called a "university". He did all alone by himself. There was nothing like large scale collaborations on a project, the universities, the laboratories, the institutions and organized science we have today. It is quite natural that scientific progress proceeded extremely slowly in those times. And yet, it is a miracle that these people could produce something tangible almost alone in the time span of a lifetime. I believe that string theoreticians should refrain from pointing at such examples, that doesn't make them look well....

arivero

Disclaimer, I am not a string theoretician. I just happen to be intrigued by the theory and my still current opinion, developed in other thread here , is that most of the important work was done in less than five years, say 1968-1973, and then they did a wrong turn towards Planck energy.

marcus

Hi Arivero, Aidyan, Mitchell, MTd2 and others thinking about the causes of observed string decline.
(on that, see post #340 http://physicsforums.com/showthread....10#post3854410 )

I'd been favoring the "no-fault" idea that there's nothing wrong that wasn't already known over 10 years ago and the cooling of interest could be attributed to the appearance of non-string alternative approaches to QG. But Matt Visser posted a paper yesterday that I think could represent real substantive trouble.

Visser cites stringy black hole work by people like Strominger, Horowitz,... and calls their results/conjectures into question as unphysical. In other words he finds actual physical fault, not merely failure to be predictive. I'd appreciate any comment on this. Does anyone see reasons to dismiss or minimize Visser's argument?

http://arxiv.org/abs/1204.3138
Quantization of area for event and Cauchy horizons of the Kerr-Newman black hole
Matt Visser (Victoria University of Wellington)
(Submitted on 14 Apr 2012)
Based on various string theoretic constructions, there have been repeated suggestions that the areas of black hole event horizons should be quantized in a quite specific manner, involving linear combinations of square roots of natural numbers. It is important to realise the significant physical limitations of such proposals when one attempts to extend them outside their original framework. Specifically, in their most natural and direct physical interpretations, these specific proposals for horizon areas fail for the ordinary Kerr-Newman black holes in (3+1) dimensions, essentially because the fine structure constant is not an integer. A more baroque interpretation involves asserting the fine structure constant is the square root of a rational number; but such a proposal has its own problems. Insofar as one takes (3+1) general relativity (plus the usual quantization of angular momentum and electric charge) as being paramount, the known explicitly calculable spectra of horizon areas for the physically compelling Kerr-Newman spacetimes do not resemble those of currently available string theoretic constructions.
15 pages

Here are papers Visser cites, which he appears to be shooting down:
[1] G. T. Horowitz and A. Strominger, “Counting states of near extremal black holes”, Phys. Rev. Lett. 77 (1996) 2368 [hep-th/9602051].
[2] E. Keski-Vakkuri and P. Kraus, “Microcanonical D-branes and back reaction”, Nucl. Phys. B 491 (1997) 249 [hep-th/9610045].
[3] G. T. Horowitz, J. M. Maldacena and A. Strominger, “Nonextremal black hole microstates and U duality”, Phys. Lett. B 383 (1996) 151 [hep-th/9603109].
[4] E. Halyo, B. Kol, A. Rajaraman and L. Susskind, “Counting Schwarzschild and charged black holes”, Phys. Lett. B 401 (1997) 15 [hep-th/9609075].
[5] G. T. Horowitz and J. Polchinski, “A Correspondence principle for black holes and strings”, Phys. Rev. D 55 (1997) 6189 [hep-th/9612146].
[6] F. Larsen, “A String model of black hole microstates”, Phys. Rev. D 56 (1997) 1005 [hep-th/9702153].
[7] M. Cvetic and F. Larsen, “General rotating black holes in string theory: Grey body factors and event horizons”, Phys. Rev. D 56 (1997) 4994 [hep-th/9705192].
[8] M. Cvetic and F. Larsen, “Greybody Factors and Charges in Kerr/CFT”, JHEP 0909 (2009) 088 [arXiv:0908.1136 [hep-th]].
[9] M. Cvetic, G. W. Gibbons, and C. N. Pope, “Universal Area Product Formulae for Rotating and Charged Black Holes in Four and Higher Dimensions”, Phys. Rev. Lett. 106 (2011) 121301 [arXiv:1011.0008 [hep-th]].
[10] A. Castro and M. J. Rodriguez, “Universal properties and the first law of black hole inner mechanics”, arXiv:1204.1284 [hep-th].

Physics Monkey

I believe "shooting down" is way too strong a phrase to use!

Many of the references singled out by marcus concern themselves with supersymmetric situations not encountered in the real world. Supersymmetry can affect, among other things, the running of the coupling constants which happens to be one of Visser's first criticisms.

Let me take pains to emphasize that I am not saying the Visser article is not interesting, I merely suggest that way more parsing of the works involved needs to be done before anyone has any right to claim that Visser is shooting down anyone.

For example, a simple explanation would be that Refs. 1-10 mostly consider susy models (or very nearly susy models) while Visser mostly considers very non-susy models. I'm only half joking when I suggest that "general" to a string theorist could mean "true among susy models".

mitchell porter

It looks to me like Visser hasn't asked any of the string theorists he cites what they make of his observation - and that's foolish. They could probably set him straight in five minutes. He's acting like one of those people who hope to show that the whole of particle physics is a fallacy because of some single calculation they made. I don't doubt that there are challenges to understanding the entropy of real black holes in stringy terms, but I'm sure this isn't one of them. I'm just a string-theory amateur, but if I figure out what he's overlooking (and whether he nonetheless has a valid point somewhere), I'll post about it.

MTd2

String theory might be a fallacy, so, what s the problem? Developing string theory as a physical theory can indeed be as tricky as programming a very complicated computer game, and that s all. N

mitchell porter

The specific problem is that Visser frames his paper as a criticism of a string theory formula, but he doesn't understand string theory and didn't consult with anyone who does.

Most of his paper elaborates on non-string formulas found in references 19-22, and I think not in a very sophisticated way. Those formulas relate black hole horizon areas to spin, mass, and charge. What Visser then does is to reexpress the charge in terms of the fine structure constant, also using the fact from particle physics of charge quantization. But he just does this at the level of elementary algebraic substitution. If he's going to talk about the impact of charge quantization on black hole entropy, shouldn't he be using quantum field theory - maybe something like 't Hooft's brick-wall model from 1985?

Anyway, most of his paper is devoted to these slight elaborations of non-string formulas. But in the paper he has released, he starts out by quoting a few string formulas (or conjectures) in which horizon areas are functions of integers. His thesis here is that these formulas can't be valid for the real world, because (so he says) they would imply that the fine structure constant equals 2, not approximately 1/137. And besides, alpha runs, so it has to pass through irrational values.

So here is the most obvious gap in what he writes: he says nothing about how to interpret the significance of a varying alpha for his own new formulas, the ones which he advances as the correct alternative to the string formulas. Is the reader to assume that black hole entropy runs with the energy scale? He's just silent on this issue. If he had bothered to develop his new formulas within a proper QFT framework, he would have an answer. As things stand, apparently the reader has to figure this out for themselves.

Similarly, he doesn't even raise the question of how running coupling constants figure in string theory. In string theory, the tree-level coupling constant depends on the VEV of the dilaton. My ignorance is such that I don't know whether this is already taken into account in the string formulas for horizon area, whether it's an invisible normalization factor, or what. But again, Visser says nothing about this issue, he evinces no awareness of it. And that is precisely why I can say that he's been sloppy in relating his new formulas to string theory - he hasn't bothered to find out how the string formulas work, he's just relying on his own personal guess as to how they work.

MTd2

I am aware that you are studying string theory for quite some time. You seem to be a smart person, yet, these details, which seem to be not so subtle, still are hard to find and even to study. More over, there is no empiricism involved here.

So, that reinforces my thoughts about giving up on strings because they are too hard with little practical reward. Perhaps, other theories are also hard, but you can cut the loose buds and simplify the teaching/exposition literature, for the reason that you can do experiments.

marcus

Since we've turned a page, I'll recap the basic input we are trying to explain. There's been a decline in string interest and activity that shows in various ways. One is a downtrend in the rate of first-time faculty hires, starting around 2001. This is visible both in terms of absolute numbers (from average about 9 per year down to around 1 per year) and also in terms of string as a fraction of total Particle Theory. It used to be that around HALF the first-time faculty hires in HEP theory were in string, now it's more like a tenth.

A physicist at the U Toronto (Erich Poppitz) charts first time faculty hires in High Energy Physics Theory (Usa and Canada) by year and keeps track of what fraction of these are in string, and which fractions are in other branches of theory.
http://www.physics.utoronto.ca/~poppitz/Jobs94-08
His chart shows 11 HEP theory hires in 2011 of which one was string.
Here I've smoothed annual rates by averaging over 3 years intervals.The source used is http://particle.physics.ucdavis.edu/rumor/doku.php

There's also been a decline in annual citations to recent string research by the theorists themselves.
Number of recent string papers making the top fifty in the annual Spires HEP topcite list
Here a paper is counted as recent if it appeared in the previous five years. Citations gauge the quality/significance of current work by how much other researchers in the field currently refer to it.

It could be that the decline in String research is not primarily due to any flaw or inadequacy in stringy physics that was not evident already, say, by 2001. It could simply be that newer approaches to QG and explaining the SM have arisen, and that researchers have a natural tendency to spread out seeking fresh ideas and new areas to work on. The String program is some 40 years old and there has been plenty of time for new ideas to germinate and new research directions to develop.

On the other hand we can try to identify inherent shortcomings in the physics which might be responsible. Arivero made the point that we should consider String separately as a source of particle models (where it seems to have limited utility) and of candidate theories of the quantum geometry of the universe.
In the latter role, does it offer promising ways to resolve the cosmological singularity and model conditions leading up to the start of expansion? Are these testable by astronomical observation? Such seem among the main things one wants from a QG theory. In short, it's still debatable whether there are valid physics reasons for the decline.