Has string theory actually done anything?

[Reaper]

Just a quick question for those who are knowledgeable in this area: in the 30+ years in it's existence, has string theory actually done anything in terms of furthering our understanding of nature?

 

I mean, there are a bunch of pop-sci books and so-called "documentaries" that all serve to overrate the merits of the theory, claiming that this will lead to the "Theory of Everything", or answering all of those annoying unanswerable questions (e.g. why *insert pseudo-philosophical topic here*?). But, after having read all of those newer books (e.g. Elegant Universe), they don't really say anything different from what Steven Hawking wrote in "A Brief History of Time", and that was written over 20 years ago! His "Universe in a Nutshell" adds a section about Branes and multi-universes, but beyond that there is nothing really new or "cutting edge" in those book.

 

And, as far as I know, they still haven't found a way to actually test the theory. Has there been any real progress at all with string theory, or it's application in anything in particular (Even if only previously unknown mathematical theorems)? If not, is there any other theory that actually seems like it might, you know, do something that pushes our knowledge of the universe further?

[Mokele]

Does it even qualify as a theory? Last time a checked, theories had experimental evidence.

[ajb]

Even if only previously unknown mathematical theorems?

 

I do not think that string theory has really be used to prove any new results in mathematics (principally I am thinking of geometry and low dimensional topology) but I know that it has been use to produce "simple physics" proofs of theorems.

 

More generally, there has been a massive exchange of ideas between mathematics and (supersymmetric) quantum field theory, statistical mechanics etc.

 

Key works include;

 

1) Witten's construction and generalisation of the Jones polynomial in knot theory using Chern-Simons theory.

 

2) Donaldson's work on smooth 4-manifolds using instantons and the subsequent generalisation by Seiberg and Witten .

 

3) The massive current interest in mirror symmetry and Calabi-Yau manifolds.

 

4) The Batalin-Vilkovisky formalism has been used in may areas of mathematics and physics. For example Maxim Konsevitch's deformation quantisation on Poisson manifolds. I myself have discovered some nice constructions based on the BV-formulism.

 

5) The so called Maldacena duality (AdS-CFT correspondence). This is a gravity-gauge theory duality that arose from string theory. Today, people are trying to use this duality to construct "gravity theories" dual to (nonSUSY) QCD. If this work string theory could indeed be a theory of hadrons!

 

So even if string theory and related things do not turn out to be useful in physics proper, they have and will continue to inspire lots interesting constructions in mathematics that themselves maybe useful in physics.

---

Merged post follows:

Consecutive posts merged

Does it even qualify as a theory? Last time a checked, theories had experimental evidence.

 

Not exactly, but there should be in principle testable predictions. These in principle at a later date be compared with experiment/observation.

 

Popper's ideas about falsifiability seem to be under great question in modern theoretical physics. Mathematical beauty and consistency seem to be the guiding principles. This is most peoples argument against string theory. No predictions.

[Mokele]

Not exactly, but there should be in principle testable predictions. These in principle at a later date be compared with experiment/observation.

 

Maybe I'm spoiled by being a lab-based guy, but IMHO if the predictions haven't been tested, it's still a hypothesis.

 

Popper's ideas about falsifiability seem to be under great question in modern theoretical physics. Mathematical beauty and consistency seem to be the guiding principles. This is most peoples argument against string theory. No predictions.

 

So is the idea "Oh, the rules of science are inconvenient, so let's change them?"

[ajb]

Maybe I'm spoiled by being a lab-based guy, but IMHO if the predictions haven't been tested, it's still a hypothesis.

 

This has come up before I am sure. Theory has a slightly different meaning in physics than the other sciences. This can be a source of confusion. Theory is synonymous with mathematical model. A "sensible" model should have predictions and a "good" model should reflect nature to some desired degree.

 

A mathematical hypothesis "if X then Y", is a statement within a theory which can be proved or disproved in general. I don't think it is any different to a conjecture.

 

So is the idea "Oh, the rules of science are inconvenient, so let's change them?"

 

Science is always evolving and if something stands in the way of progress (very hard to define progress) then why not modify/remove it?

 

Anyway, it looks like we have to live with the unfalsifiability of modern theories for the moment. Symmetries, mathematical beauty and consistency are the driving forces right now and less so experimental evidence.

 

However, any theory should in principle make predictions that in principle could be tested. It is really an open question if string theory is a "sensible" theory.

[Reaper]

Maybe I'm spoiled by being a lab-based guy, but IMHO if the predictions haven't been tested, it's still a hypothesis.

 

So is the idea "Oh, the rules of science are inconvenient, so let's change them?"

 

Not quite. I would have to agree with ajb's statement on this one. The approach to formulating theories in physics or mathematics are indeed a bit different, for the sole reason that many of the physical phenomenon cannot be directly observed or brought into the lab.

 

Indeed, the whole approach is quite different in that physicists start with a conjecture, and then just wait for the experiments to either prove or falsify them. Rather than the other way around. The basic idea is that if the mathematics are consistent with what we observe, then there must be some truth to it, even if it ultimately ends up being just an approximation, or not quite the "full story". This argument has been around since at least the time of Newton, and it is a very powerful one. I don't think even Karl Popper has been able argue against it or debunk it.

 

It is also for this reason that we still use Newton's theories even though Einstein showed him to be wrong, because they are an approximation to truth (whatever that may be....), and are still useful and easier to use. And most of all they are still valid at low velocities.

 

 

While on the other hand, biology/geology and the social sciences have way too many physical events occurring to make a proper mathematical model, and as such they must rely on experimental evidence.

---

Merged post follows:

Consecutive posts merged

1).......

2).......

3).......

4).......

5).......

 

 

So even if string theory and related things do not turn out to be useful in physics proper, they have and will continue to inspire lots interesting constructions in mathematics that themselves maybe useful in physics.

---

Merged post follows:

Consecutive posts merged

 

Ok, so there really is something substantial that string theory is doing then. I suppose the main reason I was a bit skeptical was because my only exposure to string theory (other than a couple of field equations) was primarily from pop sci books and other media hype, and I think we all know how reliable their information tends to be.

 

I'm wondering though, if it has helped provide better proofs for known physics and mathematics, has it also come out with any new or deeper underlying physical principles that may be actually experimentally tested in those areas of physics in the near future? I know you listed Maldacena duality as one that could potentially do that, but what about in the other areas that we are familiar with (e.g. statistical physics, etc.). Or are they all just tools to make doing the math easier...

---

Merged post follows:

Consecutive posts merged

 

This is most peoples argument against string theory. No predictions.

 

Well, strictly speaking it does make predictions. It's just that those predictions require a particle accelerator the size of the solar system to test for them...

 

Anyway, it looks like we have to live with the unfalsifiability of modern theories for the moment. Symmetries, mathematical beauty and consistency are the driving forces right now and less so experimental evidence.

 

I suppose I get a little impatient sometimes, especially since most of the information available is not all that reliable.

 

But of course, I think it should be mentioned that it took more than 70 years for Bose-Einstein condensates to be experimentally confirmed, so there is no reason to be discouraged if it has been more than 30+ years and we are still trying to figure out what, exactly, string theory will ultimately end up being.

Edited June 3, 2009 by Reaper
Consecutive posts merged.

[ajb]

I'm wondering though, if it has helped provide better proofs for known physics and mathematics, has it also come out with any new or deeper underlying physical principles that may be actually experimentally tested in those areas of physics in the near future? I know you listed Maldacena duality as one that could potentially do that, but what about in the other areas that we are familiar with (e.g. statistical physics, etc.). Or are they all just tools to make doing the math easier...

 

One of the deepest principles I think that string theory has made prominent is duality. Both the strong-weak (S-duality) and T-duality (long-short distance duality).

 

One of the earliest dualities discovered was S-duality in four-dimensional Yang-Mills theory with N=4 supersymmetry. It was discovered by Claus Montonen and David Olive. (I have met both of these people. David Olive was one of my undergrad lecturers.)

 

It is by now well known that in four-dimensional Yang-Mills theory with N=4 supersymmetry you get physically equivalent theories if you exchange the coupling [math]g[/math] for [math]1/g[/math]. Thus you get the same theories in both strong and weak coupling.

 

 

 

Well, strictly speaking it does make predictions. It's just that those predictions require a particle accelerator the size of the solar system to test for them...

 

String theory makes two quite general predictions (non-dependent on the vacuum)

 

1) Gravity. Closed strings have graviton states associated with them. There is no way around this. You start with a theory with no mention of gravity and you are forced to include it.

 

2) Extra dimensions. You can calculate the number of dimensions required for a non-interacting string to be consistent once quantised. There are several ways of doing this, but the outcome is 10 dimensions for superstrings.

 

 

 

Further predictions really require knowledge of the string landscape. In essence one needs to single out the true string vacua and make calculations form there. So far this has been a very challenging problem.

 

 

I suppose I get a little impatient sometimes, especially since most of the information available is not all that reliable.

 

A First Course in String Theory

Barton Zwiebach

Cambridge University Press (10 Jun 2004)

 

 

String Theory Volume I

Joseph Polchinski

Cambridge University Press (June 20, 2005)

 

String Theory Volume II

Joseph Polchinski

Cambridge University Press (July 11, 2005)

 

An Introduction to String Theory and D-Brane Dynamics

Richard J. Szabo

Imperial College Press; illustrated edition edition (July 12, 2004)

 

(Szabo I have met a couple of times, he is a nice guy.)

 

Other books I have found useful are Kaku Quantum Field Theory: A Modern Introduction and Introduction to Superstrings and M-Theory.

 

Then there is the string bible;

 

Superstring Theory: Volume 1, Introduction

Michael B. Green, John H. Schwarz, Edward Witten

Cambridge University Press (July 29, 1988)

 

Superstring Theory: Volume 2, Loop Amplitudes, Anomalies and Phenomenology

Michael B. Green, John H. Schwarz, Edward Witten

Cambridge University Press (July 29, 1988)

 

But it should be noted that these last two are pre-second string revolution and do not discuss branes.

 

 

But of course, I think it should be mentioned that it took more than 70 years for Bose-Einstein condensates to be experimentally confirmed, so there is no reason to be discouraged if it has been more than 30+ years and we are still trying to figure out what, exactly, string theory will ultimately end up being.

 

Indeed. Forget what Woit and Smolin say, string theory is not dead yet.