Answer your string questions for a nickle! II (string physics only!!!)

[vincent]

Hi everybody,

 

Let me be absolutely clear that by "strings physics only" I mean to not only exclude discussions of alternative theories, but also restrict discussion to technical and conceptual issues in string theory itself. If someone wants to discuss alternative theories in relation to discussions in this thread, they should initiate a new thread for that purpose. This thread is meant to help members (including myself!!!) deepen their understanding of string theory on it’s own merits and not in comparison to other ideas. To those members searching this statement for loopholes to exploit, I expect posters in this thread to also respect the spirit in which these remarks are intended, which should be quite clear to any half-way sentient creature posting in this forum.

[timo]

Well, let's start with a simple one: Apart from possibly giving a handle to describe quantum gravity, are there any other reasons to believe in strings that are not already features of less exotic models (most notably supersymmetry and GUTs). Which ones?

[insane_alien]

heres another thats bugging me. i'm a laymen when it comes to string theory so excuse any glaring errors.

 

it is often spouted by opponents of string theory that it makes no predictions and there will never be anyway to test it. quite bluntly, is this true?

 

this is bugging me because i really doubt there would be so much interest in string theory if there wasn't anything we could do to test it. even if the experiments may be out of our technological reach for the time being.

 

so, what would be a test of it and what sort of technology are we going to need to perform these tests?

[fredrik]

it is often spouted by opponents of string theory that it makes no predictions and there will never be anyway to test it. quite bluntly, is this true?

 

This may sound like I would defend/advocate string theory (which I ultimately don't) but in all fairness I have to say that at least some of this critics is a little bit simple minded and unfair. (Right should be right)

 

This boils down to IMO a simple first level view of the scientific method where science is about generating theories, and we allow experiment to discriminate the theories that fail to predict experiments. A necessary condition for this "strategy" to work, is that all theories are falsifiable. Ie, they have to make specific predictions in specific situations, and matching it against experiment is a basic agree or doesn't agree. Either the theory agrees, in which case it keeps beein subject to test. But what happens if the theory fails? This first level analysis leaves no clue to the logic of revising the theory.

 

These days science is more sophisticated, and some people come up with things wich is ultimately a theory of theories, wherein theories may live and evolve.

 

How do you falsify that? IMO, the scientific view must also be upgraded to be able to cope with the situation.

 

For example, how do you falsify an organism? Well, either it is successful and survives in it's environment, or it dies. If it dies, the adaptive strategy of hte organism was "wrong". This is my view of a more modern concept of science, that applies to the more modern theories that really aren't old style theories, they are frameworks and strategies.

 

I have no idea if the stringers here will recognize this defense, but let it be my opinion in any case.

 

But, this very analysis taken further is also why I don't like string theory, but I suspect this is not classified as "string questions", so I wont comment more on it in this thread.

 

/Fredrik

[BenTheMan]

Apart from possibly giving a handle to describe quantum gravity, are there any other reasons to believe in strings that are not already features of less exotic models (most notably supersymmetry and GUTs). Which ones?

 

Hi Athiest---

 

First, a story. This was a (famous?) short story by someone, who I forget. Either way, the plot basically goes that a man set out to produce a single page of ``Don Quixote'' from scratch---i.e. he never read the book. He studied Miguel de Cervantes' life, and devoted his whole life to understanding ``Don Quixote'' without ever actually reading it. Then, before he dies, he sits down and reproduces exactly the first page of the novel. After he dies, most people think he has wasted his life, except for one critic, who makes the point that what this man has done is so much more beautiful than everyone gives him credit for---he has understood Miguel de Cervantes SO WELL, he has understood the novel SO WELL, that he was able to reproduce a page of it without ever having read it.

 

String theory is like this, in some sense. It doesn't tell us that many NEW things, but it DOES explain things that we already know. For example, things like the number of generations and supersymmetry are properties of the specific compactification we use. Chiral fermions are a given! The number of dimensions is derived. Gauge symmetries are natural, and not put in by hand. All of these things we can put into our low energy theories, but when it comes to string theory, we get these things out.

 

There are some unique features of string theory that aren't present in other constructions, too. But ``unique'' often means ``consequences of specific solutions'' and not ``generic predictions''. So, for example, there could be cosmic strings (not topological defects, but actual strings) in the WMAP data. These things (I think) would seed structure formation, and this is something that is being looked for very closely. You can also have things like large extra dimensions, which would show up as Kaluza Klein towers at the LHC. Also, quite typically, there is exotic matter which you get out of string theory.

 

Of course, none of these things would be PROOF of string thoery, because what you are measuring (at least at LHC) are parameters in an effective lagrangian. And a priori it is impossible to know what fundamental theory your effective lagrangian comes from. If we COULD know these things, then physics would be solved when the higgs mass is known.

 

So to sumarize, I would say that generally there are no new features (below the planck length) of string theory that can't be incorporated into other models in an ad hoc way. One can write down an SO(10) GUT that gives good fermion masses, one can pencil in four dimensions into the calculations, etc etc. But string theory offers a framework to understand why we have all of these ideas.

 

Of course, once we understand vacuum selection (i.e. moduli stabilization) all of this is liable to change

[iNow]

it is often spouted by opponents of string theory that it makes no predictions and there will never be anyway to test it. quite bluntly, is this true?

From reading the other string question thread, it appears that sting theory does make predictions, but the energies involved in trying to perform experiments to test those is currently beyond our engineering ability. Someone please correct me if I'm wrong.

[BenTheMan]

alien---

 

I would add that there's so much we still don't know that it's hard to say. In my reply to Athiest, I said that everything would be different if we knew how to pick a vacuum, or even one consistent vacuum.

 

Once we figure out how to pick a vacuum, THEN we can try to find test. In the meantime, we can use a shot-gun approach---shoot in the general direction and see if we hit anything. This is (more or less) what I'm doing.

[vincent]

This may sound like I would defend/advocate string theory (which I ultimately don't)… I suspect this is not classified as "string questions", so I wont comment more on it in this thread.

 

You shouldn’t have commented at all. I made it quite clear that this thread is for people who are interested in understanding string theory, and you've made it quite clear that you're not one of them. So please don`t further pollute this thread with your OT and unhelpful remarks.

[Severian]

First, a story. This was a (famous?) short story by someone, who I forget. Either way, the plot basically goes that a man set out to produce a single page of ``Don Quixote'' from scratch---i.e. he never read the book. He studied Miguel de Cervantes' life, and devoted his whole life to understanding ``Don Quixote'' without ever actually reading it. Then, before he dies, he sits down and reproduces exactly the first page of the novel. After he dies, most people think he has wasted his life, except for one critic, who makes the point that what this man has done is so much more beautiful than everyone gives him credit for---he has understood Miguel de Cervantes SO WELL, he has understood the novel SO WELL, that he was able to reproduce a page of it without ever having read it.

 

I think I disagree with the analogy. Imagine instead that this guy had employed a near infinite number of monkeys to bang the keys of keyboards randomly. They would produce many many 'first pages', and given long enough, one of them would reproduce the first page of Don Quixote.

 

This is more like string theory. It can predict anything you want, as long as you are prepared to wait long enough.

[YT2095]

isn`t that what happened in the 1`st place though?

 

a good many monkeys and so many millions of years, and hey Presto, that Book exists.

although that does invoke 8 or 9`th Dimension to do exactly the same thing again if I rem correctly.

[vincent]

...string theory...can predict anything you want, as long as you are prepared to wait long enough.

 

Only logically inconsistent theories can predict anything you want, and string theory has passed every one of the numerous self-consistency checks it has ever been subjected to. Given it's complexity and the fact that we still don't know what string theory is, this must be viewed as a highly nontrivial fact.

 

Keep in mind that it took nearly two and a half millenia for the atomic hypothesis to be verified and that there were plenty of skeptics right up until it was. String theory on the other hand has been with us for less than fifty years. Given the fact that it is the only theory we have that appears to include the ingredients necessary for understanding all of the fundamental interactions in an absolutely natural, beautiful and coherent way, you'll understand why very nearly the entire physics community thinks it deserves a bit more time than you seem to think it does.

 

By the way, where did you get this pearl of wisdom? I'm guessing not from studying string theory.

[BenTheMan]

I think I disagree with the analogy. Imagine instead that this guy had employed a near infinite number of monkeys to bang the keys of keyboards randomly. They would produce many many 'first pages', and given long enough, one of them would reproduce the first page of Don Quixote.

 

This is more like string theory. It can predict anything you want, as long as you are prepared to wait long enough.

 

How about only 10^500 monkeys?

 

I disagree that string theory can give you anything you want. I'll give you two examples that I am familiar with.

 

In building SU(5) SUSY GUTs, you generally need higgses in the larger than andjoint reps, for example a 45 to get fermion masse right. But in the heterotic string, one cannot get such large reps out, that I know of. So, for example, start with E8 and break to E6, and then to SO(10) and then to SU(5). The only matter at the string scale is the adjoint of E8 which is the 248. Then you can go to Slansky and find

 

[math]248 \supset 1+ 27 + \overline{27} + 78[/math]

and

[math]27 = 1 + 10 + 16[/math]

[math]78 = 1 + 45 + 16 + 16[/math]

 

None of these SO(10) reps contain the 45 of SU(5). The case when starting from SO(32) is the same, I think. This is all in heterotic strings, where the gauge degrees of freedom are internal bosonic modes.

 

In Type IIA, I'm not sure how one can get reps larger than adjoint of SU(N). The (gauge) degrees of freedom there are strings which start on one brane and end on another, so I don't know how a stack of N branes can ever have more than N^2 - 1 degrees of freedom associated with it.

 

This is one of the main reasons that string theory actually offers an improvement over traditional SUSY GUTs---there you can add arbitrary higgs reps untill you get everything to work out right, in the string side you are limited with what you can add.

 

So while there are SPECIFIC string vacua that have features that we want, there is no single string vacuum (yet) that has all of the features we associate with our universe.

 

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

 

I just talked with a post doc about the argument I gave above about large GUT reps. For matter which is localized at orbifold fixed points, the argument is a bit more subtle, but I am assured that it still doesn't work (something about higher level Kac-Moody algebras).

[Severian]

What is wrong with the 27plet? That is big enough to hold an entire generation in it, including Higgs. All you need to do then is give the Higgs a generation index and only let the third generation gain a vev (from the heavy top).

[BenTheMan]

Nothing's wrong with the 27. I LOVE the 27.

 

But it doesn't contain a 45 of SU(5). So if the real world IS SUSY SU(5), then it is unlikely (at least from what I know) that it comes from string theory---it's hard to get big higgses like that, if not impossible, at least from standard model building.

 

As with everything though (and as you rightly pointed out), it is hard to make any firm statements.

[Severian]

I am not sure what you mean by 'big higgses'. I am thinking specifically of models like http://arxiv.org/abs/hep-ph/0510419 where E6 is broken down to the SM by way of SO(10) and SU(5). Are you saying this is in contradiction with String Theory?

[BenTheMan]

I am not sure what you mean by 'big higgses'. I am thinking specifically of models like http://arxiv.org/abs/hep-ph/0510419 where E6 is broken down to the SM by way of SO(10) and SU(5). Are you saying this is in contradiction with String Theory?

 

No no. E6 is easy to get---there is a good embedding of E6 in E8, as you know, or can look up in Slansky. The E6 model you linke to puts everything in the 27 of E6, and the 27 is easy to get from string theory. (Generation replication happens in a nice way, too!)

 

The problem is that getting SU(5) to work requires a non-minimal higgs sector. This can be seen in fermion mass predictions---the Georgi SU(5) predicts that the electron and tau have the same mass at the electroweak scale, or some similar goofy thing (I can't remember exactly). To fix this, people add new scalars with mass M_GUT in large representations, for example:

 

http://prola.aps.org/abstract/PRD/v26/i1/p312_1

 

These authors look at adding a 45 of SU(5) or a 45 and a 70, in order to get fermion mass predictions to work out at the weak scale. The point is, such large representations are difficult to get out of strings, and the post doc I work with assures me that theres a proof---it's not QUITE as straightforward as I mentioned earlier, about the branching rules.

 

So the correct statement is that large GUT higgs representations are generally not present in stringy constructions, and if we find something like SUSY SU(5) with many new GUT higgses, it probably doesn't come from string theory. The caveat is, of course, that someone will probably find (or has found) a way to do it, using some clever math.

[Severian]

OK - that is fine, but is a little bit semantic. No-one expects SU(5) to be a complete model, but we could have an SU(5) on the way down from E6 or E8. I am certainly willing to believe that string theory doesn't allow everything to be in an SU(5) rep, but I wouldn't say that is a very strong statement.

 

I am actually working on some papers looking at this E6 inspired model at the moment, so I would have been a little concerned if it was inconsistent with string theory.

[BenTheMan]

I am certainly willing to believe that string theory doesn't allow everything to be in an SU(5) rep, but I wouldn't say that is a very strong statement.

 

Fair enough, but it IS a statement:)

 

I am actually working on some papers looking at this E6 inspired model at the moment, so I would have been a little concerned if it was inconsistent with string theory.

 

The paper you linked to looked interesting (by Stephen King). I only skimmed over the text, but it looks like they fit everything into 27's with a higgs 27+b27? What is the spectrum, exactly?

 

Getting it from a heterotic string compactification might be pretty easy, if it hasn't already been done. PM me and I can send you some references, if you're interested.

[Severian]

A sample spectrum is in the figure of http://arxiv.org/abs/0708.3248 (only a conference proceedings, so not much detail yet).

[BenTheMan]

Severian---

 

Two things. I am pretty sure that I can build your model with a heterotic string compactification, and it will be one of the first things I try to do when I finish working on this stupid gauge coupling unification thing I'm doing now. (See below.) The reason it's not so hard is that the 27 of E6 is almost trivial to get out. In general, however, one will have exotic matter, which can generally be given masses near the string scale, decoupling it from the low energy theory. The challenge will be to find just the right conditions so that one gets the right number of 27's, while everything else pairs up into vector-like singlets.

 

This leads me to 2. There are certain things that are VERY difficult to get from string theory, for example, gauge coupling unification in heterotic string models. There is always a problem fitting the low energy data, and one typically has to rely on large threshold corrections, or magic at the GUT/string scale. This is what I am calculating right now, and it is a pain in the ass---i.e. not so easy.

 

In string theory the situation is much more tightly constrained. Since we're deriving everything from (essentially) first principles, the spectrum and such are much more difficult to get right.