Proving String Theory: Insights and Possibilities from New Scientist Articles"

[kurt.physics]

Hello--

Does anyone know if there is any tests to prove string theory, or a part of string theory

I have seen many articles in New Scientist about possible ways to prove string theory but i don't have access to them because I am not a subscriber

Has anyone seen them, does it seam genuine

 

[arivero]

Well, there was not a set of specific predictions, but a roadmap along the GUT mountain range. You can see it more or less here:
http://www.slac.stanford.edu/spires/find/hep/www?eprint=hep-ph/0207124
So for stringers a measurement of the success is to keep themselves in this road.

 

[BenTheMan]

kurt---

The simple answer is probably not. String theory can only be proved, not disproved. There is a difference :) For example, one could see black holes at the LHC, and this would be a sure sign that we have large extra dimensions, which fit nicely into string theory. However, the non-observation of black holes does not rule string theory out.

 

[Haelfix]

I've asked string theorists before about this, but what about the non observation of a Higgs at any scale (absent any weirdness like random neutral scalars that prevent detection). It seems like a generic prediction of most backgrounds and goes beyond any effective field theory considerations like unitarity bounds.

 

[Thomas Larsson]

AFAUI, the observation of nothing new, not even an SM Higgs, at the LHC would not only disprove string theory, but field theory itself. It would make WW scattering violate unitarity bounds. It does not seem very likely to me, but would be interesting if it were true. Time will tell.

 

[arivero]

AFAUI, the observation of nothing new, not even an SM Higgs, at the LHC would not only disprove string theory, but field theory itself. It would make WW scattering violate unitarity bounds. It does not seem very likely to me, but would be interesting if it were true. Time will tell.

Well, there was some recent review, by Chris Quigg I think, with a chapter about how to scape this problem, depending on luminosity.

 

[BenTheMan]

AFAUI, the observation of nothing new, not even an SM Higgs, at the LHC would not only disprove string theory, but field theory itself. It would make WW scattering violate unitarity bounds. It does not seem very likely to me, but would be interesting if it were true. Time will tell.

This is a pretty weak assertion :) ``If physics is non-unitary, then string theory is wrong.'' Fair enough. String theory is definitely unitary.

Haelfix---

I think that that is probably a true statement, with the caveat that the higgsing isn't accomplished by something like technicolor. I have not seen any attempts at getting technicolor out of string theory, and the post doc I work with won't even CONSIDER looking at anything less than N=1 SUSY. I think that it should be possible, though. I just haven't worked through the details.

There are other things which string theory doesn't deal well with---large representations of gauge groups are not very natural in string theory. So, for example, standard non-SUSY SU(5) can be fixed if you add a GUT higgs in the 40 of SU(5). Such large representations are not easy to get out of string theory as we currently understand it---note, I didn't say impossible, but very difficult. And this is in the corner of string theory that I can claim some understanding of, specifically the weakly coupled heterotic end. (But also the Type IIA brane constructions have a similar limitation.)

 

[AlphaNumeric2]

AFAUI, the observation of nothing new, not even an SM Higgs, at the LHC would not only disprove string theory, but field theory itself. It would make WW scattering violate unitarity bounds. It does not seem very likely to me, but would be interesting if it were true. Time will tell.

Or would it result in a huge community effort to rejig our entire understanding of the Standard Model so that a new field theory model is constructed which works better than the SM in the scenario you outline. I'm sure the vast majority of field theorists would prefer to say "The SM was fundamentally flawed" than "Field theory itself is fundamentally flawed", if it came down to a crunch like that.

Non-field theory based ideas would certainly be interesting though.

 

[BenTheMan]

Yeah, interesting indeed. I don't know how one would go about developing a model of physics that is non-unitary. I'm probably just not creative enough, though.

 

[Thomas Larsson]

This is a pretty weak assertion :) ``If physics is non-unitary, then string theory is wrong.'' Fair enough. String theory is definitely unitary.

I didn't say that. If the description of WW scattering within QFT and string theory is non-unitary, then there must be something wrong with these theories. That is a different thing.

 

[javierR]

Hello--

Does anyone know if there is any tests to prove string theory, or a part of string theory

I have seen many articles in New Scientist about possible ways to prove string theory but i don't have access to them because I am not a subscriber

Has anyone seen them, does it seam genuine

So, to summarize: There aren't any known, genuine tests that would disprove string theory. This is a central criticism of the framework (among more technical ones). The claimed tests to date would only confirm that string theory is still consistent with observations; but this is no surprise since it is a large framework, like quantum field theory. We expect that a quantum field theory will also be able to describe observations in the LHC or at Fermilab or Brookhaven, etc. I'll mention one such claim of testing string theory: Using the AdS/CFT correspondence conjecture, people have used string calculations to quantitatively determine features of a conformal field theory that they argue models the quark-gluon plasma observed in Brookhaven experiments. However, this is a test of string theory as a useful calculational tool not as a fundamental theory of stringy objects (but first one must check whether the conformal field theory used in the model *does* describe QCD in its quark-gluon plasma phase well enough).

To address another point:
The observation of small black holes in an experiment like that at Brookhaven would not be a signal of extra dimensions. There are tests you can do to look for extra dimensions, but if you don't see them, you can keep claiming they're too small to see. An observation of extra dimension(s) would not necessarily make string theory more compelling since there are field theories that do require more dimensions (e.g. higher derivative theories).

 

[BenTheMan]

I didn't say that. If the description of WW scattering within QFT and string theory is non-unitary, then there must be something wrong with these theories. That is a different thing.

I'm confused. QFT is based on being able to write down a unitary S matrix. If one cannot do that, then you have to reformulate/get rid of QFT. This is why there is an LHC ``no lose'' theorem---the unitarity must be saved, otherwise we find out we don't know anything about the universe---QFT is wrong, quantum mechanics is wrong, and we're set back 100 years or so.

So if WW scattering is TRULY non-unitary, then QFT is wrong, and it was only a coincidence that particle physics below a TeV was described by a QFT.

You can replace ``string theory'' for ``QFT'' in the above.

 

[BenTheMan]

To address another point:
The observation of small black holes in an experiment like that at Brookhaven would not be a signal of extra dimensions.

Can you find some other way to lower the Planck scale?

An observation of extra dimension(s) would not necessarily make string theory more compelling since there are field theories that do require more dimensions (e.g. higher derivative theories).

I disagree---while one can formulate field theories in higher dimensions, it was only after string theory that these ideas were fully investigated. In general, these models are not renormalizable, which means that they have no satisfactory UV completion. This means that they must be imbedded into some deeper framework. And because these models are natural in the context of string theory, it certainly makes that UV completion more compelling. Other approaches to QG have made quite a big deal about living in four space-time dimensions (even though most put this number in by hand), and observation of a large extra dimension at LHC would certainly not be good news for them, I think.

 

[Fra]

Yeah, interesting indeed. I don't know how one would go about developing a model of physics that is non-unitary.

I personally think non-unitary modelling is necessary for some quite basic reasons that I consider to be founded in philosophical consideration of the measurement problem. I've yet to see another satisfactory way around it.

My vision is that physics will still in my world still be "close to unitary" under most circumstances, which also explains why the unitary models work fine - at least until found incomplete :) and a unitary model can always be found to a given accuracy, if the change of the emergent probability space can be limited.

The main problem seem to be to find the principle that warrants stability, when allowing fundamental non-unitary behaviour. I think much of what we consider fixed stable references, like space time dimensions and whatever... are rather a dynamic steady state that is selected for stability.

I think this selection principle that yields the observed stability in a world with seemingly lack of prior bounds or "hard physical principles" on what is allowed is one of the keys.

/Fredrik

 

[josh1]

I don't know how one would go about developing a model of physics that is non-unitary.

For the statement that a theory is nonunitary to be meaningful requires that the notion of unitarity is defined in the theory. So we`re certainly going to be dealing with some sort of quantum theory. Now, conservation of probability is not on the same footing as other conservation laws. The latter can in many cases be violated without threatening the consistency of the theory. But I think that the idea that either the probability of measuring some observable to be of some given value can be greater than 1 or that the probability of measuring an observable to have any value at all can be less than 1 is fundamentally nonsensical.

 

[BenTheMan]

The latter can in many cases be violated without threatening the consistency of the theory.

Hmmm. Yeah, ok. But then you lose the ability to make predictions, right? I don't know---this is pretty confusing to me :)

 

[Fra]

But I think that the idea that either the probability of measuring some observable to be of some given value can be greater than 1 or that the probability of measuring an observable to have any value at all can be less than 1 is fundamentally nonsensical.

From the axioms of probability, a probability more than 1 is clearly baloney.

But that is IMO not the real problem. Reality isn't just about arbitrarily chosen axioms. Noone observed these axioms in nature!

If one tries to attach the elements of probability theory, probabilities and sample spaces to reality, in terms of observations and measurements, it seems the entire foundational problem that is usually "axiomatized away" is pretty shaky.

For example, if you expect one particle as per your estimated probability space, but this is later shown wrong, then the conditional probability can apparently still be larger than one. Are we seeing two indistinguishable particles in different places, or is it the same particle that seem to be "more than" all over at the same time? This also IMO, raises the issues of what reality is, and what a "particle" really is. Some may think tha the many particle interpretation is the obviously correct one, but I don't think it's that easy.

I think the reason for non-unitary thinking (ok - at least if I speak for myself), is that one questions to what extent the axioms of probabiltiy theory really make sense for a fundamental model of reality. In a theory the elements of the theory must have a sufficiently good mapping to reality, or it does not make sense to me at least. And to be specific, I am happy if the mapping is good enough to be indistinguishable from a supposedly "better" mapping.

The normal probability thinking + frequentists interpretation is not near satisfactory IMO.

/Fredrik

 

[josh1]

Hi Fredrik,

a probability more than 1 is clearly baloney.

This is the part of your post I understood.

 

[Fra]

Hi Fredrik,

a probability more than 1 is clearly baloney.

This is the part of your post I understood.

I guess what I tried to say, that _probability_ as defined in probability theory as we know it, is by definition always normalised to unity. This is trivial and not something to argue on.

But, if we can not properly measure such a probability with certainty, like I think we can't. Then what we effectively attache the status of probability in physical thinking, could in fact be just an estimated probability. And this uncertainty in the probability itself, means that the "expected probability" can violate conservation laws. But of course, then it really is not a proper probability in the first place - it's just an "estimated probability" in an estimated "probability space".

The estimate is ideally induced from past experience and current information, which is admittedly uncertain an incomplete. So one would expect that while perhaps the induced conclusion also has similar uncertainty, and that "conservation laws" are merely emergent expected conservations.

IMO, our experience induces an emergent, effective probability space, on which we do an effective probability - one question is, exactly how does this induction look like? And how dose the effective probability on the emergent probabilit space, couple to the lower level of comlpexity that evolves the probability space itself?

/Fredrik