About this whole question of when a theory is scientific

[josh1]

The question isn't whether a theory is testable now, it's whether it's testable in principle. For example, there's nothing in our current understanding of string theory that precludes it from being falsifiable. The main problem for string theory is the same problem that continues to torture the entire field, namely, that we simply don't understand the gravitational physics of the vacuum.

For string theory this means that we don't yet understand how to break supersymmetry in a way that reveals the kind of low energy phenomenology it can describe. Similarly, for the theories that go under the rubric of "quantum geometry", it's preventing the discovery of what kind of low energy gravitational behaviour it has so that we have nothing to compare it to either. Thus at present, neither theory produces numbers that we can check agains experiment.

There's another source of confusion here. It has to do with statements like, here's an experiment we can perform to see if spacetime is discrete and hence if something like LQG is correct. The confusion is that a theory doesn't have to describe spacetime in the explicitly discrete terms that LQG does to be consistent with an experimental signature indicating some kind of discretenes. Any theory with a fundamental length will also predict such a result, including string theory. Another example is the resolution of gravitational singularities of one kind or another. We already know that singularities must be resolved somehow and there are at this time a lot's of arguments about how this might be achieved being put forward based in string theory, LQG, and I'm sure other approaches.

Let me just make one final point. In the past string theory and LQG have been described as complementing each other in that LQG shows how a background independent quantum gravity theory can be constructed, and strings shows how the low energy limit might be obtained from a quantum gravity theory. But this is misleading in that we don't know for a certainty that nature has no trick when it comes to gravity that shows that background independence isn't needed. However there can be no doubt about the reality of the low energy world we live. The truth then is that LQG tries to solve a problem that may turn out to have no basis in nature, while string theory solves a problem that every quantum theory of gravity must, and that at this time only it does. In fact, it's when one comes to understand how string theory does this against the background of so many failed attempts, that the inevitability of string theory will begin to really impose itself on you.

 

[hossi]

There's another source of confusion here. It has to do with statements like, here's an experiment we can perform to see if spacetime is discrete and hence if something like LQG is correct. The confusion is that a theory doesn't have to describe spacetime in the explicitly discrete terms that LQG does to be consistent with an experimental signature indicating some kind of discretenes. Any theory with a fundamental length will also predict such a result, including string theory.

Well, then string theory should predict something! At least to me, it's not immedately clear in which sense string theory has a minimal length, or which exact experimental signature that would lead to (as going beyond effective models with a minimal length scale).

-B.

 

[marcus]

...There's another source of confusion here. It has to do with statements like, here's an experiment we can perform to see if spacetime is discrete and hence if something like LQG is correct...

I haven't heard anyone say that, Josh. Maybe you have. Or maybe you are misunderstanding what you hear LQG people say.
About getting empirically falsifiable predictions out of LQG and related theories a good paper to read would be Smolin's.
http://arxiv.org/abs/hep-th/0501091
Falsifiable predictions from semiclassical quantum gravity
As of that paper they were not there yet but you can see the goal they are working towards.

the emphasis is on DISproving LQG. If someone could show that some form of LQG implies that higher energy gamma photons go FASTER (slightly at very high energies, just barely detectable in observable gamma bursts)
and then the GLAST satellite would find that higher energy gamma actually go SLOWER then that would be progress. Because that would FALSIFY that form of LQG.

I gather people are working hard now to get some unambiguous prediction of this sort from some form of LQG before the scheduled flight of GLAST in 2007. Predictions AFTER the fact do not count.
The aim is to survive a real test (or die in the attempt---also a possibility.)

A traditional theorist tries to make theories falsifiable, ideally to propose only falsifiable theories. The more explicit, practical, and unambiguous the test, the better.

So the game is not to prove yer theory right (as in your statement to show space is "discrete"). the game is to set things up so it can be shown wrong---if things turn out that way.

It would be great if someone could extract from some form of stringy theory a prediction of how gamma dispersion might go, and then GLAST could prove that wrong too!

So far, I did not hear any string theorist bet his career and theory on any outcome of LHC (like whether SUSY would be seen there or not) or on any outcome of GLAST or AUGER or anything else. It would be great if someone did make a prediction that could refute stringy constructs! If you see any stringy article like the Smolin LQG one, about falsifiable predictions, please let us know so we can celebrate. I mean it. I'd be happy to see string get on board to some experiment or astronomical observation!
=============

BTW if they do manage to extract a prediction about gammaray dispersion from LQG before 2007. And then if GLAST flies and it does NOT refute the LQG prediction----in other words if LQG survives the test.
then you will not hear me say that this form of Quantum Gravity is correct.

I will not say something is correct merely because it passes one test. What I will say is, it had the balls to undergo a real test, and it survived it. that's all.

 

[FSC729]

String theory is in principle testable but, it is not falsifiable, for whatever the results of any experiement the string theorist can tweak one of the many parameters to make the theory fit the data.

My impression of string theory is that theorists used advanced mathematical techniques to construct an extenstion of QFT and GR with such flexiblity that it acts more like mathematics rather than physics.

This is akin to taking the theory of partial differential equations and calling it physics, yes most classical physical phenomena can be described in terms of PDF's but, not every PDF describes physical phenomena.

John G.

 

[josh1]

String theory is in principle testable but, it is not falsifiable, for whatever the results of any experiement the string theorist can tweak one of the many parameters to make the theory fit the data.

Why are you distinguishing between testifiability and falsifiability? If a theory predicts A and experiment gives B it's wrong.

By parameters you probably mean the moduli which are scalars that parametrize the compactified part of spacetime, different choices of which give different phenomenologies. Until we discover how to stably break supersymmetry in a way that does not produce a large cosmological constant in conflict with observation, the theory will not pick out the right values of these parameters. No one 'tweaks' anything. We simply don't yet know how strings ultimately chooses which vacuum to represent nature and won't know until we understand the gravitational physics of the vacuum e.g., Why is the cosmological constant a small nonvanishing number etc.

My impression of string theory is that theorists used advanced mathematical techniques to construct an extenstion of QFT and GR with such flexiblity that it acts more like mathematics rather than physics.

Impressions? What about facts?

 

[josh1]

Just to be clear, what I intended to convey in part was that there are no known reasons why string theory will not turn out to be falsifiable. The same goes for LQG etc. So our understanding of these sorts of theories isn't yet sufficient for them to be criticized on this basis.

There is one thing I should add though. This is that M-theory is unique, which is another reason to take it seriously. However, the same cannot be said about quantum geometry since it's not a theory but rather a collection of theories different theories. We would therefore need an entire family of theories to be falsifiable, not just one. This is clearly more complicated and also makes this direction of research less plausible.

 

[FSC729]

Don't take my word for it, you can reference Peter Woit's blog:

http://www.math.columbia.edu/~woit/wordpress/?p=360

and

http://www.math.columbia.edu/~woit/wordpress/?p=359

Lee Smolin's Take On the Current State of Fundemental Physics:

http://www.nyas.org/publications/UpdateUnbound.asp?UpdateID=41

John G.

 

[josh1]

Hi FSC729,

What message exactly are we suppose to take away from the links you gave? The assertion in this thread is that the criticism by members of this forum of the sorts of theories discussed here on the basis of whether or not our current understanding of them allows us to produce smoking gun predictions is meaningless. Our understanding of these theories has simply not yet reached the point where we can make categorical statements about this. In fact, there is no evidence whatsoever that these theories can never be falsified. Is there some proof in the links you provided that this not the case? If so, then you really should rush to alert the physics community with this very important news.

 

[FSC729]

Hello Josh1

Nope I don't claim anything new, it's just that Peter Woit and Lee Smolin are leading researchers in the field with a deep understanding of String theory and LQG, and I value their opinions quite highly.

The message I intended to convey is simply that string theory is in crisis and part of it is due to its apparent non-falsifiablity. One need only look at the number of all time HEP citations for post 1999 string theory papers. Not to mention that the current direction of string theory, "Landscape", is quite disturbing; philosophically speaking. Part of this stems from the seemingly non-falsifiablity of some of the theories as Lee Smolin explained in his article. I suppose I should have been more careful with my characterization of string theory.

The evidence you speak of amounts to proving that a theory will never be falsifiable. Remember physics is not like math, we can't prove something with the certainty of proving Fermat's last theorem. I can't prove that string theory will never be falsifiable. A question for you is: can you prove that the lack of categorical statements of falsifiablity is due to our lack of understanding? How do you distinguish this from something inherently incorrect in the theory?

John G

 

[f-h]

Josh1, the whole class of theories has a prediction right there. Make a meassurement of area. If you don't get the spectrum of LQG (quantitatively defined up to the Immirzi parameter) the whole approach dies. It's natural language is geometry since that the natural language of the framework.
I guess it is this you were referring to initially.

It's not clear to me whether there is such a prediction from String Theory. It's natural language is scattering, is there a scattering prediction in String theory that is reasonably unqiue to string theory, quantitative, and shared by all string theories? I guess the tower of higher excitations of string modes might be such a thing, the only free parameter being the string tension. But even then you can push the string tension to the Planck scale and say that the higher modes are nonperturbative and don't show up at scattering I guess...

 

[josh1]

QUOTE=FSC729]I suppose I should have been more careful with my characterization of string theory. [/QUOTE]

This is the part of your post I agree with.

...Peter Woit and Lee Smolin are leading researchers in the field with a deep understanding of String theory and LQG, and I value their opinions quite highly.

Not only is Peter Woit not a leading researcher in string theory, he is not a string theorist period and has never published a single scientific paper on the subject. In fact, he’s not a leading researcher in any field. Hell, he never even made tenure. He’s just a lecturer, not a professor. He’s at columbia so feel free to verify all of this.

Lee Smolin is a very creative researcher with truly terrible physical intuition. He has published a number of string-related papers, but I wouldn’t call him a string theorist, and I wouldn’t be surprised if he felt the same way. Smolin simply likes to take the road less travelled. I guess that’s his nature. I can assure you though he’s made no notable contributions to string theory.

...string theory is in crisis and part of it is due to its apparent non-falsifiablity.

String theory is not in crisis. That only happens when logical inconsistencies appear. One of the most impressive things about string theory is that despite it’s complexity, and in particular the way it allows dual descriptions of the same physical systems, it has passed every test of it’s logical consistency that has ever been made. This is very important because experiment isn’t the only thing that can falsifiy a theory. On the other hand, LQG hasn’t even gotten to the point that it could be tested this way. The current problems string researchers face are not intrinsic to string theory. We simply don’t understand the gravitatonal physics of the vacuum, and until we do neither string theory nor any other quantum theory of gravity is going to be able to make contact with the low energy world in any sort of convincing way.

One need only look at the number of all time HEP citations for post 1999 string theory papers.

Assessing theories on this basis is not a good idea. Anyone who campaigns against string theory or any theory on this basis probably doesn’t understand them. If you look at serious critiques or assessments - even when there are polemics involved - you will not see such numbers being quoted to support their point of view. Surely if someone really wanted to sway you they would give you the best arguments they have, so if they do adopt these tactics, you have to conclude that they simply don’t have better arguments.

...current direction of string theory, "Landscape", is quite disturbing…

As I posted, string theory doesn’t force the landscape on us, just as it doesn’t impose brane-world cosmology on us. I dislike both of them and believe that we will eventually discover how string theory really does deal with the problems of a positive cosmological constant (and the unnaturally large hierarchy between the strong and electro-weak scales in the latter case).

The evidence you speak of amounts to proving that a theory will never be falsifiable. … can you prove that the lack of categorical statements of falsifiablity is due to our lack of understanding? How do you distinguish this from something inherently incorrect in the theory?

Are you feeling okay? I’m not saying whether or not string theory can ever be falsified. I’m saying that we don’t yet know enough about string theory to say unequivocally that strings can never be falsified. We don’t even no what the basic principles of string theory are. There is currently no basis whatsoever for the belief that strings can never advance to the point that it makes unique but incorrect predictions of the mass of observed particles. I explained that the nonuniqueness of predictions based on the tweaking of parameters you’re worried about is simply a reflection of our inadequate understanding of this sort of physics and not endemic to string theory. When we do learn how to break supersymmetry it will give us the correct vacuum with the correct values of the parameters. This is the goal and we know of no reason why this shouldn’t be achievable.

FSC729, I have a question for you. Why if what you're saying is true does stringy research continue to completely dominate the field of high energy theory? Why are courses in string theory being taught now even at the undergraduate level? Why can we not say the same for any other such theories? What you need to do is spend time trying to understand what it is about string theory that despite this challenging period virtually everyone in the field believes that these conceptual difficulties (and really, these difficulties are basically conceptual, not technical) will be overcome.

 

[josh1]

Josh1, the whole class of theories has a prediction right there. Make a meassurement of area. If you don't get the spectrum of LQG (quantitatively defined up to the Immirzi parameter)

I’m not surprised that you view obtaining the functional form of the area operator in LQG to be more important than obtaining the value of the immirzi parameter. Most lay people would since formulae look more impressive than some constant. But physicists know how easy it is to make up interesting but nonetheless unverifiable and hence useless theories. The challenge is to construct theories that make unique predictions, not an infinite number of them. Now, the correct value of the immizi parameter would have to agree with the black hole area-entropy relation. But not only has no one ever been able to show that it does, they haven't even been able to show that it's constrained by the theory to have a unique value. Because of the simplicity of LQG compared to string theory, it’s doubtful that anyone ever will.

…is there a scattering prediction in String theory that is reasonably unqiue to string theory, quantitative, and shared by all string theories?

String theory is an extension of field theory and so should make unique predictions agreeing with the field theoretic one’s if there is a principle that chooses the correct vacuum. The different string theories actually represent different vacua of a single theory called M-theory. What one computes in string theory depends on the choice of vacuum. If one chooses spacetime with vanishing cosmological constant - i.e. an asymptotically flat spacetime - the S-matrix is what we compute. If we choose a background with a negative cosmological constant, there is no S-matrix so we compute something different. According to the AdS/CFT correspondence, we would compute local gauge-invariant correlators on the boundary of the spacetime. Backgrounds with a positive cosmological constant are a problem that the landscape is meant to address.

 

[hossi]

String theory is not in crisis.

Anyone who campaigns against string theory or any theory on this basis probably doesn’t understand them.

Our understanding of these theories has simply not yet reached the point where we can make categorical statements about this.

Whether you believe it or not, string theory IS in a crisis. And it is in a crisis, because the point where categorical statements can be made STILL is not reached. Maybe, it is reached some day, who knows? But the more time passes by, the more effort, money and people it takes, the less promising string theory looks, that's for sure.

To give a crude analogy: Communism never failed. Its defenders will point out that it was never realized in its full version, and that circumstances set by the outside world were against them. In itself, Communism is a beautiful and indeed very promising idea. It just seems to be a bit disattached from (human) nature.

Lee Smolin is a very creative researcher with truly terrible physical intuition.

I would say, everybody who at this point invests time into specifics of string theory has a truly terrrible physical intuition. I don't think you are qualified to judge on Lee's physical intuition and I don't share your opinion.

B.

 

[josh1]

...the point where categorical statements can be made STILL is not reached.

Correct, and making this point is the purpose of this thead. So maybe when you next come across posts to the effect that string theory isn't falsifiable and therefore not scientific you'll correct them by making this very same point, informing them that we don't yet know whether string theory is falsifiable.

 

[f-h]

josh, sorry, but WTF are you talking about? The immirzi parameter is one scaling parameter fixing an entire spectrum.
The spectrum up to scaling is a unique prediction common to all lines of LQG research (not CDT apparently though). You can in principle go and meassure a very small area and see if the distribution of meassuremetn outcomes you get fits the LQG prediction or not.

To my knowledge there is no similar prediction in string theory, at least nobody has pointed one out to me. Even given arbitrarily good experimental capabilities you could not say: Well simply conduct this specific experiment, if it doesn't show this and that specific result predicted by String Theory, String Theory is clearly wrong!

Ah I see your response to Hossi now that you agree that ST doesn't have such a prediction at this time. It seems then there is merely a confusion of words, you assume there is some wonderfull unique theory that unfortunately nobody understands and which is String Theory, since nobody understands it to any degree there is no point in saying it's this or that, it might be anything!
But if you call the body of work accumulated so far String Theory, instead of some hypothetical and conjectured eventually to be realized platonian idea, then it isn't falsifiable.

 

[FSC729]

Let's all just calm down, this is getting personal. Taking personal shots at Peter Woit and Lee Smolin won't help your arguement.

I’m saying that we don’t yet know enough about string theory to say unequivocally that strings can never be falsified.

You're making the implict logical assumption that given enough understanding string theory will be able to make falsifiable predictions and that one day we will gain this understanding, within the context of string theory. How do you know that this understanding won't lead to a new theory where strings are merely approximations. Or that this understanding will lead us to abandon string theory altogether? Both of your logical assumptions are unprovable.

We don’t even no what the basic principles of string theory are.

So it appears that both the proponents of string theory and the critics of string theory don't fully understand string theory. Not just schmucks without tenure.

There is currently no basis whatsoever for the belief that strings can never advance to the point that it makes unique but incorrect predictions of the mass of observed particles. I explained that the nonuniqueness of predictions based on the tweaking of parameters you’re worried about is simply a reflection of our inadequate understanding of this sort of physics and not endemic to string theory. When we do learn how to break supersymmetry it will give us the correct vacuum with the correct values of the parameters. This is the goal and we know of no reason why this shouldn’t be achievable.

Well I hope one day we can obtain this understanding, predicting the future is risky, all I'm saying is don't put all your eggs in one basket.

John G.

 

[hossi]

So maybe when you next come across posts to the effect that string theory isn't falsifiable and therefore not scientific you'll correct them by making this very same point, informing them that we don't yet know whether string theory is falsifiable.

Josh, I have no idea what your problem is. I don't mind persuing research on string theory in general. I just think that the effort put into string theory and other theories is more than slighly off balance

String theory isn't falsifiable yet. I wouldn't say, its not scientific, but so far, its got nothing to do with reality. Right now, its pure mathematics. I will surely not make any comments to anybody that might only remotly indicate that more work has to be put into string theory! In my opinion, string theory does not qualify as a theory of everything, and I would not recommend my students to join the field. I find it likely though that string theory will turn out to describe some parts of a fundamental theory. Therefore, I appreciate the recent efforts to find some string phenomenology.

B.

PS: regarding the question of falsifiability, what would you say if it turns out there are more than 11 dimensions...?

http://xxx.lanl.gov/abs/hep-ph/0503178
Black holes in many dimensions at the LHC: testing critical string theory
Authors: JoAnne L. Hewett, Ben Lillie, Thomas G. Rizzo

 

[josh1]

The immirzi parameter is one scaling parameter fixing an entire spectrum.
The spectrum up to scaling is a unique prediction common to all lines of LQG research (not CDT apparently though). You can in principle go and meassure a very small area and see if the distribution of meassuremetn outcomes you get fits the LQG prediction or not.

We need to know the value of the immirzi parameter to compute numbers. Without it, we have some pretty mathematics and that’s it. For example, without the value of immirzi, we can’t predict the spectrum of black hole emissions. This is clearly not good since LQG is supposed to be a quantum theory of gravity and the only definite number we have in relation to quantum gravity is this number. String theory on the other hand does in fact give the correct constant of proportionality 1/4 between ordinary Schwarzschild black hole entropy and mass.

Ah I see your response to Hossi now that you agree that ST doesn't have such a prediction at this time

Up until now I’ve discussed why string theory CANNOT yet make predictions about phemonmena at energy scales currently accessible to us. I also said - and this is the point - that there’s no known reason why it shouldn’t ever be able to and that stating categorically that this is not the case is unjustified at present.

However, as I just stated, string theory does give us the correct constant of proportionality in the black hole area-entropy relation. It also predicts gravity in that there is no way to formulate a quantum theory of extended objects like strings without including gravity. In fact, the requirement that the Polyakov action for strings be conformally invariant produces the vacuum einstein equations.

 

[josh1]

I don't mind persuing research on string theory in general. I just think that the effort put into string theory and other theories is more than slighly off balance

Off balance? What does that mean? There aren't a lot of Einsteins around. Most scientists work where the pasture is green and on what catches their fancy. They're only human. But you must know that. So what is it you propose we do? Should we have quotas?

String theory isn't falsifiable yet. I wouldn't say, its not scientific

This is all I was saying. Yet look at all the static I got. Where do you suppose that came from?

 

[hossi]

Yet look at all the static I got. Where do you suppose that came from?

I guess mainly from 'the inevitability of string theory' that should impose on us. As far as I am concerned, and some of my colleagues have made the same experience, string theorists have been very pampered for quite a while, supported by an unreflected optimism and a serious lack of criticism. Now that criticism gets louder, its in parts unreflected in the opposite direction. Neither will help to make progress. I strongly believe in live and let live. E.g. the LQG people could probably learn something about branes from the stringy people, whereas the stringy people might want to look into the space-time structure more.

Off balance? What does that mean? There aren't a lot of Einsteins around. Most scientists work where the pasture is green and on what catches their fancy. They're only human. But you must know that.

Yeah, indeed, thanks for reminding me that I am human, I should get something for dinner

But its people go where money go and money goes where people go. It just seems to me there is too much money in string theory. And people that spent their time making calculations for a string theory adviser will never find out whether they have the potential for being the next Einstein. What I am asking for is more diversity. More courage. And more confidence in young researchers.

So what is it you propose we do? Should we have quotas?

Actually, kind of a quota sounds indeed like a good idea! We all know that funding is distributed on vague and biased reasons. It would be good to have grants distributed in a reasonable way. Meaning, not in the first place based on the number of people or publications on the field. That certainly is ONE point, but there should be others as well.

What would you think is a reasonable indicator?


B.

 

[josh1]

Actually, kind of a quota sounds indeed like a good idea! We all know that funding is distributed on vague and biased reasons. It would be good to have grants distributed in a reasonable way. Meaning, not in the first place based on the number of people or publications on the field. That certainly is ONE point, but there should be others as well.

What would you think is a reasonable indicator?


B.

The most important new ideas in chess are generated by the top players and the best way for players of master strength and above to judge how promising new lines are is to simply study what these top playes are experimenting with. Chess has shown itself to be too complicated to improve on this.

Well, physics is just a bit more difficult than chess so I'm pretty sure if we started down the road you're talking about it wouldn't be long before we drew the same conclusion. We really have no good alternative to allowing the popularity of ideas to guide how we assess and manage research. I'm sure there are probably good examples in the history of science that support this view and that we'd simply be learning a lesson that was first taught long ago.

 

[f-h]

*We need to know the value of the Immirzi parameter to compute numbers.*

That is obviously absurdly wrong. Again the Immirzi parameter is a simple scaling parameter, it doesn't change the shape of the spectrum, therefore there is in the area operator infinitely much meassurable information predicted uniquelw by the theory, for example the all ratios of eigenvalues are uniquely predicted.

Given arbitrarily precise area meassurements we can immidiately, without a line of theoretical work, determine whether the LQG prediction is correct or not.

 

[josh1]

*We need to know the value of the Immirzi parameter to compute numbers.* … the Immirzi parameter is a simple scaling parameter, it doesn't change the shape of the spectrum, therefore there is in the area operator infinitely much meassurable information predicted uniquelw by the theory, for example the all ratios of eigenvalues are uniquely predicted.

Yes, immirzi is dimensionless, and will not affect features that depend only on the shape of the area spectrum. You know, the origin of the form of the operators in LQG isn’t some kind of miracle. They’re a consequence of building a theory of gravity based on the quantum mechanics of angular momentum. Even though researchers in LQG describe this result as an unexpected miracle, it was obvious from the outset that they would obtain a discrete theory of spacetime because they understood the kinematics of angular momentum in quantum mechanics. In string theory on the other hand, the way in which General Relativity is forced on you was completely unexpected: It’s absolutely amazing that the vacuum einstein equations appear by requiring conformal invariance.

Now, whatever their origin and beauty, the functional form of the equations of a theory are never enough. We need the value of the constants as well since they are what characterize the scale at which the effects characteristic of a theory become significant. A good example of how beautiful theories can turn out to be not so great because of this is euclidean quantum gravity. That theory predicts a new effect, wormholes, that become important at energies much lower than the Planck scale. But there is no such scale, coming from either euclidean quantum gravity or anywhere else. This is what has stopped people from extending euclidean quantum gravity, though the theory does seem to capture important issues in quantum cosmology.

The situation is analogous for LQG in that it "predicts" black hole evaporation, but doesn't specify the scale of the effect. LQG needs to produce the correct constant of proportionality between the entropy and area of a black hole. But let’s talk in experimental terms.

Suppose we analyzed the hawking emissions of a black hole and from it determined the black hole's current entropy S. Then imagine along the lines that you're talking about we measured the area A of the event horizon. We could then check whether S and A are related by the Hawking-Beckenstein relation S = (1/4)A. Now, arguments have been presented which demonstrate that LQG produces this formula only up to a dimensionless constant of proportionality - the immirzi parameter. Thus we cannot use LQG to verify S = (1/4)A. If we could, it would be the very first consistency check that’s ever been performed in LQG in that we would see whether or not we get the correct factor of 1/4.

What we need is an argument coming from within LQG that the value of the immirzi parameter is the one that gives us the factor of 1/4. It's accurate to say that if someone resolved this problem, stringy people would be on LQG like white on rice. But no one has been able to produce such an argument. There have even been attempts to determine immirizi by forgetting that LQG is suppose to be a theory of first principles that describes gravity on the most fundamental level and go outside of it, but these failed too. Actually, a few years ago, excitement ran through the string community about just such a result. What happened was that someone recognized that a dimensionless number that came up in analyzing ringing modes of a Schwarzschild black hole might detemine the value of immirzi. Unfortunately, for this to be correct would require that the result holds for all black holes. It turned out that this wasn’t the case and that it was just a conincidence. The excitement among stringy people then disappeared once again.

From the perspective of the point of this thread, this doesn’t rule out LQG because it doesn’t prove that LQG can never produce a unique value for immirzi that can be checked in the manner discussed above. But it does put in much greater doubt then this kind of problem would in string theory simply because there does not appear to be anything deeper about LQG that would resolve this, while we know for a fact that we've only seen the tip of the iceberg with strings.

 

[selfAdjoint]

The most important new ideas in chess are generated by the top players and the best way for players of master strength and above to judge how promising new lines are is to simply study what these top playes are experimenting with. Chess has shown itself to be too complicated to improve on this.

Well, physics is just a bit more difficult than chess so I'm pretty sure if we started down the road you're talking about it wouldn't be long before we drew the same conclusion. We really have no good alternative to allowing the popularity of ideas to guide how we assess and manage research. I'm sure there are probably good examples in the history of science that support this view and that we'd simply be learning a lesson that was first taught long ago.

In chess there's an objective way to tell who's a top player. Competition. In physics it's just one person's opinion against another. Witten got the Field Prize from the mathematicians but will very probably never get a Nobel prize in physics; why are physicists studying him?

 

[josh1]

In chess there's an objective way to tell who's a top player. Competition.

Correct. Now we can go ahead and use their games to differentiate between ideas that have legs and those that don't, which is what I posted.

In physics it's just one person's opinion against another.

No it's not. For example, strings are correct. Now, since you view this as simply my personal opinion, even though I'm stating it as if it we're fact, I'm sure you'll not be bothered by my remark in the slightest because if everything is just a matter of opinion there's no point in worrying about the beliefs of people who disagree with yours. But I doubt you feel this way because you know that opinions in the exact sciences do matter since they're either right or wrong and choosing between them has consequences.

 

[f-h]

josh1, so GR can't predict anything because it doesn't predict the value of G? QED is meaningless because it doesn't predict the finestructure constant?

I say we can predict B you answer but you can't predict A, and in my opinion A is far more important and the prediction of B is trivial anyways!

Fair enough, that doesn't change the fact that there are unique numerical parameterindependent and original predictions from LQG, like the form of the area spectrum.
That doesn't tell you how to meassure it, or at what scale to expect it but again, given an arbitrarily fine area meassurement apparatus we could simply meassure the immirzi parameter, then the entire rest of the spectrum is a prediction. I never talked about Blackholes, you did. IMO Blackholes play a different role to this hypothetical meassurability issues.

Of course it's all just SU(2) representation theory in the end (the main sequence anyways, that's not actually true for the full spectrum). But that you can come to that end, that the holonomy of paths can form a rigorously defined and unique Hilbertspace for diffeomorphism invariant theories is everything but trivial.

 

[hossi]

We really have no good alternative to allowing the popularity of ideas to guide how we assess and manage research. I'm sure there are probably good examples in the history of science that support this view and that we'd simply be learning a lesson that was first taught long ago.

Josh, why don't we have a basis democracy where every decision is based on the current popularity? Because it would be a complete disaster. Everybody would in the first line vote as for his/hers immediate advantage.

Instead, we have some rules that (ideally) everybody in the society has agreed on. Based on which some (hopefully) qualified people make decisions that they think are in accordance with the rules, and will on the long term lead to progress (do be defined).

I believe that was not an issue in science until recently because

a) due to the enormous growth in research it has become increasingly difficult, if not impossible, for individuals to have an overview on the whole field and

b) because science has become much closer attached to the public. Though I find that in principle a good thing, it gives an increasing importance to questions like: which theory looks better in catchy movies, who has the cooler leaders, who looks better on TV or leaves a smarter impression in interviews, etc.

The current popularity is definitely not a good basis to manage research. Indeed, I would think some of most important questions are probably also the most unpopular ones. How do you ever solve some fundamental problems when nobody dares to attack them because they might fail with it? How do you ever correct a direction of research, if no one dares to criticize because they depend on the support?

Well, people go where money goes...



B.

 

[josh1]

Josh, why don't we have a basis democracy where every decision is based on the current popularity? Because it would be a complete disaster. Everybody would in the first line vote as for his/hers immediate advantage.

By not having anything better then popularity to assess and manage research, I didn't mean we should formally adopt this as a principle and impose it on everyone as if we lived in the Soviet Union. I meant simply that people be allowed to continue making decisions on this basis (though not only on this basis, we do live in the real world), whether they're doing so for entire departments or just for themselves.

 

[josh1]

josh1, so GR can't predict anything because it doesn't predict the value of G?

Like all fundamental constants of nature, G's value must be determined experimentally. The immirzi parameter on the other hand represents a kind of “gauge” freedom in the theory. It is just a dimensionless number and there's nothing to measure. The coefficient in the black hole area-entropy relation must also be derived from theory, and was in Hawking’s historic paper.

…there are unique numerical parameter independent and original predictions from LQG, like the form of the area spectrum.

As I said, LQG was constructed to have this property: LQG is not a prediction, it’s a theory.

 

[f-h]

*It is just a dimensionless number and there's nothing to measure.*

Like the fine structure constant? Or rather, the CKM angle in QCD?

 

[josh1]

*It is just a dimensionless number and there's nothing to measure.*

Like the fine structure constant? Or rather, the CKM angle in QCD?

The fine structure constant can't be described as simply some dimensionless number. It characterizes the strength of the electromagnetic interaction and isn't measured but calculated in terms of the fundamental physical constants of QED. CKM doesn't help you either.