Tuesday, August 23, 2005 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Book that is Not Even Wrong

Let's start with my first public review of this book.



Bitter emotions and obsolete understanding of high-energy physics (1 star)

Peter Woit is the owner of a well-known blog that provides high-energy theoretical physics with the same service as William Dembski's ID blog offers to evolutionary biology: it is designed to misinterpret and obscure virtually every event in physics and transform it into poison - and to invent his own fantasies to hurt science. This makes Woit's blog highly popular among the crackpots, for example some of the reviewers of this book. The book is not identical to the author's blog but it is not too different either.

Parts of this book are fun to read, although they will be too difficult for outsiders. But the text is definitely not a trustworthy source of knowledge about physics. The book can basically be divided into two parts. The first part of the book describes physics from the early 20th century to the 1970s or so. This part covers some standard material as well as some points that have not yet appeared in the popular literature. The early chapters also honestly explain that the author has not done any important work in high-energy physics himself and that he has been isolated from research (and researchers) for the last 20 years. Because of these reasons, I originally rated the book by two stars.




As the focus of the presentation shifts to modern physics since the 1970s or so, an expert recognizes that the author misunderstands some very elementary questions.

The book contains a lot of very embarrassing errors. Let me mention a few examples. Woit originally wrote that the center-of-mass energy of the LHC beams would be 14 GeV, instead of 14 TeV: this error has been corrected after long debates in which he didn't want to admit any flaws. He incorrectly argues that the neutrinos with electroweak energies interact very weakly. He thinks that higher-dimensional rotations are associated with one-dimensional "axes". He misunderstands how SU(2) can be embedded to SO(4). In his description of the history of supersymmetry, he forgets Pierre Ramond. He writes that the supersymmetric vacua predict a higher vacuum energy than the non-supersymmetric ones.

Also, Woit seems to misunderstand that all of our knowledge of theories such as QED comes from perturbative expansions when he attacks the perturbative method as such. He also misunderstands what "background independence" means. At one point, the author also claims that the primary evidence supporting scientific theories is an authority (Edward Witten in his case). Even more seriously, he builds his case upon e-mail messages from undetermined sources that supported Woit's viewpoint. Most of these e-mails were obviously written by cranks.

Authorities play an important role and the author quotes many outsiders in high-energy physics who have criticized string theory. But he never mentions names like Weinberg, Gell-Mann, Hawking, Randall, Arkani-Hamed - famous and active physicists who are not string theorists but who believe that it is the right direction. Books by Brian Greene, Lisa Randall, and others were much more balanced in this respect. The book is a gigantic spin zone.

Woit conjectures the existence of singularities in some integrals that appear in string theory and that are known to be non-singular. Woit does not distinguish a family of theories from one theory with a massless scalar field (a modulus). He does not mention Andrew Strominger and Cumrun Vafa when the black hole entropy is discussed. Woit incorrectly believes that the "beauty" of a theory is the same thing as an experimental verification.

The author repeats poisoned remarks about string theory too many times. The second part of the book could be reduced by 60 percent or so. Moreover, most of the statements in the second part of the book are supported by no technical arguments, neither in the book nor in scientific literature. The problematic statement that string theory makes no prediction is repeated hundreds of times, and in many particular contexts, such a statement becomes not only boring but also patently false. The author is not aware (or denies) the actual mechanisms that are considered to be solutions of various puzzles - for example the doublet-triplet splitting problem.

The book is also full of inconsistencies. In one chapter, he argues that the alternatives to string theory in the field of quantum gravity should be supported. In the following chapter, he argues that they should be suppressed - the work of the Bogdanoff brothers is one of his examples. Woit's knowledge of the history of the subjects he discusses is extremely superficial, too. For example, Leonard Susskind is painted as the discoverer of the large number of vacua in string theory. Quite obviously, Peter Woit has no idea about the "discretuum" described by Bousso and Polchinski and many other concepts that have been discussed for years.

Peter Woit also offers a highly obsolete view on many concepts in theoretical physics such as the gauge symmetry; he is obsessed with the old-fashioned idea that all of physics follows from a gauge symmetry principle. He thinks that the gauge symmetry is uniquely determined by physics because he is apparently unaware of dualities and all other phenomena discovered in the last 20 years that show that his preconceptions are wrong and that gauge symmetries are only associated with a particular description of physics that does not have to be unique.

The book is called "Not Even Wrong" but the readers should know that most of the book is wrong after all. I can only recommend the book to the people who dislike theoretical physics - or at least theoretical physics of the last 20 years - and who want their opinion to be confirmed by a semi-serious source. The readers who want to learn what physics is all about may want to avoid the book because it could make them very confused. As far as modern physics goes, the author is a layman. The topics he raises have nothing to do with the actual discussions that take place among the scientists.

... [End of review] ...

A seventeen-page-long review of the book is here. An analysis of typical reviewers is here.

Comment by Lenny Susskind, one of the most original scientists of his generation:

  • You’re talking probably about some of the books and blogs that have come out in very very big criticism of [string theory]. Well, I think one would have to say that some of it is due to a certain kind of grumpiness of people who... um...
  • Well, for example, there’s one fellow who failed as a physicist, never made it as a physicist, became a computer programmer, has been angry all of his life that he never became a physicist and that physicists ignore him, so he’s now taking out his revenge by writing diatribes and polemics against string theory.
Let's continue with the old text of this blog article:

McCutcheon's book is completely wrong - and in a few days, we will officially announce the winner of the $13.08 grand prize. But there are books that are Not Even Wrong. One of them has been written by Peter Woit - and I am apparently the first one who found its page at amazon.com. He explains that the whole book is full of a bitter criticism from a person who does not know what high-energy physicists are talking about but who wants to criticize them nevertheless - which is how Peter Woit apparently envisions the ideal combination for a science writer. ;-) Using this strategy, he concludes, among other things, that supersymmetry and string theory "have failed conclusively".

If the blog is a realistic guide how the book looks like, then it is roughly a 2-star book. But of course I would have to see an actual copy of the book to give you a more accurate rating. ;-) I have nothing against the publication of silly books - except that the internet may be enough and many trees could be saved. As you can see, my opinions about the book are exactly opposite that those of Sean Carroll: I believe that Woit's criticism of the Landscape is a legitimate one (which is the real reason why he did not get 1 star from me), but his dislike for string theory as a mathematical structure is grounded in an uninformed and unscientific judgement. ;-)

Peter Woit probably can't write a scientific book on current high-energy physics, not even a popular one. The publisher apparently realizes it; this may be the reason why the book has been included in the category: Social Science - Popular Culture.

Peter also argues that the string theorists among his referees - who did not like the book, with one exception - did not have too many explicit arguments against the book and did not find too many undisputable errors. One should not believe everything that Peter Woit says; but in this case, I tend to believe him. It requires a certain amoung of training to learn all these silly pseudo-arguments that most critics who are unfamiliar with the subject are typically using. In many cases, it is incredibly difficult for some of the leaders of our field to comprehend that some critics of string theory really believe XY - because XY is often dumb beyond imagination.

The first educated referee - apparently a physics bigshot; my guess is Andy Strominger himself - mentioned that arguing against string theory is like arguing against the evolution. In my opinion, that's a bit exaggerated - but it is not too exaggerated! ;-) In both cases, the main reason to be convinced about the validity of the theory is that it is the only known (and conceivable) scientific theory that is compatible with some of the most fundamental observations, such as the existence of diverse species, their related characteristics (and DNA codes); or the existence of gravity as well as other forces described by quantum field theory. In both cases, a specific enough alternative that someone proposes is more or less guaranteed to be silly. But once again, yes, I do think that at this moment, the evidence in favor of evolution is stronger than the evidence supporting string theory.

Add to del.icio.us Digg this Add to reddit

snail feedback (26) :


reader Mesothelioma said...

I like your blog. Very nice! I should probably bookmark it for later.

I just got done putting the finishing touches to a new website about: Mesothelioma. It's got some
great information about Mesothelioma.

You should check it out sometime when you get time. :-)


reader Stephen Paul King said...

So Lubos, who, exactly, is a suitable critic for superstring theory? If we students and laymen can't point out that it fails to make any predictions that can be tested, what are we to think? I think that you are being far to defensive in your critisism of Woit's book.


reader Lumo said...

Dear Stephen,

everyone can have any opinion she or he wants. But it is not true that all opinions are created equal.

A student can definitely point out that he thinks that there are no testable predictions of a theory - and you are welcome to "point it out" (is not it a bit arrogant word?) on my blog. Also, he can "point out" that general relativity is essentially renormalizable, Peter Woit is right, supersymmetry is equivalent to Intelligent Design, SU(2) should be embedded in an off-diagonal way, and Chern-Simons theory should be considered as a replacement for the Standard Model.

(This is just a very tiny percentage of the stupidities that Peter has generated throughout the years.)

A student can do it, indeed. However, it's a job for her (or his) professor to fail her (or him) during the first or second quantum field theory exam.

Also, I believe that a student who thinks that the sentence "there are no testable predictions of string theory" reflects the important questions that an active theoretical physicist should ask in 2005 is probably a student at the level of the readers of Peter Woit's blog. And that's far too little for one to become a theoretical physicist.

I think it was an error for her (or him) to choose theoretical physics because this "Woitian" approach is certainly not one that can lead to any progress in science.

We have as many data as our experimental friends can give us. Everyone has the same data. We are trying to use the data and the previous theories to say as much as we can about the real world. With the missing Planckian scattering data, we obviously need to rely on mathematical methods and tools much more than our predecessors. When we do it properly, we end up with solving important open questions in string theory.

String theory is the most predictive theory one can imagine (and we ever had), and only those who prefer general confusing cliches over the technical analysis of the quantum gravity and other high-energy issues have not understood why.

When we do this job of making conclusions out of the data badly, we may end up writing and reading books like Not Even Wrong. Let me say that I would definitely encourage any student who has not understood this basic idea - namely that the cliches of Peter Woit are completely vacuous and misled - to quit the field.

I want to be as clear as possible in saying that this is not the type of "new spirit" that physics needs.

This is not quite the easiest period for the field, and if someone is furthermore confused about the conceptual questions that are independent of the status of the field, it could never be a pleasant experience for her (or him) to work on these questions.

And indeed, physics is not solving Peter Woit's questions because he has no interesting physics questions to offer. If someone prefers to insult scientists who are trying to do their best instead of helping them, he should become a politician or learn from Peter Woit how to earn money from meaningless books.

Thanks for your understanding.

Best wishes
Luboš


reader Lumo said...

Oops, Stephen, I forgot to answer your question Who is a suitable critic of string theory.

The most suitable critic of string theory is, of course, a person who has learned how all the essential - as well as many non-essential - questions are answered by string theory much like by all the competing (usually non-existent) theories and she (or he) can compare.

If you're afraid that this means that the most suitable critic of string theory is a leading string theorist, then you're afraid rightfully. ;-) Yes, this is the case. If you were picking a specific person, it is likely that Ed Witten would be the most suitable person to decide what's right and what's wrong in string theory. It's as simple as that. Still, he (or He) is not the omnipotent God which is why others try to think, too. ;-)

The second class but still reasonable critics are those who have been helpful in answering some important related questions in mathematics or physics, although they never learned string theory. But I think that it is obvious that their opinion about string theory - especially if their important discoveries were done 30 years ago - won't be as important as the opinion of Witten or Strominger, for example.

Does it make sense that the most meaningful critic of string theory is a string theorist? Yes, it does. But maybe it makes no sense to you. You may be imagining that theoretical physics is about a permanent struggle between the proponents and opponents of string theory. But this is a caricature of physics that you can only get by reading some silly blogs too often; and it is indeed sad if you really imagine that this is how physics can work.

The term "string theory" is not a universal enchantment that divides the scientists into infidels and believers whose main task is to fight with each other. Instead, we are trying to answer actual physical, testable questions - not linguistic and religious battles.

In quantum gravity, we are answering what happens at super-high energy scattering; whether the topology of space can change; where is the information going from the black hole interior; what happens with the electric or baryon charge in this or a different situation involving objects such as black holes, and so forth.

Once again, we are not answering the question "is string theory correct?" most of our lives because without all the "details", this is a question for simpletons like those who enjoy "Not Even Wrong".

It's simply the case the string theory is the only framework that is able to provide us with any kind of answer that can't be extracted from the simpler theories - such as classical GR - but that can still be justified rationally. In this sense, it is really analogous to the evolution in the sense that if someone proposes someone that is completely different, it is usually trivial to see why she (or he) is a crackpot.

No one is writing serious non-stringy papers trying to answer the actual questions in physics (of quantum gravity etc.) simply because this is not possible in a scientific fashion. Most likely, a different theory than string theory that can provide us with this framework does not exist.

If you think that you have one, write a paper about it. But in advance, you should know that most people who write such papers are not viewed seriously because they're missing some important insights. So I recommend you to try to learn more details before you try to generate big conclusions.

If you still feel that the string theorists are not sufficiently self-critical about their research and the research of their colleagues, I think that you are mostly wrong but let me answer your question by one more way that carries a certain albeit limited weight: a suitable critic of string theory is me. ;-)


reader Lumo said...

I've erased the spams except for the most polite one. ;-)


reader CapitalistImperialistPig said...

I think you just insulted me Lubos, but I'm not quite sure. I would mention that I am neither a critic of string theory by disposition nor qualified to be one if I wished, but I can count to one (and beyond) and I have yet to have any string theorist tell me of any confirmed prediction of string theory - and I've asked you more than once. I am willing to listen though.

It's nice that string theory doesn't contradict known facts, but that's never been the test in science before.


reader Lumo said...

Dear CIP,

your demand is logically inconsistent. If you ask string theory to have "confirmed predictions", then the word "confirmed" means that the validity of the prediction is already known, and therefore the prediction is a known fact that string theory does not contradict.

It seems that you want a prediction which is both confirmed as well as unconfirmed which is not possible.

If you want confirmed predictions, take an apple, let it fall, and you will see gravity in action. Gravity is a confirmed prediction of string theory. It is not inserted into string theory but arises as a necessary consequence of a particular state of the string theory's ingredients.

If you want yet-unconfirmed predictions of string theory, you will logically have to wait for some future experiments - when SUSY is discovered at SOME scale, macroscopic strings are found, or evaporating black holes and/or excited string states are revealed, among many other possibilities.

You may also argue that your demanded prediction should have been first unconfirmed, and then confirmed. Then it is easy to show that the confirmation must be in the future because so far, there has not been any convincing observation that goes beyond the previous layer of theories - such as the Standard Model.

You may view dark matter as a counterexample, and if you wish, its existence in the form of a new kind of particle - something that is almost obvious today - is a recently confirmed prediction of string theory.

Best
Lubos


reader Lumo said...

Fairy-tales whether something has been the test before or not are completely secondary. The progress in science is also based on the fact that it is always somewhat new.

What I say about the things that are possible and things that are not is an inevitable logical consequence of the known facts, and logic is always superior over some prejudices that things should look exactly like in the past.

Logic and mathematical consistency is also the main tool that allows the string theorists to know something beyond the Standard Model and GR even without obvious experiments while others - those who prefer empty religious statements and insults - know ... nothing, to put it politely.


reader Arun said...

CIP,
It's nice that string theory doesn't contradict known facts, but that's never been the test in science before.

Depends on what you mean by "contradict known facts". With string theory lots of work is involved in showing how it saves the appearances (e.g., we appear to live in a 4-D world), and string theory has too many ways of saving the appearances, all of which cannot be right.

Anyway, Lubos Motl is probably right that when we don't have much to go on, mathematical speculation is the only way forward. The question is still open as to when this exercise goes from being science to becoming a branch of philosophy or mathematics. Aristotle too could protest that he was rational (undoubtedly) and that the technology of the times limited him (also arguable).


reader Quantoken said...

Lubos said:
"I've erased the spams except for the most polite one. ;-)"

Funny. Those spams are generated by automatic computer codes. When did you find a computer program being polite, and you feel the need to be polite to a piece of computer code? Delete them!

Quantoken


reader CapitalistImperialistPig said...

Since you apparently misunderstood me, let me clarify. A prediction is a statement about something not yet discovered - it becomes confirmed when it is discovered. Dirac's theory predicted the discovery of the positron, Pauli predicted the neutrino, Gell-Mann and Pais predicted the K0 phenomena, Gell-Mann predicted the Omega-- and the quark. Each was subsequently predicted.

I'm fairly sure that dark matter was discovered before string theory predicted it - but maybe you can cite a paper that proves me wrong.


reader CapitalistImperialistPig said...

Sorry - that should read subsequently confirmed.


reader Lumo said...

Dear CIP,

I've answered your questions already. It may be your turn now to try to read the answers.

Still, you seem to confuse physics and history of physics. The development of Dirac's equation - as well as all other examples you mention - belong to the history of physics while string theory belongs to the current physics research.

Of course that string theory cannot be confirmed and established today in the same way as the other examples you mention because this would mean that it has become a part of history of physics, too.

String theory is, on the contrary, a currently active field in physics. When its unambiguous proofs are found, it will be almost the end of the story. We will be all heroes, the critics of string theory will have to splash themselves in the toilet, and everything will be very different.

In other words, if you want to stop research of string theory, the right strategy is not to spread confusing and bitter remarks similar to Peter Woit or some other unnamed sourballs - remarks that every educated theoretical physicist knows to be derived from a complete ignorance about the actual physics questions. The right way to stop string theory research for you is to find unambiguous evidence that it is correct.

Best
Lubos


reader Lumo said...

Let me say the same things a bit differently. We all continue to work on string theory and think about it exactly because we're *not* yet quite sure that we can prove it correct.

Why is it exactly string theory that we're working on? Because string theory is the only promising framework that goes beyond the language of effective quantum field theories. We don't know any other framework like that and it is very likely that no other framework of this kind exists.

In any other system of ideas, the predictions of new phenomena are pure guesswork.

Anyone in physics has the right to choose a completely different approach. But it is likely that any conclusions based on something else will have the character of black magic and guesswork. String theory is the only arena where things fit together and can be derived from each other scientifically.

Everyone who is smart enough and has worked on the actual physics questions of quantum gravity and other things for a sufficient amount of time knows why this argument about the uniqueness of strings is so incredibly powerful.

Of course, there are other people who don't want to be answering "small" questions such as what happens with the information, what does the high-energy scattering looks like, etc. They first want to answer the "big question" - namely "is string theory correct?" Of course, this is not possible, and the people who always answer "big questions" first are classified as crackpots, as you could know if you read my text about common crackpots' errors.

No one is interested in morons who keep on repeating stupid, boring, and obnoxious comments about the experimental unavailability of string theory - pretending that it is a problem of string theory. They're dumb, useless idiots.

The lack of experiments that would give us clear hints how to go beyond the Standard Model is not a problem of string theory; it is a problem of anyone who tries to make any progress in theoretical physics these days. If someone associates this fact exclusively with string theory, then he's a complete idiot, and those who submissively read his texts are even more manifest idiots.

What I want to be answered are actual physics questions like what happens with the information in the black hole, how does the exceptional E_{10} symmetry enters the most symmetric compactifications of quantum gravity, how can one compute the parameters of the low-energy effective field theories, and so forth. If you have other scientific ways to answer these scientific questions that don't involve string theory, try to offer them. (Both of us know very well that you have absolutely nothing, much like all the other "smart" critics.)

If you have nothing to offer, then I encourage you to shut up because your comments are completely counter-productive and idiotic.


reader Quantoken said...

Lubos:

I can agree with some of your statements but not some others.

It is OK to say that lack of prediction and experimental evidence COULD BE just temporary as string theory is still in development, and the whole field is difficult in general so progress just may take more time. It could well be that it may just take more time, 40 years, 100 years, 500 years, before all the details of this theory is worked out.

But it is NOT OK to ignore the possibility that the whole approach is simply wrong, and when you eventually figure everything out, you find it is a dead end. Such possibility exists, and as the amount of man power piled up over the years, without seeing real progress towards verifiable predictions, you've got to admit that the possibility that it is wrong is becoming ever more likely, and the possibility that it is right becomes ever more unlikely.

So far I do not see any thing essential that adds to the support of SST. The two hypothesises that started the whole enterprise:

1.The spacetime is 10-D. 2.Fundamental particles are not point-like, but rather string like, and different vibration modes leads to different particles we see.

These two hypothesises has no evidence support so far.

The biggest claim of success of SST so far, are
1.Derived the existance of a spin-two zero-rest-mass boson, which you call graviton.

2.Derived the correct Blackhole entropy formula.

As I discussed previously, any reasonable calculation has to lead to a result with the CORRECT UNIT. And this requirement along has already locked the blackhole entropy to something proportional to its surface area in Planck unit, and multiply by a numerical constant. After than, getting the remaining trivial factor 1/4 correct, is simply a numerological coincidence. I have shown quite a few trivial arguments, right or wrong, do leads to the 1/4 factor.

As for the prediction of graviton. I can shown that graviton can NOT exist. If it exists it breaks the strict Equivalence Principle, as well as the Hawking-Bekenstein Entropy Bound. Gravity effect must be explained as purely geometry effect of spacetime curvature, and no boson is exchanged. So the fact that SST predicts graviton is actually evidence that it is clearly wrong.

Einstein says there is NO distinction between acceleration or gravity attraction. If gravity is exchanged by a boson called graviton, then the two cases can in principle be distinguished by observing whether any graviton has actually been absorbed by the object or not, breaking the equivalence principle. There can be many more gedanken experiments where the existence of graviton breaks the EP.

Quantoken


reader Leucipo said...

Why is it exactly string theory that we're working on? Because string theory is the only promising framework that goes beyond the language of effective quantum field theories.

Exactly! This is an important point. Weinberg and others impulsed the current view of Quantum Field Theory as a "language of effective quantum field theories". That any QFT refers to an scale range and that beyond this range a new QFT theory will appear, turtles and turtles. With this panorama, young stars escaped towards strings, meeting some elders trained in the hadronic dualities, and so we are where we are.

But perhaps the QFT effective is wrong.


reader Ken said...

Regarding Quantoken's argument about the graviton breaking the equivalence princicple... As I understand it, the exchanged graviton would be a virtual particle, and thus off-shell, would Quantoken's argument still stand? Just as we explain electromagnetic force as the exchange of virtual photons, does this mean that electromagnetic force also violate the equivalence principle?

As this is my first posting, I would like to thank Professor Motl for running this site. I have read this site (and also Peter Woit's site) regularly for quite some time now. And it has been quite enlightening.


reader Quantoken said...

Ken:
Nice try! But EM forces are exchanged by REAL photons, not virtual ones. Likewise, gravity force, even those between gravitons, must be exchanged by real gravitons, not virtual ones, assuming a thing called graviton exists.
And even the very idea that SOME gravity are exchanged by real gravitons, and SOME OTHER gravity are exchanged by virtual gravitons would seem very odd. It hints not all gravity are created equal. And that defeats the very notion of equivalence principle. If gravity and acceleration is indistinguishable, at least there should be no distinction between one kind of gravity and another kind. And I shall further ask what exchanges the gravity force between virtual gravitons, eh? Virtual virtual graviton? Or triple virtual one?

Any way you think about it, gravitons are troublesome. Think about a lightest photon and another lightest photon at the other side of the universe, they still need to exchange gravitons to gravitate with each other. You need a infinite number of gravitons to do that. The natue simply can not allow an infinite number of any thing to be existant.


reader Dimitri Terryn said...

"Also, I believe that a student who thinks that the sentence "there are no testable predictions of string theory" reflects the important questions that an active theoretical physicist should ask in 2005 is probably a student at the level of the readers of Peter Woit's blog. And that's far too little for one to become a theoretical physicist."

That is some argument, and rather a dishonest one Lubos.

For one thing, you take a perfectly valid concern "Does the theory I'm working on allow experimental verification" to it's logical extreme, namely that it is THE main concern. Of course it isn't, but it shoudn't be ignored either. While the proporties of the various string models are being examined (and that being the main task of the theorist), it should be obvious that at one point one should be thinking about actually applying the theory to an experimentel situation. Otherwise it's not science, it's philosophy. Or worse, scholastic.

I think it is better to actively reflect on this question then to act condescending to anyone who raises it, as seems to be your style.

- A young student doing his Master's thesis on QFT this year


reader Lumo said...

Dear Dimi,

my guess is that you must know very well that what you write is completely unrealistic.

99.9% of the hep-th arXiv contains stuff that can't and won't be verified in the next decade.

It's just not possible to generate so much stuff that is directly testable. If someone wants to work on directly testable things only, she would either have to be checking the Standard Model all the time. Fair work.

It works. Again. Another scattering. Again. Success. Again.

This is not how the majority of the hep-th community views interesting progress because the Standard Model is simply a correct effective theory.

Alternatively, she would have to be guessing all the time what can happen at the next collider, invent hypothetical chances what could still be seen at the existing colliders, and so forth. Interesting work, and this is mostly hep-ph.

These things are related to experiments, but in this case at least 99.9% of the specific models and papers are going to be ruled out. And to some extent it is fair to say that guessing phenomenology beyond the Standard Model is a lottery.

For example, all of thousands of papers written since 1975 (and essentially 1970) that were predicting a new (which, in this period, already meant beyond-the-SM) physics to be seen by 2005 have been ruled out.

The goal and strategy of the string theory community is different. It is about the development of a framework in which insights about reality are not pure guesswork because it is based on deep mathematical thoughts. It needed and probably still needs work and the path is not straightforward either.

In my comments above, I wanted to say that a student with a master degree in these things should already have some idea that the world is not that simple, and she will be checking things that everyone else considers obvious in 99.9% of the time; or she will be making specific guesses about the new physics out of which 99.9% will be shown wrong; or she will be building a path to a conceptual and reliable derivation of facts, out of which 99.9% won't be directly experimentably testable in foreseeable future.

Nothing else is possible in the bulk right now and this fact has nothing to do with any particular features of string theory.

Some of the current students may be very lucky and they would fit to neither category. But a student must already know all these things. It is also important to know that others can respect you as a physicist even for insights about old models; for new models that don't have to be correct but look deep; and for contributions to a better understanding for the overarching theory.

What I criticize is if someone has a completely unrealistic idea what is possible in the current world and what is not possible, yet he wants others to derive any implications from this unrealistic viewpoint. I don't suggest that people should be deliberately choosing questions and models that are untestable; they should be choosing the best approaches to reality and be ready that it is simply impossible that most of their work will be simultaneously new, correct, and testable.

All the best
Lubos


reader Plato said...

My Grandson is nine years old.

We talked about "prediction and guessing" last night. He wanted to know the difference.

I find this discussion here has helped to see better this morning in the one sense, and on the other?

How model apprehension, will make way for prediction. If such a model is wrong, the prediction will be wrong.

My grandson saids this is guessing?:)

I said to create such models was important, and was better then guessing. Was I right?


reader Lumo said...

Dear Plato,

creating models and calculating phenomena using the models is better than guessing the particular results of particular experiments.

It's simply because a model is something that relates many observations that would a priori be independent - and that would have to be guessed separately.

Nevertheless, when you try to say something about the future experiments, you are guessing even if you are creating the model. Your model is likely to be wrong.

But the probability that a model is correct - together with all of its predictions - is still much much higher than the probability that a random set of guesses about many individual phenomena will be uniformly correct.

Predicting things using models is therefore less of a guesswork, but it is guesswork nevertheless. The deeper theory - and more unified model - you deal with, the more rational the process of guessing becomes.

Best
Lubos


reader Leucipo said...

Let me elaborate a bit more on my previous remark. It is sometimes remarked that the pre-cooking of the main string theory techniques (including superthings, critical dimensions...) created a pool of Thesis Advisors well prepared to reenter game at the age of the String Revolutions. But this remark does not explain why young bright students were eager to enter string theory instead of pursue the refining of Quantum Field Theories.

The point, to me -and I expect Lubos and others could add some insight on it- is that QFTs, seen under the glass of the new renormalisation group (Wilson Kogut, Kadanoff...), were inherently restricted to an interval of scales, and then each model was unable to go beyond its own scale. The theories were "effective", a word whose deep sense was that the theories were uneffective to describe the world up to the infinite momentum or infinitesimal displacement.
Thus young bright students, accepting this pessimistic view of QFT, were easily lured into the only big alternative available, string theory.
Today the view of QFT is, IMAO, not so pessimistic as in the seventies. Connes-Kreimer insight of the mathematical meaning of the renormalisation group, around 1995, could help us to detach ourserves from the "effective" view. The previous models of 4+discrete dimensional theories (Connes-Lott, Coquereaux et al... and all the modern ones) could generate supersymmetry if we were able to exchange the discrete coordinate by a grassmanian one in some adequate limit.


reader Lumo said...

Hi Leucipo,

good points. The string theory community has grown during the revolutions in particular. Whatever was the composition of the physics community at that time, there were extremely good reasons to stop anything else and focus on string theory.

In 1984 Green and Schwarz showed that the main technical "rumors" that were previously used to humiliate string theory were just mathematically wrong. It is not true that one does never get a consistent gauge theory at low energies. It is not true that the gauge theories inevitably bring anomalies. If computed properly, the anomalies in string theory cancel and the cancellation looks like a sequence of miracles. Miracles that physicists can nevertheless understand pretty well.

This killed most of the people who thought that string theory was inconsistent in some way, and it killed most of the people who thought that string theory did not have the right physics to match reality. The second group diminished even more when the heterotic strings and their Calabi-Yau compactification was constructed.

It became clear in 1985 that string theory generates everything one needs to describe everything, and it was completely reasonable to expect that the "final theory" could be completed within months. I am saying it even though I realize that the following 20 years showed that the task was actually much more difficult and the optimistic expectations exaggerated.

The second boom for string theory was around 1995 when all versions of string theory got united, the seemingly awkward components of string theory - like the RR fields - got their meaning and new objects (D-branes) charged under it. The D-branes are essential for nonperturbative physics and their analysis reveals many properties of black holes as well as gauge theories and holography etc.

It's been a deep revolution that proved that string theory is almost certainly a consistent theory involving quantum black holes, well-defined strong coupling behavior, and so forth. It just eliminates the work on hypothetical alternatives because they seem ridiculous in comparison.

A great part of ambitious young theorists chose and choose string theory because this is where all the important insights seem to meet and where the most revolutionary insights are expected. There may be isolated cases in which people are "manipulated" into string theory but that can't work globally, as a systematic measurable effect. There are many physicists who are not string theorists - and it just seems that the work of string theorists has simply been more attractive (and still is).

If it were possible to generate interesting non-string-theory physics trying to solve similar questions, be sure that this direction of research would grow rapidly. Mostly everyone calls for alternatives, and so forth. No problems with funding. A huge pool of people who are potentially skeptical about string theory and who could jump on the new bandwagon. But there is no alternative, so this boom just can't start. And it will never start, I think, simply because it is not mathematically possible.

Some people may find it politically correct if there were an alternative; but mathematics does not allow for this kind of political correctness.

Best
Lubos


reader Leucipo said...

I can not tell about the 1985 revolutions and its ability to convert people from other topics (GUTS, Sugra, etc) towards string theory; but I was marginally present at 1995 and it seemed that the main feel in the revolutionary army was an enthusiam about the huge quantity of work that could be published in order to analise and understand the unifying dualities. IE: somewhere in these 10 years the desire for a final theory had become lost, or sublimated if you want, and the good reasons were starting to be more about private enjoy that about real goals.

Another thing I noticed in the 1990-2000 period was a curious corolary of the axiom of "lack of alternative": any tool of any utility should necesarily be incoporpated into string theory (If not, it is not an useful tool). Thus the belief in the absence of alternatives is strenghted via a social mechanism, suplementing the original "effective theory" argument against QFT modelling.


reader bjackk said...

Hi,
Could you address Electroweak breaking in string theory? Are there fundamental scalar to be served as Higgs from string theory? Is adjoint-Higgs breaking realistic?
Even there exist a number of CYs which give SM mass spectrum, do they pass electroweak precision tests? Can heterotic explain acceleration of the 4D universe?