String Theory Kerfuffle

Nobody made an analogy of the correctness of string theory to that of evolution. It was a misinterpretation or a remark.

It might be worthwhile to begin by noting that Sean is not a string theorist himself, and so isn't making these statements out of pure self-interest.

Sean also did not really say that string theory must be right, just that "there are very good reasons to think that something like string theory is going to be part of the ultimate understanding of quantum gravity". If you want to know how he "knows" that, you probably need to wait for the unfinished post he was advertising.

But, jumping the gun, I can add a few cents myself:

It's not true that string theory doesn't make testable predictions. There are many string models which make specific predictions. The problem is that (a) many of them aren't testable now, and (b) there are few realistic tests that can rule out all of string theory at once. To (a) I say that many of them are testable now (and have been ruled out, e.g., various braneworld and large extra-dimension scenarios), and many will be soon. To (b) I say that the same is true of quantum field theory -- what realistic experiment can rule out all QFTs at once? -- and nobody complains about it being "untestable".

Beyond testability, many of the major reasons for favoring string theory arise from consistency. Requiring that quantum theory and relativity be consistent with each other has proven to be a very strong constraint on what theories are even logically possible. (Of course, you can toss out quantum theory or relativity altogether, and some have, but within the context of "a theory of quantum gravity", which is what Sean was talking about, you assume both are valid.)

In one sense, string theory has proven to be more predictive than other attempts at quantum gravity, because in string theory the high-energy physics does not depend on the (non-)existence of unknown processes at arbitrarily high energies: the high-energy behavior is uniquely determined and you can't introduce new super-massive particles or whatever to change it. In all other known attempts at quantum gravity, the high-energy behavior of the theory can be altered by new physics that enters as close as you want to the Planck scale, and therefore the high energy behavior of the theory can never be known (being always beyond what we can test). For more on this line of reasoning, see Jacques Distler's blog.

Another point is that string theory is mathematically inseparable from quantum field theory, so if you accept the latter (which is what the Standard Model of particle physics is based on), you are pretty much forced to accept the former in one form or another -- and due to the unified nature of string theory, you can convincingly argue that "one form" of string theory is equivalent to the whole of string theory. For more on this line of reasoning, see Lubos Motl's blog.

For more detailed discussion of both points, I refer you to an old Slashdot post of mine.

For the record, I'm not a string theorist either. I just agree with Sean that most people don't understand what is so compelling about string theory, even in the absence of experimental tests. Theoretical consistency can be important too. When Pauli posited the existence of the neutrino, he did it by requiring that the experimental results be consistent with conservation laws, even though there were absolutely no tests at the time that showed any kind of direct evidence for such a particle. It's possible that the conservation laws were just wrong, but I don't think it would have been outrageous for Pauli to say that "there are very good reasons to think that a neutrino is what it needed to reconcile our theories of particle physics which at the moment appear inconsistent with each other".

"of a remark"

I think that the position the string theorists are coming from when they say that "it must be right or at the very least something like string theory is correct." Is more or less recognition of logical necessity rather than scientific hubris. That is, if we accept that there is some grand unified theory of everything that can be captured mathematically as other scientific theories in physics have been, it will necessarily have a certain type of theoretical "shape".

A large part of the whole string theory enterprise (from the perspective of a mathematician looking in from the outside) seems to be reconciling the theoretical and logical implications of things we can observe and have independent theories of through the string model, the classic example of which is reconciling general relativity with quantum theory.

I read the assertion that the truth must be "something like string theory" as simply the acknowledgment that these constraints have been pushed through the theoretical body of string theory and accounted for, any other viable theory would have to account for the same constraints.

Assistant Professor of physics at Harvard, Lubos Motl, has claimed in book reviews that criticising string theory is as irrational as criticising evolution, example:

http://www.amazon.ca/o/ASIN/0465092756/702-9001745-8839239?Subscription…

"Bitter emotions and obsolete understanding of high-energy physics, Aug 25 2006
Reviewer: Lubos Motl (Cambridge, MA United States) - See all my reviews

"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 funny thing is that Lubos' is the only review at that particular defunct Amazon page (the book is mainly being sold on the main Amazon.com and co.uk sites), so Lubos is including onlyhimself as a "crackpot...reviewer"!

String theory is having to defend itself by personal attacks and claims that anyone who demands science from a theory is a "science hater". It's really pathetic propaganda and some percentage of people reading such propaganda will see it has no scientific content, just like string theory itself.

Ukko: there's no evidence at all for any of the speculation they have, so there's no reason why their particular method is the only one possible. The failure of string theory to achieve anything, mainly because of the complexity introduced by the Calabi-Yau manifold which allows the string to take a virtually endless number of different forms each representing a different set of physics, suggests that it is a dead end and can be ruled out. Instead of this, it's becoming a religion based on the belief or faith that it is the truth, like Plato's pseudoscience. Unless you can check a theory, it's not science.

nc: I agree with you that it is necessary to be able to check a theory for it to be considered science; however, I believe we are disagreeing more in the matter of degree. As I understand the situation string theory is not untestable a priori, but rather that it is difficult to identify such a test at present. I personally don't have any problem with a given scientific program or effort taking a significant amount of time refining its theoretical underpinnings before making testable predictions about the world. That said, of course it's better to have earlier predictions in earlier tests to avoid running down a blind alley and to better inform the development of the theory in question. I just don't see gosh you guys are taken a long time as a valid criticism in this context - I mean it's taken a long time getting here and things are only going to get harder.

ukko,

The problem with string theory is that, in our current understanding of what "string theory" is, it does appear to be untestable a priori.

(see http://www.math.columbia.edu/~woit/testable.pdf)

There are very basic reasons no one has been able to identify a conventional scientific test of the theory, and the only way around these seems to be for there to be some revolution in our understanding of what string theory is, or some dramatic unexpected experimental discovery. Both seem pretty unlikely in the forseeable future.

Hi Ambitwistor (your multiple links caused the spam filter to junk your comment - I just unjunked it),

Yes as far as my limited knowledge of string theory goes, I agree that ST is testable in principle, yet with our current technology it isn't. As an experimentalist I am very weary of theories that do not add further insight.

Physics today has several problems, dark matter, dark energy, an incomplete understanding of why there are so many fundamental particles, the paradox of QM and relativity. A model is only worth anything if it can explain these disparities and give further insight - insight that leads to testable predictions. ST tries to bridge the QM/ST gap but doesn't really offer a concrete way to do so, just a hint that it might work (if we ever figure out which of the billions versions of ST is the correct one). Maybe ST is right, maybe it isn't, but to say that it is beautiful or compelling doesn't mean anything if it can't make predictions about all that's baffling around us.

From an outsider ST seemed like it was promising but as it became more prominent, it also devolved into amorphous blob. Right now it can be anything you want it to be, which is the equivalent of saying that it is nothing. To imply that the standard model relies on ST is stretching it. I guess Popper would have something to say about that.

If there is anything that I've learned from being a scientist is that we are often mislead into hubris. Reality, i.e. empirical data, throws curveballs. ST seems quite divorced from empirical science and that is worrisome.

I would object quite a bit to say that string theory offers "no insight". With respect to the specific issues you raise, all string theories resolve the "paradox" of QM and relativity. String theory has inspired models of dark matter and dark energy that would not have been conceived of within the QFT framework. I can't imagine why an experimentalist would be opposed to more models which can be tested.

It is again wrong to say that string theory cannot make predictions about anything that baffles us. The classic mistake is to compare a theoretical framework, such as string theory, with a specific concrete realization, such as the Standard Model of particle physics. You should instead compare string theory to quantum field theory in general.

"Quantum field theory" does not have anything specific to say about dark matter, dark energy, etc. (and is inconsistent with gravity to boot, unlike string theory). All it really says is "things are made out of quantum fields". Specific models within QFT can say specific things, such as the Minimal Supersymmetric Standard Model, the Standard Model with axions, extra light scalar particles, etc.

Likewise, all "string theory" says about physics is "things are made of of strings, membranes, etc." If you want to consider dark matter, dark energy, etc. you can construct specific string models which have those (e.g., dilatons, KK particles, etc.)

String theory is not really any more of an "amorphous blob that can be anything you want" than is quantum field theory; you can write down infinitely many different QFT models after all, with whatever particle content and couplings you want. String theory does give you some extra options for model building, but this is a feature, not a bug!

I did not say that the Standard Model "relies" on string theory. However, there are good reasons to believe that many if not most self-consistent QFTs are equivalent to string theories, so arguing "for" QFT and "against" string theory may well be moot.