Talk:String theory/Archive 3

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Archive 6

Unreferenced speculation

"Brane-world scenario

This section does not cite any references or sources. <etc...> (November 2008) Another possibility is that we are "stuck" in a 3+1 dimensional (i.e. three spatial dimensions plus the time dimension) subspace of the full universe. This subspace is supposed to be a D-brane, hence this is known as a braneworld theory. Some believe that some combination of the two ideas — compactification and branes — will ultimately yield the most realistic theory.[citation needed]"

Looks like some rogue physicist has tried adding some wisdom to this article :p 121.219.54.76 (talk) 04:36, 9 April 2009 (UTC)

lack of quantitative data

This article is rather disappointing to me in that it seems more of a qualitative summary of the status of string theory. In fact it reads more or less like some of the popular science children's books, where one has protons, neutrons, quarks etc next to a big shiny sign saying "what is stuff made of?" Of course there is nothing wrong with such treatments, but it does not seem appropriate to the audience that actually wants to know the driving equations behind a result ie such as advanced university undergraduate level. Not really terribly satisfying to a trained scientist. It would be helpful if there was more discussion of the precise abstract structures used, their applications, field equations from actions etc. Even if this was down in watered down language like matrix multiplication etc that would be ok - the Feynmann lectures in physics are a good example of the sort of standard that I would like to see. Perhaps an expert in the field can address this glaring inadequency. RogueTeddy (talk) 09:31, 28 February 2009 (UTC)

The reason that you don't do that is because it is a wrong way of thinking. This article could describe the particular choice of mathematical objects which were useful in the early days for describing string theory, but that would leave out the physical motivation. The physical reasoning makes it clear why it had to be that way (in hindsight).

String theory is completely different from field theory: you don't give it a local space-time description. There is no action principle, or a field equation, at least not in the usual sense. There's the RG fixed point equation, which reproduces the classical field equations, but that's a classical background equation, it's not supposed to generalize into a quantum dynamical law. The action for strings is the action for a path-integral on all string paths, which is just like a Feynman diagram in Feynman's original formulation, it is a way of calculating a unitary S-matrix. It separates out positive and negative frequencies and produces no space-time description in the usual sense of the word. Using it, you can't really say "I have a description now, what's a description of the future". You have to do it by S-matrix, which means "I have a bunch of incoming free particles, what will come out once they scatter?", or else you have to say "I have an AdS space, and some state on the boundary, how does it evolve in boundary-time". Either way, the holography gets in the way of a straightforward action principle in what you would think of as the usual space-time.

The string action on the string worldsheet should not be thought of as a space-time description, but as a summary of the particle content on the worldsheet (or as a description of the dynamics in a specially chosen AdS3 space). You can give a field theoretic Lagrangian description for the low-energy field theory approximation, and work with that, but that's just supergravity, it's not strings. If you want to do strings, you have to give up on a microscopically detailed picture of space-time.

The only real dynamical descriptions are those which are reasonable from a holographic point of view. A light-front or an AdS asymptotic space, or some other near-extremal black hole asymptotics. In those situations, you are given a special type of space-time, and the description method is not the obvious one: you don't make space-like slices of equal time and describe a state on a slice. Since the pedagogically best description of the theory is not at all clear, it is hard to know where to start with a mathematical description. It is probably best do it in historical order. But history begins with a lot of Regge theory.Likebox (talk) 17:32, 3 March 2009 (UTC)

I am not an expert in the field.Likebox (talk) 17:34, 3 March 2009 (UTC)

Thanks for your explanation Likebox, it certainly makes things a bit clearer to me. For instance I had not heard of the S-matrix before, and its role in scattering etc. I guess my particular bias, which is not necessarily right, is that for progress in physics one must really develop one's abstraction (mathematics) and physical intuition in lockstep, rather than letting one outpace the other. Nonetheless you make some very interesting comments and I appreciate your response. Best wishes, RogueTeddy (talk) 20:28, 7 March 2009 (UTC)

I agree that the mathematical development needs to follow the physical intuition closely. But in this case, I don't know the best path through the historical strata of formalism. The problem is that the mathematically clearest and most elegant formalism is the worldsheet CFT, and then there are certain things that are hard to explain properly:

1. Physical Masses: The excitation content of the string theory is a bunch of particles whose masses are given by the conformal scaling dimensions of the 2D fields. Why is that? In the modern way of thinking, it's a manifestation of holography. But it was discovered before holography historically, and it is difficult to motivate directly from the worldsheet expansion. Historically, it came from the assumption of Regge behavior and straight line trajectories, because the Regge assumption tells you the masses in terms of the angular momentum, and the angular momentum turns into a scaling dimension on the world sheet by conformal invariance. But that's so opaque physically. I am constantly amazed that those seventies people figure all that out from such miserable input.
2. Scattering Amplitudes: to get the actual scattering amplitudes from the worldsheet correlation functions you have to do some sums over channels. Historically, this came first, with Veneziano, so this wasn't mysterious. But if you start with the worldsheet description, it's hard to justify. The worldsheet amplitude needs to be "pasted on" to the boundary conditions. Since the string expansion is not derived from a Hamiltonian, the pasting needs to be laboriously justified by unitarity. In modern textbooks, since you don't need to match to experiment, the calculation of the scattering probability is usually not carried out all the way.
3. Unitarity: The string perturbation expansion is unitary, but it isn't constructed from a field Hamiltonian, so there's no fundamental reason for it to come out unitary. In AdS/CFT, in a holographic interpretation, worldvolume unitarity follows from unitarity on AdS space. But the unitarity of the string genus expansion is hard to understand. To prove it, I think you need to use non-covariant space-time 1/L0 propagators, and use the very out of date "sewing" rules. It isn't obvious. This point is glossed over in modern textbooks, which end up just assuming unitarity.
4. No Ghosts: But the textbooks make a point of proving the no-ghost theorem. The problem is, the no-ghost theorem is not hard or illuminating on the worldsheet. But historically, no-ghosts plus sewing loops guaranteed perturbative unitarity for the string perturbation expansion.

There's probably a really good way of explaining all the points that I can't see.Likebox (talk) 16:51, 9 March 2009 (UTC)

Just to make the previous comment clearer: a description of string theory starting with an action principle, like NG action, and then deriving energy momentum etc for the particle spectrum is not satisfactory for a first-principles explanation. The reason is that this way of doing things does not make unitarity obvious, nor the interactions, except if you already know all about dual models, sewing, and Regge theory. It makes more sense holographically, but ignoring that, the action principle is just a physically suggestive mnemonic to derive the string spectrum. It is useless if your goal is to compute scattering amplitudes from first principles.

For example, suppose you try to explain the theory to a martian:

You: Earthlings have discovered a theory of quantum gravity!

Martian: Thank you Earthling! We on Mars understand field theory very well, but not quantum gravity. How does it go?

You: just replace points by strings in the sum-over paths formulation of field theory!

Martian: (thinking) Ok. But what is the world-sheet action?

You: Isn't it obvious? It's area! (write NG action)

Martian: Why not include the extrinsic curvature?

You: Um... you just don't.

Martian: Why not include nongeometric worldsheet fields?

You: Um... you just don't.

Martian: Why can you use space time embedding coordinates as if they are flat, when the geometry is fluctuating?

You: That's because, as you'll see, the theory describes strings much longer than the Planck length!

Martian: So how can you call it a theory of quantum gravity?

You: (irritated) look, just calculate the spectrum.

Martian: Ok, Ok. It's the area. (Calculating) I get some masses and angular momenta. Ok. How do you describe the forces?

You: Isn't it obvious? It's by genus expansion!

Martian: Why can't I add interactions whenever the strings intersect?

You: Um... you just don't.

Martian: How do you know it's unitary?

You: Um... it just is.

There's also a grizzled old S-matrix physicist who's been overhearing the conversation:

S-matrix friend: Mr. Martian, ignore this young person. I can explain everything to your satisfaction. This theory is designed to give a theory of particle exchange, just like Feynman diagrams. The particles which are exchanged have to be present in the spectrum. The spectrum consists of linear Regge trajectories.

Martian: So what is this area action?

S-matrix friend: That's just a way of summarizing the spectrum. Once you know the spectrum, you derive the interactions by enforcing unitarity when the particles exchanged in the t-channel are those in the spectrum, the amplitudes have no point-like behavior, no power-laws, they fall off as expected from Regge theory. That's why you can't add extra stuff like interactions at the point of intersection. Also, notice that the sum over channels has this nifty world-sheet duality, let me show you (scribbles duality diagrams).

Martian: I see! So how come it's all derived from this world-sheet action?

S-matrix friend: The miracle is, that the amplitudes you get by this procedure, when you remove all the ghosts, end up being a sum over different genus surfaces of a world sheet action that is conformally invariant. I don't know why that works.

Martian: (calculating) You're right! why is that?

S-matrix friend: Stop asking why. Notice there's a tachyon. You need supersymmetry to get rid of that.

Martian: (calculating) You're right! It's a miracle.

S-matrix friend: ok. Note the graviton in the spectrum.

Martian: Another miracle. (folds tentacles) But that's not real quantum gravity--- you're just doing perturbations around flat space! Where's the fluctuating geometry?

S-matrix friend: (folds arms) You ungrateful alien! I show you a finite perturbation expansion for gravity, and you complain!

The S-matrix person and the alien are about to come to blows, but then a holography person chimes in:

Holographist: I can explain everything! It's really because the physics of the sum is revealing a property of the classical black hole limit of these strings. There's a corresponding holographic description for many different types of black hole in the theory.

Martian: Is that how come it's unitary both on the worldsheet and in space-time?

Holographist: Yes, that's because the conformal theory is really describing the dynamics of an AdS space. It's a near horizon dynamics.

Martian: Ah! I think I'm starting to understand all these miracles. Thank you Earthlings. We on Mars have been stumped by quantum gravity for two millenia. We are amazed by your wisdom and foresight. (flying saucer takes off)

The point is that if you want to explain the mathematical formulation of string theory in a logical, rather than historical, way, you need to explain unitarity and S-matrix. Otherwise, how can you justify the world sheet action and genus expansion? They really don't make any sense at all outside of S-matrix theory. Modern books just gloss over this, although it was a big deal in the seventies.Likebox (talk) 18:21, 9 March 2009 (UTC)

To restore balance, I want to add that the above discussion is really overselling the S-matrix point of view, and shortchanging the modern one, which is superior in nearly all aspects. The above is just a list of annoyances for learning string theory, which I think can be made clearer with some historical development.

Reading the above, a lot of the suggestions are just plain wrong. In particular, it is not too hard to give a modern presentation which treats all the points above, with very little S-matrix theory:

1. The NG action is natural, it's lowest order Lorentz invariant term, so you can't just modify it by higher order terms without fine-tuning the area coefficient to zero. In principle, you could do the fine tuning, but that question is the same as "how many consistent string actions are there?"
2. To justify the scale-dimension mass correspondence, all you need is conformal invariance--- translation on the a long worldsheet tube gives the mass, and it's the same as a scale transformation on the worldsheet after you map the tube into a punctured plane. That's the 1980s presentation for the mass/dimension correspondence. I personally didn't internalize it until I understood that historically, this is the Regge trajectory hypothesis, except reworked to talk about operators and CFT's instead of the spectrum directly. This type of discussion is fine if you accept that the worldsheet theory must be a CFT, but it doesn't motivate why it should be a CFT. Linear trajectories plus sewing motivates this.
3. To justify unitarity, it is enough to show that a genus expansion cut up along a tube keeping the external momenta fixed is equivalent to the sum over all operator insertions, giving a Feynman propagator for each particle in the known spectrum. This can be easily shown in the modern point of view. When a string is cut, you can do the sum over the endpoint by letting the tube shrink to a point, splitting the point in two, and including all the operators you can put at that point in the sum. This reproduces the unitarity condition.
4. I left out factorization, but that's OPE. Also natural. The only difference is that the seventies point of view constructs a consistent OPE from the spectrum assumption and soft scattering, which might motivate things better but takes more steps. The 1980s point of view is just as good for this, but is perhaps somewhat less motivated.
5. To get the actual S-matrix elements from the amplitude is easy to understand by analogy with field theory. So that's not really a big deal.

In short, like everybody says, it's really no problem to reproduce every old fasioned result from the 1980s formalism. But I still think (perhaps for no good reason) that the historical stuff is better for motivation. So maybe there's a good way to write the mathematical development after all (along the lines of Green Schwarz Witten):

1. Discuss the Regge theory semi-quantitatively, giving the explicit amplitudes as examples.
2. Discuss the transformation to the conformal world sheet point of view, and how to replace trajectories with towers of operator dimensions. I think it is best to introducing the L operators in the Regge point of view, and then changing to a CFT interpretation.
3. quickly introduce NG and Polyakov actions, because that's easy. It's never the rate limiting step.
4. Then show unitarity etc in the CFT point of view, but relating it back to S-matrix stuff.
5. Then textbook stuff.

Perhaps this should be the plan for the article. Sorry for rambling. I had no idea how to write this clearly before, perhaps because I didn't understand it well enough.Likebox (talk) 22:09, 9 March 2009 (UTC)

conceptual inconsistency

Inconsistency of string theory follows from assumption of Lorentz invariance and hidden dimensions concept, because the existence of hidden dimensions should manifest itself just by Lorentz invariance violation in analogy to wave spreading at 2D water surface (the reference frame of underwater is just a result of surface wave dispersion into third dimension). This leads into landscape of infinitelly many solution and lost of testability (http://aetherwavetheory.blogspot.com/2009/02/consistence-problem-of-string-theory.html) —Preceding unsigned comment added by 194.213.42.170 (talk) 16:29, 3 March 2009 (UTC)

History Section

Normally, I would be totally fine with splitting the history off. But in this case, look at all those red links. Each of those links represents a person who got too little credit for their work. This work was just too hard for nearly all physicists in the 70s to follow, because it relied on S-matrix theory and Regge theory, which were high falutin inventions. In the eighties, it was revived, but in the path-integral form due to Polyakov. Again, the S-matrix people who actually developed the theory were left out.

Because this is the great lasting contribution of S-matrix theory, and because S-matrix was laughed at by most physicists (even string theorists!) for the last 40 years, I think it's only fair to leave the history here, because many of the neglected people are mostly still alive and deserve a little recognition for founding this field.Likebox (talk) 03:29, 12 April 2009 (UTC)

String Theory Makes No Predictions

String theory is known to make predictions only at an energy scale which is likely to be too high to be directly probed by experiments. This is not contested by knowledgable critics, but there are people who say "string theory makes no predictions", leaving it ambiguous whether they mean in principle or in practice. The statement "string theory makes no predictions in principle" is wrong. The statement "string theory currently makes no predictions in practice" is sort of right. The purpose of the section is to make the distinction clear.Likebox (talk) 17:01, 15 April 2009 (UTC)

GA Reassessment

This discussion is transcluded from Talk:String theory/GA1. The edit link for this section can be used to add comments to the reassessment.

GA Sweeps: Delisted

As part of the WikiProject Good Articles, we're doing sweeps to go over all of the current GAs and see if they still meet the GA criteria. I believe the article currently has multiple issues that need to be addressed, and as a result, I have delisted the article. Although several references are listed, there are no inline citations for a large portion of the article's content which is required under the criteria. Add additional citations from a variety of sources to provide a balanced representation of the information present. Perhaps sources can be pulled from the main articles linked to within the article. Look to books, magazines, newspaper articles, other websites, etc. The lead also needs to be reduced to four paragraphs (look to merge the paragraphs and take out some of the information) to better summarize the article. See WP:LEAD for guidelines. There are also multiple grammar issues throughout the article as well as a popular culture section that needs to be converted to prose (and cut out the non-notable occurrences). Although the article has been delisted, the article can be return to GA status by addressing the above points. Once sources are added and cleanup is done, I recommend renominating the article at WP:GAN or you can contact me on my talk page and I'll review it for you so you can bypass the month-long backlog. If you need assistance with any of these issues, please contact me on my talk page and I'll do my best to help you out. --Happy editing! Nehrams2020 (talk • contrib) 06:51, 9 June 2009 (UTC)

String theory is BS

Regarding Mintrick's request for more "string theory is BS" opinions, "Problems and Controversy" used to start out with this:

Following the appearance of two books claiming string theory has failed, a hot media debate evolved in 2007.

"For more than a generation, physicists have been chasing a will-o’-the-wisp called string theory. The beginning of this chase marked the end of what had been three-quarters of a century of progress. Dozens of string-theory conferences have been held, hundreds of new Ph.D.s have been minted, and thousands of papers have been written. Yet, for all this activity, not a single new testable prediction has been made, not a single theoretical puzzle has been solved. In fact, there is no theory so far—just a set of hunches and calculations suggesting that a theory might exist. And, even if it does, this theory will come in such a bewildering number of versions that it will be of no practical use: a Theory of Nothing." -- Jim Holt.

And then went on to say more or less what it says now: the substantive criticisms. This was deleted, I think because the quote was somewhat polemical and devoid of scientific content. But the opinion is definitely out there in the public mind, and probably should be represented somehow.

Keep in mind, though, that the mathematics and physics of string theory is both extremely beautiful, very subtle, and the product of many brilliant underrecognized physicists, some of whom lost their jobs in the 1970s in a great purge. This tragic history makes string theorists touchy about criticism. But I think the section can be balanced to your liking by reintroducing the quote.Likebox (talk) 21:25, 23 June 2009 (UTC)

I think that as long as we establish that the debate is not a scientific one, we should be OK. We can take heavy guidance from global warming and related articles. Mintrick (talk) 03:45, 24 June 2009 (UTC)

There's a differece from global warming, though. The consensus is that string theory is possible, meaning that it makes sense, not that it is true of our universe. Whether it is true for our universe, there's no consensus. I tend to believe it is, others tend to believe it isn't. The consensus for global warming, however, is that not only is it possible, but it is also happening. So that's a difference.

I reinserted the discussion. Thanks for bringing that up. It did bother me when all the criticisms were deleted (although I don't agree with them, I think they are interesting to preserve).Likebox (talk) 03:53, 24 June 2009 (UTC)

Re: global warming, I was speaking merely in terms of how one can cover the public debate, which is distinct from the scientific consensus. Global warming, as a FA, apparently does a good job of combining the two. Mintrick (talk) 03:59, 24 June 2009 (UTC)