One of the big experimental HEP conferences, the Lepton-Photon Symposium, has just ended and many of the talks are on-line. This is the 22nd of these conferences which happen every two years. New data from the Tevatron about the top quark was discussed, and a paper with the new top quark mass results has been released.

There’s a new web-site with news about the LHC.

Last month there was an Einstein Symposium in Alexandria, and presentations are on-line. They’re for the general public so pretty content-free, but it is interesting to see what Witten’s latest view of string theory is: “I’d like to believe — but of course I don’t know — that string theory is on the right track…” Michio Kaku begins his presentation with advertisements for his books, then tells the audience that testing string theory would require creating a “baby universe”, that “Mind of God = music resonating through 11 dimensional hyperspace” and that the standard model is “supremely ugly” (which strikes me as something supremely stupid to say).

If you can’t wait for next week’s Strings 2005 in Toronto, there’s a summer school on strings going on at Perimeter, and a meeting in Crete that just ended, along with many more string conferences to come. The one series of talks I won’t be able to make it to, but would love to hear would be Graeme Segal’s talks in Oporto on 2d QFT.

The DOE has just announced the award of seven new Outstanding Junior Investigator grants in high-energy theory and experiment, one of which is going to Lubos Motl.

OK – so to answer Arun’s original question [is there a purely Quantum Field Theoretical treatment of the hydrogen atom], the answer is “No”,

Not to be combative, but I pointed you to one, effective field theory methods. Indeed this is the *sane* QFT treatment, since you have things in the problem you can treat exactly.

Put another way, it doesn’t make much sense to restrict your “pure” QFT treatment of the proton to

S = \bar\psi (Dslash – m) \psi

as your action. You know you’re in the non-relativistic limit, it makes sense to use that fact to the maximum.

OK – so to answer Arun’s original question [is there a purely Quantum Field Theoretical treatment of the hydrogen atom], the answer is “No”, you’re better off with your undergraduate text books.

There’s very good evidence that asymptotically free theories like QCD are non-trivial in the continuum limit. If you ignore fermions, the lattice Monte-Carlo calculations work quite well. There’s a mass gap and the extrapolation ot zero lattice spacing and infinite volume look fine. Non-asymptotically free theories like QED are the ones that appear to have a problem in the contiuum limit, since they are becoming strongly coupled at the cutoff scale.

The notion that any RQFT becomes free in the limit of lattice spacing going to zero ties in with other results, Haag’s theorem in particular. People seem to be prepared to do the most amazing things, such as abandoning special relativity, to avoid accepting this.

You can certainly formulate QED on a lattice, and even do perturbative calculations (although they’re a lot harder since your regularization is not Lorentz invariant). The lattice technique that doesn’t work well for QED is that of Monte-Carlo computer calculations on a finite lattice. These work best in theories with a mass gap, since finite-size effects will go away for lattices much bigger than the inverse of this mass.

As far as I know, no one has really understood what happens non-perturbatively to QED, whether regularized by a lattice or by any other method. By analogy with phi^4 theory, the fact that it’s not asymptotically free suggests that no matter how you try and take the continuum limit, you’ll end up with a non-interacting theory. Whether this or something more interesting happens is not known as far as I’m aware, but I haven’t followed work in this area, maybe more is known.

Is this not just saying that lattice techniques simply do not work for QED?

No, it’s saying you have to be careful.

… at finite lattice spacing QED confines just like QCD. It has been demonstrated (numerically at least, I don’t think it’s been proven) that QED has a phase transition as the spacing goes to zero, which takes it to a non-confining phase.Is this not just saying that lattice techniques simply do not work for QED?

It may be worth listing out the highest-precision tests of each sector of the Standard Model and of General Relativity.

QED — electron g-2

Electroweak — LEP precision data

QCD — Running of \alpha_s or evolution of structure functions

GR — Hulse Taylor Binary pulsar timing data

For the electroweak theory, I’m not really sure there’s a single “showstopper” number. But the LEP data as a whole is impressive confirmation of the standard model.

For QCD, you can measure \alpha_s in a number of different experiments, and run each value to a common scale. The fact that they all agree when run to this scale is strong evidence for QCD.

There’s also the evolution of structure functions in deep inelastic scattering (i.e. deviations from Bjorken scaling). This is impressive evidence, but I’m not sure if it relies on any “non-QCD” input.

In regard to the the first two: no doubt, but this is not actually a quantum field theoretical treatment of bound states. It is more like saying, “If we could do bound states in QFT, then we would expect the following”.

It’s an effective field theory analysis. It’s more rigorous than just guesswork, it’s a systematic expansion. These are used all over modern particle physics.

It’s as field theoretical as using a Schwinger-Dyson or Bethe-Salpeter type approach, just different.

In regard to lattices, is it actually feasible to do lattice gauge theory for QED in the same way as one does it for QCD (& I am talking about QFT rather than QM) – ?

Yes, but there are some subtlties peculier to QED. The major one is that at finite lattice spacing QED confines just like QCD. It has been demonstrated (numercially at least, I don’t think it’s been proven) that QED has a phase transition as the spacing goes to zero, which takes it to a non-confining phase.

From a practical standpoint, I’m not sure how hard this makes it to do QED on a lattice. People do do it though.

IIRC there was some interest in 2+1 dimensional QED in the condensed matter community.

It may be worth listing out the highest-precision tests of each sector of the Standard Model and of General Relativity.

Hi Matthew,

In regard to the the first two: no doubt, but this is not actually a quantum field theoretical treatment of bound states. It is more like saying, “If we could do bound states in QFT, then we would expect the following”.

In regard to lattices, is it actually feasible to do lattice gauge theory for QED in the same way as one does it for QCD (& I am talking about QFT rather than QM) – ?

The usual answer is that one uses the Bethe-Salpeter equation to do bound states in QFT, but one has to ignore the infinities detonating all over the place, an inconvenience not present in the first-quantised treatment of the one-electron atom.

You can also use a non-relativistic effective theory (NRQED) properly matched to relativistic QED. The Lamb shift has been done this way.

Also, no-one has ever satisfactorily answered my question as to how a technique that only covers scattering processes can be used for bound states anyway.

There are a few techniques. The effective field theory analysis is probably the clearest. For an example of what you can do with this approach there’s http://arxiv.org/abs/hep-ph/0003277.

Then of course there is Lattice Gauge Theory. The last time I enquired – relatively recently – I was told that no-one has used the technique to calculate the Hydrogen atom. I am surprised at this as I would have thought it would provide a useful check, especially as they obviously do not balk at the much harder problem of q-qbar bound states.

It’s not a terribly useful check since lattice QCD is much more complicated than lattice QED. You can do the H atom using (numerical) path integrals in Quantum mechanics though. That’s a fun undergraduate level excercise. However, lattice field theory is not a precision tool. You could easily do the H atom, and get 5/10% accurate results. To get a ppm/b determination of the lamb shift would be very hard.

Some of these checks are to ten digit accuracyWell, as I have remarked in other places, by using quenched QED and some experimental data an efriend and myself got to fit -or to fake- six or seven of these same digits in hep-ph/0503104. I guess the real value of QFT is about the generality of its application, and not about a particular, experiment driven, quantity.

James, the problem with QCD is essentially asymptotic freedom; as momenta grow smaller and distances larger, the coouplings between quarks grow *stronger*. So you can use the well-developed techniques of perturbation theory to compute the outcomes of processes where quarks are banged into each other at very high energy, but the “perturbations” grow too large to handle that way when you try to compute the behaviour of quarks assembling into stable, bound states involving small average momentum transfers. (And as if that weren’t bad enough, existing techniques for handling bound states in QFT are quite unsatisfactory; I have little doubt that good progress on that front would open up whole new frontiers for detailed parton model building – now there’s something *useful* for all those talented mathematica physicists now slaving away at strings to look at!)

Not being able to use perturbation theory means having to solve a much, much harder problem; in practice you can only do it numerically, with lattice techniques, and then you run into all sorts of technical problems, starting from hardware limitations (a 4-dimensional lattice eats up memory and CPU time very quickly as you increase its size – double the linear size and you get 2^4=16 more points – and ideally you want to be able to take the limit of zero lattice spacing…).

Pentaquarks are such a non-perturbative, and therefore computationally (but maybe not conceptually, depending on how you want to draw the line between the two) hard problem. The 10-digit-precision claims which you quote on the other hand are from realms amenable to perturbation theory.

Peter Woit wrote, in response to my comment, “It’s just bizarre to talk about “how unpredictive the standard model” is.” Once again Peter, thank you for your response. Well Peter, you know this stuff and I don’t, so I’m just going to have to take our word for it, or spend years studying it. And I have heard and read that ten digit accuracy claim many times. That’s why the inability to predict or even agree on the interpretation or even existence of whole classes of particles at this late date seems so bizarre to me. Once again thank you.

Jim Graber

It’s just bizarre to talk about “how unpredictive the standard model” is. It makes an infinity of predictions, and every single one that experimentalists have been able to check comes out correctly within experimental errors. Some of these checks are to ten digit accuracy. It’s probably the most predictive of all scientific theories known to man. Every particle physics experiment done during the last 30 years has generated reams of data exactly predicted by the standard model. Pick any experimental high energy physics conference, look at the data reported and its precise agreement with standard model predictions.

You seem to not understand the difference between not being able to predict everything (standard model) and not being able to predict anything (string theory). There are some things the standard model inherently can’t predict, like fermion mass matrices, others it can predict in principle, but our calculational methods are not good enough to extract the answer. If the standard model predicted everything and we knew how to extract every prediction, high energy physics would just close up shop.

If you want to understand which things the standard model inherently can predict and which it can’t, that’s not very hard, just learn exactly what the theory is. If you want to understand which things are easy to calculate and which are hard, you have to spend some time really understanding what the known calculational methods are in QFT, and what are their limitations. This takes some serious work.

But it’s just completely absurd to claim that there’s any similarity between string theory and the standard model from the point of view of predictivity.

Peter,

Thank you for your response to my comment. I was not trying to imply that neutrinos were directly related to pentaquarks. My main point is the surprising (to me, at least) lack of predictivity of the standard model. Not being an expert on either the standard model or on string theory, I am continually surprised at how unpredictive both of them are. I will leave criticism of string theory to you and others. I realize that the standard model is not “not even wrong” because with solid evidence of neutrino oscillations, and thus indirect but solid evidence for neutrino masses, “everyone” agreed that the standard model needed to be changed and expanded.

(By the way, I was somewhat astounded at how long acceptance of this change took, but that’s a different issue. I am also amazed at the apparent preference for Majorana over Dirac masses. No majorana particles are yet known, so why should neutrinos be Majorana? Of course, it would be cool if they were, but it would seem like Dirac is a much more likely bet.)

And also, people think that by accurately measuring the top quark mass, they can partly predict the Higgs mass. But on the other hand, the pentaquark or the tetraquark can exist or not and the standard model seems fine either way. This seems very unpredictive to me. I have wondered about this for some time. My comment was triggered not just by the recent press release concerning Jlab’s failure to confirm the pentaquark, but also by hep-ph/0507025 which I just happened to read. In addition to your response, another anonymous poster also suggested the reason for this lack of predictivity was the difficulty of solving QCD. I can accept that, but I still wonder why some predictions are possible and others are not. It makes the standard model seem much shakier than it is usually presented as being.

Jim Graber

Chris,

Thanks! Do you think we can count this as one of the “mass of unsolved problems” that physicists have left behind “in a pursuit of an

a priorivision of what a simple world would look like” ?-Arun

Hi Arun,

The usual answer is that one uses the Bethe-Salpeter equation to do bound states in QFT, but one has to ignore the infinities detonating all over the place, an inconvenience not present in the first-quantised treatment of the one-electron atom. Also, no-one has ever satisfactorily answered my question as to how a technique that only covers scattering processes can be used for bound states anyway. Then of course there is Lattice Gauge Theory. The last time I enquired – relatively recently – I was told that no-one has used the technique to calculate the Hydrogen atom. I am surprised at this as I would have thought it would provide a useful check, especially as they obviously do not balk at the much harder problem of q-qbar bound states.

Question born of ignorance: is there a purely Quantum Field Theoretical treatment of the hydrogen atom?

Neutrinos have nothing to do with pentaquarks. For no obvious reason you’re conflating two of the unsatisfactory aspects of the standard model: the fact that it doesn’t predict fermion mass matrices, and the fact that one doesn’t know how to exactly solve QCD. If one could exactly solve QCD in the infrared, this should tell you whether or not pentaquark states exist and what their properties are.

Actually this will be a good test of the idea of using string theory to solve QCD. If anyone ever comes up with a workable string theory dual to QCD, it should predict what happens with pentaquarks.

Jim Graber wrote:

Why can’t the theoreticians even agree on a prediction?

Because non-perturbative QCD is pretty damn hard to solve? (And no, while I’m certainly no expert, I very much doubt that neutrinos, whether massive or not, can be relevant to pentaquarks.)

Perhaps this “Various and Sundry” thread is a good place to enter this question, which I have been wanting to ask for some time.

The pentaquark has been searched for for over twenty years. Recently, the pentaquark seems to have reappeared and then disappeared, but other weird particles, or particles with weird decay modes are still being seen. The arxiv is full of conflicting explanations for these particles, mostly based on the standard model. Of course, the old standard model, 321Z with _Zero neutrino masses has just been replaced with two competing new standard models, 321M with _Majorana neutrino masses and 321D with _Dirac neutrino masses. So does any of these models predict the pentaquark or not? It seems that the standard model of particle physics (SMOPP or just SM) is almost as non predictive as string theory (ST) at least as far as pentaquarks are concerned. Or maybe it does predict the pentaquark (albeit broader than the recent mirage) and is just plain wrong.

The experimentalists once again seem to be converging on the position that the pentaquark does not exist at any significant level. Why can’t the theoreticians even agree on a prediction? Or is the absence of pentaquarks and tetraquarks some kind of “superselection rule” to be axiomatically included in the SMOPP?

Jim Graber

“To the men and women who create the accelerators, the detectors and the experiments from which the concepts of particle physics spring.”

Ah yes, it’s in “Concepts of Particle Physics” (Volume II, at least), by Kurt Gottfried & Victor F. Weisskopf. Not so recent though (1986). I remember being struck by that dedication, many years ago, as quite paternalizing for all its political correctness. It’s nice to see that I evidently wasn’t the only one.

Joseph Schwartz in

The Creative Moment : How Science made itself alien to Modern Culture. Please, oh please shoot the following down.“In the late Victorian period Planck was an unusual physicist. Physics then was like biology today, with theorists being objects of scorn for their lack of contact with experimental realities. Other physicists from the period who today are known for their theoretical work were skilled experimenters. Gustav Kirchoff (1824-1887) is known now for his theoretical contributions to heat radiation and the analysis of electrical circuits and not for his experimental work with George Bunsen on optical spectra. Maxwell, famous today for his partial differential equations of the electromagnetic field, was professor of

experimentalphysics at Cambridge. H.A. Lorentz, the first theoretical physicist in Holland, was an active experimentalist doing work in optical spectra. Max Born, one of the leading pioneers of quantum theory, was an expert in experimental optics.And Einstein, known as the foremost theoretician of all time, was experienced in laboratory techniques and interested in technology. With the Habicht brothers he patented a precision volt meter in 1914. In the 1920s, in collaboration with aerodynamicist Rudolf Goldschmidt, he invented a hearing aid. With the Dutch firm, N. V. Nederlandsche Technische Handelsmaatschappy he held a patent for a gyrocompass. And with Leo Szilard he patented several refrigerating devices designed to reduce the noise levels of existing commercial machines.

But today the physics community is deeply divided between the theorists and the experimenters. {resulting in a two class system, with theorists at the center and experimentalists circling around them}

Sensitive theoreticians have been careful to soothe ruffled feathers by making it a point to acknowledge the role played by experimental work. A recent textbook concentrating on theoretical developments is dedicated to the experimentalists, “the men and women who create the accelerators, the detectors and the experiments from which the concepts of particle physics spring”.

Although these words are a much-needed acknowledgement of the fundamental source of physics in observation, they nevertheless confirm the existence of inequality. Their unavoidably patronizing connotations are reminiscent of the dedications made by male professionals of all kinds to their wives and secretaries…..

The potentially explosive tensions caused by the inequality between experimental and theoretical physicists are kept within bounds by an exceptionally strong belief in inherited intelligence. Experimental physicists accept second-class status because they feel they are not as bright as theorists. Theorists take their superiority for granted as due recognition of superior intelligence. Physics meetings are famous for their stilted, tense atmospheres as each person is afraid of asking a “stupid” question. Even informal contacts can be dominated by a competitive proving of who understands more of a subject under discussion.

{As a result of the necessarily industrial size undertaking that experimental particle physics has become} “theorists have become an elite within an elite with little or no accountability to an outside audience. There is the occasional theoretical briefing “to the experimentalists”, but in the main, theoreticians have, not unnaturally, become ingrown in their approach to physics.

…Theorists today are so divorced from real experience of nature they have no unconsciously absorbed perceptual knowledge to draw upon. Everything they know has been learned from books. And the understandings based on this secondhand knowledge have inevitably been derivative and second-rate.

—

“Once one believes that spin is a consequence of the Dirac equation, it is only a short step to try to make physics out of the thin air of mathematical guesswork….

{Previously}

“Dirac, perhaps because of his engineering training, was one of the few physicists who remained clear about where things came from. In 1962, Thomas Kuhn…spoke to Dirac about his life and work. Among other things, Kuhn was interested in how Dirac came to write down his equation for a relativistic electron with spin.

Dirac said: “I was playing around with equations and I found that Pauli’s matrices were quite a nice thing to play with. It needed quite an effort to make the further generalization (from 2×2) to 4×4 matrices, but that work did come about from playing about with the three dimensional scalar product.”

Kuhn said: “Was it a surprise that what came out were spin terms?”

Dirac said: “No, I don’t think so. Because one had the Pauli matrices in it [to begin with.” }

…what students of the 1950s were not taught was the understanding of James Franck, one of the leading experimentalists of the 1920s: “What one doesn’t put into the equation will not finally be given by the mathematics”. In its place, students absorbed a theoretical sensibility inspired, not by an attempt to understand, express, and describe, or in Bohr’s word, communicate real physical experience, but by spectacular deductions from a few well-chosen equations….

{resumes} “The patterns of thinking that now dominate theoretical physics approach the classical definition of autistic thinking: thought that is solely determined by the subject’s wishes and fantasies without reference to the environment or to realistic considerations of space and time”.

About strings 2005, any clue about these axions from Witten? Are they still related to a possible two-higgs doublet (remember it is also my bet ?

LM has never had a single idea about physics that amounted to pouring piss from a boot, and our wonderful, wonderful academic structure rewards him this way. THAT is why string theory, a vile and idiotic lie, has managed to slough itself along for 20 years, leaving a shiny trail behind it.

-drl

While I like Chris’s interpretation, I suspect that what Kaku had is mind was that in 1968 what was discovered was not a theory, but a formula for amplitudes. Only later was it understood that these amplitudes come from a (first-quantized) string theory. The continuing hope is that the current (inconsistent and unable to reproduce physics) version of string theory is an approximation to some wonderful but yet to be discovered M-theory.

Does anyone know what Kaku meant by “String theory has been moving backwards”I presume that what he means is that the more they work on it, the more that they realise that it is not going to be of any use as a physical theory.

Hi,

Does anyone know what Kaku meant by “String theory has been moving backwards” in his PowerPoint statement

= String theory has been moving backwards, since it was accidentally discovered in 1968.

?

currently acceptedis not different of the typical forms to request funds in some project, where you are basically asked what are you to discover, and when. I call this part of science, very botanic-wise, the “classification” side. The (also botanic) counterpart, “exploration”, is always more problematic. Smolin article on “New Einstein” was about this, wasn’t it?

I’m tempted to delete the previous comment, but am leaving it since I think that, if accurate, it is interesting to see that the editor of PRL is resorting to an indefensible argument in dealing with nonsense submitted to him (although the “…” may hide a more defensible argument). Please discuss this with the author of this comment on his weblog, not here. I’ll be deleting any further comments about this.

Editor of Physical Review Letters says

Sent: 02/01/03 17:47

Subject: Your_manuscript LZ8276 Cook

MECHANISM OF GRAVITY

Physical Review Letters does not, in general, publish papers on alternatives to currently accepted theories … Yours sincerely, Stanley G. Brown, Editor, Physical Review Letters

Now, why has this nice genuine guy still not published his personally endorsed proof of what is a ‘currently accepted’ prediction for the strength of gravity? Will he ever do so?

‘String theory has the remarkable property of predicting gravity’: false claim by Edward Witten in the April 1996 issue of Physics Today, repudiated by Roger Penrose on page 896 of his book Road to Reality, 1994: ‘in addition to the dimensionality issue, the string theory approach is (so far, in almost all respects) restricted to being merely a perturbation theory’. String theory does not predict for the strength constant of gravity, G! However, the Physical Review Letters editor still ‘believes in’ Edward Witten and Physics Today.

http://einstein157.tripod.com/

There was a short presentation on E Witten on CNN recently. And no, it’s not the one that’s really on Jim Simons. Seiberg, Nappi and Kaku were interviewed and asked what they think of Witten. Witten’s son was also interviewed. I think it was nice.

I can not but agree with DOE about labeling Lubos as one of the Outstanding Junior Investigators in the USA research network. I can not tell if I am happy about it. In any case, at least Lubos is worried about the spectrum.

Kaku, slide 45:

The Unfinished Theory= String theory has a new Picture: strings and membranes.

= String theory lacks a physical principle, like the equivalence principle.

= The full symmetry and mathematics of string theory are also unknown.

= String theory has been moving backwards, since it was accidentally discovered in 1968.

I am not sure how these observations lead one to the conclusion that String theory is a good thing to work on. Could someone please explain?

Pingback: Path integrals for gauge boson radiation versus path integrals for real particles, and Weyl’s gauge symmetry principle « Quantum field theory