Panel Discussion Video

Video of the panel discussion at Toronto is now available, so one can hear some of the context of the comments that were reported in the recent New York Times article. As reported, the audience voted 4 or 5 to 1 against the anthropic principle. Unfortunately the camera was not on the panel during the vote, so one can’t tell from this video how the panelists voted.

Some other things that weren’t reported: while Andy Strominger commented that he saw no reason for pessimism, he also said he thought the odds were against any data relevant to quantum gravity or string theory coming out of the LHC. Steve Shenker said that he was very much bothered by the fact that it was starting to look as if one could associate some sort of “quantum gravity” dual to any quantum mechanical system whatsoever, so any notion of uniqueness was completely gone.

There were several skeptical questions from the audience. Someone with a Russian accent pointed out that it was becoming increasingly difficult to argue the case for string theory in the physics community, and asked what argument he should use in its favor. The panel didn’t seem to want to address this, but Shenker finally said “Only consistent theory of quantum gravity”. The next question wasn’t really audible, but had something to do with “it’s been 20 years”. Shenker’s response was something like “most of us don’t want to think about this, we haven’t done as well as in other 20 year periods”. Later on someone asked “Can you imagine any experiment in the next 20 years that will falsify string theory”, getting no real response except “You’re not supposed to be asking that” from Shenker. Another question from the floor was about why none of the panelists had mentioned M-theory, which didn’t get much of an answer except from Nathan Berkovits who commented that in particle theory problems not solved in five years stop being discussed.

In their speculation about the future, many of the panelists invoked the possibility of having to change quantum mechanics. From the floor Witten speculated that quantum mechanics was only valid in asymptotic regions of space time, with something different needed to understand the interior. Also from the floor Susskind speculated that the splittings into different universes of the many-worlds interpretation of quantum mechanics were the same as the cosmological bubbling off of different baby universes. Several panelists responded that they had no idea what he was talking about.

The emphasis on vague ideas about the foundations and interpretation of quantum mechanics led Martin Rocek to point out that there was one field of study in physics that had gone nowhere in the last eighty years: the study of the interpretational issues in quantum mechanics. Lee Smolin rose to the defense of this field, claiming that it had led to recent ideas about quantum computers.

Also now available online are videos of the public talks by Dijkgraaf and Susskind. Susskind tells the audience that there is a “War” or “battle of intellects” going on between two groups of physicists, which he describes as being “like a high-school cafeteria food fight”. The two groups are the “As” (A for anthropic), and the “Es” (E for elegant). He describes the belief by the Es in mathematical elegance as “faith-based science”, and says that they are in “psychological denial” about the existence of the landscape, then goes on to give the standard arguments for the landscape and the anthropic use of it to “explain” the value of the cosmological constant. He refers to belief in the existence of a vacuum selection principle as analogous to belief in the Loch Ness monster. He ended his talk by claiming that the As were winning the war, with the Es in retreat.

Dijkgraaf’s talk was completely standard string evangelism, and except for a couple slides mentioning D-branes and black holes, could easily have been given, completely unchanged, twenty years ago.

This entry was posted in Strings 2XXX, Uncategorized. Bookmark the permalink.

69 Responses to Panel Discussion Video

  1. Kea says:

    “Steve Shenker said that he was very much bothered by the fact that it was starting to look as if one could associate some sort of “quantum gravityâ€? dual to any quantum mechanical system whatsoever…”

    Why should he be bothered? It’s wonderful. That’s the whole point. They just need to give up the idea that String theory is physics and there wouldn’t be any anthropic problem.

  2. Dave Bacon says:

    While it is certainly true that David Deutsch began dreaming of quantum computers as a way to “test the many-world’s interpretation” I’m not sure whether the connection between interpretations and quantum computers is all that tight. After this initial major step (Deutsch is the first to seriously consider the computational advantages of quantum computers over classical computers. His result with Richard Jozsa in 1992 was the result which set the quantum algorithm ball rolling.) there really has not been a huge amount of interaction between those who study interpretation and those who study quantum computation (with notable exceptions, for example, Chris Fuchs and Robert Spekkens)

    It’s very interesting to hear people comment on modifying quantum theory. From a computer science prospective this seems very dangerous. The reason is that many of the modifications of quantum theory which we have been able to dream up, lead to modified quantum computers with pretty astounding computational abilities. Now, it could be that our universe does provide such strong computation, but there are definitely reasons to be very troubled by such a turn. See for example, Scott Aaronson’s article, quant-ph/0502072: “NP-complete Problems and Physical Reality” for interesting thoughts on this matter.

  3. Wolfgang says:

    Just a really naive thought about this issue:
    The wavelength of macroscopic bodies is typically much smaller than the Planck length. One would thus assume that modifications of quantum theory for macroscopic bodies a la Penrose are possible if not likely.

  4. Quantoken says:

    Peter said:

    “In their speculation about the future, many of the panelists invoked the possibility of having to change quantum mechanics. From the floor Witten speculated that quantum mechanics was only valid in asymptotic regions of space time, with something different needed to understand the interior. ”

    I wonder what makes them think so? And if QM is to be changed at certain small scale, what they are going to change it to? A QM with a different value of hbar, or no QM at all but return to classical limit at small scale? I really don’t know what Witten was thinking. QM is all about uncertainty principle. You either HAVE an uncertainty principle, which is QM, or you DON’T have an uncertainty principle, which is classical. It’s a yes or no and there is no third possibility, so there is really nothing to be modified of QM, even at small scales.

    It is also true that QM received far more rigorous experimental tests at microscopic scales than GR does. What needs modification at small scale is GR, not QM. Graviton, as a naive prediction of quantized gravity force, does NOT exist and does not need to exist. When people DO realize that graviton can not exist, then that’s a death sentence for super string theory, since super string theory would have wrongly predicted the existence of graviton, something that just does not exist in nature.

    Quantoken

  5. Robert says:

    From what I can tell from listening to the realaudio is that someone asked “Can you imagine any experiment…, that someone is Jan de Boer and the person with the Russian accent is Sergei Ketov oder Djordje Minic.

  6. Nigel Cook says:

    “… Susskind speculated that the splittings into different universes of the many-worlds interpretation of quantum mechanics were the same as the cosmological bubbling off of different baby universes. Several panelists responded that they had no idea what he was talking about.”

    What a character Susskind is. Feynman, ‘QED’, says wavy electron orbits occur due to path integrals interference in the small atomic space, and in a footnote he says path integrals get rid of the problems with Copenhagen Interpretation. He could have added that it also gets rid of the many-worlds. Path integrals is useful, even if Feynman didn’t derive the Schroedinger equation directly from it. Bohm did that using hidden variables in 1952, using chaotic brownian motion of electrons, caused by wave interference.

  7. Lee Smolin says:

    If, as is true, quantum computers were invented by someone seeking a way to test ideas about the foundations of quantum mechanics, then that is a pretty tight connection. What I see is that quite a few of the people working in quantum information theory (QIT) were and still are motivated by issues in foundations,besides those mentioned these include Anton Zeilinger, Lucien Hardy and many others. And while there are QIT people with no interest in foundations and foundations people with no interest in QIT, I observe quite a healthy interaction among people working in the two fields. But the point I wanted to make to Martin Rocek is much simpler. When we were students together people advised us to go into high energy theory rather than foundations of quantum theory because there would be much more chance for contact with new experimental discoveries. Looking back on the last 30 years, one would have to say that there were more opportunities for a theorist to propose new phenonomena that were confirmed experimentally in foundations than in high energy theory. And if the field turned out to spawn a whole new technology, then supporting the few theorists with the courage to think about foundations will have turned out to have been a good investment.

    And I agree that Scott Aaronson’s paper is well worth reading.

  8. D R Lunsford says:

    Wow, I leave for a week and the Blog is filled with string propaganda. Peter’s become a shill for blighted stringers 🙂

    -drl

  9. Fyodor Uckoff says:

    Somebody should really call Susskind on his deliberate confounding of the existence of a landscape [generally accepted, and no big deal — did anyone criticize Maxwell or Einstein because their theories admit lots of solutions?] and the anthropic principle. The idea that there exists a probabilistic vacuum selection principle just seems like the kind of thing quantum mechanics does for us all the time, and shouldn’t be controversial. The Tye group at Cornell even have concrete ideas about how it might work, though of course their models are very basic so far. [See Tye’s presentation at Strings 2005]

  10. Peter says:

    I think Susskind did address this in his public talk. He referred to the vacuum selection principle as “the Loch Ness monster”, something that lots of people claimed to have seen, but no one actually knows anything about. If you know what the selection principle is, let’s hear about it and see you use it to, like, select a vacuum. The Tye stuff is very far from being able to do anything like this, and I see no good reason to believe it ever will.

    Susskind would say that people like you are just in denial and engaging in wishful thinking. For once, I’d agree with him.

  11. Dave Bacon says:

    Hey Lee! I don’t disagree that there isn’t more talk between the interpretation crowd (whatever that is!) and quantum information science (whatever that is!), but when I think of the main results of quantum information science (Peter Shor’s quantum algorithm for factoring, Lov Grover’s quantum algorithm for searching, the theory of fault-tolerant quantum computation, quantum cryptography, and the main results of quantum information theory (Schumacher quantum data compression, entanglement concentration, etc.)) the results aren’t “directly” related to foundational questions. On the other hand, most of the results and the people who invented them are extremely interested in foundational questions, I think. This is, I think, because a healthy understanding of foundational questions means leads one to a very good understand of what quantum theory is and what it is not. For example, in quantum teleportation, understanding what role a quantum wave function plays seems essential to making the jump to the protocol of transfering that quantum state using entanglement and classical communication. So the main results are unconnected to foundations, but by studying foundations it’s easier to come up with results that cut to the heart of what quantum theory is! (And I say this with the caveat that some of my work has been motivated by interpretation questions, but this lead directly to something which feels very non-interpretational: the communication cost of simulating quantum correlations.)

    Peter, sorry for turning your comments into a quantum computing discussion, but, you know, I just couldn’t resist 😉

  12. Dave Bacon says:

    Oh, and as for my personal views, I’m very sympathetic to the notion that the foundations of quantum theory have something important to say about the problems of quantum gravity. Just what it has to say, however, is something else!

  13. Tony Smith says:

    Peter’s blog entry said “… Witten speculated that quantum mechanics was only valid in asymptotic regions of space time, with something different needed to understand the interior …” and that “… Lee Smolin rose to the defense of … the foundations and interpretation of quantum mechanics … claiming that it had led to recent ideas about quantum computers. …”.

    Perhaps quantum theory does NOT need to be modified, but is actually a good alternative to conventional superstring theory as a “framework” for a unified physics theory.

    For instance, in http://xxx.lanl.gov/abs/quant-ph/9512022 , Negative entropy and information in quantum mechanics, N. J. Cerf and C. Adami said:
    “… Quantum information theory … allows for negative conditional entropy even though this is forbidden classically. This leads us to propose that such quantum informational processes can be described by diagrams — much like particle physics reactions — involving particles carrying negative (virtual) information. By analogy with anti-particles, we refer to them as anti-qubits. … This analysis suggests the possibility that a qubit (the fundamental quantum of information) could have an analogously defined anti-qubit (a quantum of negative information), formally equivalent to a qubit traveling backwards in time … this leads us to conjecture that these processes can be recast into reactions involving information quanta … described by diagrams, much like particle physics reactions. …”.

    Taking such an information – particle physics correspondence seriously is one of the primary motivations I have had in constructing my Clifford-algebra related physics models. It is a puzzle to me that Clifford-algebra physics models have not become a well-funded school of physics. AFAIK, the only work on Clifford-algebra physics models closely related to information theory (as opposed to just reformulating existing physics models in terms of Clifford/Geometric algebra) has been done by David Finkelstein and his students (I am one of his former students, and that is where I learned my Clifford algebra stuff).
    Maybe there might be interesting socio-political reasons why Lenny Susskind’s anthropic superstring stuff is well-funded with a horde of rabidly devoted followers, while David Finkelstein’s Clifford-algebra stuff is largely ignored (and in the case of some of those who learned from him, such as me, even blacklisted).

    Tony Smith
    http://www.valdostamuseum.org/hamsmith/

  14. rof says:

    One question: Dan Friedan, a man who is, arguably, responsible for the present-day understanding of the relationship between string theory and gravity, says that string theory neither predicts gravity nor spacetime quantum field theory (see hep-th/0204131, which is cited on your home page, Peter). I understand that the relationship between string theory and gravity is “a matter of interpretation”, meaning that the occurrence of something which looks like Einstein’s equations as the condition for the vanishing of the beta function of string theory (prescribing reparametrization scale invariance of the world sheet) , has yet to be universally accepted as a secure indication of a consistent theory of quantum gravity (indeed, the successful reproduction of the Hawking entropy formula would hardly have been considered an additional successful accomplishment, if the “vanishing of the beta function implies general relativity” argument had already been fully accepted as a proof that the theory in question is a coherent theory of quantum gravity). Nevertheless, my question is instead about the claim that string theory reproduces spacetime quantum field theory as a consequence, at low energies. Dan Friedan claims that this is unproved. I have not seen anything which even claims to be a proof of this rather important statement. A senior string theorist (who shall not be named here, unless he cares to step forward and name himself) has told me that a proof was to be found in Green, Shwartz and Witten. I inspected the said tome and instead found a section which claimed that a proof would be too unwieldy to provide in such a slender volume, and that the correct procedure was to assume in advance that there was a QFT which corresponded to the low-energy limit of a string theory, and that merely matching the S-matrices of each sufficed (obviously, duh), to establish that one was the low-energy limit of the other. Now, you’ve read Friedan’s article, and presumably GSW, so what’s your opinion? With all due respect to everybody involved, I have never heard a coherent articulation of both the question and the answer from a single individual, apart from Dan Friedan, who advocates studying his own personal theory instead of string theory.

  15. A comment concerning possible modifications of quantum mechanics mentioned by Witten and therefore belonging to the topics of discussion (as I dare hope).

    At algebraic level the modification could reduce to a choice between different types of von Neumann algebras. Von Neumann algebras allow hyper-finite factors of type II_1 which differ from factors of type III encountered in Poincare invariant relativistic quantum field theories and from type I_n and I_infty factors encountered in non-relativistic quantum theory.

    Hyperfinite Type II_1 factors emerge when you provide infinite-dimensional separable Hilbert space with spinor structure such that unit operator has unit trace. Von Neumann thought that probability interpretation requires this property and regarded factors of type III pathological. Unit trace could be defended also by the finiteness of quantum theory of infinite systems. These algebras emerge naturally when you have an infinite-dimensional configuration space with Kaehler metric and spinor structure (the space of 3-D surfaces in certain 8-D imbedding space, the “world of classical worlds”, in the case of TGD).

    The mathematics of inclusions of hyper-finite factors has a close relationship to that of conformal field theories, quantum groups, braid groups, knot and 3-manifold invariants, etc… The sequences of Jones inclusions of these algebras have interpretation in terms of sub-system-system inclusions. A model for topological quantum computation led to the idea about the quantization of hbar in terms of Beraha numbers so that hbar would become a characterizer of Jones inclusion and would be dynamical and quantized. Large hbar phase would be of obvious interest from the point of view of quantum computation.

    A generalization of braid diagrams to Feynman diagrams suggests strongly itself together with a symmetry principle generalizing string model duality. Diagrams would be classified by the topology of the lowest genus two-surface allowing the imbedding of diagram and all diagrams with homologically trivial loops at this genus are equivalent to a minimal diagram characterized by its homology class for the miminal genus. The absence of homologically non-trivial loops has in TGD framework straightforward interpretation: there is no path integral over all possible 4-surfacace since configuration space geometry assigns an almost unique space-time surface to a given 3-surface identifiable as a generalized Bohr orbit. Almost uniqueness means the failure of strict classical determinism: this makes it possible to assign space-time correlates not only to quantum states but also quantum jump sequences.

    I have been working out the consequences of this picture for more than year year with a particular emphasis on macroscopic and macrotemporal quantum coherence and understanding of dark matter as large hbar phase. See ‘What’s New’ links of various books about TGD at http://www.physics.helsinki.fi/~matpitka/ and my blog site http://matpitka.blogspot.com.

    Matti Pitkanen

  16. Fyodor Uckoff says:

    If you know what the selection principle is, let’s hear about it and see you use it to, like, select a vacuum. The Tye stuff is very far from being able to do anything like this, and I see no good reason to believe it ever will.

    Well, if you actually bother to read their papers, they do argue that their version of Hartle-Hawking selects the KKLMMT vacuum. Of course they are open about the fact that their methods are very primitive, but their results are sure a hell of a lot more impressive than Susskind’s….which are what, exactly?

    Susskind would say that people like you are just in denial and engaging in wishful thinking. For once, I’d agree with him.

    So now you find yourself in eager accord with Lubos Motl and Susskind. If that isn’t a warning sign, then what is?

  17. Thomas Larsson says:

    Vacuum selection is only a problem if you accept the premise that string theory has anything to do with quantum gravity. If the right problem rather is to quantize gravity coupled to the standard model in 4D, as experiments indicate, vacuum selection is not an issue anymore. But you have to do it right, of course, and not ignore anomalies, which we know exist already in 2D gravity.

    Besides, I observe that mr FUckoff is either too incompetent to publish even a single paper on the ArXiv, or too coward to use his real name. Either way, I don’t understand why anybody should care about his comments.

  18. Alejandro Rivero says:

    I think we should praise Toronto Committee, they started very slow, but they have made a very good work on availability of the talks.

    As for these “20 years periods”… can anyone inform me what kind of milestom do they use? I count between 30 and 35 years since last batch of significant theoretical events.

  19. Alejandro Rivero says:

    (milestone)

  20. Peter says:

    Alejandro,

    Everyone is using “20 years” to refer to the fact that it was the fall of 1984 – early 1985 when the huge amount of effort going into the study of string theory started. The last really big, successful new piece of our understanding of particle physics was probably the discovery of asymptotic freedom, over 32 years ago.

    Fyodor,

    In this war I agree with both sides. The As aren’t doing science and the Es are in denial about what has happened. There’s an obvious point of view under which they are both right….

    Matti (or anyone else),

    Please do not post here attempts to start off-topic discussions about things like your favorite alternatives to quantum mechanics. In this case, if you inside information about what Witten had in mind, it would be interesting to hear it, but otherwise please don’t do this.

  21. Thomas Larsson says:

    From Peter’s post I in fact got the impression that Witten was contemplating giving up QM, perhaps replacing it by something like ‘t Hooftian Planck-scale determinism. Instead, he pointed out the rather obvious fact that if your quantum theory is defined in terms of asymptotic data, things become murky if your spacetime does not allow for the right kind of asymptotia. He is probabably worrying about the positive CC, which leads to de Sitter spacetime, which indeed has precisely this kind of problems. Witten made this point in hep-th/0106109, and I doubt that anything really has changed since then.

  22. On my blog, http://motls.blogspot.com/ , you may find a discussion of the achievements of the interpretation of quantum mechanics research, and about uniqueness of quantum gravity and its compatibility with holography.

  23. Ben says:

    Please pardon my ignorance as a non-physicist reader of this blog, but WRT Shenker’s comments that one can associate a quantum gravity model with any quantum field theory, should one take out of this that: (a.) there is no unique quantum gravity theory, (b.) one just needs to find the right field theory, or (c.) there is a unique theory of quantum gravity, but it won’t be coming from string theory? I’m a bit confused…

  24. Peter says:

    Hi Ben,

    You’re not the only one who is confused about this, I think everyone is, thus Shenker’s worries. See Lubos’s blog posting that he mentioned for some comments about this, although I don’t think they’ll really answer your question.

    The problem right now for the string theory program is that their best argument, that strings give a quantum gravity theory, suffers from the embarassment that, if correct, it seems to give an infinite number of quantum gravity theories. I think the standard hope from 20 years ago that string theory will lead to a unique TOE has now become essentially untenable (although Lubos does his best to come up with a scenario to rescue things). People have reacted to this in two ways

    1. The anthropic/landscape scenario: there’s an infinite number of possibilities out there, all we can do is study them all and see which ones can support life, then try and get some prediction of something out of that.

    2. Deciding that any attempt to connect string theory to a TOE is just premature, arguing that the existence of these string duals to interesting QFTs is one thing that makes string theory very much worth continuting to invest time in. Maybe once one learns more about string theory a path to a TOE will become clear.

  25. Lee Smolin says:

    Regarding quantum gravity and the foundations of quantum mechanics: it can only be good news if Witten, Lobus and other string theoirsts are finally coming to terms with the possibility that the problems of extending quantum theory to cosmology may force us to revise the foundations of quantum theory. This is a welcome development, as those in the quantum gravity world have been thinking about this for decadees and have already published several concrete proposals for how quantum theory may be modified to include cosmology. These include 1) ‘t Hooft, in his original formulation of the holographic principle and subsequent work. 2) Penrose’s proposals for non-linear modifications of the Schrodinger equation motivated by quantum gravity, which by the way leads to real experiments. 3) Gell-Mann, Hartle, Butterfield ,Isham and other attempts to formulate a generalized quantum theory for quantum cosmology. 4) The proposal of relational quantum theory, by Crane, which in some ways anticipated the holographic principle, developed by Rovelli and others. 5) Quantum causal historeis, proposed by Markopoulou as an alternative formulation of quantum cosmology. 6) a hidden variable theory inspired by LQG, published by Markopoulou and myself, 7) reinterpretations of matarix models, including the BFSS model, as hidden variables theories, by Adler, Starodubtsev and myself and 8) proposals by Dowker, Sorkin et al related to causal sets.

    Perhaps if string theorist get involved in this question they can do better; that would be very welcome.

    So we don’t have to debate how useful foundations research has been, so long as we agree that the problem of quantum cosmology forces us to revisit the issue. I would only suggest that anyone wishing to think about this problem woud do well to study the literature, as in any field of sciience, to learn about the good ideas already under development and to prevent repeating mistakes already made.

    Lee

    ps references as usual in my “Invitation…”

  26. Chris W. says:

    A reminder: Stephen Adler presented his ideas on an alternative foundation for quantum mechanics and quantum field theory in a book published about a year ago. A draft appeared on arXiv.org in 2002.

  27. Pingback: The Quantum Pontiff » Foundations of Quantum Theory, Quantum Gravity, and Quantum Computing

  28. rrtucci says:

    In my opinion,

    (1)There is a good chance that Quantum Computing/Quantum Information (QC/QI)
    can shed some important insights into Quantum Gravity

    (2)QC/QI can be practiced in an interpretation neutral (i.e., shut up and calculate) way, following the standard rules of QM (i.e., no tinkering with its foundations). 99% of the QC/QI papers in arxiv are like that. So, it’s misleading to imply at this point in history a big overlap between (QC/QI) and (Foundations of QM or Interpretations of QM)

    (3)QC/QI can and will be tested in the lab MUCH sooner than String theory. Furthermore, unlike String Theory, QC/QI promises to yield useful devices.

  29. Aaron Bergman says:

    I completely agree with #2 and #3. But, especially #2. #1, who knows?

  30. Peter says:

    rof,

    I realized I didn’t respond to your question. I’ll leave that one to the string theory experts, admitting that I’m not intimately familiar with every argument in Green-Schwarz-Witten. This isn’t something I’ve ever thought much about, largely because the claim that at low energies string theory reduces to a QFT seems quite plausible. The time I’ve spent learning about string theory has been more devoted to trying to understand the claims about the theory that seem to me implausible, not the plausible ones.

  31. Regarding the question of whether quantum foundations has gone anywhere in the past 80 years: I think one needs to distinguish carefully between “results-oriented” and “non-results-oriented” research. The “results-oriented” side of quantum foundations typically involves either experiments or nontrivial theorems (even if the theorems seem trivial in retrospect — like Bell’s Theorem, the Kochen-Specker Theorem, or the dense quantum coding theorem). The “non-results-oriented” side typically involves assertions that a particular stance toward the double-slit experiment is the correct one, and that if other people fail to see that, then it must be because they’re too dense. I would characterize the results-oriented side as having made phenomenal progress (especially from the interaction with quantum information over the last 15 years), and the non-results-oriented side as having made zero progress (except when it ‘accidentally’ stimulates a result, similarly to how Mach stimulated Einstein’s work on GR). I know that sounds like a tautology, but I think it’s a useful tautology for choosing quantum foundations problems!

  32. rof says:

    Thanks, Peter. It would be good if a string expert can shed some light on the question. It seemed almost trivial to me as well that string theory would produce something like QFT at large distances until I read the article and thought about it a little harder. To say that it will look like QFT at long distances is really to extrapolate from short-distance physics to long-distance physics.

    On the subject of revising the foundations of quantum mechanics, there are two senses in which one can read the word “revise”. The writings of Heisenberg and Von Neumann are still among the clearest expositions of what quantum mechanics is all about. Somewhere over the last eighty years, a shift occurred in the attitude of physicists to the role of the observer. It’s now considered common sense that the observer should be treated as just another part of the system, but for the founders of quantum mechanics, the observer played a crucial role.

    Von Neumann said, in “Mathematical Principles of Quantum Theory” that one must always split the world into observer and observed, or else one proceeds vacuously, since the purpose of quantum mechanics is to provide relationships between the results of measurements, and without an observer, there are no measurements to relate to one another, and all that is left is empty formalism.

    One important development in the foundations of quantum mechanics over the last eighty years has been the almost universal rejection of Von Neumann’s (and Heisenberg’s and Bohr’s) understanding of what quantum mechanics is fundamentally about. It didn’t come about because of a research program, though, but rather because of the gut feeling that every physicist has that the observer should play no special role in the theory.

  33. rof says:

    One other point: The purpose of examining the foundations of any subject isn’t to produce new results and new technologies, but is rather to develop a clear understanding of what the subject is and why it is the way that it is.

    If the foundations of quantum mechanics are dismissed as unworthy of study because studies of them haven’t produced any new results lately, then this is like saying that elementary calculus shouldn’t be studied because it hasn’t produced any new results. The measure of value is inappropriate.

    It might be argued that physicists have done very well over the last eighty years without knowing much about the foundations of quantum mechanics, so why start thinking about them now? Lee provides the answer in his post above: when it comes to cosmology, there can’t be an outside observer, and so the present formulation of quantum mechanics is incompatible with the idea of taking the whole universe as a single quantum system.

  34. Nigel Cook says:

    Lee’s seven promising options include one of causality and two of hidden variables. It would be nice to have a convergence toward consensus, which is vital to avoid a disintegration of the research-education infrastructure. Without consensus, it’s very hard to teach with interest at lower levels, because it looks a bit like speculative gambling or the disintegration of central ideas. It would be nice to see some things unified carefully, like path integrals and some hidden variables. This is not regression to determinism, because you still have uncertainty, you just have a cause for it like chaos due to wave interference. ‘Caloric’, fluid heat theory, eventually gave way to two separate mechanisms, kinetic theory and radiation. This was after Prevost in 1792 suggested constant temperature is a dynamic system, with emission in equilibrium with the reception of energy. The electromagnetic field energy exchange process is not treated with causal mechanism in current QFT, perhaps if it was it would turn out to be the missing hidden variable needed.

  35. Kea says:

    “From the floor Witten speculated that quantum mechanics was only valid in asymptotic regions of space time, with something different needed to understand the interior”

    I don’t understand why people find this surprising. Asymptotic regions are, after all, what we are used to. Of course we want QM to hold in this domain, but we should also expect it to be extended (or ‘altered’ if you like) in a theory of quantum gravity. I don’t think Witten meant anything complicated. Maybe someone could ask him.

  36. Dave Bacon says:

    Kea: Not surprising. Just (1) hard to do, and (2) not the main path advocated by the majority of theoretical physicists.

  37. Chris W. says:

    Nigel,

    Lee Smolin referred to quantum causal histories and causal sets. He didn’t use the loose term “causality”.

  38. Not a Nobel Laureate says:

    This talk of the “need” to modify QM reminds me of the one-upon-a-time talk of the “need” to modify QFT to describe the Strong Nuclear force. Before the discovery of quarks/partons and the development of QCD and asymptotic freedom.

    Will String Theory suffer the same fate as Regge Pole Theory?

    It would certainly be interesting if someone could come up an alternative to QM that would survive the battery of possible experiments.

    I’ll admit to being rather “old-school”. I still think that physics is an experimental science.

  39. garrett says:

    Hey Peter,
    Have you considered starting a group weblog for physics phd’s who don’t like string theory and are actively pursuing alternatives?
    -Garrett

  40. Tony Smith says:

    Not a Nobel Laureate commented:
    “… This talk of the “needâ€? to modify QM reminds me of the one-upon-a-time talk of the “needâ€? to modify QFT to describe the Strong Nuclear force. Before the discovery of quarks/partons and the development of QCD and asymptotic freedom.
    Will String Theory suffer the same fate as Regge Pole Theory? …”

    As I have commented in another thread on Peter’s blog:
    “… David Gross said, in Chapter 11 of the book The Rise of the Standard Model – Particle Physics in the 1960s and 1970s, edited by Hoddeson, Brown, Riordan, and Dresden (Cambridge 1997): “… The bootstrap idea was immmensely popular in the early 1960s … it rested on the solid principles of causality and unitarity … It promised to … provide a unique value for all observables … This is of course false. We now know that there are an infinite number of consistent S-matrices that satisfy all the sacred principles. …â€?. …”.

    It seems to me that the 20 years of Witten’s superstrings (1984-now) is similar to the 20 years of Chew’s Bootstrap (1950s-mid1970s).

    Chew’s Bootstrap fad did not end until the Standard Model emerged in the mid-1970s.

    I doubt that Witten’s superstrings fad will end unless or until an alternative model that is as connected to experiment as the Standard Model is recognized and accepted. Barring that (unlikely in view of the blogs and comments over the past few weeks) event, I expect that bureaucratic inertia will be sufficient for conventional superstring theory to maintain its dominant position for the forseeable future (at least through Lubos’s retirement). From my point of view, that amounts to a New Dark Age of physics.

    Tony Smith
    http://www.valdostamuseum.org/hamsmith/

  41. Tony,

    I am again and again astonished by the extreme orthodoxy of both string theorists and also many of those who do not like string theory. To me it looks utterly inconceivable that a theory created during first five centuries of physics and containing obvious logical paradoxes (mention only quantum measurement problem) could be final and that the only challenges would relate to purely technical problems of M-theory as Lubos argued in complete seriousness in his blog.

    What also surprises me is the extremely hostile attitude towards new well-formulated ideas and theories making precise predictions when the authority provided by a formal academic position is lacking. It seems that all that we are allowed to do is to try to interpret casual remarks made by Witten or some other name. This extremely authoritarian atmosphere really brings in my mind dark middle age.

    Quite concretely, for some time ago Science reported a statistical survey which suggests that rate of technological discoveries per capita was at maximum at thirties and has now reached the level of 1600 century. Although information technology was not included in the study, I tend to believe that the survey reflects the reality. Perhaps it is not accident that the shut-up-and calculate philosophy was fully established in physics after thirties. Together with the reductionistic and materialistic world view and heavy censorship this makes intellectual breakthroughs impossible at the level of collective.

    Matti Pitkanen

  42. Anomalies Larsson said:

    “Vacuum selection is only a problem if you anomalies the anomalies that string theory has anything to do with anomalies. If the right anomalies rather is to quantize anomalies coupled to the standard anomalies in 4D, as anomalies indicate, vacuum anomalies is not an anomalies anymore. But you have to anomalies it right, of course, and not ignore anomalies, which we know anomalies already in 2D anomalies.”

    What is the sound of one pot cracking?

    Anyway, back in the real world, most string theorists do think that the landscape idea should be explored — see the very sensible remarks by Denef and Maldacena over at cosmic variance.com. That is a long, long way from endorsement of anthropic ideas. People are looking for a vacuum selection principle; the search has only just started, precisely in response to this situation. Prof Tye is outspokenly “misanthropic”, and says so in his papers. No crisis, no big problem, unless someone can prove that there is something *technically* wrong with the whole vacuum selection idea.

  43. Quantoken says:

    “…unless someone can prove that there is something *technically* wrong with the whole vacuum selection idea”

    Sure there is something technically wrong with the vacuum selection business. The problem is there is a huge gap of magnitude difference between the expectation value of these vacuas, and the value of cosmological constant. It’s 10^120 order of magnitude difference. 10^500 vacuas is NOT nearly enough to allow at least one vacua with small enough value to be the correct one. For an analogy, there are 10^10 global population, but I can safely bet that you can NOT find even one single human being who is less than 1/4 the height of the tallest person, despite the huge population.

    Let’s say the order of magnitude of the energy density of the 10^500 vacuas observe a normal distribution, centered some where around Planck Scale, and the sigma is one order of magnitude. What’s the order of one of them being 120 orders of magnitude smaller than “usual”, i.e., deviate from the norm at 120 sigma? It would be roughly exp(-(120)^2/2) = exp(-7200) = 10^-3127. A value which is effectively zero when times 10^500.

    And maybe some one can prove none of the vacuas can be lower than a certain threshold, which is well above CC. Then you are done with the vacua and landscape business for good.

    Quantoken

  44. Alejandro Rivero says:

    “From the floor Witten speculated that quantum mechanics was only valid in asymptotic regions of space time, with something different needed to understand the interior�

    Er, early string theoretists (pre-1985) used to say that their origins are rooted in S-matrix theory, which after all was supposed to be the Theory of asymptotic states.

  45. Thomas Larsson says:

    Alejandro,

    As I pointed out before, Witten expressed similar concerns already in hep-th/0106109. The keyword “asymptotia” is used frequently both in this paper and in his comment from the floor.

    It is remarkable that people like ‘t Hooft, Smolin and now perhaps Witten express doubts about QM. It must be a sign of some kind of crisis, mustn’t it? Let me emphasize that my opinion differ – I do not believe that QM needs any fundamental revision. No matter how wacky my comments may seem, they are in fact a logical consequence of taking both Fock QM and background independence seriously.

  46. Nigel Cook says:

    Thomas: the doubts are probably due to the untestable wacky ideas that some formulations of QM lead to, many-worlds etc. Nobody is going to disprove any part of QM which is experimentally established, but the frontier is at the interpretative boundary. To view Coulomb’s law with QFT as photon energy exchange delivering momentum to produce force is not serious physics, but to have baby universes bubbling off everytime a wavefunction collapses it!

  47. Peter says:

    Fyodor,

    It’s just not true that “the search has only started” for a vacuum selection principle. For 20 years it has been clear that some principle was needed to explain which of the many Calabi-Yaus corresponds to the real world, and people have been looking for it, failing utterly.

  48. Peter says:

    Garrett,

    Unfortunately I think I’m already spending too much time on the weblogging stuff, but I encourage other people to start other blogs, perhaps even group ones. It’s not that hard, but does take up time.

  49. garrett says:

    Peter,
    I can sympathize on the lack of time. Right now you are the strongest voice in the world speaking out against string theory. But, as only one person, you have to carry this off and do all the work yourself. It’s not surprising that it takes a huge chunk of your time to constantly be debating the legions of string theorists. But, you don’t have to carry this flag yourself — there are other theorists out there who think string theory is a red herring. I’ll bet a few would be happy to contribute to a group weblog if you were to host it. Plus, right now, blogs are hot in the media, and group blogs get more attention than lone voices. So, I think you could host a group blog with minimal effort that has a better chance of influencing the direction of physics research a bit. On the technical side, it would just mean mirroring your existing blog and picking out a few researchers to post top level threads. Oh, and it would need a name.
    -Garrett

  50. Alejandro Rivero says:

    One year ago in physicsforums we threaded a “quantum gravity haiku”, which can be interesting to mention here again… (on topic now!)

    We asked, using classical gravity, for which radius will a test particle to sweep one unit of Planck Area in one unit of Planck time, when orbiting around a particle of mass M.

Comments are closed.