Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Revival of CM ?

  1. Nov 16, 2005 #1


    User Avatar
    Science Advisor

    Anybody who can explain the details of atomic spectra by classical means will win a Nobel. Particularly, 100 years ago, quite a few have tried to build classical models of the atom, all without success. In the meantime we've put semiconductors and lasers, quantum devices both, radioactivity and nuclear magnetic resonance to work, ..... The odds for a classical rebirth look pretty slim.

    Reilly Atkinson
  2. jcsd
  3. Nov 17, 2005 #2
    Ah, Roman greek classicism came also back after the middle ages; who would have made a bet on that in 1500 ? :smile: It is unrealistic to think that realism is sacrified in one century, on the other hand the number of science fiction freaks is growing steadily every day :biggrin:
  4. Nov 17, 2005 #3


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Let's have the "new classicists" discussion here, about the possibility of reproducing QM results with classical, or neo-classical theories. Let's refrain from scattering these discussions in other threads (if you think it useful to do so, post a small reference in the other threads to this one).
  5. Nov 18, 2005 #4
    Oh don't be afraid: I shall not be this one who will make CM revival. No, I just wanted to ask: what is for you exactly the boarder between classical and not classical (quantum)? Is it the introduction of the h number (Planck), the discovery of scattering, ...? Is it "just" a technical and mathematical procedee applied to classical theories? Is it the fundamental discovery that phenomenon at very small scala are not continuous? ...
  6. Nov 18, 2005 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The superposition principle.
  7. Nov 18, 2005 #6
    I agree here, the introduction of nonlinear interactions as mutiplication operators and thereby keeping the wave equation linear is definetly something incomprehensible from a classical point of view. Also, the fact that a multiparticle Schroedinger wave lives *essentially* on configuration space is unacceptable from a spacetime point of view. :smile: The Planck number is NOT something ``quantal´´ in the sense that it is just a phemenological scale which is put in by hand in the theory (you can do the same in classical physics). Also scattering has (semi) classical analoga. By the way, QM does not say at all that phenomena on the small scale are not continuous (as long as you do not invoque reduction). It only says that measurable quantities usually have discrete spectra; but a similar phenomenon could be achieved in a classical theory where phase space is partitioned in a ``free´´ part (continuous spectrum) and a discrete set of stable attractors (note that you have to include radiation degrees of freedom here) with very small transition times between different domains of attraction.
  8. Nov 18, 2005 #7


    User Avatar
    Science Advisor
    Gold Member

    What about the Uncertainty Principle? I have seen attempts to model it from classical principles (I'm not sure any successfully); but I personally don't really think it is classical at its base.
  9. Nov 18, 2005 #8
    That does seem to be key.
    The areas beyond the border into the QM domain can be explained away in the classical as unreadable because of the A) the lack of tools to make the needed precise measurements required, and/or B) the need to find and understand the “unknown variable” that would allow for a precise interpretation of measurements when “it” is included.

    QM experiments that seem to show proof of its correctness can be explained as statistical conveniences that mathematical predict averages. But the prediction of averages alone does not mean an understanding of reality or a proof.

    Thus for QM it is the “superposition principle” is the part that offers a demonstration that QM can understand reality. Understanding superposition is the key to completing the view of reality within QM. It’s the concept of superposition that best explains the behavior of a particle, like and electron in ‘orbit’ or ‘being’ around a proton.
    But from the CM view this can rightly be considered just a successful and workable mathematical statistical convenience, not proof.
    And again in the double slit experiment that only shows statistically accumulated evidence that explained by superposition. CM can legitimately claim that this does not give a proof to superposition, only that it can provide an interpretation that fits the statistics. CM could do as well or better if only it could find the “unknown” variable, guide particle, guide wave or whatever that could classically define the results. These could be possible as yet undiscovered parts of real physics.

    Thus the key to revival of CM would be finding the unknown “whatever” that would provide those solutions. So the task would seem clear as to what CM needs to find. QM has no way to prove QM correct over CM as its results will always be statistical.

    But QM has taken another approach, by attempting to prove that any search for the needed CM “unknown” would fail. That is to prove a negative. First ‘proven’ by Johnny von Newman his math was shown to be “absurd” by John Bell in the 60’s. In its place he proposed his Bell Theorem that he hoped would show the way to the Einstein “unknown variable” and revive CM. Although another statistical proof, when tested experimentally it seems to show Bells hopes were misplaced and the idea of an “unknown variable” might be unworkable.

    The application of Bell to the entanglement experiments both by polarization and “Stern-Gerlach” devices are the only experiments that tend to “Prove” the “unknown variable” as untenable. And by implication, as the only viable idea remaining, that superposition must be correct.

    Advise me of any others, but as far as I know this is the only “proof” of superposition and therefore QM.

    So it seems to me the first step for revival of CM is finding that “unknown”.
    Or at a minimum demonstrating that the one and only negative proof against CM, and that the “unknown” could be real, is somehow flawed.
    Which would imply “superposition” could be wrong.
    Last edited: Nov 18, 2005
  10. Nov 19, 2005 #9
    Your exposition seems fair to me here. Only some comments concerning the Bell theorem. The perfect separablility is *not* a necessary hypothesis from the local realist point of view, albeit a *reasonable* one in practical setups where the dectors are far enough separated from one another (say a distance > one meter).
    Moreover, there are many Bell inequalities which result from this hypothesis, the weakest one (and never violated in any experiment where the separability assumption is *tenable*) being the Clauser Horne 74 inequality. But, as you might have seen in previous threads, the discussion always amounts to ``everyone has his own taste´´ (citation from ``Die Fledermaus´´ :smile: ). If you dismiss taste and don't bother about the argument ``QM did so much for us´´ (except the nasty cat which some would prefer to die) then Bell tests provide no proof whatsoever for now against local realism. If they would one day be perfect (and the QM prediction comes out), then one knows finally that atoms are smart, remember their past and can anticipate what the dectors are going to do. This is actually what QM tells us if you want to give a ``realist´´ interpretation to its outcome.
  11. Nov 19, 2005 #10


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I agree entirely with your analysis. However, it is often the property of a "principle" not to be demonstrable by "direct evidence". General covariance is also not demonstrable as one can always fix up an ether theory with the same predictive value. The true value in a principle lies in its ability to guide a theoretical construction, as a unifying idea, not so much in its direct observability.
    I think that's the real point. It is a matter of "taste", "belief", ... to find out whether or not one can spend time and effort looking for that "unknown whatever" in a CM that would give us the same results (at least those that HAVE been verified experimentally) as QM. One shouldn't crucify people wanting to do so, but it should also be accepted that certain people "believe" that such efforts are quite vain. This is a matter of personal taste. The only observation is, that today, we have no such theory.
    I think that the attempt to "prove a negative" is somehow a vain effort. I think we should consider Bell's theorem as interesting challenges to QM: they indicate where it is interesting to TEST QM to experiment. But the whole class of LR CM theories is too vaguely defined to be eliminated without ever leaving a small loophole - with enough sophistication, one will always invent a way to explain away THIS experiment. So I don't see these tests as definitive exclusions of whatever class of theories. I only see them (until contradiction) as further indirect experimental confirmations of QM (and hence of its underlying ideas). Each of these experiments provides also further constraints on a possible CM theory.
    Well, I would first like to *see* a CM theory that can reproduce QM results. Until then, I'm not really interested in it. There *are* some interesting partial results, in some domains, like SED. So I can understand that some people want to look for it.
  12. Nov 19, 2005 #11
    I disagree on this point.
    It isn’t fair to expect a “realist view” of QM as they are by design different things. So IF a “perfect” QM proof could one day be found, it would not show that atoms remember their past. Rather it would confirm some QM theory that allows that past ‘here’ to be irrelevant. Maybe where another dimension is real with it own ‘history’ thus the two separating partials could remain ‘together’ in that dimension with no change in history there. But when ANY interaction takes place with either particle they then separate to develop there own histories in that other dimension. The point being, the interaction could only change one of the particles – thus they have identical histories up to and including the start position for that interaction on only one of them. In the history of our dimension the other particle would remain identical to or at least within strict correlation with that starting position but not be affected by the interaction in the least, and it would not matter at all if the unaffected particle was in the past or future in our dimension.

    True that may look like memory, but memory would not be what QM is ‘telling us’.

    But this is not this thread is about. More to the point here, do I believe that QM we can expect someday to see a “perfect” proof for QM? Actually I believe the design of QM demands that it can never do so, mostly because I don’t believe in that ‘other dimension’ I just mentioned. However there are 100’s of physicists searching for extra dimensions that if they can provide such proof would change my world view a lot. To show that this search was pointless in a clear CM argument could potentially save or redirect a lot of valuable resources.

    Thus more than the point vanesch makes that CMers shouldn’t be crucified, I think the CM area should be given at least some small fair share of encouragement.
  13. Nov 19, 2005 #12
    Good point, I think too much has been made of the “negative proof” that CM cannot work. A better focus for CM is to look at the constraints the many experiments provide as data and guide posts to better focus a complete CM or any solution with.

    I also note that reilly in opening this thread did offer some encouragement to the CM crowd by declaring that if properly successful they will win a Nobel Prize. Or at least he’d use his influence to see that it was so.

    I think that comes with about $50,000 or more.
    How about we find a little more for it. Presuming the successful CM approach is complete enough to fully explain the strong force it would most likely mean proving “Yang-Mills” wrong.
    Now there is a $1,000,000 prize from the Clay Math Millennium problems to prove it correct. But the stringent rules at ClayMath.org/millennium also point out that an altranate proof (showing Yang-Mills wrong) would also qualify. (For the math wiz’s they also have six more problems)

    Now there’s a little more encouragement.
    Anybody know of any more “encouragement” out there being offered.
    Heck I’ll kick in a couple bucks, if the result brings a legit TOE even if it’s not CM.
  14. Nov 20, 2005 #13
    Following this very clever indication I have red the pages devoted to this principle in Tannoudji Diu and Laloe .... and discovered that it plays a very important role, given us the pragmatic possibility to understand the difference between statistic and quantum theory;well. But I also learned that it stays on two requisites: a) the Schrödinger equation which is linear and homogen; b) the fact that the space of states is a vector space.
    Do you know if this picture holds for other representations of the quantum theory (Heisenberg, Dirac or interaction representation, ...)?
    The spinor theory allows the formulation of a lot of Dirac's relations; the actual research tries to incorporate the notion of spin inside a Riemanian geometry; so, does this superposition principle exist and hold in such a context? I mean with this if the space of states is no more a vector space, what happens? Is it a science-fiction representation?
  15. Nov 20, 2005 #14
    I agree with this suggested way of doing; CM needs first a plausible scenario that allows a correct description of QM predictions. Some people try to do it via the statistics but the principle of superposition is exactly the principle that explain the limit of the statistical approach; so it must not be the good one or at least, it must be completed with something else. I see QM more as a pragmatical and successful modern approach and CM as a more philosophical one. QM makes good predictions, CM tries to give us a deep description of "how are the things, how do they work really". Both are complementary ways of thinking and doing. I think we actually need a mental representation of what really happens at quantum scala in a classical way of thinking. Here lies the key.
  16. Nov 20, 2005 #15
    In this sense we could state that physical phenomenon arise with a certain probability inside a fixed underground topology and consider these phenomenon as quantum behaving;
    or that these same but classical phenomenon arise with this given probability because the underground topology is permanently changing and only statisticly presenting a given configuration ... forcing them to behave as we can observe them ...
    it is a question of relativity, of point of view, ...
  17. Nov 20, 2005 #16


    User Avatar
    Science Advisor
    Homework Helper

    If you put QM into the QFT form in the position representation, the configuration space becomes trivial. That is, a position representation creation or annihilation operator is only nonzero at a single point in spacetime.

    From this point of view, the momentum representation is just a useful mathematical trick.

    What I'm saying here is that the QFT position representation is ontological and the others are just useful mathematical methods for calculation. Any reason why this is unacceptable?

  18. Nov 21, 2005 #17
    ??? Configuration space in QFT in the *Schroedinger* picture (in the canonical free field approach) consists of *all* field configurations on a slice of your favourite foliation. The Schroedinger wave essentially entangles field configurations at different (even relatively spacelike) space time locations.
  19. Nov 21, 2005 #18
    The superposition principle, the wave equation, wave functions, harmonic oscillators and resonance, etc.-- all of these indicate a similarity to classical wave phenomena which continually evolve but are trackable in a way that quantum phenomena aren't trackable.

    The terms that are used to qualitatively describe quantum phenomena all come from our macroscopic experience. But a *qualitative* picture of the essentially wavelike quantum phenomena is necessarily incomplete, because in order to 'see' quantum phenomena, they must be made to interact with macroscopically identifiable and manipulable instruments. This necessarily renders quantum phenomena 'discrete' and 'particulate' and 'random' as far as our perceptual apprehension is concerned. This is *necessary* because of limitations regarding the behavior of macroscopic instruments.

    Quantum theory was made as classical as it could be made without sacrificing internal consistency and consistency with experimental results. For those who think that qm could now (apparently because much more is known now than was known several generations ago) be made even more classical, then the question is what exactly can be changed in qm to make it more classical?

    Or, are the limitations that caused qm to be developed as it was still in effect?
  20. Nov 21, 2005 #19


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I have noticed several of these discussions going on in a number of threads, i.e. the apparent "validity" of classical description, and the idea that you cannot observe directly the manifestation of quantum phenomena. At the risk of exposing my utmost annoyance of such claims, I will try to point out a very obvious 90000-pound gorilla that almost everyone seems to have ignored - SUPERCONDUCTIVITY.

    I will cite a paragraph from Carver Mead's PNAS paper[1] that says:

    I will point out that there have been ZERO attempt at trying to describe this phenomenon classically. NADA. Zilch! With the existence of high-Tc superconductor, classical mechanics seems to have thrown up its hand up in the air and gave up. It has no hope of describing the phase diagram of the cuprates superconductors, and even less of a hope to describe the pairing symmetry of the Cooper pairs, especially the spontaneous current that emerges out of the phase of the order parameter[2].

    I have always maintain that the most convincing and evident demonstration of QM phenomena does not come from some esoteric experiments. They come from very familiar and reproducible experiments on condensed matter physics/material science. Unfortunately, the familiarity and "mundane" access of such experiments, and the fact that these are not "sexy" areas of physics made many people overlook the fact that these are QM phenomena staring right in their faces. People who dismiss QM always seem to want to tackle the "Schrodinger Cat", the "Bell-type experiments", etc.. etc. by producing alternative description that allows plenty of weasel room. NONE of them have ever attempted to describe superconductivity with the same accuracy and agreement as the BCS theory, for example, reproducing ALL of its predictions.

    Until such a time when classical physics can produce a First Principle derivation of superconductivity, I will not be convinced that there is an "alternative" to QM.


    [1] C. Mead, PNAS v.94, p.6013 (1997); or http://www.pnas.org/cgi/content/abs...&stored_search=&FIRSTINDEX=0&journalcode=pnas

    [2] D.J. Van Harlingen Rev. Mod. Phys. v.67, p.515 (1995).
  21. Nov 21, 2005 #20


    User Avatar
    Science Advisor
    Gold Member

    For the above ref in PDF format, try: Our model system is a loop of superconducting wire...Correspondence limits based on classical mechanics are shown to be inappropriate.

    As ZapperZ points out so well, it is not really fair to say the Classical (and Sem-classical) theories haven't had their day in court. They have. Even when alternative hypotheses are presented by respected scientists, they are often critiqued and found to come up short. And the one critical piece is ALWAYS missing: a useful new prediction. Please note that QM is still making useful NEW predictions every day - 75 years after introduction. I would like to see someone start with their alternative hypothesis and derive a useful new prediction.
  22. Nov 21, 2005 #21


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The discussion is, unfortunately, not about what an existing classical theory has to say about it (hence risking falsification) but whether the hope that a future "classical" theory might one day do so, is totally vain or simply tiny but existing.

    Nevertheless I'm curious what LR proponents have to say about it - although I can guess it: "a full non-linear field equation solution for something as complicated as a solid is totally out of the question given that it is even not well known for two "particles". So anything potentially goes as we can't do the calculation. Who knows that "superconductivity" is not a solution of these complicated non-linear equations ?"

    That's why I pointed to a much more modest and probably more tractable "classical" calculation: configuration interaction in quantum chemistry, on simple molecules (or even: the argon or neon spectrum!).
    Effective field theory IS what you'd expect a classical field solution to come up with (if there's a way for the different "particles" not to merge into one big blob). But configuration interaction is a sheer product of entanglement. However, because these corrections are relatively small (although very well measurable), they can still be dismissed as "well, maybe it is a GR effect we didn't take into account".

    As I said elsewhere, there's no need to crucify people who cling still on that tiny hope - after all they might think of challenging experiments - but I think that as long as they don't come up with something a bit more elaborate than SED which reproduces a bit more QM results before getting interested.
  23. Nov 21, 2005 #22
    Also, how QM prediction of simple semiconductor bandstructure and its experimental verification is already phenomenal. There's very remote chance one can reproduce those bunch of curves with one equation :surprised

    And the very reason is that QM is inherently a model for describing probabilistic outcomes makes it very versatile. Things of this nature (anything which we are absolutely ignorant of unless we measure it :rofl: ) can most probably be tackled by QM. In fact QM model is so versatile that it was being applied to new field of studies remote from physics such as consciousness and finance. Not sure if this speaks well of QM.. but it sure does give people the hint that QM is nothing more than a tool for statistic.
  24. Nov 21, 2005 #23
    So that means seeing CM as NOT philosophical but as real.

    No: If the CM approach is to be done seriously it must reject superposition as a correct solution. Once rejecting that, it must reject QM as being the “correct path” as Einstein did in the 1920’s.
    Now assuming that it would be folly to attempt to do the work by transforming QM into something “CM” How could you avoid bring along whatever it is you’ve assumed to be wrong? Einstein saw this early on so he spent a lifetime on QT as it was prior to QM and trying to combine with GR. But I don’t think that was a CM effort. Since GR by replacing three dimensions with four is not truly classical.

    . The question becomes; just where does a CM solution start from? Assume the popular current superposition paths are dead ends. That may be the biggest problem with most CM approaches, they to often start from positions that have superposition already embedded in their background assumptions. I don’t think CM can expect to be successful using that approach, a completely new perspective needs to be used.
  25. Nov 21, 2005 #24


    User Avatar
    Science Advisor
    Gold Member

    1. I do not think SED is falsifiable to its adherents.

    2. There really is no SED as a single theory - it is a morphing concept that has a single purpose - and that purpose has no use by definition. SED is an ad hoc theory intended to mimic QM while rejecting one or more quantum mechanical concepts. The author's only objective is to restore to theory the classical notion of determinism at a fundamental level while seeing indeterminism arise from the stochastics. This objective is also true of Bohmian Mechanics, which takes a different path.

    3. Most scientists do not believe that a local realistic theory such as SED can exist. It is incompatible with Bell's Theorem, and for many that is enough to exclude it a priori. And logically, given the SED program, it should be obvious that SED is doomed to fail if you accept Bell's Theorem - you don't need an experiment ("loophole-free" or not) at all to tell you this.

    4. Why should any theory be given credit before it delivers? For that matter, why should any theory be given credit before it shows any promise? SED holds out absolutely NOTHING at the end of the rainbow. If it succeeds, we will have QM. That story would be different IF it offered us something. For example, string theory at least promises a unification of GR and QFT as its premise - and that is clear and compelling!

    So I guess it goes without saying that I personally am in the "SED has no promise" camp. :biggrin:

    I realize that you want to be conciliatory to those with a differing viewpoints. (I hope my own cock-eyed ideas don't get reamed either. :uhh: ) But there ARE good reasons why more efforts are not expended on research into alternative interpretations/explanations of QM's results. Careful strongly supports the SED program; you support the MWI program (and are a fine and reasonable advocate I might add!); ttn strongly supports the BM program. Each of you might feel there should be more research into a particular area because it shows promise. OK, fine, maybe you are right (small chance of breakthrough) - and I don't think anyone wants to stifle research anyway. Yet each of you should be able to easily acknowledge that there are conceptual difficulties with this position - SED, BM and MWI all go in completely different directions! How can they all have equal promise?
  26. Nov 21, 2005 #25


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    But this is nothing but speculation. People who cling to such hopes should realize it for what it is. You can't use unverified speculation to go against something that has been verified.

    My point is that based on existing classical theory and philosophy, I see zero indication (not just small, but ZERO) of its ability to even come close to describing the superconductivity phenomenon. The idea of long-range phase coherence of charge carriers is not only difficult, but it is a non-existing concept in classical mechanics. And I haven't even brought in the issue of David Pine's "Quantum Protectorate". Again, these are extremely robust observations from condensed matter that blatantly defy classical descriptions. It is there, for all to see. It is just that not many people are aware of it.

    But I think it is QM that is riducule every time such statements are made. And more importantly, the whole field of condensed matter is ridiculed, because they are completely ignored even when obvious QM observations are there. People seem to either ignore or be ignorant of the fact that the accepted standard values for fundamental constants such as "e" and "h" are obtained from condensed matter experiments. It is an insult to this field of study each time someone claims that classical mechanics can also emulate QM results. One can only claim that by pretending that the whole field of condensed matter does not exist.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook