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

What is your favored interpretation of quantum mechanics?

  1. Shut up and calculate / Decoherence solves everything

    21.0%
  2. Copenhagen

    11.1%
  3. Statistical (ensemble)

    8.6%
  4. Conciousness causes the wave-function collapse

    6.2%
  5. Bohmian (pilot wave) / Nelson (stochastic dynamics)

    8.6%
  6. Spontaneous collapse (e.g. GRW)

    0 vote(s)
    0.0%
  7. Many world / Relative state (a softer version of many world)

    13.6%
  8. Consistent histories

    0 vote(s)
    0.0%
  9. Information theoretic / Relational / Quantum logic

    13.6%
  10. Something else / None

    17.3%
  1. Dec 4, 2006 #1

    Demystifier

    User Avatar
    Science Advisor

    As more than 10 options are not allowed, I was forced to put some different (hopefully similar) interpretations together.

    After voting for the favored interpretation, you may also indicate what is your second best interpretation, what is the worst interpretation for you, or even make the whole top-list.
     
    Last edited: Dec 4, 2006
  2. jcsd
  3. Dec 4, 2006 #2

    Demystifier

    User Avatar
    Science Advisor

    For each interpretation (including my favored) I indicate what, in my opinion, is its main weakness:

    - Shut up and calculate
    Too pragmatic. (If I wanted to be pragmatic, I would be an engineer, not a physicist.)

    - Copenhagen
    Requires an artificial separation between classical and quantum realms.

    - Decoherence solves everything
    It does not explain how a single value of the measured observable is picked up.

    - Statistical (ensemble)
    Does not even try to describe the properties of individual objects.

    - Conciousness causes the wave-function collapse
    Too antropomorphic.

    - Bohmian (pilot wave)
    Requires a preferred foliation of spacetime.

    - Nelson (stochastic dynamics)
    The relation between first and second quantization is obscure.

    - Spontaneous collapse (e.g. GRW)
    Requires an artificial modification of the Schrodinger equation.

    - Many world
    Too metaphysical. (Additional worlds are not observable even in principle.)

    - Relative state (a softer version of many world)
    The origin of the Born rule is obscure.

    - Consistent histories
    Does not even try to anser the crucial question:
    What, if anything, exists at times at which measurements are NOT performed?

    - Information theoretic
    As Bell said: Whose information? Information about what?

    - Quantum logic
    I cannot imagine anything more radical. Classical logic is the last thing to be changed.

    - Relational interpretation
    Too solipsistic.

    - Something else
    Too obscure. :biggrin:

    - None
    Too nihilistic.

    You are encouraged to do the same according to your opinion.
     
  4. Dec 4, 2006 #3
    I prefer "Shut up and calculate". Its easier on my brain than the others and there is something nice about not needing any interpretations. If I can work out some probability amplitude then I am happy.

    I don't really see the need for some interpretation to tell us that when we measure something it changes the possible futures. As far as I am concerned its intuitively obvious that when you are working with a function which produces a distribution then as it evolves it will evolve differently if you have a "predetermined" measurement as a boundary condition than if you have another distribution as a boundary condition. What happens when we actually sample from the distribution to create the measurement is probably more important... but I don't really think about. At first glance it seems like something which could be answered with some causality argument...

    This is probably just my limited understanding at work though, which just gives me more reason to stick with "Shut up and calculate!".
     
  5. Dec 4, 2006 #4
    Where do the ideas of Penrose fall ?--his comments on the matter make sense to me.
     
  6. Dec 5, 2006 #5

    Demystifier

    User Avatar
    Science Advisor

    They fall to "Spontaneous collapse" I guess.
    If not, then they certainly fall to "Something else".
     
  7. Dec 7, 2006 #6

    Demystifier

    User Avatar
    Science Advisor

    More than 100 readings and only 10 votes! :frown:
    Come on people, why don't you vote!?
     
  8. Dec 8, 2006 #7
    I prefer Bohm's interpretation (some subjective experiences make me prefer provisionally hidden variables interpretations). Some quick comments (no time to elaborate):


    - Shut up and calculate
    Too close to the old positivism (of Mach). Besides it is widely accepted now that science can be non-cumulative sometimes; there is nothing 'metaphysical' to prefer provisionally a certain interpretation (as a personal research program deserving to be pursued further).


    - Copenhagen
    Again too positivistic (too close to logical positivism), requires also the necessary introduction of macroscopic concepts.


    - Decoherence solves everything ('consistent' histories is basically the same)
    What decoherence? Even Roland Omnes agrees that one of the main drawbacks of this approach is that we cannot put it in evidence even in principle...Of course I am not arguing that it is not a valid approach, decoherence is only a theoretical concept indeed, but it 'works' acceptably (at this moment of time).


    - Bohmian (pilot wave)
    'Requires a preferred foliation of spacetime' indeed but this does not mean that it has been falsified. Besides all 'no go' theorems are not really relevant here (at least contextual approaches are fully viable).

    I happen to prefer it at this moment of time (more exactly I prefer a hidden variables interpretation; Bohm's interpretation is the only one enough successful so far).

    It seems rather degenerative at the moment indeed (and too close to the classical physics) but this is not necessarily fatal (Lorentz invariance for ex. is not something 'set in stone' forever). Philosophy and history of science are much more relevant here than physicists usually believe...indeed future may still be full of surprises.


    - Many worlds
    Inesthetic. Too many worlds.


    - Consistent/ Decoherent histories
    'Copenhagen done right'. Roland Omnes argue extensively for it and for decoherence...but personally I do not like it too much (too close to Copenhagen). Besides I am far from considering his over-optimistic epistemology (see for ex. his book 'Quantum Philosophy') based on this interpretation as really solving the problem of infinite epistemological regress once and forever (he is a foundationist).

    Anyway he has the merit of recognizing that we have a problem which has to be solved here (regarding the empirical basis of science); the vast majority of scientists take for granted naive realism, without offering a justification for this; I'm afraid that merely labeling philosophy 'sterile' does not really solve this problem (the fact that science 'works' at the pragmatic level does not automatically imply that it necessarily approaches Truth).


    Transactional Interpretation
    Too 'science fiction' (this does not mean that I consider it wrong, I just don't prefer it at the moment, the same is valid for all viable alternatives - apart from hidden variables interpretations).


    Could we make a ‘weak’ difference (non strongly prescriptive) between the different interpretations, based on other parts of physics (and the actual scientific methodology)?

    What interpretation has the greatest coherence with the other accepted parts of physics?

    Well I would say that the coherent/decoherent histories interpretation is the first choice here. But of course the most coherent hypothesis at a certain moment with the other parts of science does not necessarily indicate the 'right branch'...
     
    Last edited: Dec 8, 2006
  9. Dec 11, 2006 #8

    Demystifier

    User Avatar
    Science Advisor

    I am happy to see that (at the moment) the Bohmian/Nelson interpretation is on the second position, just after the shut-up-and-calculate/decoherence interpretation. But this may change, of course.
     
  10. Dec 11, 2006 #9

    Demystifier

    User Avatar
    Science Advisor

    By the way, nobody said which interpretation is the most meaningless to him/her. For me, it is the quantum-logic interpretation.
     
  11. Jan 19, 2007 #10

    vanesch

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I voted, as you can guess, MWI.

    I like the arguments against MWI:

    and

    Also, note that between MWI, relative state, and consistent histories, there is not really any difference.

    Finally, even by the authors of decoherence (such as Zeh), they recognize that decoherence doesn't solve the measurement problem, but is only a tool which indicates why, in a MWI setting, the classical worlds that appear as subsolutions seem to act classically (do not interfere anymore).
     
    Last edited: Jan 19, 2007
  12. Jan 19, 2007 #11

    Demystifier

    User Avatar
    Science Advisor

    Didn't you forgot to put the relational interpretation on that list too?
    BTW, I was writing the above before you explained me your view of the MWI.
     
  13. Jan 19, 2007 #12
    Unlike Einstein, I believe that God does play dice, but he doesn't cheat.

    Things at the quantum level are to small and individually inconsequential to bother with absolute physical control as in Classical Physics.

    Its on automatic and statistical chance is the rule. This allows the universe to exist and still be indeterminate.
     
  14. Jan 25, 2007 #13

    Demystifier

    User Avatar
    Science Advisor

    The assumption of hidden variables has the greatest coherence with the paradigms of other sciences. The Bohm interpretation has the greatest coherence with the other accepted parts of physics.
     
  15. Jan 29, 2007 #14

    Well no one denies the fact that Bohm's pilot-wave interpretation is the closest to a 'classical' realist view (I've stressed myself that above; some critics consider this a weakness but I argue that such a complaint is rather a 'red herring' characteristic to an obsolete very strong positivistic view, largely abandoned now in philosophy and even in the scientific QM community). But anyways it can be equally claimed that other interpretations do support realism, also fully regaining common sense at the macro level; they appear thus as compatible as Bohm's interpretation with the methodologies used by the so called 'higher level' sciences.

    Now of course the methodologies used by the current science imply much more than this and it is here where Bohm's interpretation really has (at the moment) a lower coherence.




    Bohm's interpretation needs some 'patches' which appear, at least at this moment, as rather ad-hoc. These auxiliary assumptions are not really blowing it apart (as some claim too loudly my view) but certainly make it less coherent with the other parts of physics and requirements of the actual scientific methodologies.

    How can it be that the electrons do not fall into the nucleus? Vigier's answer is such an auxiliary hypothesis, still viable of course, but it appears now as rather ad hoc since neither could we use his hypothesis to make novel predictions nor could we obtain it as a deduction from other accepted theories. Remain to be seen however, we cannot rule it out yet.

    Also it is very unlikely that a relativistic form of Bohm's interpretation will retain Lorentz invariance (which has become part of the methodology used by physics, at least since 1919; there is the requirement that new, 'deeper' theories, should retain it). Anyways even if this is possible it looks that we should use other variants of transformations, compatible with the preferred frame of reference involved by Bohm's interpretation. Consequently we should use this modified set of Lorentz transformations for GR too; this is not in itself a problem (GR can be constructed around such modified transformations, giving equally valid empirically alternatives) but the actual methodologies used by physics have the usual set of transformation at their core...

    Why should the spin be only a useful theoretical construct (not really referential) as the pilot wave hypothesis (Bohm's variant) implies? Indeed one of the reasons for supporting so enthusiasically QED was the accuracy of its predictions regarding spin...spin seems very real...And there are some other problems (the photon for example has a different status than other constituents of the micro world) etc.

    But while I agree that these problems make Bohm's interpretation (at this moment) less coherent with the other parts of physics and currently accepted methodologies it does not result yet from here that such a research program is necessarily a 'dead end', even the methodologies could change in non trivial ways (i is still fully possible that for example Lorentz invarince will be totally abandoned)...it may be lower placed now in a (non strongly prescriptive) list of existing research programs but the future may still be of surprises...
     
    Last edited: Jan 29, 2007
  16. Jan 30, 2007 #15

    Demystifier

    User Avatar
    Science Advisor

    Metacristi, I agree with you that the Bohmian interpretation contradicts relativity (and also locality) in its usual form, which makes it uncoherent with the rest of physics. However, most other interpretations deny the existence of objective reality itself, which, in my opinion, makes it uncoherent with the rest of (non-quantum) physics even more. After all, when we were young, nonrelativistic and nonlocal Newton law of gravitation or Coulomb law of electrostatics seemed quite viable to us. Many physicists (applied physics, solid state physics, ...) still use them and think of them as fundamental.
     
  17. Jan 30, 2007 #16

    vanesch

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member


    :approve: Fully in agreement with what you write.
     
  18. Jan 30, 2007 #17

    vanesch

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    MWI is an "intermediate". Although it postulates the existence of a genuinly real objective world, and even with an entirely "classical" deterministic evolution, and fully compatible with a 4-dimensional geometrical spacetime view (as long as it is not dynamical - the problem with gravity in GR is still open), it does say that what we "observe" to be reality, is only part of it, and the part which is observed is stochastically determined. This is probably the main conceptual difference with Bohmian mechanics, which "trades" the 4-dim spacetime compatibility for the benefit of being able to say that what is observed is "all" of the (particle part of the) world.
     
  19. Jan 30, 2007 #18

    Demystifier

    User Avatar
    Science Advisor

    At this point, the relational interpretation (RI) differs from MWI. If there are many parts of reality, than one can think about the whole reality, consisting of all these parts together. On the other hand, in RI, the notion of the whole reality does not make sense. Do you agree?
     
  20. Apr 12, 2007 #19

    -I-

    User Avatar

    Wave–particle duality
    From Wikipedia,
    Quote
    Because waves are non-local then all objects are non-local too (exist in many places at once) - see non-locality.
    End quote
    Measuring them makes them local.
    -I- exists in all places
     
  21. Apr 15, 2007 #20
    I would suggest that a wave is not 'something', but the description of how something moves. Try to locate a 'wave', all you will find is some form of atomic or subatomic movement that can be described as moving, or has been moving, in a particular manner that we call a wave. Equations are great, but they do not necessarily make things real.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?