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

Measurements problem - QED

  1. Aug 10, 2007 #1
    I recently watched a program called 'Atom' on bbc4 that said something along the lines of: "If you want to strike fear into the eyes a physicist; mention the measurement problem" - What is the measurement problem and why is it such a problem?
  2. jcsd
  3. Aug 10, 2007 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I must not be a physicist then, because the "measurement problem(?)" strikes no fear in my eyes - and I'm an experimentalist too and I measure things for a living!

  4. Aug 10, 2007 #3
    BBC4 is, of course, exaggerating. What is known as "the measurement problem in quantum physics" is a certain set of open interpretational and philosophical questions regarding the relation between micro- ("quantum") and macro ("clasical") levels of physical theories. It should noted that these open questions do not signal any flaw or inconsistency of modern physical theories, including quantum mechanics.
  5. Aug 10, 2007 #4


    User Avatar
    Science Advisor

    There are two kinds of physicists: practical ones and those who want to understand nature on the deepest possible level.
    The measurement problem "strikes fear into the eyes" only for the latter.

    What that problem actually is? Perhaps not everyone will agree, but I would summarize the problem by the following questions:
    Do physical observables have some values even when we do not measure them?
    If no, then how exactly measurement makes them?
    If yes, then how exactly measurement changes them?
    Last edited: Aug 10, 2007
  6. Aug 10, 2007 #5
    But why fear? I would say "fascination and excitement". Many students of quantum mechanics (including me) get "hooked" precisely by the contraintuitive nature of QM and the sense that it never completely fits into your mind.
  7. Aug 10, 2007 #6


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Ah, but you're missing another option here. There could be physicists who are "practical ones" but also interested in understanding nature "on the deepest possible level". The difference being that until there's truly something that can distinguish and answer those questions, everything that is being argued is simply a matter of tastes and would not lead to any kind of agreement.

    Physicists like Tony Leggett make testable proposals on the measurement problem rather than simply argue based on tastes. I would consider that to be very "practical".

  8. Aug 10, 2007 #7
    So is it philosophical rather than physical?
  9. Aug 10, 2007 #8
    It depends. The problem and solutions proposed in the literature are mostly philosophical. However, if a particular mechanism of solution, like Legett's [mentioned above] or Penrose's leads, to experimentally verifiable predictions in the areas not yet tested, then it is a matter of physics.
  10. Aug 10, 2007 #9


    User Avatar
    Science Advisor

    The guys like Legett (BTW, can someone provide me an appropriate reference or link?) are not really practical. Instead, they propose experimental tests just because they want to be sure that their "deep understanding of nature" is correct. They probably do not intend to do anything practical with that. In fact, I am also one of such guys:
  11. Aug 10, 2007 #10


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I guess it depends on what you mean by "practical". Leggett is "practical" in my book because (i) he's a condensed matter theorist (ii) he always focuses on experiments and even proposes what to measure (iii) his CM background allows him to come up with CM experiments that have some of the highest degree of certainty of any experiments conducted in physics. I don't associate "practical" with something that has applications.

    As for references, he wrote about the quantum measurement problem in an article in Science a few years ago (Science v. 307, p.871 (2005)). But his seminal work on this issue was a few years earlier after the Schrodinger Cat-type experiments by the groups at Stony Brook and Delft that essentially did an experiment he sugested (J. Phys.: Condens. Matt., v.14, p.R415 (2002)).

  12. Aug 10, 2007 #11


    User Avatar
    Science Advisor

    Also, my sense is that it would be hard to really develop a theory of quantum cosmology (where the entire universe is treated using quantum rules, with no observer outside to measure it) without at least partially addressing the measurement problem, so it's not purely philosophical in this sense.
  13. Aug 15, 2007 #12
    I suggest reading “The Quantum Challenge” by G Greenstein and AG Zajonc. It is good and clear presentation of the problem and it current status (2005) almost without math.

    Regards, Dany.
  14. Aug 15, 2007 #13
    Simply out-dated:
  15. Aug 15, 2007 #14
    If you don't know how to drive a car, you cannot go behind a steering wheel and pretend that you know how to drive a car, because everybody can see that you cannot drive a car.

    But if you don't know what the quantum mechanical measurement is, you can pretend that you don't have any problems with it, and nothing will force you to admit that you have no idea what it is.
  16. Aug 16, 2007 #15
    Please, define what the quantum mechanical measurement is (the definition that you use).

    Regards, Dany.
  17. Aug 16, 2007 #16
    You have no idea what the measurement problem is. It is clearly defined during last 80 years. You confuse OP and violate PF guidelines:

    Regards, Dany.
  18. Aug 16, 2007 #17
    I was talking about the process, that is mathematically described as the state being projected onto some eigenstates. This is what quantum mechanical measurement usually means.

    It is mathematically clear, but what is it physically? When does the projection occur? What initiates the projection?

    We are violating only the philosophy "shut up and calculate", but that yet is not a big crime.
  19. Aug 16, 2007 #18
    No. It is not a measurement. You are talking about the universally valid phenomenon called the collapse of the wave packet.

    Let us talk physics. Your definitions still not clear to me. What is the physics the measurement apparatus obey? Is it the classical physics only (macroscopic)? If so (CI), then why you call it the quantum mechanical measurement? If your physical system under test is macroscopic (obey laws of the classical physics), you will find that every dynamical variable is measurable (observable); if your physical system under test is microscopic (obey laws of the quantum physics), then only the dynamical variables that form the mutually commuting set of self-adjoint operators will be observable. But the measurement set-ups are the same. Clearly for me that you perform the same measurements, only the results are different. Therefore, it is the demonstration that the quantum world is not the classical world. In addition, the above type of measurements is accompanied by the collapse of the wave packet.

    R.Penrose calls it the “R-process” to make distinction from the unitary evolution of the QM system (“U-process”). But it is not a process, there is no dynamics that associated with it and it is not needed. The act of the measurement is instant.

    If you agree with me, please redefine the quantum mechanical measurement.

    At the same instant when you perform the measurement. It is the transition from the Quantum World to the Classical World. Symbolically we call it the E. Schrödinger Cat.

    Interaction with the measurement apparatus.

    Regards, Dany.
  20. Aug 16, 2007 #19
    If a Gaussian wave packet, that is badly delocalized, hits a wall, and collapses into a smaller area, when does the collapse occur? This is a measurement of the position. When does the measurement occur?

    I can be more specific. Suppose we can approximate some wave packet solution in one dimension as

    \psi(t,x) = e^{-(x-vt)^2}

    Okey, that's not a real solution of SE, but it doesn't matter. The peak of the wave function is at location [itex]x=vt[/itex] at any given instant t. Now, if there is a wall at location L, can you tell me when this "wave packet hits that wall"? An answer "at time [itex]t=L/v[/itex]" is not valid, because that is only the instant when the center of the packet hits the wall. An answer "when the packet first touches the wall" isn't valid either, because it is touching it all the time.

    So, when does the wave packet hit the wall? When does the measurement of the position occur?
    Last edited: Aug 16, 2007
  21. Aug 17, 2007 #20


    User Avatar
    Science Advisor

    Be careful before such statements. I have published several papers on this issue in peer reviewed physics journals. For one of these journals the editor is the Nobel-prize winner G. 't Hooft (who, by the way, also explores the measurement problem in QM). How about you?

    For people like you who think that all conceptual problems of QM are already solved a long time ago, I have written a review:
    And yes, it is accepted for publication in a peer reviewed journal (for which the editor is the mentioned Nobel-prize winner).
    Last edited: Aug 17, 2007
  22. Aug 17, 2007 #21


    User Avatar
    Science Advisor

    Brilliant! :smile:
    Is that written by you? Do I have a permission to cite it somewhere?
  23. Aug 17, 2007 #22
    but I also heard that some wackos actually think that measurement and discovery can change past events??? Also I don't understand how Schrödinger's cat can be anything other than pure philosophy as it doesn't change anything physically it's just our understanding of the events.
  24. Aug 17, 2007 #23
    hmhm.. yeah it's mine. I used to think this way in programming, one older programmer emphasized this to me. In social life you can always accuse others when things don't work, and nothing forces you to see your own mistake, but in programming your own program, that doesn't work, forces you to see your mistake mercilessly. And watch how the justice works here: If you refuse to find your own mistake in the code, you cannot get the program working anymore! :biggrin:

    My comment about wave function collapse was part a more general philosophy. There is usually two kinds of "not understanding". Kind of, where nothing forces you to see the problem, and kind of where it is impossible to keep pretending.

    Citing it... hmhmh... I didn't think it's that good myself :/
  25. Aug 17, 2007 #24
    I remember hearing this too. It sounds like misunderstanding of quantum mechanics. Perhaps measurements can change past events in some very exotic interpretation, but its not part of the popular Copenhagenian interpretation at least.

    I'm not sure what you are saying here, but the Shrodinger's cat paradox is a real paradox. You'll have to understand QM before it. What is paradoxical is how macroscopic and living beings could be in superposition of very different states.
  26. Aug 17, 2007 #25
    The OP question is: What is the measurement problem and why is it such a problem? There are several sessions here at PF related to that question but at least during last year nobody suggest the discussion of the roots and the present status of the problem which is now under intensive development.

    In addition, I consider CI out-dated. First of all, I try through discussion with Jostpuur to understand how the problem is defined today. Later, what was done and what remains open. Knowledge at least of W&Z “Quantum Theory and Measurement” is required to contribute to our discussion.

    Your post #4 is related to the M. Born statistical interpretation of QM and has nothing to do with the Measurement Problem. As to your warning to be careful, it may be considered by OP, that we don’t fear anything: you have no idea what the measurement problem is. Try to focus on the specific problem or just open the relevant session.

    I have no idea what you are talking about.

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