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

Fundamental postulates of QM

  1. May 5, 2013 #1
    I find the fundamental postulates of QM very ad-hoc and strange.
    Compare them to the fundamental postulates of special relativity, special relativity naturally arises out of classical electromagnetism and the equivalence of all inertial frames, while QM seems to come out of nowhere.
     
  2. jcsd
  3. May 5, 2013 #2

    tom.stoer

    User Avatar
    Science Advisor

    Can you present a list of the postulates you have in mind?
     
  4. May 5, 2013 #3
    The reason that the postulates seem to come out of nowhere is because we live in a macroscopic world and do have a built in physical intuition of quantum mechanics.

    I'll post the postulates for further conversation referance (for simplicity assume non-degenerate discrete spectrum):

    First postulate: At a fixed time t the state of a physical system is defined by specifying a ket |psi(t)> belonging to the state spce E

    Second postulate: Every measurable physical quantity A is described by an operator A actin in E; this operator is an observable.

    Third postulate: The only possible result of a measurement of a phyical quantity A is one of the eigenvalues of the corresponding observable A.

    Fourth postulate: When the physical quantity A is measured on a system in the normalized state |psi>, the probability of obtaining the non-degenerate eivenvalue an of the corresponding observable A is:

    p(an) = |<un|psi>|2

    where |un> is the normalized eigenvector of A associated with the eigenvalue an,

    Fifth postulate: If the measurement of the physical quantity A on the system in the state |psi> gives the result an, the state of the system immediately after the measurement is the normalized projection Pn|psi>/sqrt(<psi|Pn|psi>) of |psi> onto the eigen subspace associated with an.

    Sixth postiulate: The time evolution of the state vector |psi(t)> is governed by the Schrodinger equation:

    ihbar (d/dt)|ps> = H(t)|psi>

    where H(t) is the observabe associated with the total energy of the system.
     
  5. May 5, 2013 #4

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    It does?

    The postulate of SR appears "ad hoc" to me as well. After all, one HAS to make such a thing when saying that c is constant in all reference frame. Only AFTER making a logical/mathematical consequence of such a postulate, and having it match to physics observation, can one consider such a postulate to be correct.

    So I don't see this being any different than any other postulate made in physics.

    Zz.
     
  6. May 5, 2013 #5

    tom.stoer

    User Avatar
    Science Advisor

    Could it be that you are not bothered by the postulates but by the entities used in the postulates?
     
  7. May 5, 2013 #6
    More or less, my point is that there is a natural procession from classical mechanics to SR and even GR, the system of things are still comparable to each other, we still have our good old equations of motion and initial value problems.
    Whereas for quantum mechanics, it seems the entire formalism has changed. Probability is involved, complex numbers and observables becoming operators, it is completely departed from all other theories.
     
    Last edited: May 5, 2013
  8. May 5, 2013 #7
    I disagree, Einstein's postulates seem much more physical and "tangible" and so do Newtons.
    The QM postulates just aren't really motivated by experiment and they don't seem to originate from some simple physical principle.
     
  9. May 5, 2013 #8

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    I disagree. SR postulate did not originate from "some simple physical principle" either! What physical principle dictates the uniformity and the isotropic nature of c?

    What you are describing appears to be a matter of tastes, and that is what we are arguing about now.

    Zz.
     
  10. May 5, 2013 #9

    BruceW

    User Avatar
    Homework Helper

    @ HomogenousCow: I know what you mean, in that the QM theory seems wildly different to any other physical theory. But I would say that the QM postulates are motivated by experiment. I think they are just the simplest set of postulates that people could think of that matched the new idea of the quantum. (Edit: and this idea of the quantum seemed to consistently agree with experiment).

    QM is the same as other physical theories in that it has postulates, makes predictions. But it seems very different to most other of the broad physical theories. You could say this is depends on the person. Some people might think that QM is similar to other theories, and other people might think QM is very dissimilar. Neither person is necessarily wrong. All we can say is that so far QM has not been proven wrong. Therefore, it obeys the correspondence principle, so its result agrees with other theories (like classical mechanics), in the limit which those theories apply to.
     
    Last edited: May 5, 2013
  11. May 5, 2013 #10

    BruceW

    User Avatar
    Homework Helper

    yeah, HomogenousCow is not disputing that QM is a physical theory. he's just saying that it looks pretty ad-hoc and strange. And I would agree with him. Maybe this is too 'philosophical' for PF.
     
  12. May 5, 2013 #11

    vanhees71

    User Avatar
    Science Advisor
    2016 Award

    There is nothing too philosophical with quantum mechanics. It's a clearly stated physical theory about the behavior of matter, it's giving a very precise description about what's observed in nature, and so far there is no empirical evidence for a restriction of its validity. That's why it is the most fundamental theory about matter (its constituents, the elementary particles and the interactions between them). So there is no doubt that quantum theory is a very nice example for a physical theory.

    Surely, quantum theory leads to a quite drastic correction of our everyday experience about the behavior of matter, but that's only due to the fact that we are simply not confronted with situations, where the quantum effects (coherence) play a role. Of course, given the microscopical structure of matter the most important conclusion from quantum mechanics is the supposedly "simple fact" that matter in everyday life is pretty stable. Given this "simple fact" in almost all circumstances of everyday life classical mechanics and classical electromagnetism "explain" the behavior of objects around us pretty well.

    It's, of course, important for the consistency of these different layers of description of nature that all of them can be understood in terms of the most fundamental theory at hand, which is, in our context quantum theory. That's indeed the case, because the classical behavior of macroscopic objects, including the absence of the unusual properties of coherence, entanglement, "wave-particle complementarity", etc. is understandable from quantum many-body theory, where one derives the classical equations for macroscopic objects as a very good approximation to the full quantum behavior, which we cannot resolve with our senses anyway.

    This is not so different from the theory of relativity, where also many aspects appear not to be so easily comprehensible with everyday experience. This starts even with the most simple kinematical effects like the relativity of simultaneity, time dilation, length contraction, etc. Again, the reason is that we are not used to circumstances where relativistic effects play a prominent role since usually bodies around us do not move with a speed close to the speed of light and we are living only in a very weak gravitational field of the earth (and other bodies around us like the moon and the sun). So again, Newtonian mechanics is well valid in our world of everyday experience, and that's why we are unused to the fact that these phenomena are descibable by the Newtonian approximation of the more comprehensive theory of relativity.

    All that is of course not "philosophical" but at the very heart of the natural sciences and the role of model building within them.
     
  13. May 5, 2013 #12
    Right but I find the postulate that speed of light is the same in all inertial frames much more physical and "tangible" than the QM postulates.
    If you look at it, QM is very much just infinite dimensional linear algebra given physical interpretations, it baffles me why they correctly describe our world.

    The whole business of quantization gives me a weird feeling, of them all canonical quantization is the most unintuitive to me and path integral quantization is least.
     
  14. May 5, 2013 #13
    I agree but it baffles me why this model works, none of the postulates are directly motivated by experimental evidence, only after some deep digging do we find that they agree with interference and other observations.
     
  15. May 5, 2013 #14
    The postulates are a refinement of years of progress based on experimental work. Quantum mechanics wasn't birthed by writing down the postulates.
     
  16. May 5, 2013 #15

    WannabeNewton

    User Avatar
    Science Advisor

    It seems like your uneasiness with the postulates of QM is that they are mathematical and not "physical" like the postulates of SR? While I agree that the motivation for the postulates of SR from the need to make Maxwell's equations frame invariant is extremely "physical" and elegant, it is not as if the postulates of QM are completely arbitrary even if they are mathematical. Read the first couple of pages in ch2 of Ballentine and see if the motivation there is satisfactory enough.
     
  17. May 5, 2013 #16

    micromass

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    I find the QM axioms rather natural if you take the approach of C*-algebra. In that approach, you can clearly see the link between QM and classical mechanics. And you clearly see that QM and CM are the same thing, except that QM is noncommutative. The usual axioms are than derived through some difficult mathematics.
     
  18. May 5, 2013 #17

    Fredrik

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Edit: LOL. micromass always finds a way to say something similar to what I'm going to say while I'm typing. :smile: (Although, it's been a while since the last time that happened).

    There are at least two other approaches to QM (other than starting with the Hilbert space axioms) that are a bit more intuitive. Unfortunately, they require some very heavy math. I don't have a perfect understanding of either of these approaches, so it's possible that some of these details are a bit off, but I'll try:

    One approach argues that if a theory can predict the average value of a sequence of measurements on identically prepared systems, done by identical measuring devices, then we should be able to define equivalence classes of measuring devices, and some mathematical operations on the set of equivalence classes that give it the structure of a C*-algebra. The rest is just an application of the theory of representations of C*-algebras. In particular, there's a theorem that ensures that there's a homomorphism from the C*-algebra (whose members are called "observables" btw) into the set of bounded operators on a Hilbert space. This approach is called algebraic quantum mechanics.

    Apparently if the C*-algebra is commutative, there's some other theorem that ensures that what we get is a classical theory.

    Another approach argues that there should be a lattice (a partially ordered set that satisfies some additional conditions) associated with each theory, and that this lattice will need to satisfy some technical conditions in order to not be extremely hard to work with. Apparently these technical conditions are sufficient to ensure that the lattice is isomorphic to the lattice of Hilbert subspaces of a Hilbert space. This approach is called quantum logic.

    Apparently, if the lattice satisfies some additional conditions that makes it even easier to work with, it will be isomorphic to the (partially ordered) set of subsets of some set X. This set can then be interpreted as the phase space of a classical theory.

    Some not so easy references:

    Strocchi: Introduction to the mathematical structure of quantum mechanics: a short course for mathematicians.
    Araki: Mathematical theory of quantum fields.
    Varadarajan: Geometry of quantum theory.

    For the algebraic approach, you need to know lots of topology and functional analysis. The book by Varadarjan is so hard to read that it's even hard to tell what sort of things you need to know for the quantum logic approach. The book contains long sections on projective geometry, and measure theory on locally compact simply connected topological groups.
     
    Last edited: May 5, 2013
  19. May 5, 2013 #18

    BruceW

    User Avatar
    Homework Helper

    yeah, true. But if you look at the postulates of classical mechanics, then it is not immediately obvious that they are related to experimental evidence. As far as I know, these are the most commonly used set of postulates:

    1) Transformations between inertial frames of reference are described by continuous, differentiable and bijective functions.
    2)If the velocities of two freely moving particles are equal in system S, they will also be equal in system S'.
    3)All the inertial reference frames are equivalent.
    4)The space in any inertial reference system is isotropic.
    5)Simultaneity is not relative. ((This is the one that is different for special relativity))

    Maybe you would count these as postulates about the geometry we exist in, and stuff like F=ma as the postulates of mechanics... But the point is that you need these 'geometry postulates' as well as the 'F=ma' postulates, to get classical mechanics. You can't just start with F=ma. In a similar way, you need to have some postulates in QM which do not give off an obvious appearance of being motivated by experimental evidence.

    So I'd say that (in my personal opinion), the postulates of QM give as much of an appearance of 'being motivated by experimental evidence' as the postulates of classical mechanics are. (And to be clear, I know it is true a priori that both sets of postulates are motivated by experimental evidence. But the only thing I am discussing is whether it is immediately obvious from looking at the postulates that they are motivated by experimental evidence).
     
  20. May 5, 2013 #19

    WannabeNewton

    User Avatar
    Science Advisor

    You might be interested in this document: http://www.math.uchicago.edu/~may/VIGRE/VIGRE2009/REUPapers/Gleason.pdf

    Jorriss (and some others at Uchicago to be fair xD) showed it to me ages ago but I guess he forgot to link it himself; esentially it outlines the same issues that you are putting forth but then goes on to give the more "classical mechanics related" C* algebraic approach (note that the results aren't proven in this document but there are references to texts e.g. Rudin where you can see the associated proofs).
     
  21. May 5, 2013 #20

    micromass

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    Fredrik, you mention two different theories. Can you tell me which theories are described in the books you recommend?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Fundamental postulates of QM
  1. QM postulate (Replies: 3)

Loading...