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

At what point does determinism become untenable?

  1. Oct 1, 2005 #1
    1. At what point does determinism become untenable? (Why) is it incompatible with quantum mechanics?

    2. Can we solve the wave-particle duality by postulating particles that move in waves?

    These are my main questions for now. The object is to see whether determinism can be taken to the extreme - whether even quantum physics may be regarded as deterministic.

    Thanks in advance for your answers.
  2. jcsd
  3. Oct 1, 2005 #2


    User Avatar
    Science Advisor
    Gold Member

    I think the hopes for a local deterministic version of QM has pretty well been dashed at this point (although our "many worlds" brethren might see it differently).

    Sophisticated experiments on entangled particles shows that reality is shaped by observations that are performed after the fact. By most definitions of determinism, determinism is no longer viable. Those same experiments show that the wave particle duality is fundamental, and is a direct consequence of the Heisenberg Uncertainty Principle (HUP). The HUP stands in direct opposition to determinism and that is why QM is incompatible with determinism. Don't forget that the HUP has substantial experimental verification, and that cannot be ignored. Bell's Theorem limits the ability of a local deterministic theory to replicate the predictions of the HUP.

    The only way to restore determinism is to throw out locality, a move that some scientists are OK with and some scientists are not. Either way, QM and HUP stand as they are currently formulated. It simply comes down to a matter of interpretation past that.
  4. Oct 1, 2005 #3
    I do not know the principles you are referring to. I am trying to see how far one can come just by rational thinking (without empiricism).

    As I understand the problem of measurement on a quantum scale, one cannot measure the course of, for instance, a photon without shooting another photon at it and measuring the change in this second photon's behaviour after their collision. This seems to me to be comparable to driving a car into another car's side in order to measure the course of the latter. Of course the measurement radically alters the measured in this case. But does the fact that we cannot objectively tell what the exact course of a photon is mean that it does not have any? I tend towards the determinist perspective that everything has definite relations to everything else.

    An example of my trying to understand such a principle on a purely logical basis is the following. There is (or was?) the principle that there can be no greater relative speed than the speed of light: whereas, if one car moved towards another with a speed of 20 m/ph, and the other towards the one with a speed of 30 m/ph, their relative speed would be 50 m/ph, in the case of two photons heading straight for each other their relative speed would still be only once, not twice, the speed of light. My argument runs thus:

    In the example of the cars, there is a third, the observer, to whom both these cars have a relative speed. In the case of the photons, the measure is one of the photons itself. This has no speed in relation to itself, as it does not move in relation to itself. So, from the photon's "perspective", it is standing still - it is the measure of all things, the 0,0,0 coordinate in the x,y,z axis that is the three dimensions. And, from this perspective, the photon that is heading straight at it has a relative speed of c, the speed of light. Does this make any sense?
  5. Oct 1, 2005 #4
    This can work only for single particle wavefunctions and in fact was the original idea of de Broglie. There it is possible to think of the particle as being like a cork floating on water being pushed around by waves, via the de Broglie-Bohm interpretation of quantum theory. However, the wavefunction of a multi-particle state is not a wave propagating in ordinary 3 dimensional space. Instead it is defined on configuration space, i.e. it assigns an amplitude to every possible assignment of positions to the particles rather than just one amplitude for each point in space. It is still possible to derive a de Broglie-Bohm interpretation of this, but it becomes much more abstract and can't be understood in such a simple way.
  6. Oct 1, 2005 #5


    User Avatar
    Science Advisor
    Gold Member

    No, it doesn't have any beyond what the HUP allows. Why would I say that? It is because the question is one we all deal with at various point - and the answer seems so unreasonable.

    Einstein took up this question, and he thought he had it licked in favor of determininsm. He contributed to the paper now known as EPR. Later, there was Bell's Theorem, which I would recommend you check out. It says (to paraphrase):

    No physical theory of local Hidden Variables (determinism) can ever reproduce all of the predictions of Quantum Mechanics.

    Since QM is amply verified by experiment, that makes it impossible to restore determinism unless you throw one of the key tenets of relativity out the window.
  7. Oct 1, 2005 #6


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    :smile: :smile:

    Just to say it shortly: in a many-worlds-like approach, nature is 1) deterministic and 2) satisfies local dynamics BUT you only observe part of it, and it is the part you observe which is "randomly" assigned.
    The non-local appearance of Bell-like situations then arrises from our non-justified extrapolation from learning about a result that remote measurements HAD an outcome before we locally observed them (while in fact, BOTH outcomes occured simultaneously but we were going to observe only one of both).

    This sounds of course totally crazy (it probably is :smile:). However, it is what you get out of the theory when you apply its formalism totally literally to everything (including observers, people...).
    And then it turns out that what we've been trying to shove into "locality" or "determinism" is in fact a much much weirder aspect of nature, namely that we would consciously only observe one single facet of it.

    If it were not for the fact that Einstein put his finger first on the "locality/reality" issue, in fact an EPR experiment would be typically what would constitute a "proof" for the existence of the superposition of macroscopic states (such as the superposition between the states "Bob saw his detector clicking" and "Bob didn't see his detector clicking"), in the same way as interference patterns are "proofs" for the superposition of position states (|particle through slit A> and |particle through slit B>). So the conclusion would simply be, from Alice's point of view, that the statement "Bob saw his detector clicking" or "Bob didn't see his detector clicking" does not make sense until she did her measurement AND IF YOU SUPPOSE THAT BOB'S RESULT EXISITED you get "probability-violating" conditions (just as in the case when you suppose that the particle went OR through slit A OR through slit B you get problems where the interference pattern is destructive with the arrival probabilities of the particle).

    However, this leaves you with the dilemma that FOR BOB, OF COURSE HIS RESULT EXISTED. And the way you can weasel out in many worlds, is by assuming that Bob is just experiencing another facet than Alice about the world. So results only "exist" with respect to an observer. Another observer may come to totally different conclusions and that's not in contradiction, because everybody just observes HIS/HER facet of the world.
    And that's where this view REALLY gets weird.
    Last edited: Oct 1, 2005
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook