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

I Would studying MWI be a waste of time?

  1. Apr 4, 2017 #1
    There seems to be a lot of discussion on the Many Worlds Interpretation here at the present time and it's something that I personally can't get into very deeply. The reason I can't be bothered too much about it is because I can't help feeling that the whole concept is nonsensical. I base this feeling on what I think is one of the most powerful branches of physics ........... what I describe as being general knowledge, common sense and intuition.
    Does anyone else feel the same or does anyone think the MWI should be taken seriously?
     
  2. jcsd
  3. Apr 4, 2017 #2

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    Definitely, if you think MWI is nonsense, then it's a waste of time to study it. Pick something that you don't consider nonsense to spend your time on.
     
  4. Apr 4, 2017 #3

    Demystifier

    User Avatar
    Science Advisor

    MWI is one of interpretations of quantum mechanics. There are other interpretations too. But if you only want to study things which please your common sense and intuition, then I am not sure that you should study quantum mechanics at all. Or do you think that some other interpretation is compatible with your common sense and intuition? If so, then stick with that interpretation (provided that it doesn't contradict any experiments).
     
  5. Apr 4, 2017 #4

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    My feeling about MWI is that it is the most "pure" framework from which to discuss quantum issues. Every other interpretation makes dubious distinctions between measurements and other types of interactions, or between macroscopic and microscopic, or implicitly makes one choice of basis more equal than others, or one decomposition of the universe into system + environment, or implicitly makes "classical" assumptions that are not justified by quantum theory. This isn't to say that MWI is free of conceptual problems, but in my opinion, most of them are problems for QM, not just MWI--it's just that they are less obvious in other interpretations.
     
  6. Apr 4, 2017 #5

    Demystifier

    User Avatar
    Science Advisor

    MWI appears "pure" because MWI people usually write those dirty stuff (preferred basis, additional assumptions etc) in "fine print". See
    https://arxiv.org/abs/1703.08341
    Sec. 3.
     
  7. Apr 4, 2017 #6

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    I agree completely that your issues in sections 3.2 and 3.3 (3.4, "Quantum Suicide", is not central to MWI, it's just philosophical musing) are serious issues for MWI. My point is that it's possible to attack those issues head-on in MWI, while in other interpretations (such as Copenhagen), the issues are hidden behind fuzzy concepts such as "measurement".
     
  8. Apr 4, 2017 #7
    I certainly want to study QM just because I find it interesting. And I find most of QM to make sense and to be intuitive. But there are exceptions, Many Worlds and the Quantum Zeno effect being two of them. If I invest some more time and study these topics in greater detail any increased understanding I get might make the topics seem to be more sensible. But I wonder if it will be worth the effort.
    I agree with your comments about experiments but any interpretation has got to conform to observations. Also, I think that if any interpretation is to gain credibility then it should be possible, if only in principle, to make observations that can validate the predictions made.
     
  9. Apr 4, 2017 #8

    tom.stoer

    User Avatar
    Science Advisor

    That was my opinion over a many of years, too.

    The problem is that if you do not study the Everett interpretation at least in some detail, you don't get the concept right. So what you believe is nonsensical is not the Everett interpretation itself but something you believe that it could be the Everett interpretation. Many explanations on the Everett interpretation miss the point - both in the positive and in the negative. So the problems you are told the Everett interpretation have may be non-issues or misunderstandings. At the same time you miss the esssentual problem of this interpretation b/c you are bothered with non-issues.
     
    Last edited: Apr 4, 2017
  10. Apr 4, 2017 #9

    My biggest problem with it is the lack of clear meaning of probability which is vital for the QM formalism itself before we even start to talk about interpretations and metaphysics behind it.

    This is something that MWI people (including its proponents on this forum) sweep under the carpet, and others like some respected philosophers of science who tried to derive it in a meaningful way have not did it a convincing way. It seems to me that the determinism of the theory somehow blocks a natural way to derive the probability in it. Before it's done, I believe it should not be taken seriously at all.

    To understand what I mean, first try to understand the minimalistic approach and then connect it to the approach in which everything happens and then you may indeed have supstance to conclude it is a waste of time.
     
  11. Apr 4, 2017 #10

    vanhees71

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The question is what interpretation of QT are good for. I can't answer that question, because I think from the point of view of physics, it's just the minimal statistical interpretation which is needed to apply the theory to the description of phenomena and to predict outcomes of measurements. That's all what physics is about. It's not the purpose of the natural sciences to provide a world view or religion!
     
  12. Apr 4, 2017 #11

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    I would say that interpretations of a theory are not religious beliefs. Religion is something that teaches moral precepts, such as the true purpose of the natural sciences.
     
  13. Apr 4, 2017 #12

    tom.stoer

    User Avatar
    Science Advisor

    This is a problem for QM in general and by no means restricted to the Everett Interpretation. The idea to derive probability i.e. the Born rule from the remaining axioms plus other - hopefully simpler - ingredients is definitly interesting in itself. We can learn a lot from these discussions, even of we are not succesfull. The Born rule, the collapse postulate or whatever is so terribly complicated that it's hard to accept it as fundamental axiom.

    Postulating probabilities = introducing them via an axiom does provide any clear meaning of probability at all.

    In addition we do have indications that there us a chance to find probabilities w/o postulating them, e.g. the Hartle frequency operator.

    I had some discussions on this topic here in this forum where nothing has been "swept under the carpet".

    I agree.

    So you think that we should take e.g. the collapse interpretation seriously? It postulates a collapse w/o explaining why and when it happens, it does not explain what a measurement is and how it can / why it should be distinguished from ordinary unitary time evolution, it postulates a probability w/o any clue to understand where it is coming from ...

    Honestly, I like the MWI much more, simply b/c it forces us to think about these questions.
     
  14. Apr 4, 2017 #13

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    That's what I meant in #4.
     
  15. Apr 4, 2017 #14
    But the problem for the MWI interpretation is not the potential multitude of ways how to get a probabilistic rule. The problem is how to make sense of probability at all in a theory where everything possible happens. That, by definition, in some way contradicts the meaning of probability itself. That is quite different than "going minimal" like Mr Vanhees or doing some experiments and calculating frequencies in a lab. If you feel like I've missed your point, feel free to correct :smile:

    As a matter of fact, one of the most interesting threads that I've read was the one where you participated (the topic was of course MWI).

    https://www.physicsforums.com/threa...-the-many-worlds-interpretation.706927/page-5

    I was left with the impression that the probability issue was strongly swept under the carpet by introducing "hypothesis testing" etc. (not your position ofc) because it doesn't come in a natural way. It seemed to me like adding something so fundamental like the probability rule in a QM interpretation was done in a sort of laid-back 'oh, we need probability here, let's make something up' way.


    Yes, I do.

    First of all, what do you mean by saying that it is postulating probability w/o any clue of where is it coming? I believe that it is pretty simple if we count collapse as a exclusion of other "branches" except one. Or perhaps you meant something else?

    I still believe that all of those questions can be answered more naturally than introducing probability in MWI. I believe that measurements have a special meaning which allows us to give a normal way of interpreting probabilities. If that's not the case, I believe that there's something more fundamental than just the "wavefunction unitarity" which is a pretty cheap (I may as go far as to say lame) way to interpret the mystery of quantum mechanics.
     
  16. Apr 4, 2017 #15

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    Well, imagine the following toy model of a universe. It is deterministic except for the fact that there is a special nondeterministic coin which, when flipped, has a 50/50 chance of resulting in "heads" or "tails".

    So that's a situation for which probability presumably makes sense. But suppose that the way this nondeterminism is "implemented" is as follows: (I'll make this religious sounding, just to be perverse)

    Every time someone flips the coin, God first halts the universe. Then he makes two copies. Then he lets the coin land heads-up in one copy, and heads-down in the other copy. The copies no longer interact after that point.

    So there is no nondeterminism in this multiverse. What happens is completely determined, even when people flip the seemingly nondeterministic coin. So it's deterministic, but from the point of view of people's memories, it seems nondeterministic.
     
  17. Apr 4, 2017 #16
    MWI certainly does not make sense. This starts already with the meaning of probability. A probability that something happens makes no sense if everything happens. To save the game, one uses some common sense concepts about probability. The problem are not these common sense axioms themselves, but why they make sense in a situation which does not make sense.

    The definition of the worlds makes no sense. If a world is some point of the configuration space or so where the wave function is nonzero, then there is a continuum of worlds, all except a few points (the zeros of the wave function) would be worlds. A precise definition of something else (local maxima or so) I have never seen.

    Then, they use, for some strange purpose, decoherence. Decoherence presupposes some structure (subdivision into systems) which is not defined.

    If you want an interpretation which makes sense, start with the minimal one, then look at Copenhagen (but get rid of positivistic nonsense - easy to identify, because it follows the rule "what we cannot observe does not exist", so one simply has to replace all "does not exist" with "is unknown") and try de Broglie-Bohm, given that it explains nicely what the measurement process is, and what the wave function collapse is, and also shows by example that a lot of impossibility claims are wrong, and that the quantum world can be realistic, even deterministic.
     
  18. Apr 4, 2017 #17

    tom.stoer

    User Avatar
    Science Advisor

    Yes, I mean something else.

    Accepting the "collapse" means observing a statistical behavior in experiments and therefore postulating the Born rule i.e. essentially postulating the collapse. Of course this does not explain anything. Accepting the "branching" means not postulating the Born rule, not postulating the collapse, but trying to derive the observed statistical behavior and the Born rule. In this approach we either understand what causes statistical behavior to appear and what causes the Born rule to emerge, or - if all this will fail - we at least may understand why it fails.

    This is unacceptable for me b/c in practice that means you decide that a specific interaction is not treated using unitary time evolution but collaps, simply b/c this produces the intended results.

    Even if it were true, I would like to learn why :-)
     
  19. Apr 4, 2017 #18

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    Well, I would argue that that's not true. Given any probabilistic model, you can form an equivalent "ensemble" view where everything is deterministic. The deterministic ensemble model is mathematically equivalent to the probabilistic single-system model.
     
  20. Apr 4, 2017 #19
    Thanks for the great analogy mr. stevendaryl, I am buying what you are saying. At least I think so.

    But my opinion hasn't changed because of that. After the creation of the copies of the coin you mentioned, an observer should still calculate the probability of getting an outcome. And the outcome is?

    100% heads and also 100% tails!

    This is something that seems to me like the Achilees heel of MWI, each time an observer tries to include probability you get something which is circular. Saying that both copies deterministically occur is just sweeping the real problem under the carpet in a pretty uncovincing way. Saying the observer becomes both copies after the experiment doesn't give you something like "ok, this makes sense" - it is just avoiding the intrinsic nondeterminism of the theory in a unconvincing way, which makes MWI pretty cheap in my opinion.
     
  21. Apr 4, 2017 #20
    But you're not postulating the Born rule while postulating the collapse. You are postulating the collapse per se and the Born rule comes naturally. It sort of "derives itself" through non-unitarity and the existence of only one result in experiments. This is fundamentally different than deriving it in the MWI. Do you agree that Copenhagen/collapse has a natural probability rule that is sort of already built in it?

    How about irreversible decoherence? That is treated using unitary time evolution and causes apparent collapse. You are free to assume that the non-unitary part occurs for real and that the collapse is not only apparent but real as well. Why it might occur, I would also like to learn :smile:
     
  22. Apr 4, 2017 #21
    How would you define probability in one particular (single) experiment where all outcomes occur? Let's say I throw an electron through the double slit. Does it make sense to say that it's more likely that it will land somewhere in the middle of the pattern than on the edges if all spots will be hit in some world?

    If yes - how?
     
  23. Apr 4, 2017 #22

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    It's not possible to observe all possible outcomes. What it is possible to observe is a persistent record of past events. I sketched a toy model above: Every time someone tosses a coin, two copies of the universe is made---one where the result is "heads" and another where the result is "tails". To someone confined to one copy, it seems that the universe is nondeterministic.

    Is there probability in such a multiverse? Well there will be some possible universes in which the frequency of heads will approach 50%. For people living in those worlds, it makes sense to assume that the probability of a coin toss is 50/50. The usefulness of the concept of probability in one universe shouldn't be affected by the existence of the other universes.
     
  24. Apr 4, 2017 #23
    That's extremely controversial. Forget the 50-50 coin, that indeed is a toy model. Let's assume a real quantum experiment, like the decay of the radioactive nucleus at some time t.

    Suppose that there is 90 percent probability that it does not decay and 10 percent that it does.

    Both outcomes occur. Why is non-decay more likely in this case since both branches are equally real?
    Speaking of probability in this way implies that there is something wrong with your argument that it makes sense to speak of probability on each particular branch despite the existence of others.

    In a toy model, it might. But in concrete examples like mine, MWI is flirting with nonsense and circularity.
     
  25. Apr 4, 2017 #24

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    I usually avoid interpretation threads because they tend to be filled with people who cannot do an actual QM calculation but are still rooting for Team Wheeler or Team Bohm or Team Cramer etc. Furthermore, they tend to revolve around which interpretation is "right", and since by definition interpretations all produce the same outcomes (otherwise they are theories) this is not a scientific question. But I chimed in to say that MWI is perhaps the worst named interpretation. As Sidney Coleman explained to me, in MWI there are not many worlds. There is only one world.
     
  26. Apr 5, 2017 #25

    tom.stoer

    User Avatar
    Science Advisor

    The Born rule is one of the postulates of orthodox QM.

    It doesn't come out naturally; it is specified by hand w/o any reasoning except for phenomenology.

    No, I don't agree.

    There are several step: First you postulate a measurement differing from ordinary dynamics w/o further explanation when it does apply, i.e. when a system will evolve according to unitary time evolution and when it will not; this is essentially some collaps. Second you specify the probability measure associated to the collaps, i.e. that when collapsing |ψ> into some |n> you find p(n) = |<n|ψ>|2.

    So the challenge is to understand why there appears to be a collapse at all, what a measurement is, what distinguishes a measurement situation from ordinary time evolution for the same apparatus which usually follows ordinary time evolution, why a collaps appears, why we should associate a probability, and which specific probability measure.

    None of these questions are addressed in orthodox QM.

    There is no irreversible decoherence in quantum mechanics w/o introducing it by hand. The overall dynamics remains unitary (just as in statistical mechanics). And there cannot be any non-unitary part w/o contradicting with ordinary = unitary time evolution. Yes, decoherence somehow explains an "apparent collapse", but it does not explain what happens with the unobserved branches, and why. The only solution to me is that they stay real but unobserved, b/c this is what the theory predicts.

    Whether you introduce a collaps w/ of w/o decoherence makes no fundamental difference. Decoherence solves a part of the problem only.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook