Schrödinger's cat

Discussion in 'Physics & Math' started by fess, Jan 30, 2020.

  1. Schmelzer Valued Senior Member

    Messages:
    5,003
    I name it pseudo-science because in my understanding it does not predict anything.
    The worlds, for example. What defines, what identifies a world? Silence. Which structures are defined in their Hilbert space? Silence. The Copenhagen interpretation has a large amount of structure, because it has a classical part, and all the structure of the classical part can be used. So we have a configuration space, momentum space, and various operators to measure them. MWI has a Hilbert space, some Hamilton operator, and a state. This is essentially nothing, no base for deriving even a single physical prediction. Something else? Silence.
     
  2. Google AdSense Guest Advertisement



    to hide all adverts.
  3. Halc Registered Senior Member

    Messages:
    350
    That makes it an interpretation, not pseudo-science. MWI is not unique in this respect since none of them make any empirical predictions.

    Worlds don't have identities. You're using classic language and premises in non-classic interpretation, so the questions don't make particular sense. I have no idea how to parse the 2nd question. Silence is probably a good answer to word salad questions. No, I'm not an MWI adherent.

    Copenhagen is an epistemological interpretation, not a metaphysical one like most of the others. It describes what an observe might know about a system and does not actually attempt to describe the system. That said, many people don't understand that and draw metaphysical conclusions from the language.

    I could not follow any of this. Are you saying Copenhagen makes predictions that MWI doesn't? MWI does not allow a measurement of the momentum of an object for instance?
     
  4. Google AdSense Guest Advertisement



    to hide all adverts.
  5. Schmelzer Valued Senior Member

    Messages:
    5,003
    You misunderstood. Interpretations make no additional predictions in comparison with the interpreted theory. MWI makes no predictions at all. Everything happens. They claim to derive the Born rule. I have taken a look at the "proofs", they prove nothing.
    If worlds have no well-defined meaning, then MWI adherents should not talk about them.

    The word salad comes from MWI side. They claim to prove something using decoherence. Decoherence presupposes some subdivision of the world into a system and an environment. So, it presupposes an additional structure - some subdivision of the world into subsystems. Either such a subdivision exists on the fundamental level - in this case, the MWI guys have to define it precisely as an additional structure given on the fundamental level - or they cannot apply decoherence to prove something on the fundamental level.

    Some (Zurek) at least recognize the problem:
    The situation is much different for Copenhagen. In Copenhagen, the subdivision into systems is defined by the classical part. So they can use decoherence techniques whenever they like.
    Yes. There is no fundamental base, no fundamental structure in MWI which defines a configuration space and the corresponding position and momentum operators. They talk about some "preferred basis problem" and how they solve it, and the result of this solution should be the canonical operators, not postulated but derived. The derivation is, unfortunately, BS.
     
  6. Google AdSense Guest Advertisement



    to hide all adverts.
  7. Halc Registered Senior Member

    Messages:
    350
    Your intent is to be blind, so you see nothing. I can't change that. All the interpretations make the same predictions. MWI does not assert that everything happens.

    I didn't say that. I think MWI has problems, but not the problems you manufacture. RSF has far fewer problems, and none of them empirical.
     
  8. CptBork Valued Senior Member

    Messages:
    6,460
    But the evidence we have today basically all but proves that either all the different possible outcomes occur in separate universes, or a single outcome occurs based on a probabilistic selection that has not been predetermined. To date there's no viable hidden variables theory which fits even a fraction of the existing evidence, and whatever possibility remains for such a theory to be constructed is extremely limited by tight experimental constraints. When the supernova explodes, the quantum probabilities for it to have happened within a very narrow time frame of when it actually happened were nearly 100%, but that doesn't mean one can anticipate the exact state of every atom in the supernova once it blows without a direct measurement after the fact, even with perfect knowledge of the entire system and every particle in it beforehand.
     
  9. Schmelzer Valued Senior Member

    Messages:
    5,003
    Ok, add "almost". A wave function in general position is nonzero almost everywhere. If it is nonzero, given that nobody has yet defined some minimal probability, the corresponding world exists. Not?

    And, please, don't speculate about my intent, present arguments instead of excuses for not presenting them. If I don't see something, give a link to where it is defined in detail.
    Then, fine, show the solution for those "manufactured problems", it should be easy, if you, say, try to manufacture problems in QT itself or other interpretations, those who know these interpretations will simply give short adequate answers.

    As far as one can interpret the RS formulation as a variant of formulation of the Copenhagen interpretation, it may be, indeed, without serious problems. Roughly, one uses everything the Copenhagen classical part delivers without questioning it, then moves the classical-quantum split so far away that humans will be in the quantum part, and simply does not talk about the classical part anymore. The same technique of RQFT, don't talk about the problems like BI violations and they disappear.

    The problems appear if one wants to create an independent interpretation, which defines everything starting from some fundamental axioms. Then all what can be taken in a Copenhagen reformulation from the classical part without talking about that has to be defined explicitly.
     
  10. Halc Registered Senior Member

    Messages:
    350
    I'm not very familiar with DeWitt's MWI, but it seems contradictory to correspond a wave function to a world.
    But yes, if some outcome has a nonzero probability of being measured, then I think DeWitt would say that corresponding worlds must exist. If you have problems with that, then you're not alone.

    You bring up a 'wavefunction of a location' which is an interesting way to frame the fact that a missile can turn into a bowl of petunias, in any interpretation, with not even ridiculously low probability.

    I don't know enough about MWI to answer your question in a way that they probably would. RSF doesn't posit worlds, so that answers the question nicely for that interpretation.

    No, it isn't a variant of Copenhagen. At best, MWI is a variant of RSF. It is an alternative to Copenhagen if you will, and then only if Copenhagen is taken as a metaphysical interpretation, which it never was meant to be.
     
  11. Schmelzer Valued Senior Member

    Messages:
    5,003
    Ok, that's possibly the wrong word. I have cared about the structures it presumes to be given. In RSF I have not identified a place where Everett cares about where all those object/structures we obtain in the Copenhagen interpretation from the classical part will come from if there is no classical part. He simply uses them as they could be used in Copenhagen if we decide to include almost everything into the quantum part of the split.

    So, my point is the following: If one interprets RSF as simply a formulation, instead of a different interpretation, then there would be nothing to object against RSF. But if it is a separate interpretation, then it has to define everything it uses itself, and cannot use all what the Copenhagen interpretation can simply take from the classical part.
     
  12. phyti Registered Senior Member

    Messages:
    732
    Moving back in time before the human species appeared, there was no 'wave function', yet the world was existing. It still is a question of human comprehension of how the world works.
     
  13. CptBork Valued Senior Member

    Messages:
    6,460
    I don't understand what you're trying to imply here. The whole idea behind quantum theory is that everything in the universe is described by wave functions, and you can even combine them into a wave function describing the entire universe as a whole, with or without human beings around to observe it. The wave function has a reality of its own, it doesn't need to collapse before it acquires that reality. Indeed, the Heisenberg uncertainty principle proves that it's impossible to make a wave completely collapse into a definite state, measurements just narrow down the range of possible states contained in the wave.
     
  14. Halc Registered Senior Member

    Messages:
    350
    Nearest I can get to your use of the language is to say that a measured state is classic, and an unmeasured state is not, same as any interpretation. A state unmeasured is in superposition of any valid solution of the corresponding wave function. This macrosuperposition is the quantum state to which you refer, and there's no limit to the scale of it. The moon is in superposition of exists/not-exists if it has not been measured by some system. In this regard, the wording is very similar to RQM (Rovelli), my favored interpretation. If the measurement is made (the loss of coherence), the observed system (the moon say) becomes classic to the system that measured it. The measuring system is now in superposition of having measured the various states, so it only measures classic states.

    Clarify please. What's a formulation? I don't know how to respond to your intended meaning of that word.

    An example where you feel it doesn't might help.
     
  15. phyti Registered Senior Member

    Messages:
    732
    The universe was existing without human concepts, before humans appeared. It seems they think their concepts are real when their measurements match reality. For me, the concepts are just useful tools/models that mimic the physical world. The simultaneous position-momentum measurement is not possible in Newtonian physics either. So what's new!
     
  16. CptBork Valued Senior Member

    Messages:
    6,460
    The whole point of a theoretical model is to suggest that there's some hidden underlying reality which obeys certain rules to at least extremely good approximation. All the existing evidence suggests that everything in the universe has obeyed rules such as those postulated by Quantum Mechanics and Relativity since the very beginning of the universe as we know it. Whether it's purely a coincidence or whether these theories do indeed reflect an underlying reality, no mainstream physicist is claiming that the universe only started acting this way once humans arrived on the scene and started postulating theories.

    Among many other things, one of the "new" aspects found in modern physics is that the Heisenberg uncertainty principle always requires that the product of the standard deviations of the position and momentum wave functions must always be equal or greater than a certain value, which is actually a property that holds for any pair of functions related by a Fourier transform, position and momentum wave functions being only one pertinent example. It's true that in Newtonian physics, one must disturb a particle's momentum in order to measure its position, and one must disturb its position in order to measure the momentum, but there's no classical limit on how small and insignificant these disturbances can be, thus there's no classical limit on how precisely one can simultaneously measure both values. In quantum mechanics there's a constraint on how little you can disturb the momentum while measuring position and vice versa, so that the products of the position and momentum standard deviations always exceed a certain value, as claimed.
     
  17. Schmelzer Valued Senior Member

    Messages:
    5,003
    "Formulation" I have simply taken from the paper itself. But it makes sense to distinguish formulations from interpretations. If you have different formulations, you can simply translate both formulations into each other. So, a different formulation of the spacetime interpretation would be to rename "proper time" into "clock time". It would be a shift toward the Lorentz ether which distinguishes true time and clock time, with the latter being a time measurement distorted by the ether. In this sense, the RS formulation could be understood as a combination of such a translation from Copenhagen and simply not talking about some parts (the classical part of Copenhagen).

    An interpretation would, instead, be a different theory, even if the difference is only metaphysical, does not (or not yet) lead to different empirical predictions.
    Relative interpretations are always cheap, and a step in the wrong direction. Look at them from the potential of additional empirical predictions. Realist interpretation postulate some additional structure. It is, immediately, not observable. But it has some potential for observation. Last but not least, the actual theory may not be the final one, but only an approximation, and in the better theory it becomes observable. Or they define a global structure. Locally unobservable, this may have nonetheless global consequences. So, absolute space and time may not be observable. But if they exist, they forbid wormholes and causal loops, and have a preference for a flat universe which we observe in the large structure.
    So, relative interpretations simply get rid of that additional structure of realist interpretations. So, the potential of the additional structure to become observable, even if only in some indirect way, is simply thrown away, without giving any advantage (beyond compatibility with positivist philosophy which is known to be false). And relative interpretations are cheap because it is quite clear how to create it if there is a realist interpretation.
     
  18. Halc Registered Senior Member

    Messages:
    350
    That doesn't seem to be the case between say Copenhagen and something like RSF. At least they're both local, but the former posits for instance a Heisenberg cut, and there seems to be no equivalent in RSF, so I don't see how any translate can be done back and forth. Hence your questions about the distinction between classic and quantum behavior being met with unsatisfactory answers. There just isn't always a distinction.

    Hence I call them interpretations. For really different metaphysics, go with something counterfactual like what Bohm suggests, the most realist interpretation in the bunch AFAIK.

    Cheap is good. Positing extra needless premises seems to increase the probability of being wrong. I recognize your statement as mere opinion.

    They all predict the same things, so this doesn't hold water.

    If they predict some distinction someday, then all the interpretations predict something new, and some of them will be falsified. The realist interpretations are no different in that respect. Surely you must realize this.

    I don't recall absolute space/time being unique in any of these.f

    I'm still unclear as to how you're using 'structure' here. Yes, I agree that some interpretations (not only relative ones) are simpler. I suspect that's not what you mean. It's hard to figure out looking at the way you use the word in context. Subdivision of the world into subsystems, such as what the Heisenberg cut does. Yes, there is no such subdivision is something like RSF. I find that to be a point in its favor, not against.

    BTW, I'm not sure on Copenhagen's modern take on the cat. The interpretation has evolved since Bohr's contributions, but Bohr never posited the cat being in superposition. The cat was in a classic state that was simply unknown to the observer outside the imperfect box. But modern models consider a perfect box where the mathematics show the cat being in superposition to be a valid state of the closed system. Copenhagen as a metaphysical interpretation might have a hard time with the quantum state still being in superposition despite being measured by the seemingly classic cat.

    You're saying there's not much work to do if some realist interpretation has already done all the work? I imagine so, yet, just like the realist interpretation would be cheap if the relational interpretation was worked out before it. None of them need to break ground from scratch. Doesn't mean any of them are wrong because of that. RSF and RQM for instance are almost the exact same thing (just a difference in formulation as you say) except for the trivial designation of what being real means.
     
  19. Schmelzer Valued Senior Member

    Messages:
    5,003
    Yes. But RSF can be understood as a combination of formulation and omitting a part of the Copenhagen interpretation. That's the "and simply does not talk about the classical part anymore" of that idea.
    "Cheap" means there is a quite simple algorithm to create a relative interpretation if you have an absolute one. Simply identify among the propositions those which are observable and reject everything else. Ok, this is a gross oversimplification, given that you can never get rid of metaphysical propositions in physical theories (almost all propositions taken alone are metaphysical, that means, not falsifiable taken alone, you need the whole theory, and usually even several theories, to obtain falsifiable predictions). But I hope you get the point.

    To increase the probability of being wrong is a good thing. In general, this is Popper's criterion of empirical content, in the relative form: If theory A has more falsifiable predictions, it is preferable for having higher empirical content. In the case of interpretations, this is not that clear. The additional structures may not lead to additional empirical predictions.

    But I know already different examples where what seemed to be different interpretations of the same theory later appeared to be different theories.
    1.) Minkowski's spacetime and the Lorentz ether. They became different with Bell's theorem: You can prove the Bell inequalities in the Minkowski spacetime interpretation, but not in the Lorentz ether, where FTL causal influences are possible without violation of causality.
    2.) The generalization of the Lorentz ether to the Einstein equations of GR. Locally it is always possible. Globally not. The Lorentz ether additionally predicts trivial topology and no causal loops.
    3.) QT and Nelsonian stochastics. Nelson proposed his stochastics as an interpretation of QT. Then came the Wallstrom objection, and taking it into account it is an empirically different theory. Nelsonian stochastics additionally requires that the wave function has in the configuration space representation no zeros.

    This holds only at a given situation. I interpret interpretations mainly as research programs for more fundamental theories. Different interpretations suggest different places to search for better theories. With such developments, they may become different theories.
    No, I must not realize this, because it is simply wrong. You completely ignore the possibility that one interpretation makes additional predictions but the other interpretations predict nothing new. All three examples above are of this type.
    The spacetime interpretation does not have such restrictions, and wormholes and causal loops are seriously discussed in various papers about GR. How many other interpretations of the Einstein equations you have considered?
    No, the problematic thing is not the subdivision of Copenhagen into classical/quantum, but that into observer/system/environment. You need it to start decoherence methods. You don't have it on the fundamental level, so you cannot apply decoherence at the fundamental level. In my reformulation + restriction interpretation of RSF, you take it from the classical part and restrict yourself to name the place where you have taken it.
    No. Compare my Lorentz ether interpretation with the spacetime intepretation. I have harmonic coordinates as preferred coordinates. I give the various components of the gravitational field an interpretation, say, $g^{00}\sqrt{-g}$ I interpret as the ether density. These were a lot of different things which I had to do before the interpretation made sense as an ether interpretation. (These are the sort of things I name "structure".)

    Now do the same in the other direction. All you have to do is to forget all the additional things I have added to the spacetime interpretation. Cheap.

    Or compare dBB with the minimal interpretation. Both have the Schroedinger equation. But dBB has also an additional continuous trajectory, $q(t)$, and an evolution equation for it, which contains a velocity, the Bohmian velocity. These have been quite nontrivial additional objects, additional structure. It may not always exist. Say, one could not simply use, instead of $q(t)$, a continuous momentum space trajectory $p(t)$, simply because there is no continuity equation for the momentum space representation.

    In the reverse direction, all you have to do is to forget the additional structure, the continuous trajectory $q(t)$ and its velocity. Cheap.
     
  20. arfa brane call me arf Valued Senior Member

    Messages:
    7,832
    Back to the cat. All it illustrates in one sense, is that it doesn't matter if you use a hidden observer like a cat. If you leave the cat out of the chain of (classical) events, the glass vial is left to record the state of the radioactive sample.
    An equivalent experiment is determining if a single photon is reflected or transmitted by an optical beam splitter.
    The photon is in a "cat state" until it's detected (so not very long, then)

    Besides, a cat state is a one-particle system; you have a Riemann sphere's worth of quantum information, as Penrose might say.
    It's easy to map a measurement (basis) to the Bloch sphere; the result of a measurement is always a binary value.
     
  21. Halc Registered Senior Member

    Messages:
    350
    Or Copenhagen could be understood as a combination of formulation and adding new parts to RSF. I don't see the point. It's not like Copenhagen is the default starting point, especially since it, unlike the others, isn't even a physical interpretation, but only an epistemological one.
    I brought up RQM, which also has that Heisenberg cut, a distinction between quantum and classic.

    And v-v perhaps, or perhaps not. Do you consider RSF to be a relative one, since that's what the R stands for? Because I don't. It paints a picture of one absolute and hard deterministic universe, similar to Bohm in that respect.

    The point of any interpretation is to go beyond that. Anything observable are part of the theory (QM), and not interpretational at all.

    Agree. Part of it is the methodology of science: methodological naturalism and methodological realism, both metaphysical propositions.

    Are the things in the list predictions? Can a causal loop be demonstrated to be impossible, and if it was, would that invalidate GR?
    I am, BTW, unfamiliar with the Lorentz predictions on point 2.

    Reference please. Minkowski spacetime itself seems not to be sufficiently comprehensive to satisfy the conditions imposed. Apparently, only once certain theories (non-local ones in particular) are formulated using Minkowski spacetime as a background, the theories must invoke a preferred foliation.
    I'm at the edge of my knowledge here, so what I say could be BS, but I am unaware of Minkowski spacetime forbidding FTL causal influences, especially since there are plenty of experiments that can be interpreted as effect blatantly (not ambiguously) before cause.

    Anyway, point taken on the theories having more than just interpretational differences, even barring empirical falsification tests. Were that not so, there would be no real reason to favor one interpretation over another, and my choice was made on just just a line of thinking.

    Alright, I get your point.

    I've not done any such subdivision for my view. Maybe system, but the other two are not defined. Maybe they are. Not sure how you define them.

    I name them needless additions.

    Got it. I like cheap. The Ptolemaic system had all the 'structure' until the Copernican system found a cheap way to explain all the same things.
     
  22. Schmelzer Valued Senior Member

    Messages:
    5,003
    No. The point is that RSF taken alone is not a valid interpretation, it does not contain all the structure which Copenhagen takes as given and well-understood from classical observation of the classical part.
    I have given an own interpretation of RSF given of the paper alone, and the requirement that it should make sense as an interpretation, despite my objections against MWI. I would not pretend at all that this has much to do with RSF as understood by others.

    If one tries to interpret it in a hard way, it has to define everything in the hard way, the whole structure which is required to obtain something similar to the world we observe around us. Which is not done in the RSF paper.
    Correct. But those who prefer relative interpretations usually do not agree. Instead, they usually follow some sort of positivist thinking, and for them to add something is evil.
    No wonder, this is my own invention and essentially ignored. The Lorentz ether interpretation of the Einstein equations is the limit of $\Xi,\Upsilon \to 0$ of arxiv:gr-qc/0205035. See https://ilja-schmelzer.de/ether/ for an overview.

    Once $\rho = g^{00}\sqrt{-g}$ in the preferred coordinates is interpreted as an ether density, it has to be positive everywhere. But $g^{00}\sqrt{-g}>0$ means the time-coordinate is time-like. Thus, the preferred time is time-like everywhere. Thus, no causal loops are possible. This cannot invalidate GR, given that it is an additional requirement of the interpretation. It allows, instead, to falsify the ether interpretation without falsifying GR, by observing some causal loops.
    Not sure what you want, the EPR paper together with Bell's theorem would be the standard reference to what follows from Einstein causality. If one interprets BI violations as "effect blatantly (not ambiguously) before cause", then Einstein's notion of Einstein causality, as used in the EPR paper, is simply dead. What remains is much weaker signal causality. But if you allow for a preferred frame, and do not forbid causal influences as long as they don't causally influence the absolute past as defined by the preferred time, then you will fail to prove the Bell inequalities.
    There may be variants, but some subdivision is necessary.
    They are needless for a positivist. They are necessary for a realist looking for an ether interpretation. And they give some additional predictive power.
    Except that throwing these parts away destroys a lot. Not only the additional predictions like no causal loops and trivial topology. For example, it destroys also the straightforward way to quantum gravity. The one described in every textbook of quantum condensed matter theory, and which can be used if you have a completely classical Newtonian space-time background with some ether described by usual condensed matter equations (continuity and Euler equations).
     
  23. Saint Valued Senior Member

    Messages:
    4,752
    This cat is a bullshit hypothesis,
    why not you ask Does the cat itself know it is alive or dead?
     

Share This Page