Insights 9 Reasons Quantum Mechanics is Incomplete - Comments

[Adur Alkain]

Quantum mechanics is not incomplete. It is a complete description of observation. Physics does nothing but describe observation. The laws of physics are the laws of observation.

Interpretations of quantum mechanics belong to philosophy, not to physics. It is philosophy that has failed so far, not physics. (I'm a philosopher, not a physicist, btw.)
In other words, "shut up and calculate" is good advice for physicsts. It's a preposterous advice for philosophers (but that's what logical positivism did, basically).

One doesn't have to be a professional philosopher to see the clear picture that quantum mechanics gives about the nature of reality. Einstein wasn't a philosopher, and yet he clearly saw the philosophical implications of quantum theory (in its Copenhagen version) and asked the candid question: "Do you really think the moon isn't there if you aren't looking at it?"

Einstein used this philosophical argument as an objection to the Copenhagen "interpretation" (which as you very well argued in "Against "interpretations"", should rather be called the Copenhagen theory): his contention was that the theory must be faulty. The theory, as far as I know, has been experimentally proven correct.

The philosophical conclusion is unavoidable: there is no physical world independent of observation. Quantum mechanics implies Idealism.

But again, this whole discussion lies on the realm of philosophy, not physics. (All alternative theories to Copenhagen can be understood as attempts to avoid Idealism, but as far as I'm aware they have failed so far.)

Yet one can imagine that most physicsts won't welcome the realization that physics is nothing but a specialised field of psychology: the study of observation (understanding observation as a special type of conscious experience that is bound by the laws of physics - the laws of observation).

This is a typical "the Emperor has no clothes" scenario, which I find rather amusing.

 

[bhobba]

Quantum mechanics is not incomplete. It is a complete description of observation.

It's a description of observations that occur in a classical world. But since everything is quantum how does a theory that assumes a classical world explain that world? That is the issue. A lot of progress has been made in resolving it, but it is not fully resolved yet. This means QM may or may not be complete - we just do not know yet.

Thanks
Bill

 

[ftr]

Quantum mechanics is not incomplete. It is a complete description of observation.

QM is a physics theory and theories in physics must include predictions and since it does not predict everything( like mass and couplings), hence it is incomplete.

 

[DarMM]

QM is a physics theory and theories in physics must include predictions and since it does not predict everything( like mass and couplings), hence it is incomplete.

I think this mixes up the incompleteness of the Standard Model (mass and coupling parameters) with the incompleteness of Quantum Theory in general (measurement problem, meaning of contextuality).

 

[ftr]

I think this mixes up the incompleteness of the Standard Model (mass and coupling parameters) with the incompleteness of Quantum Theory in general (measurement problem, meaning of contextuality).

SM is part of QM, measurement problem is an "interpretation" issue but mass and coupling is real/clear issue. I think understanding will come when the later is addressed first.

 

[microsansfil]

Hi,

is it not possible to verify the completeness of any mathematical axiomatization of quantum mechanics?

http://www.iecl.univ-lorraine.fr/~Wolfgang.Bertram/WB-TB.pdf

/Patrick

 

[Demystifier]

I think this mixes up the incompleteness of the Standard Model (mass and coupling parameters) with the incompleteness of Quantum Theory in general (measurement problem, meaning of contextuality).

In general that's true, but in my view of Bohmian mechanics those two types of incompleteness are closely related. See the paper linked in my signature.

 

[bhobba]

is it not possible to verify the completeness of any mathematical axiomatization of quantum mechanics?

It is not possible to verify the axiomatic completeness of any system as strong as arithmetic. Note - this is verifying completeness in the Godel sense because we always have true statements that can't be proven in the system. But that has nothing to do with if a system is not complete because its axioms may contain a hidden circularity with regard to observations, which is what people that talk about QM completeness mean. You are falling for a common semantic error - because two different things use the same name you context shift. Such reasoning is invalid.

Thanks
Bill

 

[DarMM]

SM is part of QM, measurement problem is an "interpretation" issue but mass and coupling is real/clear issue. I think understanding will come when the later is addressed first.

The measurement problem is a problem in QM that interpretations try to solve, but it is a real issue as much as the masses and couplings. It might be the case that the explanation of couplings and masses is related to the measurement problem or it might not, from where we stand now they are two separate problems.

 

[microsansfil]

It is not possible to verify the axiomatic completeness of any system as strong as arithmetic.

Some examples of complete theories are: https://en.wikipedia.org/wiki/Complete_theory

/Patrick

 

[microsansfil]

I use it colloquially.

How do we measure the distance between physical theory and the ontology of nature, without having access to it? So, therefore, any physical theory will be incomplete?

/Patrick

 

[lowlize]

"There is an objective reality out there". I don't think that applies to the relational interpretation. And what about the "new" Copenhagen interpretation with decoherence taken into account?

 

[Lord Jestocost]

Regarding “the incompleteness of QM interpretations”, there is, to my mind, a subtle hint given by David Bohm and Basil J. Hiley in “The Undivided Universe: An Ontological Interpretation of Quantum Theory”:

"Several physicists have already suggested that quantum mechanics and consciousness are closely related and that the understanding of the quantum formalism requires that ultimately we bring in consciousness in some role or other (e.g. Wigner [17], Everett [18] and Squires [19]). Throughout this book it has been our position that the quantum theory itself can be understood without bringing in consciousness and that as far as research in physics is concerned, at least in the present general period, this is probably the best approach. However, the intuition that consciousness and quantum theory are in some sense related seems to be a good one, and for this reason we feel that it is appropriate to include in this book a discussion of what this relationship might be." [emphasis added by LJ]

 

[jambaugh]

There is one reason that quantum mechanics is a complete physical theory. Quantum mechanics makes a class of predictions. Those are probabilistic predictions but since you can embed logic within probability by restricting to specific cases of p=0 or 1, this is not a failing on the part of QM. And quantum mechanics also specifies that there are no stronger predictions that can be made. The uncertainty principle defined the end of the question of what more can be known about what will happen.

So any improvement upon the predictions of QM must necessarily violate QM. That's fine, GR breaks SR except as a limiting case. SR breaks Newtonian mechanics except at a limiting case. QM is not necessarily the end of all theories but it is complete in the sense that no more exact theory can be given that incorporates all of QM's predictions.

Now if you feel it is "incomplete" because it doesn't provide for a specific observably distinguishable ontological model then I would point out to you that QM is not an ontological theory. It is a description of what happens (and how often) not of what is. That's the mistake people make in misinterpreting Copenhagen. Yes the wave function collapses but the wave function is not an ontological representative. It is a QM version of a classical probability distribution. Classical distributions collapse all the time.

The dissatisfaction people have with the Copenhagen interpretation which is the positivistic interpretation is the same sort of dissatisfaction they have with the non-absolute time in the Twins paradox which makes them one to throw in an aether (with necessarily screwy dynamical properties) in order to feel better about the relative ages of twins. The relativity of time and simultaneity is likewise the positivistic interpretation eliminating as meaningless that which cannot be observed, namely the luminiferous aether.

There is no fixed objective reality in QM. There are a set of relative frame dependent realities with well defined relativity transformations relating them. Fantasies about pilot waves and infinite worlds are no more helpful than questions about the unobservable properties of the aether.

 

[Avodyne]

Yes the wave function collapses

What is the mathematical description of the collapse process?

Let's say we're going to do a Stern-Gerlach experiment. We isolate the lab completely from any external environment. We put the lab (and everything in it, including the experimenters) in an initial pure state, with the experiment about to begin. We time-evolve forward unitarily with the hamiltonian of the Standard Model.

I claim that the state of the lab splits into an "up" branch (where the measured spin was up, the dial points up, the brains of the experimenters in the lab think they saw the dial point up, etc) plus a "down" branch. (I believe this part must be correct, but maybe you disagree.)

So then, when does collapse occur? And what is the mathematical description of it?

 

[DarMM]

What is the mathematical description of the collapse process?

If you look at @jambaugh 's post he's clearly taking an epistemic view of the wavefunction where collapse is just (generalized) Bayesian conditioning.

 

[Avodyne]

I don't see why it matters whether one views the quantum state as epistemic or ontological. QFT predicts a two-branched state for the post-measurement Stern-Gerlach lab (and the people in it). At what point in time, and according to what mathematical rule, do we throw away one branch?

 

[DarMM]

I don't see why it matters whether one views the quantum state as epistemic or ontological.

If the quantum state is epistemic then throwing away one branch doesn't correspond to a physical process which needs a more detailed mathematical description, it's simply (generalized) Bayesian updating.

 

[Avodyne]

What is "(generalized)" Bayesian updating? (I know what ordinary Bayesian updating is.) Do you have a reference?

 

[DarMM]

It's just how Lüders rule is viewed in epistemic approaches, i.e. it represents a generalized form of Bayesian conditioning.

 

[Avodyne]

Lüders rule, as I understand it (see below), seems to require an observer outside the system being measured. This is not the case in QFT, where observers and everything else are comprised of excitations of the quantum fields.

http://philsci-archive.pitt.edu/4111/1/Lueders_rule_BuschLahti.pdf

 

[A. Neumaier]

Lüders rule, as I understand it (see below), seems to require an observer outside the system being measured. This is not the case in QFT, where observers and everything else are comprised of excitations of the quantum fields.

http://philsci-archive.pitt.edu/4111/1/Lueders_rule_BuschLahti.pdf

Well, for quantum field theory you may need the thermal interpretation, where observers can be part of the system and collapse happens as an approximate effect due to the ignored environment.

 

[DarMM]

Lüders rule, as I understand it (see below), seems to require an observer outside the system being measured. This is not the case in QFT, where observers and everything else are comprised of excitations of the quantum fields.

http://philsci-archive.pitt.edu/4111/1/Lueders_rule_BuschLahti.pdf

How does this affect an epistemic view? Even if you include the observer themselves within the quantum description there is no need for the quantum state to be a physical/ontic object and thus no need for a dynamical or mathematically detailed description of where the other branches "go".

For example in a dice roll there are probabilities for each roll $p(1)$, $p(2)$ and so on. I could also include the observer in this:
$$p(1, Observer\hspace{2pt} sees \hspace{2pt}1)$$
$$p(2, Observer\hspace{2pt} sees\hspace{2pt} 2)$$

However there is no need to explain where $p(2, Observer\hspace{2pt} sees\hspace{2pt} 2)$ "goes" upon the outcome $1$.

Since observers can be included within the system even in QM, this isn't particular to QFT.

 

[Avodyne]

How does this affect an epistemic view?

Because you still need rules to decide when to modify the state by chopping off branches. If you do it under the wrong circumstances, you will destroy observable quantum coherence effects.

Psi-epistemic interpretations have all sorts of other problems as well:

https://arxiv.org/abs/1303.2834

 

[DarMM]

Because you still need rules to decide when to modify the state by chopping off branches. If you do it under the wrong circumstances, you will destroy observable quantum coherence effects.

You remove the branches related only to unobtained outcomes for the systems you actually observe. This is not in contradiction with a superobserver retaining a superposed description of your entire lab. See for example Richard Healey's "The quantum revolution in philosophy" Chapter 11 for a good summary of this.

Psi-epistemic interpretations have all sorts of other problems as well:

https://arxiv.org/abs/1303.2834

That paper concerns psi-epistemic views where there is an underlying ontic space obeying the ontological framework axioms, just like the PBR theorem. It says nothing about epistemic theories in general such as for example acausal views. It's not really an issue for any of the epistemic views actually being worked on. Especially ones where there is no underlying ontic hidden variable space.

 

[stevendaryl]

I consider orthodox quantum mechanics to be incomplete (or maybe inconsistent) and I don't think it really makes any difference whether you consider the wave function to be epistemic or ontological. Wigner's friend or Frauchiger and Renner shows this incompleteness.

Rather than going through the F&R argument again, let's just take something much simpler. Alice and Bob prepare an electron so that it is a superposition of spin-up and spin-down. Alice measures the spin. According to orthodox quantum mechanics, she either gets spin-up with such-and-such probability, or she gets spin-down with such-and-such probability.

But Bob, considering the composite system of Alice + electron, will predict that that composite system will evolve into a superposition of "the electron is spin-up and Alice measured spin-up" and "the electron is spin-down and Alice measured spin-down". From Bob's point of view, Alice doesn't get a definite result, but somehow turns herself into a superposition of both results.

So it seems to me that either you're led to (1) the "fantasy" that is many-worlds, or (2) quantum mechanics is inconsistent, or (3) somehow Alice being in a superposition of states doesn't preclude her being in a definite state.

The latter possibility is realized by the Bohmian interpretation, but I would say that if you take that way out, then you are admitting that standard quantum mechanics is incomplete, because the Bohmian interpretation has additional variables (the actual positions of particles) that are not present in standard QM.

Calling the wave function "epistemic" rather than "ontological" doesn't actually accomplish anything, in my opinion. So you want to interpret Alice in her superposed state as being epistemological? Does that mean that she's ACTUALLY in some definite state, and we just don't know which? The problem with that is that there is actually a difference between "Alice is in this state or the other, we just don't know which" and "Alice is in a superposition of this state and that state". I don't think you can consistently treat the latter as epistemological. In any case, if you say that the superposition reflects your lack of information about the "true" state of Alice, that to me means again that quantum mechanics is incomplete, since it presupposes a "true" state.

 

[jambaugh]

Parsing your description of the classic thought experiment, StevenDaryl

Alice and Bob prepare an electron so that it is a superposition of spin-up and spin-down. [\quote]
fine so far.

Alice measures the spin. According to orthodox quantum mechanics, she either gets spin-up with such-and-such probability, or she gets spin-down with such-and-such probability.

Yes and so you should then taking into account that she made this measurement and recorded her result in a classical copiable memory by describing Alice + electron via a density operator:
\rho =\rho_A\otimes \left( \begin{array}{cc} 0.5&0\\0&0.5\end{array}\right)

But Bob, considering the composite system of Alice + electron, will predict that that composite system will evolve into a superposition of "the electron is spin-up and Alice measured spin-up" and "the electron is spin-down and Alice measured spin-down".

If Bob does this he is making a grave mistake. Alice is not at zero entropy and so cannot be described sharply. He, if he wishes to describe Alice plus electron before the physical and dynamic act of measurement Alice will make occurs would presumably use a density operator to include the entropy dump Alice necessarily must utilize to amplify the electron's spin component into a macroscopic record.

From Bob's point of view, Alice doesn't get a definite result, but somehow turns herself into a superposition of both results.

From Bob's point of view, if he has correctly represented the measurement process of Alice, if he correctly represents the dynamic that makes her recording of the electron spin possible, he will end up with a density operator of the form I mentioned above. In short Alice as an observer necessarily must interact with the system entropically and decoher it. At this point Bob is not describing this Alice but any such Alice electron pair given the same or equivalent initial conditions. Then Bob classically measures the record of Alice and collapses the classical probability distribution down to a single classical sharp state just the same way that the LOTTO commission does each time they bounce their ping pong balls and collapse the value distribution over the tickets to sharp "win" or "lose" states.

 

[DarMM]

Well first of all after going through the paper more I don't think FR adds anything beyond Wigner's friend.

Secondly an epistemic account will by nature be taking the view that QM is incomplete as it views a central object in it as non-representational. What the epistemic approach is supposed to add is not a complete and closed account of the ontology of nature (explicitly because they are epistemic). Rather they seek to provide an explanation of several of the properties of quantum states by saying that they are epistemic objects. So for example teleportation, local indistinguishability, no cloning, remote steering, interference, non-commutativity are viewed as having a more natural explanation if you understand $\psi$ as epistemic.

However saying that these properties of $\psi$ have a more natural explanation if $\psi$ is a sort of generalized probability distribution, does not mean you're claiming to know the fundamental ontology of nature. In fact you're explicitly not, since you are providing an epistemic account.

The two main classes of epistemic views then differ on what the "completion" of QM will be like. Roughly speaking $\psi$-statistical views will see the underlying theory as having some strange ontology like retro or acausality. $\psi$-epistemic/doxastic views like QBism, Bub or Copenhagen view a completion as impossible for reasons that differ between them.

However to reiterate, by their nature they deny that QM is a complete representational account of the world. The simply seek to explain the properties of $\psi$.

I don't think you can consistently treat the latter as epistemological

Spekken's toy model gives an epistemic set of states that have interference at the observer + system level, while having a definite outcome at the system level. Superpostion of the total system isn't incompatible with definite outcomes for subsystems. If it was epistemic views would be finished.
To explain briefly in Spekkens toy model superposition arises from the structure of the space of probability distributions under the presence of an epistemic limit. When considered over "system + environment" that limit has a different form than over just "system", so we have a superposition for the former, but a definite outcome for the latter.
Actual epistemic views are more complicated than this, but it gives a good example of how superpositions are compatible with definite subsystem outcomes.

 

[stevendaryl]

Yes and so you should then taking into account that she made this measurement and recorded her result in a classical copiable memory by describing Alice + electron via a density operator:

No, you shouldn't. That's introducing an ad hoc element to quantum mechanics. Through unitary evolution a pure state never turns into a mixed state. It becomes a mixed state through a modeling decision on OUR part. We can trace over unobservable degrees of freedom and produce a mixed state, but as I said, that's something WE do. It's not a physical effect.

 

[jambaugh]

Let me add again, the collapse is not mysterious for classical probability distributions because they too are fundamentally (in the frequentist's intepretation of probability) phenomenological descriptions. One is not describing the state of a six sided die, one is describing the behavior of that die when tossed. The phenomena of its relative frequencies of outcomes when tossed repeatedly.

I make this distinction in my Prob Stat classes by beginning the semester having the students estimate the probability that at least two class members have the same birthday. I give them two guesses which is a major hint, and write my two guesses down on a piece of paper. We then go through and verify if it is the case (highly probable for my classes of 30 or more students). I then ask them again, what's the probability as I turn over my guesses "Either it's P=1 or P=0". The point here is you can calculate the probability that an arbitrary set of 30 people will have a common birthday but that is not a description of the reality of our set of students, it is a description of the class of such sets for which ours is an instance. Saying our specific group has a certain probability distribution is shorthand for saying that group belongs to a class of such groups with the given probability distribution.