The Trouble with Quantum Mechanics

http://www.nybooks.com/articles/2017/01/19/trouble-with-quantum-mechanics/

by themgt • 10 years ago

125 119 10 years ago

nybooks.com

coldcode• 10 years ago

I had three semesters of QM in grad school and I still have trouble with it.

jessriedel• 10 years ago

Well, the OP author is arguably the greatest living physicist, so it's not surprising that 3 semesters isn't enough :)

walrus1066• 10 years ago

"Anyone who says that they understand Quantum Mechanics does not understand Quantum Mechanics" - Richard feynmann (might be paraphrased)

wolfgke3• 10 years ago

Relevant for those who are interested in the topic: Two rather unusual interpretations if quantum mechanics:

- De Brogle-Bohm theory: https://en.wikipedia.org/wiki/De_Broglie%E2%80%93Bohm_theory

- Gerard 't Hooft - The Cellular Automaton Interpretation of Quantum Mechanics: https://arxiv.org/abs/1405.1548 (see also https://en.wikipedia.org/wiki/Superdeterminism)

demonshalo• 10 years ago

the Automaton interpretation sounds like something I would enjoy reading. But considering that I do not have the time this week nor enough knowledge in the subject, could you perhaps provide me with a quick summary as to what it is about?

tps5• 10 years ago

My understanding is this: Hooft is an "instrumentalist" as described in this article. He does not believe that quantum mechanics describes "what's really going on" and is interested in classical models that explain why QM makes such good predictions.

"Superdeterminism" is a very interesting perspective that Hooft feels could resolve this dilemma. Others feel that superdeterminism is not falsifiable, so there's no point in discussing it.

I came across this interesting article on stackexchange from a few years ago: http://physics.stackexchange.com/questions/34217/why-do-peop...

The first response is from Peter Shor, presumably the discoverer of Shor's algoithm. Kinda cool that such intelligent people with the highest academic credentials use a public forum.

lisivka13• 10 years ago

- Loop quantum gravity: https://en.wikipedia.org/wiki/Loop_quantum_gravity

wolfgke3• 10 years ago

As far as I know Loop quantum gravity is an alternative to string theory and says nothing about how to interprete quantum mechanics (though it better should reproduce its predictions). But convince me to be wrong.

mathoff• 10 years ago

In philosophy, this is what we call four dimensionalism: https://en.wikipedia.org/wiki/Four-dimensionalism https://plato.stanford.edu/entries/temporal-parts/

MichailP• 10 years ago

Let me give example why physics is hard, using most simple example.

Ever encountered term observer (like observer in inertial frame)?

Well, it turns out, observer in physics means whole laboratory with established procedures of measuring time and length, not observer as you would mean in everyday life. And established procedure, what is that? Say, how do you measure time when certain event happened? It turns out, you can measure time only locally, and for this you need to have synchronized clocks spread out in your lab in points of interest. Some more philosophically inclined physicist would say that time means synchronized clocks. And how do you synchronize clocks? Well, it has to do with speed of light but I will stop here.

And this is for the most simple term. Imagine something more complicated. Now imagine hundreds of people writing papers and books with only partial understanding, and signal to noise ratio :-)

End of rant :-)

jayajay3• 10 years ago

In this musical analogy, the act of measuring the spin somehow shifts all the intensity of the chord to one of the notes, which we then hear on its own.

People can accept that measurements and observables are operators which act on the wave function (which can be expressed as a linear combination of eigenstates of the operator in question). People can even accept that some observables cannot be measured simultaneously, because they do not commute, and cannot share eigenstates.

People cannot accept that particles can only be found in eigenstates of the observable in question, without some mechanism to explain why this happens. If a particle's measured value can only be one of the eigenvalues of the operator, it begs the question "which eigenvalue is it going to show us?". Alternatively, if nature let us measure the expectation value instead of "one of the eigenvalues", then quantum mechanics would not be "weird" at all. Too bad, nature isn't like that, or maybe, be thankful that it is.

But if the corrections to quantum mechanics represented by the new terms in the Lindblad equations (expressed as energies) were as large as one part in a hundred million billion of the energy difference of the atomic states used in the clock, this precision would have been quite lost. The new terms must therefore be even smaller than this.

I always see people (physicists) complaining about String Theory, etc. because they play around in regimes which are too small for us to actually work with. We have observed macroscopic effects which are due to a build-up of quantum mechanical effects (superconductivity, solid state, etc.). It would be expected that a correction to quantum mechanics would also make macroscopic predictions...

EGreg6• 10 years ago

What is wrong with the Pilot Wave Theory intepretation? I'm ok with FTL information transfer, and so Bell's inequalities are satisfied.

And anyway the other interpretations DO NOT rule out FTL information transfer. For example if one entangled electron flies into a black hole then we would be able to know its spin by measuring the other one even if light from it won't reach us.

Also there is this:

https://en.m.wikipedia.org/wiki/Wigner's_friend

baytrailcat2• 10 years ago

I'm ok with FTL information transfer

Which will lead to the conclusion that Special Relativity (which was extensively validated) is either wrong or incomplete.

I have seen a revival of Pilot-wave theory here on HN, but the real conclusion (a real particle pushed around by the enigmatic pilot wave or quantum potential or whatever) is as bizarre or as satisfying as saying particle is in a superposition. Also, I haven't seen anyone satisfyingly explain, in a classical-deterministic sense, the Delayed Choice Quantum Eraser Experiment (also, see the the Wheeler Thought Experiment)

goatlover• 10 years ago

There is the silicone oil droplet phenomena which is a classical version of the pilot wave, and it does reproduce some quantum behavior, including the double slit. That a pilot wave has been discovered (granted at macro scale) does make one reconsider.

https://youtu.be/WIyTZDHuarQ

lisivka13• 10 years ago

Did you saw double slit experiment reproduction in 2D macro?: https://www.youtube.com/watch?v=nsaUX48t0w8

Observer breaks interference pattern.

naasking2• 10 years ago

Which will lead to the conclusion that Special Relativity (which was extensively validated) is either wrong or incomplete.

Not necessarily. The simplest relativistic pilot wave theory requires only a preferred foliation of spacetime.

This is seemingly unappealing to most physicists, but recent work has shown that the wave function itself contains just such a structure, so the needed foliation is present in every interpretation of QM.

posterboy3• 10 years ago

Wouldn't the entanglement just break? The preassure would rip the electron apart to begin with. I'm not sure if that erases entanglement already or if the quarks needed to be annihilated. Either way, are the known laws of physics applicable inside a black hole at all? I don't think so.

krastanov13• 10 years ago

The entanglement does not break. The entire black hole becomes entangled with the electron left outside (which is less exciting than it sounds).

Sure, plenty of thing break if we start talking about unknowns, like the black hole evaporation for instance, but today's physics has a pretty good idea what happens if we just drop one of the electrons in the black hole and measure the other one. We just learn the spin of both of them at the same time, but we can not choose the value (it is random) hence there is no communication and no "weird peeking into the black hole".

EGreg6• 10 years ago

An electron is a lepton, it isn't made up of quarks. What does it mean to rip it apart?

kkylin3• 10 years ago

What I personally do not like about pilot wave theory is that in the many body versions of the theory that I've seen, you still need a wave function that depends on the positions of all the particles. That is, the "pilot wave", unlike physical fields like electric or magnetic fields, is not a function of location in spacetime, nor is it attached to a particle, but to all particles. This means in a certain sense the amount of information in a given space grows exponentially with the number of particles. To me this has always been one of the mysteries of QM, and in pilot wave theory it is as much of a mystery.

(The main thing that pilot wave theory clarifies for me is wave function collapse.)

EGreg6• 10 years ago

Still better than world-splitting at every instant. Isn't it?

krastanov13• 10 years ago

There is no information transfer in your black hole example. Entanglement can not be used for FTL communication no matter the interpretation you choose.

As for why pilot wave theory is not popular: it generalises poorly. Sure, you can build a multi-particle theory or even field theory on top of it, but they are clumsy to work with and after their initial study they did not show any promising way forward that was not available to the typical approach to quantum mechanics.

Edit: Wiki article about why entanglement does not provide FTL communication even if naively it looks like it should be possible https://en.m.wikipedia.org/wiki/No-communication_theorem

thisrod3• 10 years ago

Do you understand the arguments that the ability to send a signal faster than light allows you to send that signal backwards in time? I've never met anyone who who understood special relativity but was OK with that.

ebcode2• 10 years ago

How so? Why should faster-than-light travel imply a reversal of the arrow of time? Can you explain the reasoning?

jayajay3• 10 years ago

Pilot wave theory only postpones the question of "what is a wave function and how does it collapse?", it doesn't actually say anything new. I think Wigner's Friend is really no fundamentally different than Schrodinger's Cat.

wolfgke3• 10 years ago

What is wrong with the Pilot Wave Theory intepretation?

See

https://www.quora.com/Why-dont-more-physicists-subscribe-to-...

for a discussion where people of both sides answered. For an answer that is rather critical towards the Pilot Wave interpretation see

https://www.quora.com/Why-dont-more-physicists-subscribe-to-...

Another skepticall answer is

https://www.quora.com/Why-dont-more-physicists-subscribe-to-...

nickpeterson• 10 years ago

They overcharge you if your license is from two simultaneous states.

runn1ng• 10 years ago

I know that Eliezer Yudkowsky is very controversial and some of his statements cannot be taken seriously, but I really liked his sequence of articles on QM. It made it a little bit more accessible to me, who knows very little about it.

I cannot judge how accurate it is; but from googling around, it doesn't seem he does any strong errors. Also he makes some very strong statements in the end which I found preposterous (mainly, that Bayes reasoning is better than scientific reasoning, and that Bayes reasoning must lead to Many Worlds theory).

The sequence is here

http://lesswrong.com/lw/r5/the_quantum_physics_sequence/

danharaj• 10 years ago

One of the issues with engaging with Yudkowsky's posts is how verbose his arguments are. Fortunately, for something as well formalized as quantum mechanics, there's a succinct counter-example: A formalism of quantum mechanics that makes perfect sense with neither the idea of wavefunction collapse nor some abundance of ontologically dubious and extravagant entities (alternate universes you can never observe). That formalism is relational quantum mechanics.

Accessible overview: https://plato.stanford.edu/entries/qm-relational/ In particular, this overview addresses the fundamental difference between Everett's theory and Rovelli's theory.

Original paper: https://arxiv.org/abs/quant-ph/9609002

A fluffy paper by the same author about QFT and its relational nature: https://arxiv.org/abs/hep-th/9910131

In short: The way you can evade the difficulties MWI addresses completely by changing one premise of the problem: that physical systems have states independent of their observers. In relational quantum mechanics, there is no universal wavefunction because there is no external observer of the universe. Call it the zero worlds interpretation. How parsimonious!

There is an element of Yudkowsky's sequence that deals with relative configuration spaces, but this is a subtly different concept and is really just a discussion from a clever guy unequipped with the right concepts about the difference between, say, affine spaces and vector spaces or torsors and groups. The idea that there is a configuration space independent of the observer is the assumption he misses.

In general I find Yudkowsky's extreme certainty in his own arguments to range from amusing to obnoxious. It's not hard to find places where uncertainty creeps in. It doesn't come from mistakes in his reasoning because what he considers he is usually meticulous about considering, but from what he doesn't consider.

brianberns• 10 years ago

It seems that the scientific method has led us to a theory that is beautiful, accurate, and yet utterly incomprehensible. Until we can explain exactly what happens during a wave function collapse, I think QM will remain that way.

evanb3• 10 years ago

Wave functions do not collapse in the realist view, as explained in the article. The cost, however, is a giant explosion of alternate possibilities all having equal reality, summed in superposition.

But, the point is, there is at least one interpretation of QM where the wavefunction never collapses: it just evolves unitarily according to the Schrodinger equation, forever.

akvadrako12• 10 years ago

It doesn't seem that way to me. Decoherence is understood pretty well; although not every detail is completely worked out it explains our experience and the subjective appearance of collapse. It's just approximate though and hence we get quantum interference.

The reason you think it's incomprehensible is that a lot of smart people have trouble accepting what we called reality is only one of many classical "worlds". It's a philosophical concern; technically there are no major issues and it's not that hard to understand.

The reality of QM won't be commonly accepted until we find a way for people to accept that the universe isn't what they expected or we change their expectations.

thisrod3• 10 years ago

This is a really good description of the measurement problem. It takes all the sides of the debate seriously, which is hard for most physicists to do. If you think about quantum mechanics all day every day, you're almost bound to settle on one interpretation as your working model, and it becomes hard to take the others seriously.

gpsx2• 10 years ago

Weinberg mentions the quote by Eugene Wigner regarding the importance of consciousness in the section on instrumentalists. However, I see it is being fundamental to realist's interpretation also. I subscribe the the realist approach, which I view as meaning nothing special happens when a human makes an observation - it is just quantum mechanics as usual. This has interesting implications. The key is that the observer is not outside the system, he is a part of it. His wave function becomes entangled with the system when the measurement is made. But what makes this difficult to analyze is that we don't know what it means for the observer to say "I see the result was *". What physical process allows the observer to be aware of the measurement?

We can make a simplifying and unsatisfying assumption - the observer has a register in his head which records the result of the measurement. Here, it is easy to imagine this register is part of the wave function just like the object being measured. As mentioned above, the wave function for this register is entangled with the object being measured so the measurement and register value are correlated. In our model this corresponds to the person independently thinking each of the possible results, which some people call multiple universes.

But this simple register is not what goes on inside the brain. Not knowing this makes it difficult to accurately model the measurement process.

EDIT: for clarity

evanb3• 10 years ago

In the realist interpretation, even without people, you wind up in entangled states after measurement of a spin (where by "measurement" I mean: allow a giant classical system to interact with a spin and tilt a pointer on some gauge to / wind up with some register holding + or -). It's perfectly possible to formulate the realist ideas without any consciousness or insistence that the measurements register in a brain or observer.

gpsx2• 10 years ago

Yes, I agree you can formulate the realist ideas, because the observer is nothing special at all. And I also think this is how the world works. But from an experimental verification point of view I think you really need a model for what the brain does to say you understand how a measurement is made. What is a theory without verification?

Granted his quote used the word formulate, but I interpret the comment as referring to not just writing some rules but also having some justification for them.

lisper13• 10 years ago

This article presents a very inaccurate view of the realist approach. The universe does not "split" when you make a measurement.

The measurement problem is a solved problem. The solution is that measurement and entanglement are the same physical phenomenon. Measurement is just entanglement extended to a macroscopic system through a process called "decoherence". The net result is that, when you do the math, you recover classical behavior by taking "slices" of the wave function (the mathematical operation is called a "trace"). This has been known for decades now, and provides a coherent and easy-to-understand picture of what is "really" going on. It is astonishing to me that the physics community still bifurcates into two camps: those who think this is common knowledge, and those who are completely unaware of it (or think it's a crazy idea).

References:

http://www.flownet.com/ron/QM.pdf

https://www.youtube.com/watch?v=dEaecUuEqfc (The same content in a video)

http://blog.rongarret.info/2014/09/are-parallel-universes-re...

http://blog.rongarret.info/2014/10/parallel-universes-and-ar...

iheartmemcache• 10 years ago

Well the Copenhagen interpretation traditionally requires the act of an 'observer'. I.e., you are not allowed to perform that act of summation on the set of orthonormal vectors in N dimensional Hilbert spaces until an "observer" observes. (That's where that stupid cat came into play and the 'uncertainty' of it being both dead and alive simultaneously until it's 'observed'.) Until this discrete event occurs, the wave function doesn't "collapse" and we're just living in statistical la-la land.

The disagreement lies in "I can mathematically apply the trace function, I'm just not quite sure when".

Alternatively, according to the Everett interpretation (to which I subscribe, and to which you seem to as well) quite a few reputed physicists do believe that the universe "splits" (though not in an actual literal sense, there's nothing ripping the universe physically in two) every time an "observation" occurs and there are lots and lots of occurring simultaneously taken from various frames of reference. This actually makes a lot of sense. Let Sean Carroll https://youtu.be/ZacggH9wB7Y convince (the plural) you.

lisper13• 10 years ago

The wave function never collapses. This is easily demonstrated with a simple thought experiment that is described in the first two references that I link to.

n4r96• 10 years ago

The measurement problem is a solved problem.

I think this is overselling it slightly. Yes, Everettian style interpretations have helped to shed insight into the reality of the quantum state. Yes, decoherence has helped us to understand physical systems and their interactions with the environment. But if the measurement problem was solved there wouldn't continue to be a swathe of literature on the measurement problem by respected quantum theorists. There are some very good reasons for doubting the validity of many world interpretations. We don't have much of a consensus on what's "really" going on, and I don't believe we will for quite some time.

tim3333• 10 years ago

Indeed it's not that solved. A big problem is as Weinberg puts it

Several attempts following the realist approach have come close to deducing rules like the Born rule that we know work well experimentally, but I think without final success.

which is kind of the heart of things.

opaque2• 10 years ago

Decoherence is an extremely useful and elegant mechanism for understanding quantum systems, but unfortunately it doesn't solve the measurement problem. At least not to the satisfaction of most practising physicists. The entanglement and trace operation does produce classical probabilities in the observed sub-system. However, it requires you to make a pretty arbitrary division between the observed system and the wider environment. Why is everything not just entangled with everything else up and up the chain until the entire universe is in superposition? The problem of where the quantum world ends and the classical world begins is still unanswered. This is one of the greatest open questions in physics, likely requiring a unification of general relativity and quantum mechanics before it's resolved.

beevai1422• 10 years ago

This is correct. As it is usually applied in practice in the literature, the measurement device decoheres the quantum system and becomes entangled with it --- and at the end of the calculation, Copenhagen interpretation is applied on the measurement device. It works fine in practice, but in the end it is philosophically just the Copenhagen interpretation.

If something is not accepted by the wider physics community, it's very likely the correctness of the thing in question is not that clear.

lisper13• 10 years ago

Why is everything not just entangled with everything else up and up the chain until the entire universe is in superposition?

It is.

I'm going hazard a guess that your next question is going to be: why do I not perceive myself to be in a superposition of states? And the answer is that you are not what you think you are. You think you are a human being, a classical physical object made of atoms, but you aren't. This is a very good approximation to the truth, but it is not the truth. The truth is that you (the thing engaged in this conversation) are a software process running on a human brain. You are a classical computing process, i.e. a process that can be emulated by a classical computing model like a Turing machine. The reason for this is that the kinds of things you do necessarily involves copying information (e.g. the process of reading this comment involves copying information from your computer into your brain) and quantum information cannot be copied. Only classical information can be copied. Your conscious awareness of the existence of physical processes is an emergent phenomenon of accumulating memories, i.e. copying information. Because of this you cannot become consciously aware of your quantum nature, and because of that you cannot demonstrate the quantum nature of any system (to yourself) unless it is isolated from you. You could in principle demonstrate the quantum nature of the rest of the universe if you could somehow isolate yourself from it, but that presents insurmountable practical difficulties.

The problem of where the quantum world ends and the classical world begins is still unanswered.

Because the question tacitly makes the false assumption that there is a hard boundary between the two. There isn't.

akvadrako12• 10 years ago

At least not to the satisfaction of most practising physicists.

This is true, as the parent said - it's not accepted by many physicists.

However, it requires you to make a pretty arbitrary division between the observed system and the wider environment.

This isn't true. The whole universe is in a superposition and you can reason with decoherence at any level. Literally everything in one "world" including the vacuum of space is entangled with everything else, though to a very small degree.

gaze• 10 years ago

This is going in circles. You still need to put in Born's rule by hand somewhere. If you can explain the origin of Born's rule we'll have gotten somewhere. I also don't understand why you call a trace a slice. It's an ensemble average. Try deriving the density matrix formalism by coupling a small system to some collection of ancillae and you'll see that it doesn't really add anything. The trace is just convenient notation for the operator sandwich of closed quantum systems... it's just that your closed system is now an open system tensored with a bath.

lisper13• 10 years ago

You still need to put in Born's rule by hand somewhere

I'm not sure what you mean by "putting Born's rule in by hand". We're talking about interpretations here. Born's rule is just an empirical fact. Do you mean that I need to explain why the probabilities are the square of the amplitude? That's kind of like asking me to explain why the speed of light has the value that it has. It doesn't have an explanation. It's just part of the Way Things Are.

But I can make the following heuristic arguments:

1. Outcomes are probabilistic because this is the only way that classical behavior can emerge from the wave function, and without classical behavior you can't copy information, and without copying information you can't have discussions like the one we're having. It's an anthropomorphic argument (actually it's a info-pomomorphic argument :-)

2. The probabilities are the square of the amplitude because that is how you get a useful mathematical model of reality. It is simply an empirical fact that destructive interference happens, so if you want to build a mathematical model of that you need something that can take on negative values. You can't have negative probabilities in classical reality, so the underlying reality must be something other than classical probabilities. The square root of the probability is just the simplest mathematical model that explains the observations.

Maybe this will help: the technical paper that my position is based on is here:

https://arxiv.org/pdf/quant-ph/9605002v2.pdf

akvadrako12• 10 years ago

There are a few ways to derive the Born rule within WMI. None are perfect but their assumptions are fairly reasonable and most importantly the handling of probability is better defined than in classical mechanics or Copenhagen, so it's not a reason to discount the theory.

https://en.wikipedia.org/wiki/Many-worlds_interpretation#Pro...

The Trouble with Quantum Mechanics

Related Stories