Wave superposition nothing more than abstraction?

[DrChinese]

Then you will have a rather fun time trying to explain the whole set of tests of "realism" that has been conducted via the violation of Leggett's inequality. And I'm guessing that I don't have to explain to this crowd what is meant by "realism" in this context.

J. Romero et al., N. Jour. Phys. v.12, p.123007 (2010).
S. Gröblacher et al., Nature v.446, p.871 (2007).
Cyril Branciard et al. Phys. Rev. Lett. 99, 210407 (2007).

Zz.

Second reference can also be found here:
http://arxiv.org/abs/0704.2529

Last reference can also be found here:
http://arxiv.org/abs/0708.0584

 

[Fredrik]

How is a conspiracy theory(currents conspire to act as if they went in both directions simultaneously, but they probably didn't) better than to suppose that common, inductive reasoning is right in this case?

The reason I don't like phrases like "the current went both ways" is that they suggest that QM should be interpreted as a "description of what actually happens". The phrase "the current went this way or that way" has the same problem. If we define those phrases in the only way that makes sense scientifically, the second one has the additional problem that it's just plain wrong (in this type of experiments).

 

[Maui]

The reason I don't like phrases like "the current went both ways" is that they suggest that QM should be interpreted as a "description of what actually happens". The phrase "the current went this way or that way" has the same problem. If we define those phrases in the only way that makes sense scientifically, the second one has the additional problem that it's just plain wrong (in this type of experiments).

Well, not liking what quantum mechanical expiments strongly suggest about what happens, certainly doesn't invalidate the straightfoward(imo) conclusions that flow from it. One could easily claim that classical mechanics is also not a true description of experiments at the macro scale.

Edit: I now remember reading a post of yours where you said something to the effect that all our models are likely wrong. That sheds some light on why you seem to react the way you do to the interpretational side of quantum experiments.

 

[nonequilibrium]

Well, not liking what quantum mechanical expiments strongly suggest about what happens, certainly doesn't invalidate the straightfoward(imo) conclusions that flow from it. One could easily claim that classical mechanics is also not a true description of experiments at the macro scale.

The difference is that we can see classical mechanics, so I don't think extra care in quantum mechanics is superfluous. In the respect of new discovery, it is often helpful to assume as little as possible.

 

[Maui]

The difference is that we can see classical mechanics, so I don't think extra care in quantum mechanics is superfluous. In the respect of new discovery, it is often helpful to assume as little as possible.

Matter looks solid under the classical mechanics treatment of human vision(as you say, you can see it - don't see how that's relevant, but whatever), but classical mechanics was never good enough. This argument has clearly evolved to an argument about personal tastes.

 

[JK423]

Matter looks solid under the classical mechanics treatment of human vision(as you say, you can see it - don't see how that's relevant, but whatever), but classical mechanics was never good enough. This argument has clearly evolved to an argument about personal tastes.

Have you ever observed a particle being in 2 places at once? The answer is 'no'. So, period. It's incorrect to use such phrases.

 

[Maui]

Have you ever observed a particle being in 2 places at once? The answer is 'no'. So, period. It's incorrect to use such phrases.

That depends on what you mean by "observed". We don't observe partciles but their effects after measurement. A particle interfereing with itself can be thought of as being in two places at once.

 

[JK423]

That depends on what you mean by "observed". We don't observe partciles but their effects after measurement. A particle interfereing with itself can be thought of as being in two places at once.

In the case that a particle would be in 2 places at once, we would scatter photons on the particle and we would see them being scattered, for example, both at places A and B. Which means that we would see flashes from two different points in space. That would be proof.
But, we don't observe such a thing. Everytime we make an observation, only one is the outcome and not many simultaneously.
And what do you mean with phrases like the one in bold? *can be thought of..* is just an interpretation of the formalism. We don't know anything more!

Dont forget that Bohmian mechanics have not been falsified, a deterministic theory which talks about particles always being in one place at a time! And agrees with experiment!
Not even Leggett's model can falsify Bohmian mechanics!

As a previous poster said, whch is true in my opinion, a good scientist must always assume as few things as possible, because only in this way there will be a new discovery!
And also, phrases like *being in two places at once* are mostly used to impress ourselves and others, non-physicists. However, the truth is more shameful: We don't actually know what's going on..

 

[Maui]

In the case that a particle would be in 2 places at once, we would scatter photons on the particle and we would see them being scattered, for example, both at places A and B. Which means that we would see flashes from two different points in space. That would be proof.
But, we don't observe such a thing. Everytime we make an observation, only one is the outcome and not many simultaneously.

The OP asked about superpositions, it's much much more clear what happens after the measurement than before it. I don't know why you think it's releveant to the discussion of superpositions, what happens after an observation takes place.

And what do you mean with phrases like the one in bold? *can be thought of..* is just an interpretation of the formalism. We don't know anything more!

It was from the double slit experiment - electron interfereing with itself on the detector screen


.

Dont forget that Bohmian mechanics have not been falsified, a deterministic theory which talks about particles always being in one place at a time! And agrees with experiment!
Not even Leggett's model can falsify Bohmian mechanics!

That's not a good example of a minimalist assumption interpretation!

As a previous poster said, whch is true in my opinion, a good scientist must always assume as few things as possible, because only in this way there will be a new discovery!
And also, phrases like *being in two places at once* are mostly used to impress ourselves and others, non-physicists. However, the truth is more shameful: We don't actually know what's going on..

Fair enough. I don't claim to know how to interpret superpositions, i am just gathering more opinions on a murky issue

 

[Fredrik]

Well, not liking what quantum mechanical expiments strongly suggest about what happens, certainly doesn't invalidate the straightfoward(imo) conclusions that flow from it. One could easily claim that classical mechanics is also not a true description of experiments at the macro scale.

How would you argue that the (position) function x:\mathbb R\rightarrow\mathbb R^3 in the non-relativistic classical theory of a single particle with mass m doesn't describe what is actually happening to the particle? Perhaps you meant that you could argue that it's not an exact description of what's happening to any actual particle in the real world? You would be right of course, but that's not what I'm talking about. At the very least, this theory can be interpreted as an exact description of what's going on in a fictional universe. It's not obvious to me that QM can be interpreted that way.

Of course, if we define the phrase "QM is an approximate description of reality" as "QM makes pretty accurate predictions about results of experiments", then it is obvious that QM is an approximate description of reality. But this is not what words like "describes" or "actually happens" mean to me. I think of them as primitives, not as terms to be defined using other terms. So I'm not going to be comfortable saying that "QM describes reality" until I've seen a reason to think that this statement is correct even when "describes" is the primitive that I understand intuitively.

If you want a reason to think that it's not correct, consider post 44 here. This explains why I don't think QM can describe a single world. It might however be able to describe a physical system that contains many worlds.

Edit: I now remember reading a post of yours where you said something to the effect that all our models are likely wrong. That sheds some light on why you seem to react the way you do to the interpretational side of quantum experiments.

I have spent a lot of time thinking about what definition of "theory" is the most appropriate for physics, and the definition I like the best is (loosely stated) "an assignment of a unique probability to each possible result of each member of some set of experiments". Given that definition, or any other definition that's even close to reasonable, it doesn't make sense to classify theories as "right" or "wrong". "Right" can only mean "exactly right", and none of our theories are. (The only possible exception is QM). If we choose to label them "right" or "wrong", they'd all be in the "wrong" category. The only kind of (good vs. bad) classification of theories that makes sense to me is to say that a theory is as good as its predictions. The more accurate the predictions are, the better the theory.

This explains all statements similar to "all theories are wrong" that I might have made in the past. I'm not sure what it has to do with what we're talking about now.

 

[Fredrik]

Another reason not to think of QM as a "description of what actually happens" is the discussion in this thread. Consider the wavefunction \psi that satisfies \psi(\vec x)=N\exp(-a\vec x^2) for all x. If someone insists that a wavefunction is telling us what's "actually happening", then I would interpret that statement as saying that this wavefunction represents a particle that's "actually" spread out all over space, with "most of it" near 0. But there doesn't seem to be any valid arguments against the possibility that every particle has a position (represented by just a triple of real numbers), no matter what its wavefunction is.

I'm not saying that I think particles have positions. My point is just that since there seems to be nothing in QM that rules it out, it's very hard to argue that a particle's wavefunction is telling us what's "actually happening" to the particle. I don't think QM + the experiments that check the accuracy of its predictions give us enough information to say that we know what's "really happening" between state preparation and measurement.

 

[JK423]

Another reason not to think of QM as a "description of what actually happens" is the discussion in this thread. Consider the wavefunction \psi that satisfies \psi(\vec x)=N\exp(-a\vec x^2) for all x. If someone insists that a wavefunction is telling us what's "actually happening", then I would interpret that statement as saying that this wavefunction represents a particle that's "actually" spread out all over space, with "most of it" near 0. But there doesn't seem to be any valid arguments against the possibility that every particle has a position (represented by just a triple of real numbers), no matter what its wavefunction is.

I'm not saying that I think particles have positions. My point is just that since there seems to be nothing in QM that rules it out, it's very hard to argue that a particle's wavefunction is telling us what's "actually happening" to the particle. I don't think QM + the experiments that check the accuracy of its predictions give us enough information to say that we know what's "really happening" between state preparation and measurement.

But.. If we accept that nature is probabilistic and we include decoherence in all this, then are we able to say that QM describes *what actually happens*?
For example, the question *Why do we always observe a particle to be in a specific position* is perfectly explained by decoherence. To make an observation, the particle's wavefunction interacts with a macroscopic environment (measuring device) giving rise to decoherence, which destroys the superposition of states.

How measurements work i think that has been partially explained. But, for me, the probabilistic behaviour of nature is hard to accept.

 

[Fredrik]

But.. If we accept that nature is probabilistic and we include decoherence in all this, then are we able to say that QM describes *what actually happens*?

In my opinion, no. Decoherence theory explains why it's hard to prepare superpositions, why a superposition can't be the result of a measurement, etc., but it doesn't give us any insights into what a particle with wavefunction ψ is "really doing".

 

[JK423]

In my opinion, no. Decoherence theory explains why it's hard to prepare superpositions, why a superposition can't be the result of a measurement, etc., but it doesn't give us any insights into what a particle with wavefunction ψ is "really doing".

Ok, imagine that QM was not probabilistic and that decoherence theory could predict exactly in which state the pure state vector will collapse. Then, according to what you say, you would still think that QM is not a description of "what really happens" since the probabilities isn't your issue.
Or is it?

If we don't consider the probabilistic side of QM, then the wavefunction seems to be a complete description of 'what actually happens'. There are just no particles! There are only wavefunctions.. Why we see particles? We dont... It's just that the wavefunction is localized in space, and QM predicts the wavefunction to be localized in occassions like measurements. So, everything is predicted by QM. How this is not a complete description of "what actually happens"? Classical mechanics in the classical world don't do any better than that!
It's just the probabilities that makes it difficult to accept QM as a complete theory..

 

[BruceW]

Another reason not to think of QM as a "description of what actually happens" is the discussion in this thread.

QM is the best description of what actually happens.

Whether it is the role of science to describe what actually happens is a whole other question.

 

[Fredrik]

Ok, imagine that QM was not probabilistic and that decoherence theory could predict exactly in which state the pure state vector will collapse. Then, according to what you say, you would still think that QM is not a description of "what really happens" since the probabilities isn't your issue.
Or is it?

If QM was not probabilistic, then it wouldn't be QM, so I don't really understand the question.

If we don't consider the probabilistic side of QM, then the wavefunction seems to be a complete description of 'what actually happens'. There are just no particles! There are only wavefunctions.. Why we see particles? We dont... It's just that the wavefunction is localized in space, and QM predicts the wavefunction to be localized in occassions like measurements. So, everything is predicted by QM. How this is not a complete description of "what actually happens"?

It can be a complete description of what actually happens. But in the posts I linked to above, I argued that a) this seems to lead to many worlds, and b) there's room for additional assumptions on top, like the assumption that every particle has a position. (This means that we can at least say that the fantastic agreement between theory and experiment doesn't prove that QM describes reality in the sense you're suggesting).

There's one more thing that's always in the back of my mind: QM looks so much like a toy model that someone invented just to show us that there exists a theory that assigns non-trivial probabilities (not only 0 or 1) to possible results of measurements on pure states, that it's hard for me to think of it as something more than a probability assignment. To me, QM looks like the simplest possible non-trivial probability assignment, and it would be quite remarkable if that also turned out to be an accurate description of reality.

It's just the probabilities that makes it difficult to accept QM as a complete theory..

It's not just the fact that non-trivial probabilities are involved even when the states are pure. Classical mechanics can also be interpreted as a probabilistic theory, but the theory first produces something that can clearly be thought of as an approximate description of what actually happens, and then we can use that description to assign a probability of either 0 or 1 to each possible result. QM on the other hand skips over the first step (the description of what actually happens) and goes straight for the probability assignment.

 

[Fredrik]

QM is the best description of what actually happens.

You can argue for this if you define "description" in a way that makes the claim trivially true. But this would make the word useless, so it would be better to just stop using it altogether.

 

[Fyzix]

Fredrik, I wrote a long response to your PM, but your inbox was too full to be able to send it to you.
So I'll ask here instead.

You claim that if QM is indeed a description of reality something along the lines of your own reading of Everett seem inevitable to you.
So how is it different from the other MWI's?

In the case of Schroedingers Cat, how does your "splitting" differ from the splitt/decoheret branching of Wallace/Deutsch?

Also you can still insist that QM is a description of reality without invoking collapse.
Gerard t Hooft has shown this with his models, other people have made models, deBroglie Bohm etc. etc.

 

[Fredrik]

Fredrik, I wrote a long response to your PM, but your inbox was too full to be able to send it to you.

Sorry about that. I heard the email notifications while I was typing my previous two posts in this thread, but I assumed that they were notifications of PMs, not notifications that my inbox was full, so I figured I could just look at them after I was done here. I have deleted a few messages now.

So how is it different from the other MWI's?

In the case of Schroedingers Cat, how does your "splitting" differ from the splitt/decoheret branching of Wallace/Deutsch?

I don't think I want to get into a discussion about my thoughts on how QM can be interpreted as describing many worlds. It would take more time than I have right now. Also, I'm not sure why my previous attempts to explain the "splits" in this MWI to you didn't help, so I don't know what I should be saying differently. It's possible that I will have a better way of explaining these things once I've thought them through in more detail, but I'm giving this a low priority right now.

Also you can still insist that QM is a description of reality without invoking collapse.
Gerard t Hooft has shown this with his models, other people have made models, deBroglie Bohm etc. etc.

Bohm says that particles have positions. I find it hard to accept that, and that any solution of the Schrödinger equation can describe a particle. They seem mutually exclusive to me. So if Bohmian mechanics describes reality (something I don't really believe but also haven't ruled out), I would say that QM doesn't.

 

[BruceW]

You can argue for this if you define "description" in a way that makes the claim trivially true. But this would make the word useless, so it would be better to just stop using it altogether.

The reason QM is a description of what actually happens is because it is backed up by experiment. How else could you define "description"?

The reason I said 'best description' is because the theory is very general, and can be used in many different situations. I suppose General relativity would also be a candidate for the honour of 'best description'.

 

[harrylin]

The reason QM is a description of what actually happens is because it is backed up by experiment. How else could you define "description"? [..]

It describes what is observed by measurement - that is, the phenomena or appearances. Just like relativity. However, "superposition" isn't a measurable but as Fredrik indicated, it's part of the math of the theory that is used to make predictions about what can be observed.

 

[BruceW]

So we can only say that what actually happens is what we actually measure in experiments?
And that theory should never be considered to represent what actually happens?
If that's what you were saying, then I guess I agree. Experimental results will always be true, but theories come and go.

Edit: Not trying to hate on theory itself. Even the theories that have been proven wrong (i.e. classical mechanics) are still useful in a certain limit.

 

[harrylin]

So we can only say that what actually happens is what we actually measure in experiments?
And that theory should never be considered to represent what actually happens?
If that's what you were saying, then I guess I agree. Experimental results will always be true, but theories come and go.

Edit: Not trying to hate on theory itself. Even the theories that have been proven wrong (i.e. classical mechanics) are still useful in a certain limit.

Quite so: we may consider that a theory represents what actually happens but others may disagree - especially when conclusions depend on a series of assumptions.

 

[ZealScience]

But what you are seeing might not be the truth. And you have never seen "real electrons" with naked eyes or even with microscopes. So you are inducing the phenomenon via macroscopic truth which might not be applied to or even exactly true. Perhaps, what is true requires mathematical explanation together with well-controlled experiments instead of observations with naked eyes. In addition, there are some classical mechanics which violate our intuition but absolutely true.

 

[BruceW]

If I was supporting the argument that only experimental evidence is truth, then I would say that 'electron' is a theoretical model to explain outcomes of certain experiments. So I wouldn't count an electron as a direct experimental evidence.
Also, I wasn't trying to say that experimental evidence is only things that we can see. A blind person can do lots of different experiments.

One problem with saying 'experimental evidence is truth' is that the experimenter may not write down something that is important in interpreting the outcome of the experiment, and without this, the experimental evidence is not useful. (For an extreme example, if I gave you a list of lengths as the outcome of some experiment, but I didn't tell you what the experiment was on, then it is useless).
I think this is what ZealScience meant when you said we need well-controlled experiments. (i.e. writing down all the important factors in an experiment).

 

[Fredrik]

The reason QM is a description of what actually happens is because it is backed up by experiment. How else could you define "description"?

I wouldn't define it. See post #40.

It's a term that we already understand intuitively, so why should we define it in a way that changes our understanding of the term? Because it's the "scientific" thing to do? I reject that, because we're talking about interpretations of QM, not about science. It doesn't make sense to try to make something that by definition isn't science conform to the standards of science. It also doesn't make sense to define everything. Even in set theory, two things are left undefined (what sets are, and what it means for a set to be a member of a set). Some terms must be left undefined, and this is a good one to leave alone. In my opinion, to leave terms like "description" undefined is the same thing as admitting that we believe that the concept of "reality" is more fundamental than the theories we use to try to understand it. I don't think that should be really controversial.

The reason I said 'best description' is because the theory is very general, and can be used in many different situations. I suppose General relativity would also be a candidate for the honour of 'best description'.

Those two are certainly the only candidates for "best theory". I'm inclined to say that QM is the better one, but I remain unconvinced that it makes sense to say that it describes something. (OK, I think it might make sense to say that it describes a physical system that includes many worlds, but I don't think it makes sense to say that it describes a single universe, not even a fictional one).

 

[wawenspop]

Its the wave function that actually predicts what happens in experiment.
IMO its better to conceptualize the wave function itself rather than continuing with just waves particles neither of which predict results correctly.

That old saying - is a tree in the forest there if nobody is looking at it? For a particle the answer is a definite - NO.
The term 'there' is observed as a de-coherence of the wave function in our space-time. If the wave function of any particle does not de-cohere we do not observe it, it leaves no track. Its waiting behind the scenes if you like to get asked its location by another wave function co-located.

Now, something that leaves no track of where it goes, and is not observed by anything or anybody is not 'there'. It starts out and it arrives as evidenced by observation values. 'There' implies a physical object at an x,y,z,t location. There isn't one, there is a distributed wave function which is certainly not a physical object is it?
Its path? Well, it may have been anywhere that its wave function was allowed to exist, but that does not mean it 'travelled' any path at all. But it did 'pop out' as a value on arrival.

I conceptualize this as something called a particle 'existing' outside space-time (IMO an algorithmic entity) and it simply pops in and out of our reality by giving values through its wave function boundaries when asked by an observing photon or particle (actually the wave function's of those)

I too am a Penrose fan. Is he still active or has he retired completely?

 

[xts]

something called a particle 'existing' outside space-time (IMO an algorithmic entity) and it simply pops in and out of our reality by giving values through its wave function boundaries when asked by an observing photon or particle (actually the wave function's of those)

Would you elaborate a bit more? Seems to be promising quantum ontology...?

 

[JeffKoch]

Don't you think this should be the goal of physics, and not just meta-physics/philosophy? Physics should be about deducing the rulebook of the Universe. Why should we leave out thoughts about "what's physical and what's real?" How else are we supposed to "figure out the mind of God?"...

Oh dear, this veered off into metaphysics and religion days ago, starting really from post #1.

 

[wawenspop]

Would you elaborate a bit more? Seems to be promising quantum ontology...?

I am not allowed to say on this forum because AFAIK nobody is working on it, so it comes under the forums 'personal theories'.


Looking at the entanglement equation there is no distance of separation of entangled particles in it - yet they are linked over great distances (correlation of states). So I invoke a Bohmian-like data-frame rather than Bohm's 'pilot wave'. Then that data-frame works algorithmically
in the 'matrix interpretation' where de-coherence is effected by outputs from registers through the HUP region - BUT this not accepted by the scientific community and has not been proved. I may be deleted for saying it even.