End of Realism? Experiments Test Quantum Mechanics

[dm4b]

I came across an article from 2007 talking about some experiemtns that are an extension of the experiments which originally tested for Bell's Inequality. These tests were supposedly designed to relax the locality requirements and test soley for realism. I'll quote a few pertinent sections:

" Now physicists from Austria claim to have performed an experiment that rules out a broad class of hidden-variables theories that focus on realism -- giving the uneasy consequence that reality does not exist when we are not observing it "

"Markus Aspelmeyer, Anton Zeilinger and colleagues from the University of Vienna, however, have now shown that realism is more of a problem than locality in the quantum world. They devised an experiment that violates a different inequality proposed by physicist Anthony Leggett in 2003 that relies only on realism, and relaxes the reliance on locality. "

"They found that, just as in the realizations of Bell's thought experiment, Leggett's inequality is violated – thus stressing the quantum-mechanical assertion that reality does not exist when we're not observing it."

I had some questions on the next quote here:

"However, Alain Aspect, a physicist who performed the first Bell-type experiment in the 1980s, thinks the team's philosophical conclusions are subjective. "There are other types of non-local models that are not addressed by either Leggett's inequalities or the experiment," he said."

I'm curious if anybody knows what other non-local models he is referencing? And, have these been able to be tested since 2007? What's the latest and greatest on all this?

Either way, I agree with Alain Aspect when he says at the end of this article:

"But, I rather share the view that such debates, and accompanying experiments such as those by [the Austrian team], allow us to look deeper into the mysteries of quantum mechanics."

Entire article:

http://physicsworld.com/cws/article/news/27640

 

[StevieTNZ]

The non-local hidden variable theory that wasn't ruled out is Bohm Mechanics. No experiment has ruled this theory out.

 

[dm4b]

I wasn't aware of that. I liked some of Bohm's ideas on a conceptual level, but I thought his theory either had mathematical inconsistencies or was incomplete?

Perhaps, I was mistaken, though.

 

[StevieTNZ]

I wasn't aware of that. I liked some of Bohm's ideas on a conceptual level, but I thought his theory either had mathematical inconsistencies or was incomplete?

As far as I'm aware, that isn't the case. There are still papers being published that rule out Bohm Mechanics if you agree the experimenter has free-will, for example: http://arxiv.org/abs/1005.5173

Also: http://arxiv.org/abs/1105.0133

 

[dm4b]

Interesting, I'll have to read through those. Thanks!

 

[bhobba]

The non-local hidden variable theory that wasn't ruled out is Bohm Mechanics. No experiment has ruled this theory out.

I beg to differ. Don't know about the free will thing mentioned later by StevieTNZ - but here is another paper:
http://arxiv.org/pdf/quant-ph/0206196v1.pdf

It doesn't seem to depend on free will.

Thanks
Bill

 

[StevieTNZ]

I beg to differ. Don't know about the free will thing mentioned later by StevieTNZ - but here is another paper:
http://arxiv.org/pdf/quant-ph/0206196v1.pdf

It doesn't seem to depend on free will.

Thanks
Bill

Thanks for that paper. I shall read it. Wasn't aware it existed.

 

[questionpost]

How does ruling out "hidden variables" mean there's no realism? Of course things stop at a certain point, that doesn't mean they aren't built upon in a logical manner...
Unless they are confusing realism for detirminism?
And if matter didn't stop after a certain point, wouldn't that imply there are infinite hidden variables and therefore it would be impossible to determine anything anyway?
How did the conditions for life to form even come to be if there wasn't an existence before us?

 

[akhmeteli]

I beg to differ. Don't know about the free will thing mentioned later by StevieTNZ - but here is another paper:
http://arxiv.org/pdf/quant-ph/0206196v1.pdf

It doesn't seem to depend on free will.

If you imply that dBB was ruled out experimentally, such implication seems dubious: Zeh (Physics Letters A, Volume 309, Issues 5–6, 31 March 2003, Pages 329–334 ) made this comment on the article you quoted: "Variants of Bohm's theory with photon trajectories instead of time-dependent Maxwell fields have recently been claimed to be in conflict with quantum theory and experiments [11]. While the analysis of these experiments appears doubtful, this modified Bohm theory—in contrast to the original one—seems to have never been proven equivalent to quantum theory."

 

[Ken G]

How does ruling out "hidden variables" mean there's no realism?

You can't really expect an accurate rendition of what realism is from a pop sci article, that's the problem. The term means many different things in different contexts, but in this context, it seems to be associated with "independence from observation." In other words, realism is the belief that reality exists independently of how humans tinker with it and understand it. But that term never meant anything in science, because science is all about tinkering with nature and trying to understand it, and everything that science regards as real stems exactly from that process. Relevant here is the Bohr/Einstein debates, where Bohr said that reality is only what we can demonstrate (the positivist approach) while Einstein said that "the Moon is there even when we are not looking at it." What it comes down to is, when we say a system has a set of attributes or properties, should we regard those attributes and properties as things the system really possesses, or should we regard them as tools we are using to understand the system?

Realism says that the system would have no idea what to do if it did not have those attributes and properties, but the anti-realist says that the system has no idea what we mean by attributes and properties, they are all our creations and the system does whatever it does just fine without them, we merely imagine the system has them to help us understand that system. The key point in quantum mechanics is that if we adopt the realist approach, we are always looking for hidden variables to be those attributes and properties, but if we adopt the anti-realist approach, we don't expect that the system needs any hidden variables in order to do the things it does.

How did the conditions for life to form even come to be if there wasn't an existence before us?

This is more a misconception from the pop sci article than the actual issue. If we end up deciding that QM is incompatible with realism, it means that no hidden variable theory that we can stomach will ever be able to recover its predictions. That in turn would mean that quantum systems don't do what they do because of attributes or properties, they just do what they do, and we use attributes and properties to get a good but still incomplete understanding. None of that implies that nothing could exist before there were humans, it just means that we didn't understand quantum systems before there was quantum mechanics, and we might never understand them completely because we like to think about attributes and properties.

 

[StevieTNZ]

If you imply that dBB was ruled out experimentally, such implication seems dubious: Zeh (Physics Letters A, Volume 309, Issues 5–6, 31 March 2003, Pages 329–334 ) made this comment on the article you quoted: "Variants of Bohm's theory with photon trajectories instead of time-dependent Maxwell fields have recently been claimed to be in conflict with quantum theory and experiments [11]. While the analysis of these experiments appears doubtful, this modified Bohm theory—in contrast to the original one—seems to have never been proven equivalent to quantum theory."

Was just about to post this comment!

The reference [11] is the article mentioned by bhobba.

I came across this article: http://iopscience.iop.org/1742-6596/128/1/012017

 

[questionpost]

This is more a misconception from the pop sci article than the actual issue. If we end up deciding that QM is incompatible with realism, it means that no hidden variable theory that we can stomach will ever be able to recover its predictions. That in turn would mean that quantum systems don't do what they do because of attributes or properties, they just do what they do, and we use attributes and properties to get a good but still incomplete understanding. None of that implies that nothing could exist before there were humans, it just means that we didn't understand quantum systems before there was quantum mechanics, and we might never understand them completely because we like to think about attributes and properties.

What? That's it? I even thought people thought of that like over a quarter of a century ago. It seemed more obvious even to me that after a certain point in the quantum level that things happen for no predictable reason.
Also, if there was actually a reason for everything, would that mean we would have to calculate infinite variables since there wouldn't be an "end" or "limit" to how indepthly these attributes go? As in, an atom would be infinitely complex if there were always hidden variables and thus even then we could never get accurate predictions?

 

[bhobba]

How does ruling out "hidden variables" mean there's no realism?

It doesn't - it is possible that a Quantum State exists out there in reality like say an electric field does and wave-function collapse happens by some kind of decoherence process. Trouble is, as far as I know, it hasn't been shown to resolve all the issues. So IMHO realism is still possible - but more work needs to be done. Personally I don't believe a quantum state exists like that.

Thanks
Bill

 

[Demystifier]

I beg to differ. Don't know about the free will thing mentioned later by StevieTNZ - but here is another paper:
http://arxiv.org/pdf/quant-ph/0206196v1.pdf

It doesn't seem to depend on free will.

That paper does not rule out Bohmian mechanics. Instead, it rules out an incorrect interpretation of Bohmian mechanics. See e.g. the first two paragraphs of
http://xxx.lanl.gov/abs/quant-ph/0305131
and the published references therein.

 

[Demystifier]

As far as I'm aware, that isn't the case. There are still papers being published that rule out Bohm Mechanics if you agree the experimenter has free-will, for example: http://arxiv.org/abs/1005.5173

That paper does not rule out Bohmian mechanics, because that paper assumes a form of locality (called "non-signalling condition" in the paper) similar to locality in the Bell theorem. In v3 version of the paper, you can see that on page 12, section "REMARKS ON THE NOTION OF LOCALITY". Bohmian mechanics does not obey this locality condition, so is compatible with the results of that paper.

 

[Demystifier]

I came across an article from 2007 talking about some experiemtns that are an extension of the experiments which originally tested for Bell's Inequality. These tests were supposedly designed to relax the locality requirements and test soley for realism. I'll quote a few pertinent sections:

" Now physicists from Austria claim to have performed an experiment that rules out a broad class of hidden-variables theories that focus on realism -- giving the uneasy consequence that reality does not exist when we are not observing it "

That "broad class of hidden variables" which are ruled out are actually non-contextual hidden variables. Bohmian hidden variables are contextual, so do not belong to that class. Indeed, Zeilinger et al say explicitly in their original paper that Bohmian mechanics is not ruled out by their results.

 

[bhobba]

What? That's it? I even thought people thought of that like over a quarter of a century ago. It seemed more obvious even to me that after a certain point in the quantum level that things happen for no predictable reason.

We know that an observation gives a result with 100% certainty - its just we can't predict what that result is. That could be because of some kind of chaotic behavior that in principle is perfectly predictable but in practice can't be - but then again maybe not - we simply do not know.

Also, if there was actually a reason for everything, would that mean we would have to calculate infinite variables since there wouldn't be an "end" or "limit" to how indepthly these attributes go? As in, an atom would be infinitely complex if there were always hidden variables and thus even then we could never get accurate predictions?

If we had a theory of everything where everything happens for a reason that does not lead to any problems or infinite regress as far as I can see.

Thanks
Bill

 

[Demystifier]

I came across this article: http://iopscience.iop.org/1742-6596/128/1/012017

Bohmian mechanics has the same measurable predictions as standard QM, PROVIDED THAT THE DEGREES OF FREEDOM OF THE MEASURING APPARATUS ARE TAKEN INTO ACCOUNT. This paper does not take into account the degrees of freedom of the measuring apparatus, so does not rule out Bohmian mechanics.

 

[Ken G]

Bohmian mechanics has the same measurable predictions as standard QM, PROVIDED THAT THE DEGREES OF FREEDOM OF THE MEASURING APPARATUS ARE TAKEN INTO ACCOUNT.

But note that the form of "realism" that is being talked about is a form that is not supposed to require any measurement at all, so no degrees of freedom in the apparatus are relevant to how "realistic" some system is prior to being measured. Hidden variables are sometimes imagined to be a way to allow a system to be "real" in this sense, so the issue here is, does the Bohmian description of a system, absent of any measurement, allow us to imagine the system is "real", in the sense of having its behavior completely described by its attributes independent from the attributes of any measuring apparatus? Personally, I think the question is meaningless, because I don't think the concept of "behavior of a system" means anything beyond "interaction with measuring devices." But I would not say "reality doesn't exist until we measure it", I would say "the concept of realism relies on the concept of measurement." This is not generally the way the word "realism" is used, but I think the way it is usually used makes no scientific sense. Hence, I don't think the issue of whether BM should be regarded as "realistic" or not is of any importance to its core validity.

 

[Ken G]

That's it? I even thought people thought of that like over a quarter of a century ago. It seemed more obvious even to me that after a certain point in the quantum level that things happen for no predictable reason.

There are two separate issues here. One is why things happen, the other is simply a description of what is happening. The first issue relates to determinism, and the second to realism. Although there are clear connections there (most realists are also determinists, and most anti-realists are also anti-determinists), they are not the same thing. The "realism" being discussed in that article is about whether we can regard systems as having innate properties independent of whether or not they are interacting with a measuring apparatus (we can be vague about what constitutes a measuring apparatus, that's a whole other issue). The issue of determinism is how those properties determine what happens. Most people connect those concepts, but I could imagine a "real" property of doing random things (if we adopt a stochastic interpretation of the wavefunction but also think the wavefunction is a real property, though there appear to be recently discovered problems with this approach), or a "deterministic" behavior that cannot be traced to a property (isolated objects follow inertial paths, regardless of any of their properties).

So I would say the key issue is not whether or not things have a reason for doing what they do, but rather, do the things themselves have "any idea" what are these "properties" we associate with them (if you follow my drift)? If reality exists in the way we understand it before we measure it, and the measurement just confirms that reality, that is the brand of "realism" that is being talked about. But the pop sci article oversimplifies that terrain by imagining that the possibilities are either that reality "pops into being" when the measurement occurs, or that reality is always there in the same form as the measured version. Neither of those possibilities seem very plausible to me-- much more plausible, and more consistent with quantum mechanics (a la Bohr, anyway), is that reality passes from something that is pretty much inscrutable to us, to something that we can begin to understand, once we pass it through the filter of a measuring apparatus. Is that "realism", or isn't it? The way the word is usually used, no, but I think the way it is usually used is hopelessly naive.

 

[salvestrom]

This all seems very much like the question of the falling tree making a sound if no one is there to hear it, or, in this case, if there are no microphones around to record it. Except we're asking if the tree even exists when noones looking. And there are no squirrels nearby (why this assumption of human observation - have we gone back to the days of putting the Earth at the center of the universe, only now it is the mind?).

Does the question - the one in this thread - have any relevance, in the sense of, will it make any real difference either way to what we calculate?

 

[Ken G]

This all seems very much like the question of the falling tree making a sound if no one is there to hear it, or, in this case, if there are no microphones around to record it.

This is the issue of what constitutes an "observation", which is another ball of wax. But many adopt the point of view that any macro instrument can do a measurement, including the ground that the tree hits. Others hold it requires an intelligence, still others their own intelligence. I would just say it requires that a hypothetical intelligence be present to give meaning to the words (we must be able to imagine ourselves there, even if no one actually is).

Does the question - the one in this thread - have any relevance, in the sense of, will it make any real difference either way to what we calculate?

Does this forum make any "real difference"? Its purpose is to understand.

 

[Demystifier]

But note that the form of "realism" that is being talked about is a form that is not supposed to require any measurement at all, so no degrees of freedom in the apparatus are relevant to how "realistic" some system is prior to being measured. Hidden variables are sometimes imagined to be a way to allow a system to be "real" in this sense, so the issue here is, does the Bohmian description of a system, absent of any measurement, allow us to imagine the system is "real", in the sense of having its behavior completely described by its attributes independent from the attributes of any measuring apparatus?

That's an important question. BM is realistic in the sense that one can talk about properties of the system even without measurements. However, the process of measurement in BM affects these properties significantly. In other words, BM is realistic but strongly contextual.

 

[mkserkan]

Ken G. Very insightful explanations, Thx.

 

[Maui]

But many adopt the point of view that any macro instrument can do a measurement, including the ground that the tree hits

And the measurement changes(apparently instantaneously) the wavefunction across all of spacetime to a state of 'particle'. How is this supposed to reinstate realism? There are so many holes in this position that it's not even worth arguing over it.

 

[StevieTNZ]

But note that the form of "realism" that is being talked about is a form that is not supposed to require any measurement at all, so no degrees of freedom in the apparatus are relevant to how "realistic" some system is prior to being measured.

Before measurement - in superposition (except for position)? BM needs to be contextual.

 

[salvestrom]

... many adopt the point of view that any macro instrument can do a measurement, including the ground that the tree hits.

Does this forum make any "real difference"? Its purpose is to understand.

The ground the tree hits is precisely the sort of argument I make in favour of saying "yes, the tree does make a sound". But I extend this to this realism debate by saying that any interaction constitutes a measurement.

Hmm, isn't the forum's purpose to answer questions inline with mainstream physics? Unless you are referring to the maths rather than the interpretation? But my meaning here was that realism seems to be a question of interpretation that won't have any actual impact on the results of the maths. Not unless the interpretation opens new avenues of exploration that may end up changing that maths.

 

[lugita15]

That's an important question. BM is realistic in the sense that one can talk about properties of the system even without measurements. However, the process of measurement in BM affects these properties significantly. In other words, BM is realistic but strongly contextual.

But what is the meaning and significance of measurement in Bohmian mechanics? I thought that Bohmians downplay the measurement problem, either through superdeterminism where you say that the observer had no free will in making the particular measurement he did at the time and place he did it, or through decoherence where you say that the measuring device with its large number of particles smears out the wave function (or pilot wave) of the system and makes it look collapsed. Or some combination of the two. Either way, contextuality doesn't seem that important to Bohmian thought. Am I wrong?

 

[Ken G]

BM is realistic in the sense that one can talk about properties of the system even without measurements. However, the process of measurement in BM affects these properties significantly. In other words, BM is realistic but strongly contextual.

OK, I understand and that seems very self-consistent. All the same, the empiricist in me is dubious of any concept of "realness" that is independent of (and strongly changed by) measurement. But I recognize that a more rationalistic interpretation of what "real" means in physics does not need to be bothered by that, I'm just more on Bohr's side that understanding physical reality involves connecting with our experience, not about some reality that gets changed by how we experience it. That's not really an objection to the Bohmian interpretation, because I think any interpretation is just an interpretation.

 

[Ken G]

Hmm, isn't the forum's purpose to answer questions inline with mainstream physics?

Yet what that looks like depends on the question. Some questions are clearly of the type "how do I do this calculation", which is only a question of getting the math right. But others, like this thread, are of the type "how do I understand what this calculation is doing or saying", and there the purpose is more about understanding the math than changing it. For example, the many interpretations of quantum mechanics are often topics of mainstream physics discussions, and even the "shut up and calculate" interpretation is just another such interpretation.

But my meaning here was that realism seems to be a question of interpretation that won't have any actual impact on the results of the maths. Not unless the interpretation opens new avenues of exploration that may end up changing that maths.

I agree on both counts.

 

[Demystifier]

But what is the meaning and significance of measurement in Bohmian mechanics? I thought that Bohmians downplay the measurement problem, either through superdeterminism where you say that the observer had no free will in making the particular measurement he did at the time and place he did it, or through decoherence where you say that the measuring device with its large number of particles smears out the wave function (or pilot wave) of the system and makes it look collapsed. Or some combination of the two. Either way, contextuality doesn't seem that important to Bohmian thought. Am I wrong?

BM is deterministic, but not superdeterministic. Both contextuality and decoherence play important roles in BM.

 

[Demystifier]

OK, I understand and that seems very self-consistent. All the same, the empiricist in me is dubious of any concept of "realness" that is independent of (and strongly changed by) measurement. But I recognize that a more rationalistic interpretation of what "real" means in physics does not need to be bothered by that, I'm just more on Bohr's side that understanding physical reality involves connecting with our experience, not about some reality that gets changed by how we experience it. That's not really an objection to the Bohmian interpretation, because I think any interpretation is just an interpretation.

Yes, I can understand such a view as well, especially after my recent development of an interpretation that interpolates between Bohm and Bohr [1], suggesting that these two interpretations are not so different as people usually think.

[1] http://xxx.lanl.gov/abs/1112.2034

 

[Ken G]

Yes, I can understand such a view as well, especially after my recent development of an interpretation that interpolates between Bohm and Bohr [1], suggesting that these two interpretations are not so different as people usually think.

[1] http://xxx.lanl.gov/abs/1112.2034

And it is arguable that unifying interpretations is of greater value than coming up with yet new ones! Interpolations move us toward a kind of "continuity of interpretation" of quantum mechanics.

 

[lugita15]

BM is deterministic, but not superdeterministic. Both contextuality and decoherence play important roles in BM.

Perhaps superdeterministic is a bit too strong, but do you at least agree that Bohmians generally do not believe in the free will of the experimenter, even if that is not the reason why Bell's inequality is violated? So in that sense, a measurement was a totally predetermined even in the history of the universe, and thus there's nothing special about contextuality: just like any interaction between particles can can change the states of the particles involved, a measurement of a particle can change its state.

As for decoherence, if you truly don't believe in wavefunction collapse, then how is measurement philosophically important? It just so happens that under certain circumstances the wave function gets so smeared out by a large number of particle interactions that it becomes difficult to detect wave properties; who cares?

 

[akhmeteli]

BM is deterministic, but not superdeterministic.

On the other hand, if you give up locality, then it is easy to construct a nonlocal-realistic model consistent with QM. The simplest known model of that sort is Bohmian mechanics, which also turns out to be a superdeterministic model (no free will).

Is this apparent contradiction caused by different definitions?