What exactly does quantum entanglement imply?

[.Scott]

Alright, so just to be clear of your interpretation on quantum entanglement-

You believe what occurs is that two particles become entangled, and upon measurement of one particle, the superposition of both systems instantly crashes and they become defined states.

However the instant superposition crash of both particles does not imply faster than light communication/non-locality; rather just correlation.

Is this correct?

Let me chime in here.
First, I think that the word "communication" has too much baggage to be used in describing what happens with entangled particles. It suggest that information is moving from one to the other. In fact the experimental results are statistics that do not indicate any directionality. Also, the quantum states we are working with seem to have a life of their own - independent of other quantum states of the particle - see quantum Cheshire cat http://phys.org/news/2015-06-quantum-cheshire-cat-effect-standard.html. So the common way of "explaining" the observed correlations is to describe the state shared by the entangled particles without presuming that the state exists "at" either particle.

Regarding "upon measurement of one particle, the superposition of both systems instantly crashes".
I don't like the term "instantly" - it not as bad as "simultaneously" which would be outrightly wrong, but it's almost that bad.
The problem is that there is no instant in time that is independent of the reference-frame when the collapse happens. It's not that its wrong, it just doesn't fit with the common notion of "instant".

For example, if both particles are measured at about the same time - so that from a reference frame shared by both detectors they are measure within a picosecond of each other and the detectors are a few meters apart, we would call this a space-like separation of the events. Depending on your reference frame, you could say that either measurement A precedes B or measurement B precedes A. And so what is the "instant" when the collapse occurs? And which particle collapsed first? It is meaningless to say.

Now, you might ask, what if the measurements are time-like separated - so that all reference frames agree that A was measured before B? Well, we get exactly the same statistics from time-like separation as we do from space-like separation, so do we really want to say something different happens with one then the other? So if I have one particle stored in a box and it's entangled twin was measure last year, do I want to say that the one in the box collapsed last year? Probably not. We probably don't want to talk about "when" the collapsed occurred because it didn't happen in a way that corresponds to the common experience of "when".

 

[DrChinese]

DrChinese, do you then believe that particles have properties irrespective of observation?

Not in the classical sense, no, and not in my view. For example: If you knew property p precisely, then non-commuting property q is completely indeterminate and cannot be said to have *any* value.

If you choose to call that indeterminate property to be "existing" then it is, else it isn't, but either way it would be a matter of your semantics and nothing else. It still (in my viewpoint) has no value.

 

[gjonesy]

Unfortunately, this is the classic example of local variables at work - we know that a shoe is created with a size, a color, and a handedness footedness and it has these properties even if we don't measure them. Entangled quantum particles don't behave like that..

Suppose you and I somewhere else in the universe are looking at the pairs of shoes as they come by, but we're each only allowed to measure at random one of the three properties: color (white or red), size (9 or 10), left-foot or right-foot. If, for a given pair, you measure the size and get nine and I measure the color and get red, we might conclude when we compare notes that you had a white size nine shoe of unknown footedness while I had a red size nine shoe of unknown footedness; similar logic works for all the other possible pairs of measurements. (If we both randomly measure the same property, the other two will be unknown even after we compare notes).

However, suppose that when we compare notes we discovered that the number of white size nine shoes (using the logic above, where we combine your measurement and mine to infer something about my shoe) that I saw is more than the sum of the number of white left shoes that passed me plus the the number of size nine right shoes? That's the equivalent quantum mechanical prediction; it has been confirmed experimentally and it tells us pretty clearly that the two particles in the entangled pair were not created with definite values of all three attributes.

The shoe description is an example of how simple English descriptions become very complicated, its just an illustration of why correlated pairs do not need to "communicate" faster than the speed of light. like the Alice and Bob description featured here.

http://www.livescience.com/50262-spooky-action-is-real.html

There are many more "bad" simple English descriptions that unless you really study and research the material its fairly hard to paint an accurate picture of what entanglement really is.

 

[Jeff Rosenbury]

Thanks I'll take a look at it.
Also, if you are correct and there is no communication between the particles, is non-locality not necessary then?

I don't think he claimed no communication exists. He claimed it might not exist or to put it another way, it is not needed. There are other possible solutions.

The math is the math. "Why" is a human question and says more about us than the underlying physics I think.

In any case, simple hidden variables are disallowed by Bell. Simple FTL communication is disallowed by Einstein. That doesn't mean a more complex hidden variable theory might not work, or a more complex FTL communications theory might work, or both, or something else entirely. But any such theory would simply be a way to translate our human, macroscopic understanding to math which already works.

Or I could be wrong.

 

[StevieTNZ]

You may find this interview helpful:
Four minutes 55 seconds long.
@1:44 he talks about quantum correlations (entanglement).

 

[bhobba]

You believe what occurs is that two particles become entangled, and upon measurement of one particle, the superposition of both systems instantly crashes and they become defined states.

Its the collapse of the wave-function issue in another guise. The observation doesn't occur instantaneously - but when its completed the state has changed so we say entanglement has been broken - and for simplicity its taken as instantaneously. Its like when you flip a coin - it doesn't become heads or tails instantaneously but when it does we say that the probability of a head or tail as 1/2 instantly changed to one being zero and the other one. Its of no concern because probabilities do not exist in a real sense but as a theoretical concept. The same with a superposition and quantum states in general. Note that's the formalism - interpretations can and sometimes do have a different take.

Although the math is advanced the following post may help in understanding the role of states in the QM formalism (see post 137):
https://www.physicsforums.com/threads/the-born-rule-in-many-worlds.763139/page-7

Basically they are, like probabilities, just a useful concept and do not exist in a real sense like say the results of observations that are very real.

Thanks
Bill

 

[Daniel K]

You may find this interview helpful:
Four minutes 55 seconds long.
@1:44 he talks about quantum correlations (entanglement).

I think the reason why myself and so many other individuals find entanglement confusing is because of stuff like this. This video is directly contradicting with many peoples' statements within this thread, and so it can difficult to perceive which description is the correct one.

In the video, the person being interviewed says that quantum non-locality is the single solution to the correlations found in quantum entanglement. However, within this thread, they are many people who have advocated that quantum non-locality isn't even real.

 

[bhobba]

In the video, the person being interviewed says that quantum non-locality is the single solution to the correlations found in quantum entanglement.

That's the reason we have acceptable sources around here. Videos like that generally are not acceptable. I haven't seen it, but can assure you, locality is only violated if you want realism (ie like the slips of paper) - technically its called counterfactual definiteness:
http://www.johnboccio.com/research/quantum/notes/paper.pdf

Here you get the real deal from people who have studied it in detail.

Everything I have said is standard textbook stuff.

Generally videos like you posted are opinions - but don't make clear its an opinion.

Thanks
Bill

 

[Daniel K]

That's the reason we have acceptable sources around here. Videos like that generally are not acceptable. I haven't seen it, but can assure you, locality is only violated if you want realism (ie like the slips of paper) - technically its called counterfactual definiteness:
http://www.johnboccio.com/research/quantum/notes/paper.pdf

Thanks for the paper - I'll read it in a bit!

Also before I start reading it, can you tell me if it will only deal with the bell inequality or will it also explain why quantum entanglement does not imply faster than light communication?

 

[bhobba]

Also before I start reading it, can you tell me if it will only deal with the bell inequality or will it also explain why quantum entanglement does not imply faster than light communication?

It deals with Bells Theorem that states QM does not allow both locality and realism - by realism it means QM properties are like the slips of paper - they are red or green regardless of if you observe them or not - as I have been saying in this thread. If you reject realism you can have locality ie no FTL.

Thanks
Bill

 

[stevendaryl]

I think that there is a sense of "locality" whereby QM with definite outcomes (as opposed to MWI) is nonlocal. If Bob performs a measurement confined to some region of space $R_{bob}$, then we can say that the outcome is ${local}_{daryl}$ (to distinguish it from a dozen other subtly different notions of "local") if it depends only on conditions in region $R_{bob}$, and not on conditions in other, distant regions. Conversely, we can say that the outcome is ${nonlocal}_{daryl}$ if facts about a second region $R_{alice}$ can tell us more about Bob's outcome than facts about $R_{bob}$ alone can.

Quantum mechanical correlations are certainly nonlocal in this sense. In an EPR-type experiment with anti-correlated fermions, if Alice measures spin-up for her particle along an axis $\vec{\alpha}$, then (assuming she performs her measurement slightly before Bob performs his), she knows that Bob will measure spin-down along that axis. But there is nothing about conditions local to Bob's measurement that could be used to predict this outcome ahead of time. So Bob's outcome is ${nonlocal}_{daryl}$.

Note: The classical analog of the EPR experiment uses two slips of paper, one red and one green, placed into envelopes. One envelope is sent to Alice, and the other is sent to Bob. When Alice opens her envelope and sees a red piece of paper, she knows that Bob's result will be green. But that's not ${nonlocal}_{daryl}$, because Bob's outcome is completely determined by the state of his envelope, and that state is a fact local to Bob.

 

[.Scott]

That's the reason we have acceptable sources around here. Videos like that generally are not acceptable. I haven't seen it, but can assure you, locality is only violated if you want realism (ie like the slips of paper) - technically its called counterfactual definiteness:
http://www.johnboccio.com/research/quantum/notes/paper.pdf

Here you get the real deal from people who have studied it in detail.

Everything I have said is standard textbook stuff.

Generally videos like you posted are opinions - but don't make clear its an opinion.

Thanks
Bill

One problem here is the definition of "locality" - and whether it, by definition, presumes realism. You are taking "locality" to simply exclude FTL information transfer. Others (including Wikipedia), tie it to any FTL-like influence.
Rereading the original 1979 Scientific American article (https://www.scientificamerican.com/media/pdf/197911_0158.pdf), it is difficult to resolve the definition. According to the author, Bernard d'Espagnat, QM is either unrealistic or non-local. But he defines realism as follows:

One is realism, the doctrine that regularities in observed phenomena are caused by some physical reality whose existence is independent of human observers.

By that definition, most physicists would assert that QM conforms to realism. That leaves locality, which he defines as follows:

The third premise is called Einstein separability or Einstein locality, and it states that no influence of any kind can propagate faster than the speed of light.

These seems to match the issue with QM much better than "realism". The terms "separability" and "locality" are used synonymously throughout the article. So, by those terms, QM would be a non-local realistic theory. In the article, QM is described repeatedly by combining the terms - as not a "local realistic theory".

 

[zonde]

If you reject realism you can have locality ie no FTL.

No, you can't (if I am correctly "translating" realism as counterfactual definiteness).
While it's true that Bell theorem relies on assumption of counterfactual definiteness (so that you might hope to construct valid local model of entanglement by relaxing this assumption) there are other proofs of Bell type inequalities. Say Eberhard in his derivation (http://dx.doi.org/10.1103/PhysRevA.47.R747) assumes factual definiteness and locality and still gets Bell type inequality.

 

[Nugatory]

No, you can't (if I am correctly "translating" realism as counterfactual definiteness).
While it's true that Bell theorem relies on assumption of counterfactual definiteness (so that you might hope to construct valid local model of entanglement by relaxing this assumption) there are other proofs of Bell type inequalities. Say Eberhard in his derivation (http://dx.doi.org/10.1103/PhysRevA.47.R747) assumes factual definiteness and locality and still gets Bell type inequality.

Is there a copy not behind paywall? The abstract is less than helpful

 

[stevendaryl]

No, you can't (if I am correctly "translating" realism as counterfactual definiteness).
While it's true that Bell theorem relies on assumption of counterfactual definiteness (so that you might hope to construct valid local model of entanglement by relaxing this assumption) there are other proofs of Bell type inequalities. Say Eberhard in his derivation (http://dx.doi.org/10.1103/PhysRevA.47.R747) assumes factual definiteness and locality and still gets Bell type inequality.

I have said before that I don't believe that Bell's theorem assumes counterfactual definiteness. As somebody else put it recently, in an EPR-type experiment with two experimenters Alice and Bob, the assumption needed to get Bell's theorem is this:

$P(B| \lambda, \beta, \alpha, A) = P(B|\lambda, \beta)$

where:

• $B$ is the result of some yes/no measurement performed by Bob,
  
• $\beta$ represents variables describing conditions at Bob's measuring device,
  
• $\alpha$ represents variables describing conditions at Alice's measurement device,
  
• $A$ is the result of some yes/no measurement performed by Alice,
• $\lambda$ represents variables describing conditions in the intersection of the backward lightcones of Alice's and Bob's measurements.

In other words, Bell is assuming that the only way for Alice's measurement result to reveal any information about Bob's measurement result is if both results are affected by their common past, described by $\lambda$. That isn't counter-factual definiteness, although you can derive counter-factual definiteness from that assumption, together with the fact of perfect correlation (or anti-correlation) between Alice's and Bob's results when they choose specific settings.

 

[zonde]

As somebody else put it recently

I believe this was the post https://www.physicsforums.com/threa...r-classical-fields.843270/page-4#post-5300265

 

[zonde]

Is there a copy not behind paywall? The abstract is less than helpful

I don't know of any. But I will post at least the assumptions used in Eberhard's derivation:
A theory is defined as being "local" if it predicts that, among these possible sequences of events [with the same number of events N], one can find four sequences (one for each setup [(α1,β1), (α2,β1), (α1,β2), (α2,β2)]) satisfying the following conditions:
(i) The fate of photon a is independent of the value of β, i.e., is the same in an event of the sequence corresponding to setup (α1,β1) as in the event with the same event number k for (α1,β2); also same fate for a in (α2,β1) and (α2,β2); this is true for all k's for these carefully selected sequences.
(ii) The fate of photon b is independent of the value of α, i.e., is the same in an event k of the sequences (α1,β1) and (α2,β1); also same fate for b in (α1,β2) and (α2,β2).
(iii) Among all sets of four sequences that one has been able to find with conditions (i) and (ii) satisfied, there are some for which all averages and correlations differ from the expectation values predicted by the theory by less than, let us say, ten standard deviations.

 

[bhobba]

I haven't seen it

I have now had a look at it.

Its opinon stuff.

He says we don't have a story that explain QM correlations.

That's entirely dependant on what you mean by the terms story and explain. We have a theory that explains it, and these days that theory can be presented in a very transparent an intuitive way:
http://arxiv.org/pdf/quant-ph/0101012.pdf

If that satisfies you is not a question science can answer - its purely a personal reaction. Its obvious the person in the video finds the current state of affairs unsatisfying. He is not the only one. But nature is as nature is - it doesn't have to oblige in providing something you find satisfactory.

Personally I think with the paper I linked above we have a very satisfactory 'story' - but again that's just an opinion. Opinion's are like bums - everyone has one - it doesn't make it right - especially in an area like QM where the formalism is agreed by everyone, but what it means is hotly debated.

Thanks
Bill

 

[nikman]

He says we don't have a story that explain QM correlations.

We don't have a story that explains important classical correlations either. Why does a well-conducted exit poll accurately (within the given margin of error) mirror final election results? I mean, you've got a couple of thousand randomly-selected people representing the data of millions. Where are your causal connections? Makes no sense, I tell you.

 

[Daniel K]

Can someone explain how removing realism allows quantum entanglement to keep in touch with locality?
Sorry if this seems like a very basic question.

 

[bhobba]

We don't have a story that explains important classical correlations either. Why does a well-conducted exit poll accurately (within the given margin of error) mirror final election results? I mean, you've got a couple of thousand randomly-selected people representing the data of millions. Where are your causal connections? Makes no sense, I tell you.

There are statistical theorems such as the Central Limit Theorem that explains that.

Thanks
Bil

 

[bhobba]

Can someone explain how removing realism allows quantum entanglement to keep in touch with locality?
Sorry if this seems like a very basic question.

Previously I wrote:

Imagine you have two slips of paper a red and a green one and put them in envelopes. Send one to the other side of universe and keep the other. Open the envelope and you see red - you immediately know the other is grenn, and conversely. Nothing weird or mysterious here. That's all that's going on with entanglement with a twist I will explain.

Now for the QM twist. It turns out the paper analogy is not quite the same as QM. The correlation is a bit different - its still just a correlation - but has statistical properties different to the paper example. Why the difference? The difference is in QM things do not have properties until observed to have them, whereas the slips of paper remain red or green at all times. But what if we insist it's like the slips of paper - then it turns out you need some kind of non local superluminal communication. That's really weird. But there is nothing compelling anyone to insist its like the slips of paper - simply accept QM allows a different kind of correlation and things are no longer mysterious.

I am scratching my head what is unclear in the above.

In other words, if you do not assume FTL then you can't have properties independent of observation. If you remove that requirement then you do not need FTL. That is Bells theorem I gave a link to.

Thanks
Bill

 

[nikman]

There are statistical theorems such as the Central Limit Theorem that explains that.

The CLT is mathematical. It helps to establish the consistency and validity of statistical probability as a mathematical theory. I used the word "causal" deliberately. There is no physical model describing or explaining the measured correlations between responses of the sample population of individual voters in an exit poll and the general population's final vote count on a given election day. Here may or may not lie (very deeply) hidden variables, but we'll in all likelihood never know. My point: if someone uses the "it's voodoo" argument against quantum entanglement s/he should be prepared to use it against classical correlations as well. Those people tend not to do that because macroscopic statistical sampling indisputably proves itself all the time and they'd look dumb.

 

[Weddgyr]

There are actually many physical and/or computational models which could describe or at least estimate voter preferences in elections. At the very least it can be said that hidden variable models can predict such events (e.g. The heretofore undiscovered "Whig" and "Tory" glands). Classical correlations can be both models and understood in a way that QM correlations cannot. QM correlations really are "voodoo" in the sense that while we can see the results of the magic trick before us, we have no way of grasping how it took place.

 

[Daniel K]

I am scratching my head what is unclear in the above.

In other words, if you do not assume FTL then you can't have properties independent of observation. If you remove that requirement then you do not need FTL. That is Bells theorem I gave a link to.

I understand that if you remove realism from the situation then you can keep in touch with locality.
However my question is how does removing realism keep in touch with locality?

 

[stevendaryl]

I understand that if you remove realism from the situation then you can keep in touch with locality.
However my question is how does removing realism keep in touch with locality?

I think it's more that you can't PROVE nonlocality without assuming realism.

 

[bhobba]

The CLT is mathematical..

When applied it is more than mere mathematics and hinges on what an event is in your model.

This thread however is not a thread about the foundations of probability and the use of the Kolmogerov axioms in applications. It needs a thread of its own not in the QM section - but in the probability section. But the early pages of Feller's classic on probability discusses it.

Thanks
Bill

 

[bhobba]

However my question is how does removing realism keep in touch with locality?

One more time.

There is a theorem, called Bells theorem, I gave a link to that states: If QM is true and you want counterfactual definiteness then you need FTL. It also states if you don't insist on conterfactual definitiveness then you can keep locality.

How does it do that? Its a logical deduction like any theorem.

Thanks
Bill

 

[bhobba]

QM correlations really are "voodoo" in the sense that while we can see the results of the magic trick before us, we have no way of grasping how it took place.

I don't get this, I really don't. Why people get caught up in this dialectic of how does something happen, and what 'explains' means, has me beat. Its just useless philosophical mumbo jumbo of zero scientific value.

A logical deduction from the postulates of a theory, model, whatever you want to call it, explains how something occurs and is what science is all about. Quantum correlations are fully explained by QM. End of story. The rest is philosophical mumbo jumbo and IMHO a total waste of time. You achieve nothing in a fundamental sense expect maybe find a different explanation, that also by the nature of explanation, rests on other assumptions. You are simply looking for assumptions you like better. That isn't science.

QM is not 'voodoo' - it explains in a logical way its deductions. Some people just don't like what it says. That's fine - but don't try and make out there is some essential problem with QM like 'voodoo' - there isnt.

My old statistical modelling professor would occasionally touch on such things - his eyes would roll back - he would say - its like studying Niechie - pointless really. Guess what - he was right.

Thanks
Bill

 

[stevendaryl]

I don't get this, I really don't. Why people get caught up in this dialectic of how does something happen, and what 'explains' means, has me beat. Its just useless philosophical mumbo jumbo of zero scientific value.

I think that's completely unfair. QM really is a conceptual mess, and it seems uncharitable for you to criticize people for pointing out this truth.

Why is it a conceptual mess? If you say that QM is about making predictions about the results of measurements, that's a rule of thumb but it isn't actually a coherent thing to believe. A measurement is not a separate category. A measurement is just an interaction between one system (the thing being measured) and another (the thing doing the measuring). How can measurements have definite outcomes if other interactions don't? That doesn't really make any sense. Of course, you can just take it on faith, and get on with things, because for practical purposes, we can distinguish measurements from other interactions, but it doesn't actually make sense.

The other thing that doesn't make any sense is treating improper mixed states as if they are proper mixed states. They are very different things, because in the first case, you don't have uncertainty due to lack of knowledge, you just have a very complicated pure state. To treat it as a proper mixed state (where the weights represent lack of information) is to assume something that is false.

It's a complete mess, conceptually. Usually in science, the intuition that something is a mess is a good guide in searching for a better theory. In QM, that hasn't proved to be the case, because nobody has really made any progress. But it's not an analytic truth that it is impossible to make sense of QM.

 

[bhobba]

I think that's completely unfair. QM really is a conceptual mess, and it seems uncharitable for you to criticize people for pointing out this truth.

That, IMHO, is not true. I gave a link how the formalism comes from reasonable axioms. There is no 'voodo' involved, no conceptual mess. Its a theory about observations. Everyone has an intuitive idea of what observation means and that is what's used to start with - later it becomes more precise.

All there is, is people wanting to force a theory to conform to their intuition and present that as some kind of problem. There may have been issues in the early days of QM - there isn't these days.

Its OK to not like QM and want to put it on a foundation you like better. I get that. But to present it as more than that, to present that it contains some kind of 'voodoo' when all that going on is you don't like it is wrong.

The principles of QM fully explain entanglement - end of story.

Thanks
Bill

 

[stevendaryl]

That, IMHO, is not true. I gave a link how the formalism comes from reasonable axioms. There is no 'voodo' involved, no conceptual mess.

I don't agree. The standard way of presenting QM relies on making a distinction between "measurements" and other interactions. That distinction is incoherent, because a measurement is just a complicated collection of the same sort of interactions that govern electrons and photons. Any formalism for QM that makes measurements into a separate category is relying on voodoo.

 

[stevendaryl]

All there is, is people wanting to force a theory to conform to their intuition and present that as some kind of problem.

That is not an accurate description of people's complaints with QM.

 

[bhobba]

I don't agree. The standard way of presenting QM relies on making a distinction between "measurements" and other interactions..

See the link I gave previously:
http://arxiv.org/pdf/quant-ph/0101012.pdf

See figure 1.

That's the conceptual model its based on and gets generalised and elucidated further as the theory is developed.

Thanks
Bill

 

[bhobba]

That is not an accurate description of people's complaints with QM.

What's not accurate? Saying QM correlations are 'voodoo' when they are fully explainable by the principles of QM. Sorry - but that's just plain WRONG.

Thanks
Bill