A question regarding the Copenhagen interpretation.

[DevilsAvocado]

The main point I was making is that I don't think the key issue is whether we choose to drop realism (drop that the system in some sense "contains" its own attributes and information) or drop localism (drop that the information and attributes of an object move strictly with that object), because it doesn't really matter where we "put" the attributes, what matters is that we have two observers who are going to notice Bell-inequality-breaking correlations when they compare notes. Thus what we really have to decide between is whether we regard the source of those correlations to be some special kind of signal that can propagate superluminally without propagating any information between the points, or whether we regard the system as having a holistic correlation that does not require any kind of propagation whatsoever to maintain. Once you choose between those two issues, then whether you say you are "dropping locality" or "dropping realism" becomes a rather moot issue, because depending on other assumptions I make, I could characterize your choice in either of those terms, but not in any way that matters much-- you've already chosen the key language for speaking about the source of the correlations.

I'm not sure I understand... "I could characterize your choice in either of those terms, but not in any way that matters much", to me sounds like if the question of non-locality vs. non-realism is not a "big deal"...

Huum, if we take the non-local hidden variables in Bohmian mechanics, they give us quite a "cozy world" (in the couch watching football). Yes, non-locality is a little bit strange, but it will never cause any turmoil on NYSE or TSE, with stockbrokers cheating with "EPRB-FTL-orders", creating crashes on the binary options market. This will never happen (of course Bohmian mechanics has "some" work left to do on the SR side).

On the other hand... if we take the Two-state vector formalism, in which the present is caused by quantum states of the past and future, taken in combination... god knows what will happen if any stockbroker finds out how to 'utilize' these quantum states... :)

Just as an example of quite drastic differences in possible solutions.

 

[DevilsAvocado]

@Ken G: Another good idea is, never to talk about "realism" or even "local realism" in discussions about the interpretation of quantum theory without to define very clearly what you mean by that. I've never heard a clear definition in mathematical terms, what's meant by the words "realism" or "local realism" yet. Usually it's used by philosophers in a kind of muttering rather than scientifically clearly defined terms. So I'm not able to discuss that notions properly.

I can't say it's a "global agreement", but many would adhere to that definite values is a 'prerequisite' for realism, i.e. the moon is there even when nobody is watching, and quantum particles do have definite states (before leaving a common source).

If you like this definition, then DrC has a beautiful page that will take you through the process of mathematically prove that these definite states, if local, are not compatible with the predictions of QM, in a quite simple and straightforward form.

@DevilsAvocado: Yes, that's another very clear feature of local relativistic QFTs! They all fulfill automatically the linked-cluster principle: I.e., there is no influence by space like separated events on measurements by means of local interactions.

In our case: No matter what Bob does with his photon, i.e., whether he determines its polarization state before or after Alice does her experiment (or if Bob and Alice are in relative motion to each other, no matter whether Bob's measurement act is in the future or past lightcone or space-like separated to Alice's measurement act), Alice will always simply measure a stream of unpolarized photons (supposed Alice and Bob are sent a sequence of independently prepared entangled photon pairs from the parametric-down-conversion source). That's so, because the experiment is well-described by standard QED, which is a local relativistic QFT.

Yes of course, Alice & Bob will always measure 100% random outcomes in their local apparatus, no doubt about that, i.e. there will be absolutely no difference whatsoever (at this stage) from measuring single "normal" photons. However, the 'weirdness' comes when Alice & Bob compare their data (through classical channels) and discover that (contrary to "normal" photons) their entangled photons shows correlations that can be coupled to the relative settings between their polarizers, i.e. if Alice has put her polarizer at 32.5° and Bob his polarizer at 10°, the relative angle between Alice & Bob will be 32.5 – 10 = 22.5, which gives cos²(22.5°) = 85%, i.e. if sending 1,000 entangled photon pairs, 850 will be correlated (or anti-correlated depending on type of Bell state).

This is just the way it is, and it's obviously not compatible with local realism.

 

[Jilang]

On the other hand... if we take the Two-state vector formalism, in which the present is caused by quantum states of the past and future, taken in combination... god knows what will happen if any stockbroker finds out how to 'utilize' these quantum states... :)

Just as an example of quite drastic differences in possible solutions.

I am not convinced there would be such drastic consequences. Time symmetric QM has a reverse causality, but appears to be limited in its extent to the area of uncertainty so it would not improve the stockbroker's lot! Bear in mind that there are still many possible final states so intermediate states would not be uniquely determined, just narrowed down somewhat.

http://jamesowenweatherall.com/SCPPRG/AharonovPopescuTollaksen2010PhysToday_TimeSymQM.pdf

 

[DevilsAvocado]

I am not convinced there would be such drastic consequences.

Of course you are right. As interpretations, there can't be any 'practical' difference between Bohmian mechanics and Two-state vector formalism, since they are both forced to yield the same predictions as standard quantum mechanics (otherwise it would not be interpretations).

However, in some distant future, someone could make a clever experiment that tells us the correct interpretation, which ought to make some difference... (I hope)

However, there are other "beasts" out there (can't talk about it) which makes for example Bohmian mechanics look like your "Familiar Grandma" or a walk in the park, in comparison (trust me), and this is something completely different already on the drawing board, which then makes locality vs. realism a quite significant question (i.e. if this turns out to be true in the end).

 

[Ken G]

Huum, if we take the non-local hidden variables in Bohmian mechanics, they give us quite a "cozy world" (in the couch watching football). Yes, non-locality is a little bit strange, but it will never cause any turmoil on NYSE or TSE, with stockbrokers cheating with "EPRB-FTL-orders", creating crashes on the binary options market. This will never happen (of course Bohmian mechanics has "some" work left to do on the SR side).

On the other hand... if we take the Two-state vector formalism, in which the present is caused by quantum states of the past and future, taken in combination... god knows what will happen if any stockbroker finds out how to 'utilize' these quantum states...

But that point isn't disagreeing what what I said. I'm claiming that the important choice you make there is between some kind of instantaneous signal which you regard as passing between the subsystems as you sit on the couch, or some kind of holistic supersystem that encodes the correlation into something larger than the subsystems, as in the case of your two-state vectors. I'm saying that it doesn't matter which of those you regard as the one that breaks locality, and which breaks realism, because I can claim that either can be regarded as doing either, depending on supplemental assumptions that don't matter much. If you say that Bohmian pilot waves are an example of nonlocalism, and the two-state vectors are nonrealism, I'll just say, why can't I regard the pilot wave as something unreal yet local (there is a location to the pilot wave signal at any time, and it communicates no information to any observers so it is not nonlocal, but it cannot itself be observed, so it is unreal), and the two-state vector as something real but nonlocal (it is real because I can choose to regard those two vectors as perfectly real, but I can't localize them, because they counterpropagate in time). In either case, what matters is the mechanism you are invoking to sustain the correlations, and its possible ramifications as you say, but not whether you regard that mechanism as something nonlocal, or nonreal.

 

[bhobba]

Maybe this is correct, but perhaps safest is to let Bill explain himself.

Its easy.

In Bell type setups things are so arranged that one person will get a 'black' paper and the other a 'white'. We just don't know which got which, and will not know until its observed, or even if it has the property at all prior to observation, which is the peculiar twist QM has.

What Bells theorem shows is if its analogous to the paper situation where it has properties independent of observation then locality is violated. But otherwise its exactly the same. The purpose of the envelopes is to simulate as much as possible not knowing until observed. Of course in this analogy it does have the property, there is no classical situation that will mimic it exactly, but its just to get a bit of a handle on it.

BTW its not something I came up with - its in the standard textbook I quoted.

Thanks
Bill

 

[DevilsAvocado]

In either case, what matters is the mechanism you are invoking to sustain the correlations, and its possible ramifications as you say, but not whether you regard that mechanism as something nonlocal, or nonreal.

Okay, I understand your point now, thanks.

 

[DevilsAvocado]

In Bell type setups things are so arranged that one person will get a 'black' paper and the other a 'white'. We just don't know which got which, and will not know until its observed,

I'm afraid I must oppose again. This is not the proper description of a Bell type setup, and the crucial point is that to get correlations like 85%, there must be cases where Alice & Bob gets 'black' + 'black', as well as 'white' + 'white'. This is actually the one thing that finally settled the Bohr–Einstein debates.

It looks like you and vanhees are reading the same standard textbook:

Now you can ask, what's the probability that Alice and Bob find their photons in any possible combination. According to the rules of quantum theory you get

Alice's photon      Bob's photon     Probability
---------------------------------------------------
H                       H                    0
H                       V                    1/2
V                       H                    1/2
V                       V                    0

If this is what your textbook is stating, it really needs an update ...

 

[Jilang]

I'm afraid I must oppose again. This is not the proper description of a Bell type setup, and the crucial point is that to get correlations like 85%, there must be cases where Alice & Bob gets 'black' + 'black', as well as 'white' + 'white'.

Are you sure you got this the right way round? I understood that the (opposite) correlations are greater than you would expect from non-correlated particles.

You never get a black + black , but you get a black + white more times than you would expect from just computing the cosines between an artitary angle of spin relative to an arbitrary angle of measurement. (For Black read spin up and White spin down). If you want to try it the maths is pretty easy. Grab a pencil and start drawing, evaluate the cosines and compare to what is observed - pretty amazing!

 

[bhobba]

If this is what your textbook is stating, it really needs an update ...

Run that by me again.

In Bell type experiments the spins are anti-correlated. This means it's IMPOSSIBLE to not get opposite spins - its inherent in the setup.

This is the analogue of the paper situation I described - as much as a classical situation can describe Quantum weirdness anyway.

Thanks
Bill

 

[Ken G]

To get the weirdness, you must have Alice doing a black/white observation, and Bob doing a blue/red observation, or some such thing. If they both do black/white, you get the simple table in post 78, and there's nothing weird about that. But bhobba's point still holds, if I understand his point correctly, that if Alice does black/white, and Bob does blue/red, then the correlations between those outcomes simply cannot be explained by saying that the papers already know what outcome each observation will give before the measurement is done. If Alice could choose to do white/black, or blue/red, then half the papers cannot be thinking "if she does white/black, I'll be black" while the other half think "if she does white/black, I'll be white," independently of Bob's choices. If Alice measures white/black on an ensemble, half will be white and half black, but the way those populations will correlate with Bob's blue/red measurement will require that the envelopes themselves could not have "known" which outcome they'd get, independently of Bob's choice to measure blue/red. So it's about the envelope "not knowing" what color is inside it, that's the strangeness. It requires either a signal to tell it what Bob chose to measure (without communicating any information to Alice about that choice), or it requires a holistic character of the system that it is not just an envelope with a paper in it, but two envelopes with two papers, until all the data is collected. (Of course, there's no real advantage in replacing the particles with up/down spins with envelopes with paper in them, because real envelopes with paper in them could not retain the entanglements, but I believe the purpose of the device is to say "we don't really know how spins ought to behave because we have so little experience with them, but we do know how envelopes should behave, so the contrast is why the spins are behaving weirdly.")

 

[DevilsAvocado]

Are you sure you got this the right way round? I understood that the (opposite) correlations are greater than you would expect from non-correlated particles.

Not sure I understand the question, but generally it depends on the cut-angle of the BBO crystal in the SPDC process, which generates two types of Bell states, Type I and Type II, which means that for Type I the photons share the same polarization $|HH\rangle + |VV\rangle$ and for Type II they are orthogonally polarized $|HV\rangle + |VH\rangle$, which means that if Alice & Bob use a BBO Type I crystal and set their polarizers at same angle they will always get the same results i.e. [1, 1] or [0, 0], whereas if they use a BBO Type II they will always get opposite results i.e. [1, 0] or [0, 1]*.

*[1] means the photon went through the polarizer, [0] means stopped.

You never get a black + black , but you get a black + white more times

There seems to be some fundamental confusion around this issue, to keep things as simple as possible, I'll give you this example:

• Alice & Bob are using a BBO Type II crystal.
• Alice & Bob set their polarizers at angle 0°.
• Alice & Bob run a set of 1,000 entangled photons.
• When Alice & Bob compare their results, they are always opposite i.e. [1, 0] or [0, 1].
• Now Alice keeps her setting at 0° and Bob change his to 90°.
• Alice & Bob run a new set of 1,000 entangled photons.

Question: What will the results show now?

 

[DevilsAvocado]

In Bell type experiments the spins are anti-correlated. This means it's IMPOSSIBLE to not get opposite spins - its inherent in the setup.

Are you sure? Check out post #82.

Seriously, what are we talking about – spin or correlations in measurement outcome?

This is the analogue of the paper situation I described

You wrote "one person will get a 'black' paper and the other a 'white'. We just don't know which got which" and this now means the anti-correlated spin coming out from the BBO crystal...

Is this really the main question in EPR-Bell??

 

[bhobba]

Seriously, what are we talking about

What we are talking about is EPR-Bell type experiments:
http://en.wikipedia.org/wiki/EPR_paradox
'Alice now measures the spin along the z-axis. She can obtain one of two possible outcomes: +z or −z. Suppose she gets +z. According to the Copenhagen interpretation of quantum mechanics, the quantum state of the system collapses into state I. The quantum state determines the probable outcomes of any measurement performed on the system. In this case, if Bob subsequently measures spin along the z-axis, there is 100% probability that he will obtain −z. Similarly, if Alice gets −z, Bob will get +z.'

The outcomes are anti-correlated with 100% certainty - nothing else is possible. Anti-correlated means only two outcomes are possible called + and -. If one measures + the other is automatically - and conversely.

You can never get ++ or --.

This is from the definition of the experiment. There is no ifs or buts about it - it's its very definition.

Thanks
Bill

 

[bhobba]

Is this really the main question in EPR-Bell??

The main question in EPR-Bell is Bells Theroem's consequences.

I have zero idea what you are even arguing about.

Thanks
Bill

 

[DevilsAvocado]

You can never get ++ or --.

This is from the definition of the experiment. There is no ifs or buts about it - it's its very definition.

But Bill... your link is to the 1935 EPR paradox...

Check out this instead:
http://en.wikipedia.org/wiki/Bell's_theorem

 

[DevilsAvocado]

To get the weirdness, you must have Alice doing a black/white observation, and Bob doing a blue/red observation, or some such thing. If they both do black/white, you get the simple table in post 78, and there's nothing weird about that. But bhobba's point still holds, if I understand his point correctly, that if Alice does black/white, and Bob does blue/red, then the correlations between those outcomes simply cannot be explained by saying that the papers already know what outcome each observation will give before the measurement is done.

I think there is no question (in standard QM) that the entangled photons are in a (shared) superposition (of spin); also the particles cannot be described as two "individuals", but as a "composite system" (or something like that).

Maybe Bill's "black and white paper" isn't a huge problem as such, as there is no escape from the simple formula cos²(a-b), where a and b are the space-like separated polarizer-settings of Alice & Bob.

However, it could be an issue if we only talk about "one person will get a 'black' paper and the other a 'white'" and combine this with statements like "there's nothing Earth shattering going on" and "common source explanation", etc ...

I might be exaggerating, but to me it looks like there is (at least some) risk that the "casual reader" (in worst case) could get the "wrong impression".

We don't want to do that, do we?

 

[bhobba]

But Bill... your link is to the 1935 EPR paradox...

Errrrr.

What's your point?

I have been talking about EPR type experiments as proposed by Bell, Einstein etc

What have you been talking about?

Thanks
Bill

 

[Ken G]

I think there is no question (in standard QM) that the entangled photons are in a (shared) superposition (of spin); also the particles cannot be described as two "individuals", but as a "composite system" (or something like that).

Yes, the standard interpretation takes the "holistic" approach, though Bohmians prefer the "instantaneous signal" approach because they want everything to be as locally classical as possible. Either way, it seems weird to us, in our daily lives. The usual point of invoking some classical image like black and white paper is to place the issue in classical language, to underscore how strange it seems. Granted, it is something of a cheat to make the situation appear strange by framing it in a language that has no real business applying, but it's a standard device when we want to stress how much different QM is from our daily experience, rather than when we want to just explain what QM says.

However, it could be an issue if we only talk about "one person will get a 'black' paper and the other a 'white'" and combine this with statements like "there's nothing Earth shattering going on" and "common source explanation", etc ...

I'm not sure what those elements were trying to evince. I was happy with the simple remark that the envelope can't know what color paper is inside it (which actually follows from the simple fact that it doesn't know what colors Alice will choose to look for), and what's more, it can't even know what color it will give in advance of any given choice Alice might make, because it wouldn't produce the right correlations for it to know that independently of the rest of what is happening in the entanglement. If that is what bhobba was saying, then the two of you are not disagreeing, if he was saying something else, I don't know what he would have meant.

 

[bhobba]

I was happy with the simple remark that the envelope can't know what color paper is inside it (which actually follows from the simple fact that it doesn't know what colors Alice will choose to look for), and what's more, it can't even know what color it will give in advance of any given choice Alice might make, because it wouldn't produce the right correlations for it to know that independently of the rest of what is happening in the entanglement. If that is what bhobba was saying, then the two of you are not disagreeing, if he was saying something else, I don't know what he would have meant.

What I was saying isn't hard - and yes that's part of it ie the envelope doesn't know what color is inside, nor does the person that opens it until its opened.

All I am saying, and all the people that invoke this analogy are saying, is its not this really weird thing some seem to think it is. Its very much like the envelope analogy with the twist it doesn't have the property until observed. In fact within the decoherent histories interpretation its a very close analogy using the concept of frameworks they use.

You can get the detail in the textbook I alluded to before.

Thanks
Bill

 

[atyy]

If QM is non local is very interpretation dependent - that's the import of Bells Theorem and Einsteins error. QM rules out naive-reality ie local realism. If you reject realism (ie properties do not exist independent of observation) then locality is saved. If you keep it then locality is gone. But SR is still saved since it can't be used to send information which is what's required to sync clocks.

Basically all Bell type 'experiments' are doing is observing systems with spatial extent, and because of how its arranged if one thing in the system has a property on observation, so does the other thing - but they are spatially separated.

I have two pieces of paper, one black, and one white and put them in envelopes. I randomly send one to one person, and another to a different person. If any of those people open their envelope they immediately know what the other person will get when they open their envelope. Their is nothing Earth shattering going on. Same with Bell type experiments, with the twist we can't say it has the property of blackness or whiteness until observation.

Griffiths book - Consistent Quantum Theory discusses it from this interesting perspective:
https://www.amazon.com/dp/0521539293/?tag=pfamazon01-20

Thanks
Bill

I think Griffiths's book indicates that the nonrealism has to be much stronger than just not having the properties before measurement. The nonrealism has to assume that "reality" can be described in various incompatible ways which cannot be combined http://quantum.phys.cmu.edu/CQT/chaps/cqt27.pdf. Griffiths says his interpretation is realistic and local, but if one wants to argue that it is not realistic, that seems plausible. According to an FAQ about consistent histories, "Colored slips of paper, one red and one green, are placed in two opaque envelopes, which are then mailed to scientists in Atlanta and Boston. The scientist who opens the envelope in Atlanta and finds a red slip of paper can immediately infer, given the experimental protocol, the color of the slip of paper contained in the envelope in Boston, whether or not it has already been opened. There is nothing peculiar going on, and in particular there is no mysterious influence of one "measurement" on the other slip of paper." http://quantum.phys.cmu.edu/CHS/quest.html#EPR. Similarly, Griffths's book says that measurements reveal properties of a system before the measurement took place, and further says there is an independent reality within the consistent histories framework. http://quantum.phys.cmu.edu/CQT/chaps/cqt27.pdf. So I don't think the twist is that the cards don't have the colour before the measurement, but that reality can be described in incompatible ways.

 

[bhobba]

So I don't think the twist is that the cards don't have the colour before the measurement, but that reality can be described in incompatible ways.

In consistent histories things are analysed in terms of frameworks. One must specify a framework to analyse a particular situation and can't use incompatible ones. The other aspect is a history which is a sequence of projection operators and its only when an observation is made does a particular history become real.

If you want to see the exact detail of how Consistent Histories analyses EPR then its probably best to get the book. I have a copy and have gone through it - and in that interpretation everything works out perfectly OK - but its not an interpretation that can be discussed superficially - although I will make a stab.

I like Consistent Histories, but my issue with it is it started out as a minimalist interpretation, but turned out to be a bit more complicated.

In many ways its Many Worlds without the worlds. Because of that it has to cater to some things that is trivial in MW eg all histories occur in MW, so its trivial why a particular history happens - not so in Consistent Histories. Of course the Consistent History guys would say its just what's necessary to avoid the weirdness of MW.

I dug up the following discussing EPR in terms of Consistent Histories:
http://www.siue.edu/~evailat/pdf/qm12.pdf

'Although CH allows a realist understanding of quantum mechanics, it does not follow EPR in attributing quantum mechanically incompatible properties to a system. Griffiths gives an instructive story about what happens if one insists that P intersection Q is defined even if P and Q are incompatible. Consider a spin-half particle, and to simplify the notation, let [Z +] stand for the projector associated with Sz =1, and similarly for other projectors. Suppose now that the composite property Sz =1 intersection Sx =1 existed. Then its corresponding projector would have to project onto a subspace P of the two dimensional Hilbert space H of the spin-half particle. However, no such subspace can exist.'

Basically in EPR there are two consistent frameworks - one with +1 and -1 and the converse. Others simply do not exist so no problems can arise.

One framework corresponds to one envelope getting black and the other white, and the other framework the converse.

Thanks
Bill

 

[DevilsAvocado]

The outcomes are anti-correlated with 100% certainty - nothing else is possible. Anti-correlated means only two outcomes are possible called + and -. If one measures + the other is automatically - and conversely.

You can never get ++ or --.

This is from the definition of the experiment. There is no ifs or buts about it - it's its very definition.

One framework corresponds to one envelope getting black and the other white, and the other framework the converse.

Errrrr.

What's your point?

I have been talking about EPR type experiments as proposed by Bell, Einstein etc

What have you been talking about?

Bill, you have helped me many times and you know a lot more than I do, about different issues. However, it's only human to do mistakes – just ask me – I've done terribly embarrassing mistakes many times. But this time, all facts available clearly shows that, any confusion/mistake is not on my behalf.

My experience is also that the longer one defends an obvious mistake – the worse things get...

Therefore, as a friend, it is not in my interest to prolong this "unfortunate situation". If you don't get it this time, I'm afraid there's nothing more I can do:

https://en.wikipedia.org/wiki/Bell's_theorem]The[/PLAIN] probability of *the same result* being obtained at the two locations varies, depending on the relative angles at which the two spin measurements are made, and is strictly between *zero* and *one* for all relative angles other than perfectly parallel alignments (0° or 180°). Bell's theorem is concerned with correlations defined in terms of averages taken over very many trials of the experiment. [...] if the pairs of outcomes are always the same, the correlation is +1, no matter which same value each pair of outcomes have. If the pairs of outcomes are always opposite, the correlation is -1. Finally, if the pairs of outcomes are perfectly balanced, being 50% of the times in accordance, and 50% of the times opposite, the correlation, being an average, is 0. [...]

Measuring the spin of these entangled particles along anti-parallel directions—i.e., along the same axis but in opposite directions, the set of all results is perfectly correlated. On the other hand, if measurements are performed along parallel directions they always yield opposite results, and the set of measurements shows perfect anti-correlation. Finally, measurement at perpendicular directions has a 50% chance of matching, and the total set of measurements is uncorrelated. These basic cases are illustrated in the table below.

 

[Ken G]

All I am saying, and all the people that invoke this analogy are saying, is its not this really weird thing some seem to think it is. Its very much like the envelope analogy with the twist it doesn't have the property until observed.

Yet it seems to me that what needs to be clarified, by any classical analogy, is that there are actually two twists here, one which is already quite a bit different from our daily experience, but also another which is so different it is regarded as "spooky." The first twist is just that the envelope doesn't hold a particular color, but this has nothing to do with entanglement, it is just the concept of complementarity (a pure state in regard to one measurement is not a pure state in regard to another). So we cannot say "the envelope contains a white piece of paper and not a red one" until we know if white/black or red/blue is the measurement. That is a strange enough state of affairs, but we might still wish to imagine the envelope is thinking "if Alice measures white/black, I'll be white, and if red/blue, I'll be blue", independently with anything else that could be going on elsewhere. But entanglement ruins even that-- the envelope cannot independently know what result it will give to any given measurement, because there need to be correlations with other measurements that the envelope does not know about. These two twists are often contrasted with "Bertlmann's socks", where there's a left sock and a right sock so the envelope "knows" the measurement in question is going to be about leftness/rightness and so the envelope can know which it will produce (and it will be opposite the other envelope), even if Alice does not. But note that Bertlmann's socks fails to include either of the two types of twists I mentioned above, so sometimes it is not clear there are two completely different twists there to appreciate, not just one.

 

[stevendaryl]

The "twist" that quantum mechanics places on correlated values makes all the difference in the world, it seems to me.

In the classical situation there are two envelopes, one containing a black piece of paper, and one containing a white piece of paper. Alice gets one envelope, and Bob gets the other. When Alice opens her envelope, she immediately knows what's in Bob's envelope. This is explained as purely a change in knowledge--the envelope already was in a definite, but unknown state, and opening it just revealed information about this pre-existing state.

The difference with the quantum EPR experiment is that it is not consistent (at least not without jumping through strange hoops) to assume that each particle had a definite, but unknown state prior to measuring its spin. So an explanation purely in terms of a change knowledge doesn't seem possible.

On the other hand, an explanation in terms of causal influence doesn't seem very plausible, either, since this influence would have to propagate faster-than-light (which means back-in-time for some observers). To me, it's a really tough nut to crack. Maybe retrocausality or super-determinism would explain it, or MWI.

 

[Jilang]

This paper from Studies in History and Philosophy of Modern Physics

http://www.lophisc.org/wp-content/uploads/Price.pdf

sums up very nicely the sort of choices we are faced with for the different interpretations. The gist of it seems to be that if we are any sort of realist other than an Everettian or a Bohmian there is a choice to be made between time-asymmetric ontology and retro-causality and that this is true in QM in a manner that is not true in classical physics.

 

[Ken G]

The question I ask is, why not just conclude the marriage of causation and local realism is a classical notion? Both Einstein and Bohm seemed to say we should take classical thinking as our basic paradigm, and use it whenever we can, no matter how badly we need to retrofit it. That seems kind of opposite to me-- we built classical notions from a set of experiences, now we are having new experiences, so we need to be ready to relax old notions and build new ones. Local realism, and causation, seem like they need to be chucked, and retrocausation seems to have simply not gotten the memo that the whole causation notion should have been let go, other than as a perfectly viable theory of information propagation between people doing experiments and asking questions. In that spirit, we just say we have an entangled system, so we get entangled correlations, and leave it at that. Let QM speak for itself, with classical analogs used only as devices to show how QM isn't classical, not to show what's wrong with how we think about QM that needs creative retrofitting.

 

[Jilang]

Ken, nice post. If I was a betting type I might put my money on retrocausality being able to be demonstrated by experiment. With MWI it would seem no chance would exist. If there is an accommodating universe there should always be a way to move forward. We live in interesting times!

 

[bohm2]

Ken, nice post. If I was a betting type I might put my money on retrocausality being able to be demonstrated by experiment. With MWI it would seem no chance would exist.

GRW and even Bohmian might also be experimentally distinguishable from OQM.

 

[DevilsAvocado]

Griffiths says his interpretation is realistic and local, but if one wants to argue that it is not realistic, that seems plausible. According to an FAQ about consistent histories, "Colored slips of paper, one red and one green, are placed in two opaque envelopes, [...]

I smell a rat.

On PF we've seen numerous cranky attempts with "African Doctors", "French Epidemics" and god knows what, "explaining" EPR-Bell experiments preserving local realism. Everybody has failed, catastrophically, and of course the "Slip/Envelope" case is no different.

The "Slip/Envelope" case is mentioned in the FAQ, explaining how the 1935 EPR paradox gets handled in Consistent Histories. Of course, Bell's 1964 theorem is not mentioned once, since this would ruin the case completely.

In the Brief Introduction to Consistent Histories we are served the same moldy dish:

Consistent histories can be used to analyze various quantum paradoxes, such as the interference produced by a particle passing through a double slit, or the correlated pair of particles considered by Einstein, Podolksy, and Rosen. This allows the paradox to be understood in quantum terms, without any need to invoke peculiar long-range influences or other ghostly effects.

But what does Griffiths say about this in his book Consistent Quantum Theory?

Well, the "Slip/Envelope" case gets introduced in chapter 23.4 about Stern-Gerlach and Measurements of One Spin, where we have the mutual exclusion outcome of Z⁺ or Z^-, in which the "Slip/Envelope" case naturally fit like a glove. If we combine this with the eccentric methodology of stopping time at 1935, and 'forgetting' everything about Bell's theorem, we're almost there...

Wow, Griffiths is a crackpot??

No, certainly not, because in chapter 24.4 about Bell Inequalities, the "Slip/Envelope" case is not mentioned once (only in 24.1 about the 1935 EPR Paradox). Instead we get this restriction:

Thus the point at which the derivation of (24.10) begins to deviate from quantum principles is in the assumption that a function $\alpha$(w_a ,λ) exists for different directions w_a . As long as only a single choice for w_a is under consideration there is no problem, for then the “hidden” variable λ can simply be the value of S_aw at some earlier time. But when two (excluding the trivial case of w_a and -w_a) or even more possibilities are allowed, the assumption that $\alpha$(w_a ,λ) exists is in conflict with basic quantum principles.

The phrase "the “hidden” variable λ can simply be the value of S_aw at some earlier time" is just a circumscription of the "Slip/Envelope" case, where this prerequisite is an absolute necessity, thus the "Slip/Envelope" case only works as long as "only a single choice for w_a is under consideration".

Hence, Griffiths himself, in chapter 24.4 points out that the "Slip/Envelope" case in not compatible with Bell Inequalities, and he sums up the chapter with the following:

[my bolding]

In summary, the basic lesson to be learned from the Bell inequalities is that it is difficult to construct a plausible hidden variable theory which will mimic the sorts of correlations predicted by quantum theory and confirmed by experiment. Such a theory must either exhibit peculiar nonlocalities which violate relativity theory, or else incorporate influences which travel backwards in time, in contrast to everyday experience.

Everybody, including Einstein, understands that the "Slip/Envelope" case is a hidden variable theory. End of discussion.

What's the problem?

The problem is when guys like postdoctoral fellow Vlad Gheorghiu packages the whole thing into an unrecognizable quicksand of delusion, capable of making an erudite SA swallow it hook, line, and sinker – into the hole of scientific deception.

 

[bohm2]

I think that Bohmian mechanics has some sort of "real explanation" (do you know?), but not in detail how the non-locality is 'implemented', and anyhow, there are serious trouble with RoS as soon as you make "real stuff" being there and influencing other distant "real stuff".

This is one possibility and is sort of based on Couder's experimental stuff:

We refer neither to potentials nor to a "quantum force" or some other dynamics, but show that a "systemic nonlocality" may be obtained as a phenomenon that emerges from an assumed sub-quantum kinematics, which is manipulated only by changing its constraints as determined by the changes of the apparatus.

"Systemic nonlocality" from changing constraints on sub-quantum kinematics
http://iopscience.iop.org/1742-6596/442/1/012012/pdf/1742-6596_442_1_012012.pdf

A Classical Framework for Nonlocality and Entanglement
http://arxiv.org/pdf/1210.4406.pdf

 

[atyy]

I smell a rat.

On PF we've seen numerous cranky attempts with "African Doctors", "French Epidemics" and god knows what, "explaining" EPR-Bell experiments preserving local realism. Everybody has failed, catastrophically, and of course the "Slip/Envelope" case is no different.

The "Slip/Envelope" case is mentioned in the FAQ, explaining how the 1935 EPR paradox gets handled in Consistent Histories. Of course, Bell's 1964 theorem is not mentioned once, since this would ruin the case completely.

In the Brief Introduction to Consistent Histories we are served the same moldy dish:
But what does Griffiths say about this in his book Consistent Quantum Theory?

Well, the "Slip/Envelope" case gets introduced in chapter 23.4 about Stern-Gerlach and Measurements of One Spin, where we have the mutual exclusion outcome of Z⁺ or Z^-, in which the "Slip/Envelope" case naturally fit like a glove. If we combine this with the eccentric methodology of stopping time at 1935, and 'forgetting' everything about Bell's theorem, we're almost there...

Wow, Griffiths is a crackpot??

No, certainly not, because in chapter 24.4 about Bell Inequalities, the "Slip/Envelope" case is not mentioned once (only in 24.1 about the 1935 EPR Paradox). Instead we get this restriction:
The phrase "the “hidden” variable λ can simply be the value of S_aw at some earlier time" is just a circumscription of the "Slip/Envelope" case, where this prerequisite is an absolute necessity, thus the "Slip/Envelope" case only works as long as "only a single choice for w_a is under consideration".

Hence, Griffiths himself, in chapter 24.4 points out that the "Slip/Envelope" case in not compatible with Bell Inequalities, and he sums up the chapter with the following:

[my bolding]Everybody, including Einstein, understands that the "Slip/Envelope" case is a hidden variable theory. End of discussion.

What's the problem?

The problem is when guys like postdoctoral fellow Vlad Gheorghiu packages the whole thing into an unrecognizable quicksand of delusion, capable of making an erudite SA swallow it hook, line, and sinker – into the hole of scientific deception.

Just to make clear, I do agree that one cannot have a local realistic interpretation of quantum mechanics unless one violates the assumptions of the Bell theorem. For example, one can violate the assumptions of the Bell theorem by assuming that a measurement does not yield a unique outcome, but rather all outcomes occur - even then - it is not clear if the interpretation is local - but I'll certainly grant that its nonlocality is not assured by violation of the Bell inequalities. I also agree that it is not enough to say that the measurement results don't exist before measurement in order to violate the assumptions. One has to do something drastic like saying that there are not two observers, but only one.

However, it does seem that Griffiths claims consistent histories to be "realistic" and local. You can find the claim of locality in http://quantum.phys.cmu.edu/CQT/chaps/cqt24.pdf and the claim of "realism" in http://quantum.phys.cmu.edu/CQT/chaps/cqt27.pdf. In fact, both claims are made as points 2 and 3 on p318 of the second link. I don't know enough about consistent histories to comment on the claim, but my impression is that his definition of reality isn't "common sense realism", so if his claim is correct, my bet would be that his definition of reality is actually a form of nonrealism.

 

[bhobba]

my bet would be that his definition of reality is actually a form of nonrealism.

I have studied Consistent Histories and most definitely do NOT agree its a realistic interpretation in any usual sense.

In fact why it is described that way has me beat - but people do:
http://www.siue.edu/~evailat/pdf/qm12.pdf

This semantic nit picking of calling something 'weak property realism' leaves me cold. That's probably because philosophy in general leaves me cold.

Basically the idea seems to be only one history actually exists, and is in that sense real, but we do not know it and can only predict probabilities. Like I said before its MW without the MW's, and wanting that IMHO leads to an unnecessary complication. Still its a nice interpretation - just not my favorite.

Thanks
Bill

 

[bhobba]

I smell a rat.

Well its the other thing I often say about Consistent Histories - its defining your way out of problems.

Want a realistic theory - no problem - simply define realism the way you like.

Its a 'rat' all right - but not one I particularly worry about.

Thanks
Bill

 

[DevilsAvocado]

This is one possibility and is sort of based on Couder's experimental stuff:

Thanks bohm, that's interesting.