What Is the Marseille Interpretation of Quantum Mechanics?

[marcus]

the term "Marseille" is used humorously in part, it could also be called the "Haifa Interpretation" because Asher Peres contributed and he was at the Technion

and also David Mermin at Cornell (Ithaca, NY) has written similar stories about how to interpret QM and has called his stories the "Ithaca Interpretation"

Or it could be called "Princeton" because of B. van Fraasen's contribution, or one could cite Michel Bitbol who is somewhere in France and call it the "French" interpretation. That actually sounds OK and would include the Marseille people.

==================

Anyway we need a thread that focuses just on the Marseille interpretation of QM and does not get sidetracked. I need some peace and absence of distraction in order to unfold whatever it is this has to tell us.

==================

The first thing to say is that INTERPRETATION has to do with the STORIES rather than the mathematical notation or "formalism". Two people can use the same or very similar formulations----the same subscripts and the same equations, or almost the same---and interpret them in radically different ways.

when we are discussing interpretation, similarity at the level of symbolic notation is SUPERFICIAL RESEMBLANCE and what is deep is the stories.
===================

An illustration of this is that the math of Special Rel was developed by a highly respectable Dutchman named Hendrik Lorentz, who unfortunately told the wrong STORIES about it.
One could say Einstein's contribution to Special Rel was an INTERPRETATION of formulas that were already there. He found the right philosophical spin to put on them.

In particular he found that you had to give up the intuitive notion of SIMULTANEITY.

=============
Simultaneity was deeply rooted in the human mind, for a reason I will discuss, and in order to tell the right stories about the Lorentz transformations you had to give it up-----eradicate, meaning tear out by the roots. Not easy to do. A purely philosophical step, and nevertheless necessary to the development of physics at that junction.

=============
So we have to be alert to whatever the Marseille Interpretation is asking us to give up. It will be something that seems obvious to us but for which there is no actual concrete evidence.

 

[f-h]

What do you want to talk about in this thread?

 

[marcus]

What do you want to talk about in this thread?

Smerlak Rovelli, and also
Asher Peres (who, while he may not be saying exactly the same thing, says it very nicely)

Also the Laudisa Rovelli article in the Stanford Encyclopedia sheds some light

First and foremost I want to EXPLICATE----to hear receptively what these voices are saying

In case anyone is new to this, I will get some links

==============

Oh BTW Rovelli Smerlak says repeatedly what it is we have to give up. It is Einstein's super-realism. This is a deeprooted mistake for which there is no actual evidence----the unstated assumption that there is one preferred account of the facts.

The several QM paradoxes, by now banal, show that the super-realism assumption is an UNREALISTIC EXPECTATION to have about nature.

Nothing in our experience supports the notion that there is one official account of the facts. But some people persist in believing that there should be. I want to try to understand their psychology.

Why the notion that there should be one preferred wavefunction, one bookkeeping device, should be psychologically appealing to some. I think it is the superbeing perspective, which it may be emotionally satisfying to adopt and this may explain the deep-rooted appeal of what is AFAICS unsupported by the experience of empirical science

 

[marcus]

What do you want to talk about in this thread?

Here is Rovelli Smerlak
http://arxiv.org/abs/quant-ph/0604064
Relational EPR
Matteo Smerlak, Carlo Rovelli
7 pages
"We argue that EPR-type correlations do not entail any form of "non-locality", when viewed in the context of a relational interpretation of quantum mechanics. The abandonment of strict Einstein realism advocated by this interpretation permits to reconcile quantum mechanics, completeness, (operationally defined) separability, and locality."

Here is Peres
http://arxiv.org/abs/quant-ph/0310010
Einstein, Podolsky, Rosen, and Shannon
Asher Peres
2 pages
"The EPR paradox (1935) is reexamined in the light of Shannon's information theory (1948). The EPR argument did not take into account that the observers' information was localized, like any other physical object."

Here is the Stanford Encyclopedia 2005 article of Rovelli Laudisa
http://plato.stanford.edu/entries/qm-relational/
Relational Quantum Mechanics

Miscellaneous additional links or material cited in the above
Here is James Hartle on throwing out excess baggage (as the way to make progress)
http://arxiv.org/abs/gr-qc/0508001

Here is Rovelli 1996 RQM
http://arxiv.org/abs/quant-ph/9609002

 

[f-h]

I see, I'm quite familiar with RQM by now and will just give my own personal perspective here. There are two analogies that drive it, galillean and special relativity, which both say that different observer descriptions on something particular might not agree, that the valid statements are "relational".

What is critically missing from RQM is synchronisation. Due to this RQM is IMO best understood as an account of why the formalism of QM, which escapes any consistent onthological interpretation, even as relational data, is still consistent epistemologicaly for all observers.

 

[marcus]

...formalism of QM which escapes any consistent onthological interpretation... is still consistent epistemologicaly...

Asher Peres put this nicely in "EPR Shannon" he said

Quantum states are not physical objects: they exist only in our imagination. [/color][/size]

====================

But as to what is really missing, or rather what has been EJECTED, Rovelli Smerlak tell us over and over what excess baggage they think we have to throw out: they refer to it as "EINSTEIN'S STRICT REALISM" and words to that effect---some single-minded insistence on one preferred account of the facts.

maybe it comes to the same thing as what you said

 

[Hurkyl]

I'll have to read some more to decide what I think about these...

But, I am quite pleased to see people addressing the EPR paradox from the perspective that EPR is asking some sort of "non-local question".

 

[f-h]

Well, the question is, are we done if we keep saying let's throw away the excess baggage and take quantum states as epistemological.

Imagine we have three systems A, B and C. A and B meassure C. Now if A meassures what B has meassured with respect to C it will find it consistent with it's own meassurement of C. So will B with respect to A. What they will find if the epistemological states are all there is is not related at all.
So B might observe that itself and A have found C to be in state c_1 while A might find that B and itself have found C in state c_2. There is no paradox here of course.
But in this sense this complete "epistemologisation" of physics implies an extreme soliplism.

If we "synchronise" the different experiences we get something equivalent to collapse again.

In this sense it shows first and foremost that physical predictions are insensitive to the interpretation of Quantum Mechanics, but I wouldn't really call RQM an interpretaion of QM.

 

[Hurkyl]

Now if A meassures what B has meassured with respect to C it will find it consistent with it's own meassurement of C. So will B with respect to A. What they will find if the epistemological states are all there is is not related at all.
So B might observe that itself and A have found C to be in state c_1 while A might find that B and itself have found C in state c_2. There is no paradox here of course.

I think you're still thinking externally.


I think you're imagining "Okay, A did his experiment, and saw this". And "B did her experiment, and saw that". And you, from your external viewpoint, are comparing things.


But if you do things internally, it all works out. You perform an experiment in which A and B meet to compare notes.


It's (IMHO) easy to see that the formalism predicts the only possible outcome is that A and B agree on what they saw.

 

[f-h]

Hurkyl, yes, this is correct.

But outlawing all external thinking is precisely the extreme soliplism I was talking about.

 

[Hurkyl]

I meant my post to counter your assertion that we have to add in something called "synchronization", and that it's equivalent to collapse.

I assert it's already in the formalism, and doesn't involve collapse.

 

[f-h]

Hmmm...

What I meant is that if there is a third system D asking A and B what they saw it will find that A and B saw the same thing, but the problem remains that what A saw might be different from what B saw, because what they saw has reality only with respect to themself and not with respect to D.

I don't see a mechanism in the formalism that does what you say. Could you elaborate?

 

[Careful]

I meant my post to counter your assertion that we have to add in something called "synchronization", and that it's equivalent to collapse.

I assert it's already in the formalism, and doesn't involve collapse.

Logically correct (that is basically also Patrick's position), f-h is trying to put in realism again by his synchronisation procedure (which of course I do appreciate ). Ah elaborating upon f-h's comment, suppose A and B observe an electron flying trough an SG apparatus with polarization in the z direction and psi_electron = a |up> + b |down>, then the state D observes is
a |up > | Asees up> | Bsees up> + b |down> | Asees down> | Bsees down> at least when you can assume that A,B are reliable persons. Then D sees A and B agree as is determined by the quantum *dynamics*. Now indeed, it could be that if A,B and D are doing sequentially this game (with sufficient time for causal communication) then it could be that at the N'th step A records up, B down, D sees A and B record up, A sees B record up. That is the price to pay if you give up realism and still is the reason why I reject this line of thought.

Cheers

Careful

 

[Hurkyl]

I don't see a mechanism in the formalism that does what you say. Could you elaborate?

The analysis from the D system would look like:

(a|1> + b|2>) |A's initial state> |B's initial state> |D's initial state>

If we evolve the state through the measurement of A, we get:

(a|1>|A=1> + b|2>|A=2>) |B's initial state> |D's initial state>

And if we evolve the state through the measurement of B, we get:
(a|1>|A=1>|B=1> + b|2>|A=2>|B=2>) |D's initial state>

(and we get the same result if we do it the other way around)

And then evolving through D's measurement yields:

(a|1>|A=1>|B=1> + b|2>|A=2>|B=2>) |D sees that A = B>

 

[hossi]

Quantum states are not physical objects: they exist only in our imagination.

well, I don't know about you but in my imagination it's the other way round - there exist no quantum states Quantum states are not actually observable, but is that so mysterious? Potentials (thermo, electromagn.) e.g. are (in most cases) also not observable, but still it's handy to work with them.

B.

 

[f-h]

Yes I agreed with that. As I said, there is no paradox and D sees A and B agree. But in RQM what you just wrote down was the analysis of what D sees A see which a priori has nothing to do with what A sees, that's the "problem".

I just saw that hossi pointed out the very same thing in the other thread: https://www.physicsforums.com/showpost.php?p=965093&postcount=70

Nice way to put it. Everybody always hears what they must hear to avoid paradoxes in RQM. You have to consider what "A sees B see" as independent from what "B sees" in the set up of your theory.

 

[Careful]

Yes I agreed with that. As I said, there is no paradox and D sees A and B agree. But in RQM what you just wrote down was the analysis of what D sees A see which a priori has nothing to do with what A sees, that's the "problem".

I just saw that hossi pointed out the very same thing in the other thread: https://www.physicsforums.com/showpost.php?p=965093&postcount=70

Nice way to put it. Everybody always hears what they must hear to avoid paradoxes in RQM. You have to consider what "A sees B see" as independent from what "B sees" in the set up of your theory.

I basically answered that objection, each of your different relative states give different local ``collapse´´ possibilities. To synchronise these possibilities is tantamount to treating all macroscopic systems as classical which brings you back to the realist nonlocal collapses - that is exactly why I thought one year ago ``nonsense´´ when reading this (I am a realist, cannot help it).

Cheers,

Careful

 

[f-h]

Careful, I think we are in complete agreement. According to RQM you do not synchronise of course. Not synchronising does NOT lead to paradoxes, only indigestion and vertigo.

 

[Hurkyl]

You have to consider what "A sees B see" as independent from what "B sees" in the set up of your theory.

No you don't. You make an experiment that detects what A sees B see, and what B sees, and test for a correlation.

In other words, the question as stated is aphysical -- but you can make it physical in terms of the actual experiment where C compares what C sees A see B see, and what C sees B see.

(And the correlation will be 100%)

 

[Hurkyl]

The Rovelli paper makes it sound like some sort of gauge freedom -- I'm not sure what I think of that, but meh.


It's appealing to the theorem that it doesn't matter when wavefunction collapse is applied -- you'll get the same answers out either way. So it's reasonable to posit that the choice is aphysical (and that there is some aphysical freedom in the choice of wavefunction as well).

In other words, collapsing when A saw something, and then evolving to when D watches A and B compare notes is physically the same as just evolving all the way up until D's observation and collapsing there.

 

[f-h]

Hurkyl, I agree with what you say. It doesn't touch my point. I wasn't pointing out a physical question that would require collapse (in this sense) to be consistent. I was pointing out what to me appears to be a characteristic feature of an RQM universe: Universal soliplism.

There is nothing in RQM as presented so far that prevents A from seeing green, B from seeing red, and D from seeing A and B see yellow. And E from seeing A, B see magenta and seeing D seeing A and B see magenta and so on and so on ad inf.

Each of the physical questions asked is consistently answered, every observation is consistent with every observation of an observation.

But the price is the disconectedness of all the experiences/information each system acquires about the other systems.

 

[Hurkyl]

There is nothing in RQM as presented so far that prevents A from seeing green, B from seeing red, and D from seeing A and B see yellow. And E from seeing A, B see magenta and seeing D seeing A and B see magenta and so on and so on ad inf.

Yes there is. It is physically impossible for \omega to see A see green, and to see B see red, and to see D see A and B see yellow, and so on ad inf.


Of course, I know you know that. But you're missing the point -- you don't give up "connectedness of experience/information". What you do give up is the ability to perform the sort of classically inspired external manipulations that you need to use to arrive at the conclusion of a "disconnected experience/information."


(At least, if my initial perception of these papers is accurate)

 

[f-h]

"Yes there is. It is physically impossible for \omega to see"

But that's not what I said.

I'll have to ponder this more, I know that what you say is what they are claiming to do, but to me it looks like the way they are doing it leads to the disconnectedness I described.

 

[hossi]

Ah, I think I finally have a vague idea of the interpretation!

Is it that at every point in space and time there is in principle an infinite amount of equally relative observers, each having his own history of observation, each ones history in agreement with quantum mechanics. And though each ones history might be different, no one can tell, cause there is no one who can compare them. And whenever you ask the question whether two histories agree, you have to invoce quantum mechanics, and though the observers have different histories, they hear only an answer that agrees with their opinion.

Is that about correct? Could you give me some feedback?

Does that mean that these histories are kind of classified in the sense that concerning future predictions they have to be identical. I.e. the mouse might be tasty or yukky, but it definately is a mouse and the cat will try to eat it?



B.

 

[marcus]

Ah, I think I finally have a vague idea of the interpretation!

Is it that at every point in space and time there is in principle an infinite amount of equally relative observers,...
Is that about correct? Could you give me some feedback?
...

Hi Biene. Do you want just f-h to reply? or the person you were talking to? Or should somebody else like me also give feedback?

BTW the picture you present seems a bit bizarre to me. In normal QM one does not usually picture so many observers. One may have Alice and Bob and maybe Professor Podolsky----or Wigner's friend inside the box, or maybe a Cat----in any case one has just as many observers as one needs to do the particular experiment or tell the particular story.

In RQM one would have just the same number of observers as in usual QM, I think.

Or perhaps one fewer, because one does not have an Absolute observer who has the Absolute correct state. but let us not worry about a difference of one or two.

the main thing is, I think, there is the same number of observers whether it is usual kind QM or relational QM.

maybe f-h says different?

 

[selfAdjoint]

An infinite sea of unobserved observers?

 

[marcus]

An infinite sea of unobserved observers?

No selfAdjoint. That is not RQM.

I don't know where Hossi got the idea.

I made a recap post to give the essential definition of what RQM is about
https://www.physicsforums.com/showthread.php?p=965312#post965312
it is #72 on the Elephant thread

 

[hossi]

No selfAdjoint. That is not RQM.

I don't know where Hossi got the idea.

Well, I am trying my best I did not mean to introduce more observers as in usal QM, but can't I have a continum of observers not in causal contact observing the same event? Among them Bob and Alice? And at some point they compare their stuff. It's not an infinite sea of unobserved observers, since they do observe each other (or what do you mean with unobserved?). I am having kind of problems finding that plausible, but if anyone could answer my question how Bob and Alice agree on their different measurement histories (before they were in causal contact) otherwise, please enlighten me.



B.

 

[marcus]

... can't I have a continum of observers not in causal contact observing the same event? Among them Bob and Alice? And at some point they compare their stuff...

it sounds like a creative idea. If Niels Bohr will let you have that in usual QM, then I suppose you can have the same setup in Rovelli setup.

RQM tends to equate observers with systems-----there is not much difference.

I have a hard time imagining more than a finite number of systems, and so I can't easily picture more than a finite number of observers (which are quantum systems interacting with other systems)

Biene, maybe the thing is to read Rovelli's paper "RQM" where he talks about what is an observer.

I guess if you can picture a continuum of distinct SYSTEMS not in causal contact (but I cannot) then you can say they are all observers and then you have a continuum of distinct observers.

the main thing is not to make unnecessary difficulty. the RQM is just the usual QM except you dump the notion of an absolute state:
you allow observers to disagree, which we know they do anyway, and you STOP EXPECTING that somehow they are going to agree.

so it is business as usual, except possibly minus a superbeing---comparatively easy and pleasant. I don't understand the face you are making with the tongue

=====================
Biene, I hope all is well with you.

I had a thought after I had gone to bed, and woke up. The thought was that your continuum of observers (which is so alien to RQM as I understand it) is TELLING ME SOMETHING ABOUT MANY WORLDS INTERP.
It is in manyworldsinterp. that one might have a trajectory of observers. actually it would be branching like crazy, going forwards
but if you follow it back it might look like a smooth or almost smooth curve in some Baroque monster space.

So your having this mental picture of a continuum path of observers was a way of helping me understand how manyworlds (which has never interested me so I haven't thought about much) is concretely different from relational. I don't know if you meant it that way, but thanks.

 

[koantum]

In the Quantum Physics Forum there is a thread called https://www.physicsforums.com/showthread.php?p=965424&nojs=1#goto_threadsearch".

Since hurkyl pointed me here, I went through this thread. I find myself in sympathy with what Marcus says in it. Some additional comments:

Well, the question is, are we done if we keep saying let's throw away the excess baggage and take quantum states as epistemological.

If we have thrown away the excess baggage, we are finally in a position to arrive at http://thisquantumworld.com" .

Quantum states are not actually observable, but is that so mysterious? Potentials (thermo, electromagn.) e.g. are (in most cases) also not observable, but still it's handy to work with them.

Only in most cases? The point is: they are nothing but mathematical tools handy to work with.

P.S. I wonder why this thread, which discusses an interpretation of the standard quantum formalism, comes under "Beyond the Standard Model".