How can we test the holographic principle and nonlocality in quantum mechanics?

[vanesch]

Non-Local viable theories discussed here competing to be THE solution include:
1) GR
2) MWI-QM (Strings included)
3) OQM
4) QFT
5) BM

All are non-local, and basically in competition with each other, with General Relativity having the widest differences with and all the others.

Uh ?

GR and MWI-QM (including QFT when seen that way) are strictly local (and deterministic), in the sense that the deterministic dynamical changes to the state are entirely determined by the local part of the state itself. This is somewhat harder to see in MWI-QM than in GR, where it is obvious, but one can reformulate MWI-QM in ways where this is explicit.

OQM doesn't (by definition) prescribe any underlying mechanism, so the above definition of locality doesn't apply, since OQM is just an algorithm to predict probabilities of things happening in a classically described world.
So we can only look at OQM's probabilistic PREDICTIONS, and then we have to say that OQM's probability predictions do not fit with Bell locality, but respect signal locality.

BM, in the sense that it proposes more than OQM, namely an underlying (and moreover deterministic) mechanism, is non-local in its inner workings.
That's not a "shortcoming" wrt OQM, which bluntly doesn't propose any underlying mechanism.

Essentially not being able to do it with SR is what put Einstein on the path of GR.

I hope you realize that SR and GR are the same, if you switch off gravity ?

GR is a non-local option (background independent) to complete that picture, so naturally in principle it competes with other non-local options to be the complete solution. But does anyone feel their preferred non-local theory needs to reject the basic principals in SR?

GR is background independent, not because any non-locality, but because of general covariance: the fact that any description will do, and that a priori, nothing in GR prefers any description over any other.

You can reformulate SR also in a completely background independent way, without making use of any reference to an "inertial frame".
It's pretty easy: call 4 variables, "coordinates of events", and associate a metric tensor to each of the set of 4 values of coordinates. This metric describes entirely locally, the relative distances of events. It is empirically measureable. If the Riemann tensor that derives from it is identically 0, you have a correct SR description.

 

[Careful]

I don't really follow the phrasing of your paragraph, but I'll try to answer anyways.

I would have just said "QFT" is local, because it is local in the sense we have been discussing in this thread. But I can't bring myself to do it, because it is non-local in the sense that a quantum state cannot be fully understood merely by looking at its restrictions to different spatial regions.

Right, this was a consequence of the Reeh-Schlieder theorem (1961) - which Bell must have forgotten (or thought not to be of importance) I guess.

Careful

 

[ttn]

Right, this was a consequence of the Reeh-Schlieder theorem (1961) - which Bell must have forgotten (or thought not to be of importance) I guess.

Are you guys saying that QFT is locally causal ("Bell Local")? Don't you know that collapse of the wave function is needed in QFT, just like in non-rel OQM? The textbooks never talk about this; they immediately jump into calculating scattering amplitudes and such. But the collapse rule is needed nevertheless for the same reasons you need it in OQM -- to make the cat come out definitely alive or definitely dead for example.

 

[vanesch]

Are you guys saying that QFT is locally causal ("Bell Local")? Don't you know that collapse of the wave function is needed in QFT, just like in non-rel OQM? The textbooks never talk about this; they immediately jump into calculating scattering amplitudes and such. But the collapse rule is needed nevertheless for the same reasons you need it in OQM -- to make the cat come out definitely alive or definitely dead for example.

I concur with this. "QFT" as such is just a specification of a unitary dynamics, which is local, but it is not the dynamical, unitary part which gives rise to eventual observed non-localities ; it is the projection postulate.
This is BTW also the reason why the MWI variant of QFT is local in its dynamics (while in fact, the typical NR QM dynamics is non - local, such as Coulomb interactions and the like).

However, in most QFT calculations, the non-locality of any projection is not seen, because one usually considers transition probabilities between product states (sets of particles in "in" and "out" momentum states, hence no explicit entanglement is ever considered in most scattering situations).
That's why many high energy people are ignorant of it - to my great surprise, in fact.

 

[Careful]

**Are you guys saying that QFT is locally causal ("Bell Local")? Don't you know that collapse of the wave function is needed in QFT, just like in non-rel OQM? The textbooks never talk about this; they immediately jump into calculating scattering amplitudes and such. But the collapse rule is needed nevertheless for the same reasons you need it in OQM -- to make the cat come out definitely alive or definitely dead for example. **

Two comments
(a) If you would know the content of the Reeh Schlieder theorem, then you wouldn't ask this question ! The theorem says that the Fock vacuum state is cyclic with respect to any local operator algebra, meaning that local measurements in some spacetime neighborhood influence spacelike separated regions (as well as the past). I told you already a few times that Bell locality *and* no conspiracy is violated in QFT, so why would you expect anything good to come out of IMPOSING BOTH a priori + assuming dichotomic outcomes - something a naive projection rule wouldn't say anything about either ??
(b) No, nothing says that collaps of the wave function is needed - also something which I told you already. One can take the point of view that QFT only says something about propagation of statistical properties of series of measurements (without telling what the latter are). Nothing in the world implies that a theory of single *particle measurements* (not necessarily particles !) would have anything to do *directly* with the QFT operator algebra, and a collapse rule.

Nothing says that the Bohm theory is correct, actually experimental violation of the CH74 inequality (please not the CHSH inequality - this one is too silly) could imply four things : (a) either the no-enhancement hypothesis is false (seems plausible) (b) Bell locality is wrong (c) no-conspiracy is wrong (d) counterfactual definiteness fails (phoeh what a word :yuck: ). The last would be bad for Bohmpkes like yourself of course :tongue:

And please, stop telling what we are presumed to know or not, since it seems you are missing or ignoring all the way *many* points which indicate that it could be the naive Bohmian one particle theory which fails. Science is about exploring the imaginable, not about imposing a belief which could very well be wrong.

Careful

 

[Careful]

**
That's why many high energy people are ignorant of it - to my great surprise, in fact. **

I don't know the kind of high energy people you are talking about, but by far most of them (which I know) are aware of the measurement problem in QFT. Most however do take the attitude that a sensible theory which is supposed to solve that problem is way beyond the standard formalism of QFT/QM and that for all practical purposes, one does not need to worry about it. I prefer the latter attitude btw over people who come up with ghost stories or vigorously promote a theory which does not solve the issue either.

Careful

 

[RandallB]

I concur with this (ttn post). "QFT" as such is just a specification of a unitary dynamics, which is local, but it is not the dynamical, unitary part which gives rise to eventual observed non-localities ; it is the projection postulate.

Exactly, a theory cannot claim to BE LOCAL just because a part is defined as local and the ability to account for “observed non-localities” is done in some other axiom, principle, postulate, rule, or whatever. It means the theory is Non-Local.

GR is background independent, not because any non-locality,…..

NO, non-locity does not create independence. GR is non-local because it is background independent.

I hope you realize that SR and GR are the same, if you switch off gravity ?

As the core definition of GR is gravity, I cannot even imagine gravity turned off with GR remaining. The only way to make SR and GR the same is to formulate GR in a background dependent manner. Then you would be able to rationally describe and explain it in Classical THREE-dimensional space w/ Newtonian time, where the observation of time varies relative to other, observers in motion, observations. HOWEVER, if you could do that you would have a Local Layman’s description that would be functional as a working Theory - a Local one. But, you cannot make that kind of LOCAL description because GR requires a non-local warping within at least 4 freedoms of movement (dimensions) that cannot be described in local terms without using some mystical non-local descriptions to define it. I.E. GR is Non-local.

You can reformulate SR also in a completely background independent way, ……… easy: call 4 variables …..

Well what you do expect, you take a local theory add the freedom of a fourth variable using an extra non-local dimension and what you have is a new theory, no longer local because it is not background dependent anymore.
It is much easier to keep Minkowski space-time world lines local by keeping it background dependent, it takes the addition of GR for that analogue to become non-local in background independance.

Maybe some think my definition of Local or Local Realist is too stringent, but I consider a simple general definition of Local.
I do not understand why it seems important to so many that their favorite theory can defined as “local” under the proper interpretation of that theories reality. I should think they would explicitly want their theory to not be locally causal as the only REAL experiments we have with Real results (Bell entanglement, double slit, etc.) all seem to be giving experimental evidence that reality is Non-Local. Do they mean to say their theory is experimentally invalid! Fact is most are based on explaining these experiments in what are essentially non-local terms in agreement with experimental conclusions that locally causal interpretations cannot work. How much more non-local can you get!

Or has some new Bell experimental results come along showing otherwise?

 

[Hurkyl]

GR requires a non-local warping within at least 4 freedoms of movement (dimensions) that cannot be described in local terms without using some mystical non-local descriptions to define it.

I can't make any sense of this at all.

 

[Careful]

I can't make any sense of this at all.

Me neither, actually :

***
As the core definition of GR is gravity, I cannot even imagine gravity turned off with GR remaining. The only way to make SR and GR the same is to formulate GR in a background dependent manner. Then you would be able to rationally describe and explain it in Classical THREE-dimensional space w/ Newtonian time, where the observation of time varies relative to other, observers in motion, observations. ***

First of all, I guess you mean ``I cannot imagine gravity turned off with SR remaining'' . Second, I assure you there are *intrinsically geometric* ways (that is independent of coordinate system or any background structure whatsoever) to recuperate SR with the Poincare symmetry in the weak field limit of GR. You might -for example - consider proving that the exponentials of the ortogonal 3 spaces to some timelike geodesic form approximately geodesically complete, flat hypersurfaces in the induced metric in the weak field limit (that is all scalars you can construct from the Riemann tensor -> zero). Such prescription is entirely (quasi) local once you have picked out this geodesic, so the rest of your message is entirely incomprehensible to me.

**Well what you do expect, you take a local theory add the freedom of a fourth variable using an extra non-local dimension and what you have is a new theory, no longer local because it is not background dependent anymore. **

Euh, the point is that this time function can be given a dynamical prescription. I do not need this time function whatsoever to expand my action in a free and interacting part: on the contrary I derive this time function from the full solution itself. The rest of your message makes no sense whatsoever, it is entirely possible to derive SR from a local action principle (that is known since around 1917 I think).

Careful

 

[RandallB]

I can't make any sense of this at all.

You think GR can be describe in some form of layman’s common sense?
GR demands warping of space and time in a manner we cannot describe or see in our 3D local reality. GR explanations are mystical in layman’s terms and require accepting an alternate 4D realty. Don’t most that understand GR understand that.

If someone has a Local description of GR that does not depend on accepting an unknown fourth dimension (actually four dimensions working together in a background independent way) I’d like to see it! As a not a locally real theory, I do not think it can be done.

Also, I not aware of GR solving the paradoxes of double slits or entanglement, but as a non-local as I consider it possible that someone may find a way to have GR do so as can other Non-local theories. I doubt any even currently consider looking.

 

[Careful]

**You think GR can be describe in some form of layman’s common sense?
GR demands warping of space and time in a manor we cannot describe or see in our 3D local reality. GR explanations are mystical in layman’s terms and require accepting an alternate 4D realty. Don’t most that understand GR understand that. **

Someone who knows GR is aware of the fact that a *physical* notion of space and time usually coincides with the flow of reasonable matter forms - as is certainly the case for our universe which is approximately homogeneous and isotropic on large scales. . Your 3+1 split hence originates from there. Of course you can say then that it becomes impossible to generate a twin paradox since your time function might be ill defined given the fact that the flow can develop focal points; however this can be easily remedied by considering reference times wrt to bigger objects such as local Big Ben clock time.

The double slit experiment is a one wave phenomenon, and it is well known since the 1920 ties how to construct a covariant theory unifying the one particle Klein Gordon wave with the gravitational field (no background whatsoever required). Entanglement is a different matter which has to be understood at a more basic level first.

Now, if you were talking about the graviton approximation and perturbative quantum gravity ; well this has little to do with GR by itself no ? The characterization of plane wave spacetimes *seems* to require an a priori choice of a Minkowski background metric (and frame) but this is actually not the case.

Careful

 

[Hurkyl]

You think GR can be describe in some form of layman’s common sense?

That's not what I'm talking about. When I read your posts, I see little more than a bunch of words strung together -- I cannot connect them to anything mathematical.

For example, "non-local warping". I cannot figure out what you could possibly mean by that. The "warping" of space would be described by something like the metric, or the curvature tensor, each of which are manifestly local things.

And even if you are simply trying to talk about how disturbances in the "warping" of space move around, it is known that those disturbances propagate locally. What happens at a point in space-time cannot affect anything outside of the lightcone based at that point!

 

[vanesch]

Two comments
(a) If you would know the content of the Reeh Schlieder theorem, then you wouldn't ask this question ! The theorem says that the Fock vacuum state is cyclic with respect to any local operator algebra, meaning that local measurements in some spacetime neighborhood influence spacelike separated regions (as well as the past).

Brrr, I'm far from an expert on this, but these theorems are based upon certain analycity conditions and boundedness of operator spectra. I'm not sure this is to be taken very seriously ; we know that the mathematical construction of QFT is very dubious.
I wonder if these theorems indicate anything else than that the mathematical formulation of QFT is dubious when taken too literally.

 

[vanesch]

I don't know the kind of high energy people you are talking about, but by far most of them (which I know) are aware of the measurement problem in QFT.

Then you're lucky. Some I've met told me that Bell's theorem is not applicable to QFT because in QFT, field operators commute at spacelike intervals, and hence no such correlations can appear in QFT
I've also often heard (even from professors) that in QFT, the collapse is not immediate, but "propagates outward from the measurement point at lightspeed".

 

[vanesch]

HOWEVER, if you could do that you would have a Local Layman’s description that would be functional as a working Theory - a Local one. But, you cannot make that kind of LOCAL description because GR requires a non-local warping within at least 4 freedoms of movement (dimensions) that cannot be described in local terms without using some mystical non-local descriptions to define it. I.E. GR is Non-local.

What's your definition of a local theory then ? The "principle of Local Layman" or something ? :tongue: No, seriously, what do you call "local" ? I think it is entirely different from usual notions of "local" (of which there are variants), and which include:

"nothing in the machinery of a theory can influence something at a spacetime event outside of a local neighbourhood of that spacetime point"

Variants are concerned with the exact definition of "machinery" and "something".

In a deterministic theory with a postulated ontology, it is quite clear: what will be (deterministically) true about the the ontological description at one spacetime point, can only be a function of what values the ontological description takes on in a small neighbourhood of that spacetime point.
If you mix in causality, this small neighbourhood is moreover limited to the past lightcone (and usually this is taken for granted, so locality is directly connected with causality).

In stochastic theories, we can discuss, but the extension of the above statement to "probabilities" (which are not really part of an ontological description of nature - that's the discussion point), we arrive at Bell locality.

On the other hand, if we replace "machinery" and "something" by "signals" and "information", then we arrive at information locality, which says that no signal can carry information from one event to another event if the first event is not in the past lightcone of the other.

But again, locality is related to "stuff only changes here, determined by the strict neighbourhood of "here" ".

It seems that you adhere to a totally different definition of locality.

 

[Careful]

Brrr, I'm far from an expert on this, but these theorems are based upon certain analycity conditions and boundedness of operator spectra. I'm not sure this is to be taken very seriously ; we know that the mathematical construction of QFT is very dubious.
I wonder if these theorems indicate anything else than that the mathematical formulation of QFT is dubious when taken too literally.

Well, boundedness of operator spectra is a pretty natural demand in nature no ? Concerning the analycity properties of the state, did that not follow by imposing a UV cutoff ? I do not remember the details of the theorem well either, but the latter seem rather harmless assumptions and common practice in virtually any QFT calculation.

Apart from this all, I guess it is fairly natural to interpret a measurement as having an influence in the past.

 

[Careful]

**Then you're lucky. Some I've met told me that Bell's theorem is not applicable to QFT because in QFT, field operators commute at spacelike intervals, and hence no such correlations can appear in QFT **

I see... :uhh: and these were professional researchers ?

**
I've also often heard (even from professors) that in QFT, the collapse is not immediate, but "propagates outward from the measurement point at lightspeed".**

Well, as I said, you could entertain the idea that it travels on the past lightcone (you would not have noticed it), influencing the present of the other particles ``now''. But then you are in trouble again with the reality of your fellows having crossed your past lightcone before you arrived there. Either way you see it, its all sick and points in the direction of abandonning the particle notion as a fundamental thing. Perhaps take a look at ``explicit calculations with a hidden variable spin model'' by A.O. Barut and notice the important role the detection process (*not* necessarily a fundamental detector efficiency !) can play in reaching the EPR correlations (note : this paper says much more than it might after a first reading).

Careful

 

[Farsight]

You think GR can be describe in some form of layman’s common sense? GR demands warping of space and time in a manor we cannot describe or see in our 3D local reality. GR explanations are mystical in layman’s terms and require accepting an alternate 4D realty. Don’t most that understand GR understand that.

If someone has a Local description of GR that does not depend on accepting an unknown fourth dimension (actually four dimensions working together in a background independent way) I’d like to see it! As a not a locally real theory, I do not think it can be done.

Also, I not aware of GR solving the paradoxes of double slits or entanglement, but as a non-local as I consider it possible that someone may find a way to have GR do so as can other Non-local theories. I doubt any even currently consider looking.

Huh? No they ain't mystical. And it's manner, not manor. Here's your layman's common sense General Relativity: You've got two legs. One runs slower than the other. So you go round in circles.

 

[Farsight]

There's no spooky magical non-local "action at a distance". It's just a local space/time slope.

 

[RandallB]

Vanesch – are you reading the same posts and threads I’m seeing on this?

What's your definition of a local theory then ?
No, seriously, what do you call "local" ?

- - - - locality is related to
stuff only changes here, determined by the strict neighbourhood of "here".

I see our definitions of locality as identical on this issue.
Plus the only class of proofs that experimentally address the issue are Bell – Aspect types; that so far have ALL indicated any local theory (supporting an EPR view) is unacceptable and only a non-local theory can be viable.

My problem is I’m seeing a proliferation of comments in opposition to these experimental conclusions like:
- spooky action at a distance is just a local space/time slope (GR I assume)
- there is no issue with non-locality, or realism

None supported by any rational extension of a real experiment that I can see.
Rather they seem to theoretically reinterpret what you describe as “the strict neighborhood of here”.
I do not consider a local view of “here” to include: extensions into extra dimension(s); statistical extensions, or some overlapping non-interacting 3D aether that supports a guide wave that can affect particle paths in our classical dimension.

As to GR I’ll redirect that entire issue to the Relativity Forum as:
“GR Background Independence: Indeterminate and Non-local?”

For MWI , OQM, QFT, BM and others; at various times I’ve seen claims that they are local and that such a claim is somehow important to make:
Do you have any idea what is going on? – is there something new – maybe contradicting Aspect and others that theories may soon need to claim LOCAL to be acceptable?
I honestly don’t understand how every one of the respectable non-local theories I know of at one point or another has had someone claim it to in fact be local!
Is that intended to somehow make them more respected than other theories?
Or are they showing a lack of confidence in their Non-Local credentials for some reason?

I thought this thread was about understanding Non-locality. But it seems to have turned to claiming that is not true. Are you agreeing with such claims and that one or all these theories are local as you have defined local and the Bell-Aspect conclusions are wrong?

 

[vanesch]

I thought this thread was about understanding Non-locality. But it seems to have turned to claiming that is not true. Are you agreeing with such claims and that one or all these theories are local as you have defined local and the Bell-Aspect conclusions are wrong?

The experimental situation is not 100% clear. The problem with the actual experimental situation is that it is *highly suggestive* that Bell inequalities are "ideally" violated, but it always rests on some extra assumptions, such as the fair sampling hypothesis. It is not yet 100% conclusive.
We can only say that up to now, these results are compatible with quantum predictions, and this makes us *assume* that the ideal quantum predictions are also somehow true.

But what matters here is not so much the experimental situation, but rather the quantum-mechanical *ideal* predictions (which have not yet been approached close enough by experimental situations not to need "deconvolutions" of experimental effects in order to violate directly any Bell inequality). Now, *assuming* that these ideal predictions are somehow true, does that mean that there is only the possibility of a non-local theory as Bell wants us to believe ?

The answer is no, but something has to give. Bell needs EXTRA hypotheses in order for his claim to hold. So it is sufficient that some of these extra claims do not hold, and the claim doesn't hold. In MWI, what gives, is the following: the denial of the existence of a unique outcome of a measurement. Bell NEEDS the existence of a unique measurement at A and B, to be able to talk about the unique, spacelike connected, probabilities of the outcome A and the outcome B to happen. Well, in MWI, this is not true. Outcome B happened, and outcome Not(B) ALSO happened, and it is only later, when an observer is going to compare the outcomes of A and B, that he has to pick or the B branch, or the not(B) branch. But at that point, this observer has already the choices of A and B available, so Bell's theorem is not valid here.

So there's no violation of Bell's theorem in MWI, because "the probability of event B" has no meaning, independent of an observer, given that both B and not(B) happen.

 

[RandallB]

So there's no violation of Bell's theorem in MWI, because "the probability of event B" has no meaning, independent of an observer, given that both B and not(B) happen.

Great, once we get THERE we have both outcomes (B) and (not B) and any observer of one of those two will be correctly correlated with observations of event (A) vs. (not A) at some other THERE..

How does that tie to your own definition of :
“stuff only changes here, determined by the strict neighborhood of "here".”

THERE for B and the other THERE for A are not in the neighborhood of HERE! That would be where they started.

SO I give up,
I get the impression each of the 5 theories wants to claim local, Just to make things confusing. So they can argue the other four as wrong for debating purposes, then threads like this can stretch out to hundreds of posts.

 

[Farsight]

Perhaps we need a refresher, with a link to an experiment. Vanesch?