Scholarpedia article on Bell's Theorem

[ttn]

Perhaps you didn't read my solipsistic paper carefully enough. See again Sec. 2.2, especially the first and the third paragraph.

I know you think there's a point. I was saying I don't think there's any point.

 

[ttn]

You will probably say that it doesn't matter WHO says something, but only WHAT he/she says. That's indeed true in theory, but not in practice. In practice, humans (which includes scientists) are not perfectly rational machines. They often use intuition and heuristics rather than logic.

Who are you talking about here?

For example, if I see a new title on arXiv "Proof that quantum mechanics is local", will I bother to read it? If it is written by someone who I know as a crackpot, be sure that I will not. If it is written by A. Zeilinger, I will give the paper chance, but perhaps not too much. But if it is written by you or S. Goldstein for example, be sure that I will carefully read every word of it. And if it will not make sense to me at first, I will read it again and again, because it would be hard to believe that you or Goldstein could say something like that without having a really good argument.

On the other hand, if I see a paper with a title "The solution of the mind-body problem", I would read it if it was written by you, but I would pay much more attention to it if it was written by David Chalmers. I guess I don't need to explain why.

I'm of course also more likely to read something (and/or study it seriously) if it's written by somebody I know is serious and good. But that wasn't the issue. The issue was whether you'd refrain from judging something as crazy/wrong/irrational/unscientific simply on the grounds that the author is a "respectable scientist". I hope that if I or Goldstein or Chalmers or *whoever* came out with something that was truly nuts, you'd let yourself come to that conclusion and not just go on forever doubting yourself. Study it, sure. Give it some preliminary benefit of the doubt, sure. Make sure you understand where it's coming from and aren't just reacting superficially based on your own context, sure. But at the end of the day, if it's nuts, it's nuts.

 

[harrylin]

[..] I'm of course also more likely to read something (and/or study it seriously) if it's written by somebody I know is serious and good.

I don't follow your "of course also"; so I guess he actually "needed to explain why"! See next.

But that wasn't the issue. The issue was whether you'd refrain from judging something as crazy/wrong/irrational/unscientific simply on the grounds that the author is a "respectable scientist". [..]

That is incompatible with his explanation. What matters: if you know someone to be logical/reasonable and also expert on that topic, then you will carefully consider his (or her) arguments. I fully agree with that approach.

 

[ttn]

That is incompatible with his explanation. What matters: if you know someone to be logical/reasonable and also expert on that topic, then you will carefully consider his (or her) arguments. I fully agree with that approach.

Me too. But there's such a thing as carefully considering something, even something written by a "famous scientist", and having to conclude at the end that it is nuts.

For further evidence/explanation, see anything by Bohr.

 

[harrylin]

Me too. But there's such a thing as carefully considering something, even something written by a "famous scientist", and having to conclude at the end that it is nuts.

For further evidence/explanation, see anything by Bohr.

I will have to do that!

BTW, I think that the a sensible definition of "scientist" is someone who practices the scientific method, and I started a discussion of that method in the general science forum. Regretfully ZapperZ deemed that it doesn't belong under "physics" and he moved it instead to "Social sciences" (under "Other Sciences").

 

[martinbn]

Me too. But there's such a thing as carefully considering something, even something written by a "famous scientist", and having to conclude at the end that it is nuts.

For further evidence/explanation, see anything by Bohr.

Bohr!

 

[Demystifier]

I know you think there's a point. I was saying I don't think there's any point.

That's OK, but it would be even more OK if you could give an ARGUMENT why do you think that my reason for introducing particles in Sec. 2.2 is not a valid reason to do that.

 

[Demystifier]

I'm of course also more likely to read something (and/or study it seriously) if it's written by somebody I know is serious and good. But that wasn't the issue. The issue was whether you'd refrain from judging something as crazy/wrong/irrational/unscientific simply on the grounds that the author is a "respectable scientist". I hope that if I or Goldstein or Chalmers or *whoever* came out with something that was truly nuts, you'd let yourself come to that conclusion and not just go on forever doubting yourself. Study it, sure. Give it some preliminary benefit of the doubt, sure. Make sure you understand where it's coming from and aren't just reacting superficially based on your own context, sure. But at the end of the day, if it's nuts, it's nuts.

I of course agree.

 

[ttn]

That's OK, but it would be even more OK if you could give an ARGUMENT why do you think that my reason for introducing particles in Sec. 2.2 is not a valid reason to do that.

Well, we kind of went through this some months ago in email, didn't we?

Basically, your argument in that part of the paper is that, if we assume that conscious awareness of some branch has to arise from something physical in that branch, then we need to add (at least some) particles (or something like that) to "mark" the branch, to distinguish it from the others. My counter-argument is basically just: I don't think that makes sense. The MWI people will insist, and they have a good point (!), that the wf itself is already perfectly "physical", so why should you need to add some extra stuff for consciousness to emerge from? Why can't consciousness just emerge directly from the "stuff" out of which the wf is made? So I find your motivation for adding particles there totally uncompelling. A compelling reason for adding particles, to me at least, is that we want to have something in the theory that corresponds to the ordinary 3D physical world (as opposed to merely the fantasy/delusion thereof). That is, we want the theory to include cats and tables and trees and pointer positions and planets. But if *that's* the reason to take Bohmian particles seriously, it's hardly a good idea to only have a few of them, just enough to make a brain for some poor solipsist's consciousness to emerge from.

But there is really no need to get into all of this here. It comes up only because you want to question whether Bell really proved nonlocality. And I already conceded/stipulated that, yes, Bell's proof only applies to theories which say that the actually-performed experiments have actual, physically-real outcomes (as opposed to those outcomes being merely delusions in some disembodied mind). I'm happy to concede that because I don't think it weakens my side of things one bit. The reason people (should) want to avoid nonlocality is because it conflicts with relativity's alleged prohibition on superluminal causation. That is, to be against non-locality is to be *for* the proposition that all the physical influences in the world propagate around at or slower than light, which in turn presupposes that there is a world out there with causal influences propagating around in it at some speed or other. But a solipsistic theory like yours is hardly "local". It's not non-local, but it's also not local. It says, outside of my mind (or maybe some, I think pointless, set of brain particles) there are no causal influences at all. I am deluded about the real existence of cats, tables, pointers, planets, etc., and indeed I am deluded to have thought that special relativity applied to stuff out there in the world, etc. According to your theory there is nothing out there in the world, no causal influences between the sun and the earth, or between the particles in an entangled pair, or anything like that -- the whole thing that the locality/nonlocality debate was supposed to be *about*, simply doesn't exist. So whatever else you want to say about it, such a theory is surely not a way of "saving locality".

Glad we agree about the sociological stuff.

 

[Demystifier]

Basically, your argument in that part of the paper is ... we need to add (at least some) particles (or something like that) to "mark" the branch, to distinguish it from the others.

That's correct! (Note that I erased the part of your text mentioning consciousness, because in that part of my paper I don't mention consciousness at all.)

A compelling reason for adding particles, to me at least, is that we want to have something in the theory that corresponds to the ordinary 3D physical world (as opposed to merely the fantasy/delusion thereof).

Ah, that's great! Now I finally clearly understand the source of our disagreement. Even though we both like Bohmian mechanics, we like it for totally different reasons. You like it because it provides ontology in spacetime, while I like it because it gives a simple mechanism for effective wave-function collapse.

Related to this, we both dislike MWI, but again for totally different reasons. You dislike it because it does not provide ontology in spacetime. I dislike it because, even though it does provide a mechanism for effective wave-function collapse, it does not provide a simple explanation of why this effective collapse obeys the Born rule.

Now when I understand clearly the true origin of our disagreement, it's much easier to live with it, at least for me. How about you?

Glad we agree about the sociological stuff.

Of course we do, sociology is much easier than interpretations of quantum mechanics.

 

[malreux]

(1) The MWI people will insist, and they have a good point (!), that the wf itself is already perfectly "physical", so why should you need to add some extra stuff for consciousness to emerge from?

(2)The reason people (should) want to avoid nonlocality is because it conflicts with relativity's alleged prohibition on superluminal causation. That is, to be against non-locality is to be *for* the proposition that all the physical influences in the world propagate around at or slower than light, which in turn presupposes that there is a world out there with causal influences propagating around in it at some speed or other.

Hi ttn, have been enjoying this thread immensely! Sorry for wading in at the tail - my recurrent 'problem of tails' - just a couple of points:

(1) I agree that in practice most modern MWI folks, especially of the oxonian decoherence stamp, will gesture towards some idea of emergence and possibly functionalism in the philosophy of mind. I think this is along the right lines, but it is very underdeveloped, although Wallace's latest book has a brilliant stab at it. However, note that many MWI bods, including Wallace, are not wavefunction monists. That is, they don't believe that the world is made out of wavefunction in the same way that e.g. Kim believes classical measuring apparatus are made out of quantum particles (as mereological sums?). Wallace, for example, is more interested in a coherent description of the quantum state, or rather just paying attention to the math in ones interpretation.

(2) I understand the context in which you made this point, but still feel that you have been a bit 'cut n' dried' about this here. As you know, Maudlin questions whether we are talking about causal influence when gesturing towards superluminal signalling within an EPR/Bell type scenario. So the alleged 'prohibition' might not be prohibiting the relevant factors. More importantly, does 'non-locality' really have to conflict with relativity? I would suggest that the answer is, at least, not obvious and not trivial. Additionally, I think we muddy the waters when getting hung up on 'speed' - two things spring to mind: Barbour's famous correction for relativists neglecting duration as a concept in his end of time escapade, and the fact such debates as the present are usually within the context of non-relativistic QM. Transported to the arena of QFT, where we're looking for a Lorentz-covariant unitary quantum theory, in which the primary dynamical variables are ST local operators like field strengths and in which particles are approximate and emergent, interpretations leaving the formalism intact pretty much carry through. Unlike modificatory strategies, although recent attempts have been made (e.g. modern GRW treatments). I bring up these measurement problem considerations to briefly illustrate the extent to which relativity is already present in the formalism(s) of QFT, coupling this point with the thought that 'pure' interpretations of QM (e.g. MWI) ought to carry over from non-relativistic to relativistic quantum theory.

Incidentally, how do accounts which strive for locality manage concepts that I think are very related, such as non-seperability, holism, etc.? (I think Healey analyses these related concepts well in 'Gauging what's Real'?)

 

[malreux]

I dislike [MWI] because, even though it does provide a mechanism for effective wave-function collapse, it does not provide a simple explanation of why this effective collapse obeys the Born rule.

Well, there aren't any simple explanations going out on the current MWI market's output. There are many attempts however, including formal derivations and so forth (the Deutsch approach). Or Greaves' caring measure. There have even been modifcatory strategies (again, Deutsch).

Do you think MWI proponents have any resources to recover the Born rule?

 

[Demystifier]

Well, there aren't any simple explanations going out on the current MWI market's output. There are many attempts however, including formal derivations and so forth (the Deutsch approach). Or Greaves' caring measure. There have even been modifcatory strategies (again, Deutsch).

Do you think MWI proponents have any resources to recover the Born rule?

To recover the Born rule, I think MWI proponents necessarily must introduce some additional assumptions or axioms in the theory, which destroys the main virtue of MWI: minimality of assumptions. In fact (ttn will kill me for saying it), I like to think of Bohmian interpretation as a variant of MWI which just picks one such assumption, which, unlike other attempts of that sort in MWI, seems very natural to me.

 

[ttn]

That's correct! (Note that I erased the part of your text mentioning consciousness, because in that part of my paper I don't mention consciousness at all.)

I thought it did -- you explicitly talk about "the point of view from" a particular branch. To me that kind of language is "code" for consciousness, but maybe you meant something else.

Ah, that's great! Now I finally clearly understand the source of our disagreement. Even though we both like Bohmian mechanics, we like it for totally different reasons. You like it because it provides ontology in spacetime, while I like it because it gives a simple mechanism for effective wave-function collapse.

Yes, that's interesting and clarifying. For the record, I wouldn't say I like it *just* because it gives ontology in spacetime. It's bound up with the measurement problem. Copenhagen also has ontology in spacetime, but there it is just *posited* and then we have the usual awkwardness of two separately-posited realms (classical and quantum, or micro and macro, or however you want to describe them) and the associated awkwardness of different sets of laws on the two sides (both of which are by the way set aside in favor of some new thing when the two sides "interact"), etc. So put it this way: the virtue of dBB is that it gives you the same "classical macro world" that Copenhagen has to just postulate, and it gives you that in the context of a fully precise and consistent microscopic theory whose equations don't need to be set aside when they become embarrassing. And, you know, it's empirically adequate, etc.

I don't really understand your reasons for liking it, that it provides a simple mechanism for effective wave-function collapse. That to me seems way too bound up with ideas from orthodox QM. My attitude is that if you think of *any* interpretation fundamentally as an attempt to solve some technical problem with ordinary QM, you're sort of missing what the real issues are. We are trying to figure out what's the best theory here, and it should be clear to everybody that ordinary QM ain't it. So the best approach is to *forget* ordinary QM and take the fundamental question to be: which theory does the best job, by the normal standards of judging scientific theories, of accounting for all the facts (by which I mean to include both pre-scientific facts like that there are trees and stuff outside my window, and then also the detailed quantitative results of all the various sophisticated experiments like 2-slit, atomic spectra, Bell, etc. etc. etc.).

Related to this, we both dislike MWI, but again for totally different reasons. You dislike it because it does not provide ontology in spacetime. I dislike it because, even though it does provide a mechanism for effective wave-function collapse, it does not provide a simple explanation of why this effective collapse obeys the Born rule.

Yikes, here we'd have to get deep into a discussion of what you (and MWI) even mean by "effective". But again my attitude is the above. Forget about explaining effective wave-function collapse. The question is, which theory is best at explaining what needs to be explained?

 

[ttn]

(1) I agree that in practice most modern MWI folks, especially of the oxonian decoherence stamp, will gesture towards some idea of emergence and possibly functionalism in the philosophy of mind. I think this is along the right lines, but it is very underdeveloped, although Wallace's latest book has a brilliant stab at it. However, note that many MWI bods, including Wallace, are not wavefunction monists. That is, they don't believe that the world is made out of wavefunction in the same way that e.g. Kim believes classical measuring apparatus are made out of quantum particles (as mereological sums?). Wallace, for example, is more interested in a coherent description of the quantum state, or rather just paying attention to the math in ones interpretation.

So what (according to Wallace or whoever) is the ontology of MWI? That is, what, exactly, according to MWI, is physically real?

(2) I understand the context in which you made this point, but still feel that you have been a bit 'cut n' dried' about this here. As you know, Maudlin questions whether we are talking about causal influence when gesturing towards superluminal signalling within an EPR/Bell type scenario. So the alleged 'prohibition' might not be prohibiting the relevant factors. More importantly, does 'non-locality' really have to conflict with relativity? I would suggest that the answer is, at least, not obvious and not trivial.

I agree. The relation between "locality" and "relativity" is more subtle than most people took it to be, before they appreciated that we are going to have to learn to live with nonlocality! I usually try to be careful with the words here, e.g., talking about "relativity's alleged prohibition on superluminal causation" as opposed to just stating that "nonlocality contradicts relativity". I'm don't think it necessarily does, though it's clear that attempts to *reconcile* nonlocality with fundamental relativity run into serious and difficult problems, not least of which is figuring out what the heck "fundamental relativity" should even mean.

I bring up these measurement problem considerations to briefly illustrate the extent to which relativity is already present in the formalism(s) of QFT, coupling this point with the thought that 'pure' interpretations of QM (e.g. MWI) ought to carry over from non-relativistic to relativistic quantum theory.

Yes, sure, relativity is already present in the formalism of QFT... But there are also some relevant things sorely lacking in standard formulations of QFT, e.g., clarity about what the heck it's *about*. This matters hugely, especially for discussions of locality/nonlocality, where the very concept denotes some feature of stuff happening in physical space / spacetime. Is there any stuff happening in physical space /spacetime according to ordinary QFT? Beats me. And I find this at least rather worrisome from the point of view of saying that ordinary QFT "is relativistic".

Incidentally, how do accounts which strive for locality manage concepts that I think are very related, such as non-seperability, holism, etc.? (I think Healey analyses these related concepts well in 'Gauging what's Real'?)

My own personal view is that the whole idea of "non-separability" is just a confused way of saying that a theory involves what Bell called non-local beables. In a theory of exclusively local beables (a "TELB" as I dubbed this in a recent paper) you couldn't *possibly* have "non-separability". The thing that people call "non-separabiliity" really just comes down to some kind of state description which says there are definite relations between "things at separate places" even though neither thing has definite values for the relata. But without fail, the assertion that this is happening is based on taking the quantum state (a nonlocal beable if it is a beable at all) as the description. But notice that it's tacitly assumed that this quantum state is some kind of description of "things at separate places". My view is that people talking this way typically are not at all entitled to this assumption. If the only thing in the theory is some crazy wf on some crazy abstract configuration space, what in the world makes you think there are "two particles, at different locations" (or whatever) in the picture at all? So the whole thing arises from the following confusion: mistaking what should be taken as a non-local beable, for some kind of description of local beables. If the ontology is really just the wf (no local beables) then there is no "non-separability" -- the state is perfectly local and separable in its proper space, the configuration space. On the other hand, if there are local beables in the ontology we should be a lot clearer and more explicit about what those are: put them squarely on the table before we get lost in arguing about how "separable" the ontology is...

 

[malreux]

[1]To recover the Born rule, I think MWI proponents necessarily must introduce some additional assumptions or axioms in the theory
[2] [...] which destroys the main virtue of MWI: minimality of assumptions.

[3] I like to think of Bohmian interpretation as a variant of MWI which just picks one such assumption, which, unlike other attempts of that sort in MWI, seems very natural to me.

[1] Certainly this is a traditional criticism, and if it carries through then [2] follows inevitably. However, what if instead of additional assumptions or axioms we need some good philosophy rather than new physics? I'm going put the point very crudely - there is enough 'going on' to reconstruct classical 'worlds' in my view, although a lot else besides that is very far from classical worlds. The question is - can we so reconstruct in a principled manner? And, if so, is this to be understood as principle-style physics or a theory of emergence? To me the main question in such reconstruction projects is - can the Born rule really be derived from the formalism? Of course, no one is suggesting a bunch of uninterpreted mathematics tells us much in words. But the actual numbers we call probabilities and their relation to the modulus squared is a bit of mystery without further assumptions. Wavefunction realism at least has the virtue that one can talk about QM without reference to probabilities at some level. Perhaps the weakness of MWI in this regards hints at a proper understanding of fundamental physics?

[3] I agree with this, although I doubt ttn will! However, I think this is a bad way for proponents to argue for ddb, accepting as I do the thrust of Wallace&Brown's paper.

 

[malreux]

[1]So what (according to Wallace or whoever) is the ontology of MWI? That is, what, exactly, according to MWI, is physically real?

[2] [...]which is figuring out what the heck "fundamental relativity" should even mean.

[3]Yes, sure, relativity is already present in the formalism of QFT... But there are also some relevant things sorely lacking in standard formulations of QFT, e.g., clarity about what the heck it's *about*.

[4] Is there any stuff happening in physical space /spacetime according to ordinary QFT?

[5] My own personal view is that the whole idea of "non-separability" is just a confused way of saying that a theory involves what Bell called non-local beables. In a theory of exclusively local beables (a "TELB" as I dubbed this in a recent paper) you couldn't *possibly* have "non-separability". "[Non]-separabiliity" really just comes down to [a] state description which says there are definite relations between "things at separate places" even though neither thing has definite values for the relata. But without fail, the assertion that this is happening is based on taking the quantum state (a nonlocal beable if it is a beable at all) as the description. But notice that it's tacitly assumed that this quantum state is some kind of description of "things at separate places". My view is that people talking this way typically are not at all entitled to this assumption. If the only thing in the theory is some crazy wf on some crazy abstract configuration space, what in the world makes you think there are "two particles, at different locations" (or whatever) in the picture at all? [...]If the ontology is really just the wf (no local beables) then there is no "non-separability" -- the state is perfectly local and separable in its proper space, the configuration space.

[1] Well Wallace argues that the Everett interpretation is "really just quantum mechanics itself understood in a conventionally realist fashion." But what does this mean? The main insight here is that to suppose that the linearity of quantum mechanics commits us to macroscopic objects being in superpositions, in indefinite states, is false. How? He offers an analogy - In electromagnetism, a certain conguration of the field - say, F1(x; t) (here F is the electromagnetic 2-form) might represent a pulse of ultraviolet light zipping between Earth and the Moon. Another conguration, say F2(x; t), might represent a dierent pulse of ultraviolet light zipping between Venus and Mars. What then of the state of affairs represented by:

F(x; t) = 0:5F₁(x; t) + 0:5F₂(x; t)?

Must it not represent a pulse of ultraviolet light that is in a superposition of traveling between Earth and Moon, and of traveling between Mars and Venus?Of course, this is nonsense. There is a perfectly prosaic description of F: it does not describe a single ultraviolet pulse in a weird superposition, it just describes two pulses, in different places. And this, in a nutshell, is what the Everett interpretation claims about macroscopic quantum superpositions: they are just states of the world in which more than one macroscopically denite thing is happening at once. Macroscopic superpositions do not describe indeniteness, they describe multiplicity. I think this is the main insight, as to what is physically real - the answer is sought in decoherent tales of emergence...

[2] Good point, I often think about this in the context of QG.

[3] Yes, this is all interpretation-dependent.

[4] As you suggest, very difficult to say in 'standard' QFT (whatever that is). If you think quantum theory qua spacetime theory is a way to go, I like Baez' category-theoretic approach to cobordisms. I won't bore you with the details as this is way beyond the focus of our discussion.

[5] Very interesting, I'd love to download your paper and check it out, which depository? As to the bit I emboldened, I've hinted at my own view in my presentation of Wallace's (basically I agree with him, but am unsatisfied with the lacuna in the 'emergence' picture, and I think this does relate to incomplete physical theory, not solely nice philosophy).

Further thought: the ordinary separability you suggest might tacitly rest on the assumption that configuration space is e.g. an 3N space exactly analogous to 3 space, does this follow?

 

[malreux]

Ah yes, found it, guess its http://arxiv.org/abs/0909.4553

I'll read it this eve, thanks!

 

[ttn]

Ah yes, found it, guess its http://arxiv.org/abs/0909.4553

I'll read it this eve, thanks!

Yes, that's the paper with the "TELB" idea. But note: it has nothing to do with the "separability" stuff that led me to mention it!

 

[ttn]

[1] Well Wallace argues that the Everett interpretation is "really just quantum mechanics itself understood in a conventionally realist fashion."

That's either obviously false, or obvious propaganda. It's of course controversial what exactly "just quantum mechanics itself" should mean, but everybody knows the standard textbook formulations include axioms about measurement that MWI wants to get rid of. That's of course to its credit. The point is, one should really think of MWI as an attempt to get rid of the "measurement problem" that plagues "quantum mechanics itself".

But what does this mean? The main insight here is that to suppose that the linearity of quantum mechanics commits us to macroscopic objects being in superpositions, in indefinite states, is false. How? He offers an analogy - In electromagnetism, a certain conguration of the field - say, F1(x; t) (here F is the electromagnetic 2-form) might represent a pulse of ultraviolet light zipping between Earth and the Moon. Another conguration, say F2(x; t), might represent a dierent pulse of ultraviolet light zipping between Venus and Mars. What then of the state of affairs represented by:

F(x; t) = 0:5F₁(x; t) + 0:5F₂(x; t)?

Must it not represent a pulse of ultraviolet light that is in a superposition of traveling between Earth and Moon, and of traveling between Mars and Venus?Of course, this is nonsense. There is a perfectly prosaic description of F: it does not describe a single ultraviolet pulse in a weird superposition, it just describes two pulses, in different places. And this, in a nutshell, is what the Everett interpretation claims about macroscopic quantum superpositions: they are just states of the world in which more than one macroscopically denite thing is happening at once. Macroscopic superpositions do not describe indeniteness, they describe multiplicity. I think this is the main insight, as to what is physically real - the answer is sought in decoherent tales of emergence...

It's a nice argument. But unfortunately I get off board right at the beginning. (Here I am in total agreement with at least Maudlin and Goldstein.) In the example with the electromagnetic waves, the argument works great, because each of the superposed terms genuinely describes / corresponds to "a pulse of ultraviolet light". On the quantum/MWI side of the analogy though, I do not accept that the kind of thing that is ordinarily called "a single branch of the universal wf" -- I mean supposing that this "one branch" were the whole wave function, i.e., supposing the universal wf has just one branch -- corresponds in any obvious way to "a world". You see, the problem is that even a universal wf with "one branch" is still what... some field with support only in some small region of a huge configuration space. This looks nothing like a world full of galaxies and planets and people and trees and cats and whatnot. (Sure, the quantum state is some approximate eigenstate of various operators like "there's a galaxy over there", "there's a cat-shaped lump over here", etc... But that is *not the same thing at all*! Note that in ordinary QM, where most of us develop our intuitions about when quantum states possesses definite properties and what this means, the *meaning* is always cashed out in terms of some classical/macro apparatus that somehow registers the result of an experiment. So it is totally unwarranted to extend this intuition to MWI, whose whole point is to *deny* that there is a distinct non-quantum macro/classical posited world.)

So I don't buy this kind of argument at all. If it were true that each "branch" of the universal wave function somehow corresponded to some definite sensible bit of ontology in 3 space, then I would be with the argument all the way. But it doesn't. So the MWI people I think need to explain either how they get local beables out of the wf, or they need to posit some local beables distinct from the wf. Then maybe an argument like that could fly.

Further thought: the ordinary separability you suggest might tacitly rest on the assumption that configuration space is e.g. an 3N space exactly analogous to 3 space, does this follow?

I didn't understand what you meant here.

 

[Demystifier]

The question is, which theory is best at explaining what needs to be explained?

The only problem with this is to define what exactly it means to explain something "the best". Obviously, we don't have a function f:E -> R from the set of all explanations E to the set of real numbers R, where the real number can be interpreted as a measure of "quality" of an explanation. So, in the absence of such an objective measure, different physicists use different subjective vague measures of it, which results in different interpretation of QM without a possibility to reach a consensus about which one is "the best".

The only such function that comes to my mind is the inverse number of words used in the explanation. But by this definition, the best explanation would be the shut-up-and-calculate interpretation, and I am sure you wouldn't accept that this interpretation is "really" the best. (Although, for most practical physicists it probably is.)

Or if you can propose another concrete objective measure, I would be REALLY REALLY happy to see your proposal.

Anyway, even without such a measure, I think you and I agree which interpretation is "the best" - the Bohmian one. But perhaps the problem is that we don't agree which is "the second best" or "the third best".

 

[Demystifier]

I don't really understand your reasons for liking it, that it provides a simple mechanism for effective wave-function collapse. That to me seems way too bound up with ideas from orthodox QM.

For many purposes, orthodox QM is not so bad at all. Of course, I am not satisfied with it and I want more, but it doesn't necessarily mean that a better theory should not borrow some ideas from orthodox QM. In fact, being strictly against orthodox QM is an orthodoxy itself, so by not being strictly against orthodox QM I am perhaps less orthodox than you.

which theory does the best job, by the normal standards of judging scientific theories, of accounting for all the facts (by which I mean to include both pre-scientific facts like that there are trees and stuff outside my window, and then also the detailed quantitative results of all the various sophisticated experiments like 2-slit, atomic spectra, Bell, etc. etc. etc.).

There is one related joke I think you might like:
http://abstrusegoose.com/276

 

[Demystifier]

Yes, sure, relativity is already present in the formalism of QFT... But there are also some relevant things sorely lacking in standard formulations of QFT, e.g., clarity about what the heck it's *about*. This matters hugely, especially for discussions of locality/nonlocality, where the very concept denotes some feature of stuff happening in physical space / spacetime. Is there any stuff happening in physical space /spacetime according to ordinary QFT? Beats me. And I find this at least rather worrisome from the point of view of saying that ordinary QFT "is relativistic".

On that issue, I would like to note that, in my view, ordinary QFT is not "relativistic enough", so in
http://xxx.lanl.gov/abs/0904.2287
I attempt to make it "more relativistic" than it is in the usual formulation.

 

[ttn]

The only problem with this is to define what exactly it means to explain something "the best". Obviously, we don't have a function f:E -> R from the set of all explanations E to the set of real numbers R, where the real number can be interpreted as a measure of "quality" of an explanation. So, in the absence of such an objective measure, different physicists use different subjective vague measures of it, which results in different interpretation of QM without a possibility to reach a consensus about which one is "the best".

Yes, I agree, it's hard and controversial to measure. But I still think it's best to keep one's eyes on the real goal and not get too sidetracked by the goal of fixing what's wrong with the theory one learned in school. That last attitude tends to make one focus on the wrong things and head off in wrong directions.

Re: the cartoon you shared, that's great. I do like it very much! Indeed, it captures very well a big part of my objection to MWI and other such theories. It's really ridiculous for us to conclude -- after long thought based exclusively on evidence from the senses -- that actually everything the senses give us is some kind of delusion. It's like the ultimate circular argument. So one of my fundamental priorities in trying to understand QM is to avoid doing this. That's why the ability to recover something that genuinely looks like the ordinary world of ordinary perception is so important to me, and why I am very dismissive of "solipsist" type ideas.

 

[Demystifier]

Re: the cartoon you shared, that's great. I do like it very much! Indeed, it captures very well a big part of my objection to MWI and other such theories.

I'm glad to hear that.

It's really ridiculous for us to conclude -- after long thought based exclusively on evidence from the senses -- that actually everything the senses give us is some kind of delusion. It's like the ultimate circular argument. So one of my fundamental priorities in trying to understand QM is to avoid doing this. That's why the ability to recover something that genuinely looks like the ordinary world of ordinary perception is so important to me, and why I am very dismissive of "solipsist" type ideas.

Yes, I can understand it very well, because most of the time (spent on doing science) I think the same way. But still, I find it important, amusing and intellectually challenging to try the other ways of thinking as well. If some scientists may be religious part of their time, then also a Bohmian may be a slightly Bohrian part of the time.
It's better to be Bohred than to be bored.

 

[ttn]

By the way, I wonder if mattt is still reading? Mattt, I'm still really interested in hearing your thoughts on Bell's formulation of "local causality". It seemed like you were skeptical at first, then you read my paper and bought into it completely, but then went back to skepticism. I'm really anxious to hear what caused that last transition!

 

[malreux]

[1]That's either obviously false, or obvious propaganda.

[2] The point is, one should really think of MWI as an attempt to get rid of the "measurement problem" that plagues "quantum mechanics itself".

[3] On the quantum/MWI side of the analogy though, I do not accept that the kind of thing that is ordinarily called "a single branch of the universal wf" -- I mean supposing that this "one branch" were the whole wave function, i.e., supposing the universal wf has just one branch -- corresponds in any obvious way to "a world".

[4] Note that in ordinary QM [...] the *meaning* is always cashed out in terms of some classical/macro apparatus that somehow registers the result of an experiment.

[5] So I don't buy this kind of argument at all. If it were true that each "branch" of the universal wave function somehow corresponded to some definite sensible bit of ontology in 3 space, then I would be with the argument all the way.

[6] I didn't understand what you meant here.

[1] Lol! I'm sorry, I was being deliberately controversial here, and being unfair to you (in terms of the level of debate, i.e. me reducing it to a slanging match) and Wallace (he actually carefully presents his views, albeit this is a direct quote).

[2] Sure!

[3] I agree this isn't obvious, if what you want is either an exact MWI, or even an aprox decoherent MWI. This is where I depart from proponents like Wallace, I interpret Everett in a way closer to e.g. Barrett.

[4] Agreed. However this presentation could go wrong in a way similar to how the Verification Principle goes wrong, by referring to semantics when what we want to know is what the world is like. I know the previous sentence is vague, I just don't want to go down this route. I'll interpret your 'meaning' as 'referring' if you don't mind.

[5] By the principle of charity, I would choose to believe MWI proponents aren't arguing this as is, since most don't seem bothered by problems like branch count and so forth. Clearly their urging something aprox, emergent, decoherent, fill-in-blank.

[6] Something involved that's only going to take us further away from the topic of this thread.

Talking of which: I feel we could have a really interesting discussion about interpretations of QM, or 'QM's'. However, I feel like I've steadily drawn you further and further away from the topic of this thread. Sorry about that! I mean, obviously Bell's Theorem is enmeshed with such issues, but I feel like I want to debate the merits and other re Bohm v Everett and perhaps that's not actually conducive to the topic. So I'll cease fire, for now.

 

[ttn]

[3] I agree this isn't obvious, if what you want is either an exact MWI, or even an aprox decoherent MWI. This is where I depart from proponents like Wallace, I interpret Everett in a way closer to e.g. Barrett.

I'm not sure you understood the worry I have in mind. It really has nothing to do with how exactly you'd "carve" the big wf into "branches". There are undoubtedly difficult questions for MWI there, too, but that's not at all what I meant to be expressing worry about. Let's just imagine (i.e., take for granted) that there was some clear way of defining a "branch" and that we didn't have to worry about different branches overlapping/interfering or anything like that.

Let me try to express the worry this way. Suppose the universe just had a bajillion particles in it, and suppose those bajillion particles clumped together to make a ball. Let's ignore the bajillion-3 *relative* degrees of freedom (just assume those are always in a nice "making a ball" kind of eigenstate) and just focus on the 3 'center of mass' degrees of freedom. The usual worry about MWI goes something like this (I'm trying to relate this back to the example about the two/superposed light rays):

1. If the center-of-mass degrees of freedom are in an (approximate) position eigenstate corresponding to the ball being "here", then the ball is here, and there's no problem.

2. If the center-of-mass degrees of freedom are in an (approximate) position eigenstate corresponding to the ball being "there", then the ball is there, and there's no problem.

3. But in QM, and inevitably according to the unitary evolution, we're going to end up with the state being a non-trivial *superposition* of the state mentioned in 1 and the state mentioned in 2. And then that's really weird, because the ball isn't in either place, or it's in both, or maybe there are two balls in parallel/noninteracting worlds, or whatever.

Now I think you were responding as if my worry was "in the superposition kind of situation mentioned in 3, how can you really say that there are two balls in parallel/noninteracting worlds when really the two superposed terms have tails and maybe they overlap a little bit ..." or something like that. But that isn't the worry at all. The worry is: I don't agree with 1 and I don't agree with 2. That is, if the wave function is the only thing in the game, then I don't see how the (rough) eigenstate mentioned in 1 has *anything to do with* there being a ball at some place. I think the burden is on the MWI people to explain, precisely, what the quantum state mentioned in 1 has to do with there being a ball here (i.e., some kind of lump of matter at a particular place in 3-space). I of course know of various ways you could do some mathematical thing to the wave function and arrive at something that could be interpreted as a mathematical representation of a ball here. But why in the world should I take that particular mathematical thing seriously, as yielding some "real ontological stuff", when there are many other mathematical things I could have done that (I assume??) I'm *not* supposed to interpret as giving me some "real ontological stuff".

I can never quite tell which of the following the MWI answer is supposed to be: (a) It's obvious, you project down from 3N space to 3 space in the obvious way, something like the way the "m" field is computed in GRWm or Sm, and that "m" is the local beables; or (b) no, you're missing the point, there are no local beables at all, instead what we think of as "matter in 3 space" is really just an illusion in the minds that emerge directly from the 3N-space wave function which is the only physical reality. Basically the question comes down to: what the heck are the local beables of MWI? Are there some? Or none? If some, I want to know exactly what they are, and maybe something along the lines of "why those??". Then we can have a fair comparison with other theories like dBB (without making it seem like one is "simpler" really only because half of it was left tacit!). Or if none, if (b), then we should acknowledge how weird and almost solipsist that is even though in some sense this is still a "realist" theory.

Talking of which: I feel we could have a really interesting discussion about interpretations of QM, or 'QM's'. However, I feel like I've steadily drawn you further and further away from the topic of this thread. Sorry about that! I mean, obviously Bell's Theorem is enmeshed with such issues, but I feel like I want to debate the merits and other re Bohm v Everett and perhaps that's not actually conducive to the topic. So I'll cease fire, for now.

Well, nobody's really discussing Bell here anymore, so I have no objection to chatting about other related stuff if you want to. Or email me or something.

 

[lugita15]

Well, nobody's really discussing Bell here anymore, so I have no objection to chatting about other related stuff if you want to. Or email me or something.

I'm still interested in discussing Bell. Although I haven't yet come up with a good counterargument to your "several axes" version of EPR, it still seems to me that counterfactual definiteness is important. Earlier this thread, when I brought it up you said that CFD is either metaphysical and unimportant, or insofar as it is important it is so essential for all scientific theories that it shouldn't be questioned. Yet I think that quantum mechanics does not possesses it; to wit, if you measure the polarization of a photon at 0 degrees, then in QM the question "What would have been the result if you had instead measured the polarization at 45 degrees" has no definite answer. This seems to me to be saying more than just that quantum mechanics is nondeterministic. You can have a nondeterministic theory in which given the present state of the world, you cannot determine the future state of the world, but you CAN determine what would have been the alternate present states of the world if different measurement decisions had been made. In contrast, in quantum mechanics the quantum state right now is not sufficient to tell you what the present quantum state would have been in alternate histories. I think that this is a significant fact, don't you?

 

[ttn]

I'm still interested in discussing Bell. Although I haven't yet come up with a good counterargument to your "several axes" version of EPR, it still seems to me that counterfactual definiteness is important. Earlier this thread, when I brought it up you said that CFD is either metaphysical and unimportant, or insofar as it is important it is so essential for all scientific theories that it shouldn't be questioned.

I don't think I said that latter about CFD. Or at least that's not exactly what I meant. What I think is more like this: you will realize that this whole issue of CFD simply melts away into nothingness (I mean, it becomes clear that there is no issue here at all) if you think of Bell's theorem as a constraint on *what theories say* -- as opposed to trying to think of every last character in the math as somehow referring directly to some real experimental outcome.

Yet I think that quantum mechanics does not possesses it; to wit, if you measure the polarization of a photon at 0 degrees, then in QM the question "What would have been the result if you had instead measured the polarization at 45 degrees" has no definite answer.

But this question does have a definite answer: "If you had instead measured at 45 degrees, what would the possible results have been, and what are their probabilities?" That is, the reason QM gives no definite answer to your question is only that QM is not deterministic. But that certainly doesn't matter. You can derive the Bell inequality just fine, from locality, without invoking determinism.

Of course, you might (as many people have) look at some derivation of the Bell inequality in some textbook and see that it seems to *start with* -- to *presume* -- pre-existing (deterministic) answers/outcomes to all these different possible questions/measurements. But that's just because many commentators and textbook authors confuse (what we call in the article) "Bell's inequality theorem" for the full "Bell's theorem". The full "Bell's theorem" starts just with the assumption of locality and *derives* the pre-existing (deterministic) answers/outcomes, using basically the EPR argument. So really the whole thing leading to this red herring about CFD is simply missing this, failing to realize that "Bell's inequality theorem" and "Bell's theorem" are not the same thing. Put it this way: it's true that QM is not the type of theory that is assumed in standard derivations of "Bell's inequality theorem". But this is of no real relevance whatsoever. Actually what is shows is just this: QM is not a local theory! (Because, if it were, it would have to explain the perfect correlations with pre-existing values, the way the EPR argument proves any local theory must.)

This seems to me to be saying more than just that quantum mechanics is nondeterministic.

I don't agree. It is no more than that. But the real issue is the EPR half of the argument.

You can have a nondeterministic theory in which given the present state of the world, you cannot determine the future state of the world, but you CAN determine what would have been the alternate present states of the world if different measurement decisions had been made.

I don't think so, but who cares. The real point is this: you cannot have a non-deterministic theory that explains the perfect correlations in a local way.

In contrast, in quantum mechanics the quantum state right now is not sufficient to tell you what the present quantum state would have been in alternate histories. I think that this is a significant fact, don't you?

Nope. Except, as I explained above, insofar as it's just an obscure way of confessing that QM is nonlocal. Which of course means that it is hardly some kind of counterexample to Bell's claim that nonlocality is required to generate the QM predictions.