Against Realism: Examining the Meaning of Local Realism

[DrChinese]

"Against" Realism

Travis Norsen has written an article entitled "Against Realism". In it, he argues that the phrase "local realism" is not meaningful.

Against Realism (2006)

Abstract:
"We examine the prevalent use of the phrase “local realism” in the context of Bell’s Theorem and associated experiments, with a focus on the question: what exactly is the “realism” in “local realism” supposed to mean? Carefully surveying several possible meanings, we argue that all of them are flawed in one way or another as attempts to point out a second premise (in addition to locality) on which the Bell inequalities rest, and (hence) which might be rejected in the face of empirical data violating the inequalities. We thus suggest that this vague and abused phrase “local realism” should be banned from future discussions of these issues, and urge physicists to revisit the foundational questions behind Bell’s Theorem."

-----

My questions for your consideration:

1. What does realism mean to you?

2. Einstein said: "I think that a particle must have a separate reality independent of the measurements. That is: an electron has spin, location and so forth even when it is not being measured. I like to think that the moon is there even if I am not looking at it." Is this realism?

3. In your opinion, is "realism" an assumption of Bell's Theorem? If so, where does it arise?

 

[Careful]

My questions for your consideration:

1. What does realism mean to you?

2. Einstein said: "I think that a particle must have a separate reality independent of the measurements. That is: an electron has spin, location and so forth even when it is not being measured. I like to think that the moon is there even if I am not looking at it." Is this realism?

3. In your opinion, is "realism" an assumption of Bell's Theorem? If so, where does it arise?

Me stupid realist think that such papers are silly. For me realism means the existence of an external world, independent of observation in which correlations between two events are due to processes which satisfy local equations of motion. Einsteins quote is an example of how a realist proposes a model of the world, although it might very well be that particles are not the fundamental degrees of freedom. But I like to think the moon is also there, even if we do not see it. Realism is *not* an assumption of Bell per se, there is a very specific form of realism present (for example, Bell excludes extra dimensions which could be used to define a holographic principle) which together with the screening off condition (which is not a logical consequence of locality) leads to a conflict with the *measurement* postulate of quantum mechanics. There are at least four well known local mechanisms which violate the ``logic'' in this paper : holography, polarizable media, negative ``probabilities'', predeterminsim. That is all the objective content there is to this discussion in my mind. Barely enough to fill an abstract with...

Since you dr Chinese, seem to be so adsorbed with Bell inequalities, why don't you learn a bit more about those approaches which might violate them. You like these negative probabilities and actually I do too : Feynman, Dirac and others have written beautiful papers about this subject in the context of the meaning of quantum mechanics. Feynman once said that ``the only difference between quantum and classical mechanics seems to be that probabilities can go into the negative''. Feynman never doubted the validity of ``traditional'' QM (Schrödinger wave + 100 percent measurement accuracy) but kept on searching for local computer models, very much the right spirit in my mind.

Careful

 

[chronon]

I agree with the paper that the term 'realism' in 'local realism' is pretty much meaningless. See http://www.chronon.org/Articles/localreal.html

 

[Hurkyl]

Me stupid realist think that such papers are silly. For me realism means the existence of an external world, independent of observation in which correlations between two events are due to processes which satisfy local equations of motion.

This is just local hidden variables, is it not? Each point in space-time has some parameters "attached" to it, the outcome of any measurement performed in a region can be completely determined by the values of the parameters appearing within the region, and the value of a parameter at any particular point is completely determined by the values of the parameters in any cross-section of its past light-cone. Is that a fair assessment?

 

[Careful]

This is just local hidden variables, is it not? Each point in space-time has some parameters "attached" to it, the outcome of any measurement performed in a region can be completely determined by the values of the parameters appearing within the region, and the value of a parameter at any particular point is completely determined by the values of the parameters in any cross-section of its past light-cone. Is that a fair assessment?

I would think about this in the first place, but as I explained hundred's of times, that does not imply screening off, non local correlations can be there, neither does it imply dichotomic outcomes, it allows for all possibilities I mentioned previously. One could also allow for laws where the value of x now depends on its nearest elements in the future, that would still be a local spacetime theory. So, it seems to me that the arrow realism -> nonlocality is not logically possible. This is the last thing I say about this unless you have some objections since all this kind of philosophy is turning my stomach. Physics for me is a game of playing around with possible scenario's until the puzzle fits, in such state of mind expressing one's religion leaves a rather silly impression.

 

[selfAdjoint]

For me realism means the existence of an external world, independent of observation in which correlations between two events are due to processes which satisfy local equations of motion.

And that's umm, how many independent premises?
1. External world
2. Independent of observation
3 Correlations due... to equations of motion.

Note that one can hold 3 (quantum equations being specified), without holding 1 or 2. Likewise it's possible to conceive an external world that is contingent on observation (close to Hume's position), or on one in which observation can not give us truly reliable information (not too far from Kant).

Any one or all of these must be considered independently in evaluating quantum correlations, especially in the Bell context, and I agree with the essay that a fast shuffle between them is characteristic of many who would argue from Bell against standard QM (whether Copenhagen, MWI or whatever interpretaion is used). One could illustrate with many past posts on this forum.

Generally speaking, I believe "external reality" is not an operationally meaningful concept. Has anybody seen an experiment that rules out solipsism?

 

[Careful]

Generally speaking, I believe "external reality" is not an operationally meaningful concept. Has anybody seen an experiment that rules out solipsism?

Has anybody seen an experiment to test the premise that I have 100 ghostlike unmaterial eyes which are hanging in China and Japan which unfortunately do not send any signals to my brain (so there is no way to interrogate me to check if I can see beyond my room or not)? SelfAdjoint, honestly, do you not find such kind of reasoning a bit silly ? There have been written plenty of texts by good physicists which explain the necessity of an objective reality as a necessary condition for science to be possible. I still have to wait for the first great physicist to say something positive about solipsism, do not confuse philosophers of physics with physicists. I agree with Norsen concerning the need for realism, I think to have shown on many occasions that it does not exclude a well defined form of local realism. Therefore, I conclude that if his aim is to disprove locality, then well yeh ... on the other hand, it is good that some terminology is settled, although his conclusions about many scientists not knowing what they are talking about when they speak about reality seem to be unfair to say the least.

Careful

 

[RandallB]

As he uses “Metaphysical” it strikes me as redundant and could more easily be disposed of than realism. All forms useful forms of realism should reject the imaginary realism of solipsism.
Also Travis asks we change how we use the word so why didn’t her offer a specific list of how he would see that change being made.

As to 1.) What does realism mean to me
Looks like Naive would describe me best.

2) Einstein’s real means real
Yes I think that is true – otherwise science would need to try show the Dail Lama could reach total spiritual enlighten and discard and destroy reality as he imagined it and then explain what would happen to the reality imaged by those of us remaining – too bizarre for me.

3) "realism" in Bell's Theorem - where does it arise
Maybe you can say that bell addresses a specific type of realism “local realism” of localility. And as it seems to show it may not be true that a different form of realism or reality other than what my farm boy common sense insists on believing in.
Several different theories or kinds of reality are proposed:
BM reality: where Local equations of motion solve bell by using the guidance of a guide wave traveling unseen in an extra dimension(s) or space overlapping our own “Local Reality”.
WMI reality: What ever wider reality & dimensions as needed to solve the paradox.
QM reality: With uncertain probabilistic outcomes linked though superposition the original alternate form of reality.
How many are there?

But Bell can only shed light on the (Naive) Local Realism vs. all other forms Realism some with there own version of local (guide wave, extra dimensions, collapse of entanglement, etc.).
But I don’t think Bell can help show or select which of the Non-Naive versions is correct. Leaving it to each version to find another experiment to decide between non-local ideas.

Here is an option - if we look around and find that Einstein Unknown Variable we could kick them all and go with the Naive Realisim instead. Help me out and take a look under your desk anything hidden there?
If not I’ll have to keep looking.

 

[selfAdjoint]

Has anybody seen an experiment to test the premise that I have 100 ghostlike unmaterial eyes which are hanging in China and Japan which unfortunately do not send any signals to my brain (so there is no way to interrogate me to check if I can see beyond my room or not)? SelfAdjoint, honestly, do you not find such kind of reasoning a bit silly ? There have been written plenty of texts by good physicists which explain the necessity of an objective reality as a necessary condition for science to be possible. I still have to wait for the first great physicist to say something positive about solipsism, do not confuse philosophers of physics with physicists. I agree with Norsen concerning the need for realism, I think to have shown on many occasions that it does not exclude a well defined form of local realism. Therefore, I conclude that if his aim is to disprove locality, then well yeh ... on the other hand, it is good that some terminology is settled, although his conclusions about many scientists not knowing what they are talking about when they speak about reality seem to be unfair to say the least.

Careful

And you can cite thousands of experts on the necessity of believing in this or that religion for a coherent life. So what? Nullius in Verba stands at the head of our modern understanding of how to do science. Patrick Vanesch calls relational QM a form of solipsism, yet there are respected physicists who adopt it. It's just opinion.

My statement that external reality is not an operationally defined concept was intended to suggest that it cannot be meaningfully employed in discussing physics, because there is no objective way to decide which of two definitions is preferrable. And that is what I see working in all the hundreds of posts that have been put up here in all the many threads on Bell, locality and interpretation. It isn't science, it isn't even well constructed philosophy!

 

[Careful]

** Patrick Vanesch calls relational QM a form of solipsism, yet there are respected physicists who adopt it. It's just opinion. **

And I agreed with this assesement (actually I made it myself PRIOR to Patrick's intervention), but I also remember from the discussion with f-h that these respected physicists are indeed worried about the lack of synchronisation as they should be. You might want to read what E.T. Jaynes had to say about MWI, he simply said that ``theoretical physics went bezerk''. It is not opinion, an overwhelming majority does not even want to speak about it.

**
My statement that external reality is not an operationally defined concept was intended to suggest that it cannot be meaningfully employed in discussing physics, because there is no objective way to decide which of two definitions is preferrable. **

Rubbish, first of all your operational definition of measurement requires something you cannot capture at all, nobody agrees upon what consciousness is ! In science, you always have to make a definition no matter how you turn or twist the pig. If no such agreement can be made, then you can better close the books and go into politics.

** And that is what I see working in all the hundreds of posts that have been put up here in all the many threads on Bell, locality and interpretation. **

Do you really think that practising scientists care about this ?!
I know of people who think hard about these matters (and that is already not very standard), but none of them would come up with solipsism as it is presented here. 95 percent, in my experience, takes a very pragmatic approach towards this, and with good reason.

** It isn't science, it isn't even well constructed philosophy! **

My god, so you are denying 2000 years of scientific practice ? I am afraid you have been too long amongst ``philosophers'' of science.

By the way if you live by the motto ``on basis of experiments we establish ``truth'' but not on the basis of words of anyone'' well SelfAdjoint, don't forget then that
(a) science is not about the truth, that does not exist. Science is about making a good model of nature in which all factors involved are well defined and fair predictions about the future can be made.
(b) no model should contain any assumption which in principle cannot be falsified by the very instrument of science itself.

Consciouness does not fall under (b), a theory which declares what is real belongs to the material world in principle does.
As I said, you are involved in philsophy of science as we understand it today. The problem with philosophers of science is that they usually do not know the internal dynamics of science where a world view is connected with the boundaries of our understanding. Your philosophy is an unheatlhy extrapolation from the mathematical framework of QM : an extrapolation which shows that any ``believer'' is prepared to acknowledge that he/she did not understand anything about it and is moreover not going to look for anything better ; it is the task of science to take mystery away from things. For example : you are proclaiming the relational point of view (which does contain some serious problems); if you once yourself calculated a solution of Brans Dicke theory or something similar and then imagine the trouble you would have to go through in making such theory a logical theory of relational quantum propositions, to realize in the end that you are stuck with two different notions of time, things like self measuring devices and so on to make your reasoning selfconsistent then I am sure you would pretty quickly pack your bags and try something more realistic to start with.

But again, nothing but philosophical comments, where is the real science ?

Careful

 

[Hurkyl]

This is the last thing I say about this unless you have some objections since all this kind of philosophy is turning my stomach.

Well, the problem is that you haven't answered my question at all. I was trying to translate what you wrote into something more precise, so I could see if what I think you're saying is what you think you're saying.

One could also allow for laws where the value of x now depends on its nearest elements in the future, that would still be a local spacetime theory.

Okay; let's throw in the entire past and future lightcones for simplicity, then discuss particulars when we decide it matters.

 

[Careful]

** Well, the problem is that you haven't answered my question at all. I was trying to translate what you wrote into something more precise, so I could see if what I think you're saying is what you think you're saying. **

I *literally* said that your definition of local realism is *fine*, although some alterations could be made (such as taking into account some part of the future), so that makes one wonder why you consider this as no answer at all. Your definition was in no sense more precise (!), you keep on confusing writing clear ideas in a mathematical language as a significant contribution. What I just said, you can all find in Franco Selleri ``The EPR paradox'' chapter 5 : proposed solutions (although he would call holographic solutions ``non local'', better would be apparently non local). In contrast to what some think, rigorous definitions of local realism *have* been made long time ago... Clearly adding some part of the future helps, since then Bob can send a signal to the future of Alice with the result of his measurement and vice versa, but that would be non local, but causal. Some holographic principle is clearly local and solves the issue, it is just that events which are far away on the ``base manifold'' can be arbitrary close together in the extra conal dimension (actually this idea is ``cleaner'' than the wormhole suggestion since you have no problems whatsoever here with defining a global arrow of time), etc... Now, you might want to read this chapter prior to making any further comments on what I said, which is the respectable thing to do. In contrast to some, I am not going to say after you read this, that you might be drawing your conclusions on a paper uttering some simple ideas since at least the physical context is very clear in this book (which should be studied by anyone in that field in my opinion) - Selleri is a first class scientist. EDIT : it needs to be said that the idea of negative probabilities would deserve more attention, a good survey paper on this is ``A review of extended probabilities'' Physical reports, 133, No 6 (1986), pages 337-401.

Careful

 

[RandallB]

(a) science is not about the truth, that does not exist. Science is about making a good model of nature in which all factors involved are well defined and fair predictions about the future can be made.
(b) no model should contain any assumption which in principle cannot be falsified by the very instrument of science itself.

Consciouness does not fall under (b), a theory which declares what is real belongs to the material world in principle does.

I like these; concise and to the point, descriptions.
Do they come from the Selleri (and Afriat) book “The Einstein, Podolsky, and Rosen paradox” you mentioned.
Or do you have another source for them – I ‘d like to read the whole of the ideas around them.

 

[Careful]

I like these; concise and to the point, descriptions.
Do they come from the Selleri (and Afriat) book “The Einstein, Podolsky, and Rosen paradox” you mentioned.
Or do you have another source for them – I ‘d like to read the whole of the ideas around them.

No, they come from me, it is in my blood (and I am dead serious about it) But if you insist, the introduction of the latter book contains some spicy remarks about the sociology of science. However, I all find this a bit sad, you know ... (a) first people say ``you cannot violate the Bell inequalities´´ then (b) when you point out it can be done, it is against their religion. The book of Selleri and Afriat is a true masterpiece of objectivity in science (and very informative as well), he is even slightly critical towards other well known realists. It contains as I seem to remember also a section about *single event* interpretations of the Heisenberg principle and a brief summary of proposed tests to violate it. I would recommend anyone interested in physics to buy it, even if it were just to broaden your horizon and to remain open to the possibility that local realism (or something close to it, like the value of a field in one spacetime point being determined by the field content in some finite part of the future and past lightcones) can return.

However, it does not contain any full solution to the paradox which is satisfactory *in my mind* and which has been shown to reproduce all QM effects (although one can be hopeful). Merely it outlines and explains well the ideas behind the different proposals for the solution of the latter and provides enough useful references.

Careful

 

[Rade]

...What does realism mean to you?...

Reality is that which exists. How humans know it or not is another matter--IMO.

 

[simon009988]

1. What does realism mean to you?
Realisim to me is what your mind percieves.


2. Einstein said: "I think that a particle must have a separate reality independent of the measurements. That is: an electron has spin, location and so forth even when it is not being measured. I like to think that the moon is there even if I am not looking at it." Is this realism?

This is pure supernatural and cannot be measured in any way, therefore not reality.

 

[Careful]

** 1. What does realism mean to you?
Realisim to me is what your mind percieves. **

So, what is your mind and what is perception ? By what mechanism does your mind operate ?

**
2. Einstein said: "I think that a particle must have a separate reality independent of the measurements. That is: an electron has spin, location and so forth even when it is not being measured. I like to think that the moon is there even if I am not looking at it." Is this realism? **

This is a form of realism.

** This is pure supernatural and cannot be measured in any way, therefore not reality. **

Sure it can be measured, but not in the way you imagine measurement.
So, your mind measures your mind he and stones, ants and so on have this precious commodity too. Basically, it occurs to me you are denying the practise of physics. How was an electron observed ? By following the presumed track in a bubble chamber one says ``this track was formed by the electron which passed here´´. So, one assumes the electron passed there even if no one had seen it ! We speak about particles, since pictures coming from scattering experiments suggest us such interpretation because we see something *was* there prior to our minds measuring circumstantial evidence for it.

Careful

 

[Hurkyl]

Careful,

You can't communicate very well unless the person speaking and the person listening both think the same thing was said. It happens often that two people will see the exact same words and come away with entirely different opinions about what they said.

I *literally* said that your definition of local realism is *fine*

For example, after reading this, I would have expected there to be, in your original post, a phrase that resembled "your definition of local realism is fine"... or at the very least, expected to see the word "fine".

writing clear ideas

The clarity of an idea is, of course, a subjective notion. And even if the idea is clear, the writing of the idea may not. After all, "yes" is rather clear, yet you managed to write five long sentences in post #5 without managing to convey it to me.

And the converse is true...

you keep on confusing writing clear ideas in a mathematical language as a significant contribution.
...
making any further comments on what I said
...
you might be drawing your conclusions

I was simply trying to figure out what you were saying, while trying very carefully not to "make comments" or to "draw conclusions" -- and making a "significant contribution" was the furthest thing from my mind. But, I did not manage to convey that to you.

I translated into the "mathematical language" because, IMHO, there's less "wiggle room" for conflicting interpretations -- and also because it's a generally more objective language. If we can agree on a mathematical translation, then I would have high confidence that what I think you said matches what you think you said.

But if you would rather be misinterpreted... then you can't complain when it happens.

 

[Careful]

***
I translated into the "mathematical language" because, IMHO, there's less "wiggle room" for conflicting interpretations -- and also because it's a generally more objective language. If we can agree on a mathematical translation, then I would have high confidence that what I think you said matches what you think you said.

But if you would rather be misinterpreted... then you can't complain when it happens. **

Thanks for clearing that out - I will simply add references each time I make an apparently ``strange'' statement. The reason why I gave you a broader answer in my first reply is because local realism (and physics in general) is not something static for me. If I restrict myself to your more or less traditional specification then I am cornered into the four options I mentioned (or you can also declare the measurement loophole to be fundamental); on the other hand, there has always been this interesting possibility of using advanced Green functions ... What I want to tell you is that physics consists of having mechanisms in your mind, imagining yourself how stuff works ; since quantum mechanics makes this difficult for you, the first thing to do is to clear out the latter. I have never been ``pro string theory'' but self Adjoint recently mentioned that he thought to have heard that within ST a non supersymmetric fermi model had been constructed, now that would be interesting physics. We still don't understand yet our first elementary particle (electron models are a recurrent theme in the literature - especially in the 70 ties and 80 ties, some considerable effort was put into that).

Careful

 

[vanesch]

By the way if you live by the motto ``on basis of experiments we establish ``truth'' but not on the basis of words of anyone'' well SelfAdjoint, don't forget then that
(a) science is not about the truth, that does not exist. Science is about making a good model of nature in which all factors involved are well defined and fair predictions about the future can be made.
(b) no model should contain any assumption which in principle cannot be falsified by the very instrument of science itself.

In as much as (a) is a very profound statement, you should apply it entirely logically until the end. I agree with (a), with maybe one small modification: I'm not even sure we can say that "truth doesn't exist", only, that "truth isn't accessible", and we leave in the middle whether truth exists or not. We simply know with certainty that we won't know it with certainty. The realisation of this philosophically elementary fact does - IMO - a lot of good to any of these sometimes heated discussions


And in fact it is entirely the point I'm often making: don't think that what we scientifically "knew" 400 years ago, today, or 400 years from now, has much to do with "truth". It is all about, as stated, making a model of observational data. However, in as much as this model has some logical consistency, it also has its *Platonic*, conceptual existence. As such, I don't agree at all with (b), which I find a totally arbitrary requirement. The only goal of the model is to make a model of observational data. What elements are used to set up such a model is entirely free, I'd say. Of course there can be *preferences* and Occam's rasor is a good guide: two models which are empirically equivalent, but one which has some extra elements which the other doesn't need, might be preferentially rejected in favor of the "simpler" one. But again, this is just a guiding principle, not a strict requirement.

Given that we have no access to "truth", but only to observational data, the best we can ever do is to think up (one or several) models which can explain these observational data. It would be nice too, if these models were logically consistent. But as to what a model should contain, and not, is, IMO, purely a matter of taste. It needs to work, it needs to make correct empirical predictions, it needs to be logically consistent and that's it.

As was pointed out (and as is philosophically also known) is that solipsism is irrefutable (but also not very productive as a starting point). That means that whatever you take as "reality", it is a hypothesis, and nothing more. You can never *prove* any reality, so it is always a hypothesis.
So with the word "reality" always comes a hypothesis.

As to 1.) What does realism mean to me
Looks like Naive would describe me best.

This is correct: the "realism" one is talking about here is what's philosophically called "naive realism" (but without any pejorative meaning to "naive"). It simply means that the hypothesis of reality assumed, is that what constitutes "reality" is nothing else but what is observed.
For a long time, physicists have taken this hypothesis, even without saying so ; it is only since Faraday and his "field lines" that people started talking about "things that might be real but not directly a product of our observation", and hence, a modest departure from naive realism.
It is my understanding that the word "realism" in the context of Bell/EPR/local "realists" etc... is this: a version of naive realism which ultimately makes the hypothesis that observation is what is "real".

Another possible hypothesis of reality (let's not forget that in any case it is a hypothesis), is a version of idealism. That is, we picture reality as a part of the Platonic world of abstract ideas. Observations are then nothing else but specific derivations, through "glasses", from this view. I think personally that this is the most fruitful working frame for a physicist - always keeping in mind that it is entirely hypothetical.
So what should now be taken as this "reality" in this hypothesis ? Well, nothing else but the model which gave us the logically consistent agreements with observation we started with.

Within this frame of thinking, it is of course totally unsound to declare "certain elements of the model agreeing with elements of reality", because we declared, by hypothesis, our model to BE reality. I often called that the "toy world of the theory". It is in this mindset, btw, that I defend MWI as the "correct" interpretation of quantum theory - simply because it assigns reality to the elements of the model (= quantum formalism). Within this frame of thinking, also, MWI is "locally realistic", simply because the dynamical prescriptions have some form of locality to them, for a suitable definition of locality. As an idealism, it is true that MWI is rather remote from any naive realism, which invites all the agressivity towards it by people who assume implicitly naive realism. But this comes about because the initial "hypothesis of reality" was totally different in both cases. So the entire conflict comes about when this idealism is confused with (implicitly assumed) naive realism. It is in this context that the "naive" in naive realism becomes pejorative: its proponents don't realize they already made an implicit hypothesis about what is "real" (while acknowledging also that one can't know this for sure at the same time).

And as we all seem to agree upon the fact that we'll never know what is "true", this discussion can go on for ages without ever finding any resolution.

 

[Careful]

** In as much as (a) is a very profound statement, you should apply it entirely logically until the end. I agree with (a), with maybe one small modification: I'm not even sure we can say that "truth doesn't exist", only, that "truth isn't accessible", and we leave in the middle whether truth exists or not. We simply know with certainty that we won't know it with certainty. The realisation of this philosophically elementary fact does - IMO - a lot of good to any of these sometimes heated discussions **

Knitpicker I think there exists a reality (we need some form of reality when we make theories), but that indeed we will never know that the ansatz we make about it is true or not. In a pragmatic sense the truth does not exist, since we will never know that we are right even when we are.

**
As such, I don't agree at all with (b), which I find a totally arbitrary requirement. The only goal of the model is to make a model of observational data. What elements are used to set up such a model is entirely free, I'd say. Of course there can be *preferences* and Occam's rasor is a good guide: two models which are empirically equivalent, but one which has some extra elements which the other doesn't need, might be preferentially rejected in favor of the "simpler" one. But again, this is just a guiding principle, not a strict requirement. **

So I guess you have to live with the problem of my hundred eyes in Japan and China as I mentioned in a previous post on this thread. Or you have to anwer my previous objection about how your belief does not contradict the practise of science: I am not speaking about the freedom we have in cooking up the model, but how the model deals with our ability to produce it.

** Within this frame of thinking, it is of course totally unsound to declare "certain elements of the model agreeing with elements of reality", because we declared, by hypothesis, our model to BE reality. ***

So, how would you discover the electron or anything which is NOT in your model ? I mean, you would have to avoid the argument, the electron (positron) was discovered because we observed its track through the nebula.

Careful

 

[vanesch]

So I guess you have to live with the problem of my hundred eyes in Japan and China as I mentioned in a previous post on this thread. Or you have to anwer my previous objection about how your belief does not contradict the practise of science: I am not speaking about the freedom we have in cooking up the model, but how the model deals with our ability to produce it.

I agree with your objection about the eyes in China (and for our Chinese membership: the eyes in New York ), and the fundamental objection I also have with equating *your* observations with reality while denying any "reality out there" by hypothesis. So I go for your second option: the model should indeed be such that within it, an observer should be able to reproduce it (as he obviously was trying to do).
I put that on the slide of "self-consistency" of the model, and "agreement with observation". I think the entire endeavour of MWI people is to try to do exactly that: find A (even if it is totally "bezerk") self-consistent way of deducing how an entity within the model would arrive at those observations and conclusions, as one actually did (in this case, write down the rules of the quantum formalism).

I agree that this is one of the stronger arguments for naive realism: the relationship between observation and reality being by definition (close to) 1-1, this step is evident, while it isn't in an idealist view. Nevertheless, the opposite is, IMO, not *necessary* (even though - I admit - attractive): it is sufficient to have such a self-consistent view (no matter how bezerk), for the idealist, to be satisfied.

So, how would you discover the electron or anything which is NOT in your model ? I mean, you would have to avoid the argument, the electron (positron) was observed because we discovered its track through the nebula.

Elementary, dear Watson

From an idealist view, you cannot "discover" an electron. You can only make observations, and then ponder about it, and come up with a MODEL (as you said yourself), which might then predict said observations.
The model is then "a charged particle which produces tracks in a cloud chamber and which I decide to call 'positron' ".
Once this model gets accepted/integrated/gets other observational successes, in the idealist viewpoint, it then becomes (by convention/hypothesis/majority vote...) an element of reality.

It is the success of the "model of the electron" which turns it into something we call "real".

So, in short, you cannot discover anything. You can only make observations, and think up models of those observations. This successful tandem is then the "discovery" of the entity of your new model, if it turns out to be successful/useful/socially accepted. At least according to the idealist viewpoint.

But of course the closer an observation is to our biological senses, the more intuitive this model becomes ("that's my mom") and from a certain point on, our sense of reality for certain models becomes so overwhelming that questioning this "everyday" reality triggers unfriendly statements about the mental health of the one making those questioning statements . In other words, for everyday observations, we strongly intuitively tend to take on the naive realism viewpoint. The question is whether this intuition is really so fundamental. It is the fundamental objection to an idealist hypothesis.

 

[Careful]

**
Elementary, dear Watson **

Not quite

**
From an idealist view, you cannot "discover" an electron. You can only make observations, and then ponder about it, and come up with a MODEL (as you said yourself), which might then predict said observations. **

Hmmm, I always had the strange idea that from time to time we discover something in our life . But when constructing such model, you always use classical reasoning, you assume something was there which did this and that to the nebula which you do observe now. Whether you have acces to this reality or not is some other matter, and what you could measure would probably be some emergent phenomenon and not the ``real'' stuff (even within the theory you construct). Even in quantum mechanics, you can easily uphold the view that at any time the electron is somewhere on the support of its wave function (which would be the track in the nebula), but you simply don't know where.

**But of course the closer an observation is to our biological senses, the more intuitive this model becomes ("that's my mom") and from a certain point on, our sense of reality for certain models becomes so overwhelming that questioning this "everyday" reality triggers unfriendly statements about the mental health of the one making those questioning statements . In other words, for everyday observations, we strongly intuitively tend to take on the naive realism viewpoint. The question is whether this intuition is really so fundamental. It is the fundamental objection to an idealist hypothesis. ***

Ok, so let me ask you this : if quantum theory could be explained in an entirely ``realistic'' way (which it can, by holography (seems very plausibe), the future influencing the present (requires much more work) and so on...) why uphold this ``idealist'' view ? It is not that these two models are in accordance with my intuition, nobody likes to accept a shadowworld beyond ones perceptions, but at least one is able to clearly define it, make the math much easier (a local model in a slightly more complex shadow world is still much more economic that the configuration space methods in QM) and well defined ! I mean, strictly speaking, I cannot close the door for your viewpoint (and basically, I don't care ) but it is a very difficult and unnatural position to maintain as I see it.

Careful

 

[vanesch]

But when constructing such model, you always use classical reasoning, you assume something was there which did this and that to the nebula which you do observe now. Whether you have acces to this reality or not is some other matter, and what you could measure would probably be some emergent phenomenon and not the ``real'' stuff (even within the theory you construct). Even in quantum mechanics, you can easily uphold the view that at any time the electron is somewhere on the support of its wave function (which would be the track in the nebula), but you simply don't know where.

Well, a quantum theory = a "classical" theory + superposition principle, or in other words, what we call "classical" is a basis. If we now have some arguments which make us understand (or make us hope one day we might know) why our *observations* often agree with basis-states (in other words, are classically-looking), then this "classical reasoning" is not so wrong at all. When you do that, you are implicitly reasoning in terms of basis functions, and not in terms of the entire state space, but that doesn't matter. When you say "the electron is somewhere", then this is a shortcut for "I use as a basis, states which I call 'the electron is there', and I will probably observe also states close to 'the electron is here' ", but of course the superposition principle mixes into this which first makes superpositions of the different states, and then also is supposed to explain why my observations are again limited to the one of those states.

So "classical reasoning" is entirely allowed for. It just changed meaning, and became "reasoning in a basis related to observation".

Ok, so let me ask you this : if quantum theory could be explained in an entirely ``realistic'' way (which it can, by holography (seems very plausibe), the future influencing the present (requires much more work) and so on...) why uphold this ``idealist'' view ?

Because it is just another idealist view ! It is not part of naive realism to say that the future determines the present, not any more than parallel worlds are. Simply because our experienced "flow of causality" is just as well part of our observational sense of reality as is anything else, and as such, cannot consider "the future determines the past". I'd say it is just a matter of taste to say what you like more.

Now, in as much as this is technically possible (isn't this the "transactional interpretation of QM" in some way, although I thought it had serious problems for multi-particle correlations - so I'm not well versed in all this), it is just a model amongst others. You should know very well that I'm very favorable of making alternative models.

It is not that these two models are in accordance with my intuition, nobody likes to accept a shadowworld beyond ones perceptions, but at least one is able to clearly define it, make the math much easier (a local model in a slightly more complex shadow world is still much more economic that the configuration space methods in QM) and well defined ! I mean, strictly speaking, I cannot close the door for your viewpoint (and basically, I don't care ) but it is a very difficult and unnatural position to maintain as I see it.

Well, to me they are certainly on the same levels of "bezerkness" and I don't think there's a reason to prefer one over the other, unless one formalism is mathematically much more appealing than another. If some equivalence is established, then it is entirely a matter of taste (and probably of social convention). I think it is then very interesting to look at both models. If equivalence is not established, then this is also a very interesting situation of competing models.

But all this doesn't do away with the fact that the idealist interpretation of the standard quantum formalism still smells more like MWI, and that this other formulation will have its own, different idealist interpretation, given that it will be formally different (though possibly equivalent).

Again, given that this is all matter of hypothesis, I don't see why one should fight over it, or adhere to one over the other. The more different views one is aware of, the more understanding one usually gains of the matter at hand.

This is why I don't fight (intelligent) local realists, nor do I "adhere" to any such vision. It's just interesting to know about, that's all.

 

[Careful]

** Well, a quantum theory = a "classical" theory + superposition principle, or in other words, what we call "classical" is a basis. **

That remains to be seen, as soon as one includes self interactions, it occurs to me that the superposition principle *might* go down the drain if one insists upon the theory to be well defined. Check out one of the latest papers of Thomas Elze on this (if you want to, I will look it up) where you have only a weak form of the superposition principle.

** If we now have some arguments which make us understand (or make us hope one day we might know) why our *observations* often agree with basis-states (in other words, are classically-looking), then this "classical reasoning" is not so wrong at all. When you do that, you are implicitly reasoning in terms of basis functions, and not in terms of the entire state space, but that doesn't matter. When you say "the electron is somewhere", then this is a shortcut for "I use as a basis, states which I call 'the electron is there', and I will probably observe also states close to 'the electron is here' ", but of course the superposition principle mixes into this which first makes superpositions of the different states, and then also is supposed to explain why my observations are again limited to the one of those states. ***

Nonlinearity is the key word I presume, a *dynamical* law of observation should break the superposition principle.

** Because it is just another idealist view ! It is not part of naive realism to say that the future determines the present, not any more than parallel worlds are. **

Well, not quite , first of all observations are still REAL in such world, if Careful sees a dead cat, Vanesch will too, I have no synchronisation difficulties. I have already long time said that if you want to predict the ideal quantum correlations using a local theory, then you will have to redefine what is real. But still, what is real, does not depend upon my perception, on the contrary, the reality will explain perception itself.


**Simply because our experienced "flow of causality" is just as well part of our observational sense of reality as is anything else, and as such, cannot consider "the future determines the past". I'd say it is just a matter of taste to say what you like more. **

Well, I would think the experience of the flow of causality to remain the same for heavy stuff (see my response on the other thread).

** You should know very well that I'm very favorable of making alternative models. **

Of course, this is just a friendly conversation.


**Well, to me they are certainly on the same levels of "bezerkness" and I don't think there's a reason to prefer one over the other, unless one formalism is mathematically much more appealing than another. If some equivalence is established, then it is entirely a matter of taste (and probably of social convention). I think it is then very interesting to look at both models. If equivalence is not established, then this is also a very interesting situation of competing models.**

Well you probably know this paper, but wait ... perhaps not (it contains some important arguments) : ``Simulating physics with computers´´, Richard.P. Feynman, International Journal of Theoretical Physics, Vol 21, Nos 6/7 1982

** Again, given that this is all matter of hypothesis, I don't see why one should fight over it, or adhere to one over the other. The more different views one is aware of, the more understanding one usually gains of the matter at hand. **

Indeed, no fight is necessary !

Careful

 

[setAI]

the problem is that a century later- even smart people have trouble accepting the implications of unitary QM- that is the less that an element of a system is observed- that is causally connected to an observer- the more it is in a superpostional state of ALL possibilities-and this does extend to all scales- it is just too complicated and EXPENSIVE at present to measure those affects with mezzo scale and larger systems- what we call reality is a sum-over-histories of what we are interacting with NOW yet the complexity of our interaction with the environment severly probabilistically limits the possibilites to what appears to be a singular reality and a single past- except at the edges of shadows- as David Deutsch put it

 

[selfAdjoint]

That remains to be seen, as soon as one includes self interactions, it occurs to me that the superposition principle *might* go down the drain if one insists upon the theory to be well defined. Check out one of the latest papers of Thomas Elze on this (if you want to, I will look it up) where you have only a weak form of the superposition principle.

Have you loked into the behavior of solitons? They are a characteristic behavior of nonlinear equations; e.g. it has long been known that the "inverse problem" of identifying a potential in a one-dimensional Schroedinger equation from its scattering behavior leads to Kottweg-DeVries (KdV) nonlinear ODE with soliton solutions.

And they obey a kind of quasi-superposition behavior, so you could have superposition where it counts without assuming it to be some kind of universal ordering principle.

 

[Careful]

** Have you loked into the behavior of solitons? **

Yes

** They are a characteristic behavior of nonlinear equations; e.g. it has long been known that the "inverse problem" of identifying a potential in a one-dimensional Schroedinger equation from its scattering behavior leads to Kottweg-DeVries (KdV) nonlinear ODE with soliton solutions. **

Right


**
And they obey a kind of quasi-superposition behavior, so you could have superposition where it counts without assuming it to be some kind of universal ordering principle. ***


Correct, if you have two bubbles with more or less disjoint support (that is A is very weak where B is very strong), then this automatically leads to that. But the key message here is that the AND automatically becomes an OR, there is almost no interference between the different solitons (so for example a superpostion of two disjoint soliton beams can be seen as just two different beams). In Barut self field for example one can take soliton like solutions for the single events, they behave like particles but still contain the interference information due to the solution of the linear equation - so there is no measurement problem here (actually this only works for fermions). In GRW for example, the unitarity is broken due to the introduction of stochastic noise. When I speak about a non linear extension of the Schrödinger equation, I do not mean to intend that the statistics of the particles (obviously such a psi has no good probability interpretation) is not going to satisfy a unitary evolution law (something which is a very delicate issue, vis a vis black holes), but that the hidden theory which includes single events is necessarily non linear and much more involved (remember, Vanesch and I were talking about the measurement problem) - there can be information loss at the classical level though. The quantization of gauge fields is a very different issue (and might not be necessary at all).

Careful

 

[Careful]

**the problem is that a century later- even smart people have trouble accepting the implications of unitary QM- that is the less that an element of a system is observed- that is causally connected to an observer- the more it is in a superpostional state of ALL possibilities-and this does extend to all scales- it is just too complicated and EXPENSIVE at present to measure those affects with mezzo scale and larger systems- what we call reality is a sum-over-histories of what we are interacting with NOW yet the complexity of our interaction with the environment severly probabilistically limits the possibilites to what appears to be a singular reality and a single past- except at the edges of shadows- as David Deutsch put it ***

Yes and it was for sure problematic that one Einstein could not accept absolute thinking after 250 years of Newtonian bucket dominance. Moroever, I do not have problems accepting the Schrödinger wave at all, I reject the measurement postulate though.

Careful

 

[vanesch]

** Well, a quantum theory = a "classical" theory + superposition principle, or in other words, what we call "classical" is a basis. **

That remains to be seen, as soon as one includes self interactions, it occurs to me that the superposition principle *might* go down the drain if one insists upon the theory to be well defined.

Ok, but then it is not "quantum theory" but another theory. In quantum theory, as far as I know, the strict superposition principle is postulated to hold. If you change that, you have another theory, which may be superior, and we'll then refer to quantum theory as its linear predecessor. If strict linearity leads to inconsistencies, then that only means that quantum theory is inconsistent - but it might be that it is only the quantum field model that might suffer from that, not quantum theory in all generality (for instance with just a finite number of degrees of freedom).

Check out one of the latest papers of Thomas Elze on this (if you want to, I will look it up) where you have only a weak form of the superposition principle.

Pffft... guess I have to say "thank you" for the reference. If I would read all the stuff that is suggested to me, I would spend even less time doing useful stuff

Nonlinearity is the key word I presume, a *dynamical* law of observation should break the superposition principle.

But then it is, strictly speaking, not a standard quantum theory where this superposition is to be strictly valid, no ? But rather a successor.

Well, not quite , first of all observations are still REAL in such world, if Careful sees a dead cat, Vanesch will too, I have no synchronisation difficulties.
I have already long time said that if you want to predict the ideal quantum correlations using a local theory, then you will have to redefine what is real. But still, what is real, does not depend upon my perception, on the contrary, the reality will explain perception itself.

Yes, granted. But so does "a bag of events" (which you sometimes call superdeterminism). We're back to 'correlations happen', and now they come in from the future instead of from some specific initial condition in the past.

What you probably want to say, is that in such a situation, you do not have to delve into the very painful question of exactly what constitutes "an observer", because there are still perfectly well defined events (which do not exist independent of a "branch choice" in an MWI view). Granted. You're relieved from considering this question for the definition of "events". The question is if that's a necessary requirement. Again, it can be advantageous not to have to delve into this. But as, philosophically speaking, sooner or later, you have to confront the question of what exactly is a subjective experience and how it is related to the physical world/model/theory (the famous psycho-physical parallellism von Neumann already realized was the crux in understanding QM), you only bought yourself some time by requiring a physical theory to be clear of these considerations.

So I can understand the desire to uncouple this question from "physics per se", but in my opinion, it is not more than that: something that is to be desired. It's not a hard requirement.

All the ponderings about QM opened this Pandora's box of bringing in conscious/subjective experience into a full physical theory, and once you consider it, it is hard to find a fundamental reason to put it back in the box and reject the idea on some fundamental ground. Sure, things would be cleaner if this messy problem were uncoupled from "physics", but I don't see why we can require this as an absolute requirement, especially because it doesn't fundamentally solve anything about the question of what *is* a subjective experience. It only allows you to ignore the question, not to answer it.