Are quantum fields real objects in space?

[Demystifier]

Sure, although it'll take some time, until I can have a close look at it, because the semester started, and I've to give two lectures to my teachers students (including one about quantum mechanics ;-)).

I have sent you the draft by e-mail.

 

[martinbn]

See refs. [5,6,7] in my http://de.arxiv.org/abs/1112.2034 .

Ok, I looked through them quickly, it is possible that I missed it, but I found nothing that supports your (and atyy) statement.

 

[Demystifier]

Ok, I looked through them quickly, it is possible that I missed it, but I found nothing that supports your (and atyy) statement.

Let us take, for instance, some quotes from "Message of the quantum" by Zeilinger:

But I think that the concept of reality itself is at stake ...
... it is not always possible to assign definite measurement outcomes, independently of and prior to the selection of specific measurement apparatus in the specific experiment. ...
Rather, the assumption that a particle possesses both position and momentum, before the measurement is made, is wrong. Our choice of measurement apparatus decides which of these quantities can become reality in the experiment.
... the distinction between reality and our knowledge of reality, between reality and information, cannot be made. There is no way to refer to reality without using the information we have about it. ...
Maybe this suggests that reality and information are two sides of the same coin, that they are in a deep sense indistinguishable. If that is true, then what can be said in a given situation must, in some way, define, or at least put serious limitations on what can exist. ...

Now you may say that Zeilinger does not say explicitly that things do not exist until measured, but I think that he says it implicitly. Or at least, I do not see any other coherent interpretation of his words compatible with all those sentences. If you see another coherent interpretation of his words, I would like if you could spell it out in a concise and unambiguous form.

Or take e.g. the Bell theorem, which is often presented as a statement that QM is incompatible with local reality. So to save QM and locality, one must give up reality. How would you interpret "giving up reality" if not by saying that things do not exist without observations?

 

[martinbn]

Let us take, for instance, some quotes from "Message of the quantum" by Zeilinger:

But I think that the concept of reality itself is at stake ...
... it is not always possible to assign definite measurement outcomes, independently of and prior to the selection of specific measurement apparatus in the specific experiment. ...
Rather, the assumption that a particle possesses both position and momentum, before the measurement is made, is wrong. Our choice of measurement apparatus decides which of these quantities can become reality in the experiment.
... the distinction between reality and our knowledge of reality, between reality and information, cannot be made. There is no way to refer to reality without using the information we have about it. ...
Maybe this suggests that reality and information are two sides of the same coin, that they are in a deep sense indistinguishable. If that is true, then what can be said in a given situation must, in some way, define, or at least put serious limitations on what can exist. ...

Now you may say that Zeilinger does not say explicitly that things do not exist until measured, but I think that he says it implicitly. Or at least, I do not see any other coherent interpretation of his words compatible with all those sentences. If you see another coherent interpretation of his words, I would like if you could spell it out in a concise and unambiguous form.

But he doesn't say anything that implies that the particle doesn't exists. He only says that the values of some observables cannot exists, which is very different.

 

[A. Neumaier]

If you see another coherent interpretation of his words, I would like if you could spell it out in a concise and unambiguous form.

He says:

There is no way to refer to reality without using the information we have about it. ...

This is quite unambiguous, and is the core of his statement. The subsequent conclusion begins with maybe, hence is not a claim made.

His statement is true of anything we refer to in place of reality, so of knowledge and information itself. So if we do not give the status of reality to a particle because we do not have precise information about its position and momentum, we can neither give the status of reality to knowledge because we do not have precise information about knowledge - it is conceptually more vague than a particle.

 

[Fra]

Are they things that exist in space or are they just mathematical abstractions that help use calculate things?

Is there an actual difference between the options?

There is no way to "see" reality except through a physical observer.

The only difference lies in how you may a build a theory, to relate observers. By assuming that there is a structure somewhere, that is not subject to the constraints of having to be inferred, things get easier of course. But there is no empirical or instrumental justification for that beyond that it seems like the least complicated thing to try, rather than making everything fluid at once.

/Fredrik

 

[Demystifier]

But he doesn't say anything that implies that the particle doesn't exists. He only says that the values of some observables cannot exists, which is very different.

But it's not clear what it means. What exactly are the quantities (or qualities) of the particle that do exist without measurement?

 

[Demystifier]

His statement is true of anything we refer to in place of reality, so of knowledge and information itself. So if we do not give the status of reality to a particle because we do not have precise information about its position and momentum, we can neither give the status of reality to knowledge because we do not have precise information about knowledge - it is conceptually more vague than a particle.

Is it a critique or a defense of the Zeilinger claims?

 

[A. Neumaier]

What exactly are the quantities (or qualities) that do exist without measurement?

Approximate position and approximate momentum. Just like with every classical object in the world.

If we don't know approximately where a particle is, it may not be in the lab - so how can we do an experiment with it?
And if we do not know approximately its momentum, how can we tell whether it will move along a ray where it is supposed
to move, according to a planned experiment?

Thus approximate position and approximate momentum must exist before measurement, or all of our experiments do not make sense.

Interpretational problems appear only if we want to pretend that values exist to an arbitrary precision.

 

[A. Neumaier]

Is it a critique or a defense of the Zeilinger claims?

Both.

 

[Fra]

So if we do not give the status of reality to a particle because we do not have precise information about its position and momentum, we can neither give the status of reality to knowledge because we do not have precise information about knowledge - it is conceptually more vague than a particle.

They way i see this from an inference perpective: I see one distinction here, the information an observer has about something, can not be rated by himself. It's just the defining part in a conditional part of subjective probability.

This is the conceptual trick i personally use to justify statistics, in cases where its obvious that the empirical basis for it does not exist. The statistical basis is rather than possibly only the information the observer has (wether its right or wrong as per somone else is not relevant). In this sense the illusion is always more "first hand" than the real thing. But this does not imply that we have agreement among observer. So if real i supposed to mean "all observers agree" we just transformed the problem into asking - how come the subjective inferences from different interacting observers, tend to magically agree, at least in the case of classical observers?

/Fredrik

 

[A. Neumaier]

to justify statistics, in cases where its obvious that the empirical basis for it does not exist.

In these cases it cannot be justified.

if real is supposed to mean "all observers agree"

I don't think it is supposed to mean that. Real means independent of observers and observations, no matter whether or not they agree. One needs the latter only to check what precisely is real!

 

[Peter Morgan]

There is no way to refer to reality without using the information we have about it. ...

The information we have recorded about an experiment has to be unambiguous and uncontentious, with a trail that an experimenter could follow and reproduce, otherwise Nature or Science will retract the article that depends on the raw data record. We can't get away with saying that our raw data hasn't been reviewed by Wigner's friend yet, so the editors and referees can't look at it.
The trail in a characteristically QM experiment will be from a thermodynamic transition of an engineered device embedded in the apparatus and driven by its surrounding environment, to a change of voltage on a signal line driven by that engineered device, to hardware that identifies that there has been a change of voltage definite enough to qualify as being called a measurement event, then the time of the event has to be obtained from a piece of hardware that has been engineered to act as a clock (characterized as unaffected by its surrounding environment, as far as possible, in contrast to the responsiveness of the measurement hardware), and finally a record will be made in computer memory (used to be, the record would be in a lab book, of the movement of a pointer, but modern data rates make that impossible). Note that the computer record could be less compressed: we could have stored the signal line voltage as a 12-bit number after analog-digital conversion, every nanosecond or every picosecond, allowing significantly more sophisticated post-analysis, but that would increase the data rate too much.
All of the records stored in computer memory are classical raw data, but a process of careful selection from that data and the computation of statistics for the various curated ensembles might allow us to demonstrate that, for example, Bell-CHSH inequalities are violated. This has usually been said to be because the classical raw data was not caused by a classical process, however the violation of some inequality or another as much depends on what selection and computation of statistics is done; certainly the computation from the ensemble is modeled by the operator we use (if we square values that we use an operator $\hat A$ to model, we use $\hat A^2$ to model the different computation); it seems that whether or not the raw data is caused by a nonclassical process, the choice of one operator or another must also model the different possible selection processes from the raw data that allow the computation of statistics.
I see all the above as agreeing with Arnold's #82 and inserting more detail, but perhaps he will disagree.

 

[martinbn]

But it's not clear what it means. What exactly are the quantities (or qualities) of the particle that do exist without measurement?

Well, if the particle doesn't exist, then what do you measure? So, there is no particle, I guess it is vacuum. Then you measure the spin(of what, the vacuum?) and pop the particle exists, but only for a moment, only when measure, then it doesn't exist until the next measurement. That seems a very strange way of describing the situation. And I don't see that, nor anything that would suggest it, in the papers you cited.

 

[Peter Morgan]

Well, if the particle doesn't exist, then what do you measure? So, there is no particle, I guess it is vacuum. Then you measure the spin(of what, the vacuum?) and pop the particle exists, but only for a moment, only when measure, then it doesn't exist until the next measurement. That seems a very strange way of describing the situation. And I don't see that, nor anything that would suggest it, in the papers you cited.

Suppose there are no particles, but the event devices we set up cause identifiable changes of voltage, pops, on signal lines every so often. In a dark room, there is still the "dark rate", but when we turn on the power to a source device in the room the pops happen at a different rate, indeed perhaps with quite different statistics. Suppose we have two event devices in the room, now we can ask about correlations on the signal lines and about whether identifiable pops happen at the same time, and so on.
Now, we know that how many source devices there are and where the source devices and the event devices are in the room changes the event statistics, so we can think of the sources as causing the pop statistics. If we want to say more than that, we have to introduce some new idea: we know that thinking that particles cause the pops is only sustainable if we're willing to adopt something like a de Broglie-Bohm interpretation (which we might or might not be willing to do, or perhaps only sometimes), so we ask whether we might think of the events being caused by a field in the immediate surroundings of each event device. Now there's a lot to do, but the distance between an appropriate random field and a quantum field can be shown to be not very great, so that we can think of QFT as just one way (a useful way) of doing signal analysis in the presence of noise sophisticatedly. Classical noise is correlated across time and space, and we effectively engineer those correlations when we introduce elaborate sources of the kind that cause violations of Bell inequalities, et cetera. Furthermore, noncommutative operators are commonplace in classical signal analysis, so the violation of Bell-CHSH-type inequalities is not a surprise.
I don't expect many people will read this far, and likely no-one will feel compelled by the discussion above, but I have found it very helpful to think that events pop, not particles. Events are correlated because the vacuum is already correlated and because we further engineer events to be correlated in a variety of ways. FWIW, a slogan I've been playing with of late has been to speak of field/event duality instead of wave/particle duality, though that seems sometimes too trite and sometimes just right.
Of course this leaves questions as to why the source and event devices exist as distinguishable objects with trajectories, but for me it's one thing at a time, walk before you try to run, ...

 

[Fra]

In these cases it cannot be justified.

In my understanding it can likely be justified, but it depends on modifying the theory of course, so its not just about interpretations (and its an open question).

The justification is to consider let's say a bayesian kind o observer depedent frequentist intepretation. This is what you get when all observer can not agree on pointer states and classical counters. The event counters themselves are necessarily of the nature that can not be copied. The observer "consumes" the information. Ie. the "evidence" that makes up the probabilistic foundation is observer dependent.

The exception is where we have a classical observer, but this must be a special case of something more general.

I don't think it is supposed to mean that. Real means independent of observers and observations, no matter whether or not they agree. One needs the latter only to check what precisely is real!

Its how i think of it. I can't see another meaniful definition?

To just by real mean independent of observers of observations honestly make no sense to me, because how are you to infer or verify something that is detached from observation? Like the colour of gods underwear? Such a thing would have no implications for experiments, because it would contradict the fact that it was assume to be independent of anything we can observe.

/Fredrik

 

[Fra]

Approximate position and approximate momentum. Just like with every classical object in the world.

If we don't know approximately where a particle is, it may not be in the lab - so how can we do an experiment with it?
And if we do not know approximately its momentum, how can we tell whether it will move along a ray where it is supposed
to move, according to a planned experiment?

Thus approximate position and approximate momentum must exist before measurement, or all of our experiments do not make sense.

Interpretational problems appear only if we want to pretend that values exist to an arbitrary precision.

hmm i suspect this does related to your termal interpretation? But do i understand this to be a FAPP type of interpretation? or does this interpretation supopse to be valid a the future QG theory? (just checking so i don't attempt to overinterpret this)

An approximate value sound like nothing else than the observer expectation. And this in my view exists encoded in the observers microstructure. Ie. its part of the MAP. So do you agree that what exists is an uncertain (~approximate) map, and can be identify this map with the observer? But this "map" is then effectivelt the same thing as an observers "information" about something. And the "approximation" comes only from the fact that the information is not complete and uncertain.

OTOH, if you remove the observer altogether, i don't understand whose average or expectation you refer to as real?

/Fredrik

 

[bhobba]

Can you give a reference for one that is not.

Copenhagen has evolved and changed since Bohr and Einsteins time:
https://arxiv.org/pdf/1511.01069.pdf

These days modern Copenhagenists seem to be switching to Decoherent Histories they call Copenhagen done right just as Feynman did at the end. Gell-Mann thinks it's basically MW without the many worlds ie just one world.

That seems to be the emerging consensus interpretation - but there are many others about - Copenhagen is still the most generally held view according to Sean Carrol:
http://www.preposterousuniverse.com/blog/2013/01/17/the-most-embarrassing-graph-in-modern-physics/

Thanks
Bill

 

[bhobba]

I'm not aware of a variant that says the objects don't exist before measurement, most posit "something" is there.

Trouble is in doing that you have to invoke some other process other than Schrodinger Evolution or you run into Kochen-Specker. My interpretation Ignorance Ensemble holds that as the central issue - how does a mixed state become a proper mixed state. Copenhagen ignores it completely. Decoherent Histories simply redirects the issue to histories ie sequences of projection operators.

Thanks
Bill

 

[DarMM]

Trouble is in doing that you have to invoke some other process other than Schrodinger Evolution or you run into Kochen-Specker. My interpretation Ignorance Ensemble holds that as the central issue - how does a mixed state become a proper mixed state. Copenhagen ignores it completely. Decoherent Histories simply redirects the issue to histories ie sequences of projection operators.

Thanks
Bill

I could be completely wrong here, but does one need a "process" as such. In antirealist one views $\psi$ as simply encoding probabilities to see various values of observables and since there is no $\Lambda$ and thus no assumption of no pre-existent values associated with the system this would make the Kochen-Specker theorem irrelevant I think.

My interpretation Ignorance Ensemble holds that as the central issue - how does a mixed state become a proper mixed state. Copenhagen ignores it completely. Decoherent Histories simply redirects the issue to histories ie sequences of projection operators.

(I know you know most of the below, just laying it out to hear your thoughts)

I think this is the central issue in all antirealist (a term I dislike, better would be "no system variables" or similar) interpretations. QBism, Neo-Copenhagen, Ignorance Ensemble, etc basically all say that when decoherence happens one has a value you don't know.

Another way of phrasing it is, when decoherence occurs we obtain usual (Kolmolgorov) probability theory, which we know to interpret as ignorance. The issue is how to interpret the (non-commutative) probabilities prior to decoherence. QBism for example says that prior to measurement the "fact" of the value isn't created yet and QM is a probability calculus for yet-to-exist quantities. However once the value has been created you get the proper mixed state as you simply don't know the value now.

I think you might find the views of Richard Healey interesting. He basically views decoherence as measuring when classical concepts make sense and one then has the ignorance interpretation once it has occured.

Again I think it comes back to what is the meaning of non-decohered states in QM and why is the probability calculus noncommutative.

 

[martinbn]

Copenhagen has evolved and changed since Bohr and Einsteins time:
https://arxiv.org/pdf/1511.01069.pdf

These days modern Copenhagenists seem to be switching to Decoherent Histories they call Copenhagen done right just as Feynman did at the end. Gell-Mann thinks it's basically MW without the many worlds ie just one world.

That seems to be the emerging consensus interpretation - but there are many others about - Copenhagen is still the most generally held view according to Sean Carrol:
http://www.preposterousuniverse.com/blog/2013/01/17/the-most-embarrassing-graph-in-modern-physics/

Thanks
Bill

Ok, there are versions of the interpretation. My question is according to which version the fields are not real objects in space. That was the statement that @atyy made.

 

[DarMM]

Ok, there are versions of the interpretation. My question is according to which version the fields are not real objects in space. That was the statement that @atyy made.

@atyy knows better, but some versions of Copenhagen like that of Rudolf Haag would just posit that an observer working in spacetime region $\mathcal{O}$ can measure quantities associated with the C*-algebra $\mathcal{A}(\mathcal{O})$ and that the C*-algebra has a basis consisting of observables of the form $e^{i\phi(f)}$.

This would just see observables as being formed from fields, i.e. one has a map $f: \phi \rightarrow \mathcal{A}(\mathcal{O})$, but without having fields as real objects in the theory.

 

[Demystifier]

Well, if the particle doesn't exist, then what do you measure? So, there is no particle, I guess it is vacuum. Then you measure the spin(of what, the vacuum?) and pop the particle exists, but only for a moment, only when measure, then it doesn't exist until the next measurement. That seems a very strange way of describing the situation. And I don't see that, nor anything that would suggest it, in the papers you cited.

I don't think you answered my question.

 

[Demystifier]

Thus approximate position and approximate momentum must exist before measurement, or all of our experiments do not make sense.

I disagree. For instance, beauty does not exist until one observes it (beauty is in the eyes of the beholder), yet it doesn't mean that observation of beauty does not make sense. Of course, the shape of the beautiful object exists before the observation, but the shape by itself is not beautiful.

Another, more physical example is the color. The EM wave has its wavelength even without observation, but it has a color only when someone observes it.

 

[martinbn]

@atyy knows better, but some versions of Copenhagen like that of Rudolf Haag would just posit that an observer working in spacetime region O\mathcal{O} can measure quantities associated with the C*-algebra A(O)\mathcal{A}(\mathcal{O}) and that the C*-algebra has a basis consisting of observables of the form eiϕ(f)e^{i\phi(f)}.

This would just see observables as being formed from fields, i.e. one has a map f:ϕ→A(O)f: \phi \rightarrow \mathcal{A}(\mathcal{O}), but without having fields as real objects in the theory.

How is this related to whether the fields are real objects in space or not?

I don't think you answered my question.

Do you mean this question?

What exactly are the quantities (or qualities) of the particle that do exist without measurement?

What do you mean by quantities and qualities? It exists and its state can be fully described by a vector in a certain vector space.

 

[martinbn]

But you didn't answer my question. If the particle doesn't exist, does it mean that we have vacuum i.e. empty space?

 

[Demystifier]

It exists and its state can be fully described by a vector in a certain vector space.

OK, now you answered my question so I can proceed. Essentially, you identify the particle with the corresponding state in the Hilbert space. The particle is the state in the Hilbert space. Fine, now let us see what are the consequences of this statement.

Consider a state which before the measurement is a Gaussian wave function with a very large width $\Delta x$ in the position space. So far so good. But now assume that we perform the measurement of the particle position. This means that the wave function changes to a new wave function with a new width $\delta x$ ,where $\delta x\ll \Delta x$. This change is called the wave function collapse. But the collapse is non-local, it happens faster than light. And that's the problem.

A way out of this problem is to say that collapse is not a real physical event, but only the update of our knowledge. However, you are not allowed to use that argument, because you essentially said that the wave function is the particle. This means that the collapse is a real physical event, and not only an update of knowledge.

Do you agree with my sequence of arguments? Do you accept that measurement involves a nonlocal collapse as a real physical event? Do you find such kind of nonlocality problematic?

 

[DarMM]

How is this related to whether the fields are real objects in space or not?

I don't understand, if the fields are not taken as really existing, but just taken as mathematical method for generating basis elements for an observable algebra, then they're not real objects in space right? It's directly related to it by saying they're not real. I genuinely don't understand, it seems fairly clearly related.

 

[DarMM]

I disagree. For instance, beauty does not exist until one observes it

Does this not ignore those who are objectively handsome/beautiful such as us Advisors?

 

[Demystifier]

But you didn't answer my question. If the particle doesn't exist, does it mean that we have vacuum i.e. empty space?

If you think that it is not easy to make sense of interpretation in which particle does not exist until measured, I perfectly agree with you. One needs to work hard to make sense of it. As a result of such a hard work, I wrote the paper http://de.arxiv.org/abs/1112.2034

 

[bhobba]

Does this not ignore those who are objectively handsome/beautiful such as us Advisors?

Maybe that's why I became a mentor - my mirror tells me different every morning - my shaver even started to shut down when I used it. Disconcerting actually - but I eventually sorted the shaver out - took it to the place I bought it from who puled it to pieces - even my whiskers are against me - they clogged it up despite me cleaning it every day - it turns out you have to pull it to pieces and get rid of bits in the things that spin and foul it.

Thanks
Bill

 

[martinbn]

OK, now you answered my question so I can proceed. Essentially, you identify the particle with the corresponding state in the Hilbert space. The particle is the state in the Hilbert space. Fine, now let us see what are the consequences of this statement.

No, not at all. The particle is not the state. Take for example classical mechanics. The particle is described by six numbers. The particle is not a six-tple of numbers.

I don't understand, if the fields are not taken as really existing, but just taken as mathematical method for generating basis elements for an observable algebra, then they're not real objects in space right? It's directly related to it by saying they're not real. I genuinely don't understand, it seems fairly clearly related.

You are using the same word in two different ways, which causes the confusion. The fields are the physical objects that exists in space (and time), which we want to study. The operators, which you also call fields, are the mathematical description, and of course they don't exist, it is meaningless to say that they do.

If you think that it is not easy to make sense of interpretation in which particle does not exist until measured, I perfectly agree with you. One needs to work hard to make sense of it. As a result of such a hard work, I wrote the paper http://de.arxiv.org/abs/1112.2034

But, there is no such interpretation. At least so far you havn't shown one. In all your citations there wasn't even a hint that particles/fields don't exist. And you didn't answer my question. If the particle doesn't exist, do we have empty space?

 

[DarMM]

Maybe that's why I became a mentor

Well you mentors are on a whole other level, not merely good looking but so much so that us mortals can only grasp it dimly.

 

[DarMM]

You are using the same word in two different ways, which causes the confusion. The fields are the physical objects that exists in space (and time), which we want to study. The operators, which you also call fields, are the mathematical description, and of course they don't exist, it is meaningless to say that they do.

No, I'm not using it in two sense. In Haag's description there is nothing physically real obeying field equations, hence there are no physically real fields. There is only the algebra of local observables. Fields only appear as one method of constructing the algebra.

Other methods exist:
https://arxiv.org/abs/1005.2656

 

[bhobba]

Well you mentors are on a whole other level, not merely good looking but so much so that us mortals can only grasp it dimly.

Yea right. We are so good looking we have an area we have to look at to discuss posts that need looking at. I am sure its meant to age us so whatever looks we have are soon gone - assuming they are there to begin with. The things we do so people have a reliable source of science/math/engineering . Seriously it is nice helping people with science stuff in a different way than just contributing to threads - I really enjoy it.

Thanks
Bill