Unruh effect and lessons regarding reality

[tom.stoer]

The Unruh effect claims that a quantum state which can be defined as an "empty vacuum state" by an inertial observer will become a "thermal state containing particles" for an observer with constant acceleration a; the temperature observed is T ~ a.

In contrast to other oberver dependent effects this is not just a different interpretation of one frame-independent reality but seems to be a challange for physical reality at all. Let me explain why.

Suppose we are detecting particles in different references frames, i.e. with different detectors. Each time we are detecting a particle we agree that there is a particle, but we are not in agreement regarding its energy. We are not worried about this fact b/c we can use a Lorentz transformation to get the relation between energy and momentum.

For the Unruh effect it becomes much more disturbing b/c we cannot even agree whether there there is a particle or not. So it seems that we are no longer talking about one unique quantum state with frame dependent interpretations but that we have two truly different quantum states, two different "realities". There is not one event "a particle" with different interpretations like "energies E, E', E'', ...", but there are different realities some with an event "a particle", some with "no particle".

Suppose an accelerating observer and a goup of stationary observers at rest collect the information regarding "their quantum states" over a couple minutes. Once the accelerating observer passes a stationary one they make a simultaneous measurement whether there is a particle or not. After a while the obervers at rest will not have counted any particle, but the accelerating observer will have counted many. So they disagree on the fact "whether there are particles at all".

b/c every particle can be detected only once one could get rid of the problem via the idea that the particles detected by the accelerated observer cannot be detected by the obsevers at rest, so there is no logical contradiction. This is OK, but of course the accelerating observer "knows" that there are other particles that he could detect in principle, but that they escape from his detector by whatever reason. These particles are not detected by the observers at rest, not even in principle.

So the particles detected by the accelerated observer are not and cannot be detected by the observers at rest (b/c they can be detected only once). And the particles not detected by the accelerated observer are not and need not be detected by the observers at rest. So it really seems that there is no logical contradiction - but you may understand that when talking about "reality" one may be bothered by these ideas.

Any thoughts?

 

[Vanadium 50]

I am not a GR expert, but.

Let's imagine the simplest case - I have a "perfect detector" and as soon as it sees a particle, it puts up a flag. All observers agree that this flag goes up - or not- irrespective of frame. The rocket takes off, and after a moment, the flag goes up.

If you ask the rocket traveler what happened, she will say that the rocket started accelerating, they saw Unruh radiation, and the detector sent up a flag.

If you ask the stay-at-home traveler, she will say that the rocket started accelerating, began to radiate, and the detector measured this radiation and sent up a flag.

This is in one of Unruh's papers. Now, you might say, "suppose I took great care to design a rocket that had no changing multipoles and thus will not radiate". It turns out that in that case the detector will not respond to radiation. If you want the detector to be able to detect, it needs some sort of multipole to respond to the radiation, and that very multipole begins to emit. So the answer is that what one observer characterizes as absorbtion, the other characterizes as emission.

 

[Bill_K]

So it seems that we are no longer talking about one unique quantum state with frame dependent interpretations but that we have two truly different quantum states

No, it's all one state, but observers may disagree on whether it's vacuum. Different observers have different vacuums.

what one observer characterizes as absorbtion, the other characterizes as emission.

Yes, exactly.

 

[tom.stoer]

No, it's all one state, but observers may disagree on whether it's vacuum.

yes, I know, sorry for the confusing formulation

 

[tom.stoer]

This is in one of Unruh's papers ... So the answer is that what one observer characterizes as absorbtion, the other characterizes as emission.

is this really the way to resolve the 'paradox'? thanks a lot for this information, seems quite interesting!

 

[strangerep]

In QFT, we build the theory in terms of Poincare unirreps. As soon as you transform to an accelerating FoR, you can't expect the Poincare-based theory to remain good in all aspects.

Consider a simple uniform-acceleration case. The accelerated observer is no longer in Minkowski spacetime, but rather in Rindler spacetime, with extra horizons, etc. Imho, one should then return to a group of transformations that encompass both Poincare and uniform accelerations, then redo the Wignerian analysis from scratch with the (new) Casimirs, etc, and construct unirreps of this larger group. Without such a unified framework, the accelerated and unaccelerated observers just confuse each other.

 

[tom.stoer]

In QFT, we build the theory in terms of Poincare unirreps. As soon as you transform to an accelerating FoR, you can't expect the Poincare-based theory to remain good in all aspects.

Consider a simple uniform-acceleration case. The accelerated observer is no longer in Minkowski spacetime, but rather in Rindler spacetime, with extra horizons, etc. Imho, one should then return to a group of transformations that encompass both Poincare and uniform accelerations, then redo the Wignerian analysis from scratch with the (new) Casimirs, etc, and construct unirreps of this larger group. Without such a unified framework, the accelerated and unaccelerated observers just confuse each other.

Mathematically this is clear. It's a question regarding "interpretation" and "reality", not regarding the math

 

[Demystifier]

Tom.stoer, it is possible to retain objective status of particles in the Unruh effect, provided that a preferred time exists as in Horava gravity. See
http://xxx.lanl.gov/abs/0904.3412

 

[strangerep]

[...] It's a question regarding "interpretation" and "reality", [...]

That's precisely the purpose of a group of transformations between reference frames. Without it, one observer's "reality" cannot be meaningfully compared with another's.

 

[TrickyDicky]

I am not a GR expert, but.

Let's imagine the simplest case - I have a "perfect detector" and as soon as it sees a particle, it puts up a flag. All observers agree that this flag goes up - or not- irrespective of frame.

This just shifts the issue backwards by assuming a "perfect detector that all observers agree about whether its flag is up or down".
As strangerep points out to have such an agreement a new set of transformations would be needed. And that is what Tom stoer seems to be looking for in the OP.

 

[tom.stoer]

That's precisely the purpose of a group of transformations between reference frames. Without it, one observer's "reality" cannot be meaningfully compared with another's.

Sorry to say that but reality in the philosophical sense is never about an "observer's reality"; refer e.g. to Wkipedia: "... reality includes everything that is and has been, whether or not it is observable or comprehensible". What you are talking about are perceptions.

It is absolutely clear to me that we need transformations (acting on states) to compare perceptions of observers, but not to compare realities, as there is - by definition - only one reality.

This holds in non-rel. QM if one is willing to restrict the meaning of reality. Any two observers will agree on the result of a measurement (modulo some transformations) in terms of classicasl physics.

In reality there should be one particle (one event that a particle is at a certain location) or there should be no particle. And this is independent from any observer. The Unruh effect adds an additional philosophical problem b/c the "event that there is a particle" is no longer a concept of an underlying reality - b/c there are different groups of observers some detecting particles, some don't.

The idea that the "existence of a particle" is an element of reality is questioned by the Unruh effect!

This is my problem.

I see some logical excuses (e.g. that there is not one experiment but two, one for each observer, and in one experiment there is a particle whereas in the other one there isn't) but nevertheless the Unruh effect restricts the (limited) concept of reality according to non-rel. QM.

 

[Ilmrak]

Sorry to say that but reality in the philosophical sense is never about an "observer's reality"; refer e.g. to Wkipedia: "... reality includes everything that is and has been, whether or not it is observable or comprehensible". What you are talking about are perceptions.

It is absolutely clear to me that we need transformations (acting on states) to compare perceptions of observers, but not to compare realities, as there is - by definition - only one reality.

This holds in non-rel. QM if one is willing to restrict the meaning of reality. Any two observers will agree on the result of a measurement (modulo some transformations) in terms of classicasl physics.

In reality there should be one particle (one event that a particle is at a certain location) or there should be no particle. And this is independent from any observer. The Unruh effect adds an additional philosophical problem b/c the "event that there is a particle" is no longer a concept of an underlying reality - b/c there are different groups of observers some detecting particles, some don't.

The idea that the "existence of a particle" is an element of reality is questioned by the Unruh effect!

This is my problem.

I see some logical excuses (e.g. that there is not one experiment but two, one for each observer, and in one experiment there is a particle whereas in the other one there isn't) but nevertheless the Unruh effect restricts the (limited) concept of reality according to non-rel. QM.

With that definition of reality a very simple conclusion is that particles are not real. I don't find this too disturbing, they wasn't be defined as real in the first place, particles are only our way to refer to some set of relativistic invariants.

The Unruh effect is not philosophically different to say that a field configuration appears different in different coordinates.

I think this is not so far from what happens in QFT, where field configurations change using different frame of reference. In fact a field that in some frame has a particle with momentum $p$ in some other frame has a particle with momentum $\Lambda p$.
In the Unruh effect not only the momentum changes, but that's all.

To rephrase one of your quoted sentences:
The idea of the "existence of a field" is an element of reality not questioned by the Unruh effect!

Ilm

 

[tom.stoer]

With that definition of reality a very simple conclusion is that particles are not real.

That's a way out; w/o logical inconsistencies.

I don't find this too disturbing ...

I do ;-)

The Unruh effect is not philosophically different to say that a field configuration appears different in different coordinates.

Really?

In the Unruh effect not only the momentum changes, but that's all.

Yes, but for me that's more than just different momenta. Special Relativity did not really affect our concept of realitry; QM does, and QFT does even more!

The idea of the "existence of a field" is an element of reality not questioned by the Unruh effect!

Would you really refer to a field? not just to a state?

Anyway, the concept of reality becomes even more veiled and vague ... or even void

 

[martinbn]

Why do you find it disturbing? Particles are not objects that exist, they are just labels/names for some of the states of the fields.

 

[tom.stoer]

It was disturbing to lose the concept of "reality of values of attributes of particles" in QM (you cannot assign attributes like spin, etc. to a particle before this atrtribute of the particle has been measured). But in QM there was no problem with the particle itself.

Now we do not only lose the concept of "reality of values of attributes of particles" but the concept of "reality of particles".

I think this *is* disturbing (when thinking about reality; if you stick to positivism there is no such problem ;-)

 

[Ilmrak]

"The Unruh effect is not philosophically different to say that a field configuration appears different in different coordinates."

Really?

Well, it roughly says that two observers won't generally agree on the particle content of some system. Being the particle content a label to describe a fields configurations, I'd say that my sentence is not so wrong.

"In the Unruh effect not only the momentum changes, but that's all."

Yes, but for me that's more than just different momenta. Special Relativity did not really affect our concept of reality; QM does, and QFT does even more!

Relativity does change our concept of reality (whatever it could be).

Special relativity changes for example the definition of simultaneity, which we used to think about as “real” and that is instead coordinate dependent.

General relativity shows that the gravitational field is not “real” in the sense we thought before. Two different observers won't agree on the strength of gravitational force in some point or on the gravitational energy of some system.

Roughly speaking, each time the symmetry group of our theory changes casimirs changes as well. So the value of quantity that was coordinates independent, quantity that we thought as “real”, becomes dependent by the observer point of view.

"The idea of the "existence of a field" is an element of reality not questioned by the Unruh effect!"

Would you really refer to a field? not just to a state?

I could say “state”, but I'd have to specify that I'm talking about a coordinate independent quantity, such as a state in some defined coordinate system or a state as a vector in Hilbert space of QFT, so I preferred to only say “field”.
It's just like, if talking about an electron, I could say that what's real is “an electron with momentum $\vec{p}$ with respect that coordinate system”, but I'd generally prefer (for brevity if you want) to say that what is real is only the electron.

Anyway, the concept of reality becomes even more veiled and vague ... or even void

I'd say the exact opposite, but this deserves a separate thread.

I have my own definition of reality, that in my opinion is very solid (although dynamic), but I don't think that talking about this could be useful in this discussion.

Ilm

 

[Ilmrak]

It was disturbing to lose the concept of "reality of values of attributes of particles" in QM (you cannot assign attributes like spin, etc. to a particle before this atrtribute of the particle has been measured). But in QM there was no problem with the particle itself.[/itex]

I don't agree, I think there is some problem in the usual QFT definition of particles...

Now we do not only lose the concept of "reality of values of attributes of particles" but the concept of "reality of particles".

I think this *is* disturbing (when thinking about reality; if you stick to positivism there is no such problem ;-)

Then simply take fields as fundamental entities, as I suggested before.

If you do so particles are only “attributes of fields” and you don't lose the “reality of fields” but only the “reality of values of attributes of fields”. :tongue:

Ilm

 

[tom.stoer]

The problem is that you don't detect fields but particles.

The concept of fields in quantum field theory is a mathematical model; you cannot say whether it's real b/c it never has been observed. Particles i.e. "ticks in detectors" have been observed ;-)

 

[Ilmrak]

The problem is that you don't detect fields but particles.

The concept of fields in quantum field theory is a mathematical model; you cannot say whether it's real b/c it never has been observed. Particles i.e. "ticks in detectors" have been observed ;-)

I'm sorry but I don't agree again ;-)

What we see are not, strictly speaking, particles. We see as you said “ticks in detectors”.

Ticks are caused by some property of the object observed, such as charge, momentum, position, etc.

Are these property of a field or of a particle? The ticks do not answer these question, as an answer depends on the physical theory you are using to interpret the ticks.

To my opinion QFT is not a theory of particles, is a theory of fields. Particles are only something we use to describe fields configuration, we could predict the same ticks in detectors without even naming particles.

So, in my opinion, particle content of a system has to be regarded exactly as the momentum of that system, i.e. as some label (that depends on coordinates used) describing the system configuration (within our theory).

Ilm

 

[TrickyDicky]

This discussion seems to suffer from a certain mixing up of "the map and the territory" kind of problem.
Since I think the OP concern about reality regarding the Unruh effect goes beyond that trivial confusion I'll try to distinguish it so that physics and philosophy don't get mixed up.
A useful assumption (usually a tacit one) in science is that there is just one reality but as observers we only get access to it thru a particular POV, so that as long as we agree about how to go from one POV to another everything should be fine.
This is different from the kind of explanations given by Ilmrak that are rather in the realm of the distinction between the unique reality beneath the multiple models and the models themselves.
So I agree with Tom that the Unruh effect is disturbing for the commonly used in science unique reality assumption due to the lack of the adequate mathematical tools to connect different POV's.
This doesn't happen neither in the SR relativity of simultaneity nor in flat QM-QFT.

 

[Ilmrak]

[...] So I agree with Tom that the Unruh effect is disturbing for the commonly used in science unique reality assumption due to the lack of the adequate mathematical tools to connect different POV's. [...]

I thought that in the Unruh effect we do know exactly how to connect vacua relative to different points of view... Am I missing something?

Ilm

 

[tom.stoer]

I thought that in the Unruh effect we do know exactly how to connect vacua relative to different points of view... Am I missing something?

Ilm

You are missing the basic question if there is a unique underlying reality. To my little daughter it makes not so much difference if there is a Mars or a Snickers bar. What really matters is whether there is at least one chocolate bar. Believe me or not, she will not be convinced when you are trying to explain that there is an accelerated observer which can detect (and eat) a Snickers bar.

We do not face something like that in SR, GR, QM ant even in QFT (in flat space)!

 

[Dale]

In contrast to other oberver dependent effects this is not just a different interpretation of one frame-independent reality but seems to be a challange for physical reality at all.

IMO, the biggest challenge to "physical reality" is defining what we mean by the term, particularly from an experimental/scientific perspective. Once we have a testable definition then we can apply it and see, until then we are just discussing philosophy.

Do you have a proposal for a "physical reality" test?

 

[cosmik debris]

An interesting question Tom. I think I am tending towards Ilmrak's view. The fields are the basic entities. If an excitation of that field can have have a property that ranges over some values, which is frame dependant (e.g. momentum), why can that not be zero?

 

[Ilmrak]

You are missing the basic question if there is a unique underlying reality. To my little daughter it makes not so much difference if there is a Mars or a Snickers bar. What really matters is whether there is at least one chocolate bar. Believe me or not, she will not be convinced when you are trying to explain that there is an accelerated observer which can detect (and eat) a Snickers bar.

We do not face something like that in SR, GR, QM ant even in QFT (in flat space)!

I'm sorry but I can't still understand why the unique underlying reality has to show/have the same defined number of particles in all frames of reference.

I mean, take relativistic redshift. If you have 1 red photon and 0 blue photon in some coordinates, in some other coordinates you have 0 red and 1 blue photon. The particle content change for the two points of view.

How Unruh effect rise more doubts than this on the reality of a particle?

Proceeding with you metaphor, why should she believe that the same chocolate bar that has peanuts for her but hasn't for an accelerating observer?
Moreover, maybe she is allergic to peanuts so that having a Snickers is equivalent to having nothing for her...

Ilm

 

[TrickyDicky]

You are missing the basic question if there is a unique underlying reality.

But that cannot IMO be formulated as a scientific question, but as I said it is rather a working assumption to apply the scientific method, whether or not one is aware of it. Otherwise it is not possible to trust any result, even the unassuming "shut up and calculate" no-philosophy of science uses that consistency requirement as a tacit assumption.

 

[TrickyDicky]

I thought that in the Unruh effect we do know exactly how to connect vacua relative to different points of view... Am I missing something?

Ilm

As commented, there is a mathematical, and therefore physical disconnect between the accelerated observers and the other observers.

 

[Lord Crc]

As a layman this situation reminds me a bit of the situation back when Einstein came up with SR. So, as a layman, I'll ask a rather silly question: what if the particle count is real, and the Unruh effect is wrong, at least w.r.t. particle counts?

What modifications would have to be made in this case?

 

[Maui]

We need Naive realism for the validity of biology, history, geography, etc. sciences as true descriptions of the world(otherwise we run into logical contradictions and inconsistencies). Can we return to naive realism? Ever?

 

[tom.stoer]

@Maui: we can't! We have something like "veiled reality" (Bernard d’Espagnat) which is already a strange issue, but the Unruh effects adds additional problems regarding "reality". Of course one could believe in "structural realism" using field operators and states but that has nothing to do with realism (in the naive sense) which requires at least some "correspondence" between "reality" and "perception". If you give up these concept you should no longer call it "realism" or "reality".

@Ilmrak: I see two problems
a) with one red and one blue photon we still can agree on the total number of photons, i.e. on the number of ticks registered by the detector; with the Unruh effect the number of ticks changes depending on the observer (which is logically no problem for each single tick b/c for the two observers we need to detectors and one single particle is registered in one and only one detector)
b) your position seems to be closed to positivsm, therefore you don't really care about my problem; or you tend to call something "real" b/c you use it to do calculations (like a field operator); I would call this "structural realism" and it's logically consistent, of course

 

[Ilmrak]

As commented, there is a mathematical, and therefore physical disconnect between the accelerated observers and the other observers.

Sorry I still don't understand this.
The connection between the accelerated observer and the inertial one is continuous...
Could you please explain another time what are you meaning?

@Maui: we can't! We have something like "veiled reality" (Bernard d’Espagnat) which is already a strange thing, but the Unruh effects adds additional problems. Of course one could believe in a structural realism using field operators and states but that has nothing to do with realism (in the naive sense) which requires at least some "similarities" between "reality" and "perception". If you give up these concept you should no longer call it "realism" or "reality".

@Ilmrak: I see two problems
a) with one red and one blue photon we still can agree on the total number of photons, i.e. on the number of ticks registered by the detector; with the Unruh effect the number of ticks changes
b) your position seems to be closed to positivism, therefore you don't really care about my problem; or you tend to call something "real" b/c you use it to do calculations (like a field operator); this I would call "structural realism" and it's logically consistent

How Unruh effect rise more doubts than this on the reality of a particle?

a) If your detector is only sensible to red photons then in one frame it ticks but not in the other.
I agree this is a bit different from what happens in the Unruh effect, but what I'm saying is that using a detector in an inertial frame or in presence of a gravitational field is as setting two different experiments and so seems natural to have two different results (as long as the laws that describes such results are the same in the two frames).
b) Yes, maybe what you call “structural realism” could be close to my position.

Anyway, I still have some difficulties in understanding exactly what's the point on which we are debating :-)

Sorry if I insist on this, but could you explain exactly in what is the Unruh effect more challenging then the fact that in GR the energy contained in some fixed region of space depends on the coordinate system used?
As I understand it there are not much interpretation differences besides the exchange particle$\leftrightarrow$energy, and I don't get why particles are "more real" then energy.

IMO, the biggest challenge to "physical reality" is defining what we mean by the term, particularly from an experimental/scientific perspective. Once we have a testable definition then we can apply it and see, until then we are just discussing philosophy.

Do you have a proposal for a "physical reality" test?

Maybe you're right and we should debate on what the “physical reality” is to resolve this...

Ilm

 

[Physics Monkey]

Tom,

Along the lines of what Vanadium 50 said earlier, you will probably be interested in looking at this paper: http://prd.aps.org/abstract/PRD/v29/i6/p1047_1

See also some discussion of mine here https://www.physicsforums.com/showthread.php?t=574548&highlight=unruh+effect+physics+monkey&page=3 post 42 and on.

Also intriguing IMHO is black hole complementarity where there are again serious interpretive issues.

 

[tom.stoer]

Along the lines of what Vanadium 50 said earlier, you will probably be interested in looking at this paper: http://prd.aps.org/abstract/PRD/v29/i6/p1047_1

Great, thanks. Yes, it seems that there is in fact no logical inconsistency and that even physical interpretations are OK. Let me think about it for a while ...

 

[strangerep]

Sorry to say that but reality in the philosophical sense is never about an "observer's reality"; refer e.g. to Wikipedia: "... reality includes everything that is and has been, whether or not it is observable or comprehensible". What you are talking about are perceptions.

[...] there is - by definition - only one reality.

For the record, I reject the notion of one single absolute total reality -- I find the notion insufficiently supported by experiment. Rather, I prefer the Rovellian argument that different noninertial observers can give inequivalent, nonisomorphic accounts of their perceived realities, as each measures it, due to horizons, etc.

In reality there should be one particle (one event that a particle is at a certain location) or there should be no particle. And this is independent from any observer.

No. There may indeed be a Mars bar in the accessible event set of one observer, but not the other, if a Rindler horizon intervenes.

[@Ilmrak: I don't think you're missing anything. Your explanations seem fine to me.]

 

[tom.stoer]

For the record, I reject the notion of one single absolute total reality -- I find the notion insufficiently supported by experiment. Rather, I prefer the Rovellian argument that different noninertial observers can give inequivalent, nonisomorphic accounts of their perceived realities, as each measures it, due to horizons, etc.

(sometimes) even philosophers use strict definitions - and ''perceived reality" misses the essential point. You have observer dependent perceptions (obviously) and you may have or believe in one single underlying observer-independent (!) reality. Observer dependent realities or perceived realities are a contradictio in eo ipso. You should call them simply perceptions or phenomena, not reality. Perhaps these philosophical terms may not map directly to our mathematical structures we use in QFT, but we should be clear about them.

No. There may indeed be a Mars bar in the accessible event set of one observer, but not the other, if a Rindler horizon intervenes.

This is about perceptions and physical predictions; that's fine and nothing is missed - except for a definition of reality. I would not translate "accessable event set" as reality.

The basic question is whether the concept "event" or "one particle" can be an element of an underlying reality, or if it is an element of oberver dependent perceptions. If the latter one is assumed, what is the underlying reality? or do we want to give it up completely and stick to positivism? Does the particle exist w/o any observation or detection? In QM the answer is either "yes" (for all observers) or "no" (for all observers) and we only need to consider different perceptions (momentum, color, ...) of the same particle. In QFT this fades away and it seems that we cannot say that the particle exists independent from any observation. Not even the concept of "average number of particles in a given volume" cannot be used as an element of reality w/o contradiction.