A What do physicists mean when they say photons have a "path"?

[vanhees71]

[Moderator's note: thread spun off from previous thread due to topic/level change.]

This [Ed.: the claim that photons have a "path"] is a misconception of quantum theory already for massive particles. It's even more severely misleading for massless quanta of spin $\geq 1$, which do not even allow the definition of a position observable itself. Particularly in the here discussed specifically quantum properties of entanglement any reference to paths must lead to confusion, particularly for the delayed-choice quantum erasure experiment discussed right now.

 

[DrChinese]

This is a misconception of quantum theory already for massive particles. It's even more severely misleading for massless quanta of spin $\geq 1$, which do not even allow the definition of a position observable itself. Particularly in the here discussed specifically quantum properties of entanglement any reference to paths must lead to confusion, particularly for the delayed-choice quantum erasure experiment discussed right now.

There is nothing misleading in what I have said, because it matches exactly generally accepted science (what we are supposed to advance here). As I have already referenced, the top scientific team of Walborn et al in 2001 used the term "path" 38 times in a single paper to describe how a photon travels. Maybe you should inform them that photons don't have paths, so they can correct the paper you referenced. In the meantime, I ask you to provide any substantiation to support the idea that photons - or any quantum particle - lacks a path. (And please, I already mentioned that their path is not sharply defined - any more or less than any quantum observable.)

This may shock you: the light from your CRT - what you are observing now to read this post- arrived at your eye shortly after emission from an LED. To get there, it went on a "path" from the "source" CRT to your "destination" eye, and there are no other words in any language to describe this better.

So no position observable? And yet photons are regularly found - exactly where they are expected to be! By the scientists who are performing experiments on entangled photons. How about you return to the subject of this intermediate (I) thread: paths of entangled photons, not whether photons have paths.

 

[PeterDonis]

So no position observable? And yet photons are regularly found - exactly where they are expected to be!

Note, though, that the "position observable" in question in such experiments is the one for the photon detector, not for the photon itself. Yes, of course, if a photon detector clicks, we know a photon was detected at the position of the detector (which is not a single point but a reasonably small finite region of space). That is not at all inconsistent with the fact that there is no position observable for the photon itself. It just means that, if you really want to be rigorous, you have to be more careful in formulating exactly how you model, in the math, what the photon detector is doing.

 

[PeterDonis]

the light from your CRT - what you are observing now to read this post- arrived at your eye shortly after emission from an LED. To get there, it went on a "path" from the "source" CRT to your "destination" eye

Actually, the last statement here is not necessary either to describe or to model mathematically what is going on in this scenario. And one of the general guidelines in QM is that we should be careful about such statements. (And note that this would apply to an electron as well as to a photon. The key is that in the scenario being described, we are not measuring what happens in between the source and detector.)

 

[vanhees71]

Note, though, that the "position observable" in question in such experiments is the one for the photon detector, not for the photon itself. Yes, of course, if a photon detector clicks, we know a photon was detected at the position of the detector (which is not a single point but a reasonably small finite region of space). That is not at all inconsistent with the fact that there is no position observable for the photon itself. It just means that, if you really want to be rigorous, you have to be more careful in formulating exactly how you model, in the math, what the photon detector is doing.

Yes, and that's crucial to understand what's understood as "path of photons", particularly in this specific experiment with entangled photon pairs. "Which-path information" means that by some means, in this case using the quarter-wave plates in the slits, the photon is prepared in such a way that it is possible to obtain the information through which slit it went. In this case it's by the polarization state (left- or right-handed polarized is uniquely "entangled" with the "which-way information"). As a result a such prepared ensemble of photons does not show a double-slit interference pattern on the screen. It's also important to note that the which-way information is not "read out" by any measurement here, such that the entanglement with the other photon is not destroyed. This enables the "erasure of the which-way information" by postselection using this other photon of the pair, splitting the ensemble in two sub-ensembles, each showing an accordingly shifted double-slit interference pattern.

To know through which slit an individual photon would have gone, you'd have to determine whether it's L- or R-polarized, but then the entanglement with the other photon and thus the possiblity of "erasure of which-way information" is destroyed. The "path" of the photon, i.e., through which slit it went is not due to the measurement of a "photon position" (which does not exist) but due to interaction with the quarter-wave plates in the slits, and these define a "position" in terms of "which slit".

 

[DrChinese]

Actually, the last statement here is not necessary either to describe or to model mathematically what is going on in this scenario. And one of the general guidelines in QM is that we should be careful about such statements. (And note that this would apply to an electron as well as to a photon. The key is that in the scenario being described, we are not measuring what happens in between the source and detector.)

Of course a photon had a path to your eye. I can't believe this is even a topic of any kind of debate. As I mentioned several times, it is not necessarily a specific path - and not with a specific source point or destination point. But WE DO IN FACT MEASURE that the photon took a path - in contradiction to your statement. And guess what, that general (and classical!) path is predicted in advance and confirmed with entangled photons in thousands of experiments (besides the authoritative one I referenced, which has so far been ignored). Yes, the existence of a path IS necessary to describe the experiment. So I would challenge @vanhees71 (or you) to show me any authoritative source that says otherwise.

Not only do particles move along (quantum) paths when not observed: they also have mass, charge, momentum, spin and other observables. @vanhees71 may as well be telling the OP that teaching these quantum observables are misleading. He identifies our 2 entangled particles by referring to their momentums p1 and p2, and implies that is proper. Instead: I can just as rationally identify those photons as taking paths P1 and P2 (or having frequency f1 and f2, etc). That is no more or less correct, and again I challenge you or vanhees71 to show me any authoritative source that says otherwise. On the other hand, almost every paper on entangled particles reference the photons as either signal or idler, which is a description of the path taken (and NOT the initial momentum, as it should be if he were correct).

If my Walborn reference and its usage of path in the same context as I use it isn't enough, then I will supply as many more as is necessary. I'm not trying to be contrarian, but your statements are as subject to challenge as anyone else's. So... where is ANY published experiment that says entangled photons don't take a path to a detector? Every photon detected in any entangled experiment is proof of just that... just as every ray of light from the sun is.

I ask that this portion of the discussion be split off from this thread. [Ed.: it now has been, to this thread.] Seriously, does anyone think this ridiculous discussion is assisting the OP @fluidfcs to understand my answers to his questions (which I answer in post #20)? Of course not, and anyone else reading might be wondering about their understanding as well. If this discussion belongs anywhere, it is in Quantum Interpretations/Foundations as there are valid interpretations that in fact assert particles have specific paths at all times. I don't follow those interpretations personally, but: if we are going to dissect the semantics of the word "path", I guess we'd better consider that too.

 

[vanhees71]

I think it's a mutual misunderstanding here. It is only very important, particularly in this context, not to confuse photons with something like particles. A better intuitive picture is the electromagnetic field, and this it is for light since Maxwell! Of course you have a light source, but light is not a particle moving on a path but an electromagnetic wave field propagating in space (if you have a point-like source it's a spherical wave). Your eye is, for the purpose of this discussion, a detector, and this detector registers "light" basically via the photoeffect. The localization of this "registration event" is due to your eye, it's not measuring the position of an electromagnetic wave, for which "position" doesn't make much sense. It's rather a field, i.e., at each position you have an electromagnetic field, and the intensity of the light is given by the energy density of this em. field as a function of space and time.

Now a photon is a specific state of the quantized electromagnetic field and as such it does not have a position. It doesn't even have an observable referring to position. All there is, is the probability to register the photon at a place defined by the detector (e.g., a CCD camera, where a pixel gets excited by the photon with some probability with is proportional to the energy density of the em. field).

The problem with the naive photon-particle picture, as if it were a little bullet going a specific path is going wrong, particularly when it comes to the here discussed question of specifically wave-like properties like the interference fringes due to diffraction at a double slit or even the very quantum-specific properties described by entangled two-photon states.

It's also very clear what the Walborn et al mean by "path" in their paper, and it's clearly meant in the sense of quantum optics and not in a naive classical-particle like sense of some pop-sci writeups!

 

[DrChinese]

1. I think it's a mutual misunderstanding here. It is only very important, particularly in this context, not to confuse photons with something like particles.

2. It's also very clear what the Walborn et al mean by "path" in their paper...

1. This is wrong too. Photons are quantum particles, and are well thought of as such when discussing entangled photons where the number is 2. I am not denying that photons can be thought of as "excitations of the electromagnetic field" - or "specific states of the quantized electromagnetic field"- or waves for that matter. But a photon can be called a particle as a matter of convenience, just as protons and electrons can. That is because many cases, quantum objects act as if they are classical objects.

Particle: It's a word! Further, it is quite convenient to talk of the position of a photon even if it lacks a precise observable. Again, I can predict with great precision when/where a detector click will occur due to the presence of a photon. Just as I can refer to your "mind" and discuss that, even though it is the "brain" where the mind is consider to reside. That's because it's useful to do so. Yet your mind has no position observable either. When someone asks if you have made up your mind, do you tell them that you don't have a mind? No, because we understand the limits of the word's usage and accept that for the sake of convenience.

@vanhees71: And sorry, there is NO ONE BEST WAY to refer to and teach quantum ideas. I don't care if you teach or not, you can't tell people to read an entire textbook in order to answer a basic post question. Our answers should reflect the level of understanding of the person who asks. I shudder to think of what you might answer if someone here asked the distance to the sun (PS: one good answer is 93 million miles).2. Yes, thanks for finally acknowledging the obvious. What they mean is EXACTLY the same thing as was meant in usage in this thread. And how everyone uses the term. So I repeat: entangled photons travel on paths from a source PDC crystal to detectors A and B. And Walborn agrees.

 

[vanhees71]

1. This is wrong too. Photons are quantum particles, and are well thought of as such when discussing entangled photons where the number is 2. I am not denying that photons can be thought of as "excitations of the electromagnetic field" - or "specific states of the quantized electromagnetic field"- or waves for that matter. But a photon can be called a particle as a matter of convenience, just as protons and electrons can. That is because many cases, quantum objects act as if they are classical objects.

Particularly here they do definitely not act as if they were classical objects, and photons are the least particle like of everything we call a particle. They don't even admit a position observable and thus thinking in terms of paths of pointlike particles is even worse than when doing this for massive particles, where indeed under some circumstances that's not too wrong.

Particle: It's a word! Further, it is quite convenient to talk of the position of a photon even if it lacks a precise observable. Again, I can predict with great precision when/where a detector click will occur due to the presence of a photon. Just as I can refer to your "mind" and discuss that, even though it is the "brain" where the mind is consider to reside. That's because it's useful to do so. Yet your mind has no position observable either. When someone asks if you have made up your mind, do you tell them that you don't have a mind? No, because we understand the limits of the word's usage and accept that for the sake of convenience.

It's not convenient if you use words in a situation, where it leads to misconceptions and misunderstandings. Here we discuss about physics and not much more complicated philosophical issues about, what "mind" might be and what it might have to do with the physics and chemistry of our brains.

@vanhees71: And sorry, there is NO ONE BEST WAY to refer to and teach quantum ideas. I don't care if you teach or not, you can't tell people to read an entire textbook in order to answer a basic post question. Our answers should reflect the level of understanding of the person who asks. I shudder to think of what you might answer if someone here asked the distance to the sun (PS: one good answer is 93 million miles).

I don't claim that I have a good way to teach QT. All I say is that it is for sure a bad way to teach QT, using wrong concepts.

2. Yes, thanks for finally acknowledging the obvious. What they mean is EXACTLY the same thing as was meant in usage in this thread. And how everyone uses the term. So I repeat: entangled photons travel on paths from a source PDC crystal to detectors A and B. And Walborn agrees.

You may repeat it as often as you want, but it does not become right just repeating it.

 

[PeterDonis]

Of course a photon had a path to your eye. I can't believe this is even a topic of any kind of debate.

It's a topic of debate because, as far as I can tell, you are making claims that go beyond what QM actually says. So one of us must be wrong. If it's me, I'd like to find that out. If it's you, I'd like you to find that out. Either way, there is definitely something to discuss.

As I mentioned several times, it is not necessarily a specific path - and not with a specific source point or destination point. But WE DO IN FACT MEASURE that the photon took a path - in contradiction to your statement

How do we measure that the photon took a path? Please be specific. Note that just saying, well, we know the source emitted a photon and we detect a photon at detector A, and the source and detector both have well-defined positions, is not sufficient. That's not a measurement of a path.

Also note that phenomena like, for example, being able to point a laser pointer at a specific spot on a screen are not relevant, because a laser pointer does not emit single photons, and we are talking here about experiments involving single photons (or pairs of entangled photons). The point in dispute here is not whether classical light rays have paths, but whether single photons do.

 

[vanhees71]

How do we measure that the photon took a path? Please be specific. Note that just saying, well, we know the source emitted a photon and we detect a photon at detector A, and the source and detector both have well-defined positions, is not sufficient. That's not a measurement of a path.

That's the important point: According to QED, and that's finally the theory which is today considered valid to describe the quantized electromagnetic field and thus also photons as specific asymptotic free one-photon Fock states. The photons themselves have no position observable. All you can calculate are the probabilities for detecting a photon at a place determined by the detector given the state of the electromagnetic field.

Also note that phenomena like, for example, being able to point a laser pointer at a specific spot on a screen are not relevant, because a laser pointer does not emit single photons, and we are talking here about experiments involving single photons (or pairs of entangled photons). The point in dispute here is not whether classical light rays have paths, but whether single photons do.

The laser field is described by a coherent Gaussian beam. FAPP almost always you can describe it by the corresponding classical electromagnetic field. I think here nobody doubts that there is no specific position of "a photon" within this beam. Of course it's intensity is pretty narrow in the transverse direction but in longitudinal direction it's totally delocalized.

The problem only comes when people think of single photons not in terms of waves but as if they were localized "massless particles", as Einstein thought in 1905. Einstein himself was the first to acknowledge that this is not the final answer. Maybe modern QED is also not the final answer but at least it's a much more consistent and accurate description than the old "wave-particle-duality quantum mechanics" of before 1926.

 

[DrChinese]

1. It's a topic of debate because, as far as I can tell, you are making claims that go beyond what QM actually says. So one of us must be wrong. ...

2. How do we measure that the photon took a path? Please be specific. Note that just saying, well, we know the source emitted a photon and we detect a photon at detector A, and the source and detector both have well-defined positions, is not sufficient. That's not a measurement of a path.

Also note that phenomena like, for example, being able to point a laser pointer at a specific spot on a screen are not relevant, because a laser pointer does not emit single photons, and we are talking here about experiments involving single photons (or pairs of entangled photons). The point in dispute here is not whether classical light rays have paths, but whether single photons do.

1. I will repeat that this discussion has absolutely nothing to do with the OP. It should therefore be split away, preferably to the Interpretations subforum. At best, this is a gentlemanly semantics discussion; and at worst it is distracting contest of wills that can have no winner.

@PeterDonis: if you aren't willing to provide a hard reference contradicting my reference and commentary below, what does that really say about this discussion anyway? Same question to @vanhees71... you both are acting as if your words alone are authority when challenged.

------------------------------

2. I have already explained how we know light takes a path, and provided referential support for same. And I have repeatedly specified that ENTANGLED PHOTONS (not a laser pointer) take paths that are determined in advance from a source crystal, where they originate, to a destination detector, where they are destroyed as part of the detection process.

On the other hand: you refuse to provide any reference for the (absurd) idea that entangled photons lack a path. And yet, check out this paper in which the path length difference is controlled to within a half-wavelength of the detected photons after traveling more than 8 kilometers. That's a ratio of 1:12 billion for the path length for individually detected photons. Hmmm, that's a path!

https://arxiv.org/abs/quant-ph/9806043

This is the entangled photon path, which originates as I described above in Geneva, follows a carefully laid out path and ends an entire city away, at the precise location Tittel et al selected:

"For our Franson-type test of Bell inequalities [16], we produce energy-time entangled photons by parametric downconversion (Fig. 1). Light from a semiconductor laser with an external cavity (10 mW at 655 nm, ∆ν <10MHz) passes through a dispersion prism P to separate out the residual infrared fluorescence light and is focused into a KNbO 3 crystal. The crystal is oriented to ensure degenerate collinear type I phasematching for signal and idler photons at 1310 nm [17]. Behind the crystal, the pump light is separated out by a filter F (RG 1000) while the passing down-converted photons are focused (lens L) into one input port of a standard 3-dB fiber coupler. Therefore half of the pairs are split and exit the source by different output fibers. Using a telecommunications fiber network, the photons are then analysed by all-fiber interferometers located 10.9 km apart from one another in the small villages of Bellevue and Bernex, respectively. The source, located in Geneva, was 4.5 km away from the first analyser and 7.3 km from the second, with connecting fibers of 8.1 and 9.3 km length, resp., as indicated in Fig. 1. Our interferometers use both the Michelson configuration and have a long and a short arm. In order to compensate all birefringence effects in the arms (i.e. to stabilize the polarization), we employ so called Faraday mirrors (FM) to reflect the light [18]. At the input ports, we use optical circulators (C). These devices guide the light from the source to the interferometer, but, thanks to the non-reciprocal nature of the Faraday effect, guide the light reflected back from the interferometer to another fiber, serving as second output port. The output ports of each interferometer are connected to photon counters [19]. We label the ”direct” port as ”+”, the one connected to the circulator ”-”. To control and change the phases ( δ 1, δ 2), the temperature of the interferometers can be varied.

Since the arm length difference is five orders of magnitude larger than the single photon coherence length, there is no single photon interference. However, the path difference in both interferometers is precisely the same, with a sub-wavelengths accuracy. Moreover, this imbalance is two orders of magnitude smaller than the coherence length of the pump laser. Hence, an entangled state can be produced where either both photons pass through the short arms or both use the long arms. Noninterfering possibilities (the photons pass through different arms) can be discarded using a high resolution coincidence technique."

And please take specific note of the fiber path, including the coiled portions, as that is determined by the scientists and the photons are not free to go anywhere else. Other than the PATH it actually takes, of course.
@vanhees71 said: "...in the here discussed specifically quantum properties of entanglement, any reference to paths must lead to confusion". Perhaps you should explain to this "confused" top team that it is not useful to think of these entangled particles as having a path. Because they did exactly the opposite of what you say.

And you asked: "How do we measure that the photon took a path? Please be specific." So there's my verbatim quoted answer, I don't think it gets any more specific than this solid gold reference. So... what you got that says photons don't take paths? A quoted reference would be appropriate here.

---------------------------

Lastly: I will concede the point, if you can explain the following about entangled photons:

a. I say they have a path, which length can be measured within tolerance (8.1 km +/- 655 nm in the reference).
b. I say they have a frequency/wavelength, which can be measured within tolerance (1310 nm +/- in the reference).
c. I say they have a spin, which can be measured within tolerance (per figure 2 in the reference).

Tell me: what is different about a.path as opposed to b.frequency and c.spin? Photons take paths, have a frequency, and have spin. How is saying any of this misleading in any way?-DrC

 

[mattt]

This is a matter of semantics.

DrC, you are calling "a path" a collection of two ( or a finite number at most) of spatial points, whereas Peter Donis and Vanhees are calling "a path" something like a "continuous curve".

You both are right in your assertions (within your different meanings of "path").

 

[PeterDonis]

the fiber path

In other words, we know these particular photons took a particular path because we confined them to that particular path using a fiber optic cable.

That's a perfectly good answer to the question of how we know that those particular photons have a path--we forced them to have one with a physical device.

But I understood you to be claiming that all photons--even photons traveling through free space, not confined by any physical device like a fiber optic cable--have a path. That is the claim I was disputing.

 

[vanhees71]

Take an arbitrary standard textbook about quantum optics like Scully and Zubairy for the standard modern treatment of what photons are, namely specific states of the quantized em. field. Also in a fiber you don't have classical light particles that are localized or taking a path. A fiber is rather discribed as a wave guide.

 

[DrChinese]

This is a matter of semantics.

DrC, you are calling "a path" a collection of two ( or a finite number at most) of spatial points, whereas Peter Donis and Vanhees are calling "a path" something like a "continuous curve".

You both are right in your assertions (within your different meanings of "path").

It certainly can be semantics, as I mentioned earlier ("At best, this is a gentlemanly semantics discussion"). Thank you for your confirming comment.

A comment I would add: This thread was intended to answer a questions from the OP about what happens when you modify an entangled photon path. I have answered that question, while @vanhees71 (saying use of the words "path" and "particle" are poor teaching tools) and @PeterDonis ("saying I am not following generally accepted science") took the discussion in a direction that has no bearing at all on that question.

To @vanhees71:
If someone refers to the speed of light c: it doesn't make to say that photons are not particles, or that particles don't have velocities, or that the speed of light is c only in a vacuum, or that the best way to teach quantum physics is to read a 664 page book entitled "Quantum Optics" (and intended for graduate students) - when a thread is marked Beginner or Intermediate. Just answer the question using language that the person will likely understand.

To @PeterDonis:
What are we doing here, when the discussion becomes one of semantics? If you read my posts, it should be obvious that my usage of the word "path" was correct in every instance related to the questions of the OP. I repeatedly referenced *entangled* particles; and I repeatedly specified that such photons travel in near classical paths (as described in great detail) but those paths don't have a well-defined source or destination - and that the nature of their trajectories is open for interpretation (e.g. Feynman: path integrals).

But I am disappointed that you did not at any point ask vanhees71 to provide a reference (as I did) when I challenged him. (And please, a general reference to a textbook is a far cry from a PF reference.) Is my only recourse in this situation to report the post? I didn't do that, because you as moderator were participating, and in fact said the debate about "path" was meaningful. There is absolutely no justification for asking for a reference from me, when he (and you for that matter) refuse to reciprocate.

 

[vanhees71]

To @vanhees71:
If someone refers to the speed of light c: it doesn't make to say that photons are not particles, or that particles don't have velocities, or that the speed of light is c only in a vacuum, or that the best way to teach quantum physics is to read a 664 page book entitled "Quantum Optics" (and intended for graduate students) - when a thread is marked Beginner or Intermediate. Just answer the question using language that the person will likely understand.

What has $c$ to do with the question, whether photons are particles or quanta. I quoted a textbook, because you told me to do so. Any standard textbook describes photons as specific Fock states of the quantized electromagnetic field, and it's not just semantics we are discussing here, because to answer the questions of the OP to refer to paths in the sense of old quantum mechanics as if photons were localizable particles is utterly misleading.

To @PeterDonis:
What are we doing here, when the discussion becomes one of semantics? If you read my posts, it should be obvious that my usage of the word "path" was correct in every instance related to the questions of the OP. I repeatedly referenced *entangled* particles; and I repeatedly specified that such photons travel in near classical paths (as described in great detail) but those paths don't have a well-defined source or destination - and that the nature of their trajectories is open for interpretation (e.g. Feynman: path integrals).

But I am disappointed that you did not at any point ask vanhees71 to provide a reference (as I did) when I challenged him. (And please, a general reference to a textbook is a far cry from a PF reference.) Is my only recourse in this situation to report the post? I didn't do that, because you as moderator were participating, and in fact said the debate about "path" was meaningful. There is absolutely no justification for asking for a reference from me, when he (and you for that matter) refuse to reciprocate.

I gave a reference. What's the problem?

 

[DrChinese]

1. What has $c$ to do with the question...

2. I gave a reference. What's the problem?

1. I don't think you understand the concept of "analogy".

2. A reference to a 664 page text does not qualify as a reference. I provide exact quotes from references. If you are so right (when actually you are so wrong), where's a specific quote? And again, please stop quoting yourself unless it is to peer reviewed paper. Note that textbooks are not always consider PF suitable references anyway.

Please note that I am moving further discussion to the Advisor's lounge.

 

[vanhees71]

Since when are textbooks no valid references? It's such a basic and fundamental issue that it's in any modern textbook on the subject. What else should I cite? If you want a more specific shorter quote, take the book

J. Garrison and R. Chiao, Quantum optics, Oxford University
Press, New York (2008), https://doi.org/10.1093/acprof:
oso/9780198508861.001.0001

and just read Chpt. 1, particularly Sects. 1.3. and 1.4.

It is clear that either you can use semiclassical theory (e.g., for the leading-order treatment of the photoelectric effect and Compton scattering), where the em. field is treated as a classical field and the charged particles quantum-mechanically or you need full quantum electrodynamics (including effective models for the in-medium case to describe the standard optical elements like lenses, mirrors, beam splitters, polarizers, etc.).

The naive Einsteinian particle picture of 1905 is not adequate except in a very heuristic sense for the photo or Compton effects. It's for sure inadequate to understand experiments involving entangled states, particularly the here discussed quantum-erasure experiment.

 

[DrChinese]

What else should I cite?

A actual verbatim quote from any specific citation would be a good start, after probably 5 repeated requests for same.

How about: "entangled photons don't have a path" quoted from ? Or "photons should never be referred to as anything like a quantum particle" from source ?

 

[vanhees71]

I quoted enough textbooks about the very foundations of quantum electrodynamics and quantum optics. If you want original papers on these elementary subjects, maybe you find them in these textbooks.

Edit: Maybe a bit too drastic in the conclusion to abandon the word "photon" entirely from the physics literature, but very clear to the point concerning the physical issues we discuss here (and it's NOT pure semantics!)

Lamb, W.E. Anti-photon. Appl. Phys. B 60, 77–84 (1995). https://doi.org/10.1007/BF01135846

 

[PeroK]

At the risk of getting dragged into a semantic debate. Part of the problem is whether experimental descriptions must necessarily be changed to remain in line with the relevant theory.

To an experimental physicist there must clearly be a "path" along which light may be detected. The equipment is no doubt set up in specific locations along a path in space in order to successfully detect the photon. The theory predicts the probabilistic detection along this path without the photons in any sense actually following the path.

The simplest experiment would be to direct a laser across the lab. Then set up detectors at random throughout the lab. All the detection events may take place along a straight line through the lab.

The theorist may say that the photons do not themselves in any meaningful way travel along that path. But, experimentally the set of possible detection events forms, without doubt, a path through the lab.

 

[vanhees71]

No! The experimentalists are very clear in describing their apparati. Nowhere is a path of a point particle needed in describing the here discussed quantum-eraser experiment. To the contrary it's all using the correct QED description, and it's demonstrating very specific quantum-fieldtheoretical properties of entangled photon pairs, there's nothing particle-like in this phenomenology.

I stressed more than once that of course the detection events of single photons are localized, but not because the photon has a position observable but because the detector and the atom/molecule interacting with the electromagnetic field and giving rise to the measured signal, which means to detect the photon, has one. All you can calculate with the theory are probabilities for detecting a photon (or several photons in coincidence as in the quantum-eraser experients for entangled photon pairs) at the place of the detector.

Of course your laser-beam example is perfectly well described with the standard theory. Take a laser pointer and use some dust to make it visible as a "beam". The laser light is described well by a coherent state (a socalled "Gaussian beam" in the paraxial approximation; for details see Garrison and Chiao, Chpt. 7). That predicts indeed precisely what you say: It's almost entirely along a straight line, but this must not be misinterpreted as a stream of classical particles running along this beam. To the contrary, we have a very clear manifestation of an electromagnetic-wave phenomenon here!

 

[PeroK]

Nevertheless, I would like to see proof that papers by reputable experimental physicists never indicate, implicitly or explicitly, a photon "path". In diagrams, for example.

@DrChinese claims they do. I don't know, although I'm tempted to believe him on this point.

 

[mattt]

At the risk of getting dragged into a semantic debate. Part of the problem is whether experimental descriptions must necessarily be changed to remain in line with the relevant theory.

To an experimental physicist there must clearly be a "path" along which light may be detected. The equipment is no doubt set up in specific locations along a path in space in order to successfully detect the photon. The theory predicts the probabilistic detection along this path without the photons in any sense actually following the path.

The simplest experiment would be to direct a laser across the lab. Then set up detectors at random throughout the lab. All the detection events may take place along a straight line through the lab.

The theorist may say that the photons do not themselves in any meaningful way travel along that path. But, experimentally the set of possible detection events forms, without doubt, a path through the lab.

That is what I meant by "a finite number of spatial points", referred to as "a path" by DrC ( and experimental physicists really), whereas Peter Donis and Vanhees wouldn't call it "a path".

 

[DrChinese]

I quoted enough textbooks about the very foundations of quantum electrodynamics and quantum optics. If you want original papers on these elementary subjects, maybe you find them in these textbooks.

Edit: Maybe a bit too drastic in the conclusion to abandon the word "photon" entirely from the physics literature, but very clear to the point concerning the physical issues we discuss here (and it's NOT pure semantics!)

Lamb, W.E. Anti-photon. Appl. Phys. B 60, 77–84 (1995). https://doi.org/10.1007/BF01135846

First, you seem to have completely forgotten that we were primarily discussing whether entangled photon take paths (they do!). I have challenged you on that... where's the reference?

Second, "Anti-Photon" from Lamb? Of course he is an authority on light, that's no issue (I already had this link). But his viewpoint hardly represents scientific consensus... why else would he write this article? And he clearly states it is his opinion, that he prefers the usage of the word "light" and "radiation" over photons (which by definition are light quanta). The existence of Fock states can be considered solid evidence for grouping photons as particles of spin 1. For anyone interested, here is a link to the full paper itself (no paywall):

http://www-3.unipv.it/fis/tamq/Anti-photon.pdf

The following shows that there is ongoing debate about whether or not photons should be considered particles. Note that these are not intended as references themselves, just showing that it is essentially a matter of opinion how best to refer to photons (particle, or not).

https://physics.stackexchange.com/q...-what-w-e-lamb-means-in-his-paper-anti-photon
https://scienceblogs.com/principles/2013/07/12/photons-are-here-to-stay-deal-with-it

If Lamb needed in 1995 to write an article to express his displeasure that since 1926, photons are grouped as particle but don't actually exist... well, his opinion is just as controversial today, and certainly not universally accepted. Here is a suitable reference from one 2001 paper (I got 13.5 million hits on "photon spin 1 particle"):

Kim et al, 2001
https://arxiv.org/abs/quant-ph/0103168

"It is clear that preparation of maximally entangled two particle (two-photon) entangled states, or Bell states, is an important subject in modern experimental quantum optics. By far the most efficient source of obtaining two particle entanglement is spontaneous parametric down conversion (SPDC). ... is the optical path length experienced by the o-polarized photon from the output face of the crystal..."

To the extent this discussion is about semantics, clearly the usage of "photon path" and "photons are particles of spin 1" (along with their variations) is extremely pervasive among scientific papers. To the extent that this discussion is about whether entangled photons have paths (I think we already settled that) or whether photons should be considered particles: I would say the same thing - it's pervasive among scientific papers.

 

[vanhees71]

That is what I meant by "a finite number of spatial points", referred to as "a path" by DrC ( and experimental physicists really), whereas Peter Donis and Vanhees wouldn't call it "a path".

It's pretty much like the "path" due to a charged particle in a cloud chamber, i.e., it's rather a track left due to the interaction with the vapour molecules. There is no contradiction to QT here too. It has been described in a famous paper by Mott (I think in 1929).

 

[DrChinese]

No! The experimentalists are very clear in describing their apparati. Nowhere is a path of a point particle needed...

Hahaha, tell that to anyone setting up a Bell test. They describe a classical path determination process with mirrors, beam splitters, polarizers, filters - you know, the devices that have been around for hundreds of years. That's how the experiment is made to work, they create the photon path themselves. Add in some fiber, and you can be sure the photon has precisely the desired path length and go precisely where the scientist desires. Every single such test is proof that entangled photons have a path. And it is usually called that explicitly. No other term is as useful, for obvious reasons. There basically is no better word than "path" to describe where a photon goes!

And where have I ever said a photon is a point particle? Its location can be and is constrained to small spacetime volumes all the time (as I previously demonstrated in post #32. But they do not have precise sources, nor do they have precise ending destinations.

Of course, for convenience, I DO consider them as point particles for most purposes. I would guess most every reader does too.

 

[vanhees71]

First, you seem to have completely forgotten that we were primarily discussing whether entangled photon take paths (they do!). I have challenged you on that... where's the reference?

It's simply impossible to describe these entangled photons in terms of trajectories (or paths) as you claim. In all papers and books we have discussed so far the authors use the quantum-theoretical description in terms of bras and kets or, equivalently, creation and annhilation operators, because that's the only way this situation can be described. What additional reference are you looking for?

Second, "Anti-Photon" from Lamb? Of course he is an authority on light, that's no issue (I already had this link). But his viewpoint hardly represents scientific consensus... why else would he write this article? And he clearly states it is his opinion, that he prefers the usage of the word "light" and "radiation" over photons (which by definition are light quanta). The existence of Fock states can be considered solid evidence for grouping photons as particles of spin 1. For anyone interested, here is a link to the full paper itself (no paywall):

http://www-3.unipv.it/fis/tamq/Anti-photon.pdf

As I said, I find this also too drastic, because the notion of "photon" is well-established and nowadays everywhere understood in terms of modern QED, except in popular-science writing, where an outdated naive particle picture is still propagated. Imho this is tlarable only to a certain extent. To explain the photo effect of the Compton effect it's semi-ok-ish, because it's describing the phenomenology not too wrongly. I think this narrative is however totally misguiding, when it comes to phenomena, where the field-quantization of the em. field is unavoidable, and entangled photon pairs for sure are an example.http://www-3.unipv.it/fis/tamq/Anti-photon.pdf

The following shows that there is ongoing debate about whether or not photons should be considered particles. Note that these are not intended as references themselves, just showing that it is essentially a matter of opinion how best to refer to photons (particle, or not).

https://physics.stackexchange.com/q...-what-w-e-lamb-means-in-his-paper-anti-photon
https://scienceblogs.com/principles/2013/07/12/photons-are-here-to-stay-deal-with-it

If Lamb needed in 1995 to write an article to express his displeasure that since 1926, photons are grouped as particle but don't actually exist... well, his opinion is just as controversial today, and certainly not universally accepted. Here is a suitable reference from one 2001 paper (I got 13.5 million hits on "photon spin 1 particle"):

"It is clear that preparation of maximally entangled two particle (two-photon) entangled states, or Bell states, is an important subject in modern experimental quantum optics. By far the most efficient source of obtaining two particle entanglement is spontaneous parametric down conversion (SPDC). ... is the optical path length experienced by the o-polarized photon from the output face of the crystal..."

Which paper is this? If it's a scientific paper, I'm pretty sure that "path" must be meant not in a naive classical-particle-like way. Parametric downconversion can only be understood in a QFT way (non-linear quantum optics).

To the extent this discussion is about semantics, clearly the usage of "photon path" and "photons are particles of spin 1" (along with their variations) is extremely pervasive among scientific papers. To the extent that this discussion is about whether entangled photons have paths (I think we already settled that) or whether photons should be considered particles: I would say the same thing - it's pervasive among scientific papers.

Nobody denies that photons are "massless spin-one particles", but they are understood as described necessarily as a quantized Abelian gauge field, and that's why, as Lamb writes in his (obviously still controversial essay), the classical limit can not be a point-particle description but classical Maxwell theory. Of course this slang is ubiquitous in the scientific literature. The more important it is to explain to beginners, what's really understood by this slang!

 

[DrChinese]

Nevertheless, I would like to see proof that papers by reputable experimental physicists never indicate, implicitly or explicitly, a photon "path". In diagrams, for example.

@DrChinese claims they do. I don't know, although I'm tempted to believe him on this point.

LOL how many do you want to see? Hmmm, I wonder what arxiv says?

"Photon" & "path" in the Abstract, 1691:

https://arxiv.org/search/advanced?a...cts=show&size=200&order=-announced_date_first

"Photon path" exactly in the abstract (89):
https://arxiv.org/search/advanced?a...cts=show&size=200&order=-announced_date_first

"Photon path" exactly anywhere in the text (10,400):
https://www.google.com/search?q="ph...ACAAUeIAdABkgEBM5gBAKABAcABAQ&sclient=gws-wiz

On the other hand:

"excitations of the quantized electromagnetic field" exactly anywhere in the text (137*, all papers and posts by @vanhees71):
https://www.google.com/search?q="ex...gB6gaSAQM0LjWYAQCgAQHIAQnAAQE&sclient=gws-wiz

Number of arxiv hits on "Photon" in the Abstract: 64,980
Number of arxiv hits on "excitations of the quantized electromagnetic field" in the Abstract: 0

Usually, voting doesn't make it so in science. But in this case, it actually does. It's "Photons" and "Photon Path" by a landslide!

--------------------------------------
*Just kidding - actually 1300, or about 1/8 of the number mentioning photon path. But 1260 (97%) of these references turn around and switch to the usage of photon for most of the article.

 

[DrChinese]

Which paper is this?

Oops, forgot to post the reference itself (this is a just one I picked out of the blue):

Kim et al, 2001
https://arxiv.org/abs/quant-ph/0103168

 

[DrChinese]

1. It's simply impossible to describe these entangled photons in terms of trajectories (or paths) as you claim.

2. Parametric downconversion can only be understood in a QFT way (non-linear quantum optics).

3. Nobody denies that photons are "massless spin-one particles".

1. And yet I did, in detail, in post #32, to within a fraction of a wavelength. Talk to Tittel and explain it to him, then get back to me on that.

2. That's baloney, and you should know better. Conservation and basic QM is plenty enough to describe PDC as it relates to entanglement tests. And that is only one method of creating entanglement. Early experiments created entangled photon pairs using a Cesium cascade, no QFT required.

3. As I have said over and over again, photons are particles. They are quantum particles of course, and subject to "quantum" things like the uncertainty principle. Again, no QFT required.

----------------------------

To anyone still reading this thread: I've said everything that I can say, and have already repeated myself too much. I don't think there are any useful words left for me to say.

@vanhees71: It's to you to have the last word. I thank you for the scholarly discussion. Hopefully this thread will be closed soon.

-DrC

 

[PeterDonis]

@PeterDonis ("saying I am not following generally accepted science")

Where have I said that? I explained in post #14 what particular claim I was objecting to.

 

[PeterDonis]

DrC, you are calling "a path" a collection of two ( or a finite number at most) of spatial points, whereas Peter Donis and Vanhees are calling "a path" something like a "continuous curve".

You both are right in your assertions (within your different meanings of "path").

@DrChinese, if all you mean by "path" is what is described above (two spatial points, which in the case of the experiment described in the OP, would be the exit point at the crystal and either detector A or detector B), then I agree that the photons in the experiment have a "path" in this sense (although I would not myself use the term "path" for this, but that's a matter of choice of words, not physics).

But then there was no need for you to bring in references to experiments using fiber optic cables, which led me to believe you meant "continuous curve" when you said "path"--otherwise those experiments would not even be relevant, since in the experiment in the OP there are no fiber optic cables, just free space between the exit of the crystal and either detector A or detector B.

 

[DrChinese]

@DrChinese, if all you mean by "path" is what is described above (two spatial points, which in the case of the experiment described in the OP, would be the exit point at the crystal and either detector A or detector B), then I agree that the photons in the experiment have a "path" in this sense (although I would not myself use the term "path" for this, but that's a matter of choice of words, not physics).

But then there was no need for you to bring in references to experiments using fiber optic cables, which led me to believe you meant "continuous curve" when you said "path"--otherwise those experiments would not even be relevant, since in the experiment in the OP there are no fiber optic cables, just free space between the exit of the crystal and either detector A or detector B.

Entangled photons travel on paths whether or not they are in fiber. They almost exactly follow classical trajectories. They are NOT classical paths however, for a lot of reasons. The main reason is that photons are quantum particles, not classical particles. I don't know if an individual photon travels on one path, many paths (path integral concept), different paths in different MWI worlds, exact Bohmian trajectories, are continuous or not, etc. They can do lots of things when not being observed. (Nobody I aware of on this planet has any superior understanding of the "truth" of what happens.)

And yet: every experimentalist does all test calibration as if they are observing entangled photons moving on their precisely desired "classical" path with a classically expected arrival time relative to the entangled partner. Let's just call that a path, like everyone else does. Life will be simpler and better.

If it looks like a duck, and quacks like a duck...

 

[PeterDonis]

They almost exactly follow classical trajectories.

The above seems inconsistent with this:

They are NOT classical paths

The latter I would agree with; the former I would not. I don't see how both can be true (to be clear, I'm talking about the case where there are no fiber optic cables or other devices present, just free space between source and detector).

photons are quantum particles, not classical particles. I don't know if an individual photon travels on one path, many paths (path integral concept), different paths in different MWI worlds, exact Bohmian trajectories, are continuous or not, etc. They can do lots of things when not being observed. (Nobody I aware of on this planet has any superior understanding of the "truth" of what happens.)

I agree with all of this.

every experimentalist does all test calibration as if they are observing entangled photons moving on their precisely desired "classical" path with a classically expected arrival time relative to the entangled partner.

So what do you think justifies this, given the other statements quoted above?

 

[PeterDonis]

what do you think justifies this, given the other statements quoted above?

Let me give an example of a possible justification to see if it helps: suppose I assume that a photon is released from my photon source at time $t = 0$. Say the expectation value of the wavelength of photons from my source is $\lambda$. I know the distance from the photon source to the parametric down conversion crystal, and I know the distances from the crystal to detectors A and B, where the pair of down converted photons will be detected (assuming this run of the experiment produces such a pair). Could I compute the probability amplitudes for detecting photons at A and B as a function of time, and show that those amplitudes were sharply peaked around a time $t = T$, where $T$ is the classical light travel time over the sum of the relevant distances? Would the sharpness of the peak be a function of how large the distances were as compared to $\lambda$? Has any such computation been done in the literature?

 

[Demystifier]

Before saying that photons have or not have paths, one should first say what photon is.

So what is a photon? A state in the one-photon sector of the QED Hilbert space? A click in the photon detector? A pointlike object in the Bohmian inerpretation? Something else? If we first agree on that, I think it will be much easier to agree on existence or non-existence of photon paths.

There is no doubt that experimentalists measure something that they call photon paths. But also there is no doubt that theorists can describe those experiments without dealing with a notion of photon paths. So in a sense, both sides are right.

 

[CoolMint]

In double slits experiments, it is possible for classical-like paths of observed photons to be inferred/derived.
Unobserved photons don't appear to have classical-like paths.

 

[vanhees71]

Oops, forgot to post the reference itself (this is a just one I picked out of the blue):

Kim et al, 2001
https://arxiv.org/abs/quant-ph/0103168

As expected this paper uses the standard quantum (!) optics treatment of photons in terms of the quantized electromagnetic field. You claim one could describe parametric down conversion, entangled two-photon states and all that within the naive photon picture of 1905. IMO that's simply impossible, because these phenomena need the correct QFT treatment. That's what's in all introductory chapters of modern quantum-optics textbooks, you however don't seem to accept as valid sources.

 

[DrChinese]

You claim one could describe parametric down conversion, entangled two-photon states and all that within the naive photon picture of 1905.

Wow, where did you get that? What I actually said: "Conservation [of momentum, etc.] and basic QM is plenty enough to describe PDC as it relates to entanglement tests."

You don't need to know how to build a car to drive one. That's an analogy! You don't need to know how non-linear crystals generate entangled photon pairs to perform a Bell test. Or answer a post question.

Let me give an example of a possible justification to see if it helps: suppose I assume that a photon is released from my photon source at time $t = 0$. Say the expectation value of the wavelength of photons from my source is $\lambda$. I know the distance from the photon source to the parametric down conversion crystal, and I know the distances from the crystal to detectors A and B, where the pair of down converted photons will be detected (assuming this run of the experiment produces such a pair). Could I compute the probability amplitudes for detecting photons at A and B as a function of time, and show that those amplitudes were sharply peaked around a time $t = T$, where $T$ is the classical light travel time over the sum of the relevant distances? Would the sharpness of the peak be a function of how large the distances were as compared to $\lambda$? Has any such computation been done in the literature?

Why yes! That was essentially done in the reference - and quote - I placed in my post #12 in this thread. Note that the answer is expressed differently - it's expressed as the relative difference in the arrival times at the A and B detectors (labeled differently in the paper). That's because the pair creation time cannot be well constrained to an arbitrarily small time window. (After all, it's a quantum particle. ) But note that comparing the path length to the wavelength precision, it's about 12 billion to 1 in terms of the peak expectation values.

1. The meaning of the above: It is AS IF (even though it's not) each photon travels a path that is very very nearly a continuous classical path - and nothing else! To the extent that they DON'T travel such a continuous classical path: they do such similar things that they arrive extremely closely together.

2. Further, it should be obvious from comparing these results with other Bell tests with total path lengths on the order of a laboratory room (perhaps 2 meters as compared to 8 kilometers from city to city, a ratio of 1:4000) that there is no more "quantum discontinuous" action during their travel due to the total length traversed. If there were, the much longer travel time of the Tittel et al experiment would require a proportional larger coincidence window. But it doesn't.

3. The conclusion: to the extent that entangled photons do not travel in classical paths, it is not measurable as having any dependency on path length. Please do not quote me as saying entangled photons travel on classical paths, they are quantum particles and subject to quantum rules. But there is no clear measurable evidence that they don't travel on continuous paths with current experiments.

4. One of the great and easily understandable proofs that photons don't travel on a classical path is the one in which light is reflected from a mirror to a source that measures the intensity of reflected light. Let's call that amount i1. It is possible to place small etchings at precise spots AWAY FROM the primary reflection point such that some destructive quantum interference is eliminated (which should result in MORE transmitted light). Such spots are NOT in the classical path, but are in the quantum path. We measure the intensity with the etchings in place, and it is i2. If photons travel along classical paths, i1=i2. In actual experiments, i1<i2 - defying common "sense". I have not seen a reference where a similar setup used entangled photons as the light source, but I think that would be interesting. According to Kaur and Singh (2020): "Because of path revealing quantum entanglement of particles the single particle interference is suppressed." Would that mean that entangled photons wouldn't have something like i1<i2?

 

[PeterDonis]

That was essentially done in the reference - and quote - I placed in my post #12

First, that experiment uses fiber optic cables, which, as I've already noted, physically constrains the "path"; it's not the same as having free space between the source and the detector.

Second, I don't see how any calculation of the sort I described is made in the paper. The authors assert that the relative path lengths will affect the relative arrival times at the detectors in a particular way, but they don't calculate it. If they are relying on such a calculation done elsewhere in the literature, they don't give a reference to it. I'm wondering if there is any paper in the literature that actually calculates the probability of arrival at the detector as a function of time and shows that it is sharply peaked around the expected classical light travel time from the source given the classical path length, at least for path lengths that are large compared with the wavelength of the light.

 

[PeterDonis]

One of the great and easily understandable proofs that photons don't travel on a classical path is the one in which light is reflected from a mirror to a source that measures the intensity of reflected light. Let's call that amount i1. It is possible to place small etchings at precise spots AWAY FROM the primary reflection point such that some destructive quantum interference is eliminated (which should result in MORE transmitted light). Such spots are NOT in the classical path, but are in the quantum path. We measure the intensity with the etchings in place, and it is i2. If photons travel along classical paths, i1=i2. In actual experiments, i1<i2 - defying common "sense".

Yes, IIRC these experiments were discussed by Feynman in his physics lectures, and also in the popular book of his "QED", where he gives them as an example of light not behaving classically.

 

[DrChinese]

First, that experiment uses fiber optic cables, which, as I've already noted, physically constrains the "path"; it's not the same as having free space between the source and the detector.

Second, I don't see how any calculation of the sort I described is made in the paper. The authors assert that the relative path lengths will affect the relative arrival times at the detectors in a particular way, but they don't calculate it. If they are relying on such a calculation done elsewhere in the literature, they don't give a reference to it. I'm wondering if there is any paper in the literature that actually calculates the probability of arrival at the detector as a function of time and shows that it is sharply peaked around the expected classical light travel time from the source given the classical path length, at least for path lengths that are large compared with the wavelength of the light.

1. I don't think that would make any measurable difference at all (by not constraining the path using fiber). I don't have any relevant references that would answer that either way. That is what I was mentioning in my post #41, point 4. You'd need to do a specific experiment to discern either way.

2. I am not sure anyone if working like that. There must be a lot of tuning going on in these experiments. As the photons go through varying materials, the speed through that medium varies from others. I will keep my eyes open for anything that seems to calculate that. One of the problems being that the photon creation times are random.

 

[andresB]

I'm surprised by this tread and the position of DrChinese, I thought it was consensus that in general there is no such thing as a particle path in the usual interpretation of QM (no Bohmian mechanics or anything like that).

In a standard single particle, two-slit experiment, given the original position of the source and the position of the measurement when the particle hit the detector: (1) what method from QM can we use to define a path? I don't think it exists, but even if it exists, (2) how can we be experimentally sure that was the actual path taken by the particle?

 

[DrChinese]

I'm surprised by this tread and the position of DrChinese, I thought it was consensus that in general there is no such thing as a particle path in the usual interpretation of QM (no Bohmian mechanics or anything like that).

In a standard single particle, two-slit experiment, given the original position of the source and the position of the measurement when the particle hit the detector: (1) what method from QM can we use to define a path? I don't think it exists, but even if it exists, (2) how can we be experimentally sure that was the actual path taken by the particle?

Good point AndresB, I wouldn't always take this position. But these are entangled particles. They do not produce double slit interference.

But even in the double slit, with a normal photon, I would say it took a "path". Being a quantum particle, it did not take one specific path in an experiment designed to highlight this quantum behavior. I point out another such in post #41 above, point 4: where light does not take classical paths.

But in an instance where such properties can be neglected - which was the case in the thread where this originated - of course I would talk about a photon's path. Virtually all photons that arrive anywhere do so upon what is almost perfectly a classical path - that is easily demonstrated by blocking anywhere along the most likely path. Should we need factor in the effect of gravity on light when we discuss how light moves? I would not mention that gravity bends light when discussing photon path unless there are large objects involved. To recap my position (from another post of mine above):

Entangled photons travel on paths whether or not they are in fiber. They almost exactly follow classical trajectories. They are NOT classical paths however, for a lot of reasons. The main reason is that photons are quantum particles, not classical particles. I don't know if an individual photon travels on one path, many paths (path integral concept), different paths in different MWI worlds, exact Bohmian trajectories, are continuous or not, etc. They can do lots of things when not being observed. (Nobody I aware of on this planet has any superior understanding of the "truth" of what happens.)

And yet: every experimentalist does all test calibration as if they are observing entangled photons moving on their precisely desired "classical" path with a classically expected arrival time relative to the entangled partner. Let's just call that a path, like everyone else does.

Guess what? All of the above is also true about momentum, position, etc. of any quantum particle. And yet we use the words momentum and position to discuss such particles. Obviously, if I choose to place these attributes in superposition, those words can become meaningless. So I would say there is context to consider, just as when answering a question on any subject.

 

[HomogenousCow]

What exactly is the word photon supposed to refer to in this context? Is it just any state of the quantized EM field?

 

[andresB]

And yet: every experimentalist does all test calibration as if they are observing entangled photons moving on their precisely desired "classical" path with a classically expected arrival time relative to the entangled partner. Let's just call that a path, like everyone else does.

Well, I agree that If we don't look too deep into it, if we don't try to zoom too much into the particle position over time, then we can use the word "path" without any danger. That is compatible with the fact that at the most fundamental level in (the standard interpretation of) QM, particles don't have trajectories.

 

[CoolMint]

A tennis ball is a collection of quanta and is generally assumed to have a path whether it is observed or not. In practice, all attempts to derive a classical path are successful and virtually everyone assumes an unobserved tennis ball has a definite well-established path. In QM terms, the tennis ball 'path' is the average of many trajectories that peak and average out mostly around the classical 'path'. Thus I suggest that QM treatments put commas around 'path' while practical treatments treat the path in question without commas for all practical purposes. Theory meeting practice is the unresolved issue of single outcomes in QM.
For ultimate exactness and truthfullness with theory however, the path should always be denoted as 'path'.

 

[PeterDonis]

One of the problems being that the photon creation times are random.

True, but since the arrival times at the detectors are recorded (within a fairly narrow window), one can just do the calculation in reverse and look at the amplitude for emission from the source as a function of time, given a detection within the known time window, and see if it is sharply peaked about the expected classical emission time given the total path length from source to detector.

The quantum optics literature might not have such a calculation, but that could be because earlier literature on QED more generally did a calculation something like this. Unfortunately I'm not familiar enough with the QED literature to know where to look.