Talk:Abraham–Minkowski controversy

Experiment Supports Minkowski

Light propagates slower in any medium than it does in a vacuum. The product of the frequency and wavelength of light gives the velocity of light in either case. So, what this controversy really boils down to is the question of whether it is the frequency, the wavelength or both that changes when light enters a dielectric medium.

If the wavelength does not change, then the frequency must decrease by the same fraction that the velocity of light does upon entering the medium. Likewise, if the frequency does not change, then the wavelength must decrease by the same fraction that the velocity of light does upon entering the medium. If the frequency and the wavelength both change, they must both change by some fraction greater that the fraction by which the velocity of light changes, so that the product of the fractions may equal the fraction by which the velocity of light changes.

I am sorry that I can’t offer you a reference at this time, since I am away from home and access to my files, but this question has been put to the test of experiment by measuring the wavelength of light in water using a grating. The result was that only the wavelength changes and it decreases.

Since the momentum of a photon is Planck’s constant divided by the wavelength, the momentum must increase, because the wavelength decreases. So, this experiment supports Minkowski’s hypothesis. I find the experiment of Weilong She and Sun Yat-Sen inconclusive since light leaving a silica filament is just a photon rocket! A push on the filament is to be expected in any case.Koilon (talk) 02:08, 20 February 2010 (UTC)

I just noticed that article on Refractive Index confirms that the frequency stays the same, while the wavelength shortens, when light enters a medium of higher refractive index.Koilon (talk) 01:51, 28 February 2010 (UTC):

Wavelength is inversely proportional to frequency. You can't change one and not affect the other. There are other aspect of electromagnetic radiation you can change such as group/phase velocity and amplitude. Mulletronics (talk) 14:01, 24 August 2014 (UTC)

Actually, you can change the wavelength of light without changing the frequency — all you need is a lens. Light entering a lens travels more slowly through glass, keeping the same frequency but changing wavelength; the inverse happens when that light leaves the lens. That applies to any material, even air, with an index of refraction different than that of vacuum.  Unician ∇  03:30, 25 August 2014 (UTC)

Experimental Evidence Section

The only reference in the experimental evidence section is an example of bad science. Currently what is in this section is outright false, and the user who first created the section was blocked from Wikipedia for using sock puppets to promote fringe science. There are many experiments that support both sides of the controversy, and both are correct from the proper perspective. Someone who has time could review the literature and replace the section with proper experiments, but until then this wrong information should be removed. Also, I will delete the section again unless someone tells me a good reason not too within a day. 67.255.24.254 (talk) 21:13, 10 February 2012 (UTC)

Feigel?

Should his name not be linked to his article? I would but not sure who he is.--Jrm2007 (talk) 04:57, 8 August 2014 (UTC) Let the world change for all of us, aren't you people tired of lies and death. Why is it so difficult for you to see that we all need the literature from Dr Frank WLLACE it doesn't matter how much money we have today if the rest of the world don't understand the importance and the purpose for all us to live happy will keep us safe for ever. knowledge is the must powerful weapon created for all of us to live in a better world and it as been taking away from us. It is ok to use a new strategy — Preceding unsigned comment added by 50.176.140.53 (talk) 16:10, 23 December 2014 (UTC)

Is this page for real?

The article states at the end that the following problems are still open in physics.

For example, when there exist dielectric materials in space,

• Is the principle of relativity still valid?
• Are the Maxwell equations, momentum–energy conservation law, and Fermat's principle still valid in all inertial frames of reference?
• Does the Poynting vector always represent EM power flow in any system of materials?
• Does the photon have a Lorentz four-velocity like a massive particle?

It's hardly credible that the question of the Lorentz group acting on nonisotropic media was not settled long ago.

One writes linear constitutive relations for the "macroscopic" quantities as functions of the "microscopic ones", expressing everything as differential 2-forms, and see how it transforms. Here E and H are one-forms on R^3, D = ε∗E and B = μ∗H are two-forms on R^3, ε and μ are symmetric transformations of R^3, ∗ is the Hodge star operator, and you define F = B+E∧dt and G = D+H∧dt as 2-forms on R^3,1. Then G = σ F where σ = σ(ε,μ) is in block diagonal form with nonzero entries σ_ijk0 and σ_i0kl, corresponding to ε and μ^-1 respectively.

It seems like the terms must mix under boosts, so that the block-diagonal form is not preserved. Then ε, μ would not be general enough, but there would be extra cross terms between them. That would mean that ε, μ do not provide enough flexibility to express all possible constitutive relations between (E,H) and (D,B), even in the rest frame. Then these new cross terms would have to be explained somehow. Perhaps they are extremely small in real materials? After all, constitutive relations are motivated by crystals, or other solids, and such configurations have natural "rest frames". So there could be a reason why these cross terms, if they exist mathematically, tend not to be necessary in most practical situations.

One can also do some dimension counting. Because of the symmetry of ε and of μ, they yield only 6+6=12 dimensions for the image of σ(ε,μ). On the other hand, the general term of σ is σ_[ab][cd], where a,b,c,d range from 0 to 3. The antisymmetric [ab] is a total of 10 dimensions, [cd] is 10 dimensions, so it looks superficially like 100 dimensions (!) for σ. But the symmetry of ε and of μ have to be taken into account, or maybe it gets stretched to unrecognizability by the boosts. But in any case the representation generated by the images σ(ε,μ) should have dimension lower than 100, even if it is larger than 12, after including the effect of the whole Lorentz group.

The point is not to complete this here. But I'm sure, after 100 years, someone has done this, or explained why it totally breaks down. So it must be out there in the literature somewhere, and the problem is not "open".

Is there a possibility that this article is a non-report on a non-issue, and just hasn't achieved enough attention from experts for the scientific consensus to have been expressed?

In any case I would like to be able to read somewhere about the equations of it all. 89.217.26.0 (talk) 14:38, 26 May 2015 (UTC)

This article is little more than a compilation of an obvious misconception, and tries to inappropriately relate this to the possibility of a reactionless drive. The "open questions" are not unresolved, and it is not difficult to settle the "controversy". If you look, you'll see that all the references (or at least those that I looked through) are isolated primary sources, not secondary sources. This article might have had merit as documenting something historical if secondary sources treated it as historical could be found, but any writings that try, and fail, to investigate the science behind it are not worth reading. The final "possibility" you refer to is a certainty.

A short answer to the questions is: general relativity, Maxwell's equations (in the microscopic version), along with the electromagnetic stress–energy tensor are consistent, and do not leave any of the stated questions open. The macroscopic version is merely a way of partitioning the microscopic equations so that they can be conveniently applied in certain contexts; if the predictions suddenly change, the math is in error. —Quondum 04:32, 27 May 2015 (UTC)

Thanks, it is good to read that. Somehow, the article completely lacks perspective or maturity -- your observation of "isolated primary sources" hits the nail on the head. It will be helpful to readers to see these remarks in the talk page.

Of course the microscopic equations are fine. After being initially skeptical and annoyed that there are macroscopic equation at all, I now think they express useful concepts. In the case at hand, the macroscopic equations introduce a non-Minkowski-invariant, anisotropic "material medium", represented by a linear transformation on the space of F_ij. I don't quite know how general this tensor is. In the usual coordinates it's given by an electric permittivity tensor (ε_ij) and/or a magnetic permeability tensor (μ_ij); perhaps that is all that's worth doing, and most likely it allows you to deduce the rest frame of the crystal (and the crystal *has* a rest frame), which then becomes part of the data of the medium.

Incidentally, this "abstract" point of view is necessary as soon as the underlying Lorentz manifold is not Minkowski space, i.e. a general relativity situation. In that setting, you can't proceed at all until you identify the correct Lorentz covariance of the medium. Incidentally, whether the crystal medium enjoys being stretched around by tidal "forces" in a 4-manifold with no killing field we don't ask; no doubt the crystal has its own Lagrangian, but we leave that for another day. The point of the current exercise is just to model the passage of light through it.

Anyway, thanks again.

178.38.48.224 (talk) 01:20, 29 May 2015 (UTC)

Original Research

Looking at this page, it strikes me first that it is written in a distinctly opaque and monolithic manner, but also examining the sources I believe it represents original research and places far too much weight on one author whose papers never made it past arXiV in several cases, which is hardly promising for a physics paper. It also fails to mention a recently written and extensively cited article that claimed to offer a resolution to the controversy. This article needs an extensive rewrite.--ObscureFruits (talk) 08:31, 27 November 2016 (UTC)

Oh and one editor appears to have worked only on this article and a related article, and has included discredited concepts like reactionless drive.--ObscureFruits (talk) 08:33, 27 November 2016 (UTC)

I have some comments on ObscureFruits-talk:

(1) I don't think there are any statements of unpublished "original research".

(2) How do you know "one author whose papers never made it past arXiV in several cases"? I am supprised that you could know such private things, if they were true.

(3) WIKI dictionary is everybody's dictionary. In my opinion, the authors, who have "a recently written and extensively cited article that claimed to offer a resolution to the controversy", should present their work, because the authors best understand their own contributions. You are supposed to ask them to do that.

(4) If you think "one editor ... has included discredited concepts", then you should make comments on that by citing published papers to show why it is discredited, which helps the community. — Preceding unsigned comment added by Smithwikilover (talk • contribs) 17:13, 1 December 2016 (UTC)

Ah good you replied. I can tell because they have no version besides the one on arXiV, this isn't hard to check. One or two of the papers are published, but numerous cited here are not. Secondly, I more think this is an original synthesis, and not representative of the literature, for example why is this relatively well cited article not included? arxiv.org/ftp/arxiv/papers/1208/1208.0872.pdf Thirdly reactionless drive has been so soundly discredited it doesn't need citations. --ObscureFruits (talk) 19:56, 4 December 2016 (UTC)

Reply to ObscureFruits-talk:

(1) I don’t think that you can derive “whose papers never made it past arXiV in several cases" just from “they have no version besides the one on arXiV”, because the two events have no causality. Of course, if you are arXiv-subject moderator, you can know. But in that case, you would break arXiv-private policy.

(2) To my best knowledge, arXiv papers can be used to claim authors’ original research work or ideas; a typical example is the paper by Grigori Perelman in November 2002, a proof of Thurston's geometrization conjecture, including the Poincaré conjecture as a particular case, which has never been published in a peer-reviewed journal (https://en.wikipedia.org/wiki/ArXiv). Do you think this arXiv paper by Perelman is “not representative of the literature”? I am afraid that your second criticism is not valid either.

(3) If you think “reactionless drive has been so soundly discredited it doesn't need citations”, you can take your proper action to help the community, because WIKI dictionary is everybody’s dictionary. — Preceding unsigned comment added by Smithwikilover (talk • contribs) 17:54, 5 December 2016 (UTC)

To your (1) I don't understand what's unclear about saying if there is no peer reviewed publication of an article that is on arXiv, then it didn't make it past pre-print on arXiv. Scientific papers are more or less public information, paywalls aside. Related to this and your point (2), arXiv is not peer reviewed, and doesn't guarantee science of any good quality. Hence my scepticism of papers that were never published elsewhere, which suggests a failure to pass peer review. To (3) I don't understand what you're trying to say here, but reactionless drive is not credible at all in the physics community, the most recent claim of it, the Eagleworks drive has been torn to shreds for having poor methodology. Also you have completely sidestepped by question about why one of the most cited papers on this subject is neglected from the article, and finally please stop removing the original synthesis tag just because you don't like it, disagreeing editors are supposed to reach consensus.--ObscureFruits (talk) 03:48, 6 December 2016 (UTC)

Just to clarify, I put a tag and commented on the talk page rather than diving in to try and avoid an edit war in the first placeObscureFruits (talk) 03:50, 6 December 2016 (UTC)

Reply to ObscureFruits-talk, 03:50, 6 December 2016 (UTC):

(1) How do you know “if there is no peer reviewed publication of an article that is on arXiv, then it didn't make it past pre-print on arXiv (in several cases)”? For example, I have some arXiv papers, which were never rejected by arXiv, but were never submitted to any peer-reviewed journals either. From this, I can conclude that you cannot derive “whose papers never made it past arXiV in several cases" just from “they have no version besides the one on arXiV”. Being published in arXiv and being published in a peer-reviewed journal, the two events have no causality at all. Of course, as I mentioned, if you are arXiv-subject moderator, you can know. But in that case, you would break arXiv-private policy. (Honesty is the best policy, please.)

(2) “arXiv is not peer reviewed, and doesn't guarantee science of any good quality.” ---- It’s true. arXiv papers are not peer-reviewed, but they are published papers. No journals can guarantee that every of their articles has “science of any good quality”, and no one can ensure that all arXiv articles, which have no peer-reviewed journal versions, do not have “science of any good quality”. A typical example is the paper by Grigori Perelman in November 2002, as I mentioned, which has never been published in a peer-reviewed journal (https://en.wikipedia.org/wiki/ArXiv). It’s possible that Perelman’s arXiv paper “suggests a failure to pass peer review”, but we cannot conclude that it “doesn't guarantee science of any good quality”. In a sense, the arXiv repository is a good measure to prevent peer-reviewed journals from possible academic corruptions; Perelman provided us a persuasive example.

(3) If you think “reactionless drive is not credible at all in the physics community, the most recent claim of it, the Eagleworks drive has been torn to shreds for having poor methodology”, you are encouraged to take your proper action to help the community. WIKI dictionary is everybody’s dictionary.

You mentioned

”Also you have completely sidestepped by question about why one of the most cited papers on this subject is neglected from the article, and finally please stop removing the original synthesis tag just because you don't like it, disagreeing editors are supposed to reach consensus.”

My question is: Why do you not take your action to add that article, “one of the most cited papers on this subject”? WIKI dictionary is everybody’s dictionary. All “disagreeing editors” have equal right to edit this wiki article to benefit the community. As an article about controversial physics problems, different views are supposed to be co-existent, as it is now. We are all of “disagreeing editors” each other.

The tag you put is misleading, and the comment on the talk page is ambiguous. The statement “possibly contains previously unpublished synthesis of published material that conveys ideas not attributable to the original sources” is not specific. What is “possibly”? Where is “unpublished synthesis”? Why is “not attributable”? Could you please indicate specifically what, where, and why? Based on above reasons and to avoid misleading readers, I will add two fixings: (a) “ObscureFruits argues that” and (b) “, but Smithwikilover disagrees”.

“The Abraham–Minkowski controversy is a physics debate”, as stated at the beginning of the wiki article, and different views can be co-existent; accordingly, it is not necessary that “disagreeing editors are supposed to reach consensus”. That is why I never delete or remove the different academic views provided by all other scientists. — Preceding unsigned comment added by Smithwikilover (talk • contribs) 17:26, 6 December 2016 (UTC)

To 128.135.100.105

Requesting Editor 128.135.100.105 to pay attention to the definition of Poynting vector: Your presented definition of

“the Poynting vector, given by S=μ^-1
₀E×B”

is not standard, and it is very confusing. In Poynting's original paper and in many textbooks,

the Poynting vector is defined as S=E×H.

To Editor 128.135.100.105: By citing Barnett’s 2010 PRL paper, you argue:

”the Minkowski version being the canonical momentum p_Canonical=m v+q A, where A is the magnetic vector potential”

(i) I don’t think this is what Barnett means.

(ii) Your expression is for the canonical momentum of a charged particle in a magnetic field. According to the theory of gauge invariance, this canonical momentum is not unique, or indeterminate, and it is not an observable quantity. Thus I don’t think your understanding of Minkowski photon momentum is consistent with the definition given by Barnett in his Letter, reading:

“the canonical momentum of a body is simply Planck’s constant divided by its de Broglie wavelength”.

To Editor 128.135.100.105: By citing Griffiths’ AJP paper, you argue:

”There are different ways to apportion the "kinetic" and "electromagnetic" portions of a medium-photon system's momentum, because the relativistic symmetries of Maxwell's equations do not preserve stress-energy.”

I don’t think you understand what Prof. Griffiths is talking about.

(a) What do you mean for “apportion the ‘kinetic’ and ‘electromagnetic’ portions of a medium-photon system's momentum”? --- The wording of “kinetic” and “electromagnetic” portions is confusing; it should be “matter” and “electromagnetic” portions.

(b) What do you mean for “the relativistic symmetries of Maxwell's equations do not preserve stress-energy”? Are there any causality between (a) and (b)?

Prof. Griffiths’ original statements are very clear, reading:

”In the presence of matter the electromagnetic stress-energy tensor by itself is not conserved (divergenceless). Only the total stress-energy tensor carries unambiguous physical significance, and how one apportions it between an ‘electromagnetic’ part and a ‘matter’ part depends on context and convenience.”

which are exactly expert words.