Aether theories which are experimentally indistinguishable from SR.

[quantumdude]

I cited the references,

And if I chase down the reference, will I find what I asked for? Will I find an aether theory that is experimentally indistinguishable from SR?

To everyone:
Come on guys, this is your big chance! Show us that there actually exist aether theories that yield predictions that are as good as those of SR. It shouldn't be too hard, because so many people have told me that such theories exist.

and never said that this was my preferred alternative theory. Nevertheless, I agree whole-heartedly that a successful alternative theory should describe a more interesting space-time interval invariance.

OK

I don't feel guilty of launching "just a smokescreen using passages taken out of their context", but if you can show a better context for these passages then please do.

I regard your use of the remark as a smokescreen because it puts the stamp of an authoritative figure on the general idea of "aether theory", but at no point is an attempt made to actually cite an aether theory. Without the latter, the former is pretty hollow.

As far as putting the article in context goes, I'll have to read the entire article to get Dirac's point completely. But I know full well that Dirac understood that an aether (as a preferred frame) is not required either by QFT or by the quantum mechanical equation that bears his name.

Since gravity, all of the matter in the universe, and all of the energy in the universe, for example, "live in the error bars" of all SR experiments, then it is misleading to suggest that "the high accuracies achieved by many of these experiments" has somehow foreclosed on the possibility of ever detecting a locally preffered frame.

But gravity and matter have demonstrable existential qualities. This stands in stark contrast to the phantom aether. We have a good idea of how gravity and matter affect the outcome of experiments, and in particle accelerators their affect is expected to be negligible. And not surprisingly, the results all come out to be precisely what the relativistic theory predicts.

What I want to know is this: Where is the aether theory that predicts the same things?

That shouldn't be too difficult to answer, should it?

 

[wisp]

Aether

Great! De Witte has come to interpret his results as confirmation of the one-way isotropy of the speed of light, and an anisotropy of proper time. What of this, wisp?

Inputting the ether speed V=360km/s and electrical signal speed w=200,000km/s into basic equations that model light speed as one-way anisotropic does produce a peak-to-peak phase shift of around 24nS. So I believe his hypothesis on the behaviour of light is wrong. However, his results are experimentally sound and merit further investigation.

 

[Aether]

And if I chase down the reference, will I find what I asked for? Will I find an aether theory that is experimentally indistinguishable from SR?

To everyone:
Come on guys, this is your big chance! Show us that there actually exist aether theories that yield predictions that are as good as those of SR. It shouldn't be too hard, because so many people have told me that such theories exist.

I regard your use of the remark as a smokescreen because it puts the stamp of an authoritative figure on the general idea of "aether theory", but at no point is an attempt made to actually cite an aether theory. Without the latter, the former is pretty hollow.

As far as putting the article in context goes, I'll have to read the entire article to get Dirac's point completely. But I know full well that Dirac understood that an aether (as a preferred frame) is not required either by QFT or by the quantum mechanical equation that bears his name.

Locally consistent with Lorentz transforms, it seems like; but indistinguishable from SR is so loaded. Here is a final caveat on that aether theory by P.A.M. Dirac, Nature, 169, 1952, p.702:

"The existence of an aether has not been proved, of course, because my new electrodynamics has not yet justified itself. It will probably have to be modified by the introduction of spin variables before a satisfatory quantum theory of electrons can be obtained from it, and only after this has been accomplished will one be able to give a definite answer to the aether question."

But gravity and matter have demonstrable existential qualities. This stands in stark contrast to the phantom aether. We have a good idea of how gravity and matter affect the outcome of experiments, and in particle accelerators their affect is expected to be negligible. And not surprisingly, the results all come out to be precisely what the relativistic theory predicts.

What I want to know is this: Where is the aether theory that predicts the same things?

That shouldn't be too difficult to answer, should it?

OK, any alternative theory that doesn't predict some "demonstrable existential qualities" that are at least as palpable as gravity and matter in their ability to "affect the outcome of experiments, and in particle accelerators", is a lame duck theory.

Inputting the ether speed V=360km/s and electrical signal speed w=200,000km/s into basic equations that model light speed as one-way anisotropic does produce a peak-to-peak phase shift of around 24nS. So I believe his hypothesis on the behaviour of light is wrong. However, his results are experimentally sound and merit further investigation.

This De Witte experiment claims to detect some clearly satisfying "existential qualities", but they aren't quite "demonstrable" as of yet.

wisp, De Witte says that those sinusoids are 90-degrees out of phase with what they should be if the speed of light were anisotropic. Do those basic equations you mentioned account for phase?

 

[pervect]

Until DeWitte's result is replicated, we won't be able to rule out experimental error. Here is one example of a very simple sort of error that could explain DeWitte's results

We have one group of clocks, sitting next to a honking big power amplifier, that's driving miles of cable. These clocks are all phase-locked to the power amplifier (whose phase is determined by whatever clock or signal is fed to its input), and show no phase variation relative to each other.

The clocks at the other end of the cable aren't sitting next to a large power amplifier, so they are actually keeping independent time. Because they are not perfectly temperature compensated, they day/night cycle of temperature causes them to gain or lose time with respect to the solar day in comparison with the first set of clocks.

Throw in some optimistic analysis that falsely claims to distinguish the sidereal day from the solar day, and you have DeWitte's results.

Whether or not this particular explanation is true is hard to say at this point. What we can say is that unless these results can be repeated by independent experimenters, they don't mean very much. Except perhaps as an article for "The Journal of Irreproducible Results" :-).

 

[wisp]

wisp, De Witte says that those sinusoids are 90-degrees out of phase with what they should be if the speed of light were anisotropic. Do those basic equations you mentioned account for phase?

Yes, a shift in RA of 6hrs will not affect the magnitude of the phase shift times calculated. It just means that the direction of the ether flow is not the same as the cosmic background radiation.

As far as I can see, DeWitte only gives a RA for ether direction, and doesn’t give a value for its Dec.

If the ether flow (V) is in the ecliptic plane then the simple model’s predictions confirm his experiment.

Ether E/W V=360km/s, w=200,000,000, t=25nS phase shift time
Ether E/W V=360km/s, w=100,000,000, t=99nS
Ether E/W V=500km/s, w=200,000,000, t=34nS
Ether E/W V=500km/s, w=100,000,000, t=138nS

However, if the ether wind is perpendicular to the ecliptic plane as suggested by Dayton Miller in 1933, then the model predicts a stronger ether flow (V) and a slower speed (w) for the electrical signals in the coaxial cable.

Ether N/S V=360km/s, w=200,000,000, t=4.2nS
Ether N/S V=360km/s, w=100,000,000, t=17.2nS
Ether N/S V=500km/s, w=200,000,000, t=6nS
Ether N/S V=500km/s, w=100,000,000, t=24nS

Repeating the experiment with the cable running East – West could resolve matters.

 

[JesseM]

Here's a good post on why Ether theories which are experimentally indistinguishable from SR do not really have the same explanatory power as SR:

http://groups-beta.google.com/group/sci.physics.relativity/msg/a6f110865893d962

 

[Project]

I’d like to know where Selleri’s theory can be found. I tried Googling it, but there’s nothing out there.

http://arxiv.org/abs/gr-qc/0409105

 

[wisp]

Calculations corrected

Originally Posted by Aether
wisp, De Witte says that those sinusoids are 90-degrees out of phase with what they should be if the speed of light were anisotropic. Do those basic equations you mentioned account for phase?

Yes, a shift in RA of 6hrs will not affect the magnitude of the phase shift times calculated. It just means that the direction of the ether flow is not the same as the cosmic background radiation.

As far as I can see, DeWitte only gives a RA for ether direction, and doesn’t give a value for its Dec.

If the ether flow (V) is in the ecliptic plane then the simple model’s predictions closely match his experiment for t=24nS.

Ether E/W V=360km/s, w=200,000,000, t=11nS phase shift time
Ether E/W V=360km/s, w=100,000,000, t=43nS
Ether E/W V=500km/s, w=200,000,000, t=15nS
Ether E/W V=500km/s, w=100,000,000, t=60nS

However, if the ether wind is perpendicular to the ecliptic plane as suggested by Dayton Miller in 1933, then the model predicts a stronger ether flow (V) and a slower speed (w) for the electrical signals in the coaxial cable.

Ether N/S V=360km/s, w=200,000,000, t=4.2nS
Ether N/S V=360km/s, w=100,000,000, t=17.2nS
Ether N/S V=500km/s, w=200,000,000, t=6nS
Ether N/S V=500km/s, w=100,000,000, t=24nS


The DeWitte's results suggest that the flow must be more closer to the ecliptic plane, than perpendicular to it. And larger phase shift time values occur if the electrical signals travel slower than the assumed 200,000km/s.

 

[Aether]

"Thus the much debated question concerning the empirical equivalence of special relativity and an ether theory taking into account time dilation and length contraction but maintaining absolute simultaneity can be answered affirmatively." - R. Mansouri, and R.U. Sexl, A Test Theory of Special Relativity: I. Simultaneity and Clock Synchronization, General Relativity and Gravitation, Vol. 8, No. 7 (1977), p. 512.

The Mansouri-Sexl test theory is referred to in most, if not all, of the subsequently published tests of local Lorentz invariance.

This paper does a good job of explaining that the Lorentz transformation is but one special case of a more general linear transformation, and that it is only singled out by specifying the use of Einstein's clock synchronization procedure which is a purely conventional move.

Re: the "wrong claims" thread overlooking this forum
This link http://math.ucr.edu/home/baez/RelWWW/wrong.html leads to a page entitled "General Relativity Is An Aether Theory" wherein it is claimed that: "Einstein was criticizing people who claimed, in effect, that the classical notion of the aether was such nonsense that people like Maxwell should have known better. He was saying that the problem with the classical aether was not ontological, merely that it is inconsistent with observation and experiment; hence the need for str."

I submit that this is in and of itself a demonstrably wrong claim since the aether is clearly not inconsistent with observation and experiment, and as elegant and as simplifying an approximation to local reality as it may seem to be, there is no fundamental "need for str".

 

[DrChinese]

...I submit that this is in and of itself a demonstrably wrong claim since the aether is clearly not inconsistent with observation and experiment, and as elegant and a simplifying approximation to local reality as as it may be, there is no fundamental "need for str".

This couldn't be Einstein's intent. If you believe in SR and also believe in an ether, well, OK. But SR is still useful, and the Newtonian viewpoint still needed modification. Which was Einstein's point.

You can't realistically assert that the ether exists AND we should go back to Newtonian thinking. So there is definitely a need for SR. Does SR require that the ether be non-existent? I don't see it that way at all. Especially considering that our view of a vacuum has changed in the intervening years.

 

[Aether]

This couldn't be Einstein's intent.

I presume that the wrong claim originates with Chris Hillman (the author of the linked-to article) rather than with Einstein. For example, this article starts out with "Albert Einstein, in his essay On the Aether (1924), made some injudicious comments...", then continues later with "What Einstein really meant was that the aether which had been overthrown by str (and thus was incompatible with gtr, which incorporates str) involved a a specific "preferred frame of reference"" which is a demonstrably false statement. The aether has clearly not been "overthrown by str", and is not "incompatible with gtr" as apparently Einstein himself intends, by writing this essay, to caution his readers about in the first place!

If you believe in SR and also believe in an ether, well, OK. But SR is still useful, and the Newtonian viewpoint still needed modification. Which was Einstein's point.

You can't realistically assert that the ether exists AND we should go back to Newtonian thinking. So there is definitely a need for SR. Does SR require that the ether be non-existent? I don't see it that way at all. Especially considering that our view of a vacuum has changed in the intervening years.

I could assert that "the ether system exists". On page 505 of the Mansouri-Sexl paper that I referenced earlier you will find the following statement: "Summarizing these results we may say that the following statement is in perfect agreement with all experimental evidence: A preferred system of reference, the ether system, exists.", but I would never assert that we should go back to Newtonian thinking. SR per se is not what stands between us and Newtonian mechanics, it is one special case of a larger class of theories which accomplish precisely the same thing, locally. The danger is this: SR has turned into a scientific dogma because it simplifies certain elementary computations, but that doesn't necessarily make it the right tool with which to ponder the larger Universe.

 

[CarlB]

Here's a good post on why Ether theories which are experimentally indistinguishable from SR do not really have the same explanatory power as SR:

http://groups-beta.google.com/group/sci.physics.relativity/msg/a6f110865893d962

That was a good link. The author notes that ether theories, and particularly the LET, are compatible with observation. His argument against ether theories is that ether theories don't include the symmetry that Einstein's theory did, and so would not be as predictive. This really isn't quite true.

Sure you can make an arbitrary ether theory that loses symmetry, but I think the best ether theory is the one that brings proper time into the geometry of space-time as a hidden dimension. This gives space-time 5 dimensions, 4 spatial and one temporal, and it also implies a preferred reference frame.

In that theory, Lorentz symmetry arises not as the result of an accidental symmetry, but instead due to the quite natural assumption that the hidden dimension is symmetric to the 3 spatial dimensions. That is, when you consider the four spatial dimensions of the theory, they form, for any given value of time, a manifold locally equivalent to R^4. Rotations in that R^4 correspond to the usual boosts and rotations of Lorentz symmetry.

What I'm saying here is that just because most ether theories (or perhaps all the ether theories that the author considered) do not have a natural explanation for Lorentz symmetry doesn't mean that no ether theory does.

Carl

 

[Mortimer]

Carl is referring to Euclidean special relativity and, of course, I won't let an opportunity go by to support and promote that point of view.

Rob

 

[CarlB]

Mortimer, do you agree that Euclidean special relativity is a type of Lorentz ether theory?

The reason I ask is because it is possible to read a lot of ESR papers without seeing any mention of "ether" or "preferred reference frame", but at least one other person doing work in the field also believes that a preferred reference frame is implied by ESR.

Carl

 

[JesseM]

Sure you can make an arbitrary ether theory that loses symmetry, but I think the best ether theory is the one that brings proper time into the geometry of space-time as a hidden dimension. This gives space-time 5 dimensions, 4 spatial and one temporal, and it also implies a preferred reference frame.

Are you assuming the 4th spatial dimension is a compact one? If not, why do we seem to have only 3 degrees of movement? And if so, then is the "preferred reference frame" the same as the one where the diameter is maximized?

 

[CarlB]

Are you assuming the 4th spatial dimension is a compact one? If not, why do we seem to have only 3 degrees of movement? And if so, then is the "preferred reference frame" the same as the one where the diameter is maximized?

Different people treat the proper time dimension differently. I treat it as compact for the reasons you give.

Instead of the diameter of the compact dimension, if I'm going to talk about measurements along the hidden dimension I prefer to measure its circumference. Otherwise you have to define an embedding space or some such. Unfortunately, as with other ether theories, there is no way for us to measure distances in the hidden dimension, so there is no way for us to determine the preferred reference frame (at least with normal matter and light).

What I'm trying to say here is that the death of the preferred reference frame and ether was not from Einstein's showing that it was unneeded. In fact, a preferred reference frame / ether had never been needed in that no one ever had any idea where such a frame existed. They were needed only as a philosophical item in that it could not otherwise be imagined how vibrations could move through a vacuum.

Instead, the death thrust to the preferred reference frame / ether was the requirement that vibrations in it (i.e. light) satisfy the Lorentz transformations for different reference frames. This is a very special symmetry. But if that symmetry arises naturally from a hidden dimension, then the philosophical support for the preference frame / ether returns.

I wrote up a very short and simple 2 page paper on the phase velocity of de Broglie waves and the "proper time geometry", which is what I call Euclidean relativity that may give a better idea of why one naturally thinks of a preferred reference frame when using Euclidean relativity, but does not when using special relativity:
http://brannenworks.com/a_phase.pdf
The paper looks ugly in some versions of acrobat but it prints cleanly. It's not my current thinking on the subject, but it does give a suggestion of how QM and Euclidean relativity interact.

I should note that I wrote the above before I appreciated how many other people were working in the area, so there aren't nearly enough references included. Euclidean relativity is so natural that many authors, myself included, came upon it without knowing that others had already discovered it. The newer (and better) papers deal with general relativity, but my own interest is in elementary particles and fields. So naturally, I came to Euclidean relativity through an exploration of standard model particle symmetries.

Carl

 

[Mortimer]

Mortimer, do you agree that Euclidean special relativity is a type of Lorentz ether theory?

The reason I ask is because it is possible to read a lot of ESR papers without seeing any mention of "ether" or "preferred reference frame", but at least one other person doing work in the field also believes that a preferred reference frame is implied by ESR.
Carl

I guess you're referring to Hans Montanus' AEST (Absolute Euclidean Space Time), which is such a preferred reference frame. In my opinion that is a perfectly valid mathematical framework but I am not in favor of it although I highly appreciate his articles. The point is that there still is no way to determine which frame is the preferred one. Each individual observer's frame still is equally valid as a candidate.

My own opinion on the matter is that there is indeed a preferred frame implied by ESR but it is 5-dimensional, like you said. This means that no single 4-dimensional frame can be preferred. Compare this for instance with a 3-dimensional space holding multiple 2-dimensional planes with different orientations. Each plane is on equal footing with all other planes from the perspective of the 3D space but there is only one 3D space which is the preferred frame.
For a human observer his own 4D space-time in ESR is the equivalent of such a plane, embedded in a 5D space-time that is "fixed" for each human observer. The relative orientation of the 4D environment in the 5D frame is determined by the Lorentz boost, or rotation in SO(4).

Like the 2D plane analogy, this 4D analogy only leads to a preferred frame in 5D if observational skills are limited to 4D (like for us humans). A hypothetical being with 5D observational skills would on its turn treat his 5D space-time as embedded in a 6D frame that would then be the preferred frame, and so on. You could see this as a kind of multi-dimensional fractal.

Rob

 

[CarlB]

It may be of interest that S. W. Hawking's latest paper on black holes uses Euclidean gravitation theory:

"I adopt the Euclidean approach [5], the only sane way to do quantum gravity nonperturbatively."
http://www.arxiv.org/abs/hep-th/0507171

The link is from Peter Woit's "not even wrong" blog:
http://www.math.columbia.edu/~woit/blog/

Carl

 

[clj4]

And if I chase down the reference, will I find what I asked for? Will I find an aether theory that is experimentally indistinguishable from SR?

To everyone:
Come on guys, this is your big chance! Show us that there actually exist aether theories that yield predictions that are as good as those of SR. It shouldn't be too hard, because so many people have told me that such theories exist.

Tom,

Such theories DO NOT exist. This was debated more recently in two-three other threads:

1. The aether theories are not sufficiently developed (they stop at kinematics and lack the electromagnetism)

2. There are experiments (in the electromagnetic space, of course!) that point out the differences between such aether theories and SR. So far I have identified 8 such experiments.

3. As shown by CM Will in a paper dating from 1992, some "aether" theories can be contrived to appear indistinguishable from SR via "adding AD-HOC assumptions that may or may not be true" (cited from the CM Will paper). See also point 6 in Tom Roberts posting:

http://groups.google.com/group/sci.physics.relativity/msg/a6f110865893d962Well, what else is new? We have known all along that the demise of the early aether theories was caused by the requirement of adding an ad-hoc explanation for each new experiment.

So yes, you are right, this is all a smokescreen: there is no "aether" theory that is "experimentally indistinguishable" from SR

 

[clj4]

Tom - here is one citation - don't get the idea I totally endorse everything this guy says - I think he is wrong on his views about Sagnac and GR - but - anyway:
http://redshift.vif.com/JournalFiles/V11NO1PDF/V11N1SEL.pdf.

Yes, this is published in a journal that is well known as a "fringe" (as bordering on crank). Anyways, John Baez and Zhang refer to the Mansori-Sexl theory, published in 3 papers in 1977. This is a serious aether theory that picks up where Robertson (a professor at Caltech) left off.

 

[clj4]

There has got to be an aether theorist willing to fork over the $10,000 or so required to do a tabletop 1-way test. Why hasn't one been done yet? Are they afraid of the answer?

He-he. All the one way light speed experiments (there are 8 papers on them but more are coming) are done by relativists. The results are very clear: no anisotropy.

 

[clj4]

One way velocities are strictly a statement about coordinate systems. Coordinate systems don't have any ultimate physical significance - that's why one-way velocities don't, either.

So basically getting hung up on the issue of one-way velocities is a dead end. It doesn't tell you anything more, or different, about what you can measure. Furhtermore, systems with non-isotropic one-way velocities are more complicated to deal with. As I've remarked in another thread, there can be some justification for changing units or coordinates to make the mathematics and the exposition of a theory simpler. In this case, changing to non-isotropic coordinates makes the math and the exposition of a theory harder, not simpler. So it's pretty much a lose-lose proposition.

Correct. For example: Mansouri and Sexl stopped after the kinematic section. (no dynamic and no electromagnetism). One member of thE Gagnon team, P.Chang went ahead and demonstrated how Maxwell equations transform in the framework of Mansouri-Sexl, a nightmare of ugliness.

 

[rbj]

please forgive my ignorance, but how can an aether theory be compatible with SR (or GR) since movement through the aether at sufficient velocity would result in a measurable difference in the speed of light propagating in different directions? is not the isotropic propagation of light axiomatic to SR?

i s'pose an ad hoc exception to make an aether theory work with the results of the Michaelson-Morley experiment is some theory that the aether travels along with the Earth so the Earth never moves through it at any velocity that would make the M-M experiment come out differently. unless one were to believe that the Earth is at the center of reality to explain why the aether is stuck to it, one would have to contrive a theory that the hypothetical aether sticks locally to massive objects, sort of like the curvature of space-time in GR. then, i guess to disprove that, human beings would have to spend a few zillion dollars sending an M-M apparatus up in the shuttle, as far from Earth as possible and showing that light is just as isotropic there.

to my ignorant mind (i'm just a lowly electrical engineer) aether means non-isotropic radiation of E&M for anyone moving through the aether and, besides never having properties that lend themselves to being measured, this seems totally inconsistent with the wave solution to Maxwell's Eqs. in a vacuum. this is why, i thought, that Einstein believed c to be constant, whether or not he was aware of the M-M experiment in 1905.

 

[JustinLevy]

please forgive my ignorance, but how can an aether theory be compatible with SR (or GR) since movement through the aether at sufficient velocity would result in a measurable difference in the speed of light propagating in different directions?

I have ran into some aetherists that agree with the results of SR, but have some metaphysical belief that there is a one real frame. Basically, they claim that there is one real frame and all calculations must be done in this frame (and the lengths, times, velocities in this frame are the \"real\" quantities, with those being measured in other frames just being \"mathematical conveniences\"). This is just metaphysical, so I see no reason we can\'t dismiss it immediately.

Of course the calculations can be done in any frame, so this will indeed agree with SR. But it is not a valid theory because it adds the metaphysics of claiming only one frame is real, even though we could arbitrarily choose this frame (or equivalently, their theory doesn\'t allow a way to deterime which frame is the real one).

So such a theory will agree with experiment, but is not scientific.

is not the isotropic propagation of light axiomatic to SR?

In inertial frames (frames with the Minkowski metric), yes. But for frames with other metrics, there is no requirement that the speed of light be isotropic (for example in GR). But this does not indicate that GR is an aether theory. In fact, it now allows any coordinate system to be used. The exact opposite of a perferred frame.

i s\'pose an ad hoc exception to make an aether theory work with the results of the Michaelson-Morley experiment is some theory that the aether travels along with the Earth so the Earth never moves through it at any velocity that would make the M-M experiment come out differently. unless one were to believe that the Earth is at the center of reality to explain why the aether is stuck to it, one would have to contrive a theory that the hypothetical aether sticks locally to massive objects, sort of like the curvature of space-time in GR. then, i guess to disprove that, human beings would have to spend a few zillion dollars sending an M-M apparatus up in the shuttle, as far from Earth as possible and showing that light is just as isotropic there.

Aether dragging ideas do not fit with SR (or experimental evidence).

And if you are curious, a M-M apparatus has already been sent up to space. Also, the Gravity Probe B, will be releasing its results soon which tests relativity away from the Earth's surface.

 

[Garth]

And if you are curious, a M-M apparatus has already been sent up to space. Also, the Gravity Probe B, will be releasing its results soon which tests relativity away from the Earth's surface.

See Alternative theories being tested by Gravity probe B.

Garth

 

[clj4]

please forgive my ignorance, but how can an aether theory be compatible with SR (or GR) since movement through the aether at sufficient velocity would result in a measurable difference in the speed of light propagating in different directions? is not the isotropic propagation of light axiomatic to SR?

i s'pose an ad hoc exception to make an aether theory work with the results of the Michaelson-Morley experiment is some theory that the aether travels along with the Earth so the Earth never moves through it at any velocity that would make the M-M experiment come out differently. unless one were to believe that the Earth is at the center of reality to explain why the aether is stuck to it, one would have to contrive a theory that the hypothetical aether sticks locally to massive objects, sort of like the curvature of space-time in GR. then, i guess to disprove that, human beings would have to spend a few zillion dollars sending an M-M apparatus up in the shuttle, as far from Earth as possible and showing that light is just as isotropic there.

to my ignorant mind (i'm just a lowly electrical engineer) aether means non-isotropic radiation of E&M for anyone moving through the aether and, besides never having properties that lend themselves to being measured, this seems totally inconsistent with the wave solution to Maxwell's Eqs. in a vacuum. this is why, i thought, that Einstein believed c to be constant, whether or not he was aware of the M-M experiment in 1905.

rbj

The whole thing started in 1949 when a very respected professor at Caltech (Robertson) produced a kinematic alternative to SR. It was continued in 1977 , by Mansouri and Sexl . They produced a so-called "test theory" of SR. This is also only a kinematic section only (no dynamics and no electromagnetism).
The explanation is long and complicated, the bottom line is that only "aetherists" interpret the test theories of Mansouri and Sexl as "alternatives" to SR. M&S certainly did not view their theory as an alternative to SR. One side effect of the MS theory is that one way light speed is isotropic ONLY in ONE reference frame, in all other reference frames it is anisotropic. We can certainly test this and, by showing this to be false, we can disprove the MS theory. This is very different from the alternatives to GR which are true alternatives.
Again, bottom line is that there is a handful of experiments (8 , so far but more are coming) that put very severe error bars on one way light speed anisotropy. Thus, thru the MS "test theory" we get an even better confirmation of SR's validity.
The MS theory is very clever in that it assumes one way light speed to be anisotropic. The anysotropy is "crafted" in such a way that it cancells out in two-way experiments (such as MMX). This is why one way light speed experiments have become key in refuting the MS theory (In SR, one way light speed is , of course, isotropic).
Hope that this helps.

 

[Aether]

please forgive my ignorance, but how can an aether theory be compatible with SR (or GR) since movement through the aether at sufficient velocity would result in a measurable difference in the speed of light propagating in different directions? is not the isotropic propagation of light axiomatic to SR?

One-way speeds are not measurable in a coordinate-system independent way.

i s'pose an ad hoc exception to make an aether theory work with the results of the Michaelson-Morley experiment is some theory that the aether travels along with the Earth so the Earth never moves through it at any velocity that would make the M-M experiment come out differently. unless one were to believe that the Earth is at the center of reality to explain why the aether is stuck to it, one would have to contrive a theory that the hypothetical aether sticks locally to massive objects, sort of like the curvature of space-time in GR. then, i guess to disprove that, human beings would have to spend a few zillion dollars sending an M-M apparatus up in the shuttle, as far from Earth as possible and showing that light is just as isotropic there.

The Michelson-Morley experiment measures two-way light speed isotropy. This does not distinguish between SR and Lorentz ether theory.

to my ignorant mind (i'm just a lowly electrical engineer) aether means non-isotropic radiation of E&M for anyone moving through the aether and, besides never having properties that lend themselves to being measured, this seems totally inconsistent with the wave solution to Maxwell's Eqs. in a vacuum. this is why, i thought, that Einstein believed c to be constant, whether or not he was aware of the M-M experiment in 1905.

Not at all. See T. Chang, Maxwell's equations in anisotropic space, Physics Letters 70A(1), 1 (1979). See also Eq. (5) of http://imaginary_nematode.home.comcast.net/papers/Gagnon_et_al_1988.pdf" et al. (but ignore the rest of the paper) which is the vacuum wave equation obtained from this formulation of Maxwell's equations.

 

[Aether]

Of course the calculations can be done in any frame, so this will indeed agree with SR...

In inertial frames (frames with the Minkowski metric), yes. But for frames with other metrics, there is no requirement that the speed of light be isotropic (for example in GR). But this does not indicate that GR is an aether theory. In fact, it now allows any coordinate system to be used. The exact opposite of a perferred frame.

All one-way speeds are coordinate-system dependent, but two-way speeds aren't. clj4 disagrees with this.

But it is not a valid theory because it adds the metaphysics of claiming only one frame is real, even though we could arbitrarily choose this frame (or equivalently, their theory doesn\'t allow a way to deterime which frame is the real one).

So such a theory will agree with experiment, but is not scientific.

Claiming "only one frame is real" is not scientific, and neither is claiming that experiments prove that one-way light speed is isotropic. Experiments may one day detect violations of local Lorentz symmetry, and there is no problem with theories that predict such a violation as long as they are consistent with experiments to date.

Aether dragging ideas do not fit with SR (or experimental evidence).

Correct.

 

[clj4]

One-way speeds are not measurable in a coordinate-system independent way.

Repeating false statements does not constitute physics. Nor does it make the respective false statements true.

 

[ZapperZ]

One-way speeds are not measurable in a coordinate-system independent way.

The Michelson-Morley experiment measures two-way light speed isotropy. This does not distinguish between SR and Lorentz ether theory.

Not at all. See T. Chang, Maxwell's equations in anisotropic space, Physics Letters 70A(1), 1 (1979). See also Eq. (5) of http://imaginary_nematode.home.comcast.net/papers/Gagnon_et_al_1988.pdf" et al. (but ignore the rest of the paper) which is the vacuum wave equation obtained from this formulation of Maxwell's equations.

I'm sure this paper has been pointed out, but can you please write a rebuttal to this paper in PRA and submit it to that journal? I looked at the citations to this paper, and there was not even ONE paper disputing either their physics, nor their claim, to being able to determine the one-way speed of light. And this isn't a new paper either!

P. Wolf and G. Petit, PRA v.56, p.4405 (1997).

If you think they have made an erroneous claim, then it is your responsibility to respond to it. Complaining about it on some open forum is not going to cut it.

Zz.

 

[Hans de Vries]

One might hope that this doesn't become another Mansouri-Sexl thread...
Discussing pure SR won't shed any light on this, no matter how much
one extends the dispute.In Quantum Field Theory only Lorentz Invariant wave equations are
allowed. That is, they must physically reproduce effects like Lorentz
contraction and time dilation.

Maxwell’s equations physically reproduce Lorentz contraction, not time-
dilation because it describes massless particles which do not have any
time progression. The wave equations for particles with mass however
do correctly generate time dilation as well.

Note that everything propagates in these theories. Not only the
electromagnetic field but also matter fields propagate via wave-
equations. This is the big difference with the old ether theories.Regards, Hans.

 

[pervect]

I'm sure this paper has been pointed out, but can you please write a rebuttal to this paper in PRA and submit it to that journal? I looked at the citations to this paper, and there was not even ONE paper disputing either their physics, nor their claim, to being able to determine the one-way speed of light. And this isn't a new paper either!

P. Wolf and G. Petit, PRA v.56, p.4405 (1997).

If you think they have made an erroneous claim, then it is your responsibility to respond to it. Complaining about it on some open forum is not going to cut it.

Zz.

A great deal of the argument is over semantics. It may not have made it to PRA, but there has apparently been a lot of argument over these issues in the journals. One such reference was posted earlier in the thread:

http://arxiv.org/abs/gr-qc/0409105

I think this paper makes many good points and a few bad ones (i.e. I dont' necessarily agree with everything in this paper, though I find the abstract itself pretty much on-track).

The axiomatic bases of Special Relativity Theory (SRT) are thoroughly re-examined from an operational point of view, with particular emphasis on the status of Einstein synchronization in the light of the possibility of arbitrary synchronization procedures in inertial reference frames. Once correctly and explicitly phrased, the principles of SRT allow for a wide range of `theories' that differ from the standard SRT only for the difference in the chosen synchronization procedures, but are wholly equivalent to SRT in predicting empirical facts. This results in the introduction, in the full background of SRT, of a suitable synchronization gauge. A complete hierarchy of synchronization gauges is introduced and elucidated, ranging from the useful Selleri synchronization gauge (which should lead, according to Selleri, to a multiplicity of theories alternative to SRT) to the more general Mansouri-Sexl synchronization gauge and, finally, to the even more general Anderson-Vetharaniam-Stedman's synchronization gauge. It is showed that all these gauges do not challenge the SRT, as claimed by Selleri, but simply lead to a number of formalisms which leave the geometrical structure of Minkowski spacetime unchanged. Several aspects of fundamental and applied interest related to the conventional aspect of the synchronization choice are discussed, encompassing the issue of the one-way velocity of light on inertial and rotating reference frames, the GPS's working, and the recasting of Maxwell equations in generic synchronizations. Finally, it is showed how the gauge freedom introduced in SRT can be exploited in order to give a clear explanation of the Sagnac effect for counter-propagating matter beams.

My $.02 on the issue.

Coordinate independent physics is a good thing.

It is difficult to phrase velocity in any other manner than a coordinate-dependent one, however. Measuring any sort of velocity depends, at a minimum, on adopting some standard for "fair" clock synchronization. (To measure a velocity, you need a starting point, a stopping point, a distance measurement between the two points, one clock each at the starting and stopping point, and a way to synchronize the clocks.)

Rapidity does not have this synchronization problem, interestingly enough. (You mark out a course of a known fixed distance first, then measure the time it takes to transverse the distance with a single clock located on the moving vehicle rather than two clocks in the laboratory frame). However, note that you can't measure the rapidity of light in any event - clocks cannot move at 'c'. So it would make no sense to talk about the "rapidity" of light being isotropic, as light does not have a rapidity :-(.

The Einstein clock synchronization is a reasonable and well-accepted approach to making velocity (at a single point in space-time) a coordinate independent quantity. Velocity at distant locations still runs into parallel transport issues in GR, but at least with Einstein clock synchronization, we can define the velocity at a point as a coordinate independent quantity.

Conceptually one could use other sorts of isotropy other than light as the "standard" synchronization method. However, light is IMO clearly experimentally the best method to use. Therfore it is sensible to adopt light synchronization as the experimental standard, IMO.

[add]
For instance, we noted earlier that rapidities don't have the clock synchronziation problem, so we could, conceptually, say that the way we synchronize clocks is so that equal rapidities have equal velocities. Howver, while we COULD do this, it would IMO be a poor idea. For one thing, we are making an assumption - one which is compatible with SR, but might not be compatible with other theories - that it is possible to synchronize clocks in this manner so that the velocity of ALL clocks of a certain rapidity are isotropic.

Probably the only reasonably clean alternative to Einstein clock synchronization is slow clock transport.

If we accept the Einstein clock synchronization method as a standard, then we cannot actually measure the isotropy of the speed of light. We have defined it to be constant. We can, however, measure the isotropy of other things. The issue one might have with the Wolf paper is not the contents of the paper, but rather what they chose as the title of their paper. Hopefully the paper itself makes clear what they actually measured.

 

[ZapperZ]

A great deal of the argument is over semantics. It may not have made it to PRA, but there has apparently been a lot of argument over these issues in the journals. One such reference was posted earlier in the thread:

http://arxiv.org/abs/gr-qc/0409105

I think this paper makes many good points and a few bad ones (i.e. I dont' necessarily agree with everything in this paper, though I find the abstract itself pretty much on-track)

But my point is this. The paper that I cited was published a while back. I went to look at ALL the subsequent papers that cited this paper. Unless I have an outdate citation index (Scitation seems to be quite good), not ONE challenges either its interpretation, or its physics, semantics or not.

I mean, of all the noises generated here, I'm amazed that no one who opposed such a measurement, be it either they don't buy the result, or they think these authors misinterpreted their results, didn't have one single damn thing to officially rebutt this paper.

It takes zero effort to bad-mouth something on here. It takes well-thought out argument and validity of point to rebutt a PRA paper and have it appear in print. So where are they? [I'm waiting for the "conspiracy" theory to rear its ugly head]

.. and I've only singled out ONE such paper in the literature.

Zz.

 

[clj4]

.. and I've only singled out ONE such paper in the literature.

Zz.

...You are very correct. There are at least 10 more, for a grand total of 11 :

1. 3 by Gagnon, Torr, P.Chang etc

http://prola.aps.org/abstract/PRA/v38/i4/p1767_1

2. one by Krisher

http://prola.aps.org/abstract/PRD/v42/i2/p731_1?qid=6c4ab66eee46e0e8&qseq=4&show=30

3. 2 by CM Will

http://prola.aps.org/abstract/PRD/v45/i2/p403_1?qid=630b0f834f891ba4&qseq=20&show=10

4. one by Gianfranco Spavieri

http://prola.aps.org/abstract/PRA/v34/i3/p1708_1

5. 3 new ones by A.Peters. Hermann, etc

http://www.arxiv.org/PS_cache/physics/pdf/0508/0508097.pdf

There may be more, the above were discovered in the protracted arguments with the "neo-aetherist" group of : "Aether", "wisp" and the twin sock puppets "Gregory/NotForYou".

The reference quoted by pervect is very interesting and valuable indeed. Look carefully at point (iv) in section (1.2). It has a very strong message for the deniers of the validity of one-way light speed measurements. It also has a very strong message for the people that maintain that "MS theories are experimentally indistiguishable from SR". Once we read very carefully the reference quoted by pervect, we can return to the 11 papers dealing with the isotropy of light speed and we will notice a common trend: they all make use of rotational frames! Surprise, surprise!

 

[pallidin]

My, what a fantastically interesting topic! Love it!