Is Lorentz Ether Theory a Viable Alternative to General Relativity?

[Dale]

these things go beyond mere aesthetics.

Then which of your above objections do you believe is either mathematical or empirical? Or do you believe that there is a 4th category of objections to a scientific theory?

PS By your use of the phrase "mere aesthetics" and your earlier comment about "just an aesthetic decision" I think you may incorrectly consider the term aesthetic to be somehow perjorative.

 

[lalbatros]

Tam,

This is not true:

Last, the larger problem I'm tackling is not SR, but GR. SR of course has limited applicability in the real universe because it ignores gravity and accelerating frames.

SR does not ignore gravity or accelerating frames.
Even in Newtonian physics there is a place for gravity or for acceleration.

For example, the concept of centrifugal force and Coriolis force is a result of Newtonian physics.
These concepts are obtained when translating Newtonian mechanics to an accelerating frame (rotating more precisely).
The more general concept of inertial forces is obtained similarly within Newtonian mechanics.
It goes without saying, that SR does not deliver less than Newtonian mechanics.

The motivation for going from SR to GR is more interresting to analyse than the papers from Cahill ...

 

[Vanadium 50]

Vanadium, I'm fine with the criteria listed re judging what is pseudoscience. What I take issue with is Lalbatross's application of those criteria. As a lawyer and scientist, I'm keenly sensitive to how criteria and other rules are applied. Obviously, it's a subjective process. But in this case I find it rather unfair (to say the least) that Lalbatross has labeled Cahill's work as pseudoscience - without even reading Cahill's works in any detail or examining his claims. Again, Cahill is a tenured professor of physics at a great university in Australia. I think he deserves a little credit and his ideas a little credence before being summarily rejected as wrong, or worse: as pseudoscience.

Note that I never mentioned Cahill in my post. You made a statement about Einstein and Darwin. I'm asking you to either back it up, or withdraw it.

 

[Tam Hunt]

Dale, thanks for your substantive response. In order:

Dale: 1)"An ongoing confusion in physics is that absolute motion is incompatible with Lorentz symmetry". This is incorrect, it is well known that LET has both Lorentz symmetry and absolute motion.

Tam: Cahill is referring to Einsteinian SR, not LET. Cahill is correct that the mainstream interpretation of SR denies absolute motion (the ether is “superfluous”) as a consequence of its two key postulates: the principle of relativity and the principle of the constancy of the speed of light.

Dale: 2)"the evidence is that absolute motion is the cause of these relativistic effects, a proposal that goes back to Lorentz in the 19th century. ... Einstein’s postulates regarding the invariant speed of light have always been in disagreement with experiment". Here he makes the same mistake that you and I have discussed at length. You cannot even in principle have experimental evidence which verifies Lorentz and falsifies Einstein.

Tam: Cahill (and I in my earlier posts in this thread) argues in favor of absolute motion based on MMX, Miller and his own experiments. This goes to what I have mentioned a few times now: we can distinguish LET from SR if we agree that these experiments show the Earth traveling through absolute space, as Cahill argues they do. As such, LET instead of SR may be preferred based on empirical evidence. Under this line of reasoning we must, then, consider Cahill's arguments for absolute motion and determine if they are valid.

Dale: 3)The speed of light through a moving medium (V) with a refractive index of n is given by V = c/n + v (1-1/n²). Even ignoring the Frensel drag coefficient it is V = c/n + v, not simply V = c/n.

Tam: I asked Cahill about your objection. Here's his response:

The speed through a gas is c/n when the gas is at rest. When the gas is moving with speed v relative to space the speed is c/n+(v(1-1/n^2) - this is Fresnel drag. I have done experiments checking Fresnel drag in glass! If one applies this to the Michelson gas-mode interferometer one finds that one gets the same Dt time difference between the arms, up to the overall sign, as one does when not including the Fresnel drag term. This change of sign has no significance when actually doing the experiment, or analysing the data. Of course when computing the time to travel a length L of an arm one needs the speed of that arm - v as well. All these effects have been taken into account. They are all discussed in my book.

Dale: 4)There are several post-MM experiments that were performed using some other medium besides vacuum, including one by Michelson and Morely. All verify the Fizeau experiment, which can be derived using SR velocity addition.

Tam: Cahill covers these experiments and many more in his works, particularly in his 2005 book, Process Physics, and long paper from 2003, also called “Process Physics” (see chapter 10): http://www.google.com/url?sa=t&source=web&ct=res&cd=3&url=http%3A%2F%2Fwww.scieng.flinders.edu.au%2Fcpes%2Fpeople%2Fcahill_r%2FHPS13.pdf&ei=PwDQSY3jFKWEtAOFtuSgAw&usg=AFQjCNH7pqqCVarsnqFC9Z9ZDdG1OWu-Dw&sig2=PpzM5KusU4a1OGhzZH2nUQ

Dale: 5)The MM 8 km/s result is, in fact, a statistically null result as is the 10 km/s result of Miller. In their times it was common to not perform a statistical analysis on the experimental errors, but Cahill has no such excuse. When an error analysis is done it can be seen that their results are not significantly different from 0 km/s.

Anyway, Cahill's paper does not stand scrutiny. He does not understand LET and its relationship to SR, the derivation of his mathematical framework is incorrect, he ignores experiments that directly contradict his claims, and he does not have any statistically significant experiment supporting his claims. I would recommend that you read The Experimental Basis of SR. Of course, all of the experimental support listed there also supports LET, as we have discussed.

Tam: Here is Cahill's response when I asked about this issue some time ago:

Miller in his 1933 paper reported (page 238) speed errors at the level of approx 3% and azimuth errors of 2.5 degrees. My analysis of Miller's data gives a maximum speed error of 5% for the god data - i.e that without obvious glitches. That analysis is based on the rms error from fitting the expected form (including temperature drifts, the Hick's effect and expected cos[2*angle] effect), and using the usual definition of rms error. Indeed the quality of his data is truly remarkable. Roberts simply makes up the errors - in the past he had made absurd claims - and continues to do so. His comments are not based on an actual error analysis of the data.

Tam: Re Tom Roberts' list of experimental evidence for SR, most of this evidence, as you have pointed out, supports LET just as well as SR. The difference, as I have pointed out, is that experimental evidence supportive of the notion of absolute space, if correct, would provide a definitive argument in favor of LET instead of SR. Under LET, time dilation isn't ontologically real – it's “just” clock dilation. And length contraction is not caused by the malleability of space itself; rather, it is caused by electromagnetic interaction of charged particles with the ether.

With this understanding, a generalized LET may be developed to replace GR.

 

[Tam Hunt]

Lalbatross, SR doesn't replace Newtonian physics - GR does. SR does not cover gravity or accelerating frames of reference. I'd recommend Einstein's 1916 book on these issues. The 1952 edition is quite good.

 

[Tam Hunt]

Vanadium, as I mentioned it's the application of the criteria that I objected to. By the standard Lalbatross used in applying his acceptable pseudoscience criteria, he could just as well have found that Einstein and Darwin were engaging in pseudoscience. In other words, his application of the criteria was so loose that practically anyone challenging the consensus in his/her time could be labeled a crank or pseudoscientist and, accordingly, their claims ignored.

As I think you will see from the detailed response to Dale above, there is much to Cahill's claims, and they should not simply be ignored.

 

[Dale]

As such, LET instead of SR may be preferred based on empirical evidence.

We have been over this 6 times already by my count. You are wrong. It is impossible to make progress in this discussion when you insist on rewinding it back to the beginning over and over and over again.

If you really think that it is possible to distinguish between LET and SR on the basis of experimental evidence then kindly propose any experimental set up where Lorentz would use the Lorentz transforms to predict one value for a specific measurement and Einstein would use the Lorentz transforms to predict a different value for the same measurement. Please be quantitative in your predictions and explicit in your derivation.

If you cannot produce such a set of conflicting predictions then please stop making this claim. Until you do so we cannot have a meaningful conversation.

 

[lalbatros]

Tam,

As such this statement is too imprecise if you want it to be true:

Lalbatross, SR doesn't replace Newtonian physics - GR does. SR does not cover gravity or accelerating frames of reference.

I explained you why I understand the contrary.
SR can reproduce all the results of Newtonian mechanics, including the whole classical celestial mechanics.
SR also applies in a larger domain than Newtonian mechanics since it can be used for larger speeds up to v=c.

For example, it is well known that SR leads to nearly eliptical orbits for the planets.
There is a small difference however with NM, as the orbit does not close perfectly anymore.
This leads to a drift of the Mercury perihelion in SR.
This drift is smaller than the drift predicted by GR.
We can say that accounting for proper time effect explains partly the perihelion drift.

It would be better to say that SR doesn't bring added value to gravitational physics,
but that SR does cover accelerated frames and gravitation, just as NM also does.

 

[Vanadium 50]

Vanadium, as I mentioned it's the application of the criteria that I objected to. By the standard Lalbatross used in applying his acceptable pseudoscience criteria, he could just as well have found that Einstein and Darwin were engaging in pseudoscience. In other words, his application of the criteria was so loose that practically anyone challenging the consensus in his/her time could be labeled a crank or pseudoscientist and, accordingly, their claims ignored.

In that case, it sounds like you are putting words in Labatross's mouth here.

If you really think that it is possible to distinguish between LET and SR on the basis of experimental evidence then kindly propose any experimental set up where Lorentz would use the Lorentz transforms to predict one value for a specific measurement and Einstein would use the Lorentz transforms to predict a different value for the same measurement. Please be quantitative in your predictions and explicit in your derivation.

Agreed.

 

[Tam Hunt]

Dale, I'm not sure why we're missing each other on this point. I've described a number of times how SR and LET may be (and possibly have been) distinguished experimentally: the various experiments regarding detection of absolute space/ether.

If we accept Cahill's conclusions re MMX, Miller and his own experiments, we have a rich trove of data distinguishing LET from SR.

Length contraction and time dilation effects are indeed empirically indistinguishable between LET and SR - I've never disagreed on this point. But by detecting absolute space/ether we may distinguish these two theories in terms of the cause of length contraction and time dilation. This is the difference between Lorentz and Cahill - Lorentz thought that the MMX experiments had produced a null result, but Cahill believes differently.

So, once more, the crux here is Cahill's conclusions re detection of absolute space. Accordingly, I look forward to your detailed response re my previous points.

 

[Tam Hunt]

Lalbatross, your statements re SR covering gravity and accelerating frames directly contradict Einstein himself. Are you suggesting that the understanding of SR's applicability has been expanded this much since the 1950s? Here's a relevant passage from the 1952 edition of his popular book (pp. 81-82):

I seek in vain for a real something in classical mechanics (or in the special theory of relativity) to which I can attribute the different behavior of bodies considered with respect to the reference-systems K and K prime [accelerating with respect to each other]. Newton saw this objection and attempted to invalidate it, but without success. ... It can only be got rid of by means of a physics which is conformable to the general principle of relativity.

The "special principle of relativity" applies to inertial frames. The "general principle of relativity" applies to all frames.

 

[lalbatros]

Tam,

This statement is now true:

The "special principle of relativity" applies to inertial frames. The "general principle of relativity" applies to all frames.

But it is totally different from your previous statement.

It should be clear that SR can be used to analyse the physics in an accelerated frame.
But performing this change of frame or variables will lead you to equations of motion including additional terms as compared to what they would be in an inertial frame. These terms are called "inertial forces" and examples are the centrifugal force and the Coriolis force.

In general relativity, for weak potentials (V/mc² << 1), these terms are recovered in exactly the same form. But this time, they are ready-for-use right form the beginning in the general equations of motion.

 

[Tam Hunt]

Lalbatross, can you send me any papers on what you're talking about? It seems that if you agree that the special principle of relativity is limited to inertial frames then you must also agree that SR does not consider gravity or accelerated frames. I don't see any other option, but I'm all ears. It's my understanding, from all my reading on SR and GR that the ENTIRE point of developing GR, with its general principle of relativity (as distinguished from the general theory of relativity), is to allow consideration of accelerating frames and gravity.

 

[Dale]

Dale, I'm not sure why we're missing each other on this point. I've described a number of times how SR and LET may be (and possibly have been) distinguished experimentally: the various experiments regarding detection of absolute space/ether.

No. I have rebutted that an equal number of times. Any experiment that detects an ether falsifies LET as well as SR.

If we accept Cahill's conclusions re MMX, Miller and his own experiments, we have a rich trove of data distinguishing LET from SR.

Conclusions are not empirical evidence; measurements are. That is the whole point. Because SR and LET make the exact same experimental predictions in all cases you can take any piece of experimental data and use it to conclude that either theory has been verified. There is no possible experimental distinction between SR and LET.

Again: If you really think that it is possible to distinguish between LET and SR on the basis of experimental evidence then kindly propose any experimental set up where Lorentz would use the Lorentz transforms to predict one value for a specific measurement and Einstein would use the Lorentz transforms to predict a different value for the same measurement. Please be quantitative in your predictions and explicit in your derivation.

If you cannot produce such a set of conflicting predictions then please stop making this claim. Until you do so we cannot have a meaningful conversation.

 

[Tam Hunt]

Dale, I see now why we've been missing each other. You're defining LET and SR narrowly - including in these theories only their predictions re length contraction and time dilation. As you know, the basis for LET is the assertion that the ether is the cause of such effects. As such, we may simply have to disagree that an experiment demonstrating the validity of absolute space/ether is support for LET over SR.

Alternatively, I'm happy to use a different term, "neo-LET" or "modern ether theory" (or MET, which is the term I use in my in-progress book on these topics) instead of LET. As such, MET accepts that there is strong evidence for the ether and MET is to be preferred over SR because SR assumes there is no ether (it is "superfluous" as Einstein states in his 1905 paper). Additionally, if we accept Cahill's evidence and conclusions in favor of the ether, we must also accept that SR has been falsified. It doesn't get more clear than that.

This goes back, it seems, to our disagreement over what constitutes a physical theory. I and many others maintain that every physical theory has two key components: the mathematical formalisms and the interpretation of those formalisms. You seem to hold the view that the only thing that truly comprises a physical theory is the mathematical formalism.

 

[lalbatros]

This is just introductory physics, see for example (1).
In this reference (1), you can see how non-inertial frames are tackled in Newtonian mechanics.
With the ad-hoc SR modifications, you can re-write this whole page for special relativity.

As you can see, the equations (4.1.4) and (4.1.5) are not exactly the same although they are perfectly equivalent.
This reflects the fact that Newtonian mechanics is not "generally covariant", like SR.
General relativity instead is "generally covariant".

In constrast, the equations of motion in GR would remain the same in any system of coordinates.
The equation (87.3) in Landau (2) is the equation of motion.
It is "generally covariant", by design.


(1) http://books.google.com/books?id=UAeuMKLXiWgC&pg=PA161&dq=classical+mechanics+inertial+forces
(2) https://www.amazon.com/dp/0750627689/?tag=pfamazon01-20

 

[Tam Hunt]

Lalbatross, at no point have I been discussing classical mechanics, so your cites do not address my points. My question referred to SR and your claim that SR could handle accelerating frames and gravity.

 

[lalbatros]

Tam,

... at no point have I been discussing classical mechanics, so your cites do not address my points. My question referred to SR ...

I did address your point since I justified this connection:

With the ad-hoc SR modifications, you can re-write this whole page for special relativity.

In other words:

SR relativity is more general than classical mechanics.
SR is able to reproduce any result from classical mechanics.
Therefore SR can handle accelerated frames just like classical mechanics does.
The difference is that SR does not do it in a (generally) covariant way like GR.

Discussing SR implies therefore that you are also discussing classical mechanics.
To avoid any ambiguity, note that I am discussing this statement that you made:

Lalbatross, SR doesn't replace Newtonian physics - GR does. SR does not cover gravity or accelerating frames of reference.

This statement is wrong because SR does cover accelerated frames of reference, just like NM.
SR also covers gravity, but without new contribution with respect to NP.

If we cannot agree on those things, it does not really matter as this is probably a misunderstanding.
To solve the point, you could simply explain what you meant.
My understanding was that you believed that in SR the motion of particles could not be studied in accelerated frame.
This is of course an absurd statement and I am sorry if I misinterpreted your claim.
Unfortunately, I think it was not clear, specially in a Cahill context!

 

[lalbatros]

Tam,

Dale, I see now why we've been missing each other. You're defining LET and SR narrowly - including in these theories only their predictions re length contraction and time dilation. As you know, the basis for LET is the assertion that the ether is the cause of such effects.

I think that Dale is defining these theories scientifically.
Any statement that cannot be verified is not scientific.
If SR and LET cannot be distinguished by any experiment, then they are scientifically speaking the same theory.
The only difference would be in the vocabulary used.
This difference is then easily solved by a LET-SR-LET dictionnary.

In addition, if the "LET theory" is more complicated that SR, then it has much less value.
If the LET assumes some kind of "aether" and needs an additional principle to justify that this "aether" is undetectable, then you better remove the aether from the LET and get SR instead.

As Dale pointed out, the scientific approach can only be based on verifiable differences between the two theories.
Without such difference, the discussion is futile.

 

[Tam Hunt]

Lalbatross, re LET and SR being distinguishable scientifically, I've covered this a number of times in this thread.

Again: experiments that demonstrate the existence of absolute motion/ether, if valid, distinguish LET and SR because LET relies on the ether as the basis for length contraction and time dilation. I don't know how to state this point more simply. LET is not simply the Lorentz transformations - it's a fully fleshed out theory that includes the ether.

Again: if one prefers to hew strictly to Lorentz's version of LET, and accept that the ether is undetectable (as Lorentz perhaps erroneously believed), we can call this new theory neo-LET or "modern ether theory (MET)." In MET, we accept that we have empirical data for the ether and that SR is, thus, falsified. MET is an extension of LET and accepts that there is absolute space/ether. SR becomes, then, an interesting epistemological theory, but it would be shown under this scenario to be ontologically - and thus physically - invalid.

 

[Tam Hunt]

Lalbatross, it seems that you're fundamentally misunderstanding the nature of SR - or I am. SR covers ONLY inertial frames and does not include gravity or accelerating frames. Einstein's "equivalence principle," in which he realized that gravity and acceleration could be treated as essentially the same thing, led to his general principle of relativity, which asserts that the laws of physics are the same in all moving frames. The general principle of relativity is the basis for the general theory of relativity. The special principle of relativity only applies to inertial frames, and it is the basis for the special theory of relativity (along with the principle of the constancy of the speed of light).

 

[Dale]

Additionally, if we accept Cahill's evidence and conclusions in favor of the ether, we must also accept that SR has been falsified.

This is circular reasoning. Obviously, if we accept Cahill's conclusions then we must accept that SR has been falsified since that is Cahill's conclusion. In any case, a conclusion is not empirical evidence, only measurements are.

Dale, I see now why we've been missing each other. You're defining LET and SR narrowly - including in these theories only their predictions re length contraction and time dilation. As you know, the basis for LET is the assertion that the ether is the cause of such effects. As such, we may simply have to disagree that an experiment demonstrating the validity of absolute space/ether is support for LET over SR.
...
This goes back, it seems, to our disagreement over what constitutes a physical theory. I and many others maintain that every physical theory has two key components: the mathematical formalisms and the interpretation of those formalisms. You seem to hold the view that the only thing that truly comprises a physical theory is the mathematical formalism

I am not defining either LET or SR narrowly, and I understand that the interpretation is part of a theory as are the postulates and the derivation of the mathematical framework from them. I am not saying that two interpretations are the same; I am just rejecting your fallacious assertion that there can possibly be any experimental distinction between them. There is no such thing as an interpretometer that we could use to gather direct evidence about interpretations. All we can experimentally verify or falsify is a theory's prediction about a given measurement.

I think that you actually know that it is impossible to distinguish the two experimentally. That is why you have not even attempted to work out an example of an experimental measurement that could distinguish SR from LET. It remains useless to proceed until you either do so or concede the point.

 

[Tam Hunt]

Dale, this is perhaps an interesting example of why it's so hard to change paradigms: you somehow skipped right over the word "evidence" in my sentence that you actually quoted in your response. My reasoning was not circular.

Cahill and others have presented reams of data purporting to show that there is an absolute space/ether. This is "evidence." "Conclusions" follow from this "evidence."

If we accept this evidence, and therefore, the conclusion that there is an absolute space/ether, we THEN accept that SR is falsified.

I don't know how to make this any more clear.

I like your neologism "interpretometer." :-) Fortunately for us, we don't need an interpretometer. All we need is an interferometer. Please go back to my response a few turns ago and see my answers to your previous objections to Cahill's experiments and interpretations re absolute space/ether. By continuing that discussion, we may actually arrive at a fruitful conclusion to our interesting but sometimes frustrating dialogue.

 

[lalbatros]

Tam,

Again: experiments that demonstrate the existence of absolute motion/ether, if valid, distinguish LET and SR because LET relies on the ether as the basis for length contraction and time dilation. I don't know how to state this point more simply. LET is not simply the Lorentz transformations - it's a fully fleshed out theory that includes the ether.

From the Lorentz transformation, the speed of light does not depend on the (inertial) frame of reference that is chosen to measure it.
Therefore, if LET admits the LT, LET should predict that the speed of light is invariant.
Therefore, LET should not predict that detecting an absolute motion is possible.
Therefore, if the LET derivation of LT is based on the existence of an absolute frame of reference,
then the LET must also contain another hypothesis that makes this frame finally unobservable.
In any case, your point does not point us to a way to distinguish SR and LET experimentally.

I am still open to discuss such a possible experiment, but I cannot describe one myself.
Assuming that both LET and SR agree on the LT, it seems to me that distinguishing LET and SR experimentally is not possible.

Suppose, for example, that the MMX experiment would deliver a 1000000 times better precision than before.
Suppose also it would lead to a certainty about an absolute frame of reference.
I would then conclude that both SR and LET would be invalidated.
They would be invalidated because both theories would predict the speed of light is invariant,
and the experiment would then have demonstrated the contrary.

It does not matter that a "absolute frame" was mentioned in the derivation.
Only the predictions of the theory matters.
I don't see any prediction of the LET that depends on the preffered frame hypothesis.

Conversly, we can also conclude that SR does not rule out the aether hypothesis.
SR simply makes the aether hypothesis "superfluous", as it is well known.
It is superflous as far as the LT is involved.
In this sense, the SR derivation by Einstein is a cleaned version of the LET.
It goes without saying that SR brings more than that.

 

[lalbatros]

Tam,

You need to make your statement more precise:

SR covers ONLY inertial frames

If you mean that with SR you cannot derive the mechanics in an accelerated frame, then you are wrong.
Even Newtonian mechanics allows that, why the "inertial forces" are known since long.
Actually, the Newtonian relativity was precisely aimed at distinguishing inertial forces from other forces.
The general principles of relativity goes a step further than that so as to make inertial forces and gravitational forces locally undistinguishable.
For the motivation of GR, I always refer to page 245 in Landau (1).

If you mean instead that the relativity principle is about the equivalence of inertial frames, which have no relative acceleration, then you are right.

(1) http://books.google.com/books?id=QI...cover&dq=classical+theory+of+fields#PPA245,M1

 

[lalbatros]

Tam,

This is simply wrong:

Cahill and others have presented reams of data purporting to show that there is an absolute space/ether.

Cahill has systematically biased all the conclusions.
His analysis of the historical MMX is very clear in this respect.
Just read about the experimental procedure, and you will clearly see that the assumed "positive signal" is much below the error bars of this experiment. See also the analysis of the Miller trials, which is very instructive. (1)

Concerning the De Witte coaxial experiment that he systmatically reffers to, there is simply no data available from this experiment. And Cahill will be the first to claim about a conspiracy to hide such data, proving so that these data are not available. Would a conspiracy theory replace experimental data?

Cahill is not about scientific evidence, he is about shelling any evidence against his theory.

(1) http://arxiv.org/abs/physics/0608238

 

[Tam Hunt]

lalbatros, you write:

Therefore, if LET admits the LT, LET should predict that the speed of light is invariant.
Therefore, LET should not predict that detecting an absolute motion is possible.
Therefore, if the LET derivation of LT is based on the existence of an absolute frame of reference,
then the LET must also contain another hypothesis that makes this frame finally unobservable.
In any case, your point does not point us to a way to distinguish SR and LET experimentally.

You are misunderstanding LET. LET states that the speed of light will APPEAR invariant to all observers, not that it is actually invariant. This is the whole point of the ether in Lorentz ether theory: it is through electromagnetic interaction with the ether that matter (the arms of a vacuum-mode interferometer, for example) is length-contracted in the direction of motion. This was Lorentz's hypothesis explaining the apparent null result of MMX.

To be clear: for Lorentz, the speed of light was not actually invariant, it just appeared invariant because of length contraction of all apparati built to measure it.

Cahill's work is potentially ground-breaking because he, Miller, and others have shown that we can in fact detect absolute motion by changing to gas-mode interferometers or with fiber optic experiments.

As such, Cahill, Milller, and others have shown how LET and SR may be experimentally distinguished, due to evidence supporting the existence of the ether/absolute space. As I've also mentioned previously, this would make SR an interesting epistemological theory, but not a theory that actually describes reality. You are free to say, in such a scenario, that SR remains unfalsified, but I prefer the realist school of thought, not the school of thought that accepts epistemological theories masquerading as physical theories.

I'm glad in your last post that you return to discussing the merits of Cahill's experiments and interpretations. As I wrote already in response to Dale, Cahill has fully considered error bar analysis. Here's what Cahill wrote in response to me when I asked him about this issue some time ago:


Miller in his 1933 paper reported (page 238) speed errors at the level of approx 3% and azimuth errors of 2.5 degrees. My analysis of Miller's data gives a maximum speed error of 5% for the good data - i.e that without obvious glitches. That analysis is based on the rms error from fitting the expected form (including temperature drifts, the Hick's effect and expected cos[2*angle] effect), and using the usual definition of rms error. Indeed the quality of his data is truly remarkable.

 

[Tam Hunt]

lalbatross,

In your follow up re SR and inertial frames, you seem to be confusing "inertial frames" with "inertial forces." These are very different beasts and have limited relations to each other.

Again, please read Einstein's 1916 book on these issues (it's quite short). I think you'll find it interesting and helpful.

 

[Dale]

you somehow skipped right over the word "evidence" in my sentence that you actually quoted in your response. My reasoning was not circular.

Cahill and others have presented reams of data purporting to show that there is an absolute space/ether. This is "evidence." "Conclusions" follow from this "evidence."

If we accept this evidence, and therefore, the conclusion that there is an absolute space/ether, we THEN accept that SR is falsified.

Your reasoning is completely circular. Your premise "Cahill's evidence and conclusions" explicitly contains your conclusion "SR has been falsified". Circular reasoning at its most obvious and basic. Another way to http://www.fallacyfiles.org/begquest.html" is when an argument "assumes any controversial point not conceded by the other side", which you are clearly doing here.

Additionally, it is possible to accept a scientist's evidence without accepting his conclusions. After all, this is exactly what you are doing re: all of the http://www.edu-observatory.org/physics-faq/Relativity/SR/experiments.html" . Your statement "'conclusions' follow from this 'evidence'" is simply untrue. It is always possible to find another interpretation which explains the same observation thus yielding a different conclusion from the same evidence.

And you still haven't addressed the challenge.

 

[ZapperZ]

Miller in his 1933 paper reported (page 238) speed errors at the level of approx 3% and azimuth errors of 2.5 degrees. My analysis of Miller's data gives a maximum speed error of 5% for the good data - i.e that without obvious glitches. That analysis is based on the rms error from fitting the expected form (including temperature drifts, the Hick's effect and expected cos[2*angle] effect), and using the usual definition of rms error. Indeed the quality of his data is truly remarkable.

And did you challenge him on this? If there is THAT much of a discrepancy, wouldn't this already show up in the GPS system that we have set up that clearly does not take into account such variation? I mean, it doesn't take much to throw it off, and such error will in fact accumulate very quickly.

I don't understand this thread. The Miller's paper has been discussed on here, and in literature, ad nauseum! Somehow, that work is in such HIGH regards that its worshipers seems to ignore all the subsequent tests using much more sensitive set up, such as using the Kennedy-Thorndike method. What gives? How are you able to pick-and-choose like that?

Maybe you should visit the good folks running LIGO and tell them they have been calibrating their setup all wrong.

Zz.

 

[Tam Hunt]

Dale, I'm going to give this one more shot and them I'm done - recognizing that perhaps we simply can't have a constructive dialogue on this issue.

1) Cahill, Miller, etc., have produced evidence from their experiments.

2) They interpret this evidence to show that the speed of light is inconstant - contrary to the 2nd postulate in Einstein's 1905 paper. This is a conclusion. They also interpret this evidence to show that absolute space/ether is real. This is a second conclusion. If these interpretrations/conclusions are correct (or even if only the first conclusion is correct), SR is falsified. If these interpretations/conclusions are not correct, show me why they are not correct.

3) The same evidence and conclusions could be used, in a strict approach, to falsify LET - as Lorentz himself formulated it, as you have pointed out. This is the case b/c under LET, there is no way to experimentally confirm the inconstancy of the speed of light, even though it is assumed that in actuality the speed of light is inconstant. As you know, Lorentz postulated the opposite of Einstein: the speed of light is inconstant, but we are prevented, due to length contraction of measurement equipment in the direction of motion, from detecting it.

4) Alternatively - the alternative that I prefer - we may adopt a "neo-LET" or "modern ether theory (MET)" approach, in which we adopt the Lorentzian interpretation of the Lorentz transformations, which relies on the ether as the source of length contraction and apparent time dilation, BUT we modify LET with our understanding that we can in fact detect absolute space/ether. Under MET, we find that most experimental evidence supporting SR also supports MET, with the KEY difference being that the constancy of the speed of light and consequent malleability of space and time are NOT the cause of length contraction and time dilation. Rather, electromagnetic interaction between matter and the ether is the cause of these effects.

I have answered your challenge about six times now.

 

[Tam Hunt]

ZapperZ, I'd appreciate your substantive feedback on Cahill's work. He has expanded significantly on Miller's work with his own experiments. Here's his most comprehensive discussion available online (see Chapter 10):

http://www.scieng.flinders.edu.au/cpes/people/cahill_r/HPS13.pdf

And here's his most recent work on MMX:

http://arxiv.org/abs/physics/0508174

 

[ZapperZ]

ZapperZ, I'd appreciate your substantive feedback on Cahill's work. He has expanded significantly on Miller's work with his own experiments. Here's his most comprehensive discussion available online (see Chapter 10):

http://www.scieng.flinders.edu.au/cpes/people/cahill_r/HPS13.pdf

And here's his most recent work on MMX:

http://arxiv.org/abs/physics/0508174

So in other words, you were not at all worried that THAT kind of discrepancy doesn't show up in such thing as the LIGO alignment and our GPS's? How could you let him get away with something like that?

Zz.

 

[Tam Hunt]

ZapperZ, as we've discussed at length in this thread, the predictions of SR and LET match in terms of length contraction and time dilation because they both rely on the Lorentz transformations. So GPS calibration re SR is not a factor in this discussion (though I understand that calibration is not as clearly specially relativistic as most believe, relying rather on real-time calibrations that sort of match SR's predictions).

The issue at this point in this thread is whether we have good evidence for absolute space/ether, based on MMX, Miller, Cahill and other experiments. If so, the theoretical basis for LET (or at least a neo-Lorentzian relativity) would be supported over SR and SR may be considered falsified because its postulates would be shown to be wrong.

 

[ZapperZ]

ZapperZ, as we've discussed at length in this thread, the predictions of SR and LET match in terms of length contraction and time dilation because they both rely on the Lorentz transformations. So GPS calibration re SR is not a factor in this discussion (though I understand that calibration is not as clearly specially relativistic as most believe, relying rather on real-time calibrations that sort of match SR's predictions).

The issue at this point in this thread is whether we have good evidence for absolute space/ether, based on MMX, Miller, Cahill and other experiments. If so, the theoretical basis for LET (or at least a neo-Lorentzian relativity) would be supported over SR and SR may be considered falsified because its postulates would be shown to be wrong.

You are forgetting that both the GPS systems and LIGO calibration depend entirely on the anisotropy of the speed of light. It isn't JUST the issue of time dilation and length contraction. GPS system, especially, rely on the uniformity of the speed of light as the orientation of the Earth and the satellites change with respect to "space". This is the EXACT thing that any MMX experiment measures! The same thing with the "arms" of LIGO. Any discrepancy would easily show up!

Yet, you are relying on Miller's experiment as if it is valid but somehow unconcerned that your GPS seems to work perfectly fine.

Zz.