Relativity without the aether: pseudoscience?

[Aether]

Calling your opponent supersticious just because they don't share your own certainty seems counterproductive to me.

How is it counterproductive? I have clearly demonstrated that many people have an irrational belief that a circumstance (e.g., the constancy of the speed of light) not logically related to a course of events (e.g., an experiment) influences its outcome. This reads on the definition of "superstition" that I quoted. Now I am faced with a choice: 1) keep this to myself, and risk either: a) going through the rest on my own life with the wrong impression, or b) allowing my brothers and sisters to go through life believing in a superstition; or 2) ask this question so that we all may reason together.

 

[derz]

SR is valid only within the context of intertial reference systems

Well, there is always the general theory of relativity that deals with accelerating frames.

I don't believe that the speed of light is contant because everybody says so, but because it is a logical consecuence (spelling?) of the fact that there is no absolute frame of reference. And that is what can be concluded from the MM-experiment.

 

[Hurkyl]

SR is valid only within the context of intertial reference systems.

Somewhat misleading -- just like with Newtonian physics, the formulae of SR may be modified so that they would apply to any generalized coordinate system.

 

[DrChinese]

1. The idea that the constancy of the speed of light can't be proven by experiment?

2. OK, but I didn't say that there was anything wrong with science, only with relativity.

1. That c is a constant has been verified any number of times, in any number of ways. There are some people who say there are loopholes in ANY experiement that does not agree with their hyposthesis. For example: Caroline Thompson does it with Bell tests; and apparently you take up the issue similarly with c.

2. You have not shown a single prediction of SR which is inconsistent with accepted experiment. Here's your chance to show us wrong.

By the way, shouldn't this thread be getting close to end of its run?

 

[Jimmy Snyder]

In SR, the constancy of the speed of light is a postulate. It's something that is taken for granted, not something that is proven. How can you tell the difference between someone who superstitiously believes that the speed of light is a constant, and someone who is simply following the consequences of a postulate?

 

[JesseM]

It is not valid to claim that the constancy of the speed of light is proven "by experiment", or that "experiments prove" that SR is right and that LET is wrong.

It's valid in the sense that experiments prove that the laws of physics are Lorentz-symmetric, which means that if you use any type of "natural" coordinate system involving rulers and clocks moving inertially and clocks synchronized using what Mansouri and Sexl call "system-internal clock synchronization", then you will find that the speed of light is constant. The fact that this stuff is left implicit rather than made explicit doesn't make it "invalid", although it may make the claim insufficiently spelled-out.

Question: do you think it's valid to say that "experiments demonstrate that the speed of the Earth's rotation at the equator is 1670 km/hour"? This statement also makes the same sort of implicit assumptions about using an inertial coordinate system with a system-internal clock synchronization method, doesn't it? Would you say this means the statement actually has no basis in experiments?

 

[Aether]

By the way, shouldn't this thread be getting close to end of its run?

OK. I'll answer any direct questions to me, but otherwise let everyone else have the last word.

How can you tell the difference between someone who superstitiously believes that the speed of light is a constant, and someone who is simply following the consequences of a postulate?

If they insist that the constancy of the speed of light is "proven by experiment", then they are mistaken. If they understand that this is how inertial reference systems are defined, and that the constancy of the speed of light can't be the subject of an experiment in this context, then they are correct. If they also say something to the effect that without evidence of a violation of local Lorentz symmetry, then it is usually more convenient to treat the speed of light as a constant, then that's also correct.

 

[Jimmy Snyder]

And what if they say it's a postulate?

 

[Aether]

Question: do you think it's valid to say that "experiments demonstrate that the speed of the Earth's rotation at the equator is 1670 km/hour"?

Yes.

This statement also makes the same sort of implicit assumptions about using an inertial coordinate system with a system-internal clock synchronization method, doesn't it?

No. This looks like a one-clock experiment, and there would be no need to synchronize clocks.

Would you say this means the statement actually has no basis in experiments?

No.

And what if they say it's a postulate?

I'm not exactly sure what you're asking, but if they say that "the speed of light c is the same in all inertial frames" is a postulate, then that is true.

 

[JesseM]

No. This looks like a one-clock experiment, and there would be no need to synchronize clocks.

The average speed for a full rotation might be a one-clock experiment, but the instantaneous speed at any given moment would not be. But to avoid this issue, how about a statement about velocities that does not involve circular paths, like "the Andromeda galaxy is approaching our own galaxy at a speed of 68 kilometers per second"--would you say this has any more or less basis in experiment than the statement that the speed of light is constant?

 

[Jimmy Snyder]

Well, are you proponents of SR unaware that Aether is baiting you? There is no need to prove that the speed of light is a constant. SR has already proved its worth by getting correct numerical results where Newtonian theory fails, for instance in calculating the half-life of particles in particle accelerators. The ball is in the court of your opponents. They need to come up with an experiment in which SR predicts the wrong number. Getting you to try and prove that your postulates are true is a distraction meant to cover up for this shortcoming on their part.

 

[DrGreg]

Here are three postulates:

1. All cats are completely black
2. All cats are completely white
3. At least one cat exists

There are two questions you can ask about these postulates. Are they mutually consistent? Is there any experimental evidence to support or disprove them?

I suggest there is overwhelming experimental evidence that 1 and 2 are false and 3 is true.

But if we ignore the evidence and simply consider the internal logic of my "theory", you will see that any two of the three postulates are compatible, but all three taken together are incompatible.

So there are potentially two ways to "disprove" a postulate: in one sense, provide experimental counter-evidence, or in another sense, show its logical incompatibility with other postulates (or even with itself).

Einstein's postulate of the constancy of the speed of light cannot be disproved logically, and nobody has yet been able to disprove it experimentally.

However, much of the argument in this thread has been about whether there is an alternative set of postulates that is compatible with experiment (there is), and whether such an alternative set is better or worse than Einstein's. Most of us, including me, think Einstein's is by far the best (for example, by application of Occam's Razor), but Aether disagrees with us.

 

[ZapperZ]

Here are three postulates:

1. All cats are completely black
2. All cats are completely white
3. At least one cat exists

There are two questions you can ask about these postulates. Are they mutually consistent? Is there any experimental evidence to support or disprove them?

I suggest there is overwhelming experimental evidence that 1 and 2 are false and 3 is true.

But if we ignore the evidence and simply consider the internal logic of my "theory", you will see that any two of the three postulates are compatible, but all three taken together are incompatible.

So there are potentially two ways to "disprove" a postulate: in one sense, provide experimental counter-evidence, or in another sense, show its logical incompatibility with other postulates (or even with itself).

Einstein's postulate of the constancy of the speed of light cannot be disproved logically, and nobody has yet been able to disprove it experimentally.

However, much of the argument in this thread has been about whether there is an alternative set of postulates that is compatible with experiment (there is), and whether such an alternative set is better or worse than Einstein's. Most of us, including me, think Einstein's is by far the best (for example, by application of Occam's Razor), but Aether disagrees with us.

I think you have hit the issue right on its head.

I see NO COMPELLING REASON why I should abandon a well-tested theory for all of these alternative ideas, when the best they can do is agree with SR. We certainly did not adopt QM simply because it can provide an alternative view of our world that we much "prefer" than classical mechanics. And for me, personally, there is no more compelling evidence than a clear experimental observation. Till we have that, I find all of these arguments to be rather moot.

Zz.

 

[Jimmy Snyder]

I'm sorry for completely mischaracterizing the nature of the discussion in this thread. Is the issue that the following two arguments are on equal footing?

1. No one has detected a dependence of the speed of light on the motion of the emiter relative to the observer, therefore the dependence does not exist.

2. No one has detected the luminiferous aether, nonetheless the aether does exist.

I hope I have not mischaracterized a second time, but if I have, please straighten me out.

 

[Aether]

"the Andromeda galaxy is approaching our own galaxy at a speed of 68 kilometers per second"--would you say this has any more or less basis in experiment than the statement that the speed of light is constant?

The basis in experiment for this statement is that the ratio of wavelengths for signals coming from the Andromeda galaxy compared to like signals on Earth is observed to be 0.99977. The speed of 68 km/s comes from running this observed ratio though the relativistic Doppler equation. The ratio of wavelengths is proven by experiment, but the velocity depends on one's choice of coordinate system in the same way that the speed of light does; therefore, I suspect that the ratio $$\beta =v/c=0.000227$$ could probably also be considered as being proven by experiment.

 

[DrGreg]

I'm sorry for completely mischaracterizing the nature of the discussion in this thread. Is the issue that the following two arguments are on equal footing?

1. No one has detected a dependence of the speed of light on the motion of the emiter relative to the observer, therefore the dependence does not exist.

2. No one has detected the luminiferous aether, nonetheless the aether does exist.

Well, sort of.

There is a difference between the "one-way speed of light" measured from A-to-B, and the "two-way average speed of light" measured on a round trip A-to-B-to-A again. The constancy of the 2-way speed (relative to any inertial observer A) is something that can be confirmed by experiment. The constancy of the 1-way speed depends on your convention for synchronizing the clocks at A and B. If you adopt Einstein's postulates, you must synchronize in such a way that the speed of light is constant by definition.

But you could synchronize in a different way such that all observers agree on what is simultaneous. The second way is just a change of time coordinates t_ether = t_einstein + v₀ . x / c², so still valid. (v₀ is the velocity of the observer relative to whatever you arbitrarily choose to be your ether.) But it's an awful choice because you've destroyed all the isotropy of Einstein's co-ordinates. The one thing in favour of "Ether" coordinates is that those people who fail to grasp the concept of relative simultaneity (or those who do grasp it, but philosophically reject it) might be able to accept them.

 

[DrChinese]

Of course the constancy of the speed of light has been proven by experiment. The one-way vs. two-way argument is a diversion to attempt to salvage the ether, and nothing more. We could play this game with every piece of knowledge there is.

So... according to Aether: if you can't measure it according to his standards, then c has any value he wants it to have. Is that science? Please note that he does not tell us what the speed of light is in the direction of the ether, nor across the ether, or against the ether. He has no experiments to back up the values he doesn't give us. And he dismisses all experiments that yield values consistent with c. So by his logic:

a) The gravitational constant is different in Andromeda than it is in the Milky Way.

b) The electron's charge is different on Tuesdays in the year 2044.

c) Experimental exidence in favor of a constant c is flawed because of the "fair sampling" loophole (yes, please note that the beauty of this "loophole" applies to every experiment, not just Bell tests - and I am sure we could commission Caroline Thompson to write a "chaotic ball" position paper to support the concept).

Theory is a useful tool. Aether doesn't offer one, simply says all others are wrong. Meanwhile, back at the ranch, everyone else uses c and it works perfectly - one way or two ways LOL. Aether, we all acknowledge that if every previous experiment on every previous subject is subsequently proven wrong, then they are wrong. But that hasn't happened in this particular area *yet*. So when there is some actual scientific evidence to back up your assertion, Aether, please write again. The rest is speculation and belongs in Theory Development.

-DrC

 

[ZapperZ]

How is it counterproductive? I have clearly demonstrated that many people have an irrational belief that a circumstance (e.g., the constancy of the speed of light) not logically related to a course of events (e.g., an experiment) influences its outcome. This reads on the definition of "superstition" that I quoted. Now I am faced with a choice: 1) keep this to myself, and risk either: a) going through the rest on my own life with the wrong impression, or b) allowing my brothers and sisters to go through life believing in a superstition; or 2) ask this question so that we all may reason together.

Can you show me another "superstition" that can actually come with the correction to the band structure calculation of the semiconductor you are using in your computer?

Zz.

 

[Aether]

Can you show me another "superstition" that can actually come with the correction to the band structure calculation of the semiconductor you are using in your computer?

Zz.

Yes, if SR can do that with a constant speed of light, then LET can do it with a variable speed of light. The "speed of light" per se is simply not an active ingredient of Lorentz symmetry. Others have commented that these two theories are actually the same theory using different coordinate systems.

 

[pervect]

How is it counterproductive? I have clearly demonstrated that many people have an irrational belief that a circumstance (e.g., the constancy of the speed of light) not logically related to a course of events (e.g., an experiment) influences its outcome. This reads on the definition of "superstition" that I quoted. Now I am faced with a choice: 1) keep this to myself, and risk either: a) going through the rest on my own life with the wrong impression, or b) allowing my brothers and sisters to go through life believing in a superstition; or 2) ask this question so that we all may reason together.

Your argument would be much more convincing if you actually used the "speed" you defined for light to measure the speed of material bodies.

Apparently, you don't actually do that - you use radar methods to determine a conceptually different sort of speed for material bodies, a notion of speed that is in accord with the standard defintion and not the notion of speed that you propose using to measure the "speed" of light with.

The argument hinges around a semantic issue, one of what the defintion of "speed is". One can (and you do) define oddball sorts of "speed" by playing with the clock synchronization, but such "speeds" lack many assumed physical properties (such as isotropy, the property that bodies with the same mass and the same speed moving in different directions have the same momentum), and are really not very useful.

It's OK to explore non-standard semantics and approaches to see if they do anything useful, but to use non-standards semantics and then call other people "superstitious" because they don't use your non-standard semantics but instead use standard semantics is simultaneously both argumentative and very weak.

 

[ZapperZ]

Yes, if SR can do that with a constant speed of light, then LET can do it with a variable speed of light. The "speed of light" per se is simply not an active ingredient of Lorentz symmetry. Others have commented that these two theories are actually the same theory using different coordinate systems.

You either did not answer my question to show me another superstition that can explain the corrections to the band structure of a semiconductor you are using in your electronics, or you have IMPLICATED that LET is also a superstition.

Which one do you prefer?

Zz.

 

[Aether]

You either did not answer my question to show me another superstition that can explain the corrections to the band structure of a semiconductor you are using in your electronics, or you have IMPLICATED that LET is also a superstition.

Which one do you prefer?

Zz.

Again, others have commented that these two theories are actually the same theory using different coordinate systems. Either coordinate system (SR or LET) taken by itself is prone to give a similar false impression about what measurements of the speed of light represent. Taken together, either coordinate system in contrast to the other, they describe Lorentz symmetry in a less coordinate-dependent way. This may be implicit in Mansouri-Sexl and most of the papers on local Lorentz invariance that reference their work. I would prefer a completely coordinate independent approach, as I said in post #11.

 

[ZapperZ]

Again, others have commented that these two theories are actually the same theory using different coordinate systems. Either coordinate system (SR or LET) taken by itself is prone to give a similar false impression about what measurements of the speed of light represent. Taken together, either coordinate system in contrast to the other, they describe Lorentz symmetry in a less coordinate-dependent way. This may be implicit in Mansouri-Sexl and most of the papers on local Lorentz invariance that reference their work. I would prefer a completely coordinate independent approach, as I said in post #11.

Excellent! You have refused to answer my direct question to you once again.

Your blindness in trying to defend this "LET" is clouding your understanding of my question to you. I am NOT comparing SR with LET. I am asking about your equating SR with "superstitions". You have equated those two as being the same. I asked you to name a supersition that can explain the corrections to the band structure calculation to a whole zoo of semiconductors that you are using. You blatantly refused to do so.

I strongly suggest that you retract such accusation, which in itself in very insulting to those of us in this field, especially considering that you cannot even put your money where you mouth is in justifying such accusation.

Zz.

 

[Aether]

Excellent! You have refused to answer my direct question to you once again.

Your blindness in trying to defend this "LET" is clouding your understanding of my question to you. I am NOT comparing SR with LET. I am asking about your equating SR with "superstitions". You have equated those two as being the same. I asked you to name a supersition that can explain the corrections to the band structure calculation to a whole zoo of semiconductors that you are using. You blatantly refused to do so.

I strongly suggest that you retract such accusation, which in itself in very insulting to those of us in this field, especially considering that you cannot even put your money where you mouth is in justifying such accusation.

Zz.

I retract any implied accusation equating either SR or LET with "superstitions", and apologise to those in the field. I answered your question both times: LET, in the context that I have previously explained.

 

[ZapperZ]

I retract any implied accusation equating either SR or LET with "superstitions", and apologise to those in the field. I answered you question both times: LET, in the context that I have previously explained.

No you did not answer my question!

I asked for you to show ANOTHER superstition that we use to do corrections to the band structure of semiconductors. All you did was babbled about LET and SR producing the IDENTICAL results. How did THIS answer my question? Is this how you approach all problems given to you, through some filtered glasses that allows you to modify it into what you WISH to see?

And until you care to explain why you think SR is "pseudoscience" while LET isn't (funny, since they produce the "identical" results), I also will ask you to refrain from using that again unless you want me to again ask you to show another pseudoscience that you place your LIFE on every time you fly in an airplane, thankyouverymuch!

Zz.

 

[Jimmy Snyder]

I'm sorry that I haven't read very many of the posts in this thread. Was the issue of binary stars discussed? If not ...

In a binary star system, one star is approaching us as the other recedes. If the light from these two stars were traveling at different speeds, then because of the great distance travelled, we should see the light emitted at the same time from these two stars arriving at very different times. The perceived orbits of such binaries would be grossly abnormal. Even if we were situated in such a way as to cancel out those effects for one system, it seems unlikely that we were so situated for all of them and in all seasons (we are orbiting too).

Does this constitute a one-way test of the constancy of the speed of light? This test was proposed by De Sitter in 1913.

 

[DrGreg]

I'm sorry that I haven't read very many of the posts in this thread. Was the issue of binary stars discussed? If not ...

In a binary star system, one star is approaching us as the other recedes. If the light from these two stars were traveling at different speeds, then because of the great distance travelled, we should see the light emitted at the same time from these two stars arriving at very different times. The perceived orbits of such binaries would be grossly abnormal. Even if we were situated in such a way as to cancel out those effects for one system, it seems unlikely that we were so situated for all of them and in all seasons (we are orbiting too).

Does this constitute a one-way test of the constancy of the speed of light? This test was proposed by De Sitter in 1913.

No.

I believe you are talking about whether the speed of light depends on the speed of the emitter, whether one photon could overtake another photon traveling in the same direction. We haven't discussed that here.

The "one-way" test is the question: does light traveling from Earth to Saturn travel at the same speed as light traveling from Saturn to Earth?

Einstein postulates that the answer is yes. We can't prove or disprove it except by adopting a convention for synchronising the clocks on the Cassini probe with Earth clocks. (The motion of the the Cassini probe relative to Earth means that time dilation has put its clocks out of sync, if only by a fraction of a second.)

 

[Jimmy Snyder]

The "one-way" test is the question: does light traveling from Earth to Saturn travel at the same speed as light traveling from Saturn to Earth?
Einstein postulates that the answer is yes.

In his paper "On the electrodynamics of moving bodies", he postulates the following:

Any ray of light moves in the "stationary'' system of co-ordinates with the determined velocity c, whether the ray be emitted by a stationary or by a moving body.

http://www.fourmilab.ch/etexts/einstein/specrel/www/

I could not find the postulate you mention.

 

[Jimmy Snyder]

It seems the symmetry you seek to test could be easily acheived with some half-wave glass.

 

[DrGreg]

The "one-way" test is the question: does light traveling from Earth to Saturn travel at the same speed as light traveling from Saturn to Earth?
Einstein postulates that the answer is yes..

In his paper "On the electrodynamics of moving bodies", he postulates the following:

Any ray of light moves in the "stationary'' system of co-ordinates with the determined velocity c, whether the ray be emitted by a stationary or by a moving body.

http://www.fourmilab.ch/etexts/einstein/specrel/www/

I could not find the postulate you mention.

OK, to be precise I should have said "The answer 'yes' is implied by Einstein's postulate" (as you quoted it). The postulate says that light moves at c, implicitly regardless of direction.

It seems the symmetry you seek to test could be easily acheived with some half-wave glass..

I don't understand what a "half-wave glass" is.

 

[Aether]

No you did not answer my question!

I asked for you to show ANOTHER superstition that we use to do corrections to the band structure of semiconductors. All you did was babbled about LET and SR producing the IDENTICAL results. How did THIS answer my question?

If your question supposes that there could be a person "A" who comes away from the study of SR harboring "an irrational belief that a circumstance (e.g., the constancy of the speed of light) not logically related to a course of events (e.g., an experiment) influences its outcome", but that he is nevertheless able to do corrections to the band structure of semiconductors; then my answer to your question is that there could also be a person "B" who comes away from the study of LET harboring "an irrational belief that a circumstance (e.g., the variability of the speed of light) not logically related to a course of events (e.g., an experiment) influences its outcome", but that he is nevertheless able to do corrections to the band structure of semiconductors.

And until you care to explain why you think SR is "pseudoscience" while LET isn't (funny, since they produce the "identical" results), I also will ask you to refrain from using that again unless you want me to again ask you to show another pseudoscience that you place your LIFE on every time you fly in an airplane, thankyouverymuch!

I never said that I think any such thing, and in post #11 I explicitly stated that "I think that both theories involve such an assumption: SR assumes that the one-way speed of light is isotropic, but this can not be proven by experiment; LET assumes absolute simultaneity, but this hasn't been proven either (yet)".

If this amounts to nothing more than semantics, then I would appreciate either an explanation, or a link to an explanation, of why this is so.

 

[Jimmy Snyder]

OK, to be precise I should have said "The answer 'yes' is implied by Einstein's postulate" (as you quoted it). The postulate says that light moves at c, implicitly regardless of direction.

Then why wouldn't the binary star experiment be a suitable test for your purposes?

I don't understand what a "half-wave glass" is.

More commonly known as a two-way mirror. Take two Michelson-Morley interferometers and shunt half of the light of one beam from each apparatus over to the other. If the interference fringes are the same in both interferometers, then the speed of the shunted beams was the same in both directions.

 

[ZapperZ]

If your question supposes that there could be a person "A" who comes away from the study of SR harboring "an irrational belief that a circumstance (e.g., the constancy of the speed of light) not logically related to a course of events (e.g., an experiment) influences its outcome", but that he is nevertheless able to do corrections to the band structure of semiconductors; then my answer to your question is that there could also be a person "B" who comes away from the study of LET harboring "an irrational belief that a circumstance (e.g., the variability of the speed of light) not logically related to a course of events (e.g., an experiment) influences its outcome", but that he is nevertheless able to do corrections to the band structure of semiconductors.

So then I was right! You have just implicated LET as ANOTHER SUPERSTITION. I asked you for another superstition that can explain the band theory corrections, and you came up with LET. Congratulations! You have just convinced yourself that it is nothing better than a black cat, or walking under ladders, or breaking a mirror.

Zz.

 

[DrGreg]

Then why wouldn't the binary star experiment be a suitable test for your purposes?.

Why would it? I am asking if the speed of light from the binary star to Earth is the same as the speed of light from Earth to the binary star. I can't see the relevance.

More commonly known as a two-way mirror. Take two Michelson-Morley interferometers and shunt half of the light of one beam from each apparatus over to the other. If the interference fringes are the same in both interferometers, then the speed of the shunted beams was the same in both directions.

In the MM experiment you have an observer A, two mirrors B and C. You send light from A to B to A, and also from A to C to A. You conclude that the A-B-A average speed is the same as the A-C-A average speed, wherever B and C are.

But how can I tell if the A-B speed is the same as the B-A speed? It is possible that one could be greater than the other, and the differences cancel each other out to give a constant A-B-A average.

I don't precisely understand the modification you propose and how it would help.

In his paper "On the electrodynamics of moving bodies", he postulates the following:

Any ray of light moves in the "stationary'' system of co-ordinates with the determined velocity c, whether the ray be emitted by a stationary or by a moving body.

http://www.fourmilab.ch/etexts/einstein/specrel/www/

I could not find the postulate you mention.

Before Einstein gives his postulates in Section I.2, he devotes a whole section I.1 to the Definition of Simultaneity. In the middle of this section he says

We have not defined a common ''time'' for A and B, for the latter cannot be defined at all unless we establish by definition that the ''time'' required by light to travel from A to B equals the ''time'' it requires to travel from B to A.​

This is exactly the point I am making: you must decide on a convention for synchronizing clocks simply to be able to define what the "one-way" speed is. If your convention is based on the assumption that the speed of light is isotropic (an excellent convention, by the way) then no experiment can prove or disprove that the speed of light is isotropic -- it is by definition.

You might also like to read this post from me.

 

[Jimmy Snyder]

Why would it? I am asking if the speed of light from the binary star to Earth is the same as the speed of light from Earth to the binary star. I can't see the relevance.

When you said "The answer 'yes' is implied by Einstein's postulate", I took you to mean that the one (invariance) implied the other (speed from A to B equals speed from B to A). If that is not what you meant, then my comment is wrong. But if it is what you meant, then the binary star experiment would verify the one, and implication would take care of the other.

As for the double interferometer experiment, once the two interferometers are calibrated so that there are no interference fringes, shunt one of the beams from one of the interferometers (A) over to the other interferometer (B) and use it as one of the beams there, keeping the other beam as it was. You should get an interference pattern dependent on the speed of light from A to B. (Put some glass in the A-B path and watch the pattern change in order to satisfy yourself that the pattern depends upon the speed). Now do the same from B to A. Compare the two patterns. Now rotate the experiment 180 degrees and look again at the patterns. The point here is that you would not be measuring the speed (no clock), but rather the difference between the speed from A to B and that from B to A.

All this now seems irrelevant to me in view of the definition you quoted. But I'm not sure why. Are you saying that if I defined the sun to be the moon, then every experiment would verify it?

One last question. Isn't this definition forced on us by the first postulate? If the speed of light from A to B were different from the speed from B to A, then experiments involving massive particles moving less than the speed of light in one direction would be possible that were impossible in the other direction because they involve particles moving at that direction's speed of light.