What experiments confirmed the constant velocity of light?

[Nugatory]

it is just that all those particular experiments, GPS or particle accelerators do not seem, at this moment, to be direct proofs for me .

Of course they aren't - no experiment is ever a direct proof of any physical theory. Instead, experiments disprove theories, by demonstrating results that are inconsistent with the predictions of the disproven theory.

 

[Dale]

The theory of relativity is based ( or seems to be) on the invariant speed of light. If this can be proven, than clearly all else must stand.

It is proven, not in the deductive sense of a mathematical proof, but in the inductive “beyond a reasonable doubt” sense.

For that, you should not rely merely on the three experiments I highlighted addressing your specific interest. Instead you should consider the totality of the evidence. Any alternative to relativity must explain ALL of those experiments. Not merely in vague qualitative terms, but with specific quantitative predictions matching the experimental results for each experiment.

 

[jbriggs444]

The theory of relativity is based ( or seems to be) on the invariant speed of light. If this can be proven, than clearly all else must stand.

This is an attractive way to think about things. But it is (in my opinion) also a wrong way to go about deciding where to look for evidence in favor of relativity. Or for a proof of the validity of the theory.

Yes, the theory of relativity is based on the postulated invariance of the speed of light. However, the original paper is not based on any experimental proof that the speed of light is invariant.

In mathematics, a "postulate" is not a self-evident truth. It is a statement that is accepted without proof. [One could go into depth on that and talk about things like the parallel postulate, the continuum hypothesis or the axiom of choice, but that is material for another sub-forum]

Maxwell's equations had suggested that the speed of light should be invariant. You measure a couple of parameters, take the square root of the inverse of their product and out pops the velocity of an electromagnetic wave according to those equations. Which matches the speed of light.

It would be nice if this velocity were an invariant -- the same in every reference frame. Unfortunately, according to our standard way of translating from one reference frame to another, there was no such thing as an invariant velocity. According to our standard translation scheme (Galilean relativity), if you switch to a new frame moving at +v compared to the old, all of your old velocities get v subtracted from them as a vector. The new speed of light should be c+v, c-v or something in between depending on the angle of the light compared to the relative motion of your two frames.

But what if the speed of light actually were an invariant? Always equal to c regardless of frame. Take that as a postulate. See if it is possible to build a coherent system of physics that adheres to that postulate without reaching a contradiction.

It turns out to be possible to do so. The resulting theory gives up some properties of time that we had previously taken for granted. But it does not reach an outright contradiction. The resulting theory (Special Relativity) also makes some unexpected predictions.

When experiment conforms to those predictions and fails to conform to the Newtonian predictions, that is what makes us take special relativity seriously.

 

[cianfa72]

Maybe is not entirely on topic, anyway we are actually talking about the constancy/invariance of the two-way (round-trip) speed of the light from an experimental point of view.

BTW one-speed of light is actually a matter of convention. I found this old post from @PAllen

Let's say you build two atomic clocks right next to each other, get them exactly synchronized, and move them slowly 1 meter apart, with exactly symmetric acceleration profile. You assume this keeps them synchronized. And it does - exactly equivalently to as if you synched them using Einstein's clock sync based on assuming isotropy plus invariance of two way light speed. Maybe not emphasized enough is that any procedure that relies on any assumption of isotropy is informationally worthless for resolving one way light speed, because it builds in the answer given invariance of two way light speed. The assumption that slow separation with the same acceleration profile (in opposite directions) leaves the clocks synched is an isotropy assumption [it also builds in a homogeneity assumption].

Isotropy & invariance of two-way speed of light results in constancy/invariance of the one-way speed of light, I believe.

 

[berkeman]

Thread closed for Moderation...

 

[berkeman]

After a Mentor discussion, thread is re-opened...

 

[Grampa Dee]

So for example, the most stringent astronomical constraint is:While the most stringent terrestrial constraint is:This one is fun just because the experimenter was anti-relativity but his experiment proves the point you are asking about:This is simply factually untrue. When testing relativity or measuring the speed of light, you cannot and do not assume it is c. That would defeat the whole point of the experiment. This is a completely false and very uninformed complaint, particularly when it is made so broadly such that all experiments are implicated.

Beckmann and Mandics, “Test of the Constancy of the Velocity of Electromagnetic Radiation in High Vacuum”, Radio Science, 69D, no. 4, pg 623 (1965).

A direct experiment with coherent light reflected from a moving mirror was performed in vacuum better than 10−6 torr. Its result is consistent with the constant velocity of light. This experiment is notable because Beckmann was a perennial critic of SR. Optical Extinction is not a problem.

Perfect! This is exactly what I was looking for; thank you Dale...I downloaded a description of the experiment and will try to understand the process.

One might ask why do I not simply accept a theory being over a century old, having good observational evidence ? For me, it is because of the manner the theory came into being.
After Young's double slit experiment, light was proven to be a wave; so, now, a medium was needed which was called the "ether". Next, the Earth must travel also within this ether, hence the M&M experiment. But there was a nul result, meaning no velocity was detected.
Afterwards, it seemed things started to go a little weird, in my opinion. Fitzgerald, I believe, claimed that there must be a length contraction in the direction of the Earth's movement relative to the ether.
The contraction is in fact the Lorentz transformation. Next,Einstein claimed that there is no ether, but the Lorentz transformation still stands as the velocity of light is now postulated as being invariant. Besides, there was Ritz' hypothesis that claimed the velocity of light was source dependant.
Now all three theories mathematically agreed with the M&M experimental observation.
For the ether theory, the Lorentz transformation was used, Relativity didn't need the transformation at all since everything was stationary within the experiment, and the light's velocity for the Ritz's theory was also "c" since the source was stationary as well.
So, my question was what evidence do we have for the invariance velocity of light? While the DeSitter's observation supported the invariant velocity of light, we needed to be certain that the light observed had not past through some medium such as H or He gas clouds, for then, the (c+v)(c-v) effect would have been lost.
Therefore, for me personally, an experiment highly supporting the invariant velocity of light was needed...afterwards, everything else would then follow, for all the paradoxical elements in the relativistic observations, are due to the postulate in the first place...
I am satisfied with what you've given me, I thank you again...this thread is over for me, personally..I may however answer a couple of more posts because I want to try to correct the unforgivable error I wrote in one of my post. :)

 

[berkeman]

Aaand, thread is closed again for more Moderation...