Is Special Relativity Testable?

[dbkooper]

The basic premise of special relativity is that absolute motion is not detectable. Experimentally, this involves only two general cases, (i) detection via mechanical objects and (ii) detection involving light rays. Case ii readily resolves into two sub-cases, ii-a, detection involving round-trip light rays, and ii-b, detection involving one-way rays.Since cases (i) and (ii-a) were experimentally closed prior to special relativity (SR), this left only case ii-b for SR to actively pertain to.SR's prediction in the one-way light speed case is invariance.Specifically, SR predicts that all inertial observers must obtain the exact value c, and only that value, when using two clocks to measure the one-way speed of light. Since this does not happen in the case of slow clock transport, we can immediately eliminate that case. This leaves only the case of two always-mutually-at-rest clocks (in any given inertial frame).This raises the important question, How can SR be tested in that case? That is, how can an inertial observer use two clocks to experimentally measure light's one-way speed?

 

[Dale]

Special relativity is eminently testable:
http://www.edu-observatory.org/physics-faq/Relativity/SR/experiments.html

Your restriction to tests of the one way speed of light is entirely artificial. Furthermore, the one way speed of light is c by definition. It isn't something which can be tested independently and it can simply be defined to be true.

 

[ghwellsjr]

Specifically, SR predicts that all inertial observers must obtain the exact value c, and only that value, when using two clocks to measure the one-way speed of light.

Since this does not happen in the case of slow clock transport, we can immediately eliminate that case.

What is it that doesn't happen?

 

[Drakkith]

The basic premise of special relativity is that absolute motion is not detectable.

That is only one among many premises. As has been said already, SR is easily testable.

 

[harrylin]

The basic premise of special relativity is that absolute motion is not detectable. [..]

SR's prediction in the one-way light speed case is invariance.

Not exactly, and those things are related. The one-way speed is made equal to the two-way speed by means of clock synchronization - no "absolute" or "true" synchronization can be measured. The two-way speed is predicted to be invariant between inertial frames.

[..] This raises the important question, How can SR be tested in that case? That is, how can an inertial observer use two clocks to experimentally measure light's one-way speed?

That is not very useful, except to test for consistency or anomalities. Typically SR can be tested by verifying the two-way speed (and many other tests, see the FAQ).

Compare also this discussion: http://en.wikipedia.org/wiki/One-way_speed_of_light

 

[zoki85]

Of course, SR predictions are numerous and can be tested/checked in many ways. It isn't some esoteric theory like string theory

 

[dbkooper]

Special relativity is eminently testable:
http://www.edu-observatory.org/physics-faq/Relativity/SR/experiments.html

Your restriction to tests of the one way speed of light is entirely artificial. Furthermore, the one way speed of light is c by definition. It isn't something which can be tested independently and it can simply be defined to be true.

The one-way case is the only open case because we have Einstein's word that it could "go positive" given absolutely synchronous clocks. He wrote, "The velocity of propagation of a ray of light relative to the carriage thus comes out smaller than c." (This observer used the clocks of classical physics to measure the speed of a departing ray of light.) http://www.bartleby.com/173/7.html

(FYI: It is untrue that Einstein's two results conflicted with the principle of relativity because these were *not* two different general laws as he claimed, but simply two specific instances of the *same* law, namely, c±v. But even this matters not; all that really matters is the fact that absolutely synchronous clocks will not get invariance for the one-way speed of light. What does this say about "simply defining invariance to be true"?)

 

[dbkooper]

What is it that doesn't happen?

What does not happen is that no form of clock transport (slow, very slow, very, very slow, etc.) results in exactly c. Indeed, each form of transport results in a different value!

 

[Dale]

The one-way case is the only open case

The one way case is not open. Einstein's synchronization convention ensures that the one way speed is c, by definition. It is not a matter of experiment, just a definition of what it means for two clocks to be synchronized.

Anyway, the question has been answered. Yes, SR is testable. The thread is closed.