# Reichenbach on relativity

1. Sep 8, 2014

### phellen

What do people think about the conventional element of SR? Is the assumption that the one way speed of light is always the same (for any rf) justified (as opposed the the two way speed of light)?

2. Sep 8, 2014

### pervect

Staff Emeritus
Personally, I suggest (conceptually) reformulating the two-clock concept of velocity to the one-clock concept concept of proper velocity (also called celerity).

There's some discussion of the definition of this in http://arxiv.org/abs/physics/0608040 though they do not suggest exactly what I am suggesting as far as using this concept.

I regard the definition of velocity using two clocks as being mainly inspired by the experimental difficulties in realizing accurate clocks that move along with the object whose velocity we wish to measure.

However, if we regard the concept of velocity as something that could be measured by one clock, that's carried along with the object whose velocity we wish to measure, we no longer have to worry about synchronization issues. Though we do need to note that celerity is numerically not the same as velocity, it's related only on a philosophical / conceptual level as being another way to measure "speed".

Thus, rather than dealing with all the synch issues, my suggestion is to reformulate the problem in terms of clocks that measure proper time, by using the concept of celerity to replace the concept of velocity.

This won't really answer issues related to the "one-way speed of light", because in this model the celerity of any object approaching the speed of light appraoches infinity. But I think it's more productive than agonizing over the synchronization issues, in the belief that they are fundamental and unavoidable. I believe that this prescription allows one to avoid the synchronization issues and reveals them to be actually non-fundamental.

3. Sep 8, 2014

### WannabeNewton

Well Malament proved under certain assumptions that there is a preferred and natural synchronization convention for inertial frames defined by the causal structure and this is the one in which the one-way speed of light is isotropic; this convention also agrees with slow-clock transport. So in that sense it is clear that the $\varepsilon = 1$ convention holds a special place.

Of course this is still not the only possible convention as one can still freely choose other synchronization conventions for inertial frames even if they are very unnatural. Malament's argument was that nature naturally picks out $\varepsilon = 1$ in inertial frames so it is not only justified but also preferred. Other conventions would of course also be justified but they would the calculations, amongst other things, intractable due to global time functions built out of more complicated synchronization prescriptions.