I refer here to the case of a light clock where each light pulse emitted travels forth and then reflected back between the two plates of the clock on the same axis as the direction of the moving clock:


image credits: https://www.ajuronline.org/uploads/Volume%209/Issue%202%263/92.3D-FixelleArt.pdf

Assuming you are using very short duration light pulses much shorter in duration than the time it takes them to travel to the opposite plate in the case where the clock was stationary, then when the clock is on the move and you register only the pulse moving in the same direction with the clock when it reaches the opposite plate, the result will be time dilation for the time recorded by this clock.

However my question is, if you register the time instead, only by using the reflected light pulse coming back to the initial plate, then will there be any time dilation effect?

  • 2
    $\begingroup$ Certainly there will be time dilation, but you need to be careful in this case because it gets mixed up with length contraction and relativity-of-simultaneity effects. That's why pedagogically people usually start with the transverse light clock. $\endgroup$
    – knzhou
    Commented Aug 30, 2023 at 20:58
  • $\begingroup$ My controversial take on the whole situation is that in the case of short duration light pulses and measuring the two-way path, the LLC operates under Galilean relativity and not SR. $\endgroup$
    – Markoul11
    Commented Aug 31, 2023 at 18:10

1 Answer 1


The time dilation formula in SR is a special case, and applies only to intervals between pairs of events that occur in the same place in one of the two frames being considered. For that reason, neither the outbound leg nor the return leg of the light pulse's journey will by itself give you the time dilation effect described by the equation. Indeed, what you will find is that in the non-clock frame the outbound leg will be longer than the return leg.

  • $\begingroup$ I found this related Arxiv paper arxiv.org/ftp/arxiv/papers/1103/1103.0445.pdf $\endgroup$
    – Markoul11
    Commented Aug 31, 2023 at 9:53
  • $\begingroup$ The paper contains mistakes. The second principle of relativity dos not say that it is impossible to detect motion, but that it is impossible to detect absolute motion. $\endgroup$ Commented Aug 31, 2023 at 10:35
  • $\begingroup$ Referring to the above previous comment Arxiv paper after reading it I found that this is a very strange special case where when the light pulse is very short in duration and not a continuous wave therefore there is no Doppler shift observed, then the two-way speed symmetry postulate of light is broken and can be restored only if length dilation is assumed which is an impossibility in SR. $\endgroup$
    – Markoul11
    Commented Aug 31, 2023 at 10:39
  • 1
    $\begingroup$ I don't understand what point you are trying to make. The case for SR is rock solid. The paper you cited was not peer reviewed, talks nonsense, and has gained no serious interest since it was published. So what exactly is the problem? $\endgroup$ Commented Aug 31, 2023 at 12:57
  • 1
    $\begingroup$ @robphy "...also the length of the longitudinal light clock must dilate at the same manner" last paragraph before the References. arxiv.org/ftp/arxiv/papers/1103/1103.0445.pdf $\endgroup$
    – Markoul11
    Commented Sep 3, 2023 at 11:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.