Imagine a Pulse of light traveling through space at $c$, coming towards an observer on Earth, while at the same time, Space-time fabric (metric) is continuously changing (expanding), then why is the speed of light constant throughout space-time, since the separation of the very two-points in space, light is traveling in between, is not constant?

My guess to that was $c=\lambda\nu$ but, how does Frequency of light and its Wavelength changes in just the right way so there product gives the speed of light and not a speed lesser than that of light(because space is Expanding)?

A few articles also argued about its effect on Sommerfield's constant, but I've read that String theory allows Sommerfield's constant to change over time.

I am not a GR head (yet) so, this post is bound to have a lot of things wrong (or maybe, all of them) so kindly keep your explanations as descriptive as possible. It'll be really helpful if you could provide some intuitions or examples for the same.

  • 2
    $\begingroup$ Maybe this answer and this link can help? $\endgroup$
    – Stratiev
    Commented Jun 5, 2020 at 7:49
  • $\begingroup$ I'd clarify that the experimental evidence only shows that the two-way speed of light is measured to be constant, which is a considerably less strict claim than saying the speed of light is constant. $\endgroup$
    – Steve
    Commented Jun 5, 2020 at 9:21
  • $\begingroup$ @Stratiev Thanks, those links surely clarify a lot of things. But, I couldn't find anything about the effects of Space-time expansion and variations in the speed of light. Also, Is it even a "valid" Question to ask? It'll be helpful if you can shed some light on that. $\endgroup$
    – Samarth
    Commented Jun 6, 2020 at 9:44
  • 1
    $\begingroup$ @Samarth, because in Lorentzian interpretations of relativity, the one way speed of light is not constant except in a universally preferred frame (which that interpretation postulates). The two-way speed emerges as a constant in all frames, because the so-called anisotropy of the one-way speed of light in a non-preferred frame, is balanced in opposite directions (i.e. if it's slower in one direction, then it's faster by the same amount in the other). For practical reasons, physics can only measure the two-way speed of light, so there is no specific scientific claim as to the one-way speed. $\endgroup$
    – Steve
    Commented Jun 6, 2020 at 10:45
  • 3
    $\begingroup$ Does this answer your question? Has the speed of light changed over time? $\endgroup$
    – tparker
    Commented Jun 8, 2020 at 0:19

2 Answers 2


This became too long for the comments. Before I continue, maybe you should take a look at this answer too. I don't claim the following is a good answer but maybe it gives you ideas...

So I think the point is that the speed of light will vary if you are in a reference frame, that is experiencing acceleration/gravity. If you are in an inertial reference frame, the speed of light is $c$. This is in one of the links that I mentioned in the comments, but let's just illustrate it through an example. Suppose we consider an observer in the background of a Schwarzschild black hole with a Schwarzschild radius $r_S$ and a distance from the singularity of $r$. The metric is

\begin{equation} d \tau^2 = - \left(1 - \frac{r_S}{r} \right) c^2 d t^2 + \left(1 - \frac{r_S}{r} \right)^{-1} dr^2 + r^2 d\theta^2 + r^2 \sin (\theta)^2 d\phi^2. \end{equation}

Now if we are a particle of light following a null geodesic, we have that $d \tau=0$. The instantaneous radial velocity is \begin{equation} c'=\frac{dr}{dt} = \left(1 - \frac{r_S}{r} \right) c. \end{equation}

So you see that far from the singularity, when $r\gg r_s$, we have that $c' \rightarrow c$. Whereas, in the vicinity of the black hole horizon, $c'$ can become arbitrarily small.

Now I think that to answer your question about the variability of space-time, you might have to repeat the same calculation for the FLRW metric, for example. You will get some variation, that I'm not sure how you can measure, but if you were to measure the speed of light locally, you'd still get $c$. I hope someone else can give a better answer to this.


These are different ideas. The local speed of light is constant. That is the speed of light as measured locally by an observer. This is unrelated to the increasing distance of the point from which the light was originally emitted.

To deal with large regions of space, involving expansion, we have to use maps involving scaling distortions, much as we do when we map the surface of the Earth. Usually we use coordinates in which objects (galaxies) remain the same size, and distances increase over time. In such a map the coordinate speed of light does not remain constant. An equivalent way of mapping shows galaxies getting smaller over time. In such a map the radial coordinate speed of light can be constant enter image description here

For this to work, the rate of time on has to be increasing, so that the observed laws of physics are the locally always the same. Since the rate of time increases, the observed frequency of light decreases. So basically, the answer to the question is that the local laws of physics remain always the same, and this means that the wavelength and frequency of light must change in just such a way that the local speed of light remains constant.


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.