If the mass of the universe were cut in half, would it affect the speed of light?

Would it be twice as fast?

Would it stay the same?

Do we have instruments that are sensitive enough to measure the speed of light at different positions relative to high-mass objects to empirically answer this question?

The speed of light is (something of) a universal constant, but is it really dependent on the universe or on something intrinsic to photons?


Related question:

Since gravity is a relationship between one atom and every other atom in the entire universe, and it takes all the energy in the universe to travel at the speed of light, is there something about the energy/gravity/mass of the universe that "slows" light from going a faster speed?

  • 1
    $\begingroup$ It's not possible to define whether c changes from one place to another. It's only possible to define whether a unitless parameter such as the fine structure constant changes. See Duff, 2002, "Comment on time-variation of fundamental constants," arxiv.org/abs/hep-th/0208093 $\endgroup$
    – user4552
    May 3, 2013 at 18:52
  • $\begingroup$ @BenCrowell: is that because (at c speeds) time passes in a noticeably different way that is relative with respect to an observer? $\endgroup$ May 3, 2013 at 18:55
  • $\begingroup$ No, it's for the reasons described in the paper that I linked to. $\endgroup$
    – user4552
    May 3, 2013 at 21:19
  • 2
    $\begingroup$ Where ever you heard or read "it takes all the energy in the universe to travel at the speed of light", you should not take it as a definition, it's a poetic way of saying you would need arbitrarily large amounts of energy. $\endgroup$ May 3, 2013 at 22:04

2 Answers 2


The speed of light is entirely a local concept - it does not care if there are 10 atoms or 10 billion galaxies somewhere in the Universe.

Obviously we can't go to distant galaxies to directly measure the speed of light, so in the absolutely strictest sense this is not directly empirically tested. However, the constancy of the speed of light is one of the most fundamental tenets of physics. In some sense, just about every observation we make in astronomy tests it, for if there were any variation it would manifest in all sorts of crazy ways in every single system we look at.

The confusion seems to stem from the term "universal." The word "universal" means "fundamental" or "unchanging in space and time" or "lies at the heart of our theoretical framework, permeating everything we do." It does not mean "tied to the Universe" or "depends on global properties of the Universe." Along the same lines, a scented candle could be said to have an "earthy" scent, but this has nothing to do with it being located on Earth the planet.

  • $\begingroup$ My interest was not so much the word "universal" as much as the fact that it takes all the energy in the universe to go the speed of light. $\endgroup$ May 3, 2013 at 18:53
  • 1
    $\begingroup$ @micahhoover Ahh, I see. Well, it takes more than all the energy in the universe to accelerate any massive particle to $c$, no matter what universe you are in - it just can't be done. When people say it takes all the energy in the universe, what they mean is that no finite amount of energy will suffice, and the entire universe is the "closest thing to infinity" they care to imagine for the comparison. Things that do move at $c$, like photons, are simply always doing that - nothing ever accelerated them, and nothing will decelerate them. $\endgroup$
    – user10851
    May 3, 2013 at 19:00
  • 1
    $\begingroup$ I think it's even misleading to say "it takes [amount of energy] to accelerate a massive object to the speed of light," regardless of what the amount of energy is (yes, even if it's "infinite energy"). It's just impossible to make a massive object move at $c$. No matter how much energy you have, it cannot be done. $\endgroup$
    – David Z
    May 3, 2013 at 19:05
  • $\begingroup$ @David Zaslavsky... if you were to accelerate an object to the speed of C - planck's constant or C - 1/100planck's constant botch which have enormous but finite energy requirements... could quantum mechanics (ie the combination of uncertainty and fuzziness/quantum jumping) allow you to accelerate to C or beyond it? $\endgroup$ May 4, 2013 at 1:47
  • $\begingroup$ Ignore the comment I am going to post as a separate question $\endgroup$ May 4, 2013 at 1:48

More of a speculation - doesn't red shift & time dilation cause a photon to see all the universe contracted to a point directly ahead of it? So, couldn't the speed of light be determined by the gravitational attraction on a massless particle of the whole universe? And, since the speed of light is not infinite, doesn't this imply that the universe (at least the one we can see) is not infinite, also? That light has a speed then would also mean there's a time (13.8 billion years ago, or more if latest Webb observations are confirmed) beyond which the universe becomes undefined. In fact, maybe the early inflation of the universe is an artifact - a time when the universe appears to us smaller - so with a larger Plank distance, so a less precise subdividing of time & space. As time & space got more precisely defined, they appeared bigger. Just some sci-fi speculation for entertainment purposes only - maybe.


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