There are other questions on this site about the CMB, but none of them answer my question specifically. I am not asking about fluctuations in the CMB.

How could inflation affect the CMB?

Now as far as I understand, photons do have stress-energy, and do exert gravitational pull on each other, especially in the early universe, when the CMB was very dense.

If you use General Relativity instead you'll find that photons make a contribution to the stress energy tensor, and therefore to the curvature of space.

Does a photon exert a gravitational pull?

How did space expansion overcome uniformly this gravitational pull that the CMB photons' had on each other?

Now just to clarify, imagine the early universe, very energetic photons, very dense CMB, so the photons' gravitational pull on each other must be significant. How can space expansion overcome this uniformly? Analogously space was filled with matter particles, those had gravitational pull on each other too. These matter particles clumped up into denser parts, which became galaxies, and voids inbetween.

Why didn't this happen with photons the same way? Those energetic photons in the early universe had significant gravitational pull on each other, like matter particles. But, CMB is uniform everywhere, no denser parts.

Another interesting thing is, there is even something called the geon.

In theoretical general relativity, a geon is a nonsingular electromagnetic or gravitational wave which is held together in a confined region by the gravitational attraction of its own field energy.



  1. Why didn't the CMB clump into denser parts like the matter in galaxies with voids inbetween since the photons exert gravitational pull on each other?
  • $\begingroup$ Perhaps I totally misunderstood your question. But the CMB did clump togheter forming younger stuff. $\endgroup$
    – Alchimista
    Apr 9, 2021 at 9:08
  • $\begingroup$ @Alchimista can you please elaborate on this one: "the CMB did clump togheter forming younger stuff"? $\endgroup$ Apr 9, 2021 at 15:40
  • $\begingroup$ you likely refers to photons alone. But all structures and void we have in the universe were in/at the bigbang. $\endgroup$
    – Alchimista
    Apr 10, 2021 at 12:39

2 Answers 2


The (baryonic) matter clumps because it can interact dissipatively. i.e. It can lose kinetic energy and sink into potential wells without flying out again.

If a photon "falls" into a potential well it either emerges on the other side or it is absorbed. But the whole point about the CMB formation is that the radiation decouples from the baryonic matter and thus no longer follows the matter density.


You need to pick a context. The common context for CMB discussion is General Relativity (GR). In GR gravity is not a "pull". GR defines the influence of gravity by determining the paths that things move on. In particular photons move at velocity c on null geodesics. (See https://en.wikipedia.org/wiki/Geodesics_in_general_relativity .) The CMB photons also are moving away from each other, not towards each other. Therefore they cannot "clump".

  • $\begingroup$ @ProfRob Imagine during the period of CMB creation that two protons are at a small distance apart. Assume that each proton simultaneously captures an electron and becomes a hydrogen atom, while emitting a photon. Assume the photons move directly towards each other. What happens when the photons "meet"? If they have sufficient energy equivalent to the mass of a particle and anti-particle pair, would these photons (perhaps) convert into the particle pair? I assume this might happen so rarely that they would have zero observable affect on the CMB. $\endgroup$
    – Buzz
    Apr 9, 2021 at 17:29

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.