Suppose that we have a massive stellar object in space, so massive that it can cause significant lensing. It is also theoretically possible to turn light by 180 degrees (u-turn).So, assuming this body is x light years away, suppose we oberve it, and the light it sends back. Then are we able to look back at the past? For eg: suppose light from the sun takes x years to reach the body, execute a u turn and another x years to come back to us. So technically by observing this, we will gain knowledge about the past of our sun? Is this idea correct? If yes, has it been used to study stuff like the origin of our solar system and all?

  • $\begingroup$ Is there some reason to suspect this wouldn't work? It seems quite straightforward, and I don't really understand why you wanted to ask a question about it. $\endgroup$
    – David Z
    Jul 11, 2016 at 17:17
  • $\begingroup$ You can use the formalism used in this article: arxiv.org/abs/gr-qc/0102068 $\endgroup$ Jul 11, 2016 at 17:20
  • $\begingroup$ Has this been applied anywhere? If it has been, should'nt we have progressed more than we already have? Thats all i wanted to know $\endgroup$
    – Lelouch
    Jul 11, 2016 at 17:21
  • $\begingroup$ I'm not aware of anyone having done detailed calculations, you should read the paper and then do some calculations on how well you can see the image of the Sun due to photons looping around a black hole. $\endgroup$ Jul 11, 2016 at 17:26
  • $\begingroup$ Related: physics.stackexchange.com/q/11940/2451 and links therein. $\endgroup$
    – Qmechanic
    Jul 11, 2016 at 18:28

2 Answers 2


There's nothing in principle wrong with "observing the past" -- after all, when you look in a mirror, you're seeing yourself as you looked a few nanoseconds ago. But using astronomical objects as faraway mirrors is not practical, for a few reasons:

  • An object dense enough to bend light all the way back to its source would be either a black hole or something extremely close to being one. It wouldn't emit any light itself, so the only way we could know where it is would be if it's surrounded by hot glowing matter, which would get in the line of sight to our past.

  • The image of our past would be compressed to an extremely thin ring hugging the "disk" of the black hole -- much smaller than the disk itself. But black holes are pretty small as astronomical objects go. Even a supermassive black hole such as Sgr A* is only a few times larger than our sun in diameter. We can just about resolve disks of stars of that size in our immediate galactic neighborhood with the best current instruments, but superminor details on the fringe of a disk? No way.

  • In order to get interesting information about the evolution of the solar system, we would need to see light reflected from a mirror at least several hundred million light years away, and preferably some billions. Then, again and with force, distance kills the idea. We can't even distinguish individual stars (or black holes) in galaxies at that distance. Minuscule detains on fringe of one of them? And that minuscule detail would be an image of the entire Milky Way seen from way out there, and inside that image we would need to resolve and identify our Sun, one among a hundred billion or so stars? That's so far out that even science fiction wouldn't dream of it.


I'll add to this answer. According to that answer, Earth cannot be seen as more than an inextremely faint thing by watching light that left Earth then looped around and came back. However, my question Does general relativity predict that it's possible to watch a process playing backwards? suggests that maybe general relativity does predict that it's possible to have something like the sonic boom but with light. Maybe's it's possible to see Earth's own luminous boom from Earth if its light bends around 2 black holes each of which are going a significant fraction of the speed of light in the frame of reference of Earth.

This question actually appeared in the review of that question before I posted it. Maybe it's the amazing ability of the Stack Exchange review software to appear to have this question appear in the review of that question because it has some sort of vague connection with that question related to the fact that I could reasonably include something from that question in an answer to this question.


Your Answer

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

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