Imagine a very large black hole, with a mass equal to a large number of galaxies. Assume a space station is in orbit around the black hole some distance from the event horizon at a point A. An astronaut in a ship, let's call him observer B, launches from the station toward the black hole. I have been told (correct me if I'm wrong) that from observer B's point of view, he wouldn’t feel any effects from approaching the event horizon—at least as far as he could determine from inside his spaceship. In fact, the visible event horizon, from his own point of view, would continually move away from him so that he could never actually reach it. Is this correct? If he were to watch the space station out his window, he would see it orbiting faster and faster. He would see the space stations occupants racing around. Eventually he would see many generations on the space station be born, live and die. As he approached the event horizon (as seen by him or the original Schwarzschild radius as seen by A?), he would see stars born and die and eventually galaxies form move out and disappear. He would see the entire life of the universe in an instant as he got ever closer to the black hole. Is this correct?

Mathematically, the observer on the space station could calculate these relativistic effects. In fact, observer A would calculate that on the other side of the event horizon, the space and time axes would actually be swapped relative to his own point of view. Of course, he cannot see beyond the event horizon, so this would be a strictly mathematical exercise.

Now this is based on my reading of several books on general relativity. If any of it is wrong, please tell me. Otherwise, I’m assuming I’ve interpreted things correctly. My thoughts turn to the following thought experiment.

Now, I do this thought experiment. I imagine many stars and galaxies moving around inside the event horizon of this black hole (remember it is very large). And I consider a layer S a fixed distance from the singularity, but located inside the event horizon as observed from some point, say our original station at A. Let’s assume a person is located at a point P on the surface S. Nothing between P and the singularity could be seen by this observer P. Any light between the singularity and P would move inward towards the black hole, so the observer at P would see nothing but blackness if he peered in the direction of the singularity. True?

What I want to know is if observer P peered outward, toward the universe, what would he see? I’ve been told due to gravitational lensing he would see all the light entering the event horizon concentrated somehow. I also want to know what the stars and galaxies located inside the event horizon but outside the shell S would look like to the observer at P. Is it possible that as he looked outward, he would be looking into the past? Is it possible that he would see all of space limited to his little shell S, and points further out than P would be points in his past? Is it possible that at the event horizon, he would see his entire shell contract to a point, as if it came from a big bang in his past? And is it possible that the spacetime he is unable to perceive, between P and the singularity, appears to be in his future? From his own point of view, he can’t perceive it, because nothing can reach him from that future (from closer to the singularity) without travelling faster than light.

I don’t know if any of this is correct. Just something I’ve been thinking about and it could be totally out to lunch.


He would see the entire life of the universe in an instant as he got ever closer to the black hole. Is this correct?

Absolutely not.

Check out a Penrose diagram, or just look at a black hole in e.g. Kruskal-Szekeres coordinates. These are designed to easily show what an events sees.

In Kruskal-Szekeres coordinate the singularity looks like the curve $y^2-x^2=k^2$ and and light moves on regular 45 degree lines like in Minkowksi space.

The horizon is the line $y=x$ and events to the right of it at the outside. Lines like $y=mx$ for $m$ with $|m|\lt 1$ are surfaces of constant Schwarzschild time and hyperbolas like $x^2-y^2=a^2$ are surfaces of constant Schwarzschild areal $r.$ The constant $m$ is unrelated to the mass $M$ of the black hole.

So. If the mass $M$ is very very large the intrinsic invariant curvature at the horizon is very very small. But if you look at the specific event where a specific object crossed the horizon there is a last light cone. And those are the events it sees before it crosses. Sure it sees points that have very late Schwarzsschild time coordinate, but those events are the ones super close to it. Basically stuff that got closer earlier and crossed you see the light from before it crossed. And you see it crossing when you cross. But you don't see anything cross that crosses after you cross.

So you can see things that crossed before you do but not things that cross after you. At least while you are still aren't inside. For distant parts of the universe you see events from farther in the past. Its a regular past light cone extending from the event of your crossing.

OK. Now extend your curve to the singularity. There is still an event for that. And a past light cone for that. The events outside it are events you never ever ever see. Though now we are inside so these claims are not verifiable unless maybe you go inside too. And while you could see things cross earlier than you by seeing the image of it crossing st the event where you crossed, you never ever see anything hit the singularity except possibly you yourself.

Again look at the event of you hitting the singularity and check out its past light cone. Now before you hit that singularity you get torn apart, so you can't take a nice picture of you hitting either.

But the main point is that if you push someone into a black hole and program their craft to follow a particular path and you believe they really follow it and nothing stops it from crossing then you can wait a finite amount of time and start doing things and have those events be outside their entire past light cone.

They won't see those events. The issue is they won't see those events if their ship takes that route. But you will never see them cross. You won't ever know they actually followed that trajectory. So if you started to do thing you hoped they will never see, you still might meet them and hear that they changed their mind, that the ship turned around, etcetera. And there is no finite amount of time you can wait and confirm they crossed.

So if they cross there will be things they don't see before they cross. And if they cross they will hit the singularity and based on how they hit it there will be things they never see, ever ever ever. But you will never know what they did, no matter how long you wait, unless you cross too to confirm they crossed. And then you will never see them hit the singularity so its hard to know for sure that they were crushed.

And that's because if the block hole is larger and they enter first and rush one way and you enter where the event of you crossing the horizon is outside their entire past light cone then it is possible the laws of physics are different in their region and your wouldn't know because you only see part of it, the parts before they were crushed.

So if you never see them get crushed and then you get crushed then its entirely theoretical to you that they get crushed.


The person going towards the center of the black hole would be spaghettified as any other matter would as it enters the center and die. However, an outside observer would see the person approaching the middle as slowing down before completely stopping, because it takes light longer and longer to reach this observer the closer the other astronaut approaches the middle due to the gravity. Source

  • $\begingroup$ While correct, this does not answer the posted question. $\endgroup$ – Asher Oct 22 '15 at 19:36
  • $\begingroup$ A black hole of the enormous size proposed in the question, the person in the ship would need to be deep inside, well below the event horizon before spagetification was a problem. $\endgroup$ – userLTK Oct 22 '15 at 20:26

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