Could spacetime be non-orientable?

  • I
  • Thread starter Jarek 31
  • Start date
  • Tags
    Spacetime
In summary: The conversation is about the possibility of spacetime being non-orientable and its implications for P and T symmetry. In summary, some experts believe that non-orientable spacetime may be possible, but it is still a topic of debate and further research is needed. The concept of time-orientability and its relationship to non-orientability is also discussed, as well as the implications for thermodynamics and the second law.
  • #36
stevendaryl said:
Being non-orientable isn’t exactly the same thing as having closed timelike curves.
Having gone back and looked up some things about Godel spacetime, yes, I realize I was confused on this point:

PeterDonis said:
the entirety of Godel spacetime is non-orientable
PeterDonis said:
Kerr spacetime violates property #2; once you've entered the non-orientable region
PeterDonis said:
the OP of the stack exchange thread is wrong when it says the diagrams it refers to are of the only non-time-orientable spacetime ever described "outside of de Sitter space with some identifications" (not sure what that refers to either). Godel spacetime, which I've already described in this thread, is another.
I found a recent paper that discusses (briefly) the fact that Godel spacetime is time orientable:

https://arxiv.org/pdf/1911.08602.pdf

I would expect that Kerr spacetime is too, for similar reasons (basically, that even though there are closed timelike curves, they are in the "tangential" or "going around with the rotation" direction, so to speak, so they still have a consistent future/past orientation because the rotation in these spacetimes has a definite sense--it's a rotation going around in a definite direction).

That makes me wonder if there is any known spacetime that is an actual solution of the Einstein Field Equation but is not time orientable. As I mentioned previously, the diagram in Wald does not describe one; it's just an informal example and no corresponding solution to the EFE is given. The stack exchange thread mentions "de Sitter space with some identifications", but I have not been able to find any reference that explains what identifications.
 
  • Like
Likes vanhees71
Physics news on Phys.org
  • #37
PeterDonis said:
That makes me wonder if there is any known spacetime that is an actual solution of the Einstein Field Equation but is not time orientable. As I mentioned previously, the diagram in Wald does not describe one; it's just an informal example and no corresponding solution to the EFE is given. The stack exchange thread mentions "de Sitter space with some identifications", but I have not been able to find any reference that explains what identifications.
https://link.springer.com/article/10.1007/s41114-019-0019-x

Example 1.9
 
  • Like
Likes vanhees71
  • #39
So what could be consequences of time non-orientability?
Any chance for observational verification? e.g. time-reversed star, https://en.wikipedia.org/wiki/White_hole ?

Or imagine a rocket going through such time-inverting path and returning to Earth orbit - so would it have reversed time evolution from our perspective, e.g. "egg unscrambling", lasers stimulating absorption, entropy decrease?
 
  • #40
Jarek 31 said:
So what could be consequences of time non-orientability?
First you need to understand what I have already pointed out before: time non-orientability is not the same as time symmetry or time reversibility. Nor, as @stevendaryl pointed out, is it the same as having closed timelike curves. It is something much more counterintuitive. (Just looking at the examples in the paper @martinbn linked to should make that clear; those examples look like nothing any of us have brought up in this thread.)

Jarek 31 said:
Any chance for observational verification? e.g. time-reversed star, https://en.wikipedia.org/wiki/White_hole ?
The white hole spacetime, which is just Schwarzschild spacetime, is time orientable. So it's irrelevant to your question. So are all the other spacetimes that anyone (including me) has brought up, such as Godel spacetime--as I noted in post #36, I was wrong to think this spacetime was not time orientable; it is time orientable. So is Kerr spacetime, even in the region where it has closed timelike curves.

Jarek 31 said:
Or imagine a rocket going through such time-inverting path and returning to Earth orbit - so would it have reversed time evolution from our perspective, e.g. "egg unscrambling", lasers stimulating absorption, entropy decrease?
Again, this has nothing to do with time non-orientability. It is just time reversibility, which is not the same.

I suggest looking at the paper that was linked to and the example @martinbn referenced in it. Then take a while to digest it. (I'm still digesting it.)
 
  • Like
Likes vanhees71
  • #41
So what about this rocket going through such vector inverting path and returning to Earth orbit?

You previously agreed for P symmetry, but what about time non-orientability?
 
  • #42
Jarek 31 said:
what about this rocket going through such vector inverting path and returning to Earth orbit?
First we have to figure out whether any such path exists in a non-orientable or non-time-orientable spacetime. Remember that non-time-orientable is not the same as the existence of closed timelike curves. In the examples in Fig. 1.9 of the paper @martinbn linked to, there are no closed timelike curves, but two of them are non-time-orientable. So in those spacetimes, no such path as you describe exists.

Note that your definition of non-orientability in general, that a vector gets "flipped" if it is parallel transported around a closed path, does not mean there exists such a closed path that is timelike. So even if such a closed path exists, that does not mean that a rocket could actually travel around it.

Jarek 31 said:
You previously agreed for P symmetry, but what about time non-orientability?
My previous agreement is now irrelevant, since I was under a misapprehension about what non-orientability and non-time-orientability actually imply and don't imply.

I strongly suggest, as I said at the end of my previous post, looking at the examples in the paper @martinbn linked to and then taking some time to digest them. Pretty much everything we have been saying up to now, going by those examples, is irrelevant to the topic of non-orientability and non-time-orientability.
 
  • #43
The articles I have cited used Klein-bottle-like wormhole as a basic example - like a wormhole, but additionally inverting one of vectors.
The path doesn't have to be closed in spacetime, being closed in space is already interesting - I am writing "returning to Earth orbit".
 
  • #44
Jarek 31 said:
The path doesn't have to be closed in spacetime
For the definition of non-orientable that you gave, yes, it does. Just coming back to the same "point in space" isn't enough; the underlying manifold is spacetime, not space, and the path has to be closed in the underlying manifold.
 
  • #45
For a rocket traveling through such Klein-bottle-like wormhole inverting time direction, and returning to Earth orbit, I would say that just waiting there it could close the loop in spacetime ("turns out it was already there when launching").

But if you disagree, let's start with simpler scenario of just returning to Earth orbit ...
 
  • #46
Jarek 31 said:
For a rocket traveling through such Klein-bottle-like wormhole inverting time direction, and returning to Earth orbit, I would say that just waiting there it could close the loop in spacetime
"Close the loop in spacetime" means a closed timelike curve. Which, as I have already commented several times now (and as @stevendaryl pointed out before I did), is not the same as the spacetime not being time orientable. Of course it is possible for a spacetime to have both properties; but when you talk about "close the loop in spacetime", if that is possible, it is because of a closed timelike curve being present, not because the spacetime is not time orientable. So if you want to talk about the spacetime being not time orientable, you should focus on effects that happen because of it being not time orientable.

Jarek 31 said:
let's start with simpler scenario of just returning to Earth orbit ...
First you have to tell me which non time orientable spacetime you are using. (For example, you could use one of the examples in the paper @martinbn referenced.) Then you have to tell me where "Earth orbit" is in that spacetime and what "returning to Earth orbit" means--what worldline the spaceship is going to follow.

Until you do those things, it is impossible for I, or anyone (including you) to make any valid predictions about what might happen in such a scenario.
 
  • #47
(parachuting in this conversation)
I believe chirality would be an issue in this universe.
 
  • #48
Indeed, such hypothetical path inverting spatial direction would turn all molecules into their enantiomers, life into https://en.wikipedia.org/wiki/Mirror_life ... but also matter into https://en.wikipedia.org/wiki/Mirror_matter

Even more interesting would be path inverting temporal direction - e.g. imagine rocket traveling through it and returning to Earth orbit ...

Are they possible? Could be realistically verified observationally?
 
  • #49
Jarek 31 said:
Are they possible?
I have already told you how to answer that question in post #46. Without an actual model, such as the examples in the paper @martinbn linked to, there is no way to answer such a question. Are you going to look at an actual model or not?
 
  • #50
diegzumillo said:
I believe chirality would be an issue in this universe.
What universe? What model are you talking about?
 

Similar threads

Replies
7
Views
989
Replies
2
Views
3K
Replies
1
Views
850
Replies
2
Views
1K
Replies
12
Views
2K
Replies
1
Views
2K
Back
Top