# Why nothing is observed going back in time

1. Jan 27, 2008

### A.T.

Could it be that an observed object moving back in time is indistinguishable from an object moving forward in time? To measure the signed rate of time flow we could use a light clock with 3 mirrors (A, B, C) reflecting the light beam in a triangle, in clockwise direction (A -> B -> C …). If someone would observe this clock moving back in time, he would see the light beam traveling counterclockwise (A -> C -> B …).

So it appears that we could determine if such a clock is moving backward or forward in time. But what if the entire clock was mirrored in space, when moving back in time. Then the light beam traveling counterclockwise would still hit the mirrors in the normal order (A -> B -> C …), and we could not distinguish it from a clock moving forward in time.

Now you ask - Why should a clock moving back in time be mirrored in space? One reason for this assumption are the formulas for time dilation und length contraction, which are connected by the Lorentz factor. A clock with a negative proper time (going backwards in time) would imply a negative Lorentz factor. Put into the length contraction formula, this would imply a negative scaling along one space dimension (which means mirroring). Aside of this pure algebraic approach there is also a geometrical interpretation of this formulas, presented by L. C. Epstein in his book “Relativity Visualized”. He uses a space-propertime diagram, where time dilation and length contraction are determined by simple projections on the time and space axes respectively. An interactive version can be viewed here:
From this, it can be easily seen, that a negative proper time component, would mirror the spatial representation of the object.

So maybe traveling back in time is not that exciting at all, and takes place all around us, without being noticed. The key point is that nothing can be observed to change it’s direction in time. So we can never see those objects going back in time, when they are not mirrored. Therefore we assume their mirrored appearance to be normal.

OK, enough unverifiable speculations for today.

Last edited: Jan 28, 2008
2. Feb 2, 2008

### A.T.

A have a related question about Feynman Diagrams: Could they be interpreted as space-(proper)time-diagrams, where anti particles move back in proper time?

This would be a nice parallel to the Epstein's diagrams, which suggest, that objects observed moving back in proper time are also observed to be mirrored spatially. In a certain sense a positron is just a mirrored electron, because where the left-hand-rule applies to the one, the right-hand-rule applies to the other.

3. Feb 2, 2008

### JesseM

I keep asking this when it comes up, but what do people mean by "moving back in time" or "moving forwards in time"? As I've said before, the Feynman diagram thing seems to be a sort of mathematical trick for how you perform a certain sum-over-Feynmam-diagrams to get predictions in quantum field theory, I don't think it's literally saying that particles are "moving" in time in one direction or another, that doesn't even seem meaningful to me.

4. Feb 2, 2008

### A.T.

Most people mean by "moving back in time" actually "moving back in coordinate time" so they can kill their grandfather and see what happens (Makes you afraid to have grandchildren). What I mean in this thread by "moving back in time" is "being observed as moving back in proper time". If you observe a clock going backwards, you could say it is "moving back in time".

I know that Feynman-diagrams are usually seen as graphs, and as not space-time diagrams. But if they have the time & space axes on them, they look very like space-propertime-diagrams: particles moving forward in proper time, antiparticles moving backward in proper time and photons moving only trough space.

5. Feb 2, 2008

### HallsofIvy

Staff Emeritus
Yes, it's true that Feynman diagrams are a "mathematical trick"- until someone comes up with a physical interpretation! Then it would become a physical theory that could be verified or falsified by experiment.

Actually, if I remember correctly it was Feynman himself who suggested that the motion of a positron could be interpreted as an electron "moving back in time". In fact he suggested that was why all electrons are identical. There is actually just one electron in the universe, moving backward and forward in time!

Of course, Feynman was a well known practical joker.

6. Feb 2, 2008

### peter0302

It's called a VECTOR. And the vector always points toward +t.

7. Feb 2, 2008

### JesseM

What does that mean, though? "Movement" normally means change in position over time...what does it mean to "move" in time, either forward or backwards? In a 4-dimensional picture where time is another dimension, you just have paths that go through various points in spacetime, but they don't "move" through spacetime any more than a line drawn on a 2D piece of paper moves "up the page" or "down the page".
It's easy to design a clock that ticks backwards, so you must mean a little more than that. Perhaps you are talking about the arrow of time defined by change in entropy, which is also thought to be the source of the psychological arrow, the fact that systems (such as the brain) can store records of the past but not the present. In this sense, if we see a macroscopic object with its entropic arrow reversed, including an ability to remember things that haven't happened yet, then I suppose we could say it's "moving backward in time" (although in the case of actual proposed time travel schemes like a wormhole, I don't think there's ever actually a case where one observer sees another going backwards in this way, rather the time travel involves moving through a wormhole that connects one location with another in its past or future light cone). But I don't see how this notion could be applied to individual particles which don't have a thermodynamic arrow of their own, only multiparticle systems can have such an arrow.
But even if they are seen as spacetime diagrams, what does it mean to say a particle is "moving" in one direction or another? Such a thing wouldn't make any sense if we were talking about classical particles in an ordinary spacetime diagram in SR. I think it has more to do with some mathematical aspect of how you add different diagrams, maybe vaguely similar to the difference between integrating some function from an earlier time coordinate t0 to a later time coordinate t1 as opposed to integrating -1 times the same function from t1 to t0, with both giving the same answer.

8. Feb 2, 2008

### peter0302

I guess it depends on what the meaning of is is.

"Moving" means having a worldLINE that is more than a single point. "Moving" in time relative to someone means that their worldline is perpendicular to the observer's space axis. Moving "backward" in time means a worldline that bends around such that the object is seen at two points in space at the same time relative to some observer.

This is very clear stuff. Why are you being so pedantic?

9. Feb 2, 2008

### JesseM

What vector are you talking about, exactly? The space-time 4-vector, the energy-momentum 4-vector, or something else?

10. Feb 2, 2008

### JesseM

I thought "moving backwards in time" was supposed to be an idea that applied to antiparticles, whose worldlines certainly don't form closed timelike curves. If you're talking exclusively about closed timelike curves, then this is an entirely separate issue from the stuff about Feynman diagrams and antiparticles, and you should stop confusing the two issues. But it isn't clear to me that this is all that A.T. meant.

11. Feb 2, 2008

### A.T.

What I mean by "move in time" here, is "change in proper time over coordinate time"
I mean a normal clock that ticks forwards in its own frame of reference, but is observed to tick backwards by a certain observer. Think of it, as a negative factor of time dilation.

12. Feb 2, 2008

### JesseM

But it's just a convention that proper time always advances in the same direction as coordinate time. Do you imagine there is some physical, testable difference between an object whose proper time is increasing over coordinate time vs. one whose proper time is decreasing, apart from the arrow of time defined by entropy for macroscopic systems? If so, what is this difference? What type of experiments could I do to determine it for particular objects?
It's also a convention that all frames agree on which direction is increasing time. The Lorentz transformation doesn't allow for different frames that disagree on this. To formalize your idea that an object can have time going in the opposite direction "in its own frame of reference", you'd need a different coordinate transformation to define the "frame of reference" of different objects, and in order for this to have physical meaning you'd have to give an experimental procedure for distinguishing two objects which have identical velocities (i.e. the distance between them is constant) but different frames in your sense (something which doesn't happen in ordinary SR, where objects with the same velocity share a common rest frame).

13. Feb 2, 2008

### A.T.

Yes, and by this convention a clock running back is a different type of clock, than the one running forwards. But you could also agree on a convention saying, it's the same type of clock, just moving back in proper time.

Well, in my first post I suggest there is no physical difference. But I am not sure when I consider this:

Observe a macroscopic system made completely of antimatter, and see if the entropy increases or decreases?

14. Feb 2, 2008

### JesseM

But what do you mean by "running back"? It's trivial to build a clock where the hands are sweeping counterclockwise, or a digital clock where the numbers are decreasing, but presumably this isn't what you mean. Do you mean something harder to fake, like the thermodynamic arrow of time?
Certainly current theories would certainly predict that the second law should apply equally to isolated antimatter systems as it does to isolated matter systems. The second law is just statistical, having to do with the fact that there are vastly more high-entropy states a system can be in than low-entropy states, and this is equally true for antimatter systems.

15. Feb 2, 2008

### phyti

A.T.

"If you observe a clock going backwards, you could say it is "moving back in time"."

You could say "it's running backward".

If something actually moved backward in time, wouldn't it disappear?

If you assume 2 directions in time, and you are observing an object (moving backward),
wouldn't it be moving forward for you to observe it?

16. Feb 3, 2008

### A.T.

I meant a clock, that is observed as running in different directions, by two different observers. And my question was, if such a situation is possible.

That is also my idea. That's why I think, that we will never observe something like teapots coming together from debris, and jumping back on the table.

No, because the velocity of the object is usually defined in regards to coordinate time(observer's clock). I'm talking about moving back in proper time(object's clock). The idea is that the object is mirrored in space, when moving back in proper time. Right-hand-rule, is replaced by left-hand-rule. Electrons of the object would be observed as positrons.

Last edited: Feb 3, 2008
17. Feb 3, 2008

### Staff: Mentor

[unscientific pseudophilosophical musing]

We humans always go backwards in time.

If I say "I am walking backwards" what I mean is that my eyes are looking where I have been and not where I am going. It is not a statement about my direction of travel since there is no meaning to going forwards in space or backwards in space. Instead, it is a statement about my personal and sensory orientation wrt my direction of travel, which does have an unambiguous meaning.

Since our memories are all about the past we are "looking" where we have been in time (the past) and not where we are going (the future). So we go backwards in time, not forwards.

No wonder we have so many problems and keep stumbling.

[/unscientific pseudophilosophical musing]