# Causality in relativity

1. Mar 5, 2015

### palmer eldtrich

I have heard some philosophers state that cause and effect can be simultaneous with each other.
Kant for example apparently claimed that when a bowling ball rests on a pillow creating a depression that the cause and the effect are simultaneous.
It seems to me this is forbidden in relativity. All causal relations must be transmitted at a speed no greater than c hence whilst a cause and effect might seem simultaneous that is just an illusion. Am I right?

2. Mar 5, 2015

### Khashishi

Changes in the state will be transmitted at no greater than c. If nothing is changing, and you are talking about some static equilibrium, there is no cause and effect to talk about.

Drop the bowling ball on the pillow, and the depression will form at a speed less than c.

3. Mar 5, 2015

### palmer eldtrich

Thanks for the reply. So you agree simultaneous causality is forbidden in relativity ? If it were allowed would it possible to send a signal back in time ?

4. Mar 5, 2015

### PeroK

Then you could ask Emmanuel Kant directly!

5. Mar 5, 2015

### Ibix

My understanding is that philosophers (in this context at least) aren't using "cause" in quite the same way as physicists do when talking about general relativity. Philosophy is off-topic here, and I'm hardly what you would call an expert, so I'll leave it at that.

As to what the consequences would be if simultaneous causation were possible, there isn't a way to reply to that. You aren't proposing a small change to relativity; you are throwing the whole thing out and putting an unspecified something else in its place. The properties of an unspecified something could be anything.

6. Mar 5, 2015

### Matterwave

I'd like to add though that two events A and B which are null separated can, with an appropriate Lorentz transformation, be made to occur arbitrarily close to simultaneously with each other. If, for example, A occurred 10 seconds before B in one frame, there are frames of reference in which A occurred only .001s before B or .000001s before B. They can not be made to actually be simultaneous with each other; however, nor can you reverse the order in which they occurred. This is how causality is preserved in SR. Since space-like separated points can be made, with an appropriate Lorentz transformation, to occur in either temporal order, they can not have a cause and effect relationship in SR or else there would be some frames in which an effect came before the cause.

7. Mar 6, 2015

### nitsuj

Implicit in "causality" is spacetime inbetween "happenings" until they happen. Your questions spoils causality defined.

Matterwave said it well with "causality is preserved in SR."

I deleted the second question because it's very confused.

8. Mar 6, 2015

### wabbit

Also, remember Kant was writing more than a century before Einstein, at a time when Newtonian ideas of absolute simultaneity and absolute spacetime where the gold standard in physics, so you can't really expect him to give you insights into relativity. Some earlier philosophers would actually be much more relevant but as Ibix said that's off-topic here.

9. Mar 6, 2015

### palmer eldtrich

I understand that but I have seen philosophers use this argument even today.

10. Mar 6, 2015

### wabbit

Ah I wish we could discuss that, but unfortunately if there was a philosophy section in the forum, it would probably swiftly descend into chaos so we have to do without : )

11. Mar 6, 2015

### Ibix

As I noted earlier, those philosophers are using a different definition of "cause" to the one being used by the people responding to this thread. By the relativistic definition of causality, Kant's argument is wrong. But Kant wasn't using that definition (he couldn't - as wabbit noted, he predated Einstein by a way). Why people still buy into Kant's definition of "cause" is not a topic that we can address here - you'll have to ask in a philosophy forum.

12. Mar 7, 2015

### Staff: Mentor

Given the philosophical focus of this topic, the thread is in danger of being closed. Let's stick with physics here. There are good physics concepts that can be discussed here. If we can't stick to those then I will have to close the thread.

13. Mar 7, 2015

### wabbit

Actually, there is a sense in which "simultaneous causality" is possible in relativity : remember that a photon "doesn't experience time" in the sense that it's proper time interval is zero - say, from emission to absorption, the two events being clearly causally related.
So you could say that emission and and absorption are simultaneous "from the point of view of the photon" as far as that makes sense.

14. Mar 7, 2015

### Staff: Mentor

No, this is not correct. What is correct is that the concept of "proper time" does not make sense for a photon. It only makes sense for objects that travel on timelike worldlines. A photon's worldline is null, not timelike.

Which is to say, not at all. The concept of "simultaneous" only makes sense for events which are spacelike separated. The emission and absorption events for a photon are null separated.

15. Mar 7, 2015

### wabbit

I see. The mathematical definition of proper time as the integral of the line element along the worldline does give a well-defined result of zero though, so I can't say I understand why it is incorrect to talk of "zero proper time along a null worldline".

Of course it is incorrect is to ascribe a "viewpoint" to the photon as this would be meaningful only if there was an inertial frame to represent that "viewpoint" - hence my "as far as that makes sense" - I would have said "not much" but I won't argue with "not at all".

16. Mar 7, 2015

### Staff: Mentor

That's part of the definition, but not all of it. The other part is that proper time must be an affine parameter along the worldline. Basically, that means each point on the worldline must be labeled with a different value of proper time. (There are also other technical conditions that aren't relevant here.) Arc length (the integral of the line element along the worldline) has this property for timelike worldlines, but not for null worldlines. So a null worldline can be said to have zero arc length, but not zero proper time, because arc length doesn't uniquely label points on the null worldline.

17. Mar 8, 2015

### palmer eldtrich

Can you explain the difference between different notions of time, proper time, conformal time etc?

18. Mar 8, 2015

### Staff: Mentor

Physically, proper time is the time that is recorded on an (ideal) clock that follows a particular timelike worldline. Geometrically, it is just arc length along that worldline. It is invariant, i.e., it does not depend on the choice of coordinates.

Conformal time is a particular type of coordinate time. Coordinate time in general is just the value of the "time" coordinate in a particular coordinate chart. (Some charts actually do not have a "time" coordinate; we're restricting discussion here to those that do.) Ob viously, coordinate time depends on the choice of coordinates.

19. Mar 9, 2015

### doaaron

I have read before that gravity must be propagating many orders of magnitude faster than EM waves (nearly instantaneously). The argument was that if it didn't, then our solar system would not work. I don't know much about it but I think there are some experts on this forum who could explain it. Also, there are numerous posts on this forum related to the "speed of gravity".

Aaron

20. Mar 9, 2015

### Staff: Mentor

For the most part, gravity doesn't propagate in our solar system, it is essentially a static field. For the minute part of gravity that does propagate our solar system would work fine with it propagating at c.

Please provide the reference. Either the reference is wrong or you misunderstood their point.

21. Mar 9, 2015

### doaaron

Perhaps......
Or their point isn't considered mainstream but is in fact right...in which case I wouldn't be allowed to post it anyway.

If I find the link I will consider posting it.

Aaron

edit: I found the link. Not being a physicist, I only really understood the gist of it, and I don't know how mainstream or not it is considered to be.

Last edited by a moderator: Mar 9, 2015
22. Mar 9, 2015

### Staff: Mentor

Hi doaaron,

You understood the link correctly, the link itself is flat out wrong. This is van Flandern who is a well-known crackpot, essentially anything you read from him is going to be wrong.

In this case, his basic mistake is in equating the observed lack of aberration with a very high propagation speed. The (obvious) other reason for there to be no aberration is for there to not be anything propagating. A static field does not propagate, by definition, so the lack of aberration for a static field gives no information at all about what speed the field would propagate at if it were propagating.

Last edited: Mar 9, 2015
23. Mar 9, 2015

### doaaron

I see...I hadn't heard of him until today, but either way, that sounds a bit harsh considering the paper showed an obviously high level of intelligence. Perhaps his theories are just not scientifically rigorous?

Anyway thanks for the correction.

Aaron

24. Mar 9, 2015

### doaaron

This is something I'm having a lot of trouble understanding actually. Its not really what this thread is about, so I may post a new thread once I can gather my thoughts...

Aaron

25. Mar 9, 2015

### Staff: Mentor

Perhaps it is a bit harsh. His Wikipedia page uses less harsh terms like "non-mainstream", "fringe-science", and "pseudo-science". If you prefer those terms over "crackpot", that is fine by me, but he is the first "pseudo-scientist" that I ever encountered and so to me he will always be the "standard crackpot" against which all others are measured.

Sounds good.

Last edited: Mar 9, 2015