# Relativity of simultaneity objections

1. Jun 30, 2013

### mirrormirror

Hi, I'm not a physicist, I'm 32 years old and I like physics. For the past years I've been troubled by an issue: the known thought experiment of Einstein with the train and the relativity of simultaneity notion.

Now, most probably I'm wrong or I'm missing something but I think that there is a mistake with the train thought experiment. The train experiment has two equivalent versions:

a) two lightning strikes hit the traincar both in the end-door and the front-door and the light goes towards the observer in the middle of the traincar. in this version, the event is simultaneous for the observer at the platform but not simultaneous for the observer in the traincar because the train is moving towards the light coming from the front of the wagon, thus it will reach the on-train observer faster.

b) the observer in the middle of the traincar emits light flashes going to both directions. in this version, the event ( of light reaching the doors ) is simultaneous for the train observer and not simultaneous forthe observer at the platform.

To make things simple let's take version a) of the experiment.

I hereby claim ( i most probably am wrong somewhere ) that the measurement method of whether the events are simultaneous for the TRAINCAR observer is WRONG. Yes, the light beam coming from the front will reach him faster than the beam that comes from the back of the train, but he should NOT DEDUCE that the events are NOT simultaneous. The CORRECT way to measure the simultaneity would be to install TWO SYNCHRONISED clocks at both doors ( front and back ). That way when he checks the clocks he will notice that the lightnings hit both clocks at the same time.

This is also a way for someone inside an inertial frame of reference ( a cubic black room with no windows, moving at a constant speed ) to measure whether or not his inertial frame of reference is moving or not: Just install 6 SYNCHRONISED clocks ( it's doable with 2 as well ) at each wall of the room: left, right, back, forth, up, down and then emit with a device a laser beam simultaneously to all 6 ( or two ) directions. Since the big-black-room-with-no-windows is moving with a constant velocity, there will be differences at the times measured at the clocks on each wall, thus he can find out the velocity of the room ( frame of reference ) in each axis.

What do you say ?

Andreas T.

Last edited: Jun 30, 2013
2. Jun 30, 2013

### Staff: Mentor

And how do you synchronize the clocks?

3. Jun 30, 2013

### mirrormirror

I don't know, it's a technical issue I guess. Who cares? I think we already DO synchronise clocks in experiments, when we leave a clock on earth and the other we put it on the spaceship. This is a thought experiment, I guess we shouldn't care about how we synchronise the clocks.

4. Jun 30, 2013

### Staff: Mentor

Since your approach relies strongly on synchronized clocks this is an issue that you cannot gloss over.

Einstein provided a method for synchronizing clocks, known as the Einstein synchronization convention. Using that synchronization method, a pair of clocks which are synchronized in one reference frame will not be synchronized in another. So using your "CORRECT" way to measure simultaneity still results in the same situation that the train and the embankment frames disagree regarding the simultaneity of the strikes.

5. Jun 30, 2013

### mirrormirror

I don't know what was Einstein's method of synchronising clocks. Why would they not be synchronised outside that frame of reference?

But yes, let's say that the clocks were synchronised for the in-train passenger he would find too that the events were not simultaneous. The out of train observer would NOT use the two clocks for the measurement, he would use his one clock, because he doesn't need two clocks to measure in a CORRECT way the simultaneity. So the change of method of measurement concerns only the on-wagon observer. The current out-of-train observer had initially a CORRECT method of measuring. So it doesn't matter if the clocks are not synchronised in the external's observer frame, because he doesn't use them to measure the simultaneity

Last edited: Jun 30, 2013
6. Jun 30, 2013

### Staff: Mentor

You have to be careful to specify how the clocks were synchronized. If you just say "two clocks that read the same at the same time, you're assuming what you're trying ton prove, namely that there is absolute simultaneity.

But with that said...

Yes, you are describing a correct way of establishing simultaneity. But... If the two clocks are synchronized and equidistant from the observer and both read the same time when the lightning strikes, then the flashes will both reach the observer at the same time. It has to be that way, because otherwise the observer, by doing a straightforward "speed equals distance divided by time" calculation, would find that the two flashes weren't both traveling at the speed of light - and we know from experiment that the speed of light is always c regardless of the motion of the source and destination.

Furthermore, the same has to be true for all observers: the one on the platform, or one in a plane flying at 400 kilometers/hr overhead, or an observer watching through a telescope from Mars which is moving at several miles a second relative to earth. (The Martian observer is important, because he can clearly see that the platform observer is no more "really at rest" than the train observer).

So now let's look at exactly how the two observers synchronized their clocks. Train observer, standing in the middle of the train, took two identically constructed clocks, set them both to read the same time (noon, let's say), and then instructed his two helpers to each carry a clock to each end of the train and leave it there.
Ground observer did the same thing with his two clocks.

Now try the thought experiment again, see what you see.

7. Jun 30, 2013

### WannabeNewton

You must choose a consistent scheme of simultaneity and actually define what that simultaneity scheme is. This is what DaleSpam is saying.

8. Jun 30, 2013

### mirrormirror

The clocks will stop at the moment that the lightnings hit them. Let's say they both stop at 12:05:05. So the on-train observer goes to one end and checks one clock, then at the other end and checks the other other clock. Both clocks have stopped at the same time, thus it was simultaneous. It doesn't matter WHEN the flashes of the lightning reach the observer in the middle, YES they won't reach him simultaneously, but that DOESN'T mean that the speed of light is different! Neither does it mean that the events were not simultaneous. It only means that the train is MOVING to the direction of the light beam that reached him earlier.

The nature of the experiment does not require the ground observer to use two clocks. He can just use his one clock with which he already found the events were simultaneous. Why he doesn't need to use two clocks? Because the light travelling from the doors towards the ground observer do not move in the direction of the train but vertical to it and he is not moving either.

9. Jun 30, 2013

### Staff: Mentor

On the contrary, it's huge issue. If you synchronize the clocks in a way that assumes that there is no relativity of simultaneity, then of course you can conclude that there is no relativity of simultaneity - but your argument is circular so proves nothing.

Here you are assuming that if you synchronize two clocks (put them in the same place at rest relative to one another and set them to the same time) and then set them in motion relative to another they will remain synchronized. That's a very plausible and intuitive sort of assumption, one that everyone pretty much took for granted in the pre-relativity days. It's what behind your unspoken assumption that not only are the train clocks synchronized with each other (OK) and the ground clocks synchronized with each other (also OK) but that the train clocks and ground clocks are synchronized with each other (not OK).

But it turns out that if you make this assumption, as Einstein's train experiment showed, you can't also have the speed of light equal to c for all observers. And it is, so the assumption can't be valid.

10. Jun 30, 2013

### mirrormirror

Hm, why this scheme of simultaneity is not consistent? the on-train observer needs two clocks to measure it correctly and he cannot depend on the light beams because they travel along the axis of train, thus the speed of the train will interfere with the measuring. On the other hand the ground observer doesn't need two clocks and CAN use the light beams because he is not moving and the beams are moving vertically to the axis of the train movement.

11. Jun 30, 2013

### mirrormirror

i never said that the ground clock has to be synchronised with the train clocks. it doesn't need to be. The on-train observer needs two clocks to measure it correctly and he cannot depend on the light beams because they travel along the axis of train, thus the speed of the train will interfere with the measuring. On the other hand the ground observer doesn't need two clocks and CAN use the light beams because he is not moving and the beams are moving vertically to the axis of the train movement.

There is no requirement for the ground clock to be synchronised to anything. For the ground clock, the two beams will reach it simultaneously, thus the events were simultaneous.

Last edited: Jun 30, 2013
12. Jun 30, 2013

### Staff: Mentor

You wrote: I hereby claim ( i most probably am wrong somewhere ) that the measurement method of whether the events are simultaneous for the TRAINCAR observer is WRONG. Yes, the light beam coming from the front will reach him faster than the beam that comes from the back of the train, but he should NOT DEDUCE that the events are NOT simultaneous. The CORRECT way to measure the simultaneity would be to install TWO SYNCHRONISED clocks at both doors ( front and back ). That way when he checks the clocks he will notice that the lightnings hit both clocks at the same time.

If the clocks had been synchronized on the train, then when he checks the clocks, he will find that the lightning strikes did not hit both clocks at the same time. The strikes are observed to be simultaneous only by observers on the ground (using their own set of synchronized clocks).

13. Jun 30, 2013

### mirrormirror

sorry, but why will he notice that the clocks did not stop at the same time? The clocks are AT the doors. One clock in the back door one in the front. Once they get hit by the strikes they stop.

Last edited: Jun 30, 2013
14. Jun 30, 2013

### Staff: Mentor

The bolded bit is a misunderstandng. If the two train clocks were synchronized and placed as I describe, and if they are read the same time when they were stopped, then the two flashes will reach the train observer at the same time. That time won't be 12:05:05, it'll be just a bit later to allow for light travel time (12:05:05+D/c where D is one-half the length of the train) but it will be the same time.

But the train observer says that the train isn't moving - as far he's concerned, he and the train are at rest while the platform is moving backwards. Now if we could do the experiment with the stopped clocks as you describe above and the light from the two flashes did not reach him simultaneously any time that the clocks were equidistant and read the same when they were stopped... Then we'd be able look for the one observer for whom the flashes were simultaneous, say that one is the one that is "really" at rest and the rest are "really" moving. But that's not how the world wors.

15. Jun 30, 2013

### Janus

Staff Emeritus
But the whole point is that the "speed" of the train does not interfere with the measurement of the light for the train observer. You seem to be holding on to an idea that somehow you can say that it is the train that is "really" moving. This isn't the case.

For anyone in the Train, the speed of light is a constant, just like it is for someone on the embankment. What this means is that he will measure the speed of the light coming from the rear of the Train as traveling at the same speed as the light coming from the front as measured relative to himself. If he is at the midpoint of the train and the light from the two ends reach him at different times, he knows that the light left the ends at different times, since light takes equal times to cross equal distances. Just like the embankment observer knows that the flashes occurred at the same time because he sees the lights at the same time.

16. Jun 30, 2013

### mirrormirror

The two train clocks ( one in the back, one in the front ) will stop BOTH at 12:05:05 when the lightning hits them. The light of the strike, moves from the clocks towards the observer in the middle, the beam from the front will reach t1 after the clocks stopped, the beam from the beam from the end will reach him t2 after the clocks have stopped. t1 != t2 because the train is moving*. Anyway this is of no importance to simultaneity because what matters for SIMULTANEITY is WHEN THE CLOCKS STOP. He can check the clocks whenever he wants and when he does he will see they both stopped at the same time.

* That's what I challenged too, the idea that an observer of an inertial frame of reference can't tell if his frame of reference is moving. I'm saying that he CAN.

So what I'm saying is:

Exactly because of the FACT that the speed of light is CONSTANT and does not depend on the speed of the source that produces it, it is possible to KNOW whether or not your inertial frame of reference is moving or not.

So the idea that the speed of light does not depend on the source that produces is mutually exclusive with the idea that inside an inertial frame of reference it's impossible to tell whether it's moving or not. I just described in my first post a way to measure whether an inertial frame of reference is moving. quoting it again:

Last edited: Jun 30, 2013
17. Jun 30, 2013

### mirrormirror

yes, exactly that. I'm saying that you CAN tell if the train is really moving. here's how:

18. Jun 30, 2013

### Staff: Mentor

Here are a couple of references to explain it:
http://www.fourmilab.ch/etexts/einstein/specrel/www/
http://en.wikipedia.org/wiki/Einstein_synchronisation

This isnt a legitimate way to avoid the issue To determine simultaneity of a pair of events always requires a clock of each event, and the clocks must be synchronized according to some convention.

19. Jun 30, 2013

### mirrormirror

i don't see why clocks at different frames of reference have to be synchronized in order to measure SIMULTANEITY. Yes the clocks on the trains need to be synchronized among them, the same goes for the clocks out of the train ( on the platform ). But why do the clocks on the platform need to be synchronized with the clocks on the train ? We are not measuring WHEN an event happened, we are measuring if a couple of events happened simultaneously and for that it's enough that ALL the clocks on the SAME reference frame be synchronized.

Plus, what's your opinion about my example of the big black room with the six clocks in it? Can you or can you not measure whether the room is moving ?

20. Jun 30, 2013

### Staff: Mentor

Of course you cannot. Nothing you can do within the room can tell you whether the room is moving or not.

21. Jun 30, 2013

### mirrormirror

do you care to provide some evidence or counter-arguments to my example?

22. Jun 30, 2013

### Staff: Mentor

The reason you can't measure the motion of the closed room using this technique is that no matter what the constant motion of the room is, the time it takes, as measured by synchronized clocks moving with the room, for a light beam to move from one side of the room to the other is the width of the room divided by c.

This has been experimentally confirmed by various forms of the Michelson-Morley experiment (and see also the FAQ at top of this forum on experimental support for relativity). This is the link to the most directly relevant part: http://www.edu-observatory.org/phys...iments.html#Tests_of_Einsteins_two_postulates

It's also a somewhat intuitive result when you consider that over a period of six months the earth changes from moving at some miles a second relative to the sun in one direction to moving at the same speed in the exact opposite direction - yet we do not expect the travel time for speed of light signals between various points on earth to change with the seasons.

23. Jun 30, 2013

### Staff: Mentor

An equivalent experiment to this one has been done; it's called the Michelson-Morley experiment. As others have already pointed out, it did *not* give the result you predict in what I quoted above. Given the results of the M-M experiment, the result of the experiment you give above is that there will be *no* differences in the times measured in the clocks on each wall, regardless of the state of motion of the room: light emitted from the exact center of the room will reach clocks on each wall at exactly the same time. (The M-M experiment, in your setup, would involve putting a mirror on each wall instead of a clock, to reflect the light beams back to the center, and then measuring whether the reflected beams all arrive back at the center at the same time--and they do.)

So your whole mental model is based on an incorrect prediction about what experiments will show.

24. Jun 30, 2013

### mirrormirror

The bold part is EXACTLY what's WRONG with the experiment of M-M if they did it this way with mirrors. Of course if you place mirrors, the light will take t1 seconds to go to the front wall, then t2 ( less than t1 ) to move back to the center. For the back wall it will take t2 seconds to go from center to back wall, then t1 seconds from the back wall to the center. So yes, t1+t2 = t2+t1 but the fundamental ERROR is that you have to put CLOCKS at each wall, NOT measure the roundtrip time, the roundtrip time will be equal...

25. Jun 30, 2013

### Staff: Mentor

One of the key aspects of the MM experiment is that they used two light beams sent at right angles to one another. Both light beams were reflected back by mirrors set at the same distance away from the source; and the measurement was whether if they left at the same time they both arrived back at the same time. Because the two light beams are traveling at right angles to one another, they cannot both be equally affected by the t2<t1 effect that you're expecting. (Actually they could, if the apparatus happens to be set up so that both light beams make a 45-degree angle with respect to the direction of travel - but it's easy to eliminate this possibility by rotating the entire apparatus to point in some different direction and then repeating the experiment).

The reason for doing the experiment this way is that we can measure differences in the arrival times of two light beams with far greater accuracy than we can measure the arrival time of a light beam at a clock.

(Edit: I should add that nonetheless the experiment HAS since been done with a single clock at the destination, using instruments not available to Michelson and Morley as advancing technology made it possible to build these instruments)

You'll find some more detailed explanations at the link that I posted, and by googling around.

Last edited: Jun 30, 2013