Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Can we predict which reference frame is correct?

  1. May 10, 2009 #1


    User Avatar

    I was thinking about something earlier today. One of Einstein's famous phrases is "I can not believe God plays dice with the Universe." After pondering on this for a while, something confused me. Say for example we have a group of observers who all observed the Simultaniety of an event differently. They then passed their findings over to a person who did not observe the event and ask him to determine which chain of events is correct. Can we do this?

    According to some, each and every event that this guy gets given to anylise are correct, none of them are wrong. Couldn't we then assume that it's down to the probability of which event actually occured in truth? I have a feeling that with the right maths, we would be able to determine these things without actually witnessing them, but I'm not sure.
  2. jcsd
  3. May 10, 2009 #2

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    None of them are wrong. You are assuming there exists one true answer. There isn't.
  4. May 10, 2009 #3
    ^^^That sure makes it easy to predict if you reference frame is right :wink:^^^
  5. May 10, 2009 #4


    User Avatar
    Science Advisor
    Gold Member

    Asking which of two (causally disconnected) events occurred first is a bit like asking which of two points, in 3D space, is furthest to the left. Two observers looking from different angles could completely disagree. It's not a matter of probability who is correct. There just is no uniquely correct answer, because "left" is a relative concept.
  6. May 10, 2009 #5


    User Avatar
    Science Advisor

    In relativity an "event" is something that occurs at a localized point in space and time, like one object colliding with another, or two clocks comparing readings at the moment they pass right next to one another. All frames make the same predictions about local events like these. The simultaneity of events at different points in spacetime isn't something objective, it just depends on your choice of coordinate system which assigns position and time coordinates to different local events.
  7. May 10, 2009 #6
    http://en.wikipedia.org/wiki/Rietdijk-Putnam_Argument" [Broken]

    Last edited by a moderator: May 4, 2017
  8. May 13, 2009 #7
    Suppose there's a civilisation that measures distances not in meters, or feet, or any other standardised measure, but in "sticks". There's no standard stick length, everybody just uses his own.

    Now suppose you ask two of these people how far away a tree is. Both will take out their sticks and start measuring. One says it's 200 sticks away, the other says 230. Who is correct? Both are, since there's no such thing as a standard stick and everybody is free to use his own.

    A similar situation arises in special relativity. Moving observers see things happening in a different order, at different distances, and at different times. The problem is there's no objective way of saying who is moving and who is standing still. For example, if a high speed train is moving at 1667 km/h through a train station at the equator, towards the west, the people in the train station will say that the train is "obviously" racing through their station while the station is not moving, but viewed from space, the train is standing still while the earth is rotating underneath. So is the station moving, or the train? Can we ask an independent observer to determine which is "true"? You might be tempted to say the view from space is more true, until you take into account the fact that the earth is also rotating around the sun, etc... So what is the "true" speed of the train?

    Unless you designate a particular reference system as "true" (by definition, in a law accepted by vote or something like that), you cannot say who is right about things like simultaneity, speeds, times, distances, etc...

    In general relativity, you can even use reference systems that do not correspond to any observer's point of view. The cosmological reference system that assigns local time to all places in the expanding universe, for example, is one that cannot be experienced by any actual observer. Yet when you hear cosmologists talk about distances and times in the universe, they usually mean measures in precisely that reference frame.

    There really is no way of saying whether or not two things were simultaneous unless you specify a specific coordinate system, just like there's no way of saying how far away something "really" is if you don't say what kind of stick you should use.
  9. May 13, 2009 #8
    All observers in SR use the same standard of length. Call it a meter. Observers at rest in frames moving relative to each other measure the others meters as less than their own. But if they come together into a single frame and compare their meters they will then be the same. This is not the same as each having different length sticks as their own standard.

  10. May 13, 2009 #9
    But if you were watching from the sidelines as two observers were measuring things at different speeds, you would say they have different length sticks and their watches are ticking at different rates. Of course everything smooths out once they decide to move at the same speed, but otherwise their sticks will appear to be different, and they will measure things differently.
  11. May 13, 2009 #10


    User Avatar

    Thanks for the input...

    So, lets take the moving train and two lightning strikes as an example. I'm going to take this in steps as there's a few things I want to get at.

    We have an electric detection device attatched to each of the lightning rods. The devices will tell us that the two lightning strikes struck at precicely the same time so we know what actually happened.

    We now have the two observers (A and B) who observe the strikes differently, A observes them simultanious, B observes one strike after the other. Since we have the devices to detect the lightning strikes, isn't it safe to assume that observer A saw what actually happened? Where as observer B didn't because of his motion relative to observer A?

    Now, lets say we add a single laser beam to both detection devices at precicely the center of inbetween the rods so that the information can be transmitted to observer B. How would the information differ to observer B's visual observation of the lightning strikes? Would it tell him that the lightning strikes struck at precicely the same way as he observed it on the moving train? One after the other? Or would it give him the results of what happened for observer A?


    So now lets assume that there is no true answer to what actually happened, wouldn't that mean that God does in fact play dice with S-R? No observation is correct and no observation is wrong, to someone trying to figure out what happened, it's down to the probability of what actually occured?
  12. May 13, 2009 #11

    ---There's no standard stick length, everybody just uses his own.----

    There is a standard meter for all observers. In your analogy you said each could choose his own stick. He cannot.

  13. May 13, 2009 #12


    User Avatar
    Science Advisor

    You have still not understood what everyone has been telling you. You have been consistently talking about what people observed not about "what actually happened".

    There is no such thing as an absolute time so there is no such thing as "when it happened". When something happened depends upon the frame of reference from which it is observed.

    Oh, and Einstein's remark about God not "playing dice with the universe" was not about relativity. He was talking about quantum physics.
  14. May 13, 2009 #13
    First of all you have to ask yourself the question how this electric detection device would determine whether or not the lightning strikes were simultaneous. Without using clocks, because two observers can't agree on whether particular clocks are fast or slow.

    One possible way would be to place a detector exactly in the middle between the poles, and observe the light from the strikes (or use an electric signal from detectors on the poles, but that signal will travel at the speed of light too, so that's the same thing). Problem: that just means that the strikes are simultaneous as seen by the detector, which is using A's reference frame.

    Suppose that the nose of the train is just passing the first pole while lightning strikes it, and another lightning strike also hits the nose of the train at precisely that time. Since it's the same location, everyone will agree on this simultaneity. Now at the back end the same happens, a stroke on the tail at precisely the same moment as the stroke on the nearby pole. Four strokes of lightning, which are all simultaneous as seen by A's detector. B only thinks the two strikes near the nose were simultaneous, and the two strikes near the tail as well, but not all four.

    Now imagine the train is also carrying a detector, in the middle. Since, from A's point of view, the detector is in the middle at the time of the strikes but then moves towards the first pole, it will clearly detect the forward stroke first. And indeed it does.

    Now just look at the lightning strikes on the nose and tail of the train, as seen from the train. If they really were simultaneous, the detector in the middle should surely see them at the same time?

    It does not really matter whether the lightning strikes hit the train or the poles, it just makes it easier to forget about the ground and just picture the train with the two lightning bolts hitting its front and back end.

    Now maybe you are saying "but that detector is moving, so it doesn't count"!

    Well, imagine the train is moving over the equator in a westbound direction at 1667 km/h. While the earth is rotating towards the east, the detector on the train is standing still. Everything is still the same, A says the nose was hit at the same time as the tail. But he was moving towards the tail of the train, so that doesn't count! If the strikes really were simultaneous, and A was moving towards the tail, he should see the rear strike first!

    You see how, just by looking at the very same situation from two different points of view, you arrive at opposite conclusions. There's now objective way of telling who is right.
  15. May 13, 2009 #14
    It's an analogy, it's not perfect! Somebody watching both observers will see them using different sticks. If one or both observers accelerate so they have the same speed, the sticks will suddenly be the same again.

    Of course the sticks are the same. They just don't look/act the same when in motion.
  16. May 13, 2009 #15


    User Avatar
    Gold Member

    No, not in SR.

    My opinion is not authoritative (others may correct if I go astray) but I think it may help you, just because I started with the same concerns.

    Your problem is with the word "happen". In SR all reference frames agree on which events "happen" and which do not. It is “only” that they assign different coordinates of time and distance/length to them. But when it comes to solving practical problems, each frame makes use of its own measurements and arrrives at the same conclusion as all other frames as to what happens and what does not. Thus the discrepancy about simultaneity does not lead to discrepancy about events or happenings. We could say that the disagreement is confined to a section of the thought process but it does not project over the outcome, in terms of events. Different observers (i.e., frames moving with regard to each other) follow different pointers and paths but they all reach the same target, if you understand by target the ultimate solution to a practical problem.

    Yes, the divergent simultaneity judgments of the observers are all true and correct. But you must understand in what sense. This is a sort of “instrumental” truth: in the end all observers combine their discrepant values in the appropriate formulas and reach the same conclusions about what happens. Thus SR, yes, is deterministic. In its conceptual framework causality is strictly respected and God does not play with dice.

    One could elaborate on this with the Andromeda “paradox” mentioned above, by playing with the concepts of causality, timelike and lightlike versus spacelike events, impossibility of faster-than-light motion and so on. A good numeric example should show that in the end the paradox is resolved, there is no enigma. But for sure others can do it faster and better than me.

    Yes, that is an important point. If transformation between frames is possible (if you can take the measurements of one frame and figure out, through the right formulas, the measurements that should have been obtained in your own frame), that is because, despite the discrepancies, there are also common elements. Mathematical transformation and geometrical rotation between frames can only be feasible if there is a basic homogeneity in the physical process of measurement, on the basis of which the measured values (with which you feed the equations and drawings) are obtained. But here I may be too imprecise. Perhaps someone could elaborate more technically.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook