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

Explanation of why Michelson Morley measurments of speed of light disproved ether

  1. Aug 28, 2012 #1
    I have never understood why this experiment is so generally cited as demonstrating the non-existence of "ether." (To be clear here, I am not suggesting that ether exists, I am just frustrated that I cannot seem to appreciate this landmark experiment.) My understanding of the experiment is that the speed of light was measured repeatedly at various times of the year. Each time, light was measured traveling at perpendicular angels. I understand the concept that something, (say a boat on a river) traveling parallel to a "current" will travel at one speed with the current and another speed against the current, and a third speed when traveling on a perpendicular to the current. As I understand the explanations of the experiment, when its results demonstrated a constant speed of light regardless of direction, this was universally accepted as conclusive proof that ether did not exist.

    I get lost on this point--if ether were thought to permeate space, and light traveled through the ether--then earth would be moving through it constantly at all times,--so why wouldn't the speed of light be a constant passing through it? Whatever affect ether would have on the speed of light, wouldn't it be a constant? The only explanation I can muster is that ether required a "current." I know at one time there was a theory that ether had a "wind" that "blew" through space, and that earth traveled in varying directions relative to this wind, so therefore the measurements of the speed of light were affected by the direction of the "wind" relative to earth. But I believe, that idea was discarded, and nonetheless, the Michelson Morley experiment was accepted as proving that ether, even without a "wind" did not exist.

    I've tried unsuccessfully to understand the experiment's results on my own for quite some time. Over the years I've read explanations of the experiment from Hawking, Greene, Eddington, Russell and others. Clearly, I'm missing something here that is, no doubt, obvious. So if anyone has the patience, I'd appreciate any insight that could be offered.

    With appreciation. Mark Goldman
  2. jcsd
  3. Aug 28, 2012 #2
    Do you know about reference frames?

    The general idea of the ether is that there is some 'thing' that light passes through. This means that there is a specific reference frame of the ether for which the speed of light is a constant in all directions.

    The thing is, according to the Copernican principle, this 'special' reference frame cannot be centered at Earth because that would imply something special about Earth's place in the universe. That means that as the Earth moves around the sun, regardless of what reference frame the ether is actually in, if it is anything but Earth's (and the Copernican Principle says it cannot be Earth's) the relative velocity of Earth with regards to the ether will change over the course of the year.

    This means the measured velocity of light will change over the year. The M-M experiment showed that it did not, and while the result did not single handedly end the ether theory, it was definitely its death knell.
  4. Aug 28, 2012 #3
    I think you've got it when you talk about an ether "wind". Think of the ether as a preferred, "stationary" frame in space. It does not have an inherent "wind", but Earth is accelerating through it, constantly creating an ether "wind" on Earth. I don't believe scientists of the day thought there was anything wrong with that train of thought. Imagine throwing a ball out of the window of a moving car: depending on which way you through the ball, it may travel quickly, or slowly, or travel in a curved path. The idea being that at some times, the beam of light would travel against this wind, with the wind, perpendicular to the wind, etc. And that these interactions with the ether should be measurable. They were not. Does this help?
  5. Aug 29, 2012 #4
    Vorde, Soothsayer,

    While I'd like to think about your comments before replying, I want to quickly say thank you for responding. I appreciate it.

    Mark Goldman
  6. Aug 29, 2012 #5
    That is indeed a fable, likely due to misunderstanding, as you will find if you read old papers and books on SR (to my regret I only found out when doing so!). However, I also would like to know the exact cause of that misunderstanding, as it's certainly instructive.
    Not exactly - you seem to mix up several models. But I think that you point to a plausible explanation. Perhaps there was confusion with the disproof of the Stokes ether, and all models were thrown on a heap?

    PS. As you mentioned Eddington, I now checked his 1920 book Space Time and Gravitation, in which he explained why "the aether is now only a background and not an active participant in the theory."
    Last edited: Aug 29, 2012
  7. Aug 29, 2012 #6
    One clarification on the OP's post. The Michelson Morley inteferometer did not measure the speed of light. It was not intended to do that. It's purpose (as explained by the other posters) was to detect the motion of the earth through the ether.

    Here is a layman's analogy of the experiment I wrote up for myself that might be of interest to others.

    Last edited: Aug 29, 2012
  8. Aug 29, 2012 #7
    Mark, from other posts related to this subject, I think most forum folks here would tell you that the ether has not been disproven by the Michelson-Morley experiment. The Lorentz Ether Theory (LET) is cited often here as a theory that supports the existence of the ether, and most forum members who have commented on the subject maintain that this theory has not been disproven.

    Check out posts by Vandam and you will get a sampling of attitudes strongly critical of Vandam for his assertion that LET is not a viable theory.
    Last edited: Aug 29, 2012
  9. Aug 29, 2012 #8
    I think LET is a viable theory, but I just don't really like the idea. SR seems simpler and more elegant, to me. Are there theoretically any experiments that could confirm one or the other?
  10. Aug 29, 2012 #9
    SR is the same without metaphysical explanation; only experiments that go beyond testing SR could in theory distinguish between different interpretations (and if we can rely on Einstein and Eddington, then GR doesn't change a thing). QM could in theory do so because of Bell's theorem, but in practice that isn't likely to happen because of ambiguities and too many possible interpretations of QM as well.
  11. Aug 29, 2012 #10
    I’m not sure that, at the time of the experiment, the conclusion was that the aether did not exist, only that they had failed to detect it as they had expected to. The basic reasons for believing in it remained. It served two purposes. One as the medium in which the heat and light of the Sun, which had been proven to be a wave form, reached us, and two as the absolute reference frame in Newtonian Mechanics. It was thus only when Einstein had his miracle year in 1905 that things changed. First, his paper on Special Relativity removed the need for it as an absolute reference frame. Then his paper on the photo-electric effect, which demonstrated that light was, after all, a stream of particles, removed the need for it as a medium. Suddenly the aether was redundant and it was realised that the reason that Michelson Morley had failed to detect it is because it doesn’t exist.

    I am sure I remember reading a post on another thread on this forum which discussed this persistence of the idea of an aether. Somebody pointed out that if you say that the aether does exist then you must accept that it has a certain list of properties. You can’t detect it, it isn’t made of any element that we know of, it has no mass etc, such that, ‘it exists and has these properties’ on the one hand, and ‘it doesn’t exist’ on the other hand are actually exactly the same thing.
  12. Aug 29, 2012 #11
    LET and SR are both viable theories, the only difference is what you choose to metaphysically believe:

    SR requires that you take for granted that for seemingly no reason, the universe maintains the consistency of the speed of light without any understanding of why.
    LET requires that you accept the existence of a preferred reference frame wherein dwells an undetectable field that permeates the entire universe.

    Which you choose to believe is up to you, however I believe that what SR requires you to 'deal with' is easier that what LET makes you deal with. Furthermore SR leads so naturally into GR.

    The point as relating to the M-M experiment is that M and M actually set out to try and prove the existence of the ether because they assumed they would detect a change in the speed of light as the Earth went around the sun (as someone else pointed out, they weren't actually measuring the change in the speed of light, but I think this simplification isn't detrimental to the result), and they didn't detect it. LET is a version of the ether theory which is compatible with the M-M experiment, but it seems to me that LET is a force-adapation of old theories to new results, rather than the simplicity and elegance that SR brought as a new theory.
    Last edited: Aug 29, 2012
  13. Aug 30, 2012 #12
    I do not see why one should take LET as an option as long as there is no prove that the ether exists.
    My neighbour can come up with a theory with flying pink elephants that might explain the constancy of the speed of light. Would you then take his point of view also into consideration?
    For the LET believers:
    1/ Prove me that the ether exists. You cannot. To put it bluntly: that's why Einstein is Einstein, and Lorentz the loser.
    2/ Show me how relativity of simultaney makes sense in LET. You cannot. Unless you start with abstract mathematical coordinates that have no physical meaning in the physical ether. To save the appearances you will then make the physical significance a philosophical debate. Great physics.
    LET is a mess.
  14. Aug 30, 2012 #13
    A pink elephant is no help for modeling light propagation - therefore you won't find your neighbour's model in the scientific literature. :smile:
    Please don't start such a discussion. It's irrelevant for the OP's question.
    Moreover, attempts to answer your challenge are not allowed on this forum:

    'Generally, in the science discussion forums we do not allow the following: [..] Attempts to promote or resuscitate theories that have been discredited or superseded (e.g. Lorentz ether theory); this does not exclude discussion of those theories in a purely historical context'
    Last edited: Aug 30, 2012
  15. Aug 30, 2012 #14
    -That's a good one. If you think I want to promote LET. haha...
    -I take note that LET has been dicredited and superseded. Thanks.
    -Unfortunately quite a few guys on Pf do not deal with LET in an historical context. They use it today as it suites them because they consider LET and SR as equally valid. (They even use it to discredit block universe ;)
  16. Aug 30, 2012 #15
    Instead, I warned you and other participants that you were inciting them to break the rules - which is almost as bad as (or even worse than!) breaking those rules yourself.
    A pink elephant can be a valid counter argument against a purple bear without need to promote it - if you don't understand that, please discuss it not here but in one of the threads about the block universe.
    Last edited: Aug 30, 2012
  17. Aug 30, 2012 #16
    It doesn't. The Michelson-Morley experiment measures the ether wind, relative to Earth. Actually, relative to the very spot where the experiment is performed. The ether wind was found to be always zero.
    From this you can infer:
    1) There is no ether.
    2) Ether exists, but the Earth has always zero velocity with respect to it. This is hard to believe, once you know that the Earth is not a special place in the universe – it's not fixed at its centre, and actually moves around a lot.

    Choosing 1) yields a simpler, cleaner description of the observed facts. But the experiment doesn't (and cannot) prove that ether doesn't exist. There were a host of alternative theories at the time that explained why no ether wind was to be found on Earth, and there was a long trail of experimentation going on to test them. In today's narration of the progress of science Michelson and Morley's experiment is pictured as a landmark event, but this is oversimplifying things a bit. A new theoretical framework is rarely (if ever) the result of one single experiment.
  18. Aug 30, 2012 #17
    You forgot the main inference that followed (Lorentz contraction), and which led the way to SR. It's all rather well explained in papers and books by Lorentz, Einstein, Eddington etc.
    And note that possibility 2 was already held to be disproved by then (also done by Michelson): http://en.wikisource.org/wiki/Influence_of_Motion_of_the_Medium_on_the_Velocity_of_Light
  19. Aug 30, 2012 #18
  20. Aug 30, 2012 #19
    I believe there was a theory that some of the ether was dragged along by the earth, making the ether "wind" stationary locally, but not away from the Earth, the idea that is we could put this experiment in orbit, it would find the effect that the experiment on the Earth could not. But that idea was as representative as a death knell for the ether theory than the failed experiment.
  21. Aug 30, 2012 #20
    That might be reasonable, if the ether-drag theory had been proposed after MM in order to explain it, but it was in fact proposed by George Stokes in 1845, as an alternative to the partial ether drag hypothesis of Fresnel. Fresnel realized in 1818 that some of the ether must be dragged along by matter in order to explain stellar aberration (otherwise the refractive index of the telescope lens should vary along with its velocity through the ether). The problem with that theory being that partial dragging would itself result in a chromatic aberration that was never seen. So Stokes proposed that ALL of the ether was dragged by a ponderable body, but that the velocities of the ponderable body and the ether occupying the same space were different.

    The MM experiment was meant to test the amount of ether wind in the laboratory, which would be substantial under Fresnel's theory but could be essentially zero in Stokes's theory as the Earth dragged all of the nearby ether with it. So when it was performed, the interpretation of the results was unclear, because: (a) contrary to what you read in the textbooks, MM did NOT make the test at different times of year and day (the experiment took place over the course of just a few days in July 1887), (b) the amount of data collected was actually quite small, consisting of only about 25 turns of the instrument, and (c) there were so many ether theories around (no such thing as THE ether theory) that the result seemed to confirm some (such as Stokes'), while falsifying others (such as Fresnel's).

    Obviously the experiment needed repetition, but that didn't happen for over a decade, because Michaelson's expertise at building interferometers was essentially unique, and he decided to drop the experiment after the first week. So it should not be surprising that the immediate reaction at the time of publication was lukewarm and basically a scientific shrug.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook