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

M/M experiment is not explained by Special Rel.

  1. Nov 26, 2008 #1
    Is it not true that without "ether" as the medium for light waves, the Michelson Morley experiment makes sense without further explanation? That is, in an inertial frame, the split light beam described in M/M should reconverge and return to the detector without any interference. No ether, no interference to be expected, correct? Isn't it therefore incorrect to say, as textbooks do, that Einstein's theory of special relativity explains the M/M result of no interference?
    Likewise, given that the M/M aparatus is a single inertial frame, and relativity speaks to different moving frames, is it not true that special relativity does not apply to M/M? The light source and detector and mirrors are all in one frame. Special relativity is about different moving frames of reference, and does not speak to M/M in which there there is only one frame of reference. So for this second reason is it not incorrect for textbooks to say M/M is explained by special relativity?
    (There is a version of M/M using starlight as the light source. There special relativity would apply. The textbooks are not talking about that version.)

  2. jcsd
  3. Nov 26, 2008 #2


    Staff: Mentor

    Hi ericqb, welcome to PF!

    I don't understand your question. Einstein's theory of special relativity is the theory that got rid of the aether. What other theory of the time had no aether?

    The textbooks are correct, SR explains the null result of the MM experiment. There are, of course, other theories that also explain the null result of the MM experiment, such as Lorentz's ether theory. Lorentz's ether theory is mathematically equivalent to SR, just conceptually messier.
  4. Nov 26, 2008 #3


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Is it really SR that did away with the aether? The reason that M&M did the experiment was because Maxwell came up with a theory which predicted a constant speed of light ([itex] c = \frac {1} {\sqrt {\epsilon_0 \mu_0 }} [/itex]) and which implied a medium free propagation method. The M&M experiment was supposed to show that Maxwell's theory was wrong, that really there was a medium and that the speed was not constant. In that case they could just throw out Maxwell's equations and be done with it. Unfortunately it did not work that way. They now had medium free propagation of light at a constant velocity while massive bodies seemed to follow a different set of rules. This was the "schism" in Physics which existed from the work of Maxwell in 1867 to Einstein in 1905.

    What Einstein did with SR was to show that ALL matter followed the same set of rules.

    My source for this is Maxwell's Conundrum by Walter Scheider.

    It was Maxwell who did away with the ether, not Einstein.
  5. Nov 26, 2008 #4

    D H

    Staff: Mentor

    The M/M apparatus was fixed with respect to the rotating and accelerating earth. The M/M apparatus frame is not, even in classical mechanics, an inertial frame.

    Even more importantly, the time of day, day of year, and orientation of the M/M apparatus was supposed to have led to different results of the speed of light as measured by the apparatus. It didn't.

    You are mischaracterizing special relativity here. The Lorentz transformations to which you refer are a consequence of Einstein's postulates. His second postulate is in direct conflict with the concept of a luminiferous aether.

    As Integral mentioned, it essentially was Maxwell who got rid of the concept of the luminiferous aether. The last decade of the 19th century and first decade of the 20th century was a time of immense conflict in physics. The intent of the Michelson-Morley experiment was to prove Maxwell wrong. That didn't come to pass.

    What makes the speed of light so special that it alone is the same to all observers? Why not some other velocity? Einstein did not pick the speed of light as that special velocity out of the clear blue sky. The speed of light must be independent of the speed of the observer and the receiver per Maxwell's equations. Einstein was well aware of Maxwell's work; the very first sentence of "On the Electrodynamics of Moving Bodies" references Maxwell's electrodynamics. The second part of that seminal paper is devoted to a detailed analysis of Maxwell's electrodynamics.
  6. Nov 26, 2008 #5


    User Avatar
    Staff Emeritus
    Science Advisor

    Another way of saying the same thing: Galileo's "relativity principle" said that if you were in a closed room moving at constant velocity there was no experiment you could do inside the room that would tell you what the speed was or that you were moving at all. In that sense motion is always "relative"- there is no "absolute" speed.

    But Galileo didn't know about electricity and magnetism. Maxwell's equations stated that the force due to magnetism on an charged body depended on the speed of the body, implying that there was an "absolute" speed" and you could determine your absolute speed inside a closed room by doing some kind of electro-magnetic (i.e. light) experiment. That was what The M-M experiment was designed to determine. The "aether" is just a way of saying what an "absolute" velocity would mean. If there is no aether then there is no absolute velocity.
  7. Nov 26, 2008 #6


    User Avatar
    Science Advisor

    Perhaps ericqb is referring to some ballistic model of light, where light is always emitted at the same speed relative to the emitter, but it can be different in a frame where the emitter itself is moving at relativistic speeds? If so, one reason M&M didn't consider such a possibility was that it would conflict with Maxwell's laws, since in any frame where Maxwell's laws are correct, electromagnetic waves should travel at c regardless of the speed of the emitter in that frame (in aether theory it was imagined that Maxwell's laws would only hold exactly in a single frame, the rest frame of the aether). There have been other experiments that have tested the idea that the speed of light waves is independent of the speed of the emitter, see here (the page is loading slowly now, if you can't get to it here is a copy of the page from archive.org).
  8. Dec 19, 2008 #7
    Thank you all so much for your attention to my M/M question. I should have got back to this site sooner, and I apologize. I don't have time to go the university and find professors willing to answer this question, and so I am very grateful for this opportunity. I would like to say that others expressed this point of view, engineers, students, barroom hippies. Perhaps we are naive, but the reasoning is not complicated and warrants an explanation.

    Although I am very interested in what these posts are saying, I would still like to go back to the two basic ideas I presented because I do not think they have been addessed by these posts.
    (1) Is it not true that the M/M null result -- no interference --makes sense by Newtonian physics alone? No posts have said yes or no to that question, and I would like to know if anyone agrees with me.
    To restate my question: suppose we agree light travels in a vacuum, no ether required. And suppose we are treating the M/M apparatus as an inertial frame, as the textbooks do. What is wrong with the no-interference result upon rotating the M/M device? Why should there be interference? Isn't the result justified by classical physics without any further explanation? First, please, address this simple question.
    Now -- If classical physics has no problem with the M/M result, why do we invoke special relativity?
    I understand that the apparatus is not really an inertial frame, but in textbook discussions it is treated as such, correct? (And it is my understanding that in the late 1800's people debating the M/M result were simplifying the problem, as when we ignore friction in a mechanics problem.) The frame of reference is treated as moving at a constant velocity, tangential to its curving path on the rotating, orbitting earth, for the purposes of the experiment. The student is told to ignore the earth's rotation and orbit, correct? So, given these provisos, that is, calling M/M an inertial frame as per the textbook treatments, does not the M/M null result make sense by classical physics?
    Or, for clarity, let's suppose the M/M apparatus is in the middle of space, not accelerating. Here, without having to ignore any earthbound realities, we have an inertial frame. Under this circumstance is the null result of the M/M experiment not to be expected by Newtonian physics?
    I am not disputing that Einstein said there is no ether, as one post suggested, or that Maxwell said it before him, and I don't see how I was construed to be in doubt of either fact. What I am saying is that, without the notion of the ether, M/M makes sense by classical physics, given that we treat the apparatus as moving at constant velocity.
    If M/M does not make sense by classical physics, I am asking the question, in what way does it not make sense?
    And, if we agree M/M does make sense by classical physics, then how is special relativity supposed to apply?

    (2) M/M apparatus is described as a contained unit in which the emitter and sensor are fixed with respect to each other, correct? Turning the apparatus results in new positions, in which interference caused by the effects of the ether was expected but not observed. But in any such position, the light source and the sensor are fixed iwth respect to each other. (And, in fact, their position with repect to each other has not changed with any rotation of the platform they are on.) So, in the M/M apparatus the light source and the "observer" (emitter and sensor) do not move with respect to each other. True or not true? If not true please explain. May I ask you to please address this question per se? I do not see any of the posts as directly answering this question.
    If, then, the source and observer are not moving with respect to each other, special relativity, which deals with light sources and observers in motion with repect to each other, does not apply, correct? If it applies, please explain how. What are the source and observer that would allow a discussion of different moving frames?
    Again, I am aware M/M apparatus is not an inertial frame in reality, but, again the textbooks say to think of it as such, for the sake of simplicity.
    And again, for clarity, put the M/M apparatus out in space, not accelerating, far enough away from other objects that we can ignore gravitational effects. It it not true that the source and observer (emitter and sensor) are in the same inertial frame? How can this circumstance be thought of as different moving frames? How can special relativity be said to be pertinent to this situation? What source and observer move with respect to each other?
    If we are speaking of the version of M/M in which starlight is the light source, then clearly we have a case of the source and the observer moving with respect to each other, and special relativity applies. But, if you will please speak to what is a quite straighforward question, how does special relativity apply to a situation in which the source and observer are not moving with respect to each other?


    {On a different note, I have a cute math problem. What are the odds that three points on a circle fall in a (any) semicircle? I saw this problem in a Shaum's outline series on probability. The algebra solution is long and convoluted. I would like to know if anyone could give me a simple explanation of how one uses calculus to solve it. I would like to know if anyone can direct me to a discussion of calculus and probability that might give me some insights about how one integrates to arrive at probabilities, without taking a whole course.
    And I would like to know if someone can solve the above problem for n points. I have a website, www.stopdown.net with my own intuitive solution. There is a great article about baboons under science -- the only article about baboons.}
    Last edited: Dec 19, 2008
  9. Dec 19, 2008 #8


    User Avatar
    Science Advisor

    Perhaps, see 4) below.
    Only the accelerative effects, not the velocity effects.

    1) The principle of relativity plus Newtonian physics implies Galilean relativity, ie. the existence of an equivalence class of preferred frames.
    2) Maxwell's equations plus Galilean relativity imply the existence of a single preferred frame (ie. an "aether")
    3) The MM experiment rotation detects an effect of light travelling parallel or perpendicular to movement of the earth relative to the single preferred frame.
    4) The initial MM experiment result is consistent with (i) the earth being at rest in the single preferred frame (ie. no new physics, as you suspected) or (ii) the non-existence of a single preferred frame (ie. new physics).
    5) However, the earth moves around the sun, so it cannot be at rest relative to the single preferred frame throughout the year. Hence later repetitions of the MM experiment cannot be explained by the earth being at rest relative to the single preferred frame.
    6) Maxwell's equations plus Lorentzian relativity (the new physics), compared to Galilean relativity, imply not just a single preferred frame, but an equivalence class of preferred frames - it restored things to what they were before Maxwell! Each Lorentzian preferred frame is just as good as the single preferred frame. So the MM experiment is indeed explained by the apparatus being at rest relative to a preferred frame (usually called a Lorentzian inertial frame).
  10. Dec 19, 2008 #9


    User Avatar
    Science Advisor

    Again, are you thinking of a ballistic model where light always moves at the same speed with regard to the emitter? Do you understand that in any frame where Maxwell's laws are valid, Maxwell's laws themselves say that the speed of electromagnetic waves in that frame does not depend on the speed of the emitter? This is why 19th century physicists made an analogy with sound waves in a medium like air, which all have the same speed in the rest frame of the medium--if you're standing on the ground at rest relative to the air, then if a car coming towards you at 90 mph honks its horn at the same moment it passes a person standing on the sidewalk nearby who yells something at you, both sounds would reach your ears at the same time, the sound wave from the car horn wouldn't get a "boost" of 90 mph. Because of this analogy, physicists imagined electromagnetic waves were a type of sound wave in a medium called the ether, and that Maxwell's laws would only work precisely in the rest frame of the ether. It's true that the M/M experiment does not in itself rule out a ballistic model of light, but such a model would be incompatible with Maxwell's laws in general, which had a lot of experimental support.
  11. Dec 21, 2008 #10
    THE M/M null result can only be explained by Newtonian physics if:

    (a) The speed of light depends on the speed of the emmiter.

    (b) The Earth is stationary.

    (a) has been shown to false by various experiments measuring the speed of light emmited from rapidly moving particles and by parallax measurements of light from stars.

    (b) requires we believe that everything in the universe including the Sun orbits around a stationary (non rotating) Earth. Some observations like the fact that geostationary satellites (that remain stationary relative to a point on the Earth's surface) do not fall to the surface, become hard to explain if we assume the satellite and the Earth are both stationary. It means going back to the epicyclic model of the universe devised by Ptolemy. That model has long been discarded.

    If we reject (a) and (b) then the M/M null result can not be explained by Newtonian physics alone and length contraction, time dilation and the relativity of simultaneity have to be invoked.
  12. Dec 25, 2008 #11
    Special relativity doesn't 'speak to' anything.
    Last edited: Dec 25, 2008
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: M/M experiment is not explained by Special Rel.
  1. Notation: m-m circles? (Replies: 2)