Questions on the Michelson Morley Experiment

[OneAverageGuy]

TL;DR Summary

Michelson Morley experiment - flawed experiment??

I've heard claims the Michelson Morley experiment proves light moves at the same speed in all reference frames. However, after looking through the experimental setup, I do not see how they can make this claim. The entire experiment, from my view, was conducted in Earth's reference frame. His experiment would always show light moving at the same speed in any direction (as it does). He needs to observe from 'outside' Earth's reference frame to make claims on light speed in other reference frames. Am I missing something?

 

[russ_watters]

I've heard claims the Michelson Morley experiment proves light moves at the same speed in all reference frames. However, after looking through the experimental setup, I do not see how they can make this claim. The entire experiment, from my view, was conducted in Earth's reference frame. His experiment would always show light moving at the same speed in any direction (as it does). He needs to observe from 'outside' Earth's reference frame to make claims on light speed in other reference frames. Am I missing something?

Well, you more or less have it, but the assumption at the time was that the rest frame of the universe was the one in which the "fixed stars" were stationary. So, the goal of the experiment was to measure impact of Earth's motion through that frame due to its orbit and rotation. They were surprised to learn their frame acted stationary.

 

[Dale]

Summary:: Michelson Morley experiment - flawed experiment??

Am I missing something?

Two things: First, what the MMX proves is that the speed of light is isotropic. Second, the reference frame of a point on the Earth changes over the course of a day and a year.

 

[hutchphd]

The Earth clearly moves in different directions at different times of the year. What Michelson-Morley showed was that their result was independent of these and any other various velocities (they could rotate the apparatus). By far the most direct explanation is that the round trip speed of light doesn't care about "absolute" velocity.

 

[OneAverageGuy]

The Earth clearly moves in different directions at different times of the year. What Michelson-Morley showed was that their result was independent of these and any other various velocities (they could rotate the apparatus). By far the most direct explanation is that the round trip speed of light doesn't care about "absolute" velocity.

Using the analogy of a moving train. . . If we conduct the experiment entirely on the train (regardless of direction or speed of travel of the train, or the apparatus). We are in the train's reference frame and the speed of light will be constant in all directions (just like a ball dropping from one meter will fall in a straight line and travel one meter in the train's reference). You must be an 'outside observer' to possibly see the speed of light at a different speed (or the ball traveling farther than one meter) from what is seen on the train. Michelson-Marley never had an 'outside observation' from what I understand.

 

[Nugatory]

The entire experiment, from my view, was conducted in Earth's reference frame.

Reference frames have nothing to do with it. We've done the experiment on earth, we've gotten the results we got. Someone watching through a telescope from Mars (that is, using a frame in which our lab on Earth is moving at several kilometers per second) will see the same null result on our interferometer as we do.

 

[OneAverageGuy]

Reference frames have nothing to do with it. We've done the experiment on earth, we've gotten the results we got. Someone watching through a telescope from Mars (that is, using a frame in which our lab on Earth is moving at several kilometers per second) will see the same null result on our interferometer as we do.

Only because the measurement device (interferometer) is operating in our reference frame. I can use a meter stick on the train to show the ball traveled one meter. If you use a telescope from Mars to see the meter stick you will get the same result. However, if you measure the balls actual travel distance (via observation from Mars) using a measuring device you will find the ball traveled further than one meter.

I'm not saying the speed of light is not constant in all frames of reference. Just that I don't think that Michelson Morley experiment can be used to support that theory/postulate.

 

[russ_watters]

Using the analogy of a moving train. . . If we conduct the experiment entirely on the train (regardless of direction or speed of travel of the train, or the apparatus). We are in the train's reference frame and the speed of light will be constant in all directions (just like a ball dropping from one meter will fall in a straight line and travel one meter in the train's reference).

In that example, the train is not accelerating. Earth is clearly accelerating. This, by the way, is a requirement (the second postulate) of SR.

And btw, you said it wrong in the OP, but people let it go. It's not "all reference frames", it's "all inertial reference frames".

 

[hutchphd]

I have spent my entire life in my reference frame. I wouldn't have it any other way.
But seriously your question belies a misunderstanding of the premise, as ably pointed out.

 

[PeterDonis]

our reference frame

There is no such thing as "our reference frame" as a single inertial frame. And, as @russ_watters has pointed out, the "all reference frames" in your OP should be "all inertial reference frames".

 

[PeterDonis]

I've heard claims the Michelson Morley experiment proves light moves at the same speed in all reference frames.

Can you give a specific reference to an example of such a claim?

 

[russ_watters]

Only because the measurement device (interferometer) is operating in our reference frame. I can use a meter stick on the train to show the ball traveled one meter. If you use a telescope from Mars to see the meter stick you will get the same result. However, if you measure the balls actual travel distance (via observation from Mars) using a measuring device you will find the ball traveled further than one meter.

The bouncing ball in a train experiment works if the train is not accelerating and has no windows. Measuring the speed of the ball from Mars demonstrates the problem the MMX was trying to measure; you can't consider the train stationary [not accelerating] anymore, because you can't consider Earth stationary anymore, when you recognize it is in circular motion (two of them).

Btw, I know you've constrained this to MMX performed stationary/on Earth, but I would hope you know that by now it and similar experiments have been performed moving relative to Earth's surface (though I'm unsure if the MMX itself has been performed on an actual train).

 

[FactChecker]

Only because the measurement device (interferometer) is operating in our reference frame. I can use a meter stick on the train to show the ball traveled one meter. If you use a telescope from Mars to see the meter stick you will get the same result. However, if you measure the balls actual travel distance (via observation from Mars) using a measuring device you will find the ball traveled further than one meter.

I'm not saying the speed of light is not constant in all frames of reference. Just that I don't think that Michelson Morley experiment can be used to support that theory/postulate.

The Michelson-Morley experiment was expected to show that light went through a field of luminiferous aether. That was the reference frame that was proposed to be "stationary". The motion of the Earth through the aether was expected to cause a speed difference relative to the Earth. It did not seem reasonable to think that the aether in the entire universe was tied to the motion of the Earth around the Sun. Once the aether theory was weakened, the idea that the velocity of light is constant was (very slowly and reluctantly) accepted.

 

[Nugatory]

You have been misled by a very common but sloppy English-language idiom. The phrase

Only because the measurement device (interferometer) is operating in our reference frame

is nonsense because there's no such thing as something being in one reference and not another - everything is always in all reference frames.

A reference frame is a mathematical convention for assigning coordinates to events in spacetime, and we are free to choose any reference frame we want to in analyzing any problem. In that sense everything is always in all reference frames; the Martian astronomer's analysis of the earthbound experiment is complicated by their need to allow for the Earth's motion when he's using the frame in which he is at rest but the result will be the same as we get using the coordinates in which the Earth is at rest.

Even though it's nonsense if taken literally, you'll hear experts saying "in Bob's reference frame..." all the time. Actually they mean "using the coordinates assigned by the reference frame in which Bob is at rest, which we've chosen to use at this point in the conversation because it simplifies the calculations, ...", but that's such a mouthful that we fall back on the "in Bob's frame" wording in casual conversation.

 

[OneAverageGuy]

The Michelson-Morley experiment was expected to show that light went through a field of luminiferous aether. That was the reference frame that was proposed. The motion of the Earth through the aether was expected to cause a speed difference relative to the Earth. It did not seem reasonable to think that the aether in the entire universe was tied to the motion of the Earth around the Sun. Once the aether theory was weakened, the idea that the velocity of light is constant was (very slowly and reluctantly) accepted.

Thanks. I'm aware of the original purpose of the experiment. I also agree that it demonstrates that there is no aether (at least none that effects light). Because the Earth was the fastest moving object upon which an experiment could be setup (and the interferometer the most accurate measurement device), it was the best data to measure possible variations in the speed of light around the time of the experiment. Can you name a few of the more recent experiments on speed of light? (showing constant identical speed from all reference frames). Thanks for your time.

 

[FactChecker]

I am not qualified to give a good list of experiments. There have been lists given in this forum before. You can do a search.

 

[russ_watters]

It's a sticky at the top of the forum:
https://www.physicsforums.com/threads/faq-experimental-basis-of-special-relativity.229034/

Note: It's unreasonable/not useful to artificially constrain the types of experiments to be exactly what you specify. SR is pretty much a package deal -- a test of one prediction is pretty much a test of the entire theory.

 

[Ibix]

Can you name a few of the more recent experiments on speed of light? (showing constant identical speed from all reference frames). Thanks for your time.

You can look at a variety of experiments. Michelson-Morley shows the isotropy of the speed of light. Fizeau's measurements of the speed of light in moving water show that refractive index has a peculiar dependence on speed, which is easily explained by the Lorentz transforms. De Sitter's observation of binary stars shows the source independence of light speed. See this Insight article.

Collectively, these things are easily explained in a relativistic universe, and no other theory we know of can do it. Almost all of modern physics sits on top of relativity, so one can take a lot of modern physics (notably GPS clocks and particle accelerators) as further validation of relativity.

 

[Grasshopper]

Thanks. I'm aware of the original purpose of the experiment. I also agree that it demonstrates that there is no aether (at least none that effects light). Because the Earth was the fastest moving object upon which an experiment could be setup (and the interferometer the most accurate measurement device), it was the best data to measure possible variations in the speed of light around the time of the experiment. Can you name a few of the more recent experiments on speed of light? (showing constant identical speed from all reference frames). Thanks for your time.

(Non-physicist commenting here):

Isn’t every test that confirms Lorentz invariance essentially the same thing?

If so, this article from 2006 could lead you to an experiment.

https://www.aps.org/publications/apsnews/200606/lorentz.cfm

Also, I imagine the de Sitter effect is of use. If I understand correctly, if the speed of light was not independent of its source, we’d see things a lot differently when looking to the sky (binary stars in particular).

Performed by Brecher.

https://ui.adsabs.harvard.edu/abs/1977PhRvL..39.1051B/abstract

 

[A.T.]

I've heard claims the Michelson Morley experiment proves light moves at the same speed in all reference frames.

No, it doesn't. The MM-result is consistent with the ballistic theory of light, which had to be ruled out by other observations:
https://en.wikipedia.org/wiki/Emission_theory

Can you name a few of the more recent experiments on speed of light? (showing constant identical speed from all reference frames).

There is no possible experiment showing something for all reference frames, because there are infinitely many reference frames.

 

[PeroK]

Can you name a few of the more recent experiments on speed of light? (showing constant identical speed from all reference frames). Thanks for your time.

The postulate of SR is that the speed of light in vacuum (as measured in an inertial reference frame) is independent of the motion of the source.

To confirm that postulate, you vary the source of light as much as you can and check that the speed of the emitted light is always $c$.

From this you can conclude that the speed of light is invariant as measured in any inertial reference frame. There is no separate test for this, as this is equivalent to the speed being independent of the motion of the source.

To emphasise this, suppose someone measured the speed of a light ray to be other than $c$. Then, they could take note of what the source was doing when it emitted the light and reproduce the experiment themselves. If they set-up the source doing some fancy motion, then that is the same experiment as someone else setting up the source to do some fancy motion.

There's no sense in which light physically "belongs" to one reference frame. It is emitted by a source and the motion of the source is the only variable.

 

[cianfa72]

To confirm that postulate, you vary the source of light as much as you can and check that the speed of the emitted light is always $c$.

From this you can conclude that the speed of light is invariant as measured in any inertial reference frame. There is no separate test for this, as this is equivalent to the speed being independent of the motion of the source.

Any measurements and experimental setup can actually measure just the two-way speed of light. The one-way speed --by very definition-- is actually physically undefined.

 

[PeroK]

The measurements and any experimental setup can actually measure just the two-way speed of light. The one-way speed -- by very definition-- is physically undefined.

I fail to see the relevance of this to the current discussion.

 

[cianfa72]

I fail to see the relevance of this to the current discussion.

Sure, I agree. It was just to highlight that point often not fully understood

 

[Grasshopper]

Any measurements and experimental setup can actually measure just the two-way speed of light. The one-way speed --by very definition-- is actually physically undefined.

Wait, you’ve confused me here. Isn’t choosing a synchronization convention literally defining the one-way speed of light? Or perhaps does adding the word “physically” in front simply mean that it is (as of yet) just unknowable?

 

[Nugatory]

Isn’t choosing a synchronization convention literally defining the one-way speed of light?

It is defining the one-way speed, but the word “physically” is often reserved for things whose definition does not depend on arbitary choices of convention.

 

[cianfa72]

Wait, you’ve confused me here. Isn’t choosing a synchronization convention literally defining the one-way speed of light? Or perhaps does adding the word “physically” in front simply mean that it is (as of yet) just unknowable?

It is physically unknowable just because you cannot measure it without a definition of simultaneity given by the coordinate chart chosen.

 

[Sagittarius A-Star]

However, the whole thing, in De Sitter's time, would be a flickering dot.

No, please see here for more information:
https://www.physicsforums.com/threads/what-tests-can-falsify-general-relativity.998068/post-6442753

 

[Grasshopper]

"De Sitter" remains the only decisive candidate. Is there any data on the exact observations of the Sitter's observation of binary stars? Distances, rotational speeds, diameters, etc. You would not see two versions of one star, but, if the star's orbital speed is sufficient, an ellipse shape (the orbit of the star) with a redshift in the removing part and a blueshift in the approaching part. However, the whole thing, in De Sitter's time, would be a flickering dot. Aren't these the same flickering dots as we see some stars with the naked eye? Why can't we just measure the speed of light from another galaxy with enough speed relative to the Milky Way to blow away all the fog on this subject?

There are closer to modern observations.

For example, in 1977:

https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.39.1051

 

[Ibix]

Your link sends me to an article. I am not going to buy the item. Can you describe exactly in your own words why which conclusion is based on and on what? Or are we going to play ping pong with articles?

Brecher is a re-run of de Sitter with x-ray sources.

The "fog" is the fact that it is not understandable why EVERYTHING is relative except the speed at which a photon travels from A to B.

Indeed it is not understandable, since everything is not relative. A great many things are invariant - proper acceleration, proper time, the readings on any instrument...

If I can't explain it to my mother-in-law,

Perhaps you should work on understanding the theory correctly before trying to explain it? From our various conversations you seem to have several misconceptions about it.

 

[Sagittarius A-Star]

In the same way we measure the speed of light from an earthly source. (According to the Fizeau method.) Maybe it is very difficult, but nowadays we can do quite "a lot". Measuring is knowing.

It was done by O. Heckmann in 1960 via aberration:
http://adsabs.harvard.edu/full/1960AnAp...23..410H

via:
http://www.edu-observatory.org/physics-faq/Relativity/SR/experiments.html

 

[Sagittarius A-Star]

Dickens observed in 1965 the "quasi-stellar" radio source 3C48 with evidence (via aberration) against the Ritz theory of light propagation:
http://articles.adsabs.harvard.edu//full/1965Obs...85..260D/0000260.000.html

In 1982 it was found out, the the quasar 3C48 is contained in an 3.9 billion light-years (light travel time) distant galaxy with the same redshit, indicating a radial velocity of 110,024 km/s of the quasar relative to the earth.

 

[cianfa72]

Maybe what follows has been already discussed in this thread...sorry

Consider the no dragging ether hypothesis (as I understand it the ether should not be dragged by the Earth in its motion around the Sun). Suppose the speed of light depends on the motion of the source. Since we know velocity is relative, with respect to what should the speed of light have a variable value? Thanks.

 

[Sagittarius A-Star]

In the text of your article: "We need not take into account very seriously the idea that ..." The piece is biased.

No. It only describes the motivation for the following: "So let us assume that the red-shifts are caused by real movements".

Heckmann predicted your posting #28 an did exactly, what you requested:

Why can't we just measure the speed of light from another galaxy with enough speed relative to the Milky Way to blow away all the fog on this subject?

 

[Ibix]

In the text of your article: "We need not take into account very seriously the idea that ..." The piece is biased.

You should probably read the article, rather than dismissing it in its second sentence based on what appears to be a misreading. What Heckmann is choosing not to consider is the "tired light" hypothesis, where photons are redshifted in flight for some reason or reasons unknown (their speed might be unaffected by this process). He explicitly does consider (and rules out) the "ballistic" theory of light where the speed of light depends on the speed of the source - which was exactly what you were proposing.

Edit: Also, what Heckmann says in the abstract, an explicit disclaimer that he is not considering "tired light", is a sign of intellectual honesty, not bias. If you believe in tired light, he says, this paper neither confirms nor denies it. The measurements may permit an analysis that rules it out, but he does not conduct one.