Q&A: Michelson–Morley Experiment
Michelson–Morley Experiment
Question
I wanted to clarify a point that is not completely clear to me regarding the Michelson–Morley experiment.
I saw on YouTube a course called ‘Physics at Eye Level,’ by Professor Ron Lifshitz from Tel Aviv University, and as I understood it there was an inaccuracy in what he said there. I tried to contact him and did not receive an answer. That led to some confusion for me regarding this experiment, and I would be very glad if you could explain the matter to me.
1 In one of the lectures dealing with the speed of light, the Michelson–Morley experiment is described, in which those researchers tried to measure the speed of the Earth relative to the ‘ether,’ with the simple assumption being that it was at rest and was not affected by the Earth’s motion. In the ‘interferometer’ apparatus they split a beam of light into two, so that one beam moved in the direction of the Earth’s motion and the second perpendicular to the Earth, and both beams were reflected by a mirror. At that point the lecturer stopped to explain that the simple assumption was that the light beam moving in the direction of the Earth should be faster than the light beam moving perpendicular to the Earth, based on the rule of adding and subtracting velocities, to which he devoted the entire beginning of the lecture (to the beam moving with the Earth’s direction one should add the Earth’s speed, and to the second beam not). In practice, no interference effect between the two beams was observed in the ‘interferometer,’ in different directions and in different seasons of the year, which indicated that their speeds were identical, and no solution to this phenomenon was found until Einstein’s special theory of relativity.
2 Here I wanted to comment: as far as I understand, the rule of adding/subtracting velocities applies only to the motion of bodies, not to the motion of waves. The propagation speed of a wave depends on various factors, such as the medium in which the wave moves and so on, but it is not affected by whether the source of the wave was at rest when it created the wave or was in motion. For example, there is no difference in the speed at which a sound wave will reach me between a car standing still a hundred meters away from me and honking, and a car driving and honking from that same point; in both cases the sound wave will reach me at the same speed.
3 Elsewhere on the internet I saw the experiment described somewhat differently. There it was said that the assumption was that the ‘ether’ ‘flows’ in a certain direction, and their analogy for the experiment was a boat sailing in the direction of the river’s flow and creating waves in the direction of its sailing and perpendicular to it. In such a picture it is clear that the waves advancing with the river’s flow will move faster, because in that direction, in addition to the wave’s rate of advance, there is also the advance of the medium itself (since the water is flowing), unlike waves moving perpendicular to the boat (from bank to bank), for which the rate of advance is only the speed of the wave, since the medium itself is not advancing in that direction.
4 This description answers my puzzlement about why there should be a difference between the speed of the two light beams in the Michelson–Morley experiment, but it raises a new puzzle: why was their simple assumption that the ether ‘flows’ in a certain direction like a stream? Why not assume that the ether is in absolute rest and is not affected at all by the Earth’s motion within it? (This assumption is not especially far-fetched compared to the other special properties that were attributed in those days to the substance called ether.) If we accept that the ether is in absolute rest and the Earth moves within it and relative to it at a constant speed, then there should not be any difference between the speeds of light beams sent in different directions even without resorting to special relativity, because the distance the beams traveled was identical, and in addition the rate of advance of the beams was also identical, since one should not add or subtract the Earth’s speed from the speed of the beams, because light moves in waves, and the addition and subtraction of velocities applies not to waves but only to bodies.
I would be very glad for a reply.
Answer
A question like this really belongs in a physics course, especially since I’m already a bit rusty on these matters. I’ll try to answer briefly.
- The speed of any wave does not depend on the speed of the medium, but at most on the speed of the source; however, light can travel in a vacuum whereas sound cannot. Therefore the assumption was that the speed of light would depend on the speed of the source, because it is not a function of the medium, unlike the speed of sound. Therefore they expected the sum of velocities to be as you described. The experiment showed that the speed of light does not depend on the speed of the source, because it is absolute.
- The ether was a proposal for some kind of medium needed for light waves (as air functions with respect to sound waves). This was ruled out by the experiment. As I understand it, if the ether were at rest relative to the Earth, you would gain nothing from positing ether. It would have no significance beyond a vacuum.
Discussion on Answer
1. What I meant was indeed that for a sound wave the speed depends on the medium and not on the speed of the source. If the medium moves, then indeed the speed increases. For a light wave there is no medium, and therefore the assumption was that what determines it is the speed of the source. But the explanation remains the same.
2. I am not a historian. I don’t know.
3. If the ether is at rest relative to the Earth, then the experiment is perfectly fine. In the direction in which the Earth rotates, the ether also moves, and therefore the speed in that direction will be different from that in the perpendicular direction.
1 You don’t need to be a historian to discuss this. The simple understanding for years was that there were only two possibilities for light: particles that move, or a wave (which requires assuming a medium in which the wave moves). The ideas that light is some other kind of entity, or a wave-particle combination, came much later.
Everywhere I have seen the Michelson–Morley experiment described, their assumptions were very simple: 1. Light moves in waves through the ether. 2. The Earth moves and rotates all the time, and therefore if you send two light beams, one in the direction of the Earth’s motion and the other against it, you should see an interference phenomenon because of the difference between their speeds.
So what you wrote — ‘For a light wave there is no medium, and therefore the assumption was that what determines it is the speed of the source’ — is not an assumption that appears in their experiment. To them it was obvious that light moved in a medium, namely the ether.
2. Another point: even if there is no medium for light, as is accepted today, it is still not clear how the speed of light is supposed to be affected by the speed of the source as you wrote. The only way to say that is if we assume light is a particle, but if so there should not be any interference phenomenon at all (which is characteristic only of waves), and there is nothing at all puzzling about the results of their experiment. If you mean that light is a third kind of entity, who says it should exhibit interference?
3 Incidentally, this is also what bothered me in Professor Ron Lifshitz’s lecture (this is the link to the YouTube video
At the beginning of the lesson he explains the whole issue of adding and subtracting velocities as it applies to bodies, and then (from minute thirty), by means of that introduction, he explains the Michelson–Morley experiment, without taking into account that in doing so he is trying to have it both ways: on the one hand he assumes that light ought to obey the rule of adding and subtracting velocities, which means that light is a particle (since for waves that rule is not valid), and on the other hand he tries to observe an interference phenomenon in the Michelson–Morley experiment, which is characteristic only of waves and not of particles (he does not propose there a third possibility for the nature of light).
In my opinion this is an example of an interesting logical failure. What do you think?
4. You wrote, ‘In the direction in which the Earth rotates, the ether also moves, and therefore the speed in that direction will be different from that in the perpendicular direction’ — that is exactly what puzzles me. The researchers at the time were so baffled by the results of their experiment and raised all kinds of strange hypotheses about why there was no difference at all between the speeds of the light beams (until Einstein came along), and my claim is that in order to expect a difference, one has to assume that the ether also moves, and that is an assumption that is not forced from any direction. The special properties they attributed to the ether at that time were much stranger than simply assuming that the ether is in absolute rest. So I do not understand what all the confusion was about (unless I am missing something).
Apparently the solution to their experimental results is simple: the ether is in absolute rest, light moving in waves is not affected by the motion of its source, as is the way with waves, and therefore there is no difference between the speeds of the two beams, even though one of them moves in the direction of the Earth’s motion.
I think this is indeed a claim. What do you think?
I would be glad for a reply
Apparently there are several inaccuracies in what you wrote, in my humble opinion
1 You wrote, ‘The speed of any wave does not depend on the speed of the medium, but at most on the speed of the source.’
A. I did not write that the speed of a wave depends only on the speed of the medium, since wave speed depends on all kinds of factors and properties of the medium. What is clear, though, is that if there are two waves moving in a given medium at the same speed, and one medium is moving in the direction of the wave, then the wave in that medium too will reach a given point faster. So certainly the rate of advance of the wave also depends on the speed of the medium.
B. How is wave speed affected by the speed of the source? If a car 300 meters away from me makes a honking sound, the sound will reach me about a second later. Similarly, if a car is traveling at 50 km/h and honks at that same point 300 meters away from me, the honk should take a similar amount of time, since the air, which is the medium through which the sound travels, is not affected by whether the source that emitted the sound was at rest or moving. Isn’t that right?
2 You wrote, ‘However, light can travel in a vacuum whereas sound cannot. Therefore the assumption was that the speed of light would depend on the speed of the source, because it is not a function of the medium, unlike the speed of sound. Therefore they expected the sum of velocities to be as you described.’
I don’t think that at that time there was an understanding that light moves in a vacuum as a wave. There were two possibilities: either light is a stream of particles moving through space (and then it also moves in a vacuum), but then although the rule of adding and subtracting velocities would apply to it like any body, still there would be no interference phenomena and the like in it, since those belong only to waves (and in any case the Michelson–Morley experiment should not show anything, since it is based on interference).
Or, as Michelson and Morley thought, light moves in waves within some medium, which is the ether, and therefore it has interference and so on. But alongside that, since a wave is not affected by the speed of the source that emitted it, there should be no difference between the light beams in their experiment, unless we assume that the ether moves in a certain direction. And that is exactly my question: why were they so intent on making that assumption?
3. You wrote, ‘As I understand it, if the ether were at rest relative to the Earth, you would gain nothing from positing ether. It would have no significance beyond a vacuum.’
I don’t understand what you mean. The gain in positing ether would be that apparently the only way to maintain that light moves in waves is to assume that there exists some substance serving as the medium in which the waves move; without a medium there are no waves. Therefore, if we assumed that the ether is at rest relative to the Earth, then it would be clear how light can move in waves, and on the other hand there would be no problem at all in the Michelson–Morley experiment, as I wrote above.