The Michelson-Morley interferometer was used to assert the aether does not exist. The Sagnac interferometer was used to assert the aether does exist. These experiments are discussed because they are relevant to the development of a unified model for light (phot model) and also to the debate about whether there is a universal spacetime ‘fabric’ or field responsible for the inertial properties of matter.
Introduction
Waves (water waves, sound waves, waves on a string etc,) typically travel at well-defined speeds in the medium in which they occur. By analogy, it was postulated that light waves must be occurring in an invisible “luminiferous aether” and that this aether filled the whole galaxy and that light travelled at a well defined speed relative to this aether.
By the mid 19th the wave model for light was well established. There was however a question about how fast it travelled and also about the properties of the aether. The two questions are related.
Bradley, Eotvos, Roemer and others showed that telescopes had to lean a little bit one way and then lean a little bit the other way six months later in order to maintain a fixed image fixed of a distant star. This stellar aberration was interpreted as being caused by the Earth moving through the luminiferous aether.
Various experiments to measure the speed of light, notably by Hippolyte Fizeau (1819-1896) showed that the speed of light in laboratory settings was very close to 3×108 meters per second.
According to the theories prevailing at the time, light traveled in an invisible medium called the luminiferous aether. It was expected that the speed of light would be its speed in the aether, plus any speed of the aether relative to the observer.
One of Fizeau’s experiments checked the speed of light in a pipe filled with moving water. He found that there was some extra speed moving with the water flow as compared to moving against the water flow, but the effect was less that expected. This corroborated some ideas by Fresnel in 1818 explaining an experiment by Arago in 1810.
Fresnel concluded that if light from the Sun was transmitted as waves the refractive index of a glass-air interface should vary as the glass moved through the aether. He thought that if a glass prism were moving towards the Sun it would carry some of the aether with it and this would change its refractive index. The effect would vary between morning and evening due to the rotation of the Earth.
Fresnel’s ideas explained the result of Arago’s experiment. The ideas introduced the concept of a largely stationary aether that is dragged by substances such as glass but not by air. Fizeau’s experiment supported Fresnel’s ideas. So the concept took hold that the Earth moved through a stationary aether but dragged some of the aether along with it. This was also consistent with the phenomenon of stellar aberration.
Some scientists were not convinced and favoured the idea that the Earth fully dragged an aether field with it. Scientists such as George Stokes, Martin Hoek, Heinrich Hertz, Hendrik Lorentz, George Fitzgerald and Max Planck all contributed to the issue.
In 1886-87 Michelson and Morley performed a precise and careful experiment to look for evidence that the Earth was moving through an aether field – the ‘aether drift’.
Michelson-Morley Experiment
The experimental apparatus set up by Michelson and Morley in Ohio in 1886-87 was designed to measure the effects of the Earth’s movement through the aether field. It consisted of a system of a beam splitter (a half silvered mirror angled at 45 degrees to incoming light), plus mirrors on arms orthogonal to each other creating paths exactly as long as each other (multiple reflections were used to get a path about 11m long). The whole apparatus was on mounted on a massive sandstone block floating in mercury in order to reduce vibrations and to allow everything to be rotated.
Fig 2: Suppose the apparatus is moving left through the aether. Event 1 is when light leaves the beam splitter in two directions. The light comes back again at Event 4. In the time interval between events 1 and 4 the beam splitter has moved a short distance to the left, relative to the aether. This creates a difference between the green and red light paths and should have affected the interference pattern on the screen. But it did not.
Light travelling with the aether drift and then back again would take longer to complete such journey than light travelling across the aether wind and back again (simply because the light would lose more time ‘swimming against the tide’ on the way back than it gained getting on the way forward). Any travel time differences would show up as an interference pattern. No such pattern could be generated, no matter which way the apparatus was oriented. It became perhaps the most famous null experiment in physics.
The experimental result is that there is no change in the interference pattern on the screen no matter which way the apparatus is turned. No evidence of aether drift at all.
The result was interpreted as proving that the Earth was not moving through a stationary aether, and that the speed of light through the apparatus was the same in any direction. Since other experiments had ruled out the possibility that the Earth was fully dragging an aether cloud, the conclusion was that there was no such thing as an aether. Michelson was disappointed about this.
Einstein did not acknowledge this experiment in his 1905 paper on Special Relativity but he must have known about it. He decided that if the aether was not going to show any effects in relevant experiments then it served no useful purpose. He decided to postulate that the speed of light was the same (for a particular medium such as a vacuum or a gas at a given temperature and pressure) for any inertial reference frame, no matter how fast such frame is moving, or what direction it is moving in.
(Other scientists have taken this postulate to an unsubstantiated. Even though the speed of light has only been measured in our infinitesimal corner of the Universe, on a rotating, orbiting planet which is not an inertial reference frame, and for an infinitesimal blink of time, they have assumed the speed of light has the same value throughout the whole Universe and has always has been.)
Einstein’s postulates necessarily lead to Lorentzian distance contractions and time dilations in systems that are moving relative to inertial reference frames. This postulate based approach was mathematically convenient and over the next few decades scientists gradually gave up the idea of an aether and just took the constant speed of light as an axiom. You could say that the aether was never disproved, it just became superfluous.
Note that the Michelson-Morley experiment was not designed to measure the speed of light but only to look for evidence of an aether drift.
Discussion
The Michelson-Morley experiment is so important every aspect of it requires careful consideration. This discussion will consider light to consist of discrete units of energy. You can think of these as photons, but since other essays in this series have been developing a unified model of light called phots, the term phots will be used instead.
Note that the experiment is done in a gravitational field of a rotating planet that is orbiting the Sun, and the Sun is orbiting the galactic centre and the whole galaxy is rotating as well. So it is not a perfect inertial reference frame. However, most of that does not matter significantly.
Like the Mach-Zender interferometer and many others, this experiment uses a beam splitter. However, a half silvered mirror cannot split a phot. For a discussion about what might be happening at the beam splitter see other essays (Van de Vusse, 2024). For now assume that phots move off in pairs – one to the mirror at the top and another to the mirror on the right.
Take the viewpoint of observers at rest relative to the beam splitter. As far as they are concerned some phots have travelled up and back for a total distance of 2L meters, and other phots have travelled right and left for a total distance of 2L meters. The phots arrive at event 4 at the same time. This is evidence that they have all been travelling consistently at the same identical speed.
It is vaguely possible that a phot has travelled fast or slow on one leg of the journey and has made up the difference on the way back. This is not plausible for the green path in the diagram, but it is worth checking for the red path. The aether drift idea was that the phot travelling to the right would do so extra quickly because it would be travelling ‘with the wind’, but it would face an aether ‘headwind’ on the way back. The aether wind effects would not cancel each other out because the headwind effect exerts its effect for a longer duration than the tailwind effect. Overall the aether wind would cause a slight time lag in the red path.
The green path is not entirely unaffected. It’s length would be 2L’ and not 2L.
The results of the experiment are undeniable. The experiment could not detect an aether drift. It was sensitive enough to detect such a drift but it did not.
There is a common view that the experiment proved that there is no aether drift, and some go further to assert it proved there is no aether. This essay proposes a different interpretation:
Conjecture 1: The Michelson-Morley experiment set out to discover the aether drift. What it discovered instead was relativistic distance contraction and relativistic time dilation, both in keeping with Special Relativity. These two relativistic effects ensure that the light in both paths takes the same time to get back to the beam splitter.
Conjecture 2: The relativistic distance contraction and relativistic time dilation affecting the Michelson-Morley experiment occurred precisely because the experiment was moving eastwards through a background aether field. Whether or not such field exists and whether it constitutes a luminiferous aether is a big question that will be considered in later essays.
Sagnac Interferometer
In 1913 Georges Sagnac (1869-1928) investigated what happens when light is sent in both directions around a closed loop. He was trying to detect the luminiferous aether. If two pulses of light are sent in opposite directions around a stationary loop, they arrive at the end point simultaneously and no interference pattern results. If however the loop is rotating then the pulse traveling in the same direction as the rotation travels a slightly greater distance than the pulse traveling in the opposite direction. As a result, the counter-rotating pulse arrives at the interference point slightly earlier than the co-rotating pulse and an interference effects can be generated.
Sagnac noted that turning the device to various degrees did not affect the outcome. Nor did the daily motion of the Earth produce a noticeable effect. However, if the device was set spinning then the interference pattern was affected. This is called the Sagnac effect. The faster the spin the greater the effect.
The question is – how does the device ‘know’ that it is rotating? Sagnac believed that his device was detecting ongoing rotations with respect to the aether.
Mach and many others believed that rotations and inertia effects generally only existed in relation to the “fixed stars”. Are the two ideas related? Is there some sort of connection betweenthe Machian idea and the aether idea?
Thought Experiment: Consider a rotating Sagnac interferometer from the point of view of observers who are rest on the device as it rotates. They are not inertial observers. In fact thay will need to be restrained to avoid them being flung off like mud from a bicycle wheel. However, they will still be capable of looking at the interference pattern on the screen. They will see the same results as anyone else since the outcomes of experiments are not affected by the arbitrary choice of reference frame. They wil see light which has been going around opposite to the rotation arriving back at the screen sooner than light going the other way around. To them it will look like the light going in the retrograde direction goes faster.
Discussion
Sagnac used his experiment to look for evidence of a luminiferous aether and believed he had found it. In his 1911 text on Special Relativity Max von Laue showed that the Sagnac effect is consistent with Special Relativity. And since Special Relativity rests on three postulates that side-step the existence of an aether all together, the Sagnac effect was neither proof nor disproof of an aether.
So what exactly does the Sagnac effect demonstrate? It shows that any physical system can determine whether or not it is rotating purely by the means of a local experiment. It therefore complements Foucault’s pendulum and Newton’s rotating pail of water. All three devices are responding to something very fundamental in the physics of the Universe that we do not yet properly appreciate.
It is clear that local physics experiments give the same results if a reference frame is displaced in space, or turned through a fixed angle, or started at a different time. That’s Galilean relativity. Special Relativity goes further and says that provided the reference frames do not display evidence of being accelerated or spun around then all local physics experiments will still give the same results. Einstein later added to this and said that in the presence of gravitational fields or uniform linear accelerations the Special Theory of Relativity should be replaced by the more General Theory of Relativity.
What is missing here is a good fundamental approach to ongoing rotations. Admittedly it is possible to describe rotating systems within Special and General Relativity but it is usually quite complicated. The Ehrenfest Paradox is an example of such complexity.
Rotating systems are even a problem in Newtonian physics since they require the introduction of “fictitious” centrifugal and Coriolis forces. These fictitious forces are strange. When a normal force accelerates a massive object energy is expended, and when the force stops the massive object moves off at its new increased speed. Not so the fictitious forces. Consider a smoothly gliding skater who reaches out sideways and grabs a pole fixed in the ice. Immediately the skater is swung around into a circle and feels tension in their arm as if they are being accelerated by the pole. If they let go they move off in a straight line at the same speed as before. But probably in a different direction. And no energy has been expended. It is a fundamentally different process from a linear acceleration.
Summary
The Michelson-Morley experiment set out to measure the aether drift but could not detect it. This cast serious doubt upon the existence of an aether. In fact what Michelson and Morley had discovered was Lorentz length contraction and Lorentz/Poincaré time dilation. Sagnac tried a closed path version of much the same experiment and discovered that moving the apparatus while light was going in opposite directions around a closed path did produce interference effects. It is suggested here that these results show that linear motions might be relative, but rotating motions seem to be absolute. The essay also shows that both experiments can be discussed using the unified model for light (phots) examined in earlier essays.
Reference
Van de Vusse, Sjoerd B.A., 2024, Some ideas and experiments for issues affecting modern physics, https://hereticalphysics.com.au
Author contact: SBAvan@utas.edu.au
Author’s location: Hobart, Australia
