As the 19th century came to a close, physicists were feeling pretty satisfied with the state of their science.

The great edifice of physical theory seemed complete.

A few minor experiments remained to verify everything.

But little did those physicists know that one of those experiments would bring the entire structure crashing down paving the way for the physics revolutions of the 20th century.

In 1894, the eminent physicist Albert Abraham Michelson described the state of his science like this “it seems probable that most of the grand underlying principles have been firmly established and that further advances are to be sought chiefly in the rigorous application of these principles to all the phenomena which come under our notice.” Michelson had lived this principle.

In 1887, he and fellow American Edward Morley attempted to verify the existence of one of the oldest concepts in physics.

They attempted to measure the luminiferous aether - the invisible, all-pervasive medium through which light was thought to propagate.

Their experimental design was so ingenious it couldn’t possibly fail.

And yet it did fail - failed to detect the aether, at least.

But in doing so it succeeded in cracking the supposedly unassailable foundations of 19th century physics.

The death of the aether helped open the way for the acceptance of Einstein’s theory of relativity.

So today we’re going to celebrate the life and death of an idea.

As with so many ideas, the aether was born in the minds of the ancient Greeks.

For them it was the air breathed by the Gods, and was embodied as a primordial deity of light.

In Aristotle’s cosmology, earth, air, fire and water are the physical elements of the world, while aether is the immutable and indestructible element that filled space and in crystalline state formed celestial bodies.

That’s right, aether is the fifth element.

Big ba da boom.

Medieval alchemists were all about the aether, also called quintessence.

They believed this spiritually pure element had miraculous capabilities of transmutation.

And in some cases that it could be extracted by distilling huge quantities of human urine.

It turned out that was phosphorus.

Still useful I guess.

Fast forward to the mid 17th century.

French philosopher Rene Descartes asserted that there could be no such thing as empty space, and Aristotle’s aether therefore both filled and gave reality to the space between the celestial bodies.

He imagined the aether as flowing - moving in vortices.

These flows carried the celestial bodies, for example in circles around the Sun.

The Dutch physicist Christiaan Huygens developed a detailed mathematical theory around this idea that gravity results from the fluid dynamics of the aether.

But Huygens is most famous for his wave theory of light.

By thinking of light as a wave, he was able to build a theory of optics, explaining how light refracted - bent in its path - between different substances.

And in Huygens’ mind, that theory also required the aether in order to make sense.

Consider a wave on a string: each string segment moves up and down only, tugging on neighboring segments so that a sinusoidal pattern flows forwards, while the segments themselves just oscillate in place.

In sound waves, air molecules oscillate back and forth to propagate a pattern in the density of the air.

All familiar waves - what we’ll call classical waves - are a chain reaction that propagates an energy pattern through some medium.

So if light is a wave, surely it also needs a medium.

For Huygens, that medium was also the aether.

He called it the luminiferous - or light-bearing aether.

This was in Huygen’s 1690 Treatise on Light, published only 5 years before his death.

Across the channel, a much younger Isaac Newton was sitting in Cambridge making life difficult for Huygens.

First his theory on universal gravitation conflicted with the predictions of Huygens’ aetheric gravity.

And Newton also opposed this whole wave theory for light business.

Now Newton’s case is complicated - some of his early ideas on gravity incorporated the aether, and he was also an avid alchemist, so must have studied the esoteric version of aether.

Whether he also studies his own urine isn’t clear.

But he eventually came to the position that Huygen’s luminiferous aether was bunk.

He reasoned that if any medium filled the space between the planets, surely it would cause a sort of drag that would impede their apparently perfect motion.

Newton also favoured his own corpuscular theory of light - light as tiny particles rather than waves.

Huygens versus Newton.

Light as a wave versus a particle.

Most accepted Newton - as most always did.

This was until the beginning of the 19th century when Thomas Young performed his famous double slit experiment.

He showed that light produces an interference pattern passing between a pair of slits, like water waves do.

These bright bands are points on the screen where the waveforms of light emerging from the two slits happen to line up perfectly, reinforcing each other.

In between the waveforms are out of phase and so cancel out.

We see exactly the same effect in a pair of expanding ripples on the surface of water.

So, light experiences refraction and interference like a wave.

Add to that the fact that in the 1860s, Maxwell’s equations predicted that electromagnetic waves should travel at exactly the speed of … wait for it, light.

It was becoming pretty clear that light was indeed some type of wave.

And waves need a medium, so the reality of the luminiferous aether seemed a done deal.

And so we get to the end of the 19th Century.

With Newton’s mechanics and gravity, Huygen’s optics, Maxwell’s electromagnetism, physics seemed pretty much wrapped up.

Albert Michelson and Edward Morley thought they’d help tie a bow on it by demonstrating the existence of the luminiferous aether as a classical medium for the propagation of light.

The key was to observe a change in the speed of light depending on the direction of motion.

All classical waves travel at a constant speed relative to their medium.

For example, sound waves travel at around 340 m/s, depending on conditions if you’re standing still.

But hop in a jet plane and you can chase your own sound waves so they appear to stand still.

The apparent velocity of an object - or a wave - depends in a simple way on the velocity and direction of motion of the observer.

This is Galilean relativity - after Galileo obviously.

You calculate what velocity everyone observes using the Galilean transformation, which is part of the foundation of Newton’s mechanics.

I were a smug 19th century physicist, I’d wanna hope Galilean relativity is right OK, so if light is a classical wave in some medium then we should see changes in the apparent speed of light depending on our direction of motion.

If you move relative to the aether then there should be a sort of aether wind - like when you stick your hand out of a moving car.

Light should have a fixed velocity relative to that wind, but for you it should be carried by that wind - faster moving behind you than ahead of you.

That is what Michelson and Morley sought to measure.

All they needed was a measuring device capable of traveling at a significant fraction of the speed of light, which in the 1890s I guess may have seemed challenging.

But why build a fast-moving lab when you already have a fast-moving planet?

Earth hurtles around the Sun at 30 km/s - that’s only one one hundredth of one percent of the speed of light - but it should produce a change in the speed of light of the same degree.

Michelson and Morley set out to measure that.

Thus was invented the ingenious Michelson-Morley interferometer.

In it, a beam of light is split in two by a semi-reflecting mirror and sent along two paths at right angles to each other.

The split beams bounce back and forth multiple times before being brought back together again.

And just like in Thomas Young’s double-slit experiment, these beams would then interfere with each other to produce the bright and dark bands of an interference pattern.

Once set up, a couple of things could change that interference pattern.

One would be changing the length of either interferometer arm.

That would cause the peaks and troughs to line up in different places.

Changes in length quite a bit smaller than a single wavelength of light would produce observable shifts in the fringe pattern.

And this is exactly the method that LIGO uses to detect gravitational waves.

The other way to get a shift in the interference pattern is by changing the relative speed of light along the two arms.

That would cause the wave pattern in one arm to lag behind the other, leading to a similar shift in the pattern.

So let’s say you set up the interferometer and see your interference pattern.

To start with you wouldn’t know whether the arms were exactly equal lengths or whether the aether wind was blowing one way or the other - but actually none of that matters.

You’d see the bright bands of the interference pattern at the locations where the beams happened to give constructive interference.

That interference pattern should stay fixed as long as the path lengths stay fixed and as long as the relative speed of light between the arms stays fixed.

But if the speed depends on the direction of the aether wind, then just rotate the interferometer by 90 degrees.

If the speed of light was faster in one arm than the other due to the aether wind, then the speed difference gets flipped when the device is rotated.

The interference pattern should shift to reflect this.

That is the Michelson-Morley experiment.

For extra genius points, they floated the whole setup in a tub of frictionless mercury so it would maintain a fixed orientation despite the earth rotating beneath it.

It was all so beautifully designed, and it should have detected the aether wind - if it existed.

Of course you know what’s coming - on rotating the device the interference pattern didn’t budge.

OK, so maybe the aether happens to be moving exactly with the earth - fine, try it 6 months later when the earth is moving in the opposite direction - still no pattern movement.

This may be the most famous null result in the history of physics.

There appeared to be no luminiferous aether - at least, not one that resembled a classical medium for wave propagation.

The speed of light appeared to be independent of the motion of the observer.

That contradicts Galilean relativity, and so revealed a crack in the sacred Newtonian mechanics.

This directly inspired Hendrik Lorentz to derive his Lorentz transformation - an update to the Galilean transformation that now allowed the speed of light to remain constant, no matter your velocity.

And that brings us to Einstein.

The constancy of the speed of light and the Lorentz transformation are fundamental to his special theory of relativity, which was published 8 years after the Michelson-Morley experiment.

Now Einstein’s starting motivation seems to have been the fact that the Maxwell equations are also a little broken under Galilean relativity.

But the death of the aether gave Einstein two things - it inspired the Lorentz transformation, critical to both special relativity and then his general theory of relativity that ultimately overthrew Newton’s theory of gravity, and, according to Einstein, it revealed the cracks in the foundations of physics that allowed his ideas to be accepted.

It’s commonly stated that the aether is disproved - but let’s be clear.

The aether as a classical medium for light is dead.

However the general concept of the aether sort of has an afterlife.

Einstein talked about the “new aether” as the medium of the gravitational field, and which we now think of the fabric of spacetime.

Paul Dirac suggested that a “particulate aether” could explain the near vacuum state of spacetime in which quantum physicists believe particle pairs are quickly born and destroyed.

In retrospect, Descartes seems prescient: empty space is not really empty.

Its flowing fabric is the source of the gravitational field, and its full of quantum fields.

And what about Albert Michelson?

Well it seems he could never bring himself to accept his own result.

He appears to have believed in the luminiferous aether until the end.

When Einstein visited Michelson on his deathbed in 1931 Michelson's daughter begged Einstein “not get him started on the subject of the aether.” Michelson may have mourned the death of the idea, and the passing of the sensible, down-to-earth world of 19th century physics.

But that death helped spark the revolutions of relativity and then quantum theory that revealed a much weirder, but still totally luminiferous space time.