Marcus Chown. Catching the cosmic wind

02 April 2005
From New Scientist Print Edition.

In search of the Ether

TWO hundred thousand dollars seems a small price to pay. If the most famous null result in science was right, at least we'll finally be sure. And if it was wrong, then Einstein is no longer king of the universe. No wonder Maurizio Consoli is keen to get started. This experiment could be dynamite.

Consoli, of the Italian National Institute of Nuclear Physics in Catania, Sicily, has found a loophole in the 19th-century experiment that defined our modern view of the universe. The experiment established that light always travels through space at the same speed, whatever direction it is heading in and whatever the motion of its source: there is no way to put the wind in light's sails.

Einstein used this foundation to build his special theory of relativity, but it seems his confidence may have been premature. Consoli's paper, published in Physics Letters A (vol 333, p 355), shows that there might be a wind that blows in light's sails after all: something called the ether.

Until just over a century ago, most physicists believed that this ghostly substance filled all of space. Their reasoning was straightforward enough: the prevailing opinion was that light travelled as a wave, just like sound. And just like sound waves, light waves would need something to move through. Light, they believed, was the result of oscillations in the ether.

In 1887 Albert Michelson, who had recently produced the best-ever measurement of the speed of light, teamed up with Edward Morley to design an experiment to detect this ether. If it filled all of space, they reasoned, then all celestial bodies must have some velocity relative to it . So someone standing on Earth and facing in the direction of its motion through space would have an "ether wind" rushing past their face. According to this thinking, a light wave travelling with the ether wind would seem to move faster than a light wave heading into it. And Michelson and Morley set out to prove this.

They set up an ether detector at the Case Institute of Technology in Cleveland, Ohio. Their "interferometer" measured the speed of two light beams travelling in perpendicular directions. Any motion relative to the ether would produce a difference in the speed of the light travelling down these two arms. The pair then recombined the perpendicular light beams in a telescope eyepiece, where any speed difference would show up in a striped pattern of interference fringes. To make sure they would maximise the effect of a speed difference, Michelson and Morley watched the fringes while they rotated their apparatus by 90 degrees; if the fringes then shifted their position in the eyepiece it could only be the result of the Earth's speed through the ether.

The Earth is travelling at 30 kilometres per second around the sun, not to mention racing around the centre of the galaxy. So Michelson and Morley reasoned the ether wind should reduce the speed of light travelling in the same direction as the Earth by at least 30 kilometres per second - 0.01 per cent of the speed of light. The experiment was easily sensitive enough to detect an effect of this magnitude. To the disappointment of the experimenters, it did not, and reluctantly they accepted the conclusion that there is no ether.

Many similar experiments have been performed since then: in every case the official conclusion has been the same. But not everyone has swallowed the story. In 1902, William Hicks published a study of the Michelson-Morley experiment, and claimed the results supported the existence of an ether wind blowing over the Earth at 8 kilometres per second. Although the pair had carried out their observations over a number of days, they had then averaged out their results as if the experiment's orientation to a prevailing ether wind had not changed. Hicks pointed out that this would cancel out any effect. Some years later Dayton Miller, a former colleague of Michelson's, reworked the Michelson-Morley measurements and also came out with a speed for the ether wind of about 8 kilometres per second. He then redid the experiment with Morley and obtained the same result, but this time with a much smaller error range.

In 1921 Miller took the result to Einstein, who thought there was probably some mistake. He suggested that Miller's result might be explained by slight temperature differences in the apparatus. "Subtle is the Lord, but malicious he is not," Einstein declared. So Miller repeated the experiment 1800 metres up, on the snowy summit of Mount Wilson in California. "He got exactly the same result as Michelson and Morley in the warm basement of the Case Institute," Consoli says.

According to Consoli, many interferometer experiments carried out over the past century have shown a measurable ether wind. "The textbooks say the experiments produced null results," he says. "The textbooks do not tell the truth."

And that's why he wants to carry out a definitive test, an adaptation of the most recent ether-detecting experiments (see Diagram). These used two sapphire cavities oriented at right angles to each other. Laser light bounces back and forth inside the cavities; the size of the cavity and the wavelength of the light means they resonate at an extremely precise frequency. Left to run for over a year, the existence of an ether would create a difference in resonance frequency between the two cavities. That's because the Earth's motion around the sun, and thus the changing orientation of the ether wind, would change the speed at which light moved in the cavities. When this was done at Humboldt University in Berlin, Germany, the frequency difference at the end of the run - less than 1 hertz - was within the experiment's margin of error: the ether was denied again (Physical Review Letters, vol 91, p 20401).

But hold on, Consoli says. So far, these sapphire cavity experiments have all been performed with the light passing through an extremely high vacuum. Consoli and his colleague Evelina Costanzo are now proposing to repeat the Humboldt University experiment with the cavities filled with a relatively dense gas, such as carbon dioxide. This will slow the light, and that could make a crucial difference to the outcome.

Consoli says any Michelson-Morley type of experiment carried out in a vacuum will show no difference in the speed of light in different directions, even if there is an ether. But he points out that some theories, such as the electroweak theory and quantum field theory, suggest that light could appear to move at different speeds in different directions in a medium such as a dense gas. The size of the effect would depend on the refractive index of the medium - and any motion relative to an ether.

With the Earth careering through space into an ether wind, light in one arm of the gas-filled interferometer would travel faster than light in the other, "just as was seen in the classic non-vacuum experiments of Michelson and Morley and others," Consoli says. The 8-kilometres-per-second result for the speed of the ether wind relative to the Earth came from using an interferometer filled with air, he points out. Experiments performed using helium-filled interferometers have obtained 3 kilometres per second and those using a "soft" vacuum 1 kilometre per second. The more rarefied the medium that light is shone through, the smaller the effect of the speed of the Earth's movement relative to any ether.

The cavity experiments will be even more sensitive to this. If there is an ether, Consoli predicts there will be a large jump in the frequency difference between the cavities - perhaps by a factor of 10,000, or even 100,000. The experiment will cost about $200,000 to set up and perform, but it will be worth it. "This is the crucial experiment," he says. "If such an effect is not seen, we will have closed the last experimental window."

It is not a straightforward experiment to perform, though. Experimenters have managed to produce a laser frequency stable enough to carry out experiments for hundreds of days only by cooling the cavities to close to absolute zero. If a gas is introduced at these temperatures it will freeze: it's going to take quite some ingenuity to overcome the problem. Nevertheless, a group of physicists at Humboldt University are considering taking on the challenge. "There is a good chance we will do the experiment," says Achim Peters, one of the group.

It's going to be a much-watched piece of lab work. "If someone does do it, I will be very interested in the result," says Holger Müller of Stanford University, California, who was involved in laser cavity experiments at Humboldt before moving to the US. Müller admits that a positive result would have profound implications for physics. For a start it would mean that one of Einstein's contemporaries, Hendrik Lorentz, has been denied proper recognition. Lorentz, not Einstein, would have to be credited with the definitive theory of relativity (see "Einstein the usurper").

Another implication, pointed out by Consoli, is the possibility of signalling at speeds that seem faster than light. In special relativity this is forbidden, because an object moving faster than light would appear to some observers as moving backwards in time. This can make an effect precede a cause, violating the principle of causality. If there is an ether providing the universe with an absolute reference frame, or "preferred" frame, faster-than-light signalling can happen: the view of events in the preferred frame is the correct one and all other frames must adjust their interpretation of what they see to fit in with it. Consoli suggests that, if there is a preferred frame, that might explain why physicists are able to use quantum entanglement to establish a link between subatomic particles, then have them influence each other instantaneously no matter how far apart they might be (New Scientist, 27 March 2004, p 32). Einstein famously rejected this phenomenon as impossible - he called it "spooky action at a distance" - but experiments have since shown it to be an entirely real and repeatable effect. But, Consoli points out, the effect wouldn't be instantaneous - and thus spooky - if measured in the correct reference frame.

Proof of the ether's existence would also mean that one of the most fundamental equations in physics needs adjusting. The Dirac equation is our best description of how light interacts with matter - it shows how the laws of relativity affect the properties of individual electrons. It is crucial because the passage of light through any medium other than a vacuum depends on the interaction of light with electrons of that medium. It is this interaction that slows light down and gives the medium its refractive index. As it stands, the equation does not allow a difference in the speed of light beams moving through the same medium in different directions. "If the equation broke down, it would be very big news indeed," Müller says.

He points out that the Dirac equation has already passed some stringent experimental trials: this makes him very sceptical that Consoli can truly be onto anything. He believes the experiment is misconceived and that Consoli ought to have applied the Dirac equation to individual electrons in the gas to see what effect there would be, rather than invoking the refractive index of the gas as a whole. Do this, and the argument that the light's speed will be boosted in one particular direction would fall apart, he suggests.

Robert Bluhm of Colby College in Waterville, Maine, also thinks Consoli is on a hiding to nothing. He doesn't even buy the basis of the argument, the problem with Michelson and Morley's averaged measurements. "I think it is safe to say that the textbooks are correct that the Michelson-Morley experiments gave a null result for the existence of an ether drift," he says. And even those who think Einstein's relativity does have limitations are not convinced that such a straightforward experiment can reveal anything. Researchers looking for a quantum theory of gravity - a theory that would unite relativity and quantum mechanics - suspect some form of "substrate" might underpin the universe. But they also suspect that the quantum gravity effects will only show up in experiments that probe matter over extremely short distances or at ultra-high energies.

Well maybe, Consoli says, but we ought to find out for sure. "All we are saying is that these experiments have not yet completely ruled out the possibility of a preferred frame. A small experimental window for its existence is left. We think it is worth investigating that window." And perhaps not even Einstein would argue with that.
Einstein the usurper?

The tenets of special relativity have withstood test after experimental test, so why bother searching for an ether again? Maurizio Consoli of the Italian National Institute of Nuclear Physics in Catania suggests that Einstein's theory could be a special case in a broader theory developed by one of his contemporaries. And the only thing that separates the ideas - the one test that will bestow the crown of king of the universe - is the question of an ether.

Hendrik Lorentz came up with a description of how light travels through space and time before Einstein formulated his special theory of relativity. Lorentz's theory is so similar to special relativity that it has passed the same tests. Indeed in his 1916 work, The Theory of Electrons, Lorentz commented: "Einstein simply postulates what we have deduced."

In Einstein's theory, when two observers look at each other, the intervals of space and time they see between them depend only on their relative velocity. But in Lorentz's view, the effects - and they look exactly the same - originate from the individual motion of each observer relative to an absolute reference: the ether.

Though they sound similar, the two theories are equivalent only in a vacuum, Consoli says. For the views from all reference frames to be equivalent, as special relativity requires, the maximum speed has to be unattainable by anything other than light, and that only in a vacuum. His gas-filled interferometer experiment should tell us whether Einstein usurped Lorentz's throne.