April 5 2000

One of the 20th century's great monuments, The General Theory of Relativity, may soon be debunked by a lecturer. Nigel Hawkes reports

© João Magueijo says that the speed of light varies

Is Einstein about to be dethroned?

João Magueijo says that the speed of light varies

A whiff of heresy is in the air. There are very few beliefs in science that have the force of doctrine, but among them is the conviction that the speed of light is fixed. It forms the rock upon which Einstein built one of the 20th century's most enduring monuments, the General Theory of Relativity.

But in science nothing is sacred, and several leading theoretical physicists are now toying with the delicious notion that the speed of light may not be constant after all - but may actually have varied during the 15 billion years the Universe has existed.

To some, it remains anathema. "It took us a year to get the first paper published in Physical Review," says Dr Joâo Magueijo, a young Portuguese physicist, now a lecturer at Imperial College in London. "The scientists who were asked to comment just couldn't accept it. Finally the editor stepped in and it appeared."

At the age of 11, Dr Magueijo was inspired by The Evolution of Physics, a slim volume by Einstein and Leopold Infeld. A dog-eared copy, in Portuguese, lies on his desk at Imperial College.

"The best book ever written about physics," as he calls it, set him on a course that took him to the University of Lisbon, then Cambridge as a research fellow at St John's, and now London. But if he is right, Einstein's conviction that light travels at a fixed and unalterable speed is about to be dethroned.

Dr Magueijo and colleagues with whom he has worked - Dr Andy Albrecht, of the University of California at Davis, and Professor John Barrow, of Cambridge - were not quite the first to have this idea. Dr John Moffat of the University of Toronto had floated it, but they were unaware of his work until he complained it hadn't been properly acknowledged in the Physical Review.

Of course, nobody would make such a radical suggestion unless there were something amiss with existing cosmological theories. The Big Bang and its offspring, inflation, are extremely well-established. They can explain a lot of what we see when we look into the great beyond. They claim that the Universe began from a single point, and has been expanding outward ever since. Simple as this seems, it raises some puzzling dilemmas. If the speed of light is fixed and not infinite, it means that the Universe always has an edge beyond which we cannot see, simply because light has not had time to reach us from that edge. This "horizon" at present lies some 15 billion light years away.

If we run the expansion backwards to a time soon after the Big Bang, we find that some parts of the Universe which we can now see would then have lain beyond the horizon. This is the puzzle. The Universe we can see is homogenous, the same temperature and density everywhere.

Yet things are only the same temperature if they are in contact with one another, and a finite speed of light means that such contact was impossible early on. So why is the Universe homogenous? Inflation provides an answer by saying that all the visible Universe came originally from a small area that expanded - for a brief period - very much faster, smoothing out unevenness.

But there is no real explanation for why this should have happened, nor any way of confirming that it did, since it is now long in the past. "There is too much that is ad hoc about it," says Dr Magueijo.

Equally tricky is explaining why the Universe seems to be so exquisitely balanced. The amount of matter and energy it contains seems to lie just on the cusp between a Universe which would expand for ever, and one which would stop expanding and contract again into a Big Crunch. Cosmologists call such a Universe "flat".

For this to have happened, the initial conditions, such as temperature, would have had to be exactly right - for which there is no good reason. Some argue that this question is irrelevant because if it hadn't been so, we wouldn't be here to worry about it anyway. The Universe, and life, could never have evolved. But most find this explanation rather feeble.

There are yet more problems. One is the evidence, gathered in the past few years by two teams studying supernova explosions, that the expansion of the Universe seems to be speeding up under the influence of an anti-gravitational force that pushes things apart.

Einstein believed such a force was necessary because in the static Universe he depicted, it would be needed to prevent gravity dragging everything together. He put it in his equations as the "cosmological constant" and gave it the Greek letter lambda. When, rather late in the day, he was convinced by Edwin Hubble's evidence that the Universe was actually expanding and lambda was not needed, he called it his greatest blunder.

The supernova evidence has revived lambda, but that in turn raises a new difficulty. "In any universe with reasonable starting conditions, lambda would quickly come to dominate," says Dr Maguiejo.

"The Universe would be thrown apart so quickly each galaxy would soon be isolated from all the others. But this hasn't happened. We need to know why."

All these problems can be solved by the simple assumption that the speed of light was once much greater than it is today. It solves the horizon problem because faster light would have enabled all parts of the Universe to remain in touch. And it solves the lambda problem because the energy in lambda depends on the speed of light.

"A sharp drop in the speed of light soon after the beginning of the Universe's expansion keeps lambda subservient," Dr Maguiejo says. It also provides a simpler explanation of why the Universe is flat, no longer requiring precise starting conditions.

There is just a glimmer of astronomical evidence that the theory may be right. By looking at quasars at different distances away from us - and hence at different times in the past - Professor Barrow and Dr John Webb at Cambridge, with colleagues in Australia and the US, have found a hint that the speed of light may indeed have changed.

"The most distant clouds have different spectra," says Dr Webb in an Equinox documentary, Einstein's Biggest Blunder, to be broadcast on April 17. "So either fundamental particles behaved differently in the past or the speed of light differed. We've spent a year trying to work out what we have done wrong and we can't find anything.

"If this is corroborated, it would be evidence that the speed of light did change. That would be a profound result - the first evidence that the laws of physics were not always the same."

Few will accept the theory unless it makes predictions about the Universe that can be tested by astronomical observation. Dr Maguiejo hopes to come up with such a test soon. Meanwhile, he says the "varying speed of light theory" - it will need a snappier name - answers one of the more tantalising questions about the Big Bang. What triggered it?

If the speed of light can vary, then the history of the Universe is a kind of dance to the music of lambda, he says. As our Universe ages, lambda comes to dominate, forcing it apart until galaxies form island universes, cut off from the main. At some point this process can suddenly turn energy into matter.

In other words, it sets the scene for a fresh Big Bang, or a series of bangs, which will start the whole process again. If so, the Universe keeps rising like a phoenix from the wreckage of the one before - not a one-off, but a continuous process with no beginning and no end.

--from London Sunday Times.