In the second of our Big Science Questions, Sir Martin Rees ponders the origin of the universe, while Martin Ince juggles the pieces of a cosmic puzzle.

Evidence that our universe had a hot dense beginning dates back to 1965, when radio astronomers discovered that space was not completely cold, but had a temperature about three degrees above absolute zero. Indeed, our entire universe is pervaded with microwaves, which are an afterglow of the hot dense initial state. Since 1965, the evidence has firmed, the result being that the extrapolation of cosmic history back to a stage when the universe had been expanding for a few seconds now deserves to be taken as seriously as, for instance, what geologists or palaeontologists tell us about the early history of our Earth - their inferences being just as indirect and generally less quantitative.

By contrast, we are still groping for firm clues about what happened still earlier - in the first tiny fraction of a second - when the primordial material was squeezed to such extreme densities and pressures that experiments offer no firm guide. But within the past few years, cosmologists have developed a consensus about how the universe is now expanding and what its future is likely to be.

In about 5 billion years, the Sun will die, and the Earth with it. We cannot predict what role life will by then have carved out for itself: it could have become extinct, or it could have achieved such dominance that it influences entire galaxies. Such speculations are the province of science fiction, but cannot be dismissed as absurd. After all, it has taken little more than 1 billion years - only a fifth of the Sun's remaining life - for the first multi-cellular organisms to evolve.

ÖйúAƬ

But what will happen in the more distant future? The answer depends on how much the cosmic expansion is being decelerated. Although everything exerts a gravitational pull on everything else, it is calculated that if all the atoms in the universe were spread uniformly through space, they would not, in the absence of an external force, exert enough gravitational pull to slow things down, implying that perpetual expansion is on the cards. But galaxies "feel" the gravitational pull of several times more material than we actually see: most of this material is made of "dark matter". But even taking this into account, there is not sufficient force to slow the universe enough to bring it to a halt: it is therefore forecast to continue expanding. Galaxies will fade from view as they get ever further away and their stars exhaust their fuel.

But in the past two years, evidence has emerged that expansion is not slowing, but accelerating. This implies that, on the cosmic scale, gravity is overwhelmed by some kind of repulsive force. Such a force was hypothesised by Einstein in 1917. At that time, astronomers knew only about our galaxy - not until the 1920s did a consensus develop that Andromeda and similar "spiral nebulae" were separate galaxies, each comparable to our own. It was therefore natural for Einstein to presume that the universe was static - neither expanding nor contracting. He found that a universe could not persist in a static state unless some force contradicted gravity.

ÖйúAƬ

The motivation for his hypothesis became irrelevant after 1929, however, when Edwin Hubble discovered that the universe was expanding - but that does not discredit it. On the contrary, empty space now seems anything but simple: all kinds of particles are latent in it; on an even tinier scale, it might be a seething tangle of strings. From our modern perspective, the puzzle is not why there should be cosmic repulsion, but why the energy and force latent in empty space is not much higher.

Within the past two years, a remarkable concordance has emerged between independent methods for measuring the contents of the universe. It seems that atoms provide only 4 per cent of the mass-energy in the universe. Dark matter contributes 20 to 30 per cent and the rest is "dark energy" latent in space. There seems nothing "natural" about this particular mixture. How did it arise and why is the universe expanding the way it is?

When our universe was an amorphous fireball only a second old, it could be described by just a few numbers: the proportions of ordinary atoms, dark matter and radiation, the expansion rate and so forth. This simple recipe must be the outcome of what happened earlier still, within the first tiny fraction of a second, when conditions became more extreme and unfamiliar. In the first trillionth of a second, each particle would have carried more energy than the most powerful accelerators at Cern, the European Laboratory for Particle Physics, can reach. Ideas about this ultra-early era are still tentative, but nonetheless there has been immense progress.

The most basic mystery is why our universe is expanding and why it is so vast. Analogies with an explosion can be seriously misleading. Bombs on Earth, or supernovae in the cosmos, explode because a sudden boost in internal pressure flings the ejecta into a low-pressure environment. But in the early universe, the pressure was the same everywhere: there was no empty region outside. The most plausible answers involve a so-called inflationary phase, during which the expansion was exponential; the scale doubled, then doubled and then doubled again. Within about 10-36 seconds, it is claimed, an embryo universe could have inflated enough to encompass everything we now see.

The generic idea that our universe inflated from something microscopic is compellingly attractive: it accounts also for why it is expanding - something that is simply an "initial condition" otherwise. It looks like "something for nothing", but that is not really the case. That is because our present vast universe might, in a sense, have zero net energy. Every atom has a positive energy because of its mass - Einstein's MC 2 . But it also has a negative energy because of the gravitational field of everything else. Thus it does not, as it were, cost anything to expand the mass and energy in our universe.

Inflation stretches a microscopic patch until it becomes large enough to evolve into our observable universe. Indeed, it is likely to overshoot, inflating more than necessary. Our universe then ends up being "stretched flat", rather like a part of a wrinkled surface becomes smooth if it is stretched enough.

ÖйúAƬ

Most theorists regard inflation as a beautiful generic concept that they will cling to until something better comes along, and there are intimations that something might - extra spatial dimensions beyond the usual three might lead us to another paradigm. But the details depend on uncertain physics. By observing some features of our present universe, however, we can select among the rival theories. For instance, inflation predicts the properties of the "ripples" that show up as non-uniformities in the background temperature over the sky and are the embryos of galaxies. They are quantum vibrations, generated on a microscopic scale, that have inflated to stretch across the sky - an amazing link between cosmos and microworld.

Some variants of the inflationary universe theory suggest that our big bang was not the only one. This speculation dramatically enlarges our concept of reality, turning the history of our universe into just an episode, one facet, of the infinite multiverse.

ÖйúAƬ

Astronomers are normally mere users of laboratory physics, except where gravity is concerned. Perhaps they can now return the compliment by probing "extreme physics" that cannot be checked in the laboratory. But the crescendo of discovery seems set to con-tinue. Large telescopes can now view objects so far away that their light set out when the universe was only a tenth of its present age. Other techniques can probe back to the first seconds of the big bang.

I would bet reasonable odds that within ten years, we will know what the dominant dark matter is, and other key numbers such as the age of the universe. If that happens, it will signal a great triumph for cosmology: we will have taken the measure of our universe, just as, over the past few centuries, we have learnt the size and shape of our Earth and Sun.

In the longer-term, theorists must elucidate the exotic physics of the earliest stages. The synthesis still eludes us is between gravity and the microworld - between the cosmos and the quantum. Until there is a unified theory, we will not be able to understand the fundamental features of our universe.

The smart money is on superstrings or M-theory, according to which each point in our ordinary three-dimensional space is actually a tightly-folded origami in six or seven extra dimensions. There is still an unbridged gap between this elaborate mathematical theory and anything we can measure, but such a theory might be needed before we understand the beginning of the universe or the nature of the energy latent in empty space.

But cosmology is also the grandest of environmental sciences. Another challenge is to understand the intricate processes whereby a simple fireball evolved into a complex cosmic habitat from which our species could emerge.

Sir Martin Rees is Astronomer Royal and professor of astronomy at Cambridge University.

ÖйúAƬ

<P align=left> Big Science Questions contents page