ABHAY ASHTEKAR remembers his reaction the first time he saw the universe bounce. "I was taken aback," he says. He was watching a simulation of the universe rewind towards the big bang. Mostly the universe behaved as expected, becoming smaller and denser as the galaxies converged. But then, instead of reaching the big bang "singularity", the universe bounced and started expanding again. What on earth was happening?Ashtekar wanted to be sure of what he was seeing, so he asked his colleagues to sit on the result for six months before publishing it in 2006. And no wonder. The theory that the recycled universe was based on, called loop quantum cosmology (LQC), had managed to illuminate the very birth of the universe - something even Einstein's general theory of relativity fails to do.
LQC has been tantalising physicists since 2003 with the idea that our universe could conceivably have emerged from the collapse of a previous universe. Now the theory is poised to make predictions we can actually test. If they are verified, the big bang will give way to a big bounce and we will finally know the quantum structure of space-time. Instead of a universe that emerged from a point of infinite density, we will have one that recycles, possibly through an eternal series of expansions and contractions, with no beginning and no end.
LQC is in fact the first tangible application of another theory called loop quantum gravity, which cunningly combines Einstein's theory of gravity with quantum mechanics. We need theories like this to work out what happens when microscopic volumes experience an extreme gravitational force, as happened near the big bang, for example. In the mid 1980s, Ashtekar rewrote the equations of general relativity in a quantum-mechanical framework. Together with theoretical physicists Lee Smolin and Carlo Rovelli, Ashtekar later used this framework to show that the fabric of space-time is woven from loops of gravitational field lines. Zoom out far enough and space appears smooth and unbroken, but a closer look reveals that space comes in indivisible chunks, or quanta, 10-35 square metres in size.
In 2000, Martin Bojowald, then a postdoc with Ashtekar at the Pennsylvania State University in University Park, used loop quantum gravity to create a simple model of the universe. LQC was born.
Bojowald's major realisation was that unlike general relativity, the physics of LQC did not break down at the big bang. Cosmologists dread the singularity because at this point gravity becomes infinite, along with the temperature and density of the universe. As its equations cannot cope with such infinities, general relativity fails to describe what happens at the big bang. Bojowald's work showed how to avoid the hated singularity, albeit mathematically. "I was very impressed by it," says Ashtekar, "and still am."
