IN EVERY time and in every culture, there have been stories of creation – how the universe began. In our time the story is that of the big bang, the incredibly hot, dense state from which the universe expanded. But this is not the whole story.
“To believe that the big bang is the first moment of time is more religious mysticism than science,” says Lee Smolin of the Perimeter Institute in Waterloo, Ontario, Canada. Smolin is not suggesting that the big bang never happened: astronomical observations and Einstein’s general theory of relativity leave little doubt that it did. But they don’t explain why it happened or what may have come before.
Martin Bojowald of the Max Planck Institute for Gravitational Physics in Golm, Germany, has come up with a possible solution to this problem. He has taken a theory called loop quantum gravity, first proposed by Smolin, which ascribes a complex quantum architecture to space, and used it to peer into the core of creation. What he found there was not a beginning at all, but rather a portal to a universe that came before, a universe that, as it turned out, was completely inside out.
Advertisement
The notion of the big bang arises from Edwin Hubble’s discovery in the 1920s that the universe is expanding right before our eyes. Cosmologists naturally followed the story backwards, and they now conclude that some 14 billion years ago, all the matter in the universe must have been crammed into a single, dimensionless point.
But Einstein’s equations of general relativity can’t describe what happens at this point, called a singularity – let alone what could have come before it. They can only predict that at the singularity, space warps beyond repair. So, while relativity can give us a comprehensive account of our cosmic beginnings, it cannot tell us what caused the big bang to happen.
Fortunately, general relativity is not the only theory in the cosmos. When it comes to the very small – the realm of atoms and electrons and quarks – quantum mechanics reigns. Both theories allow physicists to understand the world before them with incredible precision, and in that sense both theories are “right”. But there’s a catch: they completely contradict one another in their descriptions of the basic structure of space itself. Unlike the active, malleable fabric of general relativity, the space of quantum theory is a fixed and passive backdrop against which elementary particles dance. It can’t be both.
This problem becomes particularly pronounced when dealing with space in its most extreme condition – at the singularity that lies in the belly of the big bang. There, the intense gravity requires a description from general relativity, while the incredibly small volume brings quantum mechanics to bear. Physicists need a new theory of quantum gravity that can reconcile both these worlds.
For years string theory, which says that elementary particles have a structure that resembles tiny loops of string, was the only contender. But in the 1990s Smolin and colleagues developed an alternative – the theory of loop quantum gravity. And LQG can cope with the singularity, says Bojowald.
Smolin, working with a small group of physicists including Ted Jacobson, Abhay Ashtekar and Carlo Rovelli, developed LQG by rewriting the equations of general relativity in a quantum framework. The new framework described space as if it were made up of tiny loops a mere 10−35 metres in diameter. These loops, the team suggested, are the very building blocks of space. Understanding the structure of the universe became a matter of understanding how the loops link together. The web-like networks of the theory, called spin networks, encode on a two-dimensional map all the information needed to construct a three-dimensional quantum space. So, for example, each vertex on the web is taken to represent a volume in space, while each line represents an area. According to the theory, both the volumes and areas can only increase in discrete steps.
But how can this web-like pattern tell us anything about the origin of the universe? The key is that the passage of time can be represented as a function of the volume of the loop universe, something that is possible in other theories that attempt to construct space-time from individual quanta (èƵ, 4 October 2003, p 36). Since volume is made up of individual loops, time also hops along in discrete jumps. As Bojowald followed cosmic evolution backwards, the volume grew smaller and smaller until it reached the big bang itself. And that was where things got really interesting.
In the quantum network, areas and volumes are finite and indivisible. There cannot be a singularity, because space just cannot get that small. And since the theory no longer broke down, Bojowald could continue following time back beyond what had previously been viewed as the beginning.
There he found an entire universe on the other side of time zero, a looking-glass world where expansion is replaced by contraction – and a big crunch reflects our big bang. “When we follow the universe beyond the classical singularity, we can do so forever, until we reach negative infinity,” Bojowald explains. “Therefore, the universe does not have a beginning. It has always existed.”
The looking-glass universe would have looked very similar to the one we know, with all the same laws of physics. Except, that is, for one bizarre thing: it was inside out. Because Bojowald measured time in volume, he found that as he ventured into negative time, the orientation of space flipped so that its volume and other spatial quantities became negative.
Cosmic event
Bojowald likens the spatial flip to a balloon. If we idealise a balloon as a perfect sphere, and then deflate it, it will collapse to a single point. If we then imagine it continuing to collapse even further, all the points will pass through one another until the balloon reinflates, with the inside of the sphere now on the outside. Any object in the balloon would be reversed left to right, and that is just what happens in the universe before the big bang.
So would this make a difference? “This would be mostly imperceptible,” says Smolin, “as most properties of the universe and most of the fundamental laws are symmetric under the exchange of left for right.” But there are a few exceptions. Some reactions involving neutrinos and kaons are asymmetrical, because the reactions’ products are preferentially spinning in one direction rather than the other. In the universe on the other side of the big bang looking glass, those directions are reversed. So although in Bojowald’s model the big bang no longer marks a beginning of time, it remains a vitally significant event in cosmic history: the time when space flipped over, and left and right reversed. The universe has an eternal past, but all the details of the big bang evolution that have been worked out by cosmologists on this side of the big bang still apply.
The theory also provides a way to explain why the early universe apparently underwent the brief but extraordinarily fast period of expansion known as inflation. As soon as the universe flips from inside out to the right way round, it starts expanding. But because volume is made up of individual loops, it cannot grow smoothly. Instead, it tends to jump stepwise, and this creates a kind of outward pressure on the universe. This, it turns out, is just what is needed to get the universe inflating, and removes the need to introduce any arbitrary fields like the inflaton, without which inflation cannot be explained in standard models of the big bang.
The same scenario could solve another problem in general relativity: revealing what happens in the dark depths of black holes. Here, too, singularities resist any description in terms of classical general relativity. Relativity says at most that time stops at the centre of a black hole, and light rays halt in their tracks. In Bojowald’s picture, the space of the black hole may invert itself and open up into an entirely new inside-out universe. Smolin, for one, has long believed that black holes in our universe hide umbilical cords to a host of baby universes.
Smolin and Bojowald’s ideas remain controversial among the majority of physicists. Most, like Sean Carroll of the University of Chicago, believe that string theory is closer to the right track than LQG. “The best evidence is the incredible fruitfulness of the string theory idea,” Carroll says. From the idea of strings, physicists have been able to derive all the symmetries of space-time and the forces we see.
String theorists even have their own ideas on what caused the big bang: a collision between membranes or “branes” existing in higher dimensions (see “Can String theory solve the singularity puzzle?”). In their picture, too, the universe has always existed.
But Smolin points out that string theory cannot explain one important feature of nature: space. In LQG, general relativity – and with it, the notion of space – is built in from the start. String theory, by contrast, takes quantum mechanics as its starting point, and so the strings wiggle against a fixed spatial background that is unaccounted for by the theory.
Supporters of Bojowald’s approach say this means that applications of string theory to the singularity just don’t work as well. Too many assumptions are involved both in what strings are and in how they behave. Smolin finds Bojowald’s approach far more elegant. “Martin’s work is clean,” he says. “The only assumptions are the principles of general relativity and of quantum mechanics.”
Elegant calculations are one thing, but what about experimental evidence for LQG and the looking-glass universe? At the moment there is none, but that may change within a few years when NASA’s Gamma Ray Large Area Space Telescope (GLAST), scheduled for launch in 2006, starts getting results.
Giovanni Amelino-Camelia of Harvard University suggests using its data to track gamma-ray photons from billions of light years away. In our everyday lives the effects of loopy space are negligible, but if space is grainy on the smallest scale, as LQG says it is, then the gamma-ray photons will have accumulated a noticeable spread during their billions of years travelling through space. An instrument like GLAST should be able to observe such an effect, and when its measurements are analysed it may turn out that the big bang is just one small piece of a much bigger story

Can string theory solve the singularity puzzle?
String theorists have their own ideas about what came before the big bang – and they do not include a looking-glass universe. String theorist Gabriele Veneziano, for one, has attempted to use the finite size of strings to avoid a singularity, leading him to a universe that has existed forever (èƵ, 3 June 2000, p 24).
And physicists Paul Steinhardt of Princeton University and Neil Turok of the University of Cambridge, UK, proposed a model in which the extra dimensions of string theory are put to cosmological use (èƵ, 16 March 2002, p 26). According to their “cyclic model”, the three dimensions of space we experience actually live on the surface of a brane (short for “membrane”) that is floating in an additional dimension. Another brane hovers a microscopic distance from ours, and every few trillion years the two branes collide. What we perceive as the big bang, the model says, is just one of these collisions.
“The idea that underlies the cyclic model,” explains Steinhardt, “is that what appears to be a classical singularity in the usual 3-space plus one time dimension corresponds to a collision between branes in an extra dimension. There is a singularity in the sense that an extra dimension is disappearing, but it’s not our three dimensions that are disappearing.”
This cycle, in which the branes move toward one another, collide, and then move apart again, can repeat over and over again eternally, which means the universe may never have had a beginning. The model also makes testable predictions. In the standard model of cosmology, inflation would have stirred up gravitational waves whose imprints should still be discernible in the cosmic microwave background. The cyclic model, however, doesn’t need inflation, so it predicts no such primordial gravity waves. Experiments that look for gravitational waves may be able to distinguish between the two.