COUNTLESS times each day inside the world鈥檚 atom smashers, subatomic particles collide head-on and then break up in elaborate dances. A zillionth of a second later they reach more stable states, and then fly away, eventually reaching detectors.
It doesn鈥檛 only happen in experiments. This is also the story of the universe. Particles created in the big bang flew apart, eventually combining into larger particles, atoms, planets, trees and people as the universe expanded. It鈥檚 like what goes on inside an atom-smashing particle accelerator, only the universe is a huge, slow-motion experiment, playing out over billions of years.
Or is it? For years, physicists have used this analogy between atom smashers and the universe as a guiding principle towards the elusive theory of everything. The main reason for this is that the atom smasher idea happens to fit well with string theory 鈥� the theory that replaces subatomic particles with tiny vibrating 鈥渟trings鈥�, and is supposed to describe the whole universe and unify all of physics.
Advertisement
But now the analogy is under attack. Critics are claiming that the universe is nothing like a particle accelerator. On the cosmic scale, they say, particles are not bounded by an enclosed lab, and the experimenters are part of the experiment. Worse than that, the mysterious dark energy that seems to be blowing the universe apart will always prevent us from discovering the ultimate physical truths. If so, the grand, universal goals of physics as we know it are impossible.
But there might be one way to save physics: dark energy could be destined to disappear. That would bring back the older idea that the universe, far from flying apart ever faster, actually behaves more like a slowly flattening Pringles potato crisp (see Graphic). String theory may be able to describe such a universe some day, the critics say, but only if it gets an infusion of new ideas that go beyond the atom-smasher analogy.
Attacking the atom-smasher analogy is radical. Physicists have been extrapolating from the particle accelerator to the rest of the universe since Richard Feynman and Bryce DeWitt鈥檚 first attempts to combine quantum mechanics and gravity in the 1960s. It is based on the idea that, in an atom smasher, experiments have a clean beginning as well as a clean end. The incoming particles enter the interaction region from separate directions, and the outgoing ones leave in separate directions. The symmetry of this process is crucial for particle physics to have some level of the mathematical rigour that physicists want 鈥� it is built into the maths that underlies quantum mechanics.
Physicists have always wanted to apply the same rigour to the whole universe. Granted, the cosmological case has a much messier beginning in the big bang, when the entire contents of the observable universe were packed in a region the size of a grapefruit. But physicists thought they could live with this. And the universe does provide something that is just about like the clean ending of an atom-smashing experiment: it is expanding forever, and so the cosmic debris will have enough room to scatter.
鈥淏ousso鈥檚 conclusion? Physics as we know it is impossible in the universe we live in鈥�
But Raphael Bousso of the University of California, Berkeley, is now arguing that the atom-smasher analogy has other, more serious, problems. In 1998, astronomers discovered that the expansion of the universe is accelerating, instead of slowing down as previously thought. Improved estimates of the distances of remote galaxies showed that a sort of 鈥渄ark energy鈥� seems to inflate the universe, slowly winning over the pull of gravity that would otherwise check the universe鈥檚 expansion.
Nothing is known for sure about the nature of dark energy 鈥� hence its name. But if it turns out that the acceleration will go on forever, atom smashers may become cosmology鈥檚 white elephants.
This is because, in a universe with so much dark energy, no single observer could ever see the outcome of the cosmic experiment. Unlike in an atom smasher, when detectors at 鈥渢he end of time鈥� pick up the results, in the universe, any two regions separated by more than a certain distance would eventually fly away from each other faster than light speed. As no one can ever see an object travelling faster than light, any observers would have a 鈥渉orizon鈥� of visibility, and most drifting particles sooner or later would end up disappearing from view. 鈥淓verything that crosses the horizon is forever outside of the observable universe,鈥� Bousso explains.
At the moment, the universe is not yet large enough for us to experience such a horizon. But if the acceleration continues, in a few billion years the farthest galaxies should start passing light speed and disappearing from view.
The phenomenon would be similar to the event horizon of a black hole 鈥� the surface that marks the point of no return for anything that crosses it, including light. Each observer in the universe would eventually be completely surrounded by their own event horizon, as if the universe beyond it were an inside-out black hole.
Buzz of energy
The horizon problem could be fatal to the whole idea of measuring things in fundamental physics. Stephen Hawking showed in the 1970s that the horizons of black holes radiate energy 鈥� the so-called Hawking radiation. If the universe blows itself apart and all observers become shut inside their horizons, they will all start to see a buzz of energy coming from their horizon. This buzz will make it impossible to define physical quantities precisely. In the bath of radiation coming from the horizon, particles will never cool down to their lowest-energy state, no matter how long we wait. Even the most basic physical quantities, such as the mass of an electron, would be subject to a sort of fuzziness, because they could not be measured with arbitrary precision. Bousso鈥檚 conclusion is stark: physics as we know it is impossible in the universe we live in.
The clash between the atom-smasher analogy and dark energy has had a few researchers worried enough to try to think of ways around its horrible conclusions. Willy Fischler of the University of Texas at Austin and Tom Banks of the University of California, Santa Cruz, suggested in 2001 that an eternally inflating universe might still work like an atom smasher, but in an approximate form. Their proposal was to extrapolate to all of space-time the outcomes one observes in a local region and for finite time intervals. Physics would be possible, but it would never be exact, only a kind of 鈥済uesstimate鈥� ().
Meanwhile, Ed Witten of the Institute for Advanced Study in Princeton, New Jersey, proposed that precise physical quantities could still exist, but would only be knowable to abstract observers living outside the universe. They would not be limited by a horizon (). It鈥檚 a nice idea, if you happen to be conveniently situated outside the universe. But for today鈥檚 physicists, it鈥檚 depressing news: the ultimate physical truths would still remain inaccessible to people living inside the universe.
鈥淲e are like little kids in the forest who are walking and trying to find their way鈥�
Still, there might, just might, be a way around this. Leonard Susskind of Stanford University in California believes that after a random, and possibly enormous, amount of time, any inflating universe will naturally change into a universe with no dark energy at all. Gravity would then pull against the expansion of the universe, and the expansion would start slowing down instead of accelerating. Susskind admits that this is speculation, like much of this young field: 鈥淭his is not at the stage of mathematical rigour that we would like to have.鈥� But he says, 鈥淔or my money, the evidence is very strong.鈥�
If Susskind is right, the eventual state of the universe would be what cosmologists call an 鈥渙pen鈥� universe. Before the 1998 discoveries of accelerating galaxies, that鈥檚 exactly what our universe seemed to be like already: infinitely large, eternally expanding 鈥� although at a slower and slower pace 鈥� and curved like a Pringles crisp. Not that an observer looking at the universe from outside would see a giant starry Pringle hanging in space: the Pringle shape is just a way of imagining how the three dimensions of space could curve in an imaginary fourth dimension, just as a two-dimensional surface can curve into the shape of a Pringle in the third dimension. Witten says that, if the universe is indeed open, it would give a 鈥渘atural explanation鈥� of why physics seems so hard to formulate in a universe with dark energy. We don鈥檛 yet have a final 鈥渢heory of everything鈥� because, according to the rules of physics, there is no such thing as a final theory in an eternally inflating universe. The accelerating expansion we currently observe is just a sign of an adolescent universe going through a teenage phase. We shouldn鈥檛 worry too much, because it won鈥檛 last.
In a recent paper, Bousso argued that the open universe would make things much easier, because it does not suffer from the technical problems that make physics impossible in dark-energy universes. In an open universe, he says, there are no horizons limiting visibility, so it should be possible to define precise physical quantities, and in principle to calculate the history of the universe.
Nobody knows how to do that yet, but Bousso believes that the answer can only come from a radical overhaul of string theory. At the moment, string theory is basically a fancy version of particle physicists鈥� description of an atom-smasher experiment. Both traditional particle physics and string theory can only handle finite numbers of particles in a finite region of space. But for all we know, the open universe could be filled with an infinite amount of stuff, extending in all directions. Also, in particle physics, the experimenters are 鈥渙utside, looking in鈥� 鈥� they are not part of the system they are trying to describe. 鈥淲hat we do in cosmology is the opposite: we are inside, looking out,鈥� Bousso says (Physical Review D, DOI: 10.1103/PhysRevD.71.064024).
Shape-shifting universe
Perhaps the Pringle universe can come to the rescue. If space-time in our universe is curved like a Pringle, and eternally expanding, its geometry will gradually become flatter and flatter 鈥� as matter gets more and more spread out, it will curve space less and less. At the moment, string theory simply cannot handle this. The reason is that string theory is 鈥渂ackground dependent鈥�, meaning that all calculations depend on the geometry of the space-time in which they are performed. If the shape of the universe is changing, string theory needs to adapt too.
String theory, like Einstein鈥檚 general relativity, represents gravity as the bending of space and time. But while Einstein鈥檚 theory can explain the change in the bending of the whole universe, string theory can so far account only for localised bumps. On the cosmic scale, string theory requires that all energy and matter be concentrated in a finite 鈥渂ackground鈥� region of an infinite universe. Each time physicists consider a different geometry of space-time, they need a new flavour of string theory, says Bousso. 鈥淚n the backgrounds that string theory does know how to deal with, the fundamental formulation of the theory looks apparently very different,鈥� he says. 鈥淭hat bothers us, because it makes it look like at the end of the day there is a different theory for every background.鈥� Only a background-independent string theory, Bousso says, would truly describe the universe.
Bousso鈥檚 work is 鈥渆xtremely thought-provoking鈥� and helpful in delineating the problem, says Stephen Shenker of Stanford. But Shenker also believes that the mathematical formalism that describes the atom-smasher experiments should not be written off quite yet. 鈥淚t plays an essential role in the physics that we do, so my guess is it will not be abandoned.鈥�
But Abhay Ashtekar of Pennsylvania State University in University Park believes that string theorists have been attached to the atom-smasher formalism for way too long, in part because they come from a particle physics training. 鈥淚t鈥檚 a bit like saying, 鈥業f my only tool is a hammer, then every problem is a nail鈥�,鈥� says Ashtekar. 鈥淚t鈥檚 not always the case that techniques from one field are appropriate and successful in another field.鈥�
Bousso, Susskind and their colleague Ben Freivogel of Stanford are now concentrating their efforts on formulating string theory in the open universe. The open universe gives some clues about certain laws that a modified string theory should obey. In an open universe, for example, all particles will eventually stop interacting and will reach their lowest-energy state. Also, they should scatter out from interactions in a uniform way. Maybe these ideas will point physicists to the correct, background-independent version of string theory.
Others, such as Fischler, stress that it鈥檚 impossible to say whether or not Susskind is right and the dark energy phase of the universe will eventually turn into the slowing expansion of the open universe. 鈥淲e don鈥檛 really know,鈥� Fischler says. 鈥淲e are like little kids in the forest who are walking and trying to find their way.鈥�
But for Susskind and others, there is only one way to do physics: optimistically. The Pringle universe may not be entirely convincing on the current observational evidence. But it鈥檚 the best chance we鈥檝e got.
