OK, so nobody expects it to actually explain everything. No genius is going to slap their forehead one day and say, 鈥淥h yes, P equals Q squared minus Z. Now it鈥檚 all so clear 鈥� how the mind works, what happened to the dinosaurs, where socks disappear to鈥︹��
The 鈥渢heory of everything鈥� is only meant to explain all the particles and forces of nature. It should reveal, for example, why nuclear forces are strong enough to clamp protons and neutrons together, and why there are exactly three kinds of electron. It should set the constants of nature in stone, and explain the origin of time and space. That鈥檚 all.
So what鈥檚 taking so long? Thirty years have passed since physicists established the 鈥渟tandard model鈥� of particle physics, a set of limited theories that cover the basics of how particles and forces interact. Since then, they have been trying to weave these separate strands into a single fundamental theory. But they keep hitting snags. Some of the world鈥檚 most brilliant minds now suspect that we have been barking up the wrong tree, and only some outrageous new idea will get us further. Others are saying that the laws of the cosmos will never be fully explained by a single theory. Whatever the final answer, it is now clear that the theory of everything is not going to look anything like we thought.
Advertisement
The aim is simple enough. Physicists believe that there was only one force just after the big bang, and as the universe cooled it split into the four forces we now observe: gravity, electromagnetism and the strong and weak forces. The physicists鈥� dream is to find a theory describing this unified force 鈥� and especially to bring gravity into the quantum fold.
The most popular approach to unification is string theory, which holds that all particles are ultimately composed of strings just 10-33 metres long. The strings can vibrate in different frequencies, like notes: one note on a string makes it an electron, another note makes it a neutrino. Still other vibrations can make particles that transmit forces: photons to carry the electromagnetic force, W and Z bosons for the weak force, gluons for the strong force, and even gravitons for gravity. So far so good.
But it turns out that strings are actually too versatile. Way too versatile. String theory also involves curled-up extra dimensions, and these can be rolled up in any of a hundred million ways, each one of which lets the strings vibrate in all sorts of different ways too. 鈥淢ore work has always given more possibilities 鈥� far more than anyone wanted,鈥� says string grandee Edward Witten of the Institute for Advanced Study in Princeton. Nobody knows how many solutions of string theory there are, but it could be more than 10500 鈥� that鈥檚 more than the number of atoms in the universe, squared and then squared again. Surely the worst embarrassment of riches ever known?
It is more than just embarrassing, though. Each of the many solutions of string theory describes a very different universe. Depending on the way the extra dimensions are connected, and how they allow the strings to vibrate, you might produce a world that has 18 kinds of quark, instead of the six we have. Other solutions have no quarks at all. Some solutions have heavy photons, which means that light would be short-ranged 鈥� you wouldn鈥檛 be able to see from one side of an atom to the other. In some, the whole universe is microscopic; in others space might have nine infinite dimensions instead of three.
So does this mean that nothing is set in stone, that all we can hope for from a final theory is a huge range of possibilities?
Several heavyweights of physics certainly think so. Leonard Susskind of Stanford University in California, for example, believes that string theory鈥檚 multiple universes really exist (快猫短视频, 1 November 2003, p 34). He cites a cosmological theory called eternal inflation as supporting evidence: according to this, our patch of the universe is just one among infinitely many bubble universes, all emerging from their own local big bangs. The whole multiverse is in a constant frenzy of reproduction. Each time a new bubble universe forms somewhere, it ends up with a different kind of physics, determined by one of the different string solutions. 鈥淎s time goes on, you populate any corner of the landscape infinitely many times,鈥� Susskind says.
鈥淭heorists can prove M-theory exists, but can鈥檛 write its equations鈥�
Steven Weinberg of the University of Texas at Austin also thinks that all the string solutions are real, but rather than occupying separate regions of space, they could all coexist. The laws of quantum physics allow particles and even objects to be in many different states at once, so why not the universe? We only experience the one state that we are in, but just as Schr枚dinger鈥檚 cat exists in two contradictory states in the famous thought experiment, the many states of the universe are all equally real. 鈥淭hey are there in the same sense that Schr枚dinger鈥檚 cat is alive and dead,鈥� Weinberg says.
To Weinberg and Susskind, accepting this cornucopia of universes solves an old problem in physics: why the laws of nature are so perfectly tuned to allow life to exist (See 鈥淭aming the multiverse鈥�). But it remains deeply unsatisfying to many. 鈥淚 hope that current discussion of the string landscape isn鈥檛 on the right track,鈥� Witten says. 鈥淏ut I have no convincing counter-arguments.鈥�
This surprising result is not the only problem facing the quest for a theory of everything. The theory should explain why space and time exist, yet string theory, for one, has to assume their pre-existence. Its forces and particles need something outside the theory to supply a cosmic container in which they can live and work.
At the moment, string practitioners only have approximate versions of what they believe will be the fundamental theory that makes its own space-time, known as 鈥淢-theory鈥�. As well as strings, M-theory includes a lot of other objects called membranes (or branes for short) with up to nine dimensions. But for now, M-theory exists only as an ideal. Theorists can prove that it exists as a mathematical construction, but they can鈥檛 actually write down its equations and there is no clear route towards doing so. 鈥淲e probably need fundamentally new principles,鈥� says Lisa Randall, a string theorist based at Harvard University. 鈥淚t鈥檚 not hopeless, but it鈥檚 going to require some deep new insight that we don鈥檛 really have.鈥�
So is there an alternative to strings? Lee Smolin, a theorist based at the Perimeter Research Institute in Ontario, believes another approach to unification might work better. He and about a hundred others are working on an idea called loop quantum gravity, in which a network of abstract links and nodes define space-time on the smallest scales, rather like the digital elements in computer animation.
Loopy space-time
LQG does generate its own space and time, and shows that, at the Planck scale of around 10-35 metres, quantum fluctuations rumple the fabric of space-time, crinkling it into a jumble of humps and bumps. But it is still far from being a theory of everything. While LQG seems to work well at the Planck scale of quantum gravity, Smolin and its other proponents have not yet been able to show that the idea also produces the kind of gravity we see on large scales.
And even Smolin, who is scornful of the string landscape, thinks we may well have to accept some kind of multiverse to account for many features of nature. If we have to resort to the multiverse for explanations, that seriously demotes the theory of everything: no longer the absolute ruler, but merely part of a committee.
Is there still a chance of an all-powerful theory of everything? Witten believes that M-theory, once it takes shape, may have a unique solution that fits our universe and nails down all the constants. 鈥淗ope springs eternal,鈥� he says. And Randall has shown that theorists might be able to explain the values of some constants after all. So far, string theory has failed to explain the observed value of the cosmological constant, a number that describes how rapidly the expansion of the universe is accelerating. Susskind thinks this failure supports the idea that all possible string-theory universes exist, each with a different value for the constant, and the value we observe is simply in the narrow range that allows the existence of life and cosmologists. But Randall, working with Shinji Mukohyama of the University of Tokyo, has shown that a simple twist to the standard cosmological equations might produce something near the observed cosmological constant (Physical Review Letters, vol 92, p 211302). A similar approach might explain why we observe just three of string theory鈥檚 spatial dimensions.
鈥淚 suspect there is some right question that we鈥檙e not asking鈥�
Even so, we are less certain than ever what a theory of everything will be able to do, and what it will look like. Many physicists suspect that we need some radical new idea to get us out of this impasse (see 鈥淏right ideas鈥�). Better still would be some actual evidence from experiments 鈥� some hint at what path to take towards a theory of everything.
Such evidence has been hard to come by. 鈥淎ll attempts to go beyond the standard model have predicted things that experiments have not seen,鈥� says Smolin. Theories have predicted that protons should decay, for example, but there have been no telltale flashes of light in the great underground tanks of water built to look for the process. Others imply that every kind of particle should have a heavier mirror image called a superpartner, but none has yet been seen.
Smolin takes that as a sign that we may be pursuing the wrong line of enquiry. 鈥淚f you look back over the last 200 years, every decade or two there鈥檚 a dramatic advance, people always understand something new that couples theory and experiment,鈥� he says. It is now three decades since such a coupling has emerged. 鈥淚 suspect there is some right question that we鈥檙e not asking,鈥� he says.
Randall, though, is not surprised about the lack of experimental confirmation: we鈥檙e now formulating theories that can only be tested in extremely high-energy experiments, she points out. To see the unification of forces directly, we would have to heat up a bit of matter to around 1030 kelvin 鈥� far too hot for any conceivable device to achieve.
The next generation of experiments might take us some way there, though. The Large Hadron Collider at the CERN particle physics lab in Switzerland, for example, will be the most powerful particle accelerator in the world, and physicists are hoping that it will see the superpartners of the ordinary particles as predicted by string theory, and maybe even evidence of hidden extra dimensions. It wouldn鈥檛 be proof of string theory, but it would be encouraging.
Other experiments might also hint at the nature of that elusive final theory. It is just possible that astronomers will see giant strings in space, cousins of the microscopic strings of string theory, which would warp the light from distant galaxies and emit distinctive bursts of gravitational waves (快猫短视频, 18 December 2004, p 30).
And in a few years鈥� time, a vast cosmic-ray telescope called Auger and a new gamma-ray satellite, GLAST, might give us a clue about loop quantum gravity. LQG predicts that very high-energy radiation travels slightly faster than ordinary light, and Auger and Glast will be routinely looking for such deviations. Future telescopes might also see an imprint left by loops on the cosmic microwave radiation. Perhaps, then, we might soon find a reassuring sign that one of these paths to a unified theory is heading in the right direction.
Bright ideas
THE theory of everything is proving elusive. But perhaps the mainstream approaches are heading the wrong way, and what we need are some radical ideas鈥�
Some people have suggested that the universe is a computer, shuffling information in a cosmic program whose output is time, space and particles. In a recent paper, quantum information specialist Seth Lloyd of MIT has shown that a quantum computation automatically has one of the properties needed by any theory of gravity, in that its results don鈥檛 depend on your frame of reference. Lloyd likens black holes in the universe to subroutines in the program: they suck in matter and information and hide it from the rest of the universe, but eventually they evaporate, effectively returning their answer.
Or maybe a deeper theory will have to uproot one of the central physical principles of the past century, the notion of inherent uncertainty at the quantum level. Gravity and quantum mechanics are at odds, and the fault might lie with our understanding of the quantum. Gerard 鈥榯 Hooft of the University of Utrecht in the Netherlands has suggested that at the Planck scale, around 10-35 metres, nature is deterministic after all 鈥� there is no quantum fuzz or other weirdness.
Meanwhile, Lee Smolin and Fotini Markopolou of the Perimeter institute in Waterloo, Ontario, are developing a new version of loop quantum gravity. In this theory, space-time is woven from a mathematical network of nodes and connecting links. In the new version, a few direct links can form between nodes that are distant in ordinary space. These long-range links might be able to generate quantum uncertainty and the 鈥渁ction at a distance鈥� phenomenon of quantum entanglement.
David Deutsch of the University of Oxford has a fresh approach called qubit field theory. It reduces all variables to yes/no questions at every point in space, so the qubit field for an electron simply says whether or not an electron is there. Unlike other approaches to quantum gravity, qubit field theory says that space is ultimately smooth. It does get lumpy around the Planck scale, but if you look even closer it flattens out again.
One of these approaches might eventually get us closer to quantum gravity and a unified theory. But it is more likely that the crucial idea hasn鈥檛 occurred to anyone yet.
Taming the multiverse
PHYSICISTS call it the fine-tuning problem. The constants of nature seem ideally suited to the emergence of life: tweak any one of them and suddenly there is no nuclear fusion to power stars, or no stable atoms, or everything gets torn apart by antigravity. Why should we be living in a universe finely tuned for life?
An old argument called the anthropic principle explains this by saying that the constants must be as they are, or we wouldn鈥檛 be alive to measure them. It鈥檚 a distasteful idea to many scientists, but it has gained a lot of ground recently through string theory.
String theory allows a huge range of possibilities for the constants of nature and other basics of physics, such as the number of fundamental forces. Leonard Susskind of Stanford University in California believes that all these different universes actually exist at once, in different parts of a multiverse. Most universes have properties that don鈥檛 enable life to evolve, so they go unwitnessed, whereas ours has the particles we call protons, neutrons and electrons, which happen to build freakishly stable matter. They have allowed stars and planets to form, and eventually led to us.
To many physicists, this answer is unscientific. If these different universes lie beyond our reach, we鈥檒l never be able to see them and verify the hypothesis. On the other hand, Susskind believes that the anthropic principle can be tested. If we can understand enough about the landscape of possibilities, then we can make educated guesses about some of the properties of our own universe 鈥� perhaps predicting the masses of as yet undiscovered particles.
Lee Smolin, a theorist based at the Perimeter Institute in Waterloo, Ontario, has an alternative suggestion: universes reproduce by forming black holes. In this scheme, a black hole is the bud of a baby universe that has slightly different physical constants from its parent. Universes suitable for life could evolve by natural selection, because having a lot of kids means making a lot of black holes, and that means making a lot of stars that can nurture life. In this scenario, the existence of life in our universe would no longer be a cosmic coincidence, because the multiverse would contain many similar universes.
