杏吧原创

WHY are we here? In one sense at least, it鈥檚 just a cosmic accident. Our existence is possible only because a number of peculiar coincidences between the values of different constants of nature allow it. The speed of light, the strength of gravity and the charge of an electron, for example, fall within the narrow windows of opportunity that allow atoms to form and hold together. If their values were slightly different there would be no stars, no galaxies and, of course, no life.

But nobody has any idea why the fundamental constants of nature have the numerical values that they do. We can pin them down with impressive experimental accuracy, yet we know nothing of their origins. The constants combine our most precise experimental knowledge of the Universe with our most profound ignorance.

We can鈥檛 even be sure they are constant at all. There are hints that what we like to think are the constants of nature might be changing. If that鈥檚 true, they could eventually slip out of the range that allows life to exist 鈥� and of course, there is no reason why the Universe鈥檚 characteristics should facilitate our perpetual existence. If there鈥檚 one thing that the directed random walk known as progress in science has taught us, it鈥檚 that we are not the centre of the Universe. It doesn鈥檛 need us, and it certainly doesn鈥檛 exist to serve us.

Copernicus was the first to see this with his 鈥� at the time 鈥� heretical deduction that the Universe does not revolve around us. That taught us that the Universe guarantees us no special location in space. Then Darwin showed that we are not the culmination of any special design, and the geologist Charles Lyell discovered that most of history had gone by, eventfully, without us. Deeper still was the insight of Einstein, who showed how to express the laws of nature so that they look the same to all observers, no matter where they are or how they are moving. We can now express the basic laws of nature in forms that would be found by anyone investigating the Universe, from Vega to Vegas 鈥� wherever they are and however they are moving.

The constants of nature provide the next great distancing of science from human idiosyncrasy. We now understand that the structure of the Universe around us is determined by a collection of unchanging characteristics. These include things like the masses of the smallest subatomic particles, the strengths of the forces of nature, and the speed of light in vacuum. They have been quantified by precision measurement: in the backs of physics books the world over their latest values are listed to large numbers of decimal places.

These characteristics generally have units which are rather anthropocentric: they rely on human scale, or perhaps the properties of our Solar System. The speed of light, c, is measured in metres per second, for example. Centimetres, metres, feet and inches are conveniently related to the scale of the human frame. Our days and years are familiar celestial units of time that derive from the timing of the Earth鈥檚 orbit and spin.

Nature provides more fundamental measures of mass, length and time which identify the scales at which gravity and quantum reality collide, and where our understanding of physics is incomplete
(see 鈥淯niversal units鈥�).

But there is no need to stop there. Having taken so many steps to remove the anthropocentric bias from science, it makes sense to get rid of units altogether. Constants like the speed of light c, Newton鈥檚 gravitational constant G and Planck鈥檚 quantum constant h can be combined in ways that produce pure, dimensionless numbers. The resulting constants are much more than just a combination of various properties of the Universe. They are its fundamental descriptors: the bar codes of physical reality. They are central to our understanding of the Universe and our place in it.

Take the fine structure constant, or alpha, for example, which tells us the strength of electromagnetic forces and controls the nature of atoms and molecules. It is determined by a combination of the charge of the electron e, Planck鈥檚 constant h, traditionally divided by 2&pgr;, and the speed of light c. Denoted by the Greek letter alpha and defined by 2&pgr;e²/hc, it is currently determined to be approximately equal to 1/137.03599976鈥�

High and low

This is an important number. If it were much larger, atoms and molecules would be unable to exist; alpha鈥檚 value affects the interaction between electrons and protons, and determines their binding energy. No stars would be able to form either, because their centres would be too cool to start self-sustaining nuclear reactions; alpha dictates the ignition temperature at which these can occur. In short, if alpha were much larger we wouldn鈥檛 be here to know about it.

But no one knows why alpha has this particular value. Even more mysteriously, we have seen hints that its value can change. Over the past two years I have been part of a team, led by John Webb of the University of New South Wales in Sydney, that used new theoretical techniques to analyse the absorption of light from distant quasars by intervening dust clouds. We look at the separation between absorption lines caused by different chemical elements that depend sensitively on the value of alpha at the red shift (an astronomer鈥檚 measure of historical time) where the absorption occurs.

The light left these clouds between 5 and 11 billion years ago, so comparing the observed line separations with separations measured now, in the laboratory, provides a probe of whether alpha can have changed over the past 11 billion years. By using computational solutions of the equations of atomic structure we can determine the shifts in line spacings that would result from tiny changes in alpha and find the shift in its value between then and now that best fits all the data.

The results from observations of 147 quasars over two years were a big surprise, and have potentially far-reaching implications (New 杏吧原创, 11 May, p 28). The complicated 鈥渇ingerprint鈥� of shifts matches that expected if the value of the fine structure constant 11 billion years ago was smaller than it is now by about 7 parts in a million.

In an attempt to find other observational consequences of a changing alpha, Jo膩o Magueijo and H氓vard Sandvik of Imperial College, London, and I have investigated theories which extend Einstein鈥檚 theory of general relativity to include this possibility. It appears that alpha would only change at certain times in our cosmological history (Physical Review Letters, vol 88, p 31302). It鈥檚 controlled by our Universe鈥檚 rate of expansion. Our theory suggests that during the first 300,000 years of the Universe鈥檚 history 鈥� a period of rapid expansion 鈥� there would have been no significant change in alpha at all. After that it would have started to increase in value very slowly until about 5 billion years ago. Observations of distant supernovae suggest that, at this time, our Universe鈥檚 expansion began to accelerate. This is most likely due to a 鈥渧acuum energy鈥�, described by Einstein鈥檚 famous cosmological constant, that began to control the expansion of the Universe. Our analysis leads us to expect that, if that were the case, alpha would have stopped increasing then, and remained steady ever since.

There are other limits on a varying alpha. Geochemical data from a natural nuclear 鈥渞eactor鈥� that ran intermittently below the surface in the Oklo region of Gabon shows that alpha hasn鈥檛 changed by more than one part in ten million in the past 1.8 billion years. The halting of the change in alpha caused by the acceleration of the Universe鈥檚 expansion makes the Oklo limit compatible with the quasar observations. That鈥檚 good news for all of us. Without the vacuum energy to stop this increase in alpha鈥檚 value, there would come a time when atoms and stars could no longer exist. The Universe would no longer be unable to contain the building blocks of complexity, and life, like all good things, would have to come to an end.

The example of alpha illustrates how important the numbers on the Universe鈥檚 bar code can be. But what about the others? Can they change too? With our current technological abilities, it鈥檚 difficult to say. Even if they can, it is no surprise that variations in alpha have been seen first: that is where the highest-precision observations are possible.

Detecting whether there have been changes in Newton鈥檚 constant of gravitation, G, is far harder because gravity is so much weaker than electricity and magnetism, so our observational probes of G are a thousand times weaker than those of alpha. They tell us only that G can鈥檛 be changing faster than 1 per cent of the rate of expansion of the Universe. In the future it will be possible to probe constants like the ratio of the electron mass to the proton mass with greater sensitivity by astronomical observations at high red shift.

The familiar constants that govern atomic and gravitational phenomena are not all there is to the Universe, however. In the best current working theory of particle physics there are more than 25 other basic constants governing the masses of elementary particles and their interactions. Then there are the grander constants that pin down the structure of the Universe as a whole.

We have already mentioned Einstein鈥檚 cosmological constant. Its value is tiny, but nonetheless big enough to have caused the expansion of the Universe to accelerate. Until recently nobody was even sure that it existed, and we still don鈥檛 know why it exists or why it takes the mysteriously small value that it does. String theory, the idea that all of matter arises from the vibration of dense tubes of energy, seems to say that it should not be able to exist at all. Like alpha, the cosmological constant might well turn out to be not quite a constant; the vacuum energy it represents might vary over billions of years, or it might one day decay away into radiation. In this latter case, the Universe would cease to accelerate in the far future and alpha might start to increase again.

From alpha to omega

And then there is omega, a number that measures the present-day deviation of our Universe鈥檚 expansion rate from the 鈥渃ritical rate鈥�. The critical rate is the value that separates universes that will expand forever from those that will contract back to a 鈥渂ig crunch鈥�. Observation shows omega to be close to 1, and popular theories of the Universe, such as inflation, predict that it will be within 1 part in 100,000 of that figure. With such a small margin either way, we may never know on which side of the critical divide we lie.

Finding the complete explanation of these constants is the greatest challenge that physicists face. It might be that they are all uniquely and completely determined by some as yet unfound Theory of Everything. Or perhaps only some of them are determined in this way: the cosmological constant and omega may be bound up with the starting conditions of the Universe, while the values of other constants like alpha and G might just fall out at random as a result of processes that occur during the Universe鈥檚 very early history.

The signs are that any theory of everything that explains these constants can exist only if the world has many more dimensions of space than the three that we see. These 鈥渆xtra鈥� dimensions would have to behave very differently. Where are they? Perhaps they are imperceptibly small or their influence only manifests itself through the action of gravity. If they exist, then the true, unchanging constants of nature will only emerge from a theory that embraces and explains the characteristics of all the dimensions. The three-dimensional shadows that we call our constants of nature will be neither fundamental nor necessarily constant at all.

If it is the higher-dimensional constants that are the only true constants, and the extra dimensions they inhabit were to expand in size or move relative to one another, we would observe the shadows of our three-dimensional 鈥渃onstants鈥�, like e and G, to change at the same rate. This could have profound consequences; the properties of these extra dimensions are likely to prove crucial in our quest to understand the constants of nature. Only when we know why these constants take the values they do will we be able to say we understand the Universe.