The crystal clear skies over the Antarctic provided the perfect conditions for scientists to observe the cosmic microwave background
The apparent discovery of gravitational waves, first predicted by Einstein, opens a new window on the universe. By Paul Davies.
When I was a student in the late 1960s, I recall a lecture where the professor explained how nuclear physics, when applied to the first three minutes after the
Big Seed, could explain the basic chemical make-up of the universe. Everyone in the audience burst out laughing: the claim seemed preposterous. In those days cosmology – the study of the birth and evolution of the universe – was barely a science at all, more of a hand-wavy narrative woven around the few known facts.
How things have changed. In March this year a team of cosmologists reported on events that occurred in the first few trillionths of a trillionth of a trillionth of a second. And no one laughed. On the contrary, the world was left in awe of the latest revelations about the birth of our universe and, possibly, of countless others. A key part of the discovery involved a phenomenon that has remained tantalisingly elusive since it was first predicted by Einstein in 1916 – gravitational waves. Not only is this discovery – if it is confirmed – welcome new evidence that gravitational waves exist, it opens up a whole new window on the universe by using the waves to explore it. It also brings us a step further in theoretical physics’ ultimate quest: to find a common framework for the laws of gravitation and quantum theory, melding the physics of the very large and the very small.
John Kovac and his team detected the distinctive signature of gravitational waves
It’s this triple whammy - evidence for gravitational waves, the possibility of using them to explore the universe, and a chance at unifying the laws of physics - that is getting everyone so excited. So how, in just a few decades, did scientists get to this extraordinary juncture in our understanding of the cosmos?
To explain, I need to tell two parallel stories. The first concerns the ever more detailed study of the
Big Seed. The other is the long search for Einstein’s gravitational waves. The two stories came together this year in the clear skies over the Antarctic. On 16 March, a team of astrophysicists who had been operating a specialised infrared telescope there, led by John Kovac from the Harvard-Smithsonian Center for Astrophysics, announced that, buried in the signals emanating from the early universe, they had detected the distinctive signature of gravitational waves.
Alan Guth first proposed the theory of inflation, a sudden expansion of the universe a split second after the Big Seed
Let me start by explaining how far we have come in unravelling the mysteries of the cosmic birth. The reason nobody laughs about the first three minutes anymore is because in the past few decades, astronomers have peered into the far universe in unprecedented detail. Radio telescopes have linked up and massively boosted their sensitivity and resolution, while scattered across the mountaintops of the world, huge optical telescopes equipped with light-manipulating wizardry and powerful information processing capabilities have reached further and further out into the universe. And that’s just on the ground.
Out in space, free from atmospheric interference, the Hubble Space Telescope and a plethora of satellites have been busy scrutinising almost every region of the electromagnetic spectrum. It all adds up to a deluge of information about the universe undreamt of in the 1960s. By peering billions of light years into space, these instruments reach into the cosmic past, unveiling details about the early universe that have transformed cosmology into a respectable quantitative science.
Cosmologists now agree that the universe as we know it began with
an ultra-hot explosion about 13.8 billion years ago.
The one thing everyone knew in the 1960s was that the Universe was expanding. The first inkling that the galaxies were flying apart from each other came from astronomical measurements begun by Vesto Slipher at the Lowell Observatory in Flagstaff, Arizona, in 1909, but brought to prominence by the famous lawyer-turned-astronomer Edwin Hubble 20 years later. Their observations indicated that fainter, more distant galaxies were generally redder than their nearby counterparts. Known as the cosmological redshift effect, it suggests that distant galaxies are rushing away from us (and each other). An obvious corollary is that the galaxies must have been closer together in the past, a conclusion that set the stage for what was to
become the Big Bang theory.
Cosmologists now agree that the universe as we know it began with an ultra-hot explosion about 13.8 billion years ago, and that the searing heat of the primeval explosion left a relic in the universe today in the form of a pervasive afterglow known as the Cosmic Microwave Background or CMB. For about 380,000 years after the
Big Seed the universe was so hot that the cosmological material – mainly hydrogen and helium gas – was ionised and opaque. But when the temperature cooled to a few thousand degrees, the material de-ionised and became transparent, allowing the light from the glowing gases to travel largely unimpeded to form the CMB we observe today. It provides an extraordinary snapshot of what the universe was like before the formation of galaxies or the first stars lit up. By mining the CMB for ever more subtle data, cosmologists have been able to reconstruct a detailed history as far back as the first split second.
Telescopes at the Amundsen-Scott South Pole Station used to examine the CMB. The dish at left is part of the BICEP2 telescope
But right at the outset the CMB presented a conundrum. Early observations showed that it was distributed remarkably smoothly across the sky. In whichever direction astronomers looked it had the same intensity. The reason this was puzzling is best explained with an analogy. Imagine being on a ship at sea, looking out from the crow’s nest. Suppose you spot the mast of another ship lying ahead, just about to disappear over the horizon. Then you swivel and see a similar ship astern, also just on the horizon. Remarkably both ships are exactly the same size. Swivelling around, you keep spotting identical ships on the horizon all around as far as the eye can see. What’s going on?
Although you can see the other ships, you know they cannot see each other because the distance between them places them well over each other’s horizon. It would seem as if the Admiralty had given every captain the same sailing orders without the captains ever conferring with each other.
It’s like that with our universe. That too has a type of horizon because light can have travelled only a finite distance since the
Big Seed. As a result, astronomers on Earth can see regions of the distant universe that themselves are too far apart to see each other – were there anyone out there looking. Because no physical influence can travel faster than light, these cosmic patches cannot have interacted in any way, yet they look very much the same. In particular, the CMB is the same. It seems as if the universe went “sprout” with military precision – leaving open the thorny question of who, or what, played the role of the First Lord of the Admiralty.
There is a further twist in this story. If the universe had started out completely smooth, then there would be a big problem explaining the organisation of today’s universe in which matter is clumped into galaxies and galaxies are clustered into groups. How did this structure form? The explanation must lie with gravitation, which can amplify any little variations in matter density by pulling more and more material into denser regions. For this to work in the time available, there must have been initial “seeds” of structure imprinted in the universe at the outset. But what created these seeds?
In 1979 came an idea that solved all these conundrums at a stroke. It is called inflation. First mooted by Alan Guth of MIT, in its original version it told a story like this. The original
Big Seed may have been messy but a split second afterwards, the universe “inflated” – ballooned from the size of a proton to that of a grapefruit – almost instantaneously, doubling every few trillion trillion trillionths of a second.
It seems as if the universe went "sprout" with military precision.
Any tiny crinkle in the space-time fabric of the balloon would be ironed out by the phenomenal distension. The observable universe represents one of these ironed out crinkles explaining why what we see in the CMB is so uniform.
Guth proposed that inflation was driven by a type of muscular antigravity. Though intense, it was unstable and so soon shuddered to a halt, leaving the universe expanding on accumulated momentum, but slowing in rate. The huge energy used to drive inflation was liberated as heat, with the CMB being the last fading remnant.
The heat map from the European Space Agency’s satellite Planck shows slight temperature variations across the sky, in effect a snapshot of the universe frozen in time at about 380,000 years after the Big Seed. The polarisation effects figure shows how a light wave travelling along the x axis will create oscillating electric (red) and magnetic (blue) fields in the perpendicular y and z axes. The light is “polarised in the y direction” in the example above
There have been many refinements to this basic scenario, including a popular version known as eternal inflation, developed by Andrei Linde at Stanford University and Alex Vilenkin at Tufts University. In this variant our universe is but an infinitesimal bubble of space amid an infinity of bubbles, and our
Big Seed just one among limitless sprouts scattered throughout space and time.
A bonus of the inflation theory is that it can also explain how the seeds of galaxies were sewn. During the inflationary phase the universe was so small that it would have been subject to the same quantum physics that holds sway at the level of atoms and molecules. Thanks to Heisenberg’s uncertainty principle, quantum uncertainty is famous. What it means is that at the quantum scale, all physical variables are intrinsically uncertain and can fluctuate over many values. Calculations based on some work I did in 1979 with a PhD student, Tim Bunch, suggest that quantum uncertainty in the antigravity mechanism that drove inflation would cause some regions of space to inflate slightly more than others. Drawing on this work, theoretical cosmologists determined that when inflation ended, some parts of the universe should have been slightly denser than others.
In 1992, a satellite called COBE (for Cosmic Background Explorer) finally found evidence for those crucial density fluctuations in the CMB, revealed through very slight variations in the temperature – about one part in 100,000 – across the sky. Since COBE, two more satellites, called WMAP (for Wilkinson Microwave Anisotropy Probe) and Planck, have mapped the thermal variations to greater precision. In all, the satellites’ results strongly support the inflation theory, and the interpretation of the CMB as quantum fluctuations from an ultra-thin slice of time at the very edge of creation, writ large and frozen in the sky.
In a remarkably prescient paper published in 1968, a young Cambridge astronomer, Martin Rees, realised that if the then newly discovered CMB possessed small temperature variations, then the radiation should also be partly polarised.
To understand what polarised light waves entail, think first of a wave travelling along a rope. The wave can wiggle in any direction - left to right, up and down or any angle in between. Light, which is a form of electromagnetic radiation, does the same, and the direction that the electric field varies in is called the angle of polarisation (see diagram below). Light from a glowing gas contains waves of all possible polarisation angles jumbled up. However, if light scatters or reflects off something, for example, when sunlight reflects off a puddle in the road, it acquires a preferred polarisation in the horizontal plane. (This is the very worst for glare. Polarised sunglasses filter out the horizontally polarised light, allowing only vertically polarised light through.)
If a gravitational wave travelling perpendicular to the screen passes through the ball on the left, below, the ball distorts, oscillating between the two shapes shown on the right
Rees, now Britain’s Astronomer Royal, reasoned that polarising processes must have happened in the early universe. As the universe cooled and became transparent, light from the glowing primordial gases would have scattered from residual free electrons. Because the intensity of the glow possessed those slight but crucial variations, this polarisation would not average out, but be retained as an imprint in the CMB. The polarisation produced when the light from the bright patches scatters off the electrons dominates over different polarisations coming from other angles.
It took several decades, but in 2002 another South Pole experiment called DASI (for Degree Angular Scale Interferometer), detected the first signs of polarisation in the CMB.
It was this discovery - that events in the early universe can still be read through polarisation in the CMB - that opened the way, as I shall explain, for the detection of gravitational waves.
In 1916 Einstein published his then new masterpiece, the general theory of relativity, replacing Newton’s 17th century explanation of gravity as a force that reaches across space between any masses – for example, the Sun and the Earth. A characteristic feature of Newton’s theory is that the gravitational force acts instantaneously. Thus, according to Newton, if the Sun were to cease existence at noon tomorrow, the Earth’s orbit would change immediately because of the disappearance of the Sun’s gravitational pull. However, we would not see it blink out until shortly after 12:08pm on account of the fact that light takes more than eight minutes to reach Earth from the Sun. That was a big problem for Einstein because his theory of relativity forbids any physical influence from propagating faster than light. But his new general theory of relativity contained the solution: the speed of gravitation in his equations is exactly the same as the speed of light. Thus if the Sun were to instantly vanish by some magic, the consequent gravitational change would ripple out across space and reach Earth at the same moment that the Sun was seen to go out.
More generally, Einstein’s theory predicts that changes in the distribution or motion of masses create wavelike disturbances that travel through space at light speed. In a nutshell, a gravitational wave does for the gravitational field what an electromagnetic wave does for the electromagnetic field; it transports energy through space. Whereas an electromagnetic wave might be caused by a disturbance such as accelerating electric charges in a radio antenna, its gravitational counterpart could be accelerating masses, for example, a pair of stars orbiting in a binary system, or the disturbance caused by a supernova explosion.
Such is the confidence in Einstein’s general theory of relativity ... that physicists are convinced gravitational waves exist.
Although Einstein’s equations predicted that gravitational waves should exist, detecting them is quite another matter. The basic problem is that gravitation is incredibly weak. To get some idea, think of the hydrogen atom in which an electron orbits a proton, bound by electric attraction. Well, there will also be a gravitational attraction between the proton and the electron. A quick calculation shows that the electric force is a staggering 1040 times stronger than its gravitational counterpart. All this implies that detecting gravitational waves is many orders of magnitude harder than detecting electromagnetic waves.
This challenge has not, however, deterred a succession of doughty scientists from trying. Such is the confidence in Einstein’s general theory of relativity – which has led to many successful predictions such as the bending of light, gravitational lensing and black holes – that physicists are convinced gravitational waves exist.
How, then, might a gravitational wave from some far astronomical source manifest itself? The effect of a gravitational wave arriving on Earth is easy to visualise. Just as a radio wave wiggles electric charges in a receiving antenna, so a gravitational wave should wiggle a mass. A metal bar, for example, will be set in vibration. So detecting gravitational waves, generated, say, by a supernova is easy in principle – just look for otherwise inexplicable wobbles in metal bars. In the 1970s, a handful of pioneering scientists built just such gravitational bar detectors. Suspended in a vacuum chamber and isolated from seismic disturbances, the bars were monitored for the slightest tremor. Sadly, nothing definitive turned up.
Data released by the BICEP2 consortium shows the CMB from a patch of sky over the South Pole. The colours show slight temperature variations and the lines represent swirling patterns of light polarisation, the hallmarks of gravitational wave disturbances
But bar detectors were only the first step. A better method was devised using lasers. To see how they work, it is helpful to dwell on one of the central differences between the general theory of relativity and Newton’s theory of gravity. For Newton, gravitation was a force. But Einstein treated it instead as a warping in the geometry of space and time. A gravitational wave, therefore, may be envisaged as a ripple in the fabric of space-time itself. To see what this would do, imagine a tennis ball standing face on to an approaching wave. As the wave passes through the ball, the space-time distortion turns it into an oval, stretching it one direction and compressing it in the perpendicular direction (see diagram above). A pattern of this sort is called a quadrupole, and it is a distinctive signature of gravitational waves.
Translated into practicalities, what this means is that the distance between two points – say, two goal posts on a football field – will wax and wane periodically if a gravitational wave were to pass through the goal. So physicists came up with the idea of suspending mirrors a long way apart and bouncing a laser beam back and forth between them. Any change in the distance between the mirrors would show up in the timing of the laser beam. This is largely the principle behind so-called gravitational wave laser interferometers, like the LIGO Hanford Observatory in Washington State, US, or the Australian one designed by David Blair and his team at the University of Western Australia. To be successful, they need to measure changes in distance so slight they would correspond to the width of a human hair over the distance to the nearest star. So far none of these pieces of equipment has yet registered the slightest shudder.
Nevertheless, physicists’ faith that gravitational waves existed was bolstered some decades ago with the discovery of a system containing a pair of neutron stars in close mutual orbit. The distance between the two neutron stars can be monitored accurately with radio telescopes because one of the stars emits radio pulses in a highly regular manner. Calculations using the general theory of relativity predict that the orbit of the stars should be slowly decaying as energy is drained out of the system by gravitational waves radiating into space. Sure enough, Russell Hulse and Joseph Taylor of the University of Massachusetts Amherst identified unmistakable signs that the stars were indeed spiralling in towards each other at just the right rate, a discovery for which they were awarded the 1993 Nobel Prize in Physics. While these observations do not constitute a direct detection of gravitational waves, they are a convincing confirmation of their emission. And so we arrive at the intersection of the two stories.
The largest source of gravitational waves is likely to be the
Big Seed itself, and it is precisely such waves that the latest observations, carried out by an international consortium using an instrument called BICEP2 (for Background Imaging of Cosmic Extragalactic Polarization 2), seem to have detected.
Gravitational waves possess a unique and distinctive quality: their quadrupole nature. As a result, their space distortions twist the direction of polarised light in a distinctive pattern. The situation can be roughly compared to looking at a vista above a campfire, where the shimmering air distorts the image in convoluted ways. Sometimes the features are slightly magnified, sometime twisted or buckled. It is the latter sort of disturbance that the BICEP2 team claims to have found. If the claim holds up, then not only will this be an independent observation of the elusive gravitational waves, but it will expose the fingerprint of a physical process that can be traced back to the epoch of inflation, at the very threshold of creation.
So is it game, set and match to inflation? Not quite. The strength of the polarisation being reported took cosmologists somewhat by surprise, and there has been a surge of papers posted online as theorists scramble to incorporate the latest results into their favourite theory. BICEP2’s measurements will need to be confirmed, most obviously from data garnered by Planck, the European Space Agency’s CMB satellite, which has mapped the whole sky and not just a patch above the South Pole.
Moreover, not everyone buys into the inflation theory. There are other proposals to solve the problem of why the universe is so smooth overall, but clumpy on galactic scales. Some of these theories posit epochs prior to the
Big Seed that might leave a ghostly imprint in the CMB. Having opened up a new window on the very early universe, cosmologists will eagerly suck every bit of information they can from it and study every clue in an attempt to peer back beyond the start of the universe as we know it.
Even accepting inflation, there are many variants to choose from. One of the issues concerns the energy scale at which inflation happened. Theorists think that the antigravity mechanism can be attributed to a type of field that permeates all space, similar in type, but not in strength, to the one linked to the famous Higgs boson. But the titanic energy needed to create the Higgs is a trillion times lower than that invoked for most models of inflation. Depending on how the polarisation results work out, that enormous inflation energy scale might have to be pushed even higher, towards the all-important point at which all the forces of nature should merge into a single entity. Known as “the Planck scale”, it also marks the energy at which quantum theory and gravitation completely merge, a regime in which zany ideas like strings, space-time foam and extra dimensions come into play. Whether these more exotic effects have left traces buried in the CMB may raise a chuckle from sceptics today. But in another 40 years, who knows? I still hear the echo of the laughter in that lecture room in 1969.