Chuck Bennett remembers vividly the moment he realised the buck stopped with him – that he, and he alone, was responsible for the success or failure of a multimillion-dollar NASA space mission. The news that the project had got the green light had just come through when someone grabbed him by his jacket lapels. ‘You’ve got it now,’ they rasped. ‘Don’t fuck up!’
The need for a follow-up mission to COBE had been recognised by everyone as soon as the satellite discovered the elusive cosmic ripples in the Big Bang radiation back in 1992. It galvanised the cosmic-background community. With the temperature fluctuations in the sky detected, exactly as predicted, the next step was to measure how big they were on each length scale. This would enable the fireball radiation to be milked of every last drop of precious information it carried with it from the beginning of time.
It was immediately clear that this task did not require a space mission anywhere near as big and complex as COBE. To ‘characterise’ the cosmic ripples would involve merely improving on the measurements of the Differential Microwave Radiometer, not the other two instruments carried by COBE. Bennett saw a problem, however. ‘NASA didn’t have a launch vehicle for a small satellite,’ he says.
The trouble was, people at NASA wanted to be associated with big space missions. Bigger was deemed more important. It was where the kudos was. Bigger, more complex, more expensive missions also brought in more business for NASA. They provided more jobs.
What was needed now, Bennett realised, was a change of culture. It wasn’t just that he knew a simple experiment was all that was needed. He was impatient. ‘Smaller would be faster,’ he says. ‘COBE had hinted at the riches up there in the sky. We didn’t want to wait decades to get our hands on them.’
So it was that Bennett began to devote time and effort to convincing NASA that there was merit in modest space missions. ‘It wasn’t just a cosmic background experiment that could be done,’ he says. ‘A whole host of other scientific questions could potentially be answered with smaller, cheaper, faster missions.’
On the face of it, it was not much to ask of NASA. The agency’s incredibly successful Delta 2 rocket, which had launched COBE, derived its thrust from nine strap-on boosters known as GEMS. To launch a smaller payload, it was necessary merely to leave off a few of the boosters. ‘Three or four would be plenty,’ says Bennett. It was a simple and relatively straightforward modification. ‘However, this was not about changing rockets,’ he says. ‘It was about changing minds – nudging the juggernaut that is NASA in a marginally different direction.’
It took time, determination and perseverance. But success came in 1994. NASA approved the Medium-Class Explorer, or MIDEX, programme for payloads up to about a quarter of COBE’s 5,000 pounds. It removed a major obstacle on the road to launching a ‘Son of COBE’. Now all Bennett and his colleagues had to do was come up with a mission design that NASA would approve for a MIDEX launch. Of course, that was easier said than done.
In the beginning several groups had ideas for a follow-up cosmic background mission. For a few months, a team led by Lyman Page and Dave Wilkinson at Princeton University had been talking with another team at NASA’s Jet Propulsion Laboratory in Pasadena, California. The groups had failed to gel, however. Disheartened, the Princeton people visited the Goddard Space Flight Center to get some answers to technical questions about space missions. ‘The meeting went very well,’ said Bennett. ‘They decided to collaborate with us, with me in the lead as principal investigator.’
John Mather was initially part of the team, but then he was poached to head the project to build and launch the successor to the Hubble Space Telescope, the James Webb Space Telescope. It was a measure of the success of COBE and how far Mather’s star had risen in NASA’s firmament.
At this point, the team lived in hope rather than expectation that a space mission would actually happen, but they set about pinning down in detail exactly how they would do an experiment. And, in early 1995, the green light for a proposal came. NASA put out an ‘Announcement of Opportunity’ for the first two MIDEX missions.
Nothing was certain, nothing was in the bag. Not only was Bennett’s team vying with other scientific projects for a MIDEX launch, they were up against two other proposed cosmic background experiments as well.
At least their experiment had a name now: the Microwave Anisotropy Probe, or MAP. MAP made sense since the idea was to make a ‘map’ of the temperature of the whole of the sky. But ‘anisotropy’ was a different matter. ‘I don’t know what possessed us to use a word like that – we must have been mad,’ says Bennett. ‘It didn’t mean anything at all to the public.’
MAP, like COBE, would carry pairs of microwave-collecting horns. Each pair would compare the temperature of the sky in two different directions. One difference between MAP and COBE was that the horns would operate at five wavelengths rather than three. Another was that they would not look directly at the sky but via a magnifying ‘telescope’ – simply a concave radio dish. COBE had blurry vision. Even the finest hot spots and cold spots in its ‘baby photo’ of the Universe were large temperature splotches about seven degrees across (14 times the apparent diameter of the Moon). The aim of MAP would be to produce a far crisper, sharper image of the Big Bang radiation, with much better sensitivity.
Hot and cold spots seven degrees across and bigger corresponded to cosmic regions at the epoch of last scattering that were so far apart that no light signal could have spanned them since the first split second of the Universe. In the jargon, they were outside each other’s ‘horizon’. Since heat too is restricted to the cosmic speed limit set by light, it could not have flowed across the regions either. Consequently, the temperature of COBE’s hot and cold spots could not have been changed by such processes since the first split second. They were ‘fossils’ from the beginning of time itself, impressed on space during the inflationary epoch. ‘Everything changes, however, at a scale of about two degrees,’ says Bennett.
The reason is that hot spots and cold spots smaller than this corresponded to regions close enough at the epoch of last scattering that a light signal could have spanned them since the beginning. Because the hot and cold spots could have been changed by processes operating since the beginning of time, they would provide a ‘window’ onto the Universe when it was only 380,000 years old. ‘The plan was for MAP to image the whole sky, showing details as small as 0.2 degrees across, fifteen times finer than managed by COBE,’ says Bennett.
Faced with competition from the other proposals for cosmic background missions, Bennett’s team felt under pressure to propose a mission with improved performance. But their strong preference and commitment was to keep things simple: fast, inexpensive and low risk. ‘If anyone was claiming it would take six months to develop a particular piece of technology, it was out,’ says Bennett. ‘Experience teaches you everything takes far longer than people estimate.’
The team even did away with the need for a bulky and costly dewar of liquid helium for cooling the detectors. ‘That had been a major cost and complication on COBE,’ says Bennett. ‘Instead, we decided to use metal fins to radiate heat into space and so “passively cool” the instruments.’
But simplicity was not the team’s only weapon. It had another up its sleeve: detail. The team worked long hours, at night and at weekends, to produce the most detailed proposal NASA had ever seen. ‘I put my life into that proposal – often going to bed at 2 a.m. and getting up again at 6 a.m.,’ says Bennett. ‘When we’d finished, it was a whopping four inches thick and chock full of detail.’
It was late December 1995 when Bennett and systems engineer Clifton Jackson drove the proposal over to the copy shop at NASA Goddard. ‘There was a Christmas party in full swing when we got over there, and nobody was working in the shop,’ he says. ‘We had to sneak in, figure out how to use the equipment and do all the bound copies ourselves.’
But it was done. Bennett and Jackson drove downtown and delivered the proposal to NASA HQ. Now there was a four-month wait for NASA’s verdict. The delay might have been agonising, but Bennett busied himself analysing COBE data and setting up a comprehensive MAP website to convey to the public every aspect of the science and engineering of the mission. At last, in April 1996, came a phone call. NASA had selected MAP for a ‘definition study’. It was soon after this that Bennett was grabbed by his lapels by the guy telling him not to fuck up! ‘That was pressure,’ says Bennett. ‘I was principal investigator on this project and I knew that we would face tough challenges.’
He recalls the stress of the early days of the project. ‘Every evening I would go home with a tight knot in my stomach. Every day there was a new problem and every night I went to bed with it still running around in my head.’ After a while, though, Bennett came to realise that every problem did get solved. And, just as soon as it got solved, it was replaced by a new problem, which in its turn got solved. ‘I learnt to relax a little, to be a little more philosophical, to take it one problem at a time,’ he says.
Nevertheless, Bennett earned the name ‘mad dog’ from his wife, Renee, and his sons, Andrew and Ethan. ‘Not because of his attitude towards us – he’s a cream puff at home,’ says Renee. ‘But rather because of the ferociousness with which he attacked problems, refused to let threats to the mission quality or schedule slip by, and was generally madly protective of his satellite. He’s mostly a very happy person, but he can look pretty scary when he perceives bad decisions or practices.’
One thing that helped ease mad dog’s stress levels was the MAP team itself. ‘On COBE, I noticed that about 20 per cent of the people did about 80 per cent of the work,’ he says. ‘So for MAP I tried to choose people exclusively from the 20 per cent category.’ Bennett also realised that there was no point in employing the smartest person in the world if they were difficult to work with.
The MAP team was small compared with COBE – a core of only 100 scientists and engineers compared with several times that number on the earlier background mission. ‘The negative was that we all had to work like the dickens,’ says Bennett. On the other hand, he had picked a bunch of people that worked well together. ‘It was like a war experience. We were a band of brothers – and sisters. At the time some people complained about the workload, but it’s funny how, afterwards, almost everyone said it was the best thing they had done in their life.’
Thankfully, the team was given wide latitude by NASA HQ, largely because of the project’s small budget. There was some concern, however, when representatives from HQ came to a MAP review meeting and said to Bennett, ominously: ‘We’ll have to talk to you in the break.’ Throughout the meeting, Bennett found it hard to concentrate. ‘I thought, “Oh God. They must think we’re in deep trouble.”’ When the break came, Bennett took a deep breath and steeled himself for the worst. ‘I was told we weren’t sending enough freebies to HQ – those stickers and pens and things NASA always has for its missions!’
Finally, on 30 June 2001, came the day of MAP’s launch. ‘One of the upsetting things was that some of the team couldn’t be there to see it – they had to be back at Goddard monitoring equipment,’ says Bennett. Bennett was at Cape Canaveral in Florida. In fact, he had been there for three weeks. And it was non-stop problems. ‘Even hours before the launch, a whole load of boats were loitering in the restricted area off the coast,’ he says. ‘All of them had to be cleared out.’
Bennett’s wife and sons were guests at the launch site and saw the launch. So did Dave Wilkinson. But, ironically, Bennett did not. He was in the control room, glued to computer displays monitoring the engineering data. He glimpsed the Delta rocket only when it was already high in the blue sky atop its column of white smoke. ‘I remember crying just from the emotion,’ says Renee. ‘Up the rocket went, beautifully up and up. Everyone was cheering. We watched until we could see no more.’
Bennett had no time to admire the view. A celebration party had started and everyone wanted to talk with him. But he jumped into a car and sped over to another building, where a computer was awaiting telemetry from the MAP instrument. The screen had a series of boxes that would light up green as things turned on properly, and red if something was wrong. ‘As I watched, one by one, they turned green,’ says Bennett. ‘It was a tremendous relief.’
But there was no time to relax yet. The next morning, after a night in a rented condo, Bennett flew back home to Washington DC and headed straight to Goddard.
The fourth of July was the first day Bennett had off in a long time. But that night he could not get to sleep because of a terrible pain in his back. In the early hours, he got out of bed, trying not to alarm Renee, and lay flat on the floor. It made no difference. He got on his elliptical trainer in the hope that if he loosened up his muscles, the pain might go away. It did not. The pain got worse. He tried taking a hot shower. Still worse. He finally woke Renee and told her that something was very wrong. She called his doctor. He said that Bennett might be having a heart attack and he should get to the emergency room immediately.
By this time, his wife was alarmed. She got out the car. With Bennett doubled up in the passenger seat, she drove to the local hospital’s emergency room. ‘But she kept stopping at red lights,’ he says. ‘She never in her life met a rule she didn’t obey.’ In the end, Bennett yelled out in pain: ‘It’s 3 a.m.! There’s not a car on the road! Run the red lights!’
Tests at the hospital revealed a gallstone, widely believed to be one of the most painful afflictions known to man. ‘You must have noticed warning signs,’ said the doctor who examined him.
‘I never noticed anything,’ said Bennett.
The doctor looked at him, aghast. ‘What the hell were you doing?’ If Bennett had been in less agony, he might have explained.
Hospitalised and dosed up to the eyeballs with painkillers after having his gangrenous gall bladder removed, Bennett missed a critical rocket burn. (It went fine.) But he had no choice; he had to take more of a back seat with MAP than he had expected.
‘It was not great having Chuck end up with a gangrenous gall bladder,’ says Renee. ‘I’ve told him that’s the last body organ he’s allowed to sacrifice for science.’
The next worry for Bennett was the space probe’s orbit. One of the biggest problems COBE’s engineers had faced was shielding the delicate instruments from the overwhelming heat of the Earth. Since the satellite was in a low Earth orbit, the planet filled pretty much half of the sky. To avoid this problem and give MAP the ability to detect even fainter temperature differences than COBE, the decision was taken to place it at the ‘Lagrange-2’ point of the Earth–Sun system.
L2 is 1.5 million kilometres beyond the Earth on the extension of the line joining the Sun to the Earth. It is one of a handful of locations, discovered by the eighteenth-century French mathematician Joseph Louis Lagrange, where the gravitational force on a body is balanced by the centrifugal forces acting on it, so the body is becalmed in a kind of gravitational Sargasso Sea. ‘No one had ever put a satellite at L2 before,’ says Bennett. ‘I could control everything else but that was the one thing I had no control over.’
Bennett need not have worried. In August 2001, after a series of delicate rocket burns, MAP was gently deposited at L2. Like COBE more than a decade before, it opened its eyes to the Big Bang radiation.
MAP was looking for minuscule variations in the radio static coming from different directions in the sky. The horns and the electronics used to detect and amplify the tiny signals generated radio static themselves simply by virtue of the fact that they contained jiggling electrons. But the longer the instruments looked at the sky – in other words, the greater the quantity of data they collected – the bigger the contrast between the signal and the confusing noise. With MAP, the team chose to analyse the data only when the instruments had gathered a year’s worth of data and had observed the whole sky.
It was while the team were engaged in this data analysis in 2002 that they received bad news. Dave Wilkinson, the ‘grandfather of cosmic background experiments’ and a colleague and friend of many of those on the MAP team, had died. He had been diagnosed with cancer towards the end of the COBE project but had chosen to keep it quiet, preferring there to be no fuss and to get on with his work.
In 1965, Wilkinson had been pipped to the post by Arno Penzias and Robert Wilson, who had later carried off the Nobel Prize for the discovery of the cosmic background radiation. At the time, he had been unconcerned by being beaten, thinking that this was just the first of a series of wonderful things that was going to happen to him in his career. In his naivety, he had not realised how rarely discoveries of such magnitude came along. Wilkinson had later worked on both COBE and MAP, though he did not work full-time on either space experiment, preferring to do smaller-scale cosmic background experiments in parallel. ‘We were very much saddened by his death,’ says Bennett. ‘Dave was a true gentleman, a likeable man of high integrity.’
It was Lyman Page who came up with the idea of renaming MAP in Wilkinson’s honour. ‘We were all for it,’ says Bennett. ‘The difficulty was to convince NASA.’
It was not that NASA was averse to the idea of renaming space missions – quite the contrary. The agency commonly did it. ‘The problem was, once we raised the idea, it might decide to rename the mission after someone else entirely,’ says Bennett.
It took a fair bit of tact from members of the MAP team. But, finally, they achieved their goal, thanks to the support of Ed Weiler, NASA’s Associate Administrator for Space Science, a consistent and stalwart supporter of the mission from the early days. At the press conference to announce the first year’s results, held on 11 February 2003, MAP was formally unveiled as the Wilkinson Microwave Anisotropy Probe, or WMAP. ‘Dave’s wife and children were very happy, and I’m sure Dave would have been too,’ says Bennett.
To those on the outside of the project, it seemed like an age between the end of WMAP’s first year of observing the fireball radiation and the unveiling of the first year’s results, obtained from data recorded between August 2001 and August 2002. Bennett points out, however, that the data analysis actually took only six months – from August 2002 to February 2003. ‘This is astoundingly fast,’ says Bennett. ‘We worked through many nights and the holiday season to do it.’
The team had to search for every imaginable systematic error that might contaminate the data. They had to calibrate the data and see if there were periods of time where the data should not be used. They had to model the radio emission from the Milky Way in order to subtract it from the signal. They had to create tens of thousands of cosmological models and see which were most compatible with the data. They had to write a whole load of thick scientific papers, not only detailing the results but also exactly how they had been obtained from the data and why they should be believed, so that others could double check. ‘We worked our butts off on that data set,’ says Bennett.
But it was worth it. ‘Those are some of the most looked-at papers in the history of science,’ says Bennett.
Renee recalls the press conference at which the WMAP results were presented on 11 February 2003 as ‘incredibly exciting’. ‘I took the kids out of school and we travelled down to NASA headquarters by subway. Chuck had explained enough to me that I knew that the results from the precision measurements of the microwave background were very important.’
WMAP, unlike COBE, was sensitive to processes going on at the epoch of last scattering. At the time, the mix of photons and nuclei filling the Universe behaved like a fluid sloshing about in a bath. There are all kinds of ways the water in a bath can slosh about. For instance, it can slosh about with a big hump in the middle of the bath; with two smaller humps; with three even smaller humps; dwindling all the way down to tiny ripples on the surface of the water. Well, as it is for a bath, so it was for the early Universe. The big humps manifested themselves as big splotch-like hot spots and cold spots in the cosmic background radiation, while the tiny ripples appeared as mere freckles in the temperature map. And each of these sloshing ‘modes’ had its own characteristic temperature enhancement. Some modes were much hotter and colder than the average temperature of the sky, whereas others were hardly different from it at all.
WMAP had found big temperature enhancements for hot spots of some sizes and small ones for hot spots of other sizes. Plotted on a piece of graph paper, with the size of the hot spots getting bigger from left to right across the page, this temperature ‘power spectrum’ looked like a mountain range, with alternating peaks and valleys. The mountain range contained a vast amount of information about the Universe. ‘It was bursting with riches,’ says Bennett. ‘The riches we had been dreaming about for so long.’
In 1970, the renowned American cosmologist Allan Sandage said cosmology was about ‘the search for two numbers’: the ‘Hubble constant’, essentially the expansion rate of the Universe, and the ‘deceleration parameter’, a measure of how fast the expansion is being braked by gravity. ‘After WMAP, it was about scores of numbers,’ says Bennett.
The location and height of every peak and valley encoded priceless information about the Universe. For instance, the location of the first peak is related to the age of the Universe and the curvature of space, whereas its height is related to the number of atoms in the Universe. Remarkably, Bennett and his colleagues were pretty much able to read off the critical numbers that characterise our Universe, numbers that either nobody knew or, if they did, were known only crudely, despite decades of painstaking observations with the world’s biggest telescopes. ‘WMAP established the standard picture of cosmology,’ says Bennett.
That picture contains a number of key ingredients. The first is inflation, the burst of super-fast expansion widely believed to have happened in the Universe’s first split second of existence and which neatly explains how bits of the Universe that appear never to have been close enough to have influenced each other share the same temperature today. Since the Universe had inflated from a super-tiny region – far smaller than anyone had suspected – everything had been in contact with everything else early on.
Inflation was so fast – faster than light – that the horizon of the Universe shrank, effectively stranding the biggest temperature splotches outside of the observable Universe. Only when inflation came to an end did the horizon grow again, enabling the stranded temperature splotches to march back into the Universe again – last out, first back in. ‘What WMAP showed us in the temperature power spectrum is exactly what this picture predicts,’ says Bennett. ‘It is strong evidence that something very much like inflation really happened.’
The second ingredient of the standard picture of cosmology is dark matter. Its extra gravity is required to speed up the growth of clumps of matter in the early Universe so that galaxies as big as the Milky Way can accrue in the relatively short time available since the Big Bang. WMAP showed that exactly 23 per cent of the mass-energy of the Universe is tied up in dark matter, compared with only 4 per cent in the form of the normal – atomic – matter that makes up you, me, the planets and the stars.
If you are thinking that 23 per cent plus 4 per cent equals only 27 per cent and are wondering about the remaining 73 per cent, that is an interesting, not to say extraordinary, story. For the Universe to sit on the knife edge between expanding for ever and one day re-contracting it must contain a very special quantity of matter – the ‘critical density’. This is precisely the density predicted for the Universe by the theory of inflation. However, by the 1980s, it was clear that the matter content of the Universe – visible and dark – amounted to only about 30 per cent of the critical density
If inflation did not happen, the astronomer Neta Bahcall had pointed out, then it was hard to understand how the Universe had ended up so close to the critical density when, in theory, it was free to have any matter density it liked. If, on the other hand, inflation did happen, then there must be something else making up the remaining 70 per cent of the mass-energy of the Universe so that it had precisely the critical density.
In trying to engineer an unchanging, ‘static’ Universe, in 1917 Einstein had proposed that empty space might have a weird repulsive energy. He had later abandoned the idea when Edwin Hubble discovered that the Universe was expanding, calling it the biggest blunder of his life. If empty space did contain energy, it would have a mass equivalent. This might boost the mass of the Universe up to the critical density, as required by inflation.
In 1998, the scientific community was stunned by a discovery made by two independent teams. Both had been observing ultra-distant ‘Type Ia’ supernovae. Such supernovae, formed by the explosion of white dwarf stars with similar properties, were thought to be ‘standard candles’ – that is, of the same intrinsic luminosity. They could therefore be used to determine distances across the Universe.
What the two teams discovered was that the most distant supernovae were fainter than expected from their estimated distances, which were deduced from their red shifts. It was as if in the time the supernova light had been travelling to Earth, something had pushed the supernovae further away than expected. That something could only be the expansion of the Universe. Contrary to all expectations, that expansion appeared to have actually speeded up since the stars detonated long ago.
This was the complete opposite of what was expected. It had long been thought that the only force operating in the large-scale Universe was gravity. This was like a web of elastic, joining the galaxies and putting the brake on their headlong flight from each other. The fact that the galaxies were fleeing ever faster meant that some other force was acting in the Universe, something like the cosmic repulsion, or ‘cosmological constant’, envisaged, then dismissed, by Einstein. It was dubbed the ‘dark energy’ and it was the major mass component of the Universe. Somehow we had managed to overlook it until 1998.
The dark energy rescued inflation. Accounting for 73 per cent of the mass-energy of the Universe, it boosted the Universe to exactly the critical density. But it threw a clonking great spanner into the delicate workings of physics. Our best physical theory is quantum theory, which predicts the results of all known experiments to an obscene number of decimal places. But when quantum theory is used to predict the energy density of the empty space – the vacuum – it overestimates what is observed by a factor of 1 followed by 120 zeroes. This is the biggest discrepancy between a prediction and an observation in the history of science.
Despite this problem, dark energy was embraced relatively quickly by most astronomers. Nevertheless, by the time of the launch of WMAP, there were still many who doubted its existence. ‘How well do we really understand Type Ia supernovae?’ asked the sceptics. Perhaps they are not all the same. Perhaps they are not standard candles. ‘WMAP changed that,’ says Bennett. ‘The peaks and troughs were compatible with a Universe with precisely 73 per cent dark energy.’
As the American journal Science said in its 2003 ‘Breakthrough of the Year’ article: ‘Lingering doubts about the existence of dark energy and the composition of the Universe dissolved when the WMAP satellite took the most detailed picture ever of the cosmic microwave background.’
So now, at last, we know the exact composition of the Universe: 73 per cent dark energy, mysterious invisible stuff with repulsive gravity; 23 per cent dark matter, mysterious invisible stuff with normal gravity; and 4 per cent ordinary matter. Actually, less than half of the 4 per cent has actually been seen by astronomers with their telescopes; the rest is hidden somewhere, maybe in the form of tenuous intergalactic gas or black holes.
Perhaps the most remarkable thing is that we can be so precise about so much that is mysterious. After all, an astonishing 98 per cent of the mass of the Universe is in forms which we know pretty much nothing about.
This was not the end of WMAP’s contribution to pinning down the standard model of cosmology. Once upon a time, people used to say that the Universe was between 9 and 15 billion years old. It was WMAP that pinned it down to 13.7 billion years with an accuracy of ±1 per cent, the figure that has become standard in all cosmic discussions. ‘It even made the Guinness Book of Records for the “best determination of the age of the Universe”,’ says Bennett.
Once upon a time, people used to say that the epoch of last scattering began at a red shift of ‘about 1,000’. WMAP pinned it down to 1,098, with an astonishing accuracy of ±1. Once upon a time, people had no idea when the first stars formed after the Big Bang. WMAP found evidence of the Universe’s hydrogen being re-ionised by ultraviolet light from the first stars as early as 400 million years after the moment of creation. Earlier than anyone expected, the stars broke out like a rash across the cosmos.
Strictly speaking, the age of precision cosmology began with COBE’s measurement of the temperature of the background radiation to be an incredibly precise 2.725 degrees above absolute zero. But, ironically, there was not much more that could be done with that number. It was WMAP and its stunning characterisation of the cosmic ripples that truly ushered in the age of precision cosmology. ‘We didn’t create the idea of a “standard model” with inflation, dark matter and dark energy,’ says Bennett. ‘But we set the precise value of all the relevant numbers.’
‘The amount of knowledge about our Universe that they extracted from the infinitesimal temperature differences was – and is – just amazing to me,’ says his wife, Renee.
After the first year of WMAP data, there was a second year, and a third and a fourth, each improving on the precision of its measurements. The experiment exceeded all expectations. ‘Oh, we expected to measure what we measured,’ says Bennett. ‘But even we were stunned at the level of precision we obtained.’
For Bennett, all the incredible hard work, all the stress – even the gallstone – had been worth it. Like childbirth, most of the pain was forgotten in the euphoric aftermath. ‘I really can’t think of anything else I would rather have done with my life,’ he says. ‘It was a privilege to work on what we worked on and it was a privilege to work with the people I worked with.’
Bennett and his colleagues had done more in a decade to change our picture of the Universe than had been achieved during the rest of human history. For the first time it was possible to ask truly fundamental questions about the Universe and have a good chance of answering them in the near future. What was the Big Bang? What drove the Big Bang? What happened before the Big Bang? Why is there something rather than nothing – surely the ultimate question? Cosmology had never been in better shape.
Or so it seemed.
Although WMAP had succeeded in bolstering the standard model of cosmology, it had also thrown up a number of puzzles. ‘Nobody yet knows whether they are significant,’ says Bennett. ‘But nobody can quite dismiss them out of hand.’
The hot spots and cold spots in the cosmic background radiation mark locations in the early Universe where matter was slightly denser than average or more rarefied than average. As pointed out before, these were the ‘seeds’ from which would eventually sprout the great clusters of galaxies we see around us today. According to inflation, these seeds grew from ‘quantum fluctuations’, seething convulsions of space–time in the first split second of the Universe’s existence. Far smaller than a present-day atom, they had then been enormously magnified in size by the tremendous force of inflation.
This picture made a key prediction: since quantum fluctuations were inherently random, the hot spots in the cosmic background should be scattered about the sky completely randomly, no matter what their size.
But this was not what was seen. The biggest temperature splotches – technically referred to as the ‘quadrupole’ and ‘octupole’ moments – appeared not to be randomly distributed but instead aligned with each other. The physicist João Magueijo dubbed the direction along which they were aligned the ‘axis of evil’, and it stuck.
Recall that the photons and nuclei sloshing about in the early Universe were like water in a bath. If the bath – or the Universe – were far smaller in one direction than in others, then it might channel the sloshing modes, causing them to align with each other. Some suggested that the Universe might extend further in two directions than a third – that it might look like a flattened CD. Others even suggested that the simplest Big Bang models might be wrong.
Extraordinary claims require extraordinary evidence. ‘And the truth is we do not really know how extraordinary is the alignment of the axis of evil,’ says Bennett. ‘It could easily just be pure chance.’
There are several other WMAP puzzles, but perhaps the most interesting is the giant cold spot which appears in the cosmic background maps of the southern sky. Some radio astronomers have claimed that it coincides with a giant void, far more empty of galaxies than the surrounding space. But this is disputed. If it is a giant void, then it poses a big problem for cosmology. In inflation, under-dense regions, like over-dense ones, are the result of quantum fluctuations. But whereas small fluctuations are very likely, bigger ones leading to large cosmic voids are less so. And the giant void seen in the WMAP map is very unlikely indeed.
One mind-blowing possibility is connected with inflation. According to the theory, in the beginning there is the inflationary vacuum, growing ever faster but empty of anything except energy. All over the vacuum bits start decaying randomly – a region over here, a region over there, tiny bubbles breaking out all over the inflationary vacuum. The bubbles are normal vacuum and, inside each, the enormous energy of the vacuum is dumped into the creation of matter and into heating it to a fantastically high temperature. It creates a Big Bang universe just like our own.
It is a key feature of such inflation that it never stops – that it is ‘eternal’. The vacuum grows so fast that it is created more quickly than it is eaten away as decaying bubbles. Consequently, each Big Bang bubble universe rapidly recedes from every other, isolated for ever in an endless sea of nothingness. But what if in the inflationary vacuum two or more bubbles formed together? What if, before being dragged apart, they collided? Might they leave a mark on each other? An imprint? A cosmic fingerprint?
Could this be what the anomalous WMAP cold spot is? Is it the first evidence of the existence of another universe beyond our own? ‘Obviously, I am very cautious,’ says Bennett. ‘But, certainly, an imprint of this kind might be the most compelling evidence yet of inflation and the existence of other universes.’
In 2006 came the icing on the cake for Bennett and his colleagues. The Nobel Committee announced that the Nobel Prize for Physics had been awarded to John Mather and George Smoot ‘for their discovery of the black body form and anisotropy of the cosmic microwave background radiation’. ‘All of us who worked on COBE were extremely proud,’ says Bennett. ‘We saw the prize as for the science accomplished by the whole team.’
There were niggles, however. COBE carried three experiments into space, so there were three principal investigators, the third being Mike Hauser. ‘It would have been better if they had given the prize to all three of them,’ says Bennett. ‘Certainly John Mather, the leader of COBE and the person who instigated the whole thing back in 1974, was a no-brainer for the Nobel Prize.’
No longer was there any real animosity towards Smoot. However, it was clear from the moment COBE hit the world headlines in 1992 that there was likely to be a Nobel Prize for its discoveries, just as there had been for the discovery of the cosmic background radiation in 1965. Mather, as the father of the project, was the obvious choice. But then the prize was often awarded to more than one person. A large number of people worked on COBE, so who might share the prize with Mather? Smoot, that’s who. ‘From the beginning, George made a concerted effort to separate himself from the crowd,’ says Bennett. ‘There is no doubt about it. He campaigned long and hard to get the Nobel Prize.’
There is an interesting contrast here to be made with Bennett. In 2003, after the press conference at which the first WMAP results were triumphantly announced to the world, Bennett received a phone call from literary agent John Brockman, the man who reportedly got Smoot a $2-million advance for what became the book Wrinkles in Time. Did Bennett want to write a book? asked Brockman.
Bennett knew he could draw great attention to himself personally. ‘All I had to do was write a book about WMAP and spend the next couple of years travelling around the world, promoting myself, as Smoot had done,’ he says. But Bennett declined Brockman’s offer and passed on talk invitations to other members of the science team. ‘I would have ended up not doing science,’ he says. ‘And science is what I love doing.’
Given the chance, Bennett would do WMAP all over again. ‘I’ve been extraordinarily lucky in my life. Most people don’t get to do one space mission – I got to do two,’ he says. ‘And the best thing is, I didn’t fuck up!’