Why might the Big Bang theory be in crisis very soon?

Did the Universe have a beginning? Will it eventually come to an end? How did the Universe evolve into what we can see today: a ‘cosmic web’ of stars, galaxies, planets and, at least on one pale blue planet, what sometimes passes for intelligent life?

Not so very long ago, these kinds of existential questions were judged to be without scientific answers. Yet scientists have found some answers, through more than a century of astronomical observations and theoretical developments that have been woven together to give us the Big Bang theory of cosmology. This extraordinary theory is supported by a wide range of astronomical evidence, is broadly accepted by the scientific community, and has (a least by name) become embedded in popular culture.

We shouldn’t get too comfortable. Although it tells an altogether remarkable story, the current Big Bang theory leaves us with many unsatisfactorily unanswered questions, and recent astronomical observations threaten to undermine it completely. The Big Bang theory may very soon be in crisis.

To understand why, it helps to appreciate that there is much more to the theory than the Big Bang itself. That the Universe must have had a historical beginning was an inevitable consequence of concluding that the space in it is expanding. In 1929, observations of distant galaxies by the American astronomer Edwin Hubble and his assistant Milton Humason had produced a remarkable result. The overwhelming majority of the galaxies they had studied are moving away from us, at speeds directly proportional to their distances. To get some sense of these speeds, imagine planet Earth making its annual pilgrimage around the Sun at a sedate orbital speed of about 30 kilometres per second. Hubble and Humason found galaxies moving away at tens of thousands of kilometres per second, representing significant fractions of the speed of light.

Hubble’s speed-distance relation had been anticipated by the Belgian theorist Georges Lemaître a few years before and is today known as the Hubble-Lemaître law. The constant of proportionality between speed and distance is the Hubble constant, a measure of the rate at which the Universe is expanding. In truth, the galaxies are not actually moving away at such high speeds, and Earth occupies no special place at the centre of the Universe. The galaxies are being carried away by the expansion of the space that lies between us, much as two points drawn on a deflated balloon will move apart as the balloon is inflated. In a universe in which space is expanding, everything is being carried away from everything else.

The Big Bang story is almost as fascinating as the story of the Universe itself

To get a handle on the distances of these galaxies, astronomers made use of so-called Cepheid variable stars as ‘standard candles’, cosmic lighthouses flashing on and off in the darkness that can tell us how far away they are. But in the late-1920s, these touchstone stars were poorly understood and the distances derived from them were greatly underestimated, leading scientists to overestimate the Hubble constant and the rate of expansion. It took astronomers 70 years to sort this out.

But such problems were irrelevant to the principal conclusion. If space in the Universe is expanding, then extrapolation backwards in time using known physical laws and principles suggests there must have been a moment when the Universe was compressed to a point of extraordinarily high density and temperature, representing the fiery origin of everything: space, time, matter, and radiation. As far as we can tell, this occurred nearly 14 billion years ago. In a BBC radio programme broadcast in 1949, the maverick British astronomer Fred Hoyle called this the ‘Big Bang’ theory. The name stuck.

Of course, it’s not enough that the theory simply tells us when things got started. We demand more. We also expect the Big Bang theory to tell the story of our universe, to describe how the Universe evolved from its beginning, and how it came to grow into the cosmic web of stars and galaxies we see today. The theorists reduced this to a simple existential question: Why do stars and galaxies exist? To give a proper account, the Big Bang theory has itself evolved from its not-so-humble beginnings, picking up much-needed additional ingredients along the way, in a story almost as fascinating as the story of the Universe itself.

The Big Bang theory is a theory of physical cosmology, constructed on foundations derived from solutions of Albert Einstein’s equations of general relativity – in essence, Einstein’s theory of gravity – applied to the whole universe. Einstein himself had set this ball rolling in 1917. At that time, he chose to fudge his own equations to obtain a solution describing an intellectually satisfying static, eternal universe. Ten years later, Lemaître rediscovered an alternative solution describing an expanding universe. Although Einstein rejected this as ‘quite abominable’, when confronted by the evidence presented by Hubble and Humason, he eventually recanted.

Working together with the Dutch theorist Willem de Sitter, in 1932 Einstein presented a new formulation of his theory. In the Einstein-de Sitter universe, space is expanding, and the Universe is assumed to contain just enough matter to apply a gentle gravitational brake, ensuring that the expansion slows and eventually ceases after an infinite amount of time, or so far into the future as to be of no concern to us now. This ‘critical’ density of matter also ensures that space is ‘flat’ or Euclidean, which means that our familiar schoolroom geometry prevails: parallel lines never cross and the angles of a triangle add up to 180 degrees. Think of this another way. The critical density delivers a ‘Goldilocks’ universe, one that will eventually be just right for human habitation. By definition, the Einstein-de Sitter expanding universe is an early version of the Big Bang theory. It formed the basis for cosmological research for many decades.

But problems begin as soon as we try to use the Einstein-de Sitter version to tell the story of our own universe. It just doesn’t work.

If the post-Big Bang universe had expanded just a fraction faster or slower, stars and galaxies would not have formed

Applying Einstein’s equations requires making a few assumptions. One of these, called the cosmological principle, assumes that on a large scale the Universe is homogeneous (the same everywhere) and isotropic (uniform in all directions). But if this were true of our universe in its very earliest moments following the Big Bang, matter would have been spread uniformly in all directions. This is a problem because if gravity pulled equally on all matter in all directions, then nothing would move and so no stars or galaxies could form. What the early universe needed was a little anisotropy, a sprinkling of regions of excess matter that would serve as cosmic ‘seeds’ for the formation of stars and galaxies. Such anisotropy could not be found in the Einstein-de Sitter universe. So where had it come from?

Matters quickly got worse. Theorists realised that getting to the Universe we see from the Big Bang of the Einstein-de Sitter version demanded an extraordinary fine-tuning. If the immediate, post-Big Bang universe had expanded just a fraction faster or slower, then stars and galaxies would have never had a chance to form. This fine-tuning was traced to the critical or ‘Goldilocks’ density of matter. Deviations from the critical density of just one in 100 trillionth – higher or lower – would have delivered universes very different from our own, in which there would be no intelligent life to bear witness.

It got worse: theoretical studies of the formation of spiral galaxies and observational studies of the rotational motions of their stars led to another distinctly uncomfortable conclusion. Neither could be explained by taking account of all the matter that we can see. Calculations based only on the visible matter of stars suggested that, even if conditions allowing their formation could be met, spiral galaxies should still be physically impossible, and the patterns of rotation of the stars within them should look very different. To add insult to injury, when astronomers added up all the matter that could be identified in all the visible stars and galaxies, they found only about 5 per cent of the matter required for the critical density. Where was the rest of the Universe? There was clearly more to our universe than could be found in the Einstein-de Sitter version of the Big Bang theory.

The solutions to some of these problems could be found only by looking back to the very beginning of the story of the Universe and, as this is not a moment that is accessible to astronomers, it fell once more to the theorists to figure out what might have happened.

In the early 1980s, a group of theorists concluded that, in its very earliest instants, the post-Big Bang universe would have been small enough to be subject to random quantum fluctuations – temporary changes in the amount of energy present at specific locations in space, governed by Werner Heisenberg’s uncertainty principle. These fluctuations created tiny concentrations of excess matter in some places, leaving voids in others. These anisotropies would have then been imprinted on the larger universe by an insane burst of exponential expansion called cosmic inflation. In this way, the tiny concentrations of matter would grow to act as seeds from which stars and galaxies would later spring. To a certain extent, cosmic inflation also fixed some aspects of the fine-tuning problem. It was like a blunt instrument: no matter what conditions might have prevailed at the very beginning, cosmic inflation would have hammered the Universe into the desired shape.

The theorists also reasoned that the hot, young universe would have behaved like a ball of electrically charged plasma, more fluid than gas. It would have contained matter stripped right back to its elementary constituents, assembling atomic nuclei and electrons only when temperatures had cooled sufficiently as a result of further expansion. They understood that there would have been a singular moment, just a few hundred thousand years after the Big Bang, when the temperature had dropped low enough to allow positively charged atomic nuclei (protons and helium nuclei) and negatively charged electrons to combine to form neutral hydrogen and helium atoms. This moment is called recombination.

The light that would have danced back and forth between the charged particles in the ball of plasma was released, in all directions through space, and the Universe became transparent: literally, a ‘let there be light’ moment. Some of this light would have been visible, though there was obviously nobody around to see it. This is the oldest light in the Universe, known as the cosmic background radiation.

Much like a bloody thumbprint at a cosmic crime scene, it left a pattern of temperature variations across the sky

This radiation would have cooled further as the Universe continued to expand: estimates in 1949 suggested it would possess today a temperature of about 5 degrees above absolute zero (or -268^oC), corresponding to microwave and infrared radiation. This estimate was largely forgotten, only to be rediscovered in 1964. A year later, as physicists scrambled to build an apparatus to search for it, the American radio astronomers Arno Penzias and Robert Wilson found it by accident. This discovery changed everything. The cosmic background, witness to events that had occurred when the Universe was in its infancy, was getting ready to testify.

The tiny concentrations of matter produced by quantum fluctuations and imprinted on the larger universe by cosmic inflation would have made the cosmic background very slightly hotter in some places compared with others. This left a pattern of temperature variations in the cosmic background across the sky, much like a bloody thumbprint at a cosmic crime scene.

These small temperature variations were detected by an instrument aboard NASA’s Cosmic Background Explorer satellite, and were reported in 1992. George Smoot, who had led the project to detect them, struggled to find superlatives to convey the importance of the discovery. ‘If you’re religious,’ he said, ‘it’s like seeing God.’ The evidence was in. We owe our very existence to anisotropies in the distribution of matter created by quantum fluctuations in the early, post-Big Bang universe, impressed on the larger universe by cosmic inflation.

But cosmic inflation could not fix the problems posed by the physics of galaxy formation and the rotations of stars, and it could not solve the problem posed by the missing density. Curiously, part of the solution had already been suggested by the irascible Swiss astronomer Fritz Zwicky in 1933. His efforts had been forgotten, only to be rediscovered in the 1970s. Galaxies are much larger than they appear, suggesting that there must exist a form of invisible matter that interacts only through its gravity. Zwicky had called it dunkle Materie: dark matter. Each spiral galaxy, including our own Milky Way, is shrouded in a halo of dark matter that was essential for its formation, and explains why stars in these galaxies rotate the way they do.

This was an important step in the right direction, but it was not enough. Even with dark matter estimated to be five times more abundant in the Universe than ordinary visible matter, about 70 per cent of the Universe was still missing.

Astronomers now had pieces of evidence from the very earliest moment in the history of the Universe, and from objects much later in this history. The cosmic background radiation is about 13.8 billion years old. But nearby galaxies whose distances can be measured using Cepheid variable stars are much younger. We can get some sense of this by acknowledging that light does not travel from one place to another instantaneously. It takes time. It takes light eight minutes to reach us from the surface of the Sun, so we see the Sun as it appeared eight minutes ago, called a ‘look-back’ time. But the Cepheids are individual stars, so their use as standard candles is limited to nearby galaxies with short look-back times of hundreds of millions of years. To reconstruct the story of the Universe, astronomers somehow had to find a way to bridge the huge gulf between these points in its history.

It is possible to study more distant galaxies but only by observing the totality of the light from all the stars contained within them. Astronomers realised that when an individual star explodes in a spectacular supernova it can light up an entire galaxy for a brief period, showing us where the galaxy is and how fast it is being carried away by expansion. Look-back times could be extended from hundreds to thousands of millions of years. A certain class of supernova offered itself as a standard candle, and the distances to their host galaxies could be calibrated by studying supernovae in nearby galaxies that possessed one or more Cepheid variable stars.

The expectation was that, following the Big Bang, the rate of expansion of the Universe would have slowed over time, reaching the rate as we measure it today using the Hubble-Lemaître law. According to the Einstein-de Sitter version, it would continue to decelerate into the future, eventually coming to a halt. But when astronomers started using supernovae as standard candles in the late 1990s, what they discovered was truly astonishing. The rate of expansion is actually accelerating.

Further data suggested that the post-Big Bang universe had indeed decelerated, but about 5 billion years ago this had flipped over to acceleration. In a hugely ironic twist, the fudge that Einstein had introduced in his equations in 1917 only to abandon in 1932 now had to be put back. Einstein had added an extra ‘cosmological term’ to his equations, governed by a ‘cosmological constant’, which imbues empty space with a mysterious energy. The only way to explain an accelerating expansion was to restore Einstein’s cosmological term to the Big Bang theory. The mysterious energy of empty space was called dark energy.

I like to think of this as a period when the Universe was singing

In 1905, Einstein had demonstrated the equivalence of mass (m) and energy (E) through his equation E = mc², where c is the speed of light. It might come as no surprise to learn that when the critical density of matter is expressed instead as a critical density of mass-energy, dark energy accounts for the missing 70 per cent of the Universe. It may also seal its ultimate fate. As the Universe continues to expand, more and more of it will disappear from view. And, as the Universe grows colder, the matter that remains in reach may be led inexorably to a ‘heat death’.

How do we know? More answers could be found in the cosmic background radiation. The theorists had further reasoned that competition between gravity and the enormous pressure of radiation in the post-Big Bang ball of plasma would have triggered acoustic oscillations – sound waves – wherever there was an excess of matter. These would have been sound waves propagating at speeds of more than half the speed of light, so even if there had been someone around who could listen, these were sounds that could not have been heard. Nevertheless, I still like to think of this as a period when the Universe was singing.

The acoustic oscillations left tell-tale imprints in the temperature of the cosmic background, and in the large-scale distribution of galaxies across the Universe. These imprints cannot be modelled without first assuming a specific cosmology, in this case the Big Bang theory including dark matter and dark energy. Modelling results reported in 2013 tell us what kind of universe we live in – its total density of matter and energy, the shape of space, the nature and density of dark matter, the value of Einstein’s cosmological constant (and hence the density of dark energy), the density of visible matter, and the rate of expansion today (the Hubble constant). This is how we know.

But the story is not over yet. Astronomers continued to sharpen their understanding of the history of the Universe through further studies of Cepheids and supernovae using the Hubble Space Telescope. Because these are studies based on the use of standard candles to measure speeds and distances, they provide measurements of the Hubble constant and the rate of expansion later in the Universe’s history that do not require the presumption of a specific cosmology. The Hubble constant and rate of expansion deduced from analysis of the acoustic oscillations is necessarily a model-dependent prediction, as it is derived from events much earlier in the Universe’s history. For a time, prediction and measurement were in good agreement, and the Big Bang theory looked robust.

Then from about 2010 things started to go wrong again. As the precision of the observations improved, the predictions and the measurements went separate ways. The difference is small but appears to be significant. It is called the Hubble tension. The Universe appears to be expanding a little faster than we would predict by modelling the acoustic oscillations it experienced in infancy. Imagine constructing a bridge spanning the age of the Universe, begun simultaneously on both ‘early’ and ‘late’ sides of the divide. Foundations, piers and bridge supports have been completed, but the engineers have now discovered to their dismay that the two sides do not quite meet in the middle.

Matters have been complicated by the development of different kinds of standard candle that are a little more straightforward to analyse than the Cepheids, and rival teams of astronomers are currently debating the details. We should know in another couple of years if the tension is real. And if it is real, then one way to fix it is to tweak the Big Bang theory yet again by supposing that dark energy has weakened over time, equivalent to supposing that Einstein’s cosmological constant is not, in fact, constant. Some tentative evidence for this was published in March this year.

And there is yet more trouble ahead. The James Webb Space Telescope, launched on Christmas Day in 2021, can see galaxies with look-back times of more than 13 billion years, reaching back to a time just a few hundred million years after the Big Bang. Our understanding of the physics based on the current theory suggests that, at these look-back times, we might expect to see the first stars and galaxies in the process of forming. But the telescope is instead seeing already fully formed galaxies and clusters of galaxies. It is too soon to tell if this is a crisis, but there are grounds for considerable uneasiness.

Some cosmologists have had enough. The Big Bang theory relies heavily on several concepts for which, despite much effort over the past 20 to 30 years, we have secured no additional empirical evidence beyond the basic need for them. The theory is remarkably successful yet full of explanatory holes. Cosmic inflation, dark matter and dark energy are all needed, but all come with serious caveats and doubts. Imagine trying to explain the (human) history of the 20th century in terms of the societal forces of fascism and communism, without being able to explain what these terms mean: without really knowing what they are, fundamentally.

In an open letter published in New Scientist magazine in 2004, a group of renegade cosmologists declared:

In no other field of physics would this continual recourse to new hypothetical objects be accepted as a way of bridging the gap between theory and observation. It would, at the least, raise serious questions about the validity of the underlying theory.

This is simply the scientific enterprise at work. Answers to some of our deepest questions about the Universe and our place in it can sometimes appear frustratingly incomplete. There is no denying that, for all its faults, the present Big Bang theory continues to dominate the science of cosmology, for good reasons. But the lessons from history warn against becoming too comfortable. There is undoubtedly more to discover about the story of our Universe. There will be more surprises.

The challenges are, as always, to retain a sense of humility in the face of an inscrutable universe, and to keep an open mind. As Einstein once put it: ‘The truth of a theory can never be proven, for one never knows if future experience will contradict its conclusions.’