Chapter 4: A cosmic debate
The big bang vs the steady-state model
In this chapter we encounter a new view of the universe, the steady-state model. A great debatearose between the big bang and steady-state models, an argument that was resolved by the advent of radio-astronomy and the discovery of the cosmic microwave background .
The application of nuclear physics to Lemaitre’s theory of the early universe led Gamow et al. to a model of a universe that started as a hot, dense primordial soup of elementary particles and radiation (now known as the hot big bang model). This model made two new predictions; a universe that is dominated by the lightest elements hydrogen and helium, and a universe that might contain remnants of primordial cosmic radiation.
The steady-state universe
We saw in the last chapter that the study of cosmology went into sharp decline over the next fifteen years. However, this is not to say that physicists abandoned the subject altogether. In fact, a completely new model of the universe emerged in Britain in the late 1940s. At Cambridge University, a trio of physicists, Fred Hoyle, Hermann Bondi and Thomas Gold, became interested in a model of an expanding universe that did not originate in a primeval fireball.
Like Gamow, Fred Hoyle was a brilliant nuclear scientist who was also a rather colourful character. The son of a Yorkshire wool merchant, he went up to Cambridge as a scholarship student. In 1936, he won the Mayhew Prize in the Cambridge Mathematical Tripos and went on to a glittering postgraduate career, culminating in his election to a fellowship at St John's College in 1939. Hoyle was a noticeable figure around Cambridge as he was rather argumentative and pugnacious, with a marked tendency to rub polite Cambridge dons up the wrong way.
As a result of war work, Hoyle became friends with Bondi and Gold, two Austrian refugees from Hitler’s Europe who were also at Cambridge. The trio had a common interest in astronomy and cosmology and they often discussed developments in the field together. All three were familiar with the works of Lemaitre and the Gamow group, but they were unconvinced by their models of the early universe. The Cambridge physicists were particularly interested in the problem of the age of the universe; it seemed to them that Lemaitre’s use of the cosmological constant to overcome the problem was quite contrived. An emergingspecialist in nuclear physics, Hoyle was also unimpressed by the model of primordial nucleosynthesis of Gamowet al., pointing out that their theory could only account for the lightest elements. Finally, there was the old problem of the singularity; if the universe really began as the Friedmann/Lemaitre model suggested, at what point did the laws of physics become the laws we know today?
These questions led Hoyle and his colleagues to consider a very different model of the universe. The catalyst was when all three viewed the film The Dead of Night in a Cambridge cinema. The film features a plot that repeats itself endlessly (not unlike the more recent American film Groundhog Day) and it led Gold to a daring hypothesis: what if the universe also cyclic? The trio set to work on the idea, assuming at first that it would be easily ruled out1.
Many hours later, it seemed the question was not trivial. The core of the question was whether an expanding universe could somehow remain essentially the same, just as a river is unchanging but not static. A key characteristic of such a ‘steady-state’ universe would be that the density of matter remains constant – in contrast with the evolving model of Lemaitre where the density of matter decreases rapidly as space expands. But how could this happen? Hoyle’s daring insight was that to suppose that if matter is continually created, one could have a universe that is expanding but not changing – and such a steady-state universe need not have an origin. Of course, the continuous creation of matter might seem a rather far-fetched idea, but Hoyle was able to show that the amount of matter needed is extremely small – one atom of hydrogen for every cubic meter of space2!
The steady-state model extended Einstein’s cosmological principle (that the universe is both homogeneous and isotropic on the largest scales) to a perfect cosmological principle – that the universe is also the same at all times. This viewwas nicely in line with classical views in science and philosophy of an eternal universe3.In addition, the steady-state model avoided the empirical problem of the age of the universe, and the theoretical problem of thesingularityin the Friedmann-Lemaitre models. Finally, the new modeladdressed an old puzzle concerning the expansion;although relativity predicts an expanding universe, the physical reason for the expansion is notobvious. In the model of Hoyle et al.,it is the process of continuous creation that forces space to expand in order to make room for new matter4.
Hoyle became convinced that he and his colleagues were on the right track, and in consequence he set about an analysis of stellar nucleosynthesis(the formation of the elements by nuclear fusion processes in the stars) as an alternative to the early-universe nucleosynthesis of the Gamow group. He made several important advances in this field during the 1950s. In particular, he came up with a brilliant solution to the riddle of how carbon is formed in the stars, a problem that had dogged the field for years5. The result was atheory that successfully described how the heavier elements are formed in stars and supernovae, a model that is still in use today(Burbidge, Burbidge Hoyle and Fowler 1954).
A cosmic debate
The steady-state model emergedsoon after the model of the Gamow group and it made some impact amongst the small community of relativists, astronomers and physicists interested in cosmology. Here was a model that avoided the need for a psychotic beginning for the universe, gave a physical explanation for the expansion of space and could explain the formation of most of the chemical elements in terms ofstellar processes. Coupled with the problem of the age of the universe, serious doubts were raised concerning Gamow’s white-hot infant universe. A gifted science communicator, Hoyle lost no opportunity to promote his own model. Indeed, it was he who first coined the term ‘big bang’ in a comparison of the two models on BBC radio. The term stuck, although it is one of science’s great misnomers; as we saw in chapter 3, the model of the Gamow groupsays nothing about the bang itself.
The debate between a big bang and a steady-state universe lasted more than a decade. It probably helped revive interest in cosmology as it is the sort of debate that scientists like best. After all, the universe is either changing in time (big bang model) or it isn’t (steady-state). In particular, any evidence that our universe was in fact different in the past would effectively rule out the steady-state model. This principle of falsifiability is very important in science; as the science philosopher Karl Popperpointed out, science mainly progresses by ruling things out6. It was soon realised that astronomy couldprovide the answer – as the great telescopes gaze at the most distant objects in the sky, theyalso look back in timebecause of thefinite time it takes light to travel vast distances. By comparing measurements of the most distant galaxies withthose closeby, could astronomers settle the debate?
Astronomy to the rescue once more
They could,but not before another importantdiscovery was made. In 1952, Walter Baade, Hubble’s successor at the Mount Wilson observatory, announced that Hubble’s original measurements of stellar distance contained a significant systematic error. Hubble had underestimated the distances to the galaxies by at least a factor of two! (The problem was that there are two different types of Cepheid variable stars, a fact Hubble was unaware of). By 1956, further work by Humason, Mayall and Sandage suggested a Hubble constant almost three times smaller than that estimated in 1929. In consequence, the Hubble graph now predicted an age of at least 6 billion years for the universe, in reasonable agreement with the age of the stars as estimated from astrophysical processes. The paradox of the age of the universe that had so bedevilled the big bang model was no more!
At around the same time, the advent of radio astronomy (where physicists study the sky at radio rather than optical wavelengths) allowed astronomersto peer deeper into space than ever before. With the great cosmic debate above in mind, the Cambridge physicist Martin Ryle set about cataloguing all the new radio sources that were being discovered in the sky. By 1955, it seemed that the number of these sources was significantly higher in the furthest reaches of space. This was the first tentative evidence that the early universe was indeed different from that of the present. However, the results were somewhat controversial as the nature of the radio-sources was not fully understood. More detailed studies undertaken in 1959 and 1962 made it clear that Ryle’s results were essentially correct. By the early 1960s, there was compelling evidence that there is an excess of radio sources at the largest distances observable – in clear contradiction with the predictions of the steady-state model. The importance of this finding was recognized when Ryle and his Cambridge colleague Anthony Hewish became the first astronomers to win the Nobel prize in 1974.
An interesting spinoff of the radio-astronomy program was the discovery of quasars – bright sources at extreme redshifts, indicating incredibly powerful sources at extreme distances – and pulsars (stars that pulsate in an incredibly regular fashion). Again, these exotic objects were only seen in the most distant galaxies, suggesting a clear difference between the young universe and the present one. By the mid-1960s, most physicists considered that radio-astronomy offered strong support for the big bang model and cast serious doubt on the steady-state theory.Best of all, asso often happensin science, the new astronomy led to an unexpected discovery thatrevolutionized the field.
The discovery of the cosmic microwave background
In 1963, Arno Penzias and Robert Wilson, two physicists at Bell Laboratories in New Jersey who had both trained as astronomers, became interested in the problem of detecting weak signals at radio and microwave wavelengths. This problem had arisen as a result of the nascent satellite communication industry, and Penzias and Wilson set about the task of constructing an instrument that could act as a sensitive radio receiver. For this project, they used a unique 20-foot horn-shaped receiver previously used at Bell as part of the Echo satellite communications program (the giant horn shields the radio antenna from noise, see figure 6).
Using theirhighly sensitive instrument, the astronomers detected a ubiquitous, faint signal in the microwave region of the spectrum at the extremely low temperature of 3 Kelvin (radiation picked up by a radio receiver at a given wavelength is usually measured in units of temperature – the temperature at which an ideal black body emits at this wavelength). Taking the signal to be background noise, the duo spent a great deal of time trying to get rid of it. They did not succeed, and they eventually came to the conclusion that thesignal was of astronomical origin7. At this point, theyheard that a group at Princeton University were working on a theory of cosmic radiation emanating from the early universe, and they contacted the eminent theoretician Robert Dicke at Princeton.
Dicke was flabbergasted. Unaware of the earlier work of the Gamow group, he and his colleague Jim Peebles had been developing a theory of cosmic background radiation for some time, and had just reached the point where they and their colleagues were designing an experiment to search for it. The Dicke group took a trip to New Jersey, inspected the radio receiverand realized the Bellastronomers had hit the jackpot!8 The experimental findings of Penzias and Wilson were published in a historic issue of the Astrophysics Journal in 1965, next to an accompanying article by Dicke and Peebles explainingthe theoretical importance of the finding.
Figure 6 Penzias (L) and Wilson with their giant radio receiver in the background
The Princeton group soon followed up with their own measurement, with David Wilkinson and Pete Roll reporting the detection of the background radiation at a slightly different wavelength later that year. Several other experiments followed and by mid-1966 the news had spread throughout the world of physics; remnant radiation from the early universe had been found, strong evidence indeed for the big bang model. In particular, the detection of ubiquitousradiation at microwave wavelengths was in excellent accord with the relativistic picture of hot primordial radiation hugely red-shifted and cooled by the subsequent expansion of space9. Penzias and Wilson were later awarded the 1978 Nobel Prize in physics for their serendipitous discovery.
Soon, the search was on to measure the shape of the entire spectrum of the cosmic microwave background (CMB). This was an important test – if the radiation was truly of cosmic origin, it should exhibit the spectrum of a perfect black body10. For many years, this program took the form of sending delicate instruments aboard balloons above the atmosphere (to avoid interference from the atmosphere). These painstaking experiments did yield results although they were extremely difficult11. Each could record at one wavelength only and instruments often froze or malfunctioned at the freezing temperatures of the stratosphere. In time, the program gave way to a new generation of experiments where instruments were mounted on satellites that hovered far above the atmosphere. This approach scored a spectacular success in 1992, when instruments aboard the COBE satellite gave the first accurate measurement of the full spectrum of the cosmic radiation – the spectrum was a perfect fit to that of a black-body, confirming the primordial nature of the radiation (figure 7). Today, much of modern cosmology is concerned with the study of the CMB with ever more precision, using more and more sophisticated telescopes mounted on satellites.
Figure 7 Spectrum of the cosmic microwave background measured by the FIRAS instrument on the COBE satellite. Squares are experimental points while the solid curve is the black body spectrum predicted by theory.
On the philosophy of science
The discovery of a new scale for the Hubble graph (removing the problem of the age of a big bang universe) and the radio-source surveys of Ryle and others (indicating that the universe was different in the past) were significant triumphs for the big bang model, but it was the detection of the cosmic microwave background that clinched the deal. Thefinding marked a new era in cosmology; the evidence for a hot early phase of the universe was convincing and the study of the origin of the universe moved to centre stage in the world of physics. Cosmology was no longer an abstract, speculative subject confined to relativists and a few astronomers, but a vibrant field of science open to enquiry by astrophysicists, nuclear physicists, particle physicists and everyone else.
That said,such changesin science do not happen overnight. Alternate explanations for the background radiation were offered for some time, but none proved convincing. Meanwhile, more sophisticated versions of the steady-state model were developed; however, most physicists found these models very contrived. As the evidence accumulated, the big bang scenario seemed more and more plausible and the steady-state theory less and less so.This is the way science progresses; not by abrupt changes in world view but by a gradual process, much like a group of observers agreeing on the nature of a distant object that is gradually coming into view. At first, several possibilities are tenable, but as the object approaches, one becomes more and more likely while others are gradually ruled out. Crucially, experimentalists must consider all of the models whilst this process is ongoing, letting the data speak for itself. Indeed, long after a particular view has become dominant, it is standard practice to considernew data in the context of all the main theories, not just the current favourite. For the experimentalist, the case is never truly closed, an approachthat is in contrast with Thomas Kuhn’s view of how paradigm shifts occurin science12(see chapter 1).It should be noted thatKuhn was an eminent historian and philosopher but he was not a noted physicist, and his views are more popularin the social sciences than amongst practising scientists.