A HISTORICAL DIVERSION BY STEVEN WEINBERG PART TWO

It does not seem to me to be historically very important to pinpoint the moment when the detection of a 3° K isotropic microwave background became possible. The important point is that the radio astronomers did not know that they ought to try! In contrast, consider the history of the neutrino. When the neutrino was first hypothesized by Pauli in 1932, it was clear that there was not a shadow of a chance of observing it in any experiment then possible. However, the detection of neutrinos remained on physicists' minds as a challenging goal, and when nuclear reactors became available for such purposes in the 1950s the neutrino was sought and found. The contrast is even sharper with the case of the antiproton. After the positron was discovered in cosmic rays in 1932, theorists generally expected that the proton as well as the electron ought to have an antiparticle. There was no chance of producing antiprotons with the early cyclotrons available in the 1930s, but physicists remained aware of the problem, and in the 1950s an accelerator (the Bevatron at Berkeley) was built specifically to have enough energy to be able to produce antiprotons. Nothing of the sort happened in the case of the cosmic microwave radiation background, until Dicke and his associates set out to detect it in 1964. Even then, the Princeton group were not aware of the work of Gamow, Alpher, and Herman more than a decade earlier!

What, then, went wrong? It is possible here to trace at least three interesting reasons why the importance of a search for a 3° K microwave radiation background was not generally appreciated in the 1950s and early 1960s.

First, it must be understood that Gamow, Alpher, Herman, and Follin, et al., were working in the context of a broader cosmogonical theory. In their "big bang" theory essentially all complex nuclei, not only helium, were supposed to be built up in the early universe by a process of rapid addition of neutrons. However, although this theory correctly predicted the ratios of the abundances of some heavy elements, it ran into trouble in explaining why there are any heavy elements at all! As already mentioned, there is no stable nucleus with five or eight nuclear particles, so it is not possible to build nuclei heavier than helium by adding neutrons or protons to helium (He4) nuclei, or by fusing pairs of helium nuclei. (This obstacle was first noted by Enrico Fermi and Anthony Turkevich.) Given this difficulty, it is easy to see why theorists were also unwilling even to take seriously the calculation of helium production in this theory.

The cosmological theory of element synthesis lost more ground as improvements were made in the alternative theory, that elements are synthesized in stars. In 1952 E. E. Salpeter showed that the gaps at nuclei with five or eight nuclear particles could be bridged in dense helium-rich stellar cores: collisions of two helium nuclei produce an unstable nucleus of beryllium (Be8), and under these conditions of high density the beryllium nucleus may strike another helium nucleus before it decays, producing a stable carbon nucleus (CIZ). (The density of the universe at the time of cosmological nucleosynthesis is much too low for this process to occur then.) In 1957 there appeared a famous paper by Geoffrey and Margaret Burbidge, Fowler, and Hoyle, which showed that the heavy elements

could be built up in stars, particularly in stellar explosions such as supernovas, during periods of intense neutron flux. But even before the 1950s there was a powerful inclination among astrophysicists to believe that all elements other than hydrogen are produced in stars. Hoyle has remarked to me that this may have been an effect of the struggle that astronomers had to make in the early decades of this century to understand the source of the energy produced in stars. By 1940 the work of Hans Bethe and others had made it clear that the key process was the fusion of four hydrogen nuclei into one helium nucleus, and this picture had led in the 1940s and 1950s to rapid advances in the understanding of stellar evolution. As Hoyle says, after all these successes, it seemed to many astrophysicists to be perverse to doubt that the stars are the site of element formation.

But the stellar theory of nucleosynthesis also had its problems. It is difficult to see how stars could build up anything like a 25-30 percent helium abundance—indeed, the energy that would be released in this fusion would be much greater than stars seem to emit over their whole lifetime. The cosmo- logical theory gets rid of this energy very nicely—it is simply lost in the general red shift. In 1964 Hoyle and R. J. Tayler pointed out that the large helium abundance of the present universe could not have been produced in ordinary stars, and they carried out a calculation of the amount of helium that would have been produced in the early stages of a "big bang," obtaining an abundance of 36 percent by weight. Oddly enough, they fixed the moment at which nucleosynthesis would have occurred at a more or less arbitrary temperature of 5,000 million degrees Kelvin, despite the fact that this assumption depends on the value chosen for a then unknown parameter, the ratio of photons to nuclear particles. Had they used their calculation to estimate this ratio from the observed helium abundance, they could have predicted a present microwave radiation background with a temperature roughly of the right order of magnitude. Nevertheless, it is striking that Hoyle, one of the originators of the steady-state theory, was willing to follow through this line of reasoning, and to acknowledge that it provided evidence for something like a "big bang" model.

Today it is generally believed that nucleosynthesis occurs both cosmologically and in stars; the helium and perhaps a few other light nuclei were synthesized in the early universe, while the stars are responsible for everything else. The "big bang" theory of nucleosynthesis, by trying to do too much, had lost the plausibility that it really deserved as a theory of helium synthesis.

Second, this was a classic example of a breakdown in communication between theorists and experimentalists. Most theorists never realized that an isotropic 3° K radiation background could ever be detected. In a letter to Peebles dated June 23, 1967, Gamow explained that neither he nor Alpher and Herman had considered the possibility of the detection of radiation left over from the "big bang," because at the time of their work on cosmology, radio astronomy was still in its childhood. (Alpher and Herman inform me, however, that they did in fact explore the possibility of observing the cosmic radiation background with radar experts at Johns Hopkins University, the Naval Research Laboratory, and the National Bureau of Standards, but were told that a radiation background temperature of 5° or 10° K was too low to be detected with the techniques then available.) On the other hand, some Soviet astrophysicists seem to have realized that a microwave background could be detected, but were led astray by the language in American technical journals. In a 1964 review article, Ya. B. Zeldovich carried out a correct calculation of the cosmic helium abundance for two possible values of the present radiation temperature, and correctly emphasized that these quantities are related because the number of photons per nuclear particle (or the entropy per nuclear particle) does not change with time. However, he seems to have been misled by the use of the term "sky temperature" in a 1961 article by E. A. Ohm in the Bell System Technical Journal to conclude that the radiation temperature had been measured to be less than i° K. (The antenna used by Ohm was the same 20-foot horn reflector that was eventually used by Penzias and Wilson to discover the microwave background!) This, together with some rather low estimates of the cosmic helium abundance, led Zeldovich tentatively to abandon the idea of a hot early universe.

Of course, at the same time that information was flowing badly from experimenters to theorists, it was also flowing badly from theorists to experimenters. Penzias and Wilson had never heard of the Alpher-Herman prediction when they set out in 1964 to check their antenna.

Third, and I think most importantly, the "big bang" theory did not lead to a search for the 3° K microwave background because it was extraordinarily difficult for physicists to take seriously any theory of the early universe. (I speak here in part from recollections of my own attitude before 1965.) Every one of the difficulties mentioned above could have been overcome with a little effort. However, the first three minutes are so remote from us in time, the conditions of temperature and density are so unfamiliar, that we feel uncomfortable in applying our ordinary theories of statistical mechanics and nuclear physics.

This is often the way it is in physics—our mistake is not that we take our theories too seriously, but that we do not take them seriously enough. It is always hard to realize that these numbers and equations we play with at our desks have something to do with the real world. Even worse, there often seems to be a general agreement that certain phenomena are just not fit subjects for respectable theoretical and experimental effort. Gamow, Alpher, and Herman deserve tremendous credit above all for being willing to take the early universe seriously, for working out what known physical laws have to say about the first three minutes. Yet even they did not take the final step, to convince the radio astronomers that they ought to look for a microwave radiation background. The most important thing accomplished by the ultimate discovery of the 3° K radiation background in 1965 was to force us all to take seriously the idea that there was an early universe.

I have dwelt on this missed opportunity because this seems to me to be the most illuminating sort of history of science. It is understandable that so much of the historiography of science deals with its successes, with serendipitous discoveries, brilliant deductions, or the great magical leaps of a Newton or an Einstein. But I do not think it is possible really to understand the successes of science without understanding how hard it is—how easy it is to be led astray, how difficult it is to know at any time what is the next thing to be done.