Ken Kellermann and I obtained PhDs in radio astronomy in the 1960s. We were both supervised by John Bolton, one of the leading scientists making early discoveries in radio astronomy. Ellen Bouton was the NRAO librarian during my period working at NRAO in the 1980s and she has built up an extensive archive of radio astronomy documents that complements Ken’s personal knowledge of many of the discoveries discussed in this book.
During the 3C 273 occultations that led to the discovery of quasars in 1962/63 I was a summer student at the CSIRO Division of Radiophysics in Australia and attended radio astronomy lectures given by the scientists who were making the discoveries. In the early 1980s I attended a talk by Derek de Solla Price at Yale and was impressed with his analysis of the exponential growth of scientific discoveries when a new field of research opens up. This sparked my interest in the role of the research environment in the discovery process. In 1983 Ken Kellermann and I organized a meeting on Serendipitous Discoveries in Radio Astronomy for the 50th anniversary of Jansky’s discovery (Kellermann and Sheets Reference Beck, Kellermann and Sheets1984). Later, I became aware of the book The Travels and Adventures of Serendipity by the renowned sociologist of science, Robert Merton, and Elinor Barber, that was written in the 1950s, but not published until 2004. This book, and additional research papers by Merton, included many examples of serendipitous discoveries in science. Merton identified a “serendipity pattern” that was an excellent match to the discoveries in radio astronomy, although none of his examples were taken from this field. Martin Harwit had reached similar conclusions in his analysis of the discoveries across all astronomy in his series of books, starting with Cosmic Discovery (Reference Harwit1981). The authors have adopted Harwit’s discovery classification criteria for their analysis in the concluding chapter.
Robert Merton emphasized that chance discoveries depend on an impressive list of estimable qualities in a scientist: enterprise, courage, curiosity, imagination, determination, assiduity, and alertness. The background stories about how these discoveries were made provide many examples that reinforce this view. There is “nothing fortuitous” in so-called serendipitous discoveries. As Pasteur famously quoted “In the field of observation, chance favours only the prepared mind.”
As the authors point out in their concluding summary, astronomy is a technique-oriented observational science. Astronomers can’t do experiments; they can only observe. In 1933, less than 100 years ago, radio technology opened a new window on the universe that revealed an incredibly rich plethora of previously unknown phenomena. The authors have written the stories behind these discoveries, taking 10 chapters to cover the entire universe. The book focuses on the nature of the discoveries and Ken Kellermann uses his personal knowledge of the experiments and the scientists involved to tell the inside stories. This is not a straight history of the astronomical developments but an exploration of the circumstances leading up to each discovery, including many stories not generally known, but which provide the background and context. These details are often excised from the standard scientific narrative.
When we say “discovery” it often does not mean a single event. Contra the conventions in science that award prizes, professional respect, and recognition to individuals, discovery can be a lengthy process involving many actors, and many kinds of contributions over a period of time. Accounts of science are seriously distorted by emphasizing individuals and presenting discoveries as single eureka moments. Examples of the difficulties in identifying and classifying discoveries include aperture synthesis, that the authors attribute to Martin Ryle in 1962 and for which he was awarded a Nobel Prize in 1974. But, as pointed out in Chapter 3, the concept was first published by Pawsey’s team in 1947 and was demonstrated by Christiansen and Warburton in 1955. As also noted in Chapter 3 and in Table 12.1 it was the use of the EDSAC electronic computer in 1962 that made aperture synthesis imaging in radio astronomy practical. So, who or what date do we record? The case for the existence of dark matter, discussed in Chapter 10, is another example of a discovery for which evidence accumulated over decades and involved many players. While radio astronomy played a decisive role by extending the HI rotation curves to the outer reaches of galaxies, the first indications had been found by Zwicky from the dynamics of galaxy clusters and by Babcock and later Rubin using optical rotation curves in the inner parts of galaxies.
The discussion in Chapter 8 on interstellar atoms and molecules detected using radio spectroscopy provides a rather different story, and one that I have not seen discussed in any detail before. While some of these spectral lines, such as the famous 21 cm hydrogen line, were predicted, the actual discovery processes were quite chaotic and rarely a direct result of the predictions. Unusually, for radio astronomy, there was intense competition and secrecy between some of the early observers of molecular lines, that even resulted in unprofessional behavior as scientists strived for the initial discovery credit. However, the field has long since matured and astrochemistry and the ALMA telescope are outcomes from this unruly beginning.
Chapter 11 discusses why telescopes are built, a very good adjunct to discussion of the nature of discoveries. The authors are in an excellent position to do this based on their 2020 publication Open Skies describing the radio telescopes constructed by the NRAO in the United States, although this does lead to a greater emphasis on the role of US radio telescopes. While discoveries are serendipitous, they do depend on the development of new technology. So, is it the telescopes, the instruments connected to the telescopes, or the data analysis that leads to a new discovery? Table 12.1 includes a column specifying the key instrument or technology involved in each discovery. Although it is conventional to assign credit to the scientists involved in the discovery, in some cases the new technology would have enabled the discovery for whoever happened to make the first observations. One extraordinary consequence, that is strongly emphasized throughout this book, is that the scientific discoveries for which facilities become famous are rarely those they were built for. Given the nature of many of the discoveries described here we can see that this outcome is not unexpected. But what is surprising is that this obvious fact has had so little influence on the discussions of future facilities and concepts like “exploring the unknown” have had little emphasis.
Chapter 11 also includes the unusual case of the Sugar Grove radio telescope, that was started by the US Defense Department but never completed. The authors provide intriguing details and an alternate interpretation of “highly classified.”
The final chapter on future discoveries is based on lessons to be learned from past discoveries and speculates on the unexpected future discoveries. The authors have a warning “beware theoreticians,” noting that they “played no role and sometimes delayed discoveries.” But obtaining a theoretical understanding of discoveries that have been made is another matter and is hugely important. Without explanation the observed effects have no context and the contribution to knowledge can’t be assessed. The visionary theoretical physicist Freeman Dyson explained “why heretics who question the dogmas are needed.” He pointed out that “scientists often make confident predictions … Their predictions become dogmas which they do not question … [but] it may sometimes happen that they are wrong.”1 A view very well corroborated in many of the stories behind the discoveries described in this book.
The distribution of the ages of the scientists making discoveries (Figure 12.4) is quite dramatic, with a sharp decline at age about 40, and the authors note that it is well known that scientists generally do their best work as young men and women. But this is also the age when researchers tend to leave basic research as they become more involved in research management, writing grant applications, and serving on committees.
The cumulative discovery plot (Figure 12.5) is particularly intriguing and begs explanation. This plot covers the entire period from the beginning of the new science “radio astronomy,” that did not exist before the first entry in 1933. The initial burst of discoveries heralded the beginning of a new technology enabled science, as was discussed by Harwit. But in this early period the small number of players with the Second World War technology would have limited the growth rate. Then as the radio wavelength technology and digital signal processing developed in many more countries and organizations, especially in the United States, more scientists were involved and there was rapid growth. This burst of discoveries in radio astronomy was certainly one of the most dramatic periods of discovery in all the physical sciences (note the number of Nobel Prizes in radio astronomy). But why has the curve flattened so dramatically since the 1980s? This fascinating puzzle is addressed in the last chapter. Maybe it’s the Harwit effect as we reach a finite number of things to discover. Technology has continued to develop exponentially during this period (see Figure 12.1) so the discovery rate is not constrained by lack of new technology. Is it the consequence of the movement of astronomers out of radio astronomy as multiwavelength astronomy opened up in the space age? However, the authors make a rather compelling case that it may be a consequence of the change in culture, with the introduction of proposal reviewing and greater risk adversity in the current era. If so, this may be the most important message in this book for the future of basic research.
1 Excerpt from Many Colored Glass: Reflections on the Place of Life in the Universe (Page Barbour Lectures) by Freeman Dyson, University of Virginia Press, 2007, p. 44.
