Every atom is almost entirely made of empty space, with a tiny object called the nucleus and even tinier objects called electrons flying around inside it. Ernest Rutherford, a young New Zealander working in Manchester, England, discovered this fact about atoms in 1909. He shot fast particles at a thin film of gold and observed the way the particles bounced back. The pattern of the recoiling particles showed directly the internal structure of the atoms in the film. The discovery of the tiny nucleus came as a big surprise to Rutherford as well as to everybody else. The phrase “the fly in the cathedral” described what Rutherford discovered. The fly is the nucleus; the cathedral is the atom. Rutherford’s experiment showed that almost all the mass and almost all the energy of the atom was in the nucleus, although the nucleus occupied less than a trillionth part of the volume.

Rutherford’s discovery was the beginning of the science that came to be called nuclear physics. After discovering that nuclei of atoms exist, he continued to study their properties by bombarding them with fast particles and observing the results. The projectiles that he used to explore the nucleus were particles produced in the disintegration of radium. Radium is a naturally occurring radioactive metal that was discovered by Marie Curie in 1898. The particles are helium nuclei that are emitted at high speed when radium atoms decay. These particles made good probes for exploring the properties of nuclei because they were all alike and came with known energies. For twenty years, first in Manchester and later in Cambridge, Rutherford and his students and colleagues used natural particles with great success to learn how nuclei behave. They found that it was possible on rare occasions to change one kind of nucleus into another by adding or subtracting a particle. The twenty years between 1909 and 1929 were the era of table-top nuclear physics. Experiments were small enough to fit onto the tops of tables. Small and simple experiments were sufficient to establish the basic laws of nuclear physics.

Toward the end of the 1920s, nuclear physics got stuck. Major mysteries remained to be solved. Nobody knew what nuclei were made of or how their component parts were put together. But it was hard to think of exciting new experiments that could be done with the existing tools. The next round of experiments were minor variations of experiments that had already been done. It seemed unlikely that the mysteries of nuclear structure could be solved by such experiments. Rutherford announced in a public lecture in London in 1927 that new tools were needed if nuclear physics were to move ahead. Without new tools, research in nuclear physics would stagnate and bright young people would no longer be attracted to it. The most promising new tool would be a particle accelerator, an electrical machine that could produce a beam of artificially accelerated particles to replace the natural particles produced by radium. Artificially accelerated particles would be better than natural particles in three ways. They could be produced in greater quantities, they could have higher energies, and they would allow experiments to be designed more flexibly. The switch from natural sources of particles to accelerators would start a new era in the history of science, the era of accelerator physics.

The Fly in the Cathedral tells how the era of accelerator physics started. It is a dramatic story and Brian Cathcart tells it well. He is a journalist and not a scientist, but he understands enough of the science to get the details right. He has made a thorough study of the primary literature, he has read the papers and letters written by the participants, and he interviewed those of them who were still alive. The story begins with Rutherford’s decision in 1927 to explore the possibility of accelerating particles, and ends with the building of the first accelerator and its triumphant success as an atom-splitter in 1932. The era of accelerator physics began in 1932 and is not yet ended. The enormously powerful accelerators that are now exploring the fundamental forces of nature in Illinois and California and Switzerland and Japan are the direct descendants of machines that were built in 1932.

The story of the first accelerator was not only an important chapter in the history of science. It was also an international sporting event, driven by national pride as well as by scientific curiosity. Rutherford had competitors in many countries. The most formidable competitors were in the United States: Merle Tuve at the Carnegie Institution in Washington, Robert Van de Graaff at the Massachusetts Institute of Technology, and Ernest Lawrence at the University of California in Berkeley. Rutherford knew and respected his competitors but was determined to beat them. As a scientist he was a member of an international community, but as an old-fashioned New Zealander he was fiercely loyal to Britain and the Empire. He understood that science is an international enterprise that flourishes best when nations compete for the prizes.


The two men who actually built the first accelerator were John Cockcroft and Ernest Walton, graduate students working in the Cavendish Laboratory in Cambridge under the supervision of Rutherford. Cockcroft came to the Cavendish from Yorkshire in 1924, Walton from Ireland in 1927. When Walton arrived from Dublin, he proposed to Rutherford for his student project that he should start building an accelerator, not knowing that Rutherford had already announced his intention to do so. Rutherford was happy to give him the green light. Cockcroft had worked at Metropolitan-Vickers, the leading electric engineering company in Britain, before coming to Cambridge, so that he had some experience in working with heavy equipment and high voltages. Rutherford arranged for Walton to work full-time on the accelerator project, with Cockcroft working on it part-time to help with the engineering. For five years they struggled to create a technology of big machines in a laboratory of table-top experiments, just as the Wright brothers had struggled to create a technology of flying machines in a bicycle shop.

The obstacles that Cockcroft and Walton had to overcome were cultural as well as technical. If they were to build big machines, they would obviously need more space, but the idea of building a new building to house a new machine was not thinkable in the Cavendish culture. Rutherford was notoriously stingy and kept all expenditures to a minimum. The Cavendish was a historic building and could not be remodeled. So everything that Cockcroft and Walton built had to fit into existing rooms and pass through existing doors. This had the consequence that they could not use any commercial power equipment that could not be squeezed through the historic Gothic gateway of the Cavendish. They had to spend long months designing and testing their own equipment.

The culture of the Cavendish was strongly paternalistic. Rutherford took fatherly care of his students and imposed strict limits on their hours of work. Every evening at six o’clock the laboratory was closed and all work had to stop. Four times every year, the laboratory was closed for two weeks of vacation. Rutherford believed that scientists were more creative if they spent evenings relaxing with their families and enjoyed frequent holidays. He was probably right. Working under his rules, an astonishingly high proportion of his students, including Cockcroft and Walton, won Nobel Prizes.

They were maintaining the cul-ture of nineteenth-century gentlemen-scientists, who were supposed to pursue scholarly leisure-time activities in addition to their science. But this culture of short hours and plentiful leisure was hard to combine with heavy machine–building. Years went by while Cockcroft and Walton put together improvised vacuum systems, laboriously sealed the leaks, tried out various devices for handling high voltages, and found them to be inadequate.

It took Cockcroft and Walton five years to build a machine that worked. Finally, in April 1932, they had a machine that produced a steady stream of hydrogen nuclei with an energy of about half a million volts. They were carefully measuring the quality of the stream, in no hurry to begin doing nuclear experiments. On the morning of April 13, Rutherford came into their room, saw what they were doing, and lost his temper. He told them to stop wasting their time and do some science. The next day, Cockcroft was occupied with other business. Walton was alone in the laboratory, ready to do the first experiment, using his hydrogen nuclei to bombard a target made of the light metal lithium. The result was spectacular. The lithium nuclei were split in two and fell apart into pairs of helium nuclei. The helium nuclei came out with thirty times as much energy as the hydrogen nuclei going in. Walton ran to Rutherford’s office to tell him the news, and Rutherford happily spent the rest of the day serving as Walton’s assistant, checking the result and tidying up the details. That day, the era of table-top nuclear physics ended and the era of big machines and big projects began.

The American competitors were running close behind Rutherford and were beaten only by a few weeks. Van de Graaff had invented an electrostatic accelerator that was in many ways superior to the Cockcroft-Walton machine, and Lawrence had invented the cyclotron, which was in many ways better still. They were not forbidden to work after six o’clock in the evening, but they, too, had to struggle with cultural obstacles similar to those that hampered Cockcroft and Walton. American academic administrators, hard hit by the economic depression of the 1930s, were almost as stingy as Rutherford. Tuve and Van de Graaff had built a splendid machine at the Carnegie Institution in Washington, but it was standing on the grass out-of-doors, where dust and insects interfered with its operation. They were forced to dismantle it because the laboratory had no room large enough to house it.


Even Lawrence, who became in later years the main driving force of big-machine physics, had similar problems when he started to build cyclotrons. In 1931 he had acquired for his newest cyclotron a huge magnet that weighed eighty-five tons, but he had no building big enough for it and so the cyclotron remained unfinished. Van de Graaff and Lawrence were the hares, Rutherford was the tortoise, and the tortoise won the race.

In the few years that remained to him after 1932, Rutherford enjoyed the rebirth of nuclear physics brought about by the new machines. His students in the Cavendish continued to explore the universe of nuclei, in friendly competition with explorers in America and Europe. He died a year before his German friend Otto Hahn discovered the fission of the uranium nucleus in 1938. Walton returned to Dublin in 1934 and spent the rest of his life there peacefully as a professor of physics. Cockcroft stayed at the Cavendish until 1939, then moved into war research, and then became director of the British Atomic Energy Research Establishment at Harwell when it was established after the war. Harwell was mainly concerned with the development of reactors, for scientific research and for the nuclear power industry. Cockcroft showed me around Harwell when I visited there in the 1950s. The most conspicuous feature of Harwell at that time was a huge array of high-voltage cables connecting the laboratory to the national electricity grid. “The main reason the public supports us,” Cockcroft said with a smile, “is that they think the electricity is flowing out of the lab. Actually, of course, it is flowing in.”


Cathcart’s book ends with a discussion of the question why the tortoise beat the hares in the race to disintegrate nuclei with accelerators. He concludes that the main reason why Rutherford won was that he was not a machine builder. Van de Graaff and Lawrence were brilliant inventors, driven by passionate love for their machines. They were not so much concerned with what their machines could do when they were built. For Rutherford the machine was only a tool. He was not interested in the details of its design and trusted Cockcroft and Walton to get the details right. For him, what mattered was the science.

He had spent his life exploring the nuclei of atoms, and his driving passion was to dig deeper into the nucleus. That was why he made sure that the decisive experiment was carefully prepared and the lithium nuclei were ready and waiting to be disintegrated as soon as his machine was finished. The Americans had better machines, but Rutherford had a more single-minded concentration on the scientific goal. Two months after he won the race, Rutherford explained to a reporter from the Daily Herald why he had wanted so badly to disintegrate nuclei. “We are rather like children,” said Rutherford, “who must take a watch to pieces to see how it works.”

Alan Lightman’s book, A Sense of the Mysterious, tells a very different story. His book has no index, so I cannot be sure how many times Rutherford is mentioned. I believe he is mentioned only once, on page 133, in a list of names of famous people who were dead in the year 2002 when that page was written. I find it remarkable that we have two books designed to give nonexpert readers a feeling for the way research into the nature of the physical universe is done, and yet the central character of one book is barely mentioned in the other. How could two accounts of the same science be so different, in so many ways, in style and in substance?

Brian Cathcart is an Irish journalist with an amateur interest in science. Alan Lightman is an American, trained as a theoretical physicist, who made a mid-life change of career from scientific research to writing. Cathcart’s book is a straightforward historical narrative. Lightman’s book is a collection of essays, lectures, and book reviews, some of them published in this magazine, most of them describing individual scientists and their ideas. Cathcart is primarily interested in experiments, Lightman in theories. Cathcart sees progress in science mainly driven by new tools; Lightman sees progress driven by new concepts. Cathcart’s story is a simple drama with three heroes and no villains, a triumph of human pertinacity over technical and cultural obstacles. Lightman’s chapters are meditations on the human condition, illustrated by sketches of characters who are partly heroes and partly villains.

The main characters in Lightman’s stories are the theoretical physicists Albert Einstein, Edward Teller, and Richard Feynman, and the observational astronomer Vera Rubin. Not only is Rutherford absent, but almost all experimental scientists are absent too. The only experimenter who makes an appearance on Lightman’s stage is Joseph Weber, a brilliant and tragic figure whose experiments turned out to be wrong. The mainstream experimenters who explored the universe of particles and fields, continuing to play Rutherford’s game, taking the watch to pieces to see how it works, do not appear at all. Lightman’s title, A Sense of the Mysterious, and his subtitle, Science and the Human Spirit, do notexplain his neglect of experimenters. After all, Rutherford had as deep a sense of the mysteries of nature as Einstein. And the human spirit expresses itself as eloquently in the work of human hands as in the work of human minds. Rutherford was supreme as an experimenter and Einstein was supreme as a theorist, but each of them held the other in deep respect. Both of them understood that the human spirit is at its best when hands and minds are working together.

One theorist played an essential role in Rutherford’s thinking. George Gamow was a brilliant young Russian who came to Germany in 1928 and at the age of twenty-four started a revolution in nuclear physics. He was the first to understand how the quantum theory, which had been invented only three years earlier, could be applied to the nucleus. He used quantum theory to calculate how fast radioactive nuclei such as radium or uranium should decay, and found that the theory agreed pretty well with the observed rates of disintegration. He then made another decisive step, using quantum theory to calculate how easily a charged particle could come into a nucleus from outside. He understood that the same quantum rules apply in both directions. Easy out, easy in. If particles can escape from radioactive nuclei by quantum rules, then they can also penetrate into nuclei by quantum rules when fired at them from outside.

When Rutherford heard of Gamow’s idea, he saw at once that this dramatically improved the prospects for doing important science with the accelerator that he was planning to build. Rutherford did not pretend to understand quantum mechanics, but he understood that the Gamow formula would give his accelerator a crucial advantage. Even particles accelerated to much lower energies than the particles emitted naturally by radium would be able to penetrate into nuclei. Rutherford invited Gamow to Cambridge in January 1929. The fifty-eight-year-old experimenter and the twenty-four-year-old theorist became firm friends, and Gamow’s insight gave Rutherford the impetus to go full steam ahead with the building of his accelerator.

The same mutual admiration of experimenter and theorist was shown three years later when Einstein happened to be visiting Cambridge a few days after the triumph of Cockcroft and Walton. Einstein insisted on seeing the accelerator that had split the atom. Walton spent a morning showing him the apparatus and explaining the details of its operation. Einstein wrote a letter afterwards, expressing “astonishment and admiration” for what he had seen. “He seems a very nice sort of man,” wrote the imperturbable Walton to his fiancée in Ireland.

How is it that this mutual admiration and easy mixing of theory with experiment, which seemed natural and necessary in the 1930s, is absent from Lightman’s view of physics? Somehow it happened that the successors of Rutherford and Einstein drifted apart in the second half of the twentieth century. This was not the physicists’ fault. It resulted from the enormous growth of accelerators and the enormous proliferation of theories. Accelerators and the accompanying apparatus for detecting particles became so huge and complicated that each experiment was like a military operation. Hundreds of people with highly specialized skills were required to carry out a program planned many years in advance. Theorists became similarly specialized, some of them expert in accelerator design, some in particle interactions, some in general relativity, and some in string theory. It became difficult for theorists in different specialties to communicate with one another, let alone with experimenters. At the end of the century, accelerator physics was slowing down. Each experiment required about a decade to design and prepare. Alan Lightman, an imaginative theorist who liked to avoid narrow specialization, found such experiments unattractive. It was natural for him, following his sense of the beautiful, to move away from experimental physics and toward astronomy.

Astronomers have so far escaped the extreme specialization that has overtaken physicists. Telescopes are big, but they are not as complicated as accelerators. Observations with a big telescope can be carried out in hours rather than years. Astronomers can be skilled observers and also expert in the theory of what they are observing. That is why the astronomer Vera Rubin has a place of honor in Lightman’s book. She started her professional career as a student of George Gamow after Gamow moved to America. She spent the rest of her professional life observing galaxies and studying their dynamics. She found that the visible matter in galaxies is not heavy enough to explain the speed of their internal motions. She deduced from her observations that galaxies are pervaded by dark matter, invisible to our telescopes. Nobody knows what dark matter is. It is another deep mystery remaining to be explored. We know only that it is there, and that it weighs more than all the stuff that we can see.

Besides discovering and exploring dark matter, Rubin raised four children and crusaded publicly for the advancement of women in science. I was recently chairman of a committee that organized a scientific conference with a list of distinguished scientists as members. I received a blistering letter from Rubin, asking why we had no women on our list. She supplied me with another list of women who should have been invited. I wrote back to apologize and to thank her for her list, which I shall certainly use in the unlikely event that I ever become chairman of another such committee.

Lightman’s chapter on Edward Teller is a review of Teller’s memoirs, first published in these pages in 2002. Lightman considers Teller to be on the whole an evil character, in sharp contrast to his sympathetic portrayals of Einstein and Feynman. The title of the Teller chapter is “Megaton Man,” emphasizing the obsession with hydrogen bombs which made Teller famous. Lightman admits that there were two Tellers. He writes, “There is a warm, vulnerable, honestly conflicted, idealistic Teller, and there is a maniacal, dangerous, and devious Teller.” But his portrait of Teller shows us mostly the dark side.

Merle Tuve, the experimenter who was beaten to the punch by Rutherford in 1932, knew Teller well after Teller came to join George Gamow at George Washington University in 1935. In 1939 somebody at the University of Chicago asked Tuve for an appraisal of Teller. Tuve replied, “If you want a genius for your staff, don’t take Teller, get Gamow. But geniuses are a dime a dozen. Teller is something much better. He helps everybody. He works on everybody’s problem. He never gets into controversies or has trouble with anyone. He is by far your best choice.” That was the Teller I knew, when I worked with him every day for three months in 1956, designing a safe nuclear reactor. It was easy to disagree with him fiercely about politics, or about the details of the reactor, and remain friends. Those who had never been his friend did him a grave injustice when they tried to turn him into a demon.

Putting together the portrait of Rutherford in Cathcart’s book with my own recollections of Teller, I find striking similarities. Rutherford and Teller were both immigrants who became fiercely patriotic in defense of their adopted countries. Both often behaved like overgrown children, losing their tempers over trivialities and then regaining their equilibrium with a friendly smile. Both were father figures to their students, taking care of students’ personal problems as well as their professional education. Both were more interested in the strategy of science than in the tactics. Rutherford made the decision to explore nuclei with an accelerator, and then left the details of the accelerator to Cockcroft and Walton. Teller made the decision to build a hydrogen bomb or a safe reactor and then left the details to others. Both had a lifelong dedication to science, but spent more time helping younger people than doing research themselves. Teller published his version of the hydrogen bomb story under the title The Work of Many People. The names of Cockcroft and Walton appear on the letter to Nature announcing their discovery but Rutherford’s does not. My name appears on the patent for the safe reactor but Teller’s does not.

The most concise and original chapter in Lightman’s book is “Metaphor in Science,” an essay originally published in 1988 in The American Scholar. Illustrating his thesis with quotations from great physicists from Isaac Newton to Niels Bohr, Lightman traces the powerful influence of metaphors on their thinking. As science has become more abstract and remote from everyday experience, the role of metaphor in our descriptions of the world has become more central. The language that nature speaks, as Galileo long ago pointed out, is mathematics. The language that ordinary human beings speak, especially those of us who are not fluent in mathematics, is metaphor. Lightman ends his discussion with another metaphor: “We are blind men, imagining what we don’t see.” That is a good description of theoretical physics.

This Issue

February 24, 2005