Richard Feynman

by John Gribbin

  In 1958 she quit work and bought a one-way ticket to Geneva, the first leg of a planned round-the-world trip. At one level, the family were surprised at the decision; but they also accepted that Gweneth was always her own woman, and could not be deterred from anything she had set her mind on. In an article for Engineering and Science, she told how her friends reacted to the news – some said, ‘you’re mad’ while others said, ‘I’d like to do it too’, but nobody else did do it.18 She took only a little money with her, so that she would not be tempted to buy a ticket straight back home. She had made no arrangements to work in Switzerland, but she would have to find work if only to get the fare home and, she reasoned, if she started working to earn fare money she would be able to earn enough to live on without coming home. Then, when she had organized her finances, she planned to carry on travelling around the world. At the time she met Feynman, she was working as an au pair for her keep plus pocket money, and had only three hours free on Thursday afternoon and three hours free on Sunday afternoon. It was in one of these rare periods to herself that she met Richard.

  When Feynman learned of her circumstances, her plans to travel the world, and how little she was being paid (the equivalent of $25 per month), he suggested that she came to California. He needed someone to keep house for him – a maid, as he put it – and he could afford to pay her $20 a week, not $25 a month, plus her keep. At first, Gweneth didn’t take the notion seriously. She had two boyfriends in Geneva, and as far as she had any plans they involved going to Australia for a couple of years; she had no particular fancy to visit the United States.19 Feynman apologized for the brashness of his proposal. But she got on well with Richard; before he left Geneva she agreed to consider his offer, and they exchanged addresses.

  By November, Gweneth had decided to take up Richard’s offer, and began the process of sorting out an immigration visa for the United States. This involved a lot of tedious bureaucracy. In order to enter the United States to work, Gweneth would need a sponsor – somebody who would undertake to look after her financially if need be, until the work materialized, or if the job fell through. Feynman’s lawyer told him that it would be a bad idea for him to be the sponsor himself, for a young woman who would be living in his own house, because he might fall foul of legislation concerning the transportation of women for immoral purposes. So Richard had to persuade a physicist friend, Matthew Sands, to act as sponsor, on the understanding between them that if Gweneth really did need financial help it would come from Feynman, not Sands. Eventually, the visa came through, and Gweneth, now 25, arrived in Altadena in June 1959.

  Gweneth’s family had missed her when she went to Geneva, and were very worried when she announced that she was off to California that they would never see her again. At the end of the 1950s, even crossing the Atlantic, let alone the North American continent as well, was a big adventure.

  Everything was as Feynman had promised. He really did need someone to look after him. In an interview with Gleick, Gweneth later told how Dick had reduced his wardrobe to five identical pairs of shoes, identical dark blue suits, and white shirts that he wore with open collar. He had no TV or radio, and always kept keys, tickets and loose change in the same pockets, so that he would never have to think where they were.

  He lived in the front part of the house, and she had her own room at the back. ‘People in my hometown did not have the gumption to do something like going to Geneva or to Pasadena’, she said in the Engineering and Science article. ‘It worked fine.’

  At first, Feynman kept quiet about his new housekeeper. Scarcely anyone (except, of course, Matthew Sands and his wife) knew she was there. Then, colleagues noticed that Dick was going home for lunch, and soon there was gossip in the Athenaeum about Feynman living with a woman.20 In fact, at first, in spite of Dick’s reputation as a womanizer, their relationship really was as it had been described in Gweneth’s immigration papers. She had no intention of marrying him. ‘I had boyfriends here; I had a marvelous time. I would date Richard from time to time. Until suddenly, out of the blue, he proposed. I was never more surprised in my life.’21

  On Feynman’s side, his proposal in the spring of 1960 wasn’t a sudden decision at all. He later gave his version of the story to Leighton.22 He had realized how happy he was long before he made his proposal, and set himself a deadline, several weeks ahead, to see if his feelings changed. He decided to propose if he still felt the same way when the day he had set had arrived. The evening before the day he had chosen, he was so excited that he couldn’t wait, and kept Gweneth up on a pretext until midnight, so that he could ask her to marry him as early as possible without breaking his promise to himself to wait until that day. Gweneth was a match for him, though. She said she had to sleep on it before reaching a decision, and made him wait until the next morning before giving him her answer.

  They married on 24 September 1960, when he was 42 and she was 26, and stayed married for life. Jacqueline and her family did not travel to California for the wedding, because their son Christopher was too young at the time. They first travelled there in 1966, the year after Richard won the Nobel Prize. But from the very first, Gweneth (often with Richard) came back every year to walk in the Yorkshire Dales and visit her family, and after the first trip to California Jacqueline and her family also frequently made the trip to the West Coast, with Richard becoming part of their family. Richard and Gweneth’s son Carl was born in 1962, and in 1968 they adopted a baby daughter, Michelle. Feynman had found true happiness as a family man at last, and settled easily into the role of father figure not just to his own family but to a rising generation of physicists around the world. According to Willy Fowler,23 although everyone at Caltech had been impressed by Feynman in the 1950s, recognizing him as ‘the smartest and wisest guy in the physics division’, he was far from easy to get on with. But

  Feynman changed after his marriage to Gweneth. He became a much nicer guy. She was just such a sweet person; it was just the opposite with Mary Lou, who was “very strange”. Mary Lou antagonized everybody; everybody was relieved when Feynman divorced her. When Feynman married Gweneth we all wondered how it would be; it turned out to be wonderful.

  In amongst the turmoil of his personal life in the 1950s, though, Feynman had completed two major pieces of work in physics, as well as several lesser contributions.

  Notes

  1. See Helge Kragh, Dirac (Cambridge University Press, 1989). Kragh believes the attribution of the verse to Dirac to be apocryphal, but it aptly sums up the feelings of many physicists about their creativity.

  2. Michael Cohen, in Most of the Good Stuff.

  3. Mehra.

  4. Surely You’re Joking.

  5. Surely You’re Joking.

  6. Mehra.

  7. Surely You’re Joking.

  8. Surely You’re Joking.

  9. Engineering and Science, Caltech, November 1953.

  10. Surely You’re Joking.

  11. Leite Lopes, reported by Mehra.

  12. Mehra.

  13. Albert Hibbs, as told to Mehra.

  14. Surely You’re Joking. One reason why it was possible to meet such exciting people at Caltech was because Caltech itself was so small. Even by the early 1960s, there were roughly equal numbers of undergraduates, graduate students and faculty there – about 600 of each.

  15. Mehra.

  16. What Do You Care.

  17. Interview with MG, February 1996; Jacqueline Howarth is now Jacqueline Shaw.

  18. Gweneth Feynman, ‘The Life of a Nobel Wife’, Engineering and Science, March–April 1977.

  19. See note 18.

  20. Albert Hibbs, as told to Mehra.

  21. See note 18.

  22. Comment to JG, December 1995.

  23. Mehra.

  8 Supercool science

  Feynman had become interested in the peculiar behaviour of liquid helium while he was still at Cornell, but he had been too busy completing his version of QED to devote any real effort to the puzzle.
It was, though, a natural for him to take up once he got settled at Caltech in the early 1950s. It was a fundamental problem in physics, involving the quantum properties of particles, that he was able to tackle using his special insight into the behaviour of nature, seeing right to the heart of the problem and avoiding the thickets of mathematical complexity with which everybody else had surrounded the problem.

  In order to liquefy helium at all, you have to achieve really low temperatures. The lowest temperature it is possible to reach, even in principle, is –273.16° Celsius, defined as zero on the Kelvin (K) scale. This ‘absolute zero’ is the temperature at which each particle has the minimum amount of energy which it is allowed to possess by the quantum rules. In a sense, this is an example of quantum uncertainty at work. If a particle had zero energy, it would be completely at rest, in one place, and not going anywhere. So there would be no uncertainty about its position and its momentum. In order for there to be uncertainty, it must always have at least a little energy so that it can jiggle about in different ways. Helium only condenses from a gas to form a liquid at a temperature of 5.2K, where the amount of jiggling it can do is already getting close to the quantum minimum. But what it does with the little energy it has can be spectacular.

  Helium was first liquefied by the Dutch physicist Kamerlingh Onnes in 1908, and in further experiments he pressed on to temperatures even lower than 5K. In 1911, he discovered that something very peculiar happens to liquid helium at a temperature of just 2.2K; at about the same time, he discovered the phenomenon of superconductivity, the complete disappearance of electrical resistance in some metals when they are cooled to very low temperatures.

  The first peculiar thing that happens to liquid helium when it is cooled below 2.2K is that it expands as it is cooled further, instead of contracting. Because of this, and other changes that occur at the same temperature, the liquid below this transition temperature became regarded as a separate ‘phase’ of helium, as distinct from liquid helium at higher temperatures as the liquid itself is from the gas. The liquid above 2.2K became known as helium I, and the liquid below 2.2K was dubbed helium II. The most impressive property of liquid helium II, established some time after Onnes’ pioneering work, is that it is a superfluid – it can creep through tiny capillary tubes without seeming to meet any frictional resistance, and will even climb up the walls of a container to escape, or leak away through pores that are so small that gas cannot get through them.

  The idea that was becoming accepted by the beginning of the 1950s, and which was producing the profusion of mathematical thickets surrounding the puzzle of superfluidity, was that below the critical temperature of 2.2K liquid helium II could be treated as if it were a mixture of two separate fluids. Part of the fluid seemed to have settled into the state it would be in at the absolute zero of temperature, 0K itself, with the minimum amount of energy in each helium atom. The rest was a ‘normal’ fluid. At 0K, the fluid would all be in the minimum quantum energy state, and at 2.2K it would be all ‘normal’, with the proportions varying smoothly in between.

  The key to this interpretation of superfluidity is the way in which quantum entities behave, and it relied on treating whole helium atoms as if they were single quantum entities like electrons or photons – indeed, it specifically relied on treating them exactly as if they were photons.

  Quantum entities come in two varieties, called fermions and bosons (after the physicists Enrico Fermi and Satyendra Bose). Fermions are what we are used to thinking of as particles, such as electrons; they each have an amount of quantum spin which is a half-integer – ½, or 3⁄2, or 5⁄2, and so on. Bosons are what we are used to thinking of as waves, like photons; they each have zero or integer spin – 0, or 1, or 2, and so on. The important practical distinction between fermions and bosons is that no two fermions can exist in the same quantum state, while bosons can happily exist in the same quantum state as other bosons. This has implications, for example, for the structure of the atom. The electrons surrounding the nucleus of an atom must each be in a unique quantum state, sitting on different rungs of an energy level ‘ladder’. Two electrons are allowed to sit on the bottom rung, because they can have opposite spins (one up, one down), but additional electrons have to sit, in some sense, successively further out from the nucleus in order to avoid being in the same state as one of these two inner electrons. The situation is slightly more complicated than we have made it sound, but the important point is that each electron has its own place, like the members of a theatre audience each with their own numbered seat in the auditorium. If it were not for the exclusivity of the fermions, all the electrons in an atom – any atom – would jostle together in the lowest energy state next to the nucleus, so all atoms would have more or less the same chemical properties and there would be none of the chemical complexity that makes the world so interesting and makes life possible.

  Bosons obey different rules, and can pack together in the lowest energy state with other bosons. Rather than being placid theatregoers sitting in their numbered seats, they are more like the enthusiastic fans at a rock concert, all crammed into the space in front of the stage together. There are other differences, which affect the way in which a box full of bosons – a boson gas – behaves, making its properties different from those of a fermion gas. One of the most dramatic discoveries of theoretical physics in the 1920s was that the behaviour of light can be entirely explained in terms of photons as particles obeying the rules appropriate for a boson gas, without invoking the idea of waves at all. Albert Einstein was involved in this work, and such a boson gas is sometimes referred to as a Bose–Einstein condensate. The two-component model of superfluid helium says that below 2.2K part of the fluid is behaving as a Bose–Einstein condensate (a boson gas), in the same way that photons behave, while the rest is behaving in the way that particles such as electrons behave (a fermion gas).

  Feynman explained the superfluid behaviour of liquid helium in a series of ten scientific papers (more than he published on QED) in a five-year period (1953–8) during the 1950s – many of them based on work carried out during the turmoil of his second marriage and its aftermath. As always, he started from first principles, largely ignoring the efforts other people had made to come to grips with the problem, and thinking about the behaviour of individual atoms in the fluid – the way they jiggled about, or slid past one another, or bounced off one another. He used the path integral approach, which turned out to be just as effective here as in QED or in classical optics, producing a theory that the physicist David Pines has described as ‘that blend of magic, mathematical ingenuity and sophistication, and physical insight that is almost uniquely Feynman’s’.1 Pines also draws attention to the fact that the second paper in the series contains only a single equation, but leads the reader to certain conclusions about the behaviour of liquid helium, starting out from the fact that it is a Bose–Einstein condensate, through ‘a series of closely reasoned arguments’ alone. As well as establishing a satisfactory model of superfluidity, Feynman introduced a generation of condensed matter physicists to the use of Feynman diagrams and path integrals, making these techniques indispensable tools in that branch of physics.

  Feynman also worked on the problem of superconductivity, but for once his insight let him down, and he was unable to come up with a satisfactory explanation of the phenomenon. And yet, even this failure has gone down in scientific folklore, because Feynman’s response to it demonstrates another facet of his character, his scrupulous honesty in matters scientific. The problem was actually solved in 1957, by John Bardeen, Leon Cooper and Robert Schrieffer. Feynman was one of the first physicists to appreciate that their model (known as the BCS theory) really had solved the problem, and promptly abandoned his own efforts at explaining superconductivity while singing the praises of the BCS theory at every opportune occasion. But it was events at a conference held a year earlier, in 1956, that had impressed Schrieffer with Feynman’s unique way of tackling physics. Schrieffer, as it happens, was t
he rapporteur for that meeting, and so paid close attention to all the talks. In an interview with Gleick, he has recalled how Feynman delivered a talk on two problems – the one he had solved (superfluidity), and the one that still baffled him (superconductivity). Schrieffer had never before heard a scientist describe publicly, in such loving detail, all the steps in a failed theory. Feynman’s natural honesty helped others to avoid falling into the same traps he had fallen into, by signposting the danger areas, and showed clearly his ability to avoid deluding himself into thinking he was on the right trail when in fact he was barking up the wrong tree.

  In 1972, the BCS team shared the Nobel Prize for Physics for their theory of superconductivity. Bardeen thereby made history, becoming the first person to win two Nobel Prizes in the same field, having already shared the physics prize with William Shockley and Walter Brattain in 1956, for their discovery of the transistor effect. With hindsight, it is hard to see Feynman’s investigation of superfluidity as any less significant than the BCS theory of superconductivity, but as it happens when Lev Landau received his Nobel Prize in 1962 the citation specifically mentioned his work on the theory of liquid helium. In 1962, it was clear that Feynman’s masterwork was QED, so there was probably no real consideration given to splitting that year’s award between him and Landau; by 1972 (when the BCS team received their prize), it would have been too late, the prize for superfluidity having already been given. Otherwise, Feynman might well have shared Bardeen’s double distinction.