Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Rudolf Peierls by John L. Heilbron on 1963 June 17, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/4815-1
For multiple citations, "AIP" is the preferred abbreviation for the location.
This interview was conducted as part of the Archives for the History of Quantum Physics project, which includes tapes and transcripts of oral history interviews conducted with ca. 100 atomic and quantum physicists. Subjects discuss their family backgrounds, how they became interested in physics, their educations, people who influenced them, their careers including social influences on the conditions of research, and the state of atomic, nuclear, and quantum physics during the period in which they worked. Discussions of scientific matters relate to work that was done between approximately 1900 and 1930, with an emphasis on the discovery and interpretations of quantum mechanics in the 1920s. Also prominently mentioned are: Niels Henrik David Bohr, Bragg, Louis de Broglie, Constantin Caratheodory, Frank Clive Champion, Peter Josef William Debye, Max Delbruck, Enrico Fermi, Otto Halpern, Werner Heisenberg, Friedrich Hund, Lev Davidovich Landau, Ettore Majorana, Walther Nernst, Heinrich Ott, Wolfgang Pauli, Max Planck, Robert Wichard Pohl, Arnold Sommerfeld, Albrecht Unsold, Hermann Weyl, Wilhelm Wien; Universitat Berlin, Universitat Leipzig, Universitat Munchen, University of Cambridge, and Zeitschrift fur Physik.
May we begin by discussing the rise of your own interest in science, and so forth?
Yes. Well, you say you want to start with family background and so on. My father was a businessman, who had a fairly important administrative job in industry; he was the managing director of a factory. Later on he got a post on the board of this company. He had no university or scientific training; he didn’t know much about science, except engineering that he of course met. But generally it was a civilized and educated background. My mother was interested in music and theater and literature and politics and so on. And I think it was just at the time of the change from the Victorian ideas of education to the more modern ones, and I think we had a fairly liberal education in the sense that there was no excessive emphasis on discipline. I remember my father saying that he was born at the wrong time, because when he was young the young people — from the adolescent to the somewhat older people — didn’t count, and now that he was grown up it was the young people that mattered. Now, my education was a standard one; I went to the local high school, you would say, in a small suburb; there was actually no choice. It was not a particularly good school, and I found it extremely easy and therefore never did any work while I was at school. The art of really concentrating and doing some work had to be learned later, which is painful.
But you had no interest in science at that point?
Oh, yes. Now the way it was really, was that I was very interested in technical things; radio was then new. I remember the excitement of going for a first visit to the big local broadcasting transmitter — sorry, it wasn’t broadcasting yet — it was the big radio telegraphy transmitter that sent signals across the Atlantic. I was very interested in things like electricity and building little gadgets and so on, and I think my view at that time is that I wanted to be an engineer. That was the obvious thing; living in an industrial community that was the obvious thing. However I did not become an engineer because it was impressed on me by my family that I couldn’t possibly be an engineer for two reasons, neither of which I admit now was a good reason. One was that my eyesight was not very good — I had to wear glasses already from the age of 6 or so — and the other was that I wasn’t terribly good with my hands. These were regarded as obviously disqualifying me from being an engineer.
What kind of an engineer in particular did you think of becoming?
Well, it never came to that. I was always as a boy interested in what I could find out about scientific things; I mean, I read all popular books on — not so much on basic science; at that time I just didn’t know about that — engineering mainly. I knew everything about railways and how they worked and how the signaling system was and all that, as you would expect for that age. At school I was very good at mathematics, and in fact I think our best teacher was our mathematics teacher. I was not that good in physics; I found it very easy, but we had a terrible teacher. In fact, I took delight in proving him wrong, so that the poor man was always frightened when I raised my hand about something. Of course it’s a bad situation to get into when you expect your teacher to be wrong, because I queried many of his statements which in fact were right.
It’s not all lost anyway.
No, and in fact that’s perhaps a good thing then to be caught up and to be proved wrong; it was probably quite a healthy experience. Anyway, I then decided to be a physicist because that seemed the next best thing to an engineer, and I think I probably had in mind, if I had any conscious plans, to become a physicist and go into industry. I do not recall how definite that idea was, but I imagine if I’d been asked about the future that’s what I would have said.
What would you have thought the difference was between a physicist who worked in industry and an engineer?
Well, a physicist might work in the laboratory or might work on the principles, and he wouldn’t be making things. Well, I mean, I wasn’t clear myself about the objections for why I couldn’t be a good engineer. I just accepted this advice, and there was no similar objection to being a physicist at all. I think, however, at the time I went to the University I was already inclined to be a theoretical physicist, partly perhaps again because I wasn’t all that good with my hands in those days — I think I’ve got over it by now. And it was partly because I was interested in mathematics and found it easy. But I don’t think at that time I had any particular interest in the fundamentals, I mean, I just hadn’t met the front line or the basis of physics. My motivation I think was to find out how things worked, the practical things, and what you could do with them, and the sort of search for the laws of Nature was not something I was conscious of at that time. Then I was ready to go to the University, with an interruption because my school was unusual in that it finished at Easter; that was the wrong time to get into the University. You could, as it goes by semesters, enter the University at Easter, or the summer semester, but it seemed the wrong time to do it. Therefore, I think it was my own idea that it would be nice to spend the intervening time in industry, and I went into the research department of a factory and spent 6 months there until the beginning of the first semester.
Was that a relatively easy thing to do, or was it through your father’s connection?
It was done through my father’s connections. I think it might have been easy otherwise, too, but I don’t know. It was slightly unusual; it was the common thing to do if you were planning to be an engineer, and then you would probably spend a year. Now doing it for 6 months only and without being committed to engineering training, it was slightly unusual. I’d know whether I could have done it anyway, but it certainly was in a firm which had some connections with my father’s firm, so this is how it was arranged. That I certainly enjoyed and never regretted that time. There was a nice mixture of learning to do quite practical things, use of tools — of the lathes and what not.
You thought you could do it after all?
Yes, probably not brilliantly but I could learn it. Then doing simple experiments or designing some circuits; we then had what you now would call electronics except it had no valves in it, but just relays and so on; but there were automatic systems for signaling and so on that we designed and tested, and that was great fun.
But you had decided for physics at that point then, and you weren’t willing to use this new art you had in favor of going back to engineering?
No. No, I think by that time I had accepted the idea and was quite happy about it.
And you went to Berlin just because of proximity?
Yes. I think I was anxious to get away from home, not that there was any particular friction, but I liked independence. My parents felt that it was a bit early; I was perhaps too young to go away, so I stayed for the first year at Berlin. When there seemed no longer an objection to my moving, then I moved. Now you ask about this first year in Berlin. I did very little physics then because it was then generally considered that you should start physics with the physics lab, and there was congestion there, so that first year students couldn’t do it. As a result the idea was to pick up as much mathematics background as you could in the first year, so I did almost entirely mathematics. This was the funny German system where there is no regular course laid down, but there are just some lectures, and with whatever advice you can get hold of you choose what to do. I had some good advice from a friend of ours who was a physicist in industry and who helped me look at the list and made suggestions. Objectively it was a crazy choice because many of the things were really much too hard, but I got a lot of fun out of them. For example, in theoretical physics there was then the custom of having big series of courses in a three year cycle. Planck gave that course, and the stage at which I came in was his lectures on optics. Not having previously done Maxwell’s equations, let alone anything else, I couldn’t follow anything in detail very much. It wouldn’t have done much good anyway, because they were terrible lectures. Planck was one of the worst lecturers I knew because he essentially read word for word what was in his books. And you could sit there with the book and follow it line by line; there was no discussion, no explanation, and the books are bad.
At that point they were no longer changing either.
No, no. Well, his book on optics is not as bad as some of the others; I think Hydrodynamics is probably the worst. The one course which really opened my eyes to something new, apart from mathematics where there were a lot of new things of course; was a course by Bothe on x-rays where the word quantum was mentioned for the first time. This I think again was meant as a course for rather more advanced people, and therefore the basis was never explained very much. Things like the K-shell and the L-shell and quantum numbers were mentioned without making clear from the start that here were some revolutionary new principles. But at any rate I got used to the idea that there were some interesting new things there. That probably was my only contact in the studies with the old form of the quantum theory.
May I ask you about this question of getting advice? There were just no general ways in which you could approach someone?
Well, I suppose you could have gone to the Dean of the faculty, whose job it would have been to advise you, or you could meet with more senior students. The advice was less critical perhaps because you registered for a large number of courses and didn’t necessarily go to all of them. I recently looked at my university papers and found that in the first year — and probably in most years — I was registered for something like 36 lectures a week and I went to most of them. There were some exceptions, for example, (???) if you are aiming at the Doctor of Philosophy — science was in the faculty of Philosophy — you must therefore learn some philosophy. And I accordingly registered for two courses in philosophy. One was epistemology and I’ve forgotten what the other one was, but I could stand that only for a few weeks.
Yes. Who taught that, do you remember? Was it a professional philosopher?
Professional philosophers, yes. Do I remember the name; I think a Professor Dessoir was then a professor of philosophy. I think, yes, Liebert I believe was another one. Anyway they left no deep impact except that I didn’t want to have anything more to do with philosophy.
Did the physics department have any courses in the philosophy of science at that point? At the beginning of the century did they?
No. Well, they might have had, but I certainly wasn’t involved. I had very little contact with the physics department at that time. I did go — I believe that was in my second year — to the big introductory course to experimental physics given by Nernst, who was the great man then in experimental physics.
What kind of a course did he give?
Just the general introductory course and it was fairly easy. I remember my father knew him because he had been connected with the firm that my father was working for, and so I called on him one day. He was nice and polite, and he was in the lectures just then talking about Ohm’s law. And he said, “Well, this is very important; I hope you’ll take this seriously because this is very basic to everything.” And I thought he was joking, because to me at that time it seemed completely obvious that anyone who would get as far as getting into a University would be familiar with Ohm’s law; that one had at that stage to learn it, seemed to me a joke. And now I realize that it’s not so obvious, and that indeed there are people who come to the university without being familiar with Ohm’s law.
I understand that Nernst wasn’t always so nice and polite; that he had a reputation among students for being fierce.
Well, I wasn’t aware that Nernst had the reputation of being in any way fierce. One knew that he was extremely vain, and that often caused amusement. I remember the passage when he mentioned the Nernst lamp in his lectures; you know it was an old fashioned type of lamp which had some hopes after the first Edison thing. He introduced this by saying in the lecture, “Gentlemen, when Bunsen used to mention the Bunsen burner in his lectures, he referred to it as a non-luminous burner, and in the same spirit I’m now going to talk about the electrolytic lamp.” But I had otherwise not much to do with him; I just listened to his lectures and —.
Was he a good lecturer?
Not exceptionally; I mean these big experimental introductory courses are courses that traditionally the professors take a good deal of trouble with, with demonstrations and so on. It wasn’t on the standard of the lectures of Pohl in Gottingen or Scherrer in Zurich, but it was all right. I certainly have no recollection of details about that. What is interesting — well, perhaps that has nothing to do with the history of quantum theory —I’m quite amazed by the recollection that I must have gone to, well, on the average probably 30 lectures a week.
These were hour, or essentially hour lectures?
Well, no, 45 minutes with a 15 minute break. And I did not have the feeling of being particularly pressed for time. I had time to meet my friends, to do some homework, work on problems and so on and to go to the theater and do everything else I liked. And I was therefore a little suspicious when our students with perhaps 20 or 22 lectures a week, feel so terribly hard worked.
Were there regular exercises that one had to perform, written assignments?
Well, there were usually problem classes which would be included in the number mentioned, and you worked the problems in your own time, handed them in, and they were marked. This is quite a good system; in the actual problem class one of the students is called to the board to present his solution and then it’s discussed with help from the other students.
And you stayed in Berlin for —?
For two semesters, for one year. Then I moved on to Munich, again following advice from various people that Sommerfeld’s school was a good one. This was coupled with Munich’s being a pleasant town also.
Were you still going to be an industrial physicist?
Well, I don’t know whether I was thinking about it at the time, but I wouldn’t say that I had any very conscious plans. I mean, I think that if at about that time you had asked me what I was going to do, that would have been it.
Yes, but would one go to study under Sommerfe1d even if one were going to go into industry?
Yes. The funny structure of the courses in Germany at that time was that there was no Bachelor’s degree. The first examination you took in physics, if you did not want to be a school teacher, which requires a separate examination and which then involved some educational subjects, was your Ph.D. There was just no other way of finishing the university course, so you had to do some research. You got into research gradually; there was no time limit prescribed and no particular qualifying examination, but when you felt you were ready, you went to some professor and asked him for a problem. There was only an overriding minimum time limit of three years for the whole course; I mean, you couldn’t present your (big) thesis in less than three years. I think in physics nobody would do it in three years, but in subjects like economics or something perhaps it might be possible. I completed my Ph.D. in four years which was pretty short. And of course I had to learn many things afterwards. In Berlin I had really been a member of the community of mathematicians. The students’ mathematical society was extremely well organized and had a very good spirit and really one spent one’s time being a member of that and talking with other people there. Some physicists belonged to it too.
Did they have a comparable organization for physics students?
Formally not, I mean they must have had formally an organization, but —
But not nearly so active.
And probably I might have got involved with that if I had stayed with the physics practical.
I want to ask you one other thing about Berlin. You say that the course was so crowded that you couldn’t begin in your first year. Was this due to the insufficiency of apparatus and facilities, or was there really a large student body?
I think it was probably just space; space and apparatus. But I never learned quite why; I was just told I couldn’t do them —. Maybe the primary limitation was not facilities, but that they perhaps had a rule that it was good for people to get their basic mathematics first, and then do their physics practical. I seem to have a recollection that they said it was for that.
I just wondered if there was a marked increase of students.
Maybe there was an increase of students, but this is still the post-war period. I went to Berlin in ‘25, and things were still pretty short. For example, they had only just got over the period when the mathematics department was desperately short of foreign books because they just couldn’t get them. In fact already in this mathematical society I mentioned — or at least the reason why they were a very close and well-known community was that in the days immediately after the First World War when it was impossible to get foreign books, they would get students together who would get the loan of a copy and then copy it out in longhand. There were still some of these around, although by that time most of them had been replaced. But then one thing was then that the library was therefore a very valuable thing of course, and therefore the departmental library couldn’t be run — it was really for the use of staff. But the staff made it available to the students on the condition that they look after it. And therefore there was a rota of students who would undertake to sit there and supervise it and issue books and so on. Not issue on loan, but one could guarantee that books that came from the shelves went back again and so on. So I imagine there must have been a similar shortage of facilities in the physics department, but I never got near enough to the place to know. So we go back to Munich. There the community was really Sommerfeld’s department. I mean, that was now definitely theoretical physics, and there was very close personal contact. Sommerfeld looked after his people; the whole department, for example, would sometimes go out for a weekend skiing near Sommerfeld’s cottage. Not a weekend; they wouldn’t stay overnight, but for a Sunday. Sommerfeld, who was not young at that time; he must have been about 50 — well, that seems fairly young to me now, but it didn’t then — wasn’t a very good skier, but he would wander around on his skis, without bothering about skiing clothes. He wore ordinary trousers and braces and shirtsleeves and wandered around with everybody; this sort of thing helped to get people together of course.
How did one become a member of this circle?
By being a student in theoretical physics, I think, and by more or less declaring your interest in theoretical physics rather than experimental work.
You would just go to him and say, “I want to pursue this.”
Yes. He took a very great interest in every student; of course the number was very small. And in addition to theoretical physics — that was Sommerfeld’s lectures, again a theoretical course where I now got in at the right stage, and had his mechanics. I got in there at the beginning of the cycle; of course I never got to the end. He gave n an advanced course of lectures — probably not in my first year, but at some stage — on the electron theory of metals, which he was then involved with.
That would be in about ‘27.
Well, let’s see. I started in Berlin in ‘25, so I would have gone to Munich in October ‘26, and I was there until Easter ‘28. I was there three semesters. So, yes, probably in the middle, about ‘27, he was active in working these things out, and gave lectures about that.
And his lectures were different from the other ones. They were famous.
Oh, his lectures were excellent; they were excellent. In fact I think in many respects I liked his lectures better than his books, which are essentially based on these lecture courses. But in the way he developed them later I think some of the changes I’m not so happy about. I think the presentation we had at the beginning, at that time, suited me better anyway.
How was it mainly different?
Well, in electromagnetic theory, for example, one trivial change is that he changed over to the mks system.
Yes, which we can forever deplore.
Right. Well, I’d have to think to pin down other differences, and I don’t think they’re important basically; of course they’re still very good books. I don’t think I went to any more lectures in experimental physics, as far as I can remember. I did do the physics practical, of course. And I went to some lectures in physical chemistry. Fajans was then the physical chemist.
Was there any rule for Sommerfeld’s —
Well, you had to have three subjects. I had mathematics, physics, and chemistry, and in Munich physical chemistry was a separate subject, so you could just specialize in that. And I did the physical chemistry practical; now, I must somewhere or other have done the ordinary chemistry practical. That must have been in Munich; yes, I suppose in my first year I did an ordinary chemistry practical in my first semester and then changed to physical chemistry. No, I probably never did a physical chemistry practical, probably just did the two semesters of the chemistry practical and the lectures in physical chemistry.
Were these side subjects much emphasized when you took the practical? Were you required to have a considerable knowledge of them?
No, no. These were what we would call subsidiary subjects. But I spent rather more time on mathematics than on chemistry. There was Perron, whose lectures I did not find very exciting and Caratheodry who was great fun,
How was he?
He’s, to me, the prime example demonstrating that whether the lecturer is effective has nothing to do with whether or not he gives well organized lectures, or can in fact even lecture. His lectures certainly were extremely confused; I mean, he never had any tidy notes, and he tended to get mixed up. I remember one occasion when he appeared — we had reached the stage where something or other had to be proved, and he came in to say, “I had thought last night of a method of proving this but on walking to the University this morning I realized this was wrong and we have to do it some other way.” And he spent all the hour of the lecture messing around trying to get it in some other way, until at the end of the hour he said, “Oh, I just realized the proof I thought of last night was all right after all.” Then in the next lecture be proceeded to give that proof. But that didn’t matter; in fact I got much more out of this because, apart from the enthusiasm, getting in this way an insight in where the real difficulties are and what you do about them is far more valuable than a polished course of lectures in which everything is neatly presented.
What was it that he taught at that time?
Variational calculus was one of his courses that I attended. And another one was, I think, relativity, which he approached in a rather axiomatic way. This was the first time, for example, that I met spinors; you know they weren’t called spinors in those days, but projections from a hemisphere onto a —. Anyway this was essentially a two by two representation of the Lorentz group, and so on. He didn’t give much insight into relativity.
But into his methods of working.
Well, and into the structure of spinors and so on.
Sommerfeld of course talked about relativity too; that wasn’t the only place one –-
Yes. I don’t think however I had a course — no. I knew about relativity from somewhere; I imagine that I must have read. I remember in one of the first problem sheets that was set out in Sommerfeld’s mechanics course I caught him out in a question — him or his assistant or whoever had set it — which was wrong. He was then concerned with variable mass. He was anxious to bring in very early the idea that Newton’s law should be the time derivative of momentum, d(mv)/dt, rather than mdv/dt, and therefore we did examples quite early with variable mass. He had an example of a train, or some object, moving along from which water would evaporate. So in my answer I worked out the thing; first of all I said what was obviously intended, and then said, “Well this gives the intention, but this is not the correct solution because the steam which evaporates would of course carry the same velocity as the moving train unless you stop it somehow.” Then I gave what was the correct solution, and I remember also having another reason to see that the expected answer was wrong. That was that it would contradict the principle of relativity, because if you went to a framework where the object was at rest, then evaporating water couldn’t get it started, unless it was thrown out in some particular direction. At that time Bethe was one of the assistants; he was a year ahead of me and was employed in marking these problems, and be told me later that that made quite a stir. I was of course then called to present my solution, and so I think I was noticed as understanding something at that time. Then Sommerfeld also conducted a seminar in which everybody took part, and again, as there was no clear division between undergraduates and graduate students, many people came to it. This was up to you whether you came or not. Well, you had to register for it like for any other course, and you paid a fee which was probably something like 10 marks a year or something like that. So I went to the seminar, and then I imagine it must have been in my second semester in Munich, so my semester as a student, that I was asked to report about the transformation theory. Then the Dirac and Jordan papers had been –
They were really advanced in those days.
Yes. Perhaps because I came with a fresh mind to this; I mean I had only just read the Schrodinger papers and. therefore was willing to take Dirac and Jordan from first principles. I found it relatively easy to see through the connection of this with Schrodinger. I gave it in two seminars; it couldn’t be done in one, and I think made quite a good job of it. I remember the professor of experimental physics, Willy Wien came along, and he didn’t like the old quantum theory, let along quantum mechanics, and I think Sommerfeld dragged him in especially to this to see if it would persuade him. He tried to raise all sorts of objections and I had to argue with him.
What kind of objections did he raise?
Oh, I don’t remember, but you know that if people are used to classical physics and don’t understand quantum theory, there’s a lot of objections. I can’t remember now what they were.
No, but to the presentation itself?
No, no, no. The argument got beyond transformation theory, and it was just the idea that the concepts of quantum mechanics were inconsistent. Now I think, let’s see, yes, that must have been after the uncertainty principle.
The transformation theory?
Yes, or at least my seminar on it.
Transformation theory made a general stir.
Yes, well, as far as I was concerned. It was regarded by the other people as something extremely difficult, that one couldn’t understand. So for a second year student to explain it –-
That was Sommerfeld’s position too; he hadn’t spent much time at it?
That’s right. You see Sommerfeld always was rather more interested in practical things and how to calculate things and be never was very bothered by the sort of basic principles, except where you had to use them.
Well, how did he feel about the Schrodinger equation then? He must have been happy with it.
Oh, yes, yes. But he certainly never belonged to the people who like Schrodinger tried to stick to literal interpretation of this as real waves.
Was he satisfied at the beginning with the probability approach?
I think so. You ask some question in fact about how I first got into this, and I just simply cannot remember, because you see all the fundamental things, I mean the Schrodinger papers, and the Heisenberg papers and so on, came out at a time when I really wasn’t yet reading original papers.
You were doing mathematics in Berlin.
Yes, I certainly wasn’t trying to follow the recent literature. Therefore I really met wave mechanics only when it was presented as an accomplished thing. The various pieces of it didn’t seem to — weren’t clearly distinguished. I remember being a student in Berlin when somebody was saying that there were some exciting new papers by Heisenberg from Gottingen which were using matrices. But this was without my having the slightest idea of what this was all about.
What did Sommerfeld feel about the matrix approach?
Oh, I think he had no objection to it, but he was I think more comfortable with the Schrodinger way of calculating things. He would certainly use matrix methods wherever convenient. Now I never in Munich got to the point of actually starting on any piece of research. I probably would have been about ready for that when Sommerfeld went off for a sabbatical to America somewhere and therefore advised me to move. And on his advice I moved to Leipzig where Heisenberg had then Just about started a flourishing school.
Before we leave Munich. The students must have formed quite a close group; were there meetings other than the standard colloquia and so forth?
No, it was quite informal; we had a room where we were all sitting normally when we had nothing else to do — I mean where we did our work. And one met there and talked, but there were no organized meetings. I was then enthusiastic enough to try and organize a society more or less on the pattern of the one at Berlin, and we did a little bit. I mean, we followed the Berlin practice of sort of working groups that met occasionally alongside a course of lectures and tried to discuss things. It petered out because there wasn’t enough enthusiasm amongst the members. I mean students liked it, but nobody wanted to be an officer of the society or do any work for it, so it really didn’t come to very much.
Someone has told us that Sommerfeld used to come in every now and then and grab a student out of his room to speak with him about whatever he was currently interested in.
Yes, yes; I think that’s right. Well, I was one of the youngest and probably this wouldn’t happen to me so often, but it did happen occasionally.
This would be whatever he had on his mind particularly at the moment? You would just talk out -–
Yes, yes. I can’t recollect learning anything very definite from other students, from talks with them. Certainly when one had difficulty, if one saw something in a book, or one read something one didn’t understand, one might get somebody to try and explain it, but I have no recollection of getting any major ideas from any of the other students. This may be my memory. I think you mentioned here [on the outline] a list of people who were then at the various universities I attended. I haven’t looked over your Berlin list; very few of these names mean anything to me from that period. I went to Bothe’s lectures and Planck’s lectures, and I don’t think I met anybody else of this list at that time. I met many of the mathematics professors and lecturers and students. Munich: Ott, I remember; he was a sort of extraordinary professor probably or assistant. Wien I knew; Bechert was Sommerfeld’s personal assistant. Bethe was a student just a year ahead of me. Houston and Pauling I remember; they were visitors. Rabi and Teller I think didn’t overlap with me in Munich as far as I recall; I don’t remember. But there were others; Eckart was there. Ufford I remember. He is now the chairman at Pennsylvania; I remember him only because he was sitting at all the lectures with an enormous book in which he wrote everything down.
Was the main topic already electron theory of metals?
That was Sommerfeld’s main interest and many people were working on it, but by no means everybody. A man I remember very well from those days was Unsold the astrophysicist. I don’t think he was in astrophysics yet, but he was working with atomic transitions and so on; he was naturally in that line. Ott I think was on x-ray diffraction problems and so on. And Sommerfeld was still interested also in his old ideas about diffraction of electromagnetic waves and so on. I think there was still some work on that going on.
Was there any interest in trying to develop the theory to quantum electrodynamics or relativistic wave equations?
No. As far as I know there was no work on that going on. I mean, no doubt people were interested in those problems. You see the whole tendency of the Sommerfeld School was to pick up fields where you understood the basic principles and where you could calculate something. Sommerfeld was always ready to calculate, even when, as one now looking back sees, the principles were quite inadequate. I mean, for example, his first series of papers on the electron theory of metals work out various coefficients for which really the basic ingredients haven’t yet been understood. You get some number.
And he wasn’t particularly concerned about that if one pointed out the difficulties involved that were still in the calculation itself?
Well, no, no. I mean, he then noticed presumably that the answers didn’t agree very well with the experiment, and be said there is still something missing. No, no, I mean, he had a sound understanding. And he was very happy when, for example —. It was at first not clear; in those papers it was already indicated that the electric resistance of the metals might have something to do with the lattice vibrations, but it was not quite clear why first of all in the absence of lattice vibrations it would have an infinite conductivity. That was done later by Bloch. Or why you should have no resistance in a perfect metal at temperature zero because you still have the zero point motion of the lattice vibration. Now when the various steps came to put that right he was of course very happy; he was satisfied —.
Was the interest in the theory of metals continuous with Sommerfeld? I mean there was so much work done in that first 10 years of this century, and then it sort of passes out, and then it begins again, or seems to.
I don’t know whether Sommerfeld did any work before about 1926 or ‘27, but he must have known certainly the problems of the old Lorentz theory. Then Pauli’s work on paramagnetism showed at least that one of the contradictions could be resolved. Then Fermi generalized the statistics, and then Sommerfeld saw that you could now get over other difficulties. Let’s see, was Sommerfeld the first one to point out that you could resolve the specific heat difficulty? Yes, I think so. He knew that there was a contradiction with specific heat and that Fermi’s statistics would put that right. That led him then to say, “Well now it looks as if we are over the main contradictions, let’s see whether we can tidy up the whole field.”
Pauli is sometimes said to have been responsible for the suggestion that one can solve these older problems by applying the Fermi statistics to the electron and so on. Do you know whether Sommerfeld proceeded to that himself, or whether it was (Pauli)?
Well, I believe this was done by — wasn’t Pauli’s paper on paramagnetism written when he was in Munich?
I think so; it was quite early.
I think so. And certainly Pauli showed how the problem of paramagnetism can be resolved there; those methods can be resolved that way. I don’t know whether Sommerfeld had any part in that discussion; it might well be that Sommerfeld, for example, was aware of the problem. I’m not clear whether Pauli at that time would have realized himself that there was a contradiction; he might have. But at any rate whether then Pauli also suggested that this would work for the specific heat, or whether Sommerfeld saw that independently; I don’t know that we’ll ever know that.
Well, it’s not a matter of great importance.
Sommerfeld certainly was interested in looking at all the various interactions of magnetic and electric fields and thermal gradients — the Hall effect and the Peltier effect, etc. — and he had a long list of them. He was going around with a little book by (Baedeker) which was then the reference book about the properties of metals and definition of the various coefficients and seeing how far things could be made to agree with the theory and how far they couldn’t.
But the seminars continued of course to discuss everything?
Yes. But they were probably more what you would call today journal clubs, in that somebody usually reported on a recent paper. One would pick in general a sort of coherent story, not just one odd paper which had been published. One tried to cover a particular field.
How responsible was one supposed to be for attending to detail and so on? It seems to assign a student in the second year transformation theory and what not —
Well, I think in general the senior people present would know the papers as well as the speaker and would try to correct where necessary or explain, except for the details, I mean, they wouldn’t necessarily have looked into all the details of the calculation, but they would understand the principles and see that they came out right, came out clearly in the end.
Had you decided when Sommerfeld left what you would like to do for your dissertation?
No. I don’t think I was ready for that yet. But I just moved to what seemed to be the next best place.
Had you any notion of going at all to Gottingen, say?
No. Well, I suppose, you see, I must have been rather docile in those days because I always seem to have done what people advised me to do.
But you didn’t regret it?
No, no. I think Sommerfeld was not too impressed with the Gottingen school, its being somewhat highbrow and a little different from his outlook. Well, I don’t know; the question never came up. He just suggested that Leipzig would be the place. He knew of course Heisenberg very well as one of his pupils who had just gone there a few years previously and who he knew was building up a nice and active group, and this seemed a good idea.
Do you know very much about Sommerfeld’s attitude towards the Gottingen school, which seems to be a quite interesting matter? There certainly are differences in approach that are quite clear in the –-
No, I don’t think I have any more information about that than the impression one gets from looking at the work of the two schools. But certainly Sommerfeld tended to be much more down to earth, with the emphasis on numbers and on practical calculations. But I don’t recall any comments of his, and I certainly didn’t know him well enough to expect to hear such comments —
So there was no difficulty in switching immediately to Leipzig?
That’s right. Now there the set up was similar. There was again a seminar room where the students normally worked unless there was an actual meeting on. I spent most of my time there; I did go to some lectures in pure mathematics; I did do a practical class in radioactivity which was amusing.
That would mean experiments?
Yes, very primitive in those days of course. There was a man called [Erich] (Marx)… And I also insisted on doing an advanced practical class where you did really your own experiments, because I thought it was wrong not to have anything to do with experiments.
What did you undertake?
Oh, I remember the first job I was given was to prepare an (interferometer) plate, just a (parallel), glass plate, half silvered, and use that to observe the fine structure of one of the mercury lines. And I was essentially given the task and given the spectrometer, — not a very sensitive one because that was only for the rough division of the lines, — and a little piece of glass and told to get on with it. I first of all had to find out how you silvered a piece of glass. There are books on that which emphasize that the glass must be very clean, and in order to clean the glass and get all the grease off you immerse it in caustic soda. Being very careful and anxious to get it really clean, I put it into a stronger solution for a longer time than was stated. After I got a perfect silver coating on it by the recommended method, I put it into the beam and saw no interference fringes. Then I went to the lecturer in charge, and said, “I’m sorry I must be doing something wrong; I can’t get any fringes.” So he said, “Nonsense, you must be doing something silly,” and he went along and tried, and he couldn’t get any fringes either. So there the thing was stuck for some time; I probably left it and went back to some calculations, and then eventually it dawned on me what had happened, namely, — no I think I just did it experimentally.
I tried to see whether reducing the aperture would do it, and I took a piece of black paper with a small hole in it and put it in front of that glass plate, and there were the beautiful fringes as expected. And then it dawned on me what had happened; namely, by immersing the thing into caustic soda for too long and in too strong a solution, I had ruined the optical flatness of the surface, and therefore of course the different (patches) of the surface were no longer in phase with each other; but by taking a small aperture, just a pin hole essentially, I could get it back. Then of course having to expose my photographs for about 10 or 20 times longer than would otherwise have been necessary —. This was quite an education. Then the next experiment was to measure h with the photoelectric effect, or e/h, I suppose. There again I was given some equipment, and that included — I was given the choice of using a mercury arc where essentially you work on one line, or using filters with a fairly sharp cut off. I didn’t get very satisfactory results, and then I had the idea of using both together. Using the filter in other words to cut out all but one of the mercury lines, and then you were really working with a sharp, almost monochromatic line. Then I got beautiful results; in fact I think I came out with a determination of that number to quite a respectable accuracy, I think probably not as good as could be done at that time, but not so far from it, so I was quite pleased with that.
That’s a triumph for a non-experimenter.
Yes. But of course most of my time was spent on theory, and reading and going to lectures. I can’t remember all the lectures in detail. Certainly one of them was a lecture course by Heisenberg on thermodynamics which was just that particular phase of his three year cycle. And it became painfully clear that he did not like thermodynamics, but otherwise Heisenberg was an extremely stimulating influence. The first job I was given was rather unsatisfactory. An experimentalist [E. Rupp], in measuring canal rays, you know the rays emitted by recoiling ions had got some peculiar effect for which he had a theory which didn’t make sense. This however suggested to Heisenberg that there should be something involving the Uncertainty Principle, as these atoms pass the field of view only — for a short time and so there should be an appropriate broadening and so on. He told me to work this out. Now, I sat down and just tried to set up a general theory of this, which was difficult; I suppose it wasn’t really that difficult, but I was inexperienced and found it rather difficult. And I spend a month or two on that, and then I had occasion to check back on some numbers — that one should have done in the first place of course — and realized that the Uncertainty Principle was way out; I mean, the times of exposure for each atom were far too long to make any difference. I went back and told Heisenberg about that; be was very annoyed.
He hadn’t checked the numbers himself because he just assumed that the theory made by this experimentalist was at least not wildly out in order of magnitude, and therefore the numbers should have been appropriate. So this was dropped quickly. Then I think Heisenberg told me to look at the Hall effect in metals. Bloch was then in Leipzig, and so the Bloch theory of metals was in the air, and I think Heisenberg told me specifically to look at the Hall-effect, where the anomalous sign was unexplained at that time; I think Heisenberg must have had a fairly clear idea what he was after, because this was a time when by looking at atoms Heisenberg had noticed that if you had a complete shell minus one electron you could really describe that in terms of a full shell and a positive hole. And he wanted to extend that principle. I remember his mentioning this general principle; I’m not sure that he did mention it in connection with the Hall effect. But of course this is really what the Hall effect is. It is that if you have a band in metals which is filed and you take a few electrons out then you essentially are dealing with positive carriers, and therefore the deflection of the magnetic field has the opposite sign to what it usually has. Well, anyway I got that picture out, that you could in certain circumstances get the opposite sign and that was the first paper of any importance I published. Rather typically those papers, not only that one, but other papers of about the same time were published in a form which I would not allow my students to publish today. It was not then regarded as the responsibility of the professor to see that things were clearly written. He of course took enough interest in the work to satisfy himself that the results and the method were right, but then —
But of course you wouldn’t publish without his express consent.
Well, of course not. You would want to discuss it with him, but then at that point one lost interest. Also since the journals didn’t read papers at that time, horrible things passed.
That is, papers that came from well regarded institutions were published directly.
I think everything was published directly.
Everything; there was no refereeing whatsoever, even for —?
No. Well, for one thing the editor of the Zeitschrift fur Physik was a busy man, and the number of volumes to be published in one year was not laid down, and his fee was so much per volume. Certainly there was no financial incentive on his part to —. I don’t think that he deliberately accepted bad papers, but certainly there was no incentive from any end to economize.
And was that true also of other journals, like Physikalische Zeitschrift and Naturwissenschaften they all just published what they received?
I think so, unless it was obvious nonsense.
You say that was the first paper you published of any importance. What did you publish before?
Oh, I think I had just before that — it might have been after, I don’t remember — published one or two small notes on points of detail which —.
According to my bibliography, those did come out later. The one on ionization and the one on the existence of stationary states? Yes, those I think came later. Weil, most of the people were doing the theory of metals there, but Heisenberg at that time was involved with Pauli, wasn’t he?
Just beginning, I think. Also, that was the time when he did the theory of ferromagnetism. Well, that’s also theory of solids in a way of course. Well, I didn’t know much about that. I mean, that was Heisenberg’s personal correspondence with Pauli, and I don’t think that was discussed much.
That was done almost entirely by correspondence, was it?
Oh, everybody traveled in those days, and I don’t know where they met to discuss these things, but they must have met.
Do you recall whether he gave the impression that things were just about to be set right in quantum electrodynamics?
No, I don’t know. I don’t think the subject came up, as far as I remember. Let’s see, who else was there? Bloch certainly was interested in the theory of metals. Guido Beck was there, who I think then was mostly interested in regularities in (most) radioactive elements and things like that. He was Heisenberg’s assistant. Well, the place was unusually lively because quite a lot of people had been displaced by Sommerfeld’s not being there; people who either had been with or would have gone to Sommerfeld. Rabi, for example, was there. Placzek was there for a short visit; he was never there for very long; Halpern was there; I discussed a lot and learned a lot from him, although he always had difficulty. He had been brought up on the old quantum theory and still was a bit suspicious of the new ideas and occasionally had difficulties, and off and on, contradictions.
Do you recall what the sort of difficulty with him would have been?
I recall only one where he looked at the formula for the Lande g-factor and put specific values in and then got some apparent algebraic contradiction that two positive spins could add to something that was negative and so on. I couldn’t explain it then, but it was simply the usual misunderstanding that one is inclined to, when you talk about angular momentum vectors, say an angular momentum of magnitude L when you really mean of magnitude the square root of L times L plus 1. You say a spin’s pointing in the +z direction when you mean its component in the g direction is as large as it can be, and so on. It’s a sloppy convention in which one has to learn that words are used in different meanings, and I think the misunderstanding was just arising from that.
Of course one has inherited all that from the older quantum theory. I can see where that would be a particular difficulty.
Right. If you try to translate the new formula into the old model, well, then you get all sorts of contradictions. So this was the kind of trouble where he came in one day and said, “Have you ever noticed —”
Was there much contact with Debye’s work at some point?
Yes. Well, we were in the same building; there was a general colloquium where both the theoretical and experimental sides met. Of course occasionally experimentalists came to our seminars; there was a fair amount of contact.
The question of the interpretation of the quantum mechanical formalism was no doubt discussed at those seminars.
No, that seemed settled by that time.
That’s so surprising, that you have such isolated hold-outs, such as Schrodinger and Einstein without ‘war.’
Oh yes, I think to the people working in the field it was completely obvious that that was right. I mean there might be individual cases where you might get into logical difficulties and have to think how to set it right, but nobody, as far as I can recall, had any serious doubts.
Well, I think according to our notes anyway Slater and Teller were also there at the time.
I think I met Slater in Munich, but only for a short time. He sort of arrived as I was about to leave or something. Teller came to Leipzig, I think, shortly before I left, well, perhaps for the second semester I was there; yes, that is probable. We probably overlapped for one summer — no, one winter. I remember in those days one of the preoccupations everybody was involved with was learning group theory, because the discussion of the atomic spectra in those days was a very elaborate affair; it was based on permutation groups. This was before the days of the Slater paper which made all that trivial really.
So what did you appeal to then — to Wigner’s papers, or did you go to the mathematicians’ work on group theory?
Of course Wigner and other people had shown the general methods, but you needed for any particular problem the group characters in order to get numerical answers, and those very often you bad to go to mathematical papers to find. It seemed a very difficult art in those days.
Well, it is quite complicated to get out the group characters.
Now spectroscopy is not such a very difficult art — that’s now all so easy.
Do you recall a paper by Dirac; I think in ‘28 or ‘29 where he starts by saying that group theory is too complicated and what one really needs is just a certain amount of it, and here is all you need.
Well, where he used the permutation operator and gets the connection with spin? That certainly helped a lot, but I think the Slater approach —. I mean, with the two body problem Dirac’s approach is certainly very nice and simple; with the many-body problems Slater really makes it much easier.
Well, the papers by Heitler quickly became a little messy and were difficult to follow.
Yes. Still they were the standard thing to use until one had the simpler, more pedestrian, but not more powerful techniques.
What about the Dirac theory of the electron? Was that also much discussed and did people work to master it?
Yes. Now I have to recall when that first appeared.
I think it appeared in the first month or two ‘28.
Yes. Now let me see. I’m sorry I’m getting a bit confused about my own dates. I was ‘25 to ‘26 in Berlin; ‘26 to ‘28 in Munich and ‘28 to ‘29 in Leipzig. I have an idea that we already had some lectures about it, or seminars, in Munich. And I think everybody realized that this was a step in the right direction and that the difficulties about negative energy states had to be removed somehow or other, but of course one wasn’t sure how.
But that it would be along the lines of the Dirac approach and one had just to –-
Not immediately. I mean, the ‘hole’ theory was a separate new idea which wasn’t in the first paper.
No, it’s a year and a half later.
But did one follow the developments of that first since the Klein-Nishina work and so on?
Oh yes. There would have been seminars about this. People immediately learned the techniques of how to handle the Dirac equation. And Sommerfeld I think immediately started producing some papers working out, I think, the relativistic photo effect from the Dirac equation, if I remember right. I have an idea, but I wouldn’t be sure.
Well, the techniques are getting more elaborate every moment at that time. Was there any feeling of relative despair at this?
Oh no. One could still digest it at least. One couldn’t immediately be an expert of every one of them, but you could well understand what other people were doing.
The Hall-Effect paper was your thesis?
No. That wasn’t substantial enough for a thesis. Now Sommerfeld had probably expected that I would come back after a year when he was back, but by this time I liked it very much more in Leipzig.
May I ask why that would be — was it the atmosphere of the physics?
Yes. Well, I mean, perhaps I got more involved with the more modern things, and more seemed to be going on there. It seemed to be more up-to-date, as it were, than in Sommerfeld’s group. However, then life was again complicated because Heisenberg went off for a sabbatical year to America.
Oh yes, that was at Chicago, wasn’t it?
I don’t remember, probably. Then I imagined, with his advice that I should go to Pauli. This was at Easter of ‘29. Then almost at once Pauli suggested this problem about thermo conduction in insulators, which became my subject for my thesis. Now it is really a mystery to me, I mean I can’t quite get my time scale right, because obviously I moved to Zurich in the spring. By the late summer I had finished and submitted my Ph.D. thesis which was submitted to Leipzig because I hadn’t got the residence qualification to get a thesis in Zurich; I would have had to wait longer.
But it was written under Pauli’s direction?
It was written under Pauli, but it was submitted to Leipzig. I mean, I didn’t start on that problem until I got to Zurich. Actually I didn’t make a start very quickly, because I had then been approached by some publishers asking whether I would translate the book by de Broglie. [Introduction a l’etude de la mecanique ondulatoire (Paris, 1930); German translation by R. Peierls (Leipzig, 1929); English translation by H. T. Flint (London, 1930)] This was an amusing situation because the publishers knew that it wasn’t going to be published in French; it was too speculative. French publishers wouldn’t risk publishing something as abstruse as that.
Which book is this?
I think Particles and Waves, or something like that. That was about the second book he wrote, I think. And one knew there was an English edition coming and they were bringing out a German, and since many people would buy one or the other, it was vitally important for them to be first. So they came to me and said, “How quickly can you translate this?” Well, I said, “If you’ll pay for a typist, so I can dictate straight into the typewriter” — I wasn’t permitted this sort of thing then, but nowadays I’d say “if you’ll give me a tape-recorder —” “I can do it in a very short time.” And this I did, and this was done when I arrived in Zurich.
You certainly were busy. Why were you approached, do you know?
Well, the publishers were in Leipzig, like most German publishers at the time. They probably talked to the physicists to find if there was somebody who might be willing to do that, and they suggested my name. I had tried some other pieces of research before. Heisenberg suggested I should try to see whether one would get conductivity out of the Heitler-London model. In other words taking into account the repulsion of the electrons and (???). Each atom would tend to be neutral. It is now clear that you cannot get conductivity out of their model, because electrons can’t get past each other. They have to move bodily for which there is not enough probability, but I spent some time trying to get something out of it, and it didn’t work. That probably was before the Hall-effect, because I remember a summer — this must have been the summer of ‘28 in England. This was just vacation travel, but I thought it would be nice to visit Cambridge, and I wrote to Dirac whom I had met at some conference where I had been told to look after him and take him to the theater or whatever was necessary. And I asked whether I could visit there, and of course he said yes. And I was very impressed.
Of course this is the advantage of foreign travel, because I also met R. H. Fowler then. And although I was quite a young student, I was treated as quite a senior visitor who was coming from Heisenberg; I mean this was something interesting. They were obviously short of a speaker in their little colloquium, which is called the Kapitza Club — this was a thing which met in the college once a week in the evening. And they sort of asked, “Can you give a talk to that Club?” Now in those days my English was still rather poor, and I really didn’t have very much to talk about except to give an outline of Bloch’s theory, which they knew, and say a bit about my own attempts to get conductivity out of the Heitler-London model. Anyway I was very impressed at being treated as such a senior and important visitor. Pauli was certainly exciting from the very beginning. Now, he presented me with this problem and some handwritten notes where he had tried to work things out. This is, as far as I know, the only subject on which there is a wrong result published under Pauli’s name. But it’s only a summary; there is a report of a meeting of the German Physical Society where Pauli gave a talk about the damping of lattice waves in polar crystals, which is again more or less the same problem; what he stated there was clearly nonsense. And I think he knew it wasn’t right, and that’s why he told me to look at it and gave me some help at the start. Then I could see that there were some very interesting aspects of that problem, and I wrote out a Ph.D. thesis, which I typed myself with a typewriter that my parents who were then just visiting had given me as a present.
You learned to type and wrote your thesis at the same time?
Yes. Then I went back to Leipzig to submit it and for the oral examination, which rather frightened me.
Was Heisenberg back for that?
No, he wasn’t; Hund was my examiner. Well, the physics was all right. There one had an oral examination in three subjects, each lasting for about an hour. Hund was my examiner in physics, and he asked me lots of questions about three-dimensional rotations — spinning top and so on, about which I did not know very much, but I got through it somehow. Then at the end he explained, “I only asked you those questions so you shouldn’t think you know everything.” But the mathematics examination was delightfully easy with Lichtenberg who was an expert in differential equations; I think he started off by asking me to state, not prove, but state, the existence theorem for the solutions of the differential equation of the first order. When I stated that he said, “Well, now isn’t this wonderful; so often when these physicists come along they don’t really know any pure mathematics. It’s so nice to see one who knows something.” The rest of the questions were on that level. However, what frightened me was the chemistry because in Leipzig, chemistry was one subject, but they had three professors and you can there choose by whom you are going to be examined. I chose of course the professor of physical chemistry; I called on him two weeks before the date, introduced myself and explained that I had only learned physical chemistry. And he said, “Yes, but here chemistry is one subject, and I must ask you questions over the whole field.” I said, “Yes, but I’m afraid I don’t know any organic chemistry at all.” “Oh,” he said, “I won’t ask anything very difficult, but I of course must expect you to know this and that, and so on.” These were all things I had never heard of, so I had two weeks to ‘swat’ up the elements of organic chemistry, and I made it. So that was that, and then meanwhile Pauli had offered me the post of assistant, which I accepted with great delight.
Was it frequent that a student did not get by those oral examinations after producing a dissertation?
Not frequent, but possible. You know of course the story about Heisenberg’s Ph.D. examination, where he very nearly failed. And another case was Max Delbruck; you know his name probably; he’s well-known. He was examined by Pohl and failed the first time. Pohl thought that he didn’t know anything, and be was absolutely shocked by all this and went to Born and said that he would of course repeat the oral examination after learning some more, but please could he be examined by someone else. Born went to plead with Pohl, but Pohl said, “No, I’ll examine him again, and I promise you I’ll pass him whatever he says, provided you don’t tell him that. So it wasn’t so much a matter of —.
Was Pohl like Wien? I mean was he difficult for reasons that are rather obscure?
I don’t think so, I mean I never met him; I never was in Gottingen, but I never heard that be had a reputation for being difficult.
Had one begun to feel yet at Leipzig any of the anti-Semitic pressures that were being felt at other universities?
Well, of course it was almost always imbedded in German life anywhere, but, well, I suppose you just saw to it that you kept out of the circles where these things were common. It was certainly very common amongst the sort of organized student societies, you know, like the student corps. Most of them were anti-Semitic unless they were actually Jewish organizations who were trying to imitate the others; this was a little more ridiculous than anything else. Well, I remember in organizing a mathematics society in Munich I had one man who was a mathematics student and who was — I think he was one of the officers — a member of one of the Jewish corps, again with a ridiculous uniform and cap and so on. He apologized to me that if I wouldn’t mind when we met in the street while he was wearing his cap, he wouldn’t greet me, because a student in one of those cores is not supposed to greet a student who is not organized because that’s an inferior and so on. So it was always there, but one didn’t in those days take it seriously. One certainly didn’t guess that this would go to any — at least I didn’t guess that this would go to anything serious.
Leipzig would have been a place where there was less feeling among the faculty certainly than at some place like Heidelberg?
Probably. The other thing was that in Leipzig there was no university campus at all, but the various parts and university buildings were scattered over the town. The physics building was somewhere where there was nothing else, oh, a chemistry building next door I think. And therefore you would not normally meet any other parts of the university. You never went into the central university buildings; most physics students certainly didn’t; no doubt there were some who were anti-Semites, but certainly not the research students and staff or anything. This wasn’t noticeable.
Who had been Pauli’s assistant previous to yourself?
Bloch. So when I was there as a student, then Bloch was Pau1i’s assistant, and he then went to Leipzig to become Heisenberg’s assistant, I think.
So you almost exchanged positions.
But you weren’t quite Heisenberg’s assistant.
No, no. I was a student. I think Bloch and I were both students in Leipzig for awhile, or maybe he was just leaving when I came; I’m not quite clear. Anyway then he went to Zurich — that must be it; he left Leipzig when I came. He became Pauli’s assistant for a year and then he returned to Leipzig, and I went to Zurich.
What were the duties of Pauli’s assistant?
Well, nebulous. I mean, the function of an assistant was very largely what one would call a research associate or a research fellow nowadays, except you had some chores. First of all you were expected to be available to talk with Pauli when he wanted to discuss something. Sometimes he had some problem he would like to think aloud about and discuss with somebody, arid then of course one was there. You had to do some odd chores; it varied of course according to where you were; in the case of Pauli, I remember my first job — well, I don’t know whether it was an official duty of mine — but at any rate Pauli suggested it and I agreed to sort out his reprints. He had a large stack of reprints, and I got them into some kind of alphabetical order; I think with cross-references where there were joint authors and so on. Well, I think probably what Pauli said wasn’t that I had to do this, but would I like to do this, and if I did this I would then have the right to use the collection of reprints. There were other odd Jobs; for example when there were visitors coming there was a certain amount of practical work of organizing things and this obviously one did.
Koenig I think succeeded you, didn’t he?
No, he was earlier. He was before Bloch; he was I think Pauli’s first assistant.
Well, in any event, he has published a letter from Pauli’s invitation in which Pauli says that his job is mainly to dispute with Pauli. Was that the case?
That is true, though, that wouldn’t be restricted of course to the assistant. I mean, Pauli would talk to anybody that was around. Now, probably, I would think that at the date when Koenig went there, there may not have been many other people around. There weren’t probably any Ph.D. students, because Pauli hadn’t been there very long, so then really the point was that Pauli wanted at least one person to talk to. In my time, well, there weren’t many. It was probably true that Pauli talked to me more than to other people, partly because there was the feeling that I was paid to spend my time with him, and partly because I stayed there longer; the other people came and went. It was all on a very small scale.
Why do you think that was true that the other centers were larger than Pauli’s?
Well, they were all fairly small, arid I think it was already clear then that Pauli did not suffer fools gladly, and it took a certain amount of courage to go to him for a young student. Again the first summer I was there was very lively — the first two semesters I suppose — because now Heisenberg was on sabbatical, and the people who would have gone to Leipzig were in Zurich. So then there was Rabi and Oppenheimer and Bartlett; Bartlett had already been in Leipzig I think. No, Feenberg came later; that was not connected with Heisenberg’s absence. But what I meant really was that apart from the overseas visitors who sometimes stayed a year or so, or sometimes less — there weren’t so many students there. In fact, if I remember rightly, the time I was with Pauli there wasn’t anybody else, apart from overseas visitors. There was a man called Guettinger who later became Pauli’s assistant, who was there part of the time I was there.
How did the overseas visitors impress people? I mean, they had a somewhat different training and so on.
Well, it was hard to judge, because they were usually rather more senior, rather older people. So they knew a lot of things that I wouldn’t know. Also, very often they were experimentalists who wanted to learn some theory, like Rabi, who knew a lot of theory, but who was not by any means a theoretician. But, also, they very often were people who were on sabbatical leave and used that partly as an easy quiet time to do their own writing and reading and they did not often keep very regular hours. So one didn’t see quite as much of them as regular members of the department. This varied; Bartlett for example was sitting as long at the department as anybody else. I don’t think we noticed that his knowledge of physics was in any way different from anybody else. I remember we discussed general topics; and I remember one conversation where Bartlett suddenly discovered that in Europe it was not at all usual for young academic people to go to church. That surprised him; he wasn’t used to it and it took some time to sink in, and then he said, “But where do you meet young girls?”
Well, that’s an interesting question. Where did one?
Oh, at dances or, well, just through friends of friends.
Were these social events organized then through the whole university, or did you organize them yourselves?
Sometimes not connected with the university; particularly of course in the southern places like Zurich and Munich there was the convention of the carnival where you have a few weeks where not much serious work goes on and there are dances every evening. They’re not specially university ones, although the university students would be permitted.
You were Pauli’s assistant in late ‘29.
In October, 1929 for three years.
For three years. But much of that time you travelled?
Well, most of the vacations I travelled. I think there were three months’ vacation in the summer and about two months in the spring, and most peop1e would travel most of that time. So that’s how I got several visits to Copenhagen and to Holland.
But you did have a Rockefeller Fellowship somewhere along the line too?
Oh, that came later; that was when I left.
Oh, that’s the last year then. I see. Then the first year would be the year in which Heisenberg and Pauli must have put together their second paper on quantum electrodynamics.
Yes. And I got of course rather interested in this, and also I talked a lot with Pauli about this —. And sometime, now I don’t know exactly when, he was writing his article for the Handbuch about the general introduction to quantum mechanics. And there we discussed a lot of that, and I argued with him about points, and also I think I read some of the proofs and so on.
What did they feel about their theory at the time do you know; about its possible completeness?
Well, I think they very early discovered that the infinite self-energy was still there. This was a disappointment because, you see, so many ridiculous contradictions or difficulties of the earlier theory had been resolved by quantum theory; therefore it didn’t seem unreasonable to expect that if you looked properly at this difficulty it would also go. But it didn’t, and I think Pauli and Heisenberg always were quite clear that this was a fundamental difficulty.
But they were quite surprised to find it?
I don’t know whether they were surprised, but certainly they were disappointed. I don’t think they thought they could talk themselves out of it very easily.
About this same time then the ‘hole’ theory must have appeared. Was that taken with any seriousness at all?
Oh, yes. When it took the form that you expected the holes to be protons, then that looked a bit implausible. One didn’t, I think, then have the precise argument of charge conjugation to prove that the particle must have the same mass. One had then the feeling that this was very hard to believe that that way you could get the mass difference. Then I think it was Oppenheimer who first came out with the suggestion that the holes should in fact be positive electrons. Then of course, this looked a much more acceptable theory, except that everybody said, “That’s funny, because if such a thing existed, the experimentalists would have seen it long ago.” Of course the fact is they almost saw it, but always managed to miss it by so little. Let’s see, the discovery of the positive electron must have been about ‘31.
In fact it was ‘33.
Yes, yes. It was just when I came to Cambridge. Well, until that time the ‘hole’ theory looked attractive, but somewhat academic, because it was hard to believe that one could have missed such a point.
But as a solution to the difficulty of the Dirac equation, was it accepted?
One was a little worried about it, but it seemed very attractive. I remember once in Copenhagen Bohr saying that this was a foolish idea and contained too many difficulties.
The ‘hole’ theory?
Yes. I can’t remember now which brand of ‘hole’ theory, whether that was the idea of the proton or the positive electron. I tried to argue with Bohr saying that —. I mean, be stressed that this concept of an infinite sea of particles was too contradictory. I tried to convince him that this really was a manner of speaking, which, once you had made that assumption, led to quite a reasonable theory, a. reasonable formalism in which you could work things out quite consistently and which seemed to make sense. I think at that time, Bohr wasn’t ready to accept that, but I mean, this recollection of the conversation seems to show that everybody else that I knew at that time regarded that as —. I mean, I was so surprised that Bohr didn’t like the idea. I think most of us felt that it was all right; we weren’t convinced it was right, of course, but it was at least a possibility.
Well, at what point then did you meet Landau?
Well, Landau came to Zurich twice. I think probably the first time was in my first year as Pauli’s assistant, which will be ‘29 and ‘30. And then he was on a scholarship from the Russian government, and had no end of difficulty in staying in Zurich because the Soviet Union and Switzerland then did not have any diplomatic relations. He needed a special permit which he was given for only a short time, I think first of all a month; and then they renewed it for another two weeks and then another two weeks. Pauli and Scherrer worked very hard to get this extended, but finally they said, “Well, we’ve renewed it so often, we really can’t do it anymore.” So then Landau had to go off; I think he went then to Germany and to Cambridge, I think. Then the following year he came back with a Rockefeller Fellowship, and that was all right.
Was that sufficient?
That’s all right. If Mr. Rockefeller is willing to give them that money, then it must be all right.
Before we continue — just to finish off the Dirac theory. Do you recall how Weyl got involved in the question? I should think there might have been discussions.
Certainly Weyl was in very close touch with Pauli, and he was very interested mainly because the Dirac spinors interested him from the group theoretical angle. He was also aware of the mathematics of this representation of the Lorentz group, and the fact that this could have an application in physics was of course interesting.
But you recall nothing in particular about the discussions with Pauli and Weyl on that?
No. I don’t recall being present at discussions with Pauli, but I probably was; I can’t recall any details. Except I do know — though I was probably not present at the discussion — the fact that Weyl then wrote down the 2-component equation as a possibility and Pauli told him this was nonsense because it didn’t preserve parity, which is true. This I think, in fact, was a perfectly proper remark in those days, because it wouldn’t have done any good to discuss that equation at that time. I think in fact Pauli mentions that equation in his article in the Handbuch, saying that it’s not a practical thing because it doesn’t conserve parity. Well, I think the main point was —. I think Weyl’s motive, like everybody else’s was to try and get rid of the negative energy states, and thought it would be nice to have an equation which has only positive energy solutions. Since the Dirac equation has four solutions for any given momentum — 2 positive and 2 negative energy and you want only two, to boil down the equation to 2 components seemed an interesting thing to do. Of course, Pauli was perfectly right in pointing out that it was no good for that purpose because it was cutting out the wrong thing. It was cutting out the spin multiplicity and not the energy. Otherwise I don’t think I recall any details of the discussions between Weyl and Pauli.
Then it was during Landau’s first visit to Zurich that you began work on this approach to quantum electrodynamics?
And was that written while he was still in Zurich?
No, I think it was essentially completed while, he was there, and I think I wrote it up after he left. Now, you ask how that originated, or why one was interested. Well, I think first of all Landau came already with some such idea in his mind, and we got involved in discussing it. But, also, it was interesting because many people thought that it was impossible —. I mean, the configuration space description which had been so successful in Schrodinger’s work seemed the easier way to proceed, and in fact it usually is the easier way of solving problems when you can solve them. Many people made the statement, “You couldn’t do that with photons, because the photon number isn’t conserved.” Now it was obvious to us that this wasn’t so, but that simply you had to use, not one configuration space, but the set of them together. You get coupled equations, but there was nothing wrong with that.
You credit Pauli, or Pauli and Heisenberg or someone with the suggestion, I believe, in your paper.
You mean the suggestion that it could be done?
Through the simultaneous equations.
That may well be; that may well be. We just got interested in following that up; I don’t think we expected any special, or any known advantage from this, but it was just interesting to see if the equations could be cast in that form and what would follow. I mean, there’s always the possibility that some concepts, for example, the infinities, might be more intelligible that way. Of course we learned a good deal from this attempt, though not exactly what could have been predicted. In particular we learned that the configuration space for photons is a very bad language because the position of the photon has no natural meaning in the theory, in fact cannot be given a relativistically invariant meaning. In particular, for example, to prove that equations in the form in which we had them were Lorentz invariant in content was extremely difficult.
Did you just appeal to the identity of the –-
No, I actually worked it out; I mean it was clear there must be some operator which converts the equations in one Lorentz frame to those in another. I remember — this was already after Landau had left — I wanted to satisfy myself that I could construct this operator. You get involved with terrible integral operations because the equations themselves contain integral operations; also you change from one plane equal to a constant to another, and you get all sorts of complicated transformations. I kept making mistakes, and the thing didn’t work until finally one morning after I’d been at the dance and come back at three in the morning or something, I decided to sit down and do this again. By seven in the morning it had come out, and I was satisfied that the operator existed. I don’t think I ever wrote that into the paper.
No, in the paper, as I remember, you show the connection with the Pauli Heisenberg solution and say that they have already shown the Lorentz invariance of the approach.
Yes, but I think I also stated that we had found the operator which does this, so I didn’t write it down; it was very long. Now this was rather useful to understand, because I later got rather interested — I think encouraged by Pauli — in the question of how far the analogy between light and particles is really significant, I mean, wave mechanics starts really with this equivalence between the two. Yet it is not an accident that we meet electro-magnetic fields classically, first as fields, and then the photon concept comes later, whereas electrons for example first are particles and the wave nature is discovered later; this is not an accident. Actually I had to give a lecture in getting the permission to lecture, which is customary at the university. Well, you have to write a thesis for that purpose which I did, and then you have to give a special lecture for the purpose, and I chose that subject.
Would you tell us about it on the other side of the tape?
Yes, I was saying that the fact that the position of the photon doesn’t have a good meaning beyond the range of geometrical optics, when it belongs both in the wave and the particle picture, is one of the ingredients in this essential difference, which I’ve always been-rather interested in.
Were you about to say something about your Habilitations Vortrag?
Well, this was devoted to this question of the analogy between light and matter and its limitations. I don’t think otherwise its worth going into that.
What did Pauli think of this configuration space approach? Was be at all interested, in it?
Moderately interested, not excited of course. I think everybody came to the conclusion when we had done it, that it was possible to work that way, but it was not probably a very useful tool.
Did Landau bring with him the idea of getting the wave equation from Maxwell’s equations in some form, for the photons?
Well, the wave equation for the photons must have some relation to Maxwell’s equations, and that was his starting point, but it has to be hacked about quite a bit to get it in the form one wants.
Yes, and do you remember any of that process?
It was complicated because one gradually realized for example that one was after a complex solution, and also that in order to have anything like a density operator you have to do funny things. Well, Landau had this bold idea of introducing the square root of the Laplacian operator.
That’s really a shocking notation when you first look at it. But even so, as I remember, you don’t quite get a positive definite expression.
You can do so, but you have then to use the fourth root of the Laplacian; that I think we also do in the paper. It’s not much use, because the density, as so defined, hasn’t got the right relativistic properties; therefore it depends on the Lorentz frame.
So you were content to drop it at that point?
What about the attack on the measurement question which has interested me considerably?
Well, that I think was during Landau’s second stay in Zurich, and it’s very difficult to recall how we got into it. It may be, in fact this again was essentially originated by Landau and he just came along and posed the problem, but it may also have grown out of general discussions; I just can’t recall the details. We had noticed through the earlier work that the photon coordinate was really something that didn’t make any sense. It was therefore interesting in that respect to see whether in the relativistic domain an electron coordinate made any more sense. Now since at that time one was beginning to understand the role of negative energy states, or possibly holes —I think at that time it was still negative energy states —. Let me incidentally make one remark here relating to what you asked before. The original idea of dealing with the negative energy states, I had forgotten that, was simply that one said, “Let us prohibit, let us cut out, all the matrix elements for the transitions to negative energy states.” And it was clear that there was some doubt about whether this could be done relativistically because the states have to be defined as an integral over space, and that way you may be in some trouble. Also, of course, later it was pointed out that if you just prohibited all these transitions, you might spoil things like the Compton effect like the Klein-Nishina formula where as intermediate states they play an important part. But I think this was only discovered when in fact the remedy in the terms of holes was already there.
But you’re not very happy about the negative energy states business at all in that paper?
It’s one of the things to which you point as a possible objection which your considerations might at least enable you to grasp.
Well, no, what we’re merely saying is that the theory is in a mess. This suggests that there might be some fundamental modifications required. Discussions about possible measurements might give you some clue as to what the modifications should be; this was the object I think. The thought was that in the non-relativistic domain Heisenberg had shown that you can in fact invent academic devices which will do all the measuring that quantum theory described as possible. And it was interesting to see whether you could still do that in getting involved with radiation; once you started to think about it, it was very clear that you would meet all sorts of difficulties.
But your selection of the difficulties, as you point to at first, just included whatever was a particularly clear, unpleasant feature of the existing theory. You have there no underlying connection in the things you chose to point to.
Well, we weren’t of course trying to be complete, and we weren’t claiming anything like an axiomatics where you could then prove that (???) were the only objections. We sat down and tried to visualize some simple principle for how you might devise the experiment and then see what difficulties you would meet. Now maybe there were more, but those were the only ones we could see in those particular experiments.
You see it’s after the suggestion of the holes, a solution which you must not have thought very highly of, at least not at the moment in which this paper was written, since it gets associated with the infinite self-energy and other things which seem to be at that time recognized as fundamental difficulties.
One thing is clear: when you introduce the hole scheme, you get new kinds of infinities. In the old interpretation you have all this infinite charge density or infinite energy, and then in the more modern interpretation it comes out as the diverging vacuum polarization; that’s all the same thing.
Of course you don’t quite say that though. Was that —?
Well, I don’t think one saw it so clearly; all you could see was that here you had the scheme full of infinities which gave mathematical troubles.
But so far as you can remember, you’re not quite certain how the notion was taken up that these things might be due to the inapplicability of the wave mechanical concepts?
I don’t think we saw any very direct connection; we merely saw that there were difficulties about regarding everything as measurable which the theory in fact implicitly assumed to be measurable, and when you have a theory which also is in other troubles, you’re not displeased to find that. I don’t think we were ever able to see, although we were trying to see, any very direct connection between the particular things that were not measurable, or that seemed difficult to measure, and the things that gave difficulty.
Part of that preceding year, or perhaps while you were doing the paper, Landau had been in Copenhagen. Do you know if he had talked about these questions with Bohr?
I should think not. He might have just mentioned them, but certainly not in any serious way.
That’s curious in view of the subsequent history of that question.
Well, I think the ideas were only just beginning to take form in his mind. It might be in fact that they all developed during the Zurich time; certainly he had not got very far with that.
Then that paper was written when Landau was present in Zurich, so there wasn’t a question of a correspondence?
No, I think it was written when he was there. I think we worked together.
I have asked a question here about your criticism of the article by Jordan and Fock which considers the new uncertainties. What sort of a place in the history of your own paper does this paper of Jordan play? — Perhaps you have forgotten that.
Well, simply that when we were advanced a certain way, we noticed that other people had made remarks on the same subject.
So that you had already well begun by the time this had come out?
I think so. Maybe it had come out earlier, but we hadn’t particularly paid attention to it. But we thought it was wrong to take an electron as your test particle because, why should you?
Exactly, but that’s a dangerous doctrine.
No, what we always knew was that you might get further limitations if you assumed that the measurements necessarily would have to be done with elementary particles provided by Nature. But the formalism of quantum theory certainly postulates that any kind of —. Well, this is of course where the whole difficulty comes in. If you consider that you can invent anything, any kind of interaction Hamiltonian, then I think you would find that everything was measurable, also. But what we were trying to do was to take into account some limitations, for example, that any interaction in the electromagnetic field could only be measured by its interaction on charges of currents or something, and these would be attached to mechanical particles. So we were well aware of the fact that if you tried to get an inequality of the kind we did, you eliminated the charge of the test body. In other words you took the test body of a charge and mass which were best for the experiment. In fact there are no elementary particles, or even no nuclei, of a charge higher than a certain number.
For one thing, you would not (???), measure a field at a point. Now there I think we were in a general way aware of the points that were later worked out much more carefully by Bohr and Rosenfeld, that to give a meaning to a field you must introduce an average over a space-time area. We knew this but we didn’t pursue it in detail; you could have of course an extended test body of as big a charge as you liked, and there would be no difficulty about this. Now you might of course ask, “Am I now completely convinced that our paper was wrong and that Bohr-Rosenfeld’s was right?” And I think I would say that they are certainly right that quantum field theory is consistent in that, as long as you don’t bring in any new limitations of elementary particles, and so on, you can always measure any observable by postulating the right kind of measuring apparatus which will then have no contradiction to the laws of quantum theory. I still have a dim feeling that our paper is not entirely irrelevant because the measurements you then get to, if you look at the measurements as described in the Bohr and Rosenfeld paper, are really very different in nature from what any normal experimentalist would call a measurement. This is mainly connected with the fact that your test bodies have to fill the space completely in which you’re measuring the required quantity, and they must in fact produce sizeable fields themselves.
So that it is not at all clear whether the measurement really measures the field that would be there without the test body — in fact it doesn’t. I think the complications one gets into with thinking these things out suggest that the concept of measurement is still a degree harder in that field than elsewhere, and that therefore one might not be surprised if in some future elaboration, or future extension of the theory, these things are no longer measurable. However, that’s only a vague feeling, and I certainly agree that the arguments we used are not conclusive because one was later shown how to get around it.
How much interest was there in general in these measurement questions about that time?
Not very much. I think it was particularly, of course, at Copenhagen that there was a fundamental interest in this, and that’s, I think, one reason why Bohr was so upset when we produced a paper on this that he didn’t approve of.
He was quite upset according to this.
Yes. In fact we were in a very embarrassing position because we had of course discussed this with him. Landau then came to Copenhagen early and I came somewhat later, and there were discussions with Bohr,
This was when the paper had been published or was still in draft?
No, it was in draft. Now, it is customary when you have discussed a paper like this with a distinguished person like Bohr and have got something useful out of this conversation, that you acknowledge this at the end of the paper. At the same time, an acknowledgment at the end of the paper carries some implication that the person you quote approves of the contents of the paper. It’s very annoying if somebody tells you of an idea and you tell him that it’s complete nonsense, and then he acknowledges your assistance. Therefore I went to Bohr and said that we were proposing to put an acknowledgment at the end of the paper, but thought that might perhaps be embarrassing to him and asked what he felt. Bohr got very cross about this and said, well —. I can’t recall the exact words of course but the idea was, ‘he was of course trying to think about these things and it helped to discuss them rationally with people, but if we were in any doubt that his help was useful, and so on, then of course he would much rather we didn’t acknowledge it.’ In other words, be appeared to take offense at the thought that we were in any doubt whether to acknowledge his help.
Which was far from your intention?
This was the opposite of our intention, and getting the arguments brought onto that sort of plane was the most painful part of the thing. I mean, I wasn’t upset that Bohr appeared to disagree with our arguments because after all that’s how physics makes progress — that you disagree. But that he had such a strong reaction about this very unimportant point of an acknowledgment was —
That was quite unusual too.
Well, that accounts for the way in which you do handle it. I was quite surprised in reading the paper that several times in the first few pages you attribute to Bohr perhaps even more in the way of development of the measurement question, and subsequently say nothing at all about him which is a bit puzzling.
I can’t remember now whether the acknowledgment on the end did go in or not.
No. I think not. In the beginning you acknowledge that Bohr is the only one who has understood —.
Well, then for that we didn’t need his permission. I think probably it ended up that he insisted that we did not quote him at the end.
Well, you know this piece by Rosenfeld that gives the impression that the discussion was entirely one of physics.
Yes. Well, most of the time was spent on physics; this was one incident which was of course particularly important for me. You quote, I think, this phrase from Rosenfeld.
You “retired in a state of complete exhaustion.”
That’s right. The point is that one always retired from any discussion with Bohr in a state of complete exhaustion because he had this tremendous amount of energy and could go on for days. And of course it was so hard to get to the point of saying what you wanted to say, because Bohr tended to monopolize the conversation and to interrupt as soon as you said anything. Just for these external reasons, whatever you discussed with Bohr, you were worn out at the end. It had nothing to do with the substance.
But do you recall what his immediate objections were at the time? Of course this was all before he had had a chance to go into these matters in any detail.
No, I don’t. We obviously were not convinced by his arguments. Otherwise we wouldn’t have published the paper.
Well, were you content about the results? I mean, there is a sort of bottomless pit that opens.
Well, I wouldn’t say it was a bottomless pit; it just showed that quantum electrodynamics was not in every respect consistent. And of course it isn’t as the results show, but I mean that is not necessarily for that reason.
But this would tend to set one adrift.
No, if the argument could be borne out it would, I think, indicate that one would have to be more careful with the field concept and with the quantities you use in the description.
But it hardly suggests a way in which one could proceed.
Well, no, not immediately.
Had you any notions, yourself?
No. I was at that time always trying to see if one could get over the difficulties by introducing essentially an extended electron, in other words, using what one now calls non-local interactions. And I still haven’t given that up. However, I have certainly also not succeeded. It always seemed to me that that was entirely in the spirit of Lorentz’s picture of the electron. But that of course had nothing to do with these particular difficulties, with these measurement questions.
Do you recall how others reacted to that paper, besides Bohr?
I don’t recall; I don’t think Pauli took sides in this or was very excited about it, but I otherwise can’t recall his comments.
He was not much concerned with these questions in general?
Oh, he was, but I think what he would have said — I guess now — is that this is an interesting possibility, but that he hadn’t finally made up his mind whether he agreed with us or not. I would have thought that would seem plausible.
And all this same time you had continued work on the theory of metals?
Yes. Solids in general, not only metals.
And then came your year abroad beginning with Fermi?
Yes. You ask here why did I go to Fermi. Well, this was a time when it was relatively easy to get Rockefeller Fellowships, and Pauli agreed to recommend me for one. This of course involved spending some time abroad, I mean, not in your own country, and Switzerland one thought about as being almost the same thing as Germany in those days — it was a German speaking institution. Well, looking around the place, Fermi seemed a very interesting person to work with and Cambridge seemed interesting with Dirac and so on, and I thought it was a nice idea to divide my time. I followed there Bethe’s example, though I think not consciously, except Bethe who had held the fellowship the year before had chosen to spend the winter in Cambridge and the summer in Rome.
Certainly meteorologically a bad choice.
Yes, although I discovered the summer wasn’t a very profitable time in Cambridge; too many people are away for too much of the time. Well, of course at that time I could keep myself happy working on my own quite a lot, but I didn’t get much contact in Cambridge.
You say that Rockefeller Fellowships became easy to get; had they been more difficult?
No, no, they were easy to get at that time; it was shortly after that that the Rockefeller Foundation decided to give these standard kinds of fellowships only in the medical sciences, or in subjects important to medical sciences, and therefore it became harder after that. I meant it was still easy.
Did one need any very complete plan to submit to the Foundation?
No, in rather general terms, one needed a recommendation.
And that was essentially all?
There was an indefinite number of those scholarships?
No, presumably there was a limited number, but the competition wasn’t terribly severe, and presumably a recommendation from Pauli had carried some weight.
Almost everyone had one at some time or another.
And once you arrived in England you decided in view of the situation just to -–
Well, actually that was before I arrived in England, because while I was in Rome I got an offer of an assistantship in Hamburg. This was very attractive, because it traditionally was a sort of stepping stone to better jobs. That was the job that Pauli bad held before going to Zurich. The atmosphere in Hamburg was pleasant; Stern was still there, and Stern was a very nice person to be with on the experimental side and so on. So that I had already informally accepted this job, and in fact they were anxious to get somebody soon and therefore I had said I would agree to forego the second half of my Rockefeller grant and go straight from Rome to Hamburg. Then in view of events in Germany, that did not look reasonable. This was in fact before Hitler came into power a little, but one could see which way things were going. I think the precise moment when we took our decision was when Schleicher [Bruening], who was a relatively reasonable person, was thrown out and von Papen became the chancellor. At that point we said that was too much. Then I sat down and wrote a letter to Hamburg saying, “I’m sorry, I’m not coming.”
Had you any arrangements in England at that time?
No. In fact I had of course the Rockefeller Fellowship which would keep me going through the summer. It wasn’t until about August or September of that year that I knew where I would be in the winter.
What would you say about this business of the absorption of academic people from Germany at this time?
Well, it was of course a tremendous problem, but I would say all reasonably competent people got absorbed somehow. Now in England this was of course a time of unemployment, including academic unemployment, and therefore it was not easy for people to get regular academic posts; this would not have been popular. A few got them. There was then a post advertised in Cambridge, while I was at Cambridge. And I thought that it would be very good if one of the people from Germany could get this, and I applied for it myself. But I also wrote to Bethe drawing his attention to the advertisement suggesting that he should apply, which surprised some people. In the end neither of us got it.
Who did get it, do you know?
I can’t remember who got that particular one, but it was one of the local Cambridge people. In fact I remember that I took my application to Newman, who is now professor of pure mathematics in Manchester, and who was then the secretary of the appropriate board, and handed over my papers. I said, incidentally that he probably also had an application from Bethe, and that if any sort of factual information was needed I was glad to help. So he looked with some surprise and said, “So you’re a candidate and you’re also a referee for another candidate?” Then there was an assistant lectureship advertised in Manchester, and I applied for that. I got a letter from Bragg who was then the chairman of the physics department in Manchester saying he was very pleased to have this application and sounding as if there was little doubt that I would get it. He then struck some difficulty, because he already had invited Bethe to come on a job, and obviously to appoint two such people at the same time was more than public opinion, or university opinion would stand for, and I can understand it. However, then they started in Manchester a fund for refugees, an academic assistance council or something like that, and I was then offered a grant for a year on that. I felt a little bad about it because I wasn’t technically speaking a refugee; I hadn’t been in Germany, but I was assured that this would be all right. So I got that and held it for two years; it wasn’t very much money, but it was sufficient.
The English arrangements were more extensive than the French, would you say?
I don’t quite know how it went in France; in fact also there was some possibility of my going to Paris instead of Manchester. I think my wife would have very much preferred to go to Paris at that time. However, she didn’t try to press me, and I think in the long run we did the right thing because the absorption in Paris didn’t seem to have worked so well.
No, and it wasn’t a very permanent solution.
Well, neither was a permanent solution; it was hard to judge how things would develop. It was of course quite worrying. I remember there was then a fellowship vacant at Trinity College, Cambridge, and I don’t think this was something one would apply for — perhaps it was — I applied, or my name was mentioned informally; I don’t know. Anyway I had some hopes of getting this; I suppose somebody wrote to me and said, “Would you be interested,” and I was sort of waiting and hoping. And nothing happened. Then I remember we had tea with some other German refugees in Manchester in some other subject, and they said sort of casually in the conversation, “Don’t you think Heilbron is very lucky?” — that’s a namesake of yours who is now a professor of mathematics in Bristol. And we said, “No, why?” And they said, “Oh, didn’t you know he’s just been appointed to the fellowship in Trinity College?” And I think we were very proud that neither of us gave away that this information had considerable significance to us.
But the possibility of going to the States was never brought up?
I don’t know quite why, but it never really occurred to me. I think we were too fond of Europe. I remember there was correspondence about a professorship in Ecuador; the University of Ecuador was trying to use the availability of many people to build up —.
That would be a bit of an extreme solution to the problem.
Yes. I showed interest, but it never came to anything. In a similar way there was a chair vacant, or a new professorship created, in Raman’s institute in Bangalore, and I applied for that. I wrote in fact to Sommerfeld, who knew Raman very well, and asked whether he would be one of my referees. And Sommerfeld by way of reply sent me a copy of a letter he had written to Raman. It was very funny because this letter read approximately as follows: It said, “Mr. Peierls tells me that he applied for the chair in Bangalore, and I can recommend him. He is quite an able young man; he would have stayed to do his Ph.D. with me if I hadn’t gone to America at that time. While I am writing to you I should like to mention also the case of Mr. So and So, who is an excellent young man and I think is just right for your chair. He really needs a permanent job because he is now without a job and he has a wife and two children, whereas Peierls is single” which wasn’t even true. And I thought it very charming that be sent me a copy of it.
Well, one would suspect the secretary got things mixed up.
I don’t think so. Of course in the end neither this man nor I got this job; I don’t think they appointed anybody. And of course I’m very glad that I didn’t —.
Do you by any chance have any correspondence relevant to the history of quantum physics; perhaps with Landau or Pauli or Sommerfeld?
I don’t think I have any correspondence with Sommerfeld; I don’t think I ever did correspond with him.
You just received copies of letters?
I don’t think I kept that. I have a few letters from Pauli, but not from the early days, I think. That I could easily look up; I do have some later ones. I think I have somewhere, but that may take some finding, correspondence with Ehrenfest. You remember that Ehrenfest then wrote a paper called “Einige die Quantenmechanik betreffende Erkundigungsfragen,” or something like that, where he raised some questions about why he thought the interpretation of the structure of the quantum theory wasn’t clear and asked for views. Pauli wrote a reply to that, which was published. Well, I then wrote somewhat privately to Ehrenfest trying to make some points not contained in Pauli’s paper. I think probably this was during vacation when Pauli and I weren’t in the same place, I didn’t know Pauli had written and so I wrote some comments on the Ehrenfest thing, and I had a reply from him. Whether I have kept that, and whether I can still find it, I don’t know.
That would be of interest.
The point is, since the end of the war, I’ve had a secretary in the office, and I kept correspondence fairly tidily. In fact, so much so, that I now have the problem of what to do with it all when I move to Oxford. But in those early days, I don’t think I kept very much.
Well, if it should so develop that you did, we would be most interested in obtaining copies of them if we could.
I can look in one or two places very easily; beyond that things are most likely to come to light when I pack up to go to Oxford. Would that still be useful; that would be maybe in a month or two?
Oh, yes indeed.
Well, I think I’ve trespassed on your time quite extensively, but I should like to ask one or two questions about your own early work before we stop. It has to do with these little papers you tossed off. I was curious as to where they came from, and how the questions came up.
Well, a lot of my work usually starts when I’m annoyed with seeing something I disagree with, or when something is done in a foolish way. On the question of the paper about the ionization potential I saw some paper, I think in the Philosophical Magazine, which I probably quote, where somebody has looked at the trend of the ionization potential and seen some breaks in that curve at the closed shells, which is obvious, but then also some breaks in between which seemed curious to him and it raised (???). It seems to me quite obvious that one should be able to get that out by much simpler means. Now this was a time when the energy of the many electron system was not a trivial thing because of all the group theory complications. And I was therefore pleased to notice that you could get some combinations of things which would give fairly simple answers, and I proceeded to write that down. Whether today one would regard that as a good argument or not, I don’t know; it certainly makes the trend much more plausible than in the paper I was looking at. That was that. In the other paper about bound states I was just pleased to see one could make a general argument from very elementary means using the variation principle. There is a dreadful ‘howler’ in that paper because I discuss the one dimensional state correctly, and then I remark that in three dimensions, it is the same because the Schrodinger equation is the same in three dimensions as in one. This is true except for the little fact that in three dimensions, of course, the quantity which satisfies the simple equation is r?. Therefore, it evidently must vanish at the origin, and of course if one puts in that condition, the thing is quite different. So that one sentence in the paper is complete nonsense; everything else is all right as far as I know.
I hadn’t noticed that; one reads the one dimensional argument end assumes that the sentence, “that it’s all good for —”
Right, well, that’s how I made the mistake.
Do you remember how that came out?
No. There must have been a discussion about it and there was a little doubt under what circumstances you would or wouldn’t get a bound state. Actually I think what probably happened was that some people were aware of the fact that in three dimensions you do not necessarily get a bound state and were discussing this, and this puzzled me. And therefore I thought I had a proof to settle that argument, which of course I didn’t. But most of what it says in the paper is perfectly correct.