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 Charles Weiner on 1969 August 12,Niels Bohr Library & Archives, American Institute of Physics,College Park, MD USA,www.aip.org/history-programs/niels-bohr-library/oral-histories/4816-2
For multiple citations, "AIP" is the preferred abbreviation for the location.
Aspects of Peierls life and work in theoretical physics. Physicists and physics research in Manchester, Cambridge, Birmingham and other British institutions, beginning in 1933 after Peierls’ Rockefeller Fellowship in Rome and Cambridge. Observations on Russian physics, marriage to a Russian physicist, trips to Russia in the 1930’s. Attitude toward fission and his work with Frisch on the possibility of developing the atomic bomb; impressions of the U.S. in 1942 and his war work in atomic energy research. Other topics discussed are: Chadwick’s cyclotron; Cherwell’s accelerator at Oxford; Powell’s discovery of the pion and development of photographic emulsion techniques; background of the Fuchs case; postwar experimental work at Birmingham; his role in postwar development of quantum electrodynamics; participation in conferences; effect of atomic energy on world politics. Postwar concern with broad-based issues in academic, social and political fields.
Today is August 12th, and it’s in the morning. We’re resuming our discussion. You just mentioned that there were a few points that occurred to you.
Right. One thing: I was a little confused about dates yesterday. I said I was doubtful whether I had been to the Soviet Union in 1934. That was just a lapse. I definitely was there in ‘34, and in fact that was the time when I made the trip through the mountains with Landau that I had mentioned before. Now, 1934 was, I think, also the date of the London Conference you mentioned.
I now remember that I did go to the conference. In fact, I came back from Leningrad especially early, earlier than my wife, who was staying longer, in order to get there in time for the conference. However, it’s still true that I don’t remember much about the contents of the conference.
I was checking last night too on some notes, recalling what Bethe had said about the conference. He didn’t mention whether you were there, but he did recall going to Blackett’s house afterwards with some visitors from France and having discussions there and so forth. But that doesn’t ring a bell either?
Well, I spent so many evenings in Blackett’s house with various visitors that I would certainly not recall what the occasion of that conference was. I guess I don’t recall. There’s one other very different addition I want to make. We started talking about the way one gets involved in new investigation or a new kind of problem, and we got somehow sidetracked. I didn’t finish what I meant to say there, because what I had in mind was that very typically I tend to get involved with a problem by seeing something somebody else has done and getting irritated because either what he has done is wrong or very clumsy and I know it can be done more simply. And this sort of stirs me into doing the problem myself or doing it right or doing more or whatever. I think we already mentioned this one incident where I saw this paper by Kretschmann which just seemed nonsense and I was trying to correct it, and then in the course of that I got into some quite important difficulty. And this is rather typical. I think a lot of my work has started that way.
Is it from seeing a paper in print, or could it just as well be a colloquium talk or something of that nature?
Oh, it could be. I remember one occasion in Cambridge, for example, which is near the time that we’re talking about, about a problem relating to ferro-magnetism, or to super lattices — it’s the same mathematical situation — where it was well known that in a one dimensional problem, which can be very easily solved exactly, you do not get anything analogous to ferro-magnetism (that was known) and I heard a very intriguing seminar by a mathematician in Cambridge (I’ve forgotten who it was), who presented a proof that also in the two-dimensional case this did not happen, and therefore he actually raised the suspicion whether the current theory of the real physical problem, which of course was in three dimensions was correct, because he suspected in three dimensions this couldn’t happen either. Now, this again irritated me because I was sure his proof was wrong, and rather than try to pick a hole in his proof I decided to go the other way and see if I could prove that in two dimensions in fact the phenomenon did occur in this model. That indeed was possible and resulted in a paper published then which is still, I think, being quoted as a useful proof. It wasn’t quite correct in mathematical detail. People later amplified it and improved it, but basically that is the right way of proving the existence of such a transition in two dimensions.
This was from Cambridge?
Right. I think it was published in the Cambridge Proceedings. You have, I think, there a list of mine. It is this one: “On Isin’s Model of Ferro-Magnetism.” 37 is the number on that list that you’re using.
And you credited Bethe with devising the mathematical method for attacking the problem, thanked him for his help in that.
Oh, is that the one? No.
Yes. This is where the model for ferro-magnetism of 1925 had been replaced by Heisenberg’s model based on exchange forces. Then you said, “Since the problem of the model in more than one dimension has led to a good deal of controversy … it may be worthwhile to give its solution.”
That doesn’t seem to have been taken from my abstract. It doesn’t seem very adequate.
I see. It could have been just the abstract and not the paper.
The important point was that it is as a proof that in two dimensions this model gives the right answer.
Your explanation of it is clear enough.
This extract doesn’t quite … Anyway it doesn’t matter.
That’s an interesting point. It illustrates how we get involved in new problems. And that would account for a specific paper, a specific problem. But that’s somewhat different than themes which you’ve pursued over a long period of time because of your own…
Right. But they might still arise in that way. You see, going back to that paper by Kretschmann on transport theory, the thing that came out immediately and which I published in a short note commenting on Kretschmann’s paper was that although his arguments were wrong, he had hit on a significant point, which nobody else had noticed as far as I know, namely that the normal techniques were not really valid, the conditions for their validity were not satisfied; and then, of course, that’s not the end. Then you try to see how to put it right. Is the answer still right, or if not, what is the correct answer? And that was a problem to which I’ve come back on and off over many years.
So the source of your interest is unpredictable.
Any research in a way is unpredictable. But I think one could probably find other examples, but I just wanted to make this point, that for me at least — I think it would be true probably for Bethe as well, but not necessarily for everybody, not everybody works in that sort of way.
Would you say in a sense that this is a characteristic of the Sommerfeld school — this willingness to take a look at practical problems or solutions or even calculations and be critical about them?
No, there is something in it certainly, but… I mean I don’t really regard myself as being so largely a product of Sommerfeld’s school. I certainly did learn a lot from Sommerfeld directly and also indirectly perhaps through Bethe who was very much more typically perhaps a pupil of Sommerfeld. But remember that I spent only three semesters with Sommerfeld, and I think on the whole I probably was more influenced by Pauli and by Niels Bohr, although I never spent any appreciable time in Copenhagen. It was always short visits, very productive for me.
What I meant by the Sommerfeld approach was a down-to-earth approach as opposed to an abstract, theoretical
Well, this last thing about the ferro-magnetism was rather abstract, because it was settling a point of principle.
Were there other points that you felt you’d like to mention in relation to yesterday’s talk?
No, but could we just pursue this question we’re on for a second. I think this tendency that a problem isn’t really concrete until you can get a number out of it, the idea of pinning down things quantitatively, was certainly to some extent Sommerfeld’s influence. I haven’t picked that up nearly as well as, for example, Bethe, because I have often left investigations at the qualitative stage where I could see the principles, and then other people have got to the point of getting numbers out of it. So I’m not really an extreme exponent of that line. I have one other characteristic which probably is a personal one and not inherited from anybody, and that is that I get interested in any problem providing I can see it might be soluble and provided it has some connection with something. But I’m not a purist, for example, in thinking that only a problem which is directly related to the front line of physics is worth thinking about. So I’ve always been very interested in doing consulting work in helping people to solve whatever problems came along and trying to formulate their problems the right way. And that, of course, came in handy later during the war when one got involved with things which weren’t really theoretical physics in the proper sense or in the modern sense. So that is I think a personal characteristic, which also Bethe has very largely, but which I think doesn’t come from any teacher.
Well, this would carry over into teaching. I might ask at this point whether you get that kind of stimulation in teaching advanced students, the stimulation of being turned onto new problems or tackling new things just for the sake of the problem itself.
If by teaching you mean primarily lecturing, then I think that’s not a very strong connection except that I think anybody who teaches lecture courses gets interested in finding simple arguments — not only knowing that something is true but how do you explain it, how do you give a simple and intelligible argument to show it’s right and so on. And that I’ve enjoyed a lot, and that certainly has been stimulated, partly by teaching, partly by writing reviews or textbooks or something. But if in teaching you include also working with graduate students, with people doing research, yes. In fact, I find it extremely useful to have research students even when they’re not terribly good — in other words, when they don’t themselves contribute an awful lot to the progress, which sometimes happens. Because it’s simply that you have a lot of ideas that you might think about but then when you have started discussing a specific problem with a student, well, then that makes it concrete that you’ve taken on an obligation to make some progress with that problem. You see the student at certain times and keep in touch with him, and then when he comes in, well, then you are forced, as it were, to think about that problem; and, in fact, if it’s difficult between seeing him on one occasion and then the next one, if the thing got stuck, you start worrying the way it’s going and to get ready for the next meeting first of all you try to make some progress. In other words, it helps to keep your thoughts organized, and often if the student is reasonably good, at least meeting him next time brings the subject back and reminds you of where you were and what was the problem that you might otherwise have forgotten. There is a level of student where he has forgotten next time what you were talking about and you would have to explain it to him from scratch. Then it’s not so useful.
Then it’s an exercise.
Of course there are many students where it’s quite different, who have initiative and who, after you’ve discussed some problem and left it in a rather vague state, come back next time and have found some answers and made some progress. That’s the kind of student you like to have, of course.
I get the feeling from your past history that the first time you really came into close relationships with students was at Birmingham.
On any large scale, yes. There were a few students, yes. At Manchester I think there weren’t any working with me. At Cambridge there were a few, including Fred Hoyle.
Was he in the Mond Laboratory?
Well, the thing wasn’t compartmentalized to that extent, and I started as a supervisor when he started on research, but after a short time — maybe a year, maybe less — I moved to Birmingham, and I continued in collaboration for a while. He used to come and see me in Birmingham, and then it sort of petered out. Then I had an Indian student. That was already at Birmingham.
Well, I think I’ll get to that.
You’re quite right. On any scale, that really started at Birmingham and in fact on a reasonable scale only after the war. There were very few before that.
Other than the position you had with Pauli, which was an academic appointment, but, as you explained earlier, one with a terminal date (In other words, the fellowship took you away from that permanently), other than that it seems to me you didn’t have an academic position or a faculty position until Birmingham. The others were research assistants or resident fellows or something of that type.
Yes. Cambridge was a bit of a mixture because having this Royal Society appointment in the Mond Laboratory, I was also given a nominal titular position in the University in order to make it possible for me to give lecture courses, do a little bit of lecturing, for which you also got paid separately. To that extent, I was therefore a part time member of the University staff, if you like.
I didn’t know that you did lecture. How much of your time did that take and to whom did you lecture?
It didn’t take much time. I think I gave one course on theoretical nuclear physics, such as it was then. And that was fairly informal and I think attended mainly by graduate students — probably some undergraduates, but it wasn’t part of their regular curriculum. Perhaps I gave a course on radiation theory. I have a vague recollection that I did that, but I can’t remember. Anyway these were courses of maybe two hours a week per term, 16 lectures — something like that — not, in other words, a very major occupation.
But in the theoretical nuclear physics part of it, how large a group attended?
I don’t remember exactly. I would guess 20, 30 — something like that.
What I’m trying to get at here is how much theoretical nuclear physics was of interest at the time.
This was meant to be intelligible to the experimentalists, and there were of course considerable numbers of graduate students in Cambridge doing experimental nuclear physics, and others perhaps not working on nuclear physics but still interested in the subject.
How about the Cavendish staff people — Chadwick, Cockcroft, and I guess Ellis was there and Feather and Oliphant had perhaps left by then …
He was still there. How many of them — the regulars — would come in, or would any of them come into these courses to sit in?
Again, I can’t remember exactly. I think not many and certainly not regularly, but the arrangements were very informal, and therefore if any of them turned up, I wouldn’t have been particularly surprised or interested in the fact. I might have forgotten that. But I think in general they did not. They were too busy.
Did you have any close working relationships with the experimental nuclear physics work?
No. One was on friendly terms, and one talked in a general way from time to time. But I wouldn’t call that collaboration. They did occasionally come up with questions, and often these were questions you could answer off the cuff or make a guess about without doing any particular work. And occasionally it may have led to a bit of a calculation or something or other. Nothing really regular.
Did you get any impression at that time about the atmosphere of the Cavendish Laboratory itself and Rutherford’s role (this is towards the year of his death) of whether it was changing, how his personality and his role affected the feeling within the lab and the overall outlook of the people there?
I don’t think I can say that I noticed any particular change. Of course I had occasion to observe it only for a period of two years, because during my previous stay in Cambridge I didn’t really get much contact with the experimental side. So that’s too short a period to notice any change. And looking back now, including outside knowledge, I don’t think there was any very major change except that it was of course a time where in Cambridge physics was beginning to cost money, and one was going in for bigger equipment. Although Rutherford’s tendency had been the string and sealing wax approach, I don’t think he ever opposed the idea of using big equipment when it became sensible to do so. It was just not the way his work had developed. So to that extent there was a change in the way work was being done. But I don’t think that meant any particular change in personal relations. I mean Rutherford’s influence and his liveliness in arguing about any problem under discussion was still certainly very much there.
Was there talk at that time, since Rutherford was near the retirement age anyway, or any speculation about the succession of the leadership of the laboratory and what direction it might take after he was gone — either after he retired or after his death?
Nobody thought of Rutherford dying. That just didn’t enter anybody’s mind — in an abstract way, of course. He was so healthy and vigorous that that didn’t occur to anybody. If there were any speculations — there must have been — about a successor after his retirement, I don’t think I got involved or heard about it. I believe it would have been the general assumption that Chadwick would succeed him.
How about the spirit, the morale? Was there anything unusual about it? I guess the best way to do it would be to compare it to other institutions — Manchester, for example, although the size of the groups was somewhat different?
Just absolutely no comparison. And that was not the difference between Manchester and Cambridge, but the difference between Bragg and Rutherford. Bragg is a very charming person, but definitely not the leader of a vigorous physics department. In fact, looking back, I didn’t appreciate that at the time. I came to Manchester some years after Bragg had succeeded Rutherford there, and it must have been a very lively, productive place in Rutherford’s time, and it was pretty much a desert when I was there — as a department. There were some active individuals there, but not very much coherence, not very much leadership in physics. I mean presumably the administration of the department was run all right, although this was a bad time. One depressing thing was that there was a sort of agreement between the universities — at least among what we call the “red brick universities” in England (outside Cambridge and Oxford who are always laws unto themselves) — that young staff would first be appointed to the assistant lecturer grade, which is roughly like an instructor here, for a maximum of three years; and that they would then not in general be promoted but have to wait for a vacancy in a separate establishment, which meant in general that they could not continue in the place they were after the three years. Now, if that system is operated by all universities, there may be something to be said for it, to get people moving around. In fact, it wasn’t. There were only a very few universities, including Manchester and I think London, who were operating in that system; whereas most other universities were in fact promoting their own young people. This meant, particularly in that time of depression, people who had completed their first three years as assistant lecturers in Manchester had to get out and mostly — very often, we’ll say — didn’t have a chance in academic life but went off into industry or something. And this sort of feeling of insecurity was one factor in the atmosphere of the department. It was not only that. I think Bragg at that time was not terribly interested in physics. He was very good at giving popular lectures. I remember during the time he was preparing one of the big series of Christmas lectures for children at the Royal Institution in London, of which he later became director, and he went to enormous trouble getting up attractive demonstrations and made an excellent job of that. But that’s something other than being involved in research that’s growing. I mean that comparison shows up Cambridge as a completely different place at the time.
As you describe it, it had very little to do with the number of students involved, but rather with the general tone established by the person who’s taking the leadership role or not taking it.
Now, at the Cavendish, another difference would be the number of visitors who would come through. It would be the focal point
That’s right. It was one of the centers of modern physics, and certainly the number of people who dropped in or came for a visit or came for a few weeks in the summer to refresh their knowledge and so on were very large, and you had some of that of course in Manchester but on a completely different scale.
By this time many people had emigrated from Germany. There are two questions that that brings me to. One is on Cambridge as a place for refugees either for a visit or temporary stay during this period. And the second is on contact with physicists who were still in Germany still doing physics, whether they visited during the period or whether you had normal contacts with some of them who were in Germany.
I think we would not have been reluctant to have contact with German physicists. We generally assumed — and I think in most cases rightly — that they were not very happy about what was going on in Germany. There were a few exceptions. There were some physicists whom we knew to be Nazi supporters, but they didn’t come for visits, and they were clearly exceptions in those days. But there were not very many who came. I remember a visit probably in Cambridge (it was somewhere in England; it might have been during the Manchester period — I don’t recall) of Becker, Richard Becker, whom I knew reasonably well and whom I regarded as a very sensible and reasonable person. We were a little surprised about his comments on political affairs. I think this happened in more than one case, but I remember it in connection with Becker particularly where we said, “It’s amazing how far people who are obviously not supporters of the German regime are nevertheless influenced by the ideas and the stories floating around and don’t realize it.” Of course, if you are in such an atmosphere and, for example, all the newspapers carry garbled news…
You were talking about Becker and the effect, you thought, of the newspapers.
The point is this is a general phenomenon. If your main supply of news is in this sort of garbled way, which you yourself know isn’t right or which you don’t trust, what can you do? You either just ignore it and say, “I don’t know what’s going on in the world,” or else you make your own comments and try to pick out which of the news you believe and which you don’t. But that means you are still completely influenced by what you read. I learned later about Becker in particular that that was a slightly too optimistic view, that in fact his thinking had been more influenced by the system than we thought. That’s the only case that I remember. There must have been other German physicists visiting. There was, for example, Harteck, who was a German experimentalist who was in Cambridge at that time and who eventually returned to Germany. I don’t think he was a Nazi particularly but also not particularly opposed to the regime.
How about Heisenberg? Did you have any correspondence with him or any contact? You certainly were working on similar subjects.
I probably met him at meetings in Copenhagen where I went frequently and he did. I don’t remember any correspondence, although there might have been. I think my attitude toward Heisenberg in those days was beginning to be somewhat reserved. I had enormous respect for him as a physicist, of course, and as a teacher. I learned a lot from him. We didn’t believe that he was actually a Nazi, but he had, I think, more willingness to accept the regime than any of us thought right. For example, when it became clearly impossible for him to have anybody on his staff who was either Jewish or politically undesirable or something like that, he probably didn’t like this, but he didn’t make any particular protest, didn’t fight about this; and he certainly didn’t think of leaving the country at that point. Obviously, being Heisenberg, he could have had a job anywhere more or less for the asking. He started a fight only when the system interfered with physics — I mean when people like Stark and Lenard acquired more influence, and one was in danger of a party edict saying that quantum mechanics was wrong and so on.
In other words, specific attacks on theoretical physics per se as being no good and “a Jewish science.” I think that is essentially the phrase they used.
So if nothing else it was in self-defense.
No, I think that’s not fair. Of course, it was self-defense in the sense that if that attack had succeeded, then he would have also been in trouble with his work. But I don’t think for a moment that was the motivation. That was something that was quite obviously wrong to him, and he felt it necessary to fight that regardless of whether it would have affected him personally or not.
How about the refugees in Cambridge at the time? Was there any noticeable increase, or was it just a question of people showing up from time to time and then moving on?
There were a few people who stayed for considerable periods, mostly experimentalists, I think — generally there wasn’t much room for theoreticians or many jobs. The position of experimentalists had some quite interesting features, because in those days one of the fashionable subjects in physics was spectroscopy, because it was connected with quantum mechanics, and it had been very instrumental in leading to quantum theory; and there were still a lot of things to be cleared up. So many of the German laboratories were doing work in spectroscopy. And there you have a situation in fact a little reminiscent of nuclear physics or high-energy physics today, because although it was experimental physics, the young experimentalist didn’t build any apparatus because the big spectroscopes were there and the discharge tubes, well, he might put a new source in or something, but generally the equipment was there and he put some substance in, took his readings, and then started classifying the lines and measuring them out carefully, classifying them and so on. And it came as a great surprise to many of these young people when they moved to England or to Cambridge, that although there were of course laboratories interested in spectroscopy and fitted in with their interests, but the approach was quite different; there wasn’t apparatus provided; they were expected to make it themselves or get it made and take their coat off and roll up their sleeves and work with their hands. Of course, the better ones took very well to that. But there were many who felt very aggrieved by this primitive way of doing things.
What would they do — switch fields, or was it possible?
Well, it varied. Some would just sit there and complain, as a result of which this was clearly visible to people, and so they were not all that popular in the laboratory in which they were, which meant probably their jobs wouldn’t be extended. It wasn’t a question of throwing them out, but they had some temporary grant for a while, and if they had been very popular, someone would have gone to great effort to find some additional funds to keep them; but as it was they left. I know one man (I don’t think I want to mention the name) who in that sort of way got fed up with Cambridge, and also I think the job came to an end. I may be wrong, but it may be that he was supported by funds given by ICI, the big chemical firm, for a particular laboratory; and then they, on rather short notice, changed their minds and cut off the support.
Yes, there were 14 people who had been supported, and then they withdrew this support. These people were left high and dry.
That’s right. Now, whether the man I have in mind was one of those, I’m not a hundred percent sure. Well, he went then to the United States to join a big firm in their research department, and I was later told that that particular firm prided themselves that they never sacked anybody, but that in his case they had come very close to breaking their rule. Actually, this particular man is back in England now, and I think now has matured and is all right.
What order of magnitude are you talking about when you talk about people who came and were successfully assimilated or not? It’s not clear to me. I don’t know whether you can pin it down, whether you’re talking about six or 16 or 60.
Well, in Cambridge the numbers were very small. But I’d have to search my memory and write down lists before I could make any estimate of a likely number, but certainly not more than six.
That’s what I was getting at. I do know the total number, and I have lists and names and so forth from the Academic Assistance Council’s files, but I was just curious about the clusterings in specific institutions. This gives me something of the atmosphere. Also, at this time, what about relations with Soviet physicists? Was there much visiting in either direction? We’re talking about the period from ‘35 to ‘37.
Well, things were changing critically about that time. Generally the position had been that for Soviet physicists to get out was rather difficult and rather exceptional. Some people, as you know, like Landau and Gamow and so on, were sent out on official missions. But, for example, Ivanenko at that period never got out, and I think it had some effect on his attitude. Many other people never made it. And certainly even those who had traveled abroad couldn’t just decide they were going to go off. This was a very chancy business whether you got permission or not. So there wasn’t very much visiting that way, although there were occasional conferences or visits from people — not very much. On the other hand, it was rather easy for people from the west to go to the Soviet Union. And in fact, any reputable physicist who was willing to pay his expenses as far as the first stopping place in the Soviet Union, or, if he was really keen to save money until the border, they could arrange for some rubles to be sent him somehow to the border, could go more or less for the asking. It was quite easy to get an invitation and to spend a few weeks there and have all your expenses paid and be welcome in some laboratory. Now, not everybody did that, of course, but quite a few people, particularly theoreticians, made use of that. People like Weisskopf and Placzek and so on did a lot of traveling in that sort of way. This I think was completely cut off in 1937. There was the nuclear physics conference in Moscow in ‘37, which was the last occasion where people were invited officially. I think the plans for that had been made some time ahead; otherwise it wouldn’t have come off. After that it was practically impossible to go.
Did you go to that conference in ‘37?
Let’s talk about that. Who extended the invitation? Who was there from outside and then also the Soviet physicists you remember, the tone of the meeting, the significance of it?
Well, we have already mentioned it a bit yesterday. This was the conference where I said that it was a slightly half—hearted conference because people were too preoccupied with what was going on. I also don’t remember much in detail about the conference. It was a time when work on cyclotrons in Russia had started. People were reporting on the progress. I don’t think they had a working cyclotron yet.
No, I don’t think they could have.
No. But they were talking about their plans. They had, of course, quite competent work going on in radioactivity, and they had excellent theorists. I remember talking with Landau, not in the conference but on that occasion. I have a stronger memory of his worries about how the situation was deteriorating and how unhappy everybody was rather than about physics. We must have talked about physics, but the other left a stronger impression.
I have some of the correspondence — Bohr’s correspondence — in relation to the meeting, so some of the details are available about the planning and organization of it. One thing about the general atmosphere and conditions while we’re on that — then I’d like to get back to some specific work in the last year at Cambridge. There are some papers I’d like to talk about. Blackett and Thomson and others who had a discussion group in London: did you keep up in that group during the Cambridge period?
This was a regular thing.
I can’t remember how often this met, but I would guess three times a year, something like that or a bit more. And I wouldn’t, of course, necessarily go to every meeting, but I generally attended.
We discussed that yesterday. I just wanted to know if you kept up. Also about this time, in the realm of politics and atmosphere, there was a movement developing in England, the Science and Society movement, where a number of people in physics and biology and so forth had specific, outspoken views on the relationship of science to society and what the role of the government should be in science. Some of it was influenced by their view of what the Soviet planning for science was, but it was an indigenous kind of a movement of Blackett and others who were making public statements on that. Were there any ripples of that or discussion of it or were you aware of it?
I wasn’t aware of it really. I certainly wasn’t aware of Blackett’s interest in this. I don’t think it had ever come up in our conversations. I was aware of the attitudes of people like Bernal and Crowther and did not have much sympathy for their views.
Did that come through their published statements?
Yes, and occasional lectures. I may have heard Bernal talk about this. He was in Cambridge at the time. I don’t think I talked much with him privately, but he may have given some general talks on this, and I saw some of his books and articles and so on. This to me seemed just fairly regular party-line stuff and not at all convincing.
Was there much discussion, say, in Cambridge of these kinds of things? Was it ever really an issue?
I don’t think so, not so far as I was aware.
I just wanted to get your impression as far as its being characteristic of the period.
There were a number of people with rather Marxist views. And of course you were willing to listen and to see that occasionally they had interesting points. But in general you treated them like some strange religious sect whose views you were not particularly interested in.
Someone is doing a doctoral dissertation now on that period and an analysis of the statements and the reactions to it and the motivations. It’s quite interesting. I’ve seen glimpses of it. Let’s get now to the few specific papers in the remaining year, 1936, in Cambridge. One I’d like to talk about very briefly is your paper on the interpretation of Shankland’s experiment, which seemed to contradict Compton’s scattering law in the high energy region and the question of validity of energy and momentum conservation. The interesting thing here is the degree of seriousness with which the experiments were regarded, and also whether you were aware of what Bohr thought on this question.
Well, the experiment, of course, caused quite a stir, like you would expect of any experiment done by a reputable physicist who finds something in disagreement with the current views. I can’t say how widespread this was. I think there must have been people who from the beginning simply refused to believe the experiment. But at least evidently the people I talked to must have been prepared to consider the possibility that the experiment was right and then, if it was right, it appeared to mean a complete breakdown of energy conservation. And then I got interested in the logic of that situation and saw that there was a sort of halfway house, that you didn’t have to abandon everything, but there was some slightly intermediate position, and I wrote this short note just to make that point. In a way I might just as well have left that unsaid and waited first of all to see whether there was support for the experiment or not.
But this was another example of the kind of thing that provokes you into action?
Maybe — I mean simply in hearing this general discussion I saw that it was logically not quite complete because it left out one possibility, and so I was interested in pointing that out.
You didn’t do much more with that.
No, because fairly soon after that other people repeated the experiment with different answers, and it became fairly clear that the experiment was just wrong, to everybody’s relief.
Later in this period there was another venture into another field, astrophysics, where you were talking about the relativistic gases.
That was part of a controversy with Eddington.
I’d like to know the nature of the controversy.
I can give you a bit of background there, although there’s nothing specially to do with me. This was really quite incredible, because Eddington, who was a great man, particularly for having been the first to open up the study of astrophysics, of the interior of stars, by modern physical ideas, then went somewhat off the rails. He had realized that in some stars the densities were so high that the electrons would form a degenerate Fermi gas (it was shortly after Fermi statistics had become understood), and that this would modify the equation of state of the gas in the stars, because the pressure due to the degeneracy was very much higher than just the thermal pressure. So he came up with a new formula for the equation of state, which led to a new law for the conditions of stars. Then Chandrasekhar, who was a pupil of his, worked with him, pointed out that in some of these stars, the densities were in fact so high that the electrons would be relativistic, so that would have to change the equation of state again putting in the relativistic expression for the kinetic energy of the electrons for the non-relativistic one. That made a difference in the relation between energy and pressure. Eddington didn’t like that, because it would mean his abandoning, at least over a certain region, the formula he cherished. There was no conflict with experiment. I mean in fact the Chandrasekhar formula fitted what were known facts, at least as well and probably better than Eddington’s. So his argument wasn’t that it was empirically wrong. But he just liked the look of that formula, and therefore didn’t like the alteration; and perhaps he didn’t like a young man like Chandrasekhar coming to contradict him. And he violently opposed this, to the extent of preventing Chandrasekhar getting a job in England. I mean in those days things were so centralized that you couldn’t get an appointment in astrophysics without getting Eddington’s recommendation, and he didn’t get that. So that’s how he left and went to the United States. He might have gone anyway in the course of time.
Did he get to Copenhagen first?
Not from there. I mean he might have gone there — that I don’t remember.
Was he more of a student of Fowler’s?
That I don’t recall. He may have started with Fowler. Of course, this was also connected with statistical mechanics. But when he turned to astrophysics, he was very much in Eddington’s field. Then in the course of that business Eddington was induced to assert that you should not apply relativistic expressions to the interior of stars, even at very high kinetic energies. How anybody who had himself written a textbook on relativity could make such a statement is still very hard for me to understand. People then pointed out that after all the culmination of relativity in quantum mechanics, namely the Dirac equation, had been extremely successful in the hydrogen atom. And he said, yes, that was different. First of all he said that it depends on the symmetry; the hydrogen atom is a spherical object, and there certain symmetry conditions apply, and therefore it’s all right to use a relativistic equation; but in the base of statistical mechanics you always consider particles in a square box, and then the old Newtonian mechanics would be right. I would have thought that a star was, if anything, more like a spherical box than a square one. I mean this was irrational. And then also he created some confusion by saying that it wasn’t true that the … I mean normally the hydrogen atom in quantum mechanics is treated as if the proton was standing still and the electron is moving around it, which is approximately right. But, of course, in actual fact that proton is not standing still. It’s free or can be free. And then the uncertainty principle forces you either to allow it some spread in position or alternatively to have the energy undetermined and so on. And therefore he was trying to make out that since the relativistic two-body problem hadn’t been solved, that therefore we couldn’t really say anything about the hydrogen atom. That was a bit too much for us. And in fact I tried to write out some of the arguments to counter what Eddington was saying, and then discovered that Maurice Pryce and Dirac had also thought about it. Of course, they disagreed equally strongly with Eddington, and they had looked at different aspects of this. And so it came eventually that we wrote a joint paper putting our arguments together. This was purely controversy in the sense that we did not feel we were contributing any new knowledge in that paper, but we were simply trying to dispose of what we thought were irrelevant points made by Eddington.
There were two papers then. The one with Pryce and Dirac is very late 1941. That’s “On Lorentz Invariance in the Quantum Theory.” It’s a joint one with them. And then this one that we began to discuss was June of ‘36.
What was the title of that?
“Note on the Derivation of the Equation of State for a Degenerate Relativistic Gas.”
Right. Well, that’s all the same subject. I had forgotten that the thing was spread over so long an interval. I now remember the paper I wrote on my own was concerned with another objection of Eddington’s — namely, if you derive the equation of state in quantum statistical mechanics, you have to visualize some finite volume, because if you want to count up the states, taking the volume as an infinite the number of states becomes singular. However, to solve the equations for any specific volume with any specific boundary condition at the walls gets rather involved; and in fact would be difficult with the Dirac equation, because if you make your potential barrier sufficiently high to prevent particles from getting out, you get the so-called Klein paradox, particles going into negative energy states. So you would have to make provisions for eliminating that, and it gets terribly complicated. What is usually done is to replace the actual boundary condition by a fictitious cyclic boundary condition. You simply say, “This volume is part of a periodic structure.” Well, in one dimension this rule is very easily visualized by saying instead of a straight line, you work on the circumference of a circle. That shouldn’t make any difference. Now, doing that in three dimensions isn’t so easy, and so it becomes just a mathematical device. Everybody took it for granted that that was permissible provided you were not interested in details like surface effects when obviously you would have to deal with the surface realistically. But if you only want to have some measure of the volume and you’re not interested in the shape and so on, then it’s good enough. Now, this is one of the points queried by Eddington. And then again I thought I’d write down an argument which actually demonstrated that this was permissible.
Did you become involved in any personal way, in personal discussions, with Eddington on this?
I didn’t know Eddington very well. I met him on some occasions. I don’t think we ever communicated very well. I mean I don’t think we ever got anything across in these conversations. He was a strange man, typically a product of Cambridge, where a theoretician is primarily a mathematician from whom the physical reality is somewhat removed. I was very struck by one conversation at which I was present, though I didn’t take part. Some visitor asked Eddington, “What is now the state of the general theory of relativity? Should one believe it? Is it all right?” And Eddington said, “Oh, yes. It’s in fine shape.” So the man said, “Well, I’ve heard, however, that there’s some doubt about the accuracy of the experiments about the Mercury perihelion procession, that the effect is not really firmly established, and also that the accuracy of the observations on the deflection of light near the sun aren’t so reliable.” “Yes,” Eddington said, “I quite agree. These experiments are very inaccurate. I wouldn’t trust them for a minute.” So he said, “Then why are you so optimistic about the general theory of relativity?” “Oh,” said Eddington, “I would never accept the theory on the basis of those very poor experiments. That is not the reason. You just have to look at that theory to see that it’s right.”
That is probably the view that Einstein held.
Oh, no. Einstein would like it to be right, but he was critical enough to know that you couldn’t be sure until there was some evidence.
But I think his tendency would be to doubt any experiment that disproved it.
There never were any experiments which disproved it.
Maybe he was slightly optimistic about the interpretation of the experiments.
That’s a good point about Eddington. Chandrasekhar in a talk at the last meeting of the American Physical Society, the New York meeting in January and the beginning of February, gave an address — “Historical Notes,” he called it — and he reviewed two or three incidents in which he and Eddington were involved. First he gave what he knew about the background of the eclipse expedition; and the other was a very summarized version of this controversy. He didn’t go into the details, but he took some pride in saying that recent events had really demonstrated him to be right.
I’m sure that in a public address he wouldn’t have been so outspoken about Eddington’s behavior as I was just now.
Another paper in that same period toward the end of your stay was on heat conduction in liquid helium. It was an experimental paper really. That’s an interesting thing, that you were involved in experimental work. I want to know what part you did take actually, and what was your role with these two others, with Allen and Uddin?
Zaki Uddin. He was what we now call a Pakistani. Yes. Well, this arose very much, of course, out of my presence in the Mond Laboratory and my interest in what was being done there, and also my perhaps feeling of obligation or just general interest in trying to help the experimental program along. At that time the experiments about the thermal conductivity of helium gave what appeared to be very confusing results. They were not reproducible, and the co-efficient of thermal conductivity seemed to depend upon the temperature gradient you applied. It should be independent, but as you reduced the temperature gradient, the heat flow didn’t go down proportionally and so on. So here there was clearly something quite phenomenally wrong. And one question was: Since the conductivity was so enormous — the thermal conductivity — maybe the temperature gradients one was using weren’t small enough to get clear conditions. Maybe one could measure this with a very much smaller temperature gradient; with a small heat flow, one would get a better answer. And then I saw a trick, by which you could measure in fact very small temperature differences, and this might help to clear up this problem, and I suggested the experiment, which was picked up with enthusiasm by Allen. Allen, incidentally, was the other man. I mentioned that there were two research fellows appointed there at the Mond Laboratory after Kapitza left. He was the other one.
The trick was an experimental trick. It wasn’t theoretical at all.
Well, it had a theoretical basis. It would have been a very clever experiment except that there was a new effect which wasn’t known then and which directly related to our method. What happened was this: I decided that helium was a very good substance to use, not only in the flow experiments but also as a gas thermometer. The idea was to take a bulb with a small capillary leading out of it and immerse the whole thing in liquid helium. Now, when the thing is left to itself, the level which you can see inside the bulb would be at the same height as the level in the liquid outside. Now, you have a heating wire inside the bulb into which you feed a very small amount of current. That causes a heat flow. And therefore the substance inside the bulb gets slightly warmer than the rest. So the gas in the rest of the bulb expands, and the level drops. Helium would be about the ideal substance for this, because it has a very low density, and therefore a small pressure change means a large change in the level. It also has a very high thermal conductivity, and therefore you needn’t worry about equalizing the temperature within the bulb and so on. It seemed just perfect.
Did you take any role in actually building the apparatus?
Not building the apparatus, but I helped in actually taking the readings. The experiment had to be done, of course, when a supply of helium was available, which wasn’t all the time in those days, and then of course you had to keep at it while the helium lasted. And so one worked in shifts taking readings with the thing, and I took my turn in that. The results came out even more unintelligible than the previous experiments. The explanation was (what we didn’t know then) in the so-called fountain effect, which means that a temperature gradient in a capillary of liquid helium (in the so-called liquid helium 2 at low temperatures) in fact causes also a pressure difference, which of course completely wrecks the thing as a tool for measuring temperature differences by this method. It was only after I left Cambridge that Allen hit on the idea to see what would happen if you cut off the top of the bulb — in fact, had it open. On the principle we had assumed the level should always stay the same. In fact it didn’t, but you got actually a stream of liquid coming up the capillary, and that’s what is called the fountain effect. I think that was first discovered by Allen or at least simultaneously so that the wrong idea about getting the method of measuring something at least indirectly led to that discovery.
Which in turn stimulated a great deal of theoretical input?
This work on liquid helium was submitted in June of 1937, but your bibliography, if it’s complete, shows that there are no other papers for the preceding year — from June ‘36 to June ‘37. Do you recall what characterized that period?
Well, it was not a very successful period evidently, though it may be that I was working on some ideas then which bore fruit later. But I was trying various ideas again arising from the liquid helium work and low temperature work in the laboratory. Also, I think that is the beginning of the period where I spent more time discussing problems with other people, and when they in fact eventually published the papers resulting from that. For example, I think it was the period when I became interested again in diamagnetism of metals, on which I’d written papers earlier. I had pointed out qualitatively how to understand the so-called DeHaas-van Alphen effect, the rather striking oscillations in the diamagnetic susceptibility first found in bismuth. Now, at that time Shoenberg in the Mond Laboratory was doing more accurate measurements of this effect, and we thought it would be nice to try to make the theory more quantitative, and I thought I saw how to do that by making more specific assumptions about the nature of the motion of electrons in bismuth. Then Blackman, now at Imperial College in London, was in Cambridge — I think as a post-doctoral visitor, not as a graduate student. So we got to talking, and at my suggestion he did some work computing what magnetization curves one might expect from that model and fitting parameters to the experiments and so on. Actually later on a much more elegant way of doing that…
You were just mentioning Landau.
Yes, I was saying that at my suggestion Blackman did a calculation of the anomalous diamagnetism of bismuth, which was an extensive computing job. Later on Landau showed a very clever analytic method of doing the same thing. But anyway in the present context the point I was making was that I spent a considerable time discussing this project with Blackman, but I didn’t publish anything about that. It’s in Blackman’s paper. There are quite a few other things that happened the same way starting at that Cambridge period and going on from then. But also I was trying things on the thermodynamics of liquid helium, which were not very successful, and I wasted a lot of time. I remember one incident when I was for my own interest computing the entropy of liquid helium, and it didn’t seem to check with the third law, so that some further transformation could be expected at lower temperatures. I was getting rather interested in that and suggesting experiments of how one might check up on that and so on. And then later it was discovered that I’d simply made an arithmetic error; and if that was corrected, the thing checked perfectly well — there was no mystery.
This was not something you published.
No. I think it was also about that time that wasted a considerable period on a nuclear physics problem. This was connected with the fact that we had in our first approach always taken the range of the nuclear forces to be very small and maybe zero. And I think Wigner had the view at one time that perhaps these were forces with actual zero range. Then there was a very clever paper published by L. H. Thomas, who proved that if you assumed forces between neutrons and protons (actually at that time it was fashionable to expect forces only between unlike particles, only between neutrons and protons but not between two protons or two neutrons), that if you arranged these forces so that they would give the right binding energy for the deuteron and then computed the energy of the three body system, the triton; that as you made the range of the forces go to zero, the binding energy of the triton would go to infinity. This was a very important result if one was thinking about the forces. This again was a typical case where I thought this was maybe a bit clumsy, the method used by Thomas. Well, clearly one uses a variation method, in other words, guess a function, which would give an upper bound to the energy and therefore show it can be made as low as you like. I thought this thing was unnecessarily complicated; and partly to understand the situation and partly to simplify it, I would see whether I couldn’t invent a trial function which would do the same job more simply. I wasted several months on that, only to convince myself that it was much harder; or, in other words, that Thomas was much cleverer than I thought because all the best trial functions I was able to invent gave an inadequate answer. They didn’t give low enough energy, and were therefore useless. So that was a complete waste of time and probably not very wise because these long calculations have to be done carefully and accurately. If you make mistakes, everything is lost, and I wasn’t terribly good at that. So I was a little bit depressed at the end of that time.
Did you have to correspond with Thomas at all during this period?
I think he in fact was visiting Cambridge. Now, whether it was the summer of ‘36 or ‘37, I don’t remember. But when I was in Cambridge he was visiting. We talked about it. I told him I was very interested in his paper and his results. I may have mentioned that I would hope to get a simpler way of doing it. But he sort of said “Good luck to you. It would be nice to get it done in a simpler way.” But that was all. There were some mathematical tricks I was using where I talked with Goldstein, who was an expert in hydrodynamics and knew the kind of mathematical tricks that might have been helpful for that; but I didn’t get anything from him either that was powerful to solve my problem.
This work that you just discussed was apparently the only work in nuclear physics during your Cambridge period, which is rather interesting since there was a good deal of nuclear physics going on. I think I’m correct in that.
The only thing published, yes. I mean I did think about problems, but again…
Well, obviously, because you mentioned this one.
And then there were other things I was trying which didn’t get anywhere.
This quantum theory of atomic nuclei, but that came from Manchester really. That was the last major published paper. There were none published in the Cambridge period. But in teaching a course in theoretical nuclear physics you were aware of it.
Obviously thinking about it. Also Fred Hoyle was starting as a graduate student at the time, and at my suggestion started on the theory of beta decay, which was then developing in detail. But that must have been just in the summer before I was leaving Cambridge. It started then.
We will get to it later: you published a paper with Hoyle and Bethe on that subject in 1939. We’ll talk about that. How did the appointment to Birmingham come about?
That was simply that Oliphant had been appointed to the experimental physics chair in Birmingham. His appointment must have been early in ‘36, but he didn’t take up the chair, didn’t move to Birmingham until October ‘37. During that period when he was professor elect or may have already had the appointment but was on leave of absence to finish his work in Cambridge, he persuaded the University of Birmingham that they needed a theoretician also; and before that job was created, I think the position was that he wanted to demonstrate to the University that there were suitable candidates around and therefore talked to me, asking if I would be interested in taking such an appointment if it were created, and actually invited me to visit Birmingham with him and talk to the people there. This was not actually yet to decide whether I would get the job, but simply to show that there was at least one person interested in the job who might be suitable if they created it. And I was very interested, because I still didn’t have a permanent position, and it was a very considerable increase in seniority — I mean expressed, for example, by an increase of my salary as between Cambridge and Birmingham by a factor of two and a half, and the possibility of attracting some students and so on. I mean there wasn’t any doubt in my mind that if I could have that job, I would take it. And then later it was advertised in the ordinary, normal English way, and I applied for it. One might have thought that with these advance conversations, it would be more or less a matter of course. But it wasn’t because there was quite a strong field. On the short list there were three interviews on one day, and the other two people interviewed besides myself were Massy and H. Jones, who was then at Bristol and is now a professor in Imperial College, London. However, I got the job.
Had they developed plans for what they wanted to do? When you talked with them, did they have any program for the development of physics there?
Oliphant, of course, wanted to develop nuclear physics, and that meant, of course, getting some experimental facilities. Now, he eventually got some money from Nuffield, first to put a laboratory, a small, very modest building, and to build a sort of home-made cyclotron inside it. Now, whether he was already assured of that support when he accepted the job in Birmingham or when I went there, or whether that came later, I’m not sure. I think there must have been something in the wind. I think it may be that he wouldn’t have gone to Birmingham without some expectation that this would be forthcoming.
Did Birmingham have a reputation that you were aware of — good or bad?
It had a reasonable reputation as one of the provincial universities, which in those days of course were not trying to compete with places like Cambridge. It was somewhat younger than Manchester. I think those two; Manchester and Bristol probably would have been regarded as stronger than Birmingham, though Birmingham had a tradition of appointing good men. Poynting, who was a famous man in electromagnetic theory amongst other things, had been their first physics professor. Oliver Lodge, who I think is much more famous in England than elsewhere, was their first principal, and chancellor.
He had already died, hadn’t he?
He had died. I’m talking about the reputation of Birmingham. In general there were already good people in other departments. They had a very strong engineering tradition. They had quite a good medical school. They had good chemistry. The chemist who was there when we came was Norman Haworth, who later got a Nobel Prize for synthesizing Vitamin C. There was a very serious chemistry school. Now, in physics they were not very strong at the time. Oliphant’s predecessor was a competent but not terribly active man. There were some good people there. There was one man who had done experimental work related to the kinetic theory of gases and one of the few people who had studied the thermal diffusion effect, which later became of interest following Chapman’s theory that such an effect should exist.
Who was this?
A man called Ibbs. But it was not a strong physics department. Mathematics had a famous man — G. N. Watson, who is particularly known for his book on Bessel functions and his book with Whittaker on modern analysis, which is still one of the classics. In fact, I was appointed then as professor of applied mathematics following the Cambridge tradition — as joint head of the mathematics department with Watson. Watson was a very well-known and famous mathematician, but he did not have much of a department. He was really an ivory tower man. He came to the University on the days when he lectured, and when he finished lecturing, essentially he went home and worked at home on his own — except when there were administrative problems. He was very helpful in things like making timetables for the University and records on conflicting requirements of lecture rooms — all that sort of thing. That he enjoyed. But he didn’t really create any sort of research atmosphere in the department.
How long did your title remain the same? It was changed to include physics later, wasn’t it?
It was changed to mathematical physics in ‘45 or ‘46.
After the war.
It was changed to mathematical physics, but it wasn’t mathematics and physics.
No, mathematical physics. It was made a separate department. I found that running both pure mathematics and our work together in one department was a very difficult way of operating and having in courtesy to get Watson’s views on any appointments or policy change or something, in which he wasn’t really interested, was quite an inefficient arrangement. And so I asked for a change, and that was immediately done.
During the pre-war period I imagine you worked more closely with Oliphant than with Watson.
Well, in both pre-war and post-war our set-up was that as regards research, our relations were closest with physics. As regards undergraduate teaching — and, of course, in a department like that undergraduate teaching becomes one of the major responsibilities of the staff — we were much more closely related to mathematics. In fact the later arrangement of the separate department sort of intermediate between mathematics and physics was quite convenient for maintaining both these contacts and not being too tied to one side or the other.
Since this was the first time when you had a heavy teaching load, when the teaching responsibility was one of your major ones, if not the major one, how did it affect your work?
Well, the teaching wasn’t really heavy — at least not at first — not what I would regard as heavy. I remember the first year I was there; I obviously discussed with Watson what teaching I should do to pull my weight. He said, “Well, I’d much rather you didn’t start doing too much teaching but spent your time looking at our general program and making suggestions about how the applied mathematics teaching should be modernized,” because it had all been done by pure mathematicians and so on. But he said, “There’s only one course that I’d appreciate if you’d take over, and that’s a course on hydrodynamics.” Actually it was called mechanics and hydrodynamics, but the main content was hydrodynamics, which was three hours a week for the year. He said, “If you wouldn’t mind taking that on, that would help us very much.” So, of course, I said, “Sure, I’ll do that,” but being very conscious of the fact that I didn’t know any hydrodynamics whatever. I had had a course in my undergraduate years but hadn’t taken it very seriously and really didn’t know a thing. So the first few weeks at Birmingham I spent on sitting down with some textbooks and learning some hydrodynamics and working that into some reasonable course, which wasn’t at that time a terribly good one, but it improved over the years. But except for the initial period of getting started, it was not a heavy burden. The burden of teaching as such never was excessive except in sort of emergencies. But, of course, one had to think about the teaching in general. I was responsible not only for my own lectures but for the general scheme. Quite soon I decided that what we wanted was the possibility of a degree in an undergraduate course in theoretical physics or mathematical physics, as it’s called there, which I devised and which got started officially during the war.
While you were developing that program in that department, Oliphant was getting off the ground on his nuclear physics program, I assume.
Did you have much of a link with his work?
Well, we talked a lot, but, of course, they were preoccupied with building a cyclotron, and clearly until you have a machine you can’t do much physics with it. So they were interested in keeping up to date. We had a colloquium and discussed things, of course, but there wasn’t any very active nuclear physics there.
Did he have anyone especially advising him? I know there were contacts through Cockcroft at one time with Lawrence. Cockcroft had made two trips specifically to advise the people at the Cavendish. I imagine that Oliphant benefitted from this. But do you know that there were any visiting physicists there or even engineers?
Not engineers in general. Oliphant was a good friend of Lawrence’s, and he had visited him, and people from there may have visited Birmingham. But Oliphant had quite a strong engineering bent and designed the machine himself. This made it very cheap. It was all homemade, still really in the Cambridge tradition. It took a long time to complete, partly because it was interrupted during the war. When the war broke out and stopped the academic work, they had managed to assemble their magnet but nothing else. But whether he would have got it done any more quickly by ordering it from a firm, I don’t know, because there weren’t many firms with experience in that sort of field. It was all very much homemade. There were some incidents I remember. One was that the magnet was assembled in steel sections, which were quite heavy plates, of course; and they were in a sort of pit where they were put together; and a number of them were stacked up leaning against a wall of the trench in which they were. One of his senior collaborators and a lab boy were in that trench when one of those steel plates tipped over. It hadn’t been secured properly or somebody had made a mistake. So these people were quite badly injured — had legs broken. It was quite a mess. Fortunately both recovered — the boy quite quickly, and the other man took about a year to get back to normal. I remember one other thing. They had the coils for the magnet… It was air cooling, so they were strips of copper fairly loosely wound in a sort of spider’s web. And this assembly job was done in a hut outside the lab building, and then they were brought on a trolley into the building. It had been planned so that the trolley and these heavy coils would just go into the door of the building. But when they had the first coil ready and tried to bring it in, they discovered that the height of the door of the building was a few inches lower than the plans provided. The contractors had made a mistake. The result was that it just wouldn’t go in, and you had to design a different support where the coil could be tilted to an angle where it would go through the door. These were little tribulations of the homemade procedures.
I’m sure that even with more sophisticated procedures, the same thing could happen.
It could happen, yes.
Were there any formal weekly colloquia or the equivalent at the Journal Club at Birmingham, and what groups were involved?
There was a physics colloquium which would be attended by all the members of the physics department and myself and later any other people in my group. I mean there weren’t any to start with.
So you felt part of that. It wasn’t a question of there being departmental boundaries that would keep you away from that.
And their procedure was to talk about something from the literature, or report on it, or to invite people?
Yes, or somebody would talk on his own work, or some visitor would come in — like an ordinary colloquium.
This was about the time that Bethe’s review articles on nuclear physics were published.
That’s right, except I thought the first one came already while I was in Cambridge.
Yes. ‘36 and then the second and third in ‘37, however it worked out. What was the impact of the articles on you? Did it have any specific meaning other than another publication, or did you see it as something very different?
No, it was very important because it was the first time Bethe or anybody had made a really thorough job at putting all the facts together, thinking them through quantitatively, which other people may have done for themselves, but I didn’t have command of all the evidence. So it really gave me a solid foundation. It was the sort of publication which you never ignore once it’s out. Anything you do would be related to something in that article and you would take your information from that.
Did others react in the same way?
Yes, that was the general reaction.
Do you think it played any role in attracting people to the field. Were there specific instances of people who had been previously working in other fields using these as a way of finding out what was going on in nuclear physics and as a consequence perhaps switching into it? Do you know any instances of this?
No, I don’t. I mean what was obvious was that anybody who at that time decided he wanted to become interested in nuclear physics would sit down and read those articles. That was clear. But whether reading the articles would have induced anybody to pick up that field who had not otherwise felt attracted, I don’t know. I certainly can’t point to any case.
Did you see a tendency to use those articles as a starting point? In other words, the articles summarized a great deal of earlier literature. And it’s sometimes said of a review article that it compresses and compacts the past literature so that there’s no need any longer to refer to it. Was in fact this the case with these articles?
To some extent. But, of course, even in the size of those articles, they couldn’t include all the details. So sometimes you would use it simply as a source of reference and then went back to the original papers. But it certainly put all the existing literature in proportion, in perspective.
Did you use them at any time for teaching, similar to a text?
Well, I wasn’t doing any teaching of nuclear physics then. I don’t think I had a course on nuclear physics during that pre-war period in Birmingham, because I had no students practically, and there were some graduate students in the experimental physics department but not many; and so it wasn’t worth giving them a course.
When you say there weren’t many students, you mean in general, not just ones specializing in nuclear physics?
Well, I had no graduate students. That’s exaggerated.
You had Kapur.
Kapur I think had come with me from Cambridge. I think he had come with me from Cambridge, moved with me. There was, I think, one Norwegian who also came over from Cambridge to work with me for a time because he didn’t complete his thesis or anything. Then Hoyle, as I mentioned, remained at Cambridge, but came for occasional visits, usually hitch-hiking, to see me in Birmingham. My plan at the time was to say: “Well, it’s a small department, and I can’t expect for some time much in the way of staff.” There were no staff positions besides my own. And this was a static time. You didn’t think of growth. You didn’t say, “Well, in a couple of years we’ll have more jobs.” That wasn’t the attitude. So I was aware of the difficulty of teaching or encouraging research when you’re really the only one experienced person in the place. But my thoughts were: “Well, I’ll pick out gradually a few students — undergraduates who might stay on—and gradually build up a group that way.” This I now realize was not practical. It wouldn’t grow. A group has to have some minimum size to be viable.
You mean to keep the person there in the first place.
Not only that. You see, to build up a group of people who can help you, you’ve got to educate them; and it’s very hard to educate them when you’re the only person from whom they get their ideas. So I was lucky in a way that I had a second start, because it all dried up during the war; and then when I came back, conditions were different, and I had the possibility of persuading the University to create more positions. Then we started up with a sizeable group, and then got going.
During this period, you had some letters exchanged with Bohr in February of ‘38. There may have been other things, but the ones I’m thinking of dealt with Bohr’s theory of nuclear photo effect.
Bohr sent you a reprint asking for your opinion, and then you replied making suggestions for improving the theory.
I wanted to know a little about that — also your response to the compound nucleus model. We talked about your knowledge of it earlier, but now that it was done, what was the response? And what was your relationship to Bohr in this period?
Well, I was then visiting Copenhagen fairly frequently. I can’t pin down the dates now, but there were several occasions when I went there. I had got very interested in the compound nucleus theory, which seemed to everybody around at the time a breakthrough in the understanding of some nuclear phenomena. I had been groping a little bit in that direction. I remember I almost got there. There was then a paper — I think by Van Vleck, I’m not sure, but by some person of understanding — which discussed the question, which was then interesting, of resonance in neutron capture or neutron scattering. It gave a general theory, general enough to be valid; and then it estimated how rapidly the phase shift would vary with energy because if you know that, you know how many resonances to expect. Now, he gave an estimate of that based on a one-body problem — in other words, assuming the neutron was just moving in the potential field due to the nucleus. And I remember saying, “But surely that’s completely wrong, because it’s really a many-body problem and a many-body problem must have many more resonances or many more stationary states; and therefore, if one did that properly, the spacing of levels would be quite different.” This was a first step towards recognizing what Bohr did. But I didn’t take it much further, at least not quickly enough; and then Bohr came out and the picture was clear. So I was ready to receive that, because I’d been beginning to think in that direction.
Do you remember when that was?
No, except that it was obviously shortly before Bohr’s ideas first came out. I suppose that must have been around ‘37.
Bohr was earlier.
Earlier. Yes, it must have been earlier. It must have been at Cambridge. Then I got interested, of course, in this compound nucleus model, trying to see a way of describing it formally. Then there were quoted in Bethe’s article and also in papers by Bethe and Placzek equations to describe resonances, extending the formula of Breit and Wigner on fundamental resonance. And I realized this was not satisfactory, because the derivation always assumed weak coupling; assumed, in other words, that the interaction which would lead to the emission of a particle from the compound nucleus was some kind of weak force, but in nuclear physics there is no weak force responsible for that, so the derivation wasn’t right. The answers looked as if they must be about right, but there must be a better way of getting them. There my expectation that there should be a better way was supported by a conversation I had had some years before with Landau, who sort of said it was quite trivial — one could just see. I mean Landau was very good at asserting that things were obvious which were not necessarily obvious to other people. And I’d often pondered what he meant. I was talking about the problem that I felt should have a more satisfactory solution — to describe a many-body system with resonances in which you didn’t have any weak forces responsible for the emission of particles. Then about this time, ‘37-‘38, I began to see how one might try to do this, and I got Kapur, who was working with me then, interested in this problem. This led us to a formalism, which now is not the only possible one, and has some advantages and some disadvantages compared to alternative methods, but which, however, gave us something concrete to bite on. We could write down equations for things and compare them and see what might happen. This was about the time when Bohr raised the question, mainly in his paper about the photo-disintegration, that the most naive form of his statistical model had to be looked at with some care because it assumed the independence in all circumstances of the compound state of the way it was formed; and this was not always true when you had the possibilities of overlapping levels. There might be phase relations between these different possible component states, which would depend on how the state was formed; and in particular this would always favor the re-emission of the particle that had come in — the re-emission through the entrance channel. Some other people, whose work had attracted Bohr’s attention to this, had used the formula in the wrong way and therefore got wrong predictions; and Bohr saw how one should do it right. And then the formalism we were just playing with I think gave a way of sorting this out and showing just what you could confidently assert and where the doubts might be. Now, this was discussed partly by correspondence and partly on visits to Copenhagen, with both Bohr and Placzek, and resulted first of all in I think a brief note to Nature by the three of us, which made the point about the photo-electric effect and photo-disintegration and also already pointed out some of the more general considerations that would arise. Then we started writing a bigger joint paper, which has frequently been quoted but never published. The point was that the three of us — Bohr, Placzek and I had rather different attitudes. I was trying to bring in a lot of the formalism which I had just developed with Kapur, because it allowed one to make concrete quantitative statements on certain things. Bohr preferred rather more general philosophical arguments. And Placzek was sort of halfway. He was interested in our formal results but also interested in trying to pin the arguments down physically. So, as did any paper which involved Niels Bohr, this went through many drafting stages and partial drafts and correspondence back and forth and Placzek coming to visit me in Birmingham and our writing again a piece. Finally the war intervened and cut off any communication, at least with me and Bohr. I think Placzek got one version of the paper typed and duplicated which was done without Bohr and therefore didn’t have his authority. At one time I had a copy of that, but I don’t have it anymore, and I couldn’t reproduce a text of that. However, we had talked with many people, of course, about some of the contents, including what was then called the optical theorem, which was a very general result about scattering problems, and which is part of what is now called the unitarity relation — it’s sometimes still referred to as the optical theorem which was very satisfyingly general and was in fact known in connection with scattering of electromagnetic waves but not quite in that general form. We were quite pleased with that result, which I think arose mainly in discussions between Placzek and myself. So as a result later when people wrote up the literature, they quite often referred to this paper by the three of us as an existing paper or as a paper in the course of publication or something. But, in fact, when after the war there was a possibility of getting together again, the subject had developed so much further embodying all the results of our paper, there was no point in trying to write it again.
Was it circulated in manuscript form?
No, because there never was a paper.
It was never written then actually as such.
That’s very interesting, a case of apparent physical being as a paper. The article with Kapur was in 1938, February of ‘38, and then the note in Nature was July of ‘39.
I see. And then the war — that figures. But this was essentially the same discussion that I have referred to in the February ‘38 letters.
Yes, they’re a part of it. This went on a long time. Actually I delayed the paper with Kapur; it must have been ready a little earlier, but I delayed publication because I thought maybe I had hit exactly on what Landau had in mind and that maybe this was what Landau had known all along, and then if as a result of that, stimulated by a conversation with him, I just worked it out, it wasn’t fair to publish it without his name on it. And therefore I sent a copy to Landau. It took a long time to get an answer. There probably were mail delays, or maybe he didn’t reply straightaway. But he then said he’d completely forgotten our conversation, and certainly this formalism wasn’t at all what he had in mind; so that I felt free to publish it.
Did he have any comments on what you had done?
No. I mean he clearly didn’t object to it, but it was a different point from what he had thought about.
Who would you identify at this time as being your most relevant audience — that is, the people you would want to communicate with on a subject such as this and whose opinion would be very important to you?
Well, there was by that time a substantial number of people interested in nuclear theory, and I would also hope that some experimentalists would get interested in these ideas, although my paper with Kapur was probably not easy for an experimentalist to read even though we did our best to make it clear. If it hadn’t been so shortly before the war, I would have made an effort to talk to people and get them to appreciate what the ideas were. Partly, it was still an abstract formalism, and you had to do quite a lot more work to get something for the quantitative side and so on; so it wasn’t immediately usable in connection with experimental work, but I would have hoped to get there.
I meant in terms of individuals, people whom you’d want to show this to for their comments rather than this anonymous larger audience.
I don’t think I particularly wanted to show it to anybody for their comments, because it was clearly right — there was no doubt about that and it was potentially useful, though how far one could get with it experience would show. So, in other words, apart from checking that I wasn’t pinching anything from Landau, I didn’t feel any particular need to invite comments. I mean clearly anybody whom I met at that time I might have shown the paper to, but that was entirely accidental.
Who else would have been working on the compound nucleus? Kalckar was working with Bohr just before he died.
Well … In fact, I was wrong in something I said a minute ago, because I said Bohr felt that somebody else had made the wrong use of the relations and got a wrong answer. In fact, it was in a paper of Bohr and Kalckar where the wrong answer came out, and he then got interested in seeing why it was wrong and how it could be corrected. But I’m not sure of that. Well, Bethe was interested; Breit and Wigner were interested.
In your paper you did make a point. You thanked Landau for stimulating conversations, and then you said, “We understand from Professor Bohr that Dr. Kalckar was considering a very similar method but was prevented from continuing the work by his sudden death. Had you been in touch with Kalckar by correspondence?
No. I mean I had met him, of course, on trips to Copenhagen, not on this particularly. In fact, I was a little doubtful whether Kalckar’s work really was in that direction, but there didn’t seem any point in questioning it.
As long as we’re talking about Bohr, on the compound nucleus question, did people think this was a way out of dilemmas? Did they welcome it as very important?
Yes, it was welcome as clarifying a lot of puzzling results and seemed undoubtedly the right way of thinking about nuclei. It was a little regrettable that this made the picture of the nucleus rather more complicated than it would have been otherwise. But such is life. In fact, however, the idea was very much exaggerated at the time due to our lack of complete understanding of the situation. I think it’s probably right to say that Bohr himself carried the idea too far. The point is the following: Bohr was already at that time comparing the nucleus with the liquid drop, in which, of course, each atom or molecule interacts quite strongly with its neighbors; and any idea of the sort of shell model that one had used in the atom and later came to use in the nucleus, where you think approximately of each particle moving for a considerable period independently without being disturbed by the others, would be complete nonsense. In fact, if this extreme view of the Bohr model was right, then any idea of using a shell model for the nucleus was completely misguided and nonsensical. We now know that this is not right, and the secret is that although the density of the nucleus is pretty high and the particles are very close to each other, we are assisted by the Pauli principle, which prevents particles from going into states already occupied by other particles and therefore very much cuts down the possible interaction of one particle with another in the sense of deviating from the orbit it happens to be on. In fact, the first to recognize that somewhat later was Weisskopf, who pointed to the analogy with the electron theory of metals, which I should have known, where the same thing happens — that although there is a high density of electrons and their interaction is quite strong, the Pauli principle means that most of the electrons simply can’t be shifted from their path, because it would only throw them into orbits already occupied. And therefore at low energies, the ground state of the nucleus, it is possible that the shell model may in fact be perfectly adequate. The difference is that when you come to very highly excited states, like you have in a slow neutron, for example, hitting a nucleus is now some 8 Mev above its ground state, that then this protective effect of the Pauli principle is weakened; and there are so many states of the many-body system to go into, that then you really have the conditions of the Bohr theory. But whereas the first impression was that these conditions of a large number of closely interacting particles, strongly coupled particles, applied always — it was now much later seen that it applied only for very highly excited states. Now, this probably delayed the development of nuclear theory for a bit, this misunderstanding. For example, I’m told that just before the war Bohr was on a visit to Japan and there met a man called Yamanouchi, who was a mathematician, an expert in group theory and who had, in fact, invented some group theoretical techniques which later proved extremely useful for the development of the nuclear shell model. He had recognized that himself and was planning to devote his time now to the study of nuclear states when Bohr came to visit and persuaded him that this was entirely misguided and nothing could possibly be expected from such an effort. So he gave up, and did something else.
Was it the force of Bohr’s conviction that acted to dissuade him?
Well, certainly the force of his conviction and personality, but also arguments he gave, which convinced Yamanouchi, as they had convinced Bohr, as they had convinced me and many other people, because we didn’t quite appreciate the distinction between the high-energy and low-energy regimes.
It wasn’t a question of Bohr’s authority, but it was that it made sense.
And all of these discussions took place either in visits that you would make or that Placzek, for example, during that period would make, or in correspondence. Did Bohr come at all to Birmingham?
Well, he came on one occasion when he was given an honorary degree in Birmingham. That must have been in ‘38 or early ‘39. He spent a day there or a couple of days, and most of it was taken up with official ceremonies and so on, but I did have some time to talk with him. Otherwise he didn’t come to Birmingham in that period. He may have been to London. I may have gone to talk to him in London. I do not remember. I may mix that up with later occasions. But certainly most of the direct contacts with Bohr were in Copenhagen.
It could be one of your special conferences, your Easter conferences or something, or just…
Or just on a special visit. It was not in those days very easy to get support for traveling, not like today. But Bohr had a little bit of funds in the Institute, and when it seemed appropriate he would quite easily invite someone or would invite me to come over and at least look after my expenses in Copenhagen, and on some occasions also my traveling expenses.
This marks the renewal of your interest, at least your published output, in nuclear physics. The interest, as you pointed out, was alive in Cambridge, although there were no publications on it. There is a paper that you submitted in June 1939 on critical conditions in neutron multiplication. First of all, is this after fission?
Let’s talk about fission first, about how you first heard of it and what your reaction was to the news and what you did subsequently.
Well, the news came out, of course, in a very rapid sequence of letters to Nature and hearsay. One talked with people who had heard something further about it.
Do you recall how you first heard about it?
No. This is the sort of thing that gradually grows on you, and you can’t recall where it started. It was interesting. I was not at that point especially excited. Then one gradually realized that this raised the possibility of a chain reaction. However, it wasn’t at all clear how far that was a practical possibility. Then about this time Frisch turned up in Birmingham. I think that must have been some time in the summer of ‘39.
He left Copenhagen in July of ‘39.
Right and probably came more or less directly.
Directly to Birmingham.
Right. And he, of course, was full of ideas about fission at that time, and we talked a lot about it. And I began — I don’t know how far as a result of what I read or as a result of conversations with Frisch — to become mildly interested in this problem of the chain reaction.
The paper, though, was received before he came; was received in June, and he came in July.
That’s very strange.
Let me just remind you: “It is well known that a single neutron may cause a nuclear reaction chain of considerable magnitude. From recent experiments” — and you refer to Feather -– “it would appear that this condition might be satisfied in the case of uranium. It seems of some interest to discuss the dependence of the phenomenon on the size of the body.” And then you gave the theory as applied to a sphere of uranium and determining the critical size of the sphere.
I’m extremely surprised; I’ll tell you why I’m surprised. Well, let’s first go into the background of this paper. It was in the first place a purely mathematical interest to me. I had read a paper by Francis Perrin, who had made a fairly rough theory of this problem using the so-called diffusion approximation, which assumes that the mean-free path of the neutrons is very small compared to the size of the body. Now, it was quite clear that in a chain reaction of something like uranium, this would probably not be so but that the critical size would be of the order of the mean-free path in linear dimension; and therefore his assumptions didn’t seem quite justified. Now, I was a little skeptical of the result, not because of this assumption, because you could imagine circumstances in which that would be justified, but generally this idea of the critical size I hadn’t quite understood, and this tremendous difference between something a bit bigger and a bit smaller than the critical size; so I wanted to sort that out for myself. And doing some very simple theory, I convinced myself that this was indeed so; that now, as everybody knows as a matter of course, you have this tremendous dependence on criticality: if you have something a fraction less than critical, nothing will happen; a fraction more than critical, it becomes unstable. It took me some time to reproduce this argument for myself. And in the course of that I also found how one could extend the calculations to conditions where the mean-free path would be comparable or maybe even larger than the size of the object. That seemed a nice mathematical point, which was worth publishing. Now, at that point I had some doubts whether it was right to publish it, because one was beginning to feel that there might be some military applications of this. If so, was it right to publish such a paper openly? Not because of the power of the mathematical result it contains, because that wasn’t very deep but because one didn’t want to draw unnecessary attention to something as delicate as that. And now my recollection was that I consulted Frisch about the wisdom of this, and he drew my attention to the implications of the work of Bohr and Wheeler, according to which the effect was almost entirely due to the isotope 235, and if that was so, in ordinary uranium there would probably be no chain reaction possible; or if it was possible, it would be marginal and would have no very great explosive power. So as a result one could confidently feel that this was entirely an academic subject and there was no harm in publishing it. From what you say about the dates, my recollection is at fault.
There may be one thing that’s at fault here. I think the date of your submitting the paper is correct. But the date of Frisch’s coming to Copenhagen may be based on recollection rather than on fact, and I don’t know how good that date is.
Well, there’s another possibility.
Frisch recalled that he had gone over in July. But this may be an error.
It couldn’t have been very much earlier, and if the paper was submitted in June (and if that’s what it says on the journal, it’s undoubtedly right), it would have been too early for me to have any conversation with him about the wisdom of publishing it.
I just thought of something. You published in July 1939: you submitted it, the paper with Bohr and Placzek, the note to Nature. Is it possible that you went to Copenhagen in connection with that paper and there saw Frisch?
No. I think the explanation is probably this: that I sent off the paper for publication and then got worried. I may have consulted somebody else. I may have consulted Oliphant, and then may have got worried whether it was a stupid thing to do to send such a paper. And then when Frisch arrived and we had these conversations, I felt reassured and said, “Well, thank God there’s nothing practical in this, and so this paper will do no harm.”
Do you recall what the usual publication lag was from time of submission? This was the Proceedings of the Cambridge Philosophical Society. So if you submitted it in June, the earliest it could be published would be late July or August…
Oh, no, it wasn’t as fast as that. I would say at least three months. And I notice from this that it was published during ‘39, not ‘40; so it must have been submitted sometime in the summer.
It says this right on the paper: June. Meanwhile, though, right after that, you continued the Bohr-Placzek thing, which probably had come to fruition in the note to Nature on the nuclear reactions and the continuous energies.
Well, that really was finished long before. There were just delays with Bohr’s perfectionism in getting it retyped and retyped and probably proofs and so on. So that took a long time to come out, but it was really old hat by the time it came out.
Now, during that year there were two papers, both of them unrelated to fission. I’d like to talk about them, and also to find out what you were doing and what you were thinking, if you did anything further on fission. One of them is a paper which you described before, which had been started earlier, and which was done on the interpretation of the beta disintegration data, with Bethe and Hoyle. And then the other was the paper and reports on progress in physics, on the meson.
Well, the first one arose because the experiments about the shape of beta ray spectra had led people to believe that the original Fermi theory of beta decay was oversimplified. It assumed that the basic interaction was just a constant, whereas these theories suggested that there was some momentum factor in that which would distort the shape of the spectra. And my suggestion to Hoyle had been to look into the various alternative possibilities there, because there were many ways in which you might introduce factors like momentum just to see what were the possibilities with a view to comparing that with the experiment. Now, about that time an impression began to form that the Fermi theory might be right after all, because all the spectra which had appeared to be of the wrong shape for the Fermi theory were not simple but composite; that there were several excited states of the nucleus involved, as well as the ground state of the parent nucleus; so that what you saw was the superposition of several spectra, and this would therefore result in a more complicated shape. Now, Hoyle had come to that conclusion. As we discussed it, I mean, he had approached that conclusion. He discussed it with me, and we sort of tidied it up from simply looking at all the available evidence. And by correspondence I gathered that Bethe had come pretty much to the same conclusion. It wasn’t very easy to settle who had come to the idea first and since we had neither published it, we decided the more sensible thing was to draw it up and publish a joint note since we had the same view. This was then complicated by the fact that the experimentalist in Cambridge with whom Hoyle was in contact were a little bit upset because they were working themselves round to that view but were doing experiments to check it and test whether really the excited states were there, whether the gamma rays were there and so on. And so they took some exception to our planning to publish a note about this. Now this was slightly a delicate problem of ethics because if Hoyle had been influenced by the ideas, then clearly one should leave it to them. But by this time Bethe was involved, and Bethe said he certainly didn’t know anything about what the Cambridge experimentalists were thinking. He was certainly going to publish that conjecture, which seemed to him firmly based on his study of the published literature. He hadn’t any objection to Hoyle and joining in that, but he was going to go ahead and publish it anyway. So this caused some slightly embarrassing correspondence; and in the end with everybody’s agreement it was published.
In the paper you thanked P. I. Dee from the Cavendish “for discussion of the experimental data.”
This was the contact with Hoyle. It was Dee who was a little bit unhappy about this.
That explains why he was mentioned in the paper.
Well, clearly what we were doing under the circumstances acknowledge conversations with him.
Was there a genuine controversy prior to this regarding the validity of Konopinski’s and Uhlenbeck’s theory as opposed to Fermi’s?
No, I wouldn’t call it a controversy. What happened was: After the Fermi theory, people tried to check whether the beta spectra were in agreement with that theory; and the agreement was not impressive. Then Konopinski and Uhlenbeck came along and showed that a lot of the spectra then known agreed extremely well with the alternative version. And while it was a little more complicated than the Fermi theory, there was no fundamental objection to it. People tended to accept that. And then as the experiments proceeded, the situation got more complicated in that the shape of the spectra was variable. Some reactions showed spectra more like the original Fermi theory. Others were more in agreement with Konopinski and Uhlenbeck, and yet others didn’t fit with either. So clearly the thing was more confused. And then the result of these considerations published by Hoyle and Bethe and myself was to say, “Well, we’ve got to be very careful first of all to rule out spectra, which might be composite,” and also those which might be forbidden transitions because in certain types of forbidden transitions there is a possibility of extra factors coming in, so one shouldn’t base a comparison on that; and if you restricted yourself to only those spectra (because more and more had been measured at that time; whereas originally there had been little evidence now we had more spectra to choose from) which were known to be simple and not composite and were known to be “allowed” transitions that then there was perfect agreement with the simple Fermi theory. So I wouldn’t say there was a controversy, but the understanding developed.
The close relationship here of the theory with new experimental results, and then problems of accounting for this, is very similar to me to the work in high-energy physics today.
Oh, any new part of physics has always proceeded that way.
Well, I wonder — it sounds very modern, very contemporary — can you think of other cases in the ‘30s where it was that closely allied; where the experimental results were coming out very fast, and there was a tremendous need to bring the theory into phase with the results?
Yes. Well, you see, you would get this more strongly when a theory was incomplete. You see, in the case of the Fermi theory of beta decay, this was a conjecture — a hypothesis — stimulated by the experimental fact; and there was and is still no underlying basic theory from which you could conclude anything. It’s just an empirical, phenomenological description. And then, of course, once you have that situation — as the experimental knowledge proceeds—you have to check whether your simple description is in line with it or has to be modified. So I think that would have applied in the 1920s perhaps when experimental information came out which helped to complete the quantum theory; and in cosmic radiation, for example, where in the 1930s and later, once you knew the nature of the radiation, the nature of the processes leading to absorption and so on, it kept changing very rapidly — and where there was a very strong interplay between experimental findings and theoretical interpretation. That was less fundamental, because one was talking about known particles — at least until the meson was discovered — and known processes; but still the nature of the system and of the dynamics kept changing. So that was a similar situation. And I think it’s always been like that in physics at any period when there was a rapid growth in experimental data and the development of the theory, which usually go hand in hand.
But it is not a daily occurrence. These things have to coincide; when there’s a very rapid proliferation of data…
Sure. But, of course, there have been periods when everything seems stuck, where either the experiment has exhausted what can be done with existing techniques and you have to wait for new techniques to come along, or where the theory is stuck — as essentially it is in elementary particle physics at the moment.
This still stands out in the ‘30s, though, you’ll grant — the beta decay discussion as a good case, different from some others.
A good case, but not isolated.
You mentioned the meson, and that brings us to a discussion of the paper that you wrote — a review paper — published in 1939. I’m not sure when you worked on it. It was on that subject. I’d like to know a bit about it, because in 1940 you’re again dealing with similar things. You’re dealing with the theory of nuclear forces, and you say a lot about the problem of the meson — its relationship to Yukawa’s theory.
Right. Which by that time I’d understood.
Right. So we can start here perhaps with the discovery of the meson, what the reaction was to it; and then lead to this review paper.
Well, I don’t think I contributed very much there. The point was simply that a review journal asked me whether I would review the papers about the meson. I have a vague recollection that this was to be done in a hurry, because somebody else who had promised such a review was in the last minute unable to do it, and so it had to meet a deadline. Anyway, although I didn’t feel I had any special knowledge about the meson, I was asked to write that review and thought it was a good idea to learn the subject by reading the literature and summarizing it. That’s all I did. There was nothing original in that.
Well, to skip back to the beginning of the discovery of the meson, how did you learn of it and did you see it as immediately linked to the Yukawa particle?
Not immediately. In fact, this goes back a little while, because when I was in Cambridge, Blackett was doing cosmic ray experiments; and there was disagreement between the facts of the penetrating power; the number of particles which came through fairly thick absorbers was higher than the theory predicted. Blackett was conscious of the fact that that might mean the existence of a new particle. But being a cautious experimentalist, he did not want to invoke a new particle without exhausting all possibilities of explaining it in terms of known particles. And so I discussed that with him. Perhaps in fact my influence there was negative in delaying his realization that he was dealing with a new particle, because I, following his request to see whether possibly this could be explained, was trying to pick all sorts of holes in the standard theory which might allow one to get in certain circumstances this large penetration or to look into the statistics of the experiments to see whether there might be possible fluctuations — things like that — so that I probably contributed to doubts in Blackett’s mind whether this was really in conflict with the possible behavior of either electrons or protons, and therefore my influence was possibly preventing Blackett from discovering the meson, which was then discovered by Anderson. But, at any rate, the idea of there being a new particle in cosmic radiation was therefore not new to me. How that first got connected up with the idea that this might be the particle predicted by Yukawa, I can’t recall; but it was fairly close to the surface, because Yukawa had predicted there should be a particle of mass intermediate between the electron and the nucleon, and here was such a particle — so to think that this was the one was a fairly obvious idea, though it turned out to be wrong.
As early as 1940, in your paper on the theory of nuclear forces, you hold out the possibility that there are still problems. You pointed to this question of saturation, the exchange force explanation appeared more plausible than the positing of a repulsive core; but the most promising line of attack, to sum up, was the meson theory. Then you said: “The discovery of mesons of just such a mass in cosmic rays has very much increased our confidence in the meson theory of nuclear forces.” But then you pointed out that there were still problems such as the yet undiscovered neutral meson which appeared necessary to account for forces between like particles. It turned out that if these neutral mesons existed, it would mean “that the particles, the discovery of which was hailed as a confirmation of Yukawa’s theory, had actually no connection with the particles postulated by Yukawa.”
Could I just look at that again?
Part of it is paraphrased and the last part here is a quote.
Well, I don’t now understand what I was trying to say there. I’ll probably have to look back at the full article. I think that perhaps your summary of the previous sentence has shortened it a bit. There was at that time evidence accumulating of what we now call “the charge independence of nuclear forces,” in other words, that the forces between like and unlike nucleons were equal, which we now know to be true to quite a good accuracy. Now, the simplest way of accounting for that in terms of Yukawa’s theory would be to assume that only neutral mesons are responsible for the interaction, because that would then make no difference between like and unlike particles. It would also, however, mean that there were no exchange forces. The difficulty in our minds at that time was that if you had both neutral and charged mesons, then if you had unlike particles, you could get an exchange of either charged or neutral ones. If you had like ones, you could only get the neutral exchange. And therefore it was very hard to see how the forces could be equal. It was only the work of Kemmer later which clarified this — that you could make a very attractively simple formalism which would lead to exact equality of the forces. And what happens really is that the forces due to the neutral meson have opposite signs. They are, for example, attractive between unlike particles, and repulsive between like particles, but that repulsion is overcompensated by the attraction due to the charged ones. So in what appears from that point of view a rather complicated cancellation, you can still get the same forces — the charge independence of the forces. But that I didn’t know at that time, and therefore the charge independence would suggest that the forces were due only to neutral mesons. And if that was true, it would mean that the Yukawa theory needed neutral mesons; and the experimentalist had discovered charged mesons, and therefore they were different particles. That’s what the meaning of that remark is. It had nothing to do with the distinction between mu- and pi-mesons, which in fact make the final sentence a valid one today, but for quite different reasons.
In that same article you mentioned two approaches to finding out about the forces — either from direct observation or to derive them from simpler, more general laws. The limitation, of course, on the first method, which had been attempted since the early l930s, was the lack of accurate middle-energy scattering experiments; and the implication is lack of high enough energy accelerators, although you didn’t use that term.
I wanted to ask when it became clear that higher energies would be needed to gain information on the nuclear forces.
That was always clear.
Well, do you mean higher was the next order beyond which you could tell…
Well, you couldn’t tell. You can’t ever go too far ahead by dead reckoning in physics. The point was that it depended partly on the range of the forces — the shorter the range, the higher the energy you would need to get information about them — and the fact certainly was that the energies then available gave very little structure: I mean that the results of scattering experiments could be accounted for in terms of a couple of parameters meant certainly that you had to go higher — how much higher, you couldn’t tell. What wasn’t suspected at that time was that in addition to the normal attractive forces, there is a repulsive core at short distances. Now, to find that, the direct evidence of that comes from experiments at about 200 Mev. But that we didn’t know. So I don’t think anybody at that time would have been ready to make a statement as to how high an energy you do need to get sufficient explanation. In fact, what also wasn’t known then: not only is the direct experimental exploration of the forces difficult (or was difficult at that time because you didn’t have high enough energy beams available), we also know today that the phenomenological description of the forces is extremely complicated. So to get full information from direct experiments is a very long and uphill job. You see, if we believed of course that something like Yukawa’s ideas were right, the law would be a fairly simple one; and therefore, having a few points, you could guess the shape and you might be right. It turns out that almost every conceivable complication that you might imagine is in fact present, and therefore you have to pin down every point on the curves almost. And although now adequate particle beams are available, have been available for many years, the job of surveying the interaction completely — of doing enough scattering experiments to do the surveying — is not yet complete. Work is still going on.
That’s clear. It’s a good explanation. I think, though, that what I’m commenting on is the impression one gets in looking through the literature of the ‘30s: that in fact the understanding of the forces was obtainable; that in fact a lot of the work was done with the motivation of coming to an understanding of the forces.
Well, we were over-optimistic in both respects. The only elementary law of force we knew was Coulomb’s law, and that’s simple enough. Now, we knew these forces were more complicated at least because they were short-range, something like the Yukawa form. Now, if it was anything as simple as that, a limited range of experiments would have been sufficient to tell the whole story. And also one could have imagined that a fairly simple theory would have given the forces at least an outline. We didn’t know how complicated the situation was. We didn’t know how inadequate the quantitative part of Yukawa’s theory was. We might have guessed that because the coupling is so much stronger — I mean the constant, which in the electromagnetic case is 1/137 which has for the unclear force to be much greater, either a tenth or 15 depending on how you measure it and what you compare. But it’s certainly high enough for all sorts of high order corrections, which are minor corrections in the electromagnetic case, to be quite dominant and require very highbrow theory, which hasn’t yet been completed.
Getting back to one of the questions about the repulsive core, the feeling one gets is that physicists were reluctant to accept the repulsive core to explain saturation.
Why was that? That’s a good question. I’ve been often trying to reconstruct why that was. Why did one find the idea of exchange forces so much more reasonable? I think it was essentially this: that if you think of orders of magnitude in nuclear physics, you think of a few Mev — I mean binding energies typically, 6 to 8 Mev per particle. So that’s the order of magnitude. Now, then you discover that the forces are of very short range, and therefore to give enough effect, the strength of the potential at short distances must be much bigger — 50 to 100 Mev. You find that already surprising but unavoidable. But then if you have such an attraction outside of a repulsive core, then clearly that repulsive core must have a much shorter range still. And therefore to be effective must be much stronger still. I mean many calculations are done taking that repulsion to be of infinite strength, no doubt idealized. But you probably need something reaching toward bev magnitudes. Now, that seemed too wild. We just didn’t have the courage to postulate such models until we were actually forced into it. I think that was mainly the reason — plus the elegance of the idea of exchange forces fascinated everybody, and made one think this would be a good explanation.
What was the impact of Rabi’s discovery of the quadrupole moment of the deuteron? This is something you discussed in your paper — I mean pointed out that it was an important contribution to the force problem.
Right. Well, it gave one more insight; it was one more feature that was interesting, and it was a most ingenious experiment too. From the point of view of people who wanted to get a simple picture of the forces, it was very disappointing, because it meant a considerable additional complication that we hadn’t expected. And this is one of the things where in the course of time almost any conceivable complication that one could imagine was in fact there when you were able to look for it.
During this period (the war is already on, and I’m going to get back later to fission as a continuous story, but I’d like to talk about the other things that you also were doing), you have gone back to solid state for a while in “The Size of a Dislocation.” I was just wondering how that slips in once again. Orowan was apparently connected with this in showing you an idealized model. Where was he at the time?
He was at Birmingham.
He arrived as a refugee then, too, didn’t he?
Yes. He was for some years in Birmingham, eventually moved on to Cambridge and later to MIT. We were friends and were discussing all sorts of problems. There was a problem. Everybody then knew about dislocations, that they might be important in the mechanical properties of crystals; and there was a problem: what was the force necessary to move a dislocation? And what might be the factors preventing this movement (which had something to do with the lattice structure)? At least in a perfect lattice you could formulate this as a problem. But it was a very complex problem involving the motion of lots of atoms in a lattice. And Orowan had the thought of idealizing this by assuming that you could treat the two halves of the lattice, which were sort of sliding over each other, just as two elastic continua with a plane surface; and then write down a simplified expression for the forces arising between them due to the mismatch of the last layer of atoms which you had there. Now, he had the idea I think physically quite clearly but hadn’t yet written this down in mathematical terms to get an equation, let alone attempting to solve the equation. So he asked me whether I’d be willing to help with that. And I thought that was an interesting problem. And it turned out to be fairly easy to formulate the equation, which required refreshing my memory of the law of elasticity, which I had forgotten. And so I formulated the equation, which then came out to be a nonlinear integral equation, which is generally a sort of hopeless proposition to solve exactly and one has to try to find maybe approximate methods or something. I had a great surprise there, because I tried first of all to visualize qualitatively what the solution of that equation should look like, and it was clearly a function which took the value of 1 at one end and -1 at the other end and smooth in between. In other words, it behaved more or less like an inverse tangent. So I decided to see what happened if you assumed the solution was an inverse tangent and inserted that into the equation. And to my everlasting surprise, it turned out that this was the exact solution. So I had just guessed an exact solution to an otherwise hopeless-looking equation. This was shortly before a conference in Bristol, to which both Orowan and I went; and he suggested I should present my solution to that problem, which I did. And then it had to be written up for publication. Orowan was writing a paper on the practical side of this, and I thought he should just in the course of that mention that solution, because it was really his idea basically, or at least it should be a joint effort; but he insisted I should publish that myself, which I eventually did. If I had guessed what impact this paper would make and for how long it would be quoted, I would certainly have insisted that Orowan’s name should be on it as well. I feel a bit like the sorcerer’s apprentice.
Well, it’s hard to predict. There’s another paper about in ‘4l. And I think we talked about this a little bit before — the paper with Dirac and with Maurice Pryce, “On Lorentz Invariance in the Quantum Theory.”
This was just the delayed effect of some discussions we had much earlier.
You referred to a recent paper — that was Eddington’s paper of 1939. By this time Pryce was at Liverpool.
Dirac was still at Cambridge, and I was at Birmingham.
But this other paper of December ‘41 with Hepner… Who was Hepner, first of all?
Hepner was a refugee who had turned up in Birmingham in some way I don’t remember and who had no means of support originally. And I think that was the time when I was beginning to get involved with atomic energy and therefore was trying to get some relief from teaching duties and therefore arranged for some kind of temporary and part time appointment for Hepner to relieve me of what lecturing he could help with. And so that gave him a rather miserable but still some support to continue in Birmingham. And then while he was there, he was interested in keeping up with a little bit of research; and I suggested to him a problem which essentially was looking at what is now called the effective range approximation, which Bethe and I had used in our paper on the neutron-proton interaction, to see how far these results would be upset by the new discovery of tensor forces — I mean Rabi’s discovery of the quadupole moment of the deuteron showed that there were tensor forces present — would these invalidate the previous results? And in this work with Hepner we came to the conclusion that they wouldn’t; that if the parameters appear in the appropriate way, then essentially all the previous relationships could be maintained. That was the content of that paper. It wasn’t a terribly big thing.
You keep picking up the deuteron over the years as an approach to the force problem, it seems.
That’s right. Well, the best source of information about the forces is the two-body problem, because the mathematics of it is simple, and the connection between the interaction you assume and the observed results is fairly direct; whereas in anything beyond that in the three-body problem, even if you know the forces, you have to work very hard for years to make any firm prediction about what you should see. It was the same in atomic physics where the development of quantum mechanics depended vitally on the hydrogen atom, which is an equivalent problem there. I mean Bohr was aided in seeing the spectrum of the hydrogen atom. The unfortunate thing is that in nuclear physics the two-body problem has just one bound state and no spectrum otherwise, and you are then forced to rely on scattering experiments, which are much less clear-cut. But if there was a system analogous to the hydrogen atom in nuclear physics, we’d be playing with it all day to get what information we could.
There’s one other paper, but I think it’s part of the other story, and I’ll ask you whether it should be related to that. By “that story,” I mean fission and the bomb. That’s again a review paper: “The Bohr Theory of Nuclear Reaction.”
Nothing to do with fission.
In other words, there wasn’t any outgrowth of …
Well, no, that was an outgrowth of Bohr’s idea about the compound nucleus; my work with Bohr and Plazcek; my work with Kapur, as far as it was relevant, because this was a very simpleminded review and not very mathematical. I’ve forgotten there whether I was invited to write that review or I volunteered. I can’t remember. I probably was invited. It was an easy thing to write up, except that it was an opportunity for finding some simple arguments, simple ways to explain thin I knew were right, but which I had come to by a rather more complicated argument. But this didn’t represent any particular research activity. It was just a review on things I knew perfectly well.
Well, getting back to that other story then, we started to talk about your paper, which you discussed with Frisch as to the propriety of publishing it. And then the remainder of the year of 1939 you spent apparently working on old problems. Did you pursue the fission question at all during that year?
No. It didn’t seem to me that I had anything particularly to contribute there beyond the little mathematical point I had made, which was a sideline; and also I certainly didn’t think that there was any practical consequences that were worth bothering about — I wasn’t aware of them. Perhaps I was behind the times there, because lots of people had started speculating about chain reactions and so on. I really didn’t give it a thought at that time except in a vague way. And I was a little surprised when a colleague of mine, Philip Moon, was rather interested in this paper about this calculation and asked me for a reprint or maybe an advance copy. That I found rather surprising. I found later the explanation that he was a member of the government committee that was looking into the possibility; and in fact, after doing some rather crude experiments, they decided at the beginning of l940 that the possibilities of doing anything in the reasonably near future were negligible and decided more or less to wind up their work. But that I didn’t know. It was a secret. And so I wasn’t really interested in fission except that it was one phenomenon in nuclear physics which I was interested in. I had read the Bohr and Wheeler paper, of course, which was a very interesting piece of physics. There it was left. I had plenty of other things to worry about, because at that time, besides normal physics, I was trying to continue. I had some students left then who were doing a little bit of research work, one doing a master’s degree, another hoping to do a Ph.D. but he was called up before he got very far. Also there were all sorts of problems as I was technically an enemy alien, and there was a lot of paperwork with that. I wasn’t allowed to own a car in those days and had to transfer my car to somebody else, make it somebody else’s property, or I would have broken the law. There were all sorts of restrictions like that. Anyway, nothing happened until I had a conversation with Frisch about fission in the course of which he sort of asked: “Well, the Bohr and Wheeler paper has made it quite clear that the fission is due to 235. What would happen if one had a pure uranium 235 in a sufficient quantity? How much would you need? And if you got it, what would happen?” This seemed a rather academic question, because the idea of doing isotope separation on any practical scale with a heavy isotope like that seemed completely academic and unrealistic. But I’m always willing to try and answer questions somebody asks me. I mean he knew I had written that paper and that therefore the formula to estimate the thing was contained in that paper, but you had to make a guess about the cross section. Now, from the general Bohr theory of nuclear reactions, which I knew pretty well, and the analysis in the Bohr and Wheeler paper about fission, it seemed quite reasonable to assume that for neutrons on uranium 235, the cross section would be about of the order of magnitude of the geometrical cross section of the nucleus. In other words, every neutron that passed through the nucleus would do something. And that the dominant effect will be fission. In other words, most of the neutrons hitting the nucleus would cause fission. This turns out to be somewhat exaggerated but not very far. So with that you could on the back of an envelope take my formula and find out what the critical size was. And we were completely staggered to find this was very small — a few kilograms, in fact, at that point even less than a kilogram. Anyway it was quite a small number. Now, one was used to thinking of critical mass in terms of tons of ordinary uranium. And you don’t intuitively appreciate how much the fact that a factor of a hundred in the concentration means increasing the effective cross section by something like a hundred, which means almost a hundred in the linear dimension, and therefore a fantastic factor in the mass. Anyway so there were the numbers, and this was quite a small thing. Then the next question we asked ourselves: “Okay, now supposing you had that thing and managed to set up a chain reaction, what would happen?” In conversation I’m not sure how much of that argument was Frisch’s, how much was mine, but we came out in discussing it jointly that this would now involve a competition between the building up of the chain reaction and the blowing up of the mass of uranium as a result of the energy developed. If you did it slowly enough, of course, it would just be blown apart when it had reached enough energy to about evaporate or something; and if that was all, then once it had expanded, the chain reaction would stop; and in that case you wouldn’t get more than the power of a chemical explosion. However, if the chain reaction had gone rather far before the mass had had time to move appreciably, then it was a different story. So here was again a mathematical problem, involving a little bit of knowledge about the time scale of neutron multiplication, which was also contained in my paper in a rough way; knowing a little bit of hydrodynamics, which I had learned by virtue of giving a lecture course a few years before. And so it took a couple of pages of calculations to show that you would get a very substantial fraction of the available energy released; whether 5% or 20% or 50%, our arguments weren’t good enough to show, but that didn’t matter. Then we got excited and said, “Good heavens, if one could ever have this, it would be such a powerful weapon that it really would justify making a considerable effort to get that.” And in fact, as we said, rather conservatively: “Even if the plan to separate the isotopes would cost as much as a battleship, it would be worth having.” That was a considerable underestimation at this point of the cost of a plant, but the sentiment was right.
How long did the work with Frisch take up to this point?
A few days. I mean the first part, to estimate the critical size, once you had this idea that took a matter of minutes. Then the other argument about the efficiency of the explosion — I don’t remember how long it took, but to write it down was nothing. One had to formulate the problem a little bit and think what was involved, but I don’t think it took more than a day.
When you came to the realization of what could be done, was it clear then that you should communicate it to other people? Where did this idea come from? Was Frisch aware of the other committee?
Neither of us was aware of that. Our idea was that this was such an important fact… Of course we realized that other people might know this, too, but we also realized that the connection with the Bohr theory was not quite obvious; that other people like ourselves might not have had the courage to postulate the existence of separated isotopes and investigate the consequences. So we realized we might be the first ones to think about this. And then clearly we shouldn’t sit on that idea. In a country at war it was our job to make that available to the right authorities, having only a hazy idea who the right authorities were. So what we did was first of all to sit down and write a report of a few pages setting out our conclusions. In fact, we wrote two reports — one being a technical one saying more or less what I’ve just summarized; the other one being written in nontechnical terms more or less summarizing what the difficulties were and possibilities or effects of this would be. Both of these disappeared. I mean I had no access to copies after a while. We made very few copies for obvious reasons. I think we made only two copies of it.
Did you write it out by hand or type it?
Typed it. I typed it myself, because we couldn’t trust that to a secretary. Also, I didn’t have a secretary.
Was this the one that was subsequently published by Margaret Gowing?
That’s the technical one. The more general one had disappeared, and Margaret Gowing couldn’t find a copy. But later on Ronald Clark, also a popular writer about science history, wrote a biography of Tizard and in his papers found a copy of this more popular version or more general version of this and published it in his book on Tizard. Of course, it contains nothing that isn’t already in the technical part except on seeing it I was reminded of a phrase we had used there which I had forgotten. I’m rather pleased it’s there, because we pointed out that apart from the blast effects and so on, there would also be radioactivity locally in unpredictable amounts and probably fall-out and this would also be liable to cause casualties and so on. And we then went on to say: “It therefore seems likely that this might never be a suitable weapon for use by this country.”
From the very start you postulated that it might have these effects.
And probably this was something one should not use. We didn’t regard it as our business to see that it wasn’t used. We lived in a country whose government we thought we could trust, their decency and good sense; and as long as we made all the technical implications clear, it was up to them.
You wrote this and then brought Oliphant into it? I don’t want to go over a lot of ground that really is covered in a better, more thorough way by Margaret Gowing, but I want to get the framework of the story and any interesting implications that relate to personal reactions to things, motivations and connections with post-war work in nuclear physics of the type we’ve been discussing.
Well, this is getting rather far from nuclear physics work, of course. But anyway the mechanism was that we didn’t know where to send this or what to do with this. And so we showed it to Oliphant and asked whether he agreed that this was something we should bring to somebody’s attention, and of course he did agree. He then advised us to send it as a secret document to Tizard, which we did.
With any covering letter to him?
I don’t think so. He didn’t send it. I think I sent it myself.
I think Margaret Gowing says that he sent it.
It may be.
He sent it on to G. P. Thomson.
That may well be.
And this resulted in a small uranium sub-committee being set up of the other committee that was concerned with air warfare; in other words, the Maud committee.
Oh, that Maud committee came later.
I mean ultimately.
Yes. I think the committee that was under Thomson and which had looked at this problem but taken a rather pessimistic view of the possibilities was still formally in existence, and I think it was first brought to the attention of that committee.
That’s the implication. Now, in the events that follow, the outline of it that I know is that you were of course not invited to the committee meetings since you were … You were by this time naturalized, right?
But Frisch was not. And you did complain about never being in contact with the committee.
Was there a real threat of internment in your case? I know there was for a number…
Not really. I think I was rather fortunate because the group of enemy aliens were in the various categories — three categories. The one extreme were those who were obviously a danger to security and who were interned immediately, who might be German agents or something. Then there was a middle category of people about whom not enough was known, who were probably all right, but as a precaution were put under rather severe restrictions. And then there were those who were well known and vouched for and recommended and for whom the restrictions were minimal. I was fairly quickly placed in that category. Then, however, there were complications, because people who lived in certain areas — near military installations and so on — were interned regardless of their category. Well, some were told to leave the areas and go somewhere else, or they were just interned. Now, particularly in the summer of 1940, when there was a general panic, lots of things had become very untidy and lots of unreasonable things happened, I was fortunate in being given fairly good clearance straightaway and also in not living in any sensitive area. So there was never any very great problem.
And of course you’d been there for a while, too. When did you become naturalized? Was this ‘40 or ‘41?
I think it was February ‘40, if I remember right.
What about this period? There’s a lot to it, and I don’t know if you want to summarize this, keeping in mind the part that would fit into the kind of thing we’re doing — you know, so that we’d have some continuity, but whichever way you’d like to handle it.
Well, I think it’s probably not worth opening that whole story, because the connection is rather remote to what we’re talking about.
I could leave you with this, by the way, which is sort of a summary of the period put together mostly from Margaret Gowing. And if there are any things that are glaringly wrong with it, then correct the record.
Okay. But anyway the immediate effect was, of course, that from then on, almost immediately, I dropped everything else I was doing, except things that were urgent obligations. I mean continued lecturing for some time until somebody else could be found. I withdrew somewhat from sitting on faculty boards and so on, because there had been some ideas amongst my colleagues that it was undesirable to have an enemy alien or somebody at least of enemy origin in the counsels of the University, that that was unwise and so on; and so I decided, before it was suggested that I should withdraw from the meetings, much better if I didn’t go. So, in fact, I didn’t attend any faculty meetings, which came in very handy later when I didn’t have any time for it anyway.
A lot of people would like to have that excuse now. But they do this because they themselves objected to it? We were talking about the suggestion that you withdraw from the university councils and other administrative posts.
Yes. Well, I never quite understood this. I think it was just lack of imagination; that people think in wartime to have an enemy sitting on the committee that conceivably might discuss some delicate information seem to them the wrong thing. I think if it had been argued out, there probably would have been a minority of people with that view. But I didn’t want to take a chance on how this would come out. I think I must have had the feeling (I don’t recall this very clearly, because I got interested in more important things at that time) that I wasn’t too confident in the way the people running the faculty board would go; otherwise I probably would have relied on them to see me through. But this is guesswork.
How large was your family then? Had you other children?
We had two children by this time.
How old were they?
Well, they were born in ‘33 and ‘35. We’re now talking about 1940, so they would be 7 and 5. They were not at home at that time. Their school was evacuated to the country. Then in the middle of 1940, the possibility came up of sending them to Canada, on the invitation of the University of Toronto who invited the staffs of several universities in England to send their families. We jumped at that opportunity — not because we worried too much about bombing and things like that — but, of course, in the summer of 1940 the possibility of invasion looked very real; and with our both being Jewish, my wife of Russian background, and I German, we didn’t think very much of our chances of getting along and thought to have the children out of the way at that point was the right thing to do.
I think the Ewalds did the same thing. Did the outbreak of war — within a year anyway — stop all normal physics work? Not all papers. A theoretician doesn’t need much of an environment; but in general things were geared to the war or affected by the war.
Well, it took some time. In the physics department in Birmingham, for example, they completed the assembly of the cyclotron magnet — I think this was completed somewhat after the outbreak of war, but the feeling was probably that having a complete magnet might be useful for some purpose, and having a half-assembled magnet was just no good. They knew they couldn’t finish the cyclotron. And the physics department turned over completely to radar work — I mean the research facilities. Not every member took part in that. And, of course, undergraduate teaching continued. And those people who were not involved with actual work and war research and who, say, were continuing with their teaching duties, would of course continue with their research work in spare time insofar as they were able to. Oliphant wanted me originally to help with the radar work, but the Navy, for whom he was working, refused clearance. In fact, I had an office then in the physics building. I don’t know whether I’d moved into that office or not. I think I was just about to. But since it was a classified building, I wasn’t allowed to.
It was about this time, though, that you did get involved. What kind of a transition was it from being on the outside — knowledgeable but on the outside — and then being directly involved in the project?
You mean how did this come about?
Yes. In other words, was it a direct consequence of the memorandum that was submitted?
Oh, yes. I still never got a Navy clearance. I mean while I was working on atomic energy, I wasn’t allowed in Oliphant’s building.
Oh, I see. That didn’t change.
That didn’t change, no. In fact, I never got a written acknowledgement of our memorandum, but I got a personal message through Oliphant saying they were very grateful and very interested in this, and this was important, but of course we ought to understand that this would from now on be taken over by other people and we wouldn’t know anymore what was going. That we didn’t think quite right.
But in fact you did get involved later.
Then I wrote a letter objecting to this, and partly as a result of that, we did get involved.
Well, there were a number of papers you submitted with Frisch and so forth which were directly elucidating the specific problem on this. This is what Gowing reported. Well, it depends how you want to handle this. I have sort of a chronology of the events here.
Yes. I wonder whether it isn’t really best to skip right to the end of the war.
I can give this to you, and you can take it with you. The one thing that I can comment on before the war period is that you came into contact with people you normally wouldn’t have or worked closely with, but, you know, on different questions. Just the names that show up in the book there — Urey and Pegram, and then, of course, you’re working more closely with Chadwick, and Simon comes into the situation.
Yes, well, these were all people I knew anyway. In fact, I think probably I was responsible for bringing Simon in, because he seemed to me to have just the right kind of ability and experience to help with the isotope separation problem, and that proved right.
What was your impression of the United States when you went? I’m not talking about the project. But did you have any way of getting an impression of American physicists in this environment?
Well, of course, I still remember very vividly my first trip. I’d never been to the United States; I’d never been in a plane before. Here I found myself traveling in a converted bomber across the Atlantic. I’d been warned that this was a somewhat hazardous trip, because the planes were not properly heated — you had to wear flying clothes — and there had been accidents where people lost limbs through frostbite and so on, which didn’t deter me; but eventually, as you do a lot of waiting around since these planes were dependent on weather and so on, I tracked down those stories and found that the fact was that one man, who had worn rather too tight boots in his flying suit which had restricted his circulation, lost a couple of toes through frostbite, which is sad but not quite as bad as the stories sounded. Anyway, after a night over the Atlantic, I got to Montreal, where these planes landed in those days and found a message from G. P. Thomson, who was then in Ottawa, that he wanted to see me; so I got the next plane to Ottawa. Then going from the rigor of wartime England to the comfort of a commercial airliner from Montreal to Ottawa, and then coming down in the evening on the brilliantly lit city (from my point of view because we were used to blackouts) was an enormous thrill, and of course then to New York, which I’d never seen before, and then coming in to the department there, one was of course quite impressed with (although theoretically one had known it) the much greater scale of things and size of laboratories, the numbers of people, the wealth of equipment and so on; together with a sense of purpose and hard work. I mean I remember that it was then normal for everybody to be in the department all day Saturday, as well as all sorts of evenings; whereas we had mostly still stuck to the practice of having Saturday afternoon off. I find it now changed, because we still work Saturday mornings, and hardly anybody in this country nowadays does in universities. But, of course, I was much too busy talking about atomic energy problems to look at the normal physics that was going on, which, of course, in this country was still to some extent going on, though on a reduced scale.
When did you come?
‘42 was the first visit.
There wasn’t too much, because many people were off in the radiation lab at MIT by that time.
Oh, most of the senior people were, yes. But there were still lots of people doing normal work.
Did you come on the same plane with Simon and Oliphant and Chadwick? They came at that time. I just didn’t know if it was the same trip.
No, it was not. I can’t remember the reasons. I think Simon came by the clipper that was then going through Portugal. I think they had ruled against my going that way, because my name was known as being associated with nuclear physics; and, of course, Portugal being neutral, the passenger lists were known to everybody; and they thought that somebody connected with nuclear physics being flown out that way would attract attention. And Chadwick didn’t come that trip. He came the following year. Halban had preceded us. There were lots of jokes, of course, about the British contingent arriving to discuss things with the American scientists, consisting of Simon, Halban and myself, but none of us being very typically British.
It was put this way: “that the accents of the group were anything but Oxonian. The joke spread quickly that the British delegation did not speak English.” This was from this book, The Manhattan Project. A lot has been written about that. What I’m interested in is whether toward the end of the war, there were thoughts about returning to pre-war pursuits — whether there was a question of picking up the same things, the ones that you had dropped, and continuing the same way. Or if in fact the war had changed one’s tastes and problems, or if there had been certain breakthroughs either in thinking or in apparently unrelated things which made certain problems seem soluble.
I think there was no fundamental change. Of course, the United States got into the war later than we did, and therefore a certain amount of work had continued there while we were prevented from doing research. So one of the things was to discover what one had missed out on. Otherwise, I don’t think the nature of the problems or what was interesting changed particularly. The fission problem and things arising from that never struck me as a problem of very great intrinsic interest. It was of practical importance, of course, but nothing I would want to devote my time to once I returned to academic work — except that, of course, I remained for a time a consultant on various committees discussing the future of atomic energy in England — something I was interested in. Also I spent some of my time — and I have done since –- on general problems, in the first place explaining atomic energy to popular audiences for which there was a tremendous demand, of course, and then thinking about the post-war world and how one might prevent further atomic wars and that sort of thing. But that clearly was not a professional occupation but purely a sideline.
I think it’s worth talking about some of those questions.
Okay. But they had nothing to do with the research work.
That answers my question.
What did make a difference was that partly as a result of the war, partly for other reasons perhaps, the position of scientists was very different. Resources became very much easier to come by. There was more of a demand for trained personnel, and also there was an accumulation of students wanting research training. I had the good fortune, of course, that my name had become known through all this atomic energy work, and therefore I had a fairly strong pull in the University; and I used that first of all for the rather minor change, to change the name of my chair and my department and make it a separate unit, which was convenient but was of no great consequence, but also I used it to get an increase in staff. From the point of view of teaching duties there didn’t seem much case for an increase. I got one more member of the permanent teaching staff. But I said what we should do in order to get our research group off the ground was to create four research fellowships, post-doctoral; these people would help in the teaching, would give three lectures a week either on the graduate or post-graduate work and introduce more variety into the teaching, but their main function would be to help me build up a research group and get off the ground. And that turned out to be absolutely right.
This was what you had in mind from the start — for the research group.
No, I thought of building it up gradually without any particular resources, just training people myself.
And you pointed out that you recognized that this wasn’t possible.
How about Oliphant? I’m not clear. He returned to Birmingham, didn’t he?
And did he pick up his work on the cyclotron?
Yes. However, he lost interest in the cyclotron at that point, although the installation was completed. But he then had thought about the synchrotron principle, an idea which had come to him simultaneously with other people, Veksler and McMillan. And I don’t quite know about the priorities, but certainly Oliphant was very interested in that and got going as soon as he got back to get government money into building a 1 bev synchrotron at Birmingham.
In terms of what he had in mind to use it for, we talked about the force problem; we talked about the question of higher energies. Did this mean that just the very availability of a new principle which could give you far higher energies than the cyclotron would be in itself reason for building such a machine? Or did he have in mind its relationship to these other important theoretical problems?
Well, I don’t think that he was very specific –- rightly — because once you open up a new energy range, it’s very hard to predict exactly what you want to do in that. In fact, I think that with most of the machines you can trace it back now and see that what was said at the time the machine was in the thought stage of what would be the justification, what typically you might do with such a machine, eventually in most cases it turned out to be wrong because physics developed differently. But that on increasing the energy range, you would open up something new was fairly clear. The question at that time was how far you should go. Other people went to more modest energies. For example, Chadwick decided to go for a cyclotron of, I think, 200 mevs, if I remember right — between 2- and 300. He had a small one in operation during the war, and in fact this helped to make some measurements of cross sections which were quite important for atomic energy. But he went for a big one after the war. And this was slightly more modest than Oliphant and therefore more certain of achievement, because it was in fact similar in range to a machine that was either in existence or being built in Berkeley, so he got a lot of help for that going and so on; whereas Oliphant was building a thing on a brand-new principle and had considerably more difficulty. The question was not what kind of physics would you do, but just how far you would dare set up the energy range to get a machine in a reasonable time. This was a famous occasion when Cherwell in Oxford made a disastrous decision. Cherwell was a great personality, but not really an outstanding physicist — very arrogant and always believing he knew things better than other people. Now, the Clarendon laboratory in those days before the war had two sides. One was low temperature, and the other was nuclear physics. Now, as Cherwell’s training was in thermodynamics and low temperature physics, you would have expected him to interfere more on the low-temperature side. But, in fact, it was the opposite, because he knew enough about that side to understand that Simon and Kurti and people working there knew their stuff; and it was best to leave them alone and just help them by getting facilities and resources and so on; and that really worked very well. In nuclear physics he didn’t know enough to know his limitations, and therefore ran things on a very personal basis and interfered at every point. This resulted in bad appointments; it resulted in good people being allowed to go because no provision was made for them and so on. And then the decision about the machine, where Cherwell was very clever and said, “All these people are going for very ambitious projects and very high energies. It will take them years to get anywhere, before their machines are working. We will be modest. We will go for 50 mev betatron. That means we will have a machine that can be used for physics ages before anybody else is in the business.” It wasn’t a question of money, because money could be had for such projects for the asking at that time. And at Oxford Cherwell could have had anything. In fact, the design of that machine was bad, and it took years to make it work—long after all the other machines were in business. When it finally was operating, it was discovered there wasn’t any very useful physics to be done with it. So judgment was quite an important thing. It was a combination of the sort of engineering judgment as to what is a promising machine to try and build, together with what might be useful to physics. But you couldn’t plan on paper clearly the projects of experiments, because you wouldn’t know how things were going.
Did the radar experience during the war come in handy in the postwar work? I know in general that it did, but I mean in terms of specific people.
Oh, yes, in two ways. One was that every cyclotron or any of the later machines — synchrotrons and so on — require radio-frequency sources for their operation. And there, of course, the experience of radar gave eventually more possibilities. It was probably more important that the art of counters — developing counters and recorders, amplifiers and so on, which requires sophisticated fast electronics — was enormously assisted by the radar experience.
Would you say that one effect of the war was to train a generation of technical people who were then available to work on these machines? Would there have been this kind of expertise available before the war?
It wouldn’t have been available before the war, but of course it would also have developed. It’s very hard to speculate where physics — nuclear physics in general — would now be if it hadn’t been for the war. One thing, of course, it would not have had the resources available — the money — the popular interest in it. That’s clear. But if you discount that, I think on balance it would have been better off because we lost generations of people who didn’t get their research training at the right age, who got very skilled in the technical side — electronics and so on — and therefore were helpful in building the machines; but we were desperately short in England — we still are very short — of people who have the right insight into physics to make use of the machines. There has been a tremendous change in the relations of nuclear physics (between American and British nuclear physics), because typically the situation before the war was that American universities were wealthier and had very much better resources (very many of them had excellent cyclotrons) and on the whole the amount of good physics work done with these cyclotrons was very limited. Many people were preoccupied with improving the performance of their machines, getting the energy a bit more, getting the intensity and reliability of the beam; but they were lacking in people who were interested in physics; whereas, in those days in England it was different. Chadwick and Cockcroft and so on with very limited resources knew exactly what they wanted to do and did very much more with their very poor machines than many American departments with their very much better equipment. After the war that trend reversed completely. In the United States you not only had better facilities, but you also had better people to use them. And this was partly because our people were for so many years preoccupied with getting their gear together.
During the immediate post-war years.
That’s right and missed out on the chance of getting an understanding. That’s improving now.
Apparently the particle of major interest in the post-war period was still the meson. But would you say that would have been just picking up a thread from before? Take your review paper in 1940 which talked about the meson as the major particle.
I don’t think it was the major particle. It was a new particle, and I was instructed to write about it, so I did. I think if you’d asked me at that point the question, “Which is the most interesting particle?” — I wouldn’t have known.
Well, we’re talking about the theory of nuclear forces.
The connection hadn’t been made at that time and still hasn’t been made. We haven’t got any decent theory of nuclear forces yet.
I’m leading up to the paper you published in October of ‘46. That’s as close after the war as we can get. It was called “Fundamental Particles” and was a talk you gave at the Electronics Group of the Institute of Physics, which means, I think, a more popular talk.
You described the known fundamental particles, and you noted that there was still no direct evidence for the neutrino; and concerning the meson, you said, “One should not consider the meson as one fundamental particle but as a bewildering variety of them.” It was conceived of as charged or uncharged, with or without spin; and then you discussed the possibility of a negative proton and again referred, as you had earlier when you were talking about similar problems, to the fact that energies greater than those available at the time were necessary to obtain such particles, but you looked toward the possibility of cosmic rays. How much of a role did this play in the building of the machines? I know I asked that question before in a little different way, but I’m asking it specifically now, because here you’re talking about a specific problem — a negative proton. Was there any attempt either in the motivation of an individual or the public justification of it to say that “we’re going to build this machine because with it we can find a negative proton”?
No. Not at that time, because the energy you need to make anti-matter — anti-protons — is 6 bev. And none of the projects even talked about at that time was as high as that. And besides the mere fact of discovering the negative proton or anti-neutrons, for that matter, wasn’t an exciting step in physics in the sense that you could by that time quite confidently predict that it must exist. Rather it would be a good thing to verify. In fact, people said about that time, or a little later when it became a real possibility, that one shouldn’t give the Nobel Prize to whoever discovers the negative proton because it’s so obvious that it must exist; but if somebody proves that it doesn’t exist, he ought to get the Nobel Prize. That was the sort of attitude. It was clear that things were getting more interesting the higher the energy and that you wanted to make progress; but what exactly you would do, one didn’t think out in too great detail. One point, in fact, is that the tradition in building machines was that you first built the machine and built the beam and then you started thinking about the experiments you would do with it. Well, to some extent it’s necessary, because you don’t quite know what your beams are like and what their penetrating power and so on is. You can’t predict all that, and so any operation you build has to be modified once you’ve studied your beam. But I think, for example, both with the CERN and with the Brookhaven high-energy accelerators (the sort of 23 Bev ones), they were taken by surprise by finishing them more easily and more quickly than they had expected, and therefore didn’t have any gear ready. An exception was the machine at the Rutherford Laboratory in England. Well, this was, of course, easier, because that was built at energy where there were already machines available; and therefore you could plan more. But there they made plans to have the deflecting magnets and detectors and everything ready simultaneously with the machine—more or less as soon as the beam was there and had been surveyed and so on and they could get going on real physics. But that’s exceptional.
I have an idea of some of the things I’d like to cover now, and we can decide whether you want to go on.
Talking about planning — planning for what you do when you complete the basic machine — I’d like to talk about another kind of plan: two types actually. During the war, when the war was drawing to a close, do you recall your own thoughts and discussions with others about what would one have to do. Now, we did discuss the administrative things and how you were going to build up a group. I’m talking here about the problems in physics, the things that you felt would be the fundamental, important problems that you would like to get back to, and you’d like to work on once you were free of these other responsibilities. That’s one kind of planning. I know that in Los Alamos, for example, there were classes and seminars, refresher courses, for people to start thinking about picking up their work.
Yes, but those were on a more elementary level. I mean those would not be involved with what I would regard myself as the problems to work on.
But, for example, at one time McMillan was scheming for machines, and there were others, I’m sure, who were worrying about their own futures.
I don’t recall that I was very concerned in Los Alamos with what I was going to do next, partly because I had the feeling I had missed out on things that had been done in the meantime, and the first thing was to get acquainted with them; partly also because I had by then got firmly into the habit of working with students, graduate students; and so the mechanism was most of the time that I would find out what students were around and then I would look at the students and see what might be a suitable subject on which to work with them. It was that way around very often.
And so your post-war planning took the form of thinking about how to develop students and then people in the department?
Yes. Of course, that depended on knowing that there would be some problems around that would be worth studying. If you had asked me then (I don’t recall asking myself that question consciously, but it must have been in the back of my mind), I think I would have said, “There are lots of problems in nuclear physics which are open, which are a continuation of what we had worried about before; they need finishing and developing; and there will no doubt be new experiments. Maybe there are some new experiments already which I don’t know about. Let’s be guided by them.” But also there are the intriguing fundamental problems about relativistic quantum theory and the infinities of quantum electrodynamics which were bothering us and stopping progress. There was the old problem I had, which I mentioned yesterday, about postulating particles of finite size and whether one could make that theory consistent. And that, in fact, was one of the first problems I put a student on when I got back to Birmingham.
How soon after the war did you get back?
I arrived back … Well, I left the United States in December of ‘45 or maybe it was the 1st of January. And I then had first of all a personal problem because I had an offer of a chair in Cambridge — in fact, one also at Oxford, but I had already essentially turned that down — but Cambridge I took seriously. I turned the one at Oxford down not because I didn’t want to go to Oxford, but because for this particular chair the set-up wasn’t right. I’ve forgotten what the conditions were. But Cambridge I thought was an interesting possibility. After all, it was still the center of nuclear physics and had certainly a much better supply of students. But I felt I couldn’t decide that. I mean I knew that while I was in America, but I couldn’t decide until I had got back and could see how things were at Birmingham and discuss what the possibilities of development were and also look at Cambridge. But, on the other hand, it was immediately urgent to make a decision, because we had nowhere to live. We had given up our house. Coming back with the children, we had to find schools for them and all that kind of thing. So this was stuck until we knew whether we wanted to be in Birmingham or in Cambridge. And therefore on my return I parked my family with my brother, who lives in England, and gave myself two weeks to settle the question. Of course, on the last day of that period I still didn’t know, and then I said, “Well, I must settle it today.” The device by which I finally settled it was to take a vote with myself. I took a large sheet of paper and divided it in half — wrote on one side of it all the arguments I could think for staying in Birmingham and on the other side all the arguments for going to Cambridge. And when I had completed that exercise, I found that one side of the sheet was almost full and the other had a few writings on it. I realized, of course, that this depended very much on how you defined things — that you could call something one argument or decide to call it five different arguments — so I could have made it look differently. But anyway writing it down that way convinced me that I really wanted to stay, and so I stayed.
Yes. And so that was that, and then I had to get organized with teaching again and administrative questions, though it was not really a question of heavy teaching duties. Well, that was a term (quarter) when I really was teaching hard, because there was then one lecturer in the department. This was still a joint department of mathematics. He was a pure mathematician but was doing a lot of applied mathematics teaching — mainly for engineers and chemists and so on. He had been taken on for that purpose. But he had a personal row with people, including my colleague Watson, and as a result of that he had a nervous breakdown and pulled out of teaching duties. Now, one doesn’t ever like to be in such a position of depending on somebody that it’s a disaster if somebody pulls out. We said, “All right, that’s too bad — we’ll cope.” And the only way of coping was for me, since these were courses in the middle, which were continuing and would have to be picked up in the middle, to say, “All right, I’ll take on these courses.” And so for that quarter I found myself lecturing 13 hours a week.
That was your reward.
So I didn’t get very much else done at that time. So it was really the autumn of that year before we got going properly.
This was the end of ‘45 and ‘46.
That’s right. I mean it was the new academic year beginning in ‘46 when I pulled my department together. Meanwhile, I had advertised these four research fellowships and filled them, some with quite senior people whose presence was a great help in getting going; and we got a number of research students — I think we started out with seven research students. We had one new lecturer — one lecturer had continued during the war in my place — and so we were now a sizeable group, and it stayed that way.
Was there any coordinated planning throughout on the development of post-war physics — I mean other than government planning for uses and control of atomic energy?
Only as we got resources. There was the question that everybody wanted to build equipment and machines, and there was a government committee to deal with the allocation of funds; and at least the technical side of that was, in fact, a committee of all the nuclear physicists, who were discussing each other’s plans and vetting them and deciding how to allocate resources — which would have been very embarrassing if funds had been limited at that time, because it would have forced people to say, “You shouldn’t build your machine. I should build mine.” But, in fact, at that time there was enough money for the equipment that people’s imaginations ran to. This worked all right.
There wasn’t any grant system. There was, of course, but this was outside of the grant system — the DSIR grant system was not involved here. This was a fund for nuclear physics?
I think that was done through the DSIR.
But not on a grant basis where someone or some institution would apply for a grant?
No, no, it was that way, but since the amounts of money for nuclear physics were vastly greater than anything else, they made this a special section; and the applications came to a committee. The DSIR, in any case, works with technical committees of university scientists. This essentially was one of them.
One of the papers that you wrote in December of ‘46 — that at least you read and was later published in the Proceedings of the Physical Society — was specifically on what experiments were needed in fundamental physics. This seemed to be a pretty clear program.
Well, I mean this was a challenge. I tried to. I don’t think it was a particularly inspired talk. I don’t think it was very successful in setting out what we might want to do.
Was this a consequence of or was it related in any way to this DSIR committee, which had in effect asked the same question?
No, because they, I think, were too wise to tie the design of machines to particular planned experiments. But it was, in fact, rather more the nuclear physicists themselves who knew that they would have machines in due course and obviously wanted to think as clearly as they could about what was the most important thing to try and do with them. That was the origin. In other words, it was entirely on the scientific level, not government at all.
You mentioned earlier the other dimension of your post-war work, and that is the essential responsibility in determining how the atomic events of the war years would affect the world. Did this start in any way in the States in Los Alamos?
In a slight way. We did not really think very much about this mainly for two reasons. One was that we had the feeling that it wasn’t our job; that there were other people for whom we had more respect at the time than I would now feel in the light of what one has learned (I mean not respect as persons, but in their judgment and ability to make the right decisions), and so one felt simply that our job was to show what was technically possible and at the same time make quite clear what the implications and consequences are, and then other people who know more about the world can judge what is the right course of action. I wouldn’t now divide it up in that neat way. The other thing was that we were very conscious of the fact that sitting in Los Alamos or sitting in the United States we were foreigners, and it wasn’t our business to get involved in arguments with the United States government. If we had any matters with some political content to argue about, it should be through the British government, to whom, however, we didn’t have any very convenient access at the time. What I do remember is that I happened to be in Washington for a meeting when there was a visit from I think the Prime Minister — it must have been Attlee — accompanied by Sir John Anderson, who was the minister responsible for atomic energy. This is interesting. He was actually the Chancellor of the Exchequer (responsible for finance), but he had earlier had a position in which he was responsible for DSIR and therefore as such responsible for atomic energy; and he was very interested in this, partly because he was a scientist by training and as a young man had written a Ph.D. thesis in Leipzig under Ostwald dealing with the chemistry of uranium. So he kept this as a responsibility. And it was clear that in these discussions in Washington (this was after the end of the war — I suppose October-November of ‘45) there were various ideas around about what one should do in the way of control of atomic energy and so on. A group of us, of which I think I was the main one, decided we ought to make our views known to our government, for what use it might be in those discussions, through Anderson; and I remember I sent a report to Anderson, of which I have no idea whether it reached him or made any impression. But that was beginning to think about what would be a reasonable post-war world.
This was for steps that the British government might take. It had very little to do with the controversy here at the time — on the May-Johnson bill or the McMahon bill?
That controversy was entirely an internal American affair which had nothing to do with us.
But it turned out it affected you.
Right. But I mean it wasn’t for us to try and influence that, particularly as we could see that our American colleagues were very active in that in what seemed to be the right direction. Well, it was something for the British government to consider, obviously not for individual action but for discussion particularly with the United States government.
Chadwick was still in the country at the time, too, right?
So he would have been not necessarily in agreement but involved in these kinds of discussions.
Well, I think I showed him the document. Whether he was in agreement or not, I don’t know. I mean he didn’t violently oppose it. Otherwise I would remember it. But whether he was enthusiastic, I don’t know.
This was the origin of your concern. How did it carry over once you returned home?
Well, it resulted in first of all setting up an organization called the Atomic Scientists’ Association, which was conceived of as a counterpart to the Federation of American Scientists.
Were you the first president of that?
You were president at one time — acting president in ‘46.
Yes. I think, if I remember correctly, Massey was the first president, and I was probably acting president while he was abroad or something.
Has the origin of that been described anywhere? And, if not, it might be good to talk about it.
I can’t remember. It probably has been. The trouble is there I would probably want some documents to remind myself of what happened because it’s a long time ago.
That’s just why I ask, because I have no information on it. I’m not prepared to ask any questions except: what happened?
Well, what happened was that this did not get very far. One reason was there wasn’t anything immediately to be done, other than giving popular lectures, which we all did all the time with or without the organization. But the issue of international control: well, we could say, “We’re for international control.” Everybody was for it in England, so this wasn’t much to argue about. And clearly it was more for the United States government than for the British government to work towards that. We had the impression that the British government also favored international control if at all possible. They probably did — all successive British governments at the time. So this was not much of an issue. I think the American Federation had the experience of a very live practical issue over the McMahon Act or the May Johnson bill, which gave them something to get their teeth into. We didn’t have the corresponding thing to that. We had a rather mild thing. There was also legislation in England — Atomic Energy Act — which was on the whole inoffensive but contained a secrecy clause to which we objected very strongly, because it was sloppy in definition and would in fact have made the publication of any nuclear physics work in any university laboratory illegal. Now, when we raised that, we were told that this would be handled reasonably. For example, there was a clause there, which often is written into British legislation — that nobody should be prosecuted under this act without the authority of the director of public prosecutions, which means it can’t be done irresponsibly. But we still didn’t like the situation where everybody was breaking the law all the time and was staying out of jail or trouble only by the good grace of government officials who understood what he was trying to do. This didn’t seem satisfactory. And we made some contact with members of Parliament who were interested. Both Oliphant and I were involved in that — and other people. And we made a terrible mistake there, because one particular member of Parliament seemed to make that a special issue and seemed to be quite on our side, and so we relied entirely on that; and in fact the day before the debate in the house on this act, he was very excited and said he felt so strongly about that that he was willing to vote against the government, which for a British member of Parliament is a very serious step to take. In fact, he didn’t; and really his speech for our case was very feeble, and really he was not the right person to help us. Actually, perhaps partly as a result of all this agitation, the whole thing petered out, because there was a clause that the minister responsible can make an order exempting various activities from the operation of the secrecy clause; and, in fact, almost as soon as the law took effect, the minister made an order which for all practical purposes completely cancelled out the secrecy clause. I mean where the secrecy clause was so broadly worded in such a way that it exempted practically everything — made the whole thing useless. So from the academic point of view, there are now no objections except the very academic one that this is just an order in council which any minister of any government could revoke at any time and come back to the original position; but I think it’s been completely forgotten by now. So there was not an issue to fight.
I was thinking, too, of some letters. In 1950, there was a joint meeting of the Atomic Scientists’ Association and the Federation of American Scientists at Oxford — the meeting attended by Higinbottom and Allison and Placzek, Thomson, Cockcroft and Kowarski. There’s a reference to it in your letter to Kramers that I haven’t studied yet. Well, anyway, this is an invitation from you to Kramers about this meeting: “This was planned in the first place as a discussion between our group and representatives of FAS. We feel, however, that it would help us to have some eminent people from other countries present.” I’m just trying to think of the specific agenda here. “The meeting will be quite informal. It is our hope to avoid as far as possible an examination of the past history of international control and of what might have been done differently, but to concentrate on what, realistically, we may still hope to do, and what line to take in the face of increasing limitations on the free flow of information (this is the point that you were talking about) that are arising from so many sides. No doubt one of the points of view that we shall consider seriously is that recently expressed by Bohr.” Well, anyway, what do you recall of that meeting and whether it had any effect?
I don’t recall much and certainly it had not much of an effect. I mean discussions about international control were very academic in those days, of course, because it was fairly clear that there was not much hope of persuading the Soviet government to agree to any scheme that would have been acceptable to other people. It was also, I think, beginning to be clear that even given that, it might not be so easy to get the United States Congress to agree to a very generous scheme. But that didn’t even arise. So one came down to a discussion of what might be features of a possible scheme, but without very much hope of implementing it.
How long did the British Association of Atomic Scientists last? You called it the Atomic Scientists’ Association. How long did it last?
Oh, I can’t remember the dates. I would guess it was finally wound up some time in the ‘50s, maybe ‘56, perhaps a little earlier. When I get home I could look that up, but I don’t know it.
It will be of interest to just frame it out, I think, unless in the meantime I find a reference to this that covers it completely.
Well, we published a little journal, which I think was called the Journal of the Atomic Scientists’ Association, and that should be still in some libraries, and the last issue would announce that the thing was being closed down.
That’s an historian’s problem. [The British Atomic Scientists News was published 1947-53; the Atomic Scientists Journal 1953-56.] There are a few other things. I think we can save that for next time. I think we’re pretty much coming to the conclusion certainly of this period. I would like to talk next time about the transition — if you see it as such — to larger scale physics in terms of funding and people coming in and problems and so forth — not only in terms of the sociological phenomenon of it, but in terms of physics problems — also to give some thought to the development in electrodynamics after the war, some of which you were interested in. I have a paper of yours on it. Now, those are the kinds of things. I don’t know how much time that will take us, but we ought to stop now anyway.