J. Robert Schrieffer - Session I

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.

We encourage researchers to utilize the full-text search on this page to navigate our oral histories or to use our catalog to locate oral history interviews by keyword.

Please contact [email protected] with any feedback.

ORAL HISTORIES
Image not available
Interviewed by
Joan Warnow and Robert Williams
Interview date
Location
University of Pennsylvania, Philadelphia, Pennsylvania
Usage Information and Disclaimer
Disclaimer text

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.

Preferred citation

In footnotes or endnotes please cite AIP interviews like this:

Interview of J. Robert Schrieffer by Joan Warnow and Robert Williams on 1974 September 26,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
www.aip.org/history-programs/niels-bohr-library/oral-histories/4864-1

For multiple citations, "AIP" is the preferred abbreviation for the location.

 

Abstract

Childhood and high school education; undergraduate eduction at Massachusetts Institute of Technology, Bachelor's thesis with John Slater on energy level spacings in the multiple structure of transition metal atoms; graduate education at Urbana, Illinois, first paper under John Bardeen on the problem of transport of electrons bound to surfaces in semiconductors (Bardeen, David Pines); doctoral thesis on superconductivity, theoretical issues relevant to it; Leon Cooper's contributions, field theory, the bound state; Bardeen wins the Nobel Prize, emotional letdowns related to slow results of research; Stevens Conference on the many-body problem and American Physical Society Meeting, 1957; application of the Tomonaga variational technique, work on it with Cooper and Bardeen, problems with the second order phase transition, Bardeen's solution of the wave function; refinements of the new theory of superconductivity; feelings about working with Bardeen and Cooper; reactions of the scientific community to the new theory (Niels Bohr, Norman Ramsey); views on scientific creativity; the square dance analogy of the B-C-S theory; the Nobel Prize, 1972; American and Soviet competition for solution of superconductivity; objections to the theory based on gauge invariance properties; aftermath of discovery and Nobel Prize. Also prominently mentioned are: Jane Bardeen, John Bardeen, Nikolay N. Bogoliubov, Bohr family, Keith Allan Brueckner, Eli Burstein, Butler, Leon Cooper, Richard Phillips Feynman, Dave Frisch, Frölich, Ernest Guillemin, Douglas Rayner Hartree, Werner Heisenberg, David Hilbert, George F. Koster, Fritz London, Francis Eugene Low, Arkadii Beinusovich Migdal, David Pines, Léon Rosenfeld, Blat Schatloff, Ann Schrieffer, Frederick Seitz, Charles Slichter, Gregor Wentzel; Institute for Theoretical Physics (Copenhagen), and Niels Bohr Institutet.

Transcript

Warnow:

Very superficially, we know that you were born on May 31, 1931, in Oak Park, Illinois, and you moved to Florida somewhere along the line.

Schrieffer:

Right, that was in 1947.

Warnow:

Let's see, at that point you would have been in junior high school.

Schrieffer:

Right. I was a junior in high school. That is, not in junior high school, but I was a beginning junior in high school. So I had two years of high school in Manhasset, New York, and that's where I lived in fact from 1940 through '47. My father, who'd been in pharmaceutical drugs and cosmetics, and had his own company through the thirties, changed his business association to the old company he was with before the thirties, and that was based in New York. Hence we moved to New York. For health reasons he retired at the age of 47 and went into the citrus business. Ergo, I left what was a darned good high school in Manhasset, New York, and moved to Florida, where I continued high school in a little school that had 28 in the senior class, and basically had very limited science, mathematics, etc. It was quite a change for me, I must say.

Warnow:

It means that you had some science input, though, before you went to Florida, so you could continue.

Schrieffer:

OK, I had a general science course, as I remember, beginning as a freshman in high school. I remember the algebra teacher also was very good, and I had a course in geometry. But then going to Florida, there was no solid geometry or trig taught, and the physics course was essentially a non-course, So I was given freedom to study whatever I liked.

Warnow:

In one of the articles you wrote, you mentioned reading the sciences on your own down in Florida.

Schrieffer:

Yes.

Warnow:

The principal got involved.

Schrieffer:

That's right. The principal was a graduate of Georgia Tech, and he got involved, largely because he said, "Well, look, why don't you take the book and you do math on your own?"

And he looked the problems over, and every few weeks to a month, we got together. It worked out nicely.

Also, the high school physics teacher, Jim Riley, was an extremely warm, wonderful fellow who said, "Why don't you go to the lab and study out of a physics book on your own?" So I got the MIT freshman physics book and started reading through that, and did some experiments. Mr. Riley would come in and help me and discuss things on and off. Chemistry was a little more standard. There was a beginning chemistry course. So it seemed to me that what happened then, I started out a little bit to study on my own and find my way through textual material at an earlier age than many children have the opportunity to do so.

Warnow:

You were lucky to have their encouragement, too.

Schrieffer:

Yes. Oh, that was just great.

Warnow:

When did you get involved in electronics on your own?

Schrieffer:

Let's see now, the ham radio started — let's see if I can recall. That was about 1944, I would say. I was babysitting, and I went to the home of a CBS radio announcer who happened to be a radio amateur, and looking for something to keep myself busy, I found a RADIO AMATEUR'S HANDBOOK, opened it up and started on page 1. I kept coming back to babysit and gradually the husband would stay after he came home, and sit with the sitter, explaining things to me. Then I started building radio transmitters and got involved in that way.

Warnow:

What made you go to MIT? What were the circumstances?

Schrieffer:

Our neighbors, the Obies, had two boys, Andrew and Warren, and both of them were at MIT. One was in business. The other was in architecture, as I recall.

Warnow:

These were your neighbors in Florida?

Schrieffer:

This was in Manhasset, New York, before I left for Florida. So already I had decided that I would like to go to MIT, and I thought that was a great place. I was interested in electrical engineering because I was doing ham radio at the time.

When I went to Florida, I continued thinking about MIT.

Warnow:

Somewhere along the line at MIT you switched to physics. When was that?

Schrieffer:

That was at the end of the sophomore year, and that occurred for several reasons. First of all, I started to realize that electrical engineering and ham radio were clearly different. I took a course in DC network theory and I said, over my dead body. I didn't see the exciting part of EE at that point. I did have, however, some great physics courses. The first one I took, let's see, that was Arthur Kip, who at that time was at MIT. Arthur Kip then went to Berkeley somewhat later. But he taught the 801, 802 sequence I took. It was a very large classroom, but he was beautifully prepared. It was just so clear. We used Sears' text in mechanics, and that was great.

Then the sophomore year, I had Dave Frisch, who taught e1ectricity and magnetism out of Frank's book, and again that was really a great course. He would come in in his galoshes, and looking sort of bedraggled, but he really communicated the excitement of physics, and by that time we'd moved along a bit.

Warnow:

Did you keep any notebooks, by the way, in those years?

Schrieffer:

I probably did. They're all in sort of spiral bound notebooks. I came across some while looking through my attic recently. I know there are some of them in Florida in my parents' attic also, so those things may well be available.

Warnow:

That would be fun to see. Now, you were going on — your second year?

Schrieffer:

OK, now at the end of the second year, I shifted from EE to physics. I must say, a very important person in this whole development was Ernest Guillemin, who was, I guess, one of the most distinguished network theorists in the country. And the Guillemins, Professor and Mrs. Guillemin, were one of the rare faculty couples at MIT who went out of their way to invite students to their home. Mrs. Guillemin would cook up chicken for 50 young people, and her daughter, a very lovely girl, played party hostess as well. In any event, Prof. Guillemin and I started chatting a couple of times. I was taking the network theory course from him. I said I was very interested in physics, and I was thinking, was that a reasonable direction to go?

He said, "Hell, I came from physics and I think it's terrific. I think you ought to consider transferring over to physics."

So he was the one who gave me the encouragement, and I did it.

Warnow:

What other teachers — I wonder if you didn't come into contact with Slater?

Schrieffer:

That was after I had come to physics, and — now, let's see. I believe in the junior year I took Slater's course, what was it called, atomic and molecular physics, something along that line. I really found that a remarkable course in the sense that he was able to get through some rather complicated material in a step by step, logical, constructive way. The lectures were beautifully structured, and often things were made so clear that when one went home, you found that they weren't totally as clear as they'd appeared at the moment. But I was very much excited by that course.

Warnow:

What about your bachelor's thesis?

Schrieffer:

Well, bachelor's thesis time came around, and I looked around for a topic. I had been excited about the course with Slater and I was potentially interested in doing theoretical physics at that point. I'd been in the library and read a number of books on modern physics — I remember one book that particularly struck me was Born's ATOMIC PHYSICS. I discovered that there were appendices to that book, and all the excitement happened to be in Appendix 1 through N rather than in the rest of the text, and I read those appendices avidly. The same thing was true about Slater's book; all the excitement tended to be in the appendices I found.

So I came to Professor Slater and asked him whether I might do a bachelor's thesis, because he had asked if any people in the class were interested in writing a thesis with him to come and see him.

I had a friend, Al Switendick, who also decided to do a bachelor's thesis with Slater so the two of us worked together, on a similar project, in the group. That was my first contact with an ongoing research effort. Slater's group had an interesting arrangement of offices around a long conference room where there was a main table that people came for coffee every morning at 9:30. There was an esprit de corps in the group which was quite remarkable.

It was a team effort on some problems where Professor Slater really decided the direction of the effort', but each person had his own problem. Every day people got together over coffee and discussed their contributions, in a very informal way, except that people were often quite nervous, since Professor Slater isn't one who takes nonsense lightly. I recall in a number of cases Slater would say, "Well, what did you do yesterday? Did you look at the three center integrals?"

And someone would say, "I'm sorry, I went to the movies last night," and everybody would quickly stir their coffee because that wasn't the right thing to do.

So there was a certain sense of discipline which I appreciated was essential to doing good science.

Another impression was that there was clearly a pecking order in that group, of people who were at various levels of advancement through their training, but also different levels of accomplishment quite independent of the level of education, so that the quality factor became also very clear.

Warnow:

You also had PhD candidates in that group.

Schrieffer:

They were mainly PhD candidates, and just the two undergraduate people. So we saw sort of the PhD's in the making, plus there were post-docs present. It was a marvelously heterogeneous group of background interests, etc., but there was a clear structure.

 

Williams:

You actually worked on a genuine research project? This is the flavor I get.

Schrieffer:

To an extent, yes. It was research. Professor Slater had suggested that we look at some problems in the multiplet structure of transition metal atoms, and look at the energy level spacings. This type of work had been done for many years, but this particular part of the periodic table hadn't been worked out, and Slater thought that it would be a useful thing to do, just to learn and to contribute a little. So we did that. Switendick and I worked on different aspects of the same problem, and we produced some results which showed that in fact Slater's original calculations, dating back to 1929 or '30, were more reliable than some of the spectroscopists had believed. The discrepancies between his calculations and experiment, were due to the assignments of the atomic levels, and the calculations we did seemed to clarify that issue. So he was delighted.

Warnow:

Was it most of your senior year you were in and out of those sessions?

Schrieffer:

We started in the middle of the fall, as I recall, and continued on, finishing up the thesis at the end of the senior year. So it was a good six or seven months we were exposed to this research atmosphere.

One of the great excitements, I recall, was working late in the evening, and George F. Koster coming around and helping us a bit behind the scenes. He was in some sense second in command in the large group, and he was a great human being and a really great scientist, in my view. He took time to work with us, help us with our problems, and talk about the excitement he had in his own research. He was, I think, an assistant professor at the time.

So we saw all different types, and that spectrum was very important, because it gave me a little bit of a model in my own mind of how creativity takes place, and how this structured process of going from the known to the unknown in finite steps is very important. One can't make the intuitive leap immediately, and there's a heck of a lot of hard tough work that goes in between.

We saw people doing numerical calculations on and on. But yet you get a feeling that it was going somewhere. So it was sort of a microcosm of science that one saw in a fairly well-defined group, and my feeling was, it was a beautiful example.

One concern I had, however, was that the individual creativity in that structured society wasn't quite what I wanted. I felt that I wanted to have a graduate education which perhaps allowed more flexibility for individual creativity than was going on in the group.

That somewhat guided my decision about Urbana and what to do in graduate study.

Warnow:

All right, when you were going to Urbana, it was with the idea of studying with Bardeen?

Schrieffer:

It was, although that was not my original intent at all. In fact, I didn't intend to study solid state physics. I had decided that solid state physics, while interesting, and I enjoyed working with Professor Slater, that wasn't my bag, and the thing that excited me was a book that I found by Rosenfeld on —

Warnow:

Leon Rosenfeld?

Schrieffer:

Yes, and it was on nuclear forces. It's one of the great early books in that field. The concept of isotopic spin had been around a short time and excited me. Thus I felt that nuclear physics was where I was going.

So I'd gotten a Fulbright Fellowship to go off to England, to work with Rosenfeld and Blackett. It turned out that the Korean War was going at that time, it was the 1952-53 period, and my father had fallen a bit ill, and the family and I felt that it was best that I wasn't over in Europe at that time since it might make some difficulties in getting home quickly.

So I looked into other possibilities. I had heard outstanding things about Professor Bardeen and I had also heard about superconductivity, so I thought that Urbana sounded like an interesting possibility. I applied to graduate school there, to Stanford and a few other spots, and I received a letter back from Urbana saying in substance that we will admit you and furthermore if you like you can begin research immediately with Professor Bardeen.

So that was all I needed, and I immediately decided to do so.

Warnow:

You had no contact directly with Bardeen?

Schrieffer:

No. None whatsoever.

 

Williams:

Would you just back up a little bit? You say that after your original idea to go, to sort of follow Rosenfeld, you then somewhere in there — in other words, you had the family suggestion, but then you heard about superconductivity and Bardeen, how did that —

Schrieffer:

No, I'm sorry. I should make a little clearer that I had heard about these things certainly before deciding on nuclear physics. I knew about the transistor, and Bardeen's name was certainly one of the luminaries in the field. His name was often mentioned around the coffee table. I must say, however, that there was an impression around that group that somehow Bardeen's contributions were more in the direction of applied science rather than basic science. And yet, what little I had looked up in the library and things I had heard about superconductivity, indicated that much of his work was very basic. Superconductivity was very exciting to me at the time, I remember, but I hadn't thought in detail about it. I was aware of it. It was in the air, as it were. I knew that Professor Bardeen had gone to Illinois, and that if I wanted to do solid state physics, I was convinced that I wanted to do it with someone who was very creative and wasn't a highly calculationally oriented person. Professor Slater suggested I go off to Upsala to work with the group there, and they had been developing some techniques for molecular structure calculations, and yet I had decided I just didn't want to do calculations. I felt the thing that excited me was the conceptual part of physics, sort of the sketching out of new directions rather than the hard slogging work that was necessary in more evolved areas. I'd seen the latter, and I realized it was necessary, but that wasn't where I wanted to fit into the game.

Warnow:

So meanwhile in your application, you must have mentioned that you wanted to work with Bardeen, and Bardeen was told of your application.

 

Williams:

Well, either he or the referees; I don't know how they do that.

Schrieffer:

I said that I wanted to do solid state theory, and I don't remember if there was a question that said "Do you want to work with a particular person?" I rather doubt it because that was highly irregular arrangement for a first year student and I didn't expect it. In fact, I had expected to be a teaching assistant, as I recall, and was greatly surprised that anyone would be allowed to immediately start research. But Professor Bardeen wanted to have another student, and I guess he said at some point that he had seen some recommendations, perhaps from Professor Slater, and thought that this might be a reasonable candidate. I don't know quite how he came to his decision. It was easy for me.

 

Williams:

There was some communication somehow. Joan was just noticing that it didn't come from you.

Schrieffer:

Well, it didn't — you know, I was shocked to get this letter that I was being invited to work with Bardeen as a beginning student. So that was quite a surprise.

Warnow:

So you arrived in the fall of '53, beginning of the school year, and you had your first contact with Bardeen.

Schrieffer:

Right, yes.

Warnow:

What was your initial impression?

Schrieffer:

Well, my initial impression was, I didn't recognize who he was. It was a comedy in the sense that my mother and father came up to Urbana, we drove up from Florida and had a little visit, got settled and my father and I came to the lab. This was before classes began, and I was trying to find Professor Bardeen's office. I was directed there, and when I arrived the door was open but no one was in. I passed a gentleman on the steps three times in a row, and it happened to be Professor Bardeen.

Finally we got put together, and he was very gracious but very quiet. I had already heard that he was not a person who was overly open and aggressive in any sense in a relationship. So I didn't expect someone who would be effusive. We simply chatted quietly and he suggested that I start on a problem. He was then heading off to the Kyoto Conference on Theoretical Physics — the proceedings I guess are up here on my shelf — in any event, he was heading off to a conference for a few weeks in Japan, and to get me started he had suggested a problem, and gave me some relevant literature references. In fact the main reference was to Klaus Fuchs's article on transport in thin metal films, and he suggested I work on the problem of transport on semiconductor surfaces, which he was very much interested in at the time. He had an experimental laboratory which he set up, in addition to his theoretical work, and it was quite a major lab with perhaps 8 or 10 students in there doing experiments along with some post docs.

So I recall that I went to the library and found Klaus Fuchs's article, and I was so impressed by the fact that he had written down an equation which described how the electron distribution was influenced by the electric field and scattering events.

When I came back and met with Professor Bardeen, I said, "This is really a marvelous article.”

He said, “Yes, fine.”

I said, “Well, I want to explain this to you,” because I'd felt I knew all of physics, more or less — you know, I came from MIT, and what else is there to know except a little polishing up? So I said, “You see, this is how this goes.”

I recall that he said, "Well, that's very standard," and I was absolutely crushed. He said, "That's called the Boltzmann equation, and you know, it came a long time before Fuchs."

So I quickly realized that I'd better start changing my point of view and become a real student again. So that year I worked on the surface transport problem, and I think it was along March or April, I published the first paper on the problem of transport of electrons, bound to surfaces in semiconductors.

Warnow:

OK. Meanwhile was Bardeen mostly working with David Pines?

Schrieffer:

Yes. That's right. He was very actively involved with David on the problem of the electron-phonon interaction in metals, including the screening problem, i.e. the electron-electron interactions. Earlier, Fröelich had worked out the problem of the effective interaction between electrons, through the exchange of phonons. Bardeen had also worked on that problem. David Pines had worked on the electronic plasma problem, i.e. the electron-electron interactions and they were working on putting the two programs together into a screened electron-phonon-electron interaction.

They came out with a manuscript which Professor Bardeen gave me, and it was my first introduction to this whole complex of ideas of many-body physics, not done by the Slater point of view, but more the field theory point of view, although field theory was still a very mystical sort of thing at that time, at least to solid state physicists.

Warnow:

When was that?

Schrieffer:

Pines' manuscript with Bardeen? That was — I'd have to look it up explicitly, but it was about '54, spring of '54, I think. The paper of Bardeen and Pines, I can look it up — it was in the PHYSICAL REVIEW.

 

Williams:

It's in a reprint, not the paper but the reference. I think you refer to it in your paper, do you know?

Schrieffer:

Well, I'm sure we do, yes, it's in there.

 

Williams:

Bardeen and Pines, PHYS REV 99, 1140, 1955.

Schrieffer:

Fall of '54, yes, early spring or fall of '54. I recall struggling over that, and again that was quite exciting, because it was a totally different point of view than I was used to.

Warnow:

How much had you worked with Bardeen at the time? Had you had lots of contact with him?

Schrieffer:

Yes. We met at least weekly and often more frequently. He was always available, and I felt that there was never any problem of coming and chatting with him. He was sympathetic, always suggesting how to get around difficulties. I would say that the communication problem was a real problem in the beginning. The difficulty simply was that John would sit for five or ten minutes and not say anything. To me at first that was embarrassing, and sometimes I would get up and start to leave the room, feeling that the conversation had ended, and open the door. Then John would continue speaking as if I were in the chair, perhaps, and I would sit down again. This would go on and on and gradually it was clear that it was time to leave.

But I began again to appreciate how one should sit and think. What he said was often in cryptic little messages which were phrased in terms of the way he thought about problems, and he didn't expand on this unless one probed. Every time you said, “I don't understand,” then he would explain more, but he wouldn't volunteer more of an explanation if you looked puzzled, typically.

I think most people who became his students were able to communicate, and sometimes it wasn't always very easy. But he changed also, during the time I was a graduate student, insofar as his outward relationships with people, both scientifically and socially, it seemed to me. He became much more outgoing, in relationships — perhaps because in a university, that's part of the name of the game, while, perhaps at Bell Labs it wasn't. Also, the fact that he had to lecture and wanted to lecture put him in a position where he had to communicate in a continuous fashion, rather than some gaps and silence. Whatever it was, I don't quite know.

 

Williams:

Can you think of a time when it happened? Like he got the prize I think while you were —

Schrieffer:

— 1956, yes. Late in '56.

 

Williams:

Was it a continuous thing, did you notice?

Schrieffer:

It's hard to notice when someone changes, when you're with them every day or few days, but there certainly was —

Warnow:

It was more than just your getting used to him?

Schrieffer:

I think so, because other people mentioned it also. I don't know how to explain it, but that seems to be the fact.

Warnow:

Did you write a master's thesis?

Schrieffer:

I never officially wrote a master's thesis, but I remember I got a postcard saying “Do you want a master's degree or not?” and I checked “Yes” and then I got another card saying “Appear at an office at a certain time on a certain day.” I thought that I would be formally presented a diploma so I got all dressed up and came to a window, and a girl chewing gum handed me my degree, so that was the master's experience.

[NB: missing section of conversation of about 20-30 seconds]

Roughly, missing section was:

 

Williams:

What about Bardeen's Handbuch der Physik article on superconductivity? How were you involved?

Schrieffer:

There were to be two articles, for the new Handbuch on superconductivity; Bernie Serin was to write the experimental portion, and John was to write the theoretical portion.

[rejoining taped conversation]

My involvement must have been during my second year I was at Urbana, that was '54, '55. I'd completed the work in the transistor lab, because after I'd done the theoretical work on the semiconductor surface, John sent me to the lab for a period to test out the theory. Then I came back and said I wanted to do another problem, a theoretical problem. He'd suggested a problem of superconductivity in thin films, namely, the critical field of thin films.

During that time I had known he'd been writing the Handbuch article, and my contribution was really that of helping him do the proofreading, and that was it. It was a wonderful experience for me to learn what was going on in the field, because I had to read carefully enough to pick up misprints, and it also gave me a chance to really think and study over things.

He had kindly quoted at some point, I believe, a little calculation I'd done on the thin films, but basically I played the role of a proofreader.

 

Williams:

Was there any thought at that time that you had thought you'd like superconductivity?

Schrieffer:

Yes, that was certainly very much in my mind, and I had really hoped all the time when I went there that I would get to work on superconductivity, although I didn't discuss that with him in any detail. It was clear that he had worked on that problem. I'd known that, as I said, as an undergraduate, and while I was interested in semiconductors, I wasn't, you know, totally enamored with that direction. What I really wanted to work on basically was superconductivity.

Warnow:

OK, so now you've already been given your list of ten topics?

Schrieffer:

No, no. No.

Warnow:

'54, you have your master's degree. June '54.

Schrieffer:

That's right. Right.

Then '54, '55 was while he was writing, and I was doing a little proofreading. I had been working on this little problem in superconductivity not the fundamental theory of the phenomenon. The problem was sort of an applied problem in transport in films and critical field in thin films. It was using a sort of phenomenalogical theory, and it wasn't the basic theory of superconductivity.

At that time, I was still taking graduate courses, and I took a course in statistical mechanics by Francis Low, that I really liked. It was a great course. Only later did I find out from David Pines that Francis Low, for that semester, never stayed at parties later than 10 o'clock because he was going back to read the next ten pages of Tolman, supposedly – Francis never had a course in stat mech himself, so David claimed — it was really a beautiful course, clear, concise, etc.

I chatted with Francis a number of times; I felt he was a confidante, and I respected him enormously as a scientist and a human being, so I would often go and chat with him about various things.

It was, as I remember, exactly the date I don't recall, but I came and asked Bardeen for a real thesis problem, and I'm sure he had this in mind, and he said, "Come in and see me."

Exactly how the discussion went, I don't quite recall, but he traditionally kept in his bottom drawer a list of problems, and I remember, there were ten problems on this particular list, and the tenth was superconductivity. He said, "Why don't you think about it?"

There were a number of problems, one of which was the Kondo effect. That was many years before its time, but he thought the resistance minimum in metals was a very interesting and important problem to look at. There were a number of problems that varied all the way from sort of applied mathematical physics in solid state, all the way on to the superconductivity theory.

 

Williams:

Were they ordered in any particular way? Like superconductivity sounds like the most difficult of the ones you mentioned.

Schrieffer:

As I remember, that was the last one. I don't recall the order, but certainly in some sense that was an ascending order of difficulty. I think that they simply formed a list that he kept adding to, and maybe he thought the field had evolved to the point where a first principles theory of superconductivity was a potential thesis problem, and had added that to the list of nine that he had had before. I don't know quite how it happened.

 

In any event, I went and chatted with Francis Low about this, because I felt that I could chat with him. He was very open, and I asked him what he thought about it, should I try this?

He, I recall, asked “How old are you?” and I told him, and he said, “Well, you can waste a year of your life and see how it goes.”

So I took that as reasonable evidence that he felt there might be some chance of doing something.

 

Williams:

Was it the risk that was concerning you? Why was there hesitation? You said you were continuing to think about superconductivity?

Schrieffer:

Well, it was the major decision in my life, and if I chose a problem that was clearly impossible to approach, let alone solve, then I wasn't sure that I wanted to get involved. And it was such a big decision, I wanted to chat with someone. I didn't feel that I was well enough endowed to really make the decision. Certainly John didn't tell me what to choose. He said, “Here's a list and what do you think?”

I don't know if he's ever really told a student “Do this” for a thesis. He certainly gives some ideas of what looks reasonable.

But at that point, superconductivity, as you well know, had been worked on for a long time, and workers had met a lot of failures, so it certainly wasn't anything like a sure thinq. The question I had in my mind was: Was there something that I might do and I might contribute? The other question was whether Professor Bardeen himself would be concentrating on this area, so that we would have in some sense a useful interaction, rather than my being alone on the problem.

It was clear that having written the REVIEW article, in a way he was gearing up for another major attack on the problem, and that's sort of the feeling I had; that he was really interested in working vigorously on the problem.

So I felt that I was interested in the problem, that there was a chance of making progress and that Bardeen was going to actively work on it; thus it seemed the time was ripe.

Warnow:

How close have we gotten to the date of this conversation? We know that Cooper arrived in the fall of '55.

Schrieffer:

That's right.

Warnow:

This conversation took place before that?

Schrieffer:

It took place before that, I'm quite sure. It was probably in the late spring of '55. Exactly the date again, maybe John can remember, as I say I'm not too great on dates, but it was about that time because I had been working on superconductivity in thin film, and it was a natural evolution. Since Francis Low was there, it was probably in the late spring that I went to him, because he may well have been away for the summer. It seems to me it was in the late spring.

 

Williams:

Just to go back a little, in the previous interaction with Professor Bardeen in connection with the article that appeared in the Handbuch, did you have any conversations with him throughout that? Or was it more that you just sort of studied it on your own as he turned it out?

Schrieffer:

It was really the latter. I wasn't clearly up to the point that I could discuss in any sensible way the details of what was involved. I read it, and I learned the field, I would say, by reading through it. I did take occasion for example to read Schoenberg's book, which was the bible, at that time, on superconductivity, and I read some other papers. I remember, I passed my German exam largely by taking Schafroth's paper which was written in German, in Helvetica Physica Acta and translated the full article. So I had done a little reading in the field. But basically I served no more, no less than as someone who is correcting things, and in one small area I contributed a small thing.

 

Williams:

The one thing I wanted to tell you [Joan] about, and this might be a topic for later discussions, that particular article has an annotated bibliography. In other words, there are value judgments given to the references. And that's a specially good hint for you and the history type people. That's seldom done, as you know, but there for whatever reason, the major references — Schoenberg, for example, I remember is one of the ones very high on the list, and some little comments are made about it. I just wanted to say that.

Warnow:

When you finished talking to Low he said, “If you want to risk a year, you're young enough.” Then you had to think about whether Bardeen would get involved again.

Schrieffer:

It was actually more or less clear at that time that he was getting involved. He'd finished the HANDBUCH article, and was clear that that was something that he was very much interested in pursuing again. The main question, I would say, at least from my point of view, is how I would begin to do anything with what were clearly limited abilities, in the sense that I wasn't a professional, I hadn't worked in the field, I was a student, and many people had failed. The mathematical techniques around didn't seem like they were at all appropriate for this problem, and even how to begin on the problem was totally unclear. The work of, really of Fröhlich essentially used perturbation theory, and that didn't work out, and Bardeen had used a variational version of perturbation theory, and that hadn't worked out either; even though both approaches produced an isotopic effect, they didn't produce superconductivity. So, in a way, these theories were precursors of a complete theory, in that they got one of the effects correct, but the main phenomena of superconductivity just weren't explained.

The Bardeen-Pines work again extended the Fröhlich work to a much more realistic situation, where you could say you understood how the electrons, phonons and electron-electron interactions all combined, but then the whole problem was how to take this effective interaction and resolve the degeneracy of the electron gas, and there weren't any techniques known to even begin to address that problem.

The whole area of field theory, I recall, was very exciting to me. Wentzel’s book, for example, PARTICLES AND FIELDS, or whatever it is called, I saw as a junior at MIT. One of my lab instructors, Dave Finkelstein, had this book in the lab one day, reading on the side from his graduate course, and I asked him what it was. He gave me a small lecture, saying that psi isn't a wave function but a field operator, and that was so unbelievable I decided that was fun. So I also had this field theory excitement in the back of my mind, and that coupled me to the nuclear physics thing from before. But field theory really hadn't been introduced at all into solid state physics, basically, at that time. Fröhlich's work was one of the first contributions which used in some sense the field concept, in more than a rudimentary way. He employed a canonical transformation using field theory representation.

 

Williams:

When you say field theory, how do you mean it?

Schrieffer:

Second quantization, let me put it that way.

Williams:

OK, so like the earlier work, where they used some of the simplest ideas of non-interacting field theory, like the phonon, being the photon of, you know —

Schrieffer:

— those concepts were around for a long long time.

Williams:

That was done way earlier, and that sort of —

Schrieffer:

— absolutely.

Williams:

But it's interacting field theory.

Schrieffer:

Interacting field theory, writing down a Hamiltonian in terms of second quantization, and then using that language to proceed, say in perturbation language or what have you. The whole Green's function scheme hadn't been transcribed yet from the field theoretic version of quantum electrodynamics, over to many body problems. The techniques simply weren't available. Again, I'm only perhaps reflecting what John has said after the fact. I think that's what motivated Bardeen to asking Frank Yang whether he might have a person he could recommend to bring some of the techniques from the field theory of particle physics, over to solid state, because it was clear that solid state physics was very light on muscle when it came to many body problems. No one knew how to treat them, and that's how, to a certain extent, Leon got involved, as I understand it.

Warnow:

Well, I still don't see how we linked up, how we really got you onto the project. You must have gone back and talked to Bardeen about this, and finally settled down.

Schrieffer:

Yes. I said that I'd like to try it. He said, "OK, fine."

Warnow:

Did he talk with you about the idea of getting someone like Cooper in on this?

Schrieffer:

Not that I recall. I certainly wasn't in a position to discuss who he got in or didn't get in, but he did mention that Cooper was coming along, and that he had a field theory background and that this might be useful, something like that.

Warnow:

You knew this at the time you made the decision. You knew that he would be working with you?

Schrieffer:

I don't think so.

Williams:

After you said to Bardeen, "I'd like to try it," in your previous problems, he always sort of guided you or made suggestions, “let's go and try this.” What did he do then? Did he sketch out a program?

Schrieffer:

OK, I see we're on two different issues at the moment, maybe —

Williams:

Oh, sorry, I —(crosstalk)—

Warnow:

— trying to get at this problem of, you saw the problem, the complexity, perhaps impossibility, many other people had failed. And then you said, "Yes, I'll try it." How did you come to say, "Yes, I'll try it."

Schrieffer:

Basically because I was excited and interested in it, and I was pretty much convinced that Bardeen was going to work on the problem again vigorously, and I had enormous faith that he had a very good shooting chance at succeeding, anything he tried. Everything he'd done, clearly he had solved. He'd worked on this problem a long while, and it seemed that from at least my limited vision of the field, that there were potentialities coming along, the whole field theory business as I mentioned, seemed exciting, and there might be some chance of making progress. The Bardeen-Pines work was again building groundwork up towards where we could try the problem.

Warnow:

And in a way if he thought it was worth another try, maybe you should think it was worth another try.

Schrieffer:

Yes, you know — exactly how I made the decision, I think probably was on the grounds that it looked like the most exciting thing, the one I wanted to do. In a way, I was so predisposed to making that decision, it was only a question, was there a major reason not to make it? I would put it in that direction. And I asked Francis Low, and he said, "Give it a try," so I said, "Great."

It was in a way something I really wanted to do, and I wanted to make sure I wasn't doing a crazy thing, and when Francis said it was reasonable I said, "Great, I'll try it."

Warnow:

So now, you [Williams] wanted to know the guidance thing?

Williams:

I'm curious, what happened in terms of getting a more specific idea from Bardeen, after you had given him the decision, "Yes, I want to go down this road," then did he sketch a program for you? I was getting back to your point, did Professor Schrieffer know that Cooper had been requested? and all that.

Schrieffer:

Let's see. Exactly when I knew that Cooper was coming, I must say I don't remember, but I don't think it had anything directly to do with my decision — it was another part of the program that was being put together, and it was nice to have another member of the group. But Brueckner had been working on the theory of nuclear matter, and Professor Bardeen had read several of Brueckner's articles that had come out in THE PHYSICAL REVIEW I believe in 1954-55, using so called T matrix methods. These were a second quantization approach, first of all, and secondly, it was for an interacting fermion system, in that case nucleons rather than electrons, but it was one of the only techniques available to treat an interacting Fermi system. So he said, "Here's some new work that may or may not be relevant to the problem. We don't have much to go on." I'm sort of paraphrasing at least the ideas he had. "So why don't you read these articles and see what you think about it, and try and apply it to this problem? We've got an effective interaction, the Fröhlich-Bardeen-Pines interaction, and maybe you can apply some of the techniques that Brueckner has suggested for the nuclear problem to the electron problem in metals."

That was the idea. And so the year of 1955-56, I worked on that problem, and first of all, it took me a while to understand what the Brueckner approach was, how it worked out. It was phrased in scattering language, and I remember working on great sheets of paper, because the equations tended to be so enormous that I couldn't keep them all on a single sheet of paper. It was a bit of a clumsy formalism, and I found it very difficult to comprehend in a physical way. I kept trying to understand what's going on physically, what's going on?

All during this time, Professor Bardeen kept emphasizing some very basic concepts. In the discussion we had in the window of Bardeen's home in the fall of 1972, for the Swedish film, Leon Cooper mentioned some of those. One of these guiding concepts was — there must be an energy gap in the electronic excitation spectrum. In the HANDBUCH article Bardeen had written down the electrodynamic properties of a superconductor based on the assumption, rather than a microscopic theory, that there was an energy gap in the excitation spectrum. He deduced from this assumption, a generalization of the London questions, which turned out to look very much like the equation that Pippard had proposed for the non-local electrodynamics of a superconductor, a couple of years before.

Pippard had given a number of arguments in favor of this formula that he had written down more or less out of his head, one of which was the relation of the work which Chambers had done on the anomalous skin effect, others about energies of normal superboundaries, etc., but there were a number of physical ideas that went into Pippard's work. Bardeen's work on the energy gap model reproduced this expression from a more fundamental point of view, not from first principles but, if you like, second principles.

There are a number of arguments that there should be a gap, I think going all the way back to the London brothers. Fritz London's paper, said that there is a rigidity of the superconducting wave function with respect to magnetic perturbations. Bardeen kept saying in effect, "This system has to be rigid. There has to be a gap." But, in agreement with London, he stressed that this condensation of electrons which leads to the gap, has to be a condensation in momentum space. It's not in coordinate space in any sense. And this message was clearly stated in London's book of 1950, but the concept goes way back to the 1930's in London's work.

Williams:

When did you first see London's book? I mean, when you came onto the project did Bardeen say, this is something to read, or you'd already read it?

Schrieffer:

No, I hadn't read it, no. As I recall, the first book I read was Schoenberg, but then I got London's book and read that more or less cover to cover. Particularly in the last chapter, as I recall, there are some very provocative suggestions, how one should proceed to make a microscopic theory, and Bardeen felt very strongly that those suggestions were exactly the direction one should go. How you transcribe that into mathematical reality in detailed theory was far from clear, but it had to be along that line, so that's what we were focusing on.

Warnow:

Let's talk for a minute about who was on the project. Was David Pines still working on these problems too?

Schrieffer:

Now, let's see — David left for Princeton, I have to check the year exactly but I believe it was at the end of the semester, the year '54-'55, so when we started working on this I believe is just when he left, if I'm not mistaken. I know that he was sort of in touch with Bardeen, or I believe he was, back and forth, but he didn't have any direct contributions or interactions with our effort, at least that I'm aware of, at the time. The main direct interaction was his work with Bardeen, and that set up the problem as an electron-electron problem rather than electron-phonon and electron-electron. So that was very important, but I don't think David directly played a role in this part of the history.

Williams:

Just to refresh my mind a little bit, wasn't the Bardeen-Pines paper basically about a normal metal? They did it in a normal metal?

Schrieffer:

Yes, they did it in a normal metal, but Bardeen had this concept, which as in most cases his concepts turn out to be right, that the effective interaction is really between excitations in a normal metal, and that those interactions could lead to superconductivity. Thus, while you work out the effective interaction between normal electrons, normal and super electrons are electrons, after all, and the concept I think was that, much as in the Landau theory, once you've got the interactions between the quasi-particles in one state of matter, then you can predict how that system will evolve to another state. You base the concept on the Landau Fermi liquid ideas, although that wasn't made really explicit, because the Landau theory wasn't current if you like, in the sense people in the western world didn't know about it at that time.

In effect, Bardeen kept emphasizing “We have to think of the normal state, as like a free electron gas, with the excitations being in one to one correspondence with those of a free electron gas — the interactions between the actual electrons are enormous, but vie have to think in terms of effective excitations which include these strong interactions insofar as they enter the normal states. The residual interactions between these effective particles will then lead to superconductivity.” How these residual interactions lead to superconductivity is the problem. I think his point of view was that the effective interaction which coupled these excitations was in essence the interaction that he and David worked out, which as I mentioned is a generalization of that originally worked out by Fröhlich.

How those interactions ultimately lead to superconductivity is another part of the story. It's like saying the nucleon-nucleon interactions, arising from pion exchange, can be worked out. Once you do that, however, somebody has to work out the theory of nuclear structure with that effective interaction. I think that was the level of problem Bardeen had in mind.

You might say that the nucleon-nucleon-interaction in the nucleus is different than it is in free space, because the other nucleons perturb the basic nucleon-nucleon force.

The work of Migdal on the coupled electron-phonon system wasn't certainly known to us at the time. It occurred in 1956, and the importance of three body phonon forces hadn't really been investigated. I think the main contribution that Migdal had at that time, in retrospect, was to point out that three body forces were not of great importance, although that Migdal story is another interesting one. We'll get to that eventually.

Warnow:

So you didn't know about the Landau theory and of course you didn't know about the Migdal theory.

Schrieffer:

Right. Although I'm not quite sure if John had heard anything about the Landau theory. Certainly the general concepts of the philosophy were certainly there, but the words, as I recall, he didn't use. As far as I know, these ideas were independently arrived at by Bardeen.

Williams:

— he didn't refer you to the Landau paper, for example?

Schrieffer:

No. I didn't know of its existence at the time.

Warnow:

All right, this may be a good time to stop, because we're right about at introducing Cooper into the story.

Schrieffer:

Good, OK, sounds great. OK, now — let's see, if you'd like to stretch and get some air, we can go to the truck, or if you'd like to chat, I can pick up something for you.

Warnow:

I think that's a good idea.

Schrieffer:

OK, Yes, it gives the reference in the front. It's JETP 3, 920-925, 1957, in English translation.

Warnow:

OK, so that was not available to Bardeen at the time we were speaking of.

Williams:

Right, that specific paper. I have the impression that that concept, that Landau has there or uses in a sort of devastating way, what we would call maybe the quasi-particle concept, had kind of grown up earlier in the Russian school of solid state, and some of these articles in Uspekhi — they talk about the history of the Russian solid state people, and just the crude impression I have is that perhaps back quite far, the forties, they grew up thinking this 'day for certain reasons. But that's something to check.

Warnow:

It would be interesting to know if Bardeen read any of the articles that may have appeared earlier, with gems or germs of these ideas in them.

Schrieffer:

Yes. The concept of elementary excitations in normal metals was certainly around very strongly whenever we chatted with Bardeen, and we had to use the basis states, based on these, I don't think you'd call them quasi-particles, but effective electrons which were in one to one correspondence with the electrons of a free gas. And this was certainly the message right from the very beginning, and I'm sure that goes back to his ideas in the 1950-51 paper on superconductivity and probably those even before the wartime. But exactly when the connection came with the formal Landau theory, I'm not clear, but that wasn't the set of words that were used at the time.

William:

And you just got it all. In other words, he didn't say, "This is the idea, here's some papers — "?

Schrieffer:

No, no, not at all, because by and large there weren't many papers to sit down and work from at the time, in the direction we were trying to go, and again, the London ideas were very important, but there wasn't much London theory. There was the book, and that was all phenomenology, but the important contributions were not 90 percent of that book, but it was the last chapter that was quite important, to say how one should go about the microscopic theory.

Williams:

That leads us back into the earlier comment, where you were stressing guiding concepts that you had gotten sort of from Bardeen in the program, and the two that you did mention were the last part of London's book, and also the energy gap. Were there any others in that time?

Schrieffer:

Yes. One very important concept that he stressed was the Pippard coherence length, and he had right from the beginning a rough idea of how the condensation should occur, partly from London, saying it was in momentum space, it wasn't in direct space. That was the first thing he mentioned. Secondly, Pippard's coherence length which he knew was about a micron, 10,000 angstroms, led to certain conclusions about the nature of this state, which he made abundantly clear, Bardeen made very, very clear. And it was roughly as follows: that the condensation energy, if you look at it numerically, roughly is that corresponding to electrons within an energy kTc of the Fermi energy, each having their energies lowered by about kTc. So a part in 104 of the electrons each have their energies lowered by about a part in 104 of their energy. So he had a feeling that the condensation was a condensation in momentum space, associated only with those electrons very close to the Fermi surface, and that was one thing.

However, the number of electrons which are within kTc of the Fermi surface is very large for a large system. The main question here he was addressing was, suppose that you say a superconducting correlation length is given by Pippard's length ξ. How many superconducting electrons, happen to fall inside a volume of ξ cubed? If you do the simple calculation which he pointed out, you say you get about a million. So it's not that the condensed phase corresponds in some sense to condensed clusters forming a dilute gas in which you have little clumps of this condensate floating widely separated in space. Rather, it was clear that a highly correlated, very dense gas of condensed electrons was involved. Even though it's only one part on 104 of all the electrons, the density of electrons is so enormously great in a metal that in a volume ξ cubed there is still an enormous number of condensed electrons. So it was clear that the problem was in no sense a two, or four, or six electron problem locally, many similar problems spread out over space, and then a weak interaction between these clusters. Rather the problem corresponded to a highly concentrated situation with strong electron overlap. So again it comes back to the London idea: it's not a condensation in coordinate space, like molecules of electrons, but it was quite the reverse. It is more like a condensation in momentum space.

Williams:

For example, a counter-example to that would be the ordinary association of hydrogen atoms condensing in real space.

Schrieffer:

Yes, right, into hydrogen molecules, and those molecules form a hydrogen gas.

William:

— they weakly interact and it wasn't anything at all like that.

Schrieffer:

Again, he stressed that again and again, that is the wrong direction to go.

Now, in that regard, the J. M. Blatt, M. R. Schafroth and S. T. Butler attempts at making a theory of superconductivity, while their general theory was not necessarily restricted to the dilute gas view, they used that point of view in carrying out the mathematics.

Bardeen kept emphasizing “The phase coherence” — words that were used again and again – “is absolutely essential. There has to be a phase coherent condensed state, in momentum space, which gives long range order over a large distance in space, with a typical size of the correlations being like Pippard said.”

That and the energy gap I guess were the basic things.

Williams:

Yes. If I look over this list, I've tried to jot down some of them, I see four distinct lumps of stuff. I see the effect of electrons and their effective interaction. That's what we perhaps today would call quasi-particle viewpoint, but not in those terms then. I see an energy gap idea. I see Pippard coherence length, and then your comments about the pairs being heavily overlapping.

Schrieffer:

Of course, we didn't know they were pairs. Much later, or somewhat later, we came to that point of view.

Williams:

Oh, all right, then I should say, it was a very dense situation, many body correlations were important.

Schrieffer:

That's right.

Williams:

But all the other ideas you mention — the phase coherence and so forth, thinking about them real quick, they were all in London?

Schrieffer:

That's right.

Williams:

So it would be really the four. Then in terms of breaking it into discrete contributions, there are really four. Effective electrons, energy gap, Pippard coherence length implies many body system, and London part, the latter part, which suggests what to do, is that — ?

Schrieffer:

— yes, with condensation in momentum space, and the long range phase coherence of the wave function. Somehow these weren't of course, at least as I recall, and recollection's not always perfect, but as I recall, they weren't enumerated as distinct facts. It was part of the general folklore and things we kept chatting about, and it was the developing of a picture gradually that came out, but these were things that kept getting mentioned and hit, and we'd have to look in that direction. That's how we should approach it. Warno1v: Did he sometimes pull you back and say, “No, that's the wrong approach because I think it has to be such and such?”

Schrieffer:

Certainly that happened, yes. For example — well, we'll get into this a bit later, but this whole question, after Leon had looked at the pairs, and then we thought we'd make the pairs into bosons, then he realized that that in fact, was not the case, I think it was John who pointed out finally, “Wait a second, we're off on the wrong track here, because those things aren't bosons,” and then the whole question opened up again.

So he was certainly very positive in suggesting directions to go, but he wasn't hesitant in saying, "I don't think that's a very good idea," or "That's too complicated a way." So he played a very important role in educating, guiding, leading, all those things.

Warnow:

That was going to be my next question. Let's talk about Cooper's coming, I suppose, and then about the interaction between you and Cooper, Cooper and Bardeen, Bardeen and you. So let's take, Cooper arrives.

Schrieffer:

Right. Cooper arrives, I guess was September of '55. Let's see, it was early after his arrival that we were discussing, could we use his particular expertise to do something to help out in the problem? So Bardeen asked him to give some informal seminars or discussions about what field theory was all about, as practiced in electrodynamics, and how it might apply to superconductivity or many body problems in solids. And I recall that he gave some very clear lectures about Feynman diagrams, and forward and backward in time diagrams corresponding to virtual excitations of the Fermi gas, but he was not at all enthusiastic that these techniques would in any way begin to address the fundamental problems of forming the superconducting ground state, resolving the high degeneracy of the electron gas.

He said that the methods v1ere quite useful for the electrodynamics problem, where the perturbation was weak and there were perhaps not essential singularities which totally altered the character of the state, but there were by and large perturbative techniques rather than techniques which would lead to phase transitions or a qualitative change of the nature of the matter in question.

He thought that, while the field theory techniques might not be directly applicable, maybe some of the concepts might be applicable, and that seemed to come somewhat to a halt. That whole point of view didn't look very exciting or promising.

So the rationale for why Leon came, as I recall, was somewhat to put this input of the field theory in, and yet his message when he came was that, "I don't think we've got it, boys, this isn't the panacea that one might hope for."

He then suggested some of the Schwinger work on the functional derivative methods in field theory, and I started going off and reading a little bit of (Julian) Schwinger, which I found very tough sledding, and he said, those were not necessarily perturbative methods, the functional derivative methods, and perhaps there would be some way of utilizing these coupled Green's function equations to pursue the problem.

And in a way, one can say that's what Gorkov did several years after we finished our own work, and had one proceeded along that line directly, perhaps we would have formulated a theory in a very different direction. But we didn't, first of all, understand the Schwinger papers. They were formidable. They were in a different language. And it was very difficult to understand how you put physics into those equations, because we weren't familiar with the technique. But, the concept, the anschaulich view of what really was going on, as I've said before, is what we were totally consumed with. The mathematics wasn't there, and we didn't want to use totally unfamiliar mathematics, when we didn't know what the physics was, so it was a doubly difficult situation. So that was dropped.

And the field theory, I continued to be interested in it, and it was sort of fun and games but it wasn't something that helped us really.

Williams:

Could I ask a question in this sense. In terms of your own training as a graduate student or even as an undergraduate student, what did you know about field theory, let's say, second quantization and beyond, that style?

Schrieffer:

Well, basically what I knew came from reading say the Brueckner papers on the T matrix, which Bardeen had set me to. I'd taken a course in quantum electrodynamics from Francis Low, that was the other course I took from Francis, again an excellent course, but at that point, the whole renormalization in quantum electrodynamics was just getting set down in book form, and that was the book Jauch and Rohrlich. I remember all that semester he said, "Jauch and Rohrlich will soon be here, don't worry, your questions will all be answered." And he continued on. But I learned what I knew about Feynman diagrams and field theoretic techniques partially from that course.

Then there was the Migdal and Galitski paper, which again was very interesting, in that it transcribed much of the field theoretic methods to the many body problem, but again we didn't think that that was a direction that would help us particularly. That was later. I'm not even quite sure when that paper came. That's sort of where many body theory got developed.

Williams:

What I was really curious about is, of course, looking at the end, going out of time sequence, we'll see that field theory was very helpful.

Schrieffer:

Right.

Williams:

And I was just curious about where you learned about it and when, and whether you knew a great deal about it as a graduate student at that time.

Schrieffer:

I certainly didn't know a great deal. Not at all. I would say, the main place I learned about field theory, Green's function technique, was after I returned to Urbana in 1959, and David Pines and I started working. We worked through the Migdal-Galitski papers and we began educating ourselves as to what good these things were, etc. So we were very much novices. The transfer, if you like, from one field to another hadn't occurred. The techniques were simply not there.

Williams:

OK, I think that's a very crucial point, from a historical point of view, just when this kind of key ingredient got in, and from where. That's unanswered right now but I just wanted to know just what your background is. You did know then the Feynman diagrams technique, and of course, his two famous PHYS REV papers, I guess they were the textbook, probably, if you didn't have a textbook.

Schrieffer:

We didn't go back to the literature. We copied what Francis Low wrote on the board. You know, God was speaking and we listened and we did homework problems. There wasn't really much of a textbook around at the time. I had read the book by Wentzel, and I found that very interesting, but again, it wasn't really relevant to what we were trying to do. I couldn't see how it applied.

Warnow:

Isn't the next step Cooper?

Schrieffer:

Right, I was hoping we'd come in that direction.

Williams:

All right — just a little comment. What we've done to get ready is, we've made a chronology of all the published papers that we know by either Bardeen, Cooper or Schrieffer in this certain time period which we're trying to concentrate on. In addition we've scribbled down a few little things that were written by any one of the three of you. So this kind of helps us, we don't have to worry about the dates or anything now, but the next thing that happened after Cooper arrived, is, like almost a year went by. In the fall of '56, Cooper sent the one pair letter in and it was received 9/21/56 by PHYS REV. I guess that was the Cooper one pair. How did that happen?

Schrieffer:

OK, again, best source for how that happened is Leon. All I can do is to say my view of how that happened, and it may or may not be totally faithful.

Warnow:

Yes, you may not have gone enough into how you interacted, the three of you. I mean, we got the informal colloquia, but —

Schrieffer:

Right, OK, fine. Well, after that, really, I'd been, as I mentioned, working on the Brueckner theory, and Leon took very seriously the energy gap aspect, and focused on that, and really tried to look at what seems to be a very abstract point of view, and that is: let me study matrices, Hamiltonian matrix particularly, but he said, “Just let me look at matrices, see what class of matrices have the following property: the lowest eigenvalue of the matrix is split from the rest of the eigenvalues by a finite amount, and in particular that splitting,” he wanted to say, “was independent of the size of the matrix,” in the sense physically that if you had an energy gap in a system which is one cubic centimeter, that the gap didn’t change as you went from one to two to ten to a million cubic centimeters.

So that was in a way a mathematical game, but he thought that there might be some physics in there, if he could understand what intrinsic property of the Hamiltonian matrix would lead to that result. So it was reverse reasoning. It's somehow like I mentioned to Bardeen and — you know, you try to work backwards and forwards.

So he found that a certain class of matrices had this property. I remember, Fröhlich visited, and let's see now, Leon had told him about such matrices with all plus l's all over the matrix, and said that there was a finite gap, and I think Fröhlich felt that that was interesting but apparently he had also looked at such matrices somewhere in the past five years or what have you. He said, he knew those matrices plus a lot of others and he felt that really wasn't relevant to the problem, that that was a blind alley. And I remember, there was a little bit unhappiness on Leon's part, and it may have been that that visit occurred — I think it occurred after the, sort of, the pair problem had been studied and he found the bound state, but again Leon has to speak to that, but I remember there was that one contact. And Leon felt that this direction was a good one, that one should look for this gap in the spectrum of matrices.

As a particular case of that situation, he concentrated on the two electrons above the quiescent Fermi sea and found the bound state, and I sort of remember the day that he was very excited. He found the bound state, and the bound state was there regardless how weak was the coupling, and —

You see, I should mention that Bardeen and Cooper shared an office, and that was very important. They could wheel around their chairs and talk to each other continually, even though they probably didn't do it, it was possible.

Williams:

Physically speaking, where were you now?

Schrieffer:

I was at what was called the Institute for Retarded Study, affectionately known — and it was on the third and a half floor of the building. I guess it was the second and a half floor? But it was again, a wonderful format, like Slater's group. There were people all together in one large area. There were field theorists, there were nuclear physicists, there were all theorists came there, and if somehow you were able to move to the Institute for Retarded Study, you had made it. That was considered the greatest. And when there was a place open, a desk open, everyone would sort of scramble around to see who could get in there. But that was a melting pot, and I learned probably as much field theory from people like Gardenhaus and oh, I guess, Jerry Franklin, and people who were working for Chew and Low, as I did out of the courses. We were continually talking about our problems. Dan Mattis was there, Walter Harrison, etc.

But I would come down and say something to John, then Leon was there, and we'd get together in two or three way discussion, or if Leon was out, John and I would chat. But it was a sort of a round robin, where I think John and Leon probably didn't talk too much more than I chatted, but they were always together, and when they had a question it would come up and they would discuss it. So that was a very happy relationship, which largely came about because there weren't enough offices for everyone. They were just squeezed in. That worked out pretty well.

Warnow:

How many PhD candidates had a desk in that Institute?

Schrieffer:

Oh, something like nine, I would say. About that. It was a very low ceiling, and there was a great blackboard, and there were always two or three people at the blackboard, arguing and discussing, so that was fun. They were all students. The post-docs were on the fourth floor and the students were on the third and a half floor.

Warnow:

Then you were saying that you remembered the day when Cooper found this.

Schrieffer:

Yes, he found it, and he was obviously very excited, and we were all quite excited, saying, “Gee, here's something qualitative, it looks very promising, without a doubt.”

Then Leon I remember said, “Lookit, you're doing this Brueckner theory, you should find the same thing in your equations.”

I'd found that I couldn't find stable solutions, and after he had shown me how he'd gotten this in his two electron problem, I looked back and found that was the difficulty, why I couldn't solve the Brueckner problem. So I felt, gee, there's a tie here, ok, — Brueckner techniques don't work, but at least they show the same sickness or ill that Leon found in his problem with the bound state. In other words, bound states were trying to be pulled off in the Brueckner theory, but since it was a self-consistent theory, it would never allow for a self-consistent solution, because it's just not within the power of that theory to do it.

Leon worked on the two electron problem, and since he bounded the problem, the instability wasn't there, and the two electrons formed the bound state.

Then Leon wrote up the paper, and I'm not exactly sure again when Fröhlich's visit was. I have a feeling it was after Leon had found this, but that's something that's probably important to find out, for that tie.

Leon also worked out the problem of where the two electrons were interacting with a non-zero total momentum. The sum of the two momenta of the electrons was not zero, but varied continuously up to a large value, and found that the binding energy decreased as the momentum of that pair of electrons increased, relative to the zero momentum of the Fermi Sea.

So we started thinking about how we could make a many body theory which took into account many pairs at the same time. Bardeen's message again was loud and clear –- that this condensation was not one where molecules of electrons were condensed, but they were overlapping. It was very perplexing to us, as I remember, physically, how is it possible for pairs to be bound together, and yet so many pairs to physically overlap in space? Again, I think John kept saying, "Well, it's in momentum space, you shouldn't think about the coordinate space so much. It may not be confusing if you view it in the right language," which was momentum space. And he kept insisting on the picture with momentum space.

Then we had decided, the thing to do was to say, "Let's take the pairs which are most unstable" — that is, in Leon's picture, where you had a pair, it had the greatest binding energy when it didn't drift, it was at rest. And we set down what, I don't know when we called it “the pairing Hamiltonian.” I think the word “pairing” didn't come till much, much later. I think that really came from the Copenhagen group, Bohr and Mottelson and Pines got together, and the word “pairing” was not what we had used. We'd used the words “electron pairs” or “pairs.”

I may be wrong on that.

Williams:

What is the distinction in your mind right now between the word "pairs" and the word "pairing?"

Schrieffer:

I don't think any distinction, other than the fact that people now talk about the pairing concept, the pairing this — and I don't believe we used the word “pairing.” Now, it's a trivial difference, but I'm just trying to be historically accurate. I don't think there is any difference. Right.

So that was a conceptual problem, but what we did, we said, "OK, let's write down the problem where all electrons are treated, but we treat them in the second quantization formalism corresponding to pairs of zero momentum, and try and solve that problem."

We wrote down the Hamiltonian and looked at it, and couldn't make any progress on it. We didn't know how to approach it. Various ideas about variational methods, we thought, we tried all sorts of approximate schemes.

Let's see, one variational method, as I remember, John had was to take a distribution of the pair occupation numbers, with momentum, and make them drop off like a Fermi distribution, with some effective temperature, and use that variational parameter. That didn't seem to work.

So exactly all the techniques we tried, I don't recall. Maybe this is the part where we have to go back and see what sorts of notes are around, and that would be important, undoubtedly. But I recall that was sort of the early fall, and middle fall of 1956. And then I recall, I met John Bardeen on the street one day and as I passed by he smiled and said, “Oh, I just wanted to mention –- I won the Nobel Prize.”

And turned around and walked off.

So that was a real excitement at the time which sort of boosted us up, because I think we were feeling a little bit downtrodden, because things weren't breaking so quickly after Leon's contribution.

Warnow:

Like a losing team.

Williams:

No, it wasn't so much that as I think that when you saw Leon's, as you just said a few minutes ago, you were very excited. You said, "Here's something quantitative, we're almost there," and then, nothing. So it was really a letdown after the initial encouragement. Is that right?

Schrieffer:

I think that's right. Maybe I'm over-emphasizing the letdown, because from my naive point of view as a student, I thought, "Gee, now we can perhaps really make quick progress." And it was only over a relatively short period that this occurred anyway.

But then John went off to Stockholm for the prize, and that was at the beginning of December, December 10th it was on, and we continued working at that time, trying various approximate techniques. Let's see — I don't know quite what else to say about that period. I had, I remember, essentially wanted to give up on the problem, because I felt that the techniques we had at hand just weren't up to trying to solve the problem, and I wanted to start working on ferromagnetism, itinerant ferromagnetism, because the Brueckner technique, I started to apply that quietly on the side, hedging my one year bet — the year had passed — it was very interesting, but it hadn't produced that much. I came to Bardeen…(off tape)

I think I was mentioning that at least I personally was becoming somewhat discouraged at not being able to make significant progress, taking Leon's beautiful result, and making a many-body theory out of it. And the difficulties were multiple. One was that we had this Hamiltonian for the zero momentum pairs, and we weren't able to exactly solve it. It looked simple, and hence we thought we could do something with it. We tried many techniques, variational techniques of all sorts, and things just didn't jell.

I had been working on the Brueckner thing, and I had started to quietly work on ferromagnetism, and I had mentioned to Bardeen that I thought perhaps I would like to change the thesis topic, because I didn't quite see that we were going anywhere.

Warnow:

You must have felt very frustrated at this point.

Schrieffer:

Yes, frustrated, and it was very exciting, but it was very frustrating, needless to say. It wasn't clear what was going to happen, and I thought maybe it would be better, since this was getting to be the middle of my third year, and at that point it wasn't unreasonable to expect to get out in four years, and I had applied for a Fulbright fellowship at that point, to go off to work in Europe —

Warnow:

— again, Blackett?

Schrieffer:

No, at this point, I wanted to go with (Rudolf) Peierls in Birmingham. Then I was going to travel around to the Bohr Institute, and I had some plans for traveling, so I said, you know, here we've got eight months and this thesis — it's difficult at best — so let's try and do something.

I remember, just before John left for Stockholm, he said, “Give it another month or a month and a half, wait till I get back, and keep working, maybe something will happen, and we can discuss it a little later.”

Williams:

You also said that his getting the prize, right in this doldrum period, kind of gave you a rene1c1ed spirit?

Schrieffer:

It's so hard to remember back. We were all so excited and pleased and delighted for John, and that was certainly a bright spot at the time. To have him away was not always the best thing, because you know, he was the father confessor, the font of knowledge and the rest, so it was with mixed feelings.

We also felt we were really hot. It was sort of this mixed feeling. We were really on the trail, and it was sort of almost a schizophrenia, you know — we're going to do it, and, we're not.

John, of course, having a tremendous background of knowing how research goes, had the vision of not months but years or maybe even decades, and I as a student had a view of maybe months. I'm sure Leon had an intermediate view of things.

In any event, we proceeded on, and then there was this meeting at Stevens and the New York meeting, and that was in the middle to — I guess the end of January.

Williams:

Yes, we have here 1/28 and 1/29 in '57, January at the Stevens meeting.

Warnow:

Who went to that?

Schrieffer:

I went to that, and I guess Leon went to that, right. I think John went, but I'm not — no, I guess John didn't go. You'll have to ask John about it. As I remember, at that time, if I don't have my meetings confused, I think that's the one where Feynman spoke. Wait a second — let me not say that, because Feynman I know did speak on the West Coast, at the joint American-Japanese Theoretical Physics Conference on Superconductivity. During this period, I recall, I had the impression that Feynman was working on superconductivity, and that was sort of in the background, and John was concerned, particularly a little bit during this period, that Feynman was, you know, perhaps making great progress. That was particularly after the break came at the end of January, and that was one thing that said, “let's go, go immediately, as fast as we can and get this done.”

Warnow:

Meaning the wave function?

Schrieffer:

Yes, the wave function.

Williams:

You mean that the Cooper letter had appeared, and so like the key clue was out?

Schrieffer:

Oh, no, not at all.

Warnow:

That was much later. You mean —

Williams:

— why don’t you just say what you mean then? I don’t understand.

Schrieffer:

Well, I'm getting ahead of the story. Let me just put it that way.

OK, fine — so, when we went to the meeting – now, I don’t really remember, did Leon give a paper at that meeting?

Williams:

Yes.

Schrieffer:

That was on the pairs, was it? That is, the instability problem.

Williams:

Yes. He gave something which — he gave this paper, and it's almost like the letter. Incidentally, Joan, I didn't copy it at the time, but this came out in '63, so no one has ever referred to this.

Warnow:

Years later —

Williams:

— yes, this is '57. Years later — this was the Stevens Conference, Proceedings and Symposium on the many-body problem at Stevens, held January 28-29. For some reason, I don't understand it, but six years went by. I never really understood the Cooper letter until I read this. At least, I didn't understand it in this way, and he's got this one little section in here, where he does it all over again in Green's functions, but just talks about the form of the Green's functions, and maybe everybody else in the world knew it, but no one had ever shown me it, and it very quickly brought a lot of the physical things meaningful to me. Now, he then spoke about the many electron thing, — but in very general terms, he didn't say “we did it” or, he didn't say, pairing or anything.

Now, the problem with this of course, is, in six years, of course BCS paper was published. Well, and he refers here, in reference 7, to a thing that I think was obviously added, because it was published after the meeting. That's what I meant about the internal evidence.

Schrieffer:

Right. Right. I think probably he tried to put it a little in better perspective, so that the reader of this could hook into the later work, which then would get them into the main stream, so that seems appropriate in retrospect. It wasn't available at the time.

Williams:

Right. But to the people who were there, the impression I have in reading it, leaving out section 4 (the extension to many body), it seemed to be more like a tutorial about what his one pair letter was about. In other words, it was sort of explaining the result. Maybe you people at Illinois, this was well known to you, this way of looking at his one pair paper, but it was ne11s to anybody that I ever talked to in superconductivity. And of course, the fact that it didn't appear in print — of course he gave it as a talk to this assembled group, I gather where most of the experts, I would say, that were geographically, temporally approximate —

Schrieffer:

Brueckner and Yang were there.

Williams:

They have an attendee list in the front, that's what I meant to say before, that I didn't copy it.

Schrieffer:

One of the big discussion items, I remember, at that conference, was whether the Brueckner scheme was producing reasonable numbers for the binding energy of nuclear matter, and Yang got up and wrote on the board, the following:

“The validity of a theory is not necessarily established by the agreement of its results with experiment, at least insofar as only one number is concerned.” Something essentially to that effect.

So there was a lot of hubbub at that meeting about the nuclear matter. The superconductivity, as I recall at that meeting, didn't get an enormous push, because not that much had happened. There was the Cooper idea, but then things had gotten quiet again.

Warnow:

Did you wonder if he had used Green's — in the talk I mean?

Williams:

No, that's pretty clear. From the way it looks here, his reference 6 is to his, Cooper's reference 6 is to Cooper's one pair letter, so if you read the paper here, up until part 4, it just is like an exposition of the same thing but from a different point of view, and I'd say that for people in the audience it would have been a very helpful, different way to look at it, if he hadn't done the calculations himself.

Schrieffer:

Yes, I remember that Leon —

Williams:

— did you hear him give this paper?

Schrieffer:

I'm sure I did, although I must say it didn't make a great impact on me. I remembered it before the fact, but —

Williams:

— I just wonder if you remembered if there was a fuss, or —

Schrieffer:

I don't think so, because it's true, I don't think there's any debate — the debate might have occurred, if there was any, I must say I don't remember in detail, but suggesting how one might go beyond that, to extend it to the many-body problem. But I'm afraid I don't recall.

I remember, Leon did talk about the fact that the Green's function is different for a spectrum which corresponds to a finite density of states at zero energy, and zero density of states, and he emphasized that this leads to the split off state etc.

Williams:

It's a very beautiful way to look at it, which doesn’t –- it just is –- and to me, I, well, it’s always been a lot clearer to me than his original letter, which seemed very contrived. I mean, it’s a very special calculation — here you’re talking — it’s maybe the same thing.

Schrieffer:

To me, I would say, this is contrived, and hides the reality by making it look mathematically elegant and beautiful, and using Green's functions, and that just in a way puts the icing on what is basically what a very beautiful cake without the icing. I think it's much simpler, the simple facts can be taught in freshman first year quantum mechanics without ever hearing of Green's function, or beginning operator algebra or what have you. So in a way, this dresses up in a finer clothing what I think is more transparent, clear and much closer to what finally then gets transcribed into the BCS theory, in his original letter.

Williams:

That's true, because this doesn't go on. I mean, you don't go on from this approach.

Schrieffer:

Maybe another way of saying it, this is a transcription to scattering theorist's language, what he did for the bound state. Maybe it connects better onto the particle theorists. They would like this better.

Warnow:

Well, in any case, the next thing is going to the APS meeting in New York.

Williams:

Let me just give a few dates here from our little date book. Joan has lent us the APS Bulletin — and it turns out that it was a busy week in late January, because there was also, from January 20 to February 2 was the APS meeting in New York City, so it looked like a whole week of meetings.

Schrieffer:

Exactly.

Williams:

I mean, it would be a full week of meetings, the two meetings together. Now, who attended — well, let me just go to the written evidence — Cooper gave this paper apparently at the, which I discovered by accident last night — gave this 10 minute paper at that meeting, apparently — [pointing out the abstract published in the Bulletin].

Schrieffer:

OK. This is interesting, in a sense. I don't recall it. If I were to make any comment, I'd probably be off base, but let me say very informally, I assume what he was talking about here was the low density gas of condensed pairs, because he had talked about that, although as I've said, Bardeen and all of us recognized that that is not a direction which seems to be fruitful. That was the stumbling block. I think one just has to ask Leon what he spoke on that, because I'm pretty sure I didn't hear that paper.

Williams:

OK, well, in my own mind, you know, sort of trying to get a chronology ready — it got me confused when I saw this one. Nobody had ever told me there was this one. It isn't in Cooper's bibliography, by the way, which he sent to you, [Warnow] gave the Nobel Committee. That is not in that.

Schrieffer:

I think basically he was trying to see how he could extend the pair concept, in a continuous way, to the many-body system, by allowing a lot of pairs, but not going back to the fundamental problem of the fact that they overlap, interact, and it's almost a new game, to the real problem.

Williams:

You went to the APS meeting in New York City, right?

Schrieffer:

Yes, and to the Stevens. Well, let's see now – I probably went to some of the sessions at APS, but I'm not completely sure, because I was staying out with friends in Summit, New Jersey and commuting back and forth. I have the impression that I did go to the meeting, because I was in New York, so I must have been — I'm pretty sure I went up to the meetings to hear some sessions.

Williams:

Do you remember hearing a session by Schafroth, Butler and Blatt, or one of those people?

Schrieffer:

No. I don't think so.

Williams:

Did Pauli speak there? I remember Pauli —

Schrieffer:

— Wigner did, Wigner —

It was probably a later one.

Williams:

Just looking — I happened to look through all the papers, titles, last night, and cryogenics, there's chemical equilibrium approach to superconductivity, paper V-8, talks about Brueckner and things of this type, but you probably heard of this in other ways already.

Schrieffer:

Yes. Blatt had come to Illinois, and I believe Schafroth also had come by, so vie were certainly aware of their work, that they were trying to work on these things. Then I mentioned Fröhlich also, didn't I — he was one of those.

Williams:

Well, now, is this the meeting at which you did the subway riding and so forth?

Warnow:

Can we talk about your — you know, how you felt? I'd really like to get you back into how you were feeling, you know, you got on the subway, what it was like, (crosstalk) late at night? early in the morning? you know.

Schrieffer:

Yes. Let's see, these details again often are difficult to recall, but let me see if I can recall. I should put in perspective a little bit how I got led in that direction, because nothing happens spontaneously.

I had been interested always in the nuclear problem, the pion nucleon problem, going back to undergraduate days, and I'd read this Tomonaga paper on the intermediate coupling scheme in pion nucleon structure problem, on the nucleon structure problem.

Also, I had read the paper by Lee, Low and Pines that used the Tomonaga approach to treat the electron-phonon problem for the polaron. That is, there's a single electron interacting with a phonon rather than the electron gas.

So I had that in the back of my mind, and I was trying to figure out a way to treat this gas of non-bosons, as we finally correctly realized, they weren't bosons, and I wanted to use a variational scheme, because there didn't seem to be any other scheme that was appropriate. One had to guess the answer, if you like, and then use some sort of a variational approach.

And somehow, during that couple of days in New York, whether it was at the Stevens part of it or the APS meeting part, it was some time during that week, I started to think about the variational scheme associated with this Tomonaga wave function, which originally came from the pion nucleon problem, and the pair problem. And I'd been trying to construct variational wave functions which didn't require that a given state k was either occupied definitely or unoccupied, but I wanted to have some flexibility, so the electrons could scatter around and lower their energy.

We kept saying we'd have to form the wave function as a coherent super-position of normal state-like configurations, and the question was, how to do that?

So I said, "Well, a Hartree product, the product where you have either k occupied or unoccupied, is never going to lead to an energy lowering."

I said, "Well, lookit, there are so many pairs around that some sort of a statistical approach v1ould be appropriate."

That was sort of floating around in my mind, that there are so many pairs, they're overlapping, some sort of a statistical approach is appropriate. And then the other one was this Tomonaga wave function, all sort of crystallized in saying, “Well, suppose I put an amplitude,” I think I called it the square root of h, that the state is occupied, and a square root of 1 minus h that it's unoccupied. And then let's product these factors overall states k, and that's just what Tomonaga did, for that problem. I said, "Well, at least that allows the electrons to hop, the pairs to hop from state to state, and that seems like a reasonable guess."

So I set that down, and then I looked at it, and I realized that this state didn't conserve the number of electrons. It was a variable number of electrons, and that worried me, I remember, and so I decided, well, what I should do is multiply that wave function by a term involving E to the minus the number of particles, and just like in the grand canonical ensemble in statistical mechanics, sort of extend that idea to the wave function in quantum mechanics. I said, "Gee, I don't know if it's going to work but it seems to me like a reasonable approach, let me try it."

So I guess it was on the subway, I scribbled down the wave function and calculated the beginning of that expectation value, and I realized that the algebra was very simple. I think it was somehow in the afternoon, and that night at this friend's house I worked on it, and the next morning, as I recall, I did the variational calculation to get the gap equation, and I solved the gap equation for the cutoff potential. It was just a few hours' work.

Then I flew back to Urbana, that second night, so I think day after the original guess on the subway. Exactly whether or two days, I can't really remember, but it was —

Warnow:

Were you feeling excited, first of all, as you started this whole approach, did you have a feeling it was falling into place?

Schrieffer:

Well, I sort of did. Again, it's hard to remember in great detail, but I remember I was enormously excited, when I saw I could calculate the inner product. It was analytic, and you could write it down, and you could calculate things through. But the thing that really hit me is, when I saw the ground state energy. Leon had an expression for a single pair, which involved an exponential with a quantity E to the minus 2 over NV, [1] parameter. And when I got the ground state energy out, I found it was of the form E to the minus 1 over NV, [2] times the number of pairs, so it was exponentially lower in energy, and the variational principle says that's better, but whether that's the right answer, I didn't know, but at least it was a qualitative improvement, it seemed, not only a quantitative one. It seemed like a qualitatively different state. It didn't depend upon a superposition of pairs in space, but it was really ordered in momentum space. So it sounded like it had something.

I must say, at the time I was very dubious that this was any real solution to the problem, and it was exciting because it was fun to do, it worked out, and I met Leon then at the Champaign airport apparently he'd come in also from New York. Why we came there at the same time, I don't know, but we did. I showed him this, and he seemed very interested. He said, “Great, looks terrific, let's go and talk to John in the morning.”

So the next morning, we went and chatted with Bardeen, and very quickly, as I recall, he looked at it and he said he thought that there was something really there.

So we chatted around about that for a few hours. I think it was that same day, he was quite convinced that there would be an energy gap in the excitation spectra, because that was just the ground state.

I believe it was perhaps a day or two after that, that — now, let's see. Let me see if I can reconstruct this… let's see if I can get the chronology.

I believe we saw the gap first, or — maybe I can reconstruct, thinking carefully later, but I remember, the first substantive result that came out was the question that — yes, sure we had it, that's right — the — first came the gap, that's right. The excitation spectrum, I believe John showed, again we'll have to talk with Leon and John, but my present memory is that John showed that the gap was exactly the same parameter delta, we called it epsilon zero at that time, that I'd found entering in the ground state energy. It was something I'd called EK and EK was just a parameter that had dimensions of energy, was a function of momentum, but it occurred in the ground state, and I never thought of that as an excitation energy at all.

As I remember, John then had gone home, and it was a day or so — he was first of all convinced there was a gap. Second, he showed that EK was the same thing as occurred in the ground state, and then there was a number for the gap that came out of that.

I think we have to get the three of us together to recall exactly how that happened, but as I remember, the most exciting thing happened when I came in one morning — this was after four or five days. I think it was a Saturday, but I'm not completely sure. And John was then calculating the condensation energy, in terms of the observed critical field — that's known as the thermodynamic relationship — but also, from that, calculating the condensation energy from the energy gap. And there's a relationship between the critical field and the energy gap given by the theory.

Now, the energy gap was being measured experimentally by Tinkham and co-workers, for I believe lead or mercury, we'll have to check which of those it was at the time, but we had the experimental number from the Tinkham group for the gap. We knew the critical field, and the whole problem was converting units, because to get the density of states, it was in milli-Joules per mole degree or something, and I remember John was very upset that he couldn't get the numbers to work out. And we finally got that to work out, and the numbers turned out something like 9 compared to 11 in the appropriate units and we were really overjoyed, and sort of hit the roof. Things looked like pay dirt.

So that was about five, six days after the wave function got written down, that that happened.

Warnow:

And the three of you were interacting very intensively that week.

Schrieffer:

Yes, essentially —

Warnow:

— Leon must have been there at the same time you were there.

Schrieffer:

Yes, absolutely. He was also working on the excitation spectrum. You know, exactly who was doing what during that four or five day period, I don't know. We were all talking and working. But as I recall, I think it was John who got the relationship between the excitation energy EK and that — but you'll have to chat with Leon.

We then decided, John decided I think, all of us decided to break up tasks, because it was clear that there were a number of things that had to be done, and we shouldn't all do everything.

Warnow:

This is where the Feynman —

Schrieffer:

Right, and I remember John saying that he'd heard that Feynman was working on the problem. I know he'd been in Japan working on it, and he'd given a paper at the Japanese-American Theoretical Physics Conference about his things, and he was doing all sorts of complicated field theory, which we didn't know — or at least John and I, I don't think knew much about it — and we were concerned, or John was concerned that he might break the problem from another point of view. So we really wanted to move on it.

So then we worked out the low temperature specific heat. I was working on the thermodynamic properties. Leon was working on the electrodynamic properties, and John was working on all the transport and non-equilibrium properties of sound attenuation and etc.

I guess the — John decided there should be a letter put out, and that letter should…(break)

We tried very hard to get out the second order phase transition, and the language of the theory was kind of tortured at that time. We talked about the virtual pairs, that is the pairs that occurred in the ground state. Then there were excitations, or single particles, I think we called them — I don't think “quasiparticles” word was around, I think they were called single particles. And then there were the real pairs. And the real pairs were somehow pairs of single particles that happened to be in states of opposite momentum and spin. But they were orthogonal to the ground state pairs, so they were excitations.

One difficulty during this period is, we couldn't get out the second order phase transition, in the thermodynamics, just because we didn't treat the real pairs correctly, it turned out. It was just a question of getting the right orthogonalization to work out.

So after trying for that, a couple of weeks, to get out the thermodynamics correctly, we decided not to wait till that got worked out, and published the Letter first, and that was the PHYS REV Letter that was submitted.

Williams:

Right, and just for the record, that vias received on February 18, 1957, probably a day or two —

Schrieffer:

Yes, that was two and a half or three weeks after the New York situation and the wave function. Right.

Williams:

OK, fine.

Warnow:

I'm getting the feeling that, that week you really knew you were right, though, so all it was was ironing out things.

Schrieffer:

John felt really very strongly that this was right. I felt, — you know, I didn't believe it, really I didn't, or I didn't have enough basis to judge. You know, I was very excited, but I didn't understand it.

Warnow:

You were also very young. 25?

Schrieffer:

Yes — right. I'd seen a certain amount of physics. I didn't have perspective. And anything that someone does at that point that is right, at least I tend to be suspicious, you know. I didn't have any basis to judge right or wrong, so I assumed that this was, perhaps not wrong, but it was a beginning of another interesting idea. Like Leon had a very good idea and it worked to a certain extent. I assumed that this was perhaps a good idea, and it would move one along, but this wasn't the solution to the problem.

Warnow:

How did Leon feel? When you met him at the airport, he said, “That's good.” How did he —?

Schrieffer:

Yes — oh, he was really enthusiastic, because you know, we'd really worked as a team, and I can't imagine any more cooperative feeling. The advance of one was the advance of another. At least if we'd chat in those lines, there was never any sense of competition at all, at least that I recall. On the other hand —

Warnow:

— he trusted it more than you did?

Schrieffer:

I have a feeling, that's right. I don't know why. Maybe, you know, the father is the Doubting Thomas in many cases. But —

Warnow:

We were just saying that Cooper perhaps trusted it more than you. Could you comment on that?

Schrieffer:

Yes, I don't really know. I think you have to ask Leon what his feelings were. I think he said, “Gee, that's very interesting. It looks like a development which may be very promising.”

Again, I was excited because it seemed to be qualitatively different. It was a real many-body theory, and it was in some sense at least a partial solution to diagonalizing that Hamiltonian, we felt.

Warnow:

Now, you were earlier saying that there was a definite team effort among the three of you.

Schrieffer:

Yes.

Warnow:

And you were about to say, “On the other hand —” Is there anything there that you remember?

Schrieffer:

Oh, on the other hand, everybody had their own project. In other words, it wasn't that we consciously divided up the work, at least all during that period that led up to the end of January. Everybody was trying to follow his own lead, if you like, or, get his own ideas. It was in that sense a team of independent workers, but once John felt that this was the right direction, then it was clear that the mode of operation changed. It was really dividing up the work that had to be done, because the program really had been worked out in John's mind, I don't know, ten years before or what have you.

He had almost all the pieces, it seemed to me, assembled in his mind, as to how the theory must ultimately work out; if you had a microscopic theory to begin with, what you should work out, what theoretical predictions of which experiments you should go after, and what experimental predictions are such that they will be critical of different parts of the theory.

So he knew exactly what to do.

Williams:

If someone would have given him a wave function —

Schrieffer:

— given a wave function, off he would have gone. He didn't need us at that point really at all. I mean, I'm not sure he ever needed us, but he had this thing so nailed down on every corner, he understood the experiments, he understood the general requirements of the theory — the whole thing was more or less jelled in his mind, and then there was this stumbling block, and that was, you know, how to write down the wave function. So that happened.

Warnow:

You had the design and all you were missing was the wave function, is that right?

Schrieffer:

Yes, basically that's right. You can say the wave function, however, is a cryptic way of writing down all the physics that was missing. There are two views about it. One is that it's a formula that takes one line to write down. The other is, it's 50 years of missing physics. And in between those two limits is somewhat, reality.

If, as later Niels Bohr told me, happened to be correct — he said, "That's too simple to be true. It just can't be true. I don't believe it,” he said, “it's an interesting idea, but Nature isn't that simple.”

He told me that, and I wrote it down, and that's another thing I'd like to find, because after a conversation I had with Bohr in 1958, the one I mentioned to you, with Bohr in Copenhagen, I wrote about two hours of notes trying to recall what Bohr had said, and that struck me, and I wrote Bardeen a letter telling him that. And I don't think John was too happy about that, because he, you know, was sort of defending the fort. He was defending that "This is right" and there were an enormous number of people who had vested interests in themselves solving that problem, and in our having not yet solved it, and they wanted to be really convinced that this was a correct solution, and there wasn't any simple way of saying "This is right," starting from the known and moving through well-defined steps to our answer, because it was an intuitive leap. And any intuitive leap, you have to justify it through a lot of tie points to experiment, and ultimately you hope there's a theoretical deductive way of getting there, but it was certainly far from there, and I think even today we're not there.

So the fact that it was so very simple I think, did disturb Bohr.

But I guess the main point I wanted to make was, I thought it was too simple, and this just can't be the answer.

Williams:

But another thing is, just to take the opposite point of view for a moment — it isn't so simple. You see, it's only simple, when you decide, two other things before he makes his guess on the subway ride: l) what is the Hamiltonian? what is the system energy? Now, that is an unmentionably difficult word. It's one part in 108 in this problem, this little thing that they're talking about. Now, how do you write a second quantization Hamiltonian that gets the essence of that in? Well, that could have been death warmed over for three decades, till you could have guessed the right Hamiltonian. Secondly, that the tactic of, how do you solve that Hamiltonian? As you mentioned, you had already made some decisions, for a variety of reasons, partially Cooper's non-analytic result, I guess —

Schrieffer:

— yes —

Williams:

— that a variational scheme was the only thing that was left. So those two big, I think they're huge points —

Schrieffer:

I think you hit it pretty much on the head.

Williams:

Then in that context, making the guess — it's the leap.

Warnow:

Would you repeat that, so I could have it in your voice? Not word for word, but how you feel that that was right.

Schrieffer:

One point, coming back to what I mentioned earlier about Bardeen’s insistence on how the condensation occurred was that only one in 104 of the electrons, one part in 104 of the electrons really is significantly influenced by the condensation. Those are only the electrons near the Fermi surface. That was one thing.

Secondly, the amount of energy change of those electrons is only about one part in 104 of their energy. So it's a very subtle shift in energy of a small number of electrons.

The reliability of calculating the energy of either the normal state or the superconducting state is of order say one, while the energy you're trying to calculate is 10-8. So you're looking for one part in a hundred million difference between two energies, each of which you'll know to accuracy one, rather than 10-8, so it's clear, you have to know what you're doing, to isolate that little teeny piece that is relevant, get rid of the rest, and then treat that little piece very accurately.

And that's in essence what the problem was. That's not saying anything other than, that's the nature of the problem.

Leon's discovery, that a pair is unstable, suggested a direction we should move, to understand which hunk of the Hamiltonian we should look at, which piece of the total interaction was important, and the fact that vie concentrated on the pairs of zero momentum, rather than trying to treat all momentum pairs simultaneously, was to a certain extent out of simplicity. You know, do the simplest thing first, if it doesn't work, go on to the next most complicated. That seemed obvious.

But then the problem came, we couldn't even solve that simplest of problems, and then once that simplest of problems was apparently solved or nearly solved, we, at least I didn't have any real faith that all the rest that was left over wouldn't significantly change it, so that the theory, with that pairing wave function, wouldn't be as wrong as Leon's simple pair problem was wrong.

Suggestive, but certainly not representative of real physics as it occurs in nature.

That was the concern I had.

Warnow:

You all knew you were looking for a wave function, am I right?

Schrieffer:

Yes.

Warnow:

So for several months you all looked around for wave functions.

Schrieffer:

That's right. Garbage cans and whatever.

Warnow:

I had exactly that picture, I'll try this, I'll try that. Then you tried, you say, it was Tomonaga's —

Schrieffer:

— right —

Warnow:

So this was, what, how many tries had you gone through? Why did you feel this try might be more —?

Schrieffer:

OK, let me try to develop that a little more, because I started a bit and never —

The feeling I had, again, I used I think the v1ord statistical. I felt that there were so many pairs, and it had to be condensation in momentum space of these pairs. At least that's what we thought. Since there were so many pairs, it seemed inconceivable that details of how the pairs interacted with each other, pair say 17 with 641, or two million or whatever — wasn't important. What was important was some sort of statistical average.

All right. Then I said, to write down a statistical ensemble, where pairs are allowed to interact, but not strongly correlated? That was sort of the physical idea I had.

I said, well, let us say that a given pair, state h, has some amplitude, square root of h, to be occupied, and the square root of 1 minus h to be unoccupied, and not let that occupation or nonoccupation be correlated in any way with that for any other state k prime.

The simplest of approximations in many-body physics was the Hartree approximation, which had gone on from Hartree’s days back in the early or middle l920's. That state has k either fully occupied or it's empty, and there isn't anything in between.

So I said, that thing simply can't work for this problem. But maybe, the next step up from Hartree might work. There's a certain amplitude that it's occupied, and a complementary amplitude that it's empty, but don't allow any correlations between the various states. And that's exactly like Hartree said, except he said, "The probability it's coupled is either one or zero."

So in a sense I was trying the next simplest thing from Hartree.

The other was that I had known this scheme had worked pretty well in principle for the polaron problem. It worked somewhat for the pion nucleon problem. And I was trying to say that all the pairs condensed into a state, and that was another important feature, because that was brought out in Tomonaga's paper, the condensation was such that each pion went in the same orbital state around the nucleon. That's a feature of that wave function. And it seemed to me that all the pairs at least spatially were democratic, so I should try to put them all in the same spatial state, they were all overlapping with each other in space.

It was vague. It was intuitive. But it seemed to me that there were a number of features going for it, not the least of which was that one could actually calculate it. And so many good ideas fall by the wayside because you can't actually do anything with it and it stays an idea, but here was a concrete step, that was the next step up from Hartree, that sort of relied on the law of large numbers, that if you have a large number of things interacting, only the statistical average is important. The simplest Hartree approximation couldn't work, let's try the next step. That was sort of a logical sequence I'd gone through.

Warnow:

When you talked about it earlier you more or less said, "Well, you know, this was fun." You didn't have that feeling of Eureka, except in terms of, oh, it's fun doing this, it's working out.

Schrieffer:

Yes. That's right. The consequences weren't clear to me, or weren't important. It was really exciting, it was fun.

Williams:

For a change you could really work out a calculation, just like Cooper worked out his one pair calculation.

Schrieffer:

Right. It was sort of beautiful and elegant, things worked out, it was all algebraic and I didn't have to go to a computer – you know, there weren't terms I just threw away because I just couldn't handle them, but the whole thing was analytic. There were certain beauties, simplicity, which you might call esthetics. I think, to my mind, that's a phony word, it implies more than that — but it was sort of nice, and I liked it.

So people keep saying, "Nature is ultimately simple." I guess in some sense, it depends upon the eyes, but this was so simple I didn't believe it, and that was sort of the other side.

Warnow:

You know, we never did get — and now I'll take a little bit of a backward step, because we never did get on this tape that point where you thought they were bosons. I mean, you only referred to it a little bit. Remember?

Schrieffer:

Oh. OK. Well, that came, what, in November of '56, somewhere around there, October, November —

Warnow:

Are you lost?

Williams:

No. I don't know whether I should say — the impression I got from your course on superconductivity was that soon after Cooper's letter I assume was sent in, since you'd have known of it right away, you all three got stuck, for like four months, until late in January. And of course Bardeen was away part of the time, and so forth. But one of the reasons, you told us, that you got stuck, and I remember this without a doubt, and that is, that you wrote down the Hamiltonian, which I think is a pair Hamiltonian, the one we still use today, but you then worked on the — you then wrote down the commutation relations in a very abbreviated form.

Schrieffer:

Yes.

Williams:

And missed, didn't write them out explicitly – that showed you that they weren't bosons, and it's the ones that are in your paper and your book, I think here you say that they're self-fermions, other-bosons, they're neither full bosons nor full fermions.

Schrieffer:

Yes.

Williams:

That point has, I don't think, ever been discussed by anybody else, and I think it's very important for historical reasons, I think, because — well, I've learned a lot more field theory since then, but I think that corresponds to what's called an indefinite metric.

Schrieffer:

That's nice. I know indefinite metrics are around. I think they’re sometimes called Pauli statistics. But I’m not completely sure myself.

Williams:

Hell, the only thing is that the sign, the commutator is one or minus one, and depends on the occupation — is something that they use a great deal and fuss about, and hope that an elementary particle or something might come out of that end. That always struck me as, here's a real problem that has in this parlance an indefinite metric, and yet, it worked all the way through.

Schrieffer:

Hell, let me point out that we weren't, how shall I say, we were not unaware of the fact that you couldn't put two pairs in the same state. And every variational wave function we wrote down, we faithfully obeyed Fermi statistics. I think it was a relatively short period where there was some confusion about the fact that if you viewed these as operators, and you thought of them as products of fermions, they were bosons, and if they were bosons, then they would all fall to the ground state — you can't put a Fermi level, ergo, no gap.

That may be blown out of proportion, you know, in the historical importance. It was one of the many pitfalls along the way that we got out of, and there were a lot of other ones, but that's one simple one to explain.

Warnow:

In that instance, I think if you could go back to that, wasn't it Bardeen who said, “They're not bosons.”?

Schrieffer:

— Yeah, I think, well, you know — (crosstalk)

Williams:

— this is where the energy gap, where he said, “That's wrong because —”

Schrieffer:

Yes, that's right, I think Bardeen said that to begin with, but it was quickly, you know, corrected, and I think Bardeen himself corrected it. But we each had our own problems. That was a particularly nice one, because I think it characterizes exactly what is not the case, and that is that many people say that the pairs are bosons and you can treat it as a boson condensation, and it's the fact that they're not bosons, that there's a gap.

Williams:

That's really the reason I think that I latched onto that, is because not only is it sort of an historical roadblock that they detoured around, along with 4000 others, but it is also a still to this day, I think, a prominent misconception about the nature of the state. OK? So in other words even though they hurdle it over to do the calculations…

Schrieffer:

… — People calculate correctly but may speak incorrectly. I don’t know.

Williams:

I don't know, it's still a conceptual confusion, that's all.

Warnow:

Now, when you finally realized that everything was falling into place and you had something there, how did you feel about being a part of that?

Schrieffer:

Gee, well, all I can say is, well, two things. One, it was so fantastically exciting that we sort of worked 18 hours a day, because there was just so much to do. And John being at least in my reading, was so convinced that this was the right direction, and this was more or less the answer, not only an approximate but perhaps even in essence, some sense, an exact answer to the problem, that we were just overjoyed. But then we were working like crazy.

So we were working on two levels. One was saying, “Isn't it fantastic? It's all breaking open.” But on the other level, we were having mechanical difficulties of doing all the calculations and working and checking, etc. So it was an intensive period of intellectual activity, but also just hard work.

A very important thing, though, happened in the following way. Leon had been calculating the Meissner effect and the electromagnetic properties. John had been calculating some of the non-equilibrium properties. Leon noticed that in calculating the electromagnetic properties, that the single particles, as we now say quasi-particle excitations, didn't have matrix elements as they would ordinarily have if they were normal state electrons, and some extra strange factors came out, which were square roots of h and one minus h that came, that weren't expected. At least at that time they weren't familiar.

And it had to do with which particular process one was looking at, whether it was interaction with electromagnetic field, where Leon saw it, or when John was working out his transport properties and ultrasonic attenuation, nuclear spin relaxation, he found another combination of these factors.

An energy gap model could be constructed which would give many of the observed properties of a superconductor, by saying that they're, like in a semiconductor, there's an energy gap, and particles get excited across this gap, that will perhaps tend to reduce the gap, because of a self-consistent field effect. But the critical thing was, is this theory in some sense more than a two fluid model which has the right general features? Is there something peculiar to the pairs? Was it pairs or triplets or quadruplets or just something very general?

Well, Heibel-Slichter, Charlie Slichter and his student, Chuck Heibel, were doing experiments on superconducting aluminum, in which they were looking at the nuclear magnetic relaxation rate, as a function of temperature, as you go from normal to superconducting aluminum, and John had been calculating the rate of NMR relaxation based on these states that we had been working with. And exactly what order this occurred in, I'm not quite sure, I'd have to chat with John and Charlie, etc., but Charlie Slichter and Chuck Heibel found that the nuclear relaxation rate, instead of decreasing as you go below the transition temperature, actually started to increase. And according to all the two fluid models, and everything that was around, that was totally unexpected. One would expect the normal electrons that do the nuclear relaxing to disappear and go into the condensate, and the relaxation rate would decrease.

Since it increased, it was very peculiar. And just at this time the theory was worked out, and it predicted an increase followed by a subsequent decrease, and that was what was seen experimentally, so I think that was — there were two indications the theory looked right. One was the relation between Tinkham's energy gap and the known critical magnetic field, and that relationship worked out numerically quite nicely, and the second was, this temperature dependence of the NMR relaxation rate.

A third one was that Morse and co-workers at Brown had been looking at the acoustic attenuation as a function of temperature, and saw that that, instead of increasing like the NMR relaxation rate, it decreased, so the difference between the nuclear magnetic relaxation that Slichter and Heibel upstairs in the lab were finding, which was contrary to all the prejudgments of two fluid models and nobody would ever believe that, and the theory said that it did this strange thing — that strange behavior was exactly reversed when you came to the attenuation of sound waves in a superconductor. In that case, again the simple two fluid model would say that the sound attenuation would decrease as you go below the transition, and the question is, what would the theory say? And the theory said it decreased, so it agreed. So not only did it sort of predict the anomalous thing or agree with the anomalous thing, but it also agreed with the normal situation (i.e. acoustic attenuation) so this was case l and 2. And those were very exciting at the moment. We thought this, all of a sudden said, come on now, the wave function looks right.

Warnow:

OK, now, this is all before the Letter.

Williams:

What Letter?

Warnow:

The Letter that went into PHYS REV. The thing is, you said before that you hadn't worked out the thermodynamics, in the Letter.

Schrieffer:

Right. We did the low temperature thermodynamics, and we tried very hard to get the second order of phase transition, the jump in the specific heat, and that just didn't come out. And the reason was that we hadn't treated the case where you have two single particles or quasi-particles in opposite momentum in spin states, in a consistent way. Or as we eventually called the real pairs in the big paper. And the pairs we had written down, the excited pairs were not orthogonal to the ground state pairs, and that's verboten in quantum mechanics, although we didn't quite realize that was the problem.

Williams:

This question that you were previously speaking about, about coherence factors, there's a letter to you from Bardeen dated 9/1957, which is before the big PHYS REV paper came out, and in it, and this is in Bardeen's little write-up, in it Bardeen says to Schrieffer sort of what you did say, "And it looks so good that I'm adding a note added in proof to our” -– what eventually was the big BCS paper, which is where it sits today. So that this coherence factor feature, as you just said, did in fact sort of happen — well, he wrote it in September, the paper came out in December, so it was actually, the earlier Letter, the BCS Letter came out in mid-February, so —

Schrieffer:

I think the coherence factors occurred sort of in March. It was after the February Letter, as I recall, and in that sort of 13 days we were trying to get the thermodynamics going and the relation between the gap and the critical field, but the electromagnetic properties, the transport properties, spin relaxation, hadn't been worked out. I think that was March or April. It was during that period that all these things got put together.

Williams:

Meaning, at that time you saw them theoretically? Or both theoretically and experimentally?

Schrieffer:

I believe, both theoretically and experimentally. I recall, there was some concern about the question that the wave functions didn't conserve the number of particles. That was sort of line one. Charlie Slichter had been doing these experiments, and he said, "You know, I've got a fixed number of electrons in my specimen. How do I calculate a relaxation rate? How can I use wave functions that don't have a fixed number of electrons? Does it make any sense, particularly when you're changing one electron state only, and if you say that this wave function has, you know, 1023 minus 50, and you're only talking about the transfer of one electron, so to speak, how do you know that you have any predictive power whatsoever? Or aren't you mixing states with different numbers of electrons?"

So we had some debates about that.

Warnow:

When was this?

Schrieffer:

This was March, as I remember. Again, I think you have to check back with Charlie and Bardeen.

We had some arguments, friendly discussions about that, and when we looked at it carefully, it was clear that the operator which connected the initial and final state conserved particles. So if you viewed the initial state as a linear combination of a state with 1023 to 1023 plus 2 plus 4, etc., in the final state as the same thing, since the operator conserved particles, you never mixed states with different numbers of particles, although the way we wrote it down, it wasn't obvious that that was going on.

So that got cleared up, but there was a period where that confusion sort of suggested maybe the agreement with experiment was perhaps fortuitous — but that got straightened out quickly.

Warnow:

OK, and you finally got to the point where you understood that you had really contributed to something that was a longer lasting solution.

Schrieffer:

OK. The only other thing that I think was relevant here was the second order phase transition, and how that broke, because we worked hard trying to get it in that letter, but it didn't go.

Then I think it was about three weeks to a month later. I'd been working very hard on it and Bardeen had, and I remember it was a Wednesday, I thought I'd broken the problem, and I had made a slip of a sign. And early Saturday morning Bardeen called up and sort of said, “I’ve got it, I've got it, the whole thing's worked out!”

And at that point he had got the right sign. Everything was fine. The second order phase transition.

Then we looked at the jump in the specific heat, and that was also very quantitatively known, and we got the right jump in the specific heat, and that was another, I guess, measure of accuracy.

But that was another high point. You think of sort of breaking points, and that was such a tough one, we thought, gee, maybe there is something wrong — or at least I thought so.

Williams:

This is where you might have expected… (off tape)

Warnow:

Let's get that again. That was a Friday night, as you heard it, and then

Schrieffer:

Yes, well, I'm not completely sure of the details. You should ask Jane Bardeen about it. John probably could say also. But I think that Friday night, a distinguished Swedish scientist, Borelius, I believe, was visiting the Bardeens, and so as I recall, again, my memory may not be accurate here but Jane could say, that John was somehow off on Cloud 7 that night, and there were long gaps in the conversation where John was staring into space and the conversation was going on but in a very strange sort of way, and it was clear that John was thinking hard about something. And what he was thinking about was how to get the second order phase transition, and exactly how to write the wave function down.

So the next morning — apparently that night, he had cracked the problem, and called up the next morning. He was really excited.

Warnow:

And woke you up.

Schrieffer:

Yes, woke me up early in the morning.

Warnow:

And that was the Eureka feeling.

Schrieffer:

That was the Eureka, yes. I think at that point, he had felt, you know, everything was correct, and again I guess I'd been somewhat of a Doubting Thomas in the sense that I felt, OK, it had worked all right up to a given temperature, but the phase transition is always the most difficult thing, or generally the most difficult thing to get right, and you're changing the nature of one phase to another, and while we may have been lucky to get the ground state and the low temperature excitation, to be able to discuss in detail how you go continuously from one phase to another is usually very difficult. There are very few areas in physics where you can do that in detail. I think superconductivity is almost unique in that respect — that this molecular field approximation, if you like, works to a very high degree of accuracy. There are very few other phase transitions where that happens.

Warnow:

What was the reaction you had? Did you feel like, I don't know, going out on the town, flying a balloon, taking a nap?

Schrieffer:

Well, in a way, it was all great, and we were very, very busy, and we got all excited, and we kept being busy. There was just so much to do. Then there was the whole thing of — there were an enormous number of calculations, because the long paper was quite long, and there was quite a bit of work in there. John and Leon and I divided it up, but I'm sure John did at least more than his share, Leon did also, so we were all very busy. And it was an exciting time, because, let's see, I had to 1vrite my thesis, and had to get off on the NSF post-doc at that point — I think I may earlier have said Fulbright, I'm sorry, it was NSF. The Fulbright was to go as a beginning graduate student. This was the NSF post-doc.

But I had to write the thesis. So I went off to New Hampshire in what, the beginning or the middle of March, and I was going to write my thesis up in New Hampshire, quietly getting the thing written out.

Then came — let's see, then Fred Seitz had called Eli Burstein, who was somehow in charge or at least related with the March meeting of The American Physical Society here in Philadelphia, the solid state physics meeting, and said that a major break in the theory of superconductivity had occurred, at least John believed so, and that, was it possible to have two post-deadline papers?

And so those were arranged, and John refused himself to come to speak about the theory, because he wanted to make sure that the young people got the credit. You know, that's unbelievable — fantastic!

So Leon was able to come, and I got the word so late that I couldn't get on the plane to come, so he gave both papers together. He gave the one I was to give and the one he gave. So that was presented, I think, sort of an open audience, the first time at that March meeting.

I think this is a real commentary on John's humanity and ability to not only put in perspective his contributions, but sort of give more than ample reward to young people coming along. It's really one of the many strengths he has.

Warnow:

There was something you wrote, if I can find it, explaining the theory of superconductivity in terms of people dancing.

Schrieffer:

Oh yes.

Warnow:

Couples on a dance floor, and the wave function is pairs of electrons hooked up like a square dance.

Schrieffer:

Yeah, I've used that a number of times, and it's one way I can get it across, which I think is somewhat borne out and factual. But I find that's very helpful for women's clubs, undergraduates, my mother and father.

Warnow:

Since this is going to be partly used for undergraduates and high school students, why don't you tell us one more time, if you can?

Schrieffer:

OK. Should we take a break for a minute…? (break)

Warnow:

Cooper's talk in March, at the solid state meeting.

Williams:

This is at The American Physical Society meeting in March of 1957, and what were you going to say about that?

Schrieffer:

My comment was that this was a particularly interesting event, not only because it was announcing the theory, fairly early after its inception, if you like, in a very raw form. It had only been, what, a month and a half old. And the system responded to provide a possibility or a vehicle to get this out. But much more so, it was to my mind a remarkable insight into the personal character of John Bardeen, who, I think, in many ways has felt the intrinsic intellectual contribution he made through superconductivity in some ways superseded that which was made in the invention of the transistor. He's said this on various occasions.

And yet, for him, after struggling with the problem, with a great amount of success, and having finally come to the pinnacle of achievement in his professional life, in a sense, he steps aside for two young people, one of whom was a graduate student, just sort of began in the field a year and a half before — the other wasn't from the field at all, but was a post-doc brought in – and says, "OK, you go out and tell the world, and I will stay here in Urbana."

It's just beyond belief.

And furthermore, if there's anyone who believed in it, it was John, so it certainly wasn't out of any concern in any sense that the reaction might be negative. On the contrary.

That's probably the most striking example that can be documented of this wonderful characteristic he has of giving credit, even to an extent beyond that which is due, but also pushing young people as fast as they can to become professionals, and treating them as professionals, and making them rely on themselves.

So I think to my mind, that's probably the most exciting message of the whole thing.

Of that incident.

Warnow:

Going back, also, the fact that you could work with him from the moment you became a graduate student.

Schrieffer:

Yes. That again is — that was 20 years before its time; people keep talking about doing that now. I, in some sense, was totally unqualified, and yet how do you get qualified other than doing it? The fact that he didn't have 15 students — I guess he had one other student, Walt Etherlee, who was really working part time, and was working rather slowly on some problems — gave John plenty of time to coach me, to educate me, to help me along.

He did have the transistor lab that took up a certain amount of his time, but he was free to interact, and very giving of his time.

Warnow:

He didn't know the rules of academia, either.

Williams:

What are they?

Warnow:

Well, he'd come from Bell Labs, and he didn't know that — you know, you didn't necessarily have to work with your first year graduate students so closely.

Schrieffer:

I think he was interested in trying to get young people around him, and he certainly enjoys working with students, I know that, enormously. The other aspect is that Jane Bardeen is one of the most affable, lovable, outgoing people, and from the very beginning I was invited as part of their family. I say that in a real sense, because Betsy Bardeen used to sit in my lap. She has unfortunately stopped that as years have gone along. But I would tell her about snakes and alligators in Florida. You know, I would come around for dinner. In fact, one time I even forgot about a dinner invitation, and John said, "Gee, you missed some good roast beef last night." But it was a very relaxed warm relationship, which I think in a way is perhaps past time now. It's difficult to do that with a large research group, particularly in high energy physics. Maybe it's still there in solid state. But a relationship which in many ways is ideal, at least in my view of how education should really be done.

Warnow:

And the other thing that comes to mind is his encouragement to you to stick around for at least another month.

Schrieffer:

Yes, that's right.

Warnow:

Did Leon Cooper ever feel like dropping out of the whole project, do you know?

Schrieffer:

Well, I don't really know. Leon never mentioned it. I don't know what he was doing on the side. He may have been continuing his interest in particle physics, because presumably he was going to go back to that. I think this was a detour from his main line of interest. He has straddled the fence since that time, to a certain extent, going between one and the other.

Williams:

Your comment about sort of being part of the family life of the Bardeen family, in addition to having this professional relationship — it reminds me of the idea that education should be part of life, part of your life experience, not just learning technical tricks in philology or whatever field and get a PhD in X field, but it reminds me of the older time when I hear the stories from the quantum days, I think they had that kind of — I think their living of their families, I get that flavor from the Bohrs —

Warnow:

They lived in the Bohr Institute for many years.

Williams:

It intertwined. But even the other people — Bohr was sort of a special case, since the nation had made this house that was sort of the White House of Science — (crosstalk)

Warnow:

— I mean, originally he and his wife and family had an apartment in the Institute itself, and the Bohrs used to sort of run through the Institute with five sons.

Williams:

But in the other countries too, you read about, for instance, the beautiful biography of Hilbert, he's doing that and he's taking a break and having lunch and going out and doing the roses — and then there's the party for the students on the weekend — it's — something is different today, whether it's the numbers, the fragmentation, we don't live — in other words, there they lived together and got their education together, whereas now you kind of live apart, and you just come together for "your training." And the degree is the same, but the experience is very, very different. I mean, he remembers so clearly, for example, I can't help note you remember so clearly a lot of your family experiences with the Bardeen family and vividly and intensely over the, how many years? 20 years almost now.

Schrieffer:

We have been sort of back and forth with families and visiting, there's a close relationship.

In a way, just amplifying a little bit on what you've said, the things one learns from one's mentor are multi-faceted, but to my mind, there's an overriding one that I sort of learned from John, and that is taste in physics, and how to go about a problem — typically, to use the smallest weapon available in your arsenal to kill a monster. And that's one thing.

The other is to divide up the problem into small pieces, and attack each one, but learn how to focus your shots so that each one counts, and then re-assemble. And it's this way of taking big problems apart, and carefully deciding which is the essential piece, and then synthesizing them back again, which he is really a master at, and there's no way to learn that. He never said to do that. But it was so clear, that's how he went about everything, thinking about it carefully, and isolating those pieces which were relevant and then coming back.

But he was always very leery of someone who started out with a grand formalism and deduced proof. Rather, you should find proof differentially, and synthesize the proof back to the big picture.

The other thing is, never look too far in the future to say, "Gee, I really want to solve that big problem." Don't be afraid to solve little problems along, and then the big problem will get solved by nature if you just keep at it and do your thing, so all these messages are very philosophical, but they're important when one tries to do research for oneself.

Williams:

Just speaking of myself — growing up — they're very strange messages. I don't mean they're wrong or anything, but I don't think I've ever heard, except the breaking apart a big problem, but of course, how he does it is part of it. But certainly using the smallest weapon in your arsenal, that's a well hidden, I've never heard that. There's always been a great stylistic tendency, you saw an example of it earlier today, perhaps, unconsciously, when I displayed admiration for Cooper's more remote formalistic expression of what to you as a freshman —

Schrieffer:

— I tried to quietly suggest, that was more confusing than helpful.

Williams:

But this is to you an intuitive reaction, and I think now —

Warnow:

— you can understand that —

Williams:

— you can see from whence this flows, and you see my intuitive reaction is very distinct.

Warnow:

did you have anything further to say about that conversation with Bohr? Something that might be pertinent now?

Schrieffer:

Well, perhaps just a little background on how I happened to have the conversation. I'd come to the Bohr Institute in — let's see, this was part of the NSF post-doc —

Warnow:

— you went to Rudolf Peierls.

Schrieffer:

Peierls. That was in the fall of '57. And by January, the fog closed in, so I moved off to Italy, where the weather was a little bit better, for a couple of months, and then came through Copenhagen — let's see, I guess I was there more or less from May through August of '57.

Soon after I arrived, I was asked to give a seminar, and I gave a seminar at which Bohr appeared in the front row, and so then he said, "Well, look, we really ought to meet, because I've been interested in superconductivity for many years."

And he and Rosenfeld, Aage Bohr and Jens Lindhard had met periodically for a number of years, discussing superconductivity.

Bohr's idea seemed to be, and there is — let's see, a 1929 Zeitschrift für Physik article that he wrote but it was never published, it's in page proof, which I finally got a copy of – but the basic idea was that electrons, forming currents, interact with each other, roughly through the Coulomb forces, and the Coulomb forces set up some sort of a lattice, which then makes the whole electron gas flow as some sort of a solid, rather than the phonons being important, and he felt that the isotope effect and the phonons were somewhat of an afterthought, really weren't so important.

So we set up a series of meetings, of which I think actually two took place, and it always started out, "Well, electrons have to move in a periodic solid, now let's start out with the one dimensional Kronig-Penny model."

I remember thinking, "Oh, here we go again," and then trying to solve one electron moving in a periodic potential, and that wasn't the problem, but you know, we have to start at the basic, Bohr always said, and then we build up.

Then two hours would get over, and he hadn't finished solving that problem. Then the next time we got together, he sort of said, "Now, let's see, where were we?"

So then, there was a dinner party to which I was invited. Norman Ramsey was the scientific advisor to NATO at the time. He was at the dinner party, and there were journalists and artists and various people. After dinner, everyone was given after dinner drinks or cigars, and then he came over and said, "Now, it's time to talk."

He led me out to this great Roman arena and he said, "Now you must tell me about superconductivity."

So I started to say, "Well, we've been working on this."

He said, "Just a moment. Let me get one thing clear. In 1928, I started thinking about this…”

An hour and a half later he said, “Now, you really must tell me about superconductivity.”

And after about three or four minutes he said, "Aha, I understand everything, thank you," and that was the last we ever said.

So I think he had ideas that were so firmly fixed in his mind that, you know, like many people, they're not susceptible to other people's ideas until they've followed their ideas to their ultimate conclusion.

I later learned, in fact on the way back from the Nobel Prize, as I was asked to dedicate the Niels Bohr Auditorium at Risø, that he had been working the summer of '61 and '62, (I believe he died in November of '62), on another manuscript on superconductivity, and I think that he and Rosenfeld worked both summers on it.

It was somewhat along the pairing line, so-called pairing line, but it had slight vestigial properties of the earlier ideas also, and I haven't seen that manuscript.

Warnow:

You can see it in my library.

Schrieffer:

You have it? Great. I would love to see it.

Warnow:

I'm sure it's in there, since we have the full manuscripts on some —

Schrieffer:

— OK, beautiful, I would love to see that.

Warnow:

When did he make the statement that it was too simple? When he first heard about it? Because once he actually listened to the ideas, then he accepted it.

Schrieffer:

Well, it's not clear to what extent he ever accepted it. I don't really know. I've heard a little bit from Rosenfeld on that and a little bit from Aage Bohr. To what extent he totally accepted versus partly accepted, I don't really know. You should go back to the better sources. I have an impression that he was never completely happy.

Warnow:

But at least he understood what you were doing and accepted some parts of it probably, if he — we'd have to look at that last manuscript.

Schrieffer:

Yes. Right.

When he said he thought it was too simple was that night after I had the long conversation with him, and apparently he had reflected on the seminar I gave, and perhaps looked into the manuscript a bit, but he said he thought the idea was really much too simple to have anything to do with real science or real physics. And my guess is that he just didn't have time or inclination to study it in detail.

Warnow:

What were the reactions of some other significant people, like Seitz, for example?

Schrieffer:

Oh, he immediately said, "From what I can tell, you know, it sounds great. If John Bardeen did it, it's obviously going to be, if not correct, as correct as we can imagine." He was a very vocal advocate.

Of course, Fred was not in the field, and it's easy to advocate something where it's one of your colleagues and good friends, and no way have you been emotionally involved. That wasn't uniformly so, as I'm sure you're aware, even more than I am, but there were a number of people who were concerned.

Williams:

What about — I think some of the problems of the people in solid state physics or whatever who didn't and so forth, have been discussed many times, but let's just, to continue this idea of Bohr, what about the people who were theoretical physicists earlier, say, either in quantum mechanics or quantum electrodynamics or something, what about the reception in that wider community then? Because Bohr, even though he'd done the paper in '28 or '29, he is not thought of as a solid state physicist certainly.

Schrieffer:

In those days, there weren't solid state physicists.

Williams:

Today, if you break up the reaction to the theory by outside theoreticians, you think about those within the field and those — what about the reaction by those people?

Schrieffer:

Outside of solid state?

Williams:

Yes, like other elementary particle — maybe Dirac or, I don't, can't figure out who was up — Heisenberg was still alive, of course.

Schrieffer:

Yes. Of course, Heisenberg and Koppe had a theory which they published, and that was, in a sense, also relying on the electron-electron Coulomb interactions, and they thought about wave packets and electrons which were repelling each other, and hence setting up some sort of a lattice, vaguely similar to the Bohr idea, although Bohr's ideas were never written down, at least that I'd seen, in enough detail to tell what he really had in mind. They were some very general ideas about momentum conservation, translation and variance of the full Hamiltonian under the translation group of the lattice. But Heisenberg at least, I don't know that he communicated, maybe he communicated to John Bardeen, he certainly never communicated to me, and —

Williams:

Did you have any communications with any of the other people that are outside of the field of solid state physics about it? Usually the other stories are given most attention, but since you already mentioned the Bohr communication —

Schrieffer:

Peierls seemed to be, you know, affable, agreeable –-

Warnow:

OK, you went to Peierls, and what kind of work did you do with him?

Schrieffer:

Well, what I v1as doing there was continuing to develop work in superconductivity, and he was much interested in the problem. The so-called gauge invariance problem, which was very big at the time. And Bob [Williams] alluded to the fact that there were proofs or alleged proofs that there could be no gap, because it violated gauge invariance, fundamental fact of electrodynamics and quantum mechanics.

It turned out that the original theory, while it in principle violated gauge invariance, in practice it didn't really violate it. If one interpreted it correctly. It's one of these things that depends upon how you interpret it, and that took a while to get straightened out, but there was a big flap over this question of whether gauge invariance was or was not violated.

Wentzel got involved with that, a number of other people — Nambu, right down the hall from Wentzel, made a substantive contribution and in essence straightened it out. Simultaneously Phil Anderson… (off tape)

Williams:

You spoke earlier about Bardeen and some of the things you'd gotten from him, from living with his family and so on, — what sort of things did you get from Slater? These are two people that are among the most outstanding teachers in solid state in the United States.

Schrieffer:

Well, I think that primarily what I got from Professor Slater was an early introduction into the power of quantum mechanics and in real physical systems, and the importance of simple calculations on simple systems, as a way to understand science.

I'd say that he has made enormous contributions to atomic physics and molecular physics, but in his contributions to molecular physics, he didn't do the molecular structure of hemoglobin, he did the molecular structure of H2

and nitric oxide or what have you, and he approached the problem in a very hardnosed way, where he really calculated things through in enormous detail, and sort of the, how shall I say, the discipline which is required to carry through a calculation correctly, and assess one's errors, etc. That's sort of the message I got from him early in the game.

I think I mentioned the esprit de corps around the group.

I also, I think, learned to an extent to appreciate individual creativity, and the milieu in which that is possible, because I think that in a way, he was such a strong person, and so dominated his group, that it was somewhat difficult for his people to be individually creative.

Williams:

I notice, if vie look back, trying to reflect over the afternoon's talk and so forth, the word "creative" has come up an amazing number of times. I guess in some of the literature we've read, one of the personal little interview stories, there was some story that you had spoken to a freshman group about creative things, this is one of your interests. Would you sort of chat about it in general, perhaps in connection with BCS experience?

Schrieffer:

Sure. The process of creativity, to me, is one of the most interesting, exciting, and yet least understood of all the disciplines, in my view. Somehow creativity is what one tries to engender in people going through the university, and it's not a given body of knowledge, as you've mentioned, that one tries to engender, but it's the ability for that person to gain the appropriate knowledge to create something new.

I enjoy creating in woodwork. I enjoy creating in painting. And I guess I'm a nut to do something, and the thrill and joy of doing something myself, is what sort of spurs me on.

Creativity, I don't know quite how to put this, is an ideal which is very difficult to achieve on a day by day basis, and I feel one of the important things is to balance creativity, whatever that means, with the self-discipline of laying the base for the next creative step, and learning how to optimize the hard work and the constructive mathematical or laboratory experiments into creative insights, as to, let's not do these three calculations or experiments, but let's leap to the next one and see if we can do that — is sort of the name of the game.

But creativity in living, for example, is something which I guess all of us have concerns about, and when you try to entertain children or have fun with them, one realizes how tough it is to be creative. They certainly get bored very quickly.

But I am enamored with the whole thing, and I've done some reading. For example, Gheislin has a nice little book about that, THE CREATIVE PROCESS, from the seminar at Berkeley, I enjoyed reading through that. How one goes about —

Williams:

Excuse me, did you read Koestler's book, ACT OF CREATION?

Schrieffer:

I glanced through that, right, I glanced through a bit of that. The transmission to students of how to be creative is a concern I think every teacher really faces. Eli Burstein here in the department, many times says the earlier he can get one of his students to publish a paper, the sooner he's a professional, ergo, expected to perform on that level — this stretches the person, and it's the sort of stretching that is so important.

Williams:

Earlier when you chatted about Bardeen, you mentioned the word "taste." Then you gave the three examples, weapons in the arsenal, breaking apart, solving little problems. Did you have anything else in mind when you mentioned "taste," or were those three things sort of to define the word "taste?" Sort of a limited word.

Schrieffer:

Taste encompasses so much, and I think it's taste in human relations, in his case, and judgment of, say, service to country, as well as approach to problems. And one can say a person has impeccable taste, but it depends upon your view of what taste is, whether it's impeccable or not in your sight.

I think John has great taste in going about physics, largely because he does it to understand physics, rather than to impress himself or other people. That's why he uses the smallest machinery necessary in his arsenal to attack something.

He's truly in love with discovery and creation, rather than in love with, somehow, the formalism or the mechanism of it. It's the actuality rather than the process which I think he so enjoys, and he sort of tries to pass it on to the people who work with him. He's so very good at it, he often hits on the mark, and there's nothing like success to make happiness.

Warnow:

Did you cover the teachers?

Williams:

Yes.

Warnow:

Do you consider Bardeen your mentor?

Schrieffer:

Oh yes, I think so. I guess I have many mentors.

Warnow:

Certainly up to that point.

Schrieffer:

I'm not sure what the word "mentor" means. I don't really understand. Maybe you can develop that one.

Warnow:

To me, it is someone that you would like to style yourself after. Don't you think that's the meaning of mentor?

Williams:

I don't use the word.

Schrieffer:

OK, to some extent I would say, yes. To an extent one wants to emulate another person — clearly not in detail — but there are ideals I see there, in 1vhat I consider sufficiently pure form that I would like to aspire to many of those. Other things, everybody is an individual and human and you want to live your own life your own way. But I see so many good things that I wish to aspire to, and in that sense, I'd say, yes.

Warnow:

I have one more thing, which I've been after several times without success.

Schrieffer:

OK?

Warnow:

The dance.

Schrieffer:

Oh, the dance, OK. Get back to the dance — yes, I've used an analogy on a number of occasions which I find is a bit helpful, and so it goes as follows: If we admit that the electrons attract each other, and they attract each other for the reason of the Bardeen-Pines interaction, or more simply put, if you put a negative charge in a region where there are positive ions around, they pull in the positive charge, and another electron, instead of being repelled from the first electron, is fooled by the fact that that electron is clothed in sheep's clothing, if you like, and gets attracted to the positive charge, so it gets attracted, fine.

So electrons get attracted. Now, if one thinks of the ground state wave function in coordinate space (rather than in momentum space), where most people tend to think, that's the pairing wave function and is a very difficult concept. The difficulty is the high density of pairs. Because of this difficulty, there are about a million pairs whose centers of mass fall within the volume occupied by a given pair.

So I tried to use the following analogy. Suppose you say that you have a large number of couples on a dance floor, and every male has an up spin and a female has a down spin, so they're up and down spin electrons. They're doing a frug or whatever this dance is where they never touch each other and are very far apart, and dancing around this dance floor. OK, they may be, say, a couple of hundred feet apart, but they always know exactly to whom they are mated, who's their partner, and yet there are roughly one million other pairs dancing in the area corresponding to the space in between those two — the cube, the two-thirds root of that, about 10,000 people.

Now, these dancing couples are essentially totally covering the dance floor. There is very little space not covered by people. So when they dance, they have to do a highly intricate step of moving into a space that at that instant happens to be vacant, and this is an enormously complicated choreography, so that one doesn't trip, if you like, or hit someone else. And the electrons can't hit each other, or at least they can't occupy the same space at the same time, according to Pauli.

Fine, so they're all dancing together — by dancing, if you like, they lower their energy or make themselves happier or whatever analogy you like to make, fine.

Now, suppose that the dance floor is tipped. Or, another way of saying it, somebody starts pushing on one end of the dancing group, and the dancing group starts to drift across the dance floor, everybody still doing the same choreographed step, not in the rest frame but in the moving frame.

Suppose, however, that there happen to be some wood chips or nails or what have you, sticking up from the floor, and these correspond to the impurities or defects in the superconductor, or lattice vibrations that are thermally excited.

Then say a given mate of a pair would tend to be tripped. But unfortunately there's no space for that mate to go into, because it's occupied by another one, or if it does go into the wrong space, where it shouldn't have gone, it gets out of synchronism, or dance pattern, choreography, with its mate and can no longer dance. Ergo, its energy goes up discontinuously.

The only way to slow down the entire ensemble without increasing energy is not differentially, pair by pair by slowing down, because that increases the energy. The only way to slow it down and decrease the energy is for the entire dancing ensemble to slow down. And that's very unlikely, if they're just random bumps around the floor. You tend to be carried along by your mate. Another analogy being the marching army, where you have 20,000 troops in a big cylindrical or circular group, they're all locked in arms, and they say, "March at 3 miles an hour due east," then they're marching over a rough field, and anybody who trips gets carried along by his neighbors.

So there are various analogies, but it seems to me that that simple analogy tells you roughly what goes on.

Now, I have other analogies, like the attraction I mentioned. I sort of have one that, you can think of the attraction between two electrons is like putting two people in a rather poor bed. The first one makes an indentation. When you put the other one down, it tends to roll into the indentation made by the first. That's called pairing.

Warnow:

As opposed to bundling.

Schrieffer:

That's right, as opposed to bundling.

Warnow:

You get the square dance by having the BCS wave function, OK?

Schrieffer:

Right. Yes. So the choreographic notes, if you like, or that thing which is written down by the choreographer, to tell everybody how to dance, at least describe how they do dance — you know, it may be that God created all these people beautifully choreographed, and a 11 we did was figure out what the dance pattern is, and we wrote it down, and instead of taking 200 volumes, it turns out to take two lines. And if you get the right language, it appears enormously simple. If you have the wrong language, you probably couldn't — I'm sure you couldn't write it down in all the volumes in the entire world. If you write it down in coordinate space, there are 1023 electrons, and to write down even 1023 symbols would take more than all the paper in the universe. So you write it in a symbolic way which is enormously simple, allows you to calculate with it and make predictions, without ever writing the thing down in its gory details.

The wave function is just — I don't know how people write down dances, there must be some notation. And these are the symbols which record the dance the electrons are making.

Now, we didn't invent the dance. God invented the dance, if you like. Kammerlingh-Onnes discovered that the dance was going on, and we were the choreographers that recorded what the dance was.

Warnow:

I have one more thing. You just mentioned Kammerlingh-Onnes and I think it's delightful — by the way, I have to be at the Philosophical Society Library by closing time at 5 —

Schrieffer:

— easy, OK, we can close up in another five or ten minutes.

Warnow:

The story of Kammerlingh-Onnes taking the loop of wire –- where did you read about that?

Schrieffer:

Well, that's interesting, because I read about it in John Bardeen's write-up of the history of superconductivity. Now, it turns out that there may have been a small error in the details of that story, which then was pointed out by Jack Allen in Scotland, and wrote a note to me, which I then communicated to John Bardeen, and there was a minor correction to that story, because Jack Allen claims he was intimately involved in that process, and it was an RAF pilot that flew the thing across the Channel, etc., so it's a whole story that one can get accurately from Jack Allen.

Warnow:

It wasn't in 1914, then? Probably?

Schrieffer:

Well, that's my feeling.

Warnow:

It was a colorful story.

Schrieffer:

The story is more or less true. The details, I sort of picked up from John's writing, and John, I think, to an extent, should be the one to say how it got confused. I simply carried the message. I believe that John now says he misinterpreted an article he read. So maybe we'd trace it back that way. Jack Allen has what he claims is the true story.

Warnow:

OK, is there anything else you want to mention, Bob? You should mention it now. I think we've covered the discovery very well. I'm trying to remember, way back, you said something, something was an interesting side story, and we'd come upon it later, have we come upon it?

Schrieffer:

Yes, the March paper, maybe that was it?

Williams:

See, when he says that, I keep writing. When he gets done I check it off, because otherwise —

Warnow:

OK, one more thing on this, and that is, papers and photographs. How are we doing on that? We know you have papers in your attic.

Schrieffer:

That's right. The photographs, I think I can produce maybe a couple. There just aren't photographs. We weren't photographing. You know, the best I might be able to do is to pull up some of the Bardeen family or of me or something like that with the Bardeens, I don't know.

Williams:

Maybe you should ask the Bardeen family.

Schrieffer:

I wasn't much of a photographer —

Williams:

— in their family album, I guess that's where —

Warnow:

— anyone in that family take photographs that you remember?

Schrieffer:

I don't think people were much for photography at that time.

Warnow:

In the field of superconductivity, he took lots of photographs. There's more, though, let's see, Bardeen is one possibility, Cooper is another one in terms of pictures. I have a picture of you straightening Cooper's tie.

Schrieffer:

Oh yes.

Warnow:

And then the papers that you have, I don't know what we're going to do about that. I have a feeling that you won't find the time to go through them, without somebody like me being an irritant.

Schrieffer:

I would suggest very strongly you be an irritant. But the attic is now cool enough to get up there once again, and I —

Williams:

— one thing we ask — (crosstalk)

Schrieffer:

— not too difficult to find. I spent part of yesterday looking through my office for the stuff, and I then realized what happened, that the bottom pullout drawer was cleaned out about two years ago, and that stuff was transferred to home, and that's where I found the long yellow pages, so —

Warnow:

— can I be happily suspicious that maybe you haven't thrown out as much as you thought you have over the years?

Schrieffer:

You can be. I would say the notes definitely exist and without question, the yellow sheets where it's first written down. Much more than that, I'm not at all sure about.

Williams:

What about the envelope that you did it on in the subway, the work you did in Summit?

Schrieffer:

Those are the yellow sheets. That's within 12 hours or what have you of the original idea, so that is there and written down. I found those and I was surprised that they were still there, and I wrote, “Don't throw out, these are,” etc., on them. So I don't think that's gone, anyway.

Warnow:

Now I just can't resist, since we have maybe five more minutes, right — to get at the story of how you did win the prize, and how you learned the news, rather — and the tidbit of your son asking what the Nobel Prize was, just a week earlier. That's a wonderful coincidence.

Schrieffer:

Well, I don't remember too much —

Williams:

— apparently in one of the accounts, a popular account, not necessarily reliable, your son had for whatever reason, been asking you a week or so before the physics prize was announced, what is this about the prize? And you had discussed it with him in some general terms.

Schrieffer:

And he was asking, "Well, what do you have to do?" And after I told him, "Work hard," he didn't think he v1as too interested. Yes, I think something like that roughly happened. We had a lot of crazy conversations together. I think that's one of them.

Warnow:

How old was he at the time? Eight?

Schrieffer:

Eight, I think — seven — he's nine now, minus two let's see, January 3, I guess he was just seven at that time.

The other one was more or less told today in the movie, that I pulled up there — I guess the one thing I didn't mention is that on the way up in the turnpike, I stopped off at Howard Johnson's —

Warnow:

— with your wife getting the phone call. Don't you think so?

Schrieffer:

I didn't know a word about that.

Williams:

That's second hand.

Schrieffer:

Yes, I can't (crosstalk) — no, she got it about a half hour after I left, and she was so upset, she had to get the house cleaned up and all the rest. So they appeared while I was chatting with her on the phone and she was in her underwear. We happened to have an atrium, it's a two story glass thing filled with plants, and they were down there, having come in the front door without her knowing it, and she was talking to me on the phone, and she looked don, and here were the TV men looking up at her. So she let out a shriek, etc.

Warnow:

That's a warm reception. Anyhow, she got the call, and you're on the road.

Schrieffer:

Right. So I stopped off at Howard Johnson's to study a manuscript that I had to talk over with this friend of mine, who was consulting with me, and I was so taken up with this, because I'd found a couple of errors, I thought, some possibilities for changed direction of research, when I got there and he told me this – first of all I said, "Yeah, what else is new? That's a lot of baloney," because I'd heard this before and I thought it was a bad joke. He usually doesn't make bad jokes. Finally I was convinced that's right. But then I said, "Well, lookit now, this manuscript, you've got some errors in here," and I was so hepped up on that, that it got me through the first hour. I said, "Lookit, I want to take off an hour before I talk to the newsmen."

There was a guy, a stringer from Newark that was with the NEW YORK TIMES that had come out to the lab, and a couple of other people, I guess an internal Exxon guy was there, interview.

But we sat down for an hour and talked about this manuscript. And that got me through the first hour, if you like –- just like at the funeral the big food and what have you, gets you through the first day. And the rest went very smoothly.

I guess I talked to Ann and she said, "You've gotta come back," because there had been a premonition here and pre-knowledge that looked pretty good a week before. Somehow, something got out, I don't know how.

Warnow:

What do you mean? Who had the premonition?

Schrieffer:

I don't know. But apparently somebody here. In any event, they had socked away three cases of champagne in the faculty club. And that was a week beforehand. So I guess they figured they'd use it on any likely occasion that would come about. Anybody v1ho came in.

Warnow:

By the time you called her, she had had a busy morning.

Schrieffer:

Yes, already by that time. So I got on the Turnpike, and the people at Exxon, or Essa at that time, didn't want me to drive back. They wanted somebody to drive with me, for fear that I'd drive off the edge of the road.

Williams:

You mean, they thought you'd be so upset?

Schrieffer:

They thought I was, but you know, it seemed very nice and it was a great thing and I sort of accepted it. I don't know why. I guess any big thing that happens, one imagines that it's going to be totally upsetting, but it wasn't.

Warnow:

Did you think things like, so this is how it feels?

Schrieffer:

You know, I just felt, gee, that's great, isn't that marvelous, it's wonderful, isn't it nice? You know, I wasn't really high. I thought, I have to do this, that and the other thing. I guess when one's sort of under pressure of traveling and going places, you get used to very strange things happening, like missing planes and crises. And this is a pleasant strange thing so —

Williams:

The other thing, perhaps if you'd heard about it at home, when you were — you could have called up, canceled your visit, said, "We'll put it to another time," and you could have gotten upset with your wife.

Schrieffer:

That's right. That could happen.

Williams:

She was upset with the house cleaning and all the other activities. But it just depends where it falls on you.

Schrieffer:

We had a news conference that afternoon, and invited in, I guess we had about 300 people into the house. However, every month for a month and a half, what we call the "Danish Mafia" meets, Ann and her five or six Danish girlfriends. And that evening happened to be the Danish Mafia meeting at our house for dinner. All the Danish couples came, I guess there were ten or twelve of us, and after dinner, in fact we were in the middle of the dessert course, people arrived. Anna had called in the cleaners and called in the champagne and ordered some garbage cans filled with ice, and the Danish couples served as servants, you know, and served all the food and what have you.

Warnow:

Was that a surprise to you?

Schrieffer:

Oh yes. I just had no idea. Anna didn't tell me anybody was coming or what have you.

Warnow:

How did your kids feel?

Schrieffer:

Oh, they thought it was a great party. They were having fun running around. The other thing was that they were chosen as the Nobel Children for that year. I guess every year one family is chosen as the Nobel Family, so we happened to be —

Williams:

— what does that mean?

Schrieffer:

It's that the news or what have you, focuses on one family out of all the laureates each year, and it's usually the one with the smallest children, to make some news stories. We hadn't heard about this, so when we got to the Stockholm airport, there were all sorts of news cameras and everybody and I thought, gee, that's a pretty elaborate affair, which I guess they do for everyone, but they sort of swept me aside and Anna aside, you know, and the children were getting all the attention. I thought, that isn't totally fair.

And Erik Rudberg, you know, head of the committee, had a Donald Duck bag over one hand and our little girl's security blanket over the other — the kids — Every morning at 8 the T.V. crew would phone up and ask where the children were going to be hour by hour, and then they had something like a 20 minute TV special, which we saw in Denmark on the way back, about the children, what they did.

[1]exp -(2/NV)

[2]exp -(1/NV)