*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

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.

In footnotes or endnotes please cite AIP interviews like this:

Interview of Laszlo Tisza by Kostas Gavroglou on 1987 November 15, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/4915-1

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

The beginning of Tisza’s career, 1925-1937; recollections and comments from his two-year study at Kharkov with Lev Landau. Discussion of important work in low temperature physics, 1938-1952, especially superfluidity; recollections of Fritz London. Discussion of work in the 1950s in thermodynamics and statistical mechanics; involvement with the philosophy of physics; his recent work in applying algebraic methods to problems in particle physics.

Transcript

OK, maybe we can start from your years in Hungary and ... from there.

You start this university education. As a matter of fact, I have to go a little further back.

OK.

Because the first time I remember that I had some mysterious interest in mathematics was at the age of eight. Then on the playground, I had a friend who was not particularly intellectual, but quite a bit older than I, and I asked him to show me some notes he had on algebra, and he did that, and he taught me for a few weeks. I found it exceedingly boring. It was nothing else but various brackets and parentheses involved and that was very boring stuff, so I was very soon disappointed, but its effect is that I started. Then in a continuous fashion, I started with mathematics at the age of twelve, when I just worked through partly books in the field, of high school, and also I was quite interested in mathematic recreations [???] and all that. That was my absolutely loved to solve problems in the verbal, I don't know how to call it in English, verbal problems which led to exact solutions and things of this sort, and I know that I was quite intrigued by the idea that mathematics is something [???] that can tell you about the truth. When I was fourteen, I started in to learn some elementary calculus. And so I remember very clearly this interest. When I came to university, I was a little bit featured as a child prodigy. Budapest is not such a big place, and all that, that's why I became very well known.

When was that?

1925.

'25.

'25.

And that was the University of Budapest?

University of Budapest. And University of Budapest was not terribly good in mathematics. Lipot Fejér, well known for his fundamental theorem on the summation of Fourier series, was there, whom you know, who was a first class mathematician, but he didn't like to teach very much, so it was, his classes really were not that interesting, although there was a cross-exchange with Institute of Technology, and there were some very good mathematicians there, and I walked over the bridge with a group of students from the university to the Technology, and had their special courses. I particularly remember a seminar course by a Professor Kürshak who at that time was an old gentleman, quite close to retirement, and you may know actually as the result of such activity, that year we started to work out the problems of the Eötvös competition, the Hungarian problem book. It was actually published.

I wasn't aware of that.

I think I have one here. It was the famous competition for freshmen, in elementary mathematics, but sort of a sophisticated type, and in that year, myself, Edward Teller, who was Rudolf Fuchs — there was actually a third man, who was an engineer and a dear friend of mine, who was killed by the Nazis, so we were —

That was in 1926?

'25.

'25 again.

That year, yes, because I entered the university —

— oh, it was while you were a freshman.

I was a freshman, yes.

I see.

It was '25

So Teller was a freshman at that time as well?

Yes. Teller was also a freshman. He was at the Technology at that time, but he stayed — after the first term, he went to Germany to study there. He wanted to study either mathematics or physics, and his parents forced him to become a chemical engineer, just as Wigner was a chemical engineer. In Hungary at that time, the idea was, if you wanted to become a scientist and make a living, then you had to be a chemical engineer.

That's right.

And Teller I think went to Karlsruhe after that, and was a student of — I have a blockage on names — and — it's a very well-known X-ray physicist. He died just a couple of years ago, about 90. Peter Paul Ewald.

Ewald.

Ewald. He was in chemistry and got in contact with Ewald, and Ewald persuaded his father to let him switch to physics, as I mentioned and he started to work on physics and we got together again in Leipzig at first with Arnold Sommerfeld in Munich and then with Heisenberg. I will come to that later, in 1930, in Leipzig. So at the university I had a very nice time, in one sense. As I say, I was rather played up as a celebrity. But on the other hand, I went through a slump. I really found I am not that interested in mathematics and not that good at it, and I was good in problem solving. I was not good in inventing problems. And I was not good in a modern way of proving something. Theorems rigorously mathematical proof didn't appeal to me very much, problems of [???], so I felt uncomfortable with it, and after two years, I decided to quit. Now, some other background. My father was a book dealer, and — I had a younger brother who died in 1922. He was just past eleven then, not in his teens, and there was enormous pressure on that side, the father's side, that I should take over his bookstore. He had quite a prestigious book store. And I very often had worked there in vacation so I was quite familiar with book selling, and my mother was enormously against it, and —

Against what?

Against my becoming a book seller.

I see.

She wanted me to become a profession. Much later, she confessed to me that her dream had been for me to become a physician.

I see.

But I obviously had no interest in that, so she said, engineer or professional whatever, only not a book seller. So at any rate, I — in '27, I quitted the university, although I was very highly regarded and I was doing very well in my courses, so that was not a problem, and actually then decided to go to Germany and worked in a book store. But somehow that didn't work out very well. I tried to switch to Gottingen, and maybe be in a book store and somehow still try to get back to mathematics. So I didn't get to a book store but I did get back to mathematics, and after being there found it more stimulating, the Gottingen atmosphere sort of pepped me up, and I had in spring '28 very interesting semester in Gottingen, as a mathematician. I gave two seminar talks, one in applied mathematics. It was a very prestigious seminar by Prandtl, Director of the Institute of Fluid Dynamics in hydrodynamics assisted by the mathematicians Courant and Herglotz. Herglotz and Courant were sitting there in the seminar, and the subject matter was application of complex functions theory to plain two dimensional hydrodynamics, and I undertook to give a talk.

That was something which people used to do while they were undergraduates? They would give a talk in the seminar?

Yes.

Or was it more exceptional?

In Gottingen, the German system, there are really no undergraduates. The concept of undergraduate or graduate didn't quite exist. You entered the university for four years. Now, in these four years, if you took the correct examinations, you got a diploma for teaching in high school, carrying much more prestige than an American high school; and you earned also the right to try for a PhD thesis, and there were some very talented and fast working people who made their PhD, either at about that time or after the fourth year, maybe another year. For instance, Teller got his PhD in 1930, so five years. And certainly, too, it was a very free wheeling system, in Europe in those days. There was no question if you wanted to go to the seminar, you could do it. And second part, which I gave also in respect to my future interests, there was a talk on topology, so it was a seminar run by Alexandrov and Hopf, the famous Alexander of the Russian [???], a very famous topologist, and [???] was among the junior organizers of the seminar, and I had an instinctive liking for things like topology. On the other hand, in retrospect, I must say that it was about the first possible combinatorial topology which I could have encountered.^{[1]} I have since then developed an algebraic topological interest, also an application of physics, so I have now taste and judgment in topology, in theory of [???] and things of this sort. I'm very much interested in that. But that topology was sort of an extreme problem set topology that I undertook to report on a paper which had the leaders of the seminar quite excited about the proof of the Jordan(?) theorem, that closed curve divides the Euclidian plane into two parts, same thing in the space. In fact I think it was more in space at that time it was done. And there was a method of finding the relation which at that time was very much in fashion, extremely complicated, and I was intrigued by it and did it, and it was found acceptable and OK, but afterwards I had kind of a real let-down, of sort of proving this very difficult, theorem of Jordan which seemed intuitively obvious, which was in this time the height and the [???], and I thought of that. This was not particularly [???] but it was very interesting to have gone through.^{[2]} It was also very interesting to go through the hydrodynamic seminar, because I had to do quite a bit of analytic functions theory, and things like that. However, what happened, about the same time, I somehow got acquainted with a young Danish physicist, Mogens Pihl from Copenhagen, who was a kind of enthusiast and was in love with quantum mechanics.^{[3]} At that time, I didn't even know about quantum theory.^{[4]} That was in '28 — in Budapest it hasn't penetrated yet, quantum mechanics. Also I wasn't interested in physics. I was interested only in mathematics. Physics in Budapest at that time was extremely bad. Maybe experimental physics, I don't know, but theoretical physics was not at all interesting. A change came at about that time. A new professor, Ortvay, of theoretical physics was appointed. By that time I'd left already Budapest, and so in vacation I visited the university and got acquainted with him. He started a seminar in quantum mechanics in Budapest. In fact, I was the first speaker to talk at the seminar, and I talked two or three times later. So let me get back to it.^{[5]} Spring, '28, in Gottingen, I knew nothing about physics. Now, Mogens Pihl just knew quantum mechanics fantastically, said, "You have to learn that, that is the thing." I knew nothing about it. He said that in the fall, beginning October, that here Max Born would give a course on quantum mechanics, and I should take that. And meanwhile, I started to read up in a few books, a compulsory thing that I had to read in the summer time, [???], and then some book on analytical mechanics. I read Planck’s books on theoretical physics, the introduction, which are not very good books really, although I have very high regard for Planck, but the books were not very good. There was one book which was not very prestigious but I think extremely good for a first introduction. That was Arthur Haas. I don't know if you know the name. He was quite well known at that time for his popular writings on physics which were very good, and he had also another claim to fame, that in 1911 or 1912, he published an article or submitted an article on the interpretation of the hydrogen spectrum in terms of quantum theory, practically very closely almost identical to the Bohr.

To Bohr's —

— but this got him into complete disrepute with the University of Vienna, and he was refused "habilitacion?" because of this obviously pathological tendency.

I see.

So, by the fall of '28, I started on the course. It turned out that Max Born was sick at that time and didn't give his course himself. I talked about it the last time, actually. I don't want to repeat things —

— yes, yes —

— (crosstalk) so I don't want to,^{[6]} so I mention that, my experience with the course in quantum mechanics which was given by [???] and it struck a very strong chord in me. You can see from my present work, what really interests me is the mapping of reality onto mathematics, and I recognized in this case that there was something of this sort going on, even though the details, I couldn't understand, no one could understand, in fact the official line was at that time that it cannot be understood.^{[7]} And nevertheless I was hooked on it. I remember, for a while I was still in this idea of switching to physics but didn't quite dare to say it, because I had such a background in it, but I learned furiously at that point, took a large number of physics courses in addition to some mathematics, and I remember that at that time, someone told me that he had heard that I switched from mathematics to physics, and I found it very interesting that people think that it is not an absurd thing to do, but if it's so, maybe I should do it. And I did. So that was fortunate.

What was the situation among the mathematicians so far about the new quantum theory?

Well —

— were they interested? Was there discussion among themselves or what?

The mathematicians were intrigued by quantum mechanics, but not seriously interested. An exception was Hilbert, but this was mostly past by my time.

Yes, maybe you should repeat it.

Yes. Well, I was in touch with mathematicians during all that time. My switch to physics didn't mean that I cut off my mathematical connections. I went to mathematical colloquia, and it happened just about at that time that Hilbert who was really sick and was off of professional work for several years, I don't know exactly —. I tell you about an interesting event that I personally witnessed. For the last few years before my emergence in Gottingen, Hilbert was seriously ill of pernicious anemia and was away from his duties. Eventually he was successfully treated with an experimental American medicine. I was present in the Mathematics Club on his first return visit. He started with a little speech of introduction, that, you know, there is here a colloquium going on on the structure of matter, which incidentally was officially run by Max Born and Hilbert, it must have been before, by that time, he did not participate, but he did refer to this seminar, and he said, "Well, you know, in the seminar they talk about some newfangled mechanics. It is quantum mechanics. It is like Newtonian mechanics only it is different. And then as mathematicians, we are not getting around to dealing with this matter and studying it, but of course, one might ask whether the time is ripe or not, and recently I wanted to buy a radio, and asked an expert whether I should buy now or wait, and he said, well, better wait a couple of years. Maybe this is also the best response to quantum mechanics.”

I see.

Although Hilbert’s active period with respect to quantum mechanics was before my time, I was told by the math staff an anecdote that rings true to me that I would like to save for the readers of this interview. The issue is the contribution of Gottingen mathematicians to the foundations of quantum mechanics. A landmark event is the publication of Courant-Hilbert Methods of Mathematical Physics in 1924, vol. I, first edition. This volume focuses on linear vector spaces, function spaces and eigenvalue problems, all to become standard mathematical tools of quantum mechanics, but a year and a half before this theory emerged from Heisenberg’s bold speculation. I know that more than one mathematicians hoped to use eigenvalues in order to restructure Bohr’s old quantum theory into a mathematically satisfactory form. Although the expected role of high powered mathematics was born out in the forthcoming turning point in the conceptualization of physics, the details of radical change were unpredictable. Remarkable as the invention of eigenvalues might have been a surprising present for the hardnosed empirical physicist, there was a stubborn disharmony between fact and idea that had to be resolved to render the invention useful. A characteristic feature of the Bohr atom was that the energy spectrum of the H-atom was a combination of a discrete and a continuous spectrum and the two were separated by a limit point of the discrete spectrum. The difficult of the Courant-Hilbert volume is the limitation that all the eigenvalue problems converge to infinity leaving no room for a continuous contribution to the spectrum.

I see.

The turning point from the old quantum theory to a mathematically sound formulation was brought about by independent initiatives in the physics community. The historical priority is Heisenberg’s and he worked in the algebraic context of Hamilton-Jacobian mechanics, as did Born, Jordan. A class in himself is Schrodinger who combined this mechanical background with the eigenvalue method. I was fortunate to have been told of Hilbert’s, apparently unpublished, comment on Schrodinger’s pioneering paper: “Who would have believed that a singularity will bring the infinite limit point of the eigenvalues to the finite.” Although Gottingen was the prime example for the mathematical contributions to physics, the contributions of the respective specialities were characteristically different. Hilbert noticed the role of the singularity that extended the scope of eigenvalues. For Schrodinger, the Coulomb singularity was built into the Hamiltonian. The decisive new idea was for him de Broglie’s wave-particle duality. Another Hilbert saying that made the round was: “physics is much too difficult for the physicists.” In fact, however, it was the physicists who had the instinct for physics. Our social contact with the junior mathematics faculty kept the Hilbert anecdotes flowing. When Hilbert was told that Neugebauer turned to Babylonian history, he complained: “but history, all that had happened already!” Let me return from this excursion into history to 1928 and continue with my apprenticeship. The one-sided exposure to matrix mechanics called for other inputs. I turned to reading the papers of Schrodinger on wave mechanics and of Dirac on transformation theory, on photon processes and the relativistic electron. Johnnie von Neumann’s three brilliant papers on foundation in the Gottinger Nachrichten. These are believed to be superseded by his book on foundations, but I feel that readers of German are rewarded by a freshness of the papers somewhat lost in the book

How long did you stay in Gottingen?

I stayed in Gottingen until the winter 1930, and at that time, my friend Edward Teller came to visit. My friendship with Teller started from participation in the Eötvös competition, and then it was not very close because he soon went to Germany and I stayed at home, but every time that he came home, we did meet, and we had some common friends and all that so we had definite social contacts between us, and he came to visit in the winter of 1930, and participated in some seminars for a couple of days, and then he told me, "Well, this is a proper place, that was good at one time, but nothing goes on. You have to leave. You can't do anything here. Why don't you come to Leipzig?" And that's what I did, for May, June, July of the summer semester, so 1930 I spent in Leipzig, and since I was in close contact with Teller, who is a little younger than I but actually was much more advanced, and he had just finished his, got his degree, during that time, the work he started with Sommerfeld, but actually he finished his thesis while in Leipzig with Heisenberg, and so it was enormously instructive. And in — I have many recollections from Gottingen which were very stimulating, a learning experience, was concerned, I don't know if I talked too much?

No, I don't remember.

I think I certainly should mention here.^{[8]} For instance, in, I think it was in the fall of ‘29 that Ehrenfest spent some weeks or months in Gottingen, and he was a fantastic person. He had a restless desire to understand everything. He appealed to me enormously, in the sense that he really discussed things, I mean much more than anyone else was willing to do, and he had seminars going on from 9 o'clock till midnight, and he got into discussing the current issues in quantum mechanics, the relation of Shrodinger's wave mechanics and matrix mechanics, and I remember still his final conclusions — wave mechanics is a [???] so Schrodinger's thing just doesn't work. You see, at that time, that didn't mean that you shouldn't use a Schrodinger equation, but that Schrodinger's big idea of a realistic interpretation of the wave concept, which Heisenberg found an abomination, that it simply wouldn't, didn't work. And that was discussed then. I remember very much at one point, he raised the following question. Is it all right to speak German? You don't speak German?

I don't, no.

Because I remember the words.

Well, maybe you can repeat them and we can translate them.

Yes. (Speaks German here…) I mean, that was the German he used. We had learned in school that our physical quantities must be tensors, and here comes this Dirac and has his [???] wave functions which are not? What are they? And when Dirac was in the audience, and [???] too, he came back, (this is a spinor) and I remember that he presented, which he invented, I remember dotted indices, dots that either disappeared or they shouldn't have been, something like that, so it was a pain, and it was very interesting.^{[9]} I think here is again a historical accident, that [???] definitely influenced mathematical physics, but — the basic difficulty with this problem was that spinors had the middle ground between quantum mechanics and relativity. Now, at that time, like most mathematicians interested in physics, was very well versed in relativity, and not yet familiar at all with quantum mechanics. Of course Wandervalden then wrote a very interesting book on quantum mechanics a little later. And there was a basic decision to make, whether to start from the quantum mechanical Spinner concept introduced by Dirac, [???] and Katz, which were in that form not directly applicable to relativity, and to make them applicable to relativity, or the alternative thing was to take the tensorial method of forming invariants, and start from there, and Wandervalden did just that, and he complexified the formula to make it appropriate for quantum mechanics, and this complexification he undertook in a fashion which was unfortunate, namely, in his Spinor calculus, the transformation, wanted to consider the complex form, complex conjugate form of a transformation matrix, and he takes the complex conjugate of the elements, he has then to take the Hermitian conjugation and in my present work I'm very deeply involved in that, and the Hermitian conjugation, you can see, the conjugation plays a very important role. And the fact that it was so hard to understand, so you simply say "I cannot understand quantum mechanics," makes it impossible for the present young generation to change it, because all they have is by rote rules established at that time, and since they don't understand it, therefore there is no standard by which an understanding can be established. So that's just one of my ruminations about the state of the situation. I am not competent to tell you that. That is, first of all, it's not for publication. It should not be published before I have proved my point. In other words, what I'm saying now will be allowed only if I have proved that I can do better. Let's put it that way. But I am very glad to put it on record, in a way, how I feel about the situation. Again, I have a great admiration for Wandervaden, and SCIENCE AWAKENING is a marvelous work. But the fact is that I have a great admiration for Columbus, that he steered through the ocean in his caravels, and you don't blame him for thinking that he arrived in the Indies or Japan — but there is no excuse for us to try to maintain the same beliefs.

Yes.

And I think, I'm afraid that that is what goes on, at the^{[10]} –

Let's go back now to the semester in Leipzig.

Yes.

You spent only that semester in Leipzig?

Yes.

And then you went back to Gottingen?

No, I went back to Hungary. By that time it was 1930. I don't think that physics is essentially influenced by social economic factors, but my studies were, and there was a depression and my father said, that was enough, and I can't send you back, so that was it, it was finished. Your question is very much to the point because I haven't said anything positive yet about Leipzig, except that it was very fruitful, and it indeed was. Maybe I should go a little bit back, to Gottingen, because at that time it was clearly time for me to start on my thesis work. Now, the usual system is or was at that time completely different from the American. You decide to make a thesis, but it's entirely my decision and the professor's willingness to sponsor, to give me a problem. In other words, there was no qualifying exam, no exams of any sort, that would basically decide on my qualification. My qualifications were very low, terribly spotty, showing things, but I didn't have a real training neither in mathematical techniques nor in physics, and I was maybe not very well prepared for a real PhD, but I went up to Born and he at that time started to do a work to modernize the Born-Oppenheimer paper,^{[11]} which is a classic paper on molecular structure, and although it's terribly inelegant, but the substance of it is extremely important, and so he came back to it and wanted to, I had some idea to modernize it, and he wanted me to work on it. I don't think anything came out of it really even for him and certainly nothing for me. So I was not prepared to work on this difficult problem, and as I say, it was a problem most probably not very good, and that was it. And at that point, I went to Leipzig, and there —

Did you have any problems that you wanted to work on?

No.

You did not.

No. I was much too immature for that. And I'll tell you a little bit of a story, to what an extent it was not easy to have a problem, because nowadays probably it's one of the misconceptions of that time, of 1929-30,^{[12]} that the problems were just lying around, and I assure you that they didn't, and when I went to Leipzig and saw Heisenberg and asked him whether he could suggest something, I will try to reproduce for you what Heisenberg said. I think my memory probably for these times is fairly good, and so I'm pretty sure that what I'm saying is really true. So Heisenberg said, "You know, that's very difficult, to find a problem, because atomic spectroscopy is finished. Molecular spectroscopy is not finished, but I'm not interested in it, but Hund is, maybe you should talk to Hund. Solid state of course Bloch finished it. Maybe what you could do is to adapt the Bloch theory to thin layers, for two dimensional problems, and solve the Bloch integral differential equations in two dimensions." I said, I don't think he ever said that nuclear physics there is none because there was no nuclear physics at that time. The only problem in nuclear physics was the Gamow theory of alpha decay, and in fact I told you before that I gave a talk in the seminar in Budapest and that was on the Gamow theory of alpha decay. I think this is probably all what he said and in actual fact, I did look into the two dimensional layers, but my mathematical abilities were much too little to tackle it. I think his idea was to take the integral differential equations and which, in three big dimensions, if one dimension becomes small, then the continuum becomes discrete, and it's very complicated, and there are two things, my mathematical inability of dealing with such a problem, and a certain good sense that I think the experiments, I looked into the experiments on matter theorems, and found that they were very crude, because they depend very much on the type of microstructure of the film, and basically the main effect can be understood if one considers the mean three paths of electrons is influenced by collision, by a diffuse collision of the surface, and I wrote a letter to [???] to this effect. It was quite a simple minded paper which I did publish. I must admit that even in the short letter, there is a mistake which is corrected later. It was sloppy on my part. Now, the other thing was more interesting about molecular spectrum.^{[13]} Although I didn't go to Hund but to Teller, who, Teller at that time was primarily interested in some marvelous problems at the interface of physics and chemistry, in a very strong sense which, in spite of the fact that quite a number of quantum physicists who came from the side of chemistry, still, physicists did not have such a good sense for chemistry as Teller had, and he did some excellent applications of it to problems of physics, and then he suggested, not suggested a problem, he simply mentioned, noticed in the literature some interesting fact about the infra-red spectra of, the rotational vibrational spectrum of certain polyatomic molecules like methane or metal halides, where there is a very strong interaction between the angular momentum… Yes, so he noticed a problem that —have we talked about the molecular spectra?

I think we did, a little.

You know that from the rotational spacing you can get the moment of inertia.

That's right.

And that which of course we see no [???] symmetry, and the moment of inertia, it's ellipsoid, spherical, and should have only one moment of inertia. And what happens is that there are two infrared bands, and the two led to a sharply disparate moment of inertia, the moment of inertia, but one is twice as much as the other, so that again, it was very much in (crosstalk) — it was a surprising —

— phenomenon.

And that was well known. For instance, it shows again the different approaches of physicists and chemists. Thinking on this value, we have two moments of inertia, so it's easy to say, well, of course, the chemists were wrong, they thought it was tetrahedron but it's not at all, it's a [???] — it must be a asymmetric, asymmetric, it's a peculiar terminology, the symmetric top actually has two different moments of inertia, and if you assume that the methane is really a pyramid, then that could be explained, and theses have been written on that, on working out the methane on the base of this elongated form. But now here came into play that Teller was enough of a chemist to say, that is nonsense, could not be, and I remember that we had sessions, and it was a fascinating problem, no question about it. It was not only methane but the metal hallides, and again the metal hallides, there is a region(?) of rotation around the figure(?) axis, and again, for this one type of rotation, there were several moments of inertia which came up, so here you couldn't even work with the same device as asymmetric. There was no more symmetry to break, and it was completely open to question. And I don't remember all the details, but roughly it was like that, and Teller came up with all sorts of ideas how it could be, and somehow I said, "No, that couldn't be," and this sort of thing, and shut it down. The initiative always was his, he suggested something, and just from common sense I started to say — again, in this case, he wanted to show that there is a difficulty but there must be a solution, and I was in this case the devil's advocate and said, "No, I don't believe that it is the solution." Finally, he came up to the idea that what happens in this case that rotation would work mainly on the metal halides, it was a simple problem because it was a rigid rotator, and the rotation, the rotation of course had an angular momentum, but the vibration also had. Teller dug up an idea which I think originally was advanced by Wigner, that in a symmetric molecule, that there is a symmetry between the different vibrational states so that the superposition of the X vibration or Y vibration gives the polarization, like it's circular, superposition, like a polarized light, you can have a circular polarization, so if you have a sealed tube of molecules, then the vibration goes like horizontally and vertically to a superposition and you have a circular vibration, and the circular vibration has an internal angular momentum, and this angular momentum can interact with the rotational angular momentum, and then maybe this would simulate a variation in the moment of inertia. And that turned out to be correct. And so then we worked on this problem, and we published it. Apart from this [???] this was my first published paper, with Teller on that, and it was actually quite an interesting problem. Unfortunately it was not quite correct. We made some mistake which Teller discussed it then with molecular physics people, and I forget now the particular idea but it didn't enter very much. We made one wrong assumption about that internal angular momentum of vibration did not have to be quantized, and it could be a non-integer, and coupling constant there, and then Teller alone finished that paper. In other words, our paper is in the JOURNAL de PHYSIQUE but it is not entirely correct. Then Teller did not publish a second paper on it, but he wrote an article in the and there he quoted it and he did it correctly, so that's what happened. But at any rate, that was a very interesting experience, sort of tackling a problem even if we could not, completely, but many aspects of which were correct. Also, it started me off on the infra-red spectra of molecules, and then I had one asset, which I could use, namely, while in Gottingen, I had attended lectures by Heitler. I don't know whether I did talk about it?

You did talk about Heitler, yes.

I did talk about it.^{[14]} So Heitler gave a very nice course, on statistical mechanics, which started me off also in that direction, and another one on group theory which was incredibly useful, because it was very hard to learn group theory at that time, and for instance, Heitler himself did a great deal of work on quantum mechanical applications of group theory, and they are incredibly complicated, but in his course, he took a completely different tack, and he picked out the simple aspect of it, and that started me off, and I'm grateful for it, for what I got from that. So it paid off already at the time that I knew the group theoretical aspect, for instance, that Teller didn't know, and again, few physicists knew. Some went into it like Wigner and Heitler, and really went into it, and [???] I suppose, but for most physicists it was too complicated and they didn't do it, and through this special introduction, I managed to get into it, and Teller wanted to help me to get started on a thesis, and the following situation developed. Placzek was one of his friends, and did some very beautiful work on the Raman spectra of molecules. He did the type of work half phenomenological which was very formative for me later, in all sorts of later matters. You are familiar with Raman spectra?

Yes. Placzek was at Leipzig at that time?

No, he was not. I think he might — not during that time. I think there was a meeting, some very well attended meeting while I was there, and he was a visitor.

Oh, yes.

Now, Placzek actually laid the groundwork for the Raman spectra of complicated molecules. There his idea was that, he introduced the polarizability as a theoretical constant — the polarizability could be written down from the Bohr-Kramers dispersion theory in quantum mechanical terms, but he did it phenomenologically, assuming the polarizability of the molecule as a function of molecular coordinates. He considered them as a – agon(?) states, agon vectors of some molecular vibration, quite analogous to the quantum mechanical agon states, but that was classical, so it was basically almost a classical theory where polarizability depended on these molecular coordinates, and then developed a sort of semi-quantum mechanical theory based on it, so that the selection rules in the infra-red were simply along with the dipole moment of some molecule, so the dipole moment is a vector, he considered the polarizability which was a tensor, and that gave very different symmetry properties.

Yes.

Now, Placzek wrote a paper in the Leipzig er Vorträge so that's the reason you might think that he was in Leipzig, and I suppose he was at some times, but there was this series initiated by Debye, and maybe, about this molecular problem, somewhat on the borderline of classical and quantum, and Placzek had that paper on the Raman spectra, and he derived selection rules for different symmetrical situations, different types of molecules. Now, he belonged to the people who didn't like group theory, and did all these things sort of on his fingers, so that he had long tables and he developed these things in a completely intuitive fashion, and there were very many mistakes in his calculations, and Placzek recognized that this problems should be handled by group theory, who said of course he should do it with group theory, rather than in such an intuitive fashion, the way he did it. And [???] also, so, first of all, I just worked through things what Placzek did and checked it out, and there were very many mistakes, so it was a point to straighten it out. Then there were problems of excited states, vibrational states, and there really one needed group theory, and also other symmetries. It was curious that the main symmetries which came up were the cylindrical symmetries of the [???] molecule, and then there are three, four, four, four, six symmetries of the trihedral(?) molecules, and then in principle there was also the icosahedron. Now, there was no molecule of icosahedron symmetry at that time, but for completeness sake, I said, let's do it, and at that time no one had correctly the calculation of the icosahedron group, so I remember I was amusing myself with it, more or less at the time, figuring out how to get the characters of the icosahedron group. It was very funny that two or three years ago, it turned out that the icosahedron symmetry is now fairly important in describing — (crosstalk )

— that's right, yes —

There is a locally fivefold(?) symmetry, and people actually did recompute it, I don't think anybody picked up my own computation of it, but the icosahedron symmetry became of significance at this point. So that became my thesis.

Which you had completed?

In Budapest.

In Budapest, I see.

Yes.

I see. OK. And then you moved to where?

Well, that was —

That was when? You submitted it?

'32.

'32.

So at that time there was the deepest depression, and I was not established.^{[15]} You see, all the others, Teller and Peierls and Bethe were all established and receiving fellowships, but I was sort of not really in the crowd, and I had nothing of the sort. I stopped in Budapest for a while. I worked in my father's shop and I got a job at the insurance company in actuarial mathematics, and I was sort of drifting around in a sense. And again, Teller decided that the thing for me to do, to get a job, to join Landau, and Landau was in Copenhagen.

Where was Teller at that time?

Teller at that time was floating around, partly in Copenhagen, partly in Rome, with the Rockefeller Fellowship, and I think something partly in Copenhagen, and in Copenhagen he met Landau, and he contacted some problems there and all that, and told Landau about me, and at that time Landau was in Kharkov in the Ukrainian Physical Technical Institute, which was sort of at the beginning very promising Soviet institution, and Landau was a very interesting inspiration, to be with him, and he was easygoing, and Teller said, "Would you take him?" "Why not?" And then I got the invitation in June, '34, to visit Kharkov. There was an international meeting, the [???] and a number of British, and French physicists, I don't think any Americans went there, but Bohr himself, Solomon who was a son-in-law of Langevin.^{[16]}

Was Kapitza there?

No. No, it was strictly theoretical. And so I went there, to the meeting, talked to Landau, and he was sort of not very inquisitive, all right, sure, come on, offered me a research fellowship to be an aspirant, you know, the term aspirant, it's a doctoral candidate. Although I had my PhD, but Russia has a completely different standard. Doctor was a very high degree. Even the candidate corresponded to a PhD.^{[17]} At any rate, I had to take it again, so he offered me that, and in January, '35, I arrived in Russia. Now, in '34, when the whole deal was struck, the atmosphere was relatively relaxed. It was after the big collectivization crisis, and in '34 there was a good harvest, the first good harvest in many years, and the general feeling was that things are going better. In December, '34, there was the Kirov murder. By the time I came there in January, the situation was not nearly as relaxed, but then I was there. That's another thing, you know. I think it would probably not very interesting to go into the political affairs, it's probably not for that interview.^{[18]} But so I was with Landau, and I told you that originally when I was in Gottingen, I had a very perfunctory training in physics and mathematical physics, so going to Landau, I had a second and more thorough training in his group. He had his famous there, in other words, he had kind of a program.

That's right, yes. Did you pass those exams?

Yes.

The theoretical minimum?

Yes. Yes, I did, yes.

Apparently they were extremely difficult exams to pass.

A case of point was my test on thermodynamics that I ironically failed at first. The problem was that Landau’s idea of thermodynamics was utterly different from that of Born to which I had been exposed in his course. Born made a sharp distinction between the phenomenological and the statistical theories and he considered the former complete and no longer capable of new development. Landau downplayed the distinction between the two versions and, in essence, created a new discipline that he used creatively to arrive at new results. When Dau quizzed me in this new discipline, I did not know what he was talking about, and failed the test. I was freed from this deadlock by Pyatigorsky who gave me a short presentation of Landau’s new discipline. Not only did I easily pass the test, but I got greatly impressed by Landau’s initiative. This influence was reinforced by the fact that about the same time Landau wrote several papers on an innovative theory of the phase transitions that Ehrenfest has recently classified as a transition of second order. By that time I had already a reading knowledge of Russian and I was asked to translate Landau’s Russian manuscript into German for publication in the Physikalische Zeitschrift der Soviet Union, a journal which was edited at the UFTI. The cumulative effect of these influences was that I decided to try to orient my future research along the Landau type thermodynamics. It seemed to me a way to avoid complicated calculations as say in problems of renormalization.

That's right.

As I say, how you — (delight? crosstalk here)

— yes —

However at first I started on a research work, and Landau gave me a problem on pair production. See, I told you how difficult it was to get a problem in 1930, according to Heisenberg. After pair production came in suddenly there were almost as many problems as you wanted, you see, because pair production under different conditions, it was always a proper reach, in that it was routine but not easy, and sort of this early version of quantum electrodynamics. It became a quite different situation.

Yes.

So there were many problems, and nuclear physics started. I was not interested in nuclear physics, however, but we got problems in pair production.^{[19]}

Were there any other foreigners at that time in Krakow? Any other people from Europe?

Yes.

Who were?^{[20]}

Houtermans was there. Fritz Houtermans, he was there. Originally a theorist in astrophysics, but he worked there in nuclear physics. He started off with neutron physics, started off. And Fritz Langer was a German from the German electric company. He had a famous work on the multi-[???] of aiming the lightning in the Alps and devising high pressure, you see, high tension. At that time it was very difficult to reach high voltages, and he then on that basis you see, with his experience with high voltage, developed that method of taking condensers in parallel and charging them and switching then to series and getting high voltage, and that was a method of getting particle accelerators and he built that up in Kharkov. Then there was Alex, the Viennese engineer, who introduced low temperature physics for industrial application, and actually they built a separate institute for that.^{[21]} The idea was to use industrial gasses which they have in the Ukraine and [???] in that work. Metallurgical labs. And the various cold gasses which were byproducts of that, he convinced the government that by cryogenic methods, they could be liquefied and [???] and then used, and so that went on, then. [???] — you must know that.

— oh, their [Martin and Barbara Ruhemann] classical book —

Marvelous book.

Marvelous book, yes.

So they were both there, and I was very close with them.

I see. What was Landau's involvement with all the people who were working there?^{[22]} I mean, was Landau working in common with his people? Was he available? Did you come to see him, with the examiner?

Well, they are two different problems. Of course Landau had his own theoretical group, which was a close-knit group, and he was very much available to them, weekly journal seminars. He went to the library and looked over every week the journals which came, and checked them off, and then there were these journal seminars, where about four or five papers were reported, at each session, and he gave problems to everyone.^{[23]} He started to work on his textbook of theoretical physics at first with a number of people. Everyone dropped by the side. Only Lifshitz remained, and that was the Landau-Lifshitz team that became enormously effective, in the sense that that was behind closed doors, I don't know exactly what happened. I only saw phenomenologically what happened, that Lifshitz arrived to have his draft of paper, and showed it to Landau, and Landau gave lectures and he wrote it down and he came back and showed it, and then cut and paste, cutting the papers and gluing them together, and thus this course developed.

When you went to Kharkov was the time a little after, a few months after the London brothers had published their theory of superconductivity?

Yes.

Do you remember any discussions, were there any discussions in Landau's —?

— none whatever. And I think I told you the last time, when we talked about this thing, absolutely none. I never even heard about it. And it couldn't have been an oversight, because Landau looked through the literature, and it's a complete mystery to this day, a mystery, because everyone is dead who would clarify it. I know one thing, that Landau disliked both London and Heitler.^{[24]} And he was on excellent relations with Peierls. They were the best of friends. And to some extent, liked Teller, and I think with Bloch and Bethe —

— and Bohr, I think. He had high regard for Bohr.

Yes, that was another thing, high regard, but I'm talking about (crosstalk )... well, London was several years older than –

—than Landau.

Than Landau. Nevertheless I don't think he considered him as very much of an older generation. And I think that London's theory of superconductivity was so beautiful, and I think I would have imagined that it was a style which Landau should have liked.

Maybe however it was too phenomenological for him.

No, Landau's phase transition theory was completely phenomenological.

You're right. You're right. At that time you're right, yes.

Completely phenomenological. So that is not true. In fact, Landau was a master of phenomenology. And I can't, I don't understand why. And of course, you know it from our correspondence with London, that London was antagonistic towards Landau.

Yes.

Now, Landau was enormously abrasive. And he easily offended everyone he didn't like. So I have no idea whether there was an incident. I have no knowledge of it, but I somehow suspect there must be some reason why. Now, there is another case that I happen to know the history of a personal feeling of Landau. That was Christian Moeller(?) of Copenhagen. Landau said so, that he talked at one point to [???] in Copenhagen, and they discussed the relativistic two electron problem, on which Moeller(?) did his famous work, and Landau claimed that the basic idea came from him, and during my stay in Kharkov, Moeller came to visit, and Landau was absolutely rude to him and didn't talk to him. I talked with him at that time, which many years later resulted in an invitation of Moeller to Copenhagen which I followed in '63, so quite a bit later. So in this case, I know about it. I don't know with Landau and London, but I'm missing here something.

How long did you stay in Kharkov?

I stayed until '37. Now, Landau left Kharkov in February '37, and he went to Moscow the Institute of Physical Problems of Peter Kapitza.

That's right.

It was really I think founded for him. (...) I'd like to say in this connection — oh, you asked me about Landau's role in the Institute. So he was close with his group. Now, I think most people, for instance Shubnikov and Kikoin in cryogenics did consult him, and he was always helpful about it. Also nuclear physicists. But there were political tensions with certain groups in the Institute, which was very much split between the European-type intellectuals and the apparatchiks and all that type of a group, who worked in radar problems and things like that. I know that Landau had a very low opinion of ... (off tape) Now, I know that Landau had a very poor opinion of this radio group under Sluckin. Still and all, they worked for the first time with the magnetron. They worked on a magnetron. It became the central part of radar. I have no idea whether it was any good or not, whether it was related to the British successful version, whether Landau was right. Landau had very strong animosities, and sometimes he was right and sometimes he was not. So it is very hard to know. In February, he very dramatically announced at a meeting that he's leaving. Just the antagonism against him, between the conservative group became very sharp, and he just left. So at that time, so I neither wanted to stay in Kharkov nor could I have, because the whole political situation became very sharpening at that time, with various processes, and in '34 when I was there, foreigners were very much favored, and that was the very end of the first Five Year Plan and all that, and they were used to foreign specialists.^{[25]} By that time, it was turned around completely, and it was a very anti-foreign atmosphere, so I couldn't have stayed there, and I said I wanted to go back to Hungary, and they suggested, oh, why don't you try to look for a job somewhere in the Soviet Union? And all that. I didn't like the idea. ... I didn't like the idea, but I made some moves in this direction, but nothing was successful, and in June I returned to Hungary. It just occurs to me that I may mention about an activity in Russia which was in Kiev. Actually, it follows, that Guido Beck was an assistant of Heisenberg while I was in Leipzig, and we were working, rather a friendly relation, you know how it is, and he was at that time a professor in Odessa, together with Marcel Schine, cosmic rays man, and he invited me at one time to visit in Odessa, and give there a talk, something like that. We made a trip by boat from Odessa to Sebastopol(?) and then by car, to Yalta for a long weekend trip. And also, he had a guest professorship in Kiev, and he arranged for me to go for a month to Kiev. That was in fall of '36. And I gave a month long lecture course on optics, and at that time, I got acquainted with Nathan Rosen, part of EPR group of authors, who was invited then to Kiev, to stay there, and for this month I had good contact with him, and so that was in addition, but from the research point of view it was not remarkable, but it was a pleasant stay there, and I think that probably by itself, that period.

^{[1]}Topology comes in different varieties. Early in the century was point set topology and combinatorial topology. The two were soon to be synthesized int eh much more interesting algebraic topology. Unfortunately, this was to come after the time of my narrative and the combinatorial variety I was offered rather turned me off.

^{[2]}It was said that in combinatorial topology the obvious is false and the truth is paradox. I felt some satisfaction that I managed to penetrate a difficult proof along this line, but I didn’t feel like going on with this pursuit.

^{[3]}The rescue came from Mogens Pihl, a friend from Copenhagen who was an enthusiast of quantum mechanics and informed me that Max Born will next fall give a first ever course on this brand new subject. He made it clear enough that I could not be so foolish as to let this opportunity slip by.

^{[4]}It is hard to describe how far this was from any theoretical physics to which I had been exposed in Budapest. Although experimental physics was adequate within 19th century context, theory was way below the demands of the century.

^{[5]}A new era started in Budapest in August 1928 when Rudolf Ortvay was named professor of theoretical physics. He initiated a seminar on quantum mechanics in which I was to participate. He established a modern teaching program. This was too late for me. During the term vacation in 1928 I went through an elaborate reading program to fill in some of my wide gaps in physics education. Sommerfeld, Atombau und Spektralllinien was my mainstay, also the theoretical texts of Max Planck and Arthur Haas. The latter may be largely forgotten by now, but he had remarkable pedagogical talents.

^{[6]}So I had the minimal preparation to take Born’s course which was actually given by his assistants, a remarkable crowd: Walter Heitler, Leon Rosenfeld and Lothar Nordheim.

^{[7]}The course attracted me because it related higher mathematics with nature. Although I did not understand the details, and no one was supposed to understand it, I was confident that understanding will come in due time. I soon started to play with the idea of switching from math to physics. I was embarrassed by my ignorance, but when people started to assume that I have switched, I realized that this is not so absurd.

^{[8]}The seminars of the guest professor Paul Ehrenfest made a lasting impression on me. Among the issues discussed was the argument between Schrodinger and Heisenberg on the interpretation of the wave function. While this issue had been widely discussed by every student of the concepts of quantum mechanics, there was an issue that was first recognized by Ehrenfest to be worth of attention. He called attention to Dirac’s new relativistic theory of the electron with its four-component wave function. “We have learned in school that physical quantities are Lorentz tensors, but this wave function is no four-vector. What is it?” The young algebraist van der Waerden was in the audience and in a few days he came back with the answer: relativistic theories ought to be expressed in terms of the representations of the Lorentz group. The tacit assumption of the Einstein, Minkowski era was to consider the tensorial representations. Ehrenfest’s point was that the Dirac formalism does not fit this frame. The question was soon answer by van der Waerden: Dirac tacitly switched to another type of representation that van der Waerden called spinorial and he emphasized that the latter are two-valued, whereas the tensorial ones are single-valued. This seemed a reasonable mathematical account of the spin phenomenon. Adding spin to the orbit of an imaginary spinless electron leads to the doubling of the orbit.

^{[9]}Although the previous footnote was written in 2004, I trust it is a reasonably faithful account of the original proceedings. The present footnote injects a new point of view into the old issue. There is another way to distinguish the two kinds of representations: the tensorial ones are real, the spinorial ones are complex. This dichotomy played an important role in the canonical foundation of quantum mechanics. Applying canonical quantization to canonical phase space involves a complexification. This seemed at the time arbitrary, justified only by success. It was considered an indication that quantum mechanics was not a rational theory, only a strictly empirical account of atomic spectra. Since complexification is a natural part of the theory of Lorentz group, it would have been plausible to attempt quantization within the relativistic theory. This never happened during the early stages of the foundation problem, although the relativistic connection came to its own in the context of quantum electrodynamics. What is worth mentioning is that not only did van der Waerden fail to promote the connection of quantum mechanics and Einstein’s relativity, but he created an interesting roadblock. The complexification of a vector space formalism can take two distinct paths. One is Hermitian conjugation, another is the complex conjugation of the matrix elements. Canonical quantization operates with the first, while van der Waerden used the second in his spinor theory. This was quite unwarranted; with the spread of quantum mechanics the Hermitian method became the standard practice in the linear algebra of complex vector spaces. At the tiem van der Waerden was an algebraist, mainly in touch with Emmy Noether; he had a limited experience with quantum mechanics although familiar with the representation of the Lorentz group. This was to change later as his discovery of spinor algebra prompted him to write a book on group theory in quantum mechanics.

^{[10]}The account of the educational experience of Gottingen would not be complete without remembering the excellent courses of Max Born and Walter Heitler. Born gave an excellent course on special relativity, on a much higher level than his well-known popular book. Of particular importance for me was his course on thermodynamics to which I will come in context with landau’s thermodynamics. Heitler gave a very good course on statistical mechanics, and particularly important for me was his course on group theory. It paved the way to my thesis topic.

^{[11]}It is time that I turn to my problem of choosing a thesis topic. In this regard Gottingen was less promising than one might have expected it. There was a certain fatigue after the period of vigorous creativity. To me this was brought home in a personal context. Teller came for a short visit, partly to size up the place and partly to find out about my progress. He was quite brutal in his criticism of the current creativity. He suggested that I should transfer to Leipzig. I followed his suggestion and transferred to Leipzig for the Summer semester May, June, July. This was a good move. Not that the difference between the two universities would have been that great, but Teller passed his PhD exam the first day of the semester and he became available to generate his own research program of which I became part.

^{[12]}I went to Heisenberg to ask advice on a research program. He said that is not so easy. Atomic spectroscopy is finished; molecular spectroscopy is alive, but he is not interested in it. I might go to Frederic Hund. Felix Bloch finished the theory of metals but I might try a new wrinkle by generalizing his theory to thin layers. Gamov’s alpha-decay theory was amove toward nuclear physics without any further opening for the time being. I listened to the advice – up to a point. I wrote a short paper on the conductivity of thin layers. However, while Heisenberg envisaged a very sophisticated theory, I wrote a trivial phenomenological approximation that gave me neither experience, nor credit. Guiding me towards molecular spectra was sound, but instead of apprenticing to Hund, I had Teller on hand. This was much more interesting as Teller was about to open up a very original line of ideas. The concept of a “molecule” was the obvious link between chemistry and quantum mechanics, these disciplines had a very different approach to this concept. The chemist visualized a geometrical framework whereas the Schrodinger equation of a system of nuclei and electrons gave no support to geometry. This gape was largely filled by the Born-Oppenheimer approximation which separated the molecular motion into electronic, vibrational and rotational components. If the electronic state has a minimum energy configuration, this this is a justification of the geometric frame of the chemist. Teller’s guiding idea that he pursued during most of the 1930’s was that there are exceptional situations where the conditions of validity for the Born-Oppenheimer approximation are not exactly satisfied and the chemist will find that his geometric expectations will be replaced by anomalies of various sorts. Teller rightly felt that tracking down such anomalies will be more interesting than accounting for the routine behavior of spectra. Among the effect he studies was the Jahn-Teller effect involving a strong interaction of electronic and vibrational states. The anomalous mixture of electronic and vibrational states are also called “vibronic.” The case he presented to me at our first of many sessions involved a strong interaction of vibrational and roataitonal states. However, I am getting ahead of myself, this was already the solution of the problem. The beginning was an apparent puzzle.

^{[13]}An interesting type of a polyatomic molecule has a symmetry axis, two of its moments of inertia are equal: I1 = I2 ? I3. This so-called symmetric top has an interesting spectrum: its infrared vibrational-rotational spectrum consists of equidistant bands and the spacing of these enables us to determine the moment of inertia I3. A spherical top has three equal moments of inertia: I1 = I2 = I3. The case of the methane molecule belongs in this category. All infrared bands should yield the same spacing, the same moment of inertia. Teller presented me with the puzzle: CH4 had two infrared bands yielding widely different moments of inertia. Some physicist suggested that the chemist were wrong and methane is a pointed pyramid rather than a regular tetrahedron. Teller rejected this option out of hand and the hunt was on to find a reason for two spacings in a spherical top. In the absence of any clue this problem kept us busy for the whole term. Teller produced one idea after the other; I specialized in shooting them down. Until one idea survived: vibrations of symmetric molecules are degenerate, have an angular momentum that is strongly coupled to the angular momentum of rotation. This upsets the simple connection between spacing and moment of inertia and explains our anomaly. Working out the details of the new connection between spacing and moment of inertia and explains our anomaly. Working out the details of the new connection between spacing and moment of inertia took us the better part of the semester. We had the satisfaction of having solved a puzzle by recognizing an anomalous interaction of rotation and vibration: an early manifestation of what is now called “entanglement” of wave functions. We got a paper published in the prestigious Zeitscrift f. Physik. Later Teller published a somewhat improved version int eh Hand und Jahrbuch der Chemischen Physik. We had a good departure for new results. Teller unearthed more subtle results for the limitations of the Born-Oppenheimer approximation, he discovered the “entanglement of vibrational and electronic eigenfunctions (‘vibronic states’).” Insofar as I was concerned, I was primarily interested in locationg a problem of my own for a thesis subject.

^{[14]}The vibrational – rotational spectra of polyatomic molecules seemed to be a plausible choice for further exploration. Happily, these spectra come in several varieties: the infrared “active” and “inactive”, which do and do not produce observable bands, respectively. A different phenomenon is Raman scattering, in which a primary photon is scattered into a secondary line, the frequency of which shifted by a vibration – rotation quantum; hence a roundabout connection with infrared bands. Raman bands are also classified into observed “active” and unobserved “inactive” bands. Wigner showed how a group theoretical method enables us to classify the vibrational states of polyatomic molecules. The same method can be sharpened to yield the activity of the infrared and the Raman bands associated with the specific representations or their group characters. As a result, the group characters turn out to be essential tools for the interpretation of the spectra of polyatomic molecules. Having attended Walter Heitler’s course on group theory made me into a passable expert in group characters and I turned out to be well prepared to bring to bear this knowledge to the interpretation of the spectroscopy of polyatomic molecules.

^{[15]}I earned my PhD and the paper appeared shortly in the Z. f. Physik. An important landmark was passed, but I was without a professional position. I worked partly in my father’s book store and then as an actuary in an insurance company. A more interesting opportunity was engineered by Teller who met in Copenhagen with Lev Landau. They initiated a joint work and he also suggested to Landau that he might invite me to join his newly formed theory group at the Ukrainsky Physico Technichesky Institute (UFTI). At that time Landau was traveling in Germany and Copenhagen. Although I did not meet him during his travels, I was much aware of him during my Leipzig stay, because a remarkable paper of his on the diamagnetism of free electrons circulated there as a manuscript and created a great deal of attention. It was a remarkable result that the classical theory yielded a rigorously vanishing value for the diamagnetism of free electrons, whereas his rigorous quantum mechanical theory yielded a non-vanishing result.

^{[16]}The first result of Teller’s intervention was an invitation to a small international meeting on theoretical physics to the Ukrainsky Physico Technichesky Institute (UFTI) in May-June 1934. The most prominent member was Niels Bohr, otherwise there were younger theorists of some reputation from France and England. The recent discovery of electron pairs rendered the techniques of calculation of pair production, dealing with the gamma matrices, among the hottest issues discussed at the meeting and among the most active programs animating the theory group for a couple of years to come. A similarly formative experimental event was the discovery of the neutron which marked the beginning of an entirely new branch of nuclear physics. However, the UFTI meeting of 1934 was not only a scientific landmark, but noteworthy also from the political – social point of view. It marked also end of the agrarian crisis that had devastated the Ukraine for some 2-3 years. For the first time in many years a bountiful crop was harvested. The feeling of relief was palpable among the local contingent of foreigners who were my natural contacts. An important corner seems to have been turned. Was it really? Production and distribution of food has indeed been licked. Famine did not happen again until the sieges of the next war. However, the automatic expectation that this will bring about the mellowing of the regime, failed miserably. I returned to stay at the UFTI in January 1935. Since my earlier visit the Leningrad party secretary Kirov was assassinated and it triggered the sequence of staged trials. The atmosphere has radically changed since the previous spring. However, my main topic is Landau’s part in my professional development.

^{[17]}The theoretical department of UFTI was centered around a small well selected library that reminded me of the old mathematics and physics library in Gottingen in the Aula on Weender strasse, based on open circulation. (I still experienced the move into the new Mathematics Institute in Gottingen built on Rockefeller money.) There was a suite of rooms, one for Landau, affectionately called Dau. The rest for the staff, two desks in each room. The staff was Eugene (Zhenya) Lifshitz, Alexander (Shura) Akhiezer, Pyatigorsky. We were soon joined by Pomeranchuk, briefly Chuk, Dau’s favorite, he reminded him of his own youth. A weekly routine was for Dau to survey the new arrivals in the Library and distribute them for 15 minute reports among the staff. Dau pronounced his authoritative judgment in which he was most impressive. Another newcomer was Misha Koretz who was not much of a physicist, but Dau considered him his confidant in making personal and professional decisions.

^{[18]}The professional decisions turned around Landau’s standing among the leading theorists in the SU. He was at a crossroad. He was a promising graduate student in Leningrad. He was a junior compared to Fock, Tamm, Frenkel. He formed a trio with Gamov and Ivanenko, jeunesse doree, prepared to poke fun of the establishment. At the same time becoming head of the theory group at UFTI put him on the road to becoming a senior player. He seriously pondered the best policy to bring this transformation about. The idea was to exploit the constructive ideas he had on the reorganization of physics based on the great innovations of the century: to use special relativity (SR) to integrate classical electrodynamics (CED) with the canonical mechanics of particles (CMP); to use quantum mechanics to integrate macro- with microphysics and chemical thermodynamics with atomic mechanics. Whereas the dominant view was to stress the gulf separating the new physics from the classical tradition, Landau’s great idea was that the new physics forms the basis for an unprecedented unity. In order to make a case for this new point of view, Dau had a long-range strategy and a short-range strategy. The former is to write a multivolume text of theoretical physics in which the new unity is spelled in great detail. The short-range prerequisite for this program is to train a staff who would be capable to put Landau’s vision into print. At the time of my arrival a synopsis for this training program was available in a print-out of about ten pages. It was briefly referred to as “teorminimum” and consisted of eight segments to be a syllabus for as many oral exams by Landau. If my member is correct, the list was about as follows: 1. Classical mechanics. 2. Classical electrodynamics and special relativity theory. 3. Statistical and phenomenological thermodynamics. 4. Fluid dynamics. 5. The electrodynamics of condense media. 6. Quantum mechanics. 7. Relativistic quantum mechanics. 8. Gravitation theory and gravitation. It was an incredible opus summarizing all of theoretical physics. Taking the “teorminimum” was the prerequisite to become a member of the group. Since I had my PhD acquired elsewhere I was not presumed to be under this obligation. However, this was a unique training program and since my actual training had been spotty, I rather soon decided voluntarily to take the test. Landau kept a careful record of every successful candidate. I was #5 on his list and the only outsider of the SU; thus my report is of some general interest and it was vital for my further professional development.

^{[19]}This was not a program for immediate use. As I finished my teorminimum, Landau suggested a problem: Pair production on beta-decay. At the same time Akhiezer was assigned a much more difficult problem: “Scattering of light by light.” It involves a fourth-order perturbation calculus, whereas my problem was a much simpler first-order perturbation. We both finished our assignments sometime in 1936 and received our “candidate” degrees. This is somewhat equivalent to the PhD in the West.

^{[20]}Fritz Houtermans arrived shortly after myself. He was a German, but his best known previous work on astrophysics Atkinson. At UFTI he was engaged to initiate neutron physics. He pursued this program very successfully. The problem of neutron sources was a weak link of the program. The first solution was an expediency. A member of the Leningrad FTI brought by plan a radon source with a half-life of four days. In the long run it was hoped that Fritz Lange would produce a high-energy device. Lange had been before at the AEG, the German GE. He became famous for capturing high voltage from the lightning discharges at the Monte Generoso in the Alps. He hoped to achieve high voltage at the UFTI by charging banks of parallel condensers with available voltage and then switch the condensers to series. To start with at first he was building large numbers of condensers.

^{[21]}Then there were the older timers. Among the oldest in seniority was Alex Weissberg, a Viennese engineer of great imagination. Some years ago he convinced the relevant Commissariat that UFTI, which already had a first class cryogenic lab directed by Lev Shubnikov, should be supplemented by an applied cryogenic lab (OSGO). The guiding idea was that the metallurgical industry releases into the air industrial gases which could be captured, cooled and separated into constituents to create the basis of a chemical industry. Alex was commissioned to construct the physical plant to house OSGO. Martin and Barbara Ruhemann, two of the low-temperature physicists of UFTI were slated to direct the new lab after its completion. While waiting for this to happen they wrote their successful book on cryogenics.

^{[22]}Yes, Landau and Shubnikov were close friends and Landau kept a close watch on experimental low-temperature work. There was also a number of nuclear physicists who relied on Landau’s guidance. I personally never got involved with nuclear physics, but kept in touch with the cryogenic people.

^{[23]}Landau’s main interest was to keep his own group fruitfully occupied. I mentioned already his guiding the journal seminar, establishing the teorminimum and by keeping it as an active training principle. However, the long-range goal was the writing of the multi-volume text on theoretical physics. None of this was published during my stay at UFTI, but the writing was already proceeding. A selected member of the staff disappeared behind the closed door of Dau’s room, to reappear later with a long scroll generated by cutting and pasting the original draft. In the pre-computer times this was achieved by scissors and glue.

^{[24]}In the interview I expressed surprise that Landau never responded to the London brother’s phenomenological theory of superconductivity. I may add as even more puzzling that he completely ignored the remarkable 1934-45 Oxford work on liquid helium by Francis Simon and Fritz London. On writing these comments in 2004, I venture a guess on the origin of the Landau – London mutual dislike. When the two met in Germany around 1930, London went through a failing effort to use complicated group theory to broaden the Heitler – London theory into a basis for quantum chemistry. Assuming that London tried to persuade the visiting Landau of his point, he certainly did not get a sympathetic hearing. I do remember that he voice his low opinion of Heitler and London, whereas he had very positive relations with Peierls, Bethe, Bloch, Teller, Weisskopf, Placzek, Wigner, Onsager, Ehrenfest, Uhlenbeck, Casimir, and Kramers.

^{[25]}Gavroglu did not ask me about the political situation, but I have to give at least a short summary, because it was essential for my leaving the Landau group and to lead to my new involvement with London. There were ups and downs. In the spring of 1936 the so-called Stalin constitution was published. It was pretended to be a liberal – democratic document and for a short while all the gurus as Landau, Weissberg were taken in. It took some weeks before it was recognized to be a sham. In a few months the change affected our little circle and Landau was fired from his professorship at Kharkov University. As a protest a whole UFTI group submitted their own resignation. (I was not affected because I was teaching at a Technical School, rather than at the University.) This was worked up into a serious affair that was discussed weeks on end, involving bitter attacks against Landau. In the midst of all Landau got up and said: “I quite UFTI and leave next week for Moscow.” It was a bombshell and the tension further increased with the arrest of Weissberg, Houtermans, Shubnikov and others. The UFTI I had known was no longer. I followed up some recent job offers, mainly in Kiev where I had given a guest lecture course, but the entire job situation for foreigners had radically changed. I applied for my exit visa and in June I was headed back to Budapest.