History Home | Book Catalog | International Catalog of Sources | Visual Archives | Contact Us

Oral History Transcript — Dr. Laszlo Tisza

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.

Access form   |   Project support   |   How to cite   |   Print this page


See the catalog record for this interview and search for other interviews in our collection



Interview with Dr. Laszlo Tisza
By Kostas Gavroglou
At Tisza's Office at MIT
January 1, 1988

 
open tab View abstract

Laszlo Tisza; January 1, 1988

ABSTRACT: 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

Session I | Session II | Session III

Tisza:

Yes, to start, but you wanted to start, shouldn't we stop it when there is nothing happening?

Gavroglu:

OK.

Tisza:

Where is the button? ...

Gavroglu:

OK, now — the letters that I have between London and yourself are from 1946 to 1952. And there are only two letters before that. Now, these letters are basically between the Cambridge Conference of 1946 and the Oxford Conference of 1951, and they are really between London's presentation where he was so theoretical about it, and his admission that there are problems with the model. In between of course there is the MIT Conference of 1949. Now, before we get to the actual letters, maybe you can give me some background to your 1938 paper and your influence with Landau and your talks with Landau, and take it from there.

Tisza:

Thank you, that's a very good question indeed. Our correspondence cannot be understood without this background, and I would like to start, at least briefly, with my stay with Landau. I had my PhD when I went there, and it was in group theory, and I was very much searching about my place in physics, and Landau just wrote his phase transition paper, and that enormously excited me, because I felt that this is the type of mathematical physics which was congenial to me and which I could do, as complicated calculations of pair production, although actually I did a pair production paper with Landau, but a very simple one, and it didn't appeal to me very much, whereas thermodynamics I thought was great, and when I left Landau, he gave me a problem which was on exciton theory.

Gavroglu:

When was that?

Tisza:

In '37.

Gavroglu:

Where?

Tisza:

In Kharkov.[1] I was in Kharkov with Landau from '35 to '37.

Gavroglu:

Oh, I see.

Tisza:

In '37 Landau left for Moscow, and I decided to leave and go back to Hungary, and it took me a few months before that could be accomplished, but after I finished my pair production paper with Landau, he gave me a topic about exciton theory, about the spectra of solids. Now, that never led to anything. It was a not very well stated problem, and I couldn't do anything with it, but interestingly, I had an introduction to Landau's idea of exciton theory, so after I left him, I considered myself very much as a student of Landau's, and I was very much interested in pursuing the thermodynamic ideas, thermodynamical statistical ideas which I acquired in his group. When I left Kharkov in June '37, I spent a few months in Hungary, and then in September, '37, I went to Paris. That was an international congress there, Congrès du Palaòs de la Découverte to which it was easy to get a visa, and I went there, and through the help of Leo Szilard who wrote to Fritz London, I got then a very low level scholarship associated with the College de France, and I got very friendly with Michel Magat, a physical chemist, and with Fritz London. And so very soon, I got in close touch with Fritz. He was at a very exalted position then whereas I had a very lowly one, but he was pretty lonely. French physicists are not terribly particularly in professional contacts, or were not at that time. I think things have changed since then. He clearly enjoyed having me as a contact, and what he immediately started to do was to teach me about his superconductivity theory, of what he and Heinz London, and it fascinated me very much, because a very peculiar thing, that I never heard about it from Landau.[2] That is sort of an introduction to the later stories, that Landau seemed to have had a definite antagonism towards London, and he couldn't have overlooked the papers. His habit was to go to the library and search the literature every week.

Gavroglu:

Landau did.

Tisza:

Landau. And if he saw an article which seemed interesting to him, then he assigned it in his journal seminar, to one of us in his group, to talk about it, and London's paper never came up, and I would have thought it was very much in the phenomenological quantum mechanical direction that Landau should have appreciated, and it is remaining a mystery to me, why that wasn't the case. So in this sense, London's influence also came in very powerfully to me, and also personally he was very nice, and also he Landau was above the — high above [???] and whereas, and didn't talk about his ideas, except when he was ranting or so. You ask him a question, then he figured it out on the back of an envelope or so, but he didn't say what he was thinking about, what he didn't quite know yet, whereas London did that, and it was a wonderful thing. So that went on in September, October, '37. At that point, I would like to fill you in on something which you don't seem to know, that there was a van der Waals Conference[3] in Amsterdam, and London went to that, sometime in the fall, I don't know the exact date, and he came back very excited, saying that something marvelous happened, a young chemist that was Joseph Mayer developed a theory of condensation, and he was too ignorant to know that is an insoluble problem and he solved it, and no question about it, that it is this condensation paper of Mayer which directed London's attention to the Bose-Einstein condensation paper. I was impressed by the fact that it is a matter which couldn't be solved by sticking to hydrodynamics and the kinetic form of it, and just making another calculation, but something in the basic method had to be changed. And that was my guiding idea. And I had a crazy night, but I don't remember the details, and probably there was nothing logical about it, because various impressions and things I heard and all that sort of incorporated in a fashion I don't remember, and probably not very interesting, but by next morning, I had somehow the basic features of the two fluid model. But it was in an extremely immature and sloppy fashion. Well, I went to London, to tell him about it, and he became very disappointed and unpleasant about it. It is very understandable. I considered that although it was my idea, we worked together — it was not quite my idea, I considered it part of the common work. I worried about it, no one will give me much credit if I work with London, but that's how it is. However, London didn't see it that way at all, and he said, "That is all nonsense," and he just didn't want to accept it at all. Now, in his defense I must say that the presentation was enormously unorthodox and immature, and he had plenty of reason not to accept it. But no question, there was just an annoyance, that he's the senior man and he worked on this question for years and now comes a youngster and he's….

Gavroglu:

What was his main objection to your idea?

Tisza:

That an independent flow of two interpenetrating systems is nonsense. It's not possible.

Gavroglu:

But how did he reconcile with the fact that you were explaining qualitatively practically everything?

Tisza:

You see, he didn't address himself to this question. In fact, I had this experience in the years to come, that theorists with whom I've talked have certain standards, we should start from the usual Newtonian or the hydrodynamic equations or this or that, and making a completely new departure is simply not allowed. And in fact, to explain a phenomenon from assumptions was considered in a way sort of a dishonest. I remember a few months later, when I sort of desperately worked on the theory and tried to make it more understandable, we were at London's, and we played some game, some word game where you had to guess some sentence, and he posed a sentence to be guessed, that "I can explain all properties of liquid helium." So in a certain sense, the fact that this explained everything made it very suspicious to him. It was sort of just like, basically in such a fashion that, you could explain the fact, but not from quantum mechanical principles, which was of course absolutely correct. Now, during that time, Simon and Kurti came to periodical visits to Paris, because they worked in Bellevue at the observatory, which had a large magnet, which they didn't have in Oxford, and Kurti was a Hungarian, an old friend of mine, and we spent time together here and he introduced me to Simon. And I had a long discussion with Simon which I acknowledge in this paper on the Nernst principle, and so I had long discussions with him and learned an enormous amount about his paper on the zero point energy of liquid helium, and we discussed the paper itself. Now, Simon and Kurti at that time were into these adiabatic demagnetization phenomena and —

Gavroglu:

— was this after you sent it to NATURE?

Tisza:

About the same time. I can't tell. Probably I sent it off and it hadn't appeared yet, something like that. And I can't tell. But certainly I discussed it, and they were sort of amused. They considered it sort of ingenious, but sort of wild. And however I got an invitation to London, and there, indeed, I presented my ideas. In fact, I presented already the idea of the thermal waves, at that point. And I also spent about a week in Oxford and visited at the laboratories of Daunt and Mendelssohn who were very much engaged in liquid helium, which Simon and Kurti were not, and there was no question about it, that the two stream (fluid?) model made it possible to communicate the results of experiments. It was experimental. It was very well received, and in fact we could talk about the things in these terms. And we discussed even the possibilities of detecting total [???] waves. I don't know whether it came up, I don't know whether it was my idea or some of theirs, that we could have a paramagnetic salt in an alternating magnetic field, and produce periodical heating and cooling, and in such a fashion produce — I think much later, it was not [???]'s method, I think it might have been later produced. They didn't do it at that time. I think the official line is that the war came in between, but that was not really so. It was in July, '38, over here that the war started. It wasn't the war that interfered, it was the war between London and myself, because London presented his view that — oh, that is nonsense, and they didn't want to be associated with it really at that point, I'm afraid, at that point. Oh, I forgot one thing which is very important.[4] Shortly after my letter to NATURE, there appeared a letter by Heinz London which had one beautiful feature, and I have been always very sheepish about it, why didn't I think of that myself? The thermodynamic relation connecting the fountain effect, the entropy? temperature pressure difference. That was exactly the type of thing which I feel I should have done, but I was often sort of timid in going about and getting a quantity relation. My paper would have had a much better effect if I had had this relation. So I immediately accepted that. However, there was an unfortunate situation that Heinz London had expressed that in terms of the [???] Effect, and introduced the idea that in special properties of helium 2 are connected with the surface. It's a solid surface. And that was the main contention then between — and then Fritz accepted his brother's idea, and the next article which he wrote, I think '39 or some time, in the PROCEEDINGS, that he sort of plugged for this idea, and the main concrete difference in prediction was, my prediction of temperature rates which was a volume effect, and his prediction of the surface effect.

Gavroglu:

One small but rather important difference between London's Bose condensation paper in NATURE is that really London is interested in the lambda, whereas you yourself are much more interested in the newly found kinetic properties of helium. Your concern is more with viscosity and superconductivity, whereas London's concern is much more with finding discontinuities in the lambda transition.

Tisza:

Well, that came up very naturally. It's not quite that rigid, because London discovered, it was the re-discovery of the Einstein condensation.

Gavroglu:

Yes, that's right.

Tisza:

And that was a static effect. And so he was much more conservative. That is what he could prove. He had an actual calculation of that effect, and he could say "This is it," and I think the last sentence of his paper is that one can hope that this will have some clarifying effect on the kinetic phenomena, and in fact, then we talked about it. I am not quite sure exactly at what point it was, but I know that at that time we talked very much about the kinetic effects, so he was very much interested in the kinetic effects, only he felt, he didn't want to go into wild speculations. The Einstein condensation you see he proved mathematically. And he wanted to give himself more time to reflect on it. And in this sense, I can understand very well that he was very angry at my jumping the gun, because he felt that he had a solid idea, and instead of pursuing it now with care, I'm just rushing ahead, and I don't resent the fact that he didn't accept it. He had good reasons.

Gavroglu:

Were you encouraged by anybody before sending it to NATURE? Because despite London's criticism, you went ahead and sent it to NATURE, right?

Tisza:

I think — it's a good question. Unfortunately I don't remember exactly, because I think the only contacts I had at that time were Simon, who in a sort of semi-bemused fashion were interested in it. And I'm not sure if I had any correspondence with the Oxford group, who would indeed — maybe [???] and Mendelssohn — who did the experiments.

Gavroglu:

Because by the time of the London Conference, you had already sent the idea to NATURE.

Tisza:

Yes.

Gavroglu:

Because your article was in April and the conference was early July, so it was you believing in your idea.

Tisza:

Oh, did my article come out in April? Or, it is signed —

Gavroglu:

— the 16th of April. You sent it the 16th of April.

Tisza:

But when did it come out? May 21st.

Gavroglu:

OK, so —

Tisza:

I don't know, probably a little later, the May 21st copy of NATURE.

Gavroglu:

OK.

Tisza:

So certainly the article was out at that time. So I don't know, I think I probably sent it in without much encouragement.[5]

Gavroglu:

That's very very interesting. Why don't you mention the NATURE article in the two C. R. Notes? In the two Notes you have in the [???]

Tisza:

Oh, I don't?

Gavroglu:

You do not. You only mention London's article.

Tisza:

Oh, really? Well, I don't know. I very soon felt that this article was so naive, and so it's a very strange thing, this article is very much quoted in the literature, and I also feel that, OK, that gives it priority, but it was in many respects really such a naive paper that I.

Gavroglu:

You see, because, if somebody, say, does not see this article, the NATURE article, and goes to London, London's article in NATURE, and then to the to your two C.R. articles, one would think that it's a continuation.

Tisza:

You know, I didn't realize that.

Gavroglu:

Because this is really a singularity point. In the articles in C.R. you calculate the superconductivity and the viscosity.

Tisza:

Yes.

Gavroglu:

In the Bose-Einstein condensation model.

Tisza:

Yes.

Gavroglu:

So you know, if one doesn't have this, one would think that, you know —

Tisza:

— strange. You know, that's very interesting. You see, I had completely forgotten that. And I remember in the correspondence with London, I think at one point I suggested that he shouldn't quote [???] the NATURE article, but the [???] PHYSIQUE. You see, I was sort of proud of the [???] PHYSIQUE article.[6] And no one saw these before '45, '46, only after the war, because that was during the war. And so that was my great trouble with that. And London said, "But you are wrong. If I don't put your NATURE article there, it is my brother who wrote the first thing about it." It was not for the first time that I had such clumsy handling of my own priority problems. But as I say, you see, I went ahead pretty much on my own. I have no idea now who could have encouraged me.

Gavroglu:

It's a good thing NATURE accepted it.

Tisza:

What?

Gavroglu:

It's a good thing NATURE accepted it.

Tisza:

It's a good thing NATURE — yes. I think they accepted it rather easily. There are many articles on helium.

Gavroglu:

So there was no problem.

Tisza:

And so I was surprised that I didn't quote. I didn't remember.

Gavroglu:

Yes, because I think (crosstalk) references....

Tisza:

— I didn't remember.

Gavroglu:

I checked it up once.

Tisza:

Are you sure? Because there are two quotes, only one, I think —

Gavroglu:

No, I immediately (crosstalk)

Tisza:

I think I might have quoted it in —

Gavroglu:

Well, the [???] is — one is this.

Tisza:

Oh, I see. I wasn't aware of it. Yes. Yes. And I felt that, by that time, that those articles were much more sophisticated, although even then, I had a tremendous problem with the length of the articles. The C.R. had no referee policy. There was no problem to publish it. But it didn't have to be a single word longer than the prescribed length. And that is the reason that, you have my letter which I wrote to London in end of, in December, '38.

Gavroglu:

It's the one in German?

Tisza:

Yes. I have this letter, and from this letter it is evident that I explained to him that my language, that the Bose-Einstein, that is, an ideal Bose-Einstein gas, is only because I had no space in the C. R. Notes to go into detail. And in this letter, I express my ideas about the quantum liquid, sort of. Yes. Now, the quantum liquid is again an idea which I picked up from Landau while I was there, not in connection with helium. I don't think Landau ever mentioned helium at that time. Except once that helium seems to be a strange thing, maybe it's a liquid crystal or something like that. He didn't particularly consider it, but he did discuss the concept of quantum liquid. Let me see. (crosstalk) No, no, that was also not discussed, but in connection with the nucleus, I think.

Gavroglu:

Oh, yes?

Tisza:

That the nucleus is a quantum liquid.

Gavroglu:

That's interesting.

Tisza:

I think that was it. Anyway, the concept of a quantum liquid was familiar to me from the Landau period. And in this letter — private letter —

Gavroglu:

Oh, Landau, I mean.

Tisza:

Landau.

Gavroglu:

I'm sorry, I thought you said London.

Tisza:

Oh no, no, no.

Gavroglu:

I see, I see, that's interesting.

Tisza:

Quantum liquid, that was a Landau idea. The London idea was a macroscopic quantum effect.

Gavroglu:

Yes, right.

Tisza:

And I certainly wanted to integrate the two together.

Gavroglu:

— the two together — that's a good point.

Tisza:

And also, so, in my discussion with London, as of course he had much more to show on his superconductivity theory, but I had a little bit of an edge in the sense that Landau plays a very special role among the modern, the quantum theorists of his generation. He was practically the only one who knew and liked thermodynamics, and I learned it from him. I remember that London told me once that, "I don't know the disciplines with dynamics, that is, thermodynamics, electrodynamics and hydrodynamics." And at one point, he got a letter from von Laue in connection with his superconductivity paper. I don't know whether you know it, there was some friction.

Gavroglu:

I know.

Tisza:

London used to be Laue's assistant. And at one point, after the war, Laue felt that somehow his ideas on superconductivity didn't receive recognition. So he wrote a letter to London while I was there, and in connection with the thermodynamics of the superconductors and also the question of which thermodynamic function to take in a certain minimization problem, and London at that time asked my help, to straighten out the thermodynamic differences, which I could do, but I remember, I wrote him a long page, and all that he compressed that into two lines very nicely, but I had some input, in this sense, that I had more thermodynamic background than London had.

Gavroglu:

There was an extensive correspondence between London and von Laue about —

Tisza:

This I don't know.

Gavroglu:

I have copies of the letters.

Tisza:

I see. Except this one exchange of letters about which I knew.

Gavroglu:

OK. But tell me, London should have been quite satisfied with your [???] letters and the two JOURNAL de PHYSIQUE, right?

Tisza:

No, he wasn't.

Gavroglu:

He wasn't, even for those?

Tisza:

No.

Gavroglu:

Because there, you give full credit to London, right? I mean, you really work out the details of the Bose-Einstein condensation.

Tisza:

Yes, that's right. No, that was not the point. He just — he didn't believe — he thought that my whole thermal waves are a crazy invention. You see, he was absolutely, he told me he didn't believe because he was trapped by his brother, in the sense that with the surface effect, he sort of committed himself to the surface effect, and I think he felt that I am compromising his Bose-Einstein theory by carrying it to such an extent. So that was his real — I don't think that it was sort of an unfriendly — so, it was very unfriendly to me basically, but the motivation was his genuine concern that here he's a careful scholar and I'm jumping in and with some exaggerated ideas, and in a way spoil his Bose-Einstein theory by carrying it to improper extreme. That was his point. At that time.

Gavroglu:

OK. Let us now come to the letters.

Tisza:

Well, I think it's a little premature, to it. The correspondence came after that I came to MIT, and —

Gavroglu:

When was that?

Tisza:

In '41. I arrived here in the spring of '41 and became an instructor at MIT the next fall, and my hands were full with teaching. I had a course in thermodynamics which was immediately quite successful, and I pretty much forgot about helium. No one wanted to do any measurement on it and all that, so it was — And also, all the theorists with whom I tried to convince always said, why don't I do a kinetic gas calculation or something of this sort, some detailed calculation and model, and then they would understand it. Very few people appreciate a non-model consideration. As a matter of fact, I spent an awful lot of time in learning kinetic theory, reading the long papers of Enskog and reading the kinetic theory calculation of Bose-Einstein and Fermi-Dirac kinetic effects and things of this sort. I played with it, but I didn't get anywhere, tried to accommodate people, but I was completely unsuccessful, and then came the publication of the Landau paper, and I was very impressed, and I sort of — I forget now, there was first a short Landau paper and then the detailed, so I'm not quite sure about the timing between the two, but you have my correspondence with Uhlenbeck.[7]

Gavroglu:

Yes.

Tisza:

It is probably more than I remember at this point. But you see, I immediately started? — undoubtedly, truly — I immediately switched to Landau's version. But I did think that phenomenologically, my ideas remained valid. In other words, I withdrew at that point from the Bose-Einstein interpretation, but I thought that my two fluid interpretation should hold. And then nothing much happened until —

Gavroglu:

— can you expand a little on this point? The point of being more reserved on the Bose-Einstein condensation approach rather than the two fluid approach?

Tisza:

Well, when I read Landau's paper, which was hydrodynamics, then I thought that, in a way, I liked the idea of becoming more macroscopic, in a way, and I realized that my ideal gas concept was not really right. Of course, in the JOURNAL de PHYSIQUE I made it clear that the ideal gas is more in the [???] of an exciton type thing, which was quite in line really with Landau's idea of a quantum fluid. And of course, I had one edge over Landau, in the sense that I from the outset knew it was thermal waves, whereas Landau believed in the second sound and he thought that it would be acoustically excited, so it is a very strange thing that I felt that he didn't have the intuitive grasp of what goes on, because it was the second sound which became now completely popular. A common misconception of it. It was not a sound wave. The Lifshitz made a calculation, looked into the matter, and he showed, it's a long calculation, that actually these are thermal waves, which I had at the outset. Now, when this came out — I think it was just about, the war ended, and during the war I don't think I had any contact with London. I don't remember exactly when we resumed contact, but I remember one letter where he said, "Do you realize that your theory is quite equivalent to Landau's?" And then suddenly, he felt me as an ally, it happened at this point only and not until then. So it was only after the Peshkov experiments were in, suddenly he realized that after all my idea was OK, and everything changed, because he was very much impressed — he was very guilty about it, that he'd attacked me so much, and he could have done me a lot of harm in professional reputation and all that really. From then on we were extremely friendly.

Gavroglu:

I still want to explore a small point. When you read the article of Landau's, it might be the case that you first came in contact with Landau's article after Uhlenbeck told you about it?

Tisza:

Yes. Let me see. Let me turn it. ... I start to remember now. After the first Landau article in the PHYSICAL REVIEW, I immediately started to think about the hydrodynamic thing, and I thought basically it was a good idea to look at it from this point of view, but I considered Landau's quantization procedure too formal and unconvincing, and also that he came to the wrong result in the critical velocity of helium of 100 meters per second or something like that, whereas it actual fact it was a few centimeters per second, and this I explained by turbulence, and this is what I wrote in this letter. Yes, I wrote to Uhlenbeck about it, and then there follows this correspondence. In other words, I accepted the Landau hydrodynamic idea, modified it in the sense, I made it again more intuitive by introducing vortices and quantizing the vortices, not in canonical quantization but sort of in a more phenomenological quantization, more like Bohr. And then could explain the low critical velocity, that there are big vortices which limit the critical velocity but don't contribute much to the caloric properties.

Gavroglu:

Why, after reading Landau's article, did you move away from the notion of Bose-Einstein condensation? I can understand why you thought the two fluid model was more important, because the last criticism that you bring up against Landau, you know, very late on, 1948-49, is on the question of the necessity of the Bose-Einstein statistics.

Tisza:

Well, I suppose that was London's influence. I don't remember now exactly, but probably from the correspondence you can get more out of it than from my recollections.

Gavroglu:

But the reason that you moved away was because it wasn't necessary for Landau to base his hydrodynamics on statistics this was not necessary to be with hydrodynamics, is that why?

Tisza:

Yes.

Gavroglu:

OK.

Tisza:

Yes. And then, when I resumed contact with London, then came, "Hey, do you realize that your theory and Landau's are completely the same?" And ...(off tape)... Yes. So there was this period when I moved in the hydrodynamic direction, and I didn't quite accept Landau's version, but I thought maybe I don't need the Bose-Einstein after all. Then London contacted me, and I'm hazy about the exact date and how it was, we met on several meetings, but I do remember a letter or telephone call in which he then said, "Do you realize your theory is really quite equivalent?" And then he gave a talk in Cambridge.

Gavroglu:

1946?

Tisza:

Yes, 1946.

Gavroglu:

That's when the correspondence started.

Tisza:

Yes, that's right. And at this point, London did the following thing, which was not really fair to Landau and that is, my formula for the thermal waves there in terms of not very clear gas pressure terms, whereas Landau had it in terms of entropy, and of course Landau's was much better. And what London did, he transformed mine into Landau's, and he presented it in Cambridge as the Tisza Formula, whereas in actual fact it was the Landau Formula. But it was not my doing at all. And after that point then, I got at one time I think, I don't know, a phone call from Onsager, to clarify this situation. And Landau wrote a letter, to the PHYSICAL REVIEW, where he protests this misrepresentation of my theory.

Gavroglu:

OK, let's come to that.

Tisza:

I don't say — I suppose it was his initiative, I suppose I expected it at that point, I don't want to say that I had absolutely nothing to do with it, but basically he initiated it, and his motivation was to — partly to repay his debt to me, and partly to regain —

Gavroglu:

— the upper hand —

Tisza:

— the upper hand, in terms of the Bose-Einstein, so it's a very complicated thing, and I don't think it was too interesting now to go into all the details. I myself wouldn't be able, without much study, to completely reconstruct it, and I think it is no longer important.

Gavroglu:

Before going into letters, yet, there are two things that may be of some interest and may freshen your memory. You know among the three, as I told you in the beginning, Landau, London and yourself, Landau and London are much more aggressive towards each other, whereas your attitude is that very interesting that in Feynmann's article, in his final article in THE PROGRESS OF LOW TEMPERATURE HELIUM, he really adopts your point of view.[8] In other words, his attitude was that everything that has been done was very much towards clarification of the idea, rather than one being the right one and the other being the wrong one.

Tisza:

Yes, I think it's a fact that — you see, I had no animosity to either, I was grateful both to Landau and London, because I learned a great deal from them, and so therefore basically I had a positive relation, even if at different points I had arguments with both. But my basic attitude was very positive, and it always was, because I always acknowledged my indebtedness to both of them. Whereas I think there was some enmity between Landau and London, which we shall never find out, the origin of which went to the grave with them, and I don't know why, but I remember once a little remark. Landau liked to classify physicists. You know his famous classification?

Gavroglu:

Yes.

Tisza:

And I remember once that Heitler and London, he sort of dismissed them, they are not very important, forget it. So Landau could be very rude, and it is quite possible that at some meeting, I know that they met, but if had they met, I could imagine that he offended London, and that there was a personal animosity between them.

Gavroglu:

Another thing, in following. You know this booklet which is called "Supplement to the Helium," this is a book by Lifshitz and Andronikashvili and it is a supplement to the Russian edition of Keesorn's HELIUM.

Tisza:

I do.

Gavroglu:

There you see there is a passage on your work, and it is a passage which even though it has, you know, one sentence, it says that you were the person who started all these things, of the two fluid model, and it has an interesting page that says you were the person who started it independently of Landau. You know.

Tisza:

Yes. Lifschitz visited here a few years ago, at MIT.

Gavroglu:

This is 1948.[9]

Tisza:

Oh no, much later. I've forgotten exactly, it might have been in the 1970's.

Gavroglu:

It's interesting there, afterwards —

Tisza:

— no, it must have been later. In the late seventies. Because I know, I have some photo of his in our house. We moved in in '73 so it was probably '75, '76 or so. And then he told me that at first they knew only my NATURE article, and that was considered very naive and uninteresting, and then after the war the JOURNAL de PHYSIQUE reached Moscow, then they were very impressed. That is, how much of the material I had. And also, Lifshitz was extremely friendly to me.

Gavroglu:

It's a little peculiar, let me finish the story about the Supplement. In the Supplement, even though Lifshitz says that you have precedence and all that, then it goes on criticizing the two fluid model, and it's interesting that even though it also is Lifshitz, the criticism is in the exact words being used by Landau in the PHYSICAL REVIEW Letter.

Tisza:

I suppose, at that time Landau was dead, I suppose.

Gavroglu:

Landau was [???], but first of all, the author is Lifshitz and secondly it is written before the Landau letter in PHYSICAL REVIEW. It's about six months to a year before.

Tisza:

Oh oh, I'm sorry, when was this Lifshitz article?

Gavroglu:

1948, early 1948.

Tisza:

Oh, I see.

Gavroglu:

It was in the exact words, you know, the main criticism —

Tisza:

Landau and Lifshitz worked very close together. Now, whose words, there's no telling. You heard the famous quip, about Landau-Lifshitz THEORETICAL PHYSICS, that, not a word by Landau, not a thought by Lifshitz. Actually this is unfair, because Lifshitz was not a stooge, and he had his mind of his own, but they were very close, so I don't, that is nothing, I don't know, I mean, whose — it was Landau's, I think. On the other hand, last summer there was here at STATISTICAL PHYSICS conference here in Boston, and I was present, and I met — my memory is good enough for recent things — and I didn't write it down — I talked to a number of Russian physicists who are now in Israel, and they all asked me, how did it happen? And apparently they were under the impression that I visited him in Moscow, and at a meeting Landau stood up and said, "Why did you come all this way with this trash?" or something like that. And I think they considered it how rude Landau could be. But I told them, I was never back to Moscow, to Russia after '37, and that the whole thing was completely invented, and so I don't know how these things get around.

Gavroglu:

But you see, Landau in the letter in the PHYSICAL REVIEW says that they did not receive the Journal de PHYSIQUE until 1943.

Tisza:

Yes.

Gavroglu:

And that he had overlooked, just didn't pay any attention to the C. R. Notes.

Tisza:

Yes, sure.

Gavroglu:

Kapitza, who had already published in 1940-41, he then got in the [???]

Tisza:

If you want to have my guess how that happened, in 1938, of course, Landau was in jail, and he was not a real —

Gavroglu:

How long did he stay in jail?

Tisza:

I think he was arrested in the spring of '38. About a year. I don't know exactly, but from the spring of '38 to the spring of '39. So in this critical period, he was out of commission, and when Kapitza had the superfluidity paper, undoubtedly he must have, he quotes from my NATURE paper and the [???]

Gavroglu:

Both NATURE and [???]

Tisza:

Both, so no question about it that the experimental thesis, and he made so many experiments about the detailed aspects of the two fluid theory, which he couldn't have understood or even described before I started the two fluid concept, so I visualized the following discussion, that Kapitza says, "Look, I found the experiment, and there is this Tisza who describes it and I found it quite useful." And Landau looks at it and says, "It's nonsense." And then he went ahead and does it in his own way.

Gavroglu:

You see, this is extremely interesting, in that there is a theoretical appeal of your approach to the experimentalists (crosstalk) as well as Kapitza.

Tisza:

— but more sensitive, oh yes, I mean, it was correct, experimentalists, they couldn't do anything without it, without this idea, and I'm sure that Kapitza had it too, but then first he and London and then Landau discouraged that and he sort of withdrew from it. Kapitza visited here, a few years ago, and I introduced myself and tried to talk with him, but he acted as if he had never heard my name. He was very unfriendly. And certainly, I think in his later papers and all that, he certainly didn't give me any credit. There's no question about it that the two fluid model idea came from me, because through Kapitza, Landau heard about it. His theory is completely his. I think that, besides the idea that there are two fluids, everything what he did is his and was uninfluenced by me.[10] I am quite convinced that it's completely original, except the two fluid concept itself.

Gavroglu:

And the thermal waves?

Tisza:

No. You see, he didn't believe it.

Gavroglu:

No, no, I mean, for your work. It was needed for your work, a part of the two fluid concept.

Tisza:

Yes. Yes, I think so. And as a matter of fact, it's interesting to remark that something, that Landau has something indirectly to do with that.

Gavroglu:

I see.

Tisza:

So to some extent, it was Landau, but nevertheless still in my own personal question, because Landau didn't.

Gavroglu:

That's very interesting. OK. Now, let's come to the letters. Let me ask you first, why didn't you go to the Cambridge meeting of 1946?

Tisza:

Oh, I think it was — I don't know. Maybe if I mention, I was very strapped financially. I had a low position. Our families in Hungary needed support, so I think it didn't seem practical. Now, I don't know why MIT didn't send me or why I didn't try press them to it.

Gavroglu:

OK. Well, I don't have some letters that London sent to you, but most probably —

Tisza:

— yes, I think you (crosstalk) —

Gavroglu:

— so if I don't find —

Tisza:

OK, we can come back.

Gavroglu:

OK, yes.

Tisza:

Did I give you my package of letters? Or do I give them to you —

Gavroglu:

— no, no, we'll do it in the end. We'll do it in the end. Now, you have something when you first heard the Ashkov results, and at that time, it seems that you feel that this is absolutely crucial, to get to see — you know, the second sound or the thermal waves.

Tisza:

Yes.

Gavroglu:

Agree with you or Landau, for low temperatures.

Tisza:

Yes.

Gavroglu:

And the letter now, that discussion, which things did you think were the absolutely crucial points for deciding between yours and Landau's approaches?

Tisza:

Well, you see, at that time, it's very instructive, by and large, I was interested in the common features, but at that time I also emphasized the differences, and I was quite convinced that Landau is wrong, the way he handled the [???] and the [???] together, and therefore I thought, and that was my great mistake which then backfired, that the low temperature behavior, I thought at that time would work out the way I wanted, and of course I made a great mistake at that point, because it was the opposite.

Gavroglu:

By neglecting the [???]

Tisza:

Yes, that's right, and quite the contrary, it was then considered a great triumph for Landau, and suddenly my role was pretty much dismissed after that. But so, the fact is that there is by and large — you see, I was rather careful in my judgment. I made this one mistake, this strong mistake of thinking that the (fornons?) are not part of the (lomar?) — the normal fluid. So I think it was just a mistake, and I was rather stubborn about it, so that's why. That was one of the things in which I was stubborn.

Gavroglu:

In the letters you exchanged in November of 1946, I think there are some interesting discussions between you and London, on what macroscopic theory means, not the microscopic theory but the macroscopic theory, and London seems to be thinking that if one has a theory which explains everything that has been observed, then that is quite a satisfactory state. And you seem to be thinking that one has to go further than that, and not, explain only what there is to be actually called a microscopic theory.[11]

Tisza:

I'm sorry, I don't — may I see the letter?

Gavroglu:

Of course.

Tisza:

Because I don't remember. I don't know exactly what that means. It is this paragraph...

Gavroglu:

Yes?

Tisza:

I must admit that I occasionally fudged the “macro- versus micro-“ issue. My reliable results were always macroscopic, but success made me bolder and I tried to venture into the microscopic. I am afraid that all three of us were guilty of some fudging when our microscopic ideas were less than solid. London did such a fudging when he claimed in his Cambridge (UK) talk that my thermal wave velocity was not different from that of Landau. Of course, for this I had to pay because it gave an easy target for Landau to attack. Landau fudged when he ignored the Oxford results. However he covered up all by forceful self-assurance.

Gavroglu:

— not his —

Tisza:

Yes, that is, this is, here —

Gavroglu:

— OK, that is good.

Tisza:

This is the first.

Gavroglu:

OK. ... Now, you seem to be having a little difficulty in finishing your PHYSICAL REVIEW article on theory.[12]

Tisza:

Yes.

Gavroglu:

Because you don't seem to be very happy about it, and you seem to be writing and rewriting it.

Tisza:

Yes.

Gavroglu:

Was there any particular reason for that, do you remember? Was there any particular difficulty, or was it the whole thing that gave you —?

Tisza:

— just the whole thing. I mean, nothing specific. You see, that was a very difficult thing to write, because there was such a mixture of microscopic and macroscopic, and I tried to find the balance, and I was partially successful but certainly not quite. Also, you see, there was the first version, which was rejected.

Gavroglu:

This is another thing now I wanted to ask you. Did you leave anything substantial out of the first —?

Tisza:

— no. That happens to me sometimes. I wanted to be very pedagogical, and first I considered a special case, I don't remember exactly how, and then I started it over again, and the referee went through it, and when he came to the repetition, said, "My God, that is too much, he should cut it by a lot." Then we all agreed that the cut was helpful. In other words, I didn't leave out anything essential.

Gavroglu:

I see. But I was a little surprised with the following. London is extremely enthusiastic about your paper.

Tisza:

Yes.

Gavroglu:

And after you have the answer from the referee, he seems to have some reservations about the paper, because he wrote all things that he could have told you before. And then there is the question of urging you to send it to the JOURNAL OF CHEMICAL PHYSICS.

Tisza:

Yes.

Gavroglu:

Was there anything particular in the change of heart, or is it just something of the moment?

Tisza:

I think so, no, I don't see anything important.

Gavroglu:

I see. But you know, you sent again to the PHYSICAL REVIEW.

Tisza:

I sent it again to PHYSICAL REVIEW and it was accepted, yes.

Gavroglu:

Yes. Then right after that, London mentions the book he's going to write to you.

Tisza:

That was the Superfluids, yes.

Gavroglu:

Then I have the impression he's overly secretive about the book. You know, he says he doesn't want you to mention it to anybody.

Tisza:

Yes? I don't remember. He said, OK, if you tell me the problems.

Gavroglu:

You know, he's not very forthcoming. He also tells you to mention it to Slater because he hears that Slater might be writing a book or something, but again, there isn't anything specific.

Tisza:

I don't know. In fact, writing the SUPERFLUIDS, the second volume, he was very positive towards my contribution.

Gavroglu:

Oh yes, that is obvious in the book. Now, we come I think to the articles in the PHYSICAL REVIEW, between London and you, the two small pieces, and I want to ask you a specific — but by the way, what was it, did you have any reaction of London's on that correspondence you had between Landau and yourself, because during 1948, you do not have any correspondence with London. There's a whole year.

Tisza:

A whole year.

Gavroglu:

Where there are no letters. And it is during that year that your answer to Landau appeared in the PHYSICAL REVIEW.

Tisza:

I don't know, but would say that London must have realized that he caused me a lot of trouble, by sort of presenting Landau's formula in my name, and in such a fashion, Landau attacked me.

Gavroglu:

I see.

Tisza:

And there's no question about it, that it was his initiative, to present it in such a fashion. And you see, I didn't do it, but then again, I didn't sufficiently forcefully object. After all, London was very much my senior, in this respect, and although I often argued with him, but at any rate, whatever it was, I don't quite know how the responsibility is divided, but no question that the main part in the position which Landau so successfully attacked was put in by London. And it reflected on me. So it is quite possible that at this point he didn't feel like writing too much about it.

Gavroglu:

OK. Let me ask you know something which I don't understand too much, and maybe you can elucidate this point. In Landau's article, in Landau's letter to the PHYSICAL REVIEW, he has two footnotes. One footnote says ..."Tisza's remark, his assumption tends to modify the theory in the wrong direction, can hardly be justified." And before that, he says, "Such reasoning, also present in Tisza's recent paper, no quantum meaning can be given to such assertion as for example each vortex element can be associated with a definite mass contained in the volume in which the vorticity is different from zero." Somehow, Landau criticizes what you had in a footnote in your PHYSICAL REVIEW article.

Tisza:

If I come to these details, then I think that I basically, many of my arguments are just wrong. So I think, I'm not sure whether Landau was right, but in criticizing me it is OK. Surely I over-estimated at one point my ability of going into the microscopic theory. And I would put it like this — that later, there were papers where went into great details of the description of practical effects. And at that time, I completely gave up. I didn't feel that it was the type of calculation where I could compete with Landau, and I didn't pursue it really. I practically withdrew then from the low temperature work at that point.

Gavroglu:

Yes. But Landau's main criticism of your work seems to be that your work is not consistent with quantum mechanics, that you use a lot of things that quantum mechanically do not make sense, whereas your criticism of Landau's work is that Landau uses more assumptions than he says he's using.

Tisza:

Yes. I would think that neither of our theories really was above reproach. And I think it is at this point, I didn't want to argue for anything. I just didn't know. I think that I was very lucky in the sense that in the beginning of the helium problem, there was quite a deep question to which I could contribute something.

Gavroglu:

Right.

Tisza:

When it came to the quantitative detail, I was not able to do it, nor was I very much interested at that point in that. So if I look back at it, after all this controversy, I simply withdrew from the area. I was not able to carry on. No question that Landau was very much better to push through an argument and come to a conclusion. I'm not sure that he was always right, but he certainly was very powerful in carrying the argument through, and I was not. So what actually happened, it's sort of a very profound rearrangement of my whole research program came about, so I suppose I could talk about it at another occasion or in other context.

Gavroglu:

That's right.

Tisza:

But I would consider it, in fact, in this fashion, that one aspect of the helium problem appealed to me, and the second one didn't, and at this point I stuck out my neck too much, and I couldn't support that.

Gavroglu:

OK. Now, in your February 1949 letter to London, you say, "I think almost to the day eleven years that we took this walk around Robinson."

Tisza:

Yes.

Gavroglu:

What is this?

Tisza:

I told you about it, that is the one I meant — yes —

Gavroglu:

— I see. OK. Because it's —

Tisza:

Yes. Yes. That was this famous walk in February 1939. The night after that walk a part together I put together the essence of the two-fluid concept.

Gavroglu:

OK. Then in a long letter which you wrote to London in July of '49, you have something at the end which I don't understand. You say, "The only point where we disagree is the question of order in momentum space."[13]

Tisza:

Yes? I don't —... Yes, I remember now what it was. In the summer of '49, in fact, I remember I spent the summer in Vermont, in a farm house for two months, and I worked then in preparing a possible talk for the Cambridge meeting. I was working on the superconductivity.

Gavroglu:

The Cambridge and MIT?

Tisza:

The MIT-Cambridge meeting, on the superconductivity paper about which I write here, and also I started to work on the philosophy of quantum mechanics, about which I will talk later. Now, in this thing, the momentum, everything refers to the superconductivity paper. I considered it a half-baked thing. I just tried. You know, I mentioned to you that I like the quantitative aspect, and I felt that in helium, I was at the end of my ropes and I didn't know how to continue, and I thought I'd do something else, and that was really just an opportunistic thing. I don't think I'm very proud of it, and that's that. So it is nothing much, nothing to talk about it.

Gavroglu:

Because maybe the next time, we talk about a little on superconductivity, because you also mentioned trying to derive the isotope effect from your theory, somewhere.

Tisza:

Let's forget about it. These are all half-baked things. And I don't put any stock in it.

Gavroglu:

OK. I see. I see. ... About writing a book with [???] ?[14]

Tisza:

Yes.

Gavroglu:

What was that? I haven't seen it. It didn't come through or something?

Tisza:

Well, it was split. Herbert Callen wrote it alone. It came out. It became a very successful book. The Herbert Callen book — that is it, we started it together, and that, yes, so Herbert Callen thermodynamics, that was originally a joint project.

Gavroglu:

I see. OK. I really don't have anything else that — at this point — that I want to ask you. Now, if you think some of the things that we've talked about, that there is anything else, maybe you tell me, and then we finish now, and then we come back to the other and continue what I was telling you in the beginning.

Tisza:

I don't think of anything at this point. I think we covered it pretty well.

Gavroglu:

OK.

Tisza:

And if something comes to my mind —

Gavroglu:

— sure —

Tisza:

— it will come up again. But I think that we have covered it pretty much, so —

Gavroglu:

OK. So what I shall do, maybe we'll close it now.

[1]The usual support group went into action. Teller, Szilard, Fritz London who recently transferred from Oxford to the Institut Henri Poincare in Paris. I was advised to explore the job situation in Paris where there was an international meeting: “Congres du Palais de la Decouverte.” I was also referred to Edmond Bauer who was the vice-director. A Russian born physical chemist Michel Magat who worked ina lab in the College de France under, who was under the top directorship of Paul Langevin who had the prestigious rank of Professeur de College de France. It appeared that the dozen people or so in lab worked in condensed matter research. Quite consistent with the Landau type thermodynamics I was intent to pursue. All this looked promising. Moreover, Edmond Bauer was also in charge of a scholarship fund established for the support of émigré scientists. Sure enough in a few days I had the assurance of a modest scholarship with a desk in the Bauer Lab. It turned out a very congenial association that lasted until the occupation of Paris by the Germans in June 1940. The thermodynamic expertise I learned from Landau made me a useful consultant all over the lab and encouraged my intended orientation. One of my colleagues was a Russian born physical chemist Michel Magat who was to become a close friend. He soon introduced me to Fritz London, an association that would lead me to the problem of liquid helium, of considerable interest to Gabroglu, the biographer of London who was in possession of an extensive correspondence I had with London. However, I am getting ahead of myself. London had a very respected position in Paris, but had not much in way of personal relations. We soon developed a close friendship. A much warmer personal relationship than I ever had with Dau.

[2]Our first topic was the superconductivity paper that Fritz wrote with his younger brother Heinz. This is a very subtle paper and it was exciting to hear the subtleties from the horse’s mouth. However we were both puzzled why Landau never mentioned this paper in his teorminimum, although the paper was in harmony with his selection of papers. See footnote 24 where I propose to explain Landau’s apparent prejudice against London. This prejudice is even more striking in the context of the Oxford papers of 1934-35 by Francis Simon and Fritz. The discussion of these papers was the next on our agenda. It was an important preparation for the experimental papers soon to follow. Another important preparation was the van der Waals meeting in Amsterdam in the fall of 1937. Pages 39-40 of the interview are inaccurate in accounting for the sequence of events between the van der Waals meeting and my first discussion of the two-fluid idea with London. I think that a serious attempt of recollecting the sequence of events led to a reliable result which is summarized in the following note.

[3]The great event of the van der Waals meeting was Born’s report on Joe Mayer’s theory of the condensation of the van der Waals gas. This even recalled in several people’s mind Uhlenbeck’s 1927 critique of the Bose-Einstein condensation (BEC). I don’t know the exact time of several people’s response, but I am positive that London did not immediately think of BEC. In the next few lines I will suggest the exact time when London arrived at BEC. In his report to me on the meeting he talked only of van der Waals condensation. Helium became a topic only on receipt of the issue of Nature of January 8, 1938 in which the experimental discovery of superfluidity was reported by Kapitza, and by Allen and Misener. We were excited by this discovery and I was ready to first listen to any ideas London might advance. His first idea was that liquid helium may behave to some extent as a gas. He recalled Felix Bloch’s theory of metallic conductivity that started with the apparent freedom of electrons in a strictly periodical potential. He suggested that I play around with this idea of the Fermi gas but consider helium atoms in a self-consistent field of other atoms. I proceeded to do this without conviction and without success. Meanwhile I caught a cold and was confined to my hotel room. In a few days I had a call from Fritz inquiring about my progress. I said I had none, but he wanted to talk to me and came to my room. He first made me talk about whatever finding I may have. After having enjoyed my embarrassment, he said: “now listen to this”; Fermi statistics was of course nonsense, it should be Bose-Einstein statistics and now he invoked the Amsterdam meeting and said that Mayer in effect refuted the critique of BEC, he filled in the holes and here it is. I congratulated him and said this is wonderful. I believe this is the correct sequence. I don’t think London would have sent me on a wild goose chase had he already thought of the BEC at an earlier date. The next important date was the so-called fountain effect reported a couple of weeks later by Allen and Jones. Next Sunday we went ona hike in the Bois de Verriere. At the end of a long chat we concluded that the new experiments are so extraordinary that we should be able to recognize the non-standard assumptions they call for. The evening after the walk I couldn’t stop pondering the helium problem. While I don’t remember the details of my thinking, the main points are clear in my mind. I had the idee fixe that there was an irreducible conflict between two types of viscosity measurements: (1) the damping of an oscillating disk and (2) flow through a thin capillary or through a narrow slit. Although passing through the lambda point marked a lowering of the viscosity in (1), no value of viscosity, however small was consistent with the superfluidity manifest in (2). The latter marked the failure of the concept of fluid dynamics itself and I felt free to advance radically new ideas. An early product was the two-fluid concept with two overlapping velocity fields, one superfluid and the other viscous. For the rest of the night I convinced myself that the experimental implication of this picture agree with the observational situation. By morning I was sleepy, but satisfied that I was after something important. I went to see Fritz and was disappointed that he did not think that I had achieved anything. It took me some time to understand my “discovery” had its shady side. Theoretical physics has its core principles which are not easily abandoned even if they seem to conflict with experiment. London was more conservative. He had to live up to his scientific reputation. Not only did I not perform a careful analysis of the principles that ought to be upheld and those to be abandoned, but I did not believe that I would be capable of doing so. At the same time I was convinced that my qualitative ideas would be useful for the experimenters. In this I was right and was willing to make a fool of myself. I will outline the strategy I developed to make a case for my intuitive ideas. There were no cryogenic experimenters in Paris, but Simon and Kurti were shuttling between Paris and Oxford. I told them about my ideas; they listened with benevolent skepticism and they were in communication with Daunt and Mendelssohn who were into liquid helium, mainly surface creep experiments. They certainly communicated a certain interest in the two-fluids.

[4]An important event that Fritz’s brother Heinz London published a paper in Nature in which he gave phenomenological proof of the thermomechanical effect. This paper had a double effect. He referred to my paper, but subtly altered my point. He claimed that the superflow was a surface effect. This was adopted by Fritz who used it as an alternative to my volume supercurrent. However, as to myself, Heinz had a positive effect. He gave me a jolt, I felt I couldn’t have done that myself, and I emerged from my defeatist mood that I cannot attack any microscopic theory to bolster my phenomenological speculations. My first project was to get something quantitative about the two-fluid hydrodynamics. I remembered that in the teorminimum Landau linearized the Navier-Stokes equations to obtain the important special case of sound propagation. In the two-fluid case I should get two types of wave propagation. I had an intuitive idea of the two-fluid hydrodynamics. There is one flow that is the same as the ordinary theory. In the other flow the two fluids move in opposite direction and they involved a net energy transfer carried by the normal fluid. If it’s a wave, it is a temperature wave whereas the ordinary wave is a pressure wave. This is similar to what Landau developed three years later. I will return to the relation of our two waves later. My Nature paper ensured me an invitation to a small cryogenic conference in London in early July 1938. Herbert Frohlich presented his view of the lambda point as an order-disorder transformation. Fritz suggested the alternative of Bose-Einstein Condensation. I suggested to take the two fluids more seriously and presented my temperature waves. I even suggested an easy experimental approach using a paramagnetic salt in an alternating magnetic field as a source. Verification would have been an enormous coup. Unfortunately Fritz quietly discouraged this experiment. Later it was the outbreak of the war that was blamed for going ahead. Actually there was an interval of 14 months which does not seem that short.

[5]I published the temperature waves in a C. R. note. My two C. R. notes mark a turning point among my papers on helium. I was surprised to be reminded by Gavroglu that I failed to quote my Nature paper in the C. R. However, I do remember that I considered my first note naïve and tended to ignore it, thus risking my priority. In this I wasn’t successful as the Nature paper was so widely quoted that it could not be ignored. It is interesting that in none of these cases had I any problem with referees. It seems to me Nature did not referee its letters to the Editor. The C. R. policy was to accept notes submitted by members of the Institut. Mine were submitted by Langevin. A rigorous criterion was a limited length. I was much aware of this limit. As a result I referred to ideal gases, although I was much aware of Landau’s aversion of using this terms for real entities. I think he was much influenced by Bohr’s paper on the liquid drop model of nuclei. I think this came out in 1936, i. e. during my stay. As a result Landau started to refer to “quantum liquids,” and “electron liquids” or “Fermi liquids” in metals. This concept overlaps with London’s “macroscopic quantum state.” Immediately after the C. R. notes I started to work on a paper on the “Bose-Einstein liquid.” I submitted actually two papers in the Journal de Physique et de Radium, October 1939. This paper came out in 1940, but I saw a copy only after the end of the war. It is the best I wrote on the subject and contain points otherwise not available. Yet I resisted suggestions of reprinting it, because it has also errors and I don’t want these kept alive. I will later return to the problem of synthesizing the correct contributions of all relevant sources. This paper did not fare any better than the previous one in gaining the approval of Fritz. He was firmly opposed to the idea of two superposed velocity fields.

[6]The year between October 1929 and September 1931 was scientifically uneventful, but involved important displacements: Paris, Toulouse, Marseilles, Lisbon, Boston, culminating in my appointment as instructor at MIT and John Slater commissioning me to take over his course in thermodynamics, since he was to leave for work in the Radiation Lab. He gave me free hand to set up the course and I combined two texts: Slater: Chemical Physics and Landau – Lifshitz: Statistical Physics, supplemented with my lecture notes from Max Born. I repeated this course in ever improved from for some twenty years. By 1943 it was reasonably integrated and I had a very good class with Herbert Callen and Joaquin Luttinger who became my first PhD students. They helped me realize my old resolve of following Landau’s pioneering path along a living thermodynamics. My agenda at MIT was improving the course on thermodynamics, to keep thesis students busy and on the backburner the idea of using chemical thermodynamics for the foundation of quantum mechanics.

[7]There was however, a latent issue of superfluid helium that resurfaced with Landau’s entering the scene. During much of 1938 when many of us were involved with helium, Landau was in prison. He was freed through the efforts of Kapitza in the spring of 1939. Kapitza was the first to discover superfluidity experimentally; this was duplicated independently by Allen and Jones. It was well known that this discovery generated the two-fluid concept; Kapitza proceeded to test this idea by the most ingenious experiments. As a result the two-fluid concept turned into an experimental fact and was as such presented to Landau who had strong preconceptions against using the quantum gas concept for systems consisting of interacting particles. Kapitza’s experiments enabled Landau to make a new theoretical departure, an opportunity he seized by creating the concept of quantum hydrodynamics. His first publication was not available to me, but I had a detailed letter from Uhlenbeck to which I replied by expressing by partial conversion to the hydrodynamic view by maintaining my original view of a pressure wave and a temperature wave, whereas Landau’s first and second sound were both pressure waves and could not be experimentally verified. In 1944 the situation was clarified by a theretical analysis of Lifshitz who confirmed my pressure, temperature duality of waves which was then readily verified by Peshkov. At this point I got a message from London to the effect that the experiments verify my prediction no less than that of Landau. This ended our 16 year old feud which started with London’s opposition to my two-fluid concept. We restarted our original friendship and frequent correspondence to continue till his death. (Idon’t have the first letter to mark this break. The correspondence in my files starts in 1946 to go on till 1954. It is possible that the first communication was over the telephone. We also kept in touch at Physical Society and cryogenic meetings). The comparison of my theory with Landau’s was actually quite subtle and I will return to it in the context of the experiments that clarified the situation.

[8]Gavroglu points out that Landau and London are more aggressive against each other than I am. This is correct; first of all I felt that all of us were partly right and partly wrong. At no point did I feel that I am totally right. Second, I felt indebted to both Landau and London, I clearly learned a great deal from both. My adversary stance came about when I felt unjustly attacked which occurred in both relations. A few years ago Khalatnikov, the main collaborator of Dau in his helium work, gave a talk at the Harvard colloquium. At the end I introduced myself and he said: “In the early days we very much studied your papers. Of course, you cannot know this, based on our statements.” “Why was this so?” said I. “Landau would not let anyone have part of the helium credit.” This was of course very much evident as he ignored the excellent Oxford work of Simon and London of 1934-5.

[9]I have a photo of Lifshitz taken in a house where we lived beginning 1973. The Russian Supplement was published in 1949. Landau was very much alive and the evaluation of my theory was undoubtedly arrived at a discussion of the whole group. It’s not surprising that the words of Landau and of Lifshitz are identical.

[10] am puzzled by the term “Chandra.” I don’t understand this paragraph at all. As to “thermal waves,” Landau arrived originally quite formally at “second sound.” When experiments have failed to found it, Lifshitz arrived through a rigorous formal argument that second sound should be generated through periodical heating. I arrived at this conclusion by a mix of intuition and a not very rigorous formal derivation. Landau never had to explicitly rely on my argument. On another line I advanced an argument that Navier-Stokes viscosity can have two substantially different molecular basis: (1) a liquid type activation mechanism with negative temperature dependence; (2) gas type kinetic viscosity with positive temperature dependence. I suggested that “superfluidity” requires that both vanish. In the supplement Lifshitz claims that my distinction is extremely naïve. I think he was wrong, but I grant that the very low temperature increase of viscosity in liquid helium fits neither of my categories. I am sorry at the time of Lifshitz’s visit at MIT I did not think of fighting this out. Since both Landau and Lifshitz are death tis omission cannot be made up. It is a related symptom that according to Landau liquid helium I is a classical liquid. This is a misleading statement, but it is not as “in the face” mistake. I am afraid I have to let it be. One of the authors of the liquid type viscosity was Yasha Frenkel, whom Landau did not hold in high regard. Landau was not always right.

[11]The contrast between “macroscopic” and “microscopic” belabored in the correspondence is misleading. All three of us exploited an interplay between the two. At this junction the macro- aspect was undisputed, but several micro- aspects were undecided. At present most of the decisions are made and it is more interesting what the truth is, than our illusory beliefs at the time. I note the final verdicts in three issues: (1) Phonon controversy. I assigned phonons to the total liquid, which meant at very low temperatures the superfluid. Landau assigned them to the normal fluid. Landau was right and this was revealed in striking experiments. It was very damaging to my case. It is not worthwhile to deal with my motivations for thinking otherwise, since I was wrong. Landau was also closer to the truth in establishing his macroscopic two-fluid theory. (2) Has Bose-Einstein condensation (BEC) anything to do with superfluidity? London was for, Landau was against. The absence of superfluidity in the fermion He3 and the superfluidity of ultracold gases tipped the balance in favor of the connection. An impartial arbiter Severin Balibar (l’Ecole Normale superieure, Paris) leans for the Bose-Einstein connection, although he sees difficulties: there are 2-dimensional superfluid, although no BEC in 2 dimensions. Why was Landau so firmly opposed? I have a guess. He was much influenced with Bohr’s liquid drop model for the nucleus. He considered Bose-Einstein statistics as the affirmation of an absence of interaction and therefore unrealistic. Alternatively, one can see quantum statistics as tied with the class identity concept of chemical thermodynamics and quantum mechanics, whereas Boltzmann statistics is tied with the orbital identity of Newtonian mechanics. I do believe in the meaning of this dichotomy. There was a third issue emphasized only by London and myself. (3) In 1934 Simon provided a remarkably airtight argument for the importance of quantum zero point energy in liquid helium. A year later London developed an approximate quantitative theory to substantiate Simon’s insight. For London the interaction with Simon stimulated a radical turn from the deadlocked quantum chemistry to cryogenics, an orientation he pursued to the end of his life. It ensured him a wide recognition that eluded him in quantum chemistry. (It is interesting that emigration brought about a change of orientation also for Walter Heitler, in the event toward quantum electrodynamics.) I consider it indefensible that Landau totally ignored the “Oxford contribution.” As a result, for him He I is a classical liquid which is in conflict with the experimental account of viscosities presented in Fig. 19 in London’s Superfluids, Vol. II, p. 38. This conflict is not as flagrant as the one implied by my also phonon theory. It is easy to simply ignore it for Landau and Lifshitz.

[12]I was guilty of such an attempted fudging in my Phys. Rev. paper. I was running high on the success of my macroscopic theories and hoped to construct a microscopic extension, or maybe foundation. Actually, I think this is among my least meaningful contributions to the helium problem. The first version was rejected by the editor, but a tightened version was accepted.

[13]After the discussion with Landau I recognized that I was beyond my depth and I hoped to find some new direction. I spent the summer in a rented farmhouse and I had a lot of time to think. I came up with two ideas. One was a complete flop it was an attempt of a phenomenological theory of superconductivity. It was based on a macroscopic symmetry of the electronic wave function. It coincided with the surprising new departure of Frohlich and Bardeen based on the electron phonon coupling. It was not completely successful, but was to lead to the BCS. It was evident that I was on a completely wrong track, without any redeeming feature. I certainly didn’t come close to the isotope effect. It may be my only paper of which I wish I hadn’t written it. The second idea is a very different story. I wrote a qualitative philosophical paper on the foundation of quantum mechanics. I said that quantum mechanics appears paradoxical only because it is supposed to be based on Newton’s orbital mechanics instead of on chemical thermodynamics, as indeed Planck’s first step was along this line. I think I was completely right, but I had no idea how to put this in a mathematical language. I generated a very hostile reaction, particularly on the part of my head of department, John Slater. This discouraged me from any attempt of publishing, but it circulated in preprint version. I had some favorable responses, particularly Weisskopf and Philip Morrison. Both used the ideas in semi-popular publications. Eventually, a Hungarian translation was published in the Hungarian version of Physics Today. It is a shame that I never published it. My point was that I needed a mathematical expression. In fact, for many years my thinking was directed to a search along this line. In 1989 I published a math version in the Phys. Rev. It was accepted without any objection, but it did not generate any interest. Ever since then it is my hope to break out of this impasse. I never succeeded, but I never gave up.

[14]A book…yes. The idea of my Generalized Thermodynamics undoubtedly came from Landau, but I infused it with other ideas stemming from Slater and Born. I liked Landau’s use of the entropy maximum principle and his statistical interpretation, but this was out of sync with the foundation in terms of deterministic classical mechanics that Landau – Lifshitz invoked in the 1938 Oxford Press edition of their Statistical Physics. The idea of a book emerged as on the third trial I arrived at a happy mix of the original ingredients. It is interesting that Landau as well abandoned the 1938 foundation, but he replaced the classical mechanics anchor with that of quantum mechanics as it appeared in their 1956 Statistical Physics. This was a plausible option, but it led to a much less elementary approach than I was committed to at MIT. I first tried the Caratheodory foundation, but this was a catastrophe. Then I hit on the lucky idea of postulating the entropy, to be justified first by plausibility and eventually by the deductive inferences. The success of the approach was resounding and the need for a text, or at least lecture notes was soon apparent. Yet, my lack of English writing ability seemed forbidding. A mathematician friend William Whitmore came to the rescue. We spent two weeks on the Cape and came back with three chapters to serve as an introduction. It was very popular, even among students of other thermodynamics courses over the Institute. The idea of completing it into a book seemed tempting. Whitmore moved away from the area and Herbert Callen suggested to collaborate on this venture. The collaboration did not go well and Herb accepted a professorship at the University of Pennsylvania in Philadelphia. Eventually, he wrote a textbook alone, giving me generous credit in the Preface. He fully adopted an idea which I developed in my course; to forego the available foundations, such as Landau’s anchored in classical or quantum mechanics and the well-known Caratheodory-Born method. Alternatively I postulated the entropy as a fundamental equation and established experimental connection in a deductive context. Callen fully developed the pedagogical potential of this approach. His text became quite successful. I rather liked it, but didn’t feel that it is quite my child. At one point Carol Bowen, the Directory of MIT Press asked to write my own. I declined, but then we agreed on a compromise. They would reprint my publications, mostly with coauthors and I would write a lengthy introduction, including a historical preamble to provide a pedagogical justification for the abrupt postulational presentation. The main issue that brought about my split with Callen was that he objected to the discussion of history. My Generalized Thermodynamics had invariably good reviews, but none commented on the historical introduction. My book did not compete with Callen’s as a college text.

Session I | Session II | Session III