Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of John H. Van Vleck by Charles Weiner and Gloria Lubkin on 1966 February 28,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
For multiple citations, "AIP" is the preferred abbreviation for the location.
Developments in quantum mechanics, familiarity with the old quantum theory; Edwin C. Kemble is his thesis advisor at Harvard University, 1920-1922. Comparison of Harvard and University of Wisconsin; work and collaboration with graduate students and postdocs at. Wisconsin. Research work in Europe, 1926 and after; high-frequency paramagnetism. Paramagnetic anisotropy. Teaching at University of Michigan, Stanford University, Columbia University, and Harvard University; 1930 Solvay Congress; discussions of research work and papers, 1920s-1940s; awareness of the development of solid state physics; Linus Pauling and the ligand field theory; teaching responsibilities. War work at the Radio Research Laboratory at Harvard as head of the Theory Group; the many duties on advising and reviewing committees during World War II. Chairman of Physics Department at Harvard, 1945-1949; chairmanships and other official functions during the 1950s, excitement of the renewed interest in ligand field theory (chemists); comments on personal interests.
In your interviews with Tom Kuhn for the Project on Sources for History of Quantum Physics you left off in the late 1920's. We hope not only to pick up from there and go on to the present, but also to place your role in those developments in the establishment of quantum mechanics, in historical perspective; that is, in terms of your entire career, and in terms of about forty-five years of developments in physics. I think the best way to proceed would be perhaps to take a look at that period, at the end of the twenties and the beginning of the thirties. There were certain spots in your interview with Tom Kuhn where you wanted to go on but he restrained you because he felt that that interview had a very limited purpose, and he encouraged someone else to go on some day; and here we are, trying to do just that. I would like to ask you to connect the overall view that we're taking here with the more focused approach, and to give us an idea of the developments in quantum mechanics, and your work in it, and how this related to your overall career. This was an early thing for you. It was a major achievement. How did it relate, as you feel, looking back, to the rest of your career?
Why, surely, I was just lucky. I was working in quantum theory when quantum mechanics broke.
Lucky in what sense?
That I was familiar with the old quantum theory.
And this you got because of what?
Because I wrote my Ph.D. thesis under Professor Kemble.
I see. And so his interest in it was a factor. This wasn't true though of many of your American colleagues at that time?
There were very few places where there was much of any interest and understanding of what we now call the old quantum theory. Exceptions were, I think, Kemble at Harvard and Epstein out at Cal Tech. Those are the two I can think of. There were other people who perhaps understood the multiple periodic systems that were used in the old quantum theory, in a formal way, but I don't think they produced the same type of alert interest in comparison with experiment as, say, Kemble and Epstein. I'm sure I've forgotten somebody, but those are the two names that occur to me.
I know you were in touch with Kemble, because you were at Harvard. What about Epstein during this period? Were you in in touch with him?
Not at all.
Did you know him at all? Did you have any acquaintance with him?
Oh, I've met him, in subsequent years. No, at that time, did not. Three thousand miles away was a long distance in those days. Furthermore, I'm not quite clear when Epstein came from Germany to the United States.
I don't know either. I think he was there at CalTech in the twenties, but I don't know. 
I think he did the Stark effect with the old quantum theory in Germany, and with the new quantum mechanics in the United States. I may be wrong.
Did this conscious feeling of being one of just a few in this field have any consequences on your thought? Did you think it unusual that there were only one or two in this country working on that problem?
I don't think I probably thought very much. I just did research as it came along. I was lucky, in that Paul Foote got me to write this report on quantum theory for the National Research Council. This is my book on Quantum Principles and Line Spectra, which was out of date before it got printed, because that was just when quantum mechanics broke.
I know that the Kuhn interview covered that rather extensively. I'm curious as to why you did a thesis under Kemble.
Well, in the first place, I was not an experimentalist. I found his course in quantum theory interesting. I know my father asked Professor Birkhoff, "What about Kemble?" He was a new name to my father. He was not an older man at that time. Birkhoff spoke very highly of Kemble as a person to work under. I had my father's blessing, working under Kemble. Not that he would have exerted any veto power, but he supported what was my natural bent. At that time, I think Kemble was the only pure theorist in the Harvard department, unless you include electronic people. Bridgman, of course, had a good understanding of classical theory and a wonderful physical intuition, but he was primarily an experimentalist. It certainly is true that at that time there was nobody at Wisconsin who had any profound understanding of quantum theory. Max Mason was unsympathetic to it.
Then the choice to go to Harvard was a choice to study under Kemble. Was that it? Or was it a combination of several factors?
Well, the reason I went to Harvard in February, 1920, is very simple. Father was there as a visiting lecturer in mathematics, for a semester, and Mother had the idea that it might be a good idea if I could graduate in three and a half years, which I did, between my summer school work at the University of Minnesota, plus a reading course in Molière. I managed to get enough credits so that I graduated in February, 1920, although my degree was awarded in June, and I spent the term there at Harvard. I liked it because the professors used rather more theory and mathematics than the more experimental people at Wisconsin. I was never far enough up to have any work under either Mason or Mendenhall.
Then the trip to Harvard was made, and you stayed there. Did you notice, when you arrived and began to become established there as a student, any substantial difference in the atmosphere and environment, other than the difference that you had anticipated?
I would say that there were more students in physics, probably, than there were at Wisconsin, where all the undergraduate men tended to major in engineering. Physics was almost a girl's subject.
That's interesting. I think some people have hoped that more girls would be interested in it today.
Yes, it's quite the reverse today.
It would be interesting, I think, to take those two institutions—Wisconsin where you returned and, of course, you returned to Harvard later—and to take a look at their differences. I know we're asking you here to jump in your thinking, but what about Harvard when you returned about 14 years later? Wasn't it in '34?
Yes, you're absolutely right. I left Harvard in '23 and returned in '34, but I first went to Harvard in 1920.
Had you noticed any essential change? Of course, you were coming back with new glasses. You were looking at it in a different way.
The department had not expanded very much in the interim. They had appointed Slater and K.T. Compton lured him away to MIT, when Compton left Princeton to become president of the Institute, and that left a vacancy in the department, and they invited me there as associate professor in 1934.
Were your reasons for making a switch as professor substantially the same as in the earlier period, when you made the switch as a student?
Well, it was a very difficult decision for me, and I haven't very much to comment on that. Wisconsin could match the Harvard salary scale at that time (it was during the Depression—and they were very generous in what they offered me), only by having me in one of those "key man posts", a policy I don't very much approve of in connection with the universities. I like to see a university with a rather well-defined salary scale.
I see. In the interim, certain changes had taken place, in the fields that you contributed to. You gave us your course theses prepared for Professor Bridgman. One of them was in the area of quantum theory. This was 1920 or '21. I think I told you that about a week later I received from Ralph Winch his lecture notes of the course that you gave when he was a student at Wisconsin in 1930 or '31, on quantum mechanics.
As to the 1920-21 thesis—well, I'd say, looking back at it, it consisted probably to a large extent of nonsense, because a lot of the literature of the old quantum theory was nonsense, whereas by 1930 quantum mechanics was a fait accompli, and a well-established law of nature, whereas in 1920 we didn't have this. In connection with my switch from Wisconsin to Harvard, I think I should mention one thing, and that is that the difference in the caliber of the graduate students between Wisconsin and Harvard was not very noticeable. I had a very good group at Wisconsin. I never noticed any orders of magnitude difference, even though Harvard is our oldest educational institution. I'm talking as of that time. I think there was less difference then, in 1934, strange to say, than there was in 1920. But some of this can be a fluctuation phenomenon.
Did you have as close contact with graduate students in Wisconsin as you did at Harvard? Did the general environment of the institution allow a closer relationship in one institution than in another?
I don't think of any particular difference in that respect. At Wisconsin as an undergraduate, of course, I didn't have much contact with graduate students.
No, I mean as a professor at Wisconsin and a professor at Harvard?
Oh, I'd say, if anything there was a little closer contact at Wisconsin because it was a smaller department.
While we're on that, who would you consider, in your recollection, some of the best graduate students you produced and worked with during the Wisconsin period?
Well, of course there's Bob Serber.
He got his Ph.D. just when you left, I guess, in 1934.
Yes. Malcolm Hebb was another student of mine who followed me from Wisconsin to Harvard.
I see. But Serber completed his doctorate at Wisconsin, just as you left?
Yes. Penney, Schlapp, Albert Sherman, Wang, and Niessen were all at Wisconsin as post-doctorate students, although Penney did take an M.A. from the University of Wisconsin. He already had a Ph.D. from Imperial College in London, but he thought it would be nice to have an M.A., so he got one. I am sorry we didn't persuade Hebb to take an M.A. before he left. Wisconsin required an M.A. thesis and Harvard didn't, so it was not an automatic stepping stone towards getting a Ph.D., as it is and was at Harvard.
I notice that you did collaborate on some papers with Penney and Schlapp, but not with Serber.
You mentioned at one time that for sentimental reasons—
Yes, that it would have been nice to have. I'd suggest a thing to Serber and he'd work it out very independently. I did collaborate with Penney. I never did actually with Schlapp, did I? No. But they were closely associated with me, and they worked on problems that I assigned to them.
I see. On the subject of collaboration, is there something about the particular environment and the field as well as individual style which may enter into whether a person collaborates extensively or not. Do you find that your best mode of work is alone, to produce papers by yourself? Or do you find that in your career there's a tendency, in certain periods, to collaborate?
Oh, I think probably by temperament I am more a loner than a collaborator, but I have collaborated on some considerable number of papers.
With people nearby, within physical reach, or people in the field in other parts of the country?
I think you could find both cases. I'd have to look over my bibliography a little bit for this. I don't think I've collaborated very much with people in other places. There are exceptions—a paper that Huber and I wrote recently. He wrote one part at Wisconsin, I wrote one part at Harvard. By and large I should say the papers I wrote in collaboration are usually at my own place. There are exceptions. There was a paper I shared with Townes and Gunther Moore at Columbia. There are also articles which I wrote with Margenau and with Weisskopf.
During your Minnesota period there was only one joint paper. That was with E.L. Hill.
Hill was from Minnesota?
Yes, he wrote his Ph.D. with me.
Hill was the man who took over the Physical Review after Tate?
Let's take a look, if you like, at the period at Minnesota and some of the papers that you worked on during that period. We were talking before about a paper in 1927. I think on your bibliography it is paper No. 13, which represents your earliest research on the theory of paramagnetism.
Except the paper No. B in the appendix.
Now, how can you relate the work on this paper [No. B of the appendix] to the David Dennison article? You mentioned that you came into contact with an article by Dennison and that this stimulated you to work on your own.
Now, why I should be particularly interested in magnetic and electric polarization, I frankly don't know nor how I thought of making that application. I'd read Pauling's papers, where he tried to calculate these dielectric constant factors, multiplying 1/T in the old quantum theory, and one came out with quite puzzling and almost nonsensical results—the fault of the old quantum theory, not the author. It occurred to me to see what would come out of quantum mechanics, and I saw how to do it, thanks to Dennison's paper. At that time I didn't know any wave mechanics. I only knew matrix mechanics.
What did Dennison's paper supply?
The matrix elements, and I plugged them into the formula for the susceptibility, and got out the factor one-third.
Did you know Dennison at this time, or you had just read the paper?
I don't think I knew him. I know we exchanged letters. had his permission to use his results. I wrote him and cleared that.
I see. But you first came across his paper in your capacity as associate editor of the Physical Review?
Were you given it to referee?
Surely. Dr. Tate handed it to me across the table, so to speak. He was editor-in-chief, and I was at Minnesota. Whether I was actually associate editor at that time or just offered de facto help to Tate occasionally on theoretical articles, I wouldn't know without checking back into the precise dates of the two terms that I have served as associate editor.
Do you feel that this was your most important contribution in paramagnetism?
Well, it's hard to say. I'd say that the work from 1926 or '27 through about the Penney and Schlapp period in 1932 probably contained the most fundamental contributions. It's natural, because quantum mechanics had broken, and it's very hard to make as important contributions in later dates as in the pioneer days, when, as I think Kronig— it may have been Dirac—said, "Anybody could write a paper."
In that field.
Yes, or for that matter almost any field. There were many applications of quantum mechanics that opened up, when one had the tools of quantum mechanics.
You use the new availability of the equations of quantum mechanics and therefore the possibilities for people to write papers and to make a contribution, as one of the ingredients of a change taking place in physics in the United States during this period.
Surely. It had to be.
Was this a self-conscious feeling? Did people feel that change was definitely in the air, and that physics in this country was on a new level—in your words, coming of age?
I frankly don't know. I think the more sophisticated people realized that quantum mechanics had broken and they really had something. There were other people that thought it was just another flash in the pan, an attempt to bolster-up the old quantum theory. But it soon became pretty apparent, I think, that this was it.
What about your colleagues at Wisconsin? How did they react? How did Tate, and Mendenhall, who I think was still there?
Tate was at Minnesota.
Yes, pardon me, I meant in the thirties.
At Minnesota, Tate definitely realized that this was the real thing. He even audited some of my lectures. I think Mendenhall appreciated it, but I don't think ever tried to follow it as closely as Tate did.
But a student coming in to do his dissertation in that period would have an opportunity to work on the subject. Did any of your students, in the period at Wisconsin, work on topics related to quantum mechanics?
They all did. At Minnesota, Elmer Hutchisson, the former director of the Institute, had a rather unfortunate thesis topic because it was one of the very last days of the old quantum theory. It was an attempt to apply the quantum theory to the hydrogen molecule, and just about the time that he finished up, quantum mechanics broke, so that my other two Ph.D.'s at Minnesota, Rojansky and Hill, both had quantum mechanical theses. So I directed one thesis in the old quantum theory, the rest in real quantum mechanics.
You mentioned three doctoral students at Minnesota—Hutchisson, Rojansky and Hill. Then at Wisconsin we talked of Serber. Anyone else at Wisconsin? I know Penney and Schlapp were post-doctoral ...
Amelia Frank wrote her Ph.D. with me. Whitelaw, Stehn, Goble, Jordahl. I've probably forgotten somebody. I already mentioned Serber and Hebb.
Yes, all in the field of quantum mechanics at Wisconsin?
How did that carry over at Harvard?
Well, I think all my Ph.D. students, with two notable exceptions, have written theses concerned with quantum mechanics. There have been two Ph.D. students I've had that effectively didn't have any Planck's Constant in their theses. One was Dave Middleton, who wrote a Ph.D. thesis connected with noise problems, which was the outgrowth of studying radar jamming in the war. He was my research assistant there. Then Dudley Towne, who is now at Amherst, wrote a paper concerned with coherent and incoherent scattered radiation. We hoped we could clear the whole thing up quantum mechanically, but he was only able to get essentially the classical part. It's a problem on which I have a student working right now.
It would be interesting to have an estimate from you of the number of Ph.D.'s you have produced over the years.
It would be better if I just thought about that in private and gave you the number.
All right, fine. It would be interesting. [To Lubkin] Did you ever follow through on the question of the next development related to those earlier papers?
Yes, I think so. On papers 11 and 12, you have remarked that you stressed the importance of high-frequency paramagnetism. Was that an idea that people accepted easily?
I don't know quite how to answer that. I think, by and large, yes, but there was one theoretical physicist that I had to argue with quite a while, to convince him that the formulas which we used did not involve assuming a "Paschen-Bach" effect, that the result was invariant of that. There's a section in fine print in my book which I inserted in consequence, just to show that this type of objection was not a valid one.
Who was that, if we may ask?
Oh, I'd rather not say.
But this business of the high-frequency paramagnetism essentially was pushed in 1929? Is that the period?
It was really in my 1927 paper. It was taken up much more seriously after the war. It is sometimes called "Van Vleck paramagnetism," which I think is giving me undue credit. But the point is, that term just is there and has to be included, if you're going to calculate a susceptibility correctly, except in the very simplest cases.
You mentioned, in connection with the paper about the factor of one-third, the article being published in Nature and the editor asking you to trim it down, and Bohr's comments on it subsequently. Can you go into that again, please?
Well, let me see. I worked it out at Minnesota, before I went to Europe with my parents in June, 1926. I guess it was in Copenhagen that I learned that it had to be cut down, and I cut it down. I carried a typewriter with me in those days. I was much more zealous then than now.
Would that have accounted for the simultaneous publication of Pauli's?
It seems awfully unimportant now. Whether I would have been a week or two earlier, I don't know. I would have to check on the dates.
What was Bohr's comment on this subsequently?
He said, "Oh, I wish you'd told us about it. could have endorsed it in the original version."
And with his endorsement it would have gone through?
Probably, because, of course, his name carried a great deal of prestige.
At whose suggestion was it cut down? Was this the editor of Nature?
I see. Nature had a refereeing system?
It must have, of sorts.
You pointed out at one time that some of the European journals did not have a refereeing system, and therefore you could get something published perhaps sooner, though it didn't carry as much prestige as if it had gone through the refereeing system.
I don't know as Zeitschrift für Physik, which had very little refereeing, was especially fast. Nature was of course adapted to rapid publication. Probably I could have published it in Science, but I don't think anybody in Europe read Science.
I see, so the effort was to reach a European audience, since this was the relevant audience?
Yes, Nature and Naturwissenschaften were at that time the two rapid publication media, corresponding a little bit to Physical Review Letters and Physics Letters at present, and perhaps Comptes Rendus,
Philosophical Magazine as still considered a basic outlet for physics papers?
Yes. I was so pleased when they accepted my Ph.D. thesis in toto, in 1922.
Can we turn now to this so-called favorite paper, "The Theory of the Variations in Paramagnetic Anisotropy among Different Salts of the Iron Group"?
Was this the paper that provided the foundation of modern magnetochemistry and the ligand field theory?
In a certain sense, it did. I don't want to claim too much credit there, but it, for instance, did give the basic reason why a nearly cubic field would make the properties of nickel different from those of cobalt, although they were both in F states.
Did this idea get accepted right away by the chemists?
I don't know how much chemists were interested at that time. To be honest, I didn't realize how much interest chemists had in this type of thing until I was invited to a conference on essentially ligand field theory in Dublin, in 1958, I believe.
That was quite a time lag; had they been busy all of that time?
I think mostly in the later years. Of course, the war disrupted things a great deal, from about '39 to '45 or later. I don't remember any invitation to ever give a paper to the chemists in the United States on this type of thing, in the 1930's. They had a chance, of course. [On second thought, I remember that I did talk to the American Chemical Society in Chicago around 1933, but it may not have been on this subject.]
That's an interesting point. Did you notice significant differences in the sense of being a professional community; in physics it might have been one way, in chemistry somewhat different? Were differences apparent to you?
What was theoretical physics in the 1930's became theoretical chemistry in the 1940's. There's that difference, I should say. For instance, Griffith has a book on transition metal ions, which I would say was straight theoretical physics. It was more or less like Condon and Shortley's book, except applied to the solid state rather than the gaseous phase, but was written by a chemist. I think that whole type of thing has been very much taken over by the chemists.
This group theoretical approach which you applied on the salts of the iron group—was this in any way related to the Wigner paper on the SU(2) groups?
No. I don't even know what paper you're referring to.
This is the paper that all the SU(3) theoretical papers refer to.
In Nuclear physics?
No, as far as I know there is no relation of this to that. It was based entirely on what group theory I know, that is to say, Bethe's work.
This takes us up to 1936. Before we leave this period—I don't think we've really explored it enough) notice that 1933, 1934, just before you went to Harvard, was an especially peripatetic period for you. In that one year you taught at Michigan, Stanford, Columbia and Harvard.
That's a pure accident.
How did that come about?
Well, let's see. I had been visiting lecturer at Stanford on my honeymoon in 1927. Then later on, I think they realized that most of their students were away summers, and therefore if they were going to spend the money for a visiting lecturer, it was better spent in the winter than in the summer. So they had a visiting lecturer for the winter term, in 1934. I arranged to be away from Wisconsin and enjoy the California climate, with an appropriate cut in salary, so I could accept the Stanford offer.
Was that for a full semester?
For a quarter. Bob Serber took over my course in quantum mechanics, and Kerst my course in analytical mechanics.
They hadn't received their doctorates?
Neither one had, at that time. But I think Serber was accepted as more or less the dean of the theoretical physics students, at that time. The Columbia offer was just a visiting lectureship in summer school. They had visiting professors in their summer school. It was not nearly as pretentious a symposium as the Ann Arbor show. Then the Harvard offer came along, and that is how it came about that I had the rather unusual experience of teaching at those four schools in one calendar year.
I see. Tell us a bit about Columbia. You had the opportunity to compare because you had been at Minnesota, Wisconsin, Stanford; you were familiar with the scene in general; you had contact with European colleagues. So when you came to Columbia, even for a summer session, you must have had some reaction to it, against this background. I'm curious to know your reaction to the general environment. Tell us something about the individuals there.
Rabi was the key man in their work in quantum mechanics, experimentally, at that time. I don't think the students were very much different from what they were at Wisconsin or Harvard. Graduate students are pretty much the same everywhere.
How about the facilities?
I don't know, I'm just a theoretical physicist, all I need is time and a lead pencil and a library, and I guess they have those everywhere.
You mentioned Michigan. I know that you were in and out of the summer sessions at Michigan. How many do you recall having attended in a period from about 1927, when they started, through the thirties?
I never attended a whole Michigan summer school, except the one I taught, which was, I believe, in 1933. I dropped in on Michigan summer schools in 1928, and, I guess, 1929 and 1931—certainly not in '30 and in '32 as I was in Europe then.
You were close enough to do that.
What was your impression of the general atmosphere of the school, the function it served, the motivation of its organizers, and the people that attended?
Well, at that time, for theoretical physicists it was more or less a pilgrimage to Mecca, because it was the one place where there had been a conscious attempt to gather in one place key figures in quantum mechanics and theoretical physics. It was more in theory. It was not an experimentalist's show especially, but of course many experimental physicists went there, just to learn quantum mechanics—men like Lee DuBridge, for instance.
I have the impression it was not so much a formal program there, that much of it was informal. Is this your recollection?
Yes. Fermi and I were the two visiting professors in 33. I remember he and Uhlenbeck were collaborating on a paper. There were courses that could be taken for credit that were simultaneously going on, but I don't believe that I had any students taking my course for credit. At least I have no recollection of giving any examination. Maybe credit was given, a little bit the same system as Michelson's courses in Chicago. He never had very much in the way of examinations.
Were there other faculty people taking the courses? As well as graduate students, were there also other professors?
That's true. These lectures were attended by staff members from other institutions, including of course, those at Michigan. I should say that the lectures were rather more, in those days, for post-Ph.D. clientele than for pre-Ph.D.
Was there any other institution with a similar approach to make a self-conscious attempt to raise the level of theoretical physics in this country?
I can't think of any with quite the same éclat. Probably the nearest to it was Chicago, which through the years had a very strong summer school. I went to summer schools there in 1919, and in 1920. It was the summer school in physics, I should say—for the early 1920's. Harvard, in the best New England tradition, more or less folded up during the summer months, except for research people, so there was practically nothing in the way of advanced instruction in physics in Harvard in the summers. In the 1920's Wisconsin had a nine weeks session, a more pretentious effort than Harvard's, at one particular era. But the Michigan one was the big show. I wonder—and I frankly don't know—whether they got a grant from Kettering or the state legislature, or how they got the funds, to have this big endeavor in the summer school. I think Dennison would be the person to talk to about that.
We have. I don't recall the answer now, but we discussed this with Colby, Dennison, and Randall. I have some of the origins of it pretty well documented.
Of course, the personal element should not be neglected. I think Randall had the imagination to realize what could be done.
He indicated that it was a deliberate attempt to change the situation.
Oh, surely. I mean, it was quite a different thing, you know, from Chicago, which had had a rather elaborate graduate summer school as part of their regular instructional system. This ran four quarters, around the clock, and a professor could teach in the summer and go south in the winter—which had its points, I think.
During the period, you took a trip to Europe. You were at the Solvay Congress in 1930, and I think you explained in your interview with Tom Kuhn a bit about that Congress. But you didn't touch on your reactions to some of the individuals and some of the events during the Congress. Is that something worth exploring?
Perhaps. I don't know.
I'd like to know, for example, how many Americans attended that year, if you recall?
I can tell you. That's easy. One: J.H. Van Vleck, and he was there. If you look at the Solvay Congress Proceedings, you will see I was listed as an associate rather than as a full member, meaning they did not pay my expenses from the U.S.A. I was in Europe already.
What about the discussions? You'd had discussions with colleagues in this country, and presumably had participated in some meetings here. Was this of a different order, in the sense of the type of discussion and the type of structuring of the program?
I should say they had an extremely distinguished Congress. I'd be interested to talk to you about it some time, whether this perhaps had the most distinguished roster of any Solvay Congress, because it embraced the traditional experimental people, like Madame Curie, and then the new lights of quantum mechanics, such as Fermi and Dirac and Kramers. But I would say that there were many distinguished people at that Congress who had not worked particularly in the quantum theory of magnetism. Consequently, by and large the Congress was composed of people of very great distinction, but oftentimes without much background in magnetism, some of whom did not understand or sympathize with quantum mechanics. Weiss was a hold-out for his Weiss magneton, which is now recognized to be completely spurious, but which took a great deal of the time at that Congress.
Were there other people as young as you in the group?
Oh, surely. Fermi was younger than I was, also Dirac, and Heisenberg. Of the group of people that worked in quantum mechanics in those early days, I'm one of the older ones. I was a year or two older than most of the key figures. I think I'm a year or so older than Pauli; I don't know for sure.
Well, let's just skip back, comparing meetings. Were there any occasions for similar discussions—not necessarily at this high level— bringing international people together? Were there opportunities in the American Physical Society here, or in other groups, for this type of discussion of theoretical physics?
I don't think the American Physical Society had enough steam and knowledge to organize a symposium on the quantum theory of magnetism or something like it around 1930. About that time they might have had a symposium on "quantum theory/` It would be more nearly what they would have.
Do you recall any that they did have on any subject? Or was restricted to the traditional yearly meetings?
Oh, I think it was restricted to the annual meetings. I believe the Franklin Institute sponsored something. I had a paper in that. I couldn't tell without going back whether that was part of a Physical Society meeting, or whether that was a separate session, sponsored by the Franklin Institute. I frankly wouldn't know.
We can check on that. What is the paper number on that?
And what year was that?
1929. You see, the older physicists did not really understand quantum mechanics and this symposium, to a certain extent, was a briefing session, I should say.
Do you remember who attended?
I remember Slater gave a paper. I remember that for sure. I would have to check the Journal of the Franklin Institute to see exactly who else gave papers.
It's available; we can get that. In 1930, when you returned after the Solvay Congress, was there anything to report to your colleagues? Were there any people whom you felt you wanted to discuss it with?
I should say the big thing about my trip to Europe was taking a walk in the dunes of Holland with Kramers, where he pointed out to me Bethe's paper on group theory and the importance of crystalline Stark effect, possibly for the theory of magnetism. That stirred me up to do this work which got delayed because of writing my book, but which as I said, reached its fruition in my 1932 paper, and the papers of Penney and Schlapp which I was instrumental in getting them to work on, and supervised the work of.
In this discussion with Kramers in the sand dunes, how was that point brought up? You don't just kick some sand and think of that. How did you get started on the conversation?
I can't remember. Kramers was always very nice to me.
It was a program of mutual interest.
We'd both gotten interested in the helium atom, in the dying days of the old quantum theory. That is covered in the interview which I had with Kuhn.
That takes us back, of course, into the thirties and the Wisconsin period, and then we talked a bit about the transition to Harvard. Perhaps this is a good point to take a break for a few minutes and see where we are.
I think we should start back now to review some of the basic papers, starting in 1927. Can you give us an explanation of the origins and the significance of these papers? The numbers we refer to are the numbers of papers in the bibliography.
Yes, that's right. Now, as regards papers Nos. 11, 12, and 13, which are mentioned there on that page, they were anteceded by paper No. B of the appendix in my bibliography, which was a note sent to Nature in 1926, where I showed that the factor one-third was restored in the formula for the susceptibility, as compared to very foolish values of the constant of proportionality which were yielded by the old quantum theory. This was done for a very particular model, a rotating diatomic molecule or dumb bell. It was essentially a quadruple tie between Mensing and Pauli, Mannebach, Kronig and myself. Mensing and Pauli, also Mannebach, published in the Physikalische Zeitschrift. Kronig and I published respectively in the Proceedings of the National Academy and in Nature. This is all documented in my book with precise dates. My original letter to the editor, which I sent to Nature, was sent back with the request that it be cut down, which delayed it, I suppose, a month or so. I was pretty upset at the time. It happened to me in Copenhagen. Bohr remarked he wished he'd seen the original manuscript that was sent to Nature; he'd have been glad to have endorsed it, that it should be published in the form it was in.
You mentioned that the experimental agreement—
All right, now, these two gases, NO and 02, which are the two common paramagnetic gases,—Curie knew that oxygen obeyed Curie's law. In fact, that was how he got it, more or less. So there was experimental evidence on oxygen in 1927. On NO, there were only measurements at room temperatures. I did not realize that it would be possible to run NO gas down to somewhat lower temperatures. That was done simultaneously in three different laboratories—here again the details are recorded in my book. It was done by Wiersma at Leiden, by Bitter (at M.I.T., I presume), and by Aharoni and Scherrer at Zurich. I may have some of the names wrong. I remember at the 1930 Solvay Congress, Scherrer said he was going to murder his student for not getting out sooner his experimental results on the deviations from Curie's law for NO, which accorded with my theory.
He had seen your paper previously?
Oh, yes. You see, immediately, when this paper came out in 1927, it occurred independently to physicists at the three different laboratories I have mentioned to try and check my prediction as to the deviations from Curie's law. It was done at three different places. NO was the third known paramagnetic gas at that time, and there was some 2 Japanese work which differed with theory by a factor of five, which rather depressed me. However, I persuaded a graduate student to work on that problem, or rather I persuaded Mendenhall to have an experimental graduate student work on that problem, and that was Havens. The trouble was with the amount of polymerization, etc. When things were straightened out, the theory was completely confirmed for NO2.
Was that Havens' dissertation, or just a problem that he worked on?
It was Havens' dissertation, as I remember it. pretty sure it was. There are two Havens. This Havens, as far as I know, is unrelated to the Columbia Havens. I think he went into industry. I think we've discussed the high and low frequency terms already, haven't we?
I think we've discussed most of these questions.
Do you have any comments to add on why Paper No. 26 was your favorite?
Well, I think it was of importance experimentally, because it really cleared up this mystery of these very strange variations in magnetic behavior from one ion to another. And I liked it because it was such nice simple theoretical physics. It was very similar to the type of calculation which Goudsmit made in showing the multiplet inverted when he went from the left to the right half of an incomplete group of ions, only in the magnetic case there's essentially an inversion centered around a quarter of the period, rather than half the period. I remember how I couldn't understand why Schlapp found he had to have apparently a crystalline field of one sign for nickel, and the other for cobalt. Then it occurred to me that you could really show, in a very simple way, that these Stark patterns would turn upside down, in going from nickel to cobalt for one and the same crystalline potential. I remember explaining it to Penney and Schlapp, and they were quite amazed. They hadn't caught it. I showed how it came out.
Over how long a period of time did you develop this paper before you submitted it for publication?
This is the type of thing that, when it dawns on you, it just comes out in an hour. This is no long calculation. I mean, it is a very simple matter once you know how to do it. You just marvel that it took so long for you to realize it. I remember I felt that was an important paper. I took it down and put it on a mail train at 1 o'clock in the morning, to get it to Minneapolis to Jack Tate, who was editor of the Physical Review, and discovered later that he'd gone on a fishing trip for two or three weeks, so I didn't need to lose the sleep to mail off that paper. I really felt that was an important paper. I felt that at the time, and I believed it. When I got my theory of NO, I didn't know whether I believed it or not.
Let's talk about that hour of creation, when you knew you had it. In a sense you say it came upon you. Do you remember how this occurred? Were you thinking about it?
Well, I wanted to turn my graduate students upside down, so to speak, as regards the sign of their crystalline field as you went from one element to another. It looked purely ad hoc. And then it occurred to me that it did come out of a consistent sign in the crystalline field. Later Gorter, in a very important letter to the editor, pointed out that the absolute sign came out correctly. I didn't think of trying to see what sign of the crystalline field you'd get from fourth-order terms from the nearest neighbor's space with a sixfold coordination. Then he showed with a fourfold coordination, the inversion pattern should go the other way, and that was confirmed in India by Krisman. That pleased me very much.
About how long a period was there from the time you sent off the paper, with great excitement, on the mail train to the time of publication?
Oh, I don't know. I would guess on the order of two or three months. We could check that later.
Here it is. They did it very quickly. June 6, received; July 15, published.
Well, that's very good.
Of course, there was probably a time lag of a little bit in appearance in the Physical Review, but I'd like to say that I think John Tate was exceedingly kind to me, in the papers which I submitted to the Physical Review. They were usually published promptly. I am very anxious to be out at the dedication of the new physics laboratory to him, this June. You know about that?
Yes. On Paper 26, in the interim period—which wasn't a very long period, fortunately, in this case—did you communicate results to anyone other than the immediate colleagues and graduate students?
Let's see. This was in June, 1932, wasn't it? I think I went to Europe that summer. I am trying to remember my itinerary. I was in Holland. I presume I told some of the Dutch physicists what went on.
Was this prior to publication?
After, I guess, or while it was in press. I mean, after all, it was in August, and this article came out the 15th of July, so it didn't make much difference whether it had actually appeared or not. I remember, I got Gorter interested in this, and that's why he wrote this letter to the editor, checking the absolute sign of the behavior of crystalline fields. So I must have talked to Gorter, because he made a calculation that I should have been bright enough to have realized that I should have made myself—which he made in that letter to the editor.
Did you communicate the results by letter to others in this period?
I don't know. We didn't have very much secretarial help, and I think the general theory was never to write a letter if you could avoid it. I might have, to somebody, but it was sufficiently rapid publication so that I could wait till it came out.
Were preprints at all common in that period?
No, and I'm almost tempted to say, no, thank God they weren't. There are these tomes of preprints nowadays. Oh, it was nice and simple in those days. I kept all my reprints in one drawer of a filing cabinet in my office at home.
I've seen your filing cabinet now, and—
It's a shambles. It depresses me. Do you have my answer to Gorter in the Ann. de l'Institut Henri Poincaré article? That pretty much summarizes the climate as of 1939. gave eight lectures in French at the Institut Henri Poincaré on the theory of magnetism to a rather small audience. It's pretty hard for me to lecture in French. I had to go to the Riviera afterwards to recuperate; I don't know what the audience had to do. But I did stay in Paris and just worked on writing this paper up for publication. It was, in some cases, translated from French into French into French. I wrote it in French, and then it was edited by a woman who gave professional French lessons to foreigners at the Institut de Pantheon, I think it was called, or something of the kind. Then Proca, the editor of the Annales de l'Institut Henri Poincaré went over it, and very often he changed the French back to my own form. I was a little bit more familiar with the usage of physicists than a non-physicist literary Frenchwoman would be.
This was in 1939?
I think that the monograph holds the all-time record for how long something was in proof. I got the proofs the day the French Army collapsed at Sedan. Mrs. Kauffmann (formerly Miss Hildebrand) took the proofs through the lines between Free and Occupied France. [The University of Strasbourg had been moved to Claremont Ferrand during the war. Mrs. Kauffmann (then probably Miss Hildebrand) was transferred there from Strasburg and was able somehow to get to Paris where the printer was.] But they didn't come out until after the war.
Were these the lectures that you based the Reviews of Modern Physics paper upon? There is a footnote in the RMP article on theory of paramagnetism.
Well, that is, perhaps, more or less a rewrite of one chapter of these eight chapters in my Institut Henri Poincaré article. It was almost a book; it runs almost 200 pages.
What item in the bibliography?
I don't find it here. Oh, yes, No. 79. See, that's almost a book. It runs 130 pages.
What is the title of that, from the bibliography?
"Quelques Aspects de la Théorie de Magnetisme."
All right. Then in 1936, going back a bit, there's a paper on non-orthogonality and paramagnetism. Could you give us a bit of the background of that, if you think it's significant?
Well, let's see. Inglis claimed that the non-orthogonality of wave functions spoiled everything in the Heisenberg theory of ferromagnetism. I wrote this paper to show that things weren't as bad as they thought they were; if you included certain correction terms then you got rid of most of the troubles. Of course, it can be argued, if you go still further on you may still run into troubles. Slater always subscribed a little to that philosophy. But I do think that this showed that things weren't as bad as they looked. I know my father helped me with a critical reference there, to summing some series. I referred to him in the paper.
Where was he at the time?
At Madison. My wife and I very often spent several weeks at Madison during the summer—it's a beautiful place then. And between working and trying to write up a paper such as Nos. 41 or 42, I could go and take a swim in Lake Mendota, and reflect on the theory on the pier. It's a good place to work, all right. So that is that. There has been much more refined work done along the line of the question raised in 43 by, I believe, Carr and other physicists.
Going back to paper 43 on non-orthogonality, would you say this was a new way of solving a problem that Pauling had dealt with?
No. I'm afraid it is essentially just trying to show that a worry created by Inglis was not as bad as it looked. It's a somewhat specialized paper.
Then in 1937 in paper No. 48, "On the Anisotropy of Cubic Ferromagnetic Crystals," you attribute the anisotropy to anisotropic exchange. Was that idea received easily?
I just don't think it was received. Very few people were interested in that class of problem. The war came along. In postwar days, that is referred to as a classic paper, but I don't remember much reaction to it at the time. There wasn't much interest in these things in those days. It's curious that my supply of that particular reprint is exhausted, but I think I still have some of my 1932 paper, and also some of my 1927 papers. Strange how after the years some of your best papers are still not exhausted in their reprint supply, and some of those you don't feel too proud of are gone.
How would you account for the change of interest in this later? What had taken place in between?
Well, there was a new appreciation of what quantum mechanics can do, plus the tremendous expansion of manpower in theoretical physics and theoretical chemistry; and also a new generation of experimental physicists who understand theory in a way that experimentalists of the early days did not understand quantum theory.
You feel they developed this appreciation for theory as part of their education?
I would say so, by and large, because that generation grew up during the days when quantum mechanics was well established. I suppose you can find an occasional experimental physicist that got into the spirit of quantum mechanics after it came along, but I would say that it was hard to teach an old dog new tricks, and it was rather hard for those people to appreciate the nuances of quantum mechanics.
Did you find in your own experience that you were teaching more experimental physicists in later years, and that experimental physics was now becoming part of the education of a theoretical physicist? Did you find that reflected in your own teaching?
I don't know. Most of the experimental physicists at Minnesota took my courses. Brattain took my course in quantum mechanics. I've always been proud of the fact that I taught quantum mechanics to two of the three co-discoverers of the transistor.
Who was the other one, Shockley or Bardeen?
Bardeen: Bardeen was a very bright undergraduate at the University of Wisconsin. After graduation, he stayed for a year of graduate work in 1928-29. He was the star of the class. I cannot claim him as a Ph.D. thesis student of mine. He graduated and went into industry, and later on took his Ph.D. at Princeton, and then was in the Society of Fellows at Harvard. I wrote a paper with him then.
The whole concept of the solid state, and the terminology that we use today as solid state physics—was this a defined field of physics in the thirties? Did you see it developing as such?
I never really thought of it as such until they began to talk about a division of solid state physics in the American Physical Society. Now, when was that created? I think that would answer the question.
In l947. Smoluchowski was involved, [discussions regarding the need started during the war.]
All right. It was about that time. You see, the war came along, and first you had to have lots of people recover from the war, and then you began to get this tremendous influx of physicists and chemists into modern physics.
But essentially all through the thirties in your own work, you were working in the solid state.
Without the label.
Oh, certainly. The terminology was unheard-of at the time.
Are there any more comments you'd like to make on the first page of that summary?
Of course, in connection with ligand field theory, Pauling had a somewhat different approach, where he assumed that the positive ions (at least as I understand it) are essentially negative ions, which is rather hard to believe, except insofar as you say that any true wave function is a combination of everything in sight. Of course, Pauling did have his model of hybridization. That should be mentioned.
Did he communicate with you on this or did you communicate with him?
He was primarily then in chemistry circles, and you wouldn't ordinarily attend the same meetings.
So it would be a question of his coming across your paper in the literature. But were you in the habit of sending reprints of your papers?
I mailed out reprints after they got published. I was more zealous in those days than I am now.
Have you preserved your list, or do you think it might be somewhere in your file, of people to whom you sent reprints?
My bookkeeping system is very bad. I might have fragmentary lists, but I would like to emphasize the fragmentary.
I think it might be interesting to see who knew of your work and when.
In retrospect it looks like rather a strange list, because many of the people that I sent to have never been active in the fields involved.
What did you do with the drafts of the manuscripts, the working papers of these articles, in this period? Did you preserve them? Was there any particular system that you followed?
Certainly no system. I may have occasionally kept some of the original manuscripts. Usually when they got accepted for the Physical Review and I got the proofs and the article came out, why, I probably threw the manuscript away.
What about your working papers?
Oh, they're undecipherable to anybody.
But you have preserved them?
I don't think so.
Does that take us into page 2 of the technical summary?
I think so. I have a feeling, in connection with the top of page 2, that I identified the spectral classification of one of the ruby lines that has been very much used in lasers. I've always been proud of that, of giving it its spectral terminology, but it has nothing to do with the use in lasers because it was discovered experimentally that there were sharp lines. However, I gave the spectral classification, in terms of the parameters of cubic field, or rather the group theory nomenclature of the cubic field.
In which paper was this?
This is "Energy Levels of Chrome Alum," with R. Finkelstein.
In what sense was this a pioneer attempt? I think you used that term.
Well, in those days I don't think anybody ever tried to assign energy levels in the solid state. It seems unbelievably primitive nowadays, but it's rather remarkable in a way how big a time lag there was between 1926 and this terrific drive to apply quantum mechanics in the solid state. From '26 to '41, there you are—fifteen years. Nobody had done very much in solid state physics. Becquerel had talked a little bit about his interpretation of his experiments, but he was no theorist. So that's the reason this was a pioneer attempt, I guess.
When would you say this terrific surge occurred where everyone wanted to apply quantum mechanics to the solid state?
In the post-war period.
The thing you mentioned before, then, is related.
The jelling of this feeling of the need to form a separate division of the Physical Society.
Of course, the laser and maser have accentuated the interest in solid state levels.
How much of an influence did the availability of new experimental tools have to do with the postwar interest?
I think it had a great deal to do with it. There were various things. There was the existence of microwave devices which came out of the radar work during the war. That made the microwave spectra of solids a new and virgin field which was opened up after the war. That's one aspect. The rare earth compounds are much less expensive and easier to produce than they were in the old days. I think that's another factor. And since money's more plentiful there are more people in physics. It's a combination of effects.
The changes in production of the rare earths were a direct outcome of the war and war projects.
I don't know whether it was an outcome of the war projects especially. Didn't they find some rare earths over in India? Frankly, I don't know. All I do know is that they're not so rare as they used to be.
If we may jump back a minute, to another subject area of the thirties that you developed, the quantum mechanics of chemical binding, paper 42.
Paper no. 42 deals essentially with ligands, and you can say it's magnetism, you can say it's complex salts, whichever way you want to call it. I guess a chemist would say it's inorganic chemistry.
Were you working with anyone in this field at the time?
No. I never had any—what will I say—chemists working with me who got interested experimentally in applying any of these ideas or stirring me up on them. I should say that I wrote these papers to a considerable extent as a lone wolf. Not many people were interested in these things. The climate was very different then than it is now.
Did you lose interest in the chemical binding problem after, say, 1937?
Well, war came along, for one thing, and ligand field theory and magnetism seemed perhaps a little more interesting to me because the results were more succinct. All of these are calculations on molecules; the type of thing that I did in papers nos. 27, 28 and 31 are what I would call skeleton calculations, but not an attempt to ever nail things down accurately. There have of course been chemists that have spent their whole life work in trying to calculate elaborate wave functions and get agreement with experiment. In principle it should be possible to just put the Shrödinger equation on a computing machine and see what happens to any collection of atoms, but that they have not been able to do, as yet. Well, I think that accounts more or less for why I drifted out of the field. I'm a little bit of an opportunist. When I see a chance to write a paper on something or other, and have an idea, I write it. As it happened no particular further ideas occurred to me along those lines. So I think that is the answer.
Going back to magnetism per se, in 1941 and in 1940, you have a couple of papers: one on "The Extension of Waller's Fundamental Theory of Paramagnetic Relaxation," and another based on Casimir's and Dup ideas. Do you have anything to say about these?
Well, Waller had some wonderful ideas connected with paramagnetic relaxation that he never followed up very much, and I applied those ideas and elaborated them, in connection with the paramagnetic relaxation, particularly of titanium and chromium alum. For a while the Leiden people thought that it would be impossible for titanium alum to relax because Kramers' double degeneracy always persisted in an electric field. It could be violated only in virtue of a magnetic field, and it seemed very hard to get enough coupling to give the requisite spin-lattice effects. Kronig and I hit independently on the idea that there was enough coupling, and showed what the term was for the helium region. Kronig published first. When it comes to the liquid air region, there is a term which gives a relaxation frequency proportional to T7 or to T9, which was not in Kronig's paper, and which is the most important term in that region—unless there is also the Orbach process, which came many many years later.
The paper written in 1941 is what I call my bottleneck paper, paper no. 67. The point is that sometimes the phonons may be unable to carry the energy from the spin system to the walls of the vessel, so that the bottleneck, instead of being the coupling between the spins and the lattice, is simply that the phonons get warmed up to the temperature of the spin but can't carry that energy fast enough to the walls, even though they're moving at the velocity of sound, which can always be more or less of an upper limit. This is just an elementary calculation. If they transport energy at the velocity of sound, that's the fastest they can take the energy out. That was a paper that nobody thought about for years. Then Gorter came out about ten years later with some evidence for a phonon bottleneck, and in the last few years there's been a rash of papers, where the whole thing has been studied much more quantitatively than I ever dreamed it would be.
So there was no interest in the theory, or experimental verification?
I always felt a little hurt that nobody's ever paid much attention to that '41 paper, for about ten years, and then all of a sudden it got into the limelight.
There was another paper in 1941, in which you made some predictions for the susceptibility of antiferromagnetic materials.
Yes, that's the paper which is one-third Bitter and two-thirds Néel. Néel calculated the susceptibility below the Néel point, on the assumption that the molecular field was perpendicular to the applied field, and Bitter on the assumption that it was parallel, and in the powder—you average them, two-thirds of one and one-third of the other. I've always felt badly that I didn't hit upon the idea of anti-ferromagnetism when I wrote my book. I realized that antiferromagnetic exchange could give a term T + delta in the Curie denominator, but I did not realize it could give an ordered structure. I should have realized it. Néel was the one that pointed that out.
How close to absolute zero were the experimenters?
This two-thirds has nothing to do with obtainment of very low temperatures, because the Néel points in those early experiments of Tsai, Squire, and Bizette were around liquid air temperatures. From the standpoint of those experiments, well, all I needed to do was to go down to 20°K or something like that. Not much happened below that. Of course, there were adiabatic demagnetization experiments available in the 1930's. That's described in one chapter of the Annales de L'Institut de Henri Poincaré lectures. Kürti was one of the leading experimenters in that, and Hebb and Purcell wrote a pioneer paper on the theory of adiabatic demagnetization in cubic crystalline fields, as an outgrowth of a course that they took with me.
When was that?
That they did the paper, or took the course?
Both. I was rather pleased that the two or three times I gave that course in theory of electric and magnetic susceptibilities, it led to a paper either by the students in the course or by me.
Well, I think that we have covered many of the major papers. Let's use that as a sort of hitching point for broader discussions, for the period up to about 1941, the beginning of the war, when I am assuming a lot of basic changes occurred in physics and in your career. I think that it might be good to finish off this initial session with brief comments—not too brief—on this earlier period, representing your graduate work in the 1920's, through the early 1940's, a twenty-year period. We've covered some of the changes occurring in the period. Is there anything about it that seems characteristic to you, now that you think of it?
There are one or two things that seem rather amazing. When I went to Harvard in 1934, there was enough manpower that after I went there the course in quantum mechanics could be given every year. It had previously been given only every other year. It had been given every year at Wisconsin and Minnesota when I was there.
I see. This represented one change.
It seems very interesting in retrospect that its fundamental role was so little realized. There would be eight or nine students— such a low enrollment that you wondered whether it was worth giving every year, whereas after the war there were tidal waves of students taking Schwinger's lectures, which had to be shifted to the large lecture room, and so on.
Was the average student just as good when the volume increased?
Why, I think so. Surely. It's sort of like the athletes— the number of ones who can run a mile in under—what is it, four minutes?— seems to be steadily increasing. It's a little that way with the students.
What courses did you give at Harvard, during the thirties?
I gave quantum mechanics sometimes every year, sometimes every other year. Probably I alternated with Kemble on that, I can't remember. I gave dynamics. I gave thermodynamics. I was half in the mathematics department, so that my teaching there was always either a course in mechanics or dynamics, or a course in group theory, which is pretty much a physicist's course in group theory, as group theory is applied to quantum mechanics.
During this period you went to Princeton in 1937. What were you doing there?
I was a visiting lecturer.
For a semester, was it?
For a semester, yes.
That is one institution we have not introduced yet. Again, what comparison can you make of the situation at Princeton with your other teaching experiences?
Certainly some very able students attended my lectures. I remember a man by the name of Tukey who audited my lectures. Princeton had a very peculiar system, at least peculiar to me, in the sense that they didn't have any examinations or problem work, unlike the way we scrimmage the students at Harvard, Minnesota or Wisconsin. I perhaps introduced the Harvard system to Minnesota and Wisconsin a little bit, in the sense of expecting a good deal in the way of problem work.
Who was at Princeton on the faculty during that period that you worked there?
Well, I don't know as I worked. Gilbert King followed me down there, and he wrote a paper with me at that period, or rather he wrote a paper and insisted on putting my name on it, very generously. Feinberg was there. Johnny Wheeler was there, as assistant professor or something of the kind. I guess Einstein was there, for that matter; von Neumann was certainly there.
This was, again, Princeton's attempt to enrich and exchange a bit?
I think so, yes.
Was this going on at Harvard too? Did you have visiting professors at Harvard during this period?
No. At least I can't recall many, in particular. You see, from '34, it wasn't very long till the war came, and openings—I mean, at Harvard were scarce. It was the Depression. The budget of the university was very limited at that period. Harvard was almost unique in that they had no salary cut during the Depression. But from '34 to when the war started and so on is a relatively little compass of time.
Were there new entrants into physics faculties, with the arrival of refugee physicists from Europe?
The only one of that category that we had at Harvard was Professor Philipp Frank, who of course you know was partly in philosophy. I think his appointment was definitely the result of that situation.
Well, that really takes us up to certain changes that took place in the beginnings of the forties, and perhaps now this is a good time— after your long train ride, though you show no signs of fatigue from that— it's a good time, I think, to stop, if that is agreeable to you.
The Theory of Electric and Magnetic Susceptibilities, Clarendon Press, Word, 1932, P. 147.