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 Carl Eckart by John L. Heilbron on 1962 May 31,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
For multiple citations, "AIP" is the preferred abbreviation for the location.
Part of the Archives for the History of Quantum Physics oral history collection, which includes tapes and transcripts of oral history interviews conducted with circa 100 atomic and quantum physicists. Subjects discuss their family backgrounds, how they became interested in physics, their educations, people who influenced them, their careers including social influences on the conditions of research, and the state of atomic, nuclear, and quantum physics during the period in which they worked. Discussions of scientific matters relate to work that was done between approximately 1900 and 1930, with an emphasis on the discovery and interpretations of quantum mechanics in the 1920s. Also prominently mentioned are: Edwin Plimpton Adams, Niels Henrik David Bohr, Max Born, Frank W. Bubb, Arthur Compton, Karl Compton, Ona K. Defoe, William Duane, Paul Sophus Epstein, Werner Heisenberg, Otto Hellwig, Frank Clark Hoyt, Georg Eric MacDonnell Jauncey, Cornelius Lanczos, Hendrik Antoon Lorentz, Wolfgang Pauli, Philip Rau, Henry Norris Russell, Erwin Schrödinger, Arnold Sommerfeld, John Von Neumann, Norbert Wiener; American Physical Society meeting (Washington), California Institute of Technology, Kaffee Heck (Munich), Princeton University, Princeton University Physics Colloquium, and Washington University.
I will follow your general outline. My interest in science was awakened largely by three men. It would be possible to spend considerably more time than we have available without adequately acknowledging my indebtedness to them. The first was my uncle, Otto Hellwig, who suffered throughout his life from the after-effect of spinal meningitis, as well as from an injury to his eye which occurred during my seventh or eighth summer. This injury enforced a period of idleness on him, during which he introduced me to the use of tools. He was a superb craftsman, despite his handicap, and had educated himself in several technological areas, including steam engines and elementary electricity and magnetism. He never complained, not even when — as I now realize — he must have been in severe pain. My relation to him was probably as close as is possible for two rather shy people, and it continued throughout his life. The second man was Philip Rau whose family ran a store in “Dago Hill.” This was a depressed area at the edge of town, and the family was much more than a storekeeper and helped many of the families who were either destitute or in trouble. Philip was a self-educated entomologist who ultimately became an authority on the insects of the local area and read many papers at the St. Louis Academy of Science. We became acquainted at the St. Louis Ethical Society, where he led a nature study class. The third man was Frank W. Bubb, who was professor of applied mathematics at Washington University. Beginning in my freshman year he gave me a job as student assistant in his classes on mechanical drawing and descriptive geometry. He introduced me, together with a number of others — a rather queer assortment of people — to vector analysis and the elements of theoretical physics, making it possible for me to do independent reading in such books as Silberstein’s Electromagnetic Theory of Light and, to read Richard Tolman’s series of articles on statistical mechanics in the Journal of the American Chemical Society. On finishing my undergraduate work in June of 1922 a sharp difference of opinion developed between Frank and myself. I was determined on a career in mathematics, and he was equally determined that I should go into physics. I applied for a teaching fellowship in mathematics, to enable me to continue in graduate school. As the summer passed, it became obvious that this application would not be granted, and I became very depressed. There was some smugness and glee on Frank’s face as he urged me to go at once to the physics department and apply for a teaching fellowship there. I did so and immediately realized that he had foreseen all this and done an excellent job of preparing the ground, for the fellowship was awarded to me within fifteen minutes of my knock on Compton’s door. I departed, somewhat less depressed but decidedly not elated. Turning now to the graduate work at Washington University during the year of 1922 to ‘23. The department was very different than a physics department in the 1960’s. Its faculty consisted of four, namely Lindley Pyle, who had taught me elementary physics; Dr. Hagenow; G. E. M. Jauncey; and A. H. Compton. Newell Freeman, who was a year or two ahead of me throughout university, occupied a somewhat nebulous position, possibly as an instructor in the department. Mr. Reinhardt was the instrument maker, a position of considerably greater importance in those days than it would be today. There were four graduate students, Ona K. Defoe, Howard May, and someone whose name has escaped me. Of the four of us, only Defoe and I went on to the Ph.D. In spite of the small student body, the department’s space was overcrowded and to my good fortune I was assigned a desk in Jauncey’s office, which communicated with Compton’s office next door. Each afternoon Compton and Jauncey discussed the research work of the department, which consisted almost entirely of the study of scattered X-rays. Moreover, there was no objection to my listening in or even interrupting them with questions. Very soon these discussions began to center on Compton’s work on gamma rays at the Cavendish Laboratory in Cambridge, England. He had measured the absorption of primary and scattered gamma rays and established that the scattered rays were softer, had a greater coefficient of absorption. Sometime during these discussions a series of papers by Erwin Schrodinger in the Physikalische Zeitschrift were discussed in detail, and I remember reading them with considerable care. In the early fall an afternoon came around when it was clearly recognized by Compton and Jauncey that the essential element in Schrodinger’s papers was the assignment of a momentum to the photon. Einstein had much earlier investigated the photo-electric effect by assigning an energy to the photon and had later also shown the necessity for the momentum, but Schrodinger worked this out with detailed examples. In parenthesis: ‘photon’ is a word that was not then in use; one talked of corpuscles or light quanta or simply quanta, but I shall use the modern terminology. The total curriculum consisted of two courses, one on electrostatics, taught by G. E. N. Jauncey, the other on optics taught by Hagenow. This last included experimental work with interferometers of various kinds. The morning after the afternoon just mentioned, Jauncey’s class was taken over by Arthur Compton. Jauncey sat in the audience, and we heard for the first time the derivation of the equations of the Compton effect, including the famous .024 Angstrom unit shift of the scattered radiation.
Had the equations come to Compton in the evening of that day?
Yes. But here is a point that I am slightly hesitant to mention and which Jauncey didn’t mention to me until many years later: he remarked somewhat plaintively that he had promised to take his wife to the movies that evening, and that he often wondered whether, if Compton had gone to the movies and, he had stayed home, whether he would have derived these equations. [Note added during correction: To avoid all possibility of misunderstanding: Jauncey never expressed any personal dissatisfaction with Compton, only a certain wistfulness about the events. Certainly in retrospect the discussions between Compton and Jauncey seem to have been necessary preliminaries to the derivation of these equations.) Jauncey and I were certainly aware of the importance of the occasion when Compton gave us the equations for the first time. Compton also was fully aware of their implications and immediately began preparations for their experimental test.
Was Jauncey involved in any of this experimental work?
The experimental work was certainly entirely Compton’s, there is no question of that. In fact, even the construction of the equipment was done by him with his own hands. The first experiment was a complete failure. In order to obtain sufficient intensity of the scattered radiation, he decided to place a graphite scatterer inside a Coolidge X-ray tube. He took a General Electric tube, cut it open, and inserted this graphite scatterer on a tungsten wire support. The experiment probably would have succeeded had it not been that the support was too frail to withstand the electrostatic forces. The scatterer was drawn toward the other electrodes, short-circuited them, and made a big mess. Instead of continuing with this design, Compton adopted a bolder procedure. At that time it was considered that the large bulb of the Coolidge tube was essential to prevent puncture of the wall. Compton now constructed a much smaller X-ray tube, using the Coolidge electrodes but using a two or two and a half inch straight pyrex tube. The result was an X-ray tube of the kind now very familiar in all dental offices. He made two of these, baked them out to produce a good vacuum and set them up with an external scatterer and a spectrometer. All of this took considerable time, and if I can trust my memory, it was on Good Friday of the year 1923 that I went out to the University, found that Compton had spent the night in his laboratory and obtained the first spectrum showing the modified and unmodified scattered lines. That evening, I told my family that Compton would receive the Nobel Prize for this work; this was met with considerable incredulity. Turning now to my work with Jauncey. This began at the beginning of the school year. He was interested in the temperature coefficient of X-ray scattering, and I assisted him in making measurements and constructed a small furnace to heat the scatterers. This work was not interrupted by the admittedly much more important work that Compton was doing, but the afternoon discussions continued, and I have been in recent years particularly interested in recalling these as accurately as possible. Necessarily the difference between the modified and unmodified lines came into these discussions. Compton was very clear on this. The modified line resulted from an interaction between a photon and a loosely bound electron, so loosely bound that its binding energy was essentially negligible. Under these conditions both the conservation of energy and momentum came into the calculations and determined the change of wave length. On the other hand, if the electron was tightly bound, there were two possibilities. Either the photon was sufficiently energetic to ionize the atom, giving up almost its whole energy to the electron and giving up its momentum (but a negligible fraction of its energy) to the nucleus of the atom or a number of atoms. This is the photoelectric effect. The other possibility was that the photon did not have enough energy to ionize the tightly bound electron, in which case its momentum, and a very negligible, very small part, of its energy would be given up to the whole atom or to the lattice of the crystal. This would result in the unmodified Compton line. These ideas did not spring full grown from Compton’s head, as did the equations, but developed during the year, especially after the experiment succeeded. Today it is an interesting question, why did no one in those days think of the Mossbauer effect: The answer is probably to be found in the difference between our present attitude toward radioactivity and the attitude in those days. Very few laboratories, and certainly not the one at Washington University, were equipped to do any experiments with radioactivity. Moreover, radioactive substances in a relatively pure state were so valuable that no one would think of using them to “contaminate” baser materials such as crystalline graphite or calcite. Finally, it had not yet occurred to any one that the nucleus had excited states, even though the emission of gamma rays was known. Resonance absorption of gamma rays was, probably, not even thought of. A further point on which Compton and Jauncey were very clear was that the modified radiation was incoherent and should not produce interference effects, that is Bragg lines or Laue spots. In spite of this, apparently, Jauncey and I investigated the experimental evidence. This evidence was found to be negative and was recorded in the paper in Nature which you have cited. On rereading it, I am also curious why none of the theoretical items I’ve just mentioned are included in the paper. They were certainly very clear in Compton’s mind and it is possible that we considered that they were his to publish and not ours. Or maybe they became clear only after our paper. But I do not really remember anything about the preparation of this paper, not even why my name was attached to it. Despite Hagenow’s course, the concept of incoherence was very difficult for me but that is precisely why I do remember Compton’s attempts to make it clear to me. Ona K. Defoe obtained his doctor’s degree for a study of the modified and unmodified Compton lines, and he continued this at the St. Louis College of Pharmacy. One ought to look up his thesis and other papers in connection with the unmodified lines. He died some three or four years ago so there’s no possibility of consulting him, but his papers may shed some light on this.
Do you recall the Duane-Compton arguments, when Duane was unable to find at first the Compton effect?
The Duane-Compton argument occurred in the following year, and by that time I was in Princeton and no longer working on X-rays so that my memory of that controversy is rather vague. The one incident that I recall was at the annual meeting of the Physical Society in Washington in 1924, at which time both Compton and Duane spoke, and Duane very gracefully acknowledged that Compton’s work was correct and that his own work was essentially a study of other effects. The great majority of the audience was very pleased at this outcome, but one man made himself exceedingly unpopular by asking antagonistic questions of Duane. The audience as a whole felt that both Duane and Compton had behaved with complete professional dignity and correctness throughout the whole affair. In particular, Compton went to Duane’s laboratory and the two worked together in order to resolve the issue. Returning to the late spring of 1923, it was announced that Compton would go to Chicago the following year, and the future at Washington University was not clear. Through Arthur Compton’s recommendation I received a fellowship with his brother K. T. Compton at Princeton University. This was a fellowship created by the General Electric Company and required me to spend two summers at the laboratory of the Edison Lamp Works in Harrison, New Jersey. This laboratory was directed by Saul Dushman, and I have very warm recollections of him, both as a scientist and as a personal friend and adviser. This move to Princeton meant a shift in my research work since there was no X-ray work going on at Princeton at that time. Karl Compton was interested in what today would be called plasma, and this was also the interest of the Edison Lamp Works. It was endeavoring to develop cold white light. We used fluorescence in order to detect the presence of ultraviolet radiation from neon discharge tubes but completely missed the invention of the modern fluorescent lamp which produces white light through the medium of fluorescence. At Princeton there was a much broader selection of courses. A course on electron theory from Karl Compton using Richardson’s book The Electron Theory of Matter, courses on mechanics, statistical mechanics, and quantum theory from E. P. Adams; and several courses on tensor analysis from Oswald Veblen and Luther Eisenhardt; W. F. Magie had lived through the days when thermodynamics was at the frontier of physics and gave a very thorough and stimulating course. There were many people whose names will be familiar to physicists. H. D. Smyth was already an instructor in physics. His brother Charles P. Smyth was assistant professor in chemistry. Joe Morris and Merle Tuve were instructors in physics, both candidates for the Ph.D. Charles Zahn was a graduate student, as were a number of others who have not followed the career of physics as actively as these.
Do you recall the extent of Adams’ course on quantum mechanics?
Adams had written a report on quantum theory to the National Research Council which was in many respects highly original; this report and his course both stressed the relation between Poincare’s integral invariants and the Bohr-Sommerfeld quantum integral. He led me to study Cartan’s book. This was my first formal course in quantum mechanics. Actually, my first introduction to quantum mechanics is rather amusing in retrospect. It occurred during my junior year when Bubb or someone told me to go to the physics colloquium because the new Professor of Physics was going to talk on the new atomic theory. The new professor was Arthur Compton, the new theory was the Bohr theory of the circular hydrogen orbits. Compton later told me that he’d brought the first copy of Sommerfeld’s Atombau to this country when he returned from Cambridge. Sommerfeld also visited him in St. Louis at about that time. My reaction to this colloquium was amusing and perhaps characteristic. I went away muttering, “Even I know better than that. I know that accelerated electrons must radiate.” One of the characteristics of Princeton at that time was an intense interest in the whole literature of physics. The Zeitschrift fur Physik was new, and Harry Smyth had a complete set. He ran an informal seminar that met after dinner, in his rooms in the Graduate College. The objective here was not only to follow the literature but to allow people to report on their reading without self-consciousness and without fear of making fools of themselves; and also to encourage the reading of the German literature, which was dominant in physics in those days. Charles Zahn called this the “Princeton’sche Physikalische Eselschaft.” Henry Norris Russell was disentangling the spectrum of iron during this period, developing the details of the Russell-Saunders multiplet theory. His was another seminar which everyone including Karl Compton attended. It also met after dinner and Russell would report on his results of the previous week with an enthusiasm and energy that none of us could quite equal, so we all, including Karl Compton, at some time in the course of the evening took a nap. The climax of this series of seminars occurred one evening towards 11 o’clock when Russell having spoken for three hours sat down in a chair but began a sentence which became more and more involved. Suddenly his head dropped on his chest, and he was obviously asleep. We sat silent and embarrassed. Then his head was thrown back (a very characteristic gesture of his) he went to the beginning of the sentence, completed it with complete clarity, and we got home about 1:30 the next morning.
Was it in this other seminar that de Broglie’s papers were first discussed?
Someone called my attention to them, and I read them in the Palmer Library. It must have been in the spring of 1924 because I spent the summer of 1924 in St. Louis and there wrote a paper on a wave explanation of the Compton effect which was discussed with Frank Bubb and Jauncey. Frank wrote a note at the end of it. This paper was consciously connected with de Broglie’s publication. It was an endeavor to construct a formal ‘geometric optics’ theory to carry de Broglie’s ideas further. My enthusiasm for this was so great that I listed this as the project for a National Research Fellowship in 1925. When this was discussed with Epstein at Cal Tech, he said, “That’s fine. How do you propose to begin?” I had no idea how to begin, and that was that, for the moment. I selected Cal Tech — you asked why — precisely because Epstein was there. Immediately after arrival, I proposed to talk about de Broglie’s paper to the colloquium, but received no encouragement about this.
Had they known of it before at Cal Tech?
Epstein had certainly read it and seemed to have no very high opinion of it. By the way, one reason why de Broglie’s work was very easily understood and assimilated was that Bohr, Kramers and Slater had earlier introduced their idea of virtual radiation. The influence of this series of ideas is quite evident in my ‘24 paper, when I speak of the need of a new electromagnetic theory. This B-K-S paper also strongly influenced Heisenberg.
How did you feel about the Bohr-Kramers-Slater theory in view of the fact that Compton’s later measurements are supposed to have been of some importance in upsetting it.
By that time, by 1925, everyone including its authors had given it up. But it was still (and very correctly) felt that there was something there. After all, Bose’s derivation of the radiation formula as interpreted by Einstein gave a very sound basis for this virtual radiation theory. On the other band, there were other implications that were untenable. By 1925 however it was clear that something more radical was needed, and this is what was supplied, first in the form of physical ideas in Heisenberg’s first paper. And then in a complete mathematical form in Heisenberg-Born-Jordan. I was not greatly attracted to this, but in the late winter of 1925 or early 1926 Born came to Pasadena, and his lucid lectures aroused my interest. He had been at M.I.T., where he had delivered some lectures both on crystal structure and on quantum theory. They’ve been published. He had conferred with Norbert Wiener on operator calculus, and the interpretation of the commutator law for p and q as identical with the differentiation operator was very strongly emphasized in his lectures. The result was that I spent the spring of 1926 working rather intensively with this operator formulation and was completely familiar with what is now known as Schrodinger operator (the energy operator) before Schrodinger’s papers appeared in Pasadena. But it never once occurred to me, for reasons very obscure, which I can’t begin to recapture, to associate this with my earlier interest in the de Broglie ideas. Just why this block occurred I don’t know, but it certainly was one that Born shared. It perhaps does indicate the weakness of the whale operator idea as it was originated by Heaviside. The operator calculus certainly cannot be completely divorced from the things the operators act on, and this is characteristic of Heaviside and accounts for his difficulties with the pure mathematicians.
When did Lanczos’ paper come to your attention, if you recall?
That must have been very shortly after the Schrodinger paper. Lanczos essentially converted the Schrodinger equation into an integral equation which is the one that is now used in the famous Born approximation solution of the Schrodinger equation. At that time this wasn’t so clear, and the paper I wrote in the National Academy Proceedings has a somewhat misleading title which I wouldn’t choose today. That was rather amusing, because the morning that Schrodinger’s paper came to my attention, I immediately understood the whole thing, the connection with de Broglie well, of course this was in Schrodinger’s paper — but more importantly, the connection with Heisenberg-Born-Jordan. The paper contained the solution to the hydrogen atom. It did not contain the simple oscillator, which came in the second or third paper of Schrodinger. So, having gone over the solution to the simple oscillator problem in the Heisenberg version many times, I immediately tried to do it in the Schrodinger version and was blocked by the Hermite equation, which is deceptively simple. It’s simple enough after you have learned about differential equations, which I hadn’t at that time. In any case, it happened to be the colloquium day, and at the tea preceding the colloquium I found Fritz Zwicky and wrote the equation on the board, asking him how one could solve it. He was equally at a loss, but Paul Epstein came up behind us and rather characteristically remarked, “What have you got there?” I explained, whereupon he turned without a word, went into his office, came back and handed me a copy of his dissertation, told me that he wanted it back, it was his only copy. I opened it and found the whole theory of the Hermite polynomials neatly worked out and quite promptly applied it to the simple oscillator problem. About this time, I was urgently asked to give a talk on the de Broglie theory.
Then you immediately got the same eigen values? Had the identity of these two schemes occurred to you before this?
Not completely. The fact that the two methods gave the same energy operator and eigen values was of course suggestive, and I knew there must be a close connection. And then having been so intent on getting the matrix elements by the Heisenberg method, I naturally looked for the way of calculating them with the Schrodinger method. And I’m sure that the discussion of orthogonal functions in Epstein’s paper was very important to me in arriving at that.
It wasn’t the use of undetermined orthogonal functions in Lanczo’s paper that suggested it?
No. Well, it may … I must be careful. It is quite possible that Lanczos’ paper also gave me some ideas on that, but I really didn’t understand Lanczos’ paper. I did understand Epstein’s and relied more heavily on Epstein’s. Then the question of course was the matrices for the hydrogen atom. And this required properties of spherical harmonics which were then not very widely known. Epstein referred me to a paper on the tides by Hough — in which all these relations were worked out. I’m not sure that they were worked out there for the first time, but they were certainly collected there, and I learnt then from this paper. So that I used these (much the same way that Schrodinger did) for the second paper that I wrote in the Physical Review. The first paper, on Operator Calculus, was not submitted until after Schrodinger’s paper had been published, but before Schrodinger’s paper arrived in the mail at Cal Tech. People often ask about this. Naturally I was disappointed. On the other hand, admiring Schrodinger’s work, it was certainly nothing to argue about, and I haven’t suffered for not making loud complaints.
It seems to me that there is considerable difference between those two papers, the one in the Physical Review and the one preceding it in the NAS, but they are dated so closely together.
Yes, there surely is. This merely indicates the speed with which things were moving, and certainly the first one in the Proceedings was written in extreme haste, in an effort to obtain priority. The one in the Physical Review was not delayed, but it was written more carefully and a number of things were thought out that had been unclear in the first one.
But there’s no question that you already knew by the time you wrote the NAS one, of the formalism in which you can use the differential operators? That had already come with Born?
Yes, the matrices are there too in the calculation. So that essentially I knew the identity of the two isomorphism is the correct term of the two schemes, but didn’t quite succeed in getting this reduced to the simplest English sentences.
And what was the delay in the receipt of the Annalen?
I would have to check the dates — it influenced me to come back east because I felt that if this delay in the mails could have that much effect once, it might have it again. It was obviously before the days of aircraft and one had to add to the transatlantic steamship time the transcontinental train time and all the delays in transferring so that the mail between Europe and California was certainly not instantaneous.
Well, it must have taken you almost no time at all then once you saw the first Schrodinger article.
Well, the connection to the oscillator problem was made the same day that I found Schrodinger’s paper. It may have been in the library a day or so before I found it, and the actual solution of the differential equation and the calculation of the matrix elements followed the next day. And certainly within a week I had written the Proceedings paper and Epstein had agreed to send it in. In fact, he advised me to send it there, it was more rapid publication. The Physical Review paper took me somewhat longer though it couldn’t have been more than a week or so. I certainly was working at high speed, and everything had fallen into place so that it was merely a question of getting it on paper. And not a question of getting the main ideas.
Did Epstein immediately begin on the Stark effect again?
I don’t know. Epstein has always worked very slowly and methodically. He never allowed himself to be stampeded into a race. The next spring was rather active at Cal Tech. In the spring of ‘27 Schrodinger and Lorentz were both at Cal Tech at the same time, and at the end Michelson was also, and Dayton Miller so that it was quite an assemblage. Schrodinger and Lorentz lectured in successive hours. Lorentz first on spinning electrons, particularly the Thomas factor — there must be mimeographed notes on these lectures; and Schrodinger sat in the audience. Then the following hour Schrodinger lectured and Lorentz sat in the audience. Lorentz was within six or eight months of death at the time, but we didn’t know it, and he certainly didn’t show it. He was as mentally alert as you could expect anyone to be. He was very much interested in trying to get rid of the square root of minus one in the quantum theory, and I took this as essentially an assignment and worked it out. It can be done though when you do it the equations become non-linear, and they probably could be extended to the n-particle problem, but then it goes out of real space into configuration space. But the effort to eliminate the square root of minus one and to bring everything into three-dimensional space which finally ended in the Hartree method, was something that everybody tried to do. That was soon discouraged; the first man who really came out and said this was a waste of time was John von Neumann, a year or so later.
The atmosphere at Cal Tech must have been rather different from that at Princeton and certainly different from Washington University.
It was much more international than at Princeton. There was a constant flow of visitors. Raman had been there the year before and Einstein and Ehrenfest. Zwicky was there permanently. And Baade was already a visitor at Mt. Wilson. The elder Lauritsen was on the faculty. Bateman was still there and Bell and Richard Tolman. W. V. Houston, Linus Pauling were there, and they and I were among the first Guggenheim Fellows.
The entire period of your graduate and undergraduate training was that in which American physics first began to show some signs of life. Was there a feeling, a recognition that something new was developing?
Certainly my ambition was to study theoretical physics. With the exception of E. P. Adams and Epstein, there was no one in the country who had published on quantum theory. That was ‘23-‘24, thereabouts, or even ‘25. I spent the year ‘27-‘28 in Germany, and Houston was also in Munich. The elder Richtmeyer came and Bethe was there; Heisenberg visited, and so on. Then I went to Berlin where there were more Americans. I met von Neumann, Szilard, the low temperature Simon, Frank Hoyt, and Manuel Vallarta. There must have been others. When I went to Leipzig at the end of that time, Debye was there, and Heisenberg and Wentzel. Felix Bloch was just getting his doctor’s degree. Houston was there again, and Hoyt came down. Kenneth Cole, now of the National Institutes of Health, was there with Debye. I met briefly Wigner, Teller, and, heard of Robertson, Oppenheimer, and Morse. They had just come back to America as I was going over. By the time I came back and stopped in Princeton, Condon and Morse were already on the staff, I think as assistant professors. Slater had returned to M.I.T., but of course one had heard of him earlier. Van Vleck had returned to Wisconsin in 1927 so in that year of 1927-28 essentially the whole complexion of American physics departments, particularly on the theoretical side, changed. And that change really took place within about 12 months. I can’t say that I foresaw this but I suspect Frank Bubb did foresee something of this sort as early as 1922. No. [Note added during correction: By “old quantum theory” one should understand “the Bohr-Sommerfeld quantization.” Otherwise, of course, one must cite the theory of the Compton Effect as an American theoretical achievement that had a most profound influence on the development of physics. And certainly the Russell-Saunders coupling was not trivial.) The only original thing produced in America on the old quantum theory was E. P. Adams’ National Research Council report. Just how original one counts that depends on the point of view. From the physical standpoint there were no new ideas. From the mathematical standpoint there is the relation with the integral invariants, which is to this day a mystery. That is: the whole business of why the WKB approximation is so good is still to me — an unexplained riddle. I have only recently understood it in the case of multi-dimensional problems. If you look up Bellman’s papers in the Proceedings of the National Academy on invariant imbedding, you’ll find a discussion of the WKB method and a reference to other work. These really convert the WKB series to a convergent series and clarify a lot of questions. During Cal Tech times I published a paper on the relation between the new quantum theory and the old and Bohr’s correspondence principle. I used what later became known as the WKB method in this paper which Epstein reluctantly put in the Proceedings of the National Academy again. He was unhappy about that paper because he saw the mathematical difficulties. I’m not sure that it’s worth much more effort than has already been expended to clear up the mystery but it’s a case where the relationship seems obvious if you’re not too rigorous. The moment you try to become mathematically rigorous the relationship just evaporates. Or if you find examples you are left wondering why these examples I spent quite a bit of time on that in ‘27 and in Munich. Most others were not interested in the limit h=0 and the correspondence principle. I returned to this problem during a sabbatical in Princeton in 1960 and for a time hoped for a break-through. But the usual mathematical obstacles reappeared.
Was Sommerfeld interested at all?
Sommerfeld was quite interested in it. His big project for the year was to rework the Lorentz theory of electrons using the Fermi statistics. And he got all of us involved in that. He was, of course, a very great teacher. His principal technique was to appear dumber than any of us, and this of course spurred everyone on “to explain to the Herr Geheimrat.” He certainly was not as dumb as he pretended to be, but he had, no inhibitions about appearing dumb. Sometimes it seemed that he went out of his way to misunderstand and thus force you to become clearer. There’s only one time that he really went off the track, and that was just in the matter of the WKB approximation. The discussion in his Appendix on Wave Mechanics is completely wrong. Everybody tried to convince him of this and couldn’t. But in general his obtuseness was assumed, and this was his technique of teaching. Perhaps also why he never received the Nobel Prize.
Did he take this attitude with his pre-doctoral students as well?
I think so, of course that was a very strange sort of a place and not formally organized. Down in the basement was Karl Bechert working on X-rays and an instrument maker building models of crystal lattices, that were sold all over the world. Up above was a reading room and the Herr Geheimrat’s office. The reading room had some books and one assembled there at odd times, before the graduate lecture (das Kolleg) or the seminar which also consisted of lectures by Sommerfeld. Then there was the weekly meeting at the Kaffee Heck. The Kaffee Heck was a tradition of long standing in the Institute; it was located in a building at one corner of the Hofgarten, and on stated afternoons, everyone dropped in for coffee, which was served on the standard little marble tables, while one sat on uncomfortable chairs. Sommerfeld told me that v. Laue first wrote out his X-ray diffraction equations for him on the marble of one of these tables. That must have been about 1912, and these semi-social, semi-scientific gatherings were still going on in 1927 and later. But one dropped in to this reading room and sat around or worked. A couple of times a day, Sommerfeld would come out, grab somebody, and start a discussion with him. And particularly if one was working on a paper with him or a group were working on a paper.
In these discussions did he do most of the talking?
Sommerfeld’s prepared lectures were marvels of leisurely clarity, but in these informal discussions, he never “lectured” one. He asked questions or perhaps contributed facts from his memory. He made it seem that you were providing the ideas — but I often wonder …
I understand this was his technique of long standing, to discuss all things with the students.
Right, but the discussion was fairly general, and he drew the student out. He told about his visit to America in 1921 or ‘22 or whenever it was. He said when he’d left Munich, Pauli was his assistant. He told Pauli that the next fall they would have to write an article on relativity for the Encyclopedia. When he got back and had gotten settled, he reminded Pauli of this, and Pauli said typically, “Yes, I have written it,” and handed it to him. Sommerfeld read it, decided he couldn’t add anything to it, and sent it to the Encyclopedia under Pauli’s name without his own. The publisher didn’t want to publish it and wanted Sommerfeld to put his name on it. Sommerfeld refused, and they had to get Einstein to vouch for Pauli before it was published. This was published when Pauli was 21, and for years afterwards whenever Pauli’s name was mentioned, people would say, “Oh yes, he’s a very young man, isn’t he? He’s only 21.”
Much of the work in building up American physics was done by these various fellowships, I gather. I was wondering how they worked. Did you make arrangement beforehand, say with Sommerfeld, or did you just go?
The National Research Council and the Guggenheim Foundation both required that you have a letter from the place where you wanted to go saying that you would be welcome. And so I wrote to Epstein before going to Cal Tech and got the reply and sent a copy to the National Research Council. The trip to Europe was fairly well prepared. My original itinerary was to go to Zurich with Schrodinger, then to Munich with Sommerfeld. By the time I actually started, Schrodinger had moved to Berlin and wrote suggesting that it would be better if I first went to Sommerfeld and gave him a chance to get settled in Berlin. Then the decision to go to Leipzig also was an afterthought. There was a birthday party for Heisenberg at Sommerfeld’s home, and it may have been arranged on that occasion. In Berlin I was rather disappointed and even a little resentful of the situation. John von Neumann was lecturing on quantum theory and its statistical interpretation, and Schrodinger came to the seminars. But Schrodinger himself was not lecturing (rather he was lecturing on elasticity), and he did not argue with von Neumann. In fact, von Neumann was so abstract that at the time none of us really understood what he was talking about. It wasn’t until one got his book, which was then being written, that one began to see what his ideas were and how they related to the whole problem. But Schrodinger — well Szilard, who was Einstein’s assistant at the time — Szilard expressed it by saying that unfortunately Schrodinger is doing too much reading and not writing anything. Schrodinger also hoped to bring everything back to three dimensional space but couldn’t see how to do it and was rather stopped by this. One must read Schrodinger’s 1926 papers after reading his 1919-20 series of papers in the Physikalische Zeitschrift. Then one fully understands the independence of the man and how he was able to pick up de Broglie’s ideas and work them into the wave mechanical form that he gave them. But in 1928 he was passive and relatively uncreative. He did publish some details.
Were there any difficulties in Germany for foreign students at the time you were there?
There were no formal difficulties except at the University of Heidelberg where Lenard held forth and was already essentially a Nazi, although that name was not used much in those days. Elsewhere on the whole, one was received with considerable friendship. Every once in a while of course, one encountered someone who had been badly hurt in the war, either emotionally or physically and was repelled by them, although even that didn’t follow. There was one man who had lost an arm in the war but was perfectly outgoing towards Americans and had no lasting grudge. It was a period when I at least thought that the future was very clear, and the internationalism of science was going to bring about an era of peace. Szilard was probably the most realistic of all of us in 1928. My disillusionment because of the Nazi development is probably why I’m somewhat doubtful about such enterprises as the IGY — not doubtful about the enterprise itself but doubtful that it will contribute on a broader political front. Certainly, most of the people with whom I became friendly in Germany at that time I’m still friendly with. Most of them are now American citizens: Wigner, Teller, and a good many others.
Shall we go briefly to Chicago? You did translate, didn’t you, or help with, the Heisenberg lectures?
This is true. Heisenberg came there in 1929 and gave a summer lecture series which he wrote down as he gave them and which became the basis of the German edition. Frank Hoyt and I, prepared the English edition, essentially from our own notes of his lectures. We saw his notes and corrected our manuscript, but he took his own manuscript with him at the end of the summer before either the German or the English text was complete. And the English version was printed without our seeing the German version. The result is that there are quite a few differences between them. This is greatest in the part that Frank wrote, which we felt was essential for the American reader, while Heisenberg did not feel it was essential for the European reader. So that Frank’s contribution while it was in Heisenberg’s lectures, was not in Heisenberg’s manuscript in as great detail as in the English edition. That’s the second part of the English edition. I’ve never compared the English and German editions word by word, but there are quite considerable differences which came about because of this rather hasty way in which the job was done.
Do you recall anything about the process of acceptance of the uncertainty principle which came out when you were in Europe?
Well, it was already out at the time of Sommerfeld’s party for Heisenberg. This was a rather hilarious occasion. A gamma ray microscope was brought into the room in the form of a barrel with the ends knocked out. Peierls took the part of the photon, Bethe the part of the electron. And after colliding with the electron and going through the gamma ray microscope, Peierls presented Heisenberg with a Wellenpaket which consisted of a long scroll with waves drawn on it. There is a sharp difference between the statistical interpretation of quantum mechanics and the uncertainty principle in the narrower sense. The former now seems a straight forward generalization of the Bohr-Kramers-Slater ideas and inevitable after de Broglie and the Davisson-Germer experiment. Perhaps this is clearer now than when v. Neumann and the others first began to talk about it; the uncertainty principle with its paraphenalia of gamma ray microscopes was no help.
There were certain difficulties however in the acceptance of the uncertainty principle.
There are two things which are clearest in my own mind. I was very reluctant to accept it, and I still am reluctant today to give it as much weight as many people do. The force of the uncertainty principle is entirely a verbal one; it comes about because we use the particle theory, the discrete theory, and also the wave theory. Verbally neither of these is adequate. And the uncertainty principle merely expresses the inadequacy of both, and uses one to criticize the other. The uncertainty principle has no counterpart in actuality or reality, and if you are willing to take quantum mechanics as it is formulated in Dirac, for example, you don’t need the uncertainty relation. In fact, the uncertainty relation plays a very small part in Dirac’s book, none in the mathematics and little in the text. The statistical interpretation is the basic thing.
The process of the solidifying of the Copenhagen interpretation is what interests me in this respect, since there is now in particular beginning to be much opinion expressed against it, and this must also have occurred in the late ‘20’s.
Well, there are two things there. One is that Arthur Compton, for a time, made a great point that the uncertainty relation gave scope for free will in the world. This, Bohr and Heisenberg did not agree with, though I perhaps oughtn’t speak for them. This is one of the uses, one of the applications of the uncertainty principle that was quite unwarranted. This sort of misuse still occurs. For example, it has been done in a recent book on bio-physics. In this book it is used as an excuse for a crude sort of vitalism. The other thing is that the Copenhagen view is not a unified view. If you read Bohr, particularly if you have the privilege of speaking with him, you realize that in the course of the years, his attitude has changed greatly. He is still as interested in it and as willing to talk about it as always, but what he says today comes closer to what I can accept than did his original statements. In his original statements it was not clear that he thought of the uncertainty principle as a linguistic matter. The last time I was privileged to speak with him, which was a couple of years ago here in La Jolla, it was very clear that he was thinking of this as a linguistic matter. In fact he remarked that words such as life and death and love, which are exceedingly useful in ordinary conversation and in ordinary life, need not necessarily be explicable in terms of scientific concepts. That life, the distinction between living and dead matter may not be a physical or chemical difference, that the transition from the one to the other may not be a razor sharp transition, but a gradual one. Nonetheless, in everyday life the distinction between dead matter like a piece of tin and living matter like a live dog or a live human being, this is a very important distinction. But this is at the extremes, and if you follow back from these two starting points, it’s not sure that distinction will ultimately be a head-on collision. It may very well be that a continuous amorphous region separates them. This certainly is indicated by the viruses and things of this sort. Again, I’m perhaps giving my own views in regard to the matter and not quoting from Bohr; all that I want is to indicate my indebtedness to him, not to embarrass him.
But in the process of the original acceptance of this idea it was the lead given by the Heisenberg talk at the Como Conference that convinced many?
Yes. It really became an issue with me only in the course of that summer when Heisenberg was in Chicago. Previously I had hoped that it would go away, but at that stage I knew it was something that was here to stay, and that I would have to make my peace with it. It ultimately drove me into an interest in symbolic logic in trying to analyze the situation. But that’s quite a different story.
Would you say that your feeling was fairly representative? That is, people felt it was rather on the borderline of physics and would perhaps go away and that good physicists needn’t think about it?
Well, perhaps not — not representative. Heisenberg and I had long discussions about the way to translate anschaulich. The dictionary would translate it as “intuitive”, and we discussed this. Heisenberg knew enough English and I knew enough German so that this was not a formal discussion. Heisenberg leaned closer to the interpretation of “intuitive” but he was not fully satisfied with it because of the implication of congenital knowledge. At one time, we thought of “visualisable,” but this was rejected. We consulted with classicists and philosophers, trying to find a better translation than “intuitive”. It just occurs to me that I can’t remember now what word was used in the English edition. I think we avoided it as much as possible. Now there were many people who felt this uncertainty principle was a deep philosophical truth. This was certainly true of Compton, who took a mystical view of it. One sometimes got that same impression from Bohr, though he strongly disclaimed any intent of mysticism. There was a period when it was extremely unpopular to be too skeptical of the uncertainty principle. It was really accepted rather unquestioningly by a lot of people. In particular, the fact that these experiments were imaginary, were Gedanken-experiments in Bohr’s terminology, was not sufficiently recognized, and sometimes isn’t today. These experiments are not actual experiments, and the question of how much empiricism there is in appealing to an imaginary experiment for a decision or for the support of a theory is an important one. Bohr and Heisenberg were not the first to discuss imaginary experiments. Einstein appealed to them in his formulation of relativity theory, and Galileo and Maxwell. And there is no doubt that these imaginary experiments play a very important psychological role in the work of the theoretical physicist or of the theorist in general. And they may lead to correct conclusions. In fact if they don’t, they are usually discarded and one doesn’t hear of them. Kekule dreamt of eels catching their tails and woke up with the idea of the benzene ring. But I don’t believe that the citation of an imaginary experiment carries the same weight as the citation of an actual experiment in supporting a theory. Bridgman, paradoxically, is never clear on this point, in his “operationalism.” And this is somewhat subtle point, particularly if your only knowledge of the actual experiment is hearsay. Many non-physicists were and are unable to distinguish clearly between the actual experimental evidence and the imaginary evidence that is all that imaginary experiments can provide, but it’s a very important distinction, and one that in the early days tended to be overlooked — both in relativity and quantum theory. To make this matter even more confusing, the description of these imaginary experiments stresses the participation of an imaginary, superhuman but submicroscopic observer and his knowledge of events. This has led to their interpretation as a part of epistemology as giving ‘empirical’ evidence on which epistemology can be based. I think I have said enough so that I needn’t be vigorous in rejecting this view; as I said, Bridgman did much to add to this confusion. The one valid item in this context is Szilard’s great paper on the entropy of decision and his analysis of Maxwell’s demon (another animal in this imaginary zoo). This anticipated by 20 years modern information theory. Finally, when one analyzes the theories that these imaginary experiments support or supplant, one always finds tacit logical assumptions in the latter that are replaced by different but explicit logical assumptions in the former. This is clearest in the case of the relativistic theory of simultaneity, less obvious in the quantum theory. Moreover, these replaced tacit assumptions are of great antiquity in human thought and are not inconsistent with the experience of earlier generations nor even with the experience of most contemporaries. They are only inconsistent with a careful analysis of experiments made possible by an advanced technology. Being of such antiquity, there is always an emotional resistance to giving up these tacit assumptions. The imaginary experiments can, perhaps, help one to understand this emotion and overcome it; they cannot replace, scientifically, the logical analysis of the actual new experiments of the modern scientist. This emotional or non-logical resistance arises in an interesting way. Being of great antiquity, the tacit assumptions that are under fire have influenced the vocabulary and sometimes even the grammar of our native language. This makes it very difficult to talk about them non-mathematically or to eliminate them without using mathematical tools. The imaginary experiments can help even the mathematician. It occurs to me that the very title of Bohr’s book “Atomic Theory and the Description of Nature” gave me the first clue to these linguistic aspects of theoretical physics. Later, I was privileged to discuss this with R. Carnap. He did not understand Bohr or quantum theory, but his clear lucid exposition of related matters has strongly influenced me. I have stressed the antiquity of the common everyday description of nature. It is not irrelevant to remark that nature is even more antique and functioned long before men began to chatter and philosophize. This ought to make one incurably skeptical of both scientific theory and religious dogma.
By when would you say that uncertainty principle in its usual interpretation was generally accepted?
Well, there were of course people who never did. There are people who never accepted relativity theory also, and so on. Strangely, Einstein, who was the first to recognize Bose and de Broglie, never could reconcile himself to the statistical interpretation. But by the early ‘30’s there were many popular accounts of the uncertainty principle, and probably it was the general public or the non-technical public that became confused about the imaginary nature of the experiments, rather than the technical public. But even some of the technically competent people would get into an argument about the so-called “philosophical significance” (i.e., super-scientific) of the imaginary experiments. Today there is much less stress laid on it than there was at the time.
Yes. Well it must have been rather difficult to maintain a contrary opinion if one looks at de Broglie’s work, which as I understand it is originally an attempt to avoid, to provide some physical description. He then abandons this for lack of help or support. And now just recently has gone back.
I have not followed the latest developments in this argument of which there are many. I’ve my own peace with it and don’t particularly care what other people say anymore. Bohr once remarked, “I’ll be interested to see what Mr. So-and-so can contribute to this subject beside confusion.” I’m not.