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 Harold Urey by John L. Heilbron on 1964 March 24,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
For multiple citations, "AIP" is the preferred abbreviation for the location.
This interview was conducted as part of the Archives for the History of Quantum Physics project, which includes tapes and transcripts of oral history interviews conducted with ca. 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: Bichowsky, Raymond Thayer Birge, Walker Bleakney, Niels Henrik David Bohr, Ferdinand Graft Brickwedde, Compton, Albert Einstein, Samuel Abraham Goudsmit, Werner Heisenberg, Richard Jesse, Edwin Crawford Kemble, Hendrik Anthony Kramers, Ralph de Laer Kronig, Gilbert Newton Lewis, Hendrik Antoon Lorentz, Robert Sanderson Mulliken, J. Robert Oppenheimer, Linus Pauling, Arthur Edward Ruark, John Clarke Slater, Arnold Sommerfeld, George Eugene Uhlenbeck, Robert Williams Wood; Columbia University, Johns Hopkins University, Kobenhavns Universitet, University of California at Berkeley, and University of Montana.
Well, why don’t I talk about this from the beginning. You’re asking, for example, about my interest in science. Well, I grew up on a farm in Indiana and was bored with science in high school. I took a course in high school physics. I remember bits of it yet, but I must say I was anything but inspired. I took a course in zoology also and liked this much better, and had no course in chemistry whatever. I taught grade schools one year in Indiana and two years in Montana. Where I taught in Montana there was a young man (Bryan Wilson), and I stayed with his parents. He went off to college at Bozeman, Montana which gave me the idea that maybe I’d better do the same. So I went to the University of Montana. At the University of Montana I was going, I thought, to do psychology, but they didn’t take freshman students in psychology, so I registered for freshman chemistry and freshman zoology. They lost a psychologist right there. In zoology I had quite a remarkable teacher, A.W. (Bray). This man was a graduate of Cambridge, England, and he took a very special interest in me. Not until I went to Oxford in 1956-57 as Eastman professor did I realize that I got sort of an informal English education, and. A.W. (Bray) was my tutor as it were. This proved to be of enormous value to me. Here’s a country boy arriving at college, and he finds that the great professor takes a great deal of interest in him. This fed my ego and my self confidence enormously and, I think, contributed a great deal to my success all through my life. In three years I had my credits for my degree in biology, but I had almost as much chemistry.
You say you wanted to start as a psychologist to begin with — did you have any intention of making that your ultimate career?
That’s what I was going to do when I went to college.
What kind of opportunities were there?
I don’t know. I don’t know why I thought of it. Well then World War I came along in 1914 and I went east as a chemist at the Barrett Chemical Company during the war, and so dropped my zoology and became more seriously interested in chemistry. I went back to the University of Montana as an instructor in chemistry for two years, and then decided I must have a doctor’s degree and went to the University of California. Here of course the men who were most influential in my life there were Gilbert N. Lewis, whom I regard as one of the most remarkable teachers that I’ve ever known. He had the capacity to make everybody about the place study exceedingly important and fundamental things. He had no use really for grades at all, didn’t consider this important. In fact I’ve never learned what my grades were at the University of California as a graduate student. I had to go down to the registrar’s office to get them and I never went, so to this day I don’t know what my grades were. Another man of great importance to me was R.T. Birge who is still living. He’s the one who interested me a great deal in physics by (running) seven hours [of courses] and talking to me about the Bohr atom at that time. I first ran across the Bohr atom as an undergraduate student at Montana. I was an undergraduate student in 1913 when it came out, and my professor of chemistry, Richard Jesse, called my attention to this. I tried to read it and of course didn’t appreciate it in the slightest, but I realized it was there, and also Irving Langmuir’s extension of G.N. Lewis’s cubic atom came out in the Journal of the American Chemical Society, and as an instructor of chemistry, not as a student, I had gone over this and built an enormous model of an atom on Langmuir’s ideas at that time. So when I got to the University of California, I was much interested in the Bohr atom. I was more interested in it than I was in Lewis’s cubic atom.
Did you see them as conflicting completely?
Somewhat. No, they were two separate compartments in one’s thinking. There wasn’t any doubt but that there was a tendency on the part of atoms in the lower part of the periodic system to complete octets, that is shells of eight about the atom. On the other hand, Bohr’s theory of the hydrogen atom and Sommerfeld’s extension of it to elliptical orbits explained the spectrum of hydrogen in such detail that there simply had to be some truth in this theory. While I was there, I read the first edition of Sommerfeld’s Atombau in the original German — I wonder, do I still have it? [Urey looks for it] 1922 — this is the book that I read.
I think that’s the third edition.
Third edition, yes. I acquired this book and read it clear through. This is where I learned about atoms really, from this book.
Were other people reading it at Cal?
A few of us were reading this.
Let me return for just a moment to the contrast between the Lewis model and the Bohr atom. Was the dynamical and statistical difference considered important?
Well, you see, I just personally couldn’t believe that the electron stood still on the corners of a cube. This I didn’t believe could be true, but there wasn’t any doubt that certain chemical facts were explained by this which the Bohr atom of course couldn’t presume to explain since it dealt with a single atom of hydrogen. I was prejudiced very much in favor of the Bohr atom throughout all this time.
Did you know of and discuss the models that were invented by Lande and Born which did have cubical symmetry in which this Ellipsenverein of Sommerfeld’s was extended so instead of having all the electrons in the atom move in a plane at the corners of a polygon they moved in three dimensions at the corner from a pulsating polyhedron?
I don’t remember ever having paid attention to these at all, no.
What did the professor of chemistry at Montana have to say about the Bohr atom, do you recall? How did he happen to come across it?
I don’t know. Professor Jesse was an analytical chemist. He did his doctor’s work, I think, at the University of Illinois, and he worked on the atomic weight of chromium. For many years his atomic weight of chromium was the best value. I don’t know what the situation is at the present time in regard to it. He was an excellent teacher, sticking to the regular, old-fashioned descriptive chemistry with not too much attention to physical chemistry. He was an analytical chemist essentially. And yet he was noticing things of this kind, and he called my attention to it. That’s how I happened to run across Bohr’s theory of the atom at that time. I didn’t understand it at all, not a bit of understanding whatever at that stage.
Would the opinion you just gave of the relation between the Lewis and Bohr atoms, that is, the one obviously corresponding to physical facts, the other obviously corresponding to some chemical facts, would that reflect Lewis’s opinion at the time too?
No. Lewis didn’t wish to accept the Bohr atom. And approximately in the middle of my stay there Lewis announced one day in seminar that he believed the Bohr theory was correct.
What prompted his conversion?
Well, I think that the evidence was just overwhelming that something like the Bohr atom for hydrogen had to be correct. Birge the next day made a comment in class: “Yes, after all of us have worked on this and studied it, and considered it, then Professor Lewis, who has not made any such contributions at all, suddenly announces that he agrees with it, and expects that now that settles the question.” He was much annoyed about it. You see, Birge and Lewis were not very cordial friends. I think that is correct. They got along politely enough, but they were not very cordial. Birge has always been such an outspoken, honest, completely transparent person you know. Of course other men at the University of California had a great deal of influence on me; Joel Hildebrand was working on solutions. But I was interested in the physics side, rather than the solutions side, and William Bray — not the same Bray I mentioned before — was strictly interested in inorganic things, but there was (Branch) and others, both in physics and chemistry, that I was acquainted with. But these two personalities, Lewis and — well I would say Lewis and Hildebrand and Birge played a very important part in my education.
Was this tension between Birge and Lewis reflected in the relations between the physics and the chemistry departments at all, or was it just a private matter?
No, no. I think it was more or less an annoyance at each other, and rather an outspoken characteristic of both of them. Both of them were of course exceedingly enthusiastic people about science you see, very intense about it.
And Birge did discuss in great detail the Bohr theory?
Oh seminars and everything of the sort were going on all the time.
What were his main interests at the time?
He was interested in molecular spectra at the time, and following that went over to his interest in the values of the fundamental constants.
Do you recall what was said about the more fundamental problems of the quantum theory in Birge’s seminars; did they come up at all?
No; rather puzzles. You see at that time there were enormous puzzles about the quantum theory. The old Bohr orbits obviously were not right. Let’s see, attempts to work out the theory of the helium atom were being made. They didn’t work out very well. I think that at that time there was a paper that appeared showing that contrary to what you have in a gravitational system of forces, that is where everything attracts everything else, the electrons in — for example, going in an orbit about the nucleus, say at the ends of a diameter, would not be a stable configuration. The one would move off to a large orbit, and the other would fall to a small orbit; in fact it would produce a self ionization, if your classical mechanics occurred. This was being discussed at the time. You see, I graduated there in ‘23. Now, when did the quantum mechanics coma out? I think ‘26.
Well, Heisenberg in ‘25, and then Schrodinger in ‘26.
‘25 and ‘26. And I was at Johns Hopkins University at that time. My California days were — and my year at Copenhagen was much the same situation. Everybody realized that the Bohr theory just wasn’t the correct approach. Bohr himself realized that very much. He had tried, as he told me in later years, to make models of all sorts of things, and the models just would not work out.
Was it mainly of the helium problem that he was trying to make models?
That I do not know, what the models were. And while I was there, you see, the correspondence principle was developed by Sister and Kramers; Bohr of course was the moving spirit in this thing.
Slater came to Copenhagen with some of those ideas.
Some of those ideas, yes. He arrived — but just what the details of what he came with were, I don’t know. I never understood very well.
You were there at that time?
Oh yes. When I was in Copenhagen, Slater was there. (Cheddy) — do you remember (Cheddy)? Cheddy was a mineralogist really, and he went to Los Alamos, and was an important figure at Los Alamos for many years; he died there a few years ago. Heisenberg visited us, Pauli visited us, Rosseland of Oslo was there at that time. Who else? Oh, Hevesy was at the laboratory. We had quite an interesting group as you will see! Let’s see, people to get the Nobel prize in that place were Hevesy, Pauli, Heisenberg, myself, four of us, I think, that were there at one time. I think somebody else, but I can’t think now who it could be.
How did you decide to go to Copenhagen in the first place, and make your arrangements?
I don’t know. After getting my degree there — I think [A.R.] Olson, in the department of chemistry, was a Scandinavian, and he knew about this fellowship, and called my attention to it. So I applied for it, got this fellowship; that paid a thousand dollars by the way. I had to borrow a little money from my brother, but otherwise it took care of me for a year.
That was to include the transportation as well?
That had to include the transportation as well, and going, I went on a commercial boat. It took three weeks across the Atlantic. That was my first trip across the Atlantic.
But in making arrangements to go to the Institute, you corresponded with Bohr to be attached to the Copenhagen Institute?
After you had obtained the fellowship?
Yes. Well it was (???) that. Bohr always told me that he didn’t know I was a chemist, he thought I was a physicist. If he had known I was a chemist, he probably would not have let me in.
Oh, was he —?
Well, I mean, it was a physics institute and so forth. But he never knew I was a chemist.
Did he decide you were a physicist on the —
No, a very badly trained physicist, I think that’s what he thought.
On the strength of your hydrogen atom orbits I suppose?
Well, I think that coming from Lewis’s laboratory was the important thing.
And you had no research proposals at the time?
No. Kramers made a suggestion that I work on this problem of crossed electric and magnetic fields, and he helped me a great deal with it.
Do you recall what the status of that problem was when you took it up?
I don’t recall. If the paper says anything about it, maybe that explains it, but I don’t recall anymore about it.
How was your reaction to Kramers? Was he a pleasant person?
A wonderful person. You see, Kramers and I had lunch together, partly throughout the whole year. He was an exceedingly quick and intelligent person — spoke English beautifully, taught me a great deal about writing articles in English. He had an English vocabulary that I had difficulty in equaling.
That explains something that had puzzled me. I was glancing at that article [paper No. 3] again last night, and was struck by how much it sounds like Copenhagen English. It was quite extraordinary.
Yes, yes, that’s right. That paper was written so many times that the secretary just hated to see the sight of ma, because Kramers worked it over so. But I learned a great deal out of that. I didn’t see much of Bohr. Bohr was away a good part of the year, came to the United States, and came back. He was terribly busy with things; it was at the time that he was writing that big paper upon the general structure of atoms and the periodic system of the elements, and so forth. It was the one in which he divided the shells of eight into four and four, instead of two and six.
I think that came out just a bit earlier, while you were all at Berkeley — perhaps in ‘23? I believe that was the one in the Annalen der Physik, but I may very well be wrong.
[searching] — I’m afraid I can’t find out any reference to that.
I was curious in that connection whether the discovery of hafnium had any great effect at Berkeley. Was that impressive or not?
We were very interested in it. When I went to Copenhagen, Hevesy was very busy separating this out. Hevesy and I became lifelong friends on that trip. We have been awfully good friends ever since. I never go near a place where Hevesy is, without looking him up. You see his daughter is married to (Gustav) Arrhenius, who is professor here at this university. (Gustav) is the grandson of Svante Arrhenius. I also met (Gustav’s) father, a very fine man. As I say, Hevesy was working on the separation of zirconium and hafnium at that time.
Kramers was extremely approachable then; did he give of his time to everyone in the same way?
Well he was exceedingly — well to everyone perhaps not. You would hardly expect him to give his time to Hevesy — it was a completely different group. He worked a great deal with Slater. I don’t know how Slater’s and Kramer’s relationship was. I thought they were a little bit rough at times.
Yes, we have talked with Slater to try to unravel the order of those ideas, and there did seem to be a bit of trouble there.
I think Slater came — at least he felt he came — with a considerable amount of ideas about the whole subject on which they published. I had the impression that he felt that they did not give him as much credit for it as they should have. Of course on the other hand, it may be that very similar ideas existed in Copenhagen, and that Slater didn’t know it.
The idea, Slater says, was changed somewhat, or really almost fundamentally, from his original proposal, and then it was shown that the new proposal wouldn’t work, through the recoil experiments. Then Slater wrote another article, which I think appeared in the Physical Review presumably based on his earlier ideas, and I think there might be some cause for his annoyance.
I see. Well I think there was some annoyance there that Slater had.
Did you participate in any of those discussions?
No, I didn’t. I didn’t know much theoretical physics. As a matter of fact one of the things that I really learned at Copenhagen was that I did not have the mathematical equipment and the mathematical ability to be an effective theoretical person. This is the reason I stopped. You will see that there are a few little things afterwards, then there’s nothing in the way of theory of this kind. The reason was simply a conclusion of that kind. That has been true ever since. If the mathematical theory isn’t too complex, I can follow it all right. When the matrix mechanics came out and the wave mechanics came out, I gave seminars on this at a group in Washington, in which M.A. Tuve, Otto Laporte, and people like this were present, and I was able to review these papers of Heisenberg as they came out before a physics seminar, and do pretty well at it. In fact Otto Laporte I saw when I went to Japan a couple of years ago, and he said that he remembered these reports and found that they were very good reports. You see he was very nice and polite about it at least. But though I could read these things, I realized that people that have a real gift of this kind can do something in five minutes that I won’t do in a year. I realized that in theory, unless you’re really good, why the other fellow can beat you out so fast, you see. If you are an experimentalist, and if the other fellow is smarter than you are, it still takes him a certain length of time to get the experiment done, do you see. So I just decided that I had better get busy on non-theoretical problems to a large extent at least. I’ve always been interested in the theory, except that in recent years, the strange particles, all of the ‘strange-particle-physics’ that has come up so much since the war, is just an enormous puzzle to me. I think it is to an enormous group of people, don’t you? And I haven’t tried to keep up with that at all.
To get back to the paper which Kramers helped you with — was Kramers particularly interested in it, or in the outcome?
Well, he had a polite interest in it at least. I don’t think it was a terribly important paper, but the accidental degeneracy — what we call ‘Zufallig’ — came up, and I think it’s rather the first time it came up, isn’t it?
In this form it’s the first time I had ever seen it. I was very much interested in that.
Of course the accidental degeneracy is the sort of thing that has developed in the quantum mechanics to quite an important thing. When you have levels that lie close together, why you have an enormous interaction.
There was a good deal of theory on the subject I think already by the time you wrote this paper, but this particular example, and your suggestion of an experimental application of it interested me a good deal in terms of the model question. You expressed yourself quite clearly that you think it might settle the matter of the reality of the stationary orbits, particularl3 the relativistic precession. Did Kramers share that notion?
I think he did. I don’t think he had any objection to it. I think it was my idea, but I’m not awfully sure about that anymore; it might have been Kramers, and my conversations with him that suggested it. Or it might have developed mutually as a result of conversations. I no longer remember. You know how it is with things like that, you’re talking with somebody every day as I was with Kramers, and at the end you donut quite know where the ideas originated.
In one of these letters from yourself to Kemble [18 Feb. ‘26] there is a remark that Kramers doesn’t think that you get anything out of models, and that it’s only chance that any agreement arises. In that connection I had thought perhaps Kramers would not have attributed to this experiment quite the importance you —
Well you see, what Kramers meant is exactly what Bohr said — the whole group — that for some reason or other the models that they’re making are not right. Of course the quantum mechanics is the only thing that gave the model that was correct. That’s what they meant all the time by this. Bohr realized that models were not correct, that the great detail of the precessing orbits that were so favored by Sommerfeld and so forth, that these were not correct. There had to be something else, but what nobody knew.
But in the German article that was translated by Werner Kuhn, there you talk as if the experiment on the degeneracy will settle the matter, and quite possibly in favor of the reality of these precessing orbits.
I see. Well I was wrong.
But it’s very interesting.
Well you know how a young person is — I run across it very often — we’re starting out and trying very hard to get a little hold in science and do a little something that’s original, and you hold on to all your little beads very carefully. Later, as a result of more experience and perhaps more success, which is important, why you don’t mind admitting that that idea was a bad idea.
Except, is it not true that Bohr and Kramers kept a pretty close watch on what went out of Copenhagen, and that you would not have submitted that article without their approval?
That’s right. I was working on the orbits as a way of trying to indicate what the limit of the periodic system was; it’s Sugiura’s and my article [No. 8]. I was using integral quantum numbers for this. It was Bohr’s suggestion that I try half-integral quantum numbers and Sugiura then did it with half-integral quantum numbers, and because of the work that I had done, he put my name on the paper. This is an illustration of that, because Bohr would not have the paper published with whole quantum numbers at that time. But of course as you see they were equally bad, neither one was right.
Was his argument against your publishing with the integral numbers bad agreement with data? Was there anything more?
Well the general consideration of the whole problem: half-quantum numbers were in the air at that time, without any good logical reason for it. When the spin of the electron came along, with a half-integral quantum number, then there was reason for the half integral j’s — all of this. But up to that time there was no very good reason for half-integral quantum numbers.
Well there were the various arguments from the rotation spectra, the band spectrum — there was the old conundrum as to whether the thing goes 180 degrees, or 360 degrees and so forth.
Well you see you have the half-integral vibration quantum numbers. I think this is the place they found that you could not account for the isotope effect. Wasn’t that the case?
I thought the isotope effect didn’t really come up until after ‘25 —
When did (???) work on the HCl spectrum come out? I think that he worked on the lid spectrum at Michigan didn’t he? I think (???) did. Well, you’ve got me too.
But in any event they certainly have been turning up.
Yes. And the half-integral quantum numbers were needed in band spectra because the isotope effect wouldn’t fit. Now wait a second, let me see. Ruark’s and my book should say something about that. [Pause to look in book] I think it’s not in the HCl band, it comes up in the electronic spectrum that you have to put in the — yes, it was Mulliken I think who was working on that… There’s a reference here to Mulliken in ‘25, but it may be that he was doing some work before that, yes. That’s the case.
Some other developments occurred in that time such as the Stern-Gerlach experiment and the Compton effect. Were these discussed at Berkeley?
Yes, Compton was at Berkeley while I was a student there. He talked about this while he was there. Of course we were immensely interested in it. Now the Stern-Gerlach experiments, when were they done?
‘22. That’s while I was at California. Now what is the Einstein-Ehrenfest paradox that you’re talking about here? [p.2 of the outline].
That’s the argument that since one cannot presume that collisions between the silver atoms are responsible for the various quantum alignments, how, then, does the silver atom, moving like this, jump into and out of its proper quantum alignment in the magnetic field? — an objection they felt was quite important.
I don’t remember it, don’t remember any discussion of it.
And was there much discussion of the photon, do you recall, before Compton’s work, or was that really what made it an issue?
Oh yes. At that time the particle theory of light, as compared with the wave theory of light was discussed an enormous amount because nobody understood how to understand it. Do you think we understand it now? I mean we’ve just gotten used to it, haven’t we? Yes. Might I ask how is the capacity to behave like particles and to behave like waves —. And we accept it, don’t you think this —.
I think so. I think that’s the way things work very often. We have the same difficulty in comprehending gravity acting at a distance.
Yes, we’ve finally found that the problem is not an important one, that nature is this way, and so. Of course at that time the particle theory as compared with the wave theory puzzled everybody enormously. When did the discussion come up — it’s mentioned in our book — about the very short period of time for electrons to appear when you let light drop on a metal. But how long would it take for a light to accumulate enough energy to eject an electron from the metal — of course it would take hours — but it would appear immediately. This was an enormous puzzle to us at the time. You’re taking me back an awfully long time. I haven’t thought about these any re than has anybody else recently. We all come to accept them as finished and understood — at least we become accustomed to them — and no longer discuss them. Now you ask me to go back and think about it and I have to try to remember what went on!
Well that I think was something Lewis was quite interested in.
He was, yes. ‘How could this possibly be true?’ Oh, you ask how I got from paper No. 1 to Paper No. 2. I don’t know. They’re not related.
What was your dissertation?
The first two papers constitute my dissertation.
Do you recall any of the reasons for sending it to the Astrophysical Journal?
No, I don’t. I don’t know. I’ve used that journal recently, have been interested in astronomy recently. I say, “Of course one needs to have a union card really to work in astronomy, don’t you?” “Of course I have a better union card than you do” — I’m telling astronomers – “Now did you publish a paper in the Astrophysical Journal in 1924?” You see? [Laughter]
I was quite puzzled why it turned up there.
Yes, I don’t know why I sent it there. You see at that time I was also thinking of the application of the dissociation of atoms, very much the same thing that this Indian — what was this Indian’s name?
Saha, ye. I was working with the Saha equation at the time that Saha published that. That this should work, was suggested to me by Gilbert Lewis. He said that you ought to expect that atoms would be ionized in this way. So I worked out how much ionization one would expect from a concentration of electrons at various temperatures, pressures, all of that. I had worked it all out. And Saha’s equation appeared in the astronomical journals while I was working on it. That’s how I happened to know about Saha’s equation. As a result of that I also was interested in the thermodynamics of the distribution of the different orbits. Now you will see here – [moves away from microphone] (???). I was supposed to be try in to measure this effect. [showing Heilbron] Well that never worked out, that thesis never worked out. That’s what I was supposed to be working on, and this is what I presented instead of it. I think Lewis let me off there very easy; he should have insisted that I finish the job.
You were at Berkeley only two years?
Two years, yes.
Was that rather quick?
It so quick in those days. Two or three years was considered to be enough in those days, now four years is much more likely to be the case. But the reason I published that is because I was attempting to work on this experimental problem. I don’t know whether anything has ever been clone on it since. We all know that it does ionize, that it is conducting and all of this, but the troubles of the experiment were that it was so difficult to distinguish between the conductivity of cesium vapor, and the emission of electrons from the hot electrode. You see you heat up the electrodes in order to get a high temperature so as to get a conducting vapor, and of course then you have electrodes that are covered with cesium atoms and they’ll all emit electrons. Now what are you measuring, the emission of the electrons, or the amount of ionization there is in the vapor? I never could distinguish between the two. So far as I know, no one else ever succeeded in doing it. It was an impossible experiment as far as I know. But that’s why I got off on that. //But the first paper I got onto by my own interest; that was entirely my own: the variation in the heat capacity in hydrogen gas if you go to low temperatures// I tried to work out the heat capacity of hydrogen, and that was before we knew about ortho and parahydrogen and things of this sort. So of course I didn’t get the correct answer.
But you did have the occasion to look up the quantum mechanical, or quantum physical literature.
That’s right. I got the physical literature on the subject at the time and also this explained why all sorts of molecules for which this decrease in rotational heat capacity was unknown nevertheless had a finite entropy; and so I got into that subject that way.
You don’t recall what the relative merits of Reiche’s competing solution were supposed to be? It was his formula for rotational specific heats that you finally used.
No, I don’t remember the details of that. I’ve hardly looked at it since. Of course almost immediately the quantum mechanics came along, the correct solution to the problem became evident, and so forth. There wasn’t any use in working it out, because it was obvious that this gives the correct formula, there wasn’t anything to work on at all… Well I didn’t use the right energy levels and things like that. But that’s the way it goes you know. I suppose that I was foolish in those days; I should have gone back and tried to clean up these things and publish the correct statement. But at the time I had moved away from California to a place where no check up on problems of this kind could be made, namely at Johns Hopkins, Completely different surroundings you see. Giauque at California went on measuring the entropies of these gases, and the whole problem was sort of taken over by Giauque and I didn’t pay any more attention to it. He used the correct formula. It was all obvious what the correct formulas were and so forth.
Did you manage to learn Danish while you were at Copenhagen?
A little bit. Very little. I could ‘travel’ in Danish while I was there. Danish is a very difficult language to learn. Bohr later told me that they ran a little school at the Institute after I was there to teach Danish to people, and they found that they had to teach a corruption of Danish. The Danes speak their consonants very softly so that it’s mostly a matter of vowels: [Urey illustrates the Danish pronunciation using the street name “Amagergade”].
The group at the Institute must have been very warm.
Wonderful, yes. I’ve just loved Denmark ever since. I get back to Denmark every chance I get. My wife and I will be flying about I think the 13th of June, we’ll land in Copenhagen and stay a few days, then we’ll go on to Zurich, we’re picking up a car and spending a month in Europe.
And how did you happen to go to Hopkins?
Got a job!
I noticed that you were negotiating for another year abroad.
Well I was at Harvard for a week or two, and then this job turned up at Hopkins, and I decided to take it because, after all, Hopkins is a pretty good place, and you’ve got a job there at essentially what was an assistant professor level. In those days it was something — I think my salary was $2400 a year! Well that doesn’t look like much of a salary today but back in ‘24 that was quite a salary.
So you did in fact then renew that fellowship, or get a fellowship and started at Harvard?
And there you began working on the hydrogen molecule ion [paper No. 6], or did you begin that at Hopkins?
No, I began that at Hopkins. That’s sad. So much enthusiasm — so much enthusiasm — all wrong, complete wrong idea. But you know I learned something from it, which a great many young people don’t learn. That is, if you’re not on the side of old Mother Nature, get on her side as rapidly as you can. It never pays to stick up for a wrong idea. I wish that some of the young people that I have to deal with at the present time would learn the lesson. That was very keenly a disappointment. But it was certainly all wrong.
The wrong idea you mean by the whole model or the whole idea of models?
Well, the whole idea of models. I had no evidence, the evidence that I was comparing it with was no good.
But that wasn’t clear for a long time.
It was clear to some wiser people, and it wasn’t to me. If I had been as critical of the data that Olson got on his experiments at California as I should have been, I wouldn’t have made that mistake. Moreover I got to trying to fit up certain lines in the hydrogen spectrum with this model and I found that they wouldn’t agree, and I tried to stretch my data one way or the other you know, fooling with it. It was an experience though. It was worthwhile.
You spent a year or so at that?
Well I should judge that I spent the better part of a year.
I notice that your arguments with Kemble on the subject mainly centered on the question of the quantization you should employ.
Yes. Well I don’t know whether that was correct or not. What did I do? had the straight-line model, vibrating straight through the middle between the two nuclei, and there ought to have been elliptical orbits that should lie in the plane, and of course then it should have oscillated at the same time in the third dimension and should have been a three dimensional model. I forget what I did.
I couldn’t follow it too well from your letter to Kemble. [18 June 1925].
I published it in Philosophical Magazine.
That was in fact published? I couldn’t find any reference to it in the Physical Abstracts.
[Searching] No. I published it in the Physical Review! [“Structure of the Hydrogen Molecule Ion” Phys. Rev. 27 (Feb. 1926) pp. 216-37. Submitted 20 Oct. ‘25.] I don’t know whether I have any reprints of it or not. I think that I was so disgusted with it that I never saved a reprint. But I think it would be interesting to look it up… Well we might as well get it out and see if we can find it then.
Do you recall what happens then, because in this correspondence with Kemble, it ends by Kemble’s saying that he feels that he can’t recommend it for publication to the Physical Review unless you cut it down greatly in size.
Well I don’t know what happened… We’ll look it up. You see I didn’t refer to it in there [showing Heilbron another paper]. And Ruark said, “well aren’t you going to include a reference to your paper?” and I said, “No, I won’t include references to papers that are incorrect.” “Just don’t refer to them”. Suppose I ask my secretary to go and get that.
I notice that you have Mrs. Marcet’s Conversations on Chemistry up there on the shelf.
Oh, yes, yes, that was given to me by (Mother).
Many people in the 19th century first became acquainted with chemistry from that book. You were beginning to tell me that things were not so easy, facilities were not so good at Hopkins, and that someone at Berkeley had to continue the measurements of gas entropies.
No, it was just a changed (status) there. At California they were interested in the third law of thermodynamics, and in the calculation of the heat capacities of gases. Now at Hopkins there was just nobody interested in that. [J.C.W.] Fraser was interested in theoretical work, in catalytic things. [W.A.] Patrick was interested in surface phenomena. There was no low temperature work of any kind going on, which was needed in order to establish the character of the hydrogen gas at low temperatures. The possibility, for example, of there being somebody around who would be interested in doing the heat capacity of parahydrogen and orthohydrogen, this sort of thing, didn’t exist at all. I had no facilities, nobody else had any facilities. It was a different place. So I had to do something that was possible there, and I got interested in this theoretical work, but also then I attempted to prepare atomic hydrogen, which was a chemical job that I thought I could do there, and this turned out to be wrong. It was (Gerald Went) at Chicago that claimed that he had prepared it, and this is quite incorrect. I couldn’t repeat it at all; it led to a little paper, but with negative results. Then I went on to do other chemical jobs, the emission of light by the halogens, the absorption of light by hydrogen peroxide. I had a student, a man by the name of L.H. Dawsey who worked with me on the absorption spectrum of hydrogen peroxide, and continuous spectrum in the ultra-violet.
This man got his degree; he disappeared, and I never knew what happened to Lynn Dawsey until very recently I got a letter from him. He has spent his life working on hydrogen peroxide in an industrial company, has invented processes for producing hydrogen peroxide, has some patents of his own he’s trying to exploit and so forth — became a very successful man in the manufacture of hydrogen peroxide. And this little paper on the absorption spectrum of hydrogen peroxide is all right. We found that you got a continuous spectrum, that’s undoubtedly due, I think, to the dissociation of H2O2 into two OH groups. But I simply set out to see what there was I could do in this changed place. Then, without having felt that I accomplished very much at Hopkins, I went on to Columbia. I had quite a bit of trouble getting started. You see, I thought I was going to be a theoretical chemist when I went to Copenhagen, and I just concluded I couldn’t do this thing. So I had to get back and start a completely new line of work. And it took quite a while before I got started at it.
You hadn’t done much experimenting before that?
Did you have much contact with the physicists at Hopkins?
I always went to the physics seminar. I was a good friend of Robert W. Woods Karl Herzfeld came along shortly later while I was at Hopkins and we became very good friends and have remained good personal friends ever since. I made very good friends with (Frank Price) who stayed on at Hopkins for a while, then went to a Catholic university. In fact Hopkins had bad luck in this time. They made things unpleasant for a very considerable number of outstanding people. [Interruption: secretary brings Physical Review]… There it is — why don’t I make a copy of that and let you have it. I’d like to have a copy myself in my files.
It would be quite amusing to know what happened because I don’t know whether I sent you all that correspondence with Kemble. At the end he says you had sent it off to the Physical Review; and he had been selected as a referee, and he thought it was too long.
Well it finally got in:
What did Wood think about the modern drift of physics?
Well, you see you have to understand Wood. Wood was a very clever experimentalist, but he did not have a strong understanding of theoretical physics. Wood would be able to grasp, (absorb) intuitively, what the physical importance of a certain thing was, and it would be entirely on the basis of classical physics. He never understood quantum physics at all. I don’t think that I ever heard him say anything critical about modern physics, it just wasn’t part of his thinking at all.
And he didn’t miss, it?
He didn’t miss it. He talked about resonance spectra, and when he meant resonance spectra he meant resonance in the old sense.
Like a bell.
Yes. Resonance radiation. And he did these beautiful things by illuminating iodine with the light of the mercury arc lamp, a cool mercury arc lamp, and got these curious resonance spectra which if I remember correctly — it’s a long time since I thought about this — only amounted to this: the mercury line corresponded very very closely with the transition for some given rotational energy level of the iodine molecule to an excited rotational energy of the molecule and that one level was excited; then transitions from this to all the vibrational energy levels of the normal state occurred, but gave just two rotational lines in each band. This is what happened. And he referred to that as resonance radiation of the iodine molecule. He never did understand it, in my opinion.
Well he had been propounding those puzzles for some time.
Oh, enormously so, yes. I often tell this story about R.W. Wood. One time when I visited him at the laboratory, he was doing an experiment with a cool mercury arc. In order to cool it he put a beaker of water around the pool of mercury and of course he ran the arc for a short time and it got hot, so he took a rubber tube and ran the water into it. Well of course the water overflowed from the beaker, and ran down across the table and went through a crack in the table, and he shoved a pan underneath and went on with the experiment. It worked! And by the time the pan got full of water his experiment was finished. All the rest of us would have worked out a very complicated way of cooling the arc. [Laughter] It worked perfectly well, and he got his results and they were completely (correct).
He had a water clock as well.
He certainly did. And so this would be the way that he would work. You ought to talk to James Franck about R.W. Wood. James Franck knew him very well, and James Franck liked Wood enormously; he was very fond of Wood. But he recognized the limitations of Wood, and his great strength too. He had a great strength.
With his experimental abilities?
His experimental ability and a correct instinct for what was interesting. After all the resonance radiation in the iodine molecule was very interesting. He couldn’t explain it. I think that was explained by F.W. Loomis. Loomis I think explained the thing, and worked out exactly what was happening in the whole rotation vibration spectrum. Then you let it get warm. Well of course if you let it get warm or increase the density of the iodine paper, I forget which, and you get collisions; then you would get other rotational levels in the excited state filled, other vibrational energy levels filled, and then you would have a much more complex spectrum. But as long as you kept it cold, and the pressure low, why you got an exceedingly simple spectrum. But Wood never understood it in modern terms. I think that the modern terms are the correct way to understand it. I’m presuming that’s the case. He never understood that point.
…When did de Broglie’s work come to your attention, do you remember?
Not until after Schrodinger came through with it. De Broglie’s paper was not considered seriously, I think, in Copenhagen. I don’t know whether it was even known. What was the date of de Broglie’s paper?
His thesis was published at the end of ‘24.
Well you see I wasn’t there then. I came back during the summer of ‘24.
But there had been earlier shorter articles.
There had been. But so far as I know they had never been taken seriously.
Do you remember whether there was much discussion of the relative merits say of Sommerfeld’s approach to the whole problem of quantum physics and Bohr’s approach?
I think that Bohr and Sommerfeld were not good professional friends. I don’t understand this thing entirely. I hope you will be very careful about what I say, otherwise I can’t be –-
Please do, anything you want to say —
Sommerfeld went through these elliptical orbits, modified by giving quantized elliptical orbits, and got out the relativity splitting effect and in some way Bohr objected to this, didn’t like this model at all. I learned most of my modern physics — modern at that date — from reading Sommerfeld’s book, not from reading anything that Bohr wrote. But Bohr never liked Sommerfeld’s approach to these problems. I think myself that Bohr was a much more fundamental thinker than Sommerfeld. The problem seemed to be this, that unless you have a way of detecting a theoretical conclusion, then there is no use in discussing a theoretical model. Thus the nonrelativistic orbits of Sommerfeld, for Bohr, had no meaning at all because as long as no difference in energy, there is nothing to discuss. Unless there is something to make them distinct, then they wouldn’t exist. That was his philosophy. It’s very much the indeterminacy principle idea in a way, that unless you can observe a physical thing, then there is no use to manufacture a more complicated theory back of it — sort of Occam’s razor. I think the discussion between Sommerfeld and Bohr turned more on this philosophical point of view than anything else. As soon as you have a difference in energy because of the relativity effect, why then of course you were justified in having elliptical orbits.
But even so, the procedure of the phase integrals Bohr never liked anyway.
No, he never liked them.
Was any of this discussion explicit when you were in Copenhagen?
Not much. What Bohr had to say was expressed in that big paper on the periodic system and the origin of the shells.
He had no objection to Stoner’s regrouping?
Stoner regrouped them as two and six instead of [four and four] — yes. Well I don’t think Bohr did, no. He only said that, well, he had made the guess, had divided it in two, he saw no other reason to do it. Of course Stoner had very good reason for it, because it turned up in the g-factors for the s and p levels which I felt justified the thing very completely.
We also talked a bit about the Bohr-Kramers-Slater paper. Do you remember that there was any general discussion of that?
It was just talked all over the laboratory. Oh, yes. There were discussions about it, and Kramers often talked about it to me while they were working on it. Slater talked about it a little bit, but Sister is not a very marked conversationalist about scientific things, you probably noticed that, at least he wasn’t in those days, and I don’t think he is yet.
Did Kramers think that that was really the answer?
Well he thought it was a very important paper, but I wouldn’t think that Kramers felt that it was the final answer, no.
Did he, do you know, regard Heisenberg’s papers as a final answer?
Oh, immediately. So far as I know, I never talked to any competent physicist who didn’t almost immediately recognize the outstanding importance of Heisenberg’s and Schrodinger’s papers. There may be somebody that was fighting these things, but I never ran across them.
One can certainly believe that after Schrodinger one might be more impressed with Heisenberg, but even in the period between those two?
I think so. I think people realized that we’re now getting to the sort of mechanics that describes what we observe.
How did it happen that at Hopkins you taught the physicists, or reported to the physicists on the Heisenberg paper?
This was a seminar that we had in Washington. At Hopkins the physics. department was pretty old-fashioned. Joseph Ames was there, and he was a sort of imperial person. No one could say anything that Ames didn’t like, and Ames was very much of a classical physicist. And Wood, of course, knew nothing about modern physics. And [A.H.] Pfund was the other man. Again, he didn’t know anything about modern physics. It was pretty much of an old fuddy-duddy department in a certain way. In Washington there was Foote and Mohler; Otto Laporte was there, and M.A. Tuve — some of the young men you see — oh yes, Meggers. Then the x-ray man — what was his name — Wyckoff. He was there and Sam Allison was in Washington at the time. And then there came Bichowsky and me from Hopkins, and maybe M.A. Tuve came over at the same time; Ferdinand Brickwedde also went down there, and we had a seminar that met once a week at the geophysical lab at the Bureau of Standards, at the laboratory of terrestrial magnetism and so forth. It was at this seminar that I presented this. This was a more important seminar to us than what was going on at Baltimore. It was an interesting group. We had a reunion a few years ago at which we got together quite a number of people. M.A. Tuve was the promoter of it.
You just selected topics yourself?
Yes. When this came out, I undertook to give the report on Heisenberg’s paper.
What of the following ones, the Born-Heisenberg-Jordan papers?
Those came up also. I don’t know whether I reported on them or whether Otto Laporte did.
Was it quite clear to you from that first Heisenberg paper that, although the mathematics might not have been so clear, the general import was?
The general import was correct. We all took it very seriously indeed.
Was there a general preceding discussion at Copenhagen say about the necessity for going from derivatives to differences as the way to —
Not that I know of. Possibly John Slater would know more about that than I do, but I don’t remember anything of the sort during ‘23-‘24 when I was there,
Did other people share enthusiasm about the Bohr-Kramers-Slater paper? Did it strike you as a step forward?
I thought it was a step forward, yes. About other people, I don’t know, I haven’t heard anyone say much about it. You see, it only lasted a short time.
Yes. And it was those experiments that are reputed to have done it in that in fact did it?
I don’t know the answer to that. Almost immediately, within a couple of years, the quantum mechanics — the matrix mechanics and the wave mechanics — were out. Then there wasn’t any need for it at all…
Did Lande’s discussion of the relativistic versus the magnetic doublets come at Copenhagen? You make a good deal of that in your piper with Bichowsky.
Oh, is that so? I guess probably we do. Well of course the end was that they’re both magnetic and relativistic, aren’t they?
That’s the way it finally turned out. But before the spin there was a considerable problem. Do you recall how you got together with Bichowsky on this business?
Bichowsky and I had offices next to each other, and one day I was puzzling about what these doublets meant, and I was talking about this spectroscopic problem and so forth, and Bichowsky said, “Well, probably the electron is magnetic, it’s rotating.” And I grabbed on the idea, and we went to work on it. Bichowsky was trying to work on some magnetism business at the time, do you see, I’ve forgotten what that was; and I never could get a really good understanding of what he was talking about. And I don’t think he ever got the thing straightened out in a clear way. I don’t think he ever published anything except this little paper, did he?
I don’t know. This is on the magnetic properties of bodies in general?
Yes, it was some magnetic properties of bodies, the origin of magnetism and things like that. Then this got us to working, and I think — I forget what we put in that paper. I ought to get out that paper and look at it. But we saw how we could get the doublet properties, didn’t we?
Yes, and how you can get, particularly, the Z4 dependence.
Yes, that’s right. But that’s how it started, it was just a little conversation with us one day, and that’s it. Bichowsky is the one who brought up the magnetic electron and not me.
Do you know if he did it in connection with Lewis’s notions? Did Lewis talk about magnetic electrons?
I don’t think so. Compton talked about magnetic electrons.
Compton did too. But I think, in that book Lewis published in 1923, Valence — I forget what it’s called — he suggests that maybe the paired electron orbits are paired because of some magnetic property of the electron.
Oh does he?
It’s not very clear what it’s about, but he does make the suggestion. That had, then, nothing to do with Bichowsky’s idea?
I don’t think so.
Nor did Compton’s earlier suggestion?
I don’t think so. We found I think after Bichowsky had made the suggestion, and after we had worked on it for a while, then I think we discovered Compton’s paper. We referred to Compton’s paper didn’t we?
Didn’t we? Then we must have run across Compton’s magnetic electron afterwards.
So did Goudsmit and Uhlenbeck, and perhaps you saw it there, because they found it later also, and you probably did follow their publications.
Probably so. We just hadn’t seen it I guess.
The main object then was to explain the origin of these doublets.
Of the doublets, yes. ‘Why were the p levels double.’ In sodium, for example, the p levels are double.
You mention in that paper the fact that Heisenberg’s scheme, that is, a “Rumpf” that has angular momentum, the energies of the levels come out backwards, the higher one lying lower; and that you can turn it around by giving the electron its own spin, its own magnetic properties.
Oh is that so. I had forgotten this.
I was curious whether that had been the puzzle with the doublets.
Well you see, I may have been reading some of these things. I might even have been reading Lande I don’t recall that I was. Do I mention it?
Oh, yes, that comes up.
We mentioned Lande. Then undoubtedly I was reading Lande but I’ve forgotten it, and some of these problems in connection with it.
Then of course you select the quantization that makes the azimuthal quantum numbers half integral.
Yes, I suppose we would.
Were you uncomfortable with that conclusion, or at the time was ft satisfactory?
I remember. It wasn’t a very good paper. I’ve always wished that Bichowsky and I could have done better on that paper. I’ve always been a little bit sorry we published it, because I think it prevented a Nobel prize for the spinning electron to Goudsmit and Uhlenbeck. I always feel sorry about it, and I wrote to the Nobel Committee telling them so. Because I don’t think we added much beyond what Goudsmit and Uhlenbeck did, maybe we only added confusion. It was a completely original idea with us, this I always insist on, but I’m sorry that we just didn’t shut up. It was a matter of young people, you know, as I often say, trying to get ahead, and so forth. If I had been a little older and a little more mature I wouldn’t have done it.
Well then of course there was this Compton suggestion, and there was an earlier one too by a man named [A.L.] Parson, do you remember?
Oh, Parson, I guess so.
And I suppose there might be others.
I suppose so. So I suspect it was a pretty generally thought-of idea. And the specific thing that made it important was the spectroscopic application. That was the evidence.
…But it must have been annoying to see that paper by Goudsmit and Uhlenbeck.
Yes it was. On the other hand I think that ours was more annoying to them than was theirs to us. Well we’ve all been very good friends ever since. Goudsmit, Uhlenbeck and I, we’re never cross at each other, even though — at least I never felt anything. They never exhibited any animosity towards me or anything of the sort.
Have you talked with them about it?
We talked a little bit about it. When they were given a prize somewhere in New York I wrote a letter congratulating them on it, and telling them I was sorry we had published our paper and so forth.
They had gone to talk to Lorentz about it after they had written a note or something and Lorentz convinced them it was impossible.
Was it Lorentz? Wasn’t it Ehrenfest?
No. Ehrenfest had sent it in. After talking with Lorentz about it, as I remember the story, they were so unhappy that they had decided not to submit it; but Ehrenfest had already sent it in.
I think that there is another difficulty. I think that Ralph de L. Kronig had also thought of it… And he was convinced by somebody else that it true.
Pauli in fact.
Pauli. Well, I’ve often felt badly that we hadn’t stayed out of it. Perhaps Goudsmit and Uhlenbeck would have gotten the prize for it, but probably not, because of the Ralph de L. Kronig problem, and the fact that Compton had suggested magnetic electrons before. Probably all of these things showed that there were many people right on the verge of it, and whenever you get to a place where you can’t spot a person uniquely, well then the Nobel prize committee can’t award a prize for it, because it immediately becomes a problem of who really did discover it, and they do have a lot of arguments about their prizes. They will not give a prize, but once they’ve given it they don’t want an argument as to whether it was deserved or not.
As you say, the Kronig problem was really insuperable. I think that on those grounds alone that —
I think that Ralph de L. Kronig really probably deserves credit for discovering it. Of course it’s a curious thing about Kronig isn’t it? What has he done since?
He did some pretty good work on molecular spectra and molecular structure — I don’t know what else.
He hasn’t kept up –- I wonder what’s happened to Ralph?
He’s in Delft.
He’s in Delft. But he apparently just got back to do teaching and to be a professor at the university and not do much research of any kind, isn’t that true?
Well he did do this molecular work as I say, through the early ‘30’s, but I can’t recall what happened then. Then he went to Groningen. I don’t know what he did after the early 30’s, but I think he did do quite good work on molecular spectra. He’s been at Delft for quite a long time.
He didn’t keep on producing. But of course Goudsmit and Uhlenbeck just kept right on going all the years after that.
Kronig and Goudsmit wrote a paper together on intensities in ‘25 just before the spin, which makes the issue even more sticky.
You mention here the work on intensities of spectral lines and so forth, formulation of sum rules.
I had thought that since you evidently had some connection with Werner Kuhn you might have been interested in that question.
I don’t know what the answer to that question is.
Do you recall when the exchange theory first came out, the theory of non-polar binding?
Let me see, Heitler-London theory, yes I remember that.
Did chemists react to that antagonistically generally?
I don’t think so. I think chemists were completely bland about the matter, didn’t understand it, and largely, except for Pauling, nobody paid any attention to it. Electron diffraction I see you put down here [page 5, outline]. I must tell my experience about that. I went up to New York to a Physical Society and saw Gregory Breit, and I said, “Gregory, the electron should have wave-like properties; from the wave mechanics, it must have wave-like properties. You ought to be able to diffract it on crystals.” He says, “Haven’t you seen the paper by Davisson and Germer on this subject on the program?” Well no, I’d never seen it. [Laughter] Another good idea gone bad.
Well they had a head start on you by quite a few years.
Yes. Now you ask about the uncertainty principle and the general statistics approach to the new mechanics [p.5, outline]. Well, I don’t know what the general reaction of people was; mine was unfavorable. I didn’t like it.
Were you in favor of the Schrodinger approach?
The wave-packet approach. I thought this was a good idea.
And the paper with Ruark is an example of your reluctance.
That’s right. Now this thing of Bohr’s ‘extension-in-depth’ — complementarity — you mean by that his extending it to other things besides physics?
Well, I think the ideas are not quite the same either.
I don’t quite understand the origin of the indeterminacy principle. I think Bohr was a little annoyed about it; at least I heard it was the case that he had felt that this was implicit in his conversations then taking place in Copenhagen and that Heisenberg sort of just went off and formulated them in an exact way and published them. Have you heard that?
I know there was a good deal of, let’s say annoyance, over the problem because there were these conversations between Bohr and Heisenberg, as you say. Bohr went skiing — it must have been in the winter of ‘26 or ‘27 — and while Bohr was away Heisenberg evidently concocted the paper in which he described the uncertainty principle.
Well, you see, I’ll tell you what my feeling about it is. Bohr never expressed himself very clearly. I think it’s entirely possible that Bohr had these ideas himself; and I think it’s entirely possible that Heisenberg didn’t know it. You see? So I think that this may be a pure misunderstanding on both parts.
Well, the ideas are somewhat different though, as well. Bohr was evidently annoyed not only because Heisenberg wrote this paper but also because he didn’t agree with the formulation — so we were told. Bohr hoped for a more general approach, not pinning the thing down to the proper ties which can be analyzed with a microscope and so forth.
Bohr discussed some of his ideas in my presence in the years following that in the United States in which he was talking, for example, about how animals have the ability to find directions and to locate themselves on the surface of the earth and things like this; that as intelligence grows and as we have an ability to invent optical instruments and view the sun and make magnets that tell us what direction is north and so forth, we lose a certain sense of direction that animals have retained. We do not have this. He talked about things of this sort as a sort of complementarity principle as nearly as I understood them. I personally was never greatly impressed with these ideas. You see, Bohr and I in a way never understood each other. Bohr was enormously admired by the best theoretical physicists, and they felt that they got an enormous amount from him, but I personally never felt I got much from him. Now I’m perfectly willing to say that’s my fault, and not Bohr’s, because I greatly respect the opinion of people like Robert Oppenheimer, Pauli and Heisenberg and people like this that are outstanding theoretical physicists. I really think that when it comes to this difficulty, on Heisenberg’s part it was probably complete misunderstanding. If Bohr gave it to him, he didn’t know it.
Yes, well I don’t think it was a question of his giving it to him, or of Heisenberg’s having taken anything. It probably did grow out of these conversations, and it’s understandable that Bohr should be annoyed at Heisenberg’s having written it up while he was skiing, or however it happened. When you say you got little out of Bohr, do you mean his writings or his immediate conversations?
His immediate conversations.
He was not present much though, you say, or accessible, anyway, during that year you were in Copenhagen.
But he — well, first of all, he always speaks any language badly. He had very poor enunciation. But he also was talking in what I thought was a very profound way. In a way I’m bothered with Robert Oppenheimer in the same way. Robert Oppenheimer gave a speech at the National Academy of Sciences centennial celebration that I didn’t understand at all. It didn’t mean anything to me.
What was it about?
Oh, just generalities about science. I can’t even tell you what it was about. I happen to know that there are other well-known scientists that feel exactly the same way, that they didn’t understand anything that Oppenheimer said. But there are people that are intensely enthusiastic about what Oppenheimer had to say.
It was Bohr-like?
Bohr-like in this way.
And do you think many people were impressed the way you were with Bohr? I mean, one has the theoretical physicists who appreciated him greatly and people like yourself who were involved in more concrete problems, who perhaps didn’t. Was there much of a fight over complementarity in the United States?
No, I don’t think so. I don’t think so. And the indeterminacy principle was accepted immediately by everybody. As soon as I read it and studied it and so forth, I was sure it was right; and I felt it was a very, very fundamental principle. It’s the more general attempt at applying this to biological things and all this sort of thing has never meant anything at all to me really. Yes, Heisenberg’s indeterminacy principle, I just felt was quite correct from the very beginning.
But Bohr’s extension of that — just for physics — namely the principle that one must always describe one’s measuring apparatus in terms of classical physics and deductions which followed from Chat and so forth; this degree of complementarity was considered the same thing as indeterminacy?
That’s right. Maybe it should have been the other way around. We regarded Bohr’s remarks as an extension Heisenberg’s indeterminacy principle when as a matter of fact maybe it ought to have been that Heisenberg’s indeterminacy principle was only a special example of complementarity. Maybe this is the way it should have been. But in my mind the indeterminacy principle was the thing that I accepted and all the others were elucidations of this. This is the way it seemed to me.
Do you of any people who later were impressed by Bohr’s extensions of the argument?
You’d have to talk to better —. My answer to that is no. But it may be that again, the best theoretical physicists did understand the distinction between what Heisenberg was talking about and what Bohr was talking about and appreciated the difference. I would suggest you talk to some of these people. I can’t think of the man’s name at Cal Tech, a theoretical physicist, not Gell-Mann, the other man, older than Gell-Mann. I know him perfectly well; I can’t think of his name now.
Where then does your paper with Ruark [paper No. 9] fit into this scheme?
Well, I must tell you. Ruark and I were writing our book, and Ruark and I got to figuring about what we could expect for conservation of spin — angular momentum — between atoms and light. It was entirely a matter that grew out of our attempts to work out what happened in the emission of light in a magnetic field and what happens in the Zeeman effect. I’ve forgotten what we said, you see, but it was entirely a matter of talking about quanta as though they were particles that had to carry away angular momentum, as I think they must. It had nothing to do with these general principles in any way — just that we were writing a book in which we came across the problem of the Zeeman effect, the emission of light in a magnetic field and how this must conserve angular momentum some way, and the light must carry off angular momentum. That’s all it was. It had no connection with this more fundamental discussion of indeterminacy principle and things of this sort.
Well, you’d be interested to look at this paper with Ruark. I read it as antagonistic to the probability, or statistical interpretation.
Oh, is that so?
I think so. In the first page or two you point out that there exists a statistical way of looking at things, but that you’re not going to pay any attention to that.
Well, I’ve forgotten it then. You see, I’ve forgotten it completely. That was a long time ago. That was when, ‘26, ‘27?
It was submitted in September of ‘27 to the Proc. Nat. Ac. Sci. which means that it was written after the indeterminacy paper — but just after; probably you hadn’t seen it yet. You have some quite elegant arguments in that paper, I think.
You ought to talk to Ruark about it. Are you going to see Ruark?
Well, I’m not quite certain. Where is he?
He’s at the Atomic Energy Commission in Washington. He has charge of the fusion project, the fusion of heavy hydrogen, what do they call it, the Sherwood project? He’s been doing that for years. I think he was much more concerned with that particular paper than I was. I think the drive to get that paper done came from Ruark.
Had he had any earlier involvement in quantum physics? Had he had any European experience?
No. I don’t know what Arthur had published before then. Arthur was, Arthur always has been, an exceedingly meticulous person. There is a new edition of the book coming out — a paperback in two volumes. It’s supposed to appear any time now. It’s simply a reprint with revisions, mistakes correct and so forth; we hope they’re corrected, anyway. But, here is Arthur; he just plugged through that whole thing with the greatest of care, avoiding all the mistakes and re-writing the constants and everything all the way through. Terribly meticulous person about this.
That’s the second edition?
The second edition, yes; but he was the same way when we wrote the book. That book owes a lot to Arthur Ruark. Well, as I say, I think the paper you were talking about was mostly Arthur Ruark and not much me.
How did that collaboration begin?
He was part of this group that met in Washington. His home was in Baltimore, and he often came up. We got acquainted that way, and we wrote the book back and forth with him, in Washington and me in Baltimore, and finally finished it up when I was in Columbia. We never were at the same place. I think if we’d been at the same place we would have revised the book long before this. He was always off in Pittsburgh or North or South Carolina and I was in Washington and New York, and we never settled down in the same place. And so we never had a chance to revise it, really.
It was a long time in the making, wasn’t it?
Three or four years we worked on that book. We started about ‘26. It was published in ‘29, I think, twenty-nine or early ‘30. Yes it’s dated ‘30. You see, we wrote the preface January, 1930.
Did Pauli and Goudsmit know of your enterprise?
Of the book? I don’t know who knew about it. I think that most people thought that these two young squirts down there wouldn’t turn out anything worthwhile anyhow. [Laughter] Say, it’s time for lunch.