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

Oral History Transcript — Dr. Samuel A. Goudsmit

This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.

This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.

Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.

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


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



Interview with Dr. Samuel A. Goudsmit
By Thomas S. Kuhn
At the Rockefeller Institute
December 7, 1963
open tab View abstract

Samuel Goudsmit; December 7, 1963

ABSTRACT: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 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: Gladys Anslow, Robert Fox Bacher, Ernst Back, P. A. Boeser, Niels Henrik David Bohr, Hendrik Brugt Gerhard Casimir, Walter Colby, Dirk Coster, G. H. Dieke, Paul Ehrenfest, Albert Einstein, Enrico Fermi, George Hartwig de Hass, Werner Heisenberg, David Inglis, Edwin Crawford Kemble, Ivan Robert King, Oskar Benjamin Klein, Ralph de Laer Kronig, Alfred Landť, Otto Laporte, T. van Lohuizen, Hendrik Antoon Lorentz, Fraulein Mensing, Edgar Meyer, Robert Andrews Millikan, J. Robert Oppenheimer, Friedrich Paschen, Wolfgang Pauli, Linus Pauling, Isidor Isaac Rabi, Harrison McAllister Randall, Adolf Smekal, Arnold Sommerfeld, Thomas, Uhlenbeck (George's father), George EugŤne Uhlenbeck, Albrecht UnsŲld, W. van der Woude, Vry, John Wulff, Pieter Zeeman; Universiteit van Amsterdam, Rijksuniversiteit te Leiden, University of Michigan, Teyler's Museum, Universitšt TŁbingen, and Universitšt Zurich.

Transcript

Session I | Session II | Session III

Kuhn:

You remembered a couple of things that you thought had been omitted[ from the previous tape, 94a].

Goudsmit:

On the record, I think, to make it complete, I ought to mention two people. First, from my early youth, one person who probably had a great influence on me but I am not sure - - - my grandfather, my fatherí s father. My grandmother was semi-illiterate. I always tell the story that when my cousins came from Paris and brought a little dog along, she absolutely refused to believe that the dog understood French. Honest! I still remember that as a child. My grandfather died before the First World War so I didnít know him well at all. I was, probably about eleven years old.

But he was a very interesting man, from a very poor family. The story is that he ran away from home and traveled a lot. When I knew him, he was the tourist guide for the two swankiest hotels in The Hague, at a time when only special, rich, important people traveled and when a tourist guide is not what it is now.

There were no tourist agencies, no buses and so on with hundreds of tourists going around, and his autograph book, which now was lost in this war, was a marvelous collection of very distinguished names. Really, a very fine man, completely self-educated, learning languages all by himself; even when he was 76 or 77, he was still studying Spanish by himself. He was the one who introduced me to the art galleries. He took me around as a little child, and said, "I want to show you these beautiful paintings.Ē In retrospect, his taste was not what my taste is now: he would have liked Whistlerís Mother. But he took me to the art galleries in The Hague.

He had traveled so much. And I remember that his death affected my father very much, it may have affected me as a result. Later on, when I went to high school and the university, my sister and I often said, If grandfather still had lived, he could have helped us, he could have told us about this, or he would have appreciated it. He would have been proud of me. So he must have had an influence on me indirectly that way, and also in introducing me to the art galleries.

Kuhn:

Your sister, I take it, did want you to go to the university, she was also pleased, and when she spoke to you about your motherís almost taking you out, she was on your side?

Goudsmit:

Completely. But she was taken out of high school by my mother. The other person I forgot again is, during my student days, the influence, the contact I had with Dieke. He belonged to the commuters, and we were together practically daily.

Kuhn:

Was he in your class in effect?

Goudsmit:

Maybe possibly a year later, I donít know exactly. But he belonged to that little inner circle around Ehrenfest. He came from Germany. He was really from Dutch parents and had a Dutch passport, but had never, as far as I know, been in Holland before until he became a student. He had relatives living in The Hague, a cousin, where he made his home and commuted. I often spent afternoons and evenings at his house, and he used to come to my house, and we worked together quite a lot and talked about physics. I still remember the day he phoned me about the hydrogen molecular spectrum, that he had found the solution. I like to tell that, because it also throws a light upon me I again must stress that I really am not a mathematical physicist or a theorist, and I did not understand all the implications and all the model but he had done something which appealed to me enormously.

The lines of molecular hydrogen fall in little groups, and people had tried to fit a formula to them, but it didnít fit very well and didnít make any sense. Dieke had gotten interested in that kind of spectroscopy just by reading about it, and once he phoned me — I donít even know whether he remembers it — and he said, ďSam, I think I have found the solution. These bands-ď— the so-called Fulcher bands because of a man, Fulcher, who had measured them and tried to fit them in a formula - ďthat is wrong.Ē

You must take the first line of each band, and that is a real band; then the second line of each band, and the third line of each band. Itís because of the hydrogen moment of inertia being so small, the rotational distances were so much larger than in any other band spectrum compared to the frequencies.Ē I did not understand the implication. I probably didnít even know quite what a moment of inertia was, with some exaggeration. But that was the kind of puzzle which immediately struck me as a great discovery, you see — solving a puzzle. By just arranging the lines differently, suddenly the whole thing cleared up, and as you know —.

Kuhn:

When was this now?

Goudsmit:

That must have been 1924 or so, you must find out from Diekeís paper. It must have been about that time, before he went to the States.

Kuhn:

You do something like that later with the calcium fluoride line or something of the sort.

Goudsmit:

Ja, but that was no good, that was no good; that was second-rate. I did that mainly in order to learn some newer experimental techniques in Zeemanís lab. I wanted to excite a molecular spectrum, to measure it; that was no good.

Kuhn:

The paper Iím thinking of —.

Goudsmit:

Is an experimental paper on the calcium fluoride bands.

Kuhn:

Yes, I guess that was the one.

Goudsmit:

But this Dieke — the way it hit me; that was the kind of thing I could recognize as important, whereas those Pauli papers I did not recognize as something which was a contribution because it was outside the way my brain worked. Thatís why I like to tell things [like this]; it was just marvelous to see people do things like this: ďNow if you take the first one here, and then the second one and call that a band, the formula fits, the constants come out to the right value.Ē So I was enthusiastic: I probably encouraged Dieke no end. You see, I had published already a number of papers on spectra, so he confided in me right away. And that became his life career - — hydrogen molecular spectrum, it was very nice. So—-.

Kuhn:

Was he in other respects also the same sort of physicist you were, or was he —?

Goudsmit:

He probably was the same sort of physicist, but he was wiser. He right away went over to experimental physics, which I should have done also, directing experiments, analyzing them. But I did not have the full opportunity, and when the spin came, henceforth, I was type-cast as a theorist. That was a pity, as far as I am concerned. But Iím sure that Ehrenfest realized it at the Ďbeginning; thatís why he sent me to Zeeman.

Kuhn:

One thing we didnít do enough of yesterday: tell me something about Zeeman, about Paschen. I donít know whether Back was there when you made your first trip or only later.

Goudsmit:

No, he was always there.

Kuhn:

Also about Ehrenfest. George has said so much about Ehrenfest I doní t know how much you want to add, but youíve said things about him. But nobodyís told us much about these other men. Iíd be interested in knowing what they were like as human beings, as scientists, but also what their attitude was towards this whole theoretical development that was going on in the time, the extent to which they knew it, the extent to which their work was guided by it. .

Goudsmit:

Zeeman was a man who was rather distant, very formal, very aloof, very friendly because he had learned to be friendly. He always looked to me like a kind of beneficial old rich man who looked down a little at you, very nice and friendly like a grandfather talks to grandchildren. He did not understand the modern physics of the day, there was no doubt about it. He was not active in physics when I knew him; he came to the laboratory at most one hour a day, gave a lecture on spectra lines which I never attended—maybe once— to the students, and looked around the lab, a little bit, patting people on the back, and then going home again.

Occasionally he had these semi-formal dinner parties where the assistants were invited and had to give the maid a large tip, more than it would cost us to pay for our own dinner. I was invited out to his country home a couple of times, but it was very distant. His influence on physics — he was just a European man; because he had done great work once, he was respected and invited and he traveled and he got honorary degrees. But he did not take part in anything and did not influence anything. What I learned in Amsterdam I learned from his assistants.

Kuhn:

Who were the assistants at that time?

Goudsmit:

The man at that time from whom I learned real experimental technique was a man Van der Mark who left to go to Phillips and did important work at Phillips with van der Pol the radio man, all experimental work. From him I learned most. He was the man who knew how to run the magnet and how to do everything, how to attend to a large switchboard, batteries, that all the normal theorists have usually been afraid of. I learned that, I learned to walk around, to talk to the men in the shop and tell them what tube to make for me. I learned more about the development of plates and measuring, and all kinds of techniques, mainly from Van der Mark.

Then there were some other fellows around there from whom I did not learn much. So that about Zeeman; itís very little. You can already see in that photograph I published of him — I donít know, that is really characteristic, canít help it. He was exactly the way he looked — awfully nice, but distant. I remember I was told that in later years he wanted as his successor a Nobel prize winner. That is why I was really never considered at a time when I might have been considered, according to Ehrenfest. This story was told me by others, and I donít know how reliable it is. But itís very lucky for me, of course, that I wasnít considered at the time, because then I would have, ended up in Auschwitz. In the early years I would not have been able to resist this particular position which really was the kind of thing I dreamt about as a student: if I ever get into academic work, thatís where I belonged. I had the mistaken idea that there was no change in physics. I think Iíve told you that, it is not on the tape, but Iíve told it in various lectures, that my dream, when I first published these papers, was that real immortality consisted in having a footnote mention in Sommerfeldís book. And that I made it.

I wrote it even in an editorial in Physics Today, so you can find it there. And how surprised I was that Sommerfeldí s book, nobody knows of it anymore! That was really my dream; I wanted a footnote in Sommerfeldí s book, and perhaps someday Iíll be good enough so that I can really get a high position in Zeemaní s lab. Maybe even be his successor. It was a kind of a dream. Some of the students who were there were assistants with me—I remember one said to me, ďSam, you look as if you belong here. When I see you come down that stairway you look just like Zeeman. You ought to be the successor here.Ē

Kuhn:

Really?

Goudsmit:

Yes. That was Vry, V -r - y, who told me that. That made me feel good; that was 1923 or so. But I never got the job offer until it was much too late. The reason was that I wasnít good enough as far as Zeeman was concerned. He probably was right; because if I had done it, theyíd still be doing spectroscopy: this is wrong. Now they do other things. Now you want to go on to other subjects. You want to ask about Paschen?

Kuhn:

Ja, Paschen and Back. Were you at Tuebingen twice before you came to this country?

Goudsmit:

I went at least three times there. Once for just a day visit, which was a very important visit, I mentioned it, where I was introduced to some of the other people. I think Back was there already. Then I was there a couple of years later for a summer. Then I was there again in 1927, and then from there I came to this country. Then I went back again after I was in this country, so I was there several times.

But when Paschen was there it was only the first two times. Paschen was one of the first really great men I knew, because he had such a liberal worldwide outlook. I do not know whether he had traveled very much, but nevertheless it was known to me already that he had been somewhat in trouble because he had kept contact with his American friends during the First World War.

Kuhn:

Really!

Goudsmit:

Yes. Many Americans had gone there. Paschen is responsible— I didnít know that at the time, of course—for the University of Michigan infra-red work. Old Randall went to Germany and learned infra-red technique with Pasehen. Paschen in those days used an old-fashioned galvanometer, a Pasehen galvnometer, and he built one for Randall, or Randall built one there under his supervision, and brought it back in his hands — so they tell at Michigan — and set it up at Michigan and that started the famous infra-red work at Michigan. I think they tried to invite Paschen a couple of times, but I donít think he traveled much. But he had a very liberal outlook. He told me already at the second visit that he had two assistants, Back and Schuler and he said, there was one thing he did not like about them: they were very narrow-minded, they were very chauvinistic. Schueler was his son-in-law, and they broke up — Schueler is still in Germany.

These people change with time — they were very friendly, absolutely. I never noticed anything of this narrow-mindedness. Back told me that he was very anti-Allies, the reason being that he came from a wealthy family in Alsace and, of course, the French —. Schueler was just German, I donít know what. Back came from Alsace. His family had large property there, which after the First World War was just taken over by the French. He lost it all, including his library, so he was always very bitter, he said. But it never showed up, because he was such an extremely friendly, kind-hearted man whoíd cry at the drop of a hat at injustice. He was exaggeratedly polite. It was very difficult to get rid of him at times. When you saw him in that long street in Tuebingen at the other end, heíd have already taken his hat off, you know, that kind of a man, a little exaggerated. He was also somewhat unworldly. I can only give it by means of an example.

He was not too happy about his position, even though he did beautiful work, you know how it is. Once he was summoned by the Minister of Education, or whoever the man was who was in charge of the state of Wurttemberg, because they wanted to promote him. He . talked with the man, and everything went off quite well. Then the man said, ďHow much do you earn now, what is your salary?Ē And Back gave him a figure . Then he said, ďO.K., weíll add so much, weíll make it higher now.Ē And then when Back came home, it turned out, of course, he had given him the wrong figure, much too low and with the additions, instead of being a promotion, it was a cut in pay. Now you see that was a typical, just a typical kind of Schlemihl Back was.

Now whether that was remedied later on —- it was most embarrasing to him. He didnít know what to do about itÖ He didnít dare to go back and say, ďHey, I made a mistake,Ē he didnít know — That was typical of him. According to John Wulff, he was also lazy, but I donít know whether that was really so. But you had to hold the whip over him to make him work. Back had worked with Paschen and was a precisionist when it came to the experiments, and had done beautiful high precision work in spectroscopy Ö higher resolution, better sources — but he was not interested in the analysis. If you told him this is interesting to do, he might do it if you kept after him; and the hyperfine structural work really only succeeded because I left a secret agent behind, John Wulff from M.I.T. After I left there was still a lot to be done.

That was now in Ď27 and Ď29. John Wulff was there, I said, ďyou get interested in that, this is nice work. You see that Back does it.Ē And he did! In those days already we communicated occasionally by cable, that was a novelty, trying to get the results sent to me by cable, But Back, I liked him, a little strange as I said, a little bit of a schlemihl, overpolite, but nice. And later he got a job in a small school where he was his own boss. He liked that for awhile, but then he noticed you couldnít do nice work, so he came back to Tuebingen later on.

Kuhn:

But this was after Paschen was gone?

Goudsmit:

In those days Gerlach had already taken over. Paschen was there only the first time I was there, and then Paschen went to the Bureau of Standards in Berlin, already in Ď26, I think. So when I was there in Ď27 Paschen wasnít there anymore, but Back was still there, and I worked mostly with Back. Lande was there, and Gerlach was there, and Fraeulein Mensing was there. Those were the people I had close contact with. Lande was difficult, Gerlach was nice and friendly, and authoritarian. Back was the one I worked with, and Fraeulein Mensing was the one I discussed physics with; she was the theorist. She worked for Lande, but she was more communicative.

Kuhn:

Are you in touch with Fraulein Mensing now, Mrs. something or other?

Goudsmit:

Mrs. Schuetz, no, she is in the Soviet zone now.

Kuhn:

Could you write?

Goudsmit:

Ja! Ja!

Kuhn:

Well, look, thereís a question I think would be worth asking her —[Interruption] Just a little bit more on Back, and also on Paschen. Again, how closely was either one of them concerned with what was going on in theory? To what extent were they—?

Goudsmit:

They were not. Back definitely not. And Paschen in those days —. No, there was quite a gap between experimentalists and theorists. The experimentalists, I had the impression, were always very pleased when their experiments were important for theorists. I believe that they had the impression that after all — I still have often — experiment is more important than the theory.

Kuhn:

But I gather from what you say that Back at least, and I think the same thingíS true of Paschen, were quite willing to have suggestions from the theorists, about what it would be worthwhile to measure?

Goudsmit:

Ja! But they also did it the other way around. They believed, like I always did, that by improving techniques you would hit upon new fields and open up new questions, and also get new answers. I am convinced that they worked in that direction: get a better source, better magnet, better field, better resolution in spectroscopy; thatís the main motivation. Improving the old work would give new results.

Kuhn:

When you worked with Back — this is obviously ahead of the, story — on the hyperfine structure, did he care at all? The papers, after all, were finally analytical papers, giving quantum numbers, values, and so on. Did he look at that and criticize that?

Goudsmit:

No, no, definitely not.

Kuhn:

Then his name is there because the data is there?

Goudsmit:

Because the data is there, and he would advise me about the reliability of the data. I would be willing to say the interval rule doesnít hold, or the interval rule holds with certain precision, and heíd say, ďNo, it does not hold, I am sure these deviations are real, because I have, measured themĒ So from that point of view he would criticize the statements I made about the data, heíd criticize very severely.

Kuhn:

But never the identification of a line with the following quantum numbers?

Goudsmit:

No, no. He was very little aware of that.

Kuhn:

Do you think Paschen was more aware of that?

Goudsmit:

Pasehen was more aware of that, ja.

Kuhn:

Very briefly, I think in 1925, you went to Copenhagen. Was that before —?

Goudsmit:

No, it was after the spin.

Kuhn:

Because you were there again in Ď27 to do the thesis, Right after the spin?

Goudsmit:

I was invited by Bohr to come to Copenhagen after he had been at Leiden to talk about the spin, so — I arrived there in January, I believe, of 1926.

Kuhn:

Then now letís go to the spin. Or not quite yet to the spin. Thereís this whole series of papers, your paper in Physica, your and Georgeí s paper in Physica, both Of them now on the contradictions in existing theory. The one with George is on the hydrogen-helium spectrum, and before one gets to spin, itís important to know how those papers come into being, because they set up so much of the problem. And as you remember in your own paper in Physica [Paper No. 20] you list sort of fiye problems, and you said four of. them you can get rid of with the Pauli theory, and then there is the fifth one, which is the riddle about the behaviour of the X-ray and optical doublet. This is the one then that spin takes care of. Those are very much of a piece.

Goudsmit:

This paper wasnít bad, I read it yesterday, this summary paper [No. 20] in Physica. Iím surprised!

Kuhn:

Now that paper clearly relates to the paper with George just before it?

Goudsmit:

Ja, but this [summary paper] was worked on, you see. The summary paper, which was published just after, was the one I was supposed to discuss with George. Ehrenfest knew I had been assigned to write this.

Kuhn:

Who gives an assignment like that?

Goudsmit:

I donít know who it was. Probably Fokker who was then the editor of Physica. I had written a summary article before; I had given a talk in Amsterdam for. the Physical Society on that stuff, and they said, ďwhy doní t you write the article?Ē

Kuhn:

Physica was the organ of the Physical Society?

Goudsmit:

Of the Dutch Physical Society. I think it was Fokker who asked me, and I was of course flattered and right away started to do it. Ehrenfest knew about that — it was such a srnall circle. So when George Uhlenbeck came back from Rome, Ehrenfest told me, ďYou get together with George, and tell him about this. Have him help you write it. In that way he can learn, can be brought up-to-date on all this vector nonsense, cosine pest, or whatever it was.Ē

Kuhn:

Was that Ehrenfestís attitude toward —?

Goudsmit:

Ehrenfestís attitude was that it was a necessary evil to know all this. He was very nice to me, he never rubbed it in to hurt me, but you could tell. He thought that was the only thing I was good for. I remember one incident. I remember the detail. He was talking in a colloqiuium or in a discusion about a statistical problem which he hadnít been able to solve. A few days later I came with the solution, and he was so surprised, He said, ďI would never have expected that you could do that.Ē Already then, you see I doní t know what it was, I think it was one of the Bose statistics formula where he wanted a model which would give this result, because he and Einstein had guessed what formula to use, but they hadnít found a model which would give, by ordinary statistics, that same result. And I found one. I do not know exactly what it was anymore.

He was so surprised. But this stuff heíd think I knew and he didnít have to know,so he left it to me. So I had been given the task to keep George up-to-date. And Iím still surprised when I see it, because I had always in my memory an idea that there existed a summary article with George, and that this was it. That the idea was that weíd be co-authors at that time. Why I am the only author here I still donít understand. When I read it now I must confess that I have the feeling that George contributed more to it than Iím willing to admit. In this summary article there are references to papers, and, you know, I used the trick everybody used, and say, ďwell, we shall not do this - derivation in this articleĒ The real reason was that I wouldnít know how to derive it. Iím sure that my reference to some of those papers by Pauli — Iím not sure, but I now guess — probably was Georgeís influence. He had read them and told me what was in them.

Iím a little bit surprised that Iím the only author here. But already I see the influence. The only way I can find out is if I had a copy of that article I wrote a year earlier by myself, whether it has similar things or not —- My dissertation is full of references, where Ehrenfest even in the final exam said, ďI know you didnít read that stuff, you wouldnít understand it. Thatís just for padding. Here are two of Dirací s papers, and so on — And I did it already then.

Kuhn:

Is it possible that when you remembered last time we talked, that you, hadnít known the Pauli paper, that this means that, in fact, George brought this in and that thatís why you use it here?

Goudsmit:

I would not be surprised. That is my sent interpretation. Itís most likely, because it was not up my alley.

Kuhn:

Did you read the Zeitschrift?

Goudsmit:

Ja, ja!

Kuhn:

Because this was generally a very theoretical journal, as compared with the sort of thing you —-

Goudsmit:

I read it. But what does reading mean? I read only those articles which were of immediate interest to me, and which made me rush home and see whether I couldnít write another paper.

Kuhn:

But you probably found more articles that did that in other journals?

Goudsmit:

No, in the Zeitschrift. All the spectroscopy was published in the Zeitschrift.

Kuhn:

But not experimental spectroscopy?

Goudsmit:

No, experimental spectroscopy. But I was interested in the interpretation, and guides to interpretation were there, and then Iíd rush to whatever it was, say the stuff that came from King in Pasadena. I think he published in the Astrophysical Journal. I said, ďGeeĒ Maybe these people in Germany havenít seen that. And Iíd get the Astroiysical Journal and see whether I couldnít get something about the iron spectrum out of that. It was very childish.

Kuhn:

What else did you get things like that out of?

Goudsmit:

Occasionally Physical Review already. I remember flattering Gladys Anslow once because I thought gee, a girl with such a beautiful name writing a paper on the doublets in sodium! She had measured — I had forgot the paper now. She was so flattered. It was one of the first papers I read on the subject. She is retired now, but she was at Smith College for many years. It wasnít until thirty years later that I finally met the beautiful Gladys Anslow; you know, it sounded to me very romantic. So I read Physical Review but mainly Zeitschrift fuer Physik, Annalen der Physik, Antrophysical Journal Phil. Mag. —those were the journals one looked through.

Kuhn:

This paper makes it very clear that thereís deep trouble. It is not clear in any of your previous work I meanitís clear that there are spectra that are, giving you trouble, and that there are multiplets that are-not responding as well as they ought to to the application of the ďg-factorĒ with Russell-Saunders coupling, this sort of thing. But here the problems clearly go deeper. Was that a surprise to you, as that something you came to realize in the course only of doing this paper?

Goudsmit:

No, I think it was my way of looking at things. I donít know, I sometimes tell my students that, when you want to get into research? that means you- want to write a paper, get your name in the Physical Review; what you do is you read other papers, but in the back of your mind, ďIt must be wrong,Ē or in the back of your mind, ďItís incomplete.í1 If you take thatí attitude — it sounds Ďlike a mean attitude - - it may lead you to something where you can make a Žontril5ution. You must not take the attitude, well, so-and-so has done it, so thatís the last word. That was an exaggeration of the Ebrenfest attitude. So whenever I write things I really always look for what isnít done yet, is there anything I can do? It probably was part of my own attitude, always looking for what hasnít been done yet.

Kuhn:

I would want to say that that describes a good deal of your earlier work very precisely, as I see it, but that here there is more in the sense that here you were pointing not to incompleteness, not to a wrong result, but to fundamental inconsistencies.

Goudsmit:

And that must have been the influence of George. I canít prove it; but it Wasnít in my nature.

Kuhn:

Even by the 1923 Bohr-Heft of Natturwissenschaften, which you surely read, there are plenty of people Ė thereís a little paper by Lande on the Aufbauprinzip and whatís the matter with it, and the definite inconsistency there. Was that a sort of problem that had worried you?

Goudsmit:

Ja, ja. I think itís mentioned here.

Kuhn:

So you donít think that there is a change in your own mind?

Goudsmit:

No. Except that itís better and deeper and I suspect that it is Georgeís influence, because in 1931, on hyperfine structure, I also wrote a paper, at that Rome Conference about the difficulties, all those things which did not fit in hyperfine structure. So it was always my attitude, because that was the only way I could hope to make a contribution, to find that somebody else had done it wrong, or incompletely. So the attitude is normal, but that it comes out so well, as I see it now, must have been George, listening to him. It made it necessary for me first to organize it better because now I had a live audience to try it out on, and the questions he asked, of course, were the most important things.

Kuhn:

Do you remember particular things that came up as you went over this area?

Goudsmit:

No. Only the skepticism he had. He didnít understand it, he didnít believe it. He knew classical physics. He saw all the classical difficulties which I honestly could not appreciate. To me these Bohr rules, ďFine, well, thatís how nature works. If we donít understand it, we have to wait a few years.Ē As I told you, I didnít know the theory of electricity and magnetim well enough to really worry about the fact that the electrons should spiral into the nuc1eus.

Kuhn:

But surely that sort of a problem was very old and even George cannot have been encountering that sort of problem for the firstótime?

Goudsmit:

No, but he couldnít get over it, and I could get over it. Letís put it that way: it didnít worry me. I didnít see that that was a deep problem.

Kuhn:

That obviously was a deep problem, but it was not the sort of problem that emerges in this paper, or that gives this paper its form, nor is it the sort of paper that leads from this to the hydrogen and helium spectrum paper and on into the spin. So if George was asking questions that were formative with respect to the direction in which this work goes, they would have been more about inconsistencies in the quantum theory.

Goudsmit:

No. Maybe it is that by his asking, I was able to clarify my thought, and I had still one dominating influence. I was able to brush these things aside and see only the positive contributions in ideas of spin and these vectors. That must have had something to do with that rumor — I donít know if it is a rumor that George wanted to withdraw that one time.

Kuhn:

The spin paper.

Goudsmit:

Ja!

Kuhn:

Well, weíll come back to this with him this afternoon. He says you both did. He said after Lorentz —.

Goudsmit:

Because I never saw the Lorentz business. I probably was in Amsterdam at that time. I didnít understand those things

Kuhn:

He tells the story, that when Lorentz came back, he showed that the electron couldnít have these properties because of the radius that would be necessary. The two of you went to — I think, he said, the two of you, Iím not sure —

Goudsmit:

Maybe Ehrenfest and he.

Kuhn:

The two of you went to Ehrenfest and said Lorentz has showed that itís nonsense. Weíd better not publish the paper.

Goudsmit:

I donít believe it!

Kuhn:

And Ehrenfest said, ďI sent it off already.Ē And then he said, ďYouíre young, you can afford a foolish mistake.Ē

Goudsmit:

To me he said, ďYou have no reputation yet, you have nothing to lose.Ē Thatí s what I remember, but I do not remember having wanted to withdraw it, because to me it wasnít so important. To me it was just another rule to understand and explain the complex system of spectra. That is all it meant to me. I never really understood it in its importance at that time. It wasnít in my nature; definitely not! It was just one of these marvelous rules; it was no more important than that phone call I got from Dieke two years earlier, saying ďIf you put these lines in that order instead of in that order, it suddenly fits. All the formulas and begins to make sense.Ē I personally never had the feeling that it was more than that.

Kuhn:

This is a little odd, in the sense that if you look at both this Review paper and the paper on the spectrum of hydrogen and helium, in both of these; I think Ö Iíd have to study it more carefully to be sure youíve got a scheme using Pauli and the electron quantum numbers which enables you to do everything qualitative that you do with spin; the advantage of spin, the remaining problem that spin is that by talking about the way magnetic interactions will behave, you can then get some quantitative results including z^4 and so on. But in that sense, for the analysis of spectra, in terms of what rules give you, youíve got it already without the spin.

Goudsmit:

No, no, you do not. Because you have to asume that duplicity, every time you add an electron something suddenly doubles-up.

Kuhn:

Youíre quite right, even at that qualitative level it would.

Goudsmit:

And Iím also so convinced Ö I canít prove it, except with my hypnotist perhaps, that it was George who said it is spin, because when I convinced him that the electron had four quantum numbers, and when I did it my way, as I, had published it in the Zeitschrift fuer Phrsik, not the Pauli way, he said. but that means it has four degrees of freedom, something like a spin.Ē

Kuhn:

Which do you mean by ďyour wayĒ as you published it in the Zeitschrift fuer Physik?

Goudsmit:

There is one paper — and here I. am just like all the other fellows — for which I never get any credit, and that is- a paper in l925, Zeitschrift fuer Physik, where I point out that if you apply the Pauli principle, not with the quantum numbers Pauli used, but with the quantum numbers which are now used, that then everything becomes so much simpler — that you get the multiplets out the proper way at once. And I give two or three examples.

Kuhn:

Ja. This is the paper ďUber die Komplexstruktur der Spektren,Ē [Zeits. f. Phys.32 (1925) 794-798].

Goudsmit:

Ja. And then it was, of course, immediately adopted by everybody, and poor Hund wrote a 1ong paper, but he gets credit for having analyzed all the spectra. But he had written this just before he saw mine. And so thatís still done in a most complicated way. The results are, of course, the same, but he has to use all kinds of sum rules, in going from a strong field to a weak field.

I was the first to say, if you put all the spins together, you get a total spin and if you put all the orbits together, you get a total orbit. But I didnít call it spin or orbits. So that was always the paper I thought a real contribution, in my eyes, being a spectroscopist and analyst of spectra. So I told. George about it.

Kuhn:

This was presumably done before George got back from Rome?

Goudsmit:

Ja, ja. It was published in May.

Kuhn:

It was submitted in May actually.

Goudsmit:

Ja, and also published I think — I donít know for sure. Anyway it was just about the time when he came. And it was that paper I had written Kronig about. Iíd sent the ideas and maybe even a copy to Copenhagen. And Iím still surprised that Kronig didnít consider this as an additional confirmation of his spin idea. But he didnít, because the answers I got from him were about entirely different things, as if he was not interested at all.

Kuhn:

At what point in this work over the paper, what role may George have had with respect to this — not yet spin, but this next big step with the hydrogen and helium spectrum, the insistence on treating hydrogen and helium the same way the alkalis are treated, which is an old idea of yours; but now in a different form because youíve got more stuff ? Whatís interesting here is that the electron quantum numbers and so forth donít show up, what this takes off from is this paper of Wentzel.

Goudsmit:

The paper of Wentzel was one I, of course, didnít understand, but which intrigued George no end. I still donít quite understand it. People were plagued by these half-quantum numbers which showed up every so often; Schroedinger had shown it already. Wentzel had invented some new rules, where you average between whole quantum numbers in the hope that that would explain the half-quantum numbers. And really Ė

Kuhn:

Now Heisenberg was doing a lot of that too!

Goudsmit:

Also, ja. But Wentzel seemed to have formalized it, and probably was all wrong; we know now itís all wrong, but it must have fascinated George. I did not understand the paper, because in retrospect you must admit that this hydrogen-helium thing we could have done without the Wentzel paper. But hanging it on to the Wentzel paper and saying this leads to the conclusion that if this is true for hydrogen where one had no trouble with half-quantum numbers, everything fitted well —. If this was really fundamental, and if Wentzel is right for the complex spectra, he also ought to be right for hydrogen. If that is so, then you come to such-and-such conclusions. I havenít read Slaterís paper in a long time, and I donít know what he based it on.

Kuhn:

My recollection is that heís got the same idea, that they ought to be the same. He doesnít have the Wentzel paper, Iím quite sure, nor does he have this thing that youíve, got. Iím quite sure heís got nothing about the lines that do and donít show up, which makes this, I think, particularly impressive.

Goudsmit:

Of course, that impressed me again, because I knew these lines so well. That I learned from Paschen.

Kuhn:

George says, as he learned this stuff from you, that he thinks his contribution to this was that from the very beginning he said it is absurd to treat hydrogen one way and the alkalis another. Does that fit with your recollection?

Goudsmit:

Quite possible. I donít remember it anymore. But itís only in retrospect that I see that this idea shows up already in my older papers. But I do not know whether that is an accident. It certainly was not consciously so, but that it is hindsight. You can say that now that he has done this, we can trace it back to his very first paper in 1921; that may be purely coincidental, and it may simply mean that I always stuck to the interpretation o spectra. I was fascinated by that.

Kuhn:

You have no real recollection? — we touched on it yesterday — of an interim period in which the development of the magnetic interaction theory bothers you about your older relativistic interpretation?

Goudsmit:

No. I believed in that of course, from the very beginning, and then believed in it even more after Millikan and Lande independently came to the same result. No, no recollection. I never found difficulties in that because I didnít know enough, and Iíve often explained the success we had with the spin as caused by our ignorance. I knew the analysis of spectra and everything like that very well.

Then my previous papers — and what I told George, if you only look from the point of view of the analysis of spectra, and the formulas, then itís a perfect fit, itís almost a natural conclusion and you donít understand why I didnít do it already in May Ď25 before George came, or why Kronig didnít do it after Iíd sent him that paper. Itís a natural extension. The other people were so much more learnedÖ They could compute the fine, structure, which we did not. But that was all they could do. They probably did not understand the Zeeman effect and the sum rules and other things. They may understand it —

Kuhn:

Heisenberg and Pauli surely did Ė

Goudsmit:

Pauli did, ja. And thatís why itís such a surprise. But he also knew the other thing. Iím not so sure that he was as well versed in all this vector business as Lande was.

Kuhn:

By then, I think, he was, but not in the spectroscopic data.

Goudsmit:

Ja. He didnít know the ďforbiddenĒ lines and things like that. And they also knew how to compute the self-energy of an electron which was a concept which I didnít understand, even though we put a footnote in there. And I just now for the first time see a letter [of 1925] from Fermi, where he writes us about that, how do you get aroundÖ? Iíve never seen that before. We simply didní t know, But that was the thing Lorťntz knew, and thatís the first thing he thought about. To me that ďforbidden lineĒ in hydrogen was far more important. So you see how ignorance helps.

Kuhn:

On this paper on the hydrogen and helium, do you remember other peopleís reactions to this? It comes so shortly before spin.

Goudsmit:

No. I donít remember at all, and I donít know why we published it in Dutch —- havenít the faintest ideaÖ

Kuhn:

It raises a question I had meant to ask you before —- maybe weíll interrupt this, but weíve got to come back right to this point —- Iím puzzled about why you published which papers where. Some of these things come out in Nature and Naturwissenschaften. Other things come out, things which in some wars look to be a good deal more important, come out just in Physica in Dutch and donít come out anywhere.

Goudsmit:

My explanation at the moment is that when you do a thing like that, you donít know how important it is, or whether it really is important. Thatís the only explanation I have. Itís my fault, ja.

Kuhn:

What about this business of publishing in both Nature and Naturwissenschaften?

Goudsmit:

Naturwissenschaften in those days was the more natural thing for a pre1iminary note. Also Ehrenfest and Hollanders were oriented at that time towards German physics. Nature I believe, was under the influence of Bohr, who was more English-oriented. Donít forget the Nature —.

Kuhn:

But your early multiplet analysis —-

Goudsmit:

The early stuff. Early multiplet, because of Fowler. That was something along the lines of Fowler in England, so I asked Lorentz, would you p1ease send that to Nature?

Kuhn:

And youíd send it to both?

Goudsmit:

I donít know, did I send —? That was also a bad habit. Some of my papers were published three or four times, which was very bad.

Kuhn:

Some of the stuff that you and Zeeman do goes to both.

Goudsmit:

Yes, Zeeman always wanted to have it everywhere.

Kuhn:

And —. Yes, NO. 9 I would say this is, ďThe Magnetic Resolution of the Scandium Lines.Ē

Goudsmit:

And 8 are practically the same, 8 and 9 are practically the same; 10 and 11 are the same; then 11 and 13 are practically the same.

Kuhn:

One of the curious things is that 8 and 9 are practically the sane; 9 goes to - Nature, and is by Goudsmit and Zeeman, 8 goes to Naturwissenachaften and is by Goudsmit.

Goudsmit:

Ja [laughter] I have no idea. I can only tell you that Zeeman had not a damn thing to do with it. That either it was one of these Ehrenfest ideas, take your hat off for a policeman, you know, like Charlie Chaplin. You have him on the job in Amsterdam, he didnít say so, but probably in his subconscious mind, he said, ďWeíd better publish with ZeemanĒ or something like that.-óI donít remember the details. Zeeman was more English-Oriented. Are these two papers identical?

Kuhn:

No. Thereís a little more detail in one or the other, but from our present point of view theyíre identical. Theyíre reports on the same piece of work.

Goudsmit:

Ja, and one is August 9, the other August 20. Do you know whether one is signed Leiden and the other Amsterdam, or are they both signed Amsterdam? I donít know, I have no recollection. Thatís politics, thatís pure politics, has nothing to do with physics.

Kuhn:

No, theyíre both signed Amsterdam.

Goudsmit:

Ja, thatís where the work was done. Iím sure itís pure politics.

Kuhn:

Well, back to the spin. Again in this paper on hydrogen and helium youíve got this terrible important suggestion that maybe the ortho series is triplets.

Goudsmit:

Were we the first to do that or not?

Kuhn:

Iím not sure. I donít know of an earlier suggestion to this effect; there may have been one. I rather think itís in the Slater paper also, and Iím not sure which of those two papers comes first.

Goudsmit:

I think we came a little earlier, but Dutch doesnít count. Might as well have been in Chinese.

Kuhn:

In any case, did you try to get somebody did you write and ask Back to try to measure that, or anybody?

Goudsmit:

No I probably tried to go through the literature and see whether there were any cases of it. And I didnít. I donít know why, either. Donít forget it was summer. I might have asked if the spin hadnít come along, and that took all our attention, and we were asked about that and had to go to Copenhagen.

Otherwise I am sure I would have done something else, I would have written to Back or to somebody, Zeeman. The Zeeman lab. Was closed at that time; this is summer work. I probably would have, But other things took our time and attention, so I never followed it up, but then, of course —-

Kuhn:

Do you know who first did that resolution and how long it took?

Goudsmit:

No.

Kuhn:

Of course, presumably with this as a lead, it would have been very easy to set up apparatus to get that much further resolution on these.

Goudsmit:

I donít know when it was. Then I got off in other directions. There was a paper which made me very mad, from a priority and credit point of view, by Sommerfeld and Unsoeld. They had apparently come to the same conclusion later. They gave us a footnote, I donít know whether they gave Slater a footnote on the first page, and from then on that paper of course was typically German. Thatís how they get all the credit, the fatter the paper, the more credit you get. Thereís nothing new in that paper. Then they talk about ďunserer theorie.Ē You see, from then on itís ďunserer theorie.Ē Typical nostrification paper that was. But I didnít like it at all.

Kuhn:

Do you have any recollection of how the idea of spin itself gets into this, and when it comes into it?

Goudsmit:

I only remember the discussion with George, where I tried to explain the Pauli principle to him, and point out that one quantum number is always a half, that there are four quantum numbers, that he talked about four degrees of freedom, which to me was even not a clear idea. You must honestly believe me, I am not modest, that I am a numerologist in some respects, and so four degrees of freedom, ďSure, four degrees of freedom, what the hell,Ē you see If I donít even know what three degrees of freedom meant precisely, so four, why not, like a spin, you see, now that meant something to me.

And then it clicked again, like the phone call from Dieke. And Iím 99 percent sure that it was Uhlenbeck who mentioned that this must mean the spin, and that I picked it up in the discussion, and said, ja, now everything falls in line, how that whole vector scheme becomes so much simpler, and how you put all those spins together. Iím 90 percent sure that thatís the way it happened. But for instance we donít know exactly when it was.

Kuhn:

Whatís your impression? The note to Naturwissenschaften is dated the 17th of October. Now how long before that do you suppose youíd had the idea?

Goudsmit:

Probably about a month. Let me see, I have some letters from Uhlenbeck at that time. That was 1925. Now letís see. [Checks through letters] No precise date, damn it, this is very rare for this is the fall of 1925, but I donít know when. It must be early fall. I wrote that down. From Ehrenfst. [See microfilm of Goudsmit Correspondence] Probably sent to me to Amsterdam. That is about the spin, you see. I wish I understood this. [Reads] Ja. See, this is an important note. And I wish the precise date. It says here, ďIn connection with the discussion of your, very nice idea, Uhlenbeck has made the remark that your formula,Ē— that is both of us, I think he means—ďin connection with the Hund formula, that there is a contradiction with the sums, which are not fulfilled. If there is a misunderstanding here, and if it points at a defect in your formulasóď And then there is a note here on the side, ďUhlenbeck has just noticed that he has made a mistake. There are probably only two small deviations. But I would like to wait until tomorrow anyway.Ē

The letter goes on here, ďI request therefore that you wait with sending off the manuscript to Naturwissenschaften until tomorrow, until Wednesday evening.Ē See, this must have been sent to Amsterdam; Wednesday evenings I came from Amsterdam to Leiden. [continues reading] ďI ask you therefore to derive the permanence of the g-sum from your idea of the electron spin: A,Ē in the case of the alkalis; B, in the case of calcium; C, in complicated cases.Ē Now for me that was duck soup. ďPerhaps there is . modification of your g1 g2, g sum synthesisĒ I donít know —- ďI have still two or three small stylistic remarks about your letter to Naturwissenschaften. Best regards, o tentaminando.Ē —-you see, I hadnít done my exams yet.

He rubs that in again —- ďyours, P. Ehrenfest.Ē So it seems that I wrote that, and that he just said, ďDonít send it off until Wednesday because the g-sum doesnít fit,Ē and then ďOh no, Uhlenbeck has made a mistake!Ē You see; I think this is significant!

Kuhn:

We must get at that letter again this afternoon with George, because this also contradicts his recollection that the paper had already gone off because Ehrenfest had, sent it in, a story which he has told often.

Goudsmit:

Ja. And here is a letter, a note from Uhlenbeck to me about things I didnít understand, 7 October 1925. [Reads] ďThanks for your letter. I have talked it over with Ehrenfest and as far as the calculation of the magnetic moment of a rotating electron, formulas from AbrahamĒ —- of course I never understood Abraham —- magnetic moments are here, you see —- ďthen if all the charges are on the surface, then you get a factor two,Ē or something of that idea. If this is right I donít know.

You see thatís not a Thomas two, thatís just a g of 2. ďIf the charge is distributed over space itís quite different,Ē and so on. ďI hope to be able to talk it over with you Saturday afternoon when Iíll come to your home around 2:30 because I come to The Hague only that late. Best regards, G. Uhlenbeck.Ē Thatís all I have of that period.

Then I got the letters from the outside, including a note from Fermi. Here is a copy of a letter which Bohr wrote to Ehrenfest, and he showed me, and which my ex-wife copied. I guess she wanted to keep it. And then on the back there are formulas—- the vector scheme and so on. Thatís probably a part of the summary article, it looks like it.

Kuhn:

That may be a part of the Phil.Mag. piece.

Goudsmit:

No, I think itís Physica;this is Physica. On the back is a letter Bohr wrote to Ehrenfest, the nice time he has had in Holland talking about the spin. I also have a copy of —-. Throw this out, thatís a bill —- oh, keep the stub —- bill for a book.

Kuhn:

What book?

Goudsmit:

Egyptology, Iím sure. Ja Egyptian dictionary.

Kuhn:

When you defended your Egyptology thesis, did anybody ask you

Goudsmit:

Ja, ja, ja. They did at that time.

Kuhn:

Well now look: the Naturwissenschaften note came out and what happened?

Goudsmit:

Then we got letters, right away! From Heisenberg — and I notice, I saw a moment ago a letter from Fermi, canít find it anymore. Probably misfiled, because it had no date on it.

Kuhn:

Theyíre in there, I found them yesterday.

Goudsmit:

And the Heisenberg letters are in here; there are copies of those too, these are the originals. And Heisenberg said, ďWhat did you do with that factor 2?Ē And of course we didnít know what to do about it, and it was at that time that of course George had to take over, because that was the kind of thing he knew how to do. Then we were lucky that Einstein and Bohr and everybody came to Leiden. I think it is they who more or less solved it. And there is that letter from Bohr—-.

Kuhn:

Now when you say Ďsolved ití what do you mean? Because they didnít get the factor of two.

Goudsmit:

No, but they knew in which direction to look —- Bohr. Thatís the impression I have. There is that letter from Bohr to Kronig, in which he says —- have you got that letter?

Kuhn:

Ja, ja. But I donít think he says in what direction to look.

Goudsmit:

Ja, he says it must be a relativistic effect in there.

Kuhn:

Gee, I donít think so. G; Ja, Iím pretty sure. I have it with me. I thought I had it among this junk because I thought it was relevant. Ja! Let me see what it says. This is [26], March 1926. ďDear Kronig,Ē itís in English ďwe have often missed youĒ was it sent to America? Probably ďI have been deeply interested in the spectral problems which you have given, so much thought. The idea of the spinning electron, and the insight it offers in all the difficulties with which we have been struggling lately has been a source of very great pleasure to me. From a discussion after your beautiful lecture on the Zeeman effect, you know how unhappy and desperate I felt over the state of things.

Certainly I had heard then that in a letter to Naturwissenschaften Uhlenbeck and Goudsmit had proposed to provide a basis for a new quantum number by ascribing a vector to the electron itself, but not understanding how this vector could materially influence the motion of the electron to which it was attached, I had thought so little about this idea that I even quite forgot to question you about this possibility in the discussion that evening. When I came to Leiden however, to the Lorentz festival, Einstein asked the very first moment I saw him, what I believed about the spinning electron.

Upon my question about the cause of the necessary mutual coupling between the spin axis and the orbital motion, he explained that this coupling was an immediate consequence of the theory of relativity. This remark has acted as a complete revelation to me, and Iíve never since faltered in my conviction that we at least were at the end of our sorrows.Ē

Kuhn:

But thatís not the same thing. Thatís not the factor of two.

Goudsmit:

No, but thatís essentially it.

Kuhn:

No. That is, youíre right that theyíre both relativistic effects, but the Einstein remark is apparently just, if you donít understand the interaction, just remember that itís the same thing as if the nucleus were moving around the electron.

Goudsmit:

That has nothing to do with relativity. That is Biot-Savart.

Kuhn:

Ja, or itís the Lorentz invariance of the Maxwell equations. But I think thatís all thatís about.

Goudsmit:

Ja, but this is what convinced Bohr that one had to look at it from that point of view, Iím sure of that. Otherwise he wouldnít have set Thomas on the job of—-.

Kuhn:

Thomasí story, which I think is probably true, is that Bohr didnít. Thomasí story is quite interesting and very plausible. Thomas says ďThey were talking about this whole subject, and I just said, ĎLook, this ought to be done relativistically.í And they said, yes, I suppose so, but we know the order of magnitude of relativistic effects, it isnít going to make any difference. A few percent.Ē

And he thought, well, he, after all, had done all this relativity stuff before —- ďthat it would be nice to do it relativistiial1y, so he sat down and did it.Ē I think he was almost as surprised as anybody else that the factor two came out. He just thought it should be, Since you knew relativity, it should be done relativistically.

Goudsmit:

ďI have since had quite a difficult time in trying to persuade Pauli and Heisenberg, who were so deep in the spell of the magical duality that they were most unwilling to greet any outway of the sort. It has even been very difficult to persuade Pauli to believe in the beautiful contribution of Thomas which made it possible to obtain full quantitative agreement. And first in the very last days he has completely surrendered.Ē ďUnder this situationĒ —- then he goes on.

Kuhn:

Did, you have arguments with either Heisenberg or Pauli? Do you remember Bohrís initial lack of conviction?

Goudsmit:

No, no. I donít remember that at all.

Kuhn:

What about Heisenberg and Pauli?

Goudsmit:

I didnít talk to Heisenberg at that time. I talked to Pauli on my way back to Copenhagen, and tried to convince him of the Thomas two in the sense that I understood it. He was not convinced, he ridiculed me.

Kuhn:

Why was he not convinced? What did he say about the Thomas two?

Goudsmit:

I didnít know enough relativity to be able to convince him. He probably made the same remarks Einstein made to me, who also wasnít convinced. Whom did I tell that the other day to? Iíll have to use the blackboard for this. [Goes to blackboard.] Iíd give this elementary derivation, which the highbrow say is wrong, but which I have learned from somebody in Copenhagen.

I donít know whom; here is the circular orbit, here is the electron, here is a co-ordinate system with respect to the nucleus and x and y. An electron is going in that direction. This figure is in the plane of the paper too, by the way. Now a moment later the electron is here, with the co-ordinate system in the orbit. You see, Iím moving that way. Now you make a Lorentz transformation.

And Einstein said, ďNo! a moment later the electron is there! You cannot apply this to the accelerated motionĒ that kind of thing, and he never let me go on. That it gave the right result didnít impress him, he said, ďno, the electron is there.Ē And I didnít know enough highbrow physics to put it into a language so that Einstein would be convinced. And I think Pauli also didnít understand my elementary —-.

Kuhn:

This was after the Thomas paper was published?

Goudsmit:

Was it after the Thomas? No, no, it wasnít published yet, no. You see, it wasnít published yet.

Kuhn:

Ah! Were you in Copenhagen when Thomas —-?

Goudsmit:

I was in Copenhagen when Thomas had made the discovery.

Kuhn:

How do you react to Thomasí story? That it was his insistence that it ought to be done relativistically, that other people said sure that would be nice but what the devil, we know that correction, we know —-.

Goudsmit:

See, Iím again the wrong man because I didnít understand it. Thomas was already in Copenhagen when I came, and working on it. During that January, while I was there, he had gotten the factor two. And I of course didnít understand it, I still didnít understand a word of what he says. And then it was either Kramers or somebody else who translated it for me into that elementary language which I tried to put on the blackboard: that much—I could understand. But Iím told that much is wrong.

Kuhn:

Was Pauli convinced and was Einstein convinced when the Thomas paper came out?

Goudsmit:

Pauli was convinced only in March, and there must be in here somewhere a postcard from Pauli in which he more or less apologizes. I donít know where it is, I canít find it in which he admits that he now is convinced that the factor two is right.

Kuhn:

Thomas was already at work on this before you got there?

Goudsmit:

Already before I got there, ja. Why he was there you must know from Thomas, but I donít know.

Kuhn:

This was fairly usual for Fowlerís students.

Goudsmit:

Ja. Somewhere there must be the postcard of Pauli. I donít see it right now; it doesnít matter. That jibes also with the letter of Bohr to Kronig. Just about this time — this is dated March — Pauli was convinced. But I saw Pauli in February, before the Thomas paper was published.

Kuhn:

Thereís the diagram, on page 7 in a reprint of the summary spin paper. So it couldní t convince anybody. I think thatís the whole story.

Kuhn:

What can you tell me about Kronig in this?

Goudsmit:

Nothing than what I heard.

Kuhn:

Did he ever write you himself about it?

Goudsmit:

He never wrote me himself about it. I can only guess that he probably thought we knew about his work, or that he thought Heisenberg had told us about him. But the Heisenberg letters, which I sent him later, two years ago whenever it was, they donít mention anybody. And in Bohrís letter to Kronig, which is much older, Bohr says Heisenberg says he had heard it from somebody but he didnít know from whom, probably indirectly. Moreover, the natural reaction of Kronig was, of course, that it was wrong. So Kronig published two-papers proving that it was wrong.

He was convinced that it was wrong. If you read Van der Waerdení s paper, then Kronig was right — it was wrong. Very simple! Though he deserves great credit for believing that it was wrong. That probably is correct, I mean, from a mathematical point of view, of course. From a logical point of view of course physics is not a logical science, it goes by evolutions as you well know. Then of course itís the same with Bichowsky and Urey. They deserve at least as much credit as Kronig.

They also had the idea. They didnít publish it until later, and also, in their note, they point out that it canít be right because it doesnít give the right fine structure. You see how lucky George and I were that we didnít know how to compute it. Itís pure luck. If we had known, we might have had hesitations. But we didnít. I realized it was something difficult, because I couldní t do it, and I thought that all the other things fit so well, that if somebody who can compute it correctly does it, it probably will come out correctly.

The main thing I remember having done when Bohr and Einstein were at Leiden, was that I clearly saw how that hydrogen spectrum, which we had published in Physica, fitted in with the spin. And I still remember the diagram on the blackboard in Ehrenfestí s study, and explaining it in detail to Einstein and to Bohr, the new interpretation of hydrogen. Because Bohr hadnít seen it, it was published in Dutch.

Kuhn:

Do you think you had not really seen that relation at all, or just not seen it clearly, or what, at the time of the Naturwissenschaften note?

Goudsmit:

Not seen it clearly. I feel that I had not seen it clearly, because I would have mentioned it in the Naturwissenschaften. But I had not seen it clearly.

Kuhn:

I forget, just when was that assembly in Leiden?

Goudsmit:

November or December.

Kuhn:

To what extent was the piece to Nature a direct product of that assembly? See, thatís quite different. The hydrogen is in there, you use more arguments, show more ways in-which it fits.

Goudsmit:

I believe that that came out of these discussions with Bohr and Einstein, the necessity of getting all the arguments in.

Kuhn:

was it Bohr who urged sending a new announcement to Nature?

Goudsmit:

No, I think it was Bohr who finally —- it was written in Copenhagen really, I think; what is the date on it?

Kuhn:

The dateís December.

Goudsmit:

Was it really sent off in December? [checks] Copenhagen in January, you see —- ja, we wrote it in December, and I took it to Copenhagen for Bohrís O.K., and in the typical Bohr way, I remember, he changed a little word here and this little word there, suggested something here and there, then wrote this appendix. And he kept it for weeks. Then we took a taxi to the post office and had to send it on special delivery.

Kuhn:

But notice that although Bohrís letter is dated January, your letter is dated December.

Goudsmit:

December. That means it was written before I went to Copenhagen essentially.

Kuhn:

Would it be dated December if you hadnít mailed it until January?

Goudsmit:

Ja. Because Iím pretty sure that they were both mailed together.

Kuhn:

Is it possible that it was the proofs that were returned in January and that Bohr wrote his letter then?

Goudsmit:

No, because I remember discussing it with Bohr.

Kuhn:

It surprises me that Nature would print a letter with a December date line that they had not received until much later.

Goudsmit:

These old date lines were not received dates. They were really when it was written. In those days people didnít care about those details very much.

Kuhn:

Youíve got eingegangen dates in the Zeitschrift.

Goudsmit:

That is true. This is not an eingegangen date, because the eingegangen date was January, and I donít know when in January, late in January. I was really a little mad at Bohr for waiting so long. But Iím pretty sure that this came out of those discussions at Leiden when it was necessary especially for me to set all the little points straight, and this I was always very proud of. See, thatís just the same picture as was in Physica.

Kuhn:

The same picture thatís still in most books on spectra. -.

Goudsmit:

Ja, and here we knew about Slater.

Kuhn:

At this point partial anticipations of some of these things begin to come up, and you begin to know about them. For example, you begin to refer to the spin Idea as the ďCompton hypothesis:Ē

Goudsmit:

That was Ehrenfest. Iíve never read it yet. It was typically Ehrenfest again.

Kuhn:

He canít have known earlier about the Compton effect. How do you suppose he heard about it?

Goudsmit:

I donít know, maybe somebody called it to his attention. But it was again: ďDonít forget to quote some important people.Ē

Kuhn:

Do you remember how the Slater thing was called to your attention?

Goudsmit:

That we probably read. Because we looked at the National Academy. I think we were aware of that ourselves. Then you hot ice that note about de Haas. Ja, thatís all.

Kuhn:

Do you know anything more about the de Haasí work? No, nothing. There were more things which Ehrenfest made us quote, I think, which were: strange to me. Canít find it at the moment. But you see here are these formulas which came in that letter from Uhlenbeck which I still have.

Kuhn:

Right, in the middle of this, or between the two papers, you start taking up the coupling problem.

Goudsmit:

Ja. Thatís what we had really been working on. We had probably started to write that before. That came also out of the summer discussions, as a result of the summary article. See, the spin was really the thing which threw us off because of the interest of Einstein and Bohr. I must honestly admit that I at the moment didnít know that was any more important than just the coupling business. The coupling thing we had really discussed and were writing an article about. I think in the coupling paper thereís also a footnote to the spin which clearly shows that it was written conceived before.

Kuhn:

Iím not clear myself where this whole business of various forms of coupling starts. It starts before this, some of it.

Goudsmit:

I have no reprint of it at the moment, so I donít know.

Kuhn:

But youíve done —-. Youíve got a paper, either No. 22 or No. 25, in which you begin to talk about other coupling schemes.

Goudsmit:

One didnít know that they were coupling schemes at that time, I believe. It wasnít clear what the meaning was of the hŲhere Stufe.

Kuhn:

At least you call it coupling in the title of No. 24, which is the one you do with George. [ďDie Kopplungsmoeglichkeiten der Quantenvectůren im Atom,Ē Zs. f. Phys. 35 (1926), 618-625] And I suspect that you call it coupling earlier,because you speak of uncoupling. Now this is Heilbronís translation of paper No. 22 —- [reads] ďIt may however happen that the force of the ion is insufficient to uncouple R and K.Ē Already in paper No. 22. Iím not sure, but I know that thereís been some talk about —-.

Goudsmit:

Do I quote other people in that one?

Kuhn:

If you do, it isnít noted here, and I think—-.

Goudsmit:

Itís a short note, it seems, itís a letter to Naturwissenchaften.

Kuhn:

I think it would be.

Goudsmit:

October, Ď25. Ah, but that explains it, of course! That they were simultaneous. That must have been written simultaneously, you see, that whole— we did a lot of—-.

Kuhn:

I ask this because all of these things come at once and I was wondering how theyíre related to eachother.

Goudsmit:

I donít know why this letter to Naturwissenschaften was not with George T. Uhlenbeck, letís put it that way. Itís probably again selfishness on my part.

Kuhn:

I wouldnít think necessarily.

Goudsmit:

But anyway, you see paper No. 22 and paper No. 21, were probably thought about during the same time. That paper No. 21, was received so late in November was only because we had been too busy with the spin. But it was probably written, or at least a draft must have been written, very much earlier, because that was the main result, so to say, of our collaboration during that summer. It was a natural result, writing this article about the present state of spectra, this was a natural result to write about the doublet coupling schemes.

Kuhn:

In your thesis, you make a remark that Heisenberg in Ď25 first points to the possibility of other couplings.

Goudsmit:

Ja, I think that is right. I think he was really the first one to do it.

Kuhn:

You donít remember where the coupling thing really came in here? You think it was just another by-product of the summary article?

Goudsmit:

It was a more natural by-product of the summary article, the kind of thing I would have expected. No great step, but good systematization. It didnít originate with us, as you point out, it was already Heisenberg who had done it first, and I had been in this business. That was the kind of thing which was in the air, and it was natural, after studying that summary article, to write something like that.

Kuhn:

The generalization of the coupling schemes. You were probably more acutely aware than Heisenberg of how badly the other possibilities were needed in order to handle the spectroscopic results. I donít really remember, I think Iíve looked at it; but I donít remember that Heisenberg paper at all.

Goudsmit:

I donít remember it either, I donít remember it at all; but thereís no doubt that he was the first to point out.

Kuhn:

I hate to leave spin entirely behind us, though Ií m not clear that Iíve got anything else to ask.Iíll ask one sort of question. I take it that the conversion of Bohr and Einstein, this support came so quickly that there was never any problem about their resistance? I never noticed that it wasnít a conversion. I was just telling them, and they asked questions which made it necessary to put my thoughts straight, and to bring in everything I ever knew about spectra.

Kuhn:

Now what about the Heisenberg-Pauli hold-out? Did this other you?

Goudsmit:

No. First of all, I did not learn about it; I was only told about it later on, and then I noticed it when I tried to convince Pauli on my way back from Copenhagen. But I was such a beginner. That meeting with Pauli was so strange.

Kuhn:

Had you known him before?

Goudsmit:

No. I never met him before. It was a very disagreeable experience. Later on we became good friends. That first experience was a very disagreeable experience. I have told many anecdotes about it....

Kuhn:

Tell a few now.

Goudsmit:

First of all, he took me out to lunch with people, and then told jokes at my expense and then he had the gallery guffawing, which I didnít like. I didnít always understand the jokes, I didnít understand all the German at that moment. Then he was very much upset about one thing: he looked at me. I had been a poor student, but having been invited to Copenhagen by Bohr, my uncle had said, ďYou canít go like this.Ē And he gave me money to buy a new coat. So I looked very neat when I came to Pauli, and he didnít like that. He said to me, ďWhat a beautiful coat you have there, what a beautiful coat!ĒAnd he said it in a way that I felt very, uncomfortable.

But then there was one face- saving feature. He looked suddenly up, and said, ďOh, but youí re wearing a theoretical physics hat.Ē [Laughter] I had one of these very old hats, you could make soup out of if you put it in a kettle with water, quite greasy. So that pleased him again. But the whole atmosphere was like that. And I never was able to convince him of anything. It was only a short stop there.

I spent probably two hours with him, thatís all, then I more or less fled, went to my hotel, and the next day got on the train and went on. But later on Pauli and I hit it off quite well whenever we met. He was still mad at me for not quoting him on hyperfine structure, that is sure. But he never took it out on me openly, never was mean to me. I traveled with him, all the way to California in 1931, I think, and he was a very good companion.

Kuhn:

Was that making jokes at other peopleís expense, and having a sharp tongue typical of Pauli, as everybody says? Iíve never heard of a case in which he led a group in laughing at somebody. Was that also typical?

Goudsmit:

That I donít know; but it was very painful to me, I wasnít used to that. I donít know who was in the group any more at that time.

Kuhn:

Itís the sort of thing one would be likely to erase. [Interruption, recorder turned off and on] I had just said you go right from this to byperfine structure.

Goudsmit:

Ja, I went right from that.

Kuhn:

Thereís a sense in which Iím puzzled that went quite so fast, because particularly with both spin now and the coupling schemes, there was a lot you could have done on multiplets that you couldnít do before. But you jumped right past multiples.

Goudsmit:

It was probably again accidental. This is what happened. I probably was overworked. Anyway, I was in a rather bad state physically at one time. In fact, Ehrenfest worried, and Mrs. Ehrenfest took me to their doctor to find out whether there was something wrong with me.

Kuhn:

What did happen? You do the summary piece on spin, and that is just about all until this piece from, Tuebingen in October of 1926. That, for you, was a slow down.

Goudsmit:

I probably had to work — didnít I have to work for exams? I probably did my doctoral — not the Ph.D., but the doctoral exam at that time. I must find the diploma for the date. That may have had something to do with it.

Kuhn:

No, you passed the doctorate. At least in your thesis you say you passed the doctoral exams in 1925.

Goudsmit:

Did I? I canít believe it. Maybe I did it wrong. Where do I say that?

Kuhn:

In the biographical account at the front.

Goudsmit:

December í25! Gee! Also everything! Oh ja, thatís why I could talk with Einstein. He was there at the time, the walk with Einstein was then. Thatís why Ehrenfest still writes to me o tentaminando, or whatever it is in Latin you see, in October. I donít know, what did I do? I think I wasnít well. Not unlikely at that period. I was somewhat overworked, and over —-I donít know. December —- I went to Copenhagen. No, I was all right then, I was still all right then. I went to Copenhagen and was fine. I worked in Copenhagen with Bohr.

Kuhn:

On what?

Goudsmit:

That was a complete fiasco. Because Bohr knew what the next problem was, and I did not understand it, it was not up, my alley. Bohr right away understood in looking at the helium spectrum that you turn over one of the two spins, and suddenly the energy is entirely different. And every day we got together in the hope that his talking to me would put some idea in my mind which would solve that problem. It didnít. So after about five or six weeks he gave me a first class ticket back to Holland on a sleeper.

I had never been in a sleeper, that was quite something for me, with the admonition that I had to get off at Hamburg and tell Pauli about that, about the Thomas factor two. But that did not come out. And I invented all kinds of mysterious magnetic interactions, and I have a letter from Bohr in which he tells me that Iím all wet.

But I still wrote him about it later. I kept on thinking about ití You know of covrse he then called Heisenberg who indeed found the solutfrn —- the anti-symmetric eigenfunctions and so on. That was beyond me. I would never have been able to guess at anything like it.

Kuhn:

When you and Bohr worked on this, in the spring of í26 —-

Goudsmit:

January Ď26.

Kuhn:

Was he at all using the new quantum theory, or was he —-.

Goudsmit:

If so, it went past me. Not yet, I think. He still tried to look in terms of the model, but he knew that —-. You see that I admired so. That was the next problem that one had to understand, and that would solve a lot of things, and I am so convinced that he must have inspired Heisenberg, and that Bohr must have done things like that many times, and doesnít get enough credit. Because the most important thing in some problems is to find the problem.

Kuhn:

But that was a very old problem, that was the one that lots of people had been breaking their heads on, the helium problem.

Goudsmit:

Not in that form.

Kuhn:

Not with the spin, but with the old quantum theory treatments, the helium problem.

Goudsmit:

There wasnít even a hint, but now there was a hint and now you could pin-point the difficulty. All you do is turn over one spin and suddenly the energy is enormously different. If you believe in the spin being just little magnets it doesnít work; then the singlet and the triplet stage should be very near together. So that was a very important problem. Bohr felt that this was not the same. The helium is always still a problem.

Kuhn:

Was he pretty convinced by the time you left that it couldnít be done?

Goudsmit:

No. He was convinced that I couldnít do it. He was not convinced that it couldnít be done because then he called Heisenberg, he thought thatís the man who can do it. I believe that Heisenberg followed me very soon when I had been kicked out. But Bohr had at the beginning confidence in me misplaced, and must have wasted a whole month just trying to talk with me every afternoon. That was marvelous as I remember it.

Kuhn:

We have been ignoring the fact that just at this time with all the excitement about spin, thereís of course all the-excitement about quantum mechanics.

Goudsmit:

It just began. Thereís one incident I remember, it was just before the spin, I believe. I ought to be able to check up on that. That Heisenberg passed through Leiden and told us about it. And I remember a strange remark by Ehrenfest that he was like Sir Isaac Newton, that not only had he invented a new mechanics, but also had to invent a new mathematics to go with it.

And then the story goes, and I do not know whether it has been recorded, but it has been told many times, that indeed Heisenberg thought so. He had been in England, I think, at that time, then he came back to Goettingen, and they said, donít you know that new mathematics you invented you learned from Hilbert? You must have been asleep in his class; thatís nothing but the rules of matrix multiplication.

Kuhn:

Do you know where that story comes from? You told me this once before, and I havenít been able to get any confirmation of it at all.

Goudsmit:

Doesnít George remember it? Was George at Leiden at the time when Heisenberg —- because I remember being together with Heisenberg.

Kuhn:

It may have been George, not you, who told me this story.

Goudsmit:

Ja, I told you that story. But I remember very well Heisenberg in Leiden at Ehrenfestís home. He may even have come to The Hague: with me, I have a vague recollection. He came to see me from time to time, probably was at my home, the home of my parents, having dinner there, and I remember taking him to the fireworks in (Stavene) at the seaside there.

It may have been that particular visit, but Iím not so sure, because he came to Holland during that period, off and on, and I used to take the foreigners around. I was the tourist guide. I must have inherited that from my grandfather. I had to take Lande around, I had to take Sommerfeld around, I had to take Heisenberg around, Pauli —.

Kuhn:

How did people feel about this idea of Heisenbergís?

Goudsmit:

At that moment, in the summer, that was before it was published. I have the impression that Ehrenfest was a little skeptical, but hopeful. The argumentation at the time was a little strange. It was that observable — and that didnít appeal too much to some of the people I heard talk about it. And in fact there was talk later on. So the argumentation was wrong, but the mechanics was right. Thatís what it amounted to.

Kuhn:

What about the Born-Jordan, and the Born-Heisenberg-Jordan pieces?

Goudsmit:

The only part I was interested in, and learned, a little about, was of course, again the spin part, the Pauli spin matrices. I still donít know it well —-.

Kuhn:

But those are much later. Those must come after the fall of í26.

Goudsmit:

I donít know. Donít forget I was only a spectroscopist, I am the wrong man to ask. The impression I got from the people I heard discuss it at Leiden was that it was an enormous amount of work, and that, it was surprising that it gave the right result, and before they really had time to digest it, Schroedinger came. And that came as a relief. I donít know how long it took to show that Schroedinger was mathematically equivalent to it.

Kuhn:

Not very long.

Goudsmit:

So then they pitied the Born-Jordan —-

Kuhn:

A couple of months.

Goudsmit:

They pitied them they said, look at all the work they have done, and how difficult it was. So that was the impression I had. They were surprised that it gave the right result, that it was so awfully difficult. They gave up the hope that they could ever learn that kind of thing. And Schroedinger came soon enough, before everybody had tried to learn matrix mechanics, that it came as a relief. But I never learned it: it was not my type of physics.

Kuhn:

When did you start learning Schroedinger?

Goudsmit:

Not until I came to the States, until I had to teach things like that; thatís the only way I learned it. No, I am an experimenter. Once I would experiment, spectroscopy, nothing else.

Kuhn:

Come back then to this question now. You say you think during parts of Ď26 you were not very well.

Goudsmit:

I believe that in parts of Ď26 I was not very well. Ja, that is right.

Kuhn:

Then what happened that took you back to Tuebingen?

Goudsmit:

I wanted to do some more work, and there was no job anyway. So Ehrenfest got me a fellowship. It was an exception, because I didnít have my Ph.D. yet. But he talked the International Education Board Rockefeller people into giving me a fellowship anyway to do some more work at Tuebingen. Again, Ehrenfest understood that I had to be near the experiments. I remember that my illness —-.

Kuhn:

You were there for the whole year were you?

Goudsmit:

No, not for the whole year. I went then —- I went early. Probably I left in September Ď26, I left for Tuebingen. I donít know the precise date. I may still have my old passport. I went early just because I hadnít been feeling well, and nervous, and thought,ďwell, Iíll go there, Iíll go away from home, maybe itís home surroundings.Ē I think in retrospect, knowing some psychiatrists, it may have had something to do with it.

There was one terrible incident; it shouldnít go on the tape. [Tape recorder turned off:] I remember still that getting on the train, the train took off, everything was opened up again. So I came to Tuebingen and still under the impression that I wasnít well. I was well at the time. I thought Iíll take it easy. I didnít want to do much work, but Iíll go to the laboratory, just sit around, have a good time, have some fun. And I had a lot of fun that time: didnít know what to do, had no plans. Then Back came to me, and said, ďYou know, I have some very strange Zeeman effects which I cannot analyze.

Do you want to look at them?Ē So I said ďSure.Ē I remember still, I didní t smoke much, but in those days I bought Ďmyself fancy long German cigarettes, and I sat there in that little room smoking one cigarette after another, against my habits, and looking at these funny plates. It was the fine structure!

Kuhn:

You had done one paper before the one in which —-

Goudsmit:

No, I had never done anything on hyperfine structure.

Kuhn:

No, excuse me. But in Tuebingen there are two papers I think that come from Tuebingen. Yes, there are two papers that come from Tuebingen.

Goudsmit:

Oh, you mean the coupling papers, ja, ja. Iíd forgotten about that. [Paper No.27] Quite true. I donít know how that came about. That I have completely forgotten. I like that paper, but I have completely forgotten that I did that with Back. Iíll have to read it in order to get the idea. But anyway the main thing that I remember from the very beginning he showed me these very beautiful, very complex structures. Then I remembered that there was a short note, either in Naturwissenschaften or in the Physikalische Zeitschrift.

I believe it was Joos, and later I also found that Ruark had done something similar. They had shown that some of these hyperfine structures look like miniature multiplets that you could ascribe initial level, and lower level, and have transitions, with selection rules. I thought, if that is so, then one ought to apply the Zeernan effect to that, and itís a kind of Paschen-Back effect. And it worked. We were able to analyze these structures, but the first thing we did was not yet the Zeeman effect, because it was a little difficult. It was in the back of my mind. I had tried it, we werenít one hundred percent successful.

Kuhn:

When you mean apply it, you mean analytically apply it, you mean actually put a field on and see —-?

Goudsmit:

Back had a field on, you see, and we applied the rules which we knew from ordinary multiplet to these miniature Multiplets. We were doing that. But before that, we noticed one thing, that these fine structures themselves, after you did a thing like Ruark did, and get the initial level and final level, that you could do the same as with the Zeeman effect. If you had several lines with fine structure in - the spectrum, and you analyzed those, and found that several had the initial level split the same, and others the final level split the same, you could analyze the spectrum.

So in our first paper on hyperfine structure, the emphasis was on the fact that this could be used, as formerly the Zeeman effect was used to analyze the spectrum in levels. That was the emphasis. We did that with bismuth because there the hyperfine structures were the largest; many were measured by Back, but we used also measurements made in Japan by Nagaoka. It was really used like a Zeeman effect to analyze the spectrum. And then in a footnote, I think, we tentatively say we believe this fine structure is due to a nuclear moment. And then the second paper we have finally succeeded, or are convinced that we had succeeded, in even interpreting the Zeeman effect.

And that clinched it. We had done it already at the beginning, but hadnít felt sure about it. Then we got more data, and he took more pictures, and then we had the Zeeman effect, and there was no doubt about, it, it was a spin, and the spin was nine halves, enormously large, and everything fitted so beautifully. Then several papers followed.

Kuhn:

The second paper is Tuebingen, December Ď27. [Paper No. 31]

Goudsmit:

It was written by remote control on new data, and that was done in part by cable, and then Bob Bacher got a thesis out of this work, out of the theoretical interpretation of these things. John Wulff was left behind in Tuebingen —- he still had a year fellowship —- to ride hard on Back. He, for a year, became a spectroscopist, but then he went back to his chemistry, and he is now I think boss of metallurgy at M.I.T. and knows nothing about spectroscopy anymore, unfortunately. But that year was really wonderful.

Kuhn:

What sort of an invo1vemeni did Back himself have with the idea of the coupling of a nuclear moment?

Goudsmit:

Nothing. His idea was to get the most beautifully resolved pictures, and do it over again whenever we wanted to, because I later discovered that some of these pictures had been published by him before. The book of Back and Landť on the Zeeman effect has a plate at the end; the last pictures on that plate were some of these strange fine structures which in the text they say we donít know what they are, here they are; bismuth line so-and-so and so-and-so, with a field and without a field. Beautiful. With a magnifying glass you could almost have made the analysis right there, but nobody had noticed it, not even I, although I had that book. In fact it had been presented to me by Lande.

Kuhn:

In the paper that Pauli felt badly about, had he pointed to this as an effect that might help with particular spectroscopic structures, or was it just something that might happen?

Goudsmit:

No, he had pointed out that these hyperfine structures might be due to nuclear spin. Definitely! And he even —- you see, there were two causes for hyperfine structure which people could think of. One was generally accepted, namely, it was isotopes: different isotopes, slightly different energy levels. And Pauli pointed out to us still a second one, and that you could tell the difference between the two by the Zeeman effect: if it was an isotope, then the Zeeman ďeffect would have the same structure as the individual line, because all the lines would split the same way; if it was magnetic moment of the nucleus, you would get a complicated structure.

And I had never noticed that paper. Probably came out during the period that I was under the weather, and Iíd never seen it at that time. Or if Iíd seen it, it was probably written in such a language that Iíd said, it doesnít mean anything to me, didnít register. And itís a sad thing that nobody called it to my attention, because from TŁbingen, at that time. I rent , to Copenhagen, and discussed all this stuff in great detail with Bohr, who was intrigued, Linus Pauling —- in fact, I thank them in the paper for discussions and they didnít call it to my attention, ďLook,that confirms what Pauli has said a year ago, two years ago.Ē They didní t. Even when the first papers came out, nobody told us, ďHey, havení t you seen Pauliís paper, why doní t you mention him?Ē Nobody! So I was not the only one who ignored Pauliís paper. It didnít come at the right time, thatís what it means again. It was a little premature, people were not interested in hyperfine structure, the people who knew. Nuclear physics wasnít interesting yet, it was chemistry.

Madame Curie was considered a chemist at the time, so who cared about it? I believe thatís the sad fate of the Pauli paper. It came at a moment when nobody cared, except a few spectroscopists, and they even hadnít noticed it, it was too learned. But I knew the paper by Joos and by Ruark where they really had found energy levels to explain the hyperfine structure, but Iíd just given one or two examples. [Interruption]

Goudsmit:

You wanted me to tell you about Mililkan. Ehrenfest told me that he had written Millikan a letter about the spin, in which he said how nice it was that the person who had been the first to do the relativistic doublets was the one to collaborate in getting the spin idea. That was referring to my very first paper, of course. Ehrenfest told me later that he had gotten an answer from Millikan, but Millikan did not like that remark at all and seemed to be rather mad about it. But he never showed me Millikanís letter. So thatí s the whole story. Probably one can find Millikanís letter in Ehrenfestís correspondence. Iím very curious about it.

Kuhn:

The only other thing to ask about your European period is your thesis. It was a survey?

Goudsmit:

It was a survey and a very badly written one. Ehrenfest did not like it. I violated several of Ehrenfestís rules, probably because it was done in a hurry. It is not properly divided, it has no proper index, no bold print; itís just like one long piece —- It goes on and on and on. He did not like it at all. But under the circumstances it was the best I could do. Linus Pauling liked it and translated it. I met him in Copenhagen in 1927 before coming to the States. We worked together there and I discussed various things with him.

And he had gotten hold of my thesis and was translating it, and he said ďWe will publish that in America.Ē But I said it wasnít good enough, so he said, ďAll right, weíll rewrite it a little.Ē And thatís how that book [Goudsmit and Pauling, Structure of Line Spectra, 1930] originated. We rewrote it more than a little bit. He wrote some new chapters, and I wrote some new chapters and sent them to him to change the English.

It was done completely by correspondence. We made the final arrangement in London —- he left a few days before I did on a different ship. In the beginning, of course, he had no time and I had no time. But the book with Linus was nothing else than that I did not want the thesis to be published, in this form.

Kuhn:

You talk less about Ehrenfest than do a number of people who worked at Leiden. What did you get from working with him?

Goudsmit:

I really didnít work with him, because I didnít do anything which was in his Field. I told you that one incident about the statisica1 formula —- he was so surprised. And essentially, he sent me away; he sent me to Zeeman. Thatís where I belonged. But he had enough interest in my work to guide me, and he advised me on every step I took in my life.

Kuhn:

Did you feel personally close to him? Was that an important human relationship for you?

Goudsmit:

He always remained a stranger, but it was a very very close relationship from me to him. There was nothing that I did not dare to discuss with him. And Iím a little worried, therefore, about his correspondence —- the letters I wrote to him —- getting into the wrong hands. There are at least a few which are too personal, which should not be published until I am dead a hundred years.

He promised to destroy some of them, and he may have. Some of my troubles had started already that far back only I didnít know it. But I confided in him. They are purely personal and have nothing to do with physics. In some respects he has had a very bad influence on me, because I have always been the kind of person who found it easier to ask him for advice about what to do than to make up my own mind.

It was not until his death that I finally learned to make my own decisions. Up to that time —-. George, travel, things like that. Youíll find it even in this correspondence. He used to call these letters Ďmeow lettersí—like a cat wanting to go out— always complaining, always asking for help and advice. So I relied very happily upon him for advice, and in that respect it was a very important relationship. But as far as physics was concerned it was not.

Kuhn:

Did you ever work with him in the way that George did?

Goudsmit:

No. No. Maybe for an hour or so on a problem, but never more than that.

Kuhn:

When you were in Copenhagen for the second time was it entirely for the thesis or did you participate in the things that were going on?

Goudsmit:

The main thing I participated in was, probably, giving a colloquium talk or discussing the hyperfine structure, because that was my interest at the moment. Mainly I discussed this with the people who were there. I believe that I am responsible for Casimir getting into that work. In fact I found, but havenít been able to locate again, a manuscript which Casimir wrote on the subject, of which he wanted me to be the co-author because I had really done it.

I believe that preceded his famous Teyler thing. But I didnít want to be the co-author because I didnít understand it. The result was, however, very sad—-.I saw Casimir recently in Copenhagen, and we talked about the history, and he reproached me. I sat on it so long, trying to understand it, that in the meantime — Fermi published exactly the same thing. But in my book I give him credit, this formula was known to me from Casimir.Ē [Interruption]

Kuhn:

Tell me about the American job. How did it come, how did you feel about it?

Goudsmit:

The original letters are still in there. What happened approximately the way I recall it or can reconstruct it, is that Randall was a far-sighted man who wanted theory at Michigan, that Colby was his talent scout, who had studied with Boltzmann and every year went back to Europe, had been in Copenhagen before anybody else, had been with Pauli in Hamburg, with Sommerfeld in Munich, had been everywhere as a kind of observer and student, never making an impression except on the Bohr family; they always liked him personally very, very much. He had brought Europeans to Ann Arbor, but they never stayed. In fact, Kronig was there for a while.

Kuhn:

Was he?

Goudsmit:

Ja.

Kuhn:

Iíd forgotten that. I knew Klein had been there.

Goudsmit:

Klein had been there for a year, Hulthťn, and they didnít stay. So Klein, who had been at Leiden, had told Colby that maybe I would be interested in going there. And I have a letter from Klein, which is in this bundle, where he says Professor Colby from Ann-Arbor is coming and he would like to have people come to the University of Michigan, and even if you are not interested, he is such a nice man, donít fail to see him when he comes. So Colby came and talked with Ehrenfest, and asked Ehrenfest why they had so much trouble in keeping people at Ann Arbor. It was then that Ehrenfest said, you know theorists donít grow alone, you have to give them some people to talk with. I have two young fellows here, maybe they both want to go.

It must have been in the summer of Ď26 when I saw Colby and spent an afternoon with him, with the girl I was engaged to. I had a very nice pleasant afternoon in The Hague. I forgot more or less about it because then that fall I went to Tuebingen. Then finally all hell broke loose because I got letters from Uhlenbeck which must be in there, ďWell, we have gotten an offer from Michigan, Iíd like to go if you go too. Answer at once.Ē Well, thatís the kind of thing—-.

My God, I was old and conservative, and even if my family traveled, and I had a cousin in America, America had never appealed to me. If it had been Egypt or something like that, I would have gone right away, or even India, I always wanted to go to exotic places, or China, but America seemed terribly dull and uninteresting, and nothing. So I didnít answer right away. Then I got telegrams which are still in there, Make up your mind at once. In the meantime I had written to my parents, to see whether they had anything against it. I donít know what they answered. Then I said, yes, but I am engaged, and I want to get married before then. Then suddenly —-.

Kuhn:

Well how did your fiancťe feel about it?

Goudsmit:

Nothing. She thought it was all right, because we still had in our minds that this was like a fellowship, you could go for two years, or one year, and then you come back, and by that time maybe Zeeman is dead or something like that, maybe a job is open. At that time I had already done the spin, so the chances of getting into academic work were a little bit better if somebody was willing to die at the right moment. I had no longer the idea that I had to go to the little high school in the little village, having had this Rockefeller fellowship; that was the first step up, so to say. But when that would come along, and where, I had no idea. And America looked like a better place to wait perhaps than to be an assistant in Leiden or Amsterdam. Thatís what it looked like to me, something which was temporary and waiting for the right job to open up.

So I didnít like it too much but, I was pressured and, as I said, I never made my own decisions. Ehrenfest made them for me. It was obvious that he said yes, and so I said, all right. So thatís really how it happened. I was at TŁbingen at the time and had to do it all by remote control, couldní t talk to the girl I was engaged to, couldnít talk to my parents. And in those days it was an enormous decision to go to America. The way I heard it from my family, people who wanted to evade the police or the draft, they went to America in those days. But nobody in his right mind— Oh yes, my— uncles went for the diamond trade, and some acquaintances of my parents had lived here.

Kuhn:

Uncles on which side, your fatherís?

Goudsmit:

My motherís side.

Kuhn:

I thought they were all milliners.

Goudsmit:

No. They were married to diamond merchants or something like that. They were all merchants, but some of them went to America for the trade. Then they had acquaintances who were in antique business, and had business in America. In fact, they had got kicked out because they sold fake paintings. So it didnít look like a very interesting place to be at all at that time. But the decision was made, as I say, itwas really almost forced upon me, as you can see from the telegrams and the correspondence. I had to make hurried arrangements to get married, so I went back Christmas vacation and got married early in January. Then I went back to TŁbingen, with my wife this time, and prepared. My fellowship was continued; in fact, it was raised.

You got $120 a month if you were single, and $182 when you were married, which was an enormous amount of money at the time for Europe, an enormous amount. Even though I hadnít gotten my Ph.D. yet, the Rockefeller people gave that money. So I went back to TŁbingen, because we had to go to America, I decided to go to Copenhagen, and then to GŲttingen; there is still a letter in there from Ehrenfest asking the Rockefeller people to allow me to change the itinerary a little bit. As a result I did not have my fellowship a full year. I had it only for eight or nine months. Then I went to Copenhagen, wrote my thesis, went back to Holland to stay with the family before going off to far away America. Then we got our degree, our Ph.D. on the same day, that was the 8th of July or something like that. Itís in the thesis somewhere.

Then in September we sailed, made all the arrangements, wrote Colby to try and borrow money. I was always worried about money, how do you pay for a trip to America? The Rockefeller fellowship fortunately had paid so well, and it was so cheap to live in TŁbingen, that we had saved some money, and then Colby, I think, loaned me some money to come over, which we paid back very easily. So we came to America as instructors at the University of Michigan at a salary of, I think it was, $2500 or so, maybe $5000, they made a special exception for us, something like that, but no more than that. We were met at the dock by Ophenheimer. I used that as his introduction when he lectured at Brookhaven. He had completely forgotten about that. But he met us here, and showed us around New York and all the things he did for us I mentioned. But I did not mention the real reason: there was also a lady on board he was interested in at the time. She was in the audience when I introduced him, and really I didnít mention it, but she came up to me latter. That was his real interestÖ She traveled on the same boat with us, and he had written to keep an eye on her —- a physicist.

I can mention on the tape. Itís now Mrs. Houtermans, (Charlotta Riefenstaum). They were friends in GŲttingen, and she came over on the same boat. He was really a wise man, even though Ehrenfest had trouble with him. He must have reasoned that the shock for these Europeans is a little too large. So he put us up at the Brovoert, which is the most European hotel you could find in New York at the time. But then he also took us to Greenwich Village and taught us how to eat corn on the cob. It was just the right time, in September —- excellent! So we stayed in New York for a few days —- I think Uhlenbeck stayed a little longer —- and we went on to Ann Arbor.

Kuhn:

When you say he [Oppenheimer] had trouble with Ehrenfest, what sort of trouble?

Goudsmit:

He was always so vague, his physics was so superficial. And Ehrenfest would put a piece of chalk in his hand, and say, ďNow write it on the blackboard,Ē but, of course, he couldnít do that! He talked in such generalities. So Ehrenfest —- of course, George knows much more about that than I do —- sent him to Pauli for discipline. Then he came back to Leiden and he was there —- both Dirac and Oppie were there when George and I got our degrees. So that was the introduction to America and how it happened.

So it was Oskar Klein who wrote me and Colby who talked to Ehrenfest, and of course then, the way Colby is, right away he said, ďoh, you need more than one.Ē Then he right away saw to it that Laporte also came. Then Dennison was just through with his European thing, and instead of going elsewhere, he was offered a job at Michigan, even though he was a Michigan graduate. They were wise, that they didnít always want to keep their own people, but there was a man with the proper background, so there was a group of four. And it worked. And then I didnít want to leave anymore. There were two very important things Ehrenfest told me. He had been here, and writing letters in the correspondence —- his early letters from America, and his impressions of America.

Of course Ehrenfest had a brother living here who was one hundred per cent American, the typical American. But he also went rough the list even with photographs of all the physicists he had met here, and told us a little story about each of them, their attitude. Not always favorable at all, but there were so many that it didnít register with me at all.

Kuhn:

Do you remember any?

Goudsmit:

Ja, one which was derogatory. He told me about one physicist that was very conservative, and even anti-Semitic. And this man, as he is now, is just the opposite absolutely the opposite! One of the most liberal and, even though he is a New Englander, has taken an active part in the Federation of American Scientists. You know him. Heís retired now.

Kuhn:

But living.

Goudsmit:

Ja, ja, ja. And itís incredible, but that impression was that Ehrenfest had on a short visit to him, in discussing with him.

Kuhn:

Iím likely to guess wrong. Do you want to tell me for these purposes who it was?

Goudsmit:

Ja. E.C. Kemble.

Kuhn:

I would have guessed right, but itís just incredible to me that Ted should have given anybody that impression.

Goudsmit:

Yes, yes. Maybe at that time. And so I canít believe it at all. But I remember that distinctly, you see, that he said this man is i1tra-conservative, even antisemitic. That I remember because the concept in Holland was rather strange. [Ehrenfest] didnít talk about that, but he, with his Viennese background, thought it was necessary to tell me that.

Kuhn:

I should guess that he was mistaken.

Goudsmit:

I should guess. That it was not a change, that he never—-.

Kuhn:

That he would see Ted as, conservative, yes. That he was antisemitic, I think he may have been put off by the haircut.

Goudsmit:

Ja. Absolutely mistaken. So Ehrenfest wasnít always right, because never—-. I met Ted Kemble early in the game when I was here, and I never never, never met a more liberal man. In every respect. But thatís typical. Anyway Ehrenfest went through that. Then the more important thing he told me was the following.

Kuhn:

Do you remember his reactions to any other Americans?

Goudsmit:

No. I donít. Because there were too many, he went through the whole list. He told me stories of reactions of Americans to Germans who had visited here, and that they were still very anti-German, that the First World War had not worn off yet, that they especially disliked the chemist Haber. But I do not know the details. But with Kemble, Iím sure he was wrong there. But I remember the name and I remember this remark. I am not wrong, my memory is not wrong in that respect. But he was.

Kuhn:

No, Iím sure he was.

Goudsmit:

But the more important thing he told me, and the thing I often repeat is the following ďWhen you get to America youíll say they do this wrong," ďAnd their physics isnít any good.Ē He said, ďThat is utterly unimportant. What is important to feel happy is not the state how it is, but the derivative. That is positive there. Only two countries at the moment where itís positive: Russia and America, where things are getting better all the time from the point of view of education, of physics.Ē He was rather narrow, he did not mean politics or anything like that. ďThen you feel that, you can contribute,Ē he said. Then he said to me, ďBut when it comes choice of those two countries, I think you would feel better in America.Ē But at that time he still said that both Russia and America were countries where things were going up and getting better all the time. And that is important, that you feel you can contribute.

Kuhn:

How did you feel when you got there? About America, about physics in America?

Goudsmit:

I felt pleased. First of all, as I told you, my educational, my home background was so that I often felt a little out of place with the people in Holland. I did not have that feeling here. Thereís one incident in my life —-it has nothing to do with quantum mechanics; it tells about me. I had done something practical, I had even made some money, just enough for a little vacation. I had found some rule for mixing oils, the viscosity of mixing oils. My now brother-in-law was in the business, he had it printed, and they sold these charts Ďtoí dealers, so that if they wanted to mix oil, they get the right viscosity.

It was even advertised in Nature at one time. I got just a little money out of it, and a lot of pleasure out of it. I told Ehrenfest about it, and he was intrigued; but he said, ďYou know, here in Leiden you donít mention that.Ē When I came to America I was very pleased to notice that they had an automobile built in the physics building in Michigan, to do applied work. It didnít shock me at all, on the contrary. In fact, at one time, I picked up that viscosity work again and tried to go to oil companies here to see whether they were interested in developing it further. It turned out that they independently had discovered similar things, and they were printed in the handbooks. I even once tried to write a paper about it, but it was rejected by the Journal of Rhelogy.

I still have the manuscript somewhereÖon oil viscosity mixes and so forth, and dependence on temperature. It made me feel good that that was not taboo. So I felt quite at home among those people. Then I like, people in general, and we were very lucky in the people we met and got in contact with in the, early days. The only black time I remember was the first Christmas vacation, because, everybody had gone somewhere; nobody was left in Ann Arbor. Our winter clothes werenít warm enough.

We had no car. We lived in two rooms, one looking out upon the hospital, the other upon the cemetery —- no kitchen, just a little gas flame to make some tea, bathroom shared with one or two other couples, icebox out on the porch with one or two other couples. It was horrible. It was at that time that there was an ad in Nature that they needed a professor in Egypt, and I wrote immediately. The correspondence is still in the file. I didnít get the job.

Kuhn:

Would you have taken it, do you think?

Goudsmit:

At that moment, yes.

Kuhn:

But you wouldnít have gotten it before April, in April would you have?

Goudsmit:

No, by then probably I wouldnít have wanted it. But I wanted it so badly. I gave as references in that letter, I think, Bohr and Einstein and Ehrenfest a whole list of the highest references. I could have gotten the job, because then I got a letter back which may be still in there, I think itís still in there, I read it not so long ago, in which they said, ďOh yes, sorry but I think somebody else (gets ???), our regards to Professor Colby.Ē [Laughter] So he was home there too, as he always is, everywhere. That was the only black spot at the time. Then we had very good luck in that we were able to go back to Europe after about two years, I think it was. Thatís very important for immigrants, because then when you go back, then you can compare. And then, I think, we made up our mind that we liked it better here.

Kuhn:

Were you there for the summer only?

Goudsmit:

No. I think it was more than the summer. I donít know whether that was already on that new scheme. I donít know when that new scheme came. Well, I had a few offers then. Finally some European offers came through, one to London. But as I was here too short a time, I donít think I wanted it, and I recommended Kronig. I think he took it at that time. Then came the offer from ZŁrich. I doní t óknow exactly what year that was, to succeed SchrŲdinger. I believe Ehrenfest wanted me to take it, the correspondence is in there, but I just did not see it, I couldnít. -

Kuhn:

Tell me more about why.

Goudsmit:

Because I believe I was subconsciously aware that I was not a theoretical physicist, that I would have broken down just trying to keep up the standards of SchrŲdinger. I could have succeeded Zeeman. That shows you what a low opinion I had of his knowledge, in spite of the Zeeman effects But I definitely felt that I was not the successor to SchrŲdinger. I know who recommended me highly and wanted me: that was Edgar Meyer, he was the experimintalist there, and he used to come to Tuebingen. He wanted a theorist he could talk to, and do spectroscopy with and things like that; he wanted me badly. In some later correspondence, which I only read recently, it turned out that even if I had wanted it, I might not have gotten it. Apparently they were negotiating with several people at the same time.

Kuhn:

I didnít know they did that.

Goudsmit:

I didnít know either, but I think that a week or two ago, goingóthrough this correspondence, I found a letter to that effect, from Ehrenfest, of which he sent me a copy or something like that, giving them hell. But I didnít want that anyway. So I remember that I sent them a telegram or a letter telling them. But it had an effect on Michigan. Randall got to work, we got a raise in pay; then we got that arrangement which the dean accepted, and which was very important, to have leaves of absence, every fifth semester, with pay. So out of the five theorists, that means Colby was included in that cycle, one was always away for a semester and a summer. That lasted for about ten years until the war finally made it impossible.

Kuhn:

George did go back.

Goudsmit:

Then George got an offer from Holland, and did go back. Youíd better ask him about it. I always had the feeling that it was for family reasons. But he didnít like it, as you know; he didnít at all, and she didnít like it. So when the opportunity arose to come back again to this country, they went back to Michigan, to the same house even which they had left before. He hates to move and change his habits. It was really a triumph. Iím very proud of it that I finally got him to leave Ann Arbor. He went back to Utrecht. But I still was under the system of every fifth semester going abroad for a semester and a summer, with pay. It was really marvelous, very important. So Laporte, Dennison, George, Colby and myself. That came about because I got that offer. No matter what Iíve done, I wouldnít have taken it, but they didnít know that. I didnít press for that; this idea was all really Randallís.

Kuhn:

How did you find the students?

Goudsmit:

Marvelous! First of all, we were so on the same level, it was so easy. The undergraduate studens at the beginning I had trouble with, because I was used to the discipline and the hostility between student and teacher of the European schools. And I treated them the wrong way, so they took the undergraduate teaching away from us at the beginning for a while, until we —-.

Kuhn:

For both you and George.

Goudsmit:

I think for both, ja. But later, after we were adjusted a little better to the American way of doing things, we got it back again. And I believe that we were very, very successful in our undergraduate teaching too.

Kuhn:

I know you both have the reputation, and in your case I know first-hand.

Goudsmit:

You mean the Harvard lectures?

Kuhn:

That Harvard semester in Physics G. I have very fond recollections of that.

Goudsmit:

But at the beginning we did it wrong. We didnít know what to do, didnít know what to require of the students, tried to introduce European systems which didnít apply at all. It took probably two years before we knew, then we were given back some undergraduate quiz sections again. And I liked it. Then in the graduate work, of course, I didnít have many students, because, like Ehrenfest, I was snooty in that respect. I wanted only students whom I could work with on an equal level, and who didnít need too much help. I didnít want to write a thesis for them, they had to come up with something, or I would give it to them and they had to do the work.

Letís see, whom did I work with? Some of them, who didnít get their degree with me, I used to work with very closely anyway. Bob Bacher was the first one. Then I had an experimental student, was [Russell A.] Fisher, hyperfine structure; (Tai Yu Wu), a Chinese — I think he got his degree with me, at least we worked quite a lot together. I believe his thesis was with me. I donít know whether (Lloyd, Young) worked with George Uhlenbeck or with me at the time. Maybe his thesis work was with Uhlenbeck, but some of his work was with me. David Inglis —- his thesis was with me.

But it was purely theoretical and I can honestly tell you that he knew more than I did. I got him interested in the problems, but he knew much more about it. In fact, I have an old notebook here I noticed that I learned it from him,some of it, —- quantum mechanics techniques.

Kuhn:

And you were learning physics all this time?

Goudsmit:

I was learning physics all that time, by teaching it, and from my students. Because I really didnít know it.

Kuhn:

You never thought of going back and taking the exams again?

Goudsmit:

I would have been scared stiff. The rest about the American episode, you know.

Kuhn:

Now tell me about some of the work you were doing. You went on with the hyperfine structures.

Goudsmit:

I went on with the hyperfine structure, ja. I think the only way to do it is to look at the papers. But that only goes up to—-.

Kuhn:

I think it goes up to about Ď31. I think already in this period itís probably fair to say that there are all sorts of fascinating puzzles in the spectra still left, but that, on the other hand, the fundamental excitement has probably gone out of it by the end of this. Already the interest in hyperfine structure is switching from an interest in spectra to an interest in the nucleus for example. The whole question is, when is one through with multiplets? Youíve got some quite ingenious stuff here to do, and gamma sum rules and so on.

Goudsmit:

There were some tricks I was very proud of. I see now, fully, how stupid they were. Because I knew so little mathematics, I always had to rely on my intuition and ingenuity arid tricks to do certain things for instance, there is a paper about guessing at the formulas going from one coupling to the other, which I thought was marvelous. I knew enough quantum mechanics to know that it had to be quadratic equations where you had two levels and cubic equations at three levels.

I knew from most elementary considerations I did that with Inglis, I think —- what it was in one extreme coupling, what it was in the other extreme coupling. That determined the coefficients without having to know any quantum mechanics And of course, the people who knew quantum mechanics thought that all very stupid. After I had done that, then they come with long formulas with all the coefficients. What upset me most: if you ever look through Condon and Shortleyís book, you find several chapters on the subject, then it has a footnote, ďaccording to Goudsmit.Ē But then you look at it: I could never have written it, I donít understand a word of it. Itís really remarkable.

Because they did it correctly, in a general way. I got the same results always by tricks whenever it was necessary, and sometimes tricks were not possible. So that was the kind of work I did. Indeed the intensities with Kronig was similar. You knew certain extreme cases, then you guessed at what you had to do. Then later people could derive it with quantum mechanics. But these London and Shortley things, itís really remarkable. I looked at it, and I said, ďGee whizz, was I that learned?Ē

Kuhn:

But there is this difference: although you were doing the same sort of work, and in many respects in the same way, now, at this date, one would say one could already have done it the other way. Whereas when you did the intensity things with Kronig, one could not have done it any way except the way in which you did it. Now how conscious were you and other people that already those problems were solved?

Goudsmit:

I was semióconscious of it, but I had the feeling the people werenít interested because these problems were only of interst to spectroscopists. The quantum mechanicians already begin to get less and less interested in spectra lines. There were other problems for them like statistical mechanics which were far more interesting than spectroscopy. So they didnít care whether it went from JJ coupling to Russel-Saunders coupling, and that there were formulas like that. And when the Condon-Shortley book came outóI think they said, ĎNow thatís the end. No more of this. There it all is; if you want something you can find it in there, but who wants it?í There was another thing, a paper I wrote later with Bob Bacher which is not on this list, where we introduced a fractional parentage.

Kuhn:

I donít even know what that is.

Goudsmit:

Ja, you see. Itís now called fractional parentage, I donít know what we called it. But again by a marvelous— I say myself it was marvelous— by an intuitive guessing game, we did the thing which they now rediscover and do with highfaluting group theory and Racah coefficients. We had Racah coefficients before Racah existed — by a similar trick. We needed that very badly. It is the followin: that if you add an electron to a spectrum, then the state of the new atom with the added electron must be related to the state of the ion.

But now the state of the ion is usually a complicated thing, it may be a triplet p and a singlet p and so on. Only in extreme cases, when can you tell on which of these original states the new state is built up. But if you have equivalent electrons, then itís a mixture. Then it is, letí s say, 6/7ths of triplet p, and 1/12th of singlet D and so on. How do you get these co-efficients? Theyíre now called Racah co-efficients, and they have them for the most complicated cases; you just have to put it into the machinery and get it out. But we needed it for the paper, and we invented them by guess-work.

It was very nice. It was the first paper where we pointed out that this existed, that you just couldnít add an electron to an ion and say what came out of it, that you had to do all this. That was again in line with what I did, but now, of course, thatí s completely forgotten because you doní t need the trick, you look it up in the tables, the Racah co-efficients.

Kuhn:

In these papers there are a number of points where you point to possible nuclear interests, electron-electron interaction, or electron-nuclear interaction, to correct or improve the Dirac equation.

Goudsmit:

Ja, I think that was nonsense. Where did I say that?

Kuhn:

Was this all window-dressing?

Goudsmit:

Absolute window-dressing, Iím sure. The only thing I was interested in was the nuclear moments. I also was probably the first ore who was able to derive the order of magnitude of nuclear moments. At least thatís what Rabi claims. But I donít know whether I was really the first. By using a Landť type of formula, from the size of the hyperfine structure and a Fermi formula, or whateveróit was, I really got the first nuclear magnetic moments for complicated spectra.

Kuhn:

Would you say you were doing this to get magnetic moments or to db spectra#?

Goudsmit:

I was doing it to get magnetic moments for the nuclei; for no other reason. I did some other work. There was one little thing I always liked because it wac different. I did that Brownian motion paper with George. That was something quite different.

Kuhn:

How did you happen to do that?

Goudsmit:

It goes way back to Amsterdam. In Amsterdam in Zeemanís lab, there was one man who was doing Brownian motion, in a rather stupid way. I donít know why people were interested at that time —- He had put a little mirror on a fiber, and photographed the deflection, and got nice curve, Brownian motion, and he didnít know what to do with it.

Kuhn:

Gerlach did this one.

Goudsmit:

I was in TŁbingen, and then there was a fellow with Gerlach who did the same, but better, because he used a rotation, which gave much better deflections than that stupid guy in Amsterdam.

Kuhn:

Did you suggest this in Tuebingen, or was it just a coincidence that they both —?

Goudsmit:

I think it was just a coincidence. I donít think I suggested it. Iím pretty sure it was a coincidence. I believe all this work came back because my guess, and probably a right guess, is that people were interested in better and better galvanometers, and that the limit was the Brownian motion. They observed Brownian motion in galvanometers and wanted to learn more about it in order to see whether they could improve it. As a good experimenter, you begin to investigate. Now the trouble was the interpretation of these various curves. I had looked at them, I had done it at Amsterdam already, and I had told. them that they ought to take a Fourier analysis, because they were different.

At different pressures they looked quite different. But I had no explanation. But I remember playing with the Fourier analysis. At the Amsterdam lab they had one of these planimeters with a Fourier thing where you put different things in —- a very primitive way of taking Fourier analysis —- to see whether the Fourier components were the same.

Kuhn:

Youíre not supposed to know anything about Fourier analysis at that time!

Goudsmit:

I .knew just enough Not much, honest, not much. But I remember playing with the instrument and trying to get the componentís. I donít know whether I did it first in TŁbingen and then in Amsterdam; probably first in Amsterdam arid then in TŁbingen. It happened that they were both working on the same problem, and it always was back in my mind. Then when we came to Ann Arbor, I had a chance we had an office together —- to discuss it with George. He may remember that better than I do.

I said, ďEven at different pressures, we know from our elementary theory that the mean-square deviation is always the same. What is different at different pressures? Why do they look so much more like pure sine curves at low pressure than at high pressure?Ē Then we began to study the Fourier components of Brownian motion. I believe that I was the one who called Georgeís attention to the problem. But he was the one who, of course, understood this field, and the computations and so on. But I learned it. I really understood that paper at that time, to my great surprise. I probably could even reproduce it yet.

Kuhn:

Your feeling is that already by the time that you got to the United States, people were no longer excited about spectra?

Goudsmit:

Yes.

Kuhn:

And that Condon and Shortley were fairly decisive?

Goudsmit:

Ja. Well not quite; I came to the United States in Ď27. I think about 1932 or so was the end as far as spectra was concerned. At least only detailed problems and only die-hards who couldnít do anything else would go on being interested in spectra from a theoretical point of view. The only thing I still thought might be interesting was hyperfine structure because of its nuclear interest. So at Michigan I had Fisher work on hyperfine structure at the time; and so did Inglis. But that was honestly the nuclear interest, not the spectroscopy interest. But it was nice that there was something in nuclear physics I could do. It was a little premature.

Kuhn:

At the time that you were doing the hyperfine structure with an eye to the nucleus, were you also following what was going on in nuclear theory, the development of nuclear theory?

Goudsmit:

That came a little later. The first things I was very much aware of were the papers of Elsasser because I understood those a little bit. I understood enough statistical mechanics, having been exposed to it, that I understood Elsasserís papers, I thought, in getting nuclear energy shells.

Kuhn:

But even before that, how were you about problems like the statistics of nucleus?

Goudsmit:

The statistics, no.

Kuhn:

The electron and the nucleus, the neutron, positron, neutrino, these problems.

Goudsmit:

I was aware of them, but only superficially, and was only aware of them in so far as they had something to do with hyperfine structure. I was not fully aware of the difficulty in the nitrogen spectrum; I knew about it. Molecular spectra was already too difficult for me. The alternating intensities which Dennison worked on, and Dieke, I understood a little bit, but not enough to make it my own, that I could ever have made a contribution.

Kuhn:

Wouldnít your own sort of techniques have also helped with molecular spectra?

Goudsmit:

Perhaps. But I am not sure. Remember that Rabi had promised me a bottle of champagne years later if I could use my ingenuity and apply to his work—- I think it was his molecular work at the time —- the same skill I did to getting nuclear moments from the atomic spectra. He said there must be some simple relation. The nuclear moments I got by comparing the hyperfine structure with the fine structure. You didnít . have to know anything else. I could prove that by that ratio you could get the nuclear moment. He said there must be something similar in the molecular business he was doing in his molecular beam, but I never succeeded. It turned out it wouldnít have been possible —- far more complicated. I doní t think that my skill would have helped with molecular spectra. So you see how narrow I was! But in the narrow field I pushed ahead.

Kuhn:

I realize thereís one question I left out earlier: it takes us back to spin. In the Nature paper oní spin you list a lot of the reasons for it. Then you make the remark, Ēso far, we have not mentioned the Zeeman effect, although the introduction of the spinning electron was primarily suggested by the analysis of the anomalous Zeeman effect, shown by the components of multiplet structures.Ē Now what does that mean? I found that hard to understand, why you would single out the anomalous Zeeman effect as the principal thing to suggest spin.

Goudsmit:

It probably is that the sentence is too condensed. I meant by that, the anomalous Zeeman effect was the reason for the vector scheme. That I knew so well, that is how it came up. I think that is what it was, and nothing else.

Kuhn:

Thatís right, thatí s certainly clear and I should think entirely correct; but it reads as though you meant it was in particular the puzzle of the anomalous Zeeman effect.

Goudsmit:

No, it was the Zeeman effect that led to the factor with the factor two, and that it could all be now condensed into one statement, electron has a spin one half, factor two, and then the whole Landť factor, the whole business, came out.

Kuhn:

Would it be right to say that if there was a key problem that went from there to spin, it was much more the relationship of the X-ray and the doublet riddle?

Goudsmit:

No. That would not be right to say. The idea of spin, Iím sure, came in because I tried to explain the Zeeman effects and Pauli principle to George Uhlenbeck. Then after we had the idea of the spin, then I said, ďOh, but now it also explains my old doublet ideas and so on, and the X-ray doublets.Ē Thatís how it went, logically, or historically. Iím sure of that.

Kuhn:

What am I forgetting to ask you?

Goudsmit:

Nothing. I still claim that these precise ideas, how they went, my memory may fail, also I still think itís unimportant. Absolutely unimportant except for a psychiatrist or a psychologist. But for the history of physics, it is unimportant.

Kuhn:

I will argue with you about that off the tape.

Goudsmit:

Another thing is that sometimes it is more logical to tell it the way it did not really happen.

Kuhn:

more logical I donít doubt.

Goudsmit:

It would have been obvious to do it from the doublets and so on. It would have been obvious for me and for Kronig to do it in May.

Kuhn:

Tell me on the tape.

Goudsmit:

The spin paper. What Ehrenfest told me at the time: ďBe sure that George Uhlenbeckí s name comes first, because you have already published many papers, George has not. And the way it goes, heíll be forgotten, or dropped.Ē And I said, ďOf course, he gets first.Ē And even so you find, especially in the early years, many references either to me alone, or even now, references where my name comes first, incorrectly. Ehrenfest was so right about that; And not only that George deserved to be first, but he made it a point, ďPlease,Ē he said, ďGoudsmit, make sure that Georgeís name comes first, otherwise heíll be chopped Off in no time.Ē You find many of the references where the names are wrong. In all our papers, George is first. But in the quotation, the other way.

Kuhn:

In the spin papers.

Goudsmit:

The spin papers, ja.

Kuhn:

Iím very glad you told me that.

Goudsmit:

Ehrenfest stressed that, you see. And if you now look at the quotations, you find it very often the wrong way around, in the old quotations.

Session I | Session II | Session III