Oral History Transcript — Dr. Edwin Crawford Kemble
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.
See the catalog record for this interview and search for other interviews in our collection
E.C. Kemble; October 1 and 2, 1963
ABSTRACT:Transfers from Ohio Wesleyan to Case. Works for Chase Machine Company. Enters physics. Research with Professor Miller, instruments used, experiments performed. Describes his coursework, graduate thesis, and mathematics at Case. Mentions Miller's research. What was known and discussed about the quantum and Millikan's unitary theory of light at this time. At Carnegie Tech. meets Knox and Hauer and runs the freshman laboratory. Describes the atmosphere at Carnegie. Begins a fellowship at Harvard in the fall of 1913. Studies under Wallace Sabine, Osgood, Maxime BÍcher, and B.O. Peirce, Bridgman, and Harvey Davis. Mentions Duane. Discusses the physics and mathematics departments at Harvard. Bridgman's potential theory course discussed. Describes unsettling problems in physics during this period. Harvard's weekly colloquium. Thesis and collaboration with Jim Brinsmaid. Ideas he had about future research topics. Classical emissions. Bohr atom as appreciated at Harvard and the reception of quantum theory following the War. Complications in physics at the time of the interview. Instructs at Harvard after the War. Researches helium spectrum, band spectra, integral pdq equaling nh. Visits Germany in 1927, his readings in German physics. Mentions his Ph.D. student, Van Vleck. Hamilton-Jacobi techniques. Discusses theoretical physics before quantum theory. Describes Duane, his research, Duane grating theory. De Broglie's paper on the wave properties of the electron. How the discovery of the Compton effect was perceived at Harvard. Webster, Webster's career following quantum theory described. Mentions the backwardness of theoretical physics and mathematics in the United States. Writing his book, his work on general education explored. Mentions his bibliography. Included is a letter from Kemble detailing his first references to Bohr.
Session I | Session II
Kuhn:I would like to get down to the record contrast between your training and your initial work up to the point of your thesis, and then what happened from that point on. Would you start it over again?
I just wanted to say that, as I reflect back on it,impressed with the fact that as I was an undergraduate student in physics, and in the early part of my graduate training, my interest was in theoretical physics. I knew there was any imaginable number of new experiments to be done and developments to be made on the basis of the existing theory. But such things as the electro-magnetic theory of light seemed to be in order, and electricity and magnetism seemed to be in order. If they weren't in order, it was because somebody hadn't understood that every thing clicked when you properly understood it. My interest was, to an appreciable extent, in finding places where something was out of order but where I was just sure that it just needed a closer examination to show that order did exist, for example, in my first paper on the electrostriction. This was an apparent difficulty between two different approaches to the same subject. They were both obviously sound and they both fitted together when one looked at it carefully. That's something Bridgman put me at. I had this business that the definition of work--. When you thought about rolling balls, sliding things and so forth obviously the proper definition is that work is the integral of V*f dt, and everything goes smoothly. But it wasn't down in the literature anywhere. The Physical Review, however, refused to publish it because somebody who was a referee said this would. be confusing. And, so it was never published.
Kuhn:Was that a problem you originated yourself?
Kemble:I think Osgood raised the question. We had been talking about it in class and I couldn't tell exactly how, but we had been dealing with trick cases. Osgood said he didn't know how to define it in general. And so this came up. Another example of a problem which doesn't look right but if you understand it properly then everything is clear, came up in my optics course, I think, the first year that I was here. I took a course in optics with Sabine. Sabine was a top-flight person; he was the Dean of the School of Applied Science, he was the fellow who had brought me here, as a matter of fact, in the first place. But optics was a side issue with him, and he taught optics ŗ la Schuster. I had read Drude, or did read it at the same time . I was taken aback to find that the problem of the quarter wave length phase difference when you calculate by Huygens' what the wave ought to be, and the problem of the obliquity factor in Huygens' principle were in his mind unsolved problems. But it was all done by Kirchhhoff's formulation and he [Sabine] wasn't familiar with it. When I went to teach optics I was interested in getting things into more elegant form but I didn't go into my graduate work in theoretical physics with any notion, really any fundamntal break at all. And when I began --.
Kuhn:Do you think at that stage of the game you had even heard of anything like Planck's Law?
Kemble:I'm not sure just when I began to hear about that. I had to look up records. I know that the newer things were reported in our colloquium from time to time, probably inadequately, perhaps not always by the right person. But we knew there was a photo-electric effect that was puzzling. You'll find, it in the University Gazette. I did look it up and found quite a record of colloquium topics that showed we weren't entirely asleep. But I had no proper orientation as a mature theoretical physicist entering into quantum theory to bring to bear everything that ought to be borne and expect to do something fundamental with it. I didn't: I was a kid. If I could find something new that nobody had thought about in the way of an application that had made some kind, of sense, I was very happy over that. I used plenty of bad judgement in the kind of thing I would follow out. This is part of the explanation of the rather crude form which the quantum theory had in my mind during the first few years. As soon as Sommerfeld's phase integrals came out, I remember that I reported to the colloquium his paper and was informed by one of my colleagues that I didn't put it across. But it seemed pretty clear to me, and this became my language from then on.
Kuhn:When you started this before the recorder went on, you spoke of this as something that had. been true of everything up to your thesis. Clearly the thesis was, at least in retrospect we would now know, another sort of problem. It isn't one that could have been at that point really brought off cleanly and tied up once and for all that way. I wonder how conscious you were that this was somehow a different sort of work?
Kemble:Oh yes, I was excited over that thesis; anything with quantum in it, with h in it was exciting. That I could put my finger on something of this kind and go into the laboratory and find something that nobody had ever dreamed of before, of course, this was very exciting. Unfortunately the first of the wars pulled me off of this just as I'd finished and before -- all the thesis was never published because the lapse of time was too great. When I got back I wanted to do something else. It seems strange to me now, but it was true for quite a long period of time, from let's say 1922 to somewhere in the 1930's, maybe longer than that, that I didn't think there was anything in physics that if I took it in my hand and worked at it long enough, I wouldn't have a contribution to make. Everything that I worked at came alive, had meaning, and I wasn't stopped entirely, although my output of material was very small.I had sleepless nights and all sorts of strain that kept me on a certain level. I couldn't compete with people like Van [Vleck] for example in doing one thing after another, but I had this feeling that there was no limit, just the same.
Kuhn:I'd like to get on another part of it. I see the sense of excitement about working with the quantum and the sense of control, given enough work. But When you started you also emphasized the fact that everything in your training and the problems you worked on up to the thesis problem had been ones that if you stuck with them long enough came out right. You got the results that the experiment had given or if not, you could explain what was the matter with the experiment if they didn't give it. When you get into the vibrator problem -- the rotator problem -- did you in the thesis do both band spectra and specific heat for the rotator or was it just Van --?
Kemble:I didn't do any specific heat.
Kuhn:This you came back to when you worked with Van. For the band spectra it isn't as clear as it is for the specific heats -- it isn't altogether clear for the specific heats in the paper with Van -- it doesn't come out quite right. It's the contrast between the sense of this sort of work, the sense of the classical work -- how does it hit somebody who thinks physics problems behave the way classical problems behave when they stop behaving that way?
I think as soon as one began to step into the quantum area, one immediately knew that there was something very fundamental about this and one didn't expect, in this area, to pick up the right answers right away. I always insisted, against things that other people were saying, that Bohr certainly never took his models seriously in the sense that he expected them to be anything more than convenient models that are to be used as stepping stones to something else. They didn't tell the story, and I think--in retrospect I think it was practically from the beginning -- that I took it for granted that there was more to it than we had got so far. On the other hand, I remember -- and this belies what I've just said -- that when we were in Europe in 1927, I was working on the problem of the interaction between vibration and rotation of the mo cule and how that would affect the spectrum. Although I knew that there was a new way of going at things and had some notion about how it went, I started to work this problem out on the Bohr basis. The Bohr theory had given me so many things that seemed to be all right. When you did. Bohr-wise, the harmonic didn't have to be an exact harmonic. I couldn't believe that this simple little interaction effect wouldn't come out right. To test it meant going to numerical values. There got to be in the theory quite a number of parameters, and I fussed and fumed and wasted, I suppose, a couple of months of time trying to make that thing come out right. And I finally got a combination of parameters to work pretty well. But I feel now that this was fortuitous. Of course, it was fortuitous. The time when I should have been busy learning new quantum mechanics I was fussing away with a problem I couldn't quite get rid of on the older basis. This was the spring of 1927.
Kuhn:To what extent did you really look for either the matrix methods or the wave mechanical method?
Kemble:Very little. I remember Pauling was in Munich, and conversations with him got me started at Schrodinger's at that time, and it was immense fun to follow up. But then shortly after, when I once got into quantum mechanics, I got perhaps taken in by Ehrenfest's remark. He came over here and instructed me as a young physicist who hopefully had a career, that the most important things are the most general and the thing for me to do was to try to get away from things that are specific and connected with particular concrete problems and think in the most general terms. So I got to writing this article for the Reviews of Modern Physics which turned out to be the book, which used up all my sap for quite a while.
Van Vleck:It was Vol. 1 in Reviews of Modern Physics, wasn't it?
Kemble:I'm not sure it might have been.
Kuhn:In a number of places people were so sure that something was basically Laa matter-by 1923-24, at least by 21., not necessarily the matrix mechanics, but Schroedinger got picked up awful fast. It therefore speaks somewhat for the Harvard environment at this point that you had got out a problem of this sort in 1927 with the Schroedinger equation and the manner of handling it still to learn. Is this fairly typical here, do you think? What have people said in late 25, early 26, about these new methods? Van comments on them quite early; he gets to use them a bit very early. You were here at that point?
Van Vleck:No, I was in Minnesota.
The department here was aware of quanta back in 1915. Everybody was for following up every new thing that there was, but each member of the department had his awn field, and it was only the young kid coming along who was free to go ahead and follow this thing. Bridgman was committed to his high pressure work, and everybody else. There weren't any loose members, and the only theoretical physicists in the country at that time were really men on whom the load of teaching all the mathematical physics courses lay, and they all spent their time teaching. It wasn't, as I remember, a constructive occupation, except with Arthur Gordon Webster, perhaps, out at Clark. He was the man who knew more about physics than anybody else, and the man who dominated the meetings at the American Physical Society. The fact that he dominated them in part means, of course, that physics was still fairly simple. It also means that the professors of the various branches of experimental physics each stuck in a narrow domain and it was the over-all man who knew the differential equations, who could talk about any figure.
Kuhn:We talked a good deal last time, and very instructively to me, about the situation before the war and immediately after the war. What I'm after this time a little bit more is the Heisenberg-Born-Jordan papers, the Schroedinger papers. By this time there were now at least a few people who were reasonably up-to-date on this later quantum theory. Some of those were quite prepared, if not for this particular way out, at least for something terribly fundamental. I wonder how you and the people around you at Harvard felt about those papers? Presumably they were reported. What you express is the attitude with which you went at this interaction problem which you would still try out on the old quantum theory and still expect to get answers that way. This has got to displace something that didn't make you immediately rush out and learn at least what the Schroedinger treatment of the hydrogen atom and the rotator had been?
Kemble:I suppose this is in part an unadventuresomeness in my mind that I wanted to finish up the thing that I'd begun. I didn't have any appreciation of how fast things were going to proceed. I thought I had time, and it wouldn't make any difference; physics wouldn't change that much in 6 months. I felt I couldn't let anything in my past interfere with getting out the new thing. This is a place where I wasn't a good policy maker.
Kuhn:Did you actually have the full year of 1926-27?
Kemble:No, only the half-year. I went over in February.
Kuhn:How did that trip come about?
Van Vleck:What year were you a summer lecturer in Chicago and Michigan, teaching people quantum mechanics?
Kemble:Those were really courses in band spectra, in both cases, I think. It was in 1928 at Michigan, and it may have been 1926 in Chicago. It was before quantum mechanics, in the days when we had the Bohr theory to work with. We were worrying about half-quantum numbers, and how the different branches of the band spectrum started off, things of that kind.
Kuhn:What sort of audience was there there then for even old quantum theory?
Van Vleck:Chicago in the late 1910's, early 1920's had the preeminent summer school in the country. Then Michigan came on later, about 1927 or 28, and brought over a large number of European stars in the particular area of --.
Kemble:Kramers was there in the summer I was there. One thing I do recall was my shock going out to Chicago. At Chicago, the habit in those days was to give an incoming graduate student a thesis topic almost right away, either in the middle or the end of his first year. Thesis topics were experimental things, and you didn't have to know too much to get started, so everybody got started right away. Whereas here it didn't happen so quickly and I myself was here 3 years, took practically all the graduate courses in physics and a bunch of math, before I began to even think about a thesis.
Van Vleck:I'd say that the caliber of the courses given by Campbell and Bridgman in that era were by and large superior to those at chicago.
Kemble:The Chicago people were not mathematical. As I said, I was the present at one their Doctor's examinations. They invited me in, and I asked the embarrassing question of the student how did you differentiate a definite integral with respect to the upper limit and he didn't know how. I guess he passed the examination in a sense. Just as soon as one began to see what the power of quantum theory was, people rushed in all over the country. We had professors of theoretical physics.Gregory Breit was getting started there one of the first, but there was a big contrast between Gregory Breit and Lee Page who was the Yale man before him.
Van Vleck:You mean that Breit was far more able?
Kemble:Yes. Page was a conventionally-minded person; perfectly good at stating the idea, and a good teacher, but he never had any new ideas about quantum theory.
Kuhn:Did he even try in that area?
Kemble:He was a member of our Molecular Spectra committee, and wrote the first section of the report, but there wasn't anything but what was quite conventional.
Kuhn:I would like to start by going back to your early interest in science and asking you about one range of issues and how they worked themselves out for you. This is where you may want to refuse to talk about it. I had known before that you were the son of a minister and you told me something about the development of your own career, that you'd originally intended to go into the ministry yourself.
Kemble:I had never made that decision.
Kuhn:At least this was a perfectly viable possibility for you. On the other hand, you've told me that you also have an older brother who was an engineer. You yourself went into the sciences. I really want to ask you in semi-personal terms about the relation, in your own case, and to the extent that you've known others in a similar position in America, about the relationship between a religious background and a scientific career. Did these two work together for you, did they conflict for you, what sort of issues have they presented?
Kemble:First of all, my father was not only a minister, but also a person interested in science. He was to some degree an inventor. My first introduction to scientific ideas came from my father. There never was any anxiety with respect to science introduced by my religious ideas. On the other hand, my general common sense and feeling for the world introduced grave religious anxiety at an early stage which never left me.
Kuhn:How early did that start?
Kemble:About 15. I talked these things over with my father. I had very good relations with him. He was something of a philosopher as well as a scientist: he worried about problems of philosophy,he worried about the contradictions within the Bible. He told me that there have been times in his life when, a committed minister, he felt himself hanging over the pit of hell. These were his words for his worry about the fundamentals of the religion he was teaching.
Kuhn:What was his denomination?
Methodist. He was a Northwestern graduate of the Garrett Biblical Institute. I never found my anxieties about religion quelled. On the other hand -- this is really a life-long matter with me -- in spite of pretty clear notion that I did not accept the religious ideas of my forefathers, especially as they related to a view of the world and how it was brought into being and who governed it, the why and wherefore, it has always remained true that there was something in the religious atmosphere that I felt was important; and I still feel it's important, but I think of the church today as a bungling oaf trying to carry a piece of fine china, in constant danger of falling down and breaking the whole thing. That's a poor analogy because things don't happen rapidly like the breaking of china. I have gradually during my life developed a fairly mature philosophy in which the need for a transformation of religion to a new set of co-ordinates seems very essential. Right down to this moment I'm engaged, whenever I get the chance, in promoting that transition. In my first year of college I went to Ohio Wesleyan, my mother's college. While I was there my older brother persuaded me that I was not going to be able to earn a living. I was pretty sure by that time that I did not want to go into the ministry. Having done well in mathematics, especially in geometry, the engineering work seemed the next-best thing, the most sensible thing to do. While I was in the Case School I stayed out of school a year to earn more money, because I was putting myself through school. I worked in an engineering office and decided that engineering would not be for me, and so I'm in physics. My interests from early times were of many varieties. It was no singleness of purpose driving me down the physics "alley" until I was already in it.
Kuhn:Clearly, as you say, your science has on occasion made trouble for your religious feeling. Do you, in spite of this, have any sense at all -- you know the literature out of which this question arises -- that your original religious, and your continuing religious orientation, has given you either any special motivation for the sciences, special interest in nature, or has had anything to do with determining the sort of science you're particularly interested in doing?
Kemble:No. The fact of the matter is that I have never been in the mood of the generation of scientists who thought that in scientific discovery that they were following the thoughts of God. This was to me a separate thing. I was studying nature; I was interested in nature, in lawfulness, in generality of any kind, in consistency between the different areas of my own personal life. My interest in religion was connected with the notion of generosity and self-sacrifice that started me off -- and I want to keep that.
Kuhn:There have been a few quite eminent Ammerican scientists who have been very much concerned with the problem of the relations of science and religion. The Comptons in particular, Millikan however also wrote on this subject; I'm not sure who else did. At a time when I would think most European scientists had long since either abandoned any greatly significant religious feelings or at least had compartmentalized them so that this was no longer a live issue, one does find American scientists of distinction paying a great deal of attention -- Arthur compton was to the end of his life. I wondered how you would have felt about this, but also your sense of the significance of that either as an aspect of or influence on American science?
Kemble:I always thought that both Compton and Millikan were very naive. They were not either of them men in whose philosophic judgments about science I would place any particular reliance. They had great ingenuity and Millikan's chief characteristic to me was always the fact that any time he found out something new, he was sure of his interpretation and extremely enthusiastic about what was just around the corner. Perhaps two weeks later he would have changed his mind and be off on a new track with the same abounding energy and enthusiasm. There's a great difference between scientific creativity and philosophic judgment in the area of the sciences. I think mostly the experimental people aren't concerned with philosophy. The theoretical people can be. That's a different story.
Kuhn:You have no reason to suppose that attitudes of this sort have ever been an issue for physicists? Would you take it that most of your own contemporaries in the sciences had taken rather the same attitude toward the Millikan-Compton writings as you had?
Kemble:Yes. Of course, I'm more conscious of this problem, think, than most American scientists. In the back of my head. there has for quite a little time -- not since when I was a young man -- the strong conviction that religion ought not to mean a system of beliefs. It's the identification of religion with a system of belief rather than with a way of living which seems to me what's all wrong with it. I would suppose, though I haven't read much of what either Millikan or Compton wrote, that they took it to be a system of belief, accepted that, and their personal idealism could only find expression that way. My personal idealism think, is perhaps of the same order of magnitude but it has to express itself in a different way. I'm interested in rationality wherever I can find it, and in seeing a little more of it within the religious life of this country.
Kuhn:You mentioned that you had yourself had a couple of years of German at some point. I wonder what other languages you yourself had studied? Would you also tell me about the role of language education in the science curriculum more generally in your experience with an eye to the extent to which foreign literature was accessible to American scientists in this period?
Kemble:When I was a graduate student there weren't good books of American authorship. Our better texts had to be either German or English. It just happened that those I first got hold of and began to use were German. After I graduated from Case with a bachelor's degree, I taught for two years at Carnegie Tech. to earn some more money. In getting ready to come on to Harvard I divided the summer between Christiansen Theoretische Physik and jeans' Dynamics of a Particle.
Kuhn:Christiansen was Bohr's teacher. He [Bohr] used that book too, though I think he used the Danish.
Kemble:I had bettor command of German than French and the German texts all used a simple vector notation that made sense to me. I have all my life been irritated both with the English and the Americans for going off in other directions and it seems to me unduly complicating their books.
Kuhn:When did you get French as a --?
Kemble:I had a year of French in college, and didn't learn very much. As a graduate student I bought Pell, a three-volume book on Rational Mechanics and read some of that and became familiar with French. That scientific French is pretty easy to read. I never spent much time on it; I can't speak it now.
Kuhn:Was there a German or French requirement for the Ph.D. when you were a graduate student?
Kemble:I think there was, but I don't remember my relation to that requirement.
Kuhn:Was it your impression that your contemporaries were really following the German literature?
Kemble:I think I read more German than the other graduate students did. I got started that way. Abraham-Foeppl in electricity and magnetism was the one that I used in preference to Jeans. I got Voigt, Lehrbuch der Kristallphysik. They're still good books.
Kuhn:Not very readily available though.
Kemble:I think nowadays with so many good American texts competing for the students' attention I suspect that our students don't get anything like the practice with foreign languages that I had.
Kuhn:You speak of American books rather than English and American books together. I notice also some sense in which there seems to have been an effort in the teens and the twenties to bring out American books. I'm not sure that they would have been duplicating English books, but at least to conceive American textbooks as being a desideratum independent of other books even in the English language. Is that true?
Kemble:Certainly American texts did develop rapidly in increasing numbers during this period. There was a lot of excitement in physics, and some kind of an urge to publish spread abroad in the land. I think there are very few English texts that got into general use. There was, of course, Love's Theory of Elasticity, I see as I look up at my shelves, and Jeans' Electricity and Magnetism and Dynamical Theory of Gases -- yes,these were used.
Kuhn:How about Richardson?
Kemble:He was widely used. I never had a copy of Richardson.
Kuhn:Was J.J. Thomson widely read?
Kemble:Yes I would say so. I noticed that you asked Van about a paper that Lyman submitted to the Phil. Mag. Lyman having been in Cambridge and knowing J.J. Thomson was a natural connecting link, and I had one paper published in the Phil. Mag.
Kuhn:The helium one?
Kuhn:What journals would one suppose that a young American physicist graduate student or recent doctorate would follow with regularity as against looking up a particular paper in one?
Kemble:I'm not sure. I suspect most graduate students don't follow any journal with very great regularity. I know I kept track, not only in a general way, of the Physical Review. Nature was very much read, and the Philosophical Magazine and the Zeitschrift fur Physik when this began to come out. The Annalen I referred to quite frequently, but I don't think I kept as close track.
Kuhn:How about the Philosophical Transactions?
Kemble:I think by some accident or other I didn't keep track very much of that.
Kuhn:Do you suppose that most of your contemporaries also followed the Zeitschrift Fuer Physik fairly closely?
Kemble:Probably less so than I because this was the place where the theoretical atomic physics was coming out mostly.
Kuhn:You just referred to the helium atom paper in the Phil. Mag., and talked a little bit about that last time, but on looking back I discover not very much. I wondered whether you remember how you got into that problem in the first place? That turns out to be a great big problem, that presses and oppresses many Europeans. Bohr worked a lot on it with Kramers --.
Kemble:This was in the days of the Bohr theory, of course, and it was intended to fit the Bohr theory in with that. I don't know what got me to thinking of these two crossed orbits. I do know that this is one of the papers that I really regretted afterwards because I didn't have sense enough, in my excitement over having it come out as well as it did, to go over and talk to Saunders about it. Saunders would immediately have told me the whole story, I would have recognized that he would have been right that the normal state of the helium atom didn't belong in that series that I was putting it in.
Kuhn:That was a fairly influential paper. It was relatively widely cited in the literature.
Kemble:Yes, I know. People at various points have given me credit for the paper that I don't fell is coming.
Kuhn:Do you remember how long after the paper you did talk to Saunders about it? The whole unraveling of the helium series, and the recognition of the parahelium and orthohelium destinction was itself just humming in this period. It's not out of the question, for example, that when that paper was written, if you had asked Saunders, he wouldn't have said that and would have only have said it later.
Kemble:I know that Saunders has said to me, "why didn't you come to me and talk about it?" I think he was always shocked, and I was shocked myself -- this is the young man with the bit in his cheek.
Kuhn:as of the time you did that paper -- would it be right to say that this was for you a problem to be solved. but that you had very little sense of the extra special difficulties with respect to quantum mechanics that the two-electron problem was presented? Two years after this, a lot of people were convinced that you just can't do it at all, and there's something the matter with quantum mechanics.
Kemble:I'm afraid I can't say anything definite about this. I know that at that time I felt you had to try, use all your ingenuity. There was something crazy about orbits that could be stable anywhere. I couldn't tell without trying whether it could be fitted in.
Kuhn:After the war you did with Van this paper on specific heat of hydrogen. In the introduction to that paper, there is a brief historical description of background which refers to your thesis and the fact that because of the war it was never published and meanwhile Reiche had come out with a theory; so that this is now an extension which takes into account the expansion of a molecule on rotating which the Reiche theory does not. The whole implication of that brief introduction is that you had in your thesis talked about the specific heat problem. I thought you had told. me yesterday that it had. been entirely band spectra, that you hadn't dealt with specific heat?
Kemble:What had previously been published about my thesis did not have anything to say about specific heat, and I don't think the thesis did, although it had to do with the distribution of angular velocities. It was very close. One thing ought to be mentioned. Pretty early in my career --. As soon as I found out about it, I bought the Solvay Conference report by de Broglie. This was a 1911 Solvay Conference with a lot about specific heat and equipartition. I bought it January 1916. My thesis was published in 1917. It was this kind of thing that was largely behind the work of the thesis.
Kuhn:Were there any particular reactions to your advanced papers? You got pretty satisfying results. Did this seem to you that other people had pretty well tied that problem up?
Kemble:We didn't have any realization of how much further complication was going to appear in it. As I remember, our results gave us a specific heat curve that didn't have a hump in it. That was the important thing. I should say that in terms of the available information at the time, it seemed like a pretty good first approximation to a satisfactory theory. As I remember, my own interests went off in another direction about that time and I didn't follow along with what Van was doing.
Kuhn:This must be about the time that you got fairly heavily involved with the molecular spectrum committee. I'd be very grateful if you'd tell me something about that committee. What was the function of the National Research council Committee on molecular spectra? Just what was the conception of this job, how did it fit into the development of American or of international science?
Kemble:I take it that this was a subject that was alive, seemed to be moving rapidly, and that the number of people in this country actively engaged in it was not very large. We didn't have any Encykloptaedie and the National Research Council thought that there was no Reviews of Modern Physics. It was the feeling that we needed something like the Reviews of Modern Physics within the Physical Society that led to appointing a committee to make a review of this subject. 'This will get together the members, who ought to be taking together; what they accomplish in the way of producing, a report would be worthwhile and ought to push the subject ahead.'
Kuhn:The assignment of the committee from the start was to produce a report that would be a survey of the field and of quantum theory?
Kemble:I think that was the intention. I had nothing to do with this. I was the junior member of the committee appointed by the chairman, because of this other work that I had done. Bridgman, if he were alive, would know about how the thing had got started. Of course, there were in those days a whole series of reports on various topics by the National Academy, or perhaps, I should say, the National Research Council. I don't know. This one by Compton and Mohler that was so important; there two people were given the job. Haw many other cases there were of reports of this kind --.
Kuhn:There were quite a few of these -- a large proportion of the Bulletin of the National Research Council, and these come out fairly rapidly.
Kemble:It seems to me to be an anticipation of the Reviews of Modern Physics.
Kuhn:How did a committee of the sort you were on function? Did the members get together a good deal or was it done largely by correspondence, or --?
Kemble:We did get together when we could. Meetings of the Physical Society -- I couldn't say now how often this was. A great deal of it was done by correspondence. My correspondence with Birge was the principal corres-pondence there was because Birge and I were certainly the most active members of the committee. This would bring out how often the meetings were. As I look back on it, there was a great deal more excitement in this than in the later committee of the physics teacher's association -- the Coulomb law's committee. But I think I put in as much sweat and energy over the Coulomb law's committee as I did over this earlier thing.
Kuhn:What function did Lee Page have on that committee? I may ask this question in part out of ignorance of Page's work. I of course, know him as Page and Adams principally. Was he really at all involved with spectroscopic problems? Or was he there as a consultant?
Kemble:No, he was just a teacher of theoretical physics, a senior teacher of theoretical physics at one of our better universities. He was interested as a spectator I would say mostly. I may be playing down his role more than I should, I'm just candidly reporting the state of my mind. As I look at the report now, with the table of contents in front of me, each of the chapters represented essentially the work of the individual who wrote the chapter with the introductory on quantum dynamics by Page being something that was really common property to us all. It was a (???).
Kuhn:I get the impression from some of the things you said that there was among American physicists in the middle 20's, 1923-1926, very little of that sense of acute crisis in quantum mechanics that was very prevalent at both Goettingen and Copenhagen. I think people at Munich were not so convinced, but still aware that other people felt this way. Dirac felt this way, although I'm not at all clear that Fowler did.
Kemble:There was almost nobody to pick up a crisis of that kind and do anything with it in this country.
Kuhn:Regardless of whether there was anybody to do anything with it,the question as to whether people really felt that there was something fundamentally astray --. I don't mean just being aware that there were a lot of unsolved problems that nobody had learned how to manipulate yet. People had pretty well decided that these problems were just not going to come out by any sort of better, higher approximations or something of the sort -- that the problems go right smack to the root of the matter.
Kemble:I think everyone who thought about Bohr atoms with 25 electrons in them was bound to be very skeptical, but the difference, I would think, was primarily just in that this was outside the range of activity of most physicists over here, something they heard about but they were not participating in. They weren't in the game.
Kuhn:How about you yourself? I can perhaps sharpen the problem a little. You say that anyone who thought about a Bohr atom. with 25 electrons was pretty skeptical, but how about the problem of a Bohr atom with 2 electrons -- the helium problem and the hydrogen problem? It's as much as anything else, though there are other sources for this, this feeling that this one has just gone totally, to pot. If one can do the hydrogen atom, and if it's the techniques or anything resembling the techniques that's at fault, you ought to at least be able to get a reasonable approximation out of helium.
Kemble:[Long pause] All I can say is that I expected something new to come along to take place for some time before it did. But it didn't cause me anxiety in one way, because I was sure that I was not the person to solve the problem as world-shaking as that -- and I wasn't. There were no --. At Copenhagen, Gottingen, and Munich, there was a group. The personal interactions there were present. My only interactions here were the interactions within this molecular spectrum committee, interactions with my own students, and discussions occasionally in colloquium which didn't really actively involve other senior members of the faculty. There were two exceptions, of course, which I should mention, both of which I viewed with considerable distaste. Both Birkhoff and G.W. Pierce very late tried their hands and trying to get classical mechanics and electrodynamics to take care of the problem of the quantized atom, and --.
Kuhn:Did they carry any physicists along with them?
Kemble:I don't think so. The point was that both of them were outside the area of the activity. They had no real notion of the tremendous complex of different things that were all being illuminated together and being related to each other by the progress that was being made. This early area, where, if you're thinking about fundamentals, carrying things through to be right or not right, it seems to me, you are apt to lose sight of the very great interest people had in taking the ideas that were available and using them to correlate such things as specific heats, the band spectrum, the dimensions of the molecule, absorption, emission, and all these things. There was an awful lot of things that you could do. And trying to get the general organization tied up with the period ical table, and so forth--this was exciting business. Because it was exciting, I was content to not hammer away.
Kuhn:You were around here at the time of the Compton effect, the time of Duane's stand against the Compton effect. Can you tell me anything about that? It must have been pretty hot around here for a little bit.
Kemble:Duane and I never did click very much together. I admired his experimental work, but for some reason or other we never really communicated very well together. I was aware in a general way of some of his theorizing, but it didn't appeal, and we remained aloof from each other. Of course, he was a great deal older than I. He came here as a mature man with a lot of reputation. He was not a mathematical physicist at a11 so our directions were kind of crossed with each other. The Compton effect certainly made a big impression. It was bowling over to think that the frequency of waves could be changed by impact with something else. It wasn't easy to accept but it didn't take very long to have it accepted.
Kuhn:Had anyone that you know of in this country previously taken the photons seriously?
Kemble:Millikan gave a talk called "The Unitary Theory of Light" at a meeting of the American Physical Society that was held at Case in Cleveland. I don't know when it was, but I would suspect that this was a full-length address. By reading this, you might get a pretty good notion.
Kuhn:Of course, Millikan himself had performed what were probably the decisive experiments in confirming the Einstein law.
Kemble:This was pretty clear, probably before he did his photoelectric experiments. I think it was after I had left Case --. Probably before 1917.
Kuhn:His experiments were out by about 1917. Would you talk a little bit about your early students, and particularly about Van and Slater and your impressions of them? How this bright group of younger theoretical physicists built up this country? They're in the first wave with you there
Kemble:They were very good men, there was no doubt about that, from the start. Slater came in as a kind of a boy phenomenon. He had already written some kind of paper connected with the bohr theory in high school which was sent ahead when he was being admitted. No, it wasn't a high school paper, but an undergraduate paper at Rochester. It wasn't scientifically sound or good but it still showed a good deal of originality. Slater, of course, was pretty impatient around Harvard in his early days as an instructor.
Kuhn:Impatient with what?
Kemble:I don't know. He felt that the department was too deliberate about the way it did things. Can't tell, he could have been impatient with me, as I was ahead of him in the theoretical line. I do remember that after he got his degree he was kind of dissatisfied and restless, and I didn't know exactly why, until he finally left and went down to Tech.
Kuhn:Did he do that of his own volition?
Kemble:Oh yes. Of course, Compton was coming up and there were visions of a new empire to be built. The Tech. department of physics had not been an aggressive modern department. It was obviously was going to be made that, and Compton persuaded Slater to go down there. There's no question but that I was extremely fortunate in getting men like Van Vleck -- Slater never did work with me, he did his thesis with Bridgman. But [C.] Zener was a top-flight person of whom I was very proud, and Hubert James. I didn't have very many, but the few who came along who wanted to do theoretical work were pretty good.
Kuhn:What was Bridgman's role as a theoretician with respect to modern physics? I've been told the story that when the first edition of Sommerfeld came out, he was in Germany, he immediately grabbed it and locked himself in the office and studied it carefully, and wouldn't let anyone else have it until he was through. If it were true, it would indicate that whether he did anything with the quantum or not, he was very much concerned with it, and might then have been involved in the department with it.
Kemble:I've never heard that story. As a matter of fact, there's stuff here on my desk -- Francis Birch and I are writing a memoir for the National Academy. A year ago I finished my part of it, which had to do with Bridgman's philosophical ideas primarily, Birch was writing up the experimental work, and we have something to do together. Birch has been slow getting his part of it together, but the question of Bridgman's failure to get into the swim of quantum theory was the thing that I took up someplace in a sentence or two. I think that he never really put his mind to the theoretical problem as a participant in the seeking of solutions. He never taught it: he was committed to his high pressure work. This had enough in it to absorb all of one side of his energy; his philosophical work seemed to be the other thing. He just never felt that the point had come for him to break away.
Kuhn:About your trip to Europe: you put in for a Guggenheim. Had you wanted to go to Europe for some time? How did this come about, how did the timing come about, and would you tell me something about what you did while you were there and how the trip worked out?
I don't know where the suggestion came from in the first place. I do know that my wife was in the hospital having her second child in November of 1926 when I suggested, "How would you like to go to Europe for a year?" She was shocked beyond measure, but she came round and we were always glad we did it. It came at a time when the work of the molecular spectrum committee was finished and I was ready to turn to something else. I had a sense of momentum that it would be good to go and visit some of the European places and get acquainted. with some of the people over there. They weren't coming here in any large numbers at that stage. Actually I went to Munich first, where I ran into Pauling. We spent about 6 weeks in Munich, and then settled down in Gottingen. I just carried on with papers I was writing at the moment. It gave me a marvelous sense of rapport with what was going on in the world. It was an enormous advantage to me. And yet I didn't attend regular lectures by anybody; my German wasn't good enough. I went to colloquia. This was in 1927. The new quantum mechanics was already out. I heard Von Neuman's first lectures; there were 8 lectures, I think he gave, on his new approach. I always feel very excited when I think of being there but it was mostly over my head, and. Max Born said it was over his head too. At that time Dirac was at Gottingen; Hogness at the University of Chicago was there. I think that's all there is to be said about it. I can't put my finger on specific things that I learned but it was part of my development.
Kuhn:Did you spend a lot of time while you were there trying to involve yourself with the new quantum theory?
Kemble:No, I didn't spend a lot of time at it. I read. a few papers. I knew that the Schrodinger theory was out. This attracted. me very much more than the matrix theory. Matrix theory was always a formalism that didn't appeal to my geometrical mind. My mind was always geometrical. The matrices didn't appeal to that; Schrodinger did.
Kuhn:Did you get involved in any debates on this subject in Gottingen, because, of course, Gottingen was all for the matrices and at this point thoroughly dead set against Schrodinger. That played no role for you?
Kemble:No, I didn't get involved.
Kuhn:Was it really the Reviews of Modern Physics [Volumes 1 and 2] piece that got you most deeply involved with the new quantum theory?
Kemble:Yes, I would say so.
Kuhn:How did that come about? Was that an assignment? Was that something you --?
Kemble:I suppose that the answer is simply that I was known to the editor as the chairman of this molecular-spectrum committee and had taken a leading part in that committee from a purely theoretical point of view, and so he asked me to make a report on this. It wasn't a very good report. I sweated about it and didn't know how to follow it through.
Kuhn:How long were you at it? It came out, I guess, in '29 and '30.
Kemble:About a year. It didn't seem interminable like the writing of the book did.
Kuhn:You yourself have been again and again much interested in the problems of the interpretations of quantum mechanics. I wondered in your own case how far back that went, and to what extent you had become deeply involved with those in the course of the book? Also what can you tell me about debates in this country about things like the statistical interpretation, the Heisenberg principle and so on -- when were those live issues? Clearly those issues stayed alive for you at times when they cease to be alive for physicists? Anything you could say about the pattern of development of that line of interest --. You weren't in Europe long enough to get to the Como Conference, were you?
No, I wasn't. I can't say very much. In writing the book I was trying to get my own mind straight, of course. My great good fortune was that I had had previous contact with Kramers, first of all, out at Michigan. The general idea of Fourier transforms and their relation to the theory I absorbed from him. This gave a theoretical point of view for the development that was what I needed, and. might have been very slow at developing for myself. Then on this philosophical side of things it seemed to me that Bridgman's operational point of view was heaven-sent. I rigorously refused to worry about whether or not I could understand anything about the theory except the way in which it predicted what we could observe. All questions of reality I thrust to one side. I was, of course, very much intrigued with the fact that the reduction of the wave packet, when you make an observation, means that you now substitute in your future computations, a brand new thing, and there was a discontinuous jump, and the discontinuous jump was what took place in your head. So a central philosophical thought that developed here was that these psi functions are the means by which we calculate what to expect. They are not to be considered, as we thought of electromagnetic waves, as force fields in the aether that are moving around.
Kuhn:Did you yourself start out by thinking of them in those terms? Schrodinger did, as you know. He found the statistical view extremely hard to stomach and lived his life in rebellion against it.
Slater was one of the first people to write a paper about the statistical view over here. I don't know whether his paper came out. I would say at the time I was thinking the same things how this was. I think very early, while I was still getting organized, the statistical point of view had been suggested, and I accepted it, and used it troughout the writing of the book to explain that the predictions of the theory always involved ideally an infinite succession of experiments, and that you must always remember that what you knew conditioned the way in which you had set up the experiments. So it didn't seem too bad that the wave function should have this property of unreality -- I call them "subjective states" in the book. Nobody else liked that, and I got into a long correspondence with Margenau. He wouldn't be convinced, and you'll find that here [in Kemble's correspondence]. We corresponded for quite a while, and I thought we were converging on agreement because he seemed to go along with what I was saying, and then at the end suddenly the conference terminated, and he was back where he was in the beginning, as far as I could see.
Kuhn:Was there anyone else you remember having long discussions of this with beside Margenau?
Much later, since the war, had correspondence with Victor Lenzen. I wrote this on -- "Reality, measurement, and the state of the system in quantum mechanics," [Philosophy of Science 18, 273-299 (1951)] It was in that connection, shortly afterwards, that I got to talking with Lenzen. Lenzen had been talking about causality, and wrote his little book. He said. when I first pushed this paper in front of him that he had got at first the idea that I had the best analysis of this subject that he had seen. Then it came out in the end that I had been --. There are two ways in which an experiment is completed, to be said to be completed. One is, when the macroscopic event that is triggered by the macroscopic event reaches your brain, then you know, then you perceive, and draw inferences. I wanted to talk about a succession of measurements in which, for example, you might have an electron that came off of a filament, and you accelerate it with a field, and it goes through a slit. Then you bend it in the magnetic field, and it goes through another slit. You could have all sorts of successive events -- it goes through various slits. Finally, you make an observation that discovers this thing someplace, or discovers an event that must have been caused by it. I felt that each of the successive slit passages is something that I wanted to call an experiment, a measurement. Of course, it doesn't reach the brain at that point. To say that when we're dealing with situations of this kind we should successively reduce the wave packet at each slit and say from here on 'What I'm talking about is what went through that slit, and these other things that got reflected off to one side are irrelevant.' Lenzen felt that really the only observation is at the end. You musn't do that, you must keep all of the waves in the picture until you make the observation. It was this kind of a difference. I think there's a real point here, and a weakness about my paper in that I was, in effect, saying that if the ordinary lingo, language, that we are accustomed to using as experimental physicists, is to be justified, then this is the way you do it. But maybe that lingo shouldn't be justified; but that's another story, and I don't know what the answer is.
Kuhn:In your earlier thinking about this, did you read, or had you already read, and thought much about things like Heisenberg's original uncertainty paper, and the Bohr papers on this subject?
I had, of course, seen Heisenberg's original paper and I had his little book. I can't tell you how much of this I read. I didn't read all of it, for sure. I never was happy about Bohrls complementarity. I think that one of the things, -- this has nothing to do with your real problem but to do with the state of my mind as I have gone on. I think a lot of people in this country have thought of Bohr's principle of cornplerarniarity as meaning a complommtarity between waves and corpuscles. That's not what the principle really said. I had the notion for quite a while that this was what Bohr meant to say and I objected to it that there is no symmetry between the two pictures. One isn't the right and the other, the left. The waves are the statistical picture and the corpuscles are the picture of individual events which are not statistical, and you can't balance these against each other at all, or in any satisfactory way. I hadn't heard anyone else talking this way, but I still have the feeling that this is correct. I've also been very much bothered by the inability, lack of strength and speed to be in the fight when the tendency has been for Schwinger and his colleagues to emphasize the reversibility of the wave equation and my saying always to myself that you can reverse the Schroedinger equation all you please, but you can't reverse the reduction of the wave packet. And in a sense time is not reversible. You can't talk about these things without writing an awful lot of words and spending an awful lot of energy, and I've never had the energy to push at my own insights.
It's a pity, because some of these points I think could use more pushing
Session I | Session II