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

Oral History Transcript — Dr. Robert Dicke

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. Robert Dicke
By Martin Harwit
At Dickeís office in the Joseph Henry Physics Building, Princeton University, Princeton, New Jersey
June 18, 1985

Oral history interviewee photo
open tab View abstract

Robert Dicke; June 18, 1985

ABSTRACT: Comments on parents and teachers; schooling in Rochester; studies at University of Rochester and at Princeton University with comments on faculty and fellow students; thesis collaboration with John Marshall; Victor Weisskopf, M.I.T. and the Radiation Laboratory during war, microwave techniques applied to atomic physics. Return to Princeton after war, Angular Momentum of Radiation; 1957 and the start of cosmology and relativity publications; Eotvos experiment and Machís principle; discussions of own and othersí experimental work; big science; George Gamow, Ralph Alpher, Robert Herman; paper with P. James Peebles Nostrums and Conundrums; National Academy of Sciences; funding in science.

Transcript

Harwit:

Bob, I remember reading some of your background history and that you were born in St. Louis. I also read all the interview you gave to Paul Forman and Joan Bromberg, but they didnít ask you about your early growing up, what it was like, what your family was like, and where you gained your early interest in physics. Perhaps you could start at the beginning.

Dicke:

I guess one starts where heís born, which you have already covered. My father was trained as an electrical engineer. And after working for a number of years in various engineering-type positions, he decided to try out for to be patent examiner in Washington. He was taken on as patent examiner, and then went on to law school, got a law degree, and ended up as a patent attorney — practicing patent attorney. He was always interested in science, and I think this is probably where I developed some of my interest too. We moved to Washington, D.C., when I was quite young and then to Rochester, N.Y., where I grew up. I was probably four or five when I arrived in Rochester; I donít remember it exactly. I ran through the usual normal, boyish scientific Interests — collecting butterflies, insects, looking at stars — the usual run of things.

Harwit:

Who was this stimulated by? Was it your father, your mother or any brothers or sisters?

Dicke:

Probably primarily my father, although I think the Interests developed more from my own reading than that. I enjoyed reading scientific things.

Harwit:

Did you have any siblings?

Dicke:

Yes, I had a brother who died at a quite young age.

Harwit:

Did your father, then, at Rochester, work for any of the larger firms?

Dicke:

He worked for the General Railway Signal Company as a patent attorney for them.

Harwit:

What about teachers? Did you have any that you wereÖ

Dicke:

Yes. My chemistry teacher in high school was supportive in my interests.

Harwit:

But you had already previously been interested in scientific things.

Dicke:

Yes, I had already been interested. Also, I developed atÖitís quite interesting my interest in mathematics first started to develop after junior high school. In fact, in the elementary school I couldnít be more bored with anything than arithmetic, which I didnít do very well in, and it was only after I started my contact with algebra and geometry, and so on, that I became interested in mathematics.

Harwit:

Were there any teachers there who particularly...

Dicke:

Yes, we had quite good teachers at the high school, not so good at the junior high level, but at high school they were quite good.

Harwit:

This was high school in Rochester itself?

Dicke:

Rochester, New York, public school system.

Harwit:

Which high school was it?

Dicke:

It was West High School where IÖwhich was a senior highÖ and Madison High was a junior high school. One thing I remember quite clearly is developing an interest in calculus and studying it on my own, which is now extremely common in high school, and high school students n seem to take calculus as a matter of course if they have any interest at all. It was a very unusual thing at that time.

Harwit:

Yes, very few high schools offered calculus in those days.

Dicke:

Yes, I think very few indeed.

Harwit:

So, that would have been in your senior year?

Dicke:

No, I read calculus. I did it completely on my own, and I read it, I think, all through the high school years.

Harwit:

Where did you get the books for that?

Dicke:

Library mainly.

Harwit:

No particular book that you remember?

Dicke:

Yes, there was one favorite book, and I canít remember the title, Iím afraid.

Harwit:

So then when you graduated, what influenced you to come to Princeton?

Dicke:

Well, I didnít go straight to Princeton. I went to the University of Rochester first for two years, and it was a bit of blind luck that I went there. My record in foreign languages was not particularly good in high school, and I thought I would be aiming too high to try to enter the University of Rochester. My chemistry teacher convinced me otherwise; so I entered there, did quite well, and enjoyed the courses. My getting into physics was another accident. This came about in the following way: In heading for college I thought naturally of learning something I could earn a living with. It never occurred to me I could earn a living with physics, which I would have liked to have studied anyway, but I was going to enter as an electrical engineer as my father had done. When I looked over the curriculum for the freshman year for the engineering department, the courses looked so boring I thought I would try something really fancy and enter as a physicist, spend a year as a physicist, then transfer to electrical engineering. Obviously, I never made the transfer. And one of the reasons I didnít, I think, were the excellent lectures I had from Lee DuBridge, who was the chairman of the department there, and taught the elementary course marvelous lecture demonstrations. Then Fred Seitz, he was teaching a senior course in theoretical physics. I took that as a sophomore; enjoyed it very much.

Harwit:

Was Victor Weisskopf there at the time?

Dicke:

He wasnít there at the time. He cane a bit later. He was there when I was a graduate student there. In the case of the course I had with Fred Seitz, Fred was responsible for my going to Princeton. He had just come shortly before from Princeton as a fresh Ph.D., and he thought I had some possibilities, so he wrote to Ed Condon suggesting they offer me a fellowship, which they did.

Harwit:

As an undergraduate?

Dicke:

As an undergraduate. I came to Princeton and spent the last two years here.

Harwit:

Was it difficult to adjust?

Dicke:

Not really. It was an exciting time at Princeton. Seemed like my junior and senior years here — everyone was here. There was Pauli.

Harwit:

Pauli came through, but was he actually here for a longer period of time.

Dicke:

Pauli was...I donít recall whether he was here the whole term or not. He actually was here for a year or two, but I think it was later, during the war, something like that. But at that time, well, letís see, the ones that were here for an extended time, one was Rabbi.

Harwit:

I see. He was not at Columbia at that point?

Dicke:

He was visiting professor here. Of course, Einstein was around.

Harwit:

Did one see him often on the streets, or did he come to colloquia?

Dicke:

Well, you used to see him on the campus. The Institute for Advanced Study didnít have their own buildings at the time. They were borrowing space in Fine Hall from the mathematicians, so Einstein was wondering around in that general area a lot. Fine Hall still is the title of the mathematicianís building here, but it was a different building at that time.

Harwit:

Do you remember any other professors you had at the time in courses?

Dicke:

Yes. H. P. Robertson gave an exciting series of lectures on general classical theoretical physics, fluid dynamics — quite interesting things. I think I sat in on a course that Wigner taught on quantum mechanics. I took a very interesting course in optics from Alan Shenstone. One of the courses I took...I tried to take at least one course in each of the major sciences, and I succeeded, I think, except for geology, which I never took. I took a single descriptive-type course in astronomy taught by John Q. Stewart, and the outcome of that was that my first publication was actually in astronomy.

Harwit:

I saw that.

Dicke:

Did you notice that?

Harwit:

I saw that, yes, on globular cluster statistical mechanics.

Dicke:

This came about in sort of a funny way: I enjoyed the course very much. Stewart assigned a term paper for each of us to write, and I thought of this idea and I wrote a term paper on that. He was impressed with this, and thought it probably should be published. He showed it to Henry Norris Russell, and he suggested that I send it into The Astronomical Journal, which I did.

Harwit:

It was a nice paper, actually — applied statistical mechanics. You had all the references in the end. Had you had a chance to look at those, if you remember?

Dicke:

Yes, I looked at the Emden Gaskugeln book was in the library there. I only found out years later that Emden was related to Schwarzschild, his uncle or something like that.

Harwit:

Thatís right. A brother-in-law.

Dicke:

I was thinking of Martin Schwarzschild.

Harwit:

I was thinking of his father, Karl Schwarzchild. Did you have any courses with, or did you know anything about Harnwell?

Dicke:

I started out, when I first arrived as a junior, signed up for a course from Harnwell. After the first lecture or two, I decided the level was wrong and dropped out.

Harwit:

Was it electromagnetic theory?

Dicke:

Yes, it was electromatic theory?

Harwit:

What about Ladenburg?

Dicke:

Ladenburg was the predecessor of the chair that I occupied for years that has that melodic ring -ó the Cyrus Fogy Brackett Chair. He was at that chair, and one of the great things about that chair, there are only two like this in the department, one the mathematicians and mathematical types occupy and the other, Arthur Wightmann has it now. Thatís the Fine chair. Then there is the Cyrus Fogg Brackett Chair. Both of these chairs by deed of gift donít require any teaching. It says all you have to do is give a few lectures in the subject of your choice, something like that. So the result is that when I finally occupied it, which was about 1956 or so, I did rather little teaching after that, which has both its bad and good points. Youíve got to keep at teaching if youíre going to be very good at it.

Harwit:

Yes. Well, also I suppose, the fringe benefit of teaching is that occasionally students ask you questions that you donít know how to answer.

Dicke:

Yes, I think this is extremely important. Iíd Ďlike to say that something you donít really understand, the best thing is to give a course in it.

Harwit:

Thatís right, yes. What about Walter Bleakney?

Dicke:

Bleakney. I had a course from him which I enjoyed very much.

Harwit:

Do you remember what it was?

Dicke:

Yes indeed. We had a laboratory course. I think it was in the senior level, could have been junior. No, Iím pretty sure I took it as a senior. Essentially all of the physics majors took this course, it was general laboratory techniques, vacuum, and that sort of thing, mercury diffusion pumps. A number of experiments we did, we had a Millikan oil-drop experiment. Wolfgang Panofsky (Pief) Wolfgang was in the course with me. As a matter of fact, I took this course a year ahead of the time I should have, so I must have been a junior, because Pief was a year after and we were in the same course. You know he had a brother and his brother was in this course too.

Harwit:

Yes, I see. His brother is a historian, I think, isnít he?

Dicke:

No, heís a meteorologist, I think. His father was a very well-known art critic.

Harwit:

Maybe thatís what I was thinking of. Iíve known that one of the Panofskyís was a humanist.

Dicke:

Yes, right. His father was on the faculty of Institute for Advanced Study, and wrote books on art history.

Harwit:

Do you remember any of the graduate students who where around at the time.

Dicke:

I donít remember Hofstadter very well; I think I do remember him. One I remember quite well was Ruby Sherr who was on our faculty here; he just retired a couple of years ago. There was chap by the name of Sampson I used to talk to occasionally. I donít know where he went; I ran into him a number of times.

Harwit:

Did you remember Bob Herman who later on crossed paths with you again?

Dicke:

I have only a vague recollection of him. I remember he was here or having seen him.

Harwit:

How about the mathematicians? I guess there was a pretty distinguished group of people here — von Neumann, Eisenhart.

Dicke:

I think I took only one...No, wait a minute now, let me just think for a moment. I probably took two courses in mathematics in the junior and senior years. Iím not absolutely sure. I remember, though, one was a course in algebra it was taught by a topologist.

Harwit:

Solomon Lefshetz, perhaps?

Dicke:

No. It starts with a ďt.Ē [Albert Tucker]

Harwit:

It doesnít matter too much.

Dicke:

It doesnít matter. I had a course in analysis and I think I had it here; and I canít remember who taught it.

Harwit:

Do you remember being influenced by students or faculty in particular directions or having your interests aroused?

Dicke:

Well, I remember being tremendously impressed by the library here. I spent a lot of time reading.

Harwit:

On your own?

Dicke:

Yes. A tremendous collection of books.

Harwit:

So you had time from you studies to...

Dicke:

I had time. I took time. I tended to concentrate pretty much on courses in mathematics and sciences, though. I admit, I didnítí have a lot of writing and reading to do in connection with the humanities. I remember when I arrived here as a transfer student, I made out a course card which was essentially all mathematics and physics, but I was required to take this to the chairman of the department and get his signature. The chairman at the time was Harry Smythe, of the Smythe Report. I remember ever so well. Iíve told him this myself, that he looked at my course card, he looked at me, he looked at my course card again, and he looked at me and said, ďDicke, donít you think you should take something besides mathematics and physics? He urged a course in music on me which I greatly enjoyed.

Harwit:

Oh, I see, good. Were there any non-scientific courses required?

Dicke:

Things were very flexible at that time. Education seems to run through cycles — great freedom and great restriction.

Harwit:

Now, when the time came to apply for graduate schools, what decided you to go back to Rochester then? Was it that it would have been less expensive than being elsewhere?

Dicke:

Well, I had a very good impression of Rochester from the two years I had been there. It would d be easier to live at home and that one of the factors, I was a little reluctant to leave Princeton because I was enjoying it very much there too; but anyway, I didnít regret the choice.

Harwit:

Were there at that time do you recall whether it was easy or difficult to get graduate-student support in the form of an assistantship or fellowship?

Dicke:

There seemed to be teaching assistantships enough to go around at Rochester. I donít know anybody that wasnít supported in some way. There were a few fellowships. In fact, there were fellowships that were provided by industry. Leroy Apker, for example, I think, had a fellowship supplied by G.E.; and he later on went to G.E. Heís been dead a good many years already — died young. But thereís a fellowship known as the Leroy Apker Prize that the American Physical Society gives every year. The money was put in by his wife.

Harwit:

Would you have had a chance to stay at Princeton?

Dicke:

I think so; I could have stayed here, I think.

Harwit:

How big were the graduate student bodies at that time? How many students typically were at Princeton or typically were at Rochester, if you remember?

Dicke:

Well, the total undergraduate student body was probably two-thirds of what it is now in size. Physics Department concentrators were probably less than half, maybe a third of the number we have now.

Harwit:

Which is?

Dicke:

Well, at that time six or eight in a year, at most, I think.

Harwit:

And graduate students at Rochester, do you remember how many fellow graduate students you had?

Dicke:

I can only give a wild guess. I donít think it could have been more than 20, 25 maybe.

Harwit:

Now, you got through within two years which seems remarkably short now. Was it remarkably short then?

Dicke:

It was. The reason was that I had at Princeton Iíd taken primarily graduate courses. I had gotten a lot of the graduate work out of the way — course work. Then another graduate student and I had thought of a nifty experimental problem to tackle, which was to look at the energy-level structure of stable nuclei by bombarding with protons, looking at the in elastically scattered protons.

Harwit:

You had an accelerator there?

Dicke:

There was a cyclotron. One of the first major cyclotrons in the East was at Rochester. This is one of the, I think, Iím not absolutely sure of this, but I think this was a condition that DuBridge imposed on the university in agreeing to go there was that they support with their own money in construction of a major device. Inelastic scattering had been used before but in a really ancient technique, using plate tracks to count protons, and so on. We set up a counter system, and in short order cranked out a substantial number of — well short order was order of a year to do the whole job quite a few energy levels In a number of stable nuclei. John Marshall was the other student. He took half of the results and I took half of them. It wrote this up as two theses. We had the agreement of the faculty this would be alright to operate this way.

Harwit:

Did you have supervision from anyone?

Dicke:

No, In fact, we had relatively little advice on this. Did it ourselves.

Harwit:

Now, you said that by that time Weisskopf already was...

Dicke:

Weisskopf was there. Mien I arrived as a graduate student, he was already in residence.

Harwit:

Did he interact much with the students?

Dicke:

Yes, quite a bit, primarily with a few students that we had with theoretical Interests. One of them was...Itís the same name as the famous mathematician. Oh devil, what is it now? His father was head of the Courant...oh, his name is Courant. (laughter) [Ernest Courant]

Harwit:

His father was head of the Courant Institute.

Dicke:

There were a couple of others. Also, I talked with Vicky quite a bit, particularly after we had results from our scattering, trying to understand what they meant.

Harwit:

Later on, when I was a graduate student at N.I.T. and Weisskopf was there, he always seemed to be one of the people on the faculty who was able to communicate best with the graduate students.

Dicke:

Furthermore, Vicky had an almost unique ability to reason physically and to express his ideas in a semiqualitative way. I mention this that I had learned this technique from him, of how you make order of magnitude estimates involved in quantum features. I had said that I had picked this up from him. At that point he said, ďWell, you know where I learned it? I got it from Ehrenfest.Ē He had Ehrenfest as a teacher, and Ehrenfest had this technique.

Harwit:

I guess there are theorists with whom one can talk and theorists with whom one canít.

Dicke:

Essentially canít communicate all.

Harwit:

That has very little to do with the quality of their theoretical abilities.

Dicke:

Youíre absolutely right.

Harwit:

Now, from Rochester, then, you went to the Radiation Lab, and Iím going to ask you relatively little about that, since Joan Bromberg and Paul Forman already talked about that period. But I did want to ask you a couple of things: One of them is, I believe you were working in a division or branch that was headed by Ed Purcell.

Dicke:

Right.

Harwit:

In a sense, I suppose, for you, this was a what would be now called a postdoc appointment almost, except that it was a wartime appointment and therefore a more serious position. Did you find that you interacted a lot scientifically with Ed?

Dicke:

Well, we were all very much concerned with the problems involving the war. The time I had arrived there the war hadnít even started yet, so the scientific questions we were involved with always had an engineering slant to them. The amazing thing about that institution...thereíre a number of things that were amazing...one is how young. Everyone was...they were children practically.

Harwit:

Yes, yes, almost everyone was under thirty there.

Dicke:

Yes, except for the few leaders like DuBridge and Rabbi and so on. Even group leaders were very young. The reason I was assigned to that group, I think, or the reason I went to the Radiation Laboratory at all was my connection with Du Bridge, who had remembered me from my undergraduate days, and also my graduate student days here. And in fact, Du Bridge had decided that he wanted me to go to M.I.T. as soon as I had my Ph.D. material put away. I hadnít even taken the final exam. He said I should finish the thesis first, which I did, and went up there and then I came back to take the exam. My other companion in the thesis, John Marshall, was invited to go to the Manhattan Project, so he was there all during the war.

Harwit:

And he stayed on, then, in bomb research later on too? Is that right?

Dicke:

Yes, he stayed with Los Alamos, I think, after the war, continued in this business, and still is as far as I know,

Harwit:

Sometimes Iím struck by the style of work similarities that people like you and Norman Ramsey, Ed Purcell, Charlie Townes exhibit a strong theoretical ability applied to incisive experiments, and. generally related to a sophisticated interaction between radiation and largely atomic matter. Iím over generalizing obviously, but the four of you have all somehow really done experiments that nobody else I know has thought of or done, and coupled them to a fundamental theoretical understanding. Dad I was wondering whether you thought that the radiation lab had any influence, whether the interaction there between...I think, all four of you were there, isnít that right?

Dicke:

Yes, I donít know that...I would say this, that the microwave techniques we developed during the war there certainly influenced...were a support for all four of us, and we used these techniques in our research after that. As for the overall approach, my feelings have always been that you start with Idea and not with an apparatus, and the idea, you feel, is sufficiently important to warrant a lot of effort, then you develop whatever apparatus you need to crack it. The result is that it sometimes takes quite a while, but at least youíre working on something which you at least think is important.

Harwit:

But one can look at the alternative. It could have been that all of you would have become experts in microwave engineering, and then continued developing ever better microwave systems after that, without ever leaving the hardware realm or without becoming interested In fundamental atomic physics effects.

Dicke:

Many at the Radiation Lab did essentially stay in the engineering side of this, and I remember some years ago someone told me that he had been riding on a plane or something like, and his companion said, ďWhat happened to this fellow Dicke?Ē He never saw my name in IEEE or anything like; he just vanished from sight. (laughter)

Harwit:

I suppose the journals you read provide enormous barriers, separating the professions.

Dicke:

I remember right after coming to Princeton after the war, I was still thinking and speaking in electrical engineering terms. If one looks at the book that we produced, the Radiation Lab series, and you look at the part that Iíve written in there, youíll see thereís an electrical engineering cast to it. I talk about terminal pairs and heavy network theory.

Harwit:

Thatís right, but they all do. The whole style of that series is that way. I think thereís very little, if I remember it, thatís devoted to fundamental physics of the type that people like you developed in the decade following the war. And itís surprising, again, how fast these developments came.

Dicke:

There actually were several fairly fundamental things that I became interested in and wrote chapters on, the use of symmetry arguments in connection with waveguide junctions, for example. Thereís a chapter devoted to that. Another thing I found fascinating is that when I first made contact with this, it seemed there was a complete dichotomy between the electrical engineerís way of looking at things — currents flowing in wires, and networks, and branching, and so on. The way physicists would look at this with waves being reflected, and so I wrote a chapter which was essentially devoted to bringing these two ways of looking at the problem together and showing that theyíre really equivalent to each other. I found that rather interesting.

Harwit:

Well, that dichotomy still exists. Itís often difficult for people who think in optical terms to become used to radio astronomers who talk about antenna temperatures, and so forth, and the impedance of free space. Itís a different way of looking at it, and sometimes itís difficult to cross that gap. When you were writing up these Radiation Lab series chapters after the war, did you interact with Henry Guerlac? Do you remember him at all?

Dicke:

I must have interacted with him somewhere there, because I remember years later after the war being in the Cosmos Club one time in Washington, and he came up and introduced himself to me and said that heíd remembered me. I canít recall what contact Iíd had with him, though, later.

Harwit:

He was official historian of the Radiation Labs through the war, guess, and later on went on and became one of the leading historians of science in this country.

Dicke:

Heís dead now, I suppose?

Harwit:

He just died three weeks ago.

Dicke:

I seem to recall him as being fairly elderly when I saw him.

Harwit:

He was not quite 75.

Dicke:

Oh, he wasnít that old.

Harwit:

He had evidently had a number of heart attacks, and the last one was fatal. He was mainly a Newton and Lavoisier scholar, so one perhaps wouldnít have heard of him as a scholar of modern science history, but he became quite eminent — started the History of Science at Cornell. Is there anything else you like to say about the Radiation Lab before we come back to Princeton?

Dicke:

Well, I guess the only thing that might be of particular interest to you would be the invention of the microwave radiometer, how that came about, the use we made of it and that sort of thing. If you want to hear anything about thatÖ

Harwit:

Yes, I would. I would have probably come back to it, in any case, when we talk about the background radiation, but letís talk about it now; it makes sense.

Dicke:

Well, I guess one should start with the use of the lock-in amplifier I had been making at the Radiation Laboratory. Iíd been using this for...

Harwit:

The Dicke switch.

Dicke:

Effectively, yes. But the lockóin amplifier itself was usually, frequently credited to me as inventor, but it actually wasnít my invention. There was a paper which I saw in, I believe, Reviews of Scientific Instruments, published by Michels, who was professor of physics at Bryn Mawr, I believe. I thought this was a cute idea of synchronous switching, to pull a small signal out of the noise. I built circuits of that kind, and Iím trying to remember now the order in which things were done. Iím pretty sure I used them in other kinds of experiments, measurements on crystal rectifiers which I was involved with and things of that kind. Also, I was looking at antenna patterns for feeds with the idea that you could break the overall... At that time there was a tendency to treat the whole antenna as one huge organic thing that you twiddled everything at once. I decided it made sense to divide that into parts, the dish part and the feed part, and then examine the feed by itself.

So I made an instrument for looking at the phase wave fronts and intensities and that involved a lockóin amplifier. Then it just occurred to me, one day, that you could build a sensitive radiometer, if you used a wide band and synchronously switched the antenna between some radiation source and some internal heat source, with just some heat source of some kind. If the bandwidth was wide enough, you could, using this synchronous detection technique and averaging in time, get quite good sensitivity that way. I explained this idea to Ed Purcell, and he was quite taken with it; thought this was an interesting thing to try. I built one device incorporating some of his ideas; but it didnít work as well as I would have liked, so I redesigned it — the part that didnít seem to be working very well. That second thing worked quite well. I remember that the standard trick, technique for seeing whether it would work, was to hold a cigarette over the feed; the radiation from the cigarette would knock the needle off scale.

Rabbi heard about this and he wanted to see it. Rabbi and Ed came in one day and we were talking about sensitivity, and so on; and Rabbi was sounding very dubious about the whole proposition. He was smoking a big cigar and said, ďWeíll just check this,Ē and put the cigar up to the feed and the needle went ďbarn,Ē right off scale. He decided it probably was working. (Laughter) The primary motivation for doing something on this was the problem that was showing up in K-band radar. It looked, off hand, like there was much more absorption due to water vapor than people had expected. Ed thought, maybe, we could get out of this problem by looking at the sun at various angles in the sky. I never did try that technique because it seemed to me a better one was to simply look at the radiation from the atmosphere itself as a function tilt/ tipping angle. Just a crude breadboard model was used on the roof of MIT, connected with weather balloon data. Weather balloons were used to give us integrated water content in the atmosphere as function of height. Then one compared the atmospheric temperature with this and got a nice straight line. I remember giving a talk on this after I had these results. A wellóknown theorists, whose name I wonít mention, said to me afterwards, ďDicke, these results are theoretically excluded.Ē (laughter)

Harwit:

Why donít you mention his name?

Dicke:

No, I wouldnít do that; heís still alive.

Harwit:

I see. Why was he dubious?

Dicke:

Well, heíd been one of the ones that had made calculations on the absorption of water vapor in the atmosphere.

Harwit:

I see. Well, those are difficult to make. I suppose people are more convinced about...

Dicke:

I was amused by the idea that the experiments are theoretically excluded.

Harwit:

Now, one of the questions that I did want to ask later on, but I might as well ask it now is, when you eventually set up your radiometer on the roof there at MIT and set an upper limit to the background radiation which, I guess, you published as about 20įK, your Physical Review paper in 1946, do you think you could at that time already have detected and established a 3įK background radiation? Thatís been a controversy that people have sometimes bandied around.

Dicke:

No, the reason was we didnít have a cold reference source; it was a room-temperature source. That meant a long extrapolation.

Harwit:

You had something called the ďdog,Ē didnít you?

Dicke:

Yes, the ďshaggy dog.Ē

Harwit:

A piece of fur, was it?

Dicke:

Well, the ďshaggy dogĒ was something I constructed myself — a piece of thin plywood. We had this what we called — I canít remember the real name anymore — it was the absorbing cloth, a cloth impregnated with something to make it conducting. We cut wedges of this and tacked them on the...

Harwit:

Something like Eccosorb?

Dicke:

Something like Eccosorb, except it was a cloth, and tacked these wedges on there, and then the idea was youíd shake this in front of the antenna, and then the time average random motion of these things, the reflected wave would be phase incoherent on time average. You wouldnít have to worry about an offset due to reflection, due to impedance of this thing being reflected back into the...See, the problem that you had with that before the circulator, there was no circulator or isolator available at that time, so you had the problem of any voltage applied to the crystal detectors themselves.

Harwit:

What is this circulator, incidentally?

Dicke:

A circulator is a waveóguide device that incorporates a magnetic rod, and it effectively can be a three-port, four-port, five-port, any number of ports. What happens is that a wave coming in, in a four-port case, a wave coming in here, goes out there. A wave going in here, comes out there, and like so. So itís always stepped over one. Youíve lost detailed balancing from a thermodynamic point of view, and you can trace this loss of detailed balancing to the existence of a magnetized material in it. It has a built-in polarity. An isolator is essentially a similar thing. In the absence of these things, any small voltage applied to the crystal detectors would act as a modulator and would come out of the antenna as a wave. When that was reflected back in, it would look like a signal. So one had to be very careful to minimize this effect. If you wanted to put in the absorber and make sure you didnít have any reflected waves from it, that was the reason for the ďshaggy dog.Ē

Harwit:

Yes, I understand. So, you say were dependent on the cold load then?

Dicke:

To get any greater precision than this 200 upper limit, we would have had to go to a cold load. It was only a byproduct of the...In fact, when I was thinking about this, it was not from the standpoint of background radiation at all, but really just thinking about radiation from galaxies. We were concerned there could be a background of microwave radiation in galaxies, and this set an upper limit on that of 20į.

Harwit:

Had you heard at that time of any of the astronomical work that Hey, in England, had done shortly after the war, or at the end of the war, with their radar equipment?

Dicke:

No, the only astronomical work that I was familiar with, except for the little bit we did, was some observations that a chap at Bell Labs — I had forgotten his name.

Harwit:

That was Southworth.

Dicke:

Southworth, right.

Harwit:

You quote him in another article you did. I donít whether the Physical Review article mentioned it.

Dicke:

I might have mentioned it; we looked at the sun during the eclipse.

Harwit:

Thatís right. Itís...you mention it there.

Dicke:

I probably made reference to it there because he did detect the solar radiation.

Harwit:

Thatís right. He had.

Dicke:

He didnít have a radiometer. He just had an ordinary receiver and looked at the offset of it.

Harwit:

Thatís right. In the paper you did on the microwave emission of the sun and the moon, you mention Southworth. Spitzer or Martin (Schwarzschild) were there yet at the time; they came shortly after. Iíll bet you know the dates.

Harwit:

No, I donít, I had assumed that Lyman Spitzer would have been here by then, but I may be completely mistaken on that.

Dicke:

This would have been late 1946, probably.

Harwit:

I thought he had come here straight from Yale, but I may be completely wrong.

Dicke:

He may have been there; I wasnít that familiar with it.

Harwit:

In fact, I know that in 1949, he was writing papers from Princeton, He did one paper on shadowing forces, that is, dust grains that are imbedded in ambient radiation bath and that are mutually attracted with an inverse square law force. I remember reading that paper which is a 1949 paper, I believe, and that was done here. I thought he was here before, butÖ

Dicke:

Well, if he wasnít here at the time, it was certainly shortly after.

Harwit:

But as a result of that, then, you went into straight physics.

Dicke:

Well, as a result of that I really cannibalized that equipment and the microwave parts, and so on, were used in trying to do some experiments.

Harwit:

Now, again, Iím going to skip a lot of the interesting experiments you did during those first years at Princeton because you have discussed them in the Bromberg/Forman interview, and jump right over to... Well, no, let me mention one piece of work which I thought was interesting. I mean, I looked at some of the papers thatÖ

Dicke:

That was a lot of stuff to look at.

Harwit:

There are a lot of things that... but theyíre really quite fascinating. One of them is the angular momentum paper which I thought was really very nice, because I remember at one time worrying about that question, not knowing that you had looked at it. And, in fact, coming across — and I want to ask you whether you had ever seen it — a note about that in Sommerfeldís book on atomic structure, you know Atombau und Spektrallinien.

Dicke:

No, I donít recall ever seeing it.

Harwit:

He talks about the angular momentum of a plane wave there in classical terms. What he does is to look at a spherical wave which is generated by a Hertz potential, and allows that to become a plane wave in the limit of infinite radius. Of course, itís still a sphere; itís only locally plane, and then he takes the Poynting vector and integrates it along a circular path on the surface of that sphere, and that always has a ratio to the energy which is constant and is just the frequency. He gets the ratio of energy to angular momentum in the wave there, and that just is the frequency. But I hadnít come across your paper at the time, and it was rather nice because it again emphasized this interest that you always evince of looking right at key physical issues, I mean, I guess, itís just an earmark of the style that you work in, isnít it?

Dicke:

The thing I found fascinating about that was that on the face of it, it looked like a trivial question, because you expand your waves in the various spherical harmonics, and each one of these is characterized by 1 and m. Then you realize that these waves of definite 1 and m donít carry angular momentum itís only in the cross terms between them and other waves that are purely present in the other values of m, for example. The cross terms between that and the other phase of m are where the angular momentum terms actually come.

Harwit:

Thatís right. I just thought it was a really neat paper. Itís interesting how many such questions really one tends to be quite fuzzy on, especially on the very fundamental things. Occasionally, one reads a paper in Physical Review Letters which goes back to some really fundamental point, let's say some stochastic interpretation of the Schr6dinger equation, or something like that, and one realizes that one uses these things all along without really understanding quite often where they come from or worrying about philosophical points. I suppose if one worried about those all the time, one would probably be mired down too much.

Dicke:

These are the kinds of points which, I think, are continually rediscovered too. Theyíre reinvented; the wheelís reinvented over and over.

Harwit:

Yes, it is surprising how much they seem to be discounted, which means, I suppose, which is implied by their being forgotten and having to be reinvented.

Dicke:

Now one interesting thing in that connection is the attitude the average physicist had about cosmology in those periods.

Harwit:

Yes, go ahead, Iíd like to hear you talk about that; it is interesting.

Dicke:

Well, I remember...this is, I think, only an example, but I think itís typical, I remember talking to Vicky (Victor Weisskopf), when I was a graduate student, about relativity. This was something that he felt was important that I should learn. He said yes, this would be a nice thing to know, but really relativity is so divorced from the rest of physics that it has very little effect, It sort of a subject by itself general relativity. Cosmology fitted nicely into this picture because cosmology at that time was only an exercise in mathematical models, isotropic, uniform spaces — all geometry and games you played. Except for tying it to the expanding universe, there was very little in there for which observations could.

Harwit:

I think Gamow has to be credited with at least introducing the chemical question into cosmology, and therefore giving it an additional dimension which previously it didnít have, and where previously one could have considered that the chemical elements had all been God-given. He worried about whether they could have not been generated.

Dicke:

Cosmology and cosmogony were important as sort of little side issues in various sciences in geophysics, for example, geology to be able to trace back and set a time scale for which things happened was important.

Harwit:

Well, let me ask you then right off. As I was looking through your publications list, it struck me that you were doing essentially fundamental but still laboratory physics experiments until about 1957, and then all of a sudden you burst on the cosmology scene with two papers in the Reviews of Modern Physics on the equivalence principle question of Machís principle, and so on; and I was wondering how you got into that. Essentially you, you know it sounds sort of trite, but you arrived on the scene as a ďfully finished product.Ē

Dicke:

Like all things sometimes the origins are hidden.

Harwit:

I suppose. Thatís why I was wondering how that had taken place. I mean, there was no period of apprenticeship that one could trace back in the papers.

Dicke:

My interest in both geophysics and astrophysics in connection with this were recognizing they provided tools for getting at the relativity question which Iíd developed an interest in. The interest in relativity came about because in 1953 or Ď54 or somewhere in there I spent a sabbatical year at Harvard, and started thinking about other things. In particular, I became interested in gravitation.

Harwit:

But there was nobody there to talk with about that.

Dicke:

I did reading on my own; I didnít really talk with people about it particularly. As an experimentalist I was fascinated with the thought that one could do a much better Eotvos experiment than had been done by Eotvos; so when I got back, I went immediately to developing that, I got interested in astrophysics and geophysics because I thought these subjects provided a tool for getting at some of the questions which my interest in general relativity were bringing up. I was interested early in a number of aspects of relativity, and one of these came about from Diracís arguments about the large numbers. Another was the interest I had in Machís principle, thinking this should be significant. Then the conclusion that I reached that Machís principle probably was not properly incorporated in general relativity, which led naturally to adding a scalar field to the tensor field, so the scalar-tensor field approach to gravitation came out of this problem that I solved visóaóvis Machís principle. In that framework you have the requirement that the gravitational constant is not a real constant but itís a function of coordinates. The cosmological solution varies with time, which carries with it obvious geophysical and astrophysical implications, if true. This looked like a means of providing the tools for getting at this question.

Harwit:

I guess much of the preparatory work you must have done then between the time that you had your sabbatical at Harvard and the two Reviews of Modern Physics articles must have just been theoretical background thinkingÖ

Dicke:

Thatís right.

Harwit:

...and applying all the physics you knew to the situation.

Dicke:

Meanwhile I was working on the Eotvos experiment which was a hard one to pull off, and it was taking a long time.

Harwit:

Sure. I see. So you had actually started working on the Eotvos experiment around 1955, Ď56.

Dicke:

Soon as I got back.

Harwit:

Oh really? I see.

Dicke:

I had a quite a sizable group working on atomic-physics-type questions at that point — and a number of graduate students. I effectively went off on this tangent all by myself. I donít recall having...I might have had a research assistant helping me, but certainly no thesis student on it.

Harwit:

One thing I wanted to ask you about anyway, and that struck me, is that almost all the papers that deal with this are authored solely by yourself, which I understand; but virtually none of the papers that the graduate students in your group wrote are coauthored by you. Whereas most thesis advisors will tend to add their names to papers written by graduate students, not exclusively or consistently but at least occasionally. Whereas, thereíre some graduate students youíve had whom you have never coauthored anything with.

Dicke:

I think this is probably a matter of principle more than anything. I probably changed my principles as time went on because I think later on there were more coauthored papers. I had an another crazy idea at the time which is an experimentalist who has an idea about an apparatus should not publish this at the idea stage. It should be reduced to practice and produce some useful results before you write anything. The result was that a number of rather good ideas were never published.

Harwit:

That tends to still be occasionally required.

Dicke:

Itís probably wrong.

Harwit:

It probably is wrong.

Dicke:

If you lose Ideas in a process, theyíre not passed on. They have in a sense to be reinvented again.

Harwit:

Thatís right, yes. You did eventually have a student working on the Eotvos experiment?

Dicke:

Yes. I donít...as I recall the matter, there was never a thesis that came out of that. There were three of us. I think there was a student who had already done a thesis; he was in a postdoc position and was helping with the group. I believe there were three of us who coauthored the main paper.

Harwit:

Yes, I think Krotkov.

Dicke:

Krotkov had been a graduate student working with Eugene Wigner and did an theoretical thing with him.

Harwit:

I see.

Dicke:

And then Curott, was he one of the other ones?

Harwit:

Let me just take a look here.

Dicke:

David Curott, I think, he was a thesis student of mine.

Harwit:

I brought copies of some of the papers with me, but then...Mark Goldenberg was never involved with that, I guess he was just involved with solar ellipicity.

Dicke:

I believe he came before that, the start of the solar oblations stuff, but I canít remember what he did. He was one of Ramseyís students.

Harwit:

Oh, I didnít know, I see, I didnít bring that particular paper in, but I do have the list you sent me. Roll was not involved in that, I believe, either.

Dicke:

In the Eotvos?

Harwit:

Yes.

Dicke:

No, I donít believe so. I think it was Curott, Krotkov and me.

Harwit:

Okay, fine.

Dicke:

I donít remember what Krotkov thesis was, but it must be up on the shelf somewhere over there, (Added note: Curott is listed with Hegyi and Dicke in the 1968 Bull. Amer. Phy. Soc. primordial helium abundance publication. Krotkov, Roll and Dicke worked on the Eotvos experiment published in 1964.)

Harwit:

How many thesis students have you had over the years?

Dicke:

Altogether? You get a pretty rough idea by counting those volumes up there.

Harwit:

Oh, I see. It looks about thirty.

Dicke:

It might be. Maybe not quite that many.

Harwit:

Was Ray Weiss one of your students?

Dicke:

Ray Weiss was a postdoc here. He came, I think, from MIT.

Harwit:

Yes, he did, but you never published anything together either?

Dicke:

No, I donít believe so. He worked on problems which I suggested here, and I donít think we ever published jointly.

Harwit:

And then, I guess, Bill Hoffman was a student of yours.

Dicke:

He was a student of mine, yes.

Harwit:

kid with him you did publish some things. I think it was mainly instrumental things, if I am correctly looking at your publications. I think it had to do with getting information from satellite orbits.

Dicke:

Oh yes, that was a crazy thing. We did an autocovariance analysis of the residuals of Jupiterís motion and found a strong periodicity in there that the astronomers had missed and published that.

Harwit:

Is there anything else you might want to say about students at this stage and postdocs and sort of your ideas about interactions that one should or should not have?

Dicke:

Perhaps itís a question of style. My feeling about working with graduate students is in so far as possible to give then their own head and let then work, out their own problems, rescue them when theyíre in trouble, and not to try to direct things very closely.

Harwit:

Have you had the luxury, perhaps, of having grants that always were sufficiently widely based so that you could do that?

Dicke:

Yes, I never had a grant or contract that required that I do things that really were outside my normal interest. In other words, they supported what I wanted to do.

Harwit:

Nowadays, a lot of the NASA contracts tend to be very specific, and you always run the danger, when you apply for a grant, that the application will be rejected on the grounds that the applicant didnít specify precisely how he was going to go about solving the problems that he indicated. Have you run into any?

Dicke:

Well, I found that both the NSF and the Navy, and I had contracts with the Navy for many years, and in both cases they were very enlightened with respect to this. But anyone who thinks you can spell out in advance how youíre going to do a piece of research, just doesnít know anything about science.

Harwit:

Well, youíve just indicted an enormous number of referees.

Dicke:

Itís just not the way to operate.

Harwit:

What do you think of the refereeing system in that case?

Dicke:

I havenít run into problems with refereeing of that kind. I think referees Iíve encountered, if I can judge from the refereeing, have had about the same ideas I had about the nature of science. But this thought that you can spell out in detail whatís going to happen at stage 1, stage 2, and stage 3 is ridiculous.

Harwit:

Yes, Iíve perhaps run into a different batch of referees, just a different subdiscipline that I work in perhaps; but my main frustration with the system tends to be along those lines, that I will get back reviews which are evenly split, perhaps, between excellent and poor with very little in between, and people saying, ďWell, he doesnít say precisely how heís going to go about this.Ē Then, I guess, the monitor at the NSF or NASA decides on his own whether Iíll get the grant or not because theyíre so split. But I agree if youíre going to do something significant, unless youíve had the good fortune of being able to bootleg it on an earlier contract, so that you know exactly what youíre proposing because youíve already done it, itís difficult to see how youíre going toÖ

Dicke:

My impression is that this has all gotten very much worse in the last ten years from what it was when I first got into the business.

Harwit:

Oh yes, in the sixties it was much easier.

Dicke:

The Navy was one of best organizations to work for in the old days, and Iíve talked with...dear, dear. Well, it doesnít really matter.

Harwit:

Can you tell me the position of the person; I might be able to get his name.

Dicke:

Letís turn off the machine a moment, One of the problems Iím facing in my old age is complete memory blanks when it comes to names, It was bad enough when I was young; itís gotten worse. Manny (Emanuel Priore) told me that when he was running the ONR, he was much more interested, (and he didnít send things out for peer review at that time; he made his own judgments), that he was much more concerned about the scientist and his reputation than the details of the proposal, and he looked for good people to support.

Harwit:

I think thatís much less the case. Iíve argued sometimes....at NASA, you know, most of the research grants are renewable every year, and Iíve tried to argue that I would much prefer to be reviewed on the basis of what Iíve done in previous years than on the basis of what Iím proposing. Itís impossible for some reason. They keep claiming that this is contractually what they have to do.

Dicke:

You have to ask yourself, I think, how this whole business of government support of research came about in the first place. As I see it, itís a direct outgrowth of the experience that the military had with World War II, where at first they were a bit reluctant to accept new techniques involving radar. They were finally convinced that this was important, and radar, the atomic bomb, and so on, played such an important role that they decided that one should try to strengthen this science in the country. To strengthen science and strengthen the scientists to support them didnít mean they had to work on military things. So what they did initially was to seek good people and support them.

Harwit:

It worked very well. Some of my colleagues still tend to talk about ďthose idiots in the militaryĒ and how they botched things up in the sixties and the fifties; but when you look at what came out of those yearsÖ

Dicke:

Things were very good at that time support-wise.

Harwit:

It certainly has been, I think, better than what later on came in the seventies, and itís clear that in the seventies we were under fiscal restraints in the sciences, but that was, I think, mainly due to the fact that the number of scientists had so greatly multiplied. When you look at the number of dollars that were involved in terms of their buying power, their adjustment for inflation, in other words, you find that you still had about the same amount of money in the seventies. And yet in astronomy, at least, the sort of the flashier things being done in the sixties — the discoveries of quasars and pulsars, gamma-ray background and all kinds of infrared sources and X-ray stars — that you donít find repeated in the seventies as much, perhaps because we had done the easier things already. But nevertheless, I think some of it may also be part of the administrative structure. Well, let me get back to Ď57, though, when, as I say, you started out on this new research area for you at the time of relativistic work, which, I guess, step by step, also led into your solar work, because of your interest in Machís principle which was tied to the scalar-tensor theory, which, in turn, was tied to a test that...

Dicke:

Öa test of relativityÖ

Harwit:

Öof relativity, for which you wanted to use the sun. Is that the correct interpretation?

Dicke:

Yes, that is very definitely the correct one.

Harwit:

Maybe you could elaborate on this.

Dicke:

Well, as far as the sun is concerned, you had the three classical tests of general relativity that Einstein, himself, had introduced, and I convinced myself that one of these wasnít really a test of relativity at all, but it was a test of much more general principles than that specific theory.

Harwit:

The red shift.

Dicke:

The red shift, yes. And so youíre left with a light deflection.

Harwit:

Because it was an equivalence principle test.

Dicke:

Right. The light deflection business at that time looked pretty bad. You had the perihelion rotation of Mercury was the original test. In fact, this wasnít really a test because this discrepancy was known to Einstein at the time he was trying to generate general relativity. If there were a distorted sun, that could influence this and provide a correction, so it was important to know whether the sun had a quadruple moment or not. And the other aspect of it, it was the thing I mentioned on the cosmological side — the idea that the gravitational constant could evolve in time would have astrophysical implications, for example, in the evolution of stars. So you got into the questions of the stellar interior.

Harwit:

When you got into this did you have any inclining that you would be still working on it 30 years later?

Dicke:

No way. I never would have believed it. Time has flown. Of course, the older you get the faster it moves, so I just never thought.

Harwit:

Well, itís not just that, I think one always underestimates how long some of these experiments take. Do you think you have gone into it if someone had said to you that 28 years later you would still beÖ

Dicke:

I would think he was crazy.

Harwit:

Well, why donít I interrupt here since weíre at the end of this tape andÖ Well, we just came back from a coffee break. Let me ask you now about this spate of theoretical work with a view, I guess, towards experimental work that you were doing in the mid-fifties — some of the particular aspects that you remember, both the theory and, I suppose, especially the experiments were very sophisticated. There must be a lot there you could talk about.

Dicke:

Letís see, youíre speaking about the mid-fifties. Now, what wasÖ

Harwit:

This was the Eotvos experiment, then the solar oblations experiment and some of the thoughts that you had about isotropy of space at the time that you wrote about, and the interpretation of the experiments in terms of what they implied about the theoretical physics, that is, the question of did or did not a certain experiment act as a verification of general relativity or was it just an expression, if that, of the equivalence principle and then also the fitting in of Machís principle into general relativity. Was it there? Was it not there? Finally, this question of weak force bosons masquerading as gravitational effects. This whole complexion that you were building up at that time.

Dicke:

Itís a complicated complexion. I think the first thing one can say is that in a sense the purely experimental parts — experiments themselves and the development of the experiments — was a separate challenge which didnít overlap all that much with the philosophical and theoretical thinkingís, I find it amusing — maybe itís a comment on the nature of science — that both the Eotvos experiment and the solar oblations experiment were very challenging in terms of the development of the necessary instrument. In both cases, it took a number of years to pull it off and design an experiment that I was really satisfied with. The reaction to the Eotvos experiment was absolute acceptance. Everybody agreed this is what we expect; everythingís fine — one part in ten to the eleventh: good. Why not?

Harwit:

Was there any reaction like, ďWhy bother,Ē when you were done?

Dicke:

In a way, yes, I think. I donít that I ever got that explicitly, but I think it must have been implicit. When it came to the oblations business where the results didnít agree with preconceived notions, the experiment was done just as carefully, the reaction was completely different. Nobody believed it. I wouldnít say that nobody believed it, but the first reaction from most people was it must be wrong.

Harwit:

Would you generalize from this to the statement that a new experiment can only be treated by the community in two ways: either itís trivial or itís wrong.

Dicke:

I think in a way thatís almost true. Itís particularly true of null experiments. I did a number of null experiments over the years, and was particularly fond of them, because the way of getting greatest sensitivity is to find nature has set up something for you, which if you make the right measurement, you get zero for your result. Itís a terribly compelling argument when you have it. With the null experiment you frequently get the reaction, ďOf course! Why did you bother?Ē

Harwit:

Yes, I know exactly what youíre saying.

Dicke:

At one time I was trying hard to measure the g-factor of the free electron, and a series of experiments were causing me a lot of trouble. A famous theorist, whose name I wonít mention, said, ďWhy bother? Weíve known for years we can calculate this. The result is two.Ē

Harwit:

Yes, I know. Letís see, the accuracy now is something like one part in on that experiment, I believe. Isnít that correct? Thereís an article in the latest Physical Review Letters which uses theÖ

Dicke:

Hans Dehmelt is the one who has done the really beautiful experiments on that. In fact, he succeeded for the first time in getting results on experiments I was trying to do but never succeeded, and really with very good control of the trapping of the individual electrons is the way he succeeded.

Harwit:

Letís see, you and Crane at Michigan were working about the same epoch on those g-2 experiments, is that correct?

Dicke:

I think I was before Crane. I started on this almost immediately when I arrived here in Ď46. What I started first was the Lamb-shift experiment until I found out Lamb was doing it.

Harwit:

Thatís right.

Dicke:

Then I switched to this other experiment which I thought was important was to try and get the g-factor of the electron, and I must confess I never really came very close to making that work.

Harwit:

You always tried difficult experiments. Can you say something about that? It must be a deliberate choice on your part.

Dicke:

Is I said before, I think, one should be driven by the importance of the experiment. If you decide that this particular measurement is important, you should work hard at doing. And that doesnít mean...Let me give you the opposite side. I donít say this as a general principle that can be used. The opposite would be you would have a big accelerator which youíve already spent a billion dollars to build — and you try to think of the best experiments to do with that accelerator. I donít particularly like that approach. The approach I like is you think of an important experiment and you ask yourself, ďWhat apparatus do I need to do this experiment?Ē Then you go about figuring out how to do the job.

Harwit:

Well, thatís sort of an interesting comment because astronomy is now heading in the direction of this first type of work.

Dicke:

Öbecause the cost of the equipment.

Harwit:

Thatís exactly right.

Dicke:

The big VLA, for example, you have an instrument there that produces pictures, and so you now try to think of what pictures can I take that will give me the information of a certain kind I want to have.

Harwit:

But itís also true and maybe even more so with the NASA spacecraft, the Hubble Space Telescope; the Gamma-Ray Observatory; the Advanced X-Ray Astrophysics Facility; and for the infrared, the Space Infrared Telescope Facility. Those all are going to be general-purpose facilities of exquisite sensitivity, but dependent on people using them well.

Dicke:

But you know, if you think back with the history of the X-ray observations, for example, you see there are aspects of the other side of this.

Harwit:

Of course.

Dicke:

The early X-ray things, one didnít know what we were going to see up there, and you gradually developed bigger and bigger....The same thing happened with the accelerators: You started out small, and only after you learned something about the nature of the science, did you let them get bigger and bigger.

Harwit:

Are you saying that youíre attracted mainly to what would have been called ďlittle scienceĒ by Price.

Dicke:

My view of the ideal experiment is one whose the apparatus fits on a desk, just about this size.

Harwit:

Do you think that weíre close to running out of such experiments?

Dicke:

I donít know. One always, I think, periodically thinks that the most interesting things have already been discovered, but then something new opens up.

Harwit:

Well, to some extent one is helped out, I suppose, in the direction quarterly reports, and they had to be in so many copies, FOBÖ(laughter) of making small experiments by the circumstance that industrially available instrumentation is so much more sophisticated now than it used to be. On the other hand, science has escalated at the same time — maybe faster.

Dicke:

This is one of the biggest changes that Iíve seen in physics over the years is the way experimental physics is done nay, in comparison to the way it was done when I was a graduate student. We could not go out and buy an amplifier at that time. What you could do, fortunately, in the early days of radio there ware radio amateurs. Because there ware radio amateurs, there were a few stores that stocked radio tubes and parts. If it werenít for that, I donít know how we would have built apparatus at all. We went out and bought some components and put them together ourselves. You could not buy a scalar, for example — pulse counting — you built your own.

Harwit:

Until just after the war, and then, I guess, Tracer labs had some that you could buy, I think.

Dicke:

When Marshall and I did our experiment, we really did it on a shoestring. There wasnít money to buy anything anyway. We, for example, made our own Selsyn motor out of a DC motor that we put some slip-rings on. We built our own amplifiers, counters. Fortunately, somebody else had taken on the project of building the scalars, so we could use the scalar that existed. Built our own counters.

Harwit:

Now, to come back to the actual experiments, did you ever have problems financing them? Two questions come to my mind, one is that the difficult experiments ware perhaps one would need unusual equipment, and the other one is the length of time it takes to develop, and the question of whether the granting agency was getting jittery about supporting something for a long period of time that might not be coming to any kind of a culmination.

Dicke:

I was always very fortunate, I think, in the kind of support I had. When I first arrived, Milt White had already, or shortly afterward, obtained support, I believe, from the Navy to rebuild the cyclotron here. And I piggybacked on his contract for a while — just a little bit of money. One day I had a call from Marcell Golay, whom you may remember as the inventor of the Golay cell, in infrared.

Harwit:

And a lot of interesting things, yes.

Dicke:

Yes, a very clever guy.

Harwit:

Very clever guy, yes.

Dicke:

He said in his Swiss accent, SwissóGerman, I guess, or was it French, I donít remember, that heíd like to come and talk to me. He effectively said, ďI read some of your papers, and Iíd like to see if we can arrange some way that the Signal Corps — the Army — can support your work.Ē He was a scientist for the Signal Corps at the time. This required a bit of doing because theyíd never tried anything like this before, and their contract officers didnít see has it could be done. Finally, they came up with a scheme, which is they bought something. They bought something. They bought four Quarterly Reports per year.

Harwit:

Thatís very nice.

Dicke:

(laughter) So we had to crank up.

Harwit:

So you really had completely.

Dicke:

It wasnít a lot of money, but it was enough to take care of the work at that time. Then after that I got aw own Navy contract, and finally NSF support, and always the amount of money seemed to be adequate to do what I wanted to do. But in part, you know, you match your desires to whatís available too. Doesnít work both ways.

Harwit:

Now, if you didnít have any graduate students working on the Eotvos experiment, how did you handle it, or did you have then working on it but not as part of their thesis? I mean, did you do the work yourself?

Dicke:

They were research assistants. In our department we still have this system where the students that come in are either teaching assistants, have a fellowship, or theyíre research assistants. They are effectively expected to put in so many hours of work ó- hours a week work — helping you with your work.

Harwit:

Now, on a sophisticated experiment like that usually a student will be helpful only after he or sheís been around two or three years, and by that time very often they already are on thesis work. How did you handle that?

Dicke:

I canít remember now, whether I gave them a fairly simple experimental task to do or what, to tell you the truth. But of the students I had working with me, any one time, I would say the order of...in the mean, I would say the order of one-half were research-type appointments. Quite often, come to think of it, when a student would finish up and write his thesis, he would stay on for a year or two after that with me, either as a postdoc in the group or as a member of the faculty. Quite frequently he started as a postdoc. Then the department needed an Instructor, and they would go into a teaching position where they might stay in for three or four years. That type of older individual was a tremendous help. I normally had quite a few of those.

Harwit:

Now, both on the solar oblations and on the Eotvos experiment you decided that the site at which you would carry them out would be here at Princeton. What were your reasons for that?

Dicke:

The first reason is that in developing a thing you have to have ready access to the shops, and so on, and laboratory, because you are making laboratory tests. So to have a remote site is — until the Instrument is fully developed — not terribly feasible. As far as the Eotvos experiment is concerned, I thought the sight here is probably as good as anywhere else, anyway. But with respect to the oblateness, itís still unclear to me just where the balance lies. One of the things I noted very early in this is that the local atmospheric distortion would possess a symmetry property if you effectively had translation invariance of your instrument. We had a nice, grassy flat plain in which we were making observations. Weíre now on Mt. Wilson, with the apparatus, where the seeing is infinitely better in terms of number of days that you have a nice sun to look at in the sky — itís beautiful, but weíre within 200 feet of a precipitous drop-off. The symmetry we saw in our Princeton observations, we no longer have. That symmetry was a tremendous help in reducing the data, so itís a bigger problem now, to take into account the...

Harwit:

How long have you had the equipment at Mt. Wilson?

Dicke:

This is our third year there now. We revised the instrument and improved it, modern electronics, and so on, and took one simmerís data here. That was a disaster; the weather was terrible that summer — very few good days did we have. In the following year, Ď83 — we were on Mt. Wilson in Ď84 — and weíre taking data again this stunner.

Harwit:

You intend to do it in the summer, mainly, is that right?

Dicke:

We have to have the sun high in the sky; otherwise, the laminar distortion gets to be too great.

Harwit:

Do you have a student involved in that? Are the Mt. Wilson people running it? Or how does it work?

Dicke:

Iím, of course, out of business now. Iím emeritus.

Harwit:

Are you emeritus? I didnít realize it.

Dicke:

It started last July.

Harwit:

I see. I didnít know that.

Dicke:

I donít have students anymore, but my last two graduate students, Jeffrey Kuhn and Ken Libbrecht, are extremely good. They were outstanding graduate students, and the three of us are now collaborating on this. Itís the only way I could continue making measurements.

Harwit:

Sure. Well, thatís alright. Are they officially at Princeton or at Mt. Wilson?

Dicke:

Jeff is on the faculty here; heís an assistant professor. Ken is an assistant professor at Caltech, and thatís nice having him located there.

Harwit:

Okay, now, when I went over your papers, I was wondering why it had been close to 20 years between the time of your initial observations here and your setting it up again at Mt. Wilson, considering that there had been so much controversy generated.

Dicke:

In part, itís the controversy; thatís the answer to the question. I was just kept so busy for so many years battling with adversaries...

Harwit:

Yes, I noticed that.

Dicke:

They were writing papers on facular contrasts and all sorts of things, that there just hadnít been time. Then the other thing was that it really required very able graduate students with an interest in this to start this up again a major problem.

Harwit:

Is the present apparatus superior to the previous one? Any significant difference?

Dicke:

A number of improvements have been made Iím very pleased with, I think the apparatus is performing beautifully. I have a problem with the site, as I just said, if youíd like a list of improvements weíve made, I can tell you very briefly what they are.

Harwit:

It would be very interesting, yes.

Dicke:

Well, we had an analog device — lock-in amplifiers initially. Weíve gone to allódigital now; this thing has more computers tied to it than...I donít know how many microprocessors we have; it must be on the order of a dozen.

Harwit:

Oh really, that many?

Dicke:

The technique as it was formerly: We projected the sun on an occulting disk which allowed the edge to protrude over the occulting disk, and then had a scanning wheel with two apertures — diametrically opposed — of slightly different size, As they scanned, the second harmonic of this was a measure of the oblations, and the first harmonic, which came about from the unequal, size provided an error signal that could be used to the center up the sun, in a servoómechanism. This particular technique required... first of all the scanning wheel ran very fast because the servo-loop required a Nyquist frequency sufficiently high that you could close the loop and have it work with enough bandwidth. The unequal size in the two apertures had the problem that a small displacement of the sun would produce a first-order fake oblations signal.

We have gone to a different technique on the servo now in which we have equal-sized apertures which in first order displacements of the sun cause no signal, and the tracking method — the serve-system — is now quite different. There is a direct fast-feedback of the sun, relative to a dummy disk, a little side disk, that is used to correct with very wide bandwidth for small displacements of the mirror. Then there is a slow scanning of the wheel; it no longer has to scan so fast. Weíre now running more like 10 to 20 cycles per second, in that range with the scanning rate of the wheel. This is used in two loops, a fast loop and a slow loop, to center up with very high precision with the sun on the disk and have it respond quickly to small disturbances. The technique is basically the same, in that two mirrors are vertically mounted on a telescope. The telescope was manually turned by hand in to two positions — 90į apart originally. We now have it motor-driven, and itís cycled into eight different positions of the telescope, so the telescope errors are taken out up to the 16th order and only even harmonics are present and the odd harmonics donít appear.

We had to deal with such things as facular signals originally by including in the least squares fit to the data, a dummy facular signal — assuming the facular signal was proportional to the area near the facular patch — and in — corporate this into a least squares fit, That worked reasonably well with the 1966 data, Weíre doing it better now, though. The scanning wheel gives us 128 different discrete signals. When a facular signal appears in one of these bins, more than three sigma, we throw that bin out and then do a least squares fit with that bin removed. So substantial facular signals are now thrown out in real time rather than doing it later on.

Harwit:

Can you tell from the preliminary data which way things are going to go, or is it too early to tell.

Dicke:

Yes, weíll be publishing a paper shortly in Nature which shows that the oblateness....Well, all these things are subject to provisos, with the assumption that any brightness signal we have is a color-dependent one. We have a good fit with about 20-sigma significance in it. It gives about 19 milliseconds of arc in comparison with 40 some we had in Ď66. Thatís the 1983 data. Iím working on 1984 data now, and it looks like itís dropped another 10 milliseconds of arc. I believe the results, and I think there is a real change occurring in the sun.

Harwit:

Well, let me then go to this 12.8-day rigid rotating solar model you described a few years ago as giving you perhaps the best fit to the 1966 paper.

Dicke:

That appears in the 1984 data, but not so strongly that I would guarantee itís significant. If you would like to just see what I wrote about it in the paper we have coming out, we can look at that later.

Harwit:

Perhaps afterwards, yes, Iíd be interested actually.

Dicke:

Effectively, itís there in two harmonics of a significant level. The frequency is to within one sigma the same; the two harmonics agree. The signalís substantially weaker; itís sufficiently weak, in fact, that you wouldnít want....these are not the strongest peaks available, There are four other peaks in the power spectrum that are stronger, although these two peaks are strong.

Harwit:

So you think there are actual changes going on in the sun.

Dicke:

I think there are.

Harwit:

Just as a side comment, I was intrigued by the stereopticon diagram you had in the Physical Review Letter article. How did you get that drawn up? Who drew it up for you?

Dicke:

I did it.

Harwit:

You did it yourself. How do you do that? Do you close one eye at a time and imagine it?

Dicke:

No, what you do is you program the spatial configuration in three dimensions and then introduce a coordinate rotation, rotate the figure, then project on to the xy plane...

Harwit:

Oh, you did it on a computer.

Dicke:

On a computer.

Harwit:

Okay, fine, aright. I thought maybe you had imagined it. Do you do a lot of computer work yourself? Or do you have students doing it?

Dicke:

I do all the programming and the stuff I want to do. My tasks are fairly simple. I use APL a great deal on the PC over there. Iím the only one in the department who uses APL.

Harwit:

What do the other people use?

Dicke:

FORTRAN mostly.

Harwit:

I see. Okay, let me go back to the reception of the experiments, particularly the solar ones, and talk with you about that. In general, when thereís a hostile community, one tends to also have difficulty with funding. Did you have that problem?

Dicke:

No. I guess mainly because...Well, letís see, it came about in a number of ways. First of all, when I started to get interested in the astrophysical and gravitational things, and so on, I had a substantial grant from the NSF. I also had Navy support at that time. We had a sizable group. Then about the late 1960s, I guess, I was appointed to the National Science Board that required that I avoid conflict of interest, so I turned over the principal investigatorship to Dave Wilkinson on this.

Harwit:

That was in Ď66, you said?

Dicke:

No, it was more like the end of the 60s; I canít remember now quite when I was on that board, but I think I was off of it already by Ď72 or so, and it was a six-year appointment; so somewhere in there. The directorship of the research group, to the extent that it ever had one, was more now in the hands of Dave Wilkinson. That meant I was not directly responsible for the funding the way I would have been before. But also by the time all this controversy arose, the apparatus was already built and had been operated, and thatís where the money came. I had very little expenses after that. I donít think that question ever cane up, whether it should be supported or not. I canít recall that question being raised.

Harwit:

When you got your 10-11 upper limits for the Eotvos experiment and that was accepted in general, did you then have plans for going on with that?

Dicke:

No. It was a hard job to do that well. In the process I saw some improvements that might have been made, but it looked like it would be harder to go another factor of 10.

Harwit:

What about the isotropy of space? Youíve thought about that quite a bit and did a paper with Jim Peebles on that. Is there any money to be made on that?

Dicke:

I was puzzled about ten years or so ago. In fact, I was at Cornell and gave a talk on...this is one of your Bethe Lectures.

Harwit:

Yes, thatís right, yes. You gave the Bethe Lectures.

Dicke:

I think one of the talks I gave there was on this, but it puzzled me for some time that although there were really, I think, three principal things I found puzzling. One was this causality business, that if you look over here and look over here, you look at two objects which could never have been in contact with each other, according to standard cosmology, and they still seem to be causally connected to each other. Mother was the business of space being very nearly flat. If you simply go back in time, itís a great, tremendous accident the way it appears that things were so nearly flat.

Harwit:

Yes, I think you and Jim Peebles had pointed that out in a different paper which I think you called, ďNostrums and Conundrums.Ē

Dicke:

Yes, thatís right. Actually, I had pointed it out earlier in my little volume on gravitation and cosmology.

Harwit:

I see. Okay. Usually, itís this ďNostrums and ConundrumsĒ paper which is cited.

Dicke:

Yes, thatís the one. I donít think — we should have made reference to the earlier citation, I think.

Harwit:

So that was in your book already?

Dicke:

It was in that little book.

Harwit:

The ďGravitation and the Universe.Ē

Dicke:

Yes, itís only about two or three sentences.

Harwit:

Or is it in ďThe Theoretical Significance of Experimental Relativity.Ē

Dicke:

No, itís in ďGravitation of the Universe.Ē

Harwit:

What I was referring to was really more the Hughes-Drever experiment and the isotropy there, not the isotropy of the microwave background that I was talking about, and can you talk about that.

Dicke:

Well, this goes back to the Hughes-Drever experiment was in response to an idea Ed Salpeter and Cocconi had, which they published as a test of general relativity. ThenÖ

Harwit:

Are you sure it went that way? I thought Hughes and Drever did the experiment first, and Salpeter and Cocconi came up with...

Dicke:

I may be wrong, but I seem to recall it the other way around, that the theoretical paper on this understanding of nature of Machís principle was, I thought, discussed theoretically first. I may be wrong, and then Hughes and Drever separately and independently did experiments essentially of the same type to get this. I wrote a little note in Nature in which I pointed out that a different interpretation of this result could be made.

Harwit:

Yes, itís called ďExperimental Tests of Machís Principle.Ē Thatís in Physical Review Letters in 1961, and it starts out, ďIn this note it will be shown that contrary to the suggestion of Cocconi and SalpeterÖĒ

Dicke:

There must be a reference to the Cocconi and Salpeter paper there. We could find out whetherÖ

Harwit:

Yes, itís just that I donít remember whether the Hughes-Drever thing was earlier or not. You may haveÖOh, you have both of them. Yes, you do in fact. Let me take a look. Okay. The Cocconi and Salpeter paper is in 1960, Physical Review Letters, 4, page 176. Then thereís, I guess, an experiment by Hughes, Robinson and Beltrolopez (?) which is only a few hundred pages later. I doubt that they could have come up with it that fast.

Dicke:

No, youíre right.

Harwit:

But they may have been in justÖ

Dicke:

Dreverís later, I believe, isnít it?

Harwit:

And Drever is in 1961 in the Philosophical Magazine. So youíre right. Good. Yes, well, itís on that score, I mean, I remember looking at the Salpeter things at the time. I think he also won one of these Gravitation Foundation prizes on that work.

Dicke:

Which I was always very reluctant to try to win.

Harwit:

One wonders who referees the things. But at any rate, I also was puzzled by this question of whether you couldnít simply define the isotropy or anisotropy away by a proper choice of scaling where potentials and masses scaled in the same way, I guess, that in essence is your argument.

Dicke:

Yes, I guess the argument basically is that the kind of effect of a variable mass, for example, with a mass being a function of coordinates, you can use a conformal transformation to move that dependence of the mass onto the metric tensor, and then you just evoke the Einsteinís arguments about the general coordinates.

Harwit:

Now, youíve been motivated throughout your work — at least you always refer back to Machís principle — which seems to have a very firm, sort of a grip, on the way you look at things.

Dicke:

It might have had a firm grip, but I must confess I found it a very slippery principle.

Harwit:

It is, yes, of course, but maybe you could elaborate on that because it is a particular way of looking at the universe which can be interpretedÖI mean, there a lot of different versions of Machís principle, of course. But fundamentally, I guess, the idea that the universe as a whole determines certain local physical properties such as masses of particles and things like that. You very often have referred back to that, and, of course, itís not the only way of looking at things. I thought maybe you could talk about why you feel strongly that way; why you have been motivated by that? How itís guided your outlook on experiments to do, and so forth.

Dicke:

I guess one would have to say that if you go back to the time of Newton and the arguments about the nature of space that occurred at that time — discussions between Newton and Bishop Berkeley, for example, there were really two basic way of looking at the universe: One is that space was essentially an empty void that you put things in and that the coordinates in this were fixed in the space, and you could simply move particles from one place to another in this predetermined coordinates. Then the other way, which would say that the space itself has no structure to it. The only thing that is meaningful is the position of one particle relative to another particle. From that standpoint, relativity and the ideas of relativity come in very very early. If this is true that it is only the relative positions of particles with respect to other particles that are physically meaningful, then you run into paradoxes in certain problems that you see in general relativity. As for example, youíre familiar with the Lens-Thirring argument, that a small hollow mass of shell rotating will induce a rotation in the coordinates inside. From the standpoint of strict relativity, we should be equally able to describe this in a coordinate frame which is rotating for which the precessing gyroscope, if you like, is no longer rotating. Where the little inner sphere is rotating this way, the rest of the matter in the universe is rotating in the opposite direction, much more slowly, but thereís a lot of matter out there. Then you understand from that point of view, that the reason the gyroscope is not moving is that it is experiencing a tug from the light, rapidly moving shell, which is just balanced by the tug, induced by the slowly rotating mass of the shell. Thereís a nice balance there. If you accept that argument, then youíre soon in trouble, because ordinary general relativity doesnít do this, except for certain choices of the solution. Itís only for...

Harwit:

I guess Dennis Sciama did linearized...

Dicke:

Dennis Sciama made a kind of an analog theory in electromagnetic-type terms and tackled this problem.

Harwit:

Yes, thatís right.

Dicke:

But then, there were others that looked at this from the standpoint of Greenís Function approach in the framework of order in general relativity, and so on. What I saw in this was you could get around this problem of the argument involving the precessing of the rotating spheres, if you simply had a scalar field added. Then youíre ďled down the garden paths after you accept that.

Harwit:

When you say, ďLed down the garden path,Ē are you trying to say that you consider that maybe you were duped by this?

Dicke:

Not necessarily.

Harwit:

Or that you regret it?

Dicke:

No, itís just that once you incorporate a scalar field into the equations of general relativity, there are implications that go with this, and these implication demand investigation. The implication, for example, that the cosmological solution leads to a timeóvarying gravitational constant. That seems to be a firm one.

Harwit:

Now, how do you at this point feel about the scalar-tensor of the Barns-Dicke theory?

Dicke:

I donít know what to say about it. Thereís certainly no support for it that I can see. On the other hand, itís a theory for which thereís an adjustable parameter, and one way of looking at it is that itís a broader itís a more general theory than general relativity. It incorporates general relativity as a special case. Under those conditions you can hardly say that it is wrong and general relativity is right because it is always a special case. An interesting question is whether there are observable phenomenon where it makes a difference. I think there could be. Maybe not observable but in early cosmology there can be a real difference, and still hardly be observable now.

Harwit:

Now, given that the various experiments that have looked for time variations in the gravitational constant are now at a precision where, I think, you set upper limits of the order, is it one part in 1010/yr?

Dicke:

Well, that was with the value of this constant omega which is, I think, too small to be tenable now. If you look at the other tests of general relativity and look at the light deflection one, various forms which this takes, if you look at that, that sets a limit to omega which is quite high. That high value of omega also implies a very slow rate of change of the gravitation constant, so, I think, one is well under the point of having a test there. It would be very nice to have an orbiter around Mercury for a number of reasons, for relativity-type experiment.

Harwit:

What would the orbiter around Mercury do?

Dicke:

Without having to depend on radar on the surface, irregularities of Mercury and all the problems they cause, an orbiter around Mercury would give you a very precise location — determination of the way the planet is moving.

Harwit:

I see.

Dicke:

You have a good test of the varying rotation and other things, including the change of period with time.

Harwit:

kid youíd be able to compensate out, somehow, for the mass concentrations within Mercury?

Dicke:

Yes, the complications that you get in the orbit of the satellite going around Mercury, as you know, thereís all these moments of the mass distribution — everyone perturbs the orbit. But after all, if you make very accurate measurements at the location of that satellite, then you can reduce those equations and determine the internal mass distribution of Mercury and correct for it. Computers are wonderful.

Harwit:

Thereís one class of relativistic experiments that you seen to always have avoided, and I was wondering whether it was deliberate, and that is gravitational wave experiments.

Dicke:

Well, it was deliberate in a way. I always regarded it as too hard.

Harwit:

They are. (laughter)

Dicke:

I donít think Iíve changed my judgment on that. I think they are too hard.

Harwit:

Eventually, perhaps. To get back to the Machís principle, though, you did one paper in which you essentially were showing that thereíre certain invarianceís that come out of that, and I think thatís been a way of looking at the question which was novel, if I am interpreting that correctly.

Dicke:

Which invarianceís were you thinking of?

Harwit:

Itís a paper where you show various scale invarianceís that arise from...letís see if I brought it with me here. ďMachís Principle and Invariance under Transformation of Units.Ē

Dicke:

Oh, that one.

Harwit:

That represented a consequence of Machís principle which, I think, wasnít normally thought of as much.

Dicke:

I donít think that particular paper adds any new perspective on Machís principle as such, but what it does do is something quite interesting I think. I wouldnít call scale invarianceís. Itís a scale covariance argument, that if you have the same physical system and you write down the equations of motion for particles of that system, there is, if you apply a scale transformation in the right way to the equations of motion, they will take a completely different form. Theyíll look like new phenomena are there, but the matter will be moving just the way it was before. In other words, youíre looking at the same system but with different color glasses. That tells you something which is that our tendency to think that a particular picture of the world that works must be the right one — the right one. It may be wrong because there may be many pictures that are equivalent, that are all equally right.

Harwit:

Well, one has that feeling In any case, I suppose, and in a lot of different ways of looking at the universe, whether physics is geometry or not. And also the question of the extent to which local phenomena influenced by the distribution of matter in the universe at large, and whether cosmic horizon comes into that, and so on.

Dicke:

I wonder whether we should perhaps break nowÖ

Harwit:

Okay, fine.

Dicke:

Öand go get lunch. Weíll probably lose less time if we go early.

Harwit:

Okay, good. I was in any case thinking of...I think this is a good place to come to a stop, and then after lunch Iíd very much like to talk about the microwave background radiation.

Dicke:

Alright. Fine.

Harwit:

Weíve just returned from a very nice lunch, and weíll continue now, Iíd mentioned before that I was going to ask you about microwave background radiation, and how you first got started thinking about that. This must have been Ď63, Ď64.

Dicke:

I remember it rather clearly. It was one of those intuitive things. There were certain cosmological problems that had worried me for some time, and one of the worst of these was the problem of how the matter in the universe got started. Did it arise initially fullóblown? At the time this was a question the problem of singularities in general relativity was not known, and there was work by the Russians in which they discussed bouncing universes in which there was great randomness in the matter distribution that would permit the solutions to bounce — at least they thought so. It occurred to me, one day, that one possible way of accounting for the matter in the universe would be to have an oscillating universe, which at every collapse of the previous universe would be raised to sufficiently high temperature to get rid of all the heavy elements that were present — iron, everything would disappear. You would end up with nothing but protons, neutrons, mesons, electrons and what not. The universe would bounce; and in this new fresh, nascent state expand, stars would form, heavy elements would develop; and again it would bounce to some maximum size and collapse. The problem of increasing entropy that has plagued this idea of an oscillating universe for a long time is avoided in this scheme, if you, in the process of making new matter with every collapse, you make enough new matter to permit the entropy per nucleon to stay roughly constant. Well, the implication of this is, then, every bounce of the universe is more energetic than the previous one, contains more matter, if you go backwards in time, the universe gets smaller and smaller and finally ends up as a single quantum fluctuation or something we donít understand. But at least youíre not faced with the problem of generating all that matter initially in one burst.

Harwit:

Did you think of it in entropy terms at the time, or was that later?

Dicke:

I think that probably came later, because the entropy problem had been one for oscillating universes all the time. I donít really recall when that argument came up; but youíre probably right; it probably was later. Now, letís see, I saw some evidence that might support this in that the young stars were dirtied up and the old stars were clean, like the implications the universe started off in a quite clean state. I made a rough estimate of what the temperature of the universe might be now, as the background radiation cooled off and got some 35 or 40į, simply on the grounds that much higher temperature than this would produce so much energy density in radiation that it would drive the acceleration and affect the acceleration parameter of the universe noticeably. So itís sort of a crude upper bound.

Harwit:

If you wanted to have an oscillating universe then, it would have to be closed, but would have had to be able to expand out to the extent that it has now minimally; and so that gave you an idea of the accelerations that must have been involved at different times, and you related to those to the density through the Einstein equations.

Dicke:

Itís just that if you take the present matter density, the matter you actually see; and then if you were to add to that the blackbody radiation of some 40 or 50į, it gets to be an appreciable contribution,

Harwit:

Now, during none of this time had you run into the work of Gamow and Alpher, for example, on element formation.

Dicke:

Not quite. Iíd heard Gamow give a colloquium talk here, and I think it was either a colloquium or possibly a Sigma Xi talk or something of that kind, in which he described his ideas about heavy element formation in the early universe. But the way he presented thisÖ Whether this was a preliminary view or what, I distinctly remember him describing this as a completely neutron-filled universe and starting out cold, so the idea of it being hot hadnítÖI didnít realize he had that idea. We should have taken this up, but just didnít.

Harwit:

His universe always was hot, but it was filled with neutrons initially.

Dicke:

But hot also?

Harwit:

I think his universes also were always hot because he had to form the heavier elements, and so he would at least had to have energies that would have allowed the nuclear reactions to take place. Right from the beginning, the work that Alpher did for his thesis, which started out with a neutronófilled universe, had to be hot. Whether it had to have energies much higher than a few tens of MeVís, I donít know, probably that wasnít necessary. But that sort of temperature, at least, one would have had to have had. Go ahead though.

Dicke:

Thereís not an awful lot more to be said as far as my contribution is concerned, but I gathered together a group of several of us and explained these ideas Jim Peebles, Dave Wilkinson and a few others. Jim went ahead and made, as you well know, a better estimate of what the temperature might be.

Harwit:

Do you remember when you got together.

Dicke:

It was very shortly after I thought of it.

Harwit:

But you donít remember what year and month, or something like that?

Dicke:

Gee, thatís an interesting question. My feeling is that we all decided that this was something important to look for, and got going rather quickly in trying to think of an instrument to do the job. We sort of met as a group on that for probably a few weeks, anyway, before we built anything. I have some dates, as a matter of fact, on when various parts of the experimental program were done, when the antenna was tested and things of that kind, I wrote those in a letter to Bob Wilson (Robert W. Wilson), I think, could get some hard dates for you.

Harwit:

Yes, that would be interesting. It doesnít have to be now. Maybe when we get done, it would be useful. Its always interesting to know on something like this, you know, within a few days, or a week when these things were actually taking place just to see how long these things take to do.

Dicke:

I would guess if you were to take Dave Wilkinsonís publication on actual experimental results, we can do better than that, I would guess maybe withinÖ

Harwit:

Well, in May 1965, you wrote the first paper, these two letters that were back to back, between you and the Bell group.

Dicke:

The whole business might have been started anywhere from six months to a year earlier than that, I would guess.

Harwit:

Good, So David Wilkinson went off and built the radiometer, I guess.

Dicke:

For quite a while we met as a group in designing this thing, but the actual work, real work was done by Dave and Peter Roll, I think.

Harwit:

And Jim Peebles did some of the more detailed calculations on what temperature to expect?

Dicke:

Yes, Just about that time, though, I got interested in the solar oblations problem and devoted my primary efforts to that, though I still met with them occasionally to see what1s going on.

Harwit:

Now, do you recall any of the circumstances dealing with the Bell Labs group?

Dicke:

Yes, I have a clear recollection of that, of receiving a telephone call from Penzias, saying that he heard that we were getting set to look for this radiation and they had something strange that they didnít understand. Several of us decided to go over. Dave, and, I think, Peter Roll and I went, and we looked at what they had, and I gave a, I remember, a short blackboard talk on what our ideas were.

Harwit:

This is informally in the group there?

Dicke:

Yes, I donít think there was anyone but Penzias and Wilson present in their group, if I remember right, and in ours there were the three of us. We saw their apparatus and saw what they had done. I think we were rather convinced that they had seen something significant. Iím not sure what we didnít have some trouble convincing them, but...(laughter)

Harwit:

Yes, to some extent they seemed like rather reluctant heroes. The initial paper that they published in tandem with yours wouldnít say much more than ďa possible explanationĒ for this noise they were measuring. Do you find that amusing in view of the fact that they then got these enormous rewards?

Dicke:

Well, I take a philosophical view of that. In a way, one would think that pure serendipity shouldnít be so strongly rewarded, but from the standpoint of the Nobel Prize committee, what else could they do. They couldnít very well give a prize to a committee. How many ways can you split it? They did, after all, discover it.

Harwit:

Well, there could have been the alternative of giving it to Alpher and Herman.

Dicke:

That, I would have found objectionable for some reason, I donít know why.

Harwit:

Oh, why donít you elaborate because Iím interested in that. You know, to some extent the physics community has tended to pride itself on valuing an incisive theoretical prediction, more highly often than even the measurements which then verify that prediction. And when the measurements are not only are made by chance; but have to be explained to the experimenters by somebody else, it seems to me that the choice between those two might be quite clear-cut.

Dicke:

Okay, I think Iím rather almost inclined to agree with you on this, that in one case youíre rewarding pure serendipity and in the other case youíre rewarding a model thatís an interesting one, ultimately turned out to be more or less right but wasnít really pushed. In fact, they gave it up when they realized there was a problem with the beryllium problem, and so on. When they realized that you needed the stellar interiors to produce the heavy elements, thereís no evidence really that they had continued to believe this after that time, unless Iím wrong.

Harwit:

Well, I donít think thatís true. The realization that you couldnít go beyond mass five, I think was due to Fermi and Turkevich, around 1950, I believe. Thereíre papers by Hayashi and then later still by Alpher and Herman, together with another collaborator, Follin, in 1952 and Ď53, where they not only refine the data on the hydrogen/helium transition, and sort of are mainly interested in that, with I think, no mention, of the heavier elements, if Iím not wrong. They also in that paper talk about the radiation density now as compared to earlier times, both for photons and for neutrinos. The only thing they donít do in that paper, which Peebles referred to in some of his calculations, is they just barely allude to the earlier work where they only had been dealing with the radiation, I mean, what they do is they talk about the radiation in terms of relative densities from the earlier times to from the earlier times to the present, but donít come out with a 5į K figure which the earlier paper did have in it, or actually theyíd mentioned it in several different places. So Iíd been struck actually by...Well, to be sure that they had initially done a lot of the things with a view to explain the heavy elements and that fell through altogether. Youíre completely right there. But this part of it with the radiation seemed to hang in there.

Dicke:

Well, I probably shouldnít speak because Iím not that much up on...not having read any of the stuff in recent years at all.

Harwit:

But at the time, evidently it didnít make a very good impression on you.

Dicke:

Well, Iíll tell you the one I was impressed by in looking at the papers in the old days was Gamow. Gamow was an amazing character — brilliant ideas. Seldom pushed them through to completion. Seldom did the hard calculations. But when it came to ideas, he had them.

Harwit:

Well, itís an interesting sort of a....I mean, Iíve been trying to figure this out. Thatís why a lot of these interviews. Iíve been trying to understand the whole sequence of events and peoplesí perceptions about them, Gamow, the only person who is not alive anymore in that whole group, is perhaps the most enigmatic. I mean, he certainly had a lot of these ideas. On the other hand, the radiation idea does not seem to have been his, by any data that Iíve come across.

Dicke:

You may know more about this than any of us.

Harwit:

Well, you know, Iím trying to understand it. But the other thing about him which is interesting in this context is that the temperatures for the radiation that he was citing sort of went all over the place, even well after Alpher and Herman had made these detailed calculations. He forever, I guess, was making back-of-the-envelope calculations, the kind you referred to that was him, I was looking at one of his popular books which came out in the early fifties, Ď52, Ď53. I think itís called the Creation of the Universe, or something like that, and he talks about a 50įK temperature there, So when people talk about Gamow is really being the person who was behind this, and you look at these very detailed calculations of Alpher and Herman who come up with the modest 5įK figure, and then look at Gamowís figures that range all the way from 7į up to 50į in various places, I guess, it tends substantiate the idea maybe that it is Alpher and Herman rather than Gamow who on that very specific point, never mind Gamowís really much more broad scope. Thatís not, you know, under contest here, I mean, he certainly had that, and I think everybody agrees.

Dicke:

How does the Hoyle/Tayler paper figure into this?

Harwit:

Well, Hoyle and Tayler are interesting to me because they indicate that some people, at least, remembered the Alpher and Herman work, although they tend to also refer more to the Alpher/Herman/Follin work, which dealt mainly with the chemical evolution, the hydrogenóhelium work, But they do worry about the background radiation, I remember actually my own work with the rocket infrared and looking for background radiation there around Ď63, Ď64 when we started, being motivated in part by a remark of Fred Hoyleís at one time — I had postdocted with him for a year — in which he was saying, you know, thereís all this energy thatís liberated in stars in the hydrogen-to-helium conversion, that ought to be somewhere in the universe, kid here youíve got to remember that he was arguing as a steady-state theorist, so he wanted to produce everything in stars. He realized that there was energy there which ought to be part of the background radiation; and since it wasnít in the visible, it could be in the infrared, just red shifted. When we started doing rocket observations with cool telescopes, one thing to look for since we were going above the atmosphere where we could make absolute measurements with a liquidóhelium or liquid-nitrogen cooled telescope — and we built both, nitrogen first and helium afterwards — was this background radiation, because it ought to be reasonably substantial. We should have been able to....

Dicke:

Were you looking for this in a rotational spectrum?

Harwit:

No, we just thought that at that time it was a very crude age. We didnít know at what wavelength it would be. It would have been continuum stellar radiation.

Dicke:

So you were looking at background in general not thermal necessarily.

Harwit:

Thatís right. We didnít know it would be thermalized by any means. It shouldnít have been, in fact, in a steady state.

Dicke:

Well, itís hard to understand how it took us so long to find out just the scope of the Gamow/Alpher/Herman work. As I say, I remember only that one talk. I remember sicking Peebles onto this to look into it, and I donít know why his search didnít turn up something more substantial.

Harwit:

Well, Iíve talked to Jim about that, and he says he always enjoyed calculating things so much that he didnít feel like...He didnít want...You know, he would do his calculating for himself, I suppose, if Iím paraphrasing it right or quoting him correctly. He really was — and Iíd have to look into his interview to get it exactly right. He enjoyed doing this stuff, and he wasnít that interested in priorities. I suppose he went through a literature search; didnít find it. He knew about the Alpher/Follin/Herman paper, although, I think, somewhat later only.

Dicke:

I think he knew about some of these papers, but I donít think he quite dug all the significance out of them.

Harwit:

Well, I think for Alpher and Herman the tragedy, if there is one, would come from their not having referred to their background radiation work specifically in the Alpher/Follin/Herman paper, except in passing. I think the reference is there. Itís mentioned at the beginning of the paper where, you know, very often you say previous work by, and then there are four references, has discussed this; and then you never mention them again. They donít mention in that paper that they had a specifically derived 5įK temperature that they had previously gotten. The other question, and maybe I would sort of like to get your feelings on this. I mean, I have asked some people who knew about them at the time. They were, in fact, interested in following it up. They went around the Washington area; went to NRL, which was one of the few places at the time that was doing radio astronomy; and tried to see whether anybody would be willing to make these observations.

Dicke:

They never saw my paper?

Harwit:

They never saw your paper. Just as you never saw theirs. (laughter) Evidently. You know, itís strange, I suppose they were just looking in the theoretical literature. Its even stranger because Herman had done an experimental thesis dealing with infrared radiation when he was at Princeton for his Ph.D. work — sort of a solidóstate thing.

Dicke:

Who did he work with, do you know?

Harwit:

I think it also was one of these thesesÖ

Dicke:

It could have been Ladenburg?

Harwit:

With whom?

Dicke:

Ladenburg.

Harwit:

No, I think he also was doing this on his own and somebody else took responsibility, but thatís my understanding. He and another graduate, just as you had, had also worked on an experiment of their own which dealt with infrared bolometers, I think, and they made some solid-state measurements — solid-state effect measurements. He knew something about the techniques, I mean, he had switched over to being a theorist in the meantime, but I think he knew...Itís a puzzle. I think one just doesnít...I asked him about that one time, and asked him whether Bob Herman, the theorist, was forgetting about Bob Herman, the experimentalist; and he looked at me ruefully and said, "Well, there must be something to that." You kind of compartmentalize even within yourself sometimes. I was going to ask you, Alpher and Herman both went into industry shortly after the mid-fifties for various personal reasons. I think in the case of Herman it was his need for money. I think he has a daughter who was born retarded and a great financialÖor maybe even twins, Iím not sure. Thatís quite a financial burden. Alpher also went into industry.

Dicke:

Where did Alpher do his graduate work?

Harwit:

Alpher did his graduate work with Gamow at George Washington University.

Dicke:

I see.

Harwit:

But then he and Herman did this...they were both working at the Johns Hopkins Applied Physics Research Labs, and they were doing this in their spare time in the evenings, working on other things, military-more things in the daytime. Now, do you think that the fact they went into industry and perhaps were lost from the university scene, that that could have had any influence on the recognition they received?

Dicke:

I think thatís quite possible. There is, after all, the network of visiting back and forth between labs, and so on. You donít have that much of that if your in industry — within industry itself if two companies of related business might have people going back and forth, but I donít think youíd have quite the visiting as you would have between universities. Without that sort of personal contact of going to international conferences — you know the works — youíre not as apt to be as well known.

Harwit:

What was your reaction initially when...How soon did you find out that some of this work had been done by Gamow, by Alpher, and by Herman — this whole complexion of synthesis of helium from hydrogen in the early universe and also the question of background radiation?

Dicke:

I think it was quite a little while. It was not that first paper but a second one that was being written that I looked over. I donít remember who the authors were anymore. It was certainly Peebles and Wilkinson who were involved in that. I raised the question, and they pointed out that these temperatures had appeared in a paper by Alpher and Herman. That was the first Iíd heard of it. It must have been at least six months after that publication.

Harwit:

Well, itís interesting that Peebles sent in the hydrogen-to-helium conversion in the early universe, that paper to Physical Review in March, before you people sent off your paper on the microwave background radiation at the beginning of May.

Dicke:

I donít have any clear recollection of that paper. I know that it came out but I donít remember it.

Harwit:

That was rejected by the Physical Review. Pasternack had known, I guess, whom to send this sort of thing to, and the reviewer was fully aware of the previous work by Alpher, Follin and Herman...

Dicke:

I see.

Harwit:

...and rejected that paper on the grounds that there was not sufficient new material in there. So, there were some people around...

Dicke:

It might have been at that time that I found this out for the first time, but I didnít even know about that paper being submitted.

Harwit:

Well, you know, itís just interesting. Itís a reasonably well-established fact, I guess, that memories are fairly short in scientific papers. Thereís a series by Helmut Abt of publication citations which shows that they peak around five years and then decline; and by the time fifteen years pass, which is about the length of time that was involved here, thereíre very few citations to a paper anymore.

Dicke:

I ran into a remarkable exception that rule the other day. I just happened to be in the library, killing time, and I looked up the citation index of the members of our faculty, and looked at the Bargman citation list. It was remarkable. They were mainly old papers, but each one was cited over and over a list like this on each paper. In other words, the few definitive papers heís written just keep going and going.

Harwit:

Well, I guess, Abt did find that some papers, the most cited papers, kept getting cited for a longer period of time than the others, I mean, this was despite the fact that by that time they must have been incorporated into review articles, which is — normally people say, ďWell, one doesnít cite any original papers anymore because after a while you just cite the review articles.Ē That doesnít seem to be the reason there. Once you did the microwave background work, then you really devoted yourself almost entirely to the solar oblations.

Dicke:

I sort of shifted my effort over there even before much of an apparatus was built. We had gone through the preliminary design arguments and all the rest, and something was underway. I think about the time they were testing antennas, and so on, I was beginning to devote almost full time to design of the instrument.

Harwit:

Was that because Dave Wilkinson was quite capable.

Dicke:

Yes, it was in competent hands.

Harwit:

But you werenít tempted to do some of the other things isotropy?

Dicke:

I was torn because I considered both problems important.

Harwit:

Yes, sure, sure. Let me see, Iím going to probably want to go to the next tape in a few minutes. Are there any things that I havenít asked you about any pieces of work, any particular points of view. There are a couple of things that I would still like to ask you, but I was wondering whether there was a specific piece of work that you enjoy thinking about that we havenít covered.

Dicke:

Iíve considered physics to be a great game all my life. Thereís lots of things like that, but I canít put my finger on any one.

Harwit:

If you look back, what is one of your favorite pieces of work, or cite two or three, any number that you have really enjoyed and feel proud of?

Dicke:

Well, I think in a way the effect on the spectroscopic line width of putting a gas in a bottle with good walls, and hydrogen hyperfine measurement that we made with Jim Wittke, my student on that one, was one of the high points I got a big kick out of. That was a nice piece of work.

Harwit:

I didnít ask you about those because you had already discussed them with Joan Bromberg. You enjoyed that more than some of the solar oblations things, for example?

Dicke:

Thereíre a number of thingsÖ

Harwit:

Or is it a shell-shocked reaction to the communityís response to

Dicke:

In a way, this solar oblations is like being seated on a wild mare — you hang on for dear life, and you canít get off. (laughter)

Harwit:

So, itís nothing that you look back on with pleasure.

Dicke:

Nothing that I get any great pleasure out of now, but it has to be done.

Harwit:

Yes, sure. Youíve also been involved in a lot of honors over the years. Youíre in the National Academy, I guess, and youíre also have been awarded a Rumford Premium, I believe, isnít that right? What was that awarded for?

Dicke:

I canít tell you what the citation was something involving radiation and light, and it was, I think, primarily the radiometer and things like that.

Harwit:

Not the microwave background?

Dicke:

No, it was before that. That was quite a long time ago.

Harwit:

Okay, I didnít realize that. I thought maybe it was...Have you been cited for the microwave background radiation work, in any of these?

Dicke:

I donít think explicitly. It appears by implication, at least as a component, in a couple of citations. I think the National Medal of Science is one. Letís see what else; maybe thatís the only one. No, Iíll tell you another one, the Comstock award of the National Academy, is one. I think in both cases itís sort of recognition of a broad range of things.

Harwit:

Let me go to one more tape here. We were just talking about prizes, and I wanted to get your general reaction to the idea of prizes in the community. What their purposes are and whether you think theyíre a good idea; whether you think they are abused sometimes; whether theyíre beneficial or not?

Dicke:

Well, off the top of my head, I would guess theyíre a good idea, as long as thereíre not too many of them, and as long as a competent committee does a good job of appraising a situation. Itís probably inevitable that the stakes be made, or at least the choice be not the perfect one, but only a good one. And politics being what they are, that probably plays a role too.

Harwit:

Well, by politics do you mean that people sort of actively seek out prizes?

Dicke:

Well, not necessarily actively, but take two individuals who are equally competent as scientists. One can be very outgoing, go to all the meetings, always make a bright remark at the right time; and the other can be very quiet, and sit in the back corner. Itís obvious whoís going to get the prize.

Harwit:

Yes. Well, this is one of the objections that has been raised from time to time to that; and do you feel that in spite of this, perhaps, the stimulus that prizes provide is worth it?

Dicke:

My feeling is that is in fact the point, that despite the objections I just raised, I think the value in providing a goal for the scientific community, a stimulus is such that it far outweighs the undesirable effects.

Harwit:

What about as one of the possible undesirable factors, what about, in particular, the Nobel prize, which in prestige is so enormously greater than almost any other. Well, in fact, I think one could say itís the only one Time magazine has heard of, more or less. To some extent one has the feeling that once a person has been awarded a Nobel prize, that he or she suddenly is elevated to a position which is quite unrealistic.

Dicke:

Usually he, I might point out.

Harwit:

Usually he, yes, is elevated to quite unrealistic position which carries over into such things as advisory committees, and so on.

Dicke:

I think thereís little doubt about it. I think it does have the effect of loading on the individual in question chores that he wouldnít otherwise have.

Harwit:

I was thinking of it the other way around — loading on the community one more individual who is capable of making pronouncements that could then perhaps null out that of a fairly large body of people. Or do you see that as that as

Dicke:

Well, I see it in a slightly different way. Thereís been a tendency, at least, in some cases, after the Nobel prize is awarded, the individual is less productive than he was before. Despite objections of that kind, I think itís still the overall effect of providing that goal, for the whole community to shoot at outweighs the disadvantages.

Harwit:

Now letís see, youíve mentioned that as a goal.

Dicke:

Not necessarily a goal; that isnít the right word — challenge, if you like.

Harwit:

Well, even as a challenge. I mean, are you saying that perhaps the prizes are challenges that people are going after in some deliberate fashion?

Dicke:

I would hope not, but a challenge in the sense that it provides an impetus for doing your best work and being as productive as you can. The question you were just alluding to is the factor, that it is possible, like studying for an exam. It is possible to work toward a prize.

Harwit:

No doubt.

Dicke:

I would hope there would not be much of that, but Iím sure it is a factor. I can even think of some examples.

Harwit:

I imagine, yes. The other thing is that I suppose everybody who is in a community where he is surrounded by people who are very good and are potential contenders, that one tenth to see around the beginning of October a lot of people who are quite nervous and unhappy.

Dicke:

Yes.

Harwit:

And itís a disease which strikes only the very best members of the community, and so you start wondering whether to some extent that isnít a counterproductive...when a prize gets to be that prestigious that perhaps it starts to be counterproductive, because those people whom you would like to reward most are the ones who are also the unhappiest.

Dicke:

I know of really only one case of an individual being quite jittery about Nobel prize time, and I donít think itís a serious problem but I think it exists.

Harwit:

If you grant the existence of these very prestigious prizes, would you feel that they ought to be awarded for a lifetimeís work or for an individual piece of work, or one shouldnít lay down rules?

Dicke:

I would bet that you out to leave it to a committee to decide, or at least some group to decide on principles and procedures.

Harwit:

Have you ever looked at any of the historical studies of the Nobel awards that have been made?

Dicke:

No, but itís interesting.

Harwit:

Itís a mareís nest. Yes, in fact, there are only one or two articles out, but they really make you shudder. Itís quite interesting reading.

Dicke:

I know some of the early Nobel prizes were for really quite menial things — a new buoy for sailors and warning bell or something like this.

Harwit:

I didnít know that. I see. No, itís certainly true, I think, at least most people who have been given a Nobel prize in recent years certainly are quite outstanding individuals. Itís just that the community is so enormous now, compared to what it was in 1900 when the prizes were first given out. Itís just terribly competitive, I suppose.

Dicke:

In a way you can say there are some of the same objections be raised to some of the honorary societies that exist. The competitionís there.

Harwit:

Are you thinking of the academies?

Dicke:

The academy, in particular, Iím thinking of is, when it comes to the time for new members, there are some people who get quite wrought-up over whether a certain individual should be in out out, and so on.

Harwit:

Really.

Dicke:

Well, not necessarily directed so much at an individual, as a battle between divisions. There ought to be more physicists elected than chemists, that type of thing.

Harwit:

I see. Do you feel that there is a conflict in this country in having an academy which is officially appointed by Congress to provide advice to the Government, but which while doing that is largely dedicated to providing honorific appointments to people, not for the purpose of advising the Congress, because many of the academicians donít?

Dicke:

I would bet that the two aspects of the National Academy complement each other, that the members that are elected feel at least some obligation to serve once in a while. And the existence of a group of capable scientists for national emergency, from the stand point the Government, should be very useful.

Harwit:

I see. I think Iíve asked everything that I had. Would you like to add anything at all about personal views on research, structure of universities of research efforts as a whole, funding of research, relationships with students, with colleagues.

Dicke:

I think Iíve already expressed the opinion that the way funding was handled some two and three decades ago was probably superior to what we have now. Even in the NSF you see too much writing of huge proposals, the ones I get to review — big fat documents. To support a good scientist to do his thing is better for the Government and better for the country, I think, than to support him to do something which some bureaucrat thinks he should be doing.

Harwit:

So you would be against having too strong an applied motive?

Dicke:

Yes, thatís another thing. I think thereís good reason for the NSF to be supporting engineering research or fundamental developments at an engineering level, but to be supporting applied science for the benefit of certain groups, applied science, as such, I think, would be a very bad idea.

Harwit:

Where do you draw the distinction between engineering and applied science?

Dicke:

Well, engineering science — you could probably find some crude examples — one might take the development of network theory in a very broad, generic form, that would be useful to the whole electric engineering community. Thatís sort of a development that the whole community benefits from. Itís perfectly appropriate, I think, if the development, whatever it is, is for a very narrowly drawn group, I donít think the NSF should be in the business.

Harwit:

The only thing that is a little problematic is, as far as I see it, in the question of supporting people who are good in their own rights is the question then where do you draw the line? I mean, are you speaking solely of the natural sciences, and if so, what about the excellent people in the social sciences? Thereíre excellent people in the humanistic disciplines. Does the Government have an obligation to support then?

Dicke:

Possibly, but I donít think the NSF does. I think, when you have a good institution established and itís taken years of effort to do it, you donít want to jeopardize it.

Harwit:

Oh, that makes sense. Itís certainly is true, I suspect, that an institution would be better at funding a discipline about which it is informed in an intelligent way rather than one about which it isnít, so I can see where maybe a mandate to the NSF to do social sciences or the humanistic disciplines would be misplaced. When you say, one really wants to support excellent people, then there are also excellent scholars in those areas.

Dicke:

I see no objection as long as you have some objective criterion, for deciding on excellence. I see no objection to other institutions similar to NFS taking on humanities. We already have one in the case of the arts.

Harwit:

National Endowment of the Arts and National Endowment for the Humanities. The social sciences seem to fall between the cracks at the moment, and I think thatís somewhat of a detriment, because, again, social scientists could tackle broad questions, similar to the network ones that youíre talking about, without getting involved in individual political questions.

Dicke:

And at times in social sciences some of them almost seen scientific.

Harwit:

Thatís right. I had one other question. Nowadays thereíre increasing number of collaborations that one sees between people of different institutions, people in different countries; and to some extent sometimes the agencies fund things even have overt policies where they prefer funding joint ventures which they then see particularly desirable because thereíre so many different people wanting to do this. Iíve noticed that I donít think thereíre any papers at all of yours, or very few anyway, where you have collaborated with anybody except your immediate group. That must be a deliberate choice on your part. I was wondering how you feel about these different approaches.

Dicke:

I guess itís always within a group. I was thinking of one case where the group was rather broad, and thatís the papers on the laser-ranging to the moon.

Harwit:

Thatís true, yes.

Dicke:

Thatís rather a large group.

Harwit:

Yes, thatís a large group.

Dicke:

I think except for that, there hasnít been very often collaboration outside the University. I think, I canít remember now, did I ever write a paper jointly with Purcell on the spin exchange in atomic hydrogen.

Harwit:

I didnít see it on your list, unless I missed it.

Dicke:

I think youíre right; there never was such a paper. I think we talked about it at one time.

Harwit:

That would have been part of this idea of coherence that you had. No, this was an idea of Edís, that relaxation in the proton resonance, in actually the hyperfine transition of atomic hydrogen, would take place through electron exchange in a close collision between two atoms.

Harwit:

Oh, I see.

Dicke:

When I was on sabbatical leave there, I worked that up more broadly.

Harwit:

I was wondering, in fact, whether when you did the paper on coherence in spontaneous radiation processes, whether you had ever thought about an astrophysical application, because with hydrogen number densities of the order of unity in interstellar space, you would have quite a number of hydrogen atoms per cubic wavelength.

Dicke:

I had thought about spontaneous breakdown without mirrors in an excited system, but never thought about this in astronomical terms. Thereís a paper that I gave at the Physical Society one time in which I talked about a radiating system in form of a needle — a needle array of atoms — and showed what would happen if they were all excited and that you got a very strong radiation along the length of the needle, I think thatís probably where the words super-radiance first appeared.

Harwit:

Okay, yes, because of the stimulated emission. Thatís one wayÖ

Dicke:

I look at it in a quite different way. Theyíre related to each other, but the way I looked at it is this: You start out in a pure state in which all the atoms are excited, then one atom emits a photon, that direction of emission is now preferred for the next photon as well.

Harwit:

Thatís right.

Dicke:

You can show this in general.

Harwit:

Sure.

Dicke:

And that once you have a configuration like a needle, this is strongly preferred and then you get aÖthereís no talk of stimulated emission at all in that anywhere.

Harwit:

But when you say preferred direction, one of the aspects of stimulated emission is that it comes off at exactly the same direction.

Dicke:

In the case of stimulated emission, itís the radiation by one atom that acts on another atom to induce a transition. Thatís what you have in mind. The picture I have is quite a different one. Itís a completely quantum mechanical picture in which there are no photons present initially in which the atoms are all excited, and now a photon is produced. Imagine that photon absorbed, so we now again have no photons present, but the remaining atoms are in a coherent state now. That excited atom is distributed throughout the whole system.

Harwit:

I see, I see, yes.

Dicke:

In the next transition that occurs, that distribution of excitation is remembered, and the next photon that comes off — thatís a preferred direction — absorb that photon. Thereís never any talk of stimulated emission in this at all, and that gives you the right answers too.

Harwit:

But they must be equivalent in some way?

Dicke:

That equivalence has never been completely worked out, that I know of.

Harwit:

I see. I didnít ask you at all about the laser ranging, and I should have, I guess.

Dicke:

Yes, there are some amusing stories about that.

Harwit:

Go ahead.

Dicke:

Our research group used to meet once a week regularly at night, and we had discussions and free ranging — crazy ideas. One night we were discussing the possibility of a pulse searchlight. We had veins you could open and close, like this, and send light up to a satellite that had corner reflectors on it to come back so we could get a precise distance to the satellite, and back and get a good orbit for it. I think the next meeting Jim Faller, one of my graduate students (heís been at the Bureau of Standards for a long time now at Boulder), came in carrying a rubber ball about so big that he had hollowed out, and inserted a glass prism in and rolled it across the floor. The thing wobbled around and stopped with the mirror pointing up. He said what weíd do is, on the moon as a big ship goes by, youíd throw these out on the moon, the retroflectors all pointing up, and then you...I guess this was...I didnít think at the time we discussed this first there was a laser yet. I know some of the early thought about it was prior to lasers. Maybe at this time the laser already existed.

Harwit:

Well, laser radar existed around 1961, Ď62 for atmospheric probing. I think itís ďlidar.Ē

Dicke:

Yes, I think this must have been after the laser. But in any case, we had more discussion about this, and Carrol Alley...Well, we met once in Washington, the whole group of us, and several from outside our immediate group, as kind of a Physical Society meeting and had lunch together. We decided we ought to organize in some loose way, and some one of us ought to take some responsibility of approaching NASA to see if we could get funding for things.

Harwit:

Now how did this group originate?

Dicke:

I donít know who ever decided on...I canít remember now.

Harwit:

There was a deliberate effort to have laser radar.

Dicke:

It was mainly our own group at Princeton.

Harwit:

You then called up some other people who mitt potentially be interested in...

Dicke:

I suspect that Carrol Alley probably added ones outside. Anyway, we met at the Physical Society meeting, and it was mainly at that meeting, as I recall, that they were almost all Princeton — people from our own group — some had already left the group. I think Carrol was already gone; he was at the University of Maryland.

Harwit:

Oh, he had been here before that, is that right?

Dicke:

He was a graduate student of mine.

Harwit:

Okay, yes.

Dicke:

Then we broadened this group some and went to NASA. He got a contract from NASA. There was no group recognition at that time. It didnít officially exist as far as NASA was concerned, and some of us found this rather objectionable. We were finally organized within NASA as a so-called ďLunar Ranging Team.Ē We used to meet periodically, and went ahead and designed a silver reflector system and laser system, and so on, but always with rather little money. It went on year after year — quite a long time. Finally when the first Apollo was ready to go, it was within a month or two of launch time, NASA suddenly decided that they werenít going to let the astronauts tromp around outside long enough to set up the ALSOP (an acronym) package of experiments that they were supposed to take. I donít know what that stood for, but there were seismometers and temperatures sensing and things like this. They werenít going to allow them out of the spacecraft long enough to set this up. If theyíre not going to set them up, there is no point in taking them up there — that means thereís space available. What can we put in that space thatís simple? And somebody thought of these reflectors. Thatís the reason that reflector package went up on the first flight, otherwise it would have never gone. But as soon as that decision was made, money flowed like water. (laughter)

Harwit:

That sounds like NASA, yes.

Dicke:

So that group existed. That team existed for a number of years, itís now completely disbanded, but theyíre still ranging on the moon.

Harwit:

Who is doing that now?

Dicke:

Itís The University of Texas. Whatís his name, the chairman down there?

Harwit:

Harland Smith.

Dicke:

Harland Smith has some kind of a responsibility in it, I know. They still have a group there involved in it. Thereís a new telescope starting up in Hawaii now on the volcano.

Harwit:

And theyíre going to continue doing this laser ranging?

Dicke:

Yes.

Harwit:

You already got some rather interesting results out of that.

Dicke:

Oh yes, quite interesting, I think.

Harwit:

Do you feel that one can get additional information?

Dicke:

Well, itís like all these things I think. Itís an important scientific number to have, and to have these with precision you have a system for getting it, you like to keep going.

Harwit:

So you think you can get another factor of ten or so eventually with better lasers and betterÖ

Dicke:

Conceivably. Ranging has improved now substantially with the new laser, and last I heard it was almost a factor of ten better than it had been. And there are some very subtle complications involved in the moonís motion problems involving the physical libration, yielding, the hysteresis in that yield, and the effect of minor lunar quakes.

Harwit:

Do you have any hopes that one might learn more about the history of the origin of the moon from, a question that you had thought about at one timeÖ

Dicke:

Yes, I havenít thought about it from that point of view specifically.

Harwit:

Well, there use to be questions about whether the moon was spiraling in, spiraling away, that kind of stuff.

Dicke:

I think thatís rather clear now from those measurements that the moon is spiraling outÖ

Harwit:

Well, thank you very much for your willingness to talk so long.

Dicke:

There were a few things you wanted to see, I think, I canít remember what they all were now.

Harwit:

I think one of them was a letter you said you had written to Bob Wilson.