Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
We encourage researchers to utilize the full-text search on this page to navigate our oral histories or to use our catalog to locate oral history interviews by keyword.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Robert Frosch by David DeVorkin on 1981 July 10,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
For multiple citations, "AIP" is the preferred abbreviation for the location.
This interview reviews Frosch's early schooling in the New York City Public School system, his education at Columbia University and, in detail, his varied career as a physicist and a science administrator, beginning with his work as a research scientist at Hudson Laboratory and then as Assistant Director and Director of the Theoretical Division. Other topics and affiliations discussed include: Advance Research Project Agency (ARPA); United States Navy; United Nations Environmental Programme; Woods Hole Oceanographic Institute; National Aeronautics and Space Administration (NASA) (administration, research and development techniques, business practices, reorganization, communication); Naval Research Laboratory (NRL); Navy Radiological Laboratory; National Academy of Sciences (NAS); Wallops Center; Dryden Flight Research Facility; Goddard Institute of Space Studies; Office of Naval Research (ONR); JASON Foundation for Education; Jet Propulsion Laboratory (JPL); Institute for Defense Analyses (IDA); Office of Management and Budget (OMB); SOLRAD (SOLar RADiation Satellite Program); Enterprise; Mansfield Amendment; CLOT test (combined loads orbiter test); Gamma-Ray Observatory (GRO); VELA Program; Atomic bomb; relationships between US Navy and NASA; militarization of space exploration; space exploration policy; sonar navigation; musical theory and acoustics; project management theory; satellite communication systems; underwater acoustics and modelling; remote sensing; seismology; shuttle flight testing and preparation; I.I. Rabi; Henry Foley; Charles H. Townes; Ivan Tolstoy; Frank Press; Jimmy Carter, among others.
Dr. Frosch. I would like to start out with some biographical information. I know you were born in 1928 in New York City, but I would like to know more about your family background.
How far back do you want to go?
Your father and your mother.
My father was a physician who was born in Europe in 1894 in the Austro-Hungarian Empire. It is now most accurately described as Southern Poland. He went through gymnasium, through part or all of secondary school there, and came over when he was fourteen. He went to the New York City public schools, and then to Columbia College, and the College of Physicians and Surgeons. My mother was born in the U.S. on the lower East Side. She was a librarian. They met in the library. Curiously enough, their families had been neighbors in Europe. I was born somewhere in the East Bronx. When I was four years old — I don't remember exactly — we moved to a combined office and apartment, Tremont Avenue and the Grand Concourse, in the middle of the Bronx. And I grew up there, embedded in my father's medical practice, because it was a combined office and apartment.
Was he in general practice?
He had done some research, but had given up on it. He was a general practitioner and internist. He specialized at one time in hematology, but I would say he principally worked in cardiovascular and internal medicine. Earlier, of course, he had been a general practitioner, and a pathologist at one time. From the period I can remember, he was a consultant. While he had some patients who were his general practice, most of his time was spent responding to requests from other physicians to give an opinion. I guess he characterized himself as a diagnostician, and he described himself as a Sherlock Holmes of medicine. He would get the cases when nobody else could make a differential diagnosis, because it was too confusing, or get a treatment situation where it was not clear what to do, and so on. So, for biographical purposes, there was a good deal of medical scientific discussion at the dinner table.
Were you an only child?
No. I have a younger brother who is four years younger than I am. He became a physician. as a matter of fact. He's a physician and a psychiatrist in New York City.
Could you give me the names, your father's full name, your brother's full name, and your mother's maiden name?
My father's full name was Herman Louis Frosch. My mother's maiden name was Rose Bernfeld. My brother's name is William Arthur, although I think his birth certificate says Arthur William.
That's not unusual. Your father's occupation and professional interests were dinner table conversations. What other interests did your family have as you were growing up?
Well, my father particularly was a considerable reader, as was my mother. He had essentially learned English out of a set of Dickens, literally reading himself into the language. That was his account of it. He was particularly a great reader of essayists, of Montaigne and Emerson. In later years, his interests became rather eclectic. For reasons I have never been able to untangle, because I can't read the man, he would read William James and Henry James, and was a quoter of both. He ran through all the novels of Henry James at one point.
Was he very pragmatic?
It's a little complicated to describe. He had a very orthodox Jewish religious background, and was a proper Talmudist. In fact, in order to get his father's permission to study medicine, he had to agree to continue to study Talmud at the same time. I could go through a lot of anecdotal stuff. At any rate, he essentially broke his faith, broke with his religion sometime when he was in medical school. Of course, he never quite lost it. We used to go regularly on High Holy Days to a conservative synagogue, and my father took great delight in discussing things with the rabbi. There was a very strong ethical and moral flavor to the conversation. My father had very strong feelings of morality and ethics and justice, particularly with regard to his profession, to professions in general. He was fond of quoting the Bible, especially Micah.
Did you say that you picked up some of these characteristics?
Yes. I picked up a good deal of it, just by osmosis. For example, I would say there was never any question about whether either my brother or I would go to college, probably graduate school, and become some kind of an intellectual professional. The only question was, which kind. That sort of ambience.
Was there any discussion of which kind?
Oh yes, quite a lot. It was clear my father would very much have liked me to be a physician for a variety of reasons, which I am not even quite sure of. I was never terribly attracted to the medical profession. Although, at one time I was very interested in being a naturalist. If there had then been, as there is now, a more defined ecological kind of profession, I might have gone into that. As a kid I went to camp, and got heavily involved in nature studies, because I had a strong biological background.
You were living in the Grand Concourse Area. I imagine you went to camp in Upstate New York somewhere?
Yes. I went for a long time to a camp at Mahopac in Putnam County. It's not very far out of New York. It's an identifiable place on the map, still.
Did they have a science program, particularly?
No, they had the usual nature program. There was a counselor who was responsible for "nature" and there was one cabin that was a quasi-museum. One could take activities such as leaf collection, bug collection, frogs, snakes, etc.
No. But there is another early biographical strand. Children in New York City at that time were very independent. As soon as I was old enough — which probably meant ten — I was given a dime for the subway, and I could go and do things. In 1940, when I was 12, the World's Fair was held and I would spend the day there by myself. But to answer the question, I frequently would go down to the American Museum of Natural History. There was a long period in which I saw every monthly planetarium show. I saw most of them enough times so that I got selective and wouldn't go back and see it unless they changed. So I was rather science-oriented. It's hard to say whether it was more science or literature, but the professional ambience was science.
What was your schooling like? Where did you go to school?
As it says in the biographies I went to New York City public schools. I went to P.S. 38, which was around the corner. I don't have any very extraordinary recollections of that. At one point I skipped a grade. I can't remember what grade it was; whatever grade in which you learned the multiplication tables, I skipped. They just decided in those days that one skipped, and so I skipped. And I do remember coming into a class in which they obviously were at the beginning of the term doing the next stage in multiplication. It was very dark day, because I didn't have a clue of what was going on. I guess I was a fairly bright kid. I went to the Joseph H. Wade Junior High School, which I remember as being rather more rigid and confining than the elementary school, probably because of the principal. There were a lot of very bright kids in that school. They collected them from a lot of areas. I remember that.
Was there a selectivity system in the New York City public schools?
Yes. they basically tended to filter students, not by tracking as much as sorting them by talent. It was still the era where one identified the letter of the class with the level of talent. It was good to be in 7A and bad to be in 7D, you know; it was a fairly transparent system. That is my recollection of it. So I went through junior high school. I don't really remember much about it, except I had grown up with libraries. My mother was a librarian. My aunt was a librarian. In fact, my aunt was a children's librarian in the local library, and I would often go to the library. I was quite used to libraries, and used to having the run of the library. When I was old enough I got an adult card, and went upstairs instead of downstairs. And I remember going to the junior high school library was supposed to be an event. You came in and sat quietly, and were told about the library. The librarian was the kind who believed that a proper library had all the books on the shelves, so you didn't take a book off the shelf. I found that most peculiar and galling, and complained about it. I was supposed to go to a new high school, William Howard Taft High School. There was a lot of discussion whether I should do that or try to go to Bronx Science, which nobody particularly wanted me to do, and I didn't particularly want to do.
Who were these people, your teachers?
My parents. There was not great pressure to do it.
Was it a difficulty getting there physically?
No, because as it turned out, I went to a school that was more difficult to get to physically. The point was, I was zoned to go into Taft, which was local, within a few blocks, but it was a new school. And for some reason or other, we were skeptical. My father having gone to Clinton, I wanted to go to Clinton. DeWitt Clinton. When my father went, it was on 59th Street. When I went there it was on Mosholu Parkway in North Bronx. For no particular reason, we went through the school system and kind of pushed. The system was malleable enough to say: "Well, if somebody's crazy enough to want to go to Clinton because his father went to Clinton, I guess we can adjust to it. Clinton was what would now be thought of as a bussed school, in the sense that it took students from all over the Bronx and all the way down into Harlem. I am not quite sure why it had this peculiar longitudinal zone, but it did. It may have had something to do with the war years. It was a very large school running on several shifts.
You were there during World War II, then?
Yes. I got out of there in '44. Now that school had an interesting structure. It was a very large school with a fairly heterogenous population, coming from a large part of the city. It would be regarded now as an integrated school, only in the sense that the total population was represented. That didn't necessarily mean there was very much internal integration, although there was a fair effort at it. The way they ran their tracks, so to speak, was to have what they called the "Honor School." They embedded a smaller school with its own classes, its own curriculum and its own faculty in the large school. No freshman automatically got into that. The teachers of freshmen classes were supposed to be in the talent search business, looking for bright kids to push into the honor school. The search was really very eclectic. In some cases, quite a strong attempt was made with kids who weren't sure they wanted to try that, and go on to college. The teachers had to convince them and convince their parents. Anyway. I went into the honor school, and that was a very good high school. The faculty was first class; a large number of them had doctor's degrees. Remember, this is New York City just post-depression. The depression had unemployed many people of a higher educational level than would be teaching secondary school. These people would be normally teaching college or graduate school, or out in the professions. So the teaching was very good. There was a lot of competition intellectually.
I wanted to ask you about the competition. When did you start sensing competition among your fellow classmates, high school, or junior high school?
Right at the beginning of junior high school. The junior high school picked up kids from a lot of elementary schools, and since they sorted them in this track system, you suddenly discovered that although you may have been the brightest in elementary school, there were ten or twenty others who had come out of the same way. Things were very interesting.
Did it stimulate you to work harder, or did it cast a pall on your hopes for the future?
I don't recall that it did anything much, one way or the other, except that it was clear that a lot of other people could do this sort of thing pretty easily, too. That's the best way to describe it. Before I had the feeling that there were only a few of us who could obviously do this school thing very easily, and a lot who couldn't. And then suddenly it was clear that there were many others who could compete.
Right, that's true. You mentioned the depression. Was your family affected economically in any way?
I don't know. The reason I say I don't know is that there were other factors. My father had a reasonable struggle to establish a practice. He got out of Physicians and Surgeons in 1918. It was not easy to build up a private practice, but he was doing it. The complication was that about '32, when I was four or five, my mother became very ill. She had an extreme hyperthyroid condition. In those days the condition was reduced with radium — direct radon needle treatment in a lead collar. She was very ill for a long time. The complications of taking care of her meant that the practice went to pot for awhile. Things were very, very difficult for a period, but I was so young that I am not sure that I had any clear sense of it. I would say it was not a direct poverty kind of thing. What had begun as a reasonable, middle class medical practice went into real difficulty for several years. I suspect that they were in debt, but then it built back. My mother recovered, then things got back on a more even keel. I think it was just the strain of having to take care of my mother and the family, and not having much access to help and so on.
So that by the time you graduated in 1944, there was the question of either the draft or college?
In '44, remember, I was fifteen.
That's right; you skipped ahead in school.
Yes, I graduated early. Because of the way they ran the school during the war, I graduated January 1944. I remember that I wanted to go to College at Columbia.
That was your choice?
I applied to other places. I was rejected by Harvard, but I don't remember where else I applied.
Rejected on what grounds, did they say?
English, I think, which puzzled me. But I was accepted at Columbia. They didn't want to admit a fifteen and one-half year old, I think. They were willing to have me start in July, because the college was on a three term session. I started in July. During the interim, I took a post graduate high school course, probably one of the few people in the world to do so. It was a question of getting a job, and there weren't any obvious jobs around. So I went back and took analytical chemistry in high school. The school gave things like analytical. They didn't give calculus, but they gave analytical chemistry, qualitative analysis.
As you entered college, what were your feelings? Were you still somewhat open as to what you wanted to do?
I described the naturalist interest already. I received a book for a bar mitzvah present. I don't remember the name; it was a popularized kind of paleontological history, about dinosaurs, etc. I read that and began to haunt the part of the Museum of Natural History with the Cretaceous, Triassic dinosaurs, etc. I decided that's what I wanted to do, geology or paleontology, or something of that kind. That was my announced intention, then I had physics, and decided that's what I really wanted to do.
Did you have a dynamic teacher?
I had a very good teacher, who taught, in addition to the usual stuff, what amounted to introductory quantum mechanics, the theory of the atom, but without the mathematics and so on.
What was his name?
That's a good question. I don't remember. Incidentally, I had a first class biology teacher too, and a very good history teacher. At any rate. I was turned on to physics, and I took to reading Jeans and Eddington's popular books, The Mysterious Universe, Stars and Atoms and so on.
You read as many as you could find?
Well, whatever I could get my hands on; the school had a reasonably good library. I can't remember all the titles. I would go to the shelf and pick the ones I hadn't read.
Stars and Atoms?
Oh yes, all of that. I was, in fact, reading about as much popular cosmology as I was reading physics.
Did you know about Hans Bethe's successful determination of the source of energy of the sun in the late 1930s?
Yes, yes. that was the kind of thing that I was aware of, the nuclear chain theory of the sun.
These are things that decided you?
Yes. I was particularly interested in the cosmology. So by the time I went to college, it was my announced intention to be a theoretical physicist, and there it was. I suppose I had a reasonable idea of what that meant, at least in terms of what the subject was. Certainly I knew about quantum mechanics, relativity, the Hubble red shift, and so on.
There was no resistance from your family?
Your father never said, "Can you make money at this?"
Yes, he would raise the question, presuming that I would be a college professor. So I started Columbia with that intention. An interesting series of things happened. Being a very innocent, straightforward type, I proceeded to read the college catalog. The course catalog says that freshmen should read the catalog, lay out a four-year program, and be prepared to discuss it with the advisor. So I marched in to my advisor, who was Professor Robert von Nardoff, Professor of Physics, or Associate Professor of Physics, at Columbia. And I think I appalled him by saying, "Well, you know, I've done what it says, and here's what I'm going to take." Of course. I had laid out a program that ran me through all the mathematics, all the physics and all the chemistry. He looked at it and smiled, and suggested that I ought to try a few other things. He tried to push me away from the heavy physics course. Now Columbia, then as now, even early back to my father's time, never gave up a rather elaborate required core curriculum. It's based on two major course sequences that freshmen and sophomores have to take. One is called An Introduction to the Rise of Contemporary Civilization in the West, known as CC. It is a series of readings oriented toward intellectual history, particularly political and legal and some technological. We read excerpts from the Magna Carta and John Locke and Hobbes' Leviathan and Aristotle, all of that in sequence with an encouragement to go back and read more. We just read like hell. They've made their own text book of excerpts. The idea was to read original documents and then discuss them so that was one strain of course work. The other sequence was called Humanities, which is sort of the literary version of the same. It's a kind of "a book a day" indefinitely. You know, you read the Iliad and you read the Odyssey and you read Aristophanes, and you kind of work your way through. Then in the second year you take some music. and you take some art, and so forth.
Where does science come into this?
It doesn't. It wasn't part of it. Columbia has been struggling for forty years to do something like this with science, and they keep trying, and it comes apart. For people who want to be scientists, it's great. It stretches them the other way. For the people who don't there is a science requirement, but it's never been a very satisfactory arrangement.
How did you react to that core?
I got very interested in it. There was a period of several years in which I kept taking physics and math, but had given up the idea of becoming a physicist. I was toying with other things, and took art courses and Shakespeare, some advanced humanities. My coursework was very broad. I got my Bachelor of Arts, which is what you get from Columbia College even with concentration in the Physics Department. One did not "major" in Columbia College in the normal sense. Meanwhile, I had decided to go to graduate school in physics. But the department decided I didn't have enough physics in my undergraduate courses to be allowed to matriculate as a graduate student.
Had you maintained contact with von Nardoff?
I had some contact with von Nardoff, because he was my advisor, but it was a rather loose thing. His role was to make sure that you got a good education and not to ensure that you could get into graduate school. Nobody seemed very upset that I didn't want to go into physics. By the second half of my junior year, I had resumed my interest in physics. But when I got my degree, the department said that I didn't have enough physics to matriculate. They suggested I take courses as a nonmatriculating student six months or a year, promising to give credit for some courses and not for others. I did that. In six months it all got sorted out, and I think the only thing I didn't get graduate credit for was thermodynamics.
Was there anything about the events of the war, the final events of the war in Japan, that could have caused you to think twice about physics?
What was your reaction to the atomic bomb?
A funny reaction, because as soon as I heard about it, I knew about it in the following sense. As soon as the news was announced, I had remembered reading about the possibility of such a bomb in our standard textbook called Modern Physics, (Floyd Karker Richtmyer and Earle Hesse Kennard, Introduction to Modern Physics 3rd ed. (London: McGraw Hill, 1942)). Nobody knew quite what to do about the textbook during the war, because it had a paragraph describing the fission experiments, and went on to speculate that one could construct an uncontrolled chain reaction and make an explosive.
This is an American textbook?
Yes. It's in that olive-green series of physics textbooks that McGraw-Hill put out. I've still got the textbook. And so I knew that. I would have been aware, of course, because I had been taking a lot of physics courses. I was aware there was something funny going on in the basement, because you couldn't go down there, and there was a guard stationed at the cyclotron.
That was one of my questions.
We knew about that. I knew that some of the junior faculty or the advanced graduate students would go rushing around, and somebody would come in very excited and do something on the blackboard, and say, "Oh, my god!" and go out. They were having these private conversations. All of that clicked together when the bomb announcement was made. I was editing the college humor magazine, "Jester of Columbia." and did some kind of an article on it. I may have somewhere in the attic.
This is after the bomb was dropped.
Yes. It was a semi-humorous piece — what-to-do about radioactivity, wear lead galoshes, this kind of thing.
Do you recall a feeling of excitement?
Oh yes, positive excitement on the whole partly because we had found a way to end the war.
Sensing that something was going on in the basement, and people were running around talking excitedly in small groups, do you remember any people in particular who were involved there?
Oh. I remember Jim Havens was there, and Jim Rainwater; they were both older. I didn't really know them. They were instructors, and I don't remember whether I even had a class or a lab section with one of them. I think I recall having Jim Rainwater as a lab instructor. I took most of the regular physics sequence in college. I would have to go back and look at what the courses really were, but they were whatever would have been standard, with a couple of omissions, and the usual mathematics, and so forth.
As we move on to the graduate work in physics, were you beginning to look at a specialty?
Only the idea of doing theoretical physics.
It was only a question of taking a few more physics courses. and then you could go in. Was there any problem of financial support at this time?
Were you supported completely by your father?
My father had put money away, and so that was no major problem. After my first year in college and graduate school, I lived in the dorms. But if there had been a problem, I could have lived at home and commuted. Remember, those were the years in which a year's college tuition cost $600. That was more money then than it would be now, but even so, my father had put enough money away. He had had some money in the stock market in '29 and lost it, not much. He became very distrustful of that kind of investment, and never invested again. He took to buying U.S. Government bonds. He bought E Bonds; he just stashed enough bonds away so that he could use that.
For both you and your brother?
Both my brother and myself, yes.
Did you ever think of other graduate schools, or you were happy with Columbia?
I was happy with Columbia. It was obviously first class, and had a very good reputation. It was convenient and sensible, so that's what I thought I would do. I began graduate school in the summer of '47. After about six months, I must have talked to somebody at the departmental office, and said, "Well, I've taken these courses. What do I do now?" My impression is that things were simpler in those days. Somebody looked at my record and said, "Okay, we'll give you credit for this, and we won't give you credit for that. Go fill out the forms." So I was then matriculated.
So you had to take more physics courses at that point?
I took a couple of courses for which I didn't get graduate school credit. Thermodynamics is the only one I remember.
You were matriculated in the spring of '48.
I got out of college in June of '47, and then I started right in doing some graduate courses. Remember the first year or two was 3 terms per year. Somewhere in the spring, I was then a formally matriculated student.
There was an Ernest Kempton Adams Fund.
Did you ever take part in that, even though you were supported by your father?
The only thing the EKA Fund ever did was pay for the Physical Review publication of the article based on my thesis, which is largely what it is used for.
You acknowledged that in one of your papers.
Yes, but it was for the publication of my thesis, and I think it was really an acknowledgement of the funding for the publication. I never had a scholarship or a fellowship, never applied for one. There was one term I took a lab assistantship, but that didn't last very long.
You didn't like teaching?
Well, it wasn't that. I was teaching a lab section and I wasn't very good at it. I probably would have turned out to be perfectly all right after a while. I think the problems was that they didn't have enough lab assistantships, and they really wanted them for those who needed the work. And so everybody was perfectly happy that I didn't.
Were you doing any research with any particular professors?
The first year or two was straight course work, until the doctor's qualifying exam. That's the first mark point. The physics department's philosophy was that everybody who gets a doctorate in physics at Columbia becomes a theoretical physicist f irst, even if he becomes an experimentalist afterwards. That is, there were no tracks. You didn't have the opportunity, for instance, to become an experiment physicist until you passed the qualifying exams and began research. For the first couple of years. everybody took a heavy dose of mathematical physics. You took quantum mechanics, electromagnetic theory, mechanics, god knows what. I had a marvelous course in physical optics, a demonstration course.
Who was that from?
Hermon W. Farwell. I remember him well, because he taught me freshman physics. The demonstration course in physical optics consisted of a three-hour lecture on Wednesday afternoons. In the course of the term, he conducted every demonstration that you will find in R.W. Wood's Physical Optics. It was really a tour de force.
You must have loved that.
Yes, and it was really a fascinating thing to see. It took him and an assistant all day every Tuesday and Wednesday up to class time to prepare the lecture. Farwell even did the one in R.W. Wood where he creates a mirage with a heated iron sheet. That was there, as well as a microscopic demonstration of internal and external conical refraction, anisotropic crystal optical phenomena, and so on. These experiments were very difficult to set up.
That must have been a very exciting course. Did it get you back into more non-theoretical physics?
No. I never had any temptation whatever to be an experimenter. I'm clumsy — that's straightforward enough, isn't it? So I just had no direct interest at all. Later I did a lot of experimental work, but not hands-on stuff, because I'm really not good at it.
Were your Ph.D. qualifiers written or oral?
Okay, let me describe that process. You take exams, and then you come up against the qualifying, which was the exam. It was a set of written exams and generally regarded as a great hurdle; some people would take it a couple of times before they would pass it. It was a difficult exam covering all the standard physics you were supposed to have had. After about two years you took it. I must have taken it in late spring of '49.
You had your M.A. in '49. Would that have been as a result of the qualifying?
Yes, that would have been as a result of the qualifying. That exam was regarded as the great hurdle, or the first great hurdle, at any rate. So I took that. and I guess I passed it reasonably well. I don't remember much about the exam, except doing one problem in electromagnetic theory by a method that I later decided was perfectly all right, but rather peculiar.
Did you pass it the first time?
Yes, I passed it the first time. And then there were no more courses to take. But obviously the next question is research.
There were more courses beyond the qualifer?
Yes, but then you may be taking specialized courses, quantum field theory or whatever. There were also visiting lecturers. I took a course in either meson theory or quantum field theory with Yukawa, who was a visiting professor then. It was useless, because he lectured into the blackboard with a thick Japanese accent. He was a very nice man, but ... This must have been in '50 or '51, somewhere in there.
Was there any problem with the Korean War, any possibility of being inducted?
No, because as a graduate student in physics I was automatically deferred. I'll come to that. That issue didn't arise until about 1951 or '52 when I was at Hudson Labs. Up to this point the educational deferments were automatic. Then I was going to embark on thesis work, but I really didn't have any particular idea of what I wanted to do. I had a vague ambition which seems obviously ridiculous to me now, to study the quantization of relativity, how to put those two things together. I never did anything with it, but it was a subject that was in the back of my mind as an obvious kind of problem.
Who were you thinking of working with?
Well, I wasn't thinking of working with anybody in particular. I had wandered through graduate school, and I wasn't quite sure what I wanted to do. I talked to a lot of professors, but I wasn't particularly close to anybody.
What of your student colleagues at the time?
Those who wanted to do experimental work had decided already. There were three major areas at Columbia. There was high energy physics, Charlie Townes's microwave lab, and Poly Kusch's and Rabi's molecular beams lab. The only thing I was pretty sure of was that I didn't want to do experimental work. I didn't know what to do, so I went to Rabi who was the head of the department. I formally made an appointment and went to the head of the department and said, "So here I am. I have passed my qualifying. I am interested in doing a thesis. I don't have any strong feelings of my own. I want to shop around and see what would be interesting to work on." Rabi pulled my folder. His comment was beautiful. He looked at the folder, and said: "Well you're no Schwinger."
Oh, that's funny.
And he said, "Why don't you do an experiment?" I was pretty firm. I said, "No. I really don't want to do that. I'm interested in learning how to do theoretical work." He suggested I go to Henry Foley. Henry was just beginning to take graduate students. Either Andy Sessler was his first graduate student, or I was his first graduate student. Anyway, we were his first two. I went to see Henry, and he agreed to take me on. I was not a bad student. It was just that Rabi had a clear idea of what he wanted. Henry and I talked for awhile, and then Henry gave me an apprentice problem. It was a straightforward quantum mechanics matrix theory computation problem. You know, get the matrix elements of something or other. I probably have some notes in box. It was essentially a computational problem, and I set out to work on it. It happened to be one that Rabi was interested in. They wanted the answer; they didn't have those matrix elements and they wanted them. And I set out to do this; it was a three month problem. I sweated over it for awhile, and figured out how to do it, and did it. There is an interesting story related to this. Rabi knew who I was, and he knew I was working on the problem. One day I ran into him and talked about the problem. And he said the answer is going to come out like this. I said, "Yes, really, why is that?" And he gave me an analysis, kind of a hand waver, saying the spins are this way and that way and so on. This was my first brush with Rabi. Later I realized that he had given me the correct answer. Rabi always tells you the correct answer, but his reasoning was absolutely wrong. I later proved, through the proper quantum mechanical analysis, that his answer was right. I have come to the conclusion that Rabi's most important quality is that he knows the answer, but he arrives at it through some process to which he does not have access.
Then he tries to construct some reason for it, and usually the reason is kind of nutty. It's really very interesting. He's fantastically brilliant and right, but his rationalization of why he is right is frequently not. I remember a theoretical colloquium which described the problem of anomalous lifetime of the mu meson. Rabi wore a tuxedo. It was a five o'clock colloquium, but he had a dinner to go to. He had wandered in, wearing black tie, and he sat in the front row, carving his initials. He did that sort of thing. He listens to this thing: a lot of theoretical discussion, great hand waving and elaborate quantum field theory discussions, and so on. Finally, Rabi says, "You know, fellows, I don't understand much of about this. But suppose the meson doesn't go into an S-state. It goes into a P-state, then hangs around in the P-state until the P-state decays into the S-state, and then everything you're talking about would happen. So that the lifetime would be the P-state to S-state lifetime, and not the nuclear physics lifetime at all." Everybody said, "Rabi, you don't know anything about this subject." But he gave some argument. It turned out that he gave the right explanation, the key to the mesonium business.
It's an interesting kind of thing. He used to do that. He was a very brilliant guy, but people would get impatient with him during elaborate theoretical discussions, because he is intuitive.
A brilliant experimentalist, but he wasn't doing much of that when I was there. Anyway back to Foley. I embarked on this problem with Foley, and did it. Then he said that since I had demonstrated I could do this kind of work. that we should talk about a real problem. One question involved the quantum theory of molecules. I wanted to see if I could do a relativistic quantum mechanics theory of a diatomic molecule. I kind of horsed around with that for a month or so. I don't even remember why I wanted to do it, by the way, but it seemed a good problem. It became clear that I didn't have an idea of what to do. I stumbled onto the fact that the whole hyperfine structure business had been elaborated quite completely in atomic theory, but nobody really had looked systematically at the effect of nuclear properties at all in molecules. There was good but unsystematically theoretical physics that Norman Ramsey had done. Norman Ramsey had done a couple of papers in which he had taken a particular case, and elaborated it.
Those were papers around. I decided that the problem that I wanted to undertake was to do a complete elaboration of the theory of hyperfine structure in diatomic molecules. For a lot of diatomic molecules, there isn't any. There is no net magnetic field, no molecular magnetic field, because they are symmetric. That decided to be a good idea, and I embarked on that. It really was a perfectly good, original problem given the starting place, which was the Pauli Equation, essentially, the Pauli approximation of the Dirac equation. You then beat the problem to death with matrix mechanics. You just take the whole thing apart, and you make a taxonomy, so that you don't lose anything. Then you find all the matrix elements. and do systematic classifications, not a very exciting problem. It took a long time.
It was a highly cited paper.
Well, there was a reason for that. A piece of straightforward theoretical work was needed. Just then Townes's laboratory began to do NO and 016-018 — some work with radicals. They were beginning to do the microwave structure of all these net magnetic moment molecules, and they wanted all the results. Essentially, it was a straightforward piece of theory that produced something that was of use in the experimental business. It didn't change theoretical physics in any way.
Right, but you were aware of this. Foley was aware of this.
Well, I wasn't really aware of it when I started. Henry may have been aware of it, But I was a little vague on it. It was only after a while that it became clear to me that there was a market for this. I have always been a very restless person. I would work for awhile, and then I would wander around the laboratories. I was always in and out of people's laboratories. Polycarp Kusch once said. "Hey Frosch, you spend so damn much time in this place, wouldn't you like to do an experiment on the side?"
But you weren't interested in the experiments. You were talking with the people.
I was talking with the people. No, but I was learning about the experiments. I wasn't interested in the experimental technique particularly. I was picking some of that up, but I wasn't interested in doing it. I was interested in what they were measuring and why, and how. I would wander in to the microwave labs and talk to some of those people. I knew roughly what was going on.
I know you finished your thesis at Hudson Labs during this time.
Well, I didn't really. I took my doctor's oral in the spring of '51, and I passed the oral, but one of the examiners wanted rewriting done. The starting place of my work was the Pauli Equation. I wanted to be systematic, so I started with the Dirac Equation; I wanted to have a little derivation in where my starting point was. I went through the literature and found that a way to get from the Dirac Equation to the Pauli Equation was in a paper by Foldy and Woutheysen, which used a transform method. When I wrote my dissertation, I cited that. One of the theoretical physicists didn't happen to like that approach and he objected strenuously to that piece of the thesis. So I had to go back and rewrite the first chapter in order to put it into his way of doing it, which was to manipulate commutator operators. I rewrote the chapter over the summer and into the fall. The degree was awarded the next commencement, which was January '52. I had essentially finished the work and turned in the final copy of the dissertation sometime during the summer. This was before I actually started at Hudson Labs, as I recall.
Well, the question automatically arises, how did you get to Hudson Laboratories?
Obviously, the next thing for me to do was get a job. What I wanted was an interim job. I guess I hadn't finished the dissertation. I was going to finish the dissertation in time for January. I was tired of rewriting.
Was it disheartening to have to go back and do something by a different method, just because somebody wanted it?
It was annoying. I did not give a very good defense. It was probably the worst speech I have ever given. I just froze up. It was perfectly all right, but I can do better than that, obviously. I don't quite know what happened, but I just got scared of it. It was rather tough.
Was Rabi at your defense?
Yes, but Rabi's a nice guy. You know, you knew who was difficult and who wasn't. It was just the situation, the formality of the situation. Come to think of it, there was an oral part to the qualifying exam. I don't remember much about it, except that Willis Lamb is a very nice guy. Somebody asked a question, and it must have been clear from the look on my face that I didn't have the faintest idea of what they meant. There was something about the phrasing of the question that faked me out. Willis muttered some phrase that instantly made clear what it meant. Anyway I wanted an interim job, assuming I would later look for a university job. I talked to John Dunning, because I was taking his first course in nuclear physics. That's what it was. He's an interesting guy, too. I had met him once or twice before.
One of his talents was that he never forgot a face or a name. I wanted him to take me on as an assistant for a period, but he didn't have any jobs. The only thing he knew of was that there was a new lab being put together by Eugene T. Booth, Professor of Physics at Columbia. They might have a place. In fact, a classmate of mine, Frank Pollock, had gone to work there. and so I talked with him and then to Bill Nierenberg. It was through Dunning and then Pollock that I made contact with Hudson Labs. Everybody was away in Bermuda doing their first experiment. The only one who was there was Bill Nierenberg, now director of the Scripps Institute of Oceanography.
This was all experimental work?
Well, no, it wasn't. They were looking for a theoretical physicist. In fact, Frank Pollock is a theoretical physicist. They were just looking for people, they were just starting. They had almost nobody.
Can we go back now and talk about your marriage? When was that?
I got married in June of '49. While in college, I was in a fraternity. I was in ZBT, Zeta Beta Tau. I met my wife through a fraternity brother, and fell in love very intensely, and we decided to get married. She was a Smithy. We got married while she was still at Smith and I was at Columbia. Nobody was very enthusiastic about it, because I was in graduate school and she was in college. Everybody decided that it could be managed and we got married. The interesting thing was that she was a physics major, probably would have been a better physicist than me. But in the course of our marriage she switched from physics to anthropology. She later got a doctorate in anthropology.
While you were married?
She actually completed it after we were divorced. And so I was commuting to Smith for about a year, not commuting, but going up when I could. We had an apartment in New York. Then she graduated and began anthropology at Columbia. We were married in June of 1949, and it would have been in the summer and fall of '49 when I was doing the apprenticeship problem. I want to make a funny point. Joan is a very intelligent woman. She had been a physics major, shifting into anthropology. I was struggling with this matrix mechanics problem. One of the things that I had missed was that the problem was highly degenerate. When I say I missed it, I mean I worked out the matrix mechanics and the transforms without noting the degeneracy. I had these great matrices, you see, and I was struggling and screaming and tearing my hair out in the apartment. And Joan came and looked over my shoulder and said. "All the numbers are the same; isn't it degenerate?" This was very nice. Then it turned out to be a few two-by-two matrices. All very straightforward, accounting for the degeneracy. The reason it was interesting was that there was a question of the breaking of the degeneracy by something else in one of the experiments. Anyway, so we were married in this period, a very loving, very fighting, probably over-emotional couple. It was a very stormy marriage, which broke up after eight years. I don't think she had gotten her doctorate before we were divorced. Anthropology departments are not like physics departments. Physics departments seem to hustle the kids through as fast as they can make it. Anthropology departments don't seem to do that.
Was she in social or physical anthropology?
Cultural anthropology, an ethnologist.
Did you have children?
We had no children, that was one of the issues. I wanted children. She didn't want any children. We were an early version of not knowing how to work out the two careers.
Okay. Well, you've gone to Hudson Labs now as a research physicist.
Eugene T. Booth was the first director of the laboratory. He was later head of the physics department at Stevens. He was a professor of physics at Columbia, and had been heavily involved in the gaseous diffusion project (to separate U-235). His deputy was Jack Nafe, John E. Nafe. Nafe had been a physicist at Columbia, and later was a geophysicist — still is, I think — at Lamont Geophysical Observatory. Gene and Jack had gone off to Bermuda to do their first experiment with the first device they had built, called the "Diving Duck."
Let me ask you, going back just a bit: when you talked with Frank Pollock, did he tell you what the basic purpose of Hudson Labs was?
He said it was a Navy thing, and had to do with undersea warfare; that's all he could tell me. Frankly, he didn't know much more, either, because he had just got hired on himself. It looked like an interesting thing to do for awhile. I went off to be interviewed by Bill Nierenberg. Bill and I have been friends ever since. In fact, that's how he got to be chairman of the NASA Advisory Council. He worked for Gene Booth on the gaseous diffusion project. He was then, I guess, at Michigan, as an associate professor of physics. He was an old molecular beamer. He had decided to spend the summer at Hudson, helping set up this organization. I would have to say that it was a typical interview with Bill Nierenberg: I talked only for a few minutes. Then said, "Let's take a walk. It's a nice afternoon." This must have been August or July. We went walking along the river in Dobbs Ferry. Bill talked at me for an hour, at the end of which he said, "Well, it sounds like you would do very well. Why don't you try this?" You know, I had only spent five, maybe ten minutes describing who and what I was.
He was actually talking to you.
Yes, which is normal for Bill.
You were going to talk about the origin of Hudson Labs.
There was a big Navy examination of antisubmarine warfare sometime around 1949-1950. The Navy got very uncomfortable about what had happened with the antisubmarine war in the Atlantic towards the end of the war and concluded they had a potential convoy protection problem. I don't recall whether I have ever actually seen the study. I certainly haven't seen it since 1951 or '52. It was a secret study. One of their conclusions was that they ought to set up a university laboratory that would do some basic work in the subject.
Is this the experience that OSRD had during the war?
It came very much from scientists in World War II, the Columbia group. and the Woods Hole group. Remember, this was the period when ONR was one of the new inventions, and everybody was bullish on that. ONR and what I guess was then the Bureau of Ships were to set up this laboratory. It was distinctly ONR in the lead with some BuShips money. Exactly how they settled on Columbia, I don't know. Columbia had a lab that did some undersea warfare, but my impression is that ONR wanted to get a great physics institution to oversee this. It is clear that the Navy had in mind that this would be heavily embedded in the physics department at Columbia, which it never was. I can only give you a hearsay anecdote about that. At the final meetings when they were signing on the details of the agreement an so on, Rabi was there, and Booth, and some others. The Admiral turned to Rabi and said, "Now, Professor Rabi, one thing I'm not clear on is how long you propose to be director of this laboratory." And Rabi looked at him and said, "Director, me? I don't propose to have anything to do with it." At any rate, they went ahead, with Gene Booth as the director, and it never was embedded in the physics department. Gene had been the head of the department after Rabi.
Was there a reason why it was at Dobbs Ferry?
There was a building that they could get their hands on. Nevis (The Nevis Cyclotron Laboratory of Columbia) had started up the line that way. It was clear it couldn't be on the campus. They wanted a place with a pier. We never used the pier very much, but the idea was that you could do some things in the river. There was a very good building. The site originally had been a brewery, but then a large building had been constructed in the late '20s or early '30s by the Methodist Book Company as their printing house. It had a very heavy floor-loading. It was a reinforced concrete building, a very solid, gigantic building that was available, and it was leased. I started there in the middle of September 1951, September 15th. I think. Bill Nierenberg had meanwhile gone back to the University of Michigan after the summer. There was Gene Booth, Jack Nafe. Frank Pollock, me, and there was Marion Johnson, the business manager, and his assistant Mike Fritz. There were one or two other people, but that was all. (Among others: Henry C. Beck (mechanical engineer) and Harry Sonnemann (electrical engineer)). We were going to build a research lab, a research institution.
I see. So you were going to fill it in then?
We were going to fill it in and we gradually did. I can't remember the details, but over the next few years we hired more people. Of course, immediately we began to ask what is it we are going to do? Here we are. What's the problem? What are we going to do? Gene Booth established the philosophy very clearly by saying that the problem is to find out what it is that Nature is doing in the ocean. What are the natural requirements and possibilities for underwater acoustics? How does the ocean affect it? And then we will see how to use it. The idea was to find out what the natural situation is. Let's first understand the physics of the thing, and then we will know what can be done.
I've got two questions. First, you had taken this on with the idea that this is going to be a temporary position.
Yes. I just never got around to doing anything else.
So you were basically happy there.
Yes, yes; it was interesting. I had never done anything like it before.
And second: Did the Navy simply dump money on you and then leave you alone? Were there particular directives?
No, those were the days in which things were done in the old ONR way. They wrote a contract with Columbia University with a rather general task statement, and put Gene Booth in charge. They agreed on the amount of money for the year. We went off and did what we thought was the right thing to do and kept them informed.
We wrote a proposal for the next year, and we sent them reports. We would go to Washington and talk to people and so on. I can get more into the flavor of that, but it was not a question of writing a proposal to finance everything you wanted to do. In fact, all the time I was there we never had more than two contract tasks. Most of the time I was there, you had the one task to do, and you had the annual money and you lived on it, and you did your work. You justified yourself by what it was you had done in the course of the year and what was being developed. This is entirely different from how current proposals are done.
You justified yourself to Booth, not to the Navy?
Booth had to justify us to the Navy. The director justified us to the Navy, but we were working as a group, essentially. Here is an interesting description. When I went to work on the 15th or the 16th, there was a meeting. Everybody just got together in a room and started talking about what we were going to do. They had chosen a particular project just to sort of get everybody's feet wet. This was the "Diving Duck." which was a way of measuring ambient noise. It was clear to a physicist what the problems were. You had to find out how to make sounds and how they moved around in the ocean. You had to find out what the background noise was that you worked against; all of these were classical problems.
How the topography of the ocean affected them?
Yes, how the oceanography, the environment, the bottom photography and so on.
Is this what you would classify as marine physics?
Yes, it's a variety of marine acoustics; underwater sound is what we actually called it. There was this "Diving Duck" project, and then the question of propagation, and so on. I went into this first conference and had a very interesting time. The next day I had an idea that had come to me in the meantime and I wanted to check something. So I asked if there were any minutes, if anybody kept notes. "Oh yes, there are formal minutes." "Well, can I see them?" "No, they are classified, and you haven't got your clearance yet."
So you couldn't see them?
It hadn't occurred to anybody during the meeting that it was going to end up classified, and I didn't have a clearance. Anyway, after a while I got a clearance. The first problem we had was the propagation problem, and the principal document was Memoir 27 of the Geological Society of America, which contained, among other things, the Ewing-Worzel Paper on the Deep Sound Channel. Morris Ewing and Joseph Worzel. Memoir 27 had two important articles in it. I think it was the same memoir, but it may have been another one. That memoir had the Ewing-Worzel one on the deep sound channel and the ray tracing. Also, in the same one or or a different one was an article by Chaim Pekeris on modal theory in layered media.
Is the publication you are referring to "Preliminary Ray Computation of the Ocean South of Sable Island"?
No. but, I'm coming to that. That was one early thing that I had done. So the first thing I started to do was take on the problem of numerical ray tracing.
Does this mean computers?
Well, it meant computation. I'll come to the first computer in a minute. What it really amounted to was systematically taking Snell's law and figuring out how to make a reasonable model of the ocean that you could trace rays through. Then you had to invent some means for computing intensity from rays, and understanding how accurate the ray theory was, and how were you going to deal with the caustics and shadow zones and all of that. Basically, it was pretty primitive numerical stuff. We were breaking the ocean into horizontal layers, because it's pretty well temperature-stratified. The ocean is a stack of homogeneous layers, and so you take a straight line, and then refract it at the next boundary, and just keep doing that, until you can't punch the Frieden any more. That's in fact what we had. We had the old Friedens and the Marchants, the clickety-clack machines.
These certainly were electrics?
Yes, they were electrics. I did a certain amount of that, and learned the game. Then it was clear we were going to have to do a hell of a lot of this.
Did you do your own calculating at first?
Well, yes. There wasn't anybody else to do it. I just sat and did it, not by the Sable Island thing. When we understood how to do that, it was clear we were going to have a hell of a lot of that to do, and so we agreed we had to have some sort of computation team. There was no such thing as a computer at the time. This was a little previous to that. The only thing around were card program calculators and we weren't up to that yet. We bought or rented several years later. No one was doing much of that kind of computation on the CPC. And so we did what everybody was doing. You hired some people to run the calculating machines. We put an ad out — I don't know how the ad was phrased — but a bunch of young ladies answered the ad. And so my first computer consisted of eight girls in a room. They were high school graduates. A couple of them were married women who wanted a part-time job. In fact, I had the problem of setting out the computational system. Literally, my first computer was eight girls with Friedens. The question is, how do you systematically arrange the algorithms and the flow of work to get as many rays traced as accurately as possible? We set up a system. I gave lectures on Snell's law and how to do the computation, and set up flow sheets, primitive versions of all that stuff.
Did you consult with anyone on the flow technique? I know a lot of that was done during the war in the ballistic agencies.
Nobody was around that seemed to know much more about it than I did. We all knew the numerical computation of integrals. We had all done the various interpolation schemes and we were using that. It was pretty straightforward, only one formula, just Snell's law. It was a question of chaining it and getting accuracy. Now, about the Sable Island thing. I was simply asked to find some way to do this problem. The problem was that this was going to be a place that might be used for a listening station. The ocean is not horizontally homogenous going out from there, because there are parts of the Gulf Stream and of Artic water, so that the approximation of a lot of horizontal layers doesn't work. You have to find some way of tracing your way through an ocean that is inhomogeneous in two dimensions. That was my problem. I tackled it in a very simple way. There were two obvious ways to do this, and they were different, but not very different. If the answer was reasonably the same by both methods it was probably correct. One way to find out is to do it both ways. We had two models and I computed rays in both. My first approximation was to use a set of horizontal layers, followed by a transition to a new set of horizontal layers. Then there's another set of layers and another transition, and I just stepped my way through that. I had to teach the girls how to do ray tracing, through one set of layers and then when the case came to boundary, trace through the boundary and zone. The second way was to make the approximation that sometimes the layers aren't horizontal, but tilted, and to use an angular transformation and trace it that way. We did both of those. From a theoretical, computational stability stand point, you could pretty well overlay both ray tracings and conclude you've got the same result. With that agreement we concluded the result was a reasonable approximation to a more correct solution. At that point, I played with trying to get a more analytical computation of the thing. But it was clear that was a long theoretical job, and we wanted a result. It was as simple as that. It might be worth going back and working the analysis problem in general, but we really wanted to have a result in a few months, and so this was a way of doing the computation.
Why was there a time problem on this?
We were trying to solve theoretical problems, and meanwhile, Bell Labs and the Navy were trying to build systems that needed the theoretical problems. So this was a case where our work had already become useful to somebody, faster than we had thought. To make more elaborate ray tracing predictions than had been made before was interesting, but it also fit into some things that the Navy wanted to do.
How much contact did you have with your users, Bell Labs and the Navy?
A lot, in the sense that from the very beginning we found ourselves embedded in a community of laboratories and people who were working on the same class of problems. There was a group at Bell Labs that had system building responsibilities. They were also doing ray tracings. We had some arguments about physics with them.
Well, about the physics of the ocean. There were people at Woods Hole who were involved, and we would visit people at ONR. So we were embedded immediately in a community of people who were doing classified work on the same thing. All of this was secret then.
But the communication was good.
Yes, you just went and talked to people.
It wasn't like the OSRD years, where there was duplication, and people didn't know what others were doing, even if they were working on the same thing.
No, we were encouraged to travel around and visit people, and so on. There was duplication, but I would say it was purposeful. I think the Navy wanted a counterweight group to Bell Labs. They didn't want to completely depend on the Bell Lab's point of view.
Hudson was a counterweight group in that regard.
Yes, I think they viewed us as being another force in this community. There were Navy labs involved, too: USN Underwater Sound Laboratory and others.
Sure. I asked you about the Korean War before. Was there any particular need or rush for this kind of technology because of the war?
No. I don't think anybody was terribly worried about that. They were really worried about the convoying problem only in the sense that if the war expanded, they would begin to worry about the Pacific. We never got involved directly in any Korean War ventures.
There was still a sense of worry over convoys in the fifties in the Atlantic Ocean?
Yes, the Atlantic and the Pacific, but we were worrying about the Atlantic.
And of course, this was because of Russia and the beginning of the Cold War.
Yes. My sense of the time was that they were worrying about the last war, and therefore worrying about that problem for the next one.
This was general preparedness.
This was general preparedness. It was a problem that they didn't feel that they had solved. It was clear the Soviets were going to build submarines. So could they solve the problems vis-a-vis the Soviets?
As you were working underwater sound, the Hudson Labs was building.
You had good contact in the community of classified workers. What did the future look like for Hudson, and for your career in the early fifties? Your research was continuing. This was all applied research. Did that ever enter your mind?
Yes, but not very much, I must say. I was reading other physics and so on. What I was doing seemed interesting and useful and exciting. I was having a good time. I don't know that I thought very much at that point of going elsewhere and doing something else. It may have been just sheer inertia. Whatever it was, my work was interesting. People seemed to value it. The fact that it wasn't cosmology bothered me not very much, and it seemed to go on all right.
In that regard. then, Hudson Labs was expanding.
Oh yes, it was continually expanding.
And you began pretty much on top, in a way.
Yes, there isn't any bottom when there are four of you. It was clear also that I was doing reasonably well. You know, simple as that Sable Island computation is, it was some kind of a landmark at the time, because nobody had figured out how to do that before. And even through I did it by a very crude set of methods, it still brought a new practical capability in. We began to have some very strong ideas of our own. You have to understand some of the views of the ocean at the time in order to understand what it was we were saying. The first point was that we were at low frequencies; we were dealing down in the few hundreds, below a kilohertz. The ocean was generally viewed, because of the experiences at high frequency, at ten, twenty, fifty kilohertz, as being a random medium.
Even with the layers, there was enough variation, enough happening, to regard everything as not terribly coherent. It was not coherent enough to build an array of great length. There was a strong feeling that the length of a coherent array was a strictly limited item, a few hundred feet. That was it. You could not get a big coherent aperture, you could not expect great accuracy in measurements of time or frequency or whatever. This was a very tough business. We developed the idea that we could not see why this was the case. It seemed to us that from what we could tell of the oceanography, these frequencies and wavelengths ought to be predictable. You ought to be able to trace arrays (and of course from rays you can't get phase), but you ought to be able to predict the sound intensities. in fact, when you deal with the sounds, you ought to be able to make very large antennas (arrays), if you wanted to. There was another set of ideas that had to do with the noise characteristics.
The ocean is, to a first approximation, organized in horizontal layers. The question was, where does the noise come from? If the noise is organized it comes from places that are organized differently in the vertical plane than the ray tracings of the sources that you want to hear from. If you put in some vertical antenna capabilities, then you can filter signal from noise, maybe with some significant signal to noise ratio advantage. We were arguing that you could go to larger horizontal arrays, and it might pay to put in vertical directivity. You could get much higher antenna factors than anybody was talking about. I'm going to be hampered in this discussion, because I don't know even now what some of the classification details are, so I may be heavy on theory, and light on results.
I appreciate that. When you say you were arguing for this, were you trying to support funding by the Navy for these antenna arrays?
No. We were trying to say that there are possibilities and experiments and equipment that ought to be pursued, which current systems aren't exploiting. Our task was to prove it. We embarked on a whole pack of experiments dealing with coherence and noise characteristics. The array theory was the principal product for a long period of time. A whole classified literature developed — I don't even know where it is. It can probably be found at NRL, or some place like that.
Yes. The interesting thing is that we did manage to find a number of your early citations that are all secret and confidential, but they are still cited.
Where are they cited?
Well, they're cited in a good number of different types of journals that may, themselves, be techreports. But they are still searched in the science citation studies, because the abstracts are available.
That's interesting. There was an entire classified literature, that is, not only report literature. There were one or two classified journals, or journals that had classified annexes. And there was the equivalent of a classified underwater sound society, and a symposium, and a bunch of committees and people who met and worked. I was heavily involved in that community. Later I became a leader in the community.
You were involved in building up Hudson Labs. When did it come time to start hiring Ph.D. level assistants for your own division. You became assistant director in '53?
Well, somewhere in '53. 1 was told I was the theoretical division, or assistant director for theory. I think it was Frank Pollock and me. And we hired a couple of other people in that group. By then, there were enough people around, so I suppose that we must have had a dozen physicists with doctorates, mostly Columbia people. The place was gradually building up.
Were you living in the Dobbs Ferry area?
No, I was living in New York, and anti-commuting. Through this period to '55, I was living in upper Manhattan. And now, somewhere in there, I got mixed up in experimental work. The first thing that happened was that after the ray tracing work, we decided we had better do something about the shallow water areas. Clearly that was a different theoretical area. Pekaris had developed this model theory initially to make computations for prediction. Here again, the question was, could you in fact predict the propagation, or was it all much too complicated geologically in terms of water properties, so that you couldn't fit that kind of model? And again, our assertion was that you could in fact fit that kind of model, but it might be terribly complicated. We wanted to know if we could make some measurements that would show the right behavior. I ended up heavily involved in designing those experiments, and then doing them. One of them involved a very good sound source. We went out on Long Island Sound and hired a self-propelled lighter, a barge with a crane on it. It's engaged in lighterage, namely, taking cargo off ships that are not next to piers and bringing it to the piers. That's called a lighter. We hung the sound source off the side of the lighter and used another boat, the T boat, to move the hydroplanes. We went out and measured the sound propagation in various directions, and made some explosive measurements to get the layering at the bottom. Then we went back and compared it with the theory, and decided, yes, believe it or not, a layered medium behaves like a layered medium.
The sound source stayed in one place?
Yes, you moved the hydrophones. Hydrophones are easier to move than sound sources. We did that. Meanwhile, we put in a couple of hydrophones off an abandoned Coast Guard Station, on Fire Island, and we did some offshore work there using the T-boat. Sometime in this period, the Navy had funded us to put a hydrophone off San Juan, Puerto Rico in the southern side of the Puerto Rico Trench. We did a lot of sound propagation work down there.
You have, in 1953, Frosch, Guthrie, Lowell and Poss, "Preliminary Report on Sound Transmission at San Juan."
Oh yes, I'd forgotten about a lot of those.
That answers how you got down to San Juan.
Yes, we had a field station down there. Being physicists, we were looking for situations where you could get a simple enough geometry, or a simple enough characteristic, so that you could decide whether the theoretical tools you had fitted the data. And then you would work up to more complicated situations. Now, the Puerto Rico Trench has the nice property of being deep enough so that, if you stay inside the trench, it's a semi-infinite medium. You can forget about the bottom.
It's not so wide.
It's not so wide, but you can run down the length. There you could examine how do you deal with bottom bounce, and the conversion from the the bounces into other rays. We began to work these elaborate problems where you not only had a layered median, but had a slope, and when you came to the slope, you bounced. You could see roughly what the theory was like, but we wanted to demonstrate that in fact this was what was happening, so that it could be used in another context. We were working all those problems across the slope and along the slope. We were short enough in people, particularly senior people, that I was taking on some of that, running cruises and so forth. You know, it was a small enough place so that everybody did everything. You didn't get pigeon-holed.
You were doing engineering problems, too.
We didn't have equipment, so I began to worry about what kind of equipment I could use. Again, everybody was consulting everybody on everything, because it was a small group and you'd work the same problems together. Then we began to have to worry about logistics of cruises, and so on.
Did you become a group that was of such size and complexity that there was a concern for the logistics of doing the best science? Were you beginning to concern yourself with establishing a systematic approach to the problems, given the group that you had at hand?
Yes, at this point, we had a mechanical engineering department with a chief mechanical engineer, and an electronics department with a chief electronics engineer. We had engineers building and designing equipment. We were using long cables and heavy equipment, and we had rigging problems, and boatswain mates' problems, and electronics problems. It had begun to get differentiated in that sense, and was beginning to be a management and a planning problem. There were no experts, fortunately. We were able to design the thing in terms of what we were really doing.
You say fortunately?
Yes, yes. We were very lucky in our business manager. He was very good, and did not believe in running the place from the business office. I remember going to Marion once when I was director, and saying, "You know, I've been talking to people about how one ought to run a place like this, and they seem to think we ought to have a lot of project cost accounting, so that we know exactly what individual things cost." And Marion said, "Well, if you really want to do something like that, I'll set it up for you, but I don't think it's a very good idea. What is it you want to know the cost of and I'll tell you."
That's marvelous, in this day and age.
If we really wanted to know what a cruise cost, we could find out what a cruise cost. But there wasn't any sense in setting up that kind of accounting system for the purpose, because that wasn't the problem that we were faced with.
Was ONR always meeting your budget requests?
Oh, we'd go and haggle some, but when the year began we knew roughly how much money we had. We knew the accounting system. We knew how much money we were spending on people, how much we were spending on rent. We didn't have a ship budget because the Navy was providing the ships, but we could estimate what that cost, and we knew roughly what the out-of-pocket costs were to do a particular experiment or cruise. It didn't seem necessary to us to know any more than that. Indeed, there was no particular reason why we should have been able to say that it cost us thus and so to do that experiment. What would it have told us? So there was no great compulsion to have all that. I'll tell you a funny story about that. I'll tell you about the day I heard about PERT, you know, program evaluation, the great management system, the PERT diagram business. It was the one that was supposed to have been used to do the Polaris program. In fact, it wasn't. You ought to read Harvey Sapolsky's book on it. It was one of the great new management inventions. I was chairing one of our periodic meetings at Hudson where we were planning a particular experimental cruise. We went around the table, and I was probably arguing with Hank, who was our chief mechanical engineer, or one of the engineers. I wanted to know if we would get to sea on time. I was going systematically through everything that needed to be ready to do the experiment and the analysis, and so on. I got interrupted by a visitor from Washington. It was very important that he explain to me this new management technique.
Right in the middle of your meeting?
Right in the middle. So he explains this new management technique, which is in fact this business of drawing critical path diagrams and so on. And I just looked at him, and I said, "What do you mean, new management? What do you think we are doing in there?" You know, this business of new management techniques is, in a large measure, a lot of crap. People who have never dealt with that kind of thing suddenly come upon it, and invent some scheme they think is great. I am death on modern management because it consists of a lot of inventions that have done harm, mostly because the idea comes from the outside, from people who haven't been involved.
This man came from Washington and it was important enough to take you out of a meeting. Did you feel obligated to follow their guidelines for funding or anything else?
No. This was just a suggestion. Nobody was obligated. Nobody was arguing that we were doing good or bad. This was just my comment that somebody thought he had made a great invention. But to us it wasn't an invention at all. The only way to plan a cruise was to worry about whether everything would get to the ship on time, and to inquire: what is it that you are most worried about getting there on time? What is it that's not getting finished?
You simply wanted to make sure that your own division chiefs knew what they were doing, and knew what was needed.
Yes, know what was needed. We would systematically go through it. That kind of planning is usually described as an exercise in being systematic, but it isn't. It is really an exercise in imagination. There is no problem in being systematic, provided somebody's thought of everything. The problem is making sure that you have thought of everything. That's an exercise in imagination. You have to crawl through the experiment from today, all the way through to the end.
When did you first consciously think of these things in these terms? When you were director of Hudson Labs, certainly.
By then I was thinking in these terms.
When was that? When did you become director?
Oh. I became director in 1956. It's worth talking a little bit about the sequence of people. Gene Booth was director for two years.
Just two years?
Yes. And then Bill Nierenberg came and was director for a year. Gene went at that point back to the physics department at Columbia, and then after a while to Stevens. Bill came to be director, and he was director for a year, from 1953-154. Sometime by the end of the summer of '54, Bill went to Berkeley. He wanted me to become director then. There was great discussion whether that was a good idea.
Well, Bill and the Navy and Columbia, and so on.
Because of what?
Because I was too young. Bill roundly asserted that I could perfectly well be director. I was not his deputy. He had had an assistant director, Frank Levin, who was then with Jersey Standard Production. I think he is now with EXXON in research. He wanted to go back to Jersey Standard; he was on a year leave of absence.
As a scientist?
Oh, yes; he's a physicist, a geophysicist. I was doing a lot of work around the place, running experiments and cruises and going to sea, and so on. They finally concluded they couldn't do that, that that was not right. So Al Guthrie became the director. I guess I was his assistant director.
You did become assistant.
From '54, or somewhere in there.
You had always been a sort of assistant director.
I had been assistant director for theoretical physics. Now, it was assistant director of the laboratory. Remember, this was a very informal place. I mean, I was Al Guthrie's assistant. We each had our own office. They were on the fifth floor which was kind of a penthouse, just a couple of little offices. It was nice. It had a view over the river. Because it was an industrial building with glass windows, it was hot in the summer and cold in the winter. I did not then, and do not now, think of myself as an administrative type. I viewed administration as something that the business manager did. The business manager worried about administration, and the director and the assistant director worried about the science and the engineering. This included deciding who to hire, salary structure and so on. I didn't worry about the detailed stuff that Marion Johnson worried about. We didn't worry about whether the accounting was straight, or in fact, how the accounting should be done.
If Marion had come and said that we were going to have project accounting, we would have said, fine, Marion, set it up; we'll have project accounting. And if Marion said that it doesn't make any sense to have project accounting, it would be fine with us. We didn't have any trouble. I'll tell you two anecdotes. When we were being set up, the late George Pegram, who had been dean of the graduate faculties at Columbia, was still involved in this project. Pegram came up to visit us within a year after the place was set up. He was still active up to then. We just kind of wandered around the place and wandered through the stockroom. He said idly, "Do you have any trouble with losses from the stockroom?" There were three of us following Pegram around. Everybody turned to Marion Johnson, and Marion said, "Oh, not particularly. As far as I can tell, we've issued twice as many electricians' knives as people who have ever worked here, but you know, minor things like that." And Pegram said, "That's all right, I wouldn't worry about that much." He said, "Once when I was setting up Manhattan Project, a guy from the General Accounting Office came and said: 'Dean Pegram, do you know that you lost $40,000 worth of small tools last year?' You know what I said to him? I looked him in the eyes, and I said, 'And what did it cost you to find that out?'" That was kind of the flavor of the place. I remember Bill Nierenberg once opening the security safe in his office, taking out a piece of paper, and saying, "I want everybody to see this. It's the last you'll ever see it. This is the official organizational diagram of this laboratory. The Navy says there has to be one." Then he put it back. Now, after awhile we had organization, gradually, after there were certain numbers of people to organize. You have to have an engineering department with technicians, you have to have supervisors. There were enough physicists around that some people were leading others, particularly as we hired some junior people. Nobody objected to doing that. It's just that nobody was terribly interested in making a big fetish out of it. Nobody was very committed to the idea of management.
How often did those meetings take place?
We'd have weekly staff meetings. We'd have cruise meetings pretty regularly. You'd have a planning meeting when you'd outline what you were going to do, and where, and when. You'd parcel out responsibilities, which were very easy to parcel out, because they were obvious. It was somebody's experiment, so the scientist in charge and his people had their assignments. The various engineering departments had their jobs to do, too. They would all get to work on it. In a month it would be time for another meeting, which would get more intense as cruise loading approached.
Did you have the system where you would actually have one of your division people make a prediction as to how their progress would be, and then match the performance to the prediction?
Not in a very formal way. We'd ask, for example, when will the cable be in-house to start splicing? Somebody would write a note that's going to be on the 23rd. Around the 23rd, somebody would remember to call up, or Hank or one of the guys would call up and say, "They are not going to deliver the cable until next week. But I think we invented a way to do the splices twice as fast, because Eddie went and worked the production line scheme." That kind of thing.
Fair enough. That happens all the time.
There was a lot of that going on. Maybe Hank Beck had charts, but it was never systematic. The most we had was a large, stretched-out calendar where we figured cruise times. The problem we had more trouble with was getting the ship scheduled with the Navy's administrator for requirements, and making sure that people's schedules and vacations and all the rest fit together. I don't remember spending great agonizing periods over that. Somebody else worried about that.
Were you still working as a research physicist when you were assistant director to the laboratory?
Yes, sure. I was spending more of my time working on experiments and problems. When I was director and Al Berman was deputy director, we conducted a major experiment in which we were the chief scientists.
Yes, that's what I want to identify. So you never made a conscious decision at that time to be an administrator, as you indicated before.
I made a very conscious decision that I was not going to be an administrator, you see.
That's what fascinates me.
I would say I recognized that, on the scale of the experimental and theoretical work we were doing, you couldn't do it by yourself. The instant you have to work with a ship, you're not a guy sitting in a bench in a private laboratory any more. And so I regarded it not as going away from physics, but as doing science on a scale where my tool was not my computer, my experiment or my machine. It was all that stuff. My tool, this laboratory, these people, these ships.
Yes, so that was the system you were dealing with.
The experimental system was the ships and the hydrophones and the divers, and the guys, and all the rest of it.
There's one paper that you did with Berman, Clay and Sherry in 1957, "A Preliminary Report on the Correlation of CW signals at Large Hydrophone Separations." Was this related to the Project Artemis at the time?
Yes, fairly directly. We'll have to get back to that.
We're at the point where I think we should finish, because of your schedule. You have to leave. So, on the tape for the record, for next time, we want to continue on at this point in the mid-'50's where you were picking up Project Artemis.
I'll talk a little about the coherence experiment.
I'd like to know how computers were brought into the labs, and what role you had in actual machines in, other than people.
Thank you very much for this first session.
Phys. Rev., 88, pp. 1337 ff. (1952).
Hideki Yukawa, Nobel Prize, 1949.
Polykarp Kusch, Nobel Prize, 1955.
I. I. Rabi, Nobel Prize, 1947.
Julian Schwinger, Later Professor of Physics at Harvard.
Willis H. Lamb, Nobel Prize, 1955.
A tabletop mechanical calculator.
Refer to R. Frosch working file, SS & E Department, National Air and Space Museum, Smithsonian Institution, Washington, D.C.
Tech. Report AD21-402, October 1953, TR12.
Harvey Sapolsky, The Polaris System Development: Bureaucratic and Programmatic Success in Government (Cambridge: Harvard University Press, 1972).
A. Berman, C. S. Clay, and H. B. Sherry, April 25, 1957; 31 p. Secret. AD3129672.