Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Carl Romney by Kai-Henrik Barth on 1998 January 20, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/22786-1
For multiple citations, "AIP" is the preferred abbreviation for the location.
This interview focuses on Carl Romney’s involvement in the technical and political debates related to seismic detection of underground nuclear explosions during the 1950s and 1960s. Trained in seismology at Berkeley (Thesis Advisor: Perry Byerly). Romney began in 1949 to work for Beers and Heroy, an Air Force contractor, which developed methods for the detection of Soviet nuclear explosions. There he became involved in the Department of Defense’s early efforts for the seismic detection of underground explosions. As the DOD’s leading seismic detection specialist Romney played a central role in a number of technical conferences involving U.S. and Soviet scientists. These conferences, which aimed at defined the state of the arts in seismic detection techniques, had direct consequences for the diplomatic negotiations for a nuclear test ban. Romney describes in detail the technical and political controversies of the Geneva Conference of Experts (1958), the Technical Working Group 2 (1959), and the Seismic Research Program Advisory Group (1960). During the 1960s Romney became a key figure in the development of new seismic detection systems for the DOD’s Project Vela Uniform, which aimed at the improvement of U.S. seismic detection.
Tuesday, January the 20th of 1998. I am sitting here in the office of Dr. Carl Romney in Arlington. It's about 2:05 in the afternoon. Dr. Romney, I would like to ask you to give some background how you became interested in seismology.
Well, as happens to many people, I sort of followed a path of opportunity. As a graduate student in physics at Berkeley I was not really very interested in [electrical] discharge through gases, which was what was proposed to be my line of endeavor, and after describing the kind of applied, more applied, physics I was interested in, it was recommended that I go investigate the seismology group under Professor Byerly. Which I did, and shortly thereafter had a research assistantship, which was also important to me at the time, and entered the department and found that I greatly enjoyed the work and I enjoyed working with Professor Byerly. So it's not that I was totally ignorant of geophysics. I have to give some credit for my interest to Beno Gutenberg, who had given me my first introduction to seismology actually when I was a student at Caltech. Beno Gutenberg taught a course in the physics of the upper atmosphere and much of that he described as seismology turned upside down because one of the chief tools at that time for investigating the upper atmosphere was propagation of sound through the stratosphere. And he invited several of us to come to the geophysical laboratory at Caltech and gave us the grand tour and demonstration. So I was impressed by that also. But basically it was, as I said, I followed the opportunity at the time.
Which years were you studying? Romney: I started as a graduate student at the University of California at Berkeley in 1946 and I think my first contact with Byerly was probably at the end of the first semester, so let's say January 1947 roughly. And my contact with Beno Gutenberg would have been in 1944 or '45.
And what kind of classes were you taking at this time in Berkeley?
In Berkeley I began in the physics department with rather expected sort of things for a graduate student. I took atomic physics, and my instructor there was [Emilio] Segre, a Nobel Prize winner, and classical analysis, advanced mathematics, whatever it's called, and advanced mechanics also, classical mechanics. Those are the things I remember. When I transferred to geophysics I began to take a course which we called "Great Earthquakes I Have Known" which was a kind of descriptive seismology under Professor Byerly, and participated in the graduate seminars and also took some geology, and I can't remember just which geology courses I took. That was the beginning.
In the end you took your Ph.D. with Perry Byerly.
Can you tell us a little bit about the process between starting with him and working on your Ph.D. and finishing your Ph.D.?
Yeah. Byerly advised me not to bother with a Master’s degree. He said that as long as you're interested in a doctorate; let's just go straight for that. So that's what I did. Along about 1949 I was beginning to run out of money. I was married and with a young daughter by then, and the money I had saved from my several years, well, basically three years in the Navy, and the GI Bill of Rights backed up the money I got as a research assistant. But as that money, my savings and the GI Bill ran out, I was beginning to get rather short of money. And just at that time Professor Byerly had participated in the review of the early work of the firm of Beers & Heroy. Byerly was a member of the panel on seismology, which is part of the Research and Development Board, and he became aware of the fact that a group back in Troy, New York, were working on investigating seismic detection methods but had no seismologists at all on their staff. And so the two, my running out of money plus Byerly's recognition that Beers & Heroy needed a seismologist to make the kind of progress they wanted to make in nuclear test detection, the country wanted to make in nuclear test detection, he saw that these forces were moving in the same direction, so he suggested I go work for Beers & Heroy for a while and accumulate some money so I could come back and finish my graduate work. And that is the course that followed. In September 1949, my wife and daughter and I moved to upstate New York near Troy, New York, and I began to work with the firm of Beers & Heroy on the more or less fundamental aspects of the generation, propagation and detection of seismic waves from bombs.
Can you tell a little bit more specifics about the work at Beers & Heroy?
There were two or three things going on. One was, and the part that I was most involved in, was essentially reviewing the state of the art of seismology as it applied to the detection of nuclear explosions. Much of it was paper studies. A second thing that was going is that Beers & Heroy were testing a number of instruments that had been recently invented and proposed for nuclear test detection by, chiefly by the Navy as I recall. There was the what we called the DTMB [David Taylor Model Basin seismometer], and an instrument called the Stanley seismometer [built under contract by the Stanley Aviation corporation for the Department of Terrestrial Magnetism, Carnegie Institution of Washington], and these had been proposed for the long range detection of nuclear explosions by various advisory boards to the government. Beers & Heroy was testing those instruments, had a number of them in operation in the field in small buildings that were constructed to house the seismometers. And the third thing that was going on was a more general look at instrumentation itself, what really should be done for nuclear test detection. So my job was not only to survey the state of the art from a more or less literature standpoint, but also to take the data recorded by the field stations and do analyses on that, on the recordings, for the purpose of helping to evaluate those seismometers. The long and the short of that was that both the instruments we were testing proved to be unsatisfactory. They were quite unstable and in fact so unstable that when the great Assam earthquake of 1950 I believe it was occurred, all of the operators at all of the stations simultaneously turned their instrument off, thinking that it was nothing but instability that they were used to, so they were quite unsatisfactory.
Were you familiar at this point already with Benioff's instrument? Was it used?
Oh yes, yes. Yes, I had analyzed data from Benioff seismometers in my position as a research assistant at the University of California, and I knew about their potential for recording P-waves at great distances, teleseismic P-waves. And I was also aware from the literature that these Benioff seismometers had been successful at detecting the Trinity explosion out to around a thousand kilometers and the Bikini-Baker explosion, underwater, out to distances of around seven or eight thousand kilometers.
Did the Berkeley network have Benioffs at this point, or?
They did. Not at all stations. They had them at Berkeley itself and at Mt. Hamilton, at several stations. I can't any longer remember exactly which ones. But typically the setup at Berkeley was two components, north-south and east-west, horizontal components, Wood-Anderson seismometers and a Benioff vertical seismometer for detecting the P-waves. So we could measure the local magnitude with Wood-Andersons and record P-waves. The body wave magnitude scale had not really been invented yet.
This came only in the later '50s.
Well, I can't recall exactly when the first hint of it came. It may have come in, I believe there was a paper, Gutenberg and Richter, in about 1948, and there possibly was a hint of it there, or maybe even the first proposal. I've kind of forgotten that, but it didn't come to be really important. It was not proposed to be the main way of measuring magnitude until 1956.
Until the paper in Annali di Geofisica [B. Gutenberg and C.F. Richter, “Magnitude and Energy of Earthquakes,” Annali di Geofisica, vol. 9 (1956), 1-15]
Yeah, that one, yes.
The '56 paper by Gutenberg.
This is still the time before your dissertation we just talked about, before your dissertation was done at Beers & Heroy or …?
Yes, I went back after, well, after two years, a little more than two years at Beers & Heroy in Troy, we had selected the Benioff to be used and had gotten the detectors, including seismometer arrays of four elements, deployed at several places, three places in Wyoming and one place in Alaska, and were successful at recording the nuclear tests that occurred in the spring of 1951, Operation GREENHOUSE. And GREENHOUSE-GEORGE, which was the 225 kt explosion, took place during that time, and it was well recorded seismically. I believe that DOG was also recorded, and it had a yield of 81 kilotons. There were instruments that were installed either for those tests or shortly thereafter at a number of other locations around the world too, but at any rate the success of the instruments and the arrays that we had developed led to a formal incorporation of these seismic stations into the U.S. Atomic Energy Detection System. They were installed and some of them operated by contractors, but mainly Beers & Heroy. And some of them were operated by Air Force teams for the Greenhouse series and they immediately became operational stations thereafter and formed the initial nucleus of the seismic system of the Atomic Energy Detection System.
So only with GREENHOUSE does the seismic component play a major role in the use of …?
That's when they — prior to that they were considered experimental.
After that they became operational stations of the Atomic Energy Detection System. So there was like an initiation or a baptism or some acceptance into the family at any rate. And following that it became, following the Greenhouse experiment in the spring of '51, it became apparent that to go operational we had to have a place to analyze the data. And we proposed, I proposed, and others, that we establish an analysis center in Laramie, Wyoming, which we did in the early fall, late summer really, of 1951. Laramie was selected because of its proximity to the three Wyoming stations, and the operational concept was that the film recordings — which was the state of the art at the time — the film recordings from those three stations would be rushed to Laramie where they could be jointly analyzed. And with the three stations, which we called a tripartite station, or a tripartite network, with the three stations we could combine the data and get a rough direction of arrival, calculating the wave front direction as it crossed the tripartite array. And from the speed of which it crossed the array we could get an estimate of the distance. So that gave us an initial epicenter for each seismic event, at least each detected at all three stations. We then could use this information as the input to locating the event using data that came to us from other places around the world: Turkey; Camp King, Germany was one place. And this other data was telegraphic data. The way the Air Force operated at that time, each team as they called them at each station would record the data for 8 hours, then change records, analyze the records, and send a telegraphic message to the Laramie Analysis Center. But just times of arrivals and amplitudes, and the analysts I guess, in some cases, as to what the phase was. It was very difficult to put that kind of information together without having a place to start the process, and the place to start came from the tripartite. And so that was quite successful I think and allowed us to locate many of the events that were recorded. I should say here that a good seismic station detects around 30 teleseismic events a day, so quite a lot of data came in when you had quite a number of signals were reported [sic], and in many cases it is possible to incorrectly associate signals recorded by stations thousands of kilometers apart. But by having a starting place and we say, oh yes, these now, these signals from six stations, let's say, seem to go together and are consistent with this initial epicenter. So that would explain a good many of the signals, and then we'd have to struggle just with alphanumeric data on the others, which was more complicated but sometimes successful. And we operated in that mode for several years. Our first success was the JOE-4 [the fourth Soviet nuclear explosion] event in August of 1951, uh, '53, August of 1953.
And we, the initial information about JOE-4 came to us from acoustic information that had been recorded by the Atomic Energy Detection System, and we received a message informing us, a message from the headquarters of the acoustic group, which was in Washington at that time, informing us that this large explosion had taken place, and I believe we were not successful, as I recall, no, we certainly were not successful in locating the event until we received the [seismogram] films from stations, including the one in Alaska and the one in Turkey and another location I can't reveal, and I believe we had four stations that recorded the signals. And I think when we had the films together we put the story together, located the epicenter, and it was quite a good epicenter. We had a good azimuthal distribution of stations so that we could, our triangulation process in locating the event was, we thought, quite accurate. And we did make a good location, much more accurate, probably at least an order of magnitude more accurate than could be done with the acoustic system.
Which means about 200 square kilometers order of magnitude or …?
I estimated at the time that we probably were within about 10 kilometers of the correct location. Now that, in retrospect, I think was fairly optimistic about how good we were, but it turned out to be correct. Divert a little bit, but years later, probably right around 1956, I was taken into a highly classified meeting in a facility in Washington, I don't know if I can talk about it yet, and briefed a bunch of people on, that's when I told them I thought the accuracy was within about 10 kilometers. And end of story, I was ushered out, not cleared to understand or be told the real objective of it. But in the early '60s when the Russians began testing at high altitude I was on a panel and a gentleman came up to me and told me that he had remembered that briefing and that he was the U2 pilot that was to go and photograph this site. And he didn't believe he would ever get even close to the actual test site he said, but he said, "Sure enough, it was within 10 kilometers of where you said it was." So we lucked out on that one. At any rate, that was our first real success. And then in 1954, yeah, 1954 I went back. I had by then accumulated enough money to afford another year's school, and all I had left to do was a thesis and French language exams. So I went back, and so I went back to Berkeley for a year and got my Doctor's degree in 19, well, I finished my work in 1955, by September '55, and the degree was awarded early '56.
Do you remember any interaction with students, seismology students, and graduate students at this time?
Oh yes. There were a number of students who became well known seismologists: Don Tocher for one and Jack Evernden for another. And I think those are the, I remember others but those were the two that I was contemporary with and that I think are the most well-known seismologists at that time.
The seismological training was primarily provided by Perry Byerly, or was anybody else teaching you or mentoring you in this respect?
On my thesis work, Jack Evernden, who hadn't run out of money and had finished his Doctor's degree, had returned to Berkeley after I think a couple years with Standard, I believe it was Standard Oil Company. He had returned to Berkeley, and he became my thesis advisor, so I actually worked very closely with Jack during that time. Other classes I took were by instructors or professors of geology, and the only other geophysicist was John Verhoogen, who taught things like electricity and magnet-, no, magnetism, geomagnetism I guess is the right word, and heat flow and that kind of thing in geophysics. But Perry was the only instructor of seismology really.
How was working with Perry Byerly?
Oh, I thought he was a great guy. I had a great personal affection for him. I thought his approach to teaching and his philosophy of life and so on were just excellent. He was a good lecturer, he was very well prepared, a lot of humor, a lot of stories. I urged him at one time to write down his stories because I thought they shouldn't be lost to seismology. Tales of the famous English physicist Knott for instance, he had stories about him. [Note added by Carl Romney: C. G. Knott, one of a small band of British scientists working in the then exotic land of Japan beginning about 1875. This group has been described in some of the literature of seismology as the founders of modern physical seismology.] He had studied with Harold Jeffreys at Cambridge, and he had a famous tale. I'll tell just one, but he said he was punting on the Thames or the Cam, no I think it's the Cam is the river near Cambridge, and he was punting with Harold Jeffreys one day and out in this boat in the middle of the Cam Sir Harold Jeffreys turned to him and says, "You know, Perry, to a first approximation the punt is as wide as the Cam." So that, Perry was full of stories like that from his experience. And he declined to record them because he said he would not be able to document them, and he felt that to be a scholar you really had to be able to document these tales. But he was a firm advocate of believing the data. "The data is the thing," he would say, and "Don't worry about the theory; theory will follow the data" was his philosophy, and in my life I tried to, I think I was very much influenced by that, that the theory will follow the data. And while I succeeded in being misled by theory from time to time, in the end, we really did learn most everything we learned about nuclear test detection through data.
In your experience, was there a kind of rivalry between Caltech, the seismology program, and Berkeley? Because I mean Gutenberg was much more theory-oriented than Byerly apparently.
Gutenberg and Byerly were friendly antagonists, let's put it that way. Yes, there was rivalry. I don't think it got to the point of rancor or anything like that, but I think Byerly felt that Gutenberg wrote so much that, he was so prolific, that any new discovery that ever came along Gutenberg could always say, "Well, I myself have already discovered this."
[Byerly could] be a friendly critic of Gutenberg, and he did have people investigate aspects of some of Gutenberg's papers that he thought maybe Gutenberg had been too casual about putting out some idea or other, and say, "This deserves further investigation," and he would do that. I don't know that the converse was true, because my connections to Caltech are not at that level where I could say whether they reinvestigated problems that Berkeley had done.
So when we think about Caltech and Berkeley, there was not much of a graduate student exchange, or that Gutenberg would come to Berkeley and lecture there and Byerly would go to Caltech and lecture there?
Not during the time I was there. We met fairly often. We met at the scientific meetings, the Seismological Society meetings and sometimes the American Geophysical Union meetings and I of course knew the graduate students at Caltech at any given time while I was a graduate student.
Who comes to mind?
Well, the one that comes to mind was Ben Howell, Ben F. Howell, Jr.
Who went to Penn?
Went to Penn State, yeah. And I can't remember whether he was a little ahead of me or a little behind me. I think he was a little ahead of me. And certainly he got his Doctor's degree before I did. Ben Howell's thesis work was a kind of a follow on to the work that L. Don Leet did at Trinity, and he investigated the funny waves that Leet thought he found as…
The lonely Ps [P-waves] or…?
No, it was a wave he called C, a wave which was the ôcoupledö wave, and a wave he called H, the hydrodynamic wave at Trinity. And some of Ben Howell's work was to make measurements on chemical explosions in California. I can't remember where, but not too far from Los Angeles. And Ben found some confirmation of those waves. Anyway, Ben is the one that comes to mind. We probably knew others if I reflected.
We should come back to Ben Howell later, because I was always wondering why he didn't play a major role in Vela Uniform. He just didn't seem to appear very much, although he had this background is explosion seismology. But we can come to this back later.
So you worked at Beers & Heroy until 1955 and then began your career at the Air Force Technical Applications Center [AFTAC].
In '55, yes.
Can you tell us something about the transition and your early years, if this is open information?
Well, essentially while at Beers & Heroy we were working for what became AFTAC. It was then called AFOAT-1. And so the instrumentation work was largely under Ben Melton, and the other work was under a man named J. Allen Crocker. And both were relatively frequent visitors to us as contractors, both in upstate New York and in Laramie. So I knew the people, I knew Doyle Northrup from several visits that he made, and visits — I would be called into Washington to, well, brief the government and their advisory panels, including I believe the Boner panel, but certainly the panel on seismology of the RDB [Research and Development Board], on progress on our work on a couple of occasions. So I was well acquainted with those people.
So you had a clearance already since your beginning at Beers & Heroy?
Yes, very shortly after, yeah. Yes, I attended meetings, classified meetings in Washington, both at the time I lived in Troy, New York, and at the time I lived in Laramie. I'd have to travel back to Washington and meet with those people.
The Laramie station at this time was also classified, so there was no access for non-cleared seismologists.
Right, yes. We were in rental space at the University of Wyoming, and I believe they were essentially temporary classrooms set up during the war years. The University of Wyoming may have had an Army training program or something there, and had built some rather flimsy wooden buildings, and they were vacated, and Beers & Heroy rented them for the purpose of establishing the analysis center there. The analysts that we had at the time, people who read the seismograms and made the measurements, a couple of them were people who had come from Troy, New York, and they had prior experience in exploration seismology with Geotech, with the Geotechnical corporation. But a good number of them were either wives of students or professors at the University of Wyoming or in some cases they were students moonlighting, working the evening shift or something and going to school during the day. But they were a good resource, educated people, people that we could train easily, and they became quite good seismic analysts. Anyway, during that time I became quite well acquainted with Allen Crocker, and he, shortly before I finished my degree, my work at Cal, he invited me to apply for a job. And I think he felt that it would be best to get that analysis function moved back to Washington from Laramie, and I was to be the initiator of that transition, which did indeed take place a few years later. It took a while to raise the right staff.
Were you concerned at this point to forego a chance of becoming an academic seismologist?
Yes. My intention — I came to Washington thinking I'd be here about three years and then probably return to California and go into academic seismology. In fact the first several years I worked for the government I was a temporary employee. I never applied for a full-time career status, because I didn't intend to stay. And there was some advantage in not being an employee, a slight financial advantage I thought at the time. But my three years stretched into four, and by four years we had been through the Experts Conference, and so I became a career [government seismologist]. It seemed like there was much more to do, and that it was going to be a very interesting time with the, shortly after the Experts I mean, we had to organize the HARDTACK II measurement program, seismic measurement program, and that data was very important to the treaty negotiations that were to take place shortly after the experiments. And one thing led to another, and anyway my three years stretched into more than 40 anyway.
Maybe we can come from your beginning at AFTAC to your participation in Conference of Experts, then Berkner Panel, Technical Working Group 2, and the Seismological Research Program Advisory Group work. Why don't we start with the 1958 Conference of Experts. What was your affiliation with this group, and how did you become involved in this meeting of seismologists from East and West in Geneva?
The events at the time, the international events at the time, were such that at least Eisenhower and, well, Bulganin and then Khrushchev, were convinced that things were getting a bit out of hand, I think. The Russians' nuclear program was proceeding at an enormous pace, much faster than the U.S. program and their buildup of conventional forces in Europe, tanks and aircraft and so on, were very threatening. And I think both leaders and other leaders of the world felt that some kind of first step toward disarmament was needed. And I believe, also from public statements and so on, that people at the time thought a good place to start would be the nuclear test detection. People think of nuclear tests as enormous releases of energy and therefore infinitely detectable, which of course is not the case, as we know now, and some of us even knew then. But it seemed like a natural place to start. My involvement, to get back to the direct question, my involvement really began when Eisenhower created a panel under Hans Bethe to look into questions of the desirability of a nuclear test ban. And that included the verifiability of the nuclear test ban. But I don't think verification was the big issue for the Bethe panel. I think the big issue for the Bethe panel had to do with the balance, the relative developments or relative capabilities in the nuclear arena of the two countries, and that was the big question. But an important question was if we do have a test ban how well can we verify it. And I was invited to, I was not a member of the panel, but I was invited to, I have firm records of two meetings, possibly a third, but I can't be sure at this moment. And during that time, or during the time of the Bethe panel's deliberations, which I think started sometime toward the end of '57 or certainly early '58, I was invited to a couple of the meetings and quizzed on my understanding of how well we could detect, what we could detect, and a little bit on the possibilities of concealment of nuclear tests. And I did, I had I should say, jointly with Les Bailey studied the first underground nuclear test which took place in September 1957, the RAINIER explosion. And Bailey and I had collected the seismograms that were available from the U.S. and I believe some Canadian sources, but I can't guarantee that, at least U.S. and Alaskan sources. And so we had done the only study of the detectability of underground, an actual underground nuclear explosion. Because of my involvement with AFTAC I had also studied large chemical explosions when they took place, and here the interest of course was in explosions large enough to be recorded at really long ranges, certainly a thousand kilometers, but most of the interest was in events detected at many thousand kilometers. And so that was the kind of background that I had at the time in meeting with and advising the Bethe panel.
Were you the only seismologist who was quizzed by the Bethe panel?
As far as I know I was the only one that was quizzed by the Bethe panel. Certainly there were no others at the meetings I attended.
That's very interesting, because your name doesn't appear in the report, not even in the Appendix, which was written by Harold Brown and Hans Bethe as far as I recollect on the seismic issues. So now I understand much better where they had the information from. That's interesting.
Yeah. They had what information was available to me, and we continued to, when it became really very clear in the spring of '58 that there was going to be, or very probably was going to be, a meeting between East and West, we in AFTAC continued our studies of capabilities, what capability existed in the Atomic Energy Detection System and what would we want to ask for in terms of additional stations, maybe including stations inside the Soviet Union for verification purposes. And these kinds of studies were conducted within my group on seismic capability, but there were similar studies in acoustic capability and the detection of radioactivity, EMP detection, all the technologies that were functioning at that time. We were assessing their capability to monitor an agreement, should it take place, and also making estimates of how much better they could do with additional stations from place to place. And within both the USSR and China. We considered at the time you really couldn't separate the USSR and China, that if China didn't come along the United States shouldn't go along with the Test Ban Treaty because the relations between the Chinese and the Soviets were so good at that time. But at any rate we continued to do studies and got very heavily involved in the thinking of what does it take to monitor, and including some studies for Jim Fisk after he was selected by Eisenhower to head the U.S. experts group and ultimately the Western Experts Group. And so I just, I guess I was just one of the natural choices to go along with the experts, and of course they invited other people who were experts in seismology, Frank Press and Jack Oliver, and…
Norman Haskell, Perry Byerly, and all of these people had done some studying of results from the explosions. Oliver and his group for instance had, they had looked at surface waves for instance from the HARDTACK II series. I mean the RAINIER and the Pacific explosions in particular. WIGWAM I believe was the underwater explosion in the Pacific, and also some of the very large explosions at Bikini and Enewetak, and they had done studies of the seismic waves from these. So they were not exclusively focused on this kind of problem as I was, but they were knowledgeable of aspects of detection.
Whom do you recall sitting on the Eastern panel's side as a seismological expert?
The Eastern side. Well, the ones I knew were [Alois] Zatopek from Czechoslovakia, [Y. V.] Riznichenko and [V. I.] Keilis-Borok from the Soviet Union. We knew a little bit. [Ivan P.] PasechnikÆs name was known, but more as an exploration geophysicist. We didn't know him, I know I certainly didn't and I don't believe the others did, know him as having anything to do with nuclear test detection. And I believe [Mikhail A.] Sadovsky was totally unknown to us. Not that Sadovsky was a seismologist, but he was very knowledgeable of seismology and he was very knowledgeable of rock mechanics and a bunch of other related things. As far as the other countries were concerned on the East I don't believe there was anyone that, certainly no one that I knew of, and I don't believe there were any others that were really seismologists. In the UK group, just to round it out, [Sir William] Bill Penny was very well known to us, and there had been excellent cooperation between the Atomic Weapons Research Establishment, which he was a very senior member of, and he might have been the director of Aldermaston at least at that time. Penny was well known to us, and Penny had done some work on acoustics, although that wasn't his field I don't believe, that wasn't his primary field. But I don't recall that anyone else during the Experts was very well known to us in the nuclear test detection category. Sir Edward Bullard came as a British expert, and he was very well known to us, because he's a very highly respected British geophysicist, but less on seismology than on the, I believe one of his fields was geomagnetism. But at any rate, and high pressure stuff, high pressure rock mechanics, and no practicing seismologists, I don't believe.
How did the daily life of an advisor in Geneva look like during this kind of Conference of Experts? Would you meet during the day for a couple of hours for an official session and then meet just the American delegation in order to fix some or to recalculate some seismological problems or…?
Well, a typical day would start with a delegation meeting in the morning, and during which we'd sit around a big table in a conference room in the U.S. Consulate, and go over what the plan for the day was, what we were going to propose to talk about. There would be some writing assignments, and groups or individuals would go off and draft papers, technical papers or discussion papers, and circulate them to others who could help. And then I believe we would have another meeting, maybe right after lunch, to review what had been done and make general comments. I mean everybody could talk about anything. Anybody, regardless of their qualifications, could talk about anybody else's paper. It was quite open. And then the formal sessions took place in the afternoon. The Russians like to meet after lunch, and of course that was their main meal of the day so it couldn't be rushed. My recollection is that was like 2 or 3 o'clock in the afternoon, and the formal sessions would go on until we ran out of things to talk about for the day. At least usually a couple of hours, 1 to 2 hours would probably be typical. Then another delegation meeting back at the Consulate, .S. Consulate, to compare notes on what we had heard and what we had understood and what we thought the other side might be driving at and then make some decisions as to what's next for us. Somebody had to go off and write a cable describing the meeting to send back to Washington of course during this time. By then it was dinnertime for us and we would usually then adjourn and go have some dinner. And then most of us would come back after dinner at maybe 8 or 9 o'clock and work until midnight to 1 o'clock. It was very hard work, and I personally got nearly exhausted during that six or eight weeks, six or seven weeks I guess, seven weeks. I lost 10 lbs., and my wife said I looked like I just got out of a prisoner of war camp when I came back. My wristwatch was rattling on my wrist and I looked pretty bad. It was very difficult work, and it went on, except for a couple of weekends near the end, it went on during weekends as well as weekdays, although without the formal meeting. We worked very hard.
You said there were cables written to Washington. In the published documents of the Geneva Conference and later in comments by people like Robert Gilpin, for example, there is always this charge that the American delegation was basically left alone fighting for themselves without political guidance from Washington. Did you…?
I believe there was very little guidance coming back. Now I probably was not privy to, maybe not even very much of it, I don't know, but I think there was very little guidance coming from Washington to Geneva. Mostly we were just reporting factually as we could on what we heard, or thought we heard, back to Washington, and they were following our progress. But I don't think trying to torque us in any way.
Was there any kind of meeting before the Geneva Conference with people from the State Department or Department of Defense asking or making some recommendations how the negotiations should be led, how the scientists should behave in the interaction with the Russian colleagues or ?
We were briefed by the State Department at one stage, but it seemed that that was more protocol. And the thing I remember most about it is I was advised to buy a tux, because there would surely be some festive meetings, some rather formal dinners and things of that nature which of course turned out to be absurd. I mean, the Russians couldn't afford tuxes in those days, so they wore rather baggy looking business suits and they weren't about to waste their money on a tux. But it was, that's the more the spirit of what I remember is the dress code and how to behave. Oh, security also. We were told that, "Do not trust your hotel room to be a secure place to talk, and even though Geneva is supposedly neutral."
...January 20th, 1998. It's about 10 minutes past 3:00 in the afternoon. I'm sitting in the office of Dr. Carl Romney in the Center for Monitoring Research in Arlington, Washington, D.C., and this is part two of the interview with Dr. Romney, and we continue our conversation about the Geneva Conference of Experts. Dr. Romney, what else can you remember from the interaction between the American and Soviet scientists at this time?
I neglected to mention one important thing about the Soviet scientists. The Soviet Union and I can't define exactly how this came about, but they wanted to put a condition on the meeting. They wanted us to agree in advance that the purpose of the meeting was to, that the meeting was going to lead to negotiations on the Test Ban Treaty. And of course the United States refused to agree to that as a precondition; we were simply going to see if we had a common understanding of the technical means for nuclear test detection. And so we didn't, they said they were not going to waste their money on a conference without knowing it was going to lead to something concrete. So as of the time we left the United States, we had no knowledge whether the Russians would even show up. And so when we first met them, there was great surprise and wonderment on our part. And we didn't know who these characters were. We had no idea who these characters were. But, as I mentioned some of them and I won't even guarantee they were all there, Riznichenko and Keilis-Borok that we knew fairly well, I'm not certain they were there at the opening meeting. They might have been. But anyway, we knew some of them, but others were total unknown quantities to us, so it was a great surprise. When we did have social interactions with the Soviets initially, I've forgotten who sponsored a reception, afternoon reception, get acquainted reception, it may well have been the Soviets. But they were very cordial, and when we found out more about the Soviet delegation, we found them to be people of really great competence and they were pleasant to us. I think I mentioned before, we found their humor to be very much like American humor; we found their way of thinking socially to be very similar to the way the U.S. thought, that is they were rather rough and ready folks, very little formality. And so the social parts were quite comfortable for us. And as I say we knew they were quite competent.
Coming back to the debates about seismological issues. What were the major issues debated during the Geneva Conference of Experts?
I really should look at the agenda of the meeting. I have kind of forgotten the formal agenda. We finally, it took us a while to agree on an agenda, because Federov, the leader of the Soviet delegates, tried to insist for several days that we pre-agree to that this was to go on to negotiations. Which we were unable to do. But when we did address the actual agenda, we started to take each technique in sequence, each possible detection technique like seismic or acoustic or electromagnetic pulse, and define the phenomena that we were dealing with — how one detected in the case of seismology the kind of waves that would be involved in detect-, you'd have to record, how you recorded, and talked about the capabilities, how far away you could detect signals from a bomb of some particular size. We started, I believe we started doing this in the formal sessions and then began to discover that it was going to take us forever to work our way through all of the different possible techniques and talk and talk about the capabilities and apparatus, that was the word we used, apparatus for detection. And so after a while we split into informal working groups, and to discuss the techniques of detection and the capabilities for detection. After we had reached agreement, then the results of the joint informal working groups would be reported in the main sessions of the and usually with a document, some kind of a description of the technique. And so then they would be blessed by the conference in a kind of a formal sense. In seismology which of course I was deeply involved in almost to the exclusion of involvement in the other technologies, we found that there was quite a difference in estimates of capability. I would characterize the Russian tactic as kind of seizing on the most favorable report. If RAINIER, for example, had been recorded fairly well out to about a thousand kilometers and then almost not at all until 30 degrees, 3,500 I think it was kilometers at College, Alaska, if you talked about that particular event, the Russians would want to say, "Well, you can detect 1.7 kilotons out to 3,500 kilometers." Which of course is not true in general. Yes, it did happen at one station, but no, you can't take an isolated instance like that and use it as the basis for designing a system for monitoring nuclear test detection. You have to focus more on the distance to which you can detect reliably.
This is basically a question of the risk which one is willing to take in having a monitoring system. Was there some kind of consensus about what kind of risk would be acceptable among the Western delegation? Was there a search for a 100 percent foolproof system versus let's say a 90 or 75 percent?
As far as the U.S. was concerned, I believe we tended to think in terms of like 90 percent probability. We thought that's the kind of capability one ought to have. Federov made it very plain at one stage I think fairly early, at least in connection with the seismic, that the Soviet view was that you don't need to build a system that is capable all by itself of doing the job of monitoring, and he would say things like, "After all, you know, all tests up to this time have been detected." Which wasn't quite true. But, and so he would go on to say that what we want to do is supplement the existing seismic system, that should be our objective, what do we need to do to improve things a little bit, incrementally maybe or. That's not his words, but that was the flavor of what he was proposing to us. We were thinking about a system that would operate, and we didn't trust the stuff that was already there. We didn't want to rely in any way, any important way, on seismic stations in the Soviet Union operated by Russians. Our identification criterion at the time, the best, the only two we had were determining the focal depth and determining the direction of first motion, and we knew how easy it would be for Russians to reverse the polarity of their seismometers and create artificial rarefactions. And so we wanted no part of that. We wanted a system which was reliable and guaranteed in some ways in that it would be operated in a way you could believe the data. And I think we were modeling our system, our, my mental, let's put it that way, my mental model of the system was something like an expanded Atomic Energy Detection System, USAEDS, that that's the kind of system we really wanted to have. Very reliable. Americans on the scene during the operation. But the Russians, they, and I think this is true in all the techniques, but I can't be as explicit in the others, but they would seize upon this longest range or lowest yield and cite that as the capability of the technique, of the method.
So coming to some specific points like first motion. In the end there was an agreement that a signal-to-noise ratio of 2:1 [two to one] would be sufficient. How was the negotiation about this particular issue? Because this made a lot of problems later on in the Technical Working Group 2.
Yeah. I don't know that that was agreed. You see, the Russians presented their — well, let me make it a little more continuous with what we've been talking about before. After agreeing on the capabilities of these various techniques, then it was the experts' task to propose a system or a set of systems employing these techniques. Now the U.S. thought we should describe systems, we should describe a system with a low capability and a system with a high capability. The Russians wanted only to discuss a system, and that in their view was the recommended system. We had no authority to recommend anything. We had the authority to describe systems and their capability, but governments would decide what the recommended system was. They wouldn’t accept that. The Russians tabled the first [description of a monitoring] system, but they never defended it, they never gave a basis for their estimates. If they described its capability in terms of some kiloton number I don't remember it. I don't think they did. They just presented it as something that would be acceptable and adequate in their view. We on the U.S. delegation developed a rather formal method of evaluating systems. We had information. Let's start from a more nuclear point of view. We developed a model for the amplitude as a function of distance in seismology and we developed a model for the yield versus magnitude in seismology. We were wrong in all probability on these things, but we used our experience to develop a noise model. And so with a noise model, a signal model and an assumption about signal-to-noise ratio adequate for detection, you can then begin evaluating a system. You choose places to put the stations, make guesses as to what their noise levels were going to be at each location, some would be noisy, some would be quiet. Then have trial places here and there where a bomb might be detonated and calculate the signal and calculate the signal-to-noise ratio at each of the stations; then you know how many stations you need to get a reasonable location, and it's like four or five stations, and you know things about the signal-to-noise ratio for detection, and it's like two for a minimal detection, and then you can begin to calculate the compound probability of n stations detecting signals. So we did that, and we also tried to estimate the identification capability. And we do that by an assumption for depth that we can determine the depth if it's greater than 50 kilometers or something like that. We make an assumption that we can see the direction of first motion if the signal-to-noise ratio is something, and we, the U.S. group, assumed that a 2:1 [two to one] signal-to-noise ratio would be adequate. That turns out — when we had much more data after HARDTACK II explosions — that turned out to be a very bad assumption, a totally minimal assumption. But we made it, and maybe thinking, well, you know there will be progress in science and the data from RAINIER was not very good in this respect to what it really does take for a bomb to…
Was there concern among the Western delegation that extrapolating from a single data point would be a major problem down the road?
Oh, I think we were very uneasy about the fact that we had so little data. We had a lot of data on earthquakes, which are, of course one of the main problems is getting rid of the earthquakes or successfully identifying them so you don't have to suspect them, so we would have liked to [have] had much more data. We did have data from a few chemical explosions that were fairly large, but I think, yes, we felt that the data was very tenuous in making these estimates. But you do the best you can with the information at your disposal at the time, and I think that's what we did. But I do think there was some wishful thinking that maybe we could force things to improve beyond what we really knew. I think also we thought that RAINIER might not be very well coupled. Our knowledge was that sophisticated, that there probably were better coupled explosions. So that also would tend to allow us to make an optimistic assumption on signal-to-noise ratio.
Who were the driving seismologists in the U.S. team? Who was primarily involved in hammering out these papers?
The capability estimates that we made, well, the initial presentation on seismology leaning heavily on RAINIER and the experience in the AEDS was by me, and it was a broad discussion of seismic capability and estimating detection distances and that kind of thing. The initial work on estimating system capabilities I think came from Norman Haskell, who — he had done some previous thinking on that kind of thing and had written a paper, I have a copy of this somewhere, on the maximum detectability of signals, a quite theoretical paper. He had a good grasp of the kind of statistics that one needs to crank into a problem of the type I've rather casually described. And he wrote out an approach, he developed an approach to how one goes about combining these detection estimates, the compound probability aspect of it. So he set the stage. The people who then did the main capability estimates were [Harold] Brown and Dick Latter, both of which had, both men had a really good grasp of these kinds of problems. I don't know quite why. I can't point to what in their experience is what I am saying led to that, but they did have a very good grasp of that kind of problem, and they began to do the actual formal assessments. Now probably people like me, I probably gave him the propagation model, the amplitude-distance model. I probably, with Jack Oliver and Frank Press maybe. I remember Oliver and I don't remember Frank doing this. Frank might have gone home by that time. I don't know. I do remember Jack and I and Norman spent quite a lot of time. We probably said well this station is probably going to be one of the quietest, and this station is probably going to be one of the noisiest, and we had a language in our, agreed [report] with the Russians. There were stations that were considerably more quiet than average and there were quiet stations and so our noise model, the way we approached it we had three levels of stations: we had quiet stations, average stations and noisy stations, or [rather] considerably more quiet than average, average and noisy stations. And we — so the seismologists probably defined which was which and what the mean noise and what the standard deviation of the noise was for a quiet station. And then, that kind of thing, and then Harold Brown and Dick Latter did the actual calculations. We seismologists probably also said things like you need, how did this go? You need to have at least five stations to be sure. I think that was it: you need to detect first motion at five stations to have a good chance of getting a rarefaction on an earthquake. And we may have said something about azimuthal distribution, but I don't remember that.
I mean, if the five stations will be in one quadrant, it wouldn't help you much.
Yeah. It wouldn't help you much. So we probably had some logic like that that we suggested be cranked into it. And then Dick [Latter] and Harold [Brown] did the calculations, and they presented it at the conference too, their evaluations of the network. And I remember the Russian, the proposed Russian network was probably not effective for smaller than around 20 kilotons [according to the U.S. analysis] and the network we had to go to get down to about 1 kiloton for detection identification was the famous 600-station network that we were, that we did refer back to the United States before we presented it. We could see the potential for the Russians to really raise hell about that.
I think this is easy to see. On the other hand also, I mean, 600 stations with at least 10 to 30 personnel would be a major, major system.
Oh yeah. Enormous. Enormous. Yeah.
And what came back from Washington? They said, well?
They said go ahead.
Go ahead. So that was, that was, we said, "Well, here's a network that will get you down to 1 kiloton." And then the Brits came in, we I might have mentioned this is one of my e-mails to you, but Brits came in with the 170-station network, and they tabled that. Now of course they worked with us, there is no doubt about it. They worked with us, and we probably did the evaluation of the network. Because we had the methodology. But Penny was looking for a middle ground and…
He might have felt a pressure from Macmillan. I mean, they were very much interested in coming to an agreement. Macmillan was hard pressed because of the fallout issue, because of; he needed something tangible to come to an arms control agreement.
Yeah, well it's quite possible, and as I say, I'm sure that the scientists there felt that 600 was impractical. You know, I mean, I had enough experience in operating the AEDS to know that — that was under 15 stations at that time I think, and that's a big job just analyzing the data from them.
And this is under your own control. If you have to set up an international control organization this would be even harder.
So I don't know. I can't, I no longer remember whether, how far the Brits went in actually defining [details of] this network, but they worked with us on this middle ground network and they tabled it.
In mid-August 1958 the report was made public as far as I recollect. What was the feeling among the American delegation? Was it a feeling of great accomplishment, success? How would you describe the reactions among the American scientists?
I think in general they were pleased. I felt that the agreement for on-site inspection was a very important thing. We actually had an agreed document where the Soviets agreed with us that on-site inspection was needed in some [cases], there had been some talk about inspection zones in connection with surprise attack I think. ...some talk about the necessity of inspection under some circumstances. We had it agreed, and I think we thought that — most of the American scientists thought that — was a really important agreement. We really pinned them down and they signed on the dotted line. I was personally, thought that we had overestimated the capability of the network, and that's where my statements about optimism — that things are going to improve and that RAINIER was really not very well coupled — came into play. But I had some doubts that the system as we were describing it would do what we said it would do. I didn't think we were off by an order of magnitude or anything like that, but I was quite prepared for us to be off by a factor of two or three let's say.
Was there a range of opinions and discussions among the American experts talking about on what kind of system capability one should come down to and agree on? I mean, was there a feeling among people like let's say Harold Brown versus Hans Bethe that things have not been accomplished which should have been accomplished?
Well I think Harold Brown for instance would say that this system that we're talking about here isn't adequate because 1 kiloton is damned important, and even the experts don't think it will get down to 1 kiloton except in some unlikely circumstances or some unusual circumstances. But I don't think we had much discussion about how well we were assessing the capability of the network. But as I say, I thought I could agree to our assessment only in this framework of optimism that things will; I don't know what other people thought, whether they thought, "Oh God, we're exaggerating the difficulties," or I don't believe so. I don't believe so.
So there was no internal dispute between proponents and opponents of the test ban agreement in general?
No. And I think everyone who had anything to do with this knew that the range of uncertainty was very large. And we had tried to get some of that kind of stuff into the report, but Federov didn't want numbers, and he resisted. We tabled some language [from] time to time that contained ranges and things like that, but he didn't want that. That just confused the politicians, or whatever his reason was, that's part of his reasoning. But…
In retrospect do you think that would have been helpful to have a senior State Department official sitting with the scientists on the American delegation in Geneva, like Tsarapkin did for the Russians?
Not as far as the discussions at the Experts took place. That wouldn't have done any good. It was important. It was an important advantage that the Russians had when the diplomatic negotiations really began on the treaty, because Tsarapkin had been wallowing in this stuff for two months prior to the start of the meetings and he knew the language and he knew what was in the report, the Experts' report, he could quote it like the Bible, and so it was an advantage for him, but there was no disadvantage during the Experts [conference] in not having a more senior guy there. Ron Spiers was our senior guy, he was young and he was very bright and he supported the delegation in the right ways as far as protocol and who to contact in Washington if you needed some advice like on the question “do we table the 600-station network?” He knew all that kind of stuff.
So he could help with language of certain proposals?
Because I remember that in Robert Gilpin's evaluation of the Conference of Experts he pointed out that Tsarapkin was of substantial help in making distinctions between tests in the oceans and in the open sea and diplomatic language differences of this sort, and Gilpin charged that the American scientists were left alone and basically had to fight
Who made this charge?
Robert Gilpin. He wrote this book [Robert Gilpin, American Scientists and Nuclear Weapons Policy (Princeton, NJ: Princeton University Press, 1962)]
Yeah, yeah, I've got the book. I haven't read it for ages, but, well, I don't think that's, I don't remember it like that. I think Ron was quite capable of giving us that kind of help.
It's been a long time, but I think he could give us that kind of help.
Well, I think this covers the Conference of Experts for the moment. What was the development after that? So the expert conference was over in August 21st, 1958, and then in October we had the underground nuclear tests as part of Operation HARDTACK II. Can you describe the time in between? You were at AFTAC and you knew that Operation HARDTACK II would include a number of underground tests. How did you become involved in the measurements, seismological measurements, of these new tests?
Well, we knew the importance of these tests, I mean before the Experts Conference ever started. Once we embarked on this thought process, of thinking, of defining systems which we did in AFTAC, even before the Experts Conference, we knew the importance of more data to do it really right. And what I didn't know at the time is that also Fisk and Bethe, the Bethe panel itself, had recognized that the data from these underground tests were going to be quite important to the United States, or potentially quite important. I didn't realize that, but I did realize how valuable they would be to me in my job in AFTAC. By the end of the Conference of Experts though it became really crystal clear to me and to everybody else that these were really vital, these tests could be almost infinitely important to solve the kind of questions that the Russians had raised, quibbles about our interpretation, or maybe we were wrong. But we knew that we had to do something about that. And I had not been able to determine the time at which we really started the actual planning of the experiment, but it might well have been initiated from Geneva itself. Certainly we knew, I knew by the end of that time that I had strong support from Fisk and Bacher — Fisk particularly — and Hans Bethe and other senior people, which we should do what we could to get the data. We AFTAC, because there was no one else; that we had a contractor group and we had military people trained to install seismographs and get data. So somewhere, either instantly after it — the Experts Conference — or during the conference, we began the organizing of this experiment. And up until the end of October I don't think I was able to do very much of anything else except be involved in this program. The Russians of course were testing very actively during that time, and so I was also involved in monitoring the Soviet Union. And they were testing big tests that we could detect seismically, so a lot was going on. But I think my major focus was in planning and getting money out to contractors, getting our military teams committed to do it, and doing it in the right way. So we organized, we put together as many teams as we could and they were all equipped with seismometers that we knew could be calibrated and organized the experiment and carried it out, analyzed the data as quickly, as fast as it came in, which in those days was by mail. And so it was a very busy period, but focused on HARDTACK II.
And then the data were evaluated by you and an AFTAC team?
Yes. We did the initial evaluation of the data. I believe we probably had some help in this analysis out of Geotech. I believe the Laramie Analysis Center was still going at that time, and it's possible they did some of the preliminary analysis. I don't, my recollection is not that clear on that. I do know they participated, but I don't know quite how they participated. But we put together the conclusions, the summaries of the data, and then the interpretations of the data, and by, just before Thanksgiving I was called over to Geneva by my boss, Doyle Northrup, to brief them on what we had found so far. And so I went packing off to Geneva carrying as much data as I could and then the results of our interpretation and briefed the ambassadors. I briefed both [James J.] Wadsworth and [David] Ormsby-Gore, I don't remember whether there were two briefings or one briefing. Probably two. I mean, knowing how diplomats think, I would think our — but I don't remember. I know I briefed them both on the results of that. And it was clear that the capabilities for detecting 5 kt were not as great as we had said in the Experts report, and it particularly was clear that the first motion was going to be a problem, detecting the first motion was going to be a problem. We've had clear rarefactions at not very many hundred kilometers — or apparent clear rarefactions, where we know that the compression was out in front but it was lost in the noise — at not very many hundred kilometers. And so we knew that criterion was going to be much less effective. We also thought that instead of a two-thirds power relationship between amplitude and yield, which was our assumption guided by some theoretical work of Bethe's, that we had used during the Experts Conference, and now it looked like it was a first power relationship between amplitude and yield. Which meant, as you go to low amplitudes, it meant that it fell off more rapidly than we had been assuming. That was the significance of that slope change as far as our HARDTACK-thinking was concerned, so that if you can do something at 5 kt, it's going to be harder to do it in 1 kt than we had previously thought. So that much was clear. And that meant that we would have an increased number of unidentified events at low yield. And so that was my Thanksgiving Day message to our delegation. I remember it was Thanksgiving Day.
How did they receive this message?
The technical people there, Doyle Northrup and I can't remember who else on the U.S. delegation, I draw a blank on that, but technical people anyway understood it. And I don't think Wadsworth had much appreciation for it. I think he realized, of course, that it was negative in the sense that we were going to have to start arguing with the Soviets who wanted us to just never look at anything beyond the Experts report. That was the Bible, and we were to follow its injunctions or its wisdom explicitly and completely and only. But I don't think Wadsworth had been involved in the problem long enough to have a good appreciation for what a ruckus was going to finally result. I may be wrong, but I had an impression that he took it fairly calmly. And at any rate it was very clear that we needed to get our analysis completed and we were going to have to present this to the public and to the Soviets and probably ask for some reassessment of the Experts report in this respect.
Which didn't make President Eisenhower very happy either.
I'm sure. He probably understood it pretty soon. I don't know that I've read what his reaction was. Have you?
I think he said something along the line that the scientists have failed him, so that now new scientific insights would make nuclear arms control even harder than it would be in any case. Well, but this was probably the first reaction. In the end of course he agreed that there has to be Technical Working Group 2.
Yeah. Well he must have appreciated that very early on. He'd be not using his science advisor and his PSAC [President’s Science Advisory Committee] very well if he didn't appreciate it quite quickly, and those groups of course, they understood. All of the technical people really understood instantly that this was a serious disagreement with the conclusions we'd come to earlier.
Let me just break here for a second.