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 Stephen Lukasik by Finn Aaserud on 1987 April 21, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/25656
For multiple citations, "AIP" is the preferred abbreviation for the location.
Topics discussed include: Lukasik's publication records; family history and early scientific interest; scientific education at Renssalaer Polytechnic Institute; interest in acoustics; expectations of eventual career; education and atmosphere at Massachusetts Institute of Technology (MIT); job at Stevens Institute of Technology, work in hydrodynamics; move to Advanced Research Projects Agency; supervision of program in nuclear test detection, the Vela program; interest in computer science at ARPA; systems analysis at ARPA; Deputy director of ARPA, then director; reason for leaving ARPA; move to Xerox, and relations between Xerox and ARPA; changes at ARPA; work at Rand; moving between defense and non-defense careers; first meeting with Jason group; expectations for management of Jason; relations between Jason, Institute for Defense Analyses, and ARPA; need for Jason to have better management than IDA would supply; move of Jason from IDA to Stanford Research Institute; Jason's annual study cycle and ARPA's input into it; changes in Jason's operating procedures; decline in goverment's willingness to listen to science advisors; Jason's work on the McNamara Wall; current Jason studies; Jason giving technical advice vs. policy advice; Jason's position in world of science advisors; physicists' importance in science advising.
We usually ask people first, before we start with their careers, about papers, you know, the existence of your papers, their eventual accessibility, whether something has been arranged for their access to historians, stuff like that. Of course, you may be a little too young for us to ask that question now. I hope you're not insulted by that.
Well, my publication record falls into several phases. There's the period from graduate school through 1966, and that was the period when I would call myself a working scientist. And I have papers in the JOURNAL OF GEOPHYSICAL RESEARCH, THE JOURNAL OF APPLIED PHYSICS, JOURNAL OF ACOUSTICAL SOCIETY OF AMERICA, several journals, perhaps ten or a dozen papers, and also papers at meetings. I'm not counting those because those are just abstracts. I'm talking about full papers. When I went into government, obviously I stopped being a working scientist. Now there's a set of publications, which are more apt to be about national security issues or testimony on a given subject. I was in the nuclear test detection business. There are several major testimonies, and so then there was that. Then, after that, it's sort of catch as catch can. When I was in the Federal Communications Commission, there was also a lot of testimonies that were generally my own, sometimes heavy staff involvement. And then there's a set of lectures and papers and think pieces and so on, so that there are several phases, depending on whether one was in government or out of government, working in national security or whatever. So yes, there is a collection of things, and they range from reports and speeches and hearings and so on, rather than the conventional scientific, which really only covers '56 through '66.
Yes. Your papers would also include correspondence, manuscripts, everything that's been put on paper in some way or other, because you know, historians like to peak into the —
Yes, there are a few things. When I was at Rand, I did an interesting paper on technology and arms control. So there are a few things there.
But correspondence in connection with your appointment would be at the institutions, I suppose.
Yes. It's my files but, you know — I would be honored if somebody wanted this stuff, but it's all sort of, what's public records, or what's private, what's personal records.
Yes, right. So you know, we're interested in — that's part of the business of the Center anyway, to try to take care of records, of the history of physics. For my own benefit, would you know anything about any papers of your own or other that have to do with the Jason ARPA thing, that might be accessible to an unclassified person like me?
Not really. Probably this report, since it is the results of each person — each person's chapter was the result of extensive interviews. That's probably as good a statement. That is, they've accurately captured my views, so I'd say my chapter in this volume is probably about, you know, says a lot about my approach to ARPA. There's probably very little about, I don't think I've ever written about the Jasons. I'm willing to talk about them, but I don't think I've ever had occasion to write anything organized about them.
No, right, so that oral history would be one thing anyway. So, OK — let's start with the beginning, so to speak. I usually start out with family background, motivation for starting in physics, education and all that, and then turn to the working career.
History, family history has no science, no professionals in it, typical second generation American, on my father's side, but on my mother's side, going back to New York City in 1680, so a series of house painters and gardeners and butchers and industrial workers, the usual set of things. [Note: Added in 2004 — all this family history was unknown to me in 1987]. Nothing at all professional or distinguished in any sense. My father's family came to the US in late 1900, and his mother and father, and on my mother's side, it was a generation earlier, perhaps 1870, German and Irish on her side, Polish on my father's side. And the family lived in Staten Island, New York, so we're East Coasters, New Yorkers, and both families had been there for a long period of time, for as long as they'd been in the United States, and my interest in science seems to have no logical antecedent. I was born in 1931 —
What was the educational background of your parents?
Oh, my father was an accountant. At that time, you didn't get a bachelor's degree for accounting, so he went to accounting school. My mother did not graduate from high school. She had to go to work. And no other education, in their siblings. So I was born in 1931, and displayed an early history in science. I read a lot, read a lot of science among other things, and was always interested in chemistry. I had a chemistry laboratory when I was in high school. Of course chemistry is an easy thing to do at home, because the equipment and the chemicals can be procured. As I read in chemistry and prepared for a career in chemistry, it became clear to me that the parts of chemistry that I was interested in were really physics. That is, I was interested in reactions, reaction rates, atomic structure, why things were the way they were, so rather than applied chemistry, I was interested in what one would call theoretical chemistry. And so I said, "Well, I'm really interested in the why of chemistry. I guess that's really physics." So therefore in my last year of high school, I switched over from an intention of majoring in chemistry to what became a physics major. That turned out to be exactly right.
Were there any specific inputs that led to your interest, first in chemistry and then in physics, or was it mostly self-motivated?
Probably the chemistry, at that time, I was always interested in science, and there was a magazine called POPULAR SCIENCE, and it had a chemistry column in the back, home experiments. I also had a friend who later went into medicine who also had a home chemistry laboratory, and so I probably followed his lead, and, inspired by this reading in popular magazines about chemistry and the ease of setting up a laboratory, I went the chemistry laboratory route, which really had the effect. I found myself doing experiments like measuring dissociation, energies of molecules and things like that. I was doing that kind of thing, and also microscopy. I was interested in biology. I should mention that. The Barrett Park zoo on Staten Island had a woman veterinarian who ran twice a week biology classes. She would essentially have the high school students in while she did dissections of the animals that died, and then we would do blood chemistry and sectioning and gross pathology and skeletal structure and so on, so that really, there was a fairly broad education in both biology and chemistry, leading into physics.
That was connected with your schooling?
Not really. It was really the municipal zoo. It was a municipal activity rather than school activity.
You had to know about it.
Right. I was in the biology honor society. So all through high school, there was this broad base of science, focusing on the why of things — that was physics. So that's how I got into physics. Then of course I went to college.
Your parents were supportive of this?
Yes. Yes, they were quite helpful. They didn't have a lot of money and weren't lavish but were quite supportive, yes. I never did anything stupid like setting the house on fire. They had no reason to be negative about it. It kept me off the streets. So it is interesting that the early focus was always on science, and science broadly, and we'll return to that as we follow the career, because I don't — I view myself as a scientist trained in physics, rather than as a physicist. And therefore I've been interested in science and science policy, and it's always been not just what's good for physicists or particle accelerators or telescopes; it's been, what's good for science. And that breadth, that wide bandwidth has been a factor.
Yes. Do you have any siblings?
Not really. Curiously enough, my mother gave birth to a daughter when I was in the second year of college, so in effect there were two only children in the family. So shall I go on to the schools then? I went to Renssalaer Polytech.
Yes, unless there are some other special influences you would like to talk about.
No. Let's see, I was accepted by MIT, then won a full tuition scholarship to Renssalaer, thought for a week actually about accepting it, because I really wanted to go to MIT, in fact went to RPI, then went to MIT for graduate work, so I got it all in the end I wanted it. And I think there was nothing noteworthy about that period, except I felt I learned my science, and particularly learned my physics and the surrounding areas.
Any particular teachers that you would point to at Renssalaer?
No. A number of good ones. My general reaction is one of effective teaching, effective learning, good facilities, good courses. I'm always puzzled by students that spend all of their time complaining about the administration, the faculty, the courses, the homework, everything, they always seem to be complaining about things. My recollection is very positive. It's a great place. I learned a lot and everything seemed to be well structured and organized. Now, remember, this was 1947 to 1951, and of course students weren't quite as complaining in those days as they turned out to be in — after the sixties. So I think it was just a good solid bread and butter education.
What did you learn there in the late forties, early fifties, in terms of physics?
Well, the Renssalaer curriculum was a very broad curriculum, heavy focus on what one would call classical physics. That is, there really wasn't any quantum mechanics. There was a nuclear physics course. The head of the department, his field was geometrical optics, so we had a lot of geometrical and physical optics, which turned out to be very good, because we did photographic processes and we did lots of machining and building things. I thought it was very practical. So I rather liked it. We had an acoustics course and so on. But it was classical physics. I went to MIT, had a research assistantship, and it was in the acoustics lab. That was just an accident. I knew some — I knew Dick Bolt (?) who was the head of the acoustics lab, and he took me on in the lab, and I had an assistantship there. So, although I did my thesis in quantum mechanics, it was essentially, you know, low energy quantum mechanics. I've never worked in particle physics or high energy physics or field theory or those subjects. It's always been in the classical physics/engineering/systems kinds of things — again, the breadth rather than the extreme depth in a particular thing where I'm only one of a dozen people in the world who really understand some subject.
That's yourself rather than the schools, you would say.
I think that's right, because it really goes back to the very earliest approach to science, yes. And during that same period, I was just as interested in meteorology, model aircraft, and that sort of thing. So the teaching then — and I guess I did, while I was a graduate student, I worked for the firm of Bolt, Beranek and Newman, which at that time was acoustical consulting. They've since become a computer company. So I learned some engineering. I made noise measurements near airports, and reducted data and interpreted them in terms of effects on people and so on. So I always had an engineering — in fact, in my high school yearbook, I described myself as going into engineering. And I would argue that even though most engineers consider me a scientist, most scientists consider me an engineer. That is, I've always been involved in things and their management and construction and operation and use, and so on, for which good common sense and understanding of quantitative relationships is important, not necessarily to know everything something. I'm at a meeting today studying anti-particle production and that sort of thing. So it's always been a fairly classical physics and on into engineering and management, rather than intense science.
Yes. Did you support yourself during your student time?
Yes. You see, at RPI, there was not only the full tuition scholarship but also then there was a New York State Scholarship, and I was in the ROTC, and that paid a small subsistence. And then when I was a graduate student, I was working at Bolt, Beranek, and Newman and had a research assistantship. So yes, I would say that virtually the whole time, I did not represent anything more than an occasional drain, when my parents had to supply some money because I didn't have the $500 that particular month to pay the dormitory bill or something like that. It was self-supporting but not by direction, more by accident. I was always, in my high school class, I don't remember how large my high school class was — probably 500 — I was always in the top ten of that, so scholarships were natural things.
Were there any particular teachers, courses, fellow students at MIT that you would point to?
Well, the man that I worked with was also as graduate student, was a Canadian, Keith Hoyt, and I remember him with a great deal of fondness. First of all, when it looked like I wasn't going to get the assistantship, he was the one who sort of pushed the administration to see that I got one, and I think I learned a great deal of sort of professional behavior — I mean, one worked hard, and so on. Dick Bolt headed the laboratory and he was a person I admired. He was a professor at MIT. Bill Allis was my thesis professor. And I would say that there are a number of professors at MIT that I remember, and some professors at RPI, Yes, I would say that all of the professor role models were positive. I didn't think any of them were bad people or dumb people, and so, yes, there were at least half a dozen. The head of the physics department at RPI, G. Howard Carragan. He used to wear shirts of the most ghastly colors, but he was a fine man. He was a metrologist, actually, interested in theory of measurement, and he was really very good, and that was another good influence, too, this notion of physics as an experimental science. I've always been an experimentalist, and I've always, if I've done theory, I've been much more interested in working with data than in doing theory.
Right. Well, maybe you could describe your thesis, and how that fits in both to your own earlier interest and the environment at MIT.
Yes. Let's see, now, we're now talking about typically 1954, is when I picked the topic, and at that time, a subject of some considerable interest in acoustical circles was the so-called anomalous absorption of sound. It was understood in a classical sense, that is, a sound wave propagated through a medium, through each of the compression rarefaction cycles there was work done, and that resulted in an attenuation of the sound, and that was increased with frequency, because obviously you were doing more work per unit time, and therefore using more energy. When people made some experiments, in the thirties, and then no one did anything in the forties, and then picked up all this work after World War II, they noticed that quite apart from the so-called classical mechanisms, that are just easily calculable, there was some additional energy loss. No one understood what it was. It was believed that it had to do with a slow equilibration of vibrational and rotational degrees of freedom, and that energy was feeding in and out of those degrees of freedom, and that's what was causing extra loss. This turned out to be the mechanism. But it was very exciting in those days, because people weren't sure, and so I did — my thesis was a quantum mechanical calculation of — now, in this case, you see, you had to have, well, just, say my thesis was a calculation of the rotational vibrational degrees of freedom and the rate in which they equilibrated, and calculating this so-called anomalous absorption coefficient. And doing that for a variety of temperatures and materials, and getting, assembling the data from the literature, and bringing some sense into it. So it was a nice graduate student thesis. Nothing earth shaking.
Right, but it leads neatly into the interest at MIT.
That's right, because I was in the acoustics laboratory and I was interested in sort of classical kinds of things. Yes, and yet it was topical in a way. In fact, I was very pleased because when I finished up, I was asked to give several invited papers on the subject at the Acoustical Society, which was sort of a Good Housekeeping Seal of Approval.
Yes. Once you were well along in your education and started your dissertation, and all that, what were your expectations for a career? What did you want to do or expect to do?
Well, of course, when I was in high school, maybe even in undergraduate school, I had the notion of winning the Nobel Prize in physics. And it became clear, quite honestly, you know, that I was competent and had a number of virtues, but probably a Nobel Laureate in physics was not likely to be one of them. And so I converted to the professional ethic of simply understanding, teaching, doing a good job and helping where I could. So I would say professionalism replaced a more egocentric notion of — that I'm going to be awarded this great prize; a much more outgoing one of simply doing the right things. Now, as to where I would work, I never shared the enthusiasm of my both my professors and fellow graduate students for the academic life. Most graduate students — well, I shouldn't say that. Many graduate students see that the highest good is to win tenure in a good university and to do research and publish the rest of your life. I never quite saw that. In fact, I don't think I ever particularly did. When I got out, I just naturally interviewed with industry, and I don't think I ever thought about it a lot, except to say that I probably wanted to do something, not replicate myself in more students.
How untypical among your fellow graduate students at MIT, to have that inclination? Or not to have that inclination?
Well, it's hard to say. MIT has always had — obviously, it's an engineering school, though it has a good school of science, but it's more of an engineering place, so therefore being a graduate student at MIT, and obviously having many engineers as friends, just as many as in the physics department, perhaps more so, the notion of engineering and working in industry and going out into the world and doing something was probably the norm of the circle that I was in. It probably wouldn't have been true if I'd been in the physics department at Harvard, for example. So I would say that this was not unreasonable for the school, and so, I think it was — I don't know whether it was usual or unusual for students as a whole, but certainly for MIT it was not unusual.
So you would say that the cohesiveness among graduate students generally at MIT was greater than the cohesiveness between physics students at MIT and Harvard, for example?
Oh yes, that's right. No connection at all. No connection at all. That's right. I remember. I was working for a consulting firm, and this was on an hourly basis, and so therefore, I had a notion of the outside world, what was going on in the acoustics business, and at that time, this was just after the war, the hi-fi business, we were designing all sorts of wonderful loudspeaker enclosures and understanding how they worked. So there was really a lot of interest in the acoustics business and underwater sound and so forth, so acoustics was a much livelier thing. It seemed much more exciting than perhaps one would think of it now compared to the other exciting things going on in physics.
Yes. At that time, how interested were you in, if not science policy, science advising, science management, the way you come into it later? Is that a later thing with you or…
Yes, it's a later thing. It's purely an accident of going to work for the Advanced Research Projects Agency. I fell into it. Before talking about that, though, shouldn't we talk about Stevens a little bit?
When I got out of MIT, I went to Westinghouse without a great deal of thought. I didn't like it. I didn't like nuclear reactors. I didn't like my boss. I didn't like Pittsburgh. I wasn't that enthused about Westinghouse. I wanted to get back to New York.
How did you get there in the first place? Did you just apply?
I just interviewed, and I had three or four job offers, and did all the analysis and –
All industry. It was a toss-up between DuPont in the polymer business and Westinghouse in the reactor business, was the final choice, and the polymer business was getting back to that early chemistry interest, in fact. But I went into the reactor business, learned a lot about computers. You see, this was 1956 now. Whirlwind, the machine at MIT, which was one of the first, was done while I was a graduate student. So I was not a part of computing at MIT in the early fifties, but I got into becoming a computer user and therefore began to learn a great deal about computing in Westinghouse, and that's what I really learned, not neutron physics but computing, although I learned some neutron physics too. So I left, after about two years, and I was — I took essentially the first job I was offered back in the New York area. That's not quite right. I interviewed at Los Alamos, and I interviewed at what's now TRW, at that time it was called Thompson Ramo Woolridge, here in Los Angeles, and decided that I would go back to New York. I was married at this time and my wife preferred that we go back to New York rather than Los Angeles or Los Alamos.
She was also a New Yorker?
She was also a New Yorker. So we went back to New York, and I went to work for Stevens Institute of Technology, not in a full time faculty position, but in a thing that at that time was called the Experiment Towing Tank — it's now called the Davidson Laboratory, and it's a hydrodynamics laboratory. One could probably call it a ship hydrodynamics laboratory but I didn't do much ship hydrodynamics in it, but it was a hydrodynamics laboratory. I did lots of things, and that's where all my interesting publications go back to. There's a little bit in the reactor business but not much because I wasn't there that long.
No. Two years.
There's one or two papers, but nothing of note.
Just to interject, you got your PhD in '56. You were at Westinghouse from '55 to '57, now starting in '57.
Yes. Just in case anyone cares, what happened was, I left MIT in December of '55 and went to work for Westinghouse in January of '56, even though I didn't get my degree until June of '56. So that's why the '55 and'56 is kind of confused there. And so I took this job offer at New York, and was at a university, and I told them that I wanted to have a connection with the physics department, but I did not want a faculty position, so I had a connection with the physics department and I did full time research in this hydrodynamics laboratory. Eventually I was given a charter to form my own group, and so I formed a group called the Fluid Physics Division, and we worked in things like plasma physics and explosive phenomena and oceanography, and we even did a little bit of government work, some things related to mine sweeping, so it was a broad range of hydrodynamics things. We were interested in flow visualization, in boundary layer, attenuation of shallow water ocean waves. We did experiments off of Block Island. It was a wonderful period. It was science that was fun. It was the way science ought to be.
One of the articles from that period, '63.
Yes. And that was just sheer fun, and I by that time had my own group, and we would raise a couple of hundred thousand dollars a year, and we had three or four or five contracts, and I had half a dozen people working with me, and it was just a very good period.
You were fairly independent.
I was quite independent, could work on what I wanted as long as I could get money. At that time in the early sixties, although money was getting harder to get from the government, it was still pretty good as far as research funding was concerned. NASA had been formed and so on. During that time, I started to do some consulting. Oh, I should say about the physics department, I entered with the exalted rank of adjunct lecturer in physics. And eventually became a research associate professor. I'm sorry, research assistant professor and then research associate professor in the physics department, and was in fact as voting member of the faculty by the time I left. And so although I did not have, I was one of about three voting members of the faculty, of people who were associated with Stevens but not as a position in a department, but I was in the physics department. I also taught fluid dynamics in the mechanical engineering department.
So you did teaching.
So, I did teaching, yes. No, I did a lot of teaching. I probably did almost as much teaching as a full time professor. I probably had two courses a term. I taught optics and geophysics and undergraduate physics and modern physics, and did graduate courses in mechanics of deformed bodies out of Sommerfeld and just very very good period, both — I had thesis students. I sat on oral committees. I mean, I was a full and functioning member of the physics department. But I was really paid by the Davidson Laboratory, and just paid a little bit for the teaching that I did. One of the faculty members, a man by the name of Samuel Koslov, asked me, he did a lot of consulting and wanted to know if I wanted to help him. He had more than he could handle. I said yes, and that brought me into the nuclear test detection business. And I found myself interacting with an agency called the Advanced Research Projects Agency, and on a visit to the Advanced Research Projects Agency, they asked me if I wanted a job.
Which was when?
This was 1965, when they asked me, and it was rather interesting because the man who asked me, was interested, wanted to hire me to work for him, and when his boss found out that I was interested, I ended up becoming hired to be the boss of the man who was trying to hire me. Which had some sensitive elements to it. But this is just to say that I entered this Advanced Research Projects Agency now as what's called an office director. I was director for nuclear test detection. I had a little bit of research management experience, obviously, because I had to keep half a dozen people, supported and directed. But now I all of a sudden found myself running a 40 million dollar research program.
How long a connection did you have before you were asked?
About three years or something, '62 to '65, and then I went to work for them in 1966.
What was your reason for saying yes?
Well, I had been at Stevens now for eight years. Stevens is an interesting place. In fact, I'm currently a member of its board of trustees. But Stevens is not a large place, a wealthy place. It doesn't have huge intellectual depth. And after eight years, I really felt that I had exhausted the place. I knew the people there. I knew what it could do. I was having a fine time, but really, I was no longer growing. If I had stayed another twenty years I would be still doing the same things, and many of the people I was with are still there, still doing the same thing, twenty years later. So it was really a question of, it was time to leave, and eight years is an honorable enough time to work for any place. So this very interesting offer, — at this time, 1965, when I was approached, ARPA was only about six years old, so this was a new vibrant agency. It's now of course much older. So it was an exciting agency. Going to Washington, which was where all the money was coming from, and therefore was an opportunity to participate in that management, not as a spender of money but as a decider of what money should be spent on, that's the real difference between in government and out of government, to decide what should happen, and so I did, and I found myself — and of course, I'd been working in the area of nuclear test detection. Now, where that was so interesting and so important in changing what I've done is that it was in the national security business, but it was in the arms control part of the national security business. It was nuclear test detection. Nuclear test detection was not pursued for reasons of the collection of intelligence information, although that's obviously a part of it, but it had to do with the verification of comprehensive test ban treaties, measurements of yield, numbers, cheating scenarios — the whole technical side of the arms control business. So it was arms control and arms control verification, and it therefore brought into play the whole set of technical issues about weapons systems, the policy issues about why one wanted those weapons, but then the control of those weapons, the policies relating to the importance of controlling those weapons, and then overlaying that, the science and technology that enabled you to verify the agreements. So it was, of all the parts of the national security business, it was really one of the most interesting, and that also has now shaped my entire outlook into strategic issues, as well as intelligence issues and science and technology and the weapons system, where I'm currently employed now. So that was a much more interesting part of that agency than most of the other offices.
Yes. To what extent were you able to maintain some relationship with research, as you were used to it? Was it a complete transition from research to management or how was that?
Now, interestingly enough, I managed to add a new technical field. I said, I started with — computers came along when I was a graduate student, but I never did anything about them. I learned about computers at Westinghouse in my brief period there. When I was at Stevens, Stevens had what we would now call a minicomputer that was available, and I used that extensively. I was one of the heavy computer users at Stevens at the time. And then when I went to ARPA, at that time it was called ARPA, it's now called DARPA, but anyway, they were one of the main supports for computer science and artificial intelligence and software research and time sharing and graphics and all that sort of thing. So I have continued to drift in the computing directions, and now probably most of my professional colleagues think I'm a computer scientist, and it largely comes about from the ARPA and continuing associations at Xerox and now at Northrop, with the computing business. So I probably have added, in fact, not doing research, but rather than just chew over the same things that I entered with, I have continued to broaden, and now I would argue computer hardware and software, I can hold up my end of the conversation with most professionals in that business. But not doing research. On the other hand, I have also added something else, which is not taught in science, which, where we're sitting (Rand is a preeminent example), and that's the business called analysis, that is, policy analysis or operations analysis, but there's a sort of enterprise which is, thinking quantitatively about things for the reasons of making decisions or judgments, and that's another thing that I have therefore learned along the way. So I would say, in the computing and analysis, I've continued to grow while physics is receding.
Those things are related to projects that were going on in ARPA when you came there, is that right?
I guess Vela was the one that —
Vela was the nuclear testing, that's right. And probably I had more influence in the final analysis on the computer end of ARPA than I did the nuclear test detection end, because most of the important thoughts and notions in nuclear test detection were already on the table when I came to ARPA. We did some serious thinking about on-site inspection. We constructed seismic arrays that were — well, even then, my predecessor really got that whole business started. So I probably, in the nuclear test detection business, competently continued the work of my predecessors, but the things that I probably injected into the business were much more related to the computer area and computer networking and management of the agency, in a broader sense.
Of course you had an unpleasant experience too with that United Nations lecture and —
And that kind of thing, which might also have led you in another direction, I don't know.
No. No. No, I think that I have — you simply get into trouble once in a while, when you're thinking of the business, and when we were at the meeting in Geneva, we had this difficult set of charges about who had changed our report and all that sort of thing. Yes. As a matter of fact, when I was in the FCC, I had another problem like that, namely, having to do with the way we made a decision on the matter of stereo broadcasting using the AM radio band, and I got wrapped around the axle there too. It happens. I don't consider that to be — that was just part of it. That was sort of fun, in hindsight. If people aren't throwing rocks at you, I guess they don't notice you. There's a certain amount of it. You can be too controversial. But that was all right.
Yes. So that was, having nuclear detection in the Department of Defense perhaps could be interpreted as controversial too, of course.
Quite. It turns out that what most people didn't realize was that ARPA, this agency, was housed in the Department of Defense, but really was quite objective and neutral. Now, people always say, the person who pays the bills is the one who's really in charge, and I suppose it is certainly true that you lean more to the people who are paying the bills than the people who aren't paying the bills. There was merit in the charge, actually. But what can one do? You're a creature of your surroundings, I mean, you know, chemicals flow in, information flows in, and that's what you are, and if it's in the Defense Department, it's in the Defense Department. Yes, I was never ashamed of it.
No. So then there was the computer side of it that you got more into.
Was that through other projects in ARPA that were going on, or was that more something you instigated?
Well, it was a combination of two things. It started out, like so many of these things, with an accident, and that is, when my office in ARPA was in a particular location in the Pentagon, and the computer science office as across the way, and so therefore I ran into them more than I ran into let's say the counter-insurgency people, who were in the other wing. So I ended up talking to the computer people more simply because they were across the hall. And then I had a few — for example, at the beginning of the ARPA net, this was late one afternoon, and the head of that office came by and said, "We've finally got our contractors to agree to form a computer network," and another time I was called in because ARPA was going to make a major investment in a parallel processor, Illiac IV, so I was around at the beginning, and it's one thing to join something when the ideas are well in place. It's another thing to be there at the beginning became the deputy director of ARPA, the way the director and I divided it up was, he did some things and I did some things, and I drew the science part of ARPA to be in charge of. So therefore I was in a position, I was not only interested but I was in a position to influence, i.e. assist, what they were doing. I have always viewed my role as a manager as assisting the people who are doing things, rather than imposing my will and directions on them. Now, inevitably you have to make choices and decisions. Of course, that's what you're hired to do. But my whole philosophy has been, if you're the experts and I'm the manager, then I will try to find out what's the best ideas among you, and assist those people who have the best ideas. So it's always been one of assisting rather than directing.
Yes, but there's some element of hands-on there.
Yes. Yes, of course.
OK. You also talked about systems analysis, and that's a third thread, I guess. They're all interconnected of course. I don't know if that came out of ARPA too.
Yes, that came out of ARPA, a different part of ARPA. We had a — ARPA did a lot of science, science and engineering and technology. It also did studies. The studies weren't very expensive, because you can buy a lot of study for a few hundred thousand dollars, whereas it may take you twenty million dollars to build a new telescope or something. So I always paid a lot of attention to the studies, because the studies guided you to decide how you should spend the other money. The studies were like the trigger. So if a study said, the problem is in vulnerability of aircraft instead of the standoff weaponry, then that was sort of important, because then you would spend ten million dollars on this subject instead of that subject. So I always paid a lot of attention to the study work, and I did, not because they were intrinsically interesting, although they were, but simply because they enable you to make better decisions about alternatives. Inevitably then you learn to — you meet the people who do the studies. You now have to look in detail at the methodology. You have to be concerned about the data bases. You have to understand what the military requirements are. And so the studies are the mechanism — they're rather like physics, actually. They're the thing that enables you to look at the deeper going, less visible mechanisms, so you can decide, these things are happening here and those things are happening there, but it's the studies that enable you to show what the connections are. If you can understand that process, then you can find out if we want to make more of this or less of that, or defend ourselves here, or whatever. We'll have a better sense of it. So the studies were, one went into studies for the same reason that one went into physics instead of chemistry, because it was getting at the essentials of the thing, more empirical than theoretical, but that's all right, physics is an empirical science too. So there's really a very close relationship between studies, systems analysis, and physics.
Yes. So what was the place of the studies within ARPA?
Oh, they were done in, they were spread over — for example, in the counter-insurgency office, we did a lot of studies about insurgency, counter-insurgency, why populations felt the way they did, how they reacted to internal and external events. In the strategic business, it was defense, offense, technology, detection technology and so on. So all of the studies were like a few percent of every office, and I tended to watch those few percent more than the rest of it, because if you got those right, lots of people could spend five or ten million dollars well, but it was a question of what. So the focus was always on what to do and why, and the why is really very important. I mean, you can't figure out the what until you can figure out the why. So the entire orientation here has been, very much like science, you see, to try to understand what the problem is, try to understand what the mechanisms are, pick some reasonable mechanism for getting the job done, like picking some reasonable tool to work with in the machine shop, and so that's how the whole thing — and it's a very scientific approach in that sense, although it recognizes the uncertainties and the human factors and the politics. That's what the arms control business is.
Right. ARPA of course doesn't follow the projects all the way towards their development. That's part of the nature of ARPA too, right?
It's 6-1 and 6-2, I think it's called in the book there.
Yes. Although it's changed a lot since those books. Those — the history cut off there like 1974 when I left ARPA. That's now 13 years old. And in that time period, ARPA has moved much more into the feasibility demonstrations, operational test beds, experimental prototypes, 6.3 in DOD terms, so ARPA has gotten much more into the guns and the flying and the floating part of the business, than when I was there. And the budget has gone up correspondingly because those are lot more expensive enterprises. It's gone from about 200 million to 800 million?
Oh it has, now? It's gone down more than up during the period covered by the study of course.
Yes, it was going down then. That was the minimum. After I left the budget increased.
OK, that period isn't covered. I don't know what that proves. But yes, it was difficult economic times, and that colored your time there.
The Vietnam period, yes.
We might get to that when we talk about Jason. I think in general terms it's covered in Huff's book anyway, Huff and another person I think too who wrote it, correct? So that the main transition then, within ARPA for you, happened when you became a director?
Well, I was a deputy director. I was deputy director for one year, and then into the front office, and that's when I broadened my view, to have responsibility or purview of the entire set of activities. That was the big change. And I was in the front office, you see, for seven years.
Yes, and then Rechtin was the DDR and E.
So that you were actually director for the longest period of all, until then, anyway.
OK, so I left because salaries had been stationary for several years. That was a period of heavy inflation in the United States, so one was getting poorer year by year. There were several children. Expenses were going up, and it simply financially was not viable. And even getting a promotion wouldn't have helped, because even the next level up or two, those salaries were all roughly the same. So I had to go out, and I went to Xerox.
Yes. Was that a problem for that kind of position generally, do you think, at the time, that the salaries weren't high enough?
Yes. Yes, I think there was probably a flow of people out. But it's hard to say. So I went to Xerox. This was actually a continuation of the computer theme, because Xerox, while it's a copier company, I joined the part of it that was doing electronics systems. I was also interested to know what was going on in Xerox because in the early seventies, the vice president for research at Xerox named Jack Goldman had a great deal of foresight and recognized that the analog copier business had some limits, and the world was going digital, and that the whole office systems and copiers and word processors were going to be office systems. So he sets up this marvelous laboratory called the Palo Alto Research Center (PARC), and a number of ARPA principal investigators vanished into this black box, and I was always curious to know what was going in there. So I was quite interested to join Xerox and see what was so interesting there.
What was the pre-history for your joining? Did you apply in the normal way or did you have some connections before? Were you asked?
I was asked. When people knew I was available, they approached me. I had talked to a couple of other people, and that seemed to be the best opportunity. So —
It was clearly Xerox that you wanted to go to.
That's right. I must have looked at a couple of other people, and I'm having trouble remembering. I guess, I forget what my job options were, but anyway it was Xerox and perhaps one or two other companies. As a matter of fact, it was someone who had been in ARPA. In other words, it was all the ARPA people, even the ARPA managers, not just the ARPA contractors but the ARPA managers. It had become an ARPA-like place. It was logical for me to join my ARPA colleagues, some of them, at Xerox. But I stayed there for only two years. Xerox wasn't serious about the office systems, the digital business. I still think they're not serious. The copier business makes too much money. And I decided that I was not going to waste my time with them. If they weren't serious I wasn't going to stay. I feel it's very important to work for a group that's serious.
Yes. Was that a flow both in and out? I mean, did you flow out with other people too?
After me, for several years, there was a flow of people like myself out of the research center into Xerox and then out of Xerox, disillusioned, hoping Xerox would do something with their technology. So yes, I guess that's right. I was probably one of the first. I probably had less patience than everyone else. So that's a good point. I hadn't really recognized that, but that's true.
How large a part of Xerox was it anyway?
Well, Xerox PARC probably had 400 people working for it. I was not at PARC. I was brought in to set up a new organization, which would hire people away from Xerox PARC and get them to work on designing digital office system products for Xerox. So I was to be part of a technology transfer. In fact, I ought to stop and say a word about technology transfer. Technology transfer, is used in at least two different ways. One has to do with transfer of technology from a research environment into a production environment. The word is also used in policy circles to mean the transfer of technology from country X to country Y. I'm talking about the former, not the latter, although there are some similarities between the two processes. Along the way, I have developed another skill or area of expertise, I guess I'd call it acting as an agent to move technology from one organization to another. ARPA has to transfer its ideas to the military services .In Xerox, I was concerned with the transfer of technology from Xerox PARC into the product line. In Northrop, which we haven't gotten to yet but I headed up the research lab and my current title of vice president for technology has to do with moving technology into Northrop from other places, and moving technology from Northrop's corporate laboratory into the rest of the enterprise. So I have along the way, in at least three different job settings, been involved in the process of moving technology to accomplish some practical purpose. So I had the ARPA experience, now the Xerox experience, and those two were really quite important in terms of shaping my approach to life.
It seems to me, from what I've read in that book there, that transfer of technology, in this case to the services I suppose, mostly, became more and more conscious — I mean, beginning with Rechtin (?) perhaps and then with you —
— absolutely —
— partly for economical reasons, I'm sure, but maybe for other reasons as well. I don't know what you want to say about that, if that was transfer becoming —
Yes. It wasn't just for economic reasons. It was for other reasons. You're absolutely right. You've read this study correctly. It was Rechtin who put the focus on it, and it was Rechtin and his boss Johnny Foster who said — ARPA 's been in existence since 1959, or '58, and it's now 1967, and that's eight years, and nothing's coming out of it. Just the same way as the people at Xerox said, Jack Goldman set up Xerox PARC in 1971, it's now 1975, and we're not getting anything out of it. Industry is not as patient as government. And so, in both cases, I responded very positively to that notion that science, technology, research, ideas are not something to be kept and elaborated and built up and so on, but are something to be moved out the door and given to someone else so that they also can do something. And so while I'm perfectly happy to see a high quality institution operate for a hundred years doing its research, I'm not against research laboratories, but I see very much that, if an idea is good and exciting to me, it ought to be good and exciting to you. I want it because of the way it completes an intellectual structure. But I'm perfectly happy if you take it and build a product around it, and that product is then used by other people.
Yes. But in the case of ARPA, it became an added responsibility, maybe became half the responsibility, not only to develop an idea but to transfer it out and make it usable.
Yes. But that's just to say that an ARPA that was just good at creating ideas really wasn't doing anything for the Defense Department. It just got bigger and bigger and had more and more ideas. So the whole notion of utility, that ideas should be moved so that other people could use them, and that's what it makes it all worthwhile. That's why the taxpayer is paying the bill, or the shareholders, in terms of dividends foregone to be plowed into the business. So I thought that in fact what Rechtin did and Foster did was simply to illuminate a neglected part of ARPA. Now, one could argue, everyone had been focused on getting it started, and that was quite right, but when they came along, it was also quite right to say, now that we've given you lots of time to think about these things, let's see you move some of these ideas on. So it made sense. It wasn't just a political necessity or an economic necessity, although it was that too, but it really made sense, which is why I embraced it so wholeheartedly.
Yes. How new was this kind of thinking generally?
Brand new. The director before Rechtin, Charlie Hertzfeld, very good guy, recognized he had to do it, but he perhaps gave it, a weighting of one on a scale of ten, whereas Rechtin gave it a rating of six on a scale of ten. That was all the difference. It was a matter of focus and incentive and direction and guidance and so on.
Yes. I wasn't thinking just within ARPA but generally speaking, was that?
Oh, in the United States in general?
Oh, it wasn't a new idea. There's a fair literature on how to manage corporate R and D, how to manage corporate labs, how to do innovation. We probably, in the late sixties, didn't quite have the intense preoccupation with innovation, but we'd been thinking about it. Remember the brain drain from Europe, particularly the UK, was a major issue around at the time. So I think people have always been pretty conscious of innovation. It's just that we were doing so well in the United States in those days that we probably weren't quite so self-conscious about it and quite so introspective about it. Now, of course, we're totally preoccupied with it because of international competition and trade deficits and the like, so we're — so it was probably a relatively new idea at that time. Good point.
I'm not well read in management, it's one of my weak places, that's why I'm asking you.
What I meant by good point was not that you made it but that you caused me to focus on it. Now, there was one other thing that was going on at that time, and that was the Vietnam War, and everybody, particularly in the early stages before the intense political public dissatisfaction with the war, the thinking in 1966 and '67 and'68, in the R and D community, within the Defense Department, was, we want to help. Some people are fighting, some people are planning, some people are doing this, some people are doing that, — we want to bring our ideas to it. We want to save people, introduce night sights, whatever, the Barrier. When we talk about Jason we'll talk about the so-called McNamara Barrier. And so we were quite anxious to transfer technology into that particular conflict. So that was another driver — let's get some things out of these laboratories, let's get them into Vietnam, let's try them out. If they're any good, maybe they can help. Maybe they can find people digging tunnels or whatever. Yes.
In part it was an element of saving ARPA I suppose too, when it started, when Rechtin took over.
Rechtin definitely went into it as a defensive mode. That's true, yes. But it always made a lot of sense to me. It made more sense than its origins perhaps would have had it deserve.
Yes, OK, well, that's partly covered in the book, and we will come back to aspects of this when we talk about Jason, so maybe we should go to Xerox again. That's how we picked up on it anyway.
Yes. Anyway, Xerox wasn't serious, and I left after two years, and I set up in consulting. And that may not even show in the record, but there was a period of ten months when I was a consultant.
Yes, and you jumped right from '76 when you quite the Xerox Corporation, and then '77 when you joined —
— yes, well, it turns out there was a period of July of '76 to May of '77, when I came to Rand, and so it sort of gets lost in the enumeration of years. But it is the reason why I came to Rand. Now, I had said when I was in ARPA that I was always interested in studies, and Rand was one of the major institutions doing those studies. So was Hudson Laboratory and Stanford Research Institute and Lincoln Laboratories. There were a number of laboratories. Rand was one of them, and Rand was, I thought, the best. So I always had a warm feeling. When I went to consulting, Rand was a client. And then when a position opened up here, the vice president for national security affairs went to Washington.
"Here" being Rand, now.
He was at Rand and I was consulting. So I took his job. I switched over from a consultant to a full time employee.
So you were a free-lance consultant.
I was a free-lance consultant for like ten months or so, something like that. It's not listed because it doesn't quite fit. So I was here for about two years, and it was one of the most valuable two years of my life. It was really my graduate education in analysis. Now, in fact, I'm manager of the place, I'm critiquing the work, I'm guiding the work, I'm critiquing the briefings, I'm selling the studies, and so on. I'm participating in the studies. But I'm also learning. And I consider Rand to be probably the premier organization in the United States for doing policy analysis, in terms of capability, breadth, objectivity. Not perfect, but the best. And while I learned a lot it was my two year Masters's degree, like going to Harvard Business School for an MBA. Well, I did my two year thing in policy analysis here, although I was working as the manager of the place. And so, that's how I came here, because I was interested in policy analysis.
So it was clear, you got out of it what you wanted to get out of it.
That's right. And my family wanted to go back to Washington. My family didn't like the West Coast. So we moved back to Washington. And I took therefore the first job that was offered back in Washington, and it was in government, and it was in a policy position. It was the Chief Scientist of the Federal Communications Commission. And that's where I was for four years.
How different was that from Rand?
Well, of course the quality of people was much lower. But it was not unlike Rand, but it was rather more operational, because we were involved in domestic spectrum allocation with preparing for international spectrum management meetings, and handling agenda items and having a small laboratory, so it was not as intellectually exciting. Policy always seems great when you stand back and analyze and study, but when you're in a policy making position or a policy implementing position, it's a much messier sort of thing. But it was a policy job.
Yes, right, but you've maintained that kind of one foot in theory and another in practice in policy as you had in the old days in physics.
That's right. Exactly. And of course, many ideas, you see, many of the things that are technology and very advanced in ARPA in 1974 were in fact red hot ideas to be put into the civilian market in 1980, and therefore needed an understanding within the Commission of these things. So I found myself doing things like packet radio and spread spectrum and computer networks, things that I had been doing ten years before in ARPA, but now they'd gotten out of the Defense Department and into the civilian world, and even some things that Xerox was doing, I found myself involved in now, on the Commission side, and of course I knew all the Xerox people from having been there.
Right, so you went with the transfer, so to speak.
And in fact, one other theme in all of this is the Defense-non-Defense flavor in my life. You see, when I was at Stevens, I was some Defense and some non-Defense. When I was in ARPA —
You're talking about consulting being the Defense?
Yes, consulting. Also some of our contracts were in mine sweeping and things like that, so I had a concern with military subjects, but with just scientific subjects.
Even without the consulting.
Right, and we were working at Aberdeen Proving Grounds. Some of the explosives work was done at Army explosive ranges. ARPA was all obviously national security, but the technology was very broad and general. Xerox was non-Defense, so I had no conflict of interest, because it wasn't as if I'd left a Defense business to go with a Defense contractor. Then Rand was Defense, but the Federal Communications Commission was non-Defense, and now I'm with Northrop which is in the Defense business, so I've always moved back and forth between Defense and non-Defense, and technology, not specific, and then very specific military subjects, and so there's been quite a mixing of the two sides. It again goes back to that general broad bandwidth type of thing, but not necessarily great depth.
Is that typical or untypical, would you say?
It is not typical in a career. Most people are specialists. And most people who are industrial managers, tend to be industrial managers in industry X. One might be a manager in the aircraft industry or a manager in the chemical industry. You won't see a manager in the chemical industry work for a communications company, whereas I could go to either one, you see. Now, I think I have probably been a broader person across a wider range of technologies and a wider range of applications of those technologies — yes, and that's less usual.
Yes. Well, if you look at ARPA directors, for example, I would think that many of them came from — well, non-government, anyway, I guess most of them.
Most of them. That's right. Yes.
Well, I don't have a listing of their careers generally speaking, but that might be actually a place where you can find people, you know, going from one thing to the other.
Yes. There are people from universities going to ARPA. Yes. The set of ARPA directors I think are an interesting set of people, in the sense that they are different from most other career patterns. Both where they came from and what they did later.
Yes. So that's a historical study. And then, from the FCC you went to Northrop, and you went back West again, at least in part.
Northrop. That's right. By that time, what had happened, since you asked a little bit about history, is, while I was in the FCC, I was divorced. So my wife at the time who didn't like it is not the wife who's with me now, who does. So that at least explains in part why I like the West now and why I wasn't happy here in '77 through '79.
So family has played a role.
Yes. If family isn't happy, you're hard-pressed to exist some place or to continue some place.
Because you have an impressive list. As we've talked about, you've gone from place to place, and of course, — so the family's part of that.
Yes. Family influences and desires have been strong factors.
Of course there are other professional activities too — consulting, other things. I don't know if there are other things than your positions as such that you would like to talk about.
Well, the list is there. I'm on a university board of trustees. I do a lot of government committee work, listed there, that largely relates back to where I've been. I suppose one of the activities I've become interested in, in fact it is illustrative of this general bandwidth thing, is that I've been involved in planning the next scientific examination of the Shroud of Turin. So I now find myself associated with art conservators, historians, archeologists, and as we think about what should do — what are the hypotheses, what should be done, how not to harm the Shroud. Very interesting subject.
Yes. How did that come about?
Oh, that was an accident. I read a book on the subject, and became interested in it, and found that the architect who was making some renovations on my house was interested in Biblical archeology and was running a meeting on the Shroud of Turin, and invited me to meet all the scientists who had been involved in it. I told them what my hypotheses were and they asked me to join their group, and now I'm active in that circle.
Some surprising things come up. OK, maybe we should turn to Jason then.
Let's turn to Jason. Well, let's just say a little bit about Northrop. We didn't say much about Northrop. It's in the same pattern, so it will be easy to treat. When I was leaving the FCC, I'd been there for four years, there had been an election. The party changed power. The chairmanship changed and the new chairman wanted his own staff, so I was on the way out, so that was easy enough to understand. I came out here to Northrop, headed up their corporate laboratory. I describe Northrop's corporate laboratory as Northrop's ARPA. You see. Look around, do everything, transfer it into the business, advanced, far out and so on. So in that sense, it's back to ARPA. It's not as large as ARPA. We have a 20 million dollar budget. Then after about two years, I was promoted, and my current position is vice president for technology, which has the research lab reporting to me, but now I have cognizance over all of the research and technology going on at Northrop, and depending on what you count, that's easily l00 million or 200 million or 300 million dollar set of activities, which I don't direct, obviously — and I worry about scientific personnel, and I worry about, are we doing the right things, and I worry about, are we laying the right foundation for the future, and are we transferring technology, and are other divisions equal and so on. So I'm really in a national security domain again, worrying about warfare in the future, military requirements, worrying about all the things I worried about in ARPA, and it's just like ARPA. So to that extent, it's a very direct linear continuation of earlier steps, although not like the FCC and not like Xerox.
OK, so you found —
Yes, let's turn to Jason. That will bring us back in time, of course, but let us bring it back as far as we can. What was your first experience of and reaction to Jason?
Oh, I remember it well. It was a negative one. I had just gone to ARPA. I was the head of the Nuclear Test Detection Office at the time. Part of my program dealt with on-site inspection. That is, the inspection of a suspected nuclear test, illegal nuclear test, and how would you know, how would you locate it, how would you decide it really was an illegal nuclear test. And I was visited by the head of the Jasons one day.
By his initiative ?
On his initiative.
So this was in 196 —?
This was in 1967.
It was Goldberger, probably. [Note: it may have been 1967, in fact, the more I think about it, it must have been 1967].
No, it was Hal Lewis. And he came to my office, and he told me that he was going to help me that summer. He and the Jasons were going to do X and Y and Z, and in order to do that they wanted access to places A, B and C, and that's what they were going to do. And I told him, "Thank you very much, but I don't want your help." Now, the reason was, not that I have any disrespect for the intellectual capacity of the Jasons, but I had just come aboard ARPA, and I had received access to certain intelligence information, and it became clear to me immediately, and I was one of the first people in my position to have access to this kind of information, because it was very new, and I was one of the first directors to be cleared for that, and it was clear that what we were doing in this on-site inspection program was absolute nonsense. Just no relation to anything useful. So I was in the process of stopping what was going on, and I was going to re-start a program which would be, as we would now call it, a black program, and so I didn't want the Jasons, because the Jasons were not cleared, and I didn't want them elaborating and working in an area that I intended to change, to diminish or even eliminate, and get some other things started. [Noted: added in 2004: the “intelligence information” alluded to was the fact of, and the capabilities of photography from space].
It became clear after an hour or so that Hal Lewis and I were simply not communicating, that he was going to do X, Y and Z, and I was not going to have him do X, Y and Z. It was a standoff, and he left. That was really the first time that I had interacted with them. A week or two or three later, he was back in the office with the vice president of IDA, Institute for Defense Analyses, Gordon McDonald, the one who was in charge of the Jasons, the Jason Division, who was himself a Jason. The Jasons would only have Jasons manage them.
Yes. Well, I think he formally was out of Jason by then. He was vice president of...
Yes, but the point is, they would only accept their kind as being of stature and —
Yes, it was Townes, Brueckner and McDonald, I think.
Yes. I think after that they got away from the idea that the manager of Jason should be an ex-Jason, but Gordon was still in that. Now, Gordon, it turns out, had access to this information that I had received, and I was very, very pleased that he supported me. As a matter of fact, it turned out that Hal Lewis did too, but since we didn't know what it was that we were responding to, we were not able to communicate, so I couldn't tell him why he couldn't have access, and he couldn't tell me that he had access, because that's not the right way to do it. It was a marvelous example of secrecy and how it absolutely complicates communications. So the three of us got together, and we now agreed that I was right, that is, that what they were going to do didn't make any sense.
You had by that time found out that the other person was allowed to know this.
Yes. But just Hal Lewis, not the other Jasons, you see. The conclusion that that was not a useful subject for Jason was a correct conclusion, but the point was that we had to get to it in an uncomfortable way, you see.
So Jason never did that.
So Jason never did that. They went off and did other things. So that was my first introduction to the Jasons, and it is — I'm sure in your interviews, you probably had the word "arrogant" associated with Jasons.
Oh yes, a lot of times, even by themselves. Prima donnas. There's a lot of it.
Yes. And you see, it's interesting that that particular vignette is in fact of the pattern of arrogance. That is, "We will do X." "No, you will not do X. " Except we came to equilibrium because in fact we were all intelligent people, so it worked out all right, but with more arrogance in the beginning, and having nothing to do with the Jasons, this stupidity about classification that enabled well-meaning people not to communicate on what the subject was.
Yes. We'll get back to how this self-assuredness or whatever we should call it affects the impact of the group, Jason, and that kind of thing. First I would like to ask you the general question of, what has been the nature and degree of your personal interest and involvement in Jason? How big has it been? Of what nature has it been generally?
Well, we had — in the ARPA period — the reason is that as you got from this Lee Huff study, one of the things it says there is that I probably paid more attention to the internal management and workings of ARPA than any other director, and the Jasons were one of those things I had to worry about. And I had the feeling, as this little story I gave you was 1967, but as the years went by, and particularly when I became deputy director in '67, and by '68 and '69 it was clear I had responsibility for, among other things, the Jasons. And there were some things wrong with the Jasons, and so, what I tried to do through the late sixties and early seventies was to fix those things. They largely had to do with the fact that it had become a self-sustaining, self-appointed group. They pick their members. And so they had in 1969 the same members they had in 1959.
Essentially, yes, that's true.
And I felt that there should be a flow-through, because there were people who were no longer interested, weren't coming to meetings. A smaller number of them were doing all the work. They were really theoretical physicists. They didn't have any computer scientists. They didn't have any materials scientists. So they were a group of people who were aging at the rate of one year per year, who weren't bringing new people in, who didn't necessarily represent the set of disciplines that were perhaps relevant. Remember, your computers and command and control problems and issues of that sort were just major issues that were happening in the sixties, and they were a bunch of theoretical physicists, circa 1957. So I was concerned about that, and I wanted to make some changes in fixed terms of appointment, review of the effectiveness of members, the bringing in of new members, having a hand in the choice of subjects, and making them "more relevant," less self-directed, self-appointed.
Which fitted very neatly into your general management position within ARPA generally.
That's right, and one was trying to use resources in the best possible way, not leaving to chance that this group would do so. So starting with that first 1967 incident and then moving into these management changes I was trying to make, I probably was always seen as an enemy of the Jasons, a negative factor, or a problem or a pain in the neck or however I was viewed.
And perhaps vice versa.
That's right. Well, yes, to some extent.
OK. Well, it's a complicated relationship here, it seems to me, between ARPA, IDA and Jason, that whole set-up. Maybe we should talk a little bit about that before we go to the specifics of it. Just to talk about who was responsible for what within that constellation, that's my general question.
OK. Well, one can simplify the question by moving IDA out of the picture. IDA had practically no influence on Jason. I'll return to that topic. But it's not a three body problem, it's a two body problem.
OK, in the first approximation.
Yes, in the first approximation, it's a two body problem. And I would say in the early days of Jason, from the time of formation through about 1968, it really was only a one body problem, it was the Jasons. They picked their members. They decided what to do. And they told us about it. In fact, I think we used to have a little saying that the Jasons really only talk to God, but they'd be willing to talk to the Secretary of Defense, that was the best they would do. And ARPA was simply supposed to pay the bills. And IDA had an even more menial task. IDA simply had to get the reports published and handle the clearances and handle the checks and see that there was a lunch served at the meetings and that sort of thing.
Yes, coffee and doughnuts.
That's right. And so that's really the way it was, ARPA paid the bills, and they didn't expect that ARPA would tell them what to do or ask them what to do, and even if ARPA did, they weren't going to pay any attention. They were smarter than we.
Not much. They largely felt they talked to DDR and E, to whom the director of ARPA reported, and the Secretary of Defense. They largely felt that they worked for national authorities, for which the Secretary of Defense and DDR and E was the agent, and ARPA was simply paymaster. ARPA was simply to pay the bills. And of course, if they were used effectively that would be a perfectly reasonable thing to do. What became clear to me was that the Secretary of Defense wasn't paying any attention. Well, if the President isn't paying any attention and the Secretary of Defense isn't paying any attention and the DDR and E isn't paying any attention, then in fact my attitude was, well, I guess ARPA ought to pay attention. We're paying for them and we might as well use them. So really, they ended up working for ARPA by default. That's not the way it was planned.
Was the communication just as bad when it started, or was that something that had deteriorated when you came in? Both with DOD and with the DDR and E and ARPA?
I don't know. You'd have to probe those earlier days. When I came in, in 1966, the prevailing view in ARPA was, we pay them, they're very bright, in fact I think one, either Hertzfeld or Frosch, said, "Well, they only cost us about $400,000 a year, and if you get one good idea from them, it's worth $400,000 a year, " which is not a bad investment philosophy. I tried to manage the fund a little bit better. I tried to get two or three good ideas for my $400,000.
Well, we've talked about the first approximation of the three body problem, essentially a two body problem. Now, the first mention I think of you that I've been able to find in the minutes of the Jason steering committee, which is 1970, I don't have a full set, but there "Lukasik expressed his interest in the internal management of Jason and suggested a stronger coupling of Jason to IDA." I don't know if that rings a bell at all. Of course it is taken out of context.
No, it does. Let me tell you what's behind that remark, because I do remember it or the ideas behind it. At that time, I was deputy director and obviously trying to manage this thing. Strictly speaking, ARPA gave the money to IDA and then IDA ran the Jasons. That is, the Jasons were IDA consultants and held their clearances, and the Jason reports came out as the IDA Jason Division. And so when I was unhappy about the aging of the Jasons, the discipline mix, the choice of topics and so on, I went to IDA. I went through channels, as it were. They're the people we gave the money to, they're the people responsible for it. And what the IDA management essentially said is, "Oh, we keep hands off the Jasons. We just arrange for the lunches and pay the bills." So when I was saying a larger IDA management — of course, what I meant was, "IDA, you ought to manage these people." So I was essentially asking the contractor to manage the group that on paper they were managing. And they said, no, they didn't want to. And so that led to a bit of conflict, because I clearly had IDA between, as we put it, a rock and a hard place. The Jasons didn't want to be managed by IDA, and I was leaning on IDA to manage them, and the IDA management was caught in the crossfire between the two of us.
So it became a two body problem by default, in a way.
Yes, IDA put it back to me.
It's interesting, the heading for this item here is "ARPA Review" and the first sentence of it reads, "E. Rechtin discussed the just concluded ARPA review of Jason." What was that?
That was presumably my review. I'm less certain of that, but I believe that, going back to this notion of me as the manager type, which I was, and so, I had probably — frankly, I forget whether I suggested to Rechtin that I look into the management of Jason, or he suggested it to me. I don't really remember where the initiative lay. But in any event, I remember that that was that 1969 - 1970 period, when I went over questions of, how long had they been members? How actively do they participate? Do you have any discipline, X, Y or Z? Why aren't you working on subject X? Can you help us on subject Y?" And so Rechtin must have reported to them my findings. I'm sure they were my findings. It was my analysis of it that must be reference there.
Is that available, do you think, or was that classified?
Well, it wouldn't be classified. Over the years, I'm a great squirrel, I keep papers, and when I get tired of lugging them around I throw them out, and I will be willing to guess that I've probably thrown those out, but there's perhaps a 10 percent chance the papers have survived. I'd have to go through all my papers, including some that I haven't unpacked in six or seven years. I'd be perfectly willing if I ever found them to put them in the record, because I know what will be in them. It will be membership lists, meeting dates, analysis of fields, discussion of subjects. It would be the kind of thing I've talked about, because I know what I was doing at the time.
Yes. There are some other indications here too of your involvement with Jason. In December 3, 1971, for example, during the fall meeting, you gave a talk on underground test monitoring, which I guess served as a briefing in a sense.
Must have, yes. I don't remember.
And then when Jason transferred to SRI in 1973, you held a talk on the background and purposes of the study group, which I suppose is a general evaluation of Jason. I don't know if that is available. That would be interesting.
As I say, I would just have go to through a lot of papers. If I ever find them I would of course be willing to turn them over to you.
OK, great. It might require a lot of work to — systematically —
— yes, I'd have to find them. Now, what happened, I notice you're holding a piece of paper that has an SRI cover on it. There's more of a story to that. In fact (crosstalk ) — yes, because it's all part of the same thing. But the people that the Jasons felt that they were working for, the President, the National Security Council, the Secretary of Defense, the DDR and E, really didn't care about them. That hurt their feelings.
Do you have any examples of that, incidents that showed that?
Yes, well, I'll tell you one thing that really turned a lot of people off. Remember the 1968 ABM debate?
The Secretary of Defense and the President, for whatever reasons, decided we were going to deploy a thin ABM system.
Yes. The Sentinel?
Sentinel, I guess is what it was called.
Yes, that was when McNamara left just after that, yes.
Several Jasons, who after all were professors, were consultants, had their own independent writing and lecturing, and were Jasons, adopted anti-ABM positions. It wouldn't work, it was a bad idea, it was just destabilizing the strategic balance — the same arguments we hear about SDI today, as a matter of fact. A number of Jasons — two or three or five, but of the bigger names — came out against the ABM. Well, Johnny Foster, who was the DDR and E at the time, and after all, even before that, he hadn't really used these people or consulted them, and now in fact they were simply making his life more difficult — in fact, it's not unlike the most recent what was then called USDA, Don Hicks, who said about the SDI opponents (under Secretary of Defense Acquisition), "If you're not going to support the President's SDI program, then I'm going to see that you don't get any Defense Department money. Why should I support you if you're making my life harder? "It's the same thing twenty years later. So Foster really didn't care for that, thought they were a pain in the neck. They were technically wrong, they were politically obnoxious, they were arrogant — not that all of us in DDR and E didn't have any of the same characteristics, but the fact of the matter is I was simply trying to get the country, the taxpayers, the Defense Department's money's worth from a very bright group of people whom nobody really much cared about. And that's what it amounted to. And that being the case, one should do these management things — the right mix of people, let's talk about the subjects, let's see who comes to the meetings, let's make sure that it's the most important things, let's get you connected to the right people, agencies, and programs. There were some positives. Let's make sure you have access to sensitive classified information, if that's what you need.
There's a problem with that, they didn't have all the —
I think they always did fairly well, because they had so many connections, they generally knew what was going on, but the fact of the matter is that we helped them.
Yes. Were you under any pressure by Foster to change things?
No, his attitude was, "If you can use them, fix them, Steve, if you can't, do your best to ignore them." He had more important problems.
So he didn't go out of the way to try to dissolve them (crosstalk) or, oh yes, right.
He had some very good friends who were Jason members, but there were other people who were taking anti-Defense Department positions, so he had a mixture of positive, strong positive and strong negative reactions.
But it was not strong one way.
No. That was a nitty gritty problem for me to work. So here we had the Jasons that had to be used, could be used, could be really quite useful. For the record, let me say, by used I mean, helping the Defense Department, not used in any perjorative sense. But they needed some management changes to make them more useful to communicate to us before they decided what they were going to work on, and then communicate with us and others after they've done it, to see that we now understand what they've done and get benefit from it.
Yes, transfer, so to speak.
And IDA was totally unresponsive and uncooperative, because IDA saw them as a political hot potato, something to be held, not hurt, but don't let it interfere with the rest of IDA. And by the way, the rest of IDA wasn't all that enthused about them. The rest of IDA saw the Jasons as being the favored children in the family, while they did all the hard work, so there was a certain amount of internal dissatisfaction with these people among the rest of the IDA staff. So I was interested in some way of bringing further pressure to bear to achieve these management objectives, and I repeat that these were management objectives. At no time did I ever try to tell them what to conclude, or tell them — this was not a question of using them in any sense, just do sensible important things, rather than things that someone else has already solved but you don't have access to, or they're not the most important problems, or you think this particular bit of science is really hot but frankly we'd prefer that some other thing that we think is terribly important be pursued.
It would have effect on what they were doing.
It would. It would have the effect of turning some things down and turning some things up. I will admit that there's a very difficult problem here, because after all, if you got back to their early history, especially as explained by Herb York and other contemporaries, they were brought in to be very smart and to tell us on the inside what needed doing. The notion was that they weren't a bunch of hired hands, they were a bunch of outside people who brought viewpoints, and I often view that on my own consulting, my own committee work, that is, "Hey, I'm here to tell you what you don't understand from being inside the system, because I'm outside the system." So it's a difficult sort of thing. Nevertheless, at that point I felt I knew more than they did, about what was needed, and I proceeded to do that. Anyway, IDA wasn't being very helpful, that is, was not assisting me in this.
McDonald was still the vice president?
No, by that time he had left and had been replaced by an awfully nice guy, but who only did staff work.
OK, it was not Montro (?), he was not a Jason any more.
No, he was not at Jason any more. He did have a PhD. His name was Joel Bengston. Anyway, he was good, but again, he wasn't going to get out on a limb or anything. So at that time, we had a totally stupid problem arise, one of those typical things that come up in Washington that results in far more attention than it merits intellectually, and that was the notion of “ceiling” — ceiling for these organizations called federal contract research centers, of which IDA was one. Now, for a set of reasons that would take us the rest of the night and I still wouldn't be able to explain it rationally, some organizations were called federal contract research centers and some others were not called federal contract research centers.
That was from before.
Right. And the Congress said, "Defense Department, you can only put so much money into federal contract research centers." So here is IDA, that has a full time staff and does very good work and is very well appreciated by both Joint Chiefs of Staff as well as the Office of Secretary of Defense, and there's this, by this time it was probably 700 thousand dollars a year going into the Jasons. Well, since IDA wasn't managing them anyway, it was simply a holding company for them, I conceived the notion of, I'm not sure that I did, maybe someone suggested it to me — why don't we move the Jasons out of IDA, and then since nobody will notice their departure, we'll have six or seven or eight hundred thousand dollars to spend in that federal contract research center, and we'll attach the Jasons to some other place that is not so designated. And so Stanford Research Institute was one of those. It had separated from Stanford University in the late sixties, in that period of unrest in the US. And in any event, leaving all that, that was another bit of nonsense that bedeviled me, but in any case —
You were involved in that too.
Yes, I was involved in managing the balance of activities in FCRCS, where ceiling was tight, and other organizations that could do it without needing “ceiling”. And in any event, so, we looked around and we asked, who would like to take on the Jasons under the same terms and conditions, with the understanding that you'll manage them a bit more? SRI agreed to do that.
Yes. Of course, ARPA and IDA had the contract, about Jason, from the early days.
So are you saying that contract wasn't sufficient to keep the Jasons in place or some way, or is it not as simple as that? Did you want a different contract, in other words?
Well, it wasn't so much a contract. We had a contract with IDA for let's say five million dollars, of which 500 thousand is the Jason Division, and meanwhile of course we wanted to put more and more work in with IDA — inflation, you had to pay more each succeeding year for the same amount of work — so the notion was, we'd keep the IDA contract for five million dollars, but now the money that was previously consumed by Jason could be consumed by IDA. We'd take other money and give it to Jason. So it was a way of moving Jason out from under the FCRC ceiling umbrella. It was a purely bookkeeping thing. But along the way, as we shopped around for an organization to take these people, we wanted them to exert more management oversight than IDA had. It wasn't easy to find. We did not exactly have a long line of people at our door who eagerly wanted to take over the Jasons.
No. Batelle was in question, I remember.
Batelle was I'm sure considered. I've forgotten now. But there were very few. It was probably Batelle and SRI.
Actually John Wheeler left Jason at that time, because he was a trustee of Batelle. He thought that would be a conflict of interest for him. So it must have been serious.
It must have been serious. I had forgotten that, but that certainly would be a possibility.
There's also the possibility of Jason becoming an independent association.
Yes, I'm sure we considered that, incorporating them as it were.
Actually there's a report asked for by Hal Lewis and written by an IDA staff person, I think Weiner was his name, Ronald Weiner, who wrote out, you know, the possibility of making Jason an independent entity.
Yes. OK. Never saw it, didn't know it happened, but it's a logical enough thing to do.
OK, we were talking about SRI.
So that's how they happened to transfer over to SRI. SRI subsequently exerted some degree of management, not as much as one would have hoped, but moves in the right direction.
But there was a change.
There was a change, and you know, whether that was directly as a result of my urging, or the result of SRI's intervention, — I will say that in this period, as far as I'm concerned, the Jasons became absolutely useful and well managed. That is, larger spread of people, younger people — just the whole thing changed. They would talk to us about what was important, what they were going to do. I would assemble the right group of people, including all the ARPA office directors, to listen to the results of the summer study. I will say that by the time I left in 1974, as far as I was concerned, that was a good group of people. Smart, effective, working on good things, well coupled in. Now, maybe of course one tends to be pleased or satisfied with something you've done. But I would say certainly the last time I saw them, they were doing good things.
Younger, changed, yes, really quite good.
You definitely put some pressure on them to change. The letter here from you to Al Flax who was the president of IDA, pressing for discontinuation of the relationship of Jason with IDA, and I think it was discontinued 28th of February, 1973, I think. And I think Jason was hanging in the air without any institutional affiliation for a period, and then was taken by SRI before the next summer study.
Something like that. But this letter, you certainly put some pressure on them.
May I look at it?
Yes, of course. I found that in Charles Townes' papers, I think.
Yes, see, here's exactly what I — "It would clearly be in our mutual interest to stop the expenditure of IDA funds to cover Jason costs." Yes. "Since ARPA cannot find any further funds, each dollar spent for Jason must be subtracted from the funds available for other IDA research tasks. I'm particularly concerned that adequate funds might not be available for the planned electronic... "Yes. So "in our judgment the transfer to a new support organization would be of mutual benefit to IDA, ARPA and the Jasons personnel." Yes. And then here's where we were commenting on some of their ideas, and — yes, that's — Joel Bengston, that's the fellow who took over from Gordon McDonald as the IDA staff person.
Oh, did he? Yes. OK.
Right. Yes. We're now talking about two different things. He was the executive secretary for Jason.
Yes, but he was, when Gordon McDonald was there. Gordon had the rank of vice president of the Jason Division. When he left, he was replaced functionally but by a person of lesser organizational rank, that was Joe Bengston. It was just the title was changed.
OK. All right. I don't know, we have covered a little bit about how the choice of problems for Jason has been done. You were saying that Jason essentially did that itself.
In the beginning they did it themselves, and toward the end ARPA was a participant in that process.
Yes, OK. How did ARPA become a participant? What were the mechanisms?
Well, you have to go back and look at the Jason cycle. The Jason cycle was the following. They had a six week summer session, and that was the working, that was when they interacted. They then would go off and write and talk and have smaller meetings, but that was in some sense the annual coming together. OK, that's when the thoughts were exchanged and the stimulation was mutual and so on. And it would be directed to several subjects, whether they picked them or whatever. Now, what we would have would be, in the fall, we would have a meeting to "hear the results of the previous summer’s study," like say October, at which point we would say, "Did you consider this? That's very good. This is what we ought to do next year." So there was a certain continuation. Then somewhere around the fall or maybe at that same meeting, we would start to talk about subjects for next summer, some of which were continuation, and some of them were perhaps new subjects, which they would then do some thinking about and talking about and so forth. Then we would have a meeting perhaps around March or so, and they'd say, "Well, here's what we're going to do." And we would say, "Well, how about this, or less of this and more of that," and we'd kind of influence them a second time, give steering signals. And then in the beginning then around April or May, they'd start to get briefings in Washington, by government people, to equip them to focus on these new subjects, and they'd sort of do their homework, as it were. So in essence we were able to steer them from the course they were on to this course, and then, perhaps have additional conversations, because they'd meet in sub-sets and they'd meet with us and we'd meet with them and so on. So we tried, in other words, very much like steering a boat — you don't just put it on autopilot and you automatically get to where you're going. It's the wind here and the weather there and some steering here and some signals there. So we sort of tried to pilot them, recognizing that we didn't know everything, and they ought to come in and tell us some things. And they would come in and say, "Do you know about X?" "No, we don't know about X." "Well, we think we ought to look into it." "Well, that makes sense." So I would say it's as two way process. I don't know what they say, but I would say it was a shared process.
You would essentially say that you established communication where there had been none before.
Exactly. And in my view, really, that communication channel was most useful for getting from them what they had done, frankly, rather than telling them what to do. My feeling was, if you're that smart and you've done some good things, we'd better find out about it. But on the other hand, it also says that if you have X units of manpower available, we'd rather like some of it to be spent on this and this, rather than all of it on some other things of less importance, recognizing that they would be coming in and telling us some things. So I think we hit an equilibrium there, an intellectual equilibrium. I hope we did.
Yes. I'm glad to get this from your side, because I haven't sensed that discontinuity there, before. I think on the surface at any rate, you had that meeting structure before your time also. I mean, you had the summer study, the fall meeting and the spring meeting, so that you put substance into a surface that was already there.
I used it for communications, yes. Two way communications, us to suggest what they might be doing, and them to really tell us what they were up to. Which previously had been — those fall and spring meetings had been just Jason meetings, for Jasons, although they would invite government people. Again, an amusing story on this. It would be typical, the Jasons, the IDA board room which is where they used to meet has a table about twice the length o this table, maybe three times the length of this table, and they'd be sitting around, maybe 15 or 18 of them or something, and some eminent government personage would be briefing them, Assistant Secretary level or something. And it would be quite common for them to be sitting there reading a newspaper, not paying any attention to the briefer. If I want to read a newspaper, I'm going to read a newspaper. If you aren't very smart or you aren't saying anything of interest — or if they knew that, or if they didn't agree with you, they just read the newspaper. It was that kind of a group.
Actually, that reminds me of, at the Jason 25th anniversary, there was a play acted, and with younger Jasons acting as older Jasons, and part of it, I think the main theme is exactly that kind of meeting, with people actually reading newspapers or eating ice cream or sleeping or whatever.
I don't know if you've seen that.
No. But in any event, it was very much that — I thought they had terrible communications with the bureaucracy, not just with ARPA, but with the bureaucracy. In the government they were supposed to be helping, there was not a lot of mutual respect. They were viewed as, however they were viewed.
What time frame are we talking about now? Are we talking about the change taking place simultaneously with the transition to SRI?
Yes, it was all at the same — to put a time scale on it, I came on the scene in 1966, and this little vignette that I've been going through is typical of the '66, '67, '68 time period. 1967, I went into the ARPA front office, probably focused on the Jasons in '68, thought more about them and learned in '69. By '69 and '70 I was suggesting management changes. The ceiling thing was in '71, and by ‘72-‘73 we made the transition, so I'm largely talking about '67 through '73. That was the period when we went from the old style to the newer style.
How important was the SRI stronger management of it? As part of this transition?
It was a significant factor, simply because in IDA, they knew they had a sort of benevolent owner, very indulgent of their ways. They had their own vice president at IDA. They used the IDA board room for their meetings. They had the IDA management thoroughly cowed, and in its place. SRI was a totally new game. It didn't have the class of IDA. It didn't do things with the same style. They certainly didn't have any in with the head of SRI. So they were already one or two levels down in the SRI structure. So it must have been a very disorienting process, because now they have an ARPA who's trying to manage them, and they have a home organization which isn't very classy, and is also helping to manage them, and they have a bunch of new members, and some of the old guard has left by this time for one reason or another, and so yes, that must have been a very difficult period for them. Because it was almost a brotherhood, a fraternity, and the family setting changed dramatically.
Hal Lewis stayed chairman, I think, through the transition. Somebody told me that he left as a result of the transition or something to that effect. Was there anything personal in that too, I mean, in your relationship with the chairman?
As a matter of fact, it seems to me one of the management changes I suggested was that the chairman of the Jasons ought to change too.
Yes, more often. It ought to be a fixed term appointment. It could be renewed, but it ought to be a fixed term. This way it was basically a hereditary title. Yes, I felt that the change in the chairman — I'm not sure whether it was pressure from me or dissatisfaction with Hal, or a statement on his part, SRI pressure, but somehow or other it happened, and I think it was all to the good. When you're changing things, you have to change things.
I hope we'll get to some projects and examples of how projects were chosen and all that. But you started talking about choice of people, I think, and that it was getting too old an organization, during your earlier time as head of ARPA. To what extent was the change of membership formalized, as a result of the transition, or as a result of other changes?
It was probably not so much formalized as pressure, one would ask them, who are you considering for new members? Are you looking into area X or area Y? How many people didn't come to any meetings over the past year? Are you going to replace them? It was that kind of a thing. It was never management by the numbers.
Yes. I mean, there was for example a two year trial period. I don't know when that was started.
Oh, that was when they started to bring new members in, they weren't at all sure whether the new members would be right, committed, compatible and so on. So they had the notion, let's bring them in for two years and then if they work out we'll ask them to stay. Most committees do that anyway. You get re-appointed to committees continually, but if it doesn't work out, you just don't get re-appointed. That was all part of it.
What about the choice of members? Was that still something done by Jason itself?
Oh yes. We would not presume to name names or tell them who should join their fraternity. No, they just had to show good faith, that they were managing their membership. That was all we wanted.
But the membership still remained essentially physicists, I think. There was not —
They brought in some computer scientists. Yes. Now, along this period, by the way, '68, I guess, '69, I ought to mention a group called the Materials Jasons. Have you run across them?
Yes, in some papers.
OK. Well that was an explicit recognition by, in this case, the materials science office in ARPA, that they needed that kind of people, that Jasons weren't interested in materials problems, that Jasons had by and large little or no capability in material science, —
— it was less prestigious too, I suppose, material science.
Material science is less prestigious. You're right. So they said, "Look, the Jasons are fundamentally a good idea structurally. That group will never do for us what needs doing. We'll form our own." So they did, and I helped them in that process. Fred Seitz was a major factor in getting that off the ground. It was very dicey, because as I recall, it happened in the spring of '68, and that was the time when the university campuses were exploding, and here was Fred Seitz who was going to somehow keep this bunch together and committed to us, and it was really quite a different time — there were one or two defections. So there was a case where clearly we all recognized that Jason-like organizations are very good. And sometimes you just have to start a new one rather than modify an old one. So we basically did two things. We started a new one in an area which they simply couldn't perform in, and we improved Jason’s operation and performance in an area where they could perform.
Yes. That might be one of the very few organizations that could provide a comparison.
There aren't many of them.
No. [This is not to change the meaning actually, I was agreeing and “no” expresses that better, in hindsight.]
So what would you say were the differences? Of course, it was created during your period, so it was more according to your image, I suppose.
Well, in all honesty, it was the material science office that did it. I will not take a lot of credit for it. It was a fellow by the name of Rob Thompson who was the head of the material science office, who really was the sparkplug behind the idea. But nevertheless, there were some commonalties. There were about 30 or 40 people. They were more industrial people, because obviously a lot of materials is done in industry instead of just universities. An intense interaction on the selection of subject matter. Not quite as long a summer study. I think it was maybe only a two or three weeks summer study instead of the Jason six weeks. And really focusing on writing reports, and coming back, and the materials office was just as happy as it could be with the results of that.
So was that as high up as Jason? Did they report as high up?
Or was it lower level?
That tended to report in at the material science office, although the front office tried to pay attention to them, and I went to a couple of their summer studies, but frankly I was more interested in both defense policy and computing and nuclear test detection and arms control than I was in materials, so I probably wasn't as faithful. But that's right, it was one level lower.
Yes, and it didn't have as high pretensions as Jason did.
It didn't have the class. No. [Again, one can agree with either “yes” or “no” — curious nature of language.]
What happened to Mason anyway?
I think they're still going on. As far as I know, they're still going on. But you might want to look into it, depending on how your Jason study goes, it might be useful to focus in on it. As nearly as I can tell, I hear about them occasionally. They seem to be quite good.
More low key than Jason.
Yes. I will say one thing, that is worth pointing out, since this is a general lesson to be learned. Remember, the beginning of Jason and ARPA, the story that Herb York tells from the late fifties, when all this started — that was an image of science advice to government. The notion was, smart people are outside committed to the purposes of the government, people inside who do not have the time or the intellectual depth to understand all these things, but have a willingness to receive ideas from the outside, and a willingness on the outside to pour this advice into the recipients inside. And in the last twenty years, 1958, last thirty years, that situation has totally changed. The people in government see themselves as quite capable. In many cases, they are. Look at this place at Rand. This was formed because the United States Air Force said, "We don't know enough about how to lay out the future of the Air Force. We want you people to help us." Well, nowadays Rand has trouble getting the Air Force to listen to its advice. Rand can say, "The best thing to do is X, Y and Z, " and the Air Force says, "Forget about it, I have my own plans." So there really is not the willingness to open up your inner workings to the advice of some outside people. And so there's been an absolute change over the last thirty years, I feel.
The need might have become less, too, because you have got more expertise within the agencies, within government.
That's a common thing that 's said. I'm not sure that it's correct.
Right. But the basic motivation for setting up Jason in the first place was to set up a relationship between the younger generation of scientists and the government.
That's right. And that was a continuation of the scientists who piled into World War II and did wonderful things, like radar and jet aircraft and proximity fuses and nuclear weapons and operations analysis. All that stuff was done, and so they were going to keep doing it. And now we feel that we have other mechanisms for figuring out what we want to do and how we want to do it, and now the focus is on conflict of interest and special access programs and security clearances, and just a whole lot of things that tend to impede the flow of information and advice and assistance to people inside. And now there's a very high wall between the two.
Your implication is that that's bad.
Yes, that's right. And yet I'm explaining some of the mechanisms I was a part of, for bringing these unruly outsiders under control. So I recognize it — but that's the way the world is.
But they were still there. You would say that they were still giving advice.
They were still giving advice.
Even after your trying to get them in order and they were even more influential, after that.
I think they were, actually. Certainly in guiding ARPA programs. I don't think Secretaries of Defense still pay any attention to them. In fact, it would be interesting to know what was the last Secretary of Defense that recognized the name Jason.
Well, McNamara certainly did.
McNamara may be the last one.
Harold Brown, I suppose.
Harold Brown would have simply by virtue of his past associations. Clark Clifford probably didn't, and beyond that, I don't think Laird would have known the Jasons if he'd tripped over them.
I said before that we should probably talk a little bit about projects that Jason took up, maybe choose one or two to look at, their impact, and how it was chosen, how it was worked on, how the relationship with government agencies at what level took place, during that stage, how the presentations were presented, how they were taken up, etc. I don't know if there's a good case of that.
I've thought about this because I knew you would want to know that. As far as I know, there's only one example that I can give you, and that's the so-called McNamara Line. It's now 1964, '65. The (Vietnam) War is heating up. The technologists are still very much in their World War II mode of helping the government. There are computers, miniaturized electronics, radio communication and at that time the prevailing notion in the case of "the insurgency" is, if we can just keep the outsiders out, the citizens of the country inside will sort out their problems. That meant essentially a barrier around South Vietnam, or particularly the DMZ, but it was later recognized, it had to cover the Ho Chi Minh Trail. So the notion is, well, let's simply seal off the country. We'll make a barrier. We'll make sure people can't get in, supplies can't get in, and then those who are inside will achieve their political, social, military equilibrium with the insurgents. And so the Jasons put that notion together, as near as I can tell — this was before I came aboard, but only a year before I came aboard. And they sold it to McNamara and he said, "Let's go." They set up a new organization called DCPG, very hush-hush, Defense Communication Planning Group, but it was really to implement that barrier, and that was a direct implementation. Lord knows how many hundreds of millions or even billions of dollars it cost before it was through. And it turned out that it never had the desired impact on Vietnam, but it did have the following major impact. It was the first time, when we began to see, since nuclear weapons and since ballistic missiles, a totally new technology, that we now called smart weapons. Now we have a whole new jargon in the defense business — target acquisition, smart sensors, internetting, this whole idea that there are sensors and there's communication and we'll sort of follow everything and we'll direct weapons and so on, this whole idea of warfare as something other than hauling ordnance from the factory to a delivery point, and massive amounts of ordnance being delivered on the heads of the enemy. That was the paradigm of war, up until the so-called McNamara Line. And I would argue that all of the current focus on sensing and internetting and smart this and unmanned that and so on, are all basically, not necessarily linear descendants in a systems sense, but certainly in an intellectual sense. It all came from that.
That's a pretty strong influence.
That's right. And in fact, they did it all themselves. This was when ARPA wasn't guiding them. Now, it turns out, in detail it didn't work out right, but that doesn't conceal the essential importance of what they did.
The implications are larger than that particular piece anyway, that's what you're saying.
Most Jasons say, I think with some right, that was a very untypical Jason project, though. They admit themselves that they had higher connections there than they had in other projects. I think they had direct connections with the Defense Department, McNamara.
Yes. They briefed the Secretary of Defense on it, that's right.
And they also collaborated with more prestigious older scientists, the MIT crowd, Kistiakowsky, before he quit, Wiesner, you know, DSPG.
DCPG, Defense Communications Planning Group.
That was a broader group than consisting of Jasons.
Oh yes, that was a major military organization, to procure, install, and operate these sensors in Vietnam. They worked in lots of other areas. Frankly, I'm hard-pressed to even identify an accomplishment. I know what happened after we started driving them. They did some very good work in the underwater sound area. They were of assistance in — we used to use them to think about new areas of technology. It didn't always work, but that was fine. Communications through the earth's lithosphere. Metallic hydrogen. There were a number of subjects that I was very pleased that they'd done some thinking about.
Yes. But were some of these continuations of earlier Jason interests?
None that I just mentioned. As far as I know, they were not. Now, they always had a continuing interest in decoupling nuclear explosions. There were a couple of sort of longer term interests. But I think these were new subjects. Could I see this for a minute, list of topics?
Sure. From the meeting?
That's your letter.
We looked at it a couple of minutes ago. This one. Let me just see.
This is the report of the meeting from... (crosstalk )
— well, let's see now, I'm just looking for reminders of — no, I don't — here it is. Underground — but in any event, it's immaterial. I was looking for some memory joggers. You had a letter — some of that?
Here's a list of their activities.
Right, air traffic. They went out and got some money from the FAA. They got money from CIA. Particle beams, they always had an interest in See-Saw, that's what that is,—(crosstalk) that was a natural. And ASW. I thought they did some very good work in the ASW area.
You know, I'm looking for a representative kind of project or set of projects, to look at as a case study of a case study, so to speak, to see how the project was chosen, how it was worked with, in briefings.
Well, there's a very definite, I think, pre and post 1967. I think that should be viewed as a dividing line in your viewing of this. Before '67, I didn't have a lot of association. Again, I had very little knowledge of them, but the McNamara Line, the battlefield sensors, ranks as just a major result. After that it really becomes less a more scientific issue. I guess this was another point: they always saw themselves as telling Secretaries of Defense and Presidents, giving them advice. Well, Secretaries of Defense and Presidents by and large are not consumers of the advice of theoretical physicists, if they're dealing with science. So that after 1967, as we went on through '69, '70, the later period, they came to accept ARPA as technical experts. In that sense, there aren't any flashing lights on the landscape. That doesn't mean that they weren't very useful. They're not revolutionizing the Defense Department any more. They're providing technical advice, as is appropriate to highly skilled scientists.
That's exactly the development of ARPA generally speaking. It started out with Presidential issues.
And you know, revolutionary kind of work, high risk projects, and then when Rechtin and then you became directors, it was more —
The President wasn't interested in ARPA. That's right. The last Presidents to be really interested in ARPA were Jack Kennedy and the counter-insurgency business and Johnson in the 1968 ABM debate. ARPA was important in the sense of what was going on in ARPA was the entire debate. The technical basis for the 1968 debate was the result of ARPA work until that time. After that, in other words — President Nixon really didn't know anything about ARPA and didn't care about it, so that represented a big change, 1968.
So to what extent is this really a general trend, I mean of ARPA, of expectations of science advice, and to what extent is this something peculiar to Jason?
It's not just peculiar to Jason. It's general. Vietnam caused alienation of many parts of the intellectual community from national security matters; there was through the seventies an increase in special access programs in the Defense Department, so fewer people know what's going on — even people in government. There was a focus on policy issues, not heavily influenced by contributions from technical people. You can even argue that most of the discussion of technology transfer Co Com and so on do not have a heavy influence by the technical community. So the technical community has tended to be pushed off center stage for these policy debates. I think this is a general trend, and you see it in the bureaucratization of ARPA. You see it in the "do your technical things and we'll take care of the policy." And you can see it in the Defense Department, in its various changes. Now, some of that was Vietnam. Some of it was Watergate. There have been two major alienating influences in the late sixties, early seventies, which clearly are there. Then there was the Ford Presidency, the unsuccessful Carter Presidency, and so. It's all part of a trend there, of which the Jasons not as important now, but that doesn't mean they're not useful.
I have problems distinguishing between technical advice and science policy, and when I talked to several Jasons, they always correct me when I talk about Jason as a science policy group.
They're technically an advising group. Now, in the sixties, what you're saying, I suppose, is that technical advice and science policy were much closer. It was the idea, I think, Lee Huff says that too very nicely, that it was presumed that science policy decisions could be made on technical grounds, and that they flowed directly from technical advice.
So Jason could be seen as a science policy group in that respect.
And then those two things have grown apart.
And there's less of a pretension to —
—Yes, although even then, Jasons were definitely organized to give technical advice to the Defense Department. They were there because they were smart technical people, not that they were smart people. They are smart people, but they were there because they're smart technical people, and they therefore didn't discuss matters of the all-volunteer force or the state of NATO or Pacific basing and the like.
You could make the distinction then, too. But some Jasons did, I think, attempt to do that or wanted to do that.
Yes, for example, in the ABM debate, some of the stepped outside of their Jason or technical advisory role and said, "ABM is not a good thing." And no one really asked them whether it was a good thing. In fact, people got very irritated when people who had their technical credentials would then stand up and deliver a political or a policy judgment, which they were not particularly equipped to make. There is very much of a sense in Washington, as in all things, of one's proper place, and you are, if you are credentialed, if you are prepared, if you've done your homework, if you know what you're talking about, then you're welcome to stand up and deliver an opinion. And if you haven't thought about it and haven't studied it and don't know what you're talking about, we'd rather not hear from you. So I think the Jasons got themselves into a position certainly of delivering advice, and the technical community in general, about the war, for example. Now, it's very hard, after all, scientists are people too. That's trouble with that though. When you go to the town meeting and you talk about the pay scale for teachers, well, you ought to be listened to. You have children and you worry about the schools and so on. But that doesn't mean you're an expert in educational policy or in fiscal managing or anything. It's very hard to separate these roles, and we went through a period where people's roles — and we're still going through it where people's roles are questioned.
Of course it's hard to make the distinction completely. I'm sure that Jasons, even when they do technical problems successfully, make choices according to their political preferences in some ways.
Yes, but I think they've gotten away from that now.
OK. What would your general evaluation be of the impact of Jason, compared to say ARPA, generally speaking; compared to other kinds of advisory bodies?
Well, at the time the Jasons were invented, going back to the Herb York memoirs that I hope will appear in the files of all of this, because I think they're a very important insight into all of these questions, the Jasons were very important because that's the way we did things. And we would have been delinquent if we didn't do things that way. We would have been delinquent if we hadn't listened to them. As time went by, the government changed, for better or for worse, and so it wasn't important that we had Nobel Prize winning theoretical physicists advising us on how to defend the country. Maybe we should, but we've gotten away from that. ARPA, on the other hand, has become more effective over the years. ARPA is, I think, more effective now than when I was there. And my successor, although I differed with him on a number of points, did some very important things. He said, "Look — instead of supporting a lot of important science and technology activities, I want to go with a couple of big things." So he went with the Stealth bomber. He went with a smart weapon program called the Assault Breaker. He went with a big infra-red sensor in the sky to track aircraft and do air battle management and so on. And he focused on a couple of big things, and those are generally in the system. Some of them have failed. Some of them are sputtering. Some of them are wildly successful. But the fact of the matter is that ARPA, by moving closer to the military services, also by the way getting the military services more upset with them because they're now doing prototype development and weapons system development, all the things that the services have been afraid of losing control over, but that's been happening. The whole SDI office, as a matter of fact, the whole SDI office was invented because ARPA would not move into the next stage of development the things that ARPA had been doing. All particle beam and laser stuff, that's all been in ARPA for twenty years, and Cooper wouldn't do it, and Dick DeLaues wasn't enthused about it, so they invented a new ARPA just to do strategic defense. But in any event, this is to say that there are things happening in the Defense Department, and I would say the Jasons are largely irrelevant to this process. Not that they're not important. They are a small piece of the science activity of ARPA, and that's good, and to the extent that that community has important things to say, they are a mechanism for saying them, and the Defense Department in general and ARPA in particular need as many channels. It's like saying, if this were a control room, the more phones we have in it the better. People ought to be able to call up and tell you what's going on. So the Jasons are a line from one community into another that represent important intellectual groupings. They're just one of many, but they're there and they're effective, and they probably don't cost a lot of money. I don't know how much they cost now. It may well be up to a million and a half or something, but it's a perfectly reasonable price to pay just to have those people on tap. So they're more of a resource than necessarily, but you don’t have to ask everyday “what did you do for me?” That's not what you're supposed to ask. The point is, when there's a fire you're going to be glad to have that water system. So they are part of a process and a mechanism, and as such they are important, but the importance is in their existence, not their short-term results. Obviously they have some relevance over the long term, but you don't necessarily have to keep a box score on them.
No, no. It's — of course, I'm interested in them in part because they constitute that —
— That's right, they're more of interest as a typical part of a mechanism than as an entity in their own right. Yes. And they're a better recorded part than the run of the mill committee or study group or contract or something.
Yes. What about Jason as supplying a contact point between the academic physics community and government advising?
I think that's diminished with time. I think that in the early days, that was a major part of it, but that's not a major part now.
Of course, they're increasingly a smaller minority of the physics community too, because the physics community is growing and Jason isn't.
That's true too.
We talked a little bit about projects, of course. I feel that perhaps you haven't talked in detail about how you followed a project from the beginning to the end, so to speak.
A Jason project?
Yes, setting it up, following it during the different kinds of meetings, and then following the result into the deciding stages.
OK. My role in this is a bit of management philosophy as much as an answer to the question. My specific role was not important. Managers manage processes. The process is that 30, 40 smart people will get together under appropriate circumstances, will engage in some useful activity, and that the results of that activity will be transmitted to people who can use it. And so what I have to do is, as long as those people out there understand that this resource is available and come up with ideas, and as long as I see that they talk to the right people, I just have to stand up on one of the upper floors and watch the activity. So, while I may have a particular interest at some point, I'm really only representing one person, not as the head of the organization that's paying the bills. I never tried to have a heavy hand in that. What would concern me was, if they worked on poor problems, or didn't get any results, or nobody listened to good results, that's what would worry me. That's what I felt responsible for.
Yes. Is there anybody else like a program manager or something within ARPA that it would make sense to speak to about a specific project like that, seen from the outside?
That's interesting. I don't even know who you talk to any more. The deputy director of ARPA who just left, Jim Tegnalia used to meet with the Jasons. So I think what you'd have to do is go into the ARPA front office now, the director is Cliff Duncan, there's now two deputies, Bob Moore and Craig Fields, and you have to ask them. I don't know how it operates these days.
I'm essentially interested in the first 15 years anyway, so I would have to talk to people, well, maybe some old timers still in ARPA, I guess.
Might be some. But there are very few of them now.
Of course, there's a lot of change. You encouraged change in ARPA.
OK. Physicists in science policy, generally speaking, that's just to broaden it off — is Jason a good example of that? And to the extent that you're a physicist in a different role, what has the importance of physicists been in science advising and how as it developed?
Well, in the past, i.e. the years immediately after World War II, through the mid-sixties or so, physicists were very important people. The great names in World War II, especially those who stayed at government interactions, seem to be more physicists than anything else, not that there weren't chemists, and of course I don't know about the medical and biological world all that much, but physicists were sort of a major part of that. That's perhaps not surprising, because again, going back to my own view of going into physics because it's the way you understand everything. Then since the world involves everything, it's natural that the physicist is a discipline that would have a better view of how those things fit together. So I think it's natural that physicists had perhaps an easier time jumping from environmental things to energy things to aircraft to basic science to astronomy. I mean, that's the set of people who understand those things. So I think there is a more central role. For example, we had a National Academy of Science for a long time before we had a National Academy of Engineering, and for a long time even now the National Academy of Engineering doesn't have the stature of the National Academy of Science. So, the physicists have always been very prominent in that area. Now, I would argue that today probably the people who can supply more insight and advice to society are the computer people, and yet you don't see them nearly as represented in the councils of government as you still see physicists. So, I think, it's an interesting period, 1945 to 1980 or so because elder statesmen of science were physicists. There was the World War II physics story. Admittedly a fair amount of electrical engineering, and to an extent the electrical engineers gave rise to computer science, then one could argue that perhaps that's the way it all worked. But electrical engineering and physics, taken together have had major impacts. One doesn't quite see it these days, I think. I think the physicists have tended to be a liberal opposition. First there was the Vietnam war, and the SDI, and there's so many of these new questions, that physicists as a group seem to be leading or at least prominent among the opposition, "This won't work because, that's not a good idea because, that will take too much resources from this," although I notice that when it comes to the Superconducting Supercollider, they're there with their four billion dollar hat out. But I think they're just another pressure group. So I think they were, and still are probably better qualified than many of the sciences, by virtue of the sort of underlying character of the discipline. Where questions involved engineering questions, they haven't done very well at all. For example, the arguments about the SST and ozone and all that kind of thing. Maybe SSTs aren't a good idea, but it wasn't for environmental reasons. So I think they have a tendency to see their insights as being dominant. What we've seen is in policy circles, the economists have moved in. The dominant paradigm of thought today is the economist's, not the physicists', and that's really what's happened over the last few years, and it's because public policy ideas get wrapped up in questions of utility and allocation of resources. Physicists are not particularly good at questions of utility and allocation of resources. They're good at questions of technical feasibility. And so as the world has gotten more difficult and requiring tighter management, the physicist has tended to fade, and other groups of people have come to the forefront — as well as, of course, the dominant influence of computers, software, communications, and those are matters that by and large physicists have not been in the forefront of. In fact, for a long time they didn't even use computers. If you were in the physics department and you computed, you were a second class physicist, because physicists only did physics.
Yes. On the other hand, physicists don't want to use computer experts for their problems. They want to do it themselves.
That's right. Now what physicists are beginning to learn is that computers are convenient places to do experiments. There are many models that you can conceive and simulate. So now, for example, things like the hypercube work at Caltech are being pushed very heavily by Mark Fox because it's a simulation tool. And so physicists are discovering this new laboratory. But they're really behind lots of other people in that regard.
Yes. So they're becoming users rather than —
Than being used. OK, good.